University of Miami
Scholarly Repository
Open Access Dissertations
Electronic Theses and Dissertations
2008-09-30
Transcriptional Regulation in the Peripheral
Nervous System and the Role of STAT3 in Axon
Regeneration
Robin Patrick Smith
University of Miami, [email protected]
Follow this and additional works at: http://scholarlyrepository.miami.edu/oa_dissertations
Recommended Citation
Smith, Robin Patrick, "Transcriptional Regulation in the Peripheral Nervous System and the Role of STAT3 in Axon Regeneration"
(2008). Open Access Dissertations. 302.
http://scholarlyrepository.miami.edu/oa_dissertations/302
This Open access is brought to you for free and open access by the Electronic Theses and Dissertations at Scholarly Repository. It has been accepted for
inclusion in Open Access Dissertations by an authorized administrator of Scholarly Repository. For more information, please contact
[email protected].
UNIVERSITY OF MIAMI
TRANSCRIPTIONAL REGULATION IN THE PERIPHERAL NERVOUS SYSTEM
AND THE ROLE OF STAT3 IN AXON REGENERATION
By
Robin P. Smith
A DISSERTATION
Submitted to the Faculty
of the University of Miami
in partial fulfillment of the requirements for
the degree of Doctor of Philosophy
Coral Gables, Florida
December 2008
©2008
Robin P. Smith
All Rights Reserved
UNIVERSITY OF MIAMI
A dissertation submitted in partial fulfillment of
the requirements for the degree of
Doctor of Philosophy
TRANSCRIPTIONAL REGULATION IN THE PERIPHERAL NERVOUS SYSTEM
AND THE ROLE OF STAT3 IN AXON REGENERATION
Robin P. Smith
Approved:
________________
Vance P. Lemmon, Ph.D.
Professor of Neurological Surgery
_________________
Terri A. Scandura, Ph.D.
Dean of the Graduate School
________________
Abigail S. Hackam, Ph.D.
Assistant Professor of Ophthalmology
_________________
John L. Bixby, Ph.D.
Professor of Pharmacology
________________
Robert W. Keane, Ph.D.
Professor of Physiology and Biophysics
SMITH, ROBIN P.
Transcriptional Regulation in the Peripheral Nervous
System and the Role of Stat3 in Axon Regeneration
(Ph.D., Neuroscience)
(December 2008)
Abstract of a dissertation at the University of Miami.
Dissertation supervised by Professor Vance Lemmon.
No. of pages in text. (203)
Several factors contribute to the failure of the central nervous system (CNS) to
regenerate after injury. These include inhibition of axonal growth by myelin and glial
scar associated molecules, as well as the intrinsic inability of adult CNS neurons to grow
long axons in environments that are permissive for younger neurons. Neurons in the
peripheral nervous system (PNS) display a much higher capacity to regenerate after
injury than CNS neurons, as shown by conditioning lesion experiments and by
microtransplantation of dorsal root ganglia neurons into CNS white matter tracts. Our
central hypothesis is that neurons of the PNS express specific regeneration associated
genes that mediate their enhanced growth response after injury. We have employed a
combination of subtractive hybridization, microarray comparison and promoter analysis
to probe for genes specific to neurons of the dorsal root ganglia (DRG), using cerebellar
granule neurons (CGN) as a reference. We have identified over a thousand different
genes, many of whose products form interaction networks and signaling pathways.
Moreover, we have identified several dozen transcription factors that may play a role in
establishing DRG neuron identity and shape their responses after injury.
One of these transcription factors is Signal Transducer and Activator of
Transcription 3 (STAT3), previously known to be upregulated in the PNS after a
conditioning lesion but not known to be specific to the PNS. Using a real time PCR and
immunochemical approaches we have shown that STAT3 is constitutively expressed and
selectively active in DRG neurons both in culture and in vivo. We show that the
overexpression of wild type STAT3 in cerebellar granule neurons leads to the formation
of supernumerary neurites, whereas the overexpression of constitutively active STAT3-C
leads to a 20% increase in total neurite outgrowth. It is hoped that the genetic delivery of
STAT3-C, potentially combined with co-activators of transcription, will improve
functional regeneration of CNS axons in vivo.
DEDICATION
To my mother Rosemary for her support up at the farm, my father Victor for his
words of wisdom, and my sister Victoria for keeping me grounded. And to Jeff and Paul
for keeping me company in the off-hours.
iii
ACKNOWLEDGEMENTS
The work contained in this thesis was made a possibility by the generous
contribution of several colleagues from our lab as well as from around the University of
Miami.
I would like to thank my mentor, Dr. Vance Lemmon, for his continuing support,
persistence and late night emails throughout the past six years. His enduring trust in my
ideas and intuition, even when they resulted in some upheaval, has been a source of great
happiness during my tenure. I would also like to thank Dr. John Bixby for lending his
critical eye to my various projects and offering creative advice and suggestions.
From the lab I would like to thank Murray Blackmore, who was always available
for a discussion about science and provided insightful suggestions on the presentation of
my data. My fellow grad student Willie Buchser has been a great friend and colleague
over the years, and I thank him for his tireless efforts to improve the way we do science
in the Lemmon lab. Much of the early work in this thesis was made possible by Jose
Pardinas, who taught me a lot about molecular biology and is to this day a great friend. I
would also like to thank Eli Weaver, who wrangled all of the mice used in this
dissertation whilst providing sage-like wisdom on a variety of topics.
From outside of the lab I would like to thank Dr. Sawsan Khuri, for providing me
with the Transfac database and her critical reading of my paper. I would also like to
thank Dr. Pantelis Tsoulfas, who has been a constant source of ideas, antibodies and
critical advice.
iv
TABLE OF CONTENTS
Page
LIST OF FIGURES AND TABLES ...........................................................................
vii
ABBREVIATIONS .....................................................................................................
ix
PREFACE
..............................................................................................................
xii
PUBLICATIONS AND MANUSCRIPTS IN PREPARATION ................................
xiii
Chapter
1 GENERAL INTRODUCTION ......................................................................
1
Injuries to the nervous system………………………………………………..
1
Barriers to regeneration……………………………………………………….
3
Improving outcomes after spinal cord injury…………………………………
6
The Conditioning Lesion…………………………………………………….. 10
PNS vs. CNS……….………………………………………………………… 13
Hunting for regeneration associated genes…………………………………… 16
Chapter Tables ................................................................................................. 23
2 AUTOMATED ANNOTATION OF PNS SPECIFIC ESTS ........................
Summary ..........................................................................................................
Background ......................................................................................................
Implementation ................................................................................................
Results and discussion .....................................................................................
Conclusions ......................................................................................................
Availability ......................................................................................................
Chapter Figures ................................................................................................
24
24
24
25
29
31
31
32
3 PNS SPECIFIC TRANSCRIPTION FACTORS AND STAT3 ......................
Summary ..........................................................................................................
Microarray analysis of laser capture microdissected DRG neurons ...............
Ontological clustering of DRG enriched genes ...............................................
Promoter analysis of DRG enriched genes ......................................................
Identification of DRG enriched transcription factor interaction networks ......
JAK/STAT pathway profiling by real-time PCR.............................................
Stat3 expression and activity in vitro ..............................................................
Stat3 expression and activity in vivo ...............................................................
Expression of Stat3 mutants in cerebellar granule neurons .............................
Overexpression of Stat3 results in supernumerary neurite formation .............
37
37
39
42
44
47
49
53
54
56
58
v
Overexpression of constitutively active Stat3 increases neurite outgrowth ....
Materials and methods .....................................................................................
Chapter Figures ................................................................................................
Chapter Tables .................................................................................................
60
61
69
92
4 GENERAL DISCUSSION .............................................................................
Candidate PNS specific transcription factors ..................................................
Stat3 in DRG neurons: survival factor of growth checkpoint? ........................
Potentiating Stat3 activity in cerebellar granule neurons ................................
Chapter Figures ................................................................................................
Chapter Tables .................................................................................................
5 SUMMARY AND FUTURE DIRECTIONS .................................................
163
163
168
172
176
177
179
WORKS CITED…………… ...................................................................................... 183
vi
LIST OF FIGURES AND TABLES
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
32
EST Express Data Pipeline
33
EST Express PlateViewer
34
EST Express Project Viewer
EST Express “New Genes” Tool
Results of analyses of the subtracted data set
35
36
69
STAT3 signaling pathway
In silico microarray analysis of genes differentially regulated
between the PNS and CNS
Overrepresented ontological annotations describing DRG enriched
genes
Overrepresented mammalian transcription factor binding sites in the
promoters of DRG enriched genes
Identification of transcription factor centered interaction networks
DRG enriched transcription factor sub-networks
70
71
72
73
74
75
Purification of DRG neurons
Upstream regulators and downstream targets of STAT3 are enriched
in DRG neurons
STAT3 is highly expressed in DRG neurons
DRG neurons express higher levels of total STAT3 than CGNs in
vitro
DRG neurons express high levels of nuclear Y705 phosphorylated
STAT3
STAT3 is expressed in cerebellar astrocytes in vitro
DRGs express higher levels of STAT3 than cerebellum and cortex in
vivo
STAT3 is expressed in DRGs in vivo
STAT3 is phosphorylated on tyrosine 705 in vivo
vii
76
77
78
79
80
81
82
83
Figure 21
Figure 22
Figure 23
Figure 24
Figure 25
Figure 26
Figure 27
Figure 28
Figure 29
Table 1
Table 2
Table 3
Table 4
Table 5
Table 6
Table 7
Table 8
Table 9
STAT3 is constitutively expressed in neurons in vivo
STAT3 colocalizes with Islet-1
Y705 phosphorylated STAT3 is present in neuronal nuclei
STAT3 can be exogenously expressed in cerebellar granule neurons
Exogenous expression of STAT3 in cerebellar granule neurons
increases primary neurite count
Exogenous expression of constitutively active STAT3 in cerebellar
granule neurons increases total neurite length
Exogenous expression of constitutively active STAT3 in cerebellar
granule neurons increases the length of the longest neurite
Overexpression of constitutively active STAT3 alters the distribution
of neurite lengths within an experiment
Expression of importins in DRG neurons and cerebellum
Previously characterized regeneration associated genes
Listing of 1,068 unique UniGene IDs identified from the subtracted
library using EST Express.
Listing of 651 Affymetrix probes found to be highly upregulated in
laser captured DRG neurons compared to cerebellum
Listing of 414 Affymetrix probes found to be highly upregulated in
cerebellar tissue compared to laser captured DRG neurons
Listing of DAVID annotations describing DRG enriched genes
Listing of transcription factor binding consensuses that are
significantly overrepresented within the promoters of DRG enriched
genes
Listing of DRG enriched transcription factor networks
Listing of real time PCR results from 84 genes in the Jak/Stat
signaling pathway
Candidate DRG enriched transcription factors
viii
84
85
86
87
88
89
90
91
176
23
92
106
122
133
141
152
160
177
ABBREVIATIONS
ATF: Activating transcription factor
BCL-2: B-cell lymphoma 2
BDNF: Brain derived neurotrophic factor
BLAST: Basic local alignment search tool
cAMP: Cyclic adenosine monophosphate
CAP23: Cytoskeletal associated protein 23
cDNA: Complementary deoxyribonucleic acid
CGN: Cerebellar granule neuron
CGRP: Calcitonin gene related peptide
CNS: Central nervous system
CNTF: Cilliary neurotrophic factor
COS7: CV-1 (simian) in origin, and carrying the SV40 genetic material (cell line)
CSMN: Cortico-spinal motor neuron
CST: Cortico-spinal tract
CTB: Cholera toxin, B subunit
DAVID: Database for annotation, visualization and integrated discovery
DREZ: Dorsal root entry zone
DRG: Doral root ganglia
E7: Monoclonal anti β-tubulin (not neuron specific)
EASE: Expression analysis systematic explorer
EDTA: Ethylenediaminetetraacetic acid
EGFR: Epidermal growth factor receptor
ix
EST: Expressed Sequence Tag
FACS: Fluorescent activated cell sorting
FBS: Fetal bovine serum
FDR: False discovery rate
FTP: File Transfer Protocol
GAP: Growth associated protein
GDNF: Glial derived neurotrophic factor
GFP: Green fluorescent protein
GP130: Glycoprotein 130
GPCR: G-protein coupled receptors
GPI: Glycophosphatidylinositol
GPL: GNU general public license
GUI: Graphical user interface
HTML: Hypertext markup language
IB4: Isolectin B4
ID: Identifying number
IL: Interleukin
IN-1: Antibody recognizing myelin associated inhibitors
IUPAC: International Union of Pure and Applied Chemistry
LIF: Leukemia inhibitory factor
MAG: Myelin associated glycoprotein
mRNA: Messenger ribonucleic acid
MSC: Marrow stromal cell
x
MySQL: My structured query language
NCBI: National center for biotechnology information
NF200: Heavy neurofilament (NF-H)
NGF: Nerve growth factor
NGR: Nogo receptor
NT-3: Neurotrophin 3
OMGP: Oligodendrocyte myelin glycoprotein
PCL: Pre-conditioning lesion
PFA: Paraformaldehyde
PHP: PHP hypertext processor
PHRED: Phil’s read editor
PKA: Protein kinase A
PNS: Peripheral nervous system
RAG: Regeneration associated gene
RGC: Retinal ganglion cell
SAGE: Serial analysis of gene expression
SNP: Single nucleotide polymorphism
STAT: Signal transducer and activation of transcription
TSS: Transcription start site
xi
PREFACE
This thesis is composed of five chapters, including a General Introduction,
Discussion and Summary. In the Introduction I have provided a basic overview of
research relating to spinal cord injury and axon regeneration, providing a context for the
data presented in Chapters 2 and 3. Chapter 2 describes the development of a software
package named EST Express, which was used to identify DNA sequences from a
peripheral neuron specific expressed sequence tag (EST) library. This work was
published in BMC Bioinformatics earlier this year and is presented here only slightly
modified from its original form. Chapter 3 describes a series of experiments conducted
to elucidate key transcriptional regulators present in PNS but not CNS neurons, as well as
specific experiments on the transcription factor STAT3. This work is currently being
prepared as a second manuscript for submission. The General Discussion in Chapter 4
addresses the main findings from the previous chapters, and potential interpretations of
the data. The final chapter outlines the impact of the experiments in Chapter 3 and
suggests several future experiments that could make use of the data presented there.
xii
PUBLICATIONS AND MANUSCRIPTS IN PREPARATION
Smith R.P., Buchser W.B., Lemmon M.B., Pardinas J.P., Bixby J.L., Lemmon V.P.
(2008). “The Transcription Factor STAT3 is PNS Specific and Causes Supernumerary
Neurite Formation When Expressed in CNS Neurons.” In preparation.
Tapanes-Castillo A., Weaver E.J., Smith R.P., Kamei Y., Caspary T., Hamilton-Nelson
K.L., Slifer S.H., Martin E.R., Bixby J.L., Lemmon V.P. (2008). “A modifier locus on
chromosome 5 contributes to L1 cell adhesion molecule X-linked hydrocephalus in
mice.” In submission.
Smith R.P., Buchser W.B., Lemmon M.B., Pardinas J.P., Bixby J.L., Lemmon V.P.
(2008). “EST Express: PHP/MySQL based automated annotation of ESTs from
expression libraries.” BMC Bioinformatics 9:186.
Newpher T.M., Smith R.P., Lemmon V., Lemmon S.K. (2005). “In vivo dynamics of
clathrin and its adaptor-dependent recruitment to the actin-based endocytic machinery in
yeast.” Developmental Cell 9(1):87-98.
xiii
CHAPTER 1: INTRODUCTION
Injuries to the nervous system
The nervous system is a highly specialized network of neurons and glia that
coordinate the body’s responses to internal and external stimuli. It consists of the brain
and spinal cord, normally referred to as the central nervous system (CNS), as well as the
nerves and ganglia innervating the periphery, termed the peripheral nervous system
(PNS). It has long been observed that unlike an injury to a peripheral nerve, which has
the ability to regenerate and restore function, an injury to the spinal cord results in
permanent damage and loss of function. This was recognized by the ancient Egyptians
who, over 3,000 years ago, wrote in a papyrus containing instructions for surgeons “One
having a dislocation in a vertebra of his neck while he is unconscious of his two legs and
his two arms, and his urine dribbles. An ailment not to be treated.” (Breasted, 1922).
Modern medicine recognizes several different types of injuries to the spinal cord
which vary in their prognosis and treatment options. All injuries fall into two major
categories: a complete injury is one in which no neural communication is possible across
the injury site, whereas in an incomplete injury some sensory or motor function is
retained, presumably as a result of unsevered axons within the cord. Patients with
incomplete spinal cord injuries have a much better chance of recovering function,
making neuroprotection and the prevention of secondary damage a major focus of
therapeutic intervention. The human spinal cord is about 18 inches long and is encased
in bony vertebrae which run from the skull to the sacrum. The level of the injury,
usually designated by the most rostral vertebra adjacent to the injury, is very helpful in
1
2
predicting what parts of the body might be paralyzed and non-functional. Injuries to the
cervical spinal cord can result in quadriplegia, with “high C” injuries, around C1-C3,
often requiring the patient to use a ventilator because of a loss of control of the
diaphragm. Patients with low cervical injuries, for example C6-C7, retain some control
over their shoulders and arms, but suffer impaired wrist and hand movement. Injuries
to the thoracic spinal cord result in normal upper limb function but varied torso function,
with only T9-T12 injured patients able to sit up straight and control their abdominal
muscles. Lumbar and sacral injuries can result in the loss of control of the hips, legs,
bladder, sexual organs and feet.
It is estimated that the incidence of spinal cord in the United States is roughly 40
per million or 12,000 cases per year (Spinal Cord Injury Facts and Figures at a Glance,
2008).
Furthermore, more than 255,000 people in the United States live with a spinal
cord injury (1 in 1200). The most common cause of injury is vehicular accidents
(40.6%), followed by falls (27.3%) and violence (15.1%). Not surprisingly, injuries
more commonly afflict younger adults, with the majority of new injuries between the
ages of 16 and 30. Gender is also linked to spinal cord injury, with 77.8% of the injured
patients since 2000 being male. Although treatment and long-term care for injured
patients have improved outcomes over the years, SCI patients still have lower life
expectancies, the extent of which is largely determined by the level of injury. SCI
patients are also faced with an enormous financial burden; care for a cervically injured
25 year old would likely exceed $3,000,000 over the course of his/her lifetime.
Despite the large burden that SCI places on individuals as well as society as a
whole, the only FDA approved treatment is the administration of the steroid
3
methylprednisolone (MP). MP evolved as a treatment for SCI during the 1990’s through
two multicenter clinical trials based on the idea that it played a neuroprotective role after
an acute injury. Recent meta-analyses of the literature have demonstrated problems with
the interpretation of the clinical trial results, and concluded that any effects of MP on
outcome are clinically insignificant (Sayer et al., 2006). A recent study by Lee and
colleagues (2008) suggested a potential mechanism for MP as an inhibitor of apoptosis
in oligodendrocytes, which could result in less myelin debris and account for the modest
effects seen in clinical trials. Regardless of the mechanism of action, there is a strong
impetus for the development of new therapies for the treatment of spinal cord injury.
Barriers to regeneration
Several factors contribute to the inability of the CNS to regenerate after injury.
In his landmark text, “Degeneration and Regeneration of the Nervous System” (1928),
Ramòn y Cajal was the first to speculate that myelin may block regeneration of axons
after injury. Many years later this idea was confirmed by several experiments showing
that neurons cultured on CNS myelin did not send out axons (Berry et al., 1982; Crutcher
et al., 1989). Immunization of mice with the 250kDa protein fraction from CNS myelin
resulted in an antibody termed IN-1, which was able to neutralize the inhibitory effects of
myelin on CNS neurite outgrowth (Caroni et al., 1988). This result provided the first
evidence of specific protein “myelin associated inhibitors” that are expressed in
oligodendrocytes and exposed after injury. Further studies went on to identify these
inhibitors as Nogo-66 (Chen et al., 2000; GrandPre et al., 2000), myelin associated
glycoprotein (MAG; McKerracher et al., 1994; Mukhopadhyay et al., 1994) and
oligodendrocyte myelin glycoprotein (OMGP; Wang et al., 2002). Recently two other
4
myelin associated inhibitors have been identified in Sema4D (Moreau-Fauvarque et al.,
2003) and ephrin B3 (Benson et al., 2005). Sema 4D and ephrin B3 are axon guidance
cues. Interestingly, while the majority of inhibitory molecules expressed during
development, such as other Sema family members and other ephrins are downregulated
in the adult CNS, Sema 4D and ephrin B3 are not (Moreau-Favarque et al., 2003; Benson
et al., 2005).
Myelin associated inhibitors bind to membrane-bound receptors on regenerating
axons, initiating signaling cascades that retard advancing growth cones. Much has been
elucidated regarding the nature of neuronal receptor complexes that bind these inhibitors,
starting with the discovery of the Nogo 66 receptor (NgR; Fournier et al., 2001), which is
expressed in many neuronal populations throughout the nervous system. Interestingly,
both MAG and OMGP also bind to NgR with high affinities (KD = 8-15nm and 5nm
respectively) despite being structurally different (Domeniconi et al., 2002; Liu et al.,
2002; Wang et al., 2002). Because NgR is a GPI-linked molecule and cannot directly
transduce signals intracellularly, several groups set about trying to discover potential coreceptors. This finding led to the discovery of the tripartite Nogo receptor complex,
which includes NgR, Lingo1(Mi et al., 2004) and p75(Wong et al., 2002) or Taj/TROY
(Park et al., 2005; Shao et al., 2005). The latter two molecules are members of the tumor
necrosis factor (TNFR) family and serve the same function with the NgR complex,
although it is believed that Taj/TROY is the more prevalent binding partner in vivo due to
its wide ranging expression in the CNS. Unlike Nogo, MAG and OMGP, the recently
discovered myelin associated inhibitors ephrin B3 and Sema4D do not bind to the Nogo
receptor complex. Each of these ligands has a cognate receptor expressed on axon
5
terminals, with the receptor for ephrin B3 being Eph A4 and the receptor for Sema 4D
being plexin B1 (Yiu et al., 2006).
After injury to the nervous system, myelin debris is immediately scattered
throughout the lesion site, inhibiting axons from reconnecting to their targets. A second
inhibitory barrier named the glial scar forms over the next few weeks, a result of the
accumulation of microglia, oligodendrocytes, meningeal cells and astrocytes at the site of
injury (Rudge et al., 1990). Although the glial scar may serve a beneficial role by sealing
off the lesion site and preventing inflammation and secondary injury (Faulkner et al.,
2004), it also hinders axon regeneration by forming a physical barrier to growth. In
addition, a portion of astrocytes at the lesion site undergo reactive gliosis (McKeon et al.,
1991), upregulating glial fibrillary acidic protein (GFAP), vimentin, as well as a class of
inhibitory molecules known as chrondroitin sulfate proteoglycans (CSPGs) (Morgenstern
et al., 2002). The CSPGs constitute a family of molecules consisting of a protein core
with multiple sulfated glycosaminoglycan (GAG) side chains which are inhibitory to
axon growth (Zuo et al., 1998).
Although the spatiotemporal pattern of CSPG expression has been well
characterized, little is known regarding how the GAG chains interact with axonal
receptors to cause inhibition. What is clear is that both myelin associated inhibitors and
CSPGs eventually converge upon the same intracellular pathways within the axoplasm,
increasing intracellular calcium and activating the small GTPase RhoA and the kinase
ROCK (Winton et al., 2002). It is unclear why the CNS has so many different classes of
inhibitory molecules that persist throughout adulthood, but the convergence upon Rho
and ROCK suggest a degree of redundancy consistent with a vital physiological function.
6
A third major barrier to regeneration is the intrinsic inability of most adult CNS
neurons to rapidly grow long axons. Early studies showed that cultured CNS neurons
lose the ability to extend axons as they age (Cohen et al., 1986; Chen et al., 1995; Dusart
et al., 1997). However, the presence of non-neuronal cells in the culture models used
made the interpretation of these experiments troublesome, as it was unclear whether the
loss in growth potential was intrinsic to neurons. Goldberg and colleagues (2002a)
addressed this issue by developing an elegant technique for culturing highly purified
retinal ganglion cells (RGCs) from different time points between E18 and P20. When
grown on a mixture of poly-D-lysine and laminin, P20 neurons displayed a reduced rate
of growth ability compared to E18 neurons. The absence of surrounding glia lends
credence to the hypothesis that over the course of development, CNS neurons undergo a
transition during which they downregulate their intrinsic growth ability.
Improving outcomes after spinal cord injury
Seminal experiments by David and Aguayo (1981), in which a piece of peripheral
nerve was grafted to the brain stem after an injury, demonstrated that some CNS neurons
retain their ability to send out long axons into adulthood if given the right environment in
which to grow. The recent explosion of information regarding the molecular mechanisms
of axon inhibition has led to numerous interventions that have improved regeneration
after injury to the CNS. In the spinal cord, where multiple white matter tracts relay
information between the brain and periphery, the result of those interventions have often
varied based on the model being used. What follows is a review of the major findings in
rodent models of spinal cord injury and their implications to the field.
7
The discovery of the neutralizing effect of IN-1 on myelin in vitro (Caroni et al.,
1988) was followed shortly after by a study in which IN-1 was delivered systemically to
rats via an ascites tumor prior to a dorsal hemisection injury (Schnell et al., 1990). The
dorsal hemisection model is commonly used because it results in a complete transection
of both the dorsal column and cortical spinal tract (CST), while sparing white and gray
matter in the ventral portion of the spinal cord. The presence of IN-1 resulted in
sprouting of CST fibers up to 7-11mm from the lesion site in IN-1 treated animals, but
not in anti-HRP treated controls. No regeneration of dorsal column fibers was observed.
Although this degree of regeneration from CST fibers may seem remarkable, the authors
only quantified the length of the longest axon. Subsequent reports have shown the
number of fibers regenerating long distances to be quite modest (Cui et al., 2004).
Identification of specific myelin associated inhibitors led to the development of a
series of knockout mice with varying regeneration phenotypes. The current consensus on
the subject appears to be that genetic deletion of MAG (Bartsch et al., 1995), Nogo
(Zheng et al., 2003), p75 (Song et al., 2004) or the Nogo Receptor (Zheng et al., 2005) do
not result in significant regeneration of the CST. Impairment of NgR, either by genetic
deletion or blocking peptides, did however result in partial regeneration of raphespinal
and rubrospinal tracts, offering some hope for future combinatorial approaches (Kim et
al., 2004; Li et al., 2005). The discordance between the IN-1 treatment and genetic
deletion experiments has led to much discussion (Zheng et al., 2005), one possible
explanation being that IN-1 binds to several epitopes, only one of which is Nogo66
(Spillman et al., 1998).
8
Treatments focused on counteracting glial scar inhibition have also enjoyed moderate
success over the past few years. Genetic deletion of both GFAP and vimentin resulted in
a significant regeneration of the CST as well as ventral serotonin-containing fiber tracts
after a thoracic hemisection injury (Menet et al., 2003). Intrathecal delivery of
Chondroitin ABC, an enzyme that cleaves GAG chains from CSPGS, results in dorsal
column and CST regeneration in addition to improved functional outcomes after a dorsal
crush injury (Bradbury et al., 2002). Knockdown of xylosyltransferase-1, an enzyme
involved in the glycosylation of proteoglycans, results in increased outgrowth of DRG
microtransplanted into the spinal cord (Grimpe et al., 2004). Mice deficient for EphA4
exhibit reduced astrogliosis and significant CST and rubrospinal tract regeneration after a
lateral hemisection of the spinal cord (Goldshmit et al., 2004). EphA4 is the putative
receptor for ephrin B3, a myelin associated inhibitor (Benson et al., 2005), suggesting the
intriguing possibility that myelin debris may induce astrogliosis.
The convergence of multiple inhibitory signals upon the RhoA pathway has made
it a favorable target for intervention. Delivery of C3 transferase, a bacterial toxin that
blocks Rho function by ADP ribosylation, improves regeneration of RGC and CST fibers
after an optic nerve crush and dorsal hemisection respectively (Lehmann et al., 1999;
Dergham et al., 2002). Moving downstream, several groups have targeted the Rho
effector ROCK using the selective inhibitor Y-27632. Administration of Y-27632 to
mice and rats after a dorsal hemisection resulted in significant regeneration of the CST
and enhanced functional recovery (Dergham et al., 2002; Fournier et al., 2003).
Although targeting of specific components mediating axon inhibition has shown
some success in vivo, several alternative approaches have also yielded interesting results.
9
Mixed results from the Nogo knockout mice led several groups to pursue high-content
screens in vitro to identify potential signaling pathways mediating axon regeneration in
vivo. One such screen identified several inhibitors of the epidermal growth factor
receptor (EGFR) as potent blockers of myelin inhibition of cerebellar granule neurons
(CGNs) growth in vitro. Gelfoam delivery of PD168393, the most potent of these
inhibitors, resulted in the regeneration of RGC axons after an optic nerve crush
(Koprivica et al., 2005). Additional experiments showed that myelin, Nogo and OMGP
induce phosphorylation of EGFR in a calcium dependent manner. The same screen also
identified Gö6976, an inhibitor of conventional protein kinase C (PKC), as a promoter of
CGN growth on myelin (Sivasankaran et al., 2004). Intrathecal infusion of Gö6976 after
a dorsal hemisection injury led to a significant increase in dorsal column but not CST
fiber regeneration (many axons > 6mm from the lesion site).
Despite the discovery of several therapeutic agents that spur regeneration of select
groups of spinal cord axons, only a handful have reported any significant regeneration of
descending motor pathways. The degree of CST growth observed with IN-1 delivery
after a dorsal hemisection (Schnell et al., 1990) suggests a very robust regeneration
phenotype, but in actual fact only the longest neurite was quantified for each treatment.
Two alternative interpretations of this result are thus apparent. First, there could be a
group of axons that respond preferentially to the treatment and do not represent the
population behavior and would thus not be an accurate indicator of functional outcomes.
The second interpretation is that the dorsal hemisection was incomplete, allowing spared
axons to take up the anterograde label. The report by David and colleagues (1981),
describing regeneration of brainstem neurons up to 30mm into a peripheral nerve graft,
10
also suggests a significant functional breakthrough. However, a follow-up paper by the
same group quantified the percentage of brainstem axons entering the graft and found
that it was less than 1% of the total population (Richardson et al., 1984). The PKC
inhibitor Gö6976 is actually only one of many interventions that lead to the regeneration
of ascending dorsal column axons, but not descending motor pathways. The next section
will discuss another intervention that sheds light on the differences between the PNS and
CNS, known as the conditioning lesion.
The Conditioning Lesion
Dorsal root ganglion (DRG) neurons offer a unique perspective on the issue of
environmental inhibition and intrinsic growth potential. Lying adjacent to the spinal
cord, DRG neurons send out two axons; one to peripheral targets as part of a sensory or
mixed nerve, the other through the dorsal root entry zone (DREZ) and the dorsal horn of
the spinal cord. A portion of these axons ascends to the cuneate and gracile nuclei of the
brain stem as part of the posterior dorsal column tract. It has long been observed that an
injury to the peripheral branch of the DRG results in full regeneration and functional
recovery (Chen et al., 2007). An injury to the central branch, however, results in limited
re-growth and no functional recovery. One possible explanation for this discrepancy is
that DRG neurons express different complements of proteins in their two axons.
Alternatively, CNS specific environmental cues might prevent the regeneration of the
central branch.
Further evidence supporting the unique role of DRG neurons in regeneration
came to light when it was discovered that dorsal column axons were 100 times more
likely to regenerate after a central injury if a simultaneous injury was also made to the
11
peripheral branch of the DRG (Richardson et al., 1984). Furthermore, if the peripheral
lesion was made 1-2 weeks prior to the central injury, robust regeneration of dorsal
column fibers was observed into and up to 3mm beyond the spinal lesion site (Neumann
et al., 1999). In the same study, DRG explants from animals that had a peripheral lesion
2 weeks prior to injury grew 4 times faster than naive controls. These results suggest that
DRG neurons are “conditioned” by a peripheral injury, upregulating growth genes in
response to an injury. Why an injury to the central branch of the DRG alone does not
turn on the same growth genes is a matter of ongoing inquiry. The most likely
explanation is that there may be some key differences in the proteomic makeup of the
two branches of the DRG.
Because there is no analogous phenomenon in the CNS, the “pre-conditioning
lesion” (PCL) has become a popular paradigm for further evaluating the intrinsic growth
potential of DRG neurons. The first major insight into the PCL mechanism was that the
levels of cyclic adenosine monophosphate (cAMP) tripled in DRG neurons 1 day after a
peripheral lesion, and that this increase corresponded with a loss of growth inhibition by
myelin in vitro (Qiu et al., 2002; Neumann et al., 2002). cAMP is a second messenger
system traditionally downstream of G-coupled protein receptors (GPCRs) and plays
important roles in a great many biological processes. Most interestingly, injection of
dibutyryl cAMP (db-cAMP) to L4 and L5 DRG 1 week prior to a dorsal column lesion
appears to mimic the conditioning lesion effect. cAMP binds to the regulatory subunit of
protein kinase A (PKA), releasing the catalytic subunit to phosphorylate specific targets
in the cytoplasm and nucleus. Pharmacological inhibition of PKA with the drug H89
blocks peripheral lesion and cAMP mediated conditioning of DRG neurons (Qiu et al.,
12
2002). One of the main targets of PKA phosphorylation is the transcription factor
CREB, which is expressed throughout the nervous system and takes part in numerous
processes. Exogenous expression of constitutively active CREB in DRG neurons is
sufficient to bring about regeneration of dorsal column axons into a dorsal column lesion,
although not to the extent seen after a PCL (Gao et al., 2004). Another downstream
target of cAMP is the enzyme Arginase I, a key regulator of polyamine synthesis.
Overexpression of Arginase I or the addition of putrescine, a polyamine, is sufficient to
allow DRGs to overcome inhibition by MAG and myelin in vitro (Cai et al., 2002),
although this effect has never been demonstrated in vivo.
What is the signal that conditions DRG neurons by elevating cAMP levels?
Possible candidates include the neurotrophins, a family of secreted proteins that are
required for the growth of embryonic neurons (Chao et al., 2003). The procedure of
cutting the central root of the DRG close to the soma, known as a dorsal rhizotomy,
normally prevents regeneration through the dorsal root entry zone, which contains strong
gradients of CSPGs. Administration of neurotrophin 3 (NT-3), NGF, GDNF but not
BDNF allows the growth of axons across the DREZ and into the spinal cord (Ramer et
al., 2000), with different neurotrophins promoting the growth of heavy neurofilament
(NF200), calcitonin-related-gene-product (CGRP) or purinergic receptor P2X3 fiber
populations. In a dorsal column transection model of spinal cord injury, delivery of NT-3
along with bone marrow stromal cells (MSCs) but not MSCs alone resulted in significant
graft penetration by dorsal column axons (Lu et al., 2004). This effect mimicked what
was seen if L4 and L5 DRGs were conditioned with cAMP injections prior to the injury.
13
A second promising candidate is the family of GP130 activating cytokines,
including interleukin 6 (IL-6) and leukemia inhibitory factor (LIF). LIF is absent from
the CNS and it upregulated by Schwann cells after a peripheral injury (Banner et al.,
1994). IL-6 is expressed in multiple cell types including both CNS and PNS neurons as
well as Schwann cells (Bolin et al., 1995; Murphy et al., 1995; Schobitz et al., 1992). IL6 knockouts exhibit delayed functional recovery after a sciatic nerve crush (Zhong et al.,
1999), whereas overexpression of IL-6 accelerates nerve regeneration after a peripheral
injury (Hirota et al., 1996). Importantly, genetic deletion of LIF and IL-6 attenuates the
conditioning lesion effect, but can be rescued through exogenous addition of either
cytokine (Cafferty et al., 2001; Cafferty et al., 2004). A further study attempting to
understand the relationship between IL-6 and cAMP showed that IL-6 was sufficient to
mimic the conditioning lesion, but not required, as the inhibition of IL-6 did not block
cAMP mediated conditioning (Cao et al., 2006).
Although researchers are rapidly identifying the mechanisms governing the
conditioning lesion effect, no reports of conditioning have been made in the CNS. In
addition, no significant regeneration of descending motor tracts has reported after
treatment with cAMP, CREB, NT-3, LIF or IL-6, suggesting that CNS neurons may lack
response mechanisms present in DRG neurons. The following section will discuss
differences between the central and peripheral nervous system and possible ways to
restore intrinsic growth states in CNS neurons.
PNS vs. CNS
The unique response of DRG neurons to a conditioning lesion or cAMP brings up
the question of whether uninjured PNS neurons constitutively express specific genes or
14
pathways that allow them to be conditioned.
If so, one would expect there to be
situations in which PNS neurons have the ability to grow in conditions in which CNS
neurons cannot. This prediction was shown to be true by a series of microtransplantation
experiments conducted in rats. The corpus callosum of the CNS contains myelinated
white matter tracts transmitting information between the two cerebral hemispheres.
When P8 and adult CGRP positive DRG neurons were transplanted into intact corpus
callosum of adult rats they were able to send out long axons, as far as 6 mm from the
transplant site (Davies et al., 1997). CNS neurons transplanted under similar conditions
led to abortive axon growth and neurites <100um in length (Tom et al., 2004).
Transplanted adult DRG neurons can also grow long distances through degenerating
white matter in the spinal cord, but stop when they come in contact with CSPGs within a
lesion site (Davies et al., 1999). Together these data support the hypothesis that DRG
neurons express specific genes that allow them to grow in situations where CNS neurons
cannot.
This hypothesis was supported by the discovery of two growth associated proteins
(Gaps) named GAP-43 and CAP-23 that are expressed specifically in regenerating
neurons (Skene et al., 1981). Nerve crush injuries were conducted on rabbit hypoglossal
nerve (PNS) and optic nerve (CNS), in addition to toad optic nerve. Unlike mammals,
amphibians such as frogs and salamanders retain the ability to regenerate particular
pathways after injury, including the optic nerve (Skene et al., 1981b). The authors found
that both rabbit hypoglossal and toad optic nerve expressed the two Gap proteins after the
crush, whereas the rabbit optic nerve did not. In addition, GAP-43 and CAP-23 were
15
expressed in all uninjured nerves from neonatal animals, supporting the idea that age is
also a variable in the loss of growth associated genes.
GAP-43 is an acidic, 43kDa protein enriched in the growth cones of growing
neurons, both during development and regeneration (Skene et al., 1989). Genetic
ablation of GAP-43 results in reduced responsiveness to cell adhesion molecules (Meiri
et al., 1998), disrupted growth cone morphology and increased susceptibility to growth
cone collapse (Aigner et al., 1995) as well as abnormal pathfinding (Strittmatter et al.,
1995). Adult CNS axons, which manage to penetrate a peripheral nerve graft, also
express GAP-43 (Campbell et al., 1991). Although GAP-43 is not expressed after an
injury to the central branch of the DRG, it is expressed in regenerating dorsal column
axons after a conditioning lesion or cAMP treatment (Qiu et al., 2002). Most
importantly, co-expression of GAP-43 and CAP-23 in DRG neurons increases the growth
of dorsal column axons by 60 fold after a dorsal hemisection injury (Bomze et al., 2001).
Such robust growth seen after the expression of only two genes suggests that axonal
growth response pathways are present and active.
GAP-43 and CAP-23 are not the only proteins expressed specifically by DRG
neurons after injury. Bonilla and colleagues (2002) identified a small proline rich protein
named SPRR1A, which is expressed after axotomy and increased neurite outgrowth in
vitro when overexpressed in embryonic and adult DRG neurons. Mice deficient for the
AP-1 transcription factor c-Jun exhibit normal hippocampal behavior and morphology,
whereas facial motor neurons fail to regenerate after a transection injury (Raivich et al.,
2004). Motor neurons deficient for c-Jun also fail to express CD44, galanin and α7β1
integrin, all of which are normally expressed in peripheral nerves after injury (Jones et
16
al., 1997; Wynick et al., 2001; Werner et al., 2000). α7β1 is emblematic of an entire
family of integrins that are upregulated after injury in sensory neurons. Unconditioned
adult sensory neurons grown in vitro on low levels of fibronectin or laminin fail to send
out long neurites but, if transfected with α1 or α5 integrin, can send out many long axons
(Condic, 2001). Together this group of genes expressed by regenerating axons is known
as “Regeneration Associated Genes”, or RAGs. A listing of well characterized RAGs is
presented in Table 1.
Despite extensive knowledge surrounding RAG expression and function in the
PNS, few reports have demonstrated RAG mediated gain-of-function in the CNS.
Expression of GAP-43 alone is not sufficient to induce sprouting of Purkinje fibers into a
peripheral nerve graft (Zhang et al., 2005). However, if co-expressed with the cell
adhesion molecule L1CAM, peripheral nerve graft penetration is significantly increased.
Although this effect was significant, the number of fibers entering the graft was
extremely modest (~10-15 fibers per 40µm section), suggesting that this phenotype might
be the exception rather than the norm. Nonetheless, along with experiments combining
GAP-43 and CAP-23 overexpression, these data have led the field to a consensus that
overexpression of multiple genes is likely to be required to induce CNS neurons to
regenerate axons.
Hunting for regeneration associated genes
The identification of specific genes products present in the axons of regenerating
neurons has lead to several systematic comparisons of gene expression using a variety of
approaches. Traditional approaches for measuring changes in gene expression, including
microarrays, subtractive hybridization and serial analysis of gene expression (SAGE),
17
require at least two samples for comparison. Reports vary greatly in the nature of the
samples, the comparison being made, as well as the reliability of the results. This section
will highlight the major findings from this literature and discuss the underlying
technologies.
By far the most common technique for detecting changes in gene expression at a
systems level is the DNA microarray. Researchers have long used a technique known as
Southern blotting to identify specific DNA samples separated by size and transferred to a
membrane using a labeled probe. This technique soon evolved into simple membranebased arrays in which colonies from two cDNA libraries were probed using radioactively
labeled oligonucleotides (Kulesh et al., 1987). The first modern microarray, reported by
Schena and colleagues (1995), used a robot to print sample cDNAs onto a glass slide, at
which point fluorescently labeled probes were used to read out the intensity of 45
Arabidopsis genes simultaneously. Current arrays typically consist of as many as 60,000
mRNA specific oligonucleotide probes covalently bound to a glass or silicon slide. The
technology has also been expanded to identify single nucleotide polymorphisms (SNPs),
regions of DNA bound to specific transcription factors identified by chromatin
immunoprecipitation (ChIP on chip), and novel transcripts by tiling an entire genome
onto a chip.
The earliest microarray analyses of spinal cord injury probed injured and naïve
sections of spinal cord for differential gene expression (Carmel et al., 2001; Fan et al.,
2001; Song et al., 2001; Tachibana et al., 2002). Because of the numerous variables
involved in animal injury models of SCI, reports varied widely both in their experimental
design and results. Such variables include the level of injury within the cord, the sex of
18
the animals used, whether only the injury site was analyzed or also distal tissue, the
number of time points after injury examined, whether the RNA was pooled, and how
many replicates were used (reviewed by Aimone et al., 2004). The end result of these
analyses was that several genes have been identified (i.e. found in multiple reports) as
being upregulated or downregulated after injury within the spinal cord. However,
because the injured spinal cord consists of multiple neuronal populations as well as
several types of glia, fibroblasts and macrophages/microglia, the results of these
experiments are difficult to interpret. In addition, because the spinal cord often does not
contain the soma of neurons that do not regenerate after injury (e.g. corticospinal motor
neurons, etc.), it is unclear that these findings would improve our understanding of
intrinsic growth mechanisms.
A second group of microarray experiments concern DRG neurons and their ability
to switch to a growth state after a conditioning lesion or cAMP administration. Bonilla
and colleagues (2002) used an 8,500 probe microarray to analyze mouse DRGs one week
after a sciatic nerve transection. Using this technique they identified the axonal growth
gene SPRR1A and confirmed the upregulation of previously known RAGs such as
neuropeptide Y (Wakisaka et al., 1991) and galanin (Villar et al., 1989). SPRR1A was
also identified the same year on an 8,799 probe rat microarray, along with several other
intriguing genes such as α2 macroglobulin, CLP36, and nerve growth factor inducible
protein VGF (Costigan et al., 2002). The discovery that cAMP could mimic the
conditioning lesion prompted Cao and colleagues (2006) to perform a custom 5,000
probe microarray using cultured DRGs treated with db-cAMP or vehicle controls. They
identified several upregulated genes, including IL-6, Arginase I and Secretory Leukocyte
19
Peptidase Inhibitor (SLPI), and confirmed the upregulation of several known RAGs
including neuropeptide Y, cAMP response element modulator (CREM) and VGF.
A third group of microarray experiments focus specifically on the developmental
loss of intrinsic growth ability by central neurons. The discovery that retinal ganglion
cells undergo a developmental loss in growth ability between E18 and P8 (Goldberg et
al., 2002a) led to a follow-up paper in which the same group used a 26,000 probe
microarray to analyze gene expression changes in cultures of purified RGCs from 17
different ages between E17 and P21. The study identified several groups of genes
regulated by RGCs during development, including a gene named CYP1B1 which plays a
role in cell survival (Wang et al., 2007). The findings from these studies have caused
many to speculate that a similar developmental transition takes place in all central
neurons. Although not specifically looking at growth ability, Arlotta and colleagues
(2005a) characterized developmental changes in E18, P3, P6 and P14 corticospinal motor
neurons using a 45,000 probe microarray. Because the cortex is a mixed population of
neurons and glia, the authors used fluorescent microspheres to retrogradely label the
corticospinal tract and corticospinal motor neurons (CSMNs), which were purified by a
fluorescently activated cell sorter (FACS) machine. The microarray analysis led to the
identification of groups of genes that play a role in CSMN identity as well as early,
intermediate and late developmental processes. Of particular interest to the authors was a
gene named CTIP2 which encodes a transcription factor vital for specifying CSMN
identity. Amazingly, genetic deletion of CTIP2 causes pathfinding errors within the
midbrain and prevents the formation of the corticospinal tract, suggesting a potential
target for efforts to enable CST regeneration (Arlotta et al., 2005b).
20
A popular alternative to DNA microarrays is a molecular biology technique
known as subtractive hybridization (Bonaldo et al., 1996; Diatchenko et al., 1996). In
this technique, a cDNA library is made from two or more populations of cells, one of
which is called the “driver” and the other the “tester”. Although many variations exist,
typically the driver is amplified by PCR and hybridized to single stranded tester to
produce a mixed population of single and double stranded DNA. Unbound, single
stranded tester can then be isolated using hydroxyapatite (HAP) chromatography,
resulting in a new cDNA library consisting of genes enriched in the tester library. There
are numerous advantages to this technique, the most obvious of which is that it produces
a cDNA library of tester-enriched genes that can be used for sequencing and expression
studies. Unlike DNA microarrays that can reveal only known genes, these genes need
not be known, and may consist of alternatively-spliced or altogether novel transcripts.
This makes subtractive hybridization of great use to researchers working with non-model
species for which a microarray has yet to be developed. Hybridization can also be
preceded by a normalization step, in which abundant mRNA populations such as β-actin
are subtracted from the tester library, allowing for rarer transcripts to have equal
representation within the library. Consequently, subtractive hybridization can reach a
level of sensitivity several times greater than microarrays (Cao et al., 2004).
Although not as common as DNA microarrays, subtractive hybridization has been
used in several reports on spinal cord injury and axon regeneration. Gris and colleagues
(2004) used spinal cord tissue from embryonic rats, as well as from uninjured and
contusion injured adults. These samples were used to perform three subtractions, the
goal being to find (1) embryonic specific genes, (2) injury specific genes and (3) genes
21
that were present in both embryonic and injured spinal cord but not uninjured. A total
of 87 genes were found to be injury specific, several of which have already been
characterized, such as c-Fos (Raivich et al., 2004), MAG, RhoA and ROCK. A similar
report used subtractive hybridization to look at genes specific to adult rat spinal cord after
a complete transection injury (Ma et al., 2006a). The authors found 73 expressed
sequence tags mapping to 40 differentially expressed genes, with little to no agreement
with the Gris et al. study.
In most birds and mammals the adult CNS has greatly reduced ability to
regenerate after an injury. This is less true for immature animals, most displaying a
developmental loss in the ability to regenerate in a species specific timeline. The
opossum serves as an interesting model because it retains the ability to regenerate until
well after birth, with injuries to the cervical cord recovering until P9 and injuries to the
thoracic spinal cord recovering until P12. Because the opossum is not a model organism
and would require a customized DNA microarray were the technology to be used,
Mladinic and colleagues (2005) used subtractive hybridization to determine genes
specific to regions of the spinal cord that support regeneration. Using this technique the
authors identified numerous ESTs that map to genes in mouse or humans known to be
involved in transcription, myelin formation, motility and cell signaling, in addition to
several novel genes.
In summary, injuries to the adult CNS are very often untreatable due to a
combination of intrinsic and extrinsic factors preventing the regeneration of severed
axons. By contrast, peripheral neurons retain their ability to regenerate after injury into
adulthood. This growth ability is at least in part due to the expression of PNS specific
22
regeneration-associated genes, which are downregulated in the CNS over the course of
development. Several models exist that have identified such RAGs, but to this date no
single gene has been found which promotes robust regeneration of the CNS. It is likely
that more RAGs exist, particularly at the level of transcription regulation where many
downstream signals originate. In this thesis I describe an effort to elucidate PNS specific
transcription factors and their role in axonal regeneration.
23
Chapter Tables
Table 1. Previously characterized regeneration associated genes (RAGs).
Listing of genes previously shown to be expressed in regenerating neurons in the CNS
and PNS by techniques measuring both mRNA and protein levels.
RAG Description Adenylate cyclase activating polypeptide 1 Arginase I Activating transcription factor 3 Brain abundant, membrane attached Basp1 (CAP‐23) signal protein 1 Bcl2 B‐cell CLL/lymphoma 2 Bcl2l1 (Bcl‐XL) BCL2‐like 1 BDNF Brain derived neurotrophic factor Calcium/calmodulin‐dependent protein Camk2a kinase II alpha cAMP responsive element binding Creb1 protein 1 Fgf2 (bFGF) Fibroblast growth factor 2 Gal Galanin Gap43 Growth associated protein 43 HSPB1/B2 (Hsp27) Heat shock protein, 27kDa IL‐6 (IL‐6) Interleukin 6 Jun (c‐Jun) Jun oncogene L1CAM (L1) L1 cell adhesion molecule Mitogen‐activated protein kinase Map2k (Mek) kinase Mapk1 (Erk) Mitogen‐activated protein kinase 1 Ncam1 (NCAM) Neural cell adhesion molecule 1 Ninj1 Ninjurin Npy (NP‐Y) Neuropeptide Y Prph Peripherin Reg2 Regenerating islet‐derived 2 Sod2 (Mn‐SOD) Superoxide dismutase 2, mitochondrial Src (c‐src/pp60src) Sarcoma viral oncogene homolog Sprr1a Small proline‐rich protein 1A Stmn2 (SCG10) Stathmin‐like 2 Vip Vasoactive intestinal polypeptide Adcyap1 (PACAP) ArgI Atf3 Reference Moller et al. (1997) Cai et al. (2002) Seijffers et al., (2007) Caroni et al. (1997) Merry et al. (1994) Gillardon et al. (1996) Tonra et al. (1998) Lund et al. (1997) Gao et al. (2004) Ji et al. (1995) Hockfelt et al. (1987) Schaden et al. (1994) Costigan et al. (1998) Murphy et al. (1995) Herdegen et al. (1991) Anderson et al. (1998) Kiryu et al. (1995) Kiryu et al. (1995) Daniloff et al. (1986) Araki et al. (1996) Wakisaka et al. (1991) Oblinger et al. (1989) Livesy et al. (1997) Rosenfeld et al. (1997) Le Beau et al. (1991) Bonilla et al., (2002) Mason et al. (2002) Zigmond et al. (1997) CHAPTER 2: AUTOMATED ANNOTATION OF PNS SPECIFIC ESTS1
Summary
Several biological techniques result in the acquisition of functional sets of cDNAs
that must be sequenced and analyzed. The emergence of redundant databases such as
UniGene and centralized annotation engines such as Entrez Gene has allowed the
development of software that can analyze a great number of sequences in a matter of
seconds. We have developed “EST Express”, a suite of analytical tools that identify and
annotate ESTs originating from specific mRNA populations. The software consists of a
user-friendly interface powered by PHP and MySQL that allows for online collaboration
between researchers and continuity with UniGene, Entrez Gene and RefSeq. Two key
features of the software include a novel, simplified Entrez Gene parser and tools to
manage cDNA library sequencing projects. We have tested the software on a large data
set (2,016 samples) produced by subtractive hybridization. EST Express is an opensource, cross-platform web server application that imports sequences from cDNA
libraries, such as those generated through subtractive hybridization or yeast two-hybrid
screens. It then provides several layers of annotation based on Entrez Gene and RefSeq
to allow the user to highlight useful genes and manage cDNA library projects.
Background
The growing trend towards high-throughput science has generated a wealth of
sequence information. In many instances specific subsets of mRNAs are isolated with
1
The contents of this chapter were published previously as: Smith R.P., Buchser W.B., Lemmon M.B.,
Pardinas J.P., Bixby J.L., Lemmon V.P. (2008) EST Express: PHP/MySQL based automated annotation of
ESTS from expression libraries. BMC Bioinformatics 9:186.
24
25
the goal of determining differences in expression between different populations of cells.
Although microarrays have been used extensively to gauge relative expression levels,
many applications such as subtractive hybridization and yeast two-hybrid libraries require
that an mRNA transcript simply be present for inferences to be made. To assist in the
analysis of expressed sequence tags (Adams et al., 1991) and data from other types of
sequencing projects, we have designed EST Express, a web-based software suite that
accepts EST sequences and gene lists and performs analyses to ascertain the identity and
function of genes expressed in a sample population.
Implementation
Software Design
EST Express uses PHP to generate dynamic HTML and Javascript. A MySQL
database records sequence and analysis information in 13 relational tables. UniGene,
Entrez Gene and RefSeq updates are downloaded from the NCBI FTP server through a
PHP script and saved in a local folder or parsed. Several dependency modules are
required for installation, including Crossmatch (Ewing et al., 1998a/b), NCBI’s BLAST
distribution (Altschul et al., 1990), and the JPGraph PHP graphics library
(http://www.aditus.nu/jpgraph). Although EST Express is designed to be run as a web
server application, it can be used in standalone mode (i.e. with no connection to the
internet) if a web server application is available. Setup requires the installation of two
modules (BLAST and Cross_match) and the configuration of a centralized PHP settings
file, but is relatively straightforward.
26
Data Analysis and Reports
EST Express accepts base calls and Phred scores in FASTA format, which it then
parses and screens for user provided contaminating vector sequence using Crossmatch
(See Figure 1). Phred scores are then used to define a window within the sequence that is
suitable for BLASTing. Sequences without high (>20) Phred scores are designated low
sequence reads, and those with predominantly vector sequence are designated vectoronly. The remaining sequences are then subjected to a similarity search against a local
copy of the UniGene database using BLASTN. The top cluster from each BLAST result
is stored and linked to the sample sequence. The “gene2unigene” conversion table
produced by NCBI (ftp://ftp.ncbi.nih.gov/gene/DATA/gene2unigene) is then used to link
UniGene clusters with the Entrez Gene database for further annotation. To simplify the
annotations of those identifiers that have many-to-many relationships, EST Express
builds a second table named “unigeneprefs” which selects the best Entrez Gene ID for
each UniGene entry based on the presence of RefSeq cross-annotations. Other analyses
listed below are then performed on the combined data and linked back to the sample.
Sequences imported into EST Express are represented as “samples” and linked
to different analyses through unique identifiers. Each sample is, in turn, part of a “plate”,
which encompasses all samples that were part of the original imported sequence file
(Figure 2). Each plate then belongs to an overall “project”, which possesses functional
characteristics that make it distinct (Figure 3). This structure was adopted because of the
nature of sequencing projects – often 96 or 384 well plates are sequenced in succession as
part of a larger project. Analyses such as batch BLAST can be performed on individual
plates or on an entire project.
27
Once samples have been loaded into a project, the underlying goal is to assign
them a UniGene cluster and a resulting Entrez Gene ID, which provides access to the vast
collection of annotations available through the Entrez Gene database. Because this
requires that a UniGene cluster database be available, the EST Express frame-work is
most relevant for projects involving model organisms (of which there were 74 at the time
of writing). Sequences from non model organisms can also be identified provided they
have sufficient sequence similarity with those of a model organism.
The Entrez Gene database (Maglot et al., 2005) is a central depot for genespecific information. EST Express makes full use of the annotations contained within,
linking UniGene cluster IDs to Entrez Gene IDs. Because of the large size of the Entrez
Gene database (>600MB for the Mus_musculus version alone) there is considerable
interest in developing utilities that can parse the provided ASN.1 files into a useable
format (Liu et al., 2005a). Many of the Entrez Gene annotations, however, can also be
found in flat text files, which are much easier to parse. Four of these files (gene_info,
gene2unigene, gene2go and gene2refseq) are downloaded by EST Express and combined
into a single MySQL table within minutes. Users can then search annotations that match
to samples using the search tool.
In many cases it is desirable to know whether a library clone contains the full
open reading frame for the gene in question. This allows for selected full-length clones
to be re-arrayed and used in a variety of expression studies. EST Express can carry out
such an analysis for Oligo(dT)-primed cDNAs that have sequence reads from the 5’ end.
Once a sample sequence has been identified, the corresponding RefSeq protein ID is
extracted from the Entrez Gene table and matched against a locally downloaded copy of
28
the RefSeq protein database. The EST is then translated into three different frames and
matched against the first 10 amino acids of the protein sequence. Using this comparison,
each annotated sequence is assigned “full-length” or “not full-length” status. Samples
that are not annotated with a RefSeq protein identifier are examined for long open
reading frames, the results of which are stored and can be queried for further analysis.
EST Express offers two tools that enable the user to assess the content of the
source library being sequenced. The first tool generates a graph of the number of novel
UniGene clusters found in each successive sequenced plate added to a project (Figure 4).
This feature is a useful indicator of library complexity as well as of how many sequences
the user can expect to obtain. The second tool reports the number of times each UniGene
cluster has been found within a project. This is a useful measure for subtracted libraries
because cDNAs sampled more frequently correspond to transcripts that are enriched in
the tester mRNA pool.
Thus far, no individual technique provides complete information about the genes
that are at work in a system. It is therefore often useful to compare lists of genes for
commonalities or differences. EST Express allows the user to generate a list of sample
IDs, UniGene clusters or Entrez Gene IDs from a project or plate based on specific
criteria. Lists of identifiers may also be uploaded as a text file originating from another
experiment (e.g. microarray, mass spectrometry). Once a list is created it can be
compared against one or more lists, the results of which can be saved as a new list. Each
list can then be exported with full Entrez Gene annotations to an Excel spreadsheet for
further analysis.
29
Results and discussion
Evaluation with subtracted library sequences
EST Express has been successfully implemented and used to identify and
annotate 4 separate libraries containing over 2,500 samples. Of these four libraries, the
largest is the “subtracted” library generated through subtractive hybridization of tissue
specific genes. For this library, 21 plates containing 2,016 samples were analyzed,
resulting in 1,068 unique cDNAs (See Figure 5a, Table 2). Of the 2,016 samples, 192
were vector-only sequences and 107 were low quality sequence reads. Of the 1,068
unique cDNAs, 914 matched Entrez Gene entries. Selection of appropriate Entrez Gene
identifiers based on RefSeq links proved efficacious: only 23 sequences match Entrez
Gene identifiers without a RefSeq link, allowing full-length analysis of 83% of samples
returning a BLAST hit (Figure 5b). Of those samples that were analyzed, 227 were
found to be full-length.
Comparison to related software packages
EST Express is similar in broad terms to other sequence pipeline software
packages, including PipeOnline 2.0 (Ayoubi et al., 2002), ESTAP (Mao et al., 2003),
EST-PAGE (Matukumalli et al., 2004) and ESTIMA (Kumar et al., 2004). However,
there are several key differences that make EST Express an attractive alternative to the
bioinformatics community. EST Express is written entirely in PHP, an open source
scripting language that is platform independent and extremely popular amongst web
developers. All four of the packages listed above are Perl based and could not be
installed on Windows based server without modifications. EST Express uses the MySQL
30
database platform for storage of sequence data and analyses. MySQL is also open source
and freely available under the GPL, contrasting with the commercial package Oracle,
which is employed by ESTAP and ESTIMA. Unlike PipeOnline 2.0, EST Express is also
freely available for download and installation, and is distributed with explicit instructions
for both Linux and Windows based machines.
The central difference between EST Express and these other packages is that it
was designed for a post genome world in which researchers have the ability to generate
specialized expression libraries and require a pipeline for identifying the mRNAs within.
EST Express is unique in that it has a built-in support for identifying full-length cDNAs,
diagnostic tools for gauging the complexity of the cDNA library, gene list tools for
comparisons with microarray data and convergence of annotations through the use of the
relatively recent Entrez Gene database.
Potential applications
Although EST Express was primarily developed to analyze libraries generated by
subtractive hybridization, it could be employed in any number of applications, some of
which are outlined below:
a)
Generic libraries in which the host organism has an established UniGene cluster
database.
b)
Libraries generated through subtractive hybridization of two or more mRNA
populations
c)
Screened yeast two-hybrid prey libraries
d)
Comparison of gene lists generated on different platforms
e)
Annotation of custom gene lists with terms from the Entrez Gene database
31
Conclusions
We have developed a valuable new tool named EST Express for the
identification, annotation and analysis of cDNA library sequences. EST Express is
unique in that it is cross-platform, is freely available, makes full use of annotations from
the Entrez Gene database and allows the user to assess the state of the cDNA library
using diagnostic tools.
Availability
EST Express has been tested on Windows NT/2000/XP as well as Redhat Linux,
although it could conceivably be used on any system with access to web server software.
EST Express can be downloaded from http://www.sourceforge.net/projects/estexpress,
and is available under the GNU General Public License
(http://www.gnu.org/copyleft/gpl.html).
32
Chapter Figures
Figure 1. EST Express Data Pipeline. Raw sequence data is imported into EST Express
along with phred scores, where it is then screened for contaminating vector sequences
and masked for quality. Good quality sequences are then batch BLASTed against a local
UniGene database and the top hit is assigned to each sample. A local copy of the Entrez
Gene database is then linked to the UniGene hits and used to annotate each sequence with
a description, Gene Ontology identifiers, RefSeq mRNA and protein links, and genomic
context. Oligo(dT)-primed sequences can then be analyzed for full-length status using a
local copy of the RefSeq protein database and the Entrez Gene cross references. The user
interface then provides several ways to browse and visualize the results from the pipeline.
33
Figure 2. EST Express Plate Viewer. Screen capture of the “Plate Viewer” page
showing details for plate JN02001X1 owned by user “rsmith” in project “robinDNQ”.
For each matched sample in the plate a UniGene identifier is listed, along with the
BLAST score and Entrez Gene and full-length annotations.
34
Figure 3. EST Express Project Viewer. Screen capture from the “Project Viewer” page
showing a graphical breakdown of ESTs within a project. “Vector” refers to sequences
designated vector-only. “Bad_sequence” refers to sequences with low quality sequence
reads. “Unknown” refers to samples that are neither vector-only nor low quality, but do
not match against the UniGene database. “Uniques” refers to the number of unique
UniGene clusters in the project and “Repeats” refers to additional instances of those
unique clusters.
35
Figure 4. EST Express “New Genes” Tool. Screen capture from the “New Gene”
library tool, showing the number of new unique UniGene clusters found with each
successive round of sequencing. Further rounds of sequencing produce progressively
fewer unique clusters. After several rounds of sequencing, previously identified cDNAs
were re-subtracted from the original library, leading to an increase in new gene discovery
(black arrow). This figure was produced dynamically within EST Express using the
JPGraph PHP graphics library.
36
Figure 5. Results of analyses of the subtracted data set. A. Distribution of BLAST
identifications made by EST Express for all 2,016 sample sequences. B. Distribution of
associations made for 1,068 distinct UniGene entries.
CHAPTER 3: PNS SPECIFIC TRANSCRIPTION FACTORS AND STAT3
Summary
Injuries to the CNS due to trauma or stroke result in the disruption of major axon
pathways, often resulting in permanently impaired function (Cajal, 1928). Several factors
contribute to the failure of the CNS to regenerate after injury, including inhibition of
axonal growth by myelin and glial scar associated molecules (Filbin, 2003; Silver et al.,
2004), as well as the intrinsic inability of adult CNS neurons to grow long axons in
environments that are permissive for younger neurons (Goldberg et al., 2002a).
Neurons in the peripheral nervous system (PNS) display a higher capacity to
regenerate after injury. A “conditioning lesion” of a peripheral nerve 1-2 weeks prior to a
spinal cord injury results in significant regeneration of dorsal column fibers (Richardson
et al., 1984; Woolf et al., 1999). Treatment with cAMP (Qiu et al., 2002) or a small
molecule inhibitor of protein kinase C (Sivasankaran et al., 2004) results in substantial
regeneration of ascending dorsal column fibers but not descending cortical spinal tract
axons. Furthermore, appropriately microtransplanted DRG neurons have the ability to
regenerate long distances through intact and degenerating CNS white matter (Davies et
al., 1997; Davies et al., 1999). CGNs transplanted under similar conditions displayed
relatively modest outgrowth (Li and Lemmon, unpublished data).
Intrinsic differences in regenerative potential between CNS and PNS neurons
imply distinct molecular mechanisms mediating axon growth and inhibition, however to
this date very few reports have demonstrated PNS specific regeneration pathways.
Molecules such as GAP-43, CAP-23 and SPRR1A are highly expressed by DRGs after
37
38
injury and the degree of their expression correlates with regeneration of dorsal column
fibers (Skene et al., 1981; Bonilla et al., 2002). All three of these genes are also
expressed after injury in the CNS but have limited effects on the re-growth of axons
(Schmitt et al., 1999; Marklund et al., 2006), even when overexpressed (Zhang et al.,
2005). This result suggests that CNS neurons may lack key downstream effectors
required for regeneration.
Signal transducer and activator of transcription 3 (STAT3) is a well characterized
transcription factor governing a wide variety of cellular functions including migration,
proliferation, apoptosis and immune responses (Levy et al., 2002). Previous reports have
shown it to be phosphorylated in the axons and somas of DRG neurons after a peripheral
injury (Eckenstein et al., 2000), leading some to speculate that it may serve as a
retrograde messenger (MacLennan et al., 2004). Pharmacological blockade of Janus
kinase 2 (JAK2), an upstream activator of STAT3 (Figure 6), results in an abolishment of
the conditioning lesion effect (Qiu et al., 2005). Furthermore, conditional ablation of
STAT3 in motor neurons resulted in increased cell death after a peripheral lesion,
suggesting that it may play an important role in neuroprotection (Schweizer et al., 2002).
In this chapter I detail a systems biology based search for PNS specific
transcription factors and provide evidence that the JAK/STAT pathway and STAT3 in
particular are intrinsically present and active at higher levels in the PNS than CNS. In
addition, we show that overexpression of constitutively active form of STAT3 in CNS
neurons leads to significantly improved outgrowth.
39
Microarray analysis of laser capture microdissected DRG neurons
Using EST Express, we obtained a list of 1,068 genes that were present in a DRG
cDNA library after subtractive hybridization with a library of age-matched cerebellar
cDNAs. One rationale for this comparison was that the dissociation of DRG neurons
severs their axons, mimicking an injury in vivo. As such the subtracted library contains
genes that have been upregulated as a result of injury, in addition to those which are
constitutively different between DRGs and cerebellum. A potential criticism of this
approach is that the transcriptome present in culture is not identical to that in vivo. This
is even more likely considering that P7-10 mouse DRG cultures, even those that have
been enriched using density gradients and drug selection, contain some non-neuronal
cells types including Schwann cells.
To address these concerns, we decided to make use of publicly available
microarray data available at NCBI’s “Gene Expression Omnibus”, or GEO
(http://www.ncbi.nlm.nih.gov/geo). The popularity of DNA microarrays has led to an
explosion in gene expression data that quickly became a problem for journals to publish
in a standardized format. Additionally, scaling of the underlying DNA chip technology
has resulted in several platforms that have probes representing the entire genome,
allowing for multiple hypotheses to be tested with the same data set. GEO addresses
these issues by representing gene expression data in a standardized manner, allowing for
post hoc analysis through a web interface. Many journals now require that microarray
data be uploaded to GEO before publication, allowing the database to grow to over
230,000 samples as of July 2008.
40
GEO’s repository contains 23 data sets containing DRG samples, almost all of
which originate from whole ganglia or mixed cells in culture. A series of datasets
(GSE2917), originating from a gastroenterology lab at Johnson & Johnson, however,
contains gene expression data from a population of laser capture microdissected DRG
neurons (Peeters et al., 2006). The authors were interested in genetic differences between
whole ganglia and laser capture microdissected neurons, obtained by injecting the
retrograde label cholera toxin B (CTB) into the peritoneum. Although understanding
spinal cord injury was completely outside of the scope of the project, this data series
yielded a treasure trove of 5 replicate samples of laser captured neurons analyzed using
the Affymetrix MG-U74Av2 platform. This is the same chip that was used for a series of
microarray experiments within our lab several years ago to study gene expression
changes in the cerebellum of wildtype and L1 knockout mice, allowing for crossexperiment comparisons to be made.
Variations in probe synthesis and deposition, sample collection and processing, as
well as fluorescence detection results in noise that is superimposed upon all microarray
gene expression signals. For that reason, samples are usually processed in biological
replicates that are then normalized and averaged before any further analysis is performed.
Normalization is the process of removing any inherent bias in a population of values (in
this case, gene expression signals) so that comparisons can be made between samples.
There are many ways of normalizing a group of values; one classic method used by
Affymetrix is to make the mean value for each sample the same by multiplying all of its
values by a correction factor. Because this technique does not correct for nonlinear
differences between chips, a preferred method for normalizing microarray values is
41
quantile normalization (Bolstad et al., 2003). The aim of quantile normalization is to
make the distribution of values within a group of microarray samples the same. This is
achieved by ranking each of the values within a sample from the lowest quantile to the
highest, and then taking the average value for each quantile as the new normalized data
set. As a result, each normalized sample has the same values, with only the mapping of
probe to signal varying between signals. Although in theory this method could eliminate
signals at the high and low end of the sample, in practice this does not prove to be a large
problem (Bolstad et al., 2003). Quantile normalization is included as part of a larger
algorithm commonly used for microarray analysis named Robust Multi-array Average
(RMA), which also corrects for background and log-transforms microarray probe
intensities (Irizarry et al., 2003).
Using the publicly available microarray software “RMA Express”
(http://rmaexpress.bmbolstad.com), we processed 8 Affymetrix CEL files (5 DRG, 3
Cerebellum) to produce 8 sets of 12,488 normalized signal intensities. These values
were then imported into the commercial software package Spotfire DecisionSite
whereupon a mean was calculated for both the DRG and Cerebellum groups (see Figure
7). A two-tailed unpaired Student’s t-test was then used to determine the significance of
a fold difference between the signals. P-values were then corrected for multiple
comparisons using a false discovery rate technique (FDR) (Hochberg et al., 1990).
Looking for genes that were 2-fold upregulated with a p value less than 0.01, we found
1,252 and 892 genes to be significantly upregulated in DRG and cerebellum respectively.
A listing of the most significant of those genes is available in Tables 3 and 4.
42
Ontological clustering of DRG enriched genes
Large lists of differentially regulated genes present a problem for further analysis
because it is virtually impossible to follow up on each gene in the list experimentally in a
systematic fashion. As such, several techniques have been developed to reduce the
complexity of a gene list and identify underlying trends associated with a particular
pathway, receptor or transcription factor. One such technique is based on a class of
annotations known as gene ontology (GO) terms. GO annotations consist of a
hierarchical, controlled vocabulary that describes gene products in 3 distinct ways. The
“Molecular Function” describes the function of the gene product according to published
reports, so for example a transcription factor might have the GO term “transcription
activator activity” within this category. “Biological process” describes multi-step
pathways and cellular functions that the gene product plays a role in, so our hypothetical
transcription factor might have a term like “temperature homeostasis” within this
category. Finally, “Cellular component” describes the localization of the gene product
within the cell, which would most likely be “Nucleus” for our transcription factor.
GO annotations by themselves do not help reduce the complexity of a large gene
list. However, like any annotation they can be analyzed en masse for a given group of
genes to see if any particular terms predominate. By comparing the results of this
analysis with a reference gene list it is possible to determine the significance of the
overrepresentation of a particular GO term within the original list. This process, known
as “Ontological clustering”, has been implemented by several groups (Castillo-Davis et
al., 2003), but most elegantly as a web-based software tool named DAVID, or “Database
for Annotation, Visualization and Integrated Discovery” (Huang da et al., 2008). DAVID
43
allows the user to enter a list of genes of interest as well as a reference gene list,
whereupon it can search for overrepresentation of more than 24,000 GO terms. DAVID
is also not limited to GO terms, and can detect perform a similar analysis for 30,000
additional annotations in 40 different categories including protein domains, chromosomal
location and tissue expression.
DAVID was used to analyze three gene lists developed from the subtraction and
microarray experiments. The first list consisted of 1,068 subtracted library genes that
were found to be enriched in cultured DRG neurons. The second and third genes lists
consisted of the 1,252 and 892 genes found by DNA microarray to be significantly
upregulated in laser capture microdissected DRG neurons and cerebellum respectively.
Of great interest to us were GO and other annotations that were overrepresented in DRG
neurons (i.e. the first two lists) but not cerebellum enriched genes (i.e. the third list). To
determine the significance of an overrepresented annotation, DAVID uses an algorithm
named EASE (Expression Analysis Systematic Explorer) which uses a modified Fisher’s
Exact test to assign a p-value. We subsequently performed a post-hoc correction on these
p-values using the Benjamini-Hochberg FDR method.
Using DAVID we assembled a list of 122 annotations (see Table 5) that were
significantly overrepresented (p<0.05) in the DRG enriched gene lists but not the
cerebellum list (p>0.05). This list includes 51 GO annotations in addition to 21
annotations from Panther (Thomas et al., 2003), a GO-like database. Although GO uses a
controlled vocabulary, many terms describe similar biological processes or molecular
functions. As such it is helpful to further reduce the complexity of this sort of analysis by
grouping related ontological terms into clusters. This was done manually by scanning the
44
list of GO and Panther terms and composing a list of 5 ontological clusters that were
overrepresented in the DRG enriched genes. Figure 8 lists these clusters along with the
number of terms comprising the cluster and the geometric mean of the p-values for the
terms in each cluster. Interestingly, there are several DRG specific terms related to
cytoskeletal organization and cellular morphogenesis, which supports the hypothesis that
DRG neurons are enriched with “growth genes”.
Promoter analysis of DRG enriched genes
Whereas gene ontology clustering can offer information regarding the general
functional trends in a gene list, it is often more helpful to obtain specific information
about key regulators that impact expression of multiple genes within the group. Even
though DRG and cerebellar granule neurons differentially regulate thousands of genes, it
is likely that the majority of those differences are due to a much smaller set of
transcription factors. This was shown to be true for mouse embryonic fibroblasts, which,
when infected with the transcription factors OCT4, SOX2, Myc and KLF4, take on an
embryonic stem cell-like phenotype and can be re-specified into different lineages
(Takahashi et al., 2006). It is therefore of great interest to determine which transcription
factors may oversee the inherent ability of DRG neurons to switch to a growth state and
regenerate after injury. Both this section and the following one will discuss efforts aimed
at identifying such factors.
Transcriptional activators and repressors bind to specific DNA regulatory
elements located in the promoter regions of responsive genes. Although regulatory
elements for a given transcription factor vary between promoters, they invariably include
particular stretches of DNA that have a high affinity for that factor. Empirical evidence
45
of transcription factor binding sites found in vivo has lead to the establishment of
“consensus” DNA binding sites, which specify what particular sequence of DNA is the
optimal binding site for a transcription factor. Promiscuous factors such as HNF4-alpha
have very loose consensuses (“AGKYCA”), whereas specialized factors such as HOX1.3
have more elaborate consensuses (“TGCNHNCWYCCYCATTAktNNDCNMNHYCN”).
As can be seen from the two consensuses listed, there is often not a consensus for a given
position, so IUPAC nomenclature is used (e.g. “N”=any base, “W” = adenosine or a
thymine, “K” = guanine or cytosine, and so on).
The largest collection of transcription factor binding consensuses currently
available is the commercially available package TRANSFAC (Biobase GmbH,
Germany). TRANSFAC 10.2 contains a list of 810 consensuses for 55 species, 504 of
which were discovered in mammals. Consensuses from TRANSFAC are traditionally
determined by in vitro selection studies, but in some cases more biologically relevant
techniques such as chromatin immunoprecipitation (ChIP) have been used. A second
source of transcription factor binding sites is a seminal paper titled “Genome-wide
analysis of mammalian promoter architecture and evolution” (Carninci et al., 2005), in
which the authors used evolutionary conservation of promoters to identify potential
binding sites. Using this approach, the authors identified 234 conserved mammalian
promoter elements, only 129 of which map to known factors. As a result, this list
contains 105 “Discovered motifs” that may bind novel transcription factors.
Using both the TRANSFAC and Carninci consensus lists, we wanted to determine
whether any particular factor was likely to bind to the promoters of genes enriched in
DRGs. Because it is impractical to manually identify and scan promoter regions for
46
thousands of genes, I developed a software package named “Promolyzer” to automate
this process. Promolyzer was developed using EST Express as a backbone and the two
programs share a web interface and many of the same functions. Instead of entering raw
sequences, however, users enter a list of Entrez Gene identifiers for the genes they would
like to analyze. Because there is no single set of criteria that defines the size and extent
of all mammalian promoters, Promolyzer defines the “promoter region” to be the 1000bp
region immediately upstream and 200bp downstream of the transcriptional start site
(TSS). The program then allows the user to create a promoter “build”, essentially a
database table filled with the promoter regions of all of the genes of interest. Multiple
builds can be created for each gene list, the main variable being the size and offset of the
promoter region that is extracted. Once a build is created the user can search for the
prevalence of an individual motif or scan a “Consensus list” that can be uploaded to the
software. The software has two methods for scanning a consensus list. In the first, the
program lists each consensus in the list and reports of matching genes. The second
method reports each gene in the list and reports which binding site consensuses match it.
Depending on the question being asked either of these options may have relevance.
Entering our three gene lists into Promolyzer, we scanned both the TRANSFAC
and Carninci consensus lists against 1000bp promoter regions for each gene. For each
list we recorded the number of genes that contained a given consensus. To determine the
significance of any particular consensus, a reference list containing all Entrez Gene
entries was used. A p-value was then determined using Fisher’s Exact Test, and
corrected using the Benjamini-Hochberg technique. Using this approach we identified
113 consensuses that were significantly enriched in the promoters of the DRG enriched
47
gene lists but not the cerebellum gene list (Table 6). 40 of these consensuses matched
both the subtraction and microarray lists, as shown in Figure 9a. To confirm that these
sites were functional, we examined the location of STAT consensuses (TCCCRGAAR)
within the promoters of subtraction and microarray genes relative to the TSS (Figure 9b).
The distribution of those sites shows a strong preference for the 100-200bp region
upstream of the TSS, which is consistent with the findings of whole genome studies
(Tharakaraman et al., 2008).
Identification of DRG enriched transcription factor interaction networks
The second method we employed to determine whether particular transcription
factors were preferentially active within DRG neurons was to make use of existing
protein interaction data to “connect” subtraction and microarray genes. The rationale for
this approach is that a transcription factor playing an active role in DRG neurons would
be expected to have a high degree of interaction with DRG enriched genes and gene
products. The transcription factor need not be present in the gene list; its activity can be
deduced merely by comparing the degree of its connectivity in the gene list of interest
with a reference list.
Over the past few decades there have been thousands of reports of protein-protein and
protein-DNA interactions which have lead to the creation of several databases seeking to
catalog them. The three most used public collections of interaction data are BIND
(Biomolecular Interaction Network Database), BioGrid (Biological General Repository
for Interaction Datasets) and HPRD (Human Protein Reference Database), all of which
have been integrated into Entrez Gene. Interaction data within these databases is
contributed by the scientific community through the submission of manuscripts as well as
48
pre-publication data sets. However, the data is curated, meaning that there are “master”
users that verify the authenticity of the data and the methods used to procure it. Although
publicly available interaction databases have greatly improved over the past few years,
there are several commercial alternatives that provide more complete datasets as well as
better tools for visualization and analysis. One such package is named Metacore, which
we used for our analysis.
Metacore consists of a unique web interface that allows the user to visualize one of
the largest curated collections of protein interaction data available. GeneGo obtains
interaction data through professional curators that comb dozens of journals looking for
novel findings. Interactions are categorized by significance (i.e. high or low trust) and
the nature of the interaction (e.g. phosphorylation, transcriptional regulation, etc.).
Although one could simply use the database to obtain information about a particular
interaction, the real strength of Metacore is its ability to construct interaction networks
based on a list of user-supplied genes. This can be done in a direct fashion, that is to
construct a network of all of the interacting genes within a data set, or indirectly, where
Metacore will attempt to fill in the gaps within a small gene list and provide potentially
novel targets. Most importantly, Metacore is able to analyze transcriptional regulation
within a user supplied gene list, attempting to connect the genes to every transcription
factor in its database to determine if there is an unusually high degree of connectivity.
We first used Metacore to visualize direct connectivity within the subtracted gene set
(Figure 10). Of the 1,068 distinct UniGene IDs we identified, Metacore was able to
internally represent 792 of them. From this number, the largest cluster of interacting
proteins contains 191 objects (Figure 10B), with the next largest having only 2. The high
49
degree of interactivity within the 191 object cluster suggests the presence of multiple
DRG specific signaling pathways that may underpin their ability to regenerate after an
injury. Using the transcriptional regulation tool we recorded transcription factors that
were significantly connected to either list of DRG enriched genes. Significance was
determined using an exact test developed by Metacore (described in the Materials and
Methods section) that calculates the probability of a particular network forming by
chance. P-values were then corrected post-hoc for multiple comparisons using the
Benjamini-Hochberg method to eliminate false positives. Using this approach we
identified 104 transcription factors that may selectively act in DRG neurons (Figure 11a,
Table 7). 30 of those factors (>50% of factors for each list) were found to interact
significantly with both the subtraction and microarray gene lists (Figure 11b),
demonstrating the usefulness of two independent data sets. In addition, several factors on
this list were also identified by the promoter analysis, including Gata3, Pax4, Hif-1A,
Irf1, Nf-kB and STAT3.
JAK/STAT pathway profiling by real time PCR
The presence of STAT3 and Irf1 (a Stat responsive gene) in the network and
promoter analyses led us to suspect that JAK/STAT pathway may be differentially
regulated between the PNS and CNS. This is consistent with reports of high levels IL-6
(Murphy et al., 1995) and GP130 (Gardiner et al., 2002) expression in DRG neurons,
although no comparisons have been made with CNS tissue. Several reports in the
literature have characterized STAT3 expression and phosphorylation in peripheral
neurons after a conditioning lesion (Stromberg et al., 2000; Lee et al., 2004). Reports of
JAK/STAT expression in the CNS have been rarer, with one report showing that mRNA
50
for JAK2, Jak3, Stat1, STAT3 and Stat5A were upregulated after injury in peripheral
hypoglossal neurons but not in non-regenerating Clarke’s nucleus neurons.
We sought to systematically characterize the differences in expression of the
entire JAK/STAT pathway between PNS and CNS neurons. Furthermore, because the
mRNA for many of these molecules increases after injury, we decided to look at
uninjured animals to gauge what intrinsic differences there might be that would allow
PNS neurons to switch to a regenerative state. However, as we discovered during the
subtraction and microarray experiments, it proved technically challenging to obtain a
purified population of DRG neurons. Several techniques were employed to purify
neurons, most relying on using the “c” fragment of tetanus toxin to selectively label
neuronal membranes. We then attempted to purify neurons using both magnetic beads
and FACS but, due to non-specific binding to flat cells and poor yields, our efforts
proved unsuccessful. Ultimately we decided to use a simpler technique, centrifuging
DRG cell suspensions through a cushion of 28% Percoll as described by Tucker and
colleagues (2006). Because DRG neurons are larger than non-neuronal cells present in a
suspension of DRG and nerve cells, they form a pellet at the bottom of the cushion and
leave everything else at the surface. Using this technique we were able to obtain a
neuronal purity of at least 80% and a sufficient yield for the analysis of different mRNA
populations (Figure 12).
By far the most common method for validating microarray results is real-time or
quantitative PCR (qPCR; Higuchi et al., 1992). In this technique, first strand cDNA
generated from total or oligo-dT selected RNA is used as a template for PCR in a mix
that also contains a fluorescent dye that intercalates into double stranded DNA. Using a
51
thermocycler equipped with a fluorescence sensor, the polymerization of a double
stranded DNA product can be visualized during each cycle in real time. Depending on
the abundance of the sample and the affinity of the primer, the fluorescent signal will
reach the “threshold” of detection somewhere within the 15-35 cycle range. This cycle
value, named the “threshold cycle” is used as an indicator of the relative abundance of the
gene of interest within the sample. Comparing the threshold cycle for a gene of interest
(GOI) in a control sample to that in a test sample allows for the determination of the fold
difference between the two samples, although generally both values first have to be
normalized to an internal housekeeping (HKG) control. Because intercalating dyes such
as SYBR-green can bind to any double stranded product including primer-dimers, primer
design is one of the most challenging aspects of performing a qPCR experiment.
Genomic DNA can also be amplified, which is why the RNA extraction process usually
involves a DNase digestion step and primers are best designed to span intronic regions.
The JAK/STAT pathway has been well characterized in other fields and validated
qPCR probes exist for most pathway genes. Furthermore, several companies sell 96 well
qPCR plates that contain pre-validated primers in each well. We decided to use the
JAK/STAT RT2 Profiler Array from Superrarray, which contains 84 pathway genes of
interest in addition to 5 housekeeping genes and genomic contamination controls. Using
purified P7-10 DRG neurons and cerebellum from the same animals, we performed
qPCR for 3 biological replicates (6 plates in total). Normalized threshold cycle values
were obtained for each GOI using the average of four housekeeping controls (Betaglucuronidase, Hsp90, Gapdh and Beta-actin). The fifth housekeeping gene,
Hypoxanthine guanine phosphoribosyl transferase 1 (Hprt1), showed preferential
52
expression in DRG neurons and was not used. The results from the qPCR arrays (Table
8) support the hypothesis that several major components of the JAK/STAT pathway are
differentially expressed between DRGs and cerebellum. In total, 16 genes were found to
be significantly enriched (fold change > 2, Benjamini-Hochberg corrected p-value<0.05)
in DRG neurons, including the core JAK/STAT proteins JAK1, STAT3 and STAT6.
Confirming predictions made by both the promoter and interaction network analyses,
both STAT3 and Irf1 were found to be highly upregulated in DRG neurons (STAT3: 6.79
fold, p=0.007; Irf1: 8.77 fold, p=0.015). Although we had suspected that JAK1 and
GP130 (IL-6st) might be upregulated in DRG neurons, several other genes were found to
be highly upregulated in DRG neurons that were not predicted, including STAT6, the
transcription factor CEBP/β and several membrane receptors. If STAT3 were
preferentially active in DRG neurons, one would expect that (i) factors regulating STAT3
and (ii) targets regulated by STAT3 would be enriched in DRG neurons. This was found
to be the case for several factors both upstream and downstream of STAT3, as depicted
by Figure 13.
Although one might expect for there to have been no preferential expression of
JAK/STAT pathway members in CGNs, we identified 15 significantly enriched genes in
this population. However, this set of genes did not contain any core JAK/STAT pathway
genes, and many of these CGN enriched genes play significant roles in other pathways.
These include the epidermal growth factor receptor (EGFR), which has been reported to
mediate the inhibitory effects of myelin in neuronal cultures (Koprivica et al, 2005), as
well as three members of the Smad family.
53
STAT3 expression and activity in vitro
Although several JAK/STAT pathway genes were identified using the qPCR
screen, we were particularly interested in further characterizing the expression of STAT3
for several reasons. STAT3 was the only JAK/STAT pathway gene besides Irf1 that was
identified in the Metacore analysis, which supports the hypothesis that STAT3 plays a
biologically relevant role. In addition, perineural administration of the JAK2 inhibitor
AG490 blocks the phosphorylation of STAT3 and abolishes the conditioning lesion effect
(Qiu et al., 2005), supporting the hypothesis that STAT3 plays an important role in
axonal regeneration.
We first sought to characterize STAT3 expression in vitro, using cultures of DRG
and CGNs. Using a pan-STAT3 antibody we observed high levels of immunostaining in
DRG neurons using confocal microscopy (Figure 14). Additionally, STAT3 is
selectively expressed in neurons but not surrounding glia, as shown by the profile trace.
In order to compare STAT3 immunostaining between different cell types, DRGs and
CGNs were harvested from the same mouse and cultured in identical culture media for 4
days on cover glasses coated with a mixture of poly-D-Lysine and laminin. After
fixation, immunochemical staining was carried out identically between the two samples
and confocal stacks were acquired using the same settings on the same day. Although
DRG neurons are easily identifiable within a mixed culture based on their size, shape and
neurites, we used the neuronal marker HuD to confirm neuronal identity. The panels in
Figure 15 show a high magnification micrograph of a single DRG neuron and 3 CGNs
stained for tubulin, STAT3 and the nuclear dye Hoechst. As can be seen from the figure,
DRG neurons exhibit much higher levels of STAT3 immunoreactivity than CGNs.
54
Within the DRG neuron, STAT3 localizes to both the cytoplasm and nucleus, as
evidenced by colocalization with Hoechst.
Although the difference in expression between DRGs and CGNs is striking, we
were also interested to see if levels of STAT3 activity varied between the two cell types.
Activation of Janus kinases causes STAT3 to be phosphorylated on tyrosine 705, which
causes a conformational change that allows it to dimerize and bind specifically to DNA
targets. Phospho-specific antibodies directed at Y705 have been used extensively to
determine the relative activity of STAT3 (Levy et al., 2002) and were the next logical
step in our characterization. Using a polyclonal Y705-specific antibody we observed
high levels of immunostaining in the nuclei of DRG neurons, but not in CGNs (Figure
16). Despite the lack of staining of cerebellar granule neurons, GFAP positive cerebellar
astrocytes also present in the CGN culture do express low levels of STAT3 (Figure 17),
providing a positive control and supporting observations made by Xia and colleagues
(2002). Together these data show that STAT3 is both preferentially expressed and active
in DRG neurons when compared to CGNs.
STAT3 expression and activity in vivo
With STAT3 expression in vitro supporting the subtraction, microarray and realtime PCR results, we decided to determine whether it is expressed constitutively in vivo.
In that vein we conducted immunoblots and immunohistochemical analysis of tissue
sections. For the immunoblots, DRG, cerebellum and cortex from the same animal were
each resuspended in lysis buffer and sonicated until homogenized. Because whole DRGs
are a mixed population of cells it was necessary to use beta-III tubulin as a neuronal
loading control. For each replicate, a blot containing titrated amounts of each sample was
55
stained with beta-III tubulin to determine the relative degree of loading for the final blot.
The final blot was probed for total or Y705-phosphorylated STAT3 in addition to the
loading control, and visualized in two channels by infrared detection. Figure 18 shows
that levels of both total and Y705 phosphorylated STAT3 are constitutively much higher
in DRGs than in cerebellum or cortex. As can be seen in the figure, DRGs express two
distinct isoforms of STAT3 named alpha and beta. These isoforms are expressed
elsewhere in the body in different ratios, the alpha form thought to predominate and the
beta form thought to be important for immune function (Yoo et al., 2002). Interestingly,
only a single band appears in the Y705 phosphorylated blot, and it is located at the same
position of the alpha isoform (89kDa). This suggests that although both STAT3 isforms
are expressed in DRG neurons, the alpha form may be more physiologically relevant.
In addition to immunoblots, the spatial aspect of STAT3 expression was
characterized by immunohistochemistry. This was accomplished by sectioning whole
DRGs and cerebellum from the same animal and staining for STAT3 in addition to
several neuronal markers. Because the dissection of dorsal root ganglia from the spinal
column is a labor intensive process and can take several hours to complete, it was
necessary to first perfuse the animal with paraformaldehyde to prevent gene expression
changes. After dissection, DRGs and cerebellum were post-fixed, equilibrated in 30%
sucrose and then embedded into the same tissue block and frozen. 20µM Cryostat
sections were cut and stained for total and Y705 phosphorylated STAT3.
At first glance, DRG sections express high levels of STAT3 immunoreactivity in
a select group of cells (Figure 19). Similar to the results in culture, expression of STAT3
is both cytoplasmic and nuclear, confirming that this was not an artifact of axotomy. By
56
comparison, STAT3 expression in the cerebellum is slight, and only barely more intense
than no-primary controls. Expression of STAT3 in the DRG is neuronal, as shown by colabeling with the neuronal marker NeuN and exclusion from Hoechst positive Schwann
cells in the nerve (Figure 21). STAT3 co-labels extensively with Islet-1, a neural crest
specific transcription factor that localizes to the nucleus (Figure 22). This is particularly
intriguing because Islet-1 has been shown to interact with both JAK1 and STAT3 (Hao et
al., 2005). Y705 phosphorylated STAT3 shows a similar pattern of expression (Figure
20), although immunoreactivity is limited to nuclei, as was seen in culture. Interestingly,
phosphorylated STAT3 co-localizes almost completely with Islet-1, but only in %60 of
neurons (Figure 23). It is unclear whether this expression pattern reflects transient
activation of a specific sub-population of neurons. In sum, these data suggest that
STAT3 is enriched in DRG neurons constitutively, and that a substantial portion of that
STAT3 is in an active state.
Expression of STAT3 mutants in CGNs
Several lines of evidence point to STAT3 mRNA and protein being significantly
upregulated in PNS compared to CNS neurons. In addition, pathway profiling by qPCR
suggests that CNS neurons may lack an entire pathway involving STAT3. The next
obvious question is whether STAT3 plays a role on neurite outgrowth. Mice lacking
STAT3 die at embryonic day 6.5-7.5, making it difficult to assess its role in the nervous
system, which develops later (Takeda et al., 1997). Conditional ablation of STAT3 in
Bal1 positive sensory nodose neurons leads to a threefold increase in pyknotic nuclei by
E18, possibly due to a significant decrease in the expression of CNTF and LIF (Alonzi et
al., 2001). Loss of STAT3 in neurofilament-light positive facial motor neurons does not
57
lead to increased cell death in naïve animals at 6 weeks or 1 year after birth, although two
weeks after a peripheral nerve injury these animals display massive neuronal cell death
(Schweizer et al., 2002). Together these results suggest that STAT3 plays a critical role
in the development of PNS neurons in addition to their response to injury. There are no
reports of the effect of genetic deletion of STAT3 on the conditioning lesion, raising the
interesting question of whether the DRG neurons that survive the peripheral nerve injury
can still regenerate after a dorsal column lesion.
No reports of STAT3 gain of function in CNS neurons exist, making this the next
obvious step for understanding whether it plays a role in neurite outgrowth. Before
moving to an in vivo model, we decided to assess outgrowth of CGNs transfected with
STAT3 and grown on different substrates. The first step in this process was to develop a
protocol for expressing and detecting STAT3 reliably. Our initial attempts at doing so
involved electroporating FLAG tagged STAT3 constructs obtained from Jim Darnell,
who published several of the original papers on STAT3 (Zhong et al., 1994). This
worked efficiently in the COS7 cell line, with FLAG expression detected by both
immunocytochemistry and Western blot. However, when we attempted to used
cerebellar granule neurons, only a small fraction of cells (<1%) were positive for the
FLAG epitope. Successful parallel transfections with green fluorescent protein (GFP)
and FLAG tagged kruppel-like factor 4 (KLF4) confirmed that this failure was most
likely due to the plasmid being used.
Because any number of factors can influence the expression of a plasmid in a
primary neuron, we decided to switch to plasmid that more resembled native STAT3, that
is containing both 5’ and 3’ untranslated regions (UTR) and no epitope. Our lab has had
58
success transfecting several plasmids from the mammalian genome collection (MGC),
confirming that its CMV promoter is sufficient to drive expression of transgenes in
primary neurons. Mus Musculus STAT3-alpha (MGC clone MMM1013-63619), was
transfected into P7-10 neurons using the Nucleofector pulse generator (Amaxa GmbH,
Germany), after which cells were plated on a mixture of poly-D-lysine and laminin and
allowed to grow for 60 hours. Taking advantage of the lack of STAT3 expression in
CGNs, we were able to detect transfected neurons using the pan-STAT3 antibody (Figure
24). Transfection efficiency was routinely above 10%, allowing for multiple cells to be
observed and quantified from a single well.
Overexpression of STAT3 in CGNS results in supernumerary neurite formation
Although we were interested in the effect of STAT3 overexpression in CNS
neurons, it was likely that in the absence of upstream activators, STAT3 might not be
able to dimerize and drive expression of its target genes. We thus sought to examine the
effect of overexpression of several mutants of STAT3-alpha in addition to the wildtype
form. Activation of receptor complexes leads to the recruitment of the Jaks, which in
turn bind to STAT3 via its SH2 domain and phosphorylate tyrosine 705 (Schindler et al.,
1992), a necessary step for STAT3 to effect gene targets. Kaptein and colleagues (1996)
showed that overexpression of a Y705F mutant lead to the inhibition of IL-6 induced
activation of STAT3 in a dominant negative manner. STAT3 can also be made
constitutively active by mutating two cysteine residues in its C-terminal SH2 domain, and
if overexpressed can cause transformation of immortalized fibroblasts and tumor
formation in nude mice (Bromberg et al., 1999). Maximal activation of STAT3 is
thought to be dependent on the phosphorylation of serine 727, although phosphorylation
59
is not required for DNA binding (Wen et al., 1997). Knock-in mice expressing a S727A
mutant form of STAT3 often die perinatally (~75% lethality), with the survivors only
reaching 50-60% of normal body mass by 1 week of age (Shen et al., 2004). This effect
is correlated with a significant reduction in STAT3 transcriptional activity as well as
reduced expression of the insulin-like growth factor 1 (Igf-1).
Site directed mutagenesis was performed on wildtype STAT3-alpha to create four
mutants: STAT3-C (constitutively active), Y705F (dominant negative), S727A (serine
dead) and S727D (serine active). Cerebellar granule neurons were transfected with GFP
or one of these five constructs and grown on glass coverslips for observation by confocal
microscopy. It was initially quite clear that STAT3 transfected CGNs take on a different
morphology from GFP-transfected and untransfected controls. Whereas CGNs grown on
laminin normally have 2-3 primary neurites, STAT3 transfected cells had many more so,
with some cells having 10 or more. This was quantified by high magnification confocal
microscopy, with 10 cells quantified per transfected plasmid with an N of 4 replicates
(240 confocal stacks in total). Maximum projection images were obtained for each stack
and auto-leveled, whereupon primary neurites were counted by hand. Counts were then
compared between experiments directly, and a Student’s unpaired two-tailed T-test was
used to assess the significance of any differences seen. Transfection of STAT3 and
STAT3 mutants, but not GFP, lead to a significant increase in the number of primary
neurites (Figure 25). Surprisingly, STAT3-Y705F led to a significant increase in primary
neurites over wildtype, constitutively active and S727A STAT3. This could possibly be
due to differing roles for STAT3 in the nucleus and cytoplasm and is covered in more
detail in the discussion section. In sum, however, these data support the hypothesis that
60
STAT3 plays a role in neurite outgrowth, particularly during the process of primary
neurite formation.
Overexpression of constitutively active STAT3 in CGNs results in increased neurite
outgrowth
In addition to primary neurite counts, we were also interested to see if STAT3
overexpression had an effect on total neurite length and branching on different substrates.
Although manual techniques are most commonly used to trace neurites, we were able to
make use of the automated tracing algorithm “Neuronal Profiling“that is part of the
Cellomics Kineticscan plate reader software. This required us to scale up the neurite
outgrowth protocols to 96 well format, but gave us the added benefit of being able to
quantify thousands of cells in each experiment. PFA fixed cells were stained for panSTAT3, E7 tubulin, and the nuclear dye Hoechst, with GFP and STAT3 visualized in the
same (green) channel. Neuronal Profiling uses the nuclear signal to identify cells and
then uses the tubulin channel to trace the neurites. The corresponding GFP or STAT3
signal is then assigned to each cell along with neurite outgrowth parameters such as total
neurite outgrowth or branch points. Cell data and images are then exported to Spotfire
DecisionSite for further analysis.
To determine the effect of transfection on neurite outgrowth, the GFP/STAT3
channel was used to threshold cells as transfected or untransfected. This is done by
having a row in the transfection plate containing untransfected cells so that a background
GFP/STAT3 signal can be determined (see Figure 24). The threshold is equal to the
average signal of the GFP/STAT3 signal for untransfected cells plus four standard
deviations. GFP and STAT3 transfected cells with signals higher than this value are
61
deemed to be transfected. A threshold is calculated for each experiment, as it can vary
greatly based on the secondary antibodies and the lamp intensity.
Each experiment yielded several thousand transfected cells which were used to
generate a population average for each of the different neurite outgrowth parameters
calculated by Neuronal Profiling. These population averages were then pooled from
replicates and normalized to GFP controls to generate a percent change for each
parameter. A Student’s unpaired two-tailed T-test was used to assess the significance of
any changes observed. On both poly-D-Lysine and laminin, overexpression of STAT3-C
led to a significant (~20%) increase in total neurite outgrowth (Figure 26). This
difference was also observed in the length of the longest neurite (Figure 27). No
significant difference was seen in other parameters, including total branch points and the
percentage of neurons with neurites. The difference in neurite outgrowth of STAT3-C
transfected neurons seen between experiments is also observable within each experiment
as a shift in the fraction of neurons with neurites longer than a given amount (Figure 28).
Taken together these data suggest that STAT3 mediated transcription leads to an
increased rate of neurite outgrowth.
Materials and methods
Microarray analysis
Total RNA was extracted from three P11 129S6/SvEvTa wild type siblings
utilizing the Qiagen RNeasy Mini kit (Qiagen GmbH, Germany). Samples were
processed at the Case Western Reserve University Cancer Center Gene Expression Array
Core Facility and hybridized to Affymetrix Murine Genome U74AV2 gene chips.
62
Affymetrix GeneChip Operating Software (GCOS) was used to process gene chip images
and generate CEL data files. CEL data from 5 replicates of laser capture microdissected
DRG neurons (described in Peeters et al., 2006) was generously provided to us by Pieter
Peeters of Johnson & Johnson (New Brunswick, NJ).
All Samples were normalized using the Robust Multi-Array average method
(Irizarry et al., 2003) using the software package RMA Express
(http://rmaexpress.bmbolstad.com). Normalized intensities were pooled in Spotfire
DecisionSite (TIBCO Software Inc., Palo Alto CA) and mean DRG/Cerebellum ratios
were calculated. Significance was determined by an unpaired two tailed Student’s T-test.
Significantly upregulated genes (i.e. >2 fold differentially regulated, p<0.01) were
included in further analyses.
Gene ontology clustering
1,086 UniGene IDs from the subtraction along with 1,252 and 892 Affymetrix
probe IDs upregulated in DRGs and cerebellum respectively were imported into DAVID
(Huang da et al., 2007). Annotations overrepresented in the two DRG enriched gene lists
were then determined using the EASE algorithm, which uses a modified Fisher’s Exact
test to determine significance. A list of all annotations and their associated p-values was
determined for the cerebellar microarray list for comparison. DAVID requires that a
reference gene list be used to establish a background for annotations. For the subtraction,
the default Mus Musculus list was used, whereas for the microarray a list of genes present
on the U74Av2 gene chip was used.
63
Promoter analysis
Subtraction and microarray gene lists were imported into the custom software
package “Promolyzer”. Entrez Gene annotations were then used to determine the
transcriptional start site for each gene and the chromosomal reference file containing its
genomic DNA. The software then used the reference files to create a promoter build
consisting of a 1000bp stretch of DNA immediately upstream of the transcription start
site for each gene. Promoters were then matched against two consensus lists: Transfac
10.2 (Biobase GmbH, Germany) and a list of conserved mammalian promoter elements
reported by Carninci and colleagues (2005). Two additional gene lists were then used to
establish a background match rate. For the subtraction, the background list consisted of
all genes containing a transcription start site annotation within Entrez Gene. For the
microarray, the background list contained all genes represented on the U74Av2 chip. A
two-tailed Fisher’s Exact Test was used to determine the significance of a consensus
matching a gene list.
Network analysis
Subtraction and microarray gene lists were imported into Metacore (GeneGo Inc.,
St. Joseph MI) and converted into Metacore network objects. List genes were then
connected to each transcription factor in the database using the “Transcriptional
Regulation” tool. To determine the significance of connectivity between a gene list and a
given transcription factor, Metacore uses an exact test that casts the problem as a
selection without replacement. In this test the intersection between a user’s list and a
generic network is assumed to follow a hypergeometric distribution and the probability of
64
that a subset of a network of size n to include r genes from a user’s gene list is
Where R is the total number of genes with the users list and N is the total number of
genes in the database. From this equation the mean and dispersion of the distribution can
be calculated and used to generate a test statistic known as the z-score, which measures
the relative deviation of r from its expected value. Metacore also calculates a p-value to
determine statistical significance, equivalent to the probability that the intersection of two
randomly selected subsets of N would have a size n or larger:
In the case of a transcription factor centered network, Metacore builds an arbitrary
network of size n around the factor and then compares the network with the genes in R.
Purification of DRG neurons
P7-10 wild type C57BL/6 mice were sacrificed and their spinal column was
removed and placed into L-15 media (Invitrogen, Carlsbad CA). Dorsal root ganglia
were then removed and incubated in a mixture of 0.05% Trypsin-EDTA, 2.4 U/ml
Dispase and 1,000 U/ml Collagenase for 40 minutes. After neutralization with 10% FBS,
DRG cells were triturated using a flame polished Pasteur pipette. Dissociated cells were
then layered on top of a 28% isotonic Percoll solution, as described previously (Tucker et
al., 2006). The cells were then centrifuged for 10 minutes at 400g at 4°C, and the
resulting pellet was resuspended in cell culture media. Neuronal purity was determined
65
using anti neurofilament-medium (160 kDa) and quantified using the Cellomics
Kineticscan plate reader.
Real-time PCR
Purified DRG neurons and dissociated cerebellar granule neurons were pooled
from two animals and then resuspended in Buffer RLT from the Qiagen RNAeasy micro
kit (Qiagen GmbH, Germany). The RNA samples were then homogenized using the
Qiashredder kit and genomic DNA was digested using RNase free DNase from the kit
according to the manufacturer’s protocol. One microgram of total RNA was then reverse
transcribed into first strand cDNA using the “ReactionReady” First Strand cDNA
Synthesis Kit from Superarray (SABiosciences, Frederick MD), followed by incubation
at 95°C to hydrolyze any remaining RNA. Diluted first strand product was then
combined with SYBR-Green master mix from Superarray and aliquoted into JAK/STAT
RT² Profiler plates, containing primers for 84 genes of interest and 5 houskeeping genes,
as well as genomic contamination and quality controls. qPCR reactions were run using
an Eppendorf MasterCycler Realplex (Eppendorf AG, London UK), and threshold cycles
were calculated by the machine. Melt curves were conducted to confirm the presence of
a single product. The fold change for a gene of interest (GOI) using a housekeeping gene
(HKG) for normalization was determined using the “∆∆Ct” method (Livak, 1997) as
follows:
66
Significance was determined using an unpaired two-tailed Student’s T-test on normalized
∆Ct values.
Antibodies
Total (#9132) and Y705 phosphorylated (#9131) STAT3 antibodies from Cell
Signaling Technology (Danvers MA) were used at 1:250 for immunohistochemistry and
1:500 for Western blotting. E7 monoclonal tubulin antibody was used for tracing
neurites (Developmental Studies Hybridoma Bank). Chick anti-BIII tubulin (PRB435P/TuJ1) was used at 1:10,000 to determine neuronal loading by Western blot
(Covance Inc., Princeton NJ). Monoclonal anti-GFAP (#G3893, Sigma-Aldrich, St.
Louis MO) was used at 1:500 to identify astrocytes. NeuN antibody was obtained from
Chemicon (Millipore Inc., Billerica MA) and used at 1:1000. Anti-Islet1 (clone 39.4D5)
was used at 1:500 and obtained from the Developmental Studies Hybridoma bank.
Cell culture
P7-10 C57BL/6 mice were euthanized and their DRGs and cerebellum were
removed. Neurons were dissociated and plated on 18mm glass coverslips (Carolina
Biological Supply, Burlington NC) or 96 well Packard plates (#6005182, Perkin Elmer,
Waltham MA) coated with 100µg/ml poly-D-lysine (Sigma #P7886) alone or with
4µg/ml laminin (Sigma # L2020). Neurons were cultured in defined Neurobasal medium
(Invitrogen #12348017) containing 100 U/ml penicillin and 100µg/ml streptomycin
(Invitrogen #15140122), 5µg/ml insulin (Sigma #I6634), 100µM sodium pyruvate
(Invitrogen #11360070), 4µg/ml triiodo-thyronine (Sigma #T6397), 200µM L-glutamine
(Invitrogen #25030149), 50µg/ml N-acetyl-L-cysteine (Sigma #A8199), 1X B27
67
supplement (Invitrogen #17504044), 100µg/ml transferrin (Sigma #T1147), 100µg/ml
bovine-serum albumen (Sigma #A4161), 63ng/ml progesterone (Sigma #P8783),
16µg/ml putrescine (Sigma #P7505) and 400ng/ml sodium selenite (Sigma #S5261).
When making comparisons between DRGs and CGNs, 50ng/ml of NGF was added to
each well. To improve CGN transfection efficiency and cell survival after transfection
we added 10ng/ml forskolin to the plating media. Transfected CGNs also received
25mM KCl. Cells were grown for 48 or 96 hours at 37°C and 5% CO2, prior to fixation
with 4% paraformaldehyde.
Western blotting
Freshly dissected whole DRGs, cerebellum and cortex from P7-10 C57BL/6 mice
were suspended in lysis buffer containing 1% NP-40, 200µM sodium vanadate and 50µM
sodium fluoride, followed by sonication. Dilutions of each sample were run on an 8-15%
gradient gel (Lonza Inc., Switzerland) and probed with TuJ1 to determine neuronal
loading. Densitometric analysis was then used to adjust sample loading for a second gel.
STAT3 and TuJ1 signals were read simultaneously using the Odyssey infrared imager
(LI-COR Inc., Lincoln NE).
Immunohistochemistry
P7-10 C57BL/6 mice were perfused with saline followed by 4% PFA, whereupon
their DRG and cerebellum were removed. The tissue was then post-fixed overnight,
followed by equilibration in 30% sucrose. The cerebellum and DRGs of the same mouse
were mounted into the same OCT block and sectioned by cryostat at a thickness of 20µm
and places on slides. Because PFA masks the Y705 antigen, we used a simple antigen
68
retrieval technique described previously (Jiao et al., 1999) and consisting of a 30 minute
incubation in 25mM sodium citrate at 80°C. Sections were incubated for 2 hours in
blocking/permeabilization buffer containing 20% normal goat serum (Invitrogen
#16210072), 0.2% Triton X-100, 18mg/ml L-lysine (Sigma # L5501) and 10mg/ml BSA.
Slides were then incubated in primary antibody overnight and then secondary antibody
for a second night, followed by mounting with ProLong Gold anti-fade reagent (P36930,
Invitrogen). Sections were imaged by conventional epifluorescence and confocal
microscopy.
CGN neurite outgrowth assay
1 million cells per condition were resuspended in 100µl mouse neuron
Nucleofector solution (Amaxa GmbH, Germany) and 3µg of plasmid DNA was added.
Cells were electroporated using program G-013 on the Amaxa Nucleofector device, after
which 900µl of media (described earlier) was added to each cuvette to prevent cell death.
Cells were then plated at a density of 20-30,000 cells per 96 well.
After 48 hours, cells were fixed with 4% PFA and incubated for 45 minutes in the
same blocking buffer described above, followed by anti-STAT3 (1:250) and E7 (1:1,000)
for 90 minutes. Images of 9 fields per well were taken using the Cellomics Kineticscan
plate reader (Thermo Scientific, Waltham MA) and neurites were traced using the
Neuronal Profiling algorithm. Tracing data was imported into Spotfire DecisionSite for
further analysis.
69
Chapter Figures
Figure 6. STAT3 signaling pathway. Binding of cytokines to cell surface receptors
triggers the recruitment of Janus Kinases (Jaks) and the eventual phosphorylation of
STAT3, leading to dimerization and the transcription of target genes.
70
Figure 7. In silico microarray analysis of genes differentially regulated between the PNS
and CNS. A. Scatter plot of probes found to be significantly overexpressed in adult laser
captured DRG neurons and P10 cerebellar tissue. Genes that were two-fold upregulated
with p<0.01 (determined by a two-tailed unpaired Student’s t-test) were selected for
analysis by DAVID and Metacore. 1294 genes were found to be upregulated in DRG
neurons whereas 897 genes were found to be upregulated in cerebellar granule neurons.
DRG: n=5, Cerebellum: n=3. B. Hierarchical clustering of probe signals from the 8 MGU74Av2 chips. Red and green bars denote regions of high and low signal respectively.
Representative mammalian transcription factors are listed on the right.
71
Figure 8. Overrepresented ontological annotations describing DRG enriched genes.
Summary of gene ontology terms found to be significantly overrepresented in DRG
enriched genes lists (MA = Microarray). Bars represent the Benjamini-Hochberg
corrected EASE p-value determined by DAVID. Terms with a corrected p-value <0.05
were considered. Example terms that were found only in the subtraction and microarray
lists are also included for comparison.
72
Figure 9. Overrepresented mammalian transcription factor binding sites in the
promoters of DRG enriched genes. A. Scatter plot of transcription factor binding
consensuses found to be significantly overrepresented in the promoter regions of DRG
enriched genes. 1200bp promoter regions were scanned for consensus binding sites from
Transfac 10.2 and from a paper by Carninci et al. (2005). Significance was determined
by a two-tailed Fisher’s Exact Test. B. Distribution of STAT binding sites within the
promoter regions of 126 DRG enriched genes. Many sites cluster within 200bp upstream
of the transcription start site (TSS).
73
Figure 10. Identification of transcription factor centered interaction networks. A. 1,068
unique UniGene IDs found to be enriched in DRG neurons by the subtraction were
imported into Metacore, whereupon direct protein-protein and protein-DNA interactions
were illuminated. Most of DRG enriched genes did not form direct connections, possibly
due to a lack of interaction data. B. 191 of the DRG enriched genes form an “interacting
cluster”, containing secreted factors, cell surface receptors, kinases and transcription
factors. C. Metacore was used to predict transcription factors that serve regulatory roles
within DRG enriched gene pools. The STAT3 centered sub-network from the subtraction
has 56 nodes, 23 of which are close neighbors (shown).
74
Figure 11. DRG enriched transcription factor sub-networks. A. Summary of
transcription factor centered sub-networks significantly overrepresented in the DRG
enriched gene pool compared to the cerebellum enriched gene pool. P-values were
determined using an exact test developed for Metacore and discussed in the materials and
methods section, followed by a Benjamini-Hochberg correction for multiple
comparisons. B. Venn diagram depicting transcription factors identified by the network
analysis from both the subtraction and microarray gene lists. Sample transcription factors
are listed in each category.
75
Figure 12. Purification of DRG neurons. P10 mouse DRG cultures contain multiple cell
types, including satellite cells, Schwann cells and fibroblasts which are resistant to antimitotic compounds such as FdUR. A. An unpurified culture of DRG neurons stained for
tubulin (red), the neuronal marker HuD (green) and the nuclear dye Hoechst (blue). B.
DRG neurons after purification by centrifugation through a cushion of 28% isotonic
Percoll. Neuronal purity of the cultures was determined to be at least 80%. Scale bars:
100um.
76
Figure 13. Upstream regulators and downstream targets of STAT3 are enriched in DRG
neurons. Summary of upstream regulators and downstream targets of STAT3 whose
mRNAs were found to be significantly enriched in purified DRG neurons compared to
cerebellum by real-time PCR. Fold enrichments are listed next to each gene.
77
Figure 14. STAT3 is highly expressed in DRG neurons. A. Maximal projection confocal
micrograph of a mixed DRG culture stained for STAT3 (green) and heavy neurofilament
(NF200, red). B. A single frame from a higher Z point in the same confocal stack,
showing nuclear localization of the STAT3 protein. C. Profile trace of the line shown in
(A), showing much higher fluorescence in the DRG than the surrounding cells. Scale
bars: 50µm.
78
Figure 15. DRG neurons express higher levels of total STAT3 than CGNs in vitro.
Cultures of dorsal root ganglia and cerebellar granule neurons from the same animal were
stained for STAT3 after being cultured for 4 days on laminin coated coverslips. A-C. A
single DRG neuron stained for (A) E7, (B) total STAT3 and (C) the nuclear dye Hoechst.
D-F. Three cerebellar granule neurons stained for (D) E7, (E) total STAT3 and (F) the
nuclear dye Hoechst. Scale bars: 10µm.
79
Figure 16. DRG neurons express high levels of nuclear Y705 phosphorylated STAT3.
Cultures of dorsal root ganglia and cerebellar granule neurons from the same animal were
fixed with ice-cold methanol and stained for Y705 phosphorylated STAT3 after being
cultured for 4 days on laminin coated coverslips. A-C. A single DRG neuron stained for
(A) E7, (B) Y705 phosphorylated STAT3 and (C) the nuclear dye Hoechst. D-F. Four
cerebellar granule neurons stained for (D) E7, (E) Y705 phosphorylated STAT3 and (F)
the nuclear dye Hoechst. Scale bars: 10µm.
80
Figure 17. STAT3 is expressed in cerebellar astrocytes in vitro. P7-10 cerebellar
cultures were stained for STAT3 after being cultured for 4 days on laminin. A. Confocal
micrograph of a culture stained for E7, showing multiple cells and processes. B. The
same field stained for total STAT3, which is restricted to a large cell in the center. C. A
merged image of the same field with a nuclear stain, showing several cerebellar granule
neurons that are negative for STAT3 (white arrows). D. Confocal micrograph of a
culture stained with the glial marker GFAP. E. The same field stained for total STAT3.
F. Merged image of the same field, showing that the STAT3 positive cell is also GFAP
positive. Scale bars: 10µm.
81
Figure 18. DRGs express higher levels of STAT3 than cerebellum and cortex in vivo. A.
Representative Western blot of lysates from P7-10 mouse dorsal root ganglia (1),
cerebellum (2) and cerebral cortex (3) probed for total STAT3. Neuronal specific β3tubulin was used as a loading control. Both the 89kDa “alpha” and 80kDa “beta” band
are visible. B. A similar blot probed for Y705 phosphorylated STAT3. Only the 89kDa
“alpha” band was observed to be phosphorylated.
82
Figure 19. STAT3 is expressed in DRGs in vivo. P7-10 mice were perfused with PFA,
after which their cerebellum and dorsal root ganglia were dissected and mounted into the
same tissue block. 20um cryostat sections containing both tissues were stained for total
STAT3 and imaged by confocal microscopy. A. A single dorsal root ganglion showing
high levels of STAT3. B. Cerebellar tissue on the same slide showing lower levels of
STAT3. C. High magnification confocal micrograph of (A), showing cytoplasmic and
nuclear localization of STAT3. Scale bars: A and B, 100µm; C, 10µm.
83
Figure 20. STAT3 is phosphorylated on tyrosine 705 in vivo. P7-10 mice were perfused
with PFA, after which their cerebellum and dorsal root ganglia were dissected and
mounted into the same tissue block. 20um cryostat sections containing both tissues were
incubated in sodium citrate to retrieve the antigen after formaldehyde fixation, and then
stained using a Y705 specific antibody. A. A single dorsal root ganglion showing high
levels of phosphorylated STAT3. B. Corresponding cerebellar tissue showing relatively
lower levels of phosphorylated STAT3. C. High magnification view of (A), showing
nuclear staining of phosphorylated STAT3. Scale bars: A and B, 100µm; C, 10µm.
84
Figure 21. STAT3 is constitutively expressed in neurons in vivo. A-C. Cross-section of
a dorsal root ganglion dissected after perfusion with PFA and stained for (A) NeuN, (B)
total STAT3 and (C) the nuclear dye Hoechst. D. Merged image showing colocalization
between STAT3 and NeuN but not between STAT3 and Hoechst. E-F. High
magnification confocal micrograph of (E) NeuN and (F) STAT3 colocalization, along
with the merged image (G). Scale bars: A-D, 100µm; E-G, 10µm.
85
Figure 22. STAT3 colocalizes with Islet-1. A-B. Confocal micrographs from a crosssection of dorsal root ganglion dissected after perfusion with PFA and stained for (A)
Islet-1 and (B) STAT3. C. Merged image showing colocalization between Islet-1 and
STAT3. D-F. High magnification confocal micrographs of the same section stained for
(D) Islet-1 and (E) STAT3, along with the merged image (F). Scale bars: A-C, 100µm;
D-F, 10µm.
86
Figure 23. Y705 phosphorylated STAT3 is present in neuronal nuclei. High
magnification confocal micrographs of P7-10 dorsal root ganglion sections subjected to
antigen retrieval and stained for A. NeuN, B,E. Y705-phosphorylated STAT3 and D.
Islet-1. E-F. Merged images showing colocalization of STAT3 with (E) NeuN and (F)
Islet-1 in subsets of DRG nuclei. Scale bars: 10µm.
87
Figure 24. STAT3 can be exogenously expressed in cerebellar granule neurons. A.
Western blot of COS7 cells transfected with GFP (1) and STAT3 (2) and stained for total
STAT3 and β3-tubulin. B. Confocal micrograph of cerebellar granule neuron cultures
transfected with STAT3-alpha and stained for β3-tubulin (red) and total STAT3 (green).
Scale bar: 50µm. C. Cellomics KSR acquired distribution of “FITC” (i.e. GFP/STAT3)
intensity in cerebellar granule neurons transfected with no plasmid, GFP, or STAT3.
88
Figure 25. Exogenous expression of STAT3 in cerebellar granule neurons increases
primary neurite count. A. Mean primary neurite counts of CGNs cultured on laminin for
2 days after transfection with GFP, wildtype STAT3 –alpha or one of four activation
mutants (CA = constitutively active A662C/N664C, DN = dominant negative Y705F,
S727A/D = STAT3 with mutations at serine 727, thought to be important for activation).
Significance was determined by an unpaired two-tailed student’s t-test (n=4). Error bars
= 95% confidence intervals, * = p<0.05, ** = p<0.01. B-G. High magnification confocal
micrographs of individual CGNs transfected with (B) GFP, (C) wildtype STAT3-alpha,
(D) constitutively active STAT3, (E) dominant negative STAT3, (F) S727D mutant STAT3
and (G) S727A mutant STAT3. Scale bars = 10µm.
89
Figure 26. Exogenous expression of constitutively active STAT3 in cerebellar granule
neurons increases total neurite length. Mean total neurite length of CGNs cultured for 2
days on A. poly-D-lysine or B. laminin after transfection with GFP, wildtype STAT3alpha or one of four activation mutants (CA = constitutively active A662C/N664C, DN =
dominant negative Y705F, S727A/D = STAT3 with mutations at serine 727, thought to be
important for activation). Significance was determined by an unpaired two-tailed
student’s t-test (n=4). Error bars = 95% confidence intervals, * = p<0.05, ** = p<0.01.
90
Figure 27. Exogenous expression of constitutively active STAT3 in cerebellar granule
neurons increases the length of the longest neurite. Mean longest neurite length of CGNs
cultured for 2 days on A. poly-D-lysine or B. laminin after transfection with GFP,
wildtype STAT3-alpha or one of four activation mutants (CA = constitutively active
A662C/N664C, DN = dominant negative Y705F, S727A/D = STAT3 with mutations at
serine 727, thought to be important for activation). Significance was determined by an
unpaired two-tailed student’s t-test (n=4). Error bars = 95% confidence intervals, * =
p<0.05, ** = p<0.01.
91
Figure 28. Overexpression of constitutively active STAT3 alters the distribution of
neurite lengths within an experiment. Representative chart showing the cumulative
distribution of total neurite lengths for 95% of all CGNs cultured for 2 days on poly-Dlysine after transfection with GFP, wildtype STAT3-alpha, constitutively active STAT3 or
dominant negative STAT3.
92
Chapter Tables
Table 2. Listing of 1,068 unique UniGene IDs identified from the subtracted library
using EST Express. Cross -references for Entrez Gene and RefSeq are also listed.
UniGene GeneID Mm.103413 Mm.281086 Mm.39265 Mm.347369 Mm.121014 Mm.133063 Mm.27608 Mm.30056 Mm.331487 Mm.332931 Mm.390979 Mm.150977 Mm.285785 Mm.268895 Mm.349174 Mm.251306 Mm.27711 Mm.25311 Mm.41451 Mm.166485 Mm.46699 Mm.77697 Mm.140672 Mm.292672 Mm.276721 Mm.383824 Mm.21450 Mm.188413 Mm.250425 Mm.41261 Mm.250193 Mm.174044 Mm.287157 Mm.41180 68350 66839 104457 106264 76897 66123 73737 74778 66185 319202 74157 71997 68964 207740 72012 67077 67326 66270 66279 66294 208501 76527 100910 72357 76429 66374 67695 116972 102122 75687 70435 214764 67236 75695 Symbol
0610009K11Rik 0610009O20Rik 0610010K14Rik 0610012G03Rik 0710005M24Rik 1110006O24Rik 1110008P14Rik 1110014J01Rik 1110037F02Rik 1200016E24Rik 1300018I05Rik 1500002O20Rik 1500010J02Rik 1500031H01Rik 1600020E01Rik 1700019N12Rik 1700037H04Rik 1810015C04Rik 1810021J13Rik 1810037C20Rik 1810043H04Rik 2010004A03Rik 2010209O12Rik 2210016L21Rik 2310007H09Rik 2310011J03Rik 2310016E02Rik 2310047D13Rik 2310065K24Rik 2310066E14Rik 2610204M08Rik 2700050L05Rik 2810452K22Rik 2900002H16Rik UniGene GeneID Symbol
Mm.274285 Mm.296545 Mm.391339 Mm.61022 Mm.44153 Mm.202257 Mm.254267 Mm.16660 Mm.34650 Mm.151640 Mm.242865 Mm.329076 Mm.390957 Mm.46016 Mm.3401 Mm.371609 Mm.247564 Mm.4146 Mm.34775 Mm.171484 Mm.343951 Mm.299399 Mm.336205 Mm.9052 Mm.28864 Mm.16898 Mm.371997 Mm.34346 Mm.274830 Mm.18539 Mm.29836 Mm.273204 Mm.6483 Mm.274158 74143 68767 207375 74309 71720 218885 170719 18453 18483 74450 100715 78309 18515 76477 18552 56330 207728 18596 18597 94184 18613 224824 18655 53328 74451 236539 236539 27280 75669 68943 18717 18797 18807 69217 Opa1 ORF19 ORF34 Osbp2 Osbpl3 Oxnad1 Oxr1 P4hb Palm Pank2 Papd4 Parc Pbx2 Pcolce2 Pcsk5 Pdcd5 Pde2a Pdgfrb Pdha1 Pdxdc1 Pecam1 Pex6 Pgk1 Pgrmc1 Pgs1 Phgdh Phgdh Phlda3 Pik3r4 Pink1 Pip5k1c Plcb3 Pld3 Plekha4 93
Mm.27344 Mm.212543 Mm.196596 Mm.3285 Mm.24652 Mm.213582 Mm.24593 Mm.179378 Mm.35583 Mm.125857 Mm.61180 Mm.323925 Mm.296838 Mm.76166 Mm.394573 Mm.390264 Mm.74750 Mm.352212 Mm.124498 Mm.210949 Mm.251789 Mm.392040 Mm.271946 Mm.405337 Mm.337771 Mm.341004 Mm.379345 Mm.258939 Mm.299636 Mm.255464 Mm.282039 Mm.356689 Mm.28521 Mm.183102 Mm.359359 Mm.268527 Mm.215814 Mm.31227 Mm.383431 Mm.313169 Mm.329684 72931 78403 76482 73218 77574 77590 97820 75317 104859 108654 71023 223739 230757 68176 70744 67412 243300 76213 268807 71617 101351 330166 338371 320516 192950 27404 26357 109934 171209 11423 104112 11430 226977 74117 11496 66548 242669 228061 195040 382643 107239 2900010J23Rik 2900041M22Rik 3110002H16Rik 3110056O03Rik 3321401G04Rik 4631426J05Rik 4833439L19Rik 4930547N16Rik 4930573I19Rik 4933403F05Rik 4933407C09Rik 5031439G07Rik 5730409E04Rik 6230427J02Rik 6330403L08Rik 6330407J23Rik 6430598A04Rik 6530406A20Rik 8230402K04Rik 9130011E15Rik A130022J15Rik A230057G18Rik A730011L01Rik A730060N03Rik AB182283 Abca8b Abcg2 Abr Accn3 Ache Acly Acox1 Actr1b Actr3 Adam22 Adamtsl5 Adc Agps AI316787 AI450948 AI790298 Mm.3741 Mm.29475 Mm.1268 Mm.3085 Mm.301655 Mm.293761 Mm.35788 Mm.197493 Mm.279781 Mm.196045 Mm.332855 Mm.221181 Mm.250508 Mm.103382 Mm.74208 Mm.40577 Mm.188709 Mm.140 Mm.259626 Mm.211211 Mm.30039 Mm.28561 Mm.71 Mm.358657 Mm.30660 Mm.2477 Mm.44183 Mm.9431 Mm.5875 Mm.257354 Mm.105331 Mm.208883 Mm.30210 Mm.21728 Mm.98 Mm.280013 Mm.273152 Mm.389499 Mm.193688 Mm.277916 Mm.258771 233765 67220 18823 67784 73078 140484 18972 67811 67486 68273 66537 18998 66812 67533 223828 242083 329251 18938 26932 74182 19084 18762 19090 28000 70052 19132 233895 19142 76560 77754 101739 26442 26443 26444 19175 70247 57296 56742 19244 19246 19280 Plekha7 Plekho1 Plp1 Plxnd1 Pmpcb Pofut1 Pold2 Poldip2 Polr3g Pomgnt1 Pomp Pou4f3 Ppcdc Ppfibp1 Pphln1 Ppm1l Ppp1r12b Ppp1r14b Ppp2r5e Prei4 Prkar1a Prkcz Prkdc Prpf19 Prpf4 Prph Prr14 Prss12 Prss8 Prune2 Psip1 Psma5 Psma6 Psma7 Psmb6 Psmd1 Psmd8 Psrc1 Ptp4a2 Ptpn1 Ptprs 94
Mm.40038 Mm.30085 Mm.6645 Mm.302724 Mm.273571 Mm.311854 Mm.274093 Mm.41989 Mm.2787 Mm.130422 Mm.275554 Mm.250989 Mm.269088 Mm.238343 Mm.274816 Mm.253090 Mm.5159 Mm.384171 Mm.25405 Mm.305796 Mm.5061 Mm.329396 Mm.1383 Mm.240298 Mm.287267 Mm.22085 Mm.29430 Mm.337038 Mm.260193 Mm.203747 Mm.22547 Mm.358931 Mm.38010 Mm.280103 Mm.276137 Mm.340818 Mm.220823 Mm.30455 Mm.231450 Mm.92705 Mm.69013 72168 58810 11651 11656 212647 11682 11685 23801 17117 217473 68420 74251 11737 12306 11764 11772 11787 11789 11838 56292 11841 71302 14570 234094 213498 56350 65105 76709 109689 216869 11886 245860 108147 11928 11946 11975 110935 54667 233871 27215 329739 Aifm3 Akr1a4 AKT1 Alas2 Aldh4a1 Alk Alox12e Aloxe3 Amacr Ankmy2 Ankrd13a Ankrd9 Anp32a Anxa2 Ap1b1 Ap2a2 Apbb2 Apc Arc Ard1 Arf2 Arhgap26 Arhgdig Arhgef10 Arhgef11 Arl3 Arl6ip4 Arpc2 Arrb1 Arrb2 Asah1 Atg9a Atic Atp1a1 Atp5a1 Atp6v0a1 Atp6v1b1 Atp8b2 Atxn2l Azi2 B430201A12Rik Mm.182563 Mm.21936 Mm.12815 Mm.87216 Mm.9286 Mm.3272 Mm.270975 Mm.298274 Mm.28231 Mm.42150 Mm.132868 Mm.4480 Mm.279194 Mm.28275 Mm.4876 Mm.279907 Mm.293588 Mm.249966 Mm.28840 Mm.260943 Mm.1116 Mm.148877 Mm.196208 Mm.44606 Mm.100348 Mm.396875 Mm.687 Mm.57052 Mm.28551 Mm.4009 Mm.288183 Mm.228903 Mm.249986 Mm.388529 Mm.262859 Mm.294120 Mm.214569 Mm.27792 Mm.247556 Mm.307846 Mm.22723 76308 216344 19344 56187 19364 51801 19387 107746 71660 19419 213464 19647 236732 19655 19672 27632 225362 19697 19712 56210 66932 19687 71729 56533 52882 52882 11852 14787 66878 19769 66089 56736 108660 80751 73469 67588 56698 66049 19893 19934 268449 Rab1b Rab21 Rab5b Rabggta Rad51l3 Ramp1 Rangap1 Rapgef1 Rarres2 Rasgrp1 Rbbp5 Rbbp6 Rbm10 Rbmx Rcn1 Rdbp Reep2 Rela Rest Rev1 Rexo1 Rfc1 Rgs12 Rgs17 Rgs7bp Rgs7bp Rhob Rhpn1 Riok3 Rit1 Rmnd5b Rnf14 Rnf187 Rnf34 Rnf38 Rnf41 Rnuxa Rogdi Rpgr Rpl22 Rpl23a 95
Mm.44239 Mm.309498 Mm.272462 Mm.203962 Mm.43745 Mm.25556 Mm.224076 Mm.11473 Mm.247335 Mm.151577 Mm.259234 Mm.26272 Mm.29236 Mm.4598 Mm.25848 Mm.6967 Mm.89667 Mm.226175 Mm.294783 Mm.288546 Mm.2436 Mm.197387 Mm.24021 Mm.104531 Mm.131074 Mm.131700 Mm.244177 Mm.393367 Mm.390374 Mm.250641 Mm.389259 Mm.128292 Mm.101504 Mm.333624 Mm.259334 Mm.331907 Mm.275934 Mm.179091 Mm.257404 Mm.379226 Mm.9772 70369 268515 53761 224727 193742 79554 217827 230789 217216 268747 232400 213056 57278 12032 12039 12050 12055 77578 72567 544971 20893 76895 233016 69168 67991 320701 77521 99061 69183 77480 109299 442847 320827 232933 235386 97130 97863 69008 104248 73660 12289 Bag5 Bahcc1 Bat2 Bat3 Bat5 BC002216 BC002230 BC008163 BC030867 BC036313 BC048546 BC049806 Bcam Bcan Bckdha Bcl2l2 Bcl7c Bcl9 Bclaf1 Bdp1 Bhlhb2 Bicd2 Blvrb Bola1 Btbd14a C130034I18Rik C130038G02Rik C130057N11Rik C1qtnf2 C330002I19Rik C330006A16Rik C330046G13Rik C530008M17Rik C530028I08Rik C630028N24Rik C77080 C78339 Cab39l Cabin1 Cabp4 Cacna1d Mm.304790 Mm.188544 Mm.391885 Mm.379006 Mm.4071 Mm.13705 Mm.288606 Mm.131949 Mm.192580 Mm.82661 Mm.3925 Mm.100144 Mm.303924 Mm.255066 Mm.297199 Mm.34562 Mm.29594 Mm.35483 Mm.201455 Mm.83840 Mm.193096 Mm.267377 Mm.247042 Mm.89981 Mm.103584 Mm.389273 Mm.41557 Mm.247457 Mm.206536 Mm.253771 Mm.246965 Mm.173119 Mm.250605 Mm.29812 Mm.274399 Mm.38450 Mm.34329 Mm.218473 Mm.2044 Mm.339676 Mm.200120 268449 103963 75617 20090 16785 81910 106298 212892 68585 237847 20198 20200 66406 224903 217125 20227 53890 77980 107767 71145 20249 20249 20264 24046 20271 20272 69938 74616 20970 68112 67680 218811 20338 93684 20362 53860 109079 26943 20317 58172 170742 Rpl23a Rpn1 Rps25 Rps29 Rpsa Rrbp1 Rrn3 Rshl3 Rtn4 Rtn4rl1 S100a4 S100a6 Sac3d1 Safb Samd14 Sart1 Sart3 Sbf1 Scamp1 Scara5 Scd1 Scd1 Scn10a Scn11a Scn5a Scn7a Scrn1 Scrn3 Sdc3 Sdccag3 Sdhb Sec24c Sel1l Sep15 Sep8 Sep9 Sephs1 Serinc3 Serpinf1 Sertad2 Sertad3 96
Mm.330524 Mm.178322 Mm.36834 Mm.318846 Mm.326847 Mm.27736 Mm.306954 Mm.40120 Mm.270703 Mm.383737 Mm.210403 Mm.101586 Mm.390150 Mm.78373 Mm.292465 Mm.275710 Mm.344071 Mm.289900 Mm.3460 Mm.29742 Mm.182412 Mm.28219 Mm.6839 Mm.142275 Mm.289427 Mm.195663 Mm.284503 Mm.24169 Mm.290563 Mm.161470 Mm.269991 Mm.288729 Mm.38910 Mm.10160 Mm.178246 Mm.333388 Mm.259916 Mm.387173 Mm.73777 Mm.168478 Mm.252145 94332 260299 227634 100072 12337 12348 23831 192160 213819 140721 104479 67433 245902 71870 269336 51938 76380 12468 12475 12484 69957 52858 12567 12569 66971 12575 110911 12607 12615 229841 545389 16328 28135 106143 216848 107932 218865 12655 67064 75786 12716 Cadm3 Cadm4 Camsap1 Camta1 Capn5 Car11 Car14 Casc3 Casd1 Caskin2 Ccdc117 Ccdc127 Ccdc15 Ccdc19 Ccdc32 Ccdc39 Ccdc46 Cct7 Cd14 Cd24a Cdc16 Cdipt Cdk4 Cdk5r1 Cdk5rap1 Cdkn1a Cds2 Cebpz Cenpa Cenpe Cep170 Cep250 Cep63 Cggbp1 Chd3 Chd4 Chdh Chi3l3 Chmp1b Ckap5 Ckmt1 Mm.250391 Mm.192111 Mm.284505 Mm.21841 Mm.292016 Mm.31537 Mm.30068 Mm.22240 Mm.33343 Mm.262320 Mm.35325 Mm.291029 Mm.33832 Mm.28632 Mm.19325 Mm.1056 Mm.288697 Mm.30087 Mm.212813 Mm.142455 Mm.21587 Mm.391941 Mm.272920 Mm.286593 Mm.274232 Mm.26412 Mm.4628 Mm.27735 Mm.24082 Mm.282800 Mm.1541 Mm.271891 Mm.28160 Mm.273379 Mm.130 Mm.5068 Mm.43375 Mm.252722 Mm.299906 Mm.24222 Mm.204969 208043 73251 234373 20382 225027 67158 52551 73723 227700 225608 68346 72198 108052 80879 66859 20514 74011 14385 106068 20540 65962 20874 17129 20586 54380 24061 20597 75788 330959 20618 20648 56440 76742 69178 12703 20410 66042 70997 56381 56632 20740 Setd1b Setd7 Sfrs14 Sfrs2 Sfrs7 Sft2d3 Sgta Sh3bgrl3 Sh3glb2 Sh3tc2 Sirt5 Skiv2l2 Slc14a1 Slc16a3 Slc16a9 Slc1a5 Slc25a27 Slc37a4 Slc45a4 Slc7a7 Slc9a3r2 Slk Smad5 Smarca4 Smarcal1 Smc1a Smpd1 Smurf1 Snapc5 Sncg Snta1 Snx1 Snx27 Snx5 Socs1 Sorbs3 Sostdc1 Spef1 Spen Sphk2 Spna2 97
Mm.270587 Mm.40132 Mm.280563 Mm.169673 Mm.3990 Mm.37984 Mm.300931 Mm.277735 Mm.249555 Mm.738 Mm.181021 Mm.7281 Mm.10299 Mm.334994 Mm.40 Mm.276859 Mm.196884 Mm.41593 Mm.272093 Mm.17616 Mm.313558 Mm.21974 Mm.30199 Mm.29873 Mm.298893 Mm.291928 Mm.255858 Mm.359633 Mm.249744 Mm.22560 Mm.29196 Mm.154358 Mm.287279 Mm.123240 Mm.29702 Mm.28207 Mm.244971 Mm.59171 Mm.258310 Mm.240830 Mm.319038 26373 64945 66864 216705 20480 215748 12817 12842 12825 12826 12827 12831 12832 53867 26572 23789 54188 171508 54394 26896 72832 12988 27373 103236 12995 12387 22083 19025 52463 109754 66445 76884 218978 223658 28146 28027 244418 434147 272636 13132 13135 Clcn7 Cldn12 Clec14a Clint1 Clpb Cnksr3 Col13a1 Col1a1 Col3a1 Col4a1 Col4a2 Col5a1 Col5a2 Col5a3 Cops3 Coro1b Cpsf4 Creld1 Crlf3 Crsp2 Crtac1 Csk Csnk1e Csnk1g2 Csnk2a1 Ctnnb1 Ctr9 Ctsa Cxxc6 Cyb5r3 Cyc1 Cyfip2 D14Ertd436e D330001F17Rik D3Ucla1 D4Wsu132e D8Ertd82e D930028M14Rik D9Ertd280e Dab2 Dad1 Mm.123110 Mm.42038 Mm.261906 Mm.274794 Mm.363404 Mm.5222 Mm.180337 Mm.130952 Mm.249934 Mm.34064 Mm.645 Mm.8235 Mm.17461 Mm.65336 Mm.391038 Mm.21612 Mm.285400 Mm.24433 Mm.245577 Mm.247956 Mm.278578 Mm.130902 Mm.260545 Mm.230301 Mm.340436 Mm.391153 Mm.34580 Mm.273210 Mm.24670 Mm.391476 Mm.26688 Mm.320644 Mm.3951 Mm.206505 Mm.4871 Mm.23987 Mm.4342 Mm.104744 Mm.24255 Mm.136511 Mm.28626 20742 24066 14794 79043 20813 75956 70356 76630 20848 20851 20866 20868 20872 57740 268980 97387 217517 20610 20935 71733 225888 68188 56403 20972 216965 209478 56480 83383 102791 21766 21826 330267 21838 21858 21859 117149 21872 21873 21887 228012 69742 Spnb2 Spry4 Spsb2 Spsb3 Srp14 Srrm2 St13 Stambpl1 STAT3 Stat5b Stim1 Stk10 Stk16 Stk32c Strn Strn4 Stxbp6 Sumo3 Surf6 Susd2 Suv420h1 Sympk Syncrip Syngr1 Taok1 Tbc1d12 Tbk1 Tcfap4 Tcta Tex261 Thbs2 Thsd7a Thy1 Timp2 Timp3 Tirap Tjp1 Tjp2 Tle3 Tlk1 Tm2d2 98
Mm.10294 Mm.148693 Mm.291037 Mm.28733 Mm.167537 Mm.298947 Mm.307515 Mm.28222 Mm.220038 Mm.280594 Mm.314923 Mm.35399 Mm.44995 Mm.221499 Mm.60526 Mm.279692 Mm.21353 Mm.288159 Mm.44736 Mm.203949 Mm.331970 Mm.336625 Mm.94371 Mm.247695 Mm.293973 Mm.330671 Mm.23916 Mm.1791 Mm.181430 Mm.249479 Mm.23296 Mm.153688 Mm.275426 Mm.4772 Mm.258927 Mm.25997 Mm.422725 Mm.180719 Mm.389421 Mm.359137 Mm.422907 13144 70248 13178 75901 69654 13202 71986 68278 13207 67379 104418 216877 208666 214240 233335 56445 230935 54152 13429 114255 53902 13518 13527 207521 224907 252864 66959 67603 13424 13427 50496 93726 70790 14745 66656 13631 209589 245190 381438 384885 433230 Dapk3 Dazap1 Dck Dcp1a Dctn2 Ddt Ddx28 Ddx39 Ddx5 Dedd2 Dgkz Dhx33 Diras1 Disp2 Dmn Dnaja2 Dnajc11 Dnalc4 Dnm1 Dok4 Dscr1l2 Dst Dtna Dtx4 Dus3l Dusp15 Dusp26 Dusp6 Dync1h1 Dync1i2 E2f6 Ear11 Edd1 Edg2 Eef1d Eef2k EG209589 EG245190 EG381438 EG384885 EG433230 Mm.379159 Mm.313244 Mm.341593 Mm.271147 Mm.8569 Mm.260360 Mm.43212 Mm.157904 Mm.290353 Mm.194225 Mm.34256 Mm.159684 Mm.1258 Mm.200792 Mm.130362 Mm.259893 Mm.87051 Mm.121878 Mm.261259 Mm.273277 Mm.43871 Mm.422769 Mm.209265 Mm.3597 Mm.258404 Mm.30435 Mm.102136 Mm.35650 Mm.20445 Mm.15312 Mm.244512 Mm.389883 Mm.235007 Mm.132172 Mm.373672 Mm.29729 Mm.379227 Mm.227260 Mm.27670 Mm.68819 Mm.180052 68581 69195 270893 51875 72309 382014 233724 224807 67887 70397 71913 21917 21937 94185 21974 235559 252838 22003 215243 72169 66854 330863 14897 56191 72133 22084 110796 67125 22110 22117 66736 73032 319953 320244 22138 57776 73710 227613 52700 72565 56550 Tmed10 Tmem121 Tmem132e
Tmem141 Tmem158 Tmem16h Tmem41b Tmem63b Tmem66 Tmem70 Tmem79 Tmpo Tnfrsf1a Tnfrsf21 Top2b Topbp1 Tox Tpm1 Traf3ip3 Trim29 Trim35 Trim67 Trip12 Tro Trub1 Tsc2 Tshz1 Tspan31 Tspyl1 Tst Ttc35 Ttc9b Ttll1 Ttll5 Ttn Ttyh1 Tubb2b Tubb2c Txnl5 Uaca Ube2d2 99
Mm.321484 545864 EG545864 Mm.337238 Mm.389923 620480 EG620480 Mm.196580 Mm.312048 626920 EG626920 Mm.360108 Mm.422746 628781 EG628781 Mm.38802 Mm.389633 666405 EG666405 Mm.284811 Mm.391207 666538 EG666538 Mm.27404 Mm.362367 667290 EG667290 Mm.44858 Mm.285021 667682 EG667682 Mm.322453 Mm.294623 66235 Eif1ay Mm.346654 Mm.217616 27103 Eif2ak4 Mm.258320 Mm.34612 108067 Eif2b3 Mm.331784 Mm.250874 54709 Eif3s2 Mm.349358 Mm.227183 26987 Eif4e2 Mm.1838 Mm.259516 13688 Eif4ebp2 Mm.28420 Mm.27955 22384 Eif4h Mm.379457 Mm.3970 15572 Elavl4 Mm.268000 Mm.236645 68519 Eml1 Mm.292983 Mm.41423 71946 Endod1 Mm.19091 Mm.30145 83965 Enpp5 Mm.1574 Mm.297516 791299 ENSMUSG00000059659 Mm.2437 Mm.387663 100038425 ENSMUSG00000074414 Mm.5110 Mm.248620 71889 Epn3 Mm.332320 Mm.59812 67456 Ergic2 Mm.223504 Mm.392449 102058 Exoc8 Mm.374909 Mm.266635 50912 Exosc10 Mm.274948 Mm.334810 72544 Exosc6 Mm.255063 Mm.41739 58193 Extl2 Mm.348649 Mm.3355 14103 Fasl Mm.275039 Mm.236443 14104 Fasn Mm.25042 Mm.334764 50788 Fbxl8 Mm.4347 Mm.24829 76454 Fbxo31 Mm.221992 Mm.28584 71538 Fbxo9 Mm.212290 Mm.196475 50754 Fbxw7 Mm.23335 Mm.263041 207278 Fchsd2 Mm.59978 Mm.280819 14158 Fert2 Mm.30205 Mm.265716 14182 Fgfr1 Mm.3360 Mm.333499 67529 Fgfr1op2 Mm.279272 Mm.3126 14199 Fhl1 Mm.233082 Mm.23049 70598 Filip1 Mm.261606 Mm.38911 23877 Fiz1 Mm.220972 Mm.193099 14268 Fn1 Mm.38195 67921 22192 66589 268470 22248 67387 67031 22282 74270 170822 22288 113848 22320 22321 269523 22352 74199 107305 73178 57315 70465 24116 215280 330319 22404 71446 74254 72322 73192 22594 22601 77929 27377 241525 228994 22631 232879 214290 72881 21769 71591 Ube2f Ube2m Ube2v1 Ube2z Unc119 Unc50 Upf3a Usf2 Usp20 Usp33 Utrn V1ra6 Vamp8 Vars Vcp Vim Vit Vps37c Wasl Wdr46 Wdr77 Whsc2 Wipf1 Wipf3 Wiz Wrb Xab1 Xpo5 Xpot Xrcc1 Yap1 Yipf6 Yme1l1 Ypel4 Ythdf1 Ywhaz ZBTB45 Zcchc6 Zdhhc4 Zfand3 Zfp251 100
Mm.78250 Mm.24783 Mm.135965 Mm.89912 Mm.30357 Mm.276298 Mm.22742 Mm.246003 Mm.4793 Mm.32191 Mm.423024 Mm.24038 Mm.88367 Mm.41757 Mm.78118 Mm.276271 Mm.30102 Mm.36745 Mm.379511 Mm.321452 Mm.370185 Mm.196464 Mm.68889 Mm.17604 Mm.29114 Mm.9392 Mm.196269 Mm.297976 Mm.252391 Mm.290834 Mm.73666 Mm.271980 Mm.30195 Mm.1090 Mm.359573 Mm.27366 Mm.138434 Mm.41665 Mm.194811 Mm.218752 Mm.294664 17425 666060 327826 107971 14325 70300 11936 14362 14387 54393 14433 14468 14585 207182 70893 20340 216456 269037 381353 23885 14673 14678 14695 14697 14712 269682 57437 14733 14555 14766 229714 67298 209318 14775 625249 52857 235283 66168 54645 14843 606496 Foxk1 Frmpd1 Frs2 Frs3 Ftl1 Fuz Fxyd2 Fzd1 Gaa Gabbr1 Gapdh Gbp1 Gfra1 Ggtl3 Glb1l3 Glg1 Gls2 Gm672 Gm996 Gmcl1 Gna12 Gnai2 Gnb3 Gnb5 Gnpat Golga3 Golga7 Gpc1 Gpd1 Gpr56 Gpr61 Gprasp1 Gps1 Gpx1 Gpx4 Gramd1a Gramd1b Grina Gripap1 Gsh2 Gsk3a Mm.107441 Mm.276296 Mm.286296 Mm.26594 Mm.200818 Mm.296100 Mm.361832 Mm.294295 Mm.381553 Mm.389873 Mm.113388 Mm.121274 Mm.1144 Mm.23458 Mm.188422 Mm.100116 Mm.101524 Mm.117413 Mm.119584 Mm.124313 Mm.14644 Mm.196508 Mm.197450 Mm.210554 Mm.2115 Mm.22699 Mm.235103 Mm.235240 Mm.241121 Mm.242459 Mm.248254 Mm.249236 Mm.252497 Mm.255075 Mm.255909 Mm.256588 Mm.260474 Mm.266191 Mm.267521 Mm.275845 Mm.278432 22688 77652 668501 240442 69752 22710 619331 69020 72306 338354 231125 320158 52915 56364 99334 80292 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A Zfp26 Zfp422‐rs1 zfp507 Zfp508 Zfp511 Zfp52 Zfp551 Zfp707 Zfp777 Zfp780b Zfyve28 Zmat4 Zmiz2 Zmym3 Zscan29 Zxdc N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 101
Mm.2662 Mm.378931 Mm.261570 Mm.289645 Mm.275654 Mm.359030 Mm.715 Mm.279188 Mm.249391 Mm.29151 Mm.28326 Mm.3879 Mm.389331 Mm.386076 Mm.30921 Mm.4438 Mm.61526 Mm.289131 Mm.5 Mm.12863 Mm.157103 Mm.39739 Mm.209419 Mm.271798 Mm.27372 Mm.193212 Mm.277464 Mm.110 Mm.29590 Mm.14825 Mm.45558 Mm.275742 Mm.10137 Mm.27567 Mm.2644 Mm.272253 Mm.262547 Mm.292942 Mm.252421 Mm.271642 Mm.105218 14860 14873 14886 17263 14936 353502 15162 15193 207304 64209 52120 15251 15259 327655 110948 15361 208715 15360 15395 23908 101502 73442 15526 15547 59026 68180 57295 15903 170718 15929 70110 16001 16170 14204 216136 66541 28019 68510 70885 226432 16362 Gsta4 Gsto1 Gtf2i Gtl2 Gys1 Hcfc1r1 Hck Hdgfrp2 Hectd1 Herpud1 Hgsnat Hif1a Hipk3 Hisppd2a Hlcs Hmga1 Hmgcs1 Hmgcs2 Hoxa10 Hs2st1 Hsd3b7 Hspa12a Hspa9 Htf9c Huwe1 Hyi Icmt Id3 Idh3b Idh3g Ifi35 Igf1r Il16 Il4i1 Ilvbl Immp1l Ing4 Ints1 Ints10 Ipo9 Irf1 Mm.28552 Mm.288924 Mm.290645 Mm.291555 Mm.296610 Mm.306805 Mm.307906 Mm.310704 Mm.316213 Mm.32595 Mm.335551 Mm.335752 Mm.337558 Mm.339905 Mm.340968 Mm.34281 Mm.347359 Mm.347406 Mm.347546 Mm.347625 Mm.358634 Mm.363736 Mm.365513 Mm.366385 Mm.368153 Mm.379187 Mm.379620 Mm.379634 Mm.380260 Mm.380678 Mm.386284 Mm.387024 Mm.389147 Mm.389669 Mm.389685 Mm.389900 Mm.390074 Mm.390080 Mm.390517 Mm.390614 Mm.391069 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 102
Mm.274237 Mm.57035 Mm.28541 Mm.289657 Mm.307239 Mm.275071 Mm.1167 Mm.28382 Mm.388924 Mm.387390 Mm.374793 Mm.4970 Mm.10800 Mm.387016 Mm.435 Mm.8256 Mm.276408 Mm.41379 Mm.30355 Mm.4651 Mm.201493 Mm.26938 Mm.275036 Mm.251013 Mm.1249 Mm.20522 Mm.248843 Mm.276076 Mm.390821 Mm.26751 Mm.258142 Mm.22831 Mm.27961 Mm.298251 Mm.124176 Mm.390323 Mm.268018 Mm.44814 Mm.339061 Mm.320353 Mm.250479 272359 16400 16450 16451 77035 16476 16477 74187 16498 16500 16512 16515 16525 16526 66989 20218 16560 16564 16572 16579 16597 16598 23849 16211 226519 14768 73158 102436 71949 16828 57275 232798 56401 56839 16886 16886 66643 109593 432482 433955 544863 Irf2bp1 Itga3 Jag2 JAK1 Jmjd5 Jun Junb Katnb1 Kcnab2 Kcnb1 Kcnh3 Kcnj12 Kcnk1 Kcnk2 Kctd20 Khdrbs1 Kif1a Kif21a Kif5a Kifap3 Klf12 Klf2 Klf6 Kpnb1 Lamc1 Lancl1 Larp1 Lars2 Lass5 Ldha Lenep Leng8 Lepre1 Lgi1 Limk2 Limk2 Lix1 Lmo3 LOC432482 LOC433955 LOC544863 Mm.391165 Mm.391173 Mm.391585 Mm.391668 Mm.391693 Mm.391711 Mm.391724 Mm.391738 Mm.391817 Mm.391951 Mm.391976 Mm.391982 Mm.392128 Mm.392254 Mm.392311 Mm.392488 Mm.392664 Mm.393317 Mm.393451 Mm.393603 Mm.393958 Mm.394206 Mm.394298 Mm.394956 Mm.395008 Mm.396144 Mm.396370 Mm.396463 Mm.396475 Mm.396476 Mm.396495 Mm.396513 Mm.396865 Mm.397148 Mm.397518 Mm.397773 Mm.397965 Mm.398032 Mm.398239 Mm.399050 Mm.399286 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 103
Mm.375464 Mm.389395 Mm.386029 Mm.244815 Mm.416138 Mm.37324 Mm.391104 Mm.329675 Mm.289666 Mm.31247 Mm.259079 Mm.257067 Mm.341945 Mm.28204 Mm.36410 Mm.330745 Mm.241355 Mm.22575 Mm.264849 Mm.359472 Mm.18905 Mm.4219 Mm.14487 Mm.19223 Mm.154614 Mm.29815 Mm.392350 Mm.334413 Mm.1639 Mm.383472 Mm.297096 Mm.283045 Mm.272998 Mm.335639 Mm.139146 Mm.7386 Mm.390345 Mm.299693 Mm.272197 Mm.24570 Mm.4406 545798 625137 626802 628101 666761 668378 67527 225875 378937 67867 268490 23936 217779 226154 228355 16658 17136 80884 99470 270118 230815 17159 26399 212679 56527 232087 17196 17207 17210 51812 17448 66999 80889 17294 76781 17150 69900 17309 23945 102423 17395 LOC545798 LOC625137 LOC626802 LOC628101 LOC666761 LOC668378 LOC67527 Lrfn4 Lrrc24 Lrrc28 Lsm12 Lynx1 Lysmd1 Lzts2 Madd Mafb Mag Maged2 Magi3 Maml2 Man1c1 Man2b1 Map2k6 Mars2 Mast1 Mat2a Mbp Mcf2l Mcl1 Mcrs1 Mdh2 Med28 Mesdc1 Mest Mettl4 Mfap2 Mfsd11 Mgat3 Mgll Mizf Mmp9 Mm.399473 Mm.399793 Mm.400058 Mm.400215 Mm.400737 Mm.400749 Mm.401157 Mm.401233 Mm.401307 Mm.401465 Mm.401804 Mm.402278 Mm.402398 Mm.402779 Mm.402801 Mm.403190 Mm.403244 Mm.403442 Mm.404724 Mm.405468 Mm.405651 Mm.406833 Mm.407266 Mm.407322 Mm.407388 Mm.407439 Mm.407458 Mm.407867 Mm.408205 Mm.408935 Mm.409170 Mm.410179 Mm.410378 Mm.411239 Mm.411269 Mm.411320 Mm.411327 Mm.412067 Mm.412146 Mm.412348 Mm.413560 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 104
Mm.3759 Mm.41355 Mm.9986 Mm.183561 Mm.290166 Mm.29974 Mm.28023 Mm.423278 Mm.247007 Mm.228797 Mm.159028 Mm.2444 Mm.4048 Mm.234502 Mm.240965 Mm.89961 Mm.290407 Mm.292687 Mm.274610 Mm.205224 Mm.233080 Mm.248778 Mm.29867 Mm.29939 Mm.2033 Mm.379154 Mm.21669 Mm.288114 Mm.298283 Mm.1956 Mm.390700 Mm.42249 Mm.380307 Mm.31274 Mm.121508 Mm.233903 Mm.423051 Mm.28203 Mm.309520 Mm.6343 Mm.103477 17428 68929 17528 244238 67681 60441 56428 17772 219135 78388 74761 17869 17872 17913 230085 17948 53605 108123 269642 269198 228869 67273 17991 66108 66495 68194 226646 67602 380684 18039 18040 18007 269116 18027 245537 18111 57741 52530 217365 18148 230103 Mnt Mospd3 Mpz Mrgpre Mrpl18 Mrpl38 Mtch2 Mtm1 Mtmr6 Mvp Mxra8 Myc Myd116 Myo1c N28178 Naip2 Nap1l1 Napg Nat8l Nbeal1 Ncoa5 Ndufa10 Ndufa2 Ndufa9 Ndufb3 Ndufb4 Ndufs2 Necap1 Nefh Nefl Nefm Neo1 Nfasc Nfia Nlgn3 Nnat Noc2l Nola2 Nploc4 Npm1 Npr2 Mm.413759 Mm.413806 Mm.414028 Mm.415231 Mm.415245 Mm.416071 Mm.416262 Mm.416365 Mm.416557 Mm.417010 Mm.417021 Mm.417316 Mm.417884 Mm.418398 Mm.418499 Mm.418517 Mm.418770 Mm.418800 Mm.418930 Mm.419022 Mm.419476 Mm.420129 Mm.420897 Mm.421238 Mm.421284 Mm.421738 Mm.421880 Mm.422061 Mm.422171 Mm.422602 Mm.422605 Mm.422614 Mm.422678 Mm.422784 Mm.422791 Mm.422795 Mm.422829 Mm.422888 Mm.422896 Mm.422968 Mm.424048 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 105
Mm.5142 Mm.10099 Mm.3507 Mm.312068 Mm.329616 Mm.3304 Mm.325732 Mm.7271 18164 53324 18227 18189 18190 18197 26425 53319 Nptx1 Nptx2 Nr4a2 Nrxn1 Nrxn2 Nsg2 Nubp1 Nxf1 Mm.424338 Mm.424951 Mm.49752 Mm.5104 Mm.56337 Mm.56935 Mm.87456 Mm.8763 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 106
Table 3. Listing of 651 Affymetrix probes found to be highly upregulated in laser
captured DRG neurons compared to cerebellum (upregulation > 3-fold, n≥3, p-value
<0.01). Cross-references for Entrez Gene are also listed.
Probe Gene ID
Symbol
Upregulation P‐Value
96079_at 160240_at 94958_at 95119_at 103260_at 95737_at 98049_at 95518_at 100877_at 94983_at 97713_at 160916_at 160872_f_at 96320_at 95501_at 102056_f_at 104119_at 95927_f_at 104403_at 95440_at 103748_at 96614_at 97448_at 95559_at 161354_f_at 92971_at 160963_at 93045_at 93867_at 93626_at 103574_at 95064_at 160207_at 160171_f_at 97456_at 101578_f_at 71678 68552 68521 68778 68813 67808 74157 N/A 67705 67726 N/A 67484 66356 69683 N/A 67513 225358 70434 78832 67063 74440 217684 66776 103712 N/A 231440 269233 19299 19300 26357 226251 52538 104112 56360 433256 11461 0610010K06Rik 1110003E01Rik 1110013L07Rik 1110038D17Rik 1110060D06Rik 1200015A19Rik 1300018I05Rik 1810015C04Rik 1810058I24Rik 1810073G14Rik 2210010A19Rik 2310005P05Rik 2310008H09Rik 2310044H10Rik 2410001C21Rik 2610002J02Rik 2610024E20Rik 2610201A13Rik 2700078E11Rik 2810432L12Rik 4933407C03Rik 4933426M11Rik 4933439C20Rik///Pisd
6330403K07Rik 6330403K07Rik 9130213B05Rik 9630050M13Rik Abcd3 Abcd4 Abcg2 Ablim1 Acaa2 Acly Acot9 Acsl5 Actb 4.22 4.60 4.54 3.35 6.17 4.66 3.43 3.12 3.61 3.15 7.01 3.44 5.12 3.99 3.89 8.48 3.76 4.56 3.34 4.15 8.07 4.61 5.56 5.49 6.09 13.91 4.90 5.78 3.27 10.56 4.90 4.14 3.95 3.42 7.80 6.64 1.909E‐05 6.003E‐06 2.884E‐05 8.204E‐05 6.116E‐05 7.242E‐08 2.288E‐05 8.107E‐03 2.603E‐05 5.375E‐04 1.028E‐03 6.653E‐05 1.233E‐05 1.368E‐05 1.386E‐03 7.801E‐06 3.130E‐04 3.931E‐04 1.283E‐03 1.356E‐06 8.531E‐05 5.168E‐05 1.334E‐05 3.185E‐04 2.725E‐06 2.942E‐05 1.213E‐06 3.092E‐04 1.868E‐04 1.162E‐03 2.027E‐07 9.323E‐06 1.187E‐07 8.312E‐05 1.473E‐03 1.094E‐07 107
104578_f_at 95097_at 96738_at 92527_at 92200_at 104605_at 93512_f_at 92374_at 98999_at 94790_at 96025_g_at 160255_at 96858_at 95022_at 93464_at 104407_at 99559_at 161401_f_at 101058_at 94954_at 92210_at 100440_f_at 93054_at 104761_at 100569_at 93083_at 94304_at 104011_at 96529_at 161439_f_at 96185_at 97526_at 93330_at 104315_at 102677_at 92185_at 95136_at 95435_at 97211_at 95144_at 93288_at 109711 56444 11502 11515 11516 68465 11534 11539 11564 77559 269378 66395 26926 83397 100986 11658 11671 N/A 11722 52206 11601 11733 68839 71914 12306 11747 11749 11761 217030 N/A 11776 55946 11826 228359 14570 320982 65105 67166 67416 56443 76709 Actn1 Actr10 Adam9 Adcy9 Adcyap1 Adipor2 Adk Adora1 Adsl Agl Ahcy Ahnak Aifm1 Akap12 Akap9 Alcam Aldh3a2 Aldh3a2 Amy1 Anapc4 Angpt2 Ank1 Ankrd46 Antxr2 Anxa2 Anxa5 Anxa6 Aox1 Ap1gbp1 Ap1m1 Ap3d1 Ap3m1 Aqp1 Arhgap1 Arhgdig Arl4c Arl6ip4 Arl8b Armcx2 Arpc1a Arpc2 4.99 3.11 4.31 4.09 50.46 3.51 4.00 9.32 3.33 3.41 3.90 20.42 3.73 3.83 3.07 10.37 3.24 4.24 3.81 3.78 3.67 3.32 3.48 14.03 13.54 4.37 3.29 3.55 3.22 3.18 3.57 3.28 5.67 4.68 4.94 4.11 3.28 6.23 4.51 3.36 3.21 3.331E‐05 3.658E‐05 6.420E‐05 2.761E‐05 1.318E‐06 7.832E‐07 1.267E‐03 1.334E‐05 2.419E‐04 2.215E‐03 1.442E‐06 1.442E‐05 6.286E‐06 4.322E‐04 7.487E‐04 5.177E‐04 1.894E‐06 2.683E‐05 6.125E‐05 1.030E‐04 1.476E‐05 8.484E‐04 3.239E‐05 4.246E‐05 1.221E‐03 6.138E‐05 4.823E‐05 1.393E‐03 2.252E‐04 1.019E‐07 1.444E‐06 9.256E‐05 1.614E‐03 3.313E‐06 3.455E‐06 4.088E‐04 7.594E‐07 1.266E‐05 3.215E‐04 1.628E‐04 3.020E‐07 108
162481_f_at 161221_f_at 101997_at 103899_at 93798_at 99579_at 95897_at 95745_g_at 102249_at 103697_at 96572_at 95913_at 160137_at 102936_at 96167_at 161980_f_at 98509_at 95517_i_at 96158_at 94979_at 103497_at 97737_f_at 161214_r_at 95940_f_at 160876_at 102727_at 99668_at 103789_at 92526_f_at 92525_i_at 160585_at 160227_s_at 92208_at 161696_f_at 97897_at 98483_at 101040_at 102316_at 93499_at 98127_at 95142_s_at 109689 N/A 67526 50770 11928 11933 235574 11964 11567 106020 N/A 338351 53625 N/A 29810 N/A 211556 80748 232164 218333 234728 270151 N/A 223433 12033 12064 N/A 27965 67991 67991 67239 N/A 232560 97130 97863 12297 12334 12337 12340 12343 12345 Arrb1 Asns Atg12 Atp11a Atp1a1 Atp1b3 Atp2c1 Atp6v1a Avil AW061234 Azi2 B230333C21Rik B3gnt2 B4galt6 Bag3 Bag3 BC002199 BC004004 BC017133 BC018507 BC025546 BC034204 BC037034 BC052328 Bcap29 BDNF Bin1 Brd4 Btbd14a Btbd14a Bxdc1 Bysl C1qdc1 C77080 C78339 Cacnb3 Capn2 Capn5 Capza1 Capza2 Capzb 4.59 3.04 3.85 11.57 6.32 3.86 4.67 3.57 21.28 14.11 3.05 5.92 3.92 4.22 8.60 141.12 3.26 3.03 5.67 3.95 4.35 4.20 3.95 4.01 5.93 11.48 3.79 5.58 19.86 11.68 3.81 3.01 6.13 9.93 3.58 3.66 7.44 6.23 3.15 3.10 3.18 3.940E‐06 8.730E‐05 9.973E‐06 1.078E‐05 4.343E‐05 7.883E‐08 1.923E‐05 2.792E‐05 4.180E‐05 4.123E‐07 9.073E‐04 2.715E‐05 2.518E‐04 5.089E‐06 4.145E‐07 4.690E‐06 2.522E‐04 2.222E‐05 1.057E‐06 1.334E‐06 7.418E‐05 1.292E‐04 1.489E‐07 1.921E‐04 7.408E‐05 1.955E‐06 4.359E‐06 2.503E‐03 9.010E‐09 7.953E‐07 3.793E‐05 2.261E‐05 1.539E‐03 6.789E‐06 1.496E‐06 1.239E‐06 1.644E‐03 7.706E‐04 2.282E‐04 1.271E‐04 3.902E‐06 109
161329_f_at 95062_at 160479_at 93546_s_at 93547_at 99080_at 92252_at 160127_at 97930_f_at 97312_at 103422_at 100600_at 101516_at 160435_at 103251_at 100006_at 160623_at 93094_at 96524_at 104627_at 160326_at 104059_at 103088_at 94464_at 99672_at 101913_at 104029_at 96122_at 94241_at 104302_f_at 95666_at 93582_at 160680_at 104274_at 93151_at 93320_at 103334_at 101593_at 97262_at 161318_f_at 92227_s_at N/A 12380 12359 12400 12400 76551 12425 12450 12476 53599 12479 12484 12509 171486 71891 12552 53886 12585 237988 110911 321022 67772 12661 12725 12727 12728 12745 69574 71743 76501 108679 12850 12877 70568 12891 12894 12909 N/A 104318 12386 12386 Capzb Cast Cat Cbfb Cbfb Ccdc6 Cckar Ccng1 Cd151 Cd164 Cd1d1 Cd24a Cd59a Cd99l2 Cdadc1 Cdh11 Cdkl2 Cdr2 Cdr2l Cds2 Cdv3 Chd8 Chl1 Clcn3 Clcn4‐2 Clcn5 Clgn Cmbl Coasy Commd9 Cops8 Coq7 Cpeb1 Cpne3 Cpne6 Cpt1a Crcp Crip2 Csnk1d Ctnna2 Ctnna2 13.84 9.83 4.01 6.07 5.34 3.92 7.20 3.13 4.08 3.02 5.59 3.43 11.84 3.15 4.14 3.07 4.84 7.13 3.21 22.98 3.42 5.44 10.77 4.81 6.28 7.36 3.54 3.97 3.22 4.76 7.98 4.32 3.75 6.09 3.58 4.07 3.48 8.53 4.27 12.86 9.20 1.043E‐10 3.743E‐04 1.707E‐05 8.905E‐07 4.508E‐08 2.199E‐03 2.448E‐03 4.832E‐05 8.513E‐04 8.723E‐04 5.656E‐04 8.234E‐06 4.561E‐05 2.622E‐04 4.337E‐06 9.437E‐03 7.497E‐05 5.724E‐05 1.763E‐06 1.559E‐05 1.375E‐06 4.613E‐06 2.289E‐07 7.998E‐05 2.066E‐03 1.153E‐04 2.839E‐04 6.968E‐04 1.983E‐04 1.022E‐04 9.743E‐07 4.281E‐05 1.536E‐05 2.791E‐04 6.850E‐04 3.640E‐06 3.069E‐05 2.428E‐07 9.092E‐06 1.927E‐07 1.010E‐06 110
92228_at 95608_at 92256_at 101963_at 92633_at 103619_at 103922_f_at 102752_at 102009_at 103467_g_at 96526_at 95657_f_at 98013_at 95025_at 98356_at 97239_at 94967_at 96829_at 104116_at 97770_s_at 103430_at 95529_at 93534_at 94300_f_at 161139_f_at 100513_at 99096_at 96717_at 161060_i_at 95688_at 160866_at 95620_at 98451_at 94422_at 104738_at 102414_i_at 102415_r_at 103225_at 104306_at 104036_at 95584_at 12386 13030 13030 13039 64138 66427 72017 20430 N/A 54151 99324 66144 210998 27883 28001 225896 226090 226178 52331 27999 56320 13169 13179 69654 13196 13196 27225 67755 69663 13244 13347 66375 56812 235567 22791 100037258 100037258 69537 75221 83768 73703 Ctnna2 Ctsb Ctsb Ctsl Ctsz Cyb5b Cyb5r1 Cyfip1 Cyfip2 Cyhr1 D030029J20Rik D13Wsu177e D15Ertd621e D16H22S680E D16Wsu65e D19Ertd721e D19Wsu12e D19Wsu162e D5Ertd593e D6Wsu176e Dbn1 Dbnl Dcn Dctn2 Ddef1 Ddef1 Ddx24 Ddx47 Ddx51 Degs1 Dffa Dhrs7 Dnajb10 Dnajc13 Dnajc2 Dnajc3 Dnajc3 Dnase1l1 Dpp3 Dpp7 Dppa2 6.54 9.72 5.21 3.84 6.10 3.09 38.01 3.26 12.25 5.03 3.09 3.98 3.93 5.85 4.75 6.47 6.18 3.43 21.33 4.62 6.11 3.85 5.29 4.54 7.99 3.14 4.41 3.40 4.10 4.95 4.39 4.14 6.30 13.17 3.99 5.80 4.22 3.89 3.32 5.08 4.99 1.992E‐07 4.646E‐05 2.717E‐04 4.955E‐05 8.928E‐05 1.437E‐04 2.480E‐06 1.033E‐05 3.703E‐05 2.799E‐05 2.471E‐04 5.212E‐06 2.943E‐04 3.316E‐06 1.493E‐05 2.148E‐05 3.089E‐05 8.243E‐05 5.518E‐05 3.047E‐06 4.752E‐06 7.381E‐05 3.907E‐03 5.019E‐07 1.139E‐05 1.453E‐03 3.223E‐04 1.925E‐06 7.297E‐03 1.856E‐07 1.399E‐06 2.852E‐05 1.093E‐05 2.518E‐05 4.616E‐06 8.132E‐03 6.907E‐03 8.366E‐04 5.533E‐04 5.000E‐05 1.478E‐04 111
102374_at 162486_f_at 97333_at 93754_at 93058_at 98953_at 96822_at 103926_at 103891_i_at 93496_at 103665_at 101560_at 100472_at 93589_at 161119_at 101587_at 96628_at 96771_at 103531_f_at 98503_at 161808_f_at 100486_at 95474_at 161364_f_at 100928_at 93217_at 92587_at 100494_at 97509_f_at 103248_at 97593_f_at 97594_r_at 95095_at 103309_at 103699_i_at 103567_at 94833_at 101892_f_at 95756_at 94827_at 93011_at 53902 N/A 13424 51798 13664 68969 224045 208643 192657 68801 170439 N/A N/A 67464 13839 13849 107508 13867 N/A 14020 N/A 14055 N/A 14084 14115 268882 14148 14164 14182 14226 14248 14248 14251 56717 212398 72313 14314 14339 56095 11936 57436 Dscr1l2 Dscr3 Dync1h1 Ech1 Eif1a Eif1b Eif2b5 Eif4g1 Ell2 Elovl5 Elovl6 Emb Enah Entpd4 Epha5 Ephx1 Eprs Erbb3 Ero1lb Evi5 Evl Ezh1 F2r Faf1 Fbln2 Fbxo45 Fdx1 Fgf1 Fgfr1 Fkbp1b Fliih Fliih Flot1 Frap1 Frat2 Fryl Fstl1 Fts Ftsj3 Fxyd2 Gabarapl1 8.33 4.50 10.94 3.03 3.24 3.26 3.20 4.61 4.26 3.10 3.44 9.64 6.39 3.15 4.41 4.06 4.97 3.53 7.10 5.27 3.31 4.29 3.27 3.70 7.34 3.36 3.38 5.83 4.72 4.51 6.54 4.72 5.40 3.81 3.25 3.09 22.05 5.68 5.99 21.27 4.47 3.482E‐05 7.068E‐07 6.238E‐07 2.462E‐04 1.814E‐04 2.507E‐06 9.035E‐04 1.169E‐03 6.648E‐06 4.842E‐05 1.207E‐05 6.326E‐05 4.846E‐06 5.930E‐05 8.468E‐05 1.959E‐03 7.148E‐06 2.435E‐05 1.094E‐04 1.374E‐03 5.388E‐06 8.869E‐08 1.910E‐05 2.556E‐05 6.220E‐06 2.133E‐05 4.982E‐07 1.882E‐03 4.298E‐04 9.890E‐06 7.519E‐05 5.021E‐05 3.300E‐06 1.299E‐04 2.797E‐04 3.385E‐04 2.149E‐07 9.333E‐05 1.021E‐03 2.854E‐05 4.875E‐06 112
100066_at 94338_g_at 160335_at 102967_at 162049_f_at 102926_at 160530_at 160900_at 100457_at 100514_at 103843_at 102060_at 94876_f_at 100573_f_at 93132_at 103762_at 98544_at 162262_f_at 97541_f_at 97540_f_at 93120_f_at 96913_at 92580_at 97318_at 98599_at 94756_at 98820_g_at 98819_at 98111_at 93875_at 92571_at 97914_at 160139_at 103671_at 104102_at 96664_at 93752_at 97422_at 101518_at 93604_f_at 160933_at 14450 14453 14630 14545 N/A 14587 N/A 56278 20340 14674 14681 54214 70231 14751 14805 98053 14923 N/A 14964 14964 14972 N/A 15115 66044 15193 97114 15429 15429 15505 193740 15525 15526 80888 53415 64704 59026 105148 381314 55978 54725 16145 Gart Gas2 Gclm Gdap1 Gdpd5 Gfra3 Ghitm Gkap1 Glg1 Gna13 Gnao1 Golga4 Gorasp2 Gpi1 Grik1 Gtf2f1 Guk1 Gyg H2‐D1 H2‐D1 H2‐K1 Hadhb Hars Hars2 Hdgfrp2 Hist2h3c2 Hoxd1 Hoxd1 Hsp110 Hspa1a Hspa4 Hspa9 Hspb8 Htatip2 Htra2 Huwe1 Iars Iars2 Ift20 Igsf4a Igtp 6.68 4.72 9.40 3.89 3.56 7.69 3.55 4.48 4.86 3.65 3.08 3.31 3.14 7.79 10.51 3.73 3.39 8.03 31.12 3.29 3.31 3.26 18.96 3.88 3.07 4.02 19.55 3.37 3.51 4.85 4.95 8.15 7.19 12.98 4.73 4.26 3.10 5.37 13.80 5.08 3.75 2.882E‐05 7.893E‐05 2.517E‐05 7.705E‐06 1.725E‐05 7.209E‐04 3.184E‐06 4.411E‐05 9.941E‐06 2.053E‐04 1.992E‐07 2.371E‐04 3.045E‐08 1.630E‐07 5.078E‐03 1.555E‐04 8.657E‐05 1.011E‐06 1.658E‐05 1.113E‐07 3.820E‐08 7.706E‐06 1.160E‐05 3.590E‐07 1.506E‐04 2.661E‐04 2.343E‐03 1.194E‐03 1.611E‐06 8.100E‐06 6.062E‐04 4.634E‐05 2.042E‐05 5.486E‐05 1.648E‐05 4.075E‐06 4.639E‐05 7.839E‐07 6.405E‐07 1.685E‐06 1.014E‐05 113
160397_at 99491_at 94345_at 104425_at 97329_at 161444_f_at 95034_f_at 95035_at 104163_at 100123_f_at 93374_at 100130_at 160971_at 102892_at 162182_f_at 98829_at 99832_at 94060_at 100982_at 103998_at 99962_at 93635_at 98083_at 104645_at 93111_at 103656_at 94260_at 98912_at 103630_at 94278_at 93682_at 160832_at 93261_at 98909_at 104634_at 160270_at 98059_s_at 95028_r_at 96890_at 98892_at 104212_at 24010 16155 16195 12695 229543 N/A 75751 75751 320727 16412 57340 16476 107250 16498 N/A 16515 16533 N/A 105440 16558 16563 16570 23849 93691 16211 14768 N/A 217980 107045 18826 16826 16835 19141 N/A 110829 70361 16905 N/A 66887 14245 72416 Ik Il10rb IL‐6st Inadl Ints3 Ints5 Ipo4 Ipo4 Ipo8 Itgb1 Jph3 Jun Kazald1 Kcnab2 Kcnab2 Kcnj12 Kcnmb1 Kctd10 Kctd9 Kif16b Kif2a Kif3c Klf6 Klf7 Kpnb1 Lancl1 Larp1 Larp5 Lars Lcp1 Ldb2 Ldlr Lgmn Lias Lims1 Lman1 Lmna LOC329575 Lonp2 Lpin1 Lrpprc 3.35 5.13 6.29 4.47 3.92 3.06 6.04 3.34 3.32 3.27 3.42 3.23 3.43 5.98 9.38 3.56 7.49 4.03 8.08 3.25 3.26 3.88 3.06 9.98 3.03 3.23 3.30 5.05 3.83 3.71 5.21 15.13 8.18 3.08 4.76 3.15 11.09 3.06 3.18 11.83 5.17 5.695E‐04 3.248E‐07 1.010E‐05 1.690E‐04 1.554E‐07 3.446E‐06 2.994E‐06 1.215E‐03 2.655E‐04 1.268E‐07 1.519E‐08 5.450E‐06 4.516E‐05 2.842E‐04 6.712E‐09 9.211E‐03 1.108E‐03 3.159E‐03 1.400E‐06 8.189E‐04 6.577E‐03 7.070E‐06 1.660E‐05 3.257E‐05 6.378E‐05 5.632E‐05 6.686E‐06 8.987E‐07 4.419E‐05 1.446E‐03 1.053E‐04 1.528E‐07 5.271E‐05 9.377E‐06 5.736E‐07 1.767E‐06 1.462E‐05 2.926E‐04 1.153E‐04 5.327E‐07 1.017E‐04 114
95386_at 92847_s_at 95565_at 104677_at 160982_at 100470_at 161575_f_at 92350_at 162332_f_at 99604_at 99458_i_at 98475_at 96767_at 161951_f_at 97451_at 95405_at 104015_at 104033_at 95135_at 93342_at 101082_at 97296_at 102058_at 97251_at 102128_f_at 100033_at 160308_at 100046_at 102360_at 100929_at 103799_at 95632_f_at 103793_at 100915_at 102108_f_at 93427_at 92277_at 96496_g_at 96495_at 98531_g_at 103345_at 225010 17113 66591 227619 66724 26414 N/A 13589 N/A 69104 13728 17181 23943 N/A 193813 67943 75624 217615 68041 67949 17436 69163 78523 64657 N/A 17685 17698 17768 17769 210376 210376 17855 78388 17886 N/A 338367 71602 17933 17933 14455 20740 Lycat M6pr Mad2l1bp Man1b1 Map3k7ip3 Mapk10 Mapk10 Mapre1 Mapre3 Mar5 Mark2 Matn2 Mbc2 Mcart1 Mcfd2 Mesdc2 Metap1 Mgea6 Mid1ip1 Mki67ip Mod1 Mrpl44 Mrpl9 Mrps10 Mrps25 Msh2 Msn Mthfd2 Mthfr Mtmr9 Mtmr9 Mvk Mvp Myh9 Myh9 Myo1d Myo1e Myt1l Myt1l N/A N/A 8.65 3.29 5.14 4.92 4.13 3.62 8.00 3.76 3.68 3.95 3.22 3.67 5.27 3.20 4.80 3.06 4.04 4.39 4.13 5.57 6.64 3.15 4.29 3.43 4.65 3.08 3.97 5.01 3.38 8.20 3.81 4.09 3.10 6.59 3.91 3.18 3.22 3.97 3.92 3.04 3.15 4.514E‐05 2.829E‐06 1.196E‐05 1.882E‐05 1.991E‐05 8.847E‐06 6.608E‐08 3.311E‐04 4.750E‐05 8.326E‐06 1.533E‐04 7.178E‐04 4.971E‐06 1.501E‐05 1.329E‐04 2.030E‐04 2.002E‐03 1.482E‐06 2.760E‐07 4.448E‐05 9.450E‐06 1.751E‐05 2.598E‐05 2.719E‐07 3.217E‐06 9.482E‐04 1.442E‐05 2.293E‐06 6.056E‐05 2.890E‐07 1.603E‐03 6.424E‐05 3.302E‐06 2.866E‐06 3.876E‐05 8.495E‐06 7.534E‐05 8.091E‐06 2.550E‐05 6.525E‐05 7.273E‐05 115
97863_at 160450_at 102476_f_at 95681_f_at 94759_at 98464_at 100464_at 160393_at 102965_at 104092_at 100958_at 104215_at 92708_at 161075_at 96299_at 103010_at 161482_f_at AFFX‐DapX‐3_at X00686_M_at 161997_f_at X00686_5_at 97923_at AFFX‐ThrX‐3_at 97485_at 162225_f_at 160906_i_at 160419_r_at 102348_at 161147_f_at 162081_f_at 100458_at 93246_at 161750_f_at 99633_at 98884_r_at 96596_at 160464_s_at 103906_f_at 103234_at 92346_at 102827_at 52120 59003 65246 66849 67062 71452 73205 75320 98999 170707 219140 226641 234733 272636 347740 381598 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 17957 74838 N/A 26562 83431 17988 17988 83814 380684 18040 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A Napb Narg1 Ncam1 Ncdn Ndel1 Ndrg1///Ndrl Ndrl Nedd4l Nefh Nefm Nek7 3.33 3.32 5.64 5.14 4.15 3.62 3.43 7.61 4.29 3.60 3.89 3.41 3.09 5.51 3.32 4.32 30.74 21.13 17.31 8.66 7.79 6.53 6.20 4.94 3.95 3.92 3.74 3.62 3.12 3.01 6.10 5.06 3.33 6.24 3.92 10.48 8.28 3.10 7.94 4.53 11.38 5.760E‐05 4.194E‐06 5.755E‐07 3.027E‐08 4.376E‐05 3.361E‐06 3.768E‐03 5.552E‐07 1.319E‐05 8.075E‐08 9.498E‐06 4.162E‐04 3.239E‐05 2.001E‐03 7.146E‐08 4.179E‐06 1.770E‐07 1.382E‐04 1.548E‐03 4.333E‐06 6.843E‐03 8.208E‐07 4.234E‐03 5.229E‐05 2.048E‐07 4.433E‐03 1.629E‐04 1.984E‐04 3.611E‐05 6.670E‐06 1.856E‐04 1.247E‐03 9.773E‐04 1.728E‐07 6.043E‐07 7.596E‐06 8.533E‐07 2.273E‐06 2.166E‐06 7.079E‐04 2.370E‐06 116
102059_at 160991_at 103493_at 92794_f_at 102980_at 102047_at 96259_at 162037_f_at 160925_at 95016_at 92892_at 162166_f_at 94450_at 94452_g_at 97434_at 94839_at 98099_at 93046_at 92848_at 104444_at 100894_at 103311_at 102255_at 99158_at 95586_at 160187_at 95470_at 160949_at 93615_at 160610_at 97756_s_at 97398_at 94347_i_at 94348_f_at 95512_at 160264_s_at 97297_at 94694_at 100139_at 102296_at 95040_at 66257 69721 68039 18102 18107 18107 19155 18167 18176 18186 22360 22360 28114 28114 72181 18220 74167 18141 18242 N/A 26429 79196 18414 20409 18438 18475 68083 26430 18516 12936 12941 74551 18537 18537 18537 N/A N/A 18548 30052 18549 18571 Nicn1 Nkiras1 Nmb Nme1 Nmt1 Nmt1 Npepps Npy2r Nras Nrp1 Nrsn1 Nrsn1 Nsun2 Nsun2 Nsun4 Nucb1 Nudt9 Nup50 Oat Ocel1 Orc5l Osbpl5 Osmr Ostf1 P2rx4 Pafah1b2 Pak1ip1 Parg Pbx3 Pcdha4 Pcdha5 Pck2 Pcmt1 Pcmt1 Pcmt1 Pcp4l1 Pcp4l1 Pcsk1 Pcsk1n Pcsk2 Pdcd6ip 3.22 3.42 10.47 4.03 5.97 4.18 3.09 5.67 6.28 23.88 6.10 3.64 8.94 5.19 4.20 6.83 8.63 3.99 3.50 3.44 3.05 4.77 12.51 7.29 9.75 3.33 4.41 3.94 3.09 4.41 14.72 3.97 8.32 7.10 4.61 10.28 8.79 4.87 4.09 3.21 4.13 1.050E‐05 1.310E‐05 3.698E‐05 7.657E‐06 5.180E‐05 8.572E‐05 2.400E‐05 2.163E‐03 1.093E‐05 4.033E‐05 1.966E‐04 1.774E‐04 1.587E‐05 4.290E‐05 6.076E‐05 8.503E‐09 2.614E‐06 2.515E‐04 6.073E‐04 4.188E‐05 3.031E‐06 5.542E‐05 7.713E‐03 2.958E‐05 6.294E‐05 2.449E‐05 2.000E‐04 6.020E‐05 2.737E‐05 2.519E‐03 3.578E‐04 1.593E‐05 3.494E‐04 1.683E‐06 7.617E‐06 1.666E‐03 1.567E‐04 7.252E‐05 8.114E‐05 1.821E‐04 4.537E‐06 117
94932_at 94207_at 102063_at 96765_at 101042_f_at 97834_g_at 97833_at 93421_at 104313_at 94801_at 103959_at 103672_at 160105_r_at 92452_at 104208_at 95358_at 104337_f_at 104343_f_at 97967_at 93360_at 102395_at 103484_at 98508_s_at 95111_i_at 98580_at 101836_at 95680_at 160948_at 99503_at 160801_at 99366_at 94957_at 102299_at 93476_at 104751_at 93953_at 99930_s_at 95448_at 100225_f_at 93971_f_at 103362_at 18590 71853 18607 N/A 18624 56421 56421 18647 72157 70804 230936 78246 N/A 18706 N/A 18718 67451 66350 67784 29858 18858 78977 19012 66053 19042 19043 66849 19057 320717 N/A 217430 N/A 18750 19090 19132 19142 19165 19181 19182 66997 19219 Pdgfa Pdia6 Pdpk1 Peg3 Pepd Pfkp Pfkp Pftk1 Pgm2 Pgrmc2 Phf13 Phf23 Pigk Pik3ca Pik4ca Pip5k2a Pkp2 Pla2g12a Plxnd1 Pmm1 Pmp22 Popdc3 Ppap2a Ppil2 Ppm1a Ppm1b Ppp1r2 Ppp3cc Pptc7 Pqlc1 Pqlc3 Prepl Prkca Prkdc Prph1 Prss12 Psen2 Psmc2 Psmc3 Psmd12 Ptger4 4.68 4.70 5.02 5.56 6.59 23.27 6.11 5.43 3.81 6.46 4.19 3.58 3.42 4.44 4.46 4.33 5.04 11.41 4.12 3.84 14.24 6.83 4.43 3.33 6.38 4.67 3.66 4.39 12.81 3.98 24.80 4.59 3.22 4.99 56.42 3.50 8.17 5.48 3.34 3.58 10.16 4.971E‐05 6.998E‐06 7.441E‐05 1.635E‐06 1.880E‐06 1.082E‐04 3.714E‐04 1.872E‐04 1.996E‐04 4.253E‐05 2.507E‐06 2.791E‐05 9.079E‐05 1.191E‐05 6.635E‐06 8.603E‐06 3.610E‐03 8.822E‐06 1.161E‐05 3.436E‐06 9.263E‐05 2.315E‐05 3.087E‐04 1.159E‐04 2.348E‐08 2.300E‐05 1.508E‐04 2.039E‐04 5.287E‐07 3.038E‐04 6.201E‐05 1.833E‐07 1.405E‐04 1.613E‐05 3.126E‐06 2.066E‐04 2.282E‐04 1.179E‐05 1.882E‐05 1.531E‐05 1.069E‐05 118
94259_at 98811_at 103451_at 99599_s_at 100976_at 101932_at 96272_at 104422_at 97325_at 97489_at 95618_at 93347_at 97058_f_at 97415_at 93070_at 98602_at 93962_at 96733_at 99032_at 102028_at 96207_at 94972_at 98335_at 102859_at 103886_at 92671_f_at 161765_f_at 96747_at 92376_at 100509_at 161814_f_at 93829_at 93724_at 161616_f_at 98007_at 102875_at 92539_at 98600_at 92770_at 101467_at 160930_at 56351 19222 N/A 84113 56294 19267 N/A 19275 218699 110078 52055 19336 N/A 19340 70572 19387 109905 229877 19416 54354 56878 56878 19687 52250 13476 19731 67865 69581 19769 30945 30945 230257 26564 N/A 20112 320119 20194 20195 20200 20203 83493 Ptges3 Ptgir Ptk2b Ptov1 Ptpn9 Ptpre Ptprf Ptprn Pxk Pygb Rab11fip5 Rab24 Rab33b Rab3d Ranbp5 Rangap1 Rap1a Rap1gds1 Rasd1 Rassf5 Rbms1 Rbms1 Recc1 Reep1 Reep5 Rgl1 Rgs10 Rhou Rit1 Rnf19 Rnf19 Rod1 Ror2 Rprm Rps6ka2 Rps6kc1 S100a10 S100a11 S100a6 S100b Sacm1l 3.91 10.92 3.47 4.69 4.01 4.07 4.06 3.38 3.56 5.06 3.21 4.06 3.56 5.08 3.82 9.04 4.08 3.30 5.11 22.85 4.52 3.84 3.08 3.55 3.47 3.18 3.93 3.04 3.59 7.41 4.09 3.64 5.15 9.39 5.25 3.41 11.55 3.60 4.05 5.45 4.22 3.591E‐06 3.504E‐04 1.997E‐05 4.859E‐06 1.051E‐05 9.903E‐05 5.098E‐05 3.155E‐05 8.596E‐05 1.009E‐03 3.450E‐04 1.346E‐05 5.406E‐06 6.785E‐05 1.953E‐04 3.550E‐06 2.737E‐06 3.433E‐04 7.861E‐04 1.605E‐06 2.451E‐05 1.482E‐06 7.859E‐05 1.518E‐04 1.273E‐07 8.848E‐04 9.535E‐05 4.697E‐03 1.325E‐07 1.322E‐05 4.591E‐05 3.343E‐05 4.349E‐06 6.079E‐05 1.173E‐07 2.495E‐05 2.115E‐05 5.129E‐03 2.558E‐05 1.037E‐05 1.757E‐05 119
103080_at 104453_at 95752_at 104106_at 160388_at 94057_g_at 94056_at 98302_at 162255_s_at 93017_at 160291_at 160563_at 102860_at 104374_at 96060_at 93574_at 96677_at 104350_at 100024_at 92722_f_at 97957_at 103845_at 99445_at 93396_at 103818_at 161006_at 102983_at 102384_at 102062_at 161077_f_at 160630_at 160794_at 162269_at 104280_at 161859_f_at 97429_at 92839_f_at 102320_at 104651_at 97818_at 100538_at 56045 50724 66711 243272 66234 20249 20249 20264 20265 53378 53421 378702 20715 20716 20719 20317 N/A 20416 27428 20471 26569 20529 68682 20534 20540 N/A 17125 67155 20588 N/A 20603 226830 N/A 20618 N/A 20623 20639 55988 N/A 69150 20655 Samhd1 Sap30l Sbds Sbno1 Sc4mol Scd1 Scd1 Scn10a Scn1a Sdcbp Sec61a1 Serf2 Serpina3g Serpina3n Serpinb6a Serpinf1 Setd8 Shc1 Shroom3 Six1 Slc27a4 Slc31a1 Slc44a2 Slc4a1ap Slc7a7 Slco3a1 Smad1 Smarca2 Smarcc1 Smarcd2 Sms Smyd2 Smyd2 Sncg Sncg Snrk Snrpb2 Snx12 Snx14 Snx4 Sod1 3.29 6.27 3.16 6.97 3.68 3.95 3.76 38.76 3.38 13.98 3.46 3.81 3.15 3.78 24.96 3.97 3.18 6.18 20.87 5.54 3.18 3.62 5.94 3.37 13.99 7.38 3.50 3.18 6.51 5.22 3.44 3.26 4.55 19.25 19.18 3.57 4.39 3.65 3.44 4.13 4.14 6.180E‐08 3.370E‐06 9.337E‐05 5.267E‐04 1.557E‐04 6.027E‐03 7.372E‐05 1.483E‐07 1.851E‐03 1.874E‐05 1.794E‐05 3.683E‐05 3.575E‐04 6.309E‐05 1.726E‐06 3.753E‐05 7.463E‐05 5.465E‐05 4.740E‐08 2.859E‐04 4.348E‐05 2.693E‐04 1.683E‐05 5.302E‐07 1.738E‐05 9.140E‐05 7.868E‐05 4.280E‐04 2.341E‐04 1.755E‐06 4.987E‐06 8.792E‐06 2.808E‐06 1.273E‐06 2.105E‐08 2.217E‐05 1.253E‐05 1.312E‐04 4.232E‐05 2.505E‐06 3.234E‐05 120
96042_at 95430_f_at 161013_f_at 101995_at 101579_at 102780_at 103504_at 94432_at 94731_at 99100_at 94331_at 97131_at 103781_at 93501_f_at 103911_at 102726_at 92339_at 103326_at 98087_at 100690_at 99057_at 100953_at 96849_at 96275_f_at 103536_at 96605_at 94882_at 96340_at 103514_at 102850_at 96674_at 160809_at 95432_f_at 160249_at 103032_at 95010_at 160724_at 97277_at 100475_at 92270_at 92265_f_at 20656 27965 74646 18412 27058 76650 66970 20440 20840 20848 N/A 71728 20909 20916 N/A 21333 21339 272589 56480 21823 21838 21853 30058 107358 56363 66058 69981 N/A 94185 51789 320938 54473 28185 21985 22021 22031 56771 69076 217069 56191 20822 Sod2 Spg21 Spsb1 Sqstm1 Srp9 Srxn1 Ssbp2 St6gal1 Stac STAT3 STAT6 Stk11ip Stx4a Sucla2 Sumf1 Tac1 Taf1a Tbcel Tbk1 Th Thy1 Timeless Timm8a1 Tm9sf3 Tmeff2 Tmem176a Tmem30a Tmem50b Tnfrsf21 Tnk2 Tnpo3 Tollip Tomm70a Tpd52 Tpst1 Traf3 Trfp Triap1 Trim25 Tro Trove2 3.80 7.15 3.26 5.68 3.44 13.30 3.81 4.23 6.68 4.16 3.08 3.40 5.56 3.56 6.31 98.56 3.88 3.18 3.74 4.60 6.94 5.68 3.89 3.77 4.29 5.96 3.90 4.88 3.79 3.26 3.20 3.26 3.04 7.54 4.04 3.63 4.95 4.08 3.88 7.50 3.76 2.702E‐05 1.854E‐07 1.986E‐04 2.454E‐06 1.012E‐06 8.834E‐06 4.316E‐05 1.033E‐05 4.030E‐04 3.937E‐05 1.251E‐05 5.883E‐04 5.811E‐08 3.634E‐05 1.759E‐05 2.231E‐05 2.829E‐05 7.835E‐05 1.156E‐06 9.030E‐05 1.331E‐05 1.016E‐03 1.551E‐06 2.529E‐05 1.406E‐04 1.240E‐04 3.630E‐06 2.108E‐06 1.132E‐07 3.067E‐06 2.414E‐04 1.690E‐06 2.350E‐03 4.639E‐07 1.980E‐07 2.658E‐05 4.916E‐07 2.251E‐07 3.946E‐04 1.916E‐07 1.001E‐05 121
103494_at 160394_at 161612_f_at 95696_at 96113_at 96577_i_at 102812_i_at 102813_f_at 93509_at 162499_f_at 161396_f_at 160164_at 94865_at 97285_f_at 96623_at 94481_at 96165_at 96061_at 95482_at 98972_at 93191_at 104501_at 94963_at 100710_at 96002_at 103557_at 93337_at 103836_at 95138_at 160739_at 161270_i_at 103415_at 162094_f_at 103717_at 103406_at 99126_at 97102_at 98767_at 161084_at 92934_at 92262_at 216350 217449 N/A N/A 27366 74383 66663 66663 22210 56550 N/A 66589 79560 N/A 22234 216558 16589 N/A 252870 N/A 53330 56491 22330 269523 66700 107305 20479 22380 N/A 232341 N/A 78903 N/A 66894 74254 213742 27377 22632 234725 22751 22401 Tspan8 Ttc15 Tubb3 Txnl2 Txnl4 Ubap2l Ube1dc1 Ube1dc1 Ube2b Ube2d2 Ube2o Ube2v1 Ublcp1 Ubxd1 Ugcg Ugp2 Uhmk1 Usp14 Usp7 Usp8 Vamp4 Vapb Vcl Vcp Vps24 Vps37c Vps4b Wbp4 Wipi2 Wnk1 Wnk1 Wrnip1 Wtap Wwp2 Xab1 Xist Yme1l1 Yy1 Zfp612 Zfp90 Zmat3 3.40 5.31 3.26 3.12 3.32 4.31 4.44 3.24 4.77 3.75 5.20 3.18 3.21 3.74 3.52 3.49 3.89 3.64 3.64 3.79 3.22 4.22 3.45 4.00 3.07 4.16 4.15 3.13 4.07 6.17 36.29 4.52 3.41 9.33 6.08 7.44 4.39 5.32 9.29 3.67 3.18 3.637E‐04 5.360E‐06 4.847E‐07 5.278E‐07 4.939E‐06 3.042E‐05 7.581E‐07 7.178E‐06 3.805E‐08 1.052E‐05 3.865E‐04 1.135E‐05 2.696E‐03 1.977E‐04 1.109E‐03 3.627E‐05 5.582E‐07 1.957E‐06 6.657E‐06 3.031E‐07 5.120E‐04 2.529E‐05 1.760E‐03 1.525E‐07 5.394E‐08 8.385E‐06 5.487E‐04 1.673E‐03 8.620E‐03 1.044E‐03 2.810E‐04 9.129E‐07 4.637E‐04 5.978E‐05 1.004E‐04 7.383E‐03 1.783E‐04 2.351E‐07 6.933E‐05 2.339E‐04 6.261E‐07 122
Table 4. Listing of 414 Affymetrix probes found to be highly upregulated in cerebellar
tissue compared to laser captured DRG neurons (upregulation > 3-fold, n≥3, p-value
<0.01). Cross-references for Entrez Gene are also listed.
Probe 98934_at 95634_at 94908_r_at 97874_at 95406_at 102234_at 97512_at 97864_at 97865_g_at 104229_at 162249_f_at 96016_at 100116_at 96658_at 95707_at 160905_s_at 95288_i_at 98064_at 95705_s_at 93560_at 100751_at 103429_i_at 101887_at 104179_at 160546_at 103688_at 96132_at 102704_at 102703_s_at 98491_at 92428_at 100984_at 93596_i_at 93014_at 99128_at 98104_at Gene ID Symbol
N/A 104457 66117 78330 67704 67704 70257 67922 67922 67149 N/A 72657 68026 72931 67267 80515 108686 227290 11461 66204 11487 104923 11606 104877 11676 237761 494504 11829 11829 67166 56495 11908 67126 27425 28080 114143 0610007P06Rik 0610010K14Rik 1110001J03Rik 1500032D16Rik 1810037I17Rik 1810037I17Rik 2010107E04Rik 2510049I19Rik 2510049I19Rik 2610200G18Rik 2610200G18Rik 2700094K13Rik 2810417H13Rik 2900010J23Rik 2900010M23Rik A030009H04Rik A430106J12Rik Aamp Actb Acyp1 Adam10 Adi1 Agt AI788669 Aldoc Ankrd43 Apcdd1 Aqp4 Aqp4 Arl8b Asna1 Atf1 Atp5e Atp5l Atp5o Atp6v0b Upregulation P‐Value
4.42 3.36 5.44 5.77 3.35 3.05 3.68 3.36 3.05 7.54 3.88 5.03 3.77 3.91 3.25 4.44 3.65 3.12 3.62 4.78 3.05 3.34 3.69 4.73 4.10 3.17 4.68 24.72 8.04 3.87 3.63 3.02 5.80 16.91 3.42 9.80 8.164E‐08 7.949E‐06 4.855E‐06 1.368E‐06 2.526E‐05 5.617E‐08 1.018E‐06 4.925E‐07 6.408E‐06 9.964E‐06 1.254E‐05 8.648E‐06 7.719E‐05 1.814E‐06 1.097E‐05 7.578E‐05 7.474E‐04 5.213E‐07 1.563E‐03 1.573E‐06 4.928E‐05 2.000E‐05 1.178E‐05 2.599E‐08 3.351E‐04 1.180E‐05 7.956E‐06 6.162E‐12 9.983E‐11 1.805E‐04 9.844E‐07 7.256E‐07 8.421E‐07 3.439E‐06 1.316E‐06 4.323E‐06 123
92597_s_at 99127_at 101105_at 93536_at 160366_at 94802_at 93021_at 95393_at 93057_at 93104_at 96305_at 94019_at 160973_at 104460_at 98133_at 102773_at 98436_s_at 92932_at 160486_at 160159_at 98478_at 93909_f_at 98446_s_at 160493_at 101017_at 95471_at 99119_at 97841_at 97110_at 93284_at 101973_at 97468_at 97527_at 100044_at 104516_at 94256_at 101564_at 93119_at 93820_at 99661_r_at 161081_at 11966 54138 23825 12028 407819 244654 406217 228662 218490 12226 231889 66882 224171 12291 12307 12319 12367 12404 N/A 268697 12452 111662 13846 12512 12567 12577 12631 68953 54371 12696 17684 54124 66197 18417 12741 29876 18983 12859 12866 12867 231207 Atp6v1b2 Atxn10 Banf1 Bax BC031181 BC060632 Bex4 Btbd3 Btf3 Btg1 Bud31 Bzw1 C330027C09Rik Cacna1g Calb1 Car8 Casp3 Cbln1 Ccdc28b Ccnb1 Ccng2 Ccrn4l Cct3 Cd63 Cdk4 Cdkn1c Cfl1 Chmp2a Chst2 Cirbp Cited2 Cks1b Cks2 Cldn11 Cldn5 Clic4 Cnot7 Cox5b Cox7a2 Cox7c Cpeb2 4.08 5.88 3.80 3.93 6.77 6.70 3.67 3.38 15.52 4.28 3.11 4.51 3.36 3.33 13.19 13.00 3.46 6.20 4.61 4.53 5.68 4.29 5.98 5.42 3.62 4.82 4.24 3.43 5.68 3.05 4.18 3.63 6.34 13.86 5.26 4.01 3.59 4.11 12.06 3.81 4.40 1.956E‐05 1.867E‐06 3.251E‐05 9.348E‐05 3.024E‐05 1.413E‐08 4.689E‐06 2.388E‐03 1.574E‐06 4.199E‐06 9.121E‐07 1.484E‐06 5.483E‐06 2.301E‐04 1.577E‐06 5.099E‐05 6.048E‐07 1.927E‐07 1.969E‐05 2.457E‐05 9.580E‐10 1.238E‐04 4.270E‐06 1.484E‐07 4.724E‐06 1.752E‐08 6.624E‐06 5.409E‐05 2.494E‐05 7.407E‐06 3.518E‐06 3.012E‐06 2.339E‐05 1.309E‐05 9.168E‐09 4.063E‐06 1.481E‐07 1.340E‐05 2.424E‐05 7.018E‐07 6.717E‐04 124
93550_at 160979_at 97255_at 160511_at 160522_at 97248_at 102307_at 92887_at 93309_at 103842_at 93493_at 160286_at 97124_at 160449_at 160857_at 101781_f_at 99191_at 94253_at 103674_f_at 100557_g_at 94001_at 94393_r_at 160531_at 98338_at 97317_at 98958_at 160379_at 103222_at 94040_at 98967_at 99546_at 98441_at 93203_f_at 100133_at 101294_g_at 92938_at 92939_at 92940_s_at 103061_at 94813_at 160860_at 13008 13017 N/A 20315 27528 13167 13193 51793 13205 26900 13207 110052 N/A 13486 13642 664960 58521 13665 26908 75705 15568 54326 N/A 13799 18606 223527 269587 13860 13877 12140 14227 14265 14236 14360 14380 14394 14399 14399 14415 14451 14560 Csrp2 Ctbp2///Zranb1 Cugbp2 Cxcl12 D0H4S114 Dbi Dcx Ddah2 Ddx3x Ddx3y Ddx5 Dek Dfna5h Dr1 Efnb2 EG238836 Eid1 Eif2s1 Eif2s3y Eif4b Elavl1 Elovl2 Emg1 En2 Enpp2 Eny2 Epb4.1 Eps8 Erh Fabp7 Fkbp2 Fmr1 Foxn2 Fyn G6pd2///G6pdx Gabra1 Gabra6 Gabra6 Gad1 Gas1 Gdf10 3.87 18.69 12.33 5.17 15.37 22.03 10.19 3.88 3.86 5.24 3.69 4.44 9.86 4.51 3.63 7.44 27.63 4.91 4.35 3.60 3.67 3.89 3.65 25.78 8.07 3.34 3.54 7.48 24.28 3.92 4.75 7.76 3.18 3.71 3.02 5.42 38.24 20.55 34.92 3.76 9.89 2.125E‐06 1.101E‐06 1.395E‐07 9.395E‐06 5.377E‐07 3.543E‐05 6.357E‐06 1.312E‐07 1.401E‐05 9.975E‐06 6.410E‐05 3.768E‐07 7.340E‐05 7.289E‐07 3.708E‐06 8.572E‐08 1.885E‐09 9.653E‐07 3.036E‐06 5.984E‐08 1.688E‐05 6.796E‐06 5.179E‐06 9.061E‐08 4.590E‐03 2.324E‐06 1.152E‐05 2.901E‐04 1.901E‐06 1.095E‐07 6.303E‐06 2.186E‐05 1.079E‐04 8.294E‐06 1.033E‐07 1.793E‐03 1.517E‐04 1.683E‐03 2.534E‐06 2.431E‐06 1.008E‐05 125
94144_g_at 100065_r_at 94391_at 104449_at 100465_i_at 94854_g_at 102305_at 94206_at 92946_f_at 98293_g_at 93954_at 93019_at 101954_at 100380_at 100708_at 94781_at 162457_f_at 101869_s_at 103534_at 98917_at 98629_f_at 99581_at 94805_f_at 92484_at 99335_at 93095_at 93250_r_at 96699_at 101589_at 93117_at 101524_at 94303_at 96084_at 94832_at 102410_at 93013_at 98623_g_at 93823_at 96068_at 97859_at 100990_g_at 14580 14609 14623 14658 67320 14688 171469 N/A 14800 19051 54195 15270 51788 15078 15081 15122 N/A 15129 15130 66631 15251 15254 319172 15273 15275 15289 97165 15312 15331 53379 N/A N/A 50926 N/A 15476 15902 16002 67781 66541 212111 16413 GFAP Gja1 Gjb6 Glrb Gm1673 Gnb1 Gpr37l1 Grcc10 Gria2 Gsbs Gucy1b3 H2afx H2afz H3f3a H3f3b Hba‐a1 Hba‐a1 Hbb‐b1///Hbb‐b2 Hbb‐b2 Hiatl1 Hif1a Hint1 Hist1h2ab Hivep2 Hk1 Hmgb1 Hmgb2 Hmgn1 Hmgn2 Hnrpa2b1 Hnrpa3 Hnrpd Hnrpdl Hnrph2 Hs3st1 Id2 Igf2 Ilf2 Immp1l Inpp5a Itgb1bp1 3.62 4.52 4.00 3.53 3.22 6.30 3.34 4.28 3.85 3.79 9.42 3.54 4.28 3.99 8.59 24.86 5.56 30.84 33.88 3.65 3.51 7.32 12.94 5.92 3.55 5.84 7.84 4.01 7.24 3.61 3.93 6.29 4.37 3.86 3.32 7.03 5.45 6.08 3.90 7.40 5.10 1.159E‐04 3.455E‐05 2.695E‐04 4.046E‐05 6.017E‐06 1.592E‐06 1.229E‐04 3.203E‐06 1.623E‐05 3.817E‐05 2.289E‐04 1.201E‐06 4.407E‐07 1.371E‐05 8.516E‐08 1.958E‐07 5.885E‐06 8.818E‐08 5.717E‐08 2.665E‐04 1.321E‐06 1.482E‐06 5.924E‐06 3.931E‐05 8.574E‐06 4.105E‐07 9.240E‐05 4.944E‐08 1.566E‐05 6.942E‐07 2.282E‐05 1.505E‐06 1.348E‐06 1.208E‐05 2.298E‐05 2.215E‐06 3.080E‐06 5.601E‐07 4.723E‐05 2.608E‐05 8.419E‐05 126
93511_at 94977_at 93895_s_at 102364_at 99339_r_at 98364_at 102335_at 104735_at 160417_at 94321_at 104078_g_at 103023_at 103370_at 97907_at 101946_at 102405_at 160550_i_at 103021_r_at 103020_s_at 103416_at 160103_at 96865_at 97203_at 96011_at 96013_r_at 99095_at 96311_at 94352_at 160561_at 95417_at 96258_at 101453_at 99457_at 95726_at 100536_at 99046_at 99047_at 99048_g_at 103987_at 98876_at 160345_at 16431 16438 16438 16478 16508 16508 16525 239217 16573 16661 66192 16816 22343 66094 18777 17136 17149 26401 26401 50772 57438 17118 17357 17184 N/A 17187 17196 103537 17242 217664 66447 12587 17345 30853 17433 17433 17433 17433 17441 66419 94065 Itm2a Itpr1 Itpr1 Jund1 Kcnd2 Kcnd2 Kcnk1 Kctd12 Kif5b Krt10 Lage3 Lcat Lin7c Lsm7 Lypla1 Mag Magoh Map3k1 Map3k1 Mapk6 Mar7 Marcks Marcksl1 Matr3 Matr3 Max Mbp Mbtd1 Mdk Mgat2 Mgst3 Mia1 Mki67 Mlf2 Mobp Mobp Mobp Mobp Mog Mrpl11 Mrpl34 10.72 11.23 7.02 3.06 21.56 4.45 3.70 33.11 4.03 4.86 3.29 4.27 6.45 4.00 4.58 3.75 3.02 6.72 5.22 4.67 3.35 19.31 19.54 9.28 5.50 3.73 5.25 3.04 3.48 5.11 4.58 7.79 3.46 3.40 13.76 10.54 7.81 5.79 12.31 3.24 3.37 5.059E‐07 9.034E‐04 1.486E‐04 1.697E‐06 3.323E‐04 8.163E‐03 1.630E‐05 5.731E‐06 1.829E‐05 1.614E‐07 3.347E‐05 6.609E‐05 1.491E‐05 7.641E‐08 1.830E‐04 2.791E‐07 5.295E‐04 1.271E‐05 7.574E‐07 3.815E‐09 3.438E‐04 1.763E‐08 8.961E‐05 1.843E‐04 4.699E‐04 8.138E‐05 1.015E‐05 3.976E‐05 8.077E‐06 3.792E‐08 1.248E‐05 1.831E‐05 1.572E‐04 7.221E‐09 2.916E‐05 1.001E‐04 7.877E‐05 5.493E‐05 3.031E‐05 3.441E‐06 2.041E‐05 127
94912_at 93573_at 99024_at 92644_s_at 160582_at 160626_at 98571_s_at 94910_at 95652_at 96909_at 101525_at 95132_r_at 92615_at 93581_at 99593_at 93519_s_at 92717_at 92893_at 160859_s_at 99440_at 101930_at 92747_at 92625_at 97520_s_at 97250_at 93831_at 99076_at 102715_at 98631_g_at 96885_at 160395_at 161000_i_at 101013_at 92271_at 102376_r_at 162452_at 160899_at 94208_at 94209_g_at 160123_at 161763_r_at 66292 17748 17122 17863 17868 17876 17938 67203 N/A 70316 68342 68198 230075 67264 595136 18002 18012 18027 18028 18028 18032 18088 18103 18111 66181 53610 353187 13865 18194 70021 N/A 108907 18245 18508 18545 N/A 18546 71853 71853 75454 117150 Mrps21 Mt1 Mxd4 Myb Mybpc3 Myef2 Naca Nde1 Ndufa7 Ndufab1 Ndufb10 Ndufb2 Ndufb6 Ndufb8 Ndufs5 Nedd8 Neurod1 Nfia Nfib Nfib Nfix Nkx2‐2 Nme2 Nnat Nola3 Nono Nr1d2 Nr2f1 Nsdhl Nt5dc2 Nudcd2 Nusap1 Oaz1 Pax6 Pcp2 Pcp2 Pcp4///Jam4 Pdia6 Pdia6 Phpt1 Pip5k2c 6.66 4.55 3.46 3.33 4.43 4.34 10.10 3.24 3.62 6.36 4.16 4.61 14.54 9.73 4.50 4.50 20.53 3.65 4.22 3.34 3.47 3.14 3.19 16.90 3.52 4.62 4.73 3.50 3.77 7.98 3.60 3.90 3.07 6.37 4.22 3.06 5.12 4.84 4.26 4.89 3.12 2.172E‐06 2.050E‐04 3.353E‐06 4.485E‐04 7.424E‐05 4.045E‐05 4.354E‐07 2.564E‐05 1.946E‐05 7.008E‐06 3.572E‐06 2.212E‐07 2.803E‐05 2.087E‐06 2.644E‐05 6.927E‐08 1.005E‐08 4.145E‐06 2.885E‐03 9.123E‐04 1.855E‐08 1.968E‐05 1.305E‐06 3.677E‐07 6.108E‐07 5.040E‐04 2.705E‐04 6.147E‐07 1.788E‐07 5.343E‐06 4.018E‐06 2.694E‐05 1.727E‐06 1.121E‐05 6.322E‐07 1.881E‐05 7.092E‐06 6.371E‐05 2.187E‐06 6.241E‐07 1.364E‐05 128
104557_at 102696_s_at 98004_at 96187_at 92801_at 92802_s_at 100927_at 95059_at 93325_at 104279_at 100089_at 100088_at 92691_at 100332_s_at 96696_at 160750_at 93988_at 94263_f_at 94088_at 92546_r_at 92545_f_at 102105_f_at 97474_r_at 94489_at 92379_f_at 92378_at 96719_i_at 95096_at 160726_at 95785_s_at 104680_at 93319_at 93281_at 96591_at 101030_at 97164_at 96653_at 104432_at 101966_s_at 100979_at 93164_at 56305 56305 18767 227937 18823 18823 18830 52830 66420 69833 19038 19046 72930 11758 N/A 212627 26444 19177 56195 19215 19215 N/A 19242 19243 19283 19283 19293 19317 19317 19350 51801 19414 26611 19699 11852 67153 100037283 11858 24017 56515 19821 Pitpnb Pitpnb Pkia Pkp4 Plp1 Plp1 Pltp Pnrc2 Polr2e Polr2f Ppic Ppp1cb Ppp2r2b Prdx6///Prdx6‐rs1 Prmt1 Prpsap2 Psma7 Psmb7 Ptbp2 Ptgds Ptgds Ptgds Ptn Ptp4a1 Ptprz1 Ptprz1 Pvalb Qk Qk Rab7 Ramp1 Rasa3 Rcn2 Reln Rhob Rnaseh2b Rnaset2 Rnd2 Rnf13 Rnf138 Rnf2 3.55 3.40 5.42 4.90 14.20 11.13 4.34 5.10 6.98 3.06 3.41 9.28 3.73 6.26 4.47 3.30 3.00 3.62 5.42 10.69 10.61 6.67 5.57 9.77 10.39 3.02 3.17 17.39 10.29 7.89 4.77 3.96 4.66 15.82 3.08 3.78 4.94 7.68 4.42 7.38 3.26 7.120E‐06 1.683E‐04 1.362E‐07 1.637E‐06 4.394E‐09 5.076E‐06 8.688E‐06 7.598E‐09 2.433E‐06 1.036E‐05 1.102E‐04 1.611E‐05 2.695E‐07 7.956E‐10 1.744E‐08 6.289E‐06 1.393E‐05 9.475E‐07 3.290E‐08 9.125E‐06 1.995E‐07 3.808E‐06 8.938E‐05 6.624E‐09 9.403E‐06 1.586E‐05 1.287E‐04 7.863E‐08 2.576E‐05 4.650E‐05 4.384E‐05 2.289E‐06 9.261E‐06 3.099E‐04 1.017E‐05 2.152E‐06 8.555E‐05 6.620E‐06 6.123E‐05 5.269E‐06 1.076E‐07 129
101889_s_at 98342_at 100711_at 102109_at 96290_f_at 93987_f_at 161127_i_at 101680_at 100727_at 92628_at 92577_f_at 100213_f_at 96575_at 102126_at 93730_at 97647_at 96358_at 98564_f_at 96300_f_at 93030_at 98085_f_at 100758_at 101664_at 100780_at 101212_at 102001_at 94181_at 103908_at 93548_at 92636_f_at 93411_at 96249_at 98905_at 94817_at 103543_at 99621_s_at 160539_at 104586_at 96609_at 95791_s_at 97890_at 19883 Rora 110954 Rpl10 19896 Rpl10a 270106 Rpl13 19933 Rpl21 100042049 Rpl22l1 68193 Rpl24 26451 Rpl27a///EG432798///LOC545487
19943 Rpl28 54217 Rpl36 67281 Rpl37 67945 Rpl41 26961 Rpl8 20042 Rps12 N/A Rps15a 20055 Rps16 66475 Rps23 27370 Rps26 N/A Rps27 N/A Rps27a 54127 Rps28 N/A Rps28 20091 Rps3a 20102 Rps4x 20115 Rps7 20135 Rrm2 20284 Scrg1 N/A Sdccag3 66212 Sec61b 20335 Sec61g 20361 Sema7a 93684 Sep15 235072 Sep7 12406 Serpinh1 56747 Sez6l 71514 Sfpq 110809 Sfrs1 110809 Sfrs1 69207 Sfrs11 20382 Sfrs2 20393 Sgk 4.42 3.66 8.17 4.23 6.23 4.26 5.15 14.29 8.89 13.37 3.35 3.29 3.42 4.20 15.16 18.95 20.11 3.71 7.36 6.51 4.32 18.61 6.94 3.50 25.52 3.15 4.25 3.06 6.56 13.42 3.27 20.13 7.75 4.02 4.52 6.65 9.83 5.21 4.05 6.42 3.53 2.811E‐05 7.898E‐08 8.087E‐08 7.510E‐07 1.968E‐07 2.675E‐05 6.357E‐08 4.414E‐09 1.878E‐07 3.801E‐08 2.715E‐06 1.410E‐06 4.163E‐07 9.766E‐07 2.666E‐08 2.069E‐06 2.435E‐08 7.501E‐09 2.635E‐08 1.554E‐06 3.205E‐07 4.874E‐06 1.376E‐08 2.515E‐07 1.530E‐07 1.703E‐08 3.255E‐07 2.720E‐07 1.810E‐07 4.381E‐06 1.385E‐04 1.971E‐06 9.683E‐06 1.376E‐07 4.099E‐09 9.225E‐07 3.657E‐05 3.416E‐07 1.973E‐05 1.888E‐05 7.009E‐05 130
93806_at 92673_at 160236_at 99845_at 100618_f_at 101420_at 161059_at 97421_at 93273_at 95049_at 93999_at 96333_g_at 93669_f_at 101631_at 100009_r_at 101430_at 96192_at 160319_at 97722_at 104249_g_at 104595_at 100878_at 92648_at 95796_g_at 95795_at 93005_at 93918_at 101528_at 101529_g_at 160402_at 98981_s_at 101959_r_at 97477_at 98920_g_at 93818_g_at 160183_f_at 94903_at 160472_r_at 94508_at 102870_at 98129_at 56726 20404 N/A 20513 11740 22348 232333 14211 20617 107686 68011 67804 20666 20666 20674 20677 20687 13602 107513 67437 20843 94186 20912 100041294 100041294 20979 108143 21399 21399 N/A 21406 21781 30057 69742 56334 66676 66676 230157 66271 N/A N/A Sh3bgrl Sh3gl2 Slain1 Slc1a6 Slc25a5 Slc32a1 Slc6a1 Smc2 Snca Snrpd2 Snrpg Snx2 Sox11 Sox11 Sox2 Sox4 Sp3 Sparcl1 Ssr1 Ssr3 Stag2 Strn3 Stxbp3a Supt4h1///Supt4h2 Supt4h2 Syt1 Taf9 Tcea1 Tcea1 Tceb2 Tcf12 Tfdp1 Timm8b Tm2d2 Tmed2 Tmed7 Tmed7 Tmeff1 Tmem126a Tmem181 Tmsb10 5.84 3.78 5.73 5.71 4.99 13.71 6.78 3.13 3.26 7.70 4.27 3.09 8.26 3.04 11.86 3.45 17.45 27.21 3.11 5.44 3.40 5.69 3.12 3.67 6.84 9.82 10.63 5.24 3.73 18.83 6.68 4.87 3.73 3.50 6.64 3.97 3.54 3.06 4.16 25.76 11.69 2.337E‐07 6.480E‐06 3.636E‐05 7.530E‐05 3.856E‐09 3.682E‐06 5.023E‐05 5.820E‐05 7.258E‐06 5.821E‐06 1.192E‐06 8.003E‐08 8.405E‐07 4.698E‐05 6.859E‐10 3.445E‐05 1.129E‐07 1.158E‐07 6.245E‐08 3.163E‐08 2.596E‐05 3.348E‐06 2.526E‐04 1.220E‐03 2.280E‐07 2.633E‐03 3.528E‐07 4.919E‐08 5.852E‐09 6.934E‐06 7.397E‐07 6.576E‐07 3.277E‐05 3.023E‐05 3.019E‐09 1.027E‐05 3.382E‐07 1.421E‐06 2.584E‐05 4.948E‐07 2.972E‐05 131
101993_at 99532_at 95694_at 98626_at 93728_at 93312_at 94267_i_at 99575_at 102002_at 94931_at 98872_at 92499_at 99618_at 95472_f_at 99115_at 93844_at 95718_f_at 160234_at 97960_at 101509_at 98549_at 104117_at 98946_at 93740_at 160475_at 94895_at 95395_at 94109_at 102309_at 93324_at 103063_at 95522_i_at 97462_at 104169_at 96041_at 98037_at 96611_at 99166_at 93908_f_at 96261_at M32599_5_at 21923 22057 21969 59005 21807 67128 66177 66447 N/A 67890 22239 22255 66594 67530 66576 22272 66477 230484 216825 22327 22370 69641 78889 22608 67864 213541 229096 226442 54367 12192 22720 24135 68036 22771 19652 68576 69875 106264 280487 433771 N/A Tnc Tob1 Top1 Trappc2l Tsc22d1 Ube2g1 Ubl5 Ubqln1 Ubqln2 Ufm1 Ugt8a Uncx4.1 Uqcr Uqcrb Uqcrh Uqcrq Usmg5 Usp1 Usp22 Vbp1 Vtn Wdr20a Wsb1 Ybx1 Yipf4 Ythdf2 Ythdf3 Zfp281 Zfp326 Zfp36l1 Zfp62 Zfp68 Zfp706 Zic1 N/A N/A N/A N/A N/A N/A N/A 4.26 5.40 3.48 5.03 5.98 5.67 17.95 8.75 4.47 4.27 3.73 5.36 10.69 4.73 7.52 3.84 3.64 8.17 8.56 6.86 3.33 3.06 3.60 3.82 3.00 3.47 6.30 3.36 4.06 3.23 4.84 3.65 4.35 127.06 3.95 3.39 8.48 3.52 4.95 4.22 61.01 1.376E‐05 1.093E‐08 8.691E‐07 5.242E‐05 1.621E‐05 5.053E‐07 2.852E‐07 3.985E‐08 4.373E‐04 9.495E‐09 1.316E‐04 2.438E‐07 6.256E‐06 9.121E‐06 2.040E‐06 2.294E‐05 7.672E‐06 7.807E‐09 1.881E‐07 3.788E‐06 5.098E‐07 1.180E‐03 4.252E‐06 4.098E‐08 2.353E‐05 8.961E‐05 1.642E‐05 5.257E‐06 2.871E‐08 7.319E‐05 6.311E‐09 3.299E‐06 2.748E‐06 3.786E‐08 1.695E‐06 4.795E‐06 1.588E‐06 2.661E‐05 2.215E‐03 4.016E‐05 4.287E‐07 132
M12481_5_at MURINE_b1_at 93568_i_at 93569_f_at M12481_M_at 97197_r_at M32599_M_at 103303_at 102144_f_at N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 25.69 17.85 14.44 11.18 9.93 8.87 6.48 4.59 4.21 5.497E‐06 5.459E‐07 3.668E‐08 3.375E‐06 1.244E‐06 6.673E‐06 8.229E‐07 8.190E‐06 9.013E‐05 133
Table 5. Listing of DAVID annotations describing DRG enriched genes. DAVID terms
found to be significantly upregulated in the subtraction and microarray DRG gene lists.
The percentage of genes mapping to each annotation is listed (“%”), along with the data
set that the gene belongs to (“DS”). P-values were determined using EASE, followed by
a Benjamini-Hochberg correction for multiple comparisons.
Category Term %
P‐value DS
CHROMOSOME 11 11.1% 4.6E‐06 S CYTOBAND 11 E2 1.9% 3.2E‐03 S 0.6% 6.6E‐03 M 12.3% 2.9E‐03 S ENSEMBL_GENE_ID ENSMUSG00000007440 GO:0006464~protein modification GOTERM_BP_ALL process GO:0006464~protein modification GOTERM_BP_ALL process 12.8% 1.4E‐02 M GOTERM_BP_ALL GO:0006512~ubiquitin cycle 4.8% 1.7E‐03 M 4.7% 2.8E‐02 M 5.4% 2.9E‐02 M GO:0006886~intracellular protein GOTERM_BP_ALL transport GO:0007010~cytoskeleton GOTERM_BP_ALL organization and biogenesis GO:0007017~microtubule‐based GOTERM_BP_ALL process 2.9% 1.3E‐02 S GOTERM_BP_ALL GO:0008104~protein localization 7.6% 3.8E‐03 M GOTERM_BP_ALL GO:0008152~metabolic process 46.1% 3.9E‐05 S GOTERM_BP_ALL GO:0015031~protein transport 7.0% 6.2E‐04 M GOTERM_BP_ALL GO:0016043~cellular component 18.9% 2.0E‐04 S 134
organization and biogenesis GO:0016043~cellular component GOTERM_BP_ALL organization and biogenesis 20.9% 8.3E‐04 M 23.9% 4.5E‐03 M 21.0% 5.3E‐03 S 7.8% 6.9E‐03 M 36.2% 8.0E‐03 S 28.5% 7.4E‐03 S 12.7% 3.2E‐03 S 13.2% 2.1E‐02 M 42.8% 3.9E‐06 S GO:0019538~protein metabolic GOTERM_BP_ALL process GO:0019538~protein metabolic GOTERM_BP_ALL process GO:0033036~macromolecule GOTERM_BP_ALL localization GO:0043170~macromolecule GOTERM_BP_ALL metabolic process GO:0043283~biopolymer metabolic GOTERM_BP_ALL process GO:0043412~biopolymer GOTERM_BP_ALL modification GO:0043412~biopolymer GOTERM_BP_ALL modification GO:0044237~cellular metabolic GOTERM_BP_ALL process GO:0044238~primary metabolic GOTERM_BP_ALL process 42.1% 6.0E‐05 S GOTERM_BP_ALL GO:0044260~cellular macromolecule 20.7% 3.0E‐03 S 135
metabolic process GO:0044260~cellular macromolecule GOTERM_BP_ALL metabolic process 23.2% 6.2E‐03 M 20.2% 6.9E‐03 S 22.5% 2.0E‐02 M GO:0044267~cellular protein GOTERM_BP_ALL metabolic process GO:0044267~cellular protein GOTERM_BP_ALL metabolic process GO:0045184~establishment of GOTERM_BP_ALL protein localization 7.2% 2.2E‐03 M GOTERM_BP_ALL GO:0046907~intracellular transport 6.8% 1.5E‐02 M GOTERM_CC_ALL GO:0005737~cytoplasm 40.3% 6.5E‐15 S GOTERM_CC_ALL GO:0005737~cytoplasm 45.2% 8.7E‐15 M GOTERM_CC_ALL GO:0044444~cytoplasmic part 27.3% 8.6E‐06 M GOTERM_CC_ALL GO:0044444~cytoplasmic part 22.9% 4.2E‐04 S GOTERM_MF_ALL GO:0000166~nucleotide binding 16.0% 1.4E‐05 S GOTERM_MF_ALL GO:0000166~nucleotide binding 15.4% 2.3E‐02 M GOTERM_MF_ALL GO:0003824~catalytic activity 36.0% 1.6E‐05 M GOTERM_MF_ALL GO:0003824~catalytic activity 33.4% 2.2E‐03 S GOTERM_MF_ALL GO:0005488~binding 62.6% 1.6E‐06 S GOTERM_MF_ALL GO:0005515~protein binding 35.9% 6.0E‐06 S GOTERM_MF_ALL GO:0005524~ATP binding 10.9% 3.8E‐03 S GOTERM_MF_ALL GO:0008639~small protein 2.1% 4.2E‐02 S 136
conjugating enzyme activity GO:0016758~transferase activity, GOTERM_MF_ALL transferring hexosyl groups 2.0% 1.4E‐02 M GOTERM_MF_ALL GO:0016874~ligase activity 3.8% 7.3E‐03 M 2.3% 2.0E‐02 S 13.9% 2.5E‐04 S 14.1% 4.0E‐03 M 11.7% 1.1E‐03 S GO:0016881~acid‐amino acid ligase GOTERM_MF_ALL activity GO:0017076~purine nucleotide GOTERM_MF_ALL binding GO:0017076~purine nucleotide GOTERM_MF_ALL binding GO:0030554~adenyl nucleotide GOTERM_MF_ALL binding GO:0030554~adenyl nucleotide GOTERM_MF_ALL binding 11.6% 1.7E‐02 M GOTERM_MF_ALL GO:0032553~ribonucleotide binding 13.2% 9.4E‐04 S GOTERM_MF_ALL GO:0032553~ribonucleotide binding 13.6% 6.5E‐03 M 13.2% 9.6E‐04 S 13.6% 6.2E‐03 M 11.0% 3.7E‐03 S GO:0032555~purine ribonucleotide GOTERM_MF_ALL binding GO:0032555~purine ribonucleotide GOTERM_MF_ALL binding GO:0032559~adenyl ribonucleotide GOTERM_MF_ALL binding 137
GO:0032559~adenyl ribonucleotide GOTERM_MF_ALL binding 11.1% 2.9E‐02 M BP00044:mRNA transcription PANTHER_BP_ALL regulation 45.7% 3.7E‐02 S PANTHER_BP_ALL BP00063:Protein modification 15.8% 9.2E‐03 S PANTHER_BP_ALL BP00064:Protein phosphorylation 14.5% 3.7E‐04 S BP00069:Protein disulfide‐isomerase PANTHER_BP_ALL reaction 4.2% 2.2E‐02 S PANTHER_BP_ALL BP00071:Proteolysis 26.4% 2.2E‐02 S BP00104:G‐protein mediated PANTHER_BP_ALL signaling 20.0% 1.6E‐03 S PANTHER_BP_ALL BP00124:Cell adhesion 5.3% 1.0E‐02 S PANTHER_BP_ALL BP00125:Intracellular protein traffic 6.4% 3.2E‐03 S PANTHER_BP_ALL BP00125:Intracellular protein traffic 7.1% 3.0E‐02 M PANTHER_BP_ALL BP00143:Cation transport 22.0% 8.3E‐05 S PANTHER_BP_ALL BP00272:Phospholipid metabolism 5.4% 7.8E‐03 S PANTHER_BP_ALL BP00285:Cell structure and motility 10.0% 9.5E‐04 S PANTHER_MF_ALL MF00093:Select regulatory molecule 8.3% 1.1E‐03 S PANTHER_MF_ALL MF00108:Protein kinase 10.4% 8.8E‐03 S PANTHER_MF_ALL MF00121:Aminoacyl‐tRNA synthetase
2.7% 1.0E‐03 M PANTHER_MF_ALL MF00131:Transferase 12.3% 1.8E‐02 S PANTHER_MF_ALL MF00178:Extracellular matrix 5.6% 4.2E‐02 S 138
MF00213:Non‐receptor PANTHER_MF_ALL serine/threonine protein kinase 26.0% 3.7E‐06 S 5.3% 3.4E‐02 S 3.3% 4.1E‐02 M PANTHER_SUBFAMILY ALPHA 1.0% 2.8E‐07 M PUBMED_ID 9371744 2.7% 7.9E‐03 S PUBMED_ID 9811942 2.4% 4.5E‐03 S PUBMED_ID 9971739 0.5% 4.7E‐02 S PUBMED_ID 10725249 17.4% 8.1E‐14 S PUBMED_ID 10725249 22.1% 1.9E‐06 M PUBMED_ID 11044609 1.8% 2.5E‐03 S PUBMED_ID 11125038 9.3% 7.9E‐12 S PUBMED_ID 11544199 6.7% 1.4E‐11 S PUBMED_ID 12465718 1.5% 3.5E‐03 S PUBMED_ID 12693553 5.1% 5.1E‐13 S PUBMED_ID 12865426 1.9% 1.6E‐03 S PUBMED_ID 14595844 1.2% 1.8E‐03 S PUBMED_ID 14621295 4.0% 1.1E‐04 S PUBMED_ID 14651853 3.8% 4.1E‐04 S MF00264:Microtubule family PANTHER_MF_ALL cytoskeletal protein MF00270:Membrane traffic PANTHER_MF_ALL regulatory protein PTHR10596:SF92~PROTOCADHERIN 139
PUBMED_ID 14672974 0.8% 4.6E‐03 M PUBMED_ID 14681479 7.1% 4.3E‐15 S PUBMED_ID 15368895 4.5% 8.5E‐08 S PUBMED_ID 15570159 1.0% 9.3E‐06 M PUBMED_ID 15618518 7.6% 2.8E‐06 S PUBMED_ID 15640798 1.0% 4.1E‐07 M PUBMED_ID 15782199 40.0% 1.9E‐112 S PUBMED_ID 15782199 43.4% 8.0E‐38 M PUBMED_ID 16452087 2.2% 9.5E‐04 S PUBMED_ID 16498405 0.9% 3.9E‐02 S PUBMED_ID 16800626 0.9% 9.2E‐03 S PUBMED_ID 16916647 1.7% 2.3E‐03 S PUBMED_ID 17114649 4.4% 3.5E‐08 S PUBMED_ID 17114649 4.5% 8.9E‐03 M PUBMED_ID 17203969 4.2% 1.3E‐09 S PUBMED_ID 17242355 10.6% 1.4E‐23 S PUBMED_ID 17242355 11.7% 8.3E‐08 M SP_COMMENT_TYPE function 49.1% 1.9E‐07 S SP_COMMENT_TYPE interaction 6.3% 2.7E‐03 S SP_COMMENT_TYPE PTM 14.8% 1.3E‐04 S SP_COMMENT_TYPE subunit 36.5% 5.7E‐14 S SP_PIR_KEYWORDS 22.4% 2.7E‐03 S alternative splicing 140
SP_PIR_KEYWORDS atp‐binding 9.6% 8.9E‐03 M SP_PIR_KEYWORDS atp‐binding 8.8% 2.5E‐02 S SP_PIR_KEYWORDS Coiled coil 10.4% 1.2E‐02 S SP_PIR_KEYWORDS cytoplasm 17.5% 1.2E‐05 S SP_PIR_KEYWORDS cytoplasm 18.8% 1.6E‐03 M SP_PIR_KEYWORDS 3.1% 1.2E‐02 M SP_PIR_KEYWORDS nucleotide‐binding 12.3% 1.7E‐03 M SP_PIR_KEYWORDS nucleotide‐binding 11.3% 2.6E‐03 S SP_PIR_KEYWORDS phosphoprotein 31.4% 4.6E‐12 S SP_PIR_KEYWORDS protein transport 5.1% 1.6E‐04 M ligase 141
Table 6. Listing of transcription factor binding consensuses that are significantly
overrepresented within the promoters of DRG enriched genes. Consensuses were
matched the promoters of DRG enriched genes from either the subtraction or microarray.
A background list of genes are then used to calculate a P-value using a two-tailed Fisher
Exact Test followed by a Benjamini-Hochberg correction for multiple comparisons. DS
= data set used in the analysis, (S)ubtraction or (M)icroarray. S = source of consensus
uses, (T)ransfac 10.2 or (C)arninci et al. (2005) * = loose consensus.
Identifier V$SP1_01 Consensus Description
GGGGCGGGGT P‐value DS
S
Stimulating protein 1 5.6E‐11 M T Zinc finger protein 143 4.2E‐09 S C ACTAYRNNNCCC
DM_4 R DM_115 YRTCANNRCGC Discovered motif 115 2.2E‐08 M C P$ABI4_01 NNGCACCGCCC Abi4 (Zea mays) 5.5E‐08 S T ACTAYRNNNCCC
DM_4 R Zinc finger protein 143 1.8E‐07 M C DM_162 TAANNYSGCG Discovered motif 162 5.2E‐07 S C 1.4E‐06 M C transcription 1.5E‐06 M C v‐rel reticuloendotheliosis viral C‐REL(*) GCGNNANTTCC oncogene homolog (avian) Signal transducer and activator of STAT TCCCRGAAR MYAATNNNNNN
DM_138 NGGC Discovered motif 138 3.0E‐06 M C F$XBP1_Q2 CTTCGAG Xbp1 (Saccharomyces cerevisiae) 7.4E‐06 M T F$STRE_01 TMAGGGGN Stress response element 1.2E‐05 S T DM_115 YRTCANNRCGC Discovered motif 115 1.5E‐05 S C 142
F$CBF1_B NRTCACRTGA Cbf1 (Saccharomyces cerevisiae) 3.5E‐05 M T Early growth response 4 6.5E‐05 M T gamma 7.0E‐05 M T Myogenin 8.8E‐05 S T 1.0E‐04 M C CCCGCCCCCRCCC
V$KROX_Q6 C Nuclear transcription factor Y, V$NFY_Q6 TRRCCAATSRN NCACCTGYYNCN
V$E2A_Q2 KN DM_118 GGCNRNWCTTYS Discovered motif 118 chorion factor 1 (Drosophila I$CF1_02 GGGGTCACG melanogaster) 1.2E‐04 M T V$TBX5_02 YNRGGTGTKV T‐box 5 1.2E‐04 S T DM_152 WCAANNNYCAG Discovered motif 152 1.4E‐04 M C 1.5E‐04 S T Discovered motif 104, CCAAT box 1.9E‐04 M C CCACCANMNNC Lim1 (Tabacco, Nicotiana P$LIM1_01 N tabacum) CCAATNNSNNNG
DM_104 CG NNNNNTGYGGTY
V$CBF_02 NNNN Core binding factor 2.1E‐04 M T F$STRE_01 TMAGGGGN Stress response element 2.3E‐04 M T AACYNNNNTTCC
DM_113 S Discovered motif 113 3.0E‐04 M C V$AP4_Q6_01 RNCAGCTGC Transcription factor Ap4 3.7E‐04 S T 143
NYSTCACGTGAB basic helix‐loop‐helix domain V$STRA13_01 NN containing, class B, 2 4.3E‐04 M T Discovered motif 123 5.2E‐04 M C 5.7E‐04 M C Discovered motif 81 6.7E‐04 M C GGCNNMSMYNT
DM_123 TG DM_63 AAGWWRNYGGC Discovered motif 63 CGGAARNGGCN
DM_81 G CGGAARNGGCN
DM_81 G Discovered motif 81 6.7E‐04 S C DM_63 AAGWWRNYGGC Discovered motif 63 7.2E‐04 S C DM_45 CCANNAGRKGGC Discovered motif 45 7.8E‐04 S C 7.8E‐04 S T Androgen receptor half site V$AR_Q6 WGAGCANRN matrix V$E12_Q6 RRCAGGTGNCV E12 (Xenopus laevis) 8.2E‐04 M T DM_114 YTCCCRNNAGGY Discovered motif 114 8.8E‐04 S C V$EGR_Q6 GTGGGSGCRRS Early growth response 3 1.0E‐03 M T V$EGR_Q6 GTGGGSGCRRS Early growth response 3 1.0E‐03 S T Discovered motif 104, CCAAT box 1.2E‐03 S C CCAATNNSNNNG
DM_104 CG AACYNNNNTTCC
DM_113 S Discovered motif 113 1.2E‐03 S C DM_91 GTGGGTGK Meningioma (disrupted in 1.2E‐03 S C 144
balanced translocation) 1 V$ATF_B NTGACGTCANYS Activating transcription factor 1.3E‐03 M T DM_102 ATGGYGGA Discovered motif 102 1.4E‐03 M C F$DDE1_B CGCTCAGCC Dde box (Neurospora crassa) 1.6E‐03 M T P$ABI4_01 NNGCACCGCCC Abi4 (Zea mays) 1.7E‐03 M T 1.8E‐03 S T NNNNNTGYGGTY
V$CBF_02 NNNN Core binding factor Signal transducer and activator of STAT TCCCRGAAR transcription 1 2.0E‐03 S C V$AP2REP_01 CAGTGGG Transcription factor Ap2 gamma 2.1E‐03 M T V$AP2REP_01 CAGTGGG Transcription factor Ap2 gamma 2.1E‐03 S T chorion factor 1 (Drosophila I$CF1_01 GGGGTCAYS melanogaster) 2.4E‐03 M T DM_162 TAANNYSGCG Discovered motif 162 2.5E‐03 M C 2.8E‐03 S T 2.9E‐03 S C Paired‐like homeodomain V$PITX2_Q2 WNTAATCCCAR transcription factor 2 Signal transducer and activator of STAT1(*) TNCATNTCCYR transcription V$MYB_Q3 NNNGNCAGTTN c‐Myb (Gallus gallus) 2.9E‐03 M T P$DOF2_01 NNNWAAAGCNN Dof2 zing finger (Zea mays) 3.2E‐03 M T V$GATA6_01 NNNGATWANN Gata‐6b (Xenopus laevis) 3.3E‐03 M T V$PAX3_B NNNNNNCGTCAC Paired box gene 3 3.3E‐03 S T 145
GSTYNNNNN NNNGGKTGCCCS
V$HIC1_03 NNNNNN Hypermethylated in cancer 1 3.8E‐03 S T Discovered motif 19 4.2E‐03 M C 4.8E‐03 S C 4.8E‐03 M T 4.9E‐03 M T GKCGCNNNNNN
DM_19 NTGAYG DM_152 WCAANNNYCAG Discovered motif 152 Scalloped (Drosophila I$SD_Q6 CATTYCN melanogaster) V$ER_Q6_02 NAGGTCANNNY Estrogen receptor NNNGGKTGCCCS
V$HIC1_03 NNNNNN Hypermethylated in cancer 1 4.9E‐03 M T MEIS1 TGACAGNY Meis homeobox 1 5.0E‐03 S C V$MYCMAX_B GCCAYGYGSN c‐Myc:Max heterodimer 5.0E‐03 M T Pif3 (Arabidopsis thaliana) 5.5E‐03 M T balanced translocation) 1 6.0E‐03 M C Discovered motif 74 6.5E‐03 M C NNNVCCACGTGG
P$PIF3_02 NMVNN Meningioma (disrupted in DM_91 GTGGGTGK GGAMTNNNNNT
DM_74 CCY GGAMTNNNNNT
DM_74 CCY Discovered motif 74 6.5E‐03 S C ER TGACCT Estrogen receptor 7.1E‐03 M C 146
ATF6 TGACGTGK Activating transcription factor 6 DM_66 RACTNNRTTTNC Discovered motif 66 7.8E‐03 M C 7.9E‐03 S C 7.9E‐03 S T NNNVCCACGTGG
P$PIF3_02 NMVNN Pif3 (Arabidopsis thaliana) DM_114 YTCCCRNNAGGY Discovered motif 114 8.0E‐03 M C P$DOF2_01 NNNWAAAGCNN Dof2 zing finger (Zea mays) 8.0E‐03 S T 8.4E‐03 S C 8.4E‐03 S T 8.5E‐03 S C SNACANNNYSYA
DM_80 GA Discovered motif 80 P$RAV1_02 NNCACCTGRNNN Rav1 (Arabidopsis thaliana) TRAnsformer (Caenorhabditis N$TRA1_01 TGGGWGGT elegans) TRAnsformer (Caenorhabditis N$TRA1_02 TGGGWGGT elegans) 8.5E‐03 S C V$ATF1_Q6 CYYTGACGTCA Activating transcription factor 1 8.5E‐03 M T v‐rel reticuloendotheliosis viral C‐REL(*) GCGNNANTTCC oncogene homolog (avian) 8.6E‐03 S C DM_99 ATCMNTCCGY Discovered motif 99 8.7E‐03 S C Myeloid zinc finger 1 9.4E‐03 S T Discovered motif 138 1.1E‐02 S C 1.1E‐02 M T KNNNKAGGGGN
V$MZF1_02 AA MYAATNNNNNN
DM_138 NGGC V$MYOD_Q6_01 CNGNRNCAGGTG Myogenin 147
NNGNAN V$P53_02 NGRCWTGYCY Tumor suppressor p53 1.1E‐02 S T CCCNNNNNNAA
DM_158 GWT Discovered motif 158 1.2E‐02 S C V$HEB_Q6 RCCWGCTG TCF12/transcription factor 12 1.2E‐02 S T 1.2E‐02 M T NNGTNRCNATRG Regulatory factor X, 1 (influences V$RFX1_02 YAACNN HLA class II expression) DM_95 GCCNNNWTAAR Discovered motif 95 1.3E‐02 S C F$GBF_Q6 TTGGGGGTG 1.3E‐02 S T Gbf (Dictyostelium discoideum) Nuclear transcription factor Y, V$NFY_Q6 TRRCCAATSRN gamma 1.3E‐02 S T V$SP1_01 GGGGCGGGGT Stimulating protein 1 1.3E‐02 S T F$DDE1_B CGCTCAGCC Dde box (Neurospora crassa) 1.4E‐02 S T P$HBP1A_Q2 GNCACGTGGC Hbp1a (Triticum aestivum) 1.4E‐02 M T NCACCTGYYNCN
V$E2A_Q2 KN Myogenin 1.4E‐02 M T V$MYB_Q6 NNNAACTGNC c‐Myb (Gallus gallus) 1.4E‐02 S T TGGNNNNNNKCC
DM_27 AR Discovered motif 27 1.5E‐02 S C IK‐1 TTTCCCANR IKAROS family zinc finger 1 1.5E‐02 S C Early growth response 4 1.5E‐02 S T CCCGCCCCCRCCC
V$KROX_Q6 C 148
T3R TGACCTY Thyroid hormone receptor 1.6E‐02 M C ESRRA TGACCTY Estrogen related receptor, alpha 1.7E‐02 M C V$GATA6_01 NNNGATWANN Gata‐6b (Xenopus laevis) 1.7E‐02 S T V$HEB_Q6 RCCWGCTG TCF12/transcription factor 12 1.7E‐02 M T DM_29 CTTTAAR Discovered motif 29 1.8E‐02 M C melanogaster) 1.8E‐02 S T Scalloped (Drosophila I$SD_Q6 CATTYCN YNNBYYNCATTCC
F$ABAA_01 NNNNNN AbaA (Aspergillus nidulans) 2.0E‐02 S T P$ROM_Q2 GCCACSTCA Rom2 (Phaseolus vulgaris) 2.0E‐02 M T V$E12_Q6 RRCAGGTGNCV E12 (Xenopus laevis) 2.0E‐02 S T 2.0E‐02 S T 2.1E‐02 S T 2.2E‐02 M T 2.2E‐02 M T Zinc finger and BTB domain V$KAISO_01 NTCCTGCNAN containing 33 BRCA1 containing protein V$BRCA_01 KTNNGTTG complex CCACCANMNNC Lim1 (Tabacco, Nicotiana P$LIM1_01 N tabacum) Thyroid hormone receptor, alpha SNNTRAGGTCAC (erythroblastic leukemia viral (v‐
V$T3R_01 GSNN erb‐a) oncogene homolog, avian)
V$ATF_B NTGACGTCANYS Activating transcription factor 2.3E‐02 S T V$BRCA_01 KTNNGTTG 2.3E‐02 M T BRCA1 containing protein 149
complex V$E2F_Q6_01 NKCGCGCSAAAN E2F transcription factor 7 2.3E‐02 M T V$GR_Q6_01 NNTGTYCT Glucocorticoid receptor 2.4E‐02 M T DM_109 YGTCCTTGR Discovered motif 109 2.5E‐02 S C AYCVDCCAATNA HAP2/3/5, yeast homolog to F$HAP234_01 NMNN vertebrate NF‐Y/CP1/CBF 2.5E‐02 S T NF‐MUE1 CGGCCATYK YY1 transcription factor 2.5E‐02 M C UF1‐H3beta 2.5E‐02 S T GGYGGGGGAGG
V$UF1H3BETA_Q6 GGC chorion factor 1 (Drosophila I$CF1_02 GGGGTCACG melanogaster) 2.8E‐02 S T DM_29 CTTTAAR Discovered motif 29 2.9E‐02 S C Myocyte enhancer factor 2A 2.9E‐02 M T factor 1 3.0E‐02 S C Abz (Lycopersicon esculentum) 3.0E‐02 M T homeobox 3.2E‐02 S T Discovered motif 165 3.3E‐02 M C KCTAWAAATAG
V$MEF2_02 M Lymphoid enhancer‐binding LEF1 CTTTGT KKNTKACGTGGN
P$ABZ1_01 NN YNNNTAATCYSM Retina and anterior neural fold V$CRX_Q4 N CCAWNWWNNN
DM_165 GGC 150
F$HAC1_Q2 KGMCAGCGTGTC Hac1 (Saccharomyces cerevisiae) 3.3E‐02 S T 3.3E‐02 S C RYTGCNNRGNAA
MIF‐1 C MIF‐1 v‐maf musculoaponeurotic fibrosarcoma oncogene family, NF‐E2 TGASTMAGC protein K (avian) 3.4E‐02 M C P$EMBP1_Q2 GCCACGTGDN EmBP‐1b (Triticum aestivum) 3.4E‐02 M T DM_34 CYTAGCAAY Discovered motif 34 3.5E‐02 M C ERRALPHA TGACCTTG Estrogen related receptor, alpha 3.6E‐02 S C NNNNNKCCNNTC
NNBCNNGGMNN
F$FACBALL_Q2 NWN Facb (Aspergillus nidulans) 3.6E‐02 S T F$XBP1_Q2 CTTCGAG Xbp1 (Saccharomyces cerevisiae) 3.6E‐02 S T SF‐1 TGACCTTG Steroidogenic factor 1 3.6E‐02 S C site matrix 3.6E‐02 S T c‐Myc:Max heterodimer 3.9E‐02 S T Thyroid hormone receptor, half V$T3R_Q6 MNTGWCCTN NNACCACGTGGT
V$MYCMAX_01 NN Androgen receptor half site V$AR_Q6 WGAGCANRN matrix 4.0E‐02 M T RGRCAWGNCY Tumor suppressor p53 4.1E‐02 M T V$P53_DECAMER_
Q2 151
NKCNGHYACGTC
P$HBP1B_Q6 AC Hbp1b (Triticum aestivum) 4.2E‐02 M T CNSTGACGTNNN
V$ATF_01 YC Activating transcription factor 4.4E‐02 M T V$OSF2_Q6 ACCACANM runt related transcription factor 2 4.4E‐02 M T DM_118 GGCNRNWCTTYS Discovered motif 118 4.5E‐02 S C CCCNNNNNNAA
DM_158 GWT Discovered motif 158 4.5E‐02 M C DM_164 GCGSCMNTTT Discovered motif 164 4.5E‐02 M C DM_46 WTTGKCTG Discovered motif 46 4.5E‐02 S C ELF‐1 RGAGGAARY E74‐like factor 1 4.5E‐02 M C Spleen focus forming virus (SFFV) proviral integration oncogene PU.1 RGAGGAARY spi1 4.5E‐02 M C F$GCR1_01 GGCTTCCWC Gcr1 (Saccharomyces cerevisiae) 4.6E‐02 M T 4.7E‐02 S T 4.9E‐02 M T LIM domain only 2 (rhombotin‐
V$LMO2COM_02 NMGATANSG V$USF_Q6_01 like 1) NRCCACGTGASN Upstream transcription factor 152
Table 7. Listing of DRG enriched transcription factor networks. Full listing of
transcription factor centered sub-networks found to connect significantly to DRG
enriched genes. Significance was determined by Metacore’s exact test (described in the
materials and methods) followed by a Benjamini-Hochberg correction for multiple
comparisons. DS = data set used in the analysis, (S)ubtraction or (M)icroarray.
Factor Description
Nodes P‐value DS
AFX1/FOXO4 Forkhead box O4 51 8.3E‐03 M AHR Aryl‐hydrocarbon receptor 53 6.6E‐04 S AHR Aryl‐hydrocarbon receptor 57 8.4E‐03 M AP‐1 Activator protein 1 59 1.3E‐03 M ARNT Aryl hydrocarbon receptor nuclear translocator
50 6.6E‐04 S AUF1 Heterogeneous nuclear ribonucleoprotein D 51 6.8E‐04 S Bcl‐6 B‐cell CLL/lymphoma 6 51 6.1E‐04 S Bcl‐6 B‐cell CLL/lymphoma 7 53 3.0E‐03 M v‐myb myeloblastosis viral oncogene homolog b‐Myb (avian)‐like 2 53 1.2E‐02 M Brca1 breast cancer 1, early onset 50 6.5E‐04 S Brca1 Breast cancer 1, early onset 51 8.5E‐03 M C/EBP zeta CCAAT/enhancer binding protein zeta 52 7.4E‐04 S C/EBP zeta CCAAT/enhancer binding protein zeta 51 1.2E‐02 M 52 1.0E‐02 M 54 6.1E‐04 S v‐maf musculoaponeurotic fibrosarcoma c‐Maf oncogene homolog (avian) c‐Rel (NF‐kB v‐rel reticuloendotheliosis viral oncogene subunit) homolog (avian) 153
Basic helix‐loop‐helix domain containing, class DEC1/Bhlhb2 B2 50 2.0E‐04 S E2F5 E2F transcription factor 5, p130‐binding 50 9.9E‐03 M EBF Early B‐cell factor 1 51 4.2E‐03 M EKLF1 Kruppel‐like factor 1 (erythroid) 52 1.1E‐02 M E74‐like factor 4 (ets domain transcription ELF4 factor) 51 8.0E‐04 S Elk‐1 LK1, member of ETS oncogene family 51 1.2E‐02 M EPAS1 Endothelial PAS domain protein 1 53 1.3E‐02 M ERR1 Estrogen related receptor, alpha 56 4.8E‐04 S ERR1 Estrogen related receptor, alpha 55 9.9E‐03 M ESR2 Estrogen receptor 2 (ER beta) 51 6.4E‐04 S ESR2 Estrogen receptor 2 (ER beta) 52 1.2E‐02 M 50 6.7E‐04 S v‐ets erythroblastosis virus E26 oncogene ETS2 homolog 2 (avian) EWS/FLI1 fusion Ewing sarcoma breakpoint region 1 / Friend protein leukemia integration 1 51 9.2E‐03 M FKHR/Foxo1 Forkhead box O1 55 5.4E‐04 S FKHR/Foxo1 Forkhead box O1 58 7.2E‐03 M FOXP3 Forkhead box P3 51 7.8E‐04 S FOXP3 Forkhead box P3 55 1.1E‐02 M Fra‐1 Fos‐like antigen 1 50 9.6E‐03 M 154
GA binding protein transcription factor, alpha GABP alpha subunit 53 1.3E‐02 M GATA‐3 GATA binding protein 3 51 8.7E‐04 S GATA‐3 GATA binding protein 3 52 1.2E‐02 M 56 4.5E‐04 S Nuclear receptor subfamily 3, group C, member GCR‐alpha 1 (glucocorticoid receptor) Nuclear receptor subfamily 3, group C, member GCR‐alpha 1 (glucocorticoid receptor) 56 9.5E‐03 M HAND1 Heart and neural crest derivatives expressed 1 51 1.0E‐02 M HIF‐1 Hypoxia‐inducible factor 1, alpha subunit 52 8.2E‐04 S HIF‐1 Hypoxia‐inducible factor 1, alpha subunit 56 9.9E‐03 M HIF1A Hypoxia‐inducible factor 1, alpha subunit 62 4.7E‐03 M 51 5.7E‐04 S Hypoxia‐inducible factor 1, alpha subunit / aryl‐
HIF1A/ARNT2 hydrocarbon receptor nuclear translocator 2 Hypoxia‐inducible factor 1, alpha subunit / aryl‐
HIF1A/ARNT2 hydrocarbon receptor nuclear translocator 2 51 8.9E‐03 M HNF1‐alpha HNF1 homeobox A 60 1.9E‐04 S HNF1‐alpha HNF1 homeobox A 79 5.7E‐04 M HNF3‐beta Forkhead box A2 53 6.9E‐04 S HOXB4 Homeo box B4 50 6.8E‐04 S HSF1 Heat shock transcription factor 1 58 3.1E‐04 S HSF1 Heat shock transcription factor 1 67 2.9E‐03 M 155
HSF4 Heat shock transcription factor 4 51 1.1E‐02 M IFNGR1 Interferon gamma receptor 1 50 7.4E‐04 S Ikaros/Ikzf1 IKAROS family zinc finger 1 51 1.2E‐02 M IRF1 Interferon regulatory factor 1 54 1.1E‐04 S IRF1 Interferon regulatory factor 1 53 1.3E‐02 M IRF2 Interferon regulatory factor 2 51 5.8E‐04 S IRF4 Interferon regulatory factor 4 50 9.7E‐03 M IRF7 Interferon regulatory factor 7 51 1.1E‐02 M IRF8 Interferon regulatory factor 8 52 7.9E‐04 S JunB Jun‐B oncogene 51 1.8E‐04 S JunB Jun‐B oncogene 51 9.7E‐03 M KLF13 Kruppel‐like factor 13 51 1.0E‐02 M KLF4 Kruppel‐like factor 4 54 5.9E‐04 S KLF4 Kruppel‐like factor 4 52 1.3E‐02 M LEDGF/PSIP1 PC4 and SFRS1 interacting protein 1 51 1.1E‐02 M Lef‐1 Lymphoid enhancer binding factor 1 51 7.3E‐04 S LZIP/CREB3 cAMP responsive element binding protein 3 51 8.2E‐04 S MBD1 Methyl‐CpG binding domain protein 1 51 1.2E‐02 M MCR Mast cell regulator 51 1.2E‐02 M MITF Microphthalmia‐associated transcription factor
51 7.4E‐04 S MITF Microphthalmia‐associated transcription factor
52 1.3E‐02 M NANOG Nanog homeobox 59 6.1E‐03 M 156
Non‐metastatic cells 1, protein (NM23A) NDPK A expressed in 50 4.4E‐03 M 51 9.3E‐03 M 52 7.7E‐04 S 52 7.0E‐04 S 55 4.9E‐04 S 53 7.5E‐04 S Non‐metastatic cells 2, protein (NM23B) NDPK B expressed in NeuroD1 (NDF1) Neurogenic differentiation 1 Nuclear factor of activated T‐cells, cytoplasmic, NF‐AT1(NFATC2) calcineurin‐dependent 2 Nuclear factor of activated T‐cells, cytoplasmic, NF‐AT2(NFATC1) calcineurin‐dependent 1 Nuclear factor of kappa light polypeptide gene NF‐kB p50/p65 enhancer in B‐cells 1 p50/p65 heterodimer Nuclear factor of kappa light polypeptide gene NF‐kB p50/p65 enhancer in B‐cells 1 p50/p65 heterodimer 55 1.1E‐02 M NF‐Y Nuclear transcription factor‐Y 68 5.6E‐05 S NFYA Nuclear transcription factor‐Y alpha 52 1.2E‐02 M 51 8.6E‐03 M 52 1.3E‐02 M Oct‐3/4 (Pou5f1) POU domain, class 5, transcription factor 1 60 5.6E‐03 M p21/Cdkn1a Cyclin‐dependent kinase inhibitor 1A (P21) 50 1.9E‐04 S p63 Transformation related protein 63 57 3.8E‐04 S Nuclear receptor subfamily 4, group A, member NOR1/Nr4a3 3 Nuclear receptor subfamily 4, group A, member NURR1/Nr4a2 2 157
p63 Transformation related protein 63 61 5.4E‐03 M PARP‐1 Poly (ADP‐ribose) polymerase family, member 1
51 6.6E‐04 S PAX4 Paired box gene 4 50 7.1E‐04 S PAX4 Paired box gene 4 51 8.8E‐03 M PAX5 Paired box gene 5 52 1.1E‐02 M PIT1/Pou1f1 POU domain, class 1, transcription factor 1 52 1.1E‐02 M Pitx2 Paired‐like homeodomain transcription factor 2
50 7.0E‐04 S Polymerase (RNA) II (DNA directed) polypeptide POLR2A A 51 9.6E‐03 M POU4F1 POU domain, class 4, transcription factor 1 51 8.4E‐04 S PU.1/Sfpi1 SFFV proviral integration 1 55 5.7E‐04 S PU.1/Sfpi1 SFFV proviral integration 1 55 9.6E‐03 M 53 7.2E‐04 S 55 9.3E‐03 M Nuclear receptor subfamily 1, group I, member PXR 2 Nuclear receptor subfamily 1, group I, member PXR 2 RAR‐alpha/RXR‐ Retinoic acid receptor, alpha / retinoid X alpha receptor alpha 53 1.2E‐02 M RARbeta Retinoic acid receptor, beta 50 6.3E‐04 S RBP‐J kappa Recombination signal binding protein for (CBF1) immunoglobulin kappa J region 50 1.0E‐02 M 53 1.1E‐04 S RelA (p65 NF‐kB v‐rel reticuloendotheliosis viral oncogene 158
subunit) homolog A RelA (p65 NF‐kB v‐rel reticuloendotheliosis viral oncogene subunit) homolog A 60 5.9E‐03 M ROR‐alpha RAR‐related orphan receptor alpha 51 6.9E‐04 S RUNX3 Runt related transcription factor 3 50 6.4E‐04 S RXRA Retinoid X receptor alpha 51 6.2E‐04 S 52 6.8E‐04 S Nuclear receptor subfamily 5, group A, member SF1/Nr5a1 1 Nuclear receptor subfamily 5, group A, member SF1/Nr5a1 1 52 1.2E‐02 M SMAD2 MAD homolog 2 (Drosophila) 51 5.9E‐04 S SMAD4 MAD homolog 4 (Drosophila) 51 6.0E‐04 S SMAD4 MAD homolog 4 (Drosophila) 51 1.1E‐02 M SNAIL1 Snail homolog 1 (Drosophila) 51 9.0E‐03 M SOX17 SRY‐box containing gene 17 51 6.5E‐04 S SOX2 SRY‐box containing gene 2 52 1.1E‐02 M SOX9 SRY‐box containing gene 9 52 7.2E‐04 S SOX9 SRY‐box containing gene 9 51 8.4E‐03 M 61 5.1E‐03 M 59 6.5E‐03 M Sterol regulatory element binding transcription SREBP1 (nuclear) factor 1 Sterol regulatory element binding transcription SREBP2 (nuclear) factor 2 159
SRF Serum response factor 63 4.5E‐03 M 56 7.3E‐05 S 53 3.3E‐03 M TCERG1 (CA150) (CA150) 51 1.1E‐02 M TEF‐1/Tead1 TEA domain family member 1 51 7.1E‐04 S TEF‐4/Tead2 TEA domain family member 2 50 7.2E‐04 S TIF1‐beta/Trim28 Tripartite motif‐containing 28 51 7.6E‐04 S TIF1‐beta/Trim28 Tripartite motif‐containing 28 51 9.4E‐03 M Tip60/Kat5 K(lysine) acetyltransferase 5 51 1.0E‐02 M TITF1/Nkx2‐1 NK2 homeobox 1 52 1.4E‐02 M TR‐alpha Thyroid hormone receptor alpha 55 1.0E‐02 M Signal transducer and activator of transcription STAT3 3 Signal transducer and activator of transcription STAT3 3 Tcerg1 transcription elongation regulator 1 Upstream binding transcription factor, RNA UBF polymerase I 50 6.9E‐04 S USF1 Upstream transcription factor 1 55 5.1E‐04 S ZNF148 Zinc finger protein 148 53 6.4E‐04 S ZNF148 Zinc finger protein 148 51 9.9E‐03 M ZNF42 (MZF1) Myeloid zinc finger 1 53 1.2E‐02 M 160
Table 8. Listing of real time PCR results from 84 genes in the JAK/STAT signaling
pathway. Fold enrichment for DRG enriched mRNAs relative to cerebellum were
calculated using the ∆∆Ct method. Significance was determined by a two tailed
Student’s t-test (p<0.01, n=3), followed by a Benjamini-Hochberg correction for multiple
comparisons. Mean ∆Ct values are listed for each sample after normalization to
housekeeping controls.
Symbol Gene A2m Sh2b2 Bcl2l1 Ccnd1 Cdkn1a Cebpb Crk Crp Csf1r Csf2rb2 Cxcl9 EGFR Epor F2 F2r Fas Fcer1a Fcgr1 Isg15 Gata3 Gbp1 Ghr Hmga1 Ifnar1 Ifng Ifngr1 Il10ra Il10rb Il20 Alpha‐2‐macroglobulin SH2B adaptor protein 2 Bcl2‐like 1 Cyclin D1 Cyclin‐dependent kinase inhibitor 1A (P21) CCAAT/enhancer binding protein (C/EBP), beta V‐crk sarcoma virus CT10 oncogene homolog (avian) C‐reactive protein, pentraxin‐related Colony stimulating factor 1 receptor Colony stimulating factor 2 receptor, beta 2, low‐affinity (granulocyte‐macrophage) Chemokine (C‐X‐C motif) ligand 9 Epidermal growth factor receptor Erythropoietin receptor Coagulation factor II Coagulation factor II (thrombin) receptor Fas (TNF receptor superfamily member) Fc receptor, IgE, high affinity I, alpha polypeptide Fc receptor, IgG, high affinity I ISG15 ubiquitin‐like modifier GATA binding protein 3 Guanylate nucleotide binding protein 1 Growth hormone receptor High mobility group AT‐hook 1 Interferon (alpha and beta) receptor 1 Interferon gamma Interferon gamma receptor 1 Interleukin 10 receptor, alpha Interleukin 10 receptor, beta Interleukin 20 Fold enrichment P‐value 1.03 ‐3.79 ‐2.25 ‐3.53 3.07 4.48 9.6E‐01 2.1E‐02 2.8E‐02 2.3E‐02 3.4E‐02 6.4E‐03 ‐1.02 ‐1.47 ‐7.11 9.5E‐01 4.7E‐01 9.4E‐03 ‐1.54 1.39 ‐25.18 ‐3.67 ‐2.28 ‐2.19 2.69 1.6E‐01 5.2E‐01 3.5E‐03 1.5E‐02 1.3E‐01 9.0E‐02 1.5E‐01 ‐1.10 ‐5.00 1.93 ‐3.30 ‐1.30 4.94 ‐1.59 1.68 ‐1.26 1.24 3.89 14.35 ‐1.38 8.5E‐01 3.4E‐02 7.6E‐02 1.6E‐02 8.0E‐01 9.8E‐03 2.7E‐02 9.9E‐03 6.2E‐01 3.8E‐01 1.6E‐02 1.7E‐03 5.9E‐01 161
Il2ra Il2rg Il4 Il4ra Il6st Irf1 Isgf3g JAK1 JAK2 Jun Junb Mmp3 Mpl Myc Nfkb1 Nos2 Nr3c1 Oas1a Osm Pdgfra Pias1 Prl Prlr Ptpn1 Ptprc Pzp Saa3 Sfpi1 Sh2b1 Sit1 Sla2 Smad1 Smad2 Smad3 Smad4 Interleukin 2 receptor, alpha chain Interleukin 2 receptor, gamma chain Interleukin 4 Interleukin 4 receptor, alpha Interleukin 6 signal transducer Interferon regulatory factor 1 Interferon dependent positive acting transcription factor 3 gamma Janus kinase 1 Janus kinase 2 Jun oncogene Jun‐B oncogene Matrix metallopeptidase 3 Myeloproliferative leukemia virus oncogene Myelocytomatosis oncogene Nuclear factor of kappa light chain gene enhancer in B‐cells 1, p105 Nitric oxide synthase 2, inducible, macrophage Nuclear receptor subfamily 3, group C, member 1 2'‐5' oligoadenylate synthetase 1A Oncostatin M Platelet derived growth factor receptor, alpha polypeptide Protein inhibitor of activated STAT 1 Prolactin Prolactin receptor Protein tyrosine phosphatase, non‐receptor type 1 Protein tyrosine phosphatase, receptor type, C Pregnancy zone protein Serum amyloid A 3 SFFV proviral integration 1 SH2B adaptor protein 1 Suppression inducing transmembrane adaptor 1 Src‐like‐adaptor 2 MAD homolog 1 (Drosophila) MAD homolog 2 (Drosophila) MAD homolog 3 (Drosophila) MAD homolog 4 (Drosophila) 1.90 ‐1.66 ‐1.10 4.11 9.69 8.77 8.4E‐02 6.4E‐02 8.4E‐01 2.1E‐02 6.8E‐03 1.5E‐02 ‐1.03 3.22 ‐1.56 1.99 5.55 ‐1.10 1.07 2.93 9.5E‐01 2.1E‐02 6.0E‐02 2.8E‐02 5.4E‐02 8.3E‐01 8.4E‐01 1.6E‐02 ‐1.50 ‐11.40 2.0E‐01 1.3E‐04 ‐1.96 1.61 1.97 9.4E‐02 4.3E‐01 1.1E‐01 ‐3.36 ‐1.59 ‐3.15 25.27 1.4E‐02 1.2E‐01 3.4E‐02 1.5E‐02 ‐1.59 ‐1.44 2.85 ‐1.10 ‐3.21 ‐1.75 1.1E‐01 4.0E‐01 6.0E‐01 8.2E‐01 1.6E‐02 2.3E‐02 ‐1.66 ‐1.10 1.02 1.44 ‐9.99 ‐3.47 1.5E‐01 8.1E‐01 9.5E‐01 1.2E‐01 1.1E‐04 1.3E‐02 162
Smad5 Socs1 Socs2 Socs3 Socs4 Socs5 Sp1 Src Stam Stat1 Stat2 Stat3 Stat4 Stat5a Stat5b STAT6 Stub1 Tyk2 Usf1 Yy1 MAD homolog 5 (Drosophila) Suppressor of cytokine signaling 1 Suppressor of cytokine signaling 2 Suppressor of cytokine signaling 3 Suppressor of cytokine signaling 4 Suppressor of cytokine signaling 5 Trans‐acting transcription factor 1 Rous sarcoma oncogene Signal transducing adaptor molecule (SH3 domain and ITAM motif) 1 Signal transducer and activator of transcription 1 Signal transducer and activator of transcription 2 Signal transducer and activator of transcription 3 Signal transducer and activator of transcription 4 Signal transducer and activator of transcription 5A Signal transducer and activator of transcription 5B Signal transducer and activator of transcription 6 STIP1 homology and U‐Box containing protein 1 Tyrosine kinase 2 Upstream transcription factor 1 YY1 transcription factor ‐4.76 2.51 2.13 2.36 ‐1.14 2.42 ‐3.83 ‐1.34 6.0E‐03 2.0E‐02 2.2E‐02 3.3E‐01 3.9E‐01 3.8E‐02 6.2E‐02 3.3E‐01 1.34 1.9E‐01 1.92 2.9E‐02 1.56 3.3E‐01 6.79 7.4E‐03 ‐1.37 6.0E‐01 ‐1.20 8.9E‐01 1.24 7.9E‐01 17.78 1.08 1.15 ‐1.93 ‐1.46 1.4E‐04 8.5E‐01 5.4E‐01 2.1E‐02 1.1E‐01 CHAPTER 4: GENERAL DISCUSSION
Candidate PNS specific transcription factors
We have identified over two hundred regulators of transcription whose expression
and/or activity is enriched in PNS neurons. Although each technique has its strengths
and limitations, it is a reasonable assumption that any key regulator of transcription
would be identified by more than one approach. Using this criterion we have compiled a
list of 33 transcriptional regulators that are strong candidates for follow-up experiments
(Table 9). This list includes a handful of factors known to play crucial roles in
regeneration such as c-Jun, several factors for which some evidence exists but are not
necessarily “RAG”s, and many novel regulators that we predict may a role in the ability
of PNS neurons to regenerate in situations where CNS neurons cannot. The following
sections outline some of the most promising candidates from this list apart from STAT3.
c-Jun
One of several factors that homo- and/or hetero-dimerize to form the AP-1
transcription factor, c-Jun has been the object of much study in the field of spinal cord
regeneration. c-Jun is expressed in the nuclei of peripheral nerves after axotomy
(Herdegen et al., 1991) and is generally associated with regenerating neurons.
Interestingly, an injury to the peripheral, but not central, branch of dorsal root ganglion
neurons results in c-Jun expression (Broude et al., 1997). Genetic deletion of c-Jun
results in reduced target innervation and functional recovery after a facial nerve
transection, consistent with a loss of Cd44, galanin and α7β1 integrin expression (Raivich
et al., 2004). c-Jun dimers can also bind the promoter of SPRR1A, suggesting that c-Jun
163
164
may indeed by a regenerative “master switch” (Pradervand et al., 2004). c-Jun is also a
well known binding partner of STAT3 and the two transcription factors cooperate in the
induction of several genes, including IL-6 induced expression of α2-macroglobulin (Ito
et al., 1989) and Cntf induced expression of vasoactive intestinal peptide (Lewis et al.,
1994).
Although our findings are consistent with these reports, we observed an
enrichment of c-Jun in uninjured DRG neurons in both the microarray and qPCR results.
This suggests that although c-Jun levels are elevated after injury in the PNS, basal levels
are still much higher than in CNS neurons. An alternative explanation is that c-Jun
mRNA is constitutively upregulated in DRG neurons, but is only translated into protein
in the event of an injury. Regardless of the mechanism by which c-Jun is expressed, it is
the clearest example of a PNS enriched transcription factor in this study, having been
identified in each and every analysis.
RelA
The nuclear factor-κB family of transcription factors consists of 5 structurally
related proteins which form homo- and heterodimers that influence gene expression
(Baldwin 1996). RelA (also known as p65) forms one half of the p50/p65 heterodimer
which comprises the so-called “classical” pathway, thought traditionally to play a role in
immunity, stress response, cell survival and proliferation. The Fas apoptosis inhibitory
molecule (Faim), known to activate the classical pathway, increases outgrowth of PC-12
cells and cortical neurons in vitro (Sole et al., 2004). Inhibition of the NF-κB pathway
through over-expression of the super-repressor IκBα or knockout of p65 abolishes this
effect. Gutierrez and colleagues (2005) showed that inhibition of NF-κB through various
165
means reduced the length and arborization of both murine sensory nodose neurons in
vitro and pyramidal neurons in slice cultures. It is interesting to note that this effect was
age-dependent, peaking between E18 and P1. Targets of NF-κB include the neural cell
adhesion molecule Ncam (Simpson et al., 2000), β1 integrin (Wang et al., 2003) and BclXL (Chen et al., 2000), consistent with the hypothesis that it plays a beneficial role in
growth and survival.
There is little evidence in the literature to suggest that RelA is expressed at higher
levels in PNS than CNS neurons. Nevertheless, RelA was identified in the subtraction as
well as in the promoter and network analyses. It is possible, however, that this reflects
the presence of glia in subtraction DRG cultures, a hypothesis supported by preliminary
NF-kB pathway profiling in purified uninjured DRG neurons (RP Smith, unpublished
observations). An equally plausible explanation is that RelA is upregulated in DRG
neurons only after injury.
Klf6
A member of the Kruppel-like family of zinc-finger transcription factors, Klf6 is a
relatively uncharacterized molecule, with only 16 known targets. A recent report by
Veldman and colleagues (2007) used morpholinos to show that Klf6 and paralog Klf7 are
necessary for the successful regeneration of retinal ganglion cells in zebrafish after an
optic nerve crush.
We identified Klf6 as one of only 9 transcription factors present in
both the subtraction and microarray experiments, suggesting that it is constitutively
expressed in DRG neurons. Perhaps due to a lack of knowledge regarding its binding
sites and interactions, Klf6 was not identified in either the promoter or network analyses.
166
Irf1
Interferon regulator factor 1 was first characterized as a viral induced
transcriptional activator binding to upstream cis elements in the promoters of interferon
alpha and beta (Miyamoto et al., 1988). Irf1 is poorly characterized in neurons, with a
handful of reports observing transient elevations in its mRNA in bulk tissue after cerebral
ischemia (Paschen et al., 1998). Irf1 is one of the major downstream targets of the Stat
family of transcription factors and is transcriptionally regulated by STAT3 (Harroch et
al., 1994). We first identified Irf1 in the subtraction, and then again in the promoter and
network analyses. Irf1 was also part of the JAK/STAT qPCR pathway profile, and was
found to be almost 9 fold upregulated in DRG neurons.
Myc
Myc, short for “myelocytomatosis viral oncogene homolog”, is a basic helix loop
helix (bHLH) transcription factor that regulates the expression of 15% of all genes
(Gearheart et al., 2007). In most contexts expression of Myc is associated with
proliferation and growth, and its misregulation is associated with several naturally
forming tumors (Hoffman et al., 1998). We identified Myc in the subtraction and
promoter analyses, suggesting a PNS specific role. Expression of Myc is rarely seen in
differentiated cells, making its presence in DRGs intriguing. Myc is a target of both Irf1
(Dror et al., 2006) and STAT3 (Kiuchi et al., 1999), and its mRNA was found to be 3
fold upregulated in DRG neurons compared to CGNs by JAK/STAT qPCR pathway
profiling.
167
Hif1a
Hypoxia-inducible factor 1a is a bHLH transcription factor found to be
upregulated in conditions of low oxygen (Wang et al., 1995). Disruption of blood vessels
during spinal cord injury invariably leads to conditions of hypoxia, so it is perhaps not
surprising that Hif1a is upregulated in spinal motor neurons after injury (Skold et al.,
2004). Hypoxic conditions during development lead to aberrant axonal pathfinding in
C.Elegans, which can be rescued by the genetic abalation of Hif1, supporting the
hypothesis that Hif1 can influence neurite outgrowth (Pocock et al., 2008). We identified
Hif1a in the subtraction, promoter and network analyses, supporting a PNS specific role
for this factor in postnatal mice.
STAT6
Signal transducer and activator of transcription 6 is the last member of the Stat
family to be identified, and was previously known as the interleukin-4 induced
transcription factor (Quelle et al., 1995). STAT6 knockouts are viable and exhibit defects
in T-helper 2 responses and Il-4/Il-13 signaling (Igaz et al., 2001). Few reports link
STAT6 to neuronal function, and to our knowledge there is absolutely no evidence in the
literature suggesting that STAT6 is expressed in DRG neurons. We identified STAT6
mRNA in the microarray as being 3 fold enriched in laser captured DRG neurons. This
result was confirmed by qPCR, which showed an 18 fold enrichment, as well as by the
promoter analysis of the subtraction genes. It is not known whether STAT3 and STAT6
interact, but this is an intriguing possibility that would be worth exploring.
168
STAT3 in DRG neurons: survival factor or growth checkpoint?
We have shown several lines of evidence supporting the hypothesis that the
transcription factor STAT3 is preferentially expressed and active in PNS compared to
CNS neurons. Furthermore, we have demonstrated by Western blot and
immunohistochemistry that total and Y705 phosphorylated STAT3 are present in DRG
neurons constitutively, and not dependent on a peripheral injury. Loss of function studies
in which STAT3 was genetically ablated from sensory nodose and spinal motor axons
suggests that it plays a neuroprotective role during development and after a peripheral
injury. However, we have shown that the overexpression of constitutively active STAT3
leads to a 20% increase in CNS neurite outgrowth, which cannot be easily attributed to an
increase in survival because a similar increase was not seen in the percentage of neuron
with neurites (data not shown). While it is possible that overexpressed STAT3 plays a
different role in CNS neurons than resting levels in DRG neurons, the questions remains
whether STAT3 is merely a survival factor or also promotes axonal growth.
Perhaps due to its role as a proto-oncogene, STAT3 has attracted much interest
from the scientific community and is the topic of more than 1,300 publications. STAT3
has been shown to transcriptionally activate 117 downstream targets (Metacore
interaction database: http://www.genego.com), several of which play key roles in
neuroprotection and survival. One of those targets, BIRC5 (also known as Survivin) acts
as an inhibitor of apoptosis expressed in many human cancers (Gritsko et al., 2006).
BIRC5 was originally found to be upregulated in proliferating cells during G2/M phase,
binding to microtubules in the mitotic spindle (Li et al., 1998). Subsequent studies have
shown that it can inhibit the activity of the pro-apoptotic proteins BAX, Caspase-3 and
169
Caspase-7 (Tamm et al., 1998). Although few reports have characterized BIRC5
expression in the nervous system, there is evidence that it is expressed in a subset of
neurons in the hippocampus after traumatic brain injury (Johnson et al., 2005).
Additionally, BIRC5 positive neurons exhibit substantially lower levels of activated
caspase and DNA fragmentation than BIRC5 negative neurons, suggesting that it may
attenuate secondary damage. A second survival target of STAT3 is the serine-threonine
kinase AKT1 (Yu et al., 2005), an end effector of the phosphoinositide 3-kinase (PI3K)
pathway. Stimulation of the PI3K cascade leads to AKT1 membrane recruitment and
activation, leading to phosphorylation of the pro-apoptotic factors Bad, Caspase-9 and
FKHRL1, decreasing their activity and promoting cell survival (Datta et al., 1999).
AKT1 can rescue CGNs transfected with Bad from apoptosis, but only when
phosphorylated on serine 136 (Datta et al., 1997). Furthermore, overexpression of AKT1
in hypoglossal motor neurons leads to a 20% increase in survival after axotomy,
paralleled by a similar increase in regenerating neurons (Namikawa et al., 2000).
AKT1 is representative of several genes that are linked both to survival and axon
regeneration. The B-cell lymphoma 2 (Bcl-2) family consists of 25 proteins expressed on
the mitochondrial membrane which regulate permeabilization and play key roles in the
initiation and prevention of apoptosis (Chipuk et al., 2008). STAT3 transcriptionally
activates two anti-apoptotic members of the Bcl-2 family, namely Bcl-2 itself (Stephanou
et al, 2000) and Bcl-2 like 1, also known as Bcl-XL (Catlett-Falcone et al., 1999). Bcl-2
is expressed throughout the CNS during development and is downregulated postnatally,
reaching a baseline level at P4 (Cho et al., 2004). Conversely, expression of Bcl-2 is
maintained in the PNS throughout adulthood (Merry et al., 1999), possibly as a result of
170
STAT3 expression. Genetic deletion of Bcl-2 results in an 80% reduction in embyronic
mouse retinal axons invading an optic tectum slice in culture (Chen et al., 1997).
Transgenic mice overexpressing Bcl-2 exhibit a 10 fold increase in tectal slice
penetration, regardless of the age of the transgenic mice (Chen et al., 1997). Chierzi and
colleagues (1999) observed 100% RGC survival in Bcl-2 transgenics after an optic nerve
crush compared to 44% in wildtype controls. However, only after the administration of
of IN-1 were transgenic retinal axons able to penetrate the lesion site, albeit it in limited
numbers. Furthermore, Bcl-2 expression in purified P8 RGCs results in increased
survival but little neurite outgrowth without the addition of neurotrophic factors
(Goldberg et al., 2002b). Together these data suggest that although Bcl-2 is necessary for
survival during development and after injury in the adult, it is not sufficient to promote
neurite outgrowth. Bcl-XL likewise plays an important role in neuronal survival in the
developing nervous system, its genetic ablation resulting in massive apoptosis in the
mouse brain, spinal cord and DRGs at E13 (Motoyama et al., 1995). Retroviral delivery
of Bcl-XL to retinal ganglion cells results in a six fold increase in survival that persists up
to 8 weeks after a transection of the optic nerve (Malik et al., 2005). Although Bcl-XL
overexpression leads to increased RGC counts and length in vitro, like Bcl-2 it is
insufficient to facilitate regeneration in vivo (Kretz et al, 2004).
That STAT3 can directly regulate Survivin, AKT1 and Bcl-2 suggests that it plays
an important role in neuronal survival. However, STAT3 also regulates several genes
implicated in neurite initiation and elongation. Chief among these is the small proline
rich protein 1a (SPRR1A), which is activated in response to IL-6 and GP130 (Pradervand
et al., 2004). SPRR1A is upregulated in the PNS after a conditioning lesion and is
171
necessary for the outgrowth of DRG axons from explants in vitro (Bonilla et al., 2002).
Also induced by GP130 activation is Wnt5a (Fujio et al., 2004), which has recently been
shown to regulate axon specification through a pathway involving atypical PKC (Zhang
et al., 2007). A third possibility is the cell cycle regulator p21/Cip1/Waf1, which is
transcriptionally activated by STAT3 (Cocqueret et al., 2000), and when overexpressed in
hippocampal neurons can increase outgrowth threefold through the inhibition of Rhokinase (Tanaka et al., 2002). STAT3 also regulates the expression of several secreted
proteins implicated in axonal regeneration, including neuropeptide Y (Hakansson et al.,
1998), matrix metalloproteinase-9 (Song et al., 2008) and inducible nitric oxide synthase
(Yu et al., 2002).
Regardless of the targets of STAT3, it is clear from our data that its expression
and activation is constitutive, and not dependent on injury. Presumably any factor
directly regulating axonal outgrowth would be active during growth, and inactive once
the axon has found its target. STAT3 is upregulated after injury as much as 3 fold
(Schwaiger et al., 2000), and levels of Y705 phosphorylated STAT3 are significantly
elevated first in the distal axon and then the soma of the injured neuron (Lee et al., 2004).
That STAT3 can upregulate GP130 (O’Brien et al, 1997) suggests the potential for a
positive feedback loop that relies on basal levels of the inactive transcription factor. In
that sense STAT3 could serve as an early checkpoint in the cascade of events leading to
successful axon regeneration
Islet-1 is expressed extensively in DRG neurons and overlaps completely with the
STAT3 positive population. In contrast, the Y705 positive population comprises only
60% of the Islet-1 positive population (n=2). Although the Y705 positive fraction of
172
cells could represent a unique cell type, I could find no cellular marker that segregated
with this fraction. NF-M labeling only partially overlaps with the Y705 fraction, with
some cells expressing NF-M and Y705, and others only presenting one of the two
antigens. It is unlikely that the Y705 fraction represents the previously characterized
NF200, IB4 and CGRP positive populations, which each constitute less than 45% of the
total neuronal population in P7 rats (Yamamoto et al., 2008). One potential label that
may overlap with the Y705 positive fraction is the receptor tyrosine kinase Ret. Ret
mRNA is expressed in 60% of mouse L4/L5 DRG neurons (Molliver et al., 1997), is not
expressed exclusively in small or large caliber cells, and can phosphorylate STAT1 in
papillary cancer cells (Hwang et al., 2004).
The large range of Y705 STAT3 phosphorylation observed in DRG neurons is
difficult to account for. Is there a specific population of cells in which Y705
phosphorylation serves a particular function, or are there transient waves of
phosphorylation that function to preserve homeostasis? One intriguing possibility posited
recently is that adult DRGs contains a population of neural crest progenitor cells that can
differentiate into neurons after injury (Li et al., 2007). In this situation the Y705 positive
population would indicate ongoing renewal of the progenitor population.
Potentiating STAT3 activity in Cerebellar Granule Neurons
Through overexpression experiments we have shown two distinct effects of
STAT3 overexpression in cerebellar granule neurons. First, that overexpression of
STAT3 increases the formation of primary neurites regardless of its activity. Of the two
outcomes this is the hardest to interpret, because presumably the majority of STAT3’s
effects are exerted through the transcriptional regulation of target genes. However, it is
173
possible that STAT3 plays a more direct role on neurite formation in the cytoplasm, as it
has been shown to bind to the microtubule associated protein Stathmin (Ng et al., 2006).
The stathmin family of proteins, consisting of Stathmin, Sclip, Scg10 and Rb3, bind and
destabilize microtubules, leading to catastrophe if overexpressed (Cassimeris, 2002).
Stathmin is expressed ubiquitously, with constitutive expression in the nervous system
persisting through adulthood (Sobel et al., 1989). Whereas Scg10 is expressed in the
central region of advancing growth cones during development and after injury, Stathmin
is localized to neuronal cell bodies, which is consistent with the hypothesis that it could
play a role in primary neurite formation (Di Paolo et al., 1997). STAT3 binds directly to
the C-terminal region of Stathmin in PC12 cells and antagonizes its ability to
depolymerize microtubules (Ng et al., 2006). Furthermore, the same study showed that
STAT3 expression is required for the stabilization of microtubules and cell migration in
mouse embryonic fibroblasts. These data are consistent with our findings that although
STAT3 overexpression increases primary neurite formation, nuclear localization was
only observed in a subset of those cells (data not shown).
The second effect of STAT3 overexpression in CGNs, seen only when the
constitutively active form was used, was a 20% increase in total neurite outgrowth and a
similar increase in the length of the longest neurite. While this effect is clearly
significant, it is unlikely that it represents the maximum increase in CGN outgrowth that
can be expected on a growth promoting substrate such as PDL. Given that not all
transfected CGNs exhibit STAT3 expression in their nuclei, it is likely that cotransfection of additional factors could improve outgrowth even further. Table 8
provides some intriguing transcription factor candidates that could be screened in a high
174
content co-transfection assay. At least four of those factors stand out due to their known
association with STAT3, including CREB1, Foxo1, c-Jun and Myc. STAT3 contains
multiple CRE sites in its promoter (Smith RP, unpublished data), and CREB1 can
physically interact with STAT3 and increase its activity (Mynard et al., 2004). Forkhead
homolog in rhabdomyosarcoma (Fkhr), also known as Foxo1, can similarly increase the
transcription of STAT3 responsive genes when overexpressed in HepG2 cells
(Kortylewiski et al., 2003). c-Jun is a well characterized binding partner of STAT3, and
the two cooperate to drive the transcription of several genes including α2-macroglobulin,
vasoactive intestinal peptide, c-fos and matrix metalloproteinase (Shuai, 2000). Myc and
STAT3 have a more storied relationship, with both factors apparently able to regulate the
other, in addition to cooperating in the transcription of genes such as Cdc25a (Barré et al.,
2005).
Another strategy that may improve STAT3-C mediated neurite outgrowth would
be to improve nuclear localization of the active transcription factor. Because of the
relatively large size of STAT3 (~90kDa), it cannot passively enter the nucleus and must
first bind to α3 (Liu et al., 2005b), α5 or α7 importin (Ma et al., 2006b). The STAT3/αimportin complex then binds to β1-importin and which shuttles STAT3 into the nucleus.
Figure 29 shows the mean DRG and cerebellar microarray signal intensities for all seven
members of the importin family. Although there is little difference between α importin
signal intensities between PNS and CNS, β1 importin is more than 3 fold upregulated in
DRG neurons. This is consistent with a report showing that β1 importin mRNA is
present at high levels in DRG axons and is locally translated after a peripheral injury
(Hanz et al., 2003). The authors also report the presence of α1, α4 and α7 importins in
175
the axoplasm, and that after injury, axoplasmic transcription factors can form a retrograde
complex with α and β1 importins that is transported by dynein to the soma. These data
suggest a model in which STAT3 protein, present in the axons of naïve DRG neurons, is
phosphorylated in response to injury and rapidly shuttled to the nucleus via an α7β1
importin complex. In addition to explaining the role of STAT3 after a peripheral injury,
this model could also explain impaired nuclear import of STAT3 in transfected CGNs.
176
Chapter Figures
Figure 29. Expression of importins in DRG neurons and cerebellum. Average
microarray signal intensities for all 6 members of the α-importin family as well as β1
importin. Scale bars represented the standard error of the mean (n ≥3, *** = p<0.001).
177
Chapter Tables
Table 9. Candidate DRG enriched transcription factors. Summary of Tables 2-8,
showing potential DRG enriched transcription factors identified by at least two different
techniques. “SUB” = found in the subtraction. “MA” = found in the microarray.
“PS/PM” = found in the promoter analysis of the subtraction (S) or microarray (M)
genes. “NS/NM” = found in the interaction network analysis of the subtraction (S) or
microarray (M) genes. “RT” = found to be significantly upregulated by real-time PCR in
DRGs (X) or cerebellum (X*).
Transcription Factor Signal transducer and activator of transcription 3 CCAAT/enhancer binding protein zeta Breast cancer 1, early onset Interferon regulator factor 1 Jun oncogene Myelocytomatosis oncogene Signal transducer and activator of transcription 6 Upstream transcription factor 1 Activating transcription factor 3 Basic helix‐loop‐helix domain containing, class B2 Hypoxia inducible factor 1, alpha subunit V‐rel reticuloendotheliosis viral oncogene homolog (avian) V‐rel reticuloendotheliosis viral oncogene homolog A (avian) SFFV proviral integration 1 (PU.1) Signal transducer and activator of transcription 1 Upstream transcription factor 2 Symbol SUB MA PS PM NS NM RT
Stat3 X X X X X X X Cebpz Brca1 Irf1 Jun Myc X X X X X X X X X X X X X X X X X X X X X X STAT6 X X X X Usf1 X X X X* Atf3 X X X Bhlhb2 X X X Hif1a X X X Rel X X X Rela X X X Sfpi1 X X X Stat1 X X X Usf2 X X X 178
Glucocorticoid receptor Steroidogenic factor 1 GATA binding protein 3 Transcription factor AP4 RIKEN cDNA 1300018I05 gene chromodomain helicase DNA binding protein 4 early B‐cell factor 1 forkhead box O1 hepatoma‐derived growth factor, related protein 2 IKAROS family zinc finger 1 Kruppel‐like factor 6 Myeloid zinc finger 1 PC4 and SFRS1 interacting protein 1 RD RNA‐binding protein replication factor C (activator 1) 1 ring finger protein 38 YY1 transcription factor Nr3c1 Nr5a1 Gata3 Tcfap4 1300018I05Rik
X X X X X X X X X X X X X* Chd4 Ebf1 Foxo1 X X X X X X Hdgfrp2 Ikzf1 Klf6 Mzf1 X X X X X X X X Psip1 Rdbp X X X X Rfc1 Rnf38 Yy1 X X X X X X CHAPTER 5: SUMMARY AND FUTURE DIRECTIONS
This thesis details a systems-based discovery effort to identify transcription
factors specifically enriched and active in the peripheral nervous system. To accomplish
this goal, gene lists from both a subtractive hybridization and microarray experiment
were analyzed, whereupon 65 transcription factors were found to be enriched in DRG
neurons. Two separate approaches were then employed to identify factors that may be
preferentially active within PNS neurons. For the first approach, we hypothesized that if
a transcription factor was preferentially active within DRG neurons, it would be expected
to bind the promoters of DRG enriched genes disproportionately. Using a custom
software package that I developed named Promoylzer, 113 transcription factor binding
consensuses were found to be overrepresented within DRG enriched promoters. The
second approach relied on the assumption that a transcription factor specific to DRG
neurons would be expected to interact with a disproportionate number of DRG enriched
genes. Using the commercial software package Metacore, 104 transcription factors were
found to interact significantly with genes from both the subtraction and microarray.
Although a result from any one approach could be called into question as a false
positive, the identification of a transcription factor in more than one analysis increases the
confidence with which one could predict its enrichment in the PNS. By combining the
results of the subtraction, microarray, promoter and interaction network analyses, I was
able to identify 34 “high confidence” transcription factors that were identified by more
than one approach. Interestingly, this list includes several factors that were neither
179
180
identified by the subtraction nor microarray, suggesting preferential activity in the PNS
but not enrichment.
Two factors stood out in this analysis, signal transducer and activator of
transcription 3 (STAT3) and CCAAT/enhancer binding protein zeta (C/EBPζ), which
were identified by all of the approaches used. STAT3 was chosen for further study as a
proof-of-concept for the predictions made in silico. The enrichment of STAT3 mRNA
and protein in DRG neurons was confirmed both in vitro and in vivo, along with several
upstream regulators and downstream targets of the transcription factor. Furthermore, the
preferential activity of STAT3 in DRG neurons was observed both in vitro and in vivo,
using an antibody that specifically recognizes tyrosine 705 phosphorylated STAT3.
To elucidate the potential functions of STAT3 in DRG neurons, exogenous
STAT3 and STAT3 activation mutants were transfected into cerebellar granule neurons
in an in vitro model. Delivery of STAT3, regardless of its activity, resulted in an increase
in primary neurite formation, suggesting a potential role in process initiation.
Transfection of constitutively active STAT3 resulted in increased neurite outgrowth on
both PDL and laminin. These data suggest that STAT3 could potentially play a role in
neurite outgrowth in vivo.
The results of this thesis suggest several potential directions that could be
followed in future studies and are outlined as follows:
1) STAT3 served as a useful proof of concept to test in silico predictions. The 33
other transcription factors listed in Table 9 could be followed up on in a similar
approach. Because this would be a large number of factors to test individually,
the preferred approach would be transfect primary neurons in high-throughput
181
using techniques outlined in this thesis and commonly used within the Lemmon
Lab. Preferable still would be to acquire or generate cDNAs for each
transcription factor in a lentiviral vector. This approach would allow for more
straightforward delivery of the transgene and make the combination of multiple
factors more feasible.
2) It is likely that the delivery of any single transcription factor would be insufficient
to promote robust regeneration of axons in vivo. The ultimate aim, therefore,
should be simultaneously deliver multiple factors to CNS neurons in the hope that
this would either (a) reboot developmental growth pathways or (b) convert an
adult CNS neuron into adult PNS neuron. This general approach has been used
successfully by several groups recently, exemplified by Takahashi and colleagues
(2006), to generate pluripotent stem cells from terminally differentiated cells.
3) Several of the transcription factors identified in Table 9 have been shown to
interact with STAT3, including c-Jun, Myc and Irf1. One potential approach
would be to attempt to potentiate the effect of STAT3 by co-transfecting one or
more of these factors.
4) Exogenous STAT3 does not accumulate in the nucleus of CGNs to the same
degree that it does in DRGs, suggesting an impairment in nuclear import or
increased nuclear export. One potential approach would be to co-transfect β1importin along with STAT3 to increase its activity.
5) Conditional ablation of STAT3 leads to apoptosis of DRG neurons during
development. This has led many to assume that STAT3 functions as a survival
factor. However, the effect of JAK2/STAT3 inhibition in the conditioning lesion
182
paradigm suggests a role in neurite outgrowth. It would be interesting to examine
the effect of a conditioning lesion in STAT3 conditional knockouts. Another,
possibly simpler, experiment would be to inhibit STAT3 activation in DRG
neurons in vitro and examine the effects on neurite outgrowth. I attempted to do
this with the pharmacological inhibitor Stattic (Tocris Bioscience #2798),
however this compound proved to be toxic to both DRGs and CGNs at the same
concentration. Other possibilities include an inhibitor peptide (Calbiochem
#573096), the JAK2 inhibitor AG490 (Calbiochem #658408), Cucurbitacin I
(Calbiochem #238590) and Withacnistin (NCI NSC 135075). One final approach
might be to transfect DRG neurons with a protein inhibitor of STAT3, such as
PIAS3 (Entrez Gene: 229615).
WORKS CITED
Aaronson, D. S. and C. M. Horvath (2002). "A road map for those who don't know JAKSTAT." Science 296(5573): 1653-5.
Aguayo, A. J., S. David et al. (1981). "Influences of the glial environment on the
elongation of axons after injury: transplantation studies in adult rodents." J Exp
Biol 95: 231-40.
Aigner, L. and P. Caroni (1995). "Absence of persistent spreading, branching, and
adhesion in GAP-43-depleted growth cones." J Cell Biol 128(4): 647-60.
Aimone, J. B., J. L. Leasure et al. (2004). "Spatial and temporal gene expression profiling
of the contused rat spinal cord." Exp Neurol 189(2): 204-21.
Alonzi, T., G. Middleton et al. (2001). "Role of STAT3 and PI 3-kinase/Akt in mediating
the survival actions of cytokines on sensory neurons." Mol Cell Neurosci
18(3): 270-82.
Altschul, S. F., W. Gish et al. (1990). "Basic local alignment search tool." J Mol Biol
215(3): 403-10.
Araki, T. and J. Milbrandt (1996). "Ninjurin, a novel adhesion molecule, is induced by
nerve injury and promotes axonal growth." Neuron 17(2): 353-61.
Arlotta, P., B. J. Molyneaux et al. (2005). "Neuronal subtype-specific genes that control
corticospinal motor neuron development in vivo." Neuron 45(2): 207-21.
Arlotta, P., B. J. Molyneaux et al. (2008). "Ctip2 controls the differentiation of medium
spiny neurons and the establishment of the cellular architecture of the
striatum." J Neurosci 28(3): 622-32.
Baldwin, A. S., Jr. (1996). "The NF-kappa B and I kappa B proteins: new discoveries and
insights." Annu Rev Immunol 14: 649-83.
Banner, L. R. and P. H. Patterson (1994). "Major changes in the expression of the
mRNAs for cholinergic differentiation factor/leukemia inhibitory factor and its
receptor after injury to adult peripheral nerves and ganglia." Proc Natl Acad
Sci U S A 91(15): 7109-13.
Barre, B., A. Vigneron et al. (2005). "The STAT3 transcription factor is a target for the
Myc and riboblastoma proteins on the Cdc25A promoter." J Biol Chem
280(16): 15673-81.
183
184
Bartsch, U., C. E. Bandtlow et al. (1995). "Lack of evidence that myelin-associated
glycoprotein is a major inhibitor of axonal regeneration in the CNS." Neuron
15(6): 1375-81.
Benson, M. D., M. I. Romero et al. (2005). "Ephrin-B3 is a myelin-based inhibitor of
neurite outgrowth." Proc Natl Acad Sci U S A 102(30): 10694-9.
Berry, M., S. Hall et al. (1992). "Regeneration of axons in the optic nerve of the adult
Browman-Wyse (BW) mutant rat." J Neurocytol 21(6): 426-48.
Bolin, L. M., A. N. Verity et al. (1995). "Interleukin-6 production by Schwann cells and
induction in sciatic nerve injury." J Neurochem 64(2): 850-8.
Bolstad, B. M., R. A. Irizarry et al. (2003). "A comparison of normalization methods for
high density oligonucleotide array data based on variance and bias."
Bioinformatics 19(2): 185-93.
Bomze, H. M., K. R. Bulsara et al. (2001). "Spinal axon regeneration evoked by
replacing two growth cone proteins in adult neurons." Nat Neurosci 4(1): 3843.
Bonaldo, M. F., G. Lennon et al. (1996). "Normalization and subtraction: two approaches
to facilitate gene discovery." Genome Res 6(9): 791-806.
Bonilla, I. E., K. Tanabe et al. (2002). "Small proline-rich repeat protein 1A is expressed
by axotomized neurons and promotes axonal outgrowth." J Neurosci 22(4):
1303-15.
Bradbury, E. J., L. D. Moon et al. (2002). "Chondroitinase ABC promotes functional
recovery after spinal cord injury." Nature 416(6881): 636-40.
Breasted, J. H., Ed. (1922). The Edwin Smith Surgical Papyrus: Hieroglyphic
Transliteration, Translation and Commentary, University of Chicago Press.
Bromberg, J. F., M. H. Wrzeszczynska et al. (1999). "Stat3 as an oncogene." Cell 98(3):
295-303.
Broude, E., M. McAtee et al. (1997). "c-Jun expression in adult rat dorsal root ganglion
neurons: differential response after central or peripheral axotomy." Exp Neurol
148(1): 367-77.
Cafferty, W. B., N. J. Gardiner et al. (2004). "Conditioning injury-induced spinal axon
regeneration fails in interleukin-6 knock-out mice." J Neurosci 24(18): 443243.
185
Cafferty, W. B., N. J. Gardiner et al. (2001). "Leukemia inhibitory factor determines the
growth status of injured adult sensory neurons." J Neurosci 21(18): 7161-70.
Cai, D., K. Deng et al. (2002). "Arginase I and polyamines act downstream from cyclic
AMP in overcoming inhibition of axonal growth MAG and myelin in vitro."
Neuron 35(4): 711-9.
Cai, D., Y. Shen et al. (1999). "Prior exposure to neurotrophins blocks inhibition of
axonal regeneration by MAG and myelin via a cAMP-dependent mechanism."
Neuron 22(1): 89-101.
Cajal, S. R. y. (1928). Degeneration and Regeneration of the Nervous System. London,
Oxford University Press.
Campbell, G., P. N. Anderson et al. (1991). "GAP-43 in the axons of mammalian CNS
neurons regenerating into peripheral nerve grafts." Exp Brain Res 87(1): 67-74.
Cao, W., C. Epstein et al. (2004). "Comparing gene discovery from Affymetrix
GeneChip microarrays and Clontech PCR-select cDNA subtraction: a case
study." BMC Genomics 5(1): 26.
Cao, Z., Y. Gao et al. (2006). "The cytokine interleukin-6 is sufficient but not necessary
to mimic the peripheral conditioning lesion effect on axonal growth." J
Neurosci 26(20): 5565-73.
Carmel, J. B., A. Galante et al. (2001). "Gene expression profiling of acute spinal cord
injury reveals spreading inflammatory signals and neuron loss." Physiol
Genomics 7(2): 201-13.
Carninci, P., T. Kasukawa et al. (2005). "The transcriptional landscape of the mammalian
genome." Science 309(5740): 1559-63.
Caroni, P. (1997). "Intrinsic neuronal determinants that promote axonal sprouting and
elongation." Bioessays 19(9): 767-75.
Caroni, P. and M. E. Schwab (1988). "Antibody against myelin-associated inhibitor of
neurite growth neutralizes nonpermissive substrate properties of CNS white
matter." Neuron 1(1): 85-96.
Caroni, P. and M. E. Schwab (1988). "Two membrane protein fractions from rat central
myelin with inhibitory properties for neurite growth and fibroblast spreading."
J Cell Biol 106(4): 1281-8.
Cassimeris, L. (2002). "The oncoprotein 18/stathmin family of microtubule
destabilizers." Curr Opin Cell Biol 14(1): 18-24.
186
Castillo-Davis, C. I. and D. L. Hartl (2003). "GeneMerge--post-genomic analysis, data
mining, and hypothesis testing." Bioinformatics 19(7): 891-2.
Catlett-Falcone, R., T. H. Landowski et al. (1999). "Constitutive activation of Stat3
signaling confers resistance to apoptosis in human U266 myeloma cells."
Immunity 10(1): 105-15.
Center, N. S. C. I. S. (2008). Spinal Cord Injury: Facts and Figures at a Glance,
University of Alabama, Birmingham.
Chao, M. V. (2003). "Neurotrophins and their receptors: a convergence point for many
signalling pathways." Nat Rev Neurosci 4(4): 299-309.
Chen, D. F., S. Jhaveri et al. (1995). "Intrinsic changes in developing retinal neurons
result in regenerative failure of their axons." Proc Natl Acad Sci U S A 92(16):
7287-91.
Chen, D. F., G. E. Schneider et al. (1997). "Bcl-2 promotes regeneration of severed axons
in mammalian CNS." Nature 385(6615): 434-9.
Chen, M. S., A. B. Huber et al. (2000). "Nogo-A is a myelin-associated neurite outgrowth
inhibitor and an antigen for monoclonal antibody IN-1." Nature 403(6768):
434-9.
Chen, Z. and Z. C. Han (2008). "STAT3: a critical transcription activator in
angiogenesis." Med Res Rev 28(2): 185-200.
Chen, Z. L., W. M. Yu et al. (2007). "Peripheral regeneration." Annu Rev Neurosci 30:
209-33.
Chierzi, S., E. Strettoi et al. (1999). "Optic nerve crush: axonal responses in wild-type
and bcl-2 transgenic mice." J Neurosci 19(19): 8367-76.
Chipuk, J. E. and D. R. Green (2008). "How do BCL-2 proteins induce mitochondrial
outer membrane permeabilization?" Trends Cell Biol 18(4): 157-64.
Cho, K. S., L. Yang et al. (2005). "Re-establishing the regenerative potential of central
nervous system axons in postnatal mice." J Cell Sci 118(Pt 5): 863-72.
Chung, C. D., J. Liao et al. (1997). "Specific inhibition of Stat3 signal transduction by
PIAS3." Science 278(5344): 1803-5.
Cohen, J., J. F. Burne et al. (1986). "Retinal ganglion cells lose response to laminin with
maturation." Nature 322(6078): 465-7.
187
Condic, M. L. (2001). "Adult neuronal regeneration induced by transgenic integrin
expression." J Neurosci 21(13): 4782-8.
Coqueret, O. and H. Gascan (2000). "Functional interaction of STAT3 transcription
factor with the cell cycle inhibitor p21WAF1/CIP1/SDI1." J Biol Chem
275(25): 18794-800.
Costigan, M., K. Befort et al. (2002). "Replicate high-density rat genome oligonucleotide
microarrays reveal hundreds of regulated genes in the dorsal root ganglion after
peripheral nerve injury." BMC Neurosci 3: 16.
Costigan, M., R. J. Mannion et al. (1998). "Heat shock protein 27: developmental
regulation and expression after peripheral nerve injury." J Neurosci 18(15):
5891-900.
Crutcher, K. A. (1989). "Tissue sections from the mature rat brain and spinal cord as
substrates for neurite outgrowth in vitro: extensive growth on gray matter but
little growth on white matter." Exp Neurol 104(1): 39-54.
Cui, Q., K. S. Cho et al. (2004). "Synergistic effect of Nogo-neutralizing antibody IN-1
and ciliary neurotrophic factor on axonal regeneration in adult rodent visual
systems." J Neurotrauma 21(5): 617-25.
Daniloff, J. K., G. Levi et al. (1986). "Altered expression of neuronal cell adhesion
molecules induced by nerve injury and repair." J Cell Biol 103(3): 929-45.
Datta, S. R., A. Brunet et al. (1999). "Cellular survival: a play in three Akts." Genes Dev
13(22): 2905-27.
Datta, S. R., H. Dudek et al. (1997). "Akt phosphorylation of BAD couples survival
signals to the cell-intrinsic death machinery." Cell 91(2): 231-41.
Davies, S. J., M. T. Fitch et al. (1997). "Regeneration of adult axons in white matter
tracts of the central nervous system." Nature 390(6661): 680-3.
Davies, S. J., D. R. Goucher et al. (1999). "Robust regeneration of adult sensory axons in
degenerating white matter of the adult rat spinal cord." J Neurosci 19(14):
5810-22.
Dergham, P., B. Ellezam et al. (2002). "Rho signaling pathway targeted to promote spinal
cord repair." J Neurosci 22(15): 6570-7.
Di Paolo, G., R. Lutjens et al. (1997). "Differential distribution of stathmin and SCG10 in
developing neurons in culture." J Neurosci Res 50(6): 1000-9.
188
Diatchenko, L., Y. F. Lau et al. (1996). "Suppression subtractive hybridization: a method
for generating differentially regulated or tissue-specific cDNA probes and
libraries." Proc Natl Acad Sci U S A 93(12): 6025-30.
Domeniconi, M., Z. Cao et al. (2002). "Myelin-associated glycoprotein interacts with the
Nogo66 receptor to inhibit neurite outgrowth." Neuron 35(2): 283-90.
Dror, N., M. Alter-Koltunoff et al. (2007). "Identification of IRF-8 and IRF-1 target
genes in activated macrophages." Mol Immunol 44(4): 338-46.
Duncan, S. A., Z. Zhong et al. (1997). "STAT signaling is active during early mammalian
development." Dev Dyn 208(2): 190-8.
Dusart, I., M. S. Airaksinen et al. (1997). "Purkinje cell survival and axonal regeneration
are age dependent: an in vitro study." J Neurosci 17(10): 3710-26.
Ehret, G. B., P. Reichenbach et al. (2001). "DNA binding specificity of different STAT
proteins. Comparison of in vitro specificity with natural target sites." J Biol
Chem 276(9): 6675-88.
Ewing, B. and P. Green (1998). "Base-calling of automated sequencer traces using phred.
II. Error probabilities." Genome Res 8(3): 186-94.
Ewing, B., L. Hillier et al. (1998). "Base-calling of automated sequencer traces using
phred. I. Accuracy assessment." Genome Res 8(3): 175-85.
Fan, M., R. Mi et al. (2001). "Analysis of gene expression following sciatic nerve crush
and spinal cord hemisection in the mouse by microarray expression profiling."
Cell Mol Neurobiol 21(5): 497-508.
Faulkner, J. R., J. E. Herrmann et al. (2004). "Reactive astrocytes protect tissue and
preserve function after spinal cord injury." J Neurosci 24(9): 2143-55.
Fausto, N. (2000). "Liver regeneration." J Hepatol 32(1 Suppl): 19-31.
Filbin, M. T. (2003). "Myelin-associated inhibitors of axonal regeneration in the adult
mammalian CNS." Nat Rev Neurosci 4(9): 703-13.
Fournier, A. E., T. GrandPre et al. (2001). "Identification of a receptor mediating Nogo66 inhibition of axonal regeneration." Nature 409(6818): 341-6.
Fournier, A. E., B. T. Takizawa et al. (2003). "Rho kinase inhibition enhances axonal
regeneration in the injured CNS." J Neurosci 23(4): 1416-23.
189
Fujio, Y., T. Matsuda et al. (2004). "Signals through gp130 upregulate Wnt5a and
contribute to cell adhesion in cardiac myocytes." FEBS Lett 573(1-3): 202-6.
Gao, Y., K. Deng et al. (2004). "Activated CREB is sufficient to overcome inhibitors in
myelin and promote spinal axon regeneration in vivo." Neuron 44(4): 609-21.
Gardiner, N. J., W. B. Cafferty et al. (2002). "Expression of gp130 and leukaemia
inhibitory factor receptor subunits in adult rat sensory neurones: regulation by
nerve injury." J Neurochem 83(1): 100-9.
Gearhart, J., E. E. Pashos et al. (2007). "Pluripotency redux--advances in stem-cell
research." N Engl J Med 357(15): 1469-72.
Gillardon, F., L. Klimaschewski et al. (1996). "Expression pattern of candidate cell death
effector proteins Bax, Bcl-2, Bcl-X, and c-Jun in sensory and motor neurons
following sciatic nerve transection in the rat." Brain Res 739(1-2): 244-50.
Goldberg, J. L., J. S. Espinosa et al. (2002). "Retinal ganglion cells do not extend axons
by default: promotion by neurotrophic signaling and electrical activity."
Neuron 33(5): 689-702.
Goldberg, J. L., M. P. Klassen et al. (2002). "Amacrine-signaled loss of intrinsic axon
growth ability by retinal ganglion cells." Science 296(5574): 1860-4.
Goldshmit, Y., M. P. Galea et al. (2004). "Axonal regeneration and lack of astrocytic
gliosis in EphA4-deficient mice." J Neurosci 24(45): 10064-73.
GrandPre, T., S. Li et al. (2002). "Nogo-66 receptor antagonist peptide promotes axonal
regeneration." Nature 417(6888): 547-51.
Grimpe, B. and J. Silver (2004). "A novel DNA enzyme reduces glycosaminoglycan
chains in the glial scar and allows microtransplanted dorsal root ganglia axons
to regenerate beyond lesions in the spinal cord." J Neurosci 24(6): 1393-7.
Gris, P., S. Murphy et al. (2003). "Differential gene expression profiles in embryonic,
adult-injured and adult-uninjured rat spinal cords." Mol Cell Neurosci 24(3):
555-67.
Gritsko, T., A. Williams et al. (2006). "Persistent activation of stat3 signaling induces
survivin gene expression and confers resistance to apoptosis in human breast
cancer cells." Clin Cancer Res 12(1): 11-9.
Gutierrez, H., V. A. Hale et al. (2005). "NF-kappaB signalling regulates the growth of
neural processes in the developing PNS and CNS." Development 132(7): 171326.
190
Haas, C. A., H. D. Hofmann et al. (1999). "Expression of CNTF/LIF-receptor
components and activation of STAT3 signaling in axotomized facial
motoneurons: evidence for a sequential postlesional function of the cytokines."
J Neurobiol 41(4): 559-71.
Hakansson, M. L. and B. Meister (1998). "Transcription factor STAT3 in leptin target
neurons of the rat hypothalamus." Neuroendocrinology 68(6): 420-7.
Hanz, S., E. Perlson et al. (2003). "Axoplasmic importins enable retrograde injury
signaling in lesioned nerve." Neuron 40(6): 1095-104.
Hao, A., V. Novotny-Diermayr et al. (2005). "The LIM/homeodomain protein Islet1
recruits Janus tyrosine kinases and signal transducer and activator of
transcription 3 and stimulates their activities." Mol Biol Cell 16(4): 1569-83.
Harroch, S., M. Revel et al. (1994). "Induction by interleukin-6 of interferon regulatory
factor 1 (IRF-1) gene expression through the palindromic interferon response
element pIRE and cell type-dependent control of IRF-1 binding to DNA."
Embo J 13(8): 1942-9.
Hattori, R., N. Maulik et al. (2001). "Role of STAT3 in ischemic preconditioning." J Mol
Cell Cardiol 33(11): 1929-36.
Herdegen, T., W. Kummer et al. (1991). "Expression of c-JUN, JUN B and JUN D
proteins in rat nervous system following transection of vagus nerve and
cervical sympathetic trunk." Neuroscience 45(2): 413-22.
Higuchi, R., G. Dollinger et al. (1992). "Simultaneous amplification and detection of
specific DNA sequences." Biotechnology (N Y) 10(4): 413-7.
Hirota, H., H. Kiyama et al. (1996). "Accelerated Nerve Regeneration in Mice by
upregulated expression of interleukin (IL) 6 and IL-6 receptor after trauma." J
Exp Med 183(6): 2627-34.
Hochberg, Y. and Y. Benjamini (1990). "More powerful procedures for multiple
significance testing." Stat Med 9(7): 811-8.
Hoffman, B. and D. A. Liebermann (1998). "The proto-oncogene c-myc and apoptosis."
Oncogene 17(25): 3351-7.
Hokfelt, T., Z. Wiesenfeld-Hallin et al. (1987). "Increase of galanin-like
immunoreactivity in rat dorsal root ganglion cells after peripheral axotomy."
Neurosci Lett 83(3): 217-20.
191
Huang da, W., B. T. Sherman et al. (2007). "DAVID Bioinformatics Resources:
expanded annotation database and novel algorithms to better extract biology
from large gene lists." Nucleic Acids Res 35(Web Server issue): W169-75.
Hwang, E. S., D. W. Kim et al. (2004). "Regulation of signal transducer and activator of
transcription 1 (STAT1) and STAT1-dependent genes by RET/PTC
(rearranged in transformation/papillary thyroid carcinoma) oncogenic tyrosine
kinases." Mol Endocrinol 18(11): 2672-84.
Igaz, P., S. Toth et al. (2001). "Biological and clinical significance of the JAK-STAT
pathway; lessons from knockout mice." Inflamm Res 50(9): 435-41.
Irizarry, R. A., B. Hobbs et al. (2003). "Exploration, normalization, and summaries of
high density oligonucleotide array probe level data." Biostatistics 4(2): 249-64.
Ito, T., H. Tanahashi et al. (1989). "Nuclear factors interacting with an interleukin-6
responsive element of rat alpha 2-macroglobulin gene." Nucleic Acids Res
17(22): 9425-35.
Ji, R. R., Q. Zhang et al. (1995). "Prominent expression of bFGF in dorsal root ganglia
after axotomy." Eur J Neurosci 7(12): 2458-68.
Jiao, Y., Z. Sun et al. (1999). "A simple and sensitive antigen retrieval method for freefloating and slide-mounted tissue sections." J Neurosci Methods 93(2): 149-62.
Johnson, E. A., S. I. Svetlov et al. (2005). "Cell-specific DNA fragmentation may be
attenuated by a survivin-dependent mechanism after traumatic brain injury in
rats." Exp Brain Res 167(1): 17-26.
Johnson, P. W., W. Abramow-Newerly et al. (1989). "Recombinant myelin-associated
glycoprotein confers neural adhesion and neurite outgrowth function." Neuron
3(3): 377-85.
Jones, L. L., G. W. Kreutzberg et al. (1997). "Regulation of CD44 in the regenerating
mouse facial motor nucleus." Eur J Neurosci 9(9): 1854-63.
Kaptein, A., V. Paillard et al. (1996). "Dominant negative stat3 mutant inhibits
interleukin-6-induced Jak-STAT signal transduction." J Biol Chem 271(11):
5961-4.
Kim, J. E., B. P. Liu et al. (2004). "Nogo-66 receptor prevents raphespinal and
rubrospinal axon regeneration and limits functional recovery from spinal cord
injury." Neuron 44(3): 439-51.
192
Kiryu, S., N. Morita et al. (1995). "Regulation of mRNA expression involved in Ras and
PKA signal pathways during rat hypoglossal nerve regeneration." Brain Res
Mol Brain Res 29(1): 147-56.
Kiryu-Seo, S., R. Kato et al. (2008). "Neuronal injury-inducible gene is synergistically
regulated by ATF3, c-Jun, and STAT3 through the interaction with Sp1 in
damaged neurons." J Biol Chem 283(11): 6988-96.
Kiuchi, N., K. Nakajima et al. (1999). "STAT3 is required for the gp130-mediated full
activation of the c-myc gene." J Exp Med 189(1): 63-73.
Koprivica, V., K. S. Cho et al. (2005). "EGFR activation mediates inhibition of axon
regeneration by myelin and chondroitin sulfate proteoglycans." Science
310(5745): 106-10.
Kortylewski, M., F. Feld et al. (2003). "Akt modulates STAT3-mediated gene expression
through a FKHR (FOXO1a)-dependent mechanism." J Biol Chem 278(7):
5242-9.
Krebs, D. L. and D. J. Hilton (2001). "SOCS proteins: negative regulators of cytokine
signaling." Stem Cells 19(5): 378-87.
Kretz, A., S. Kugler et al. (2004). "Excess Bcl-XL increases the intrinsic growth potential
of adult CNS neurons in vitro." Mol Cell Neurosci 26(1): 63-74.
Kulesh, D. A., D. R. Clive et al. (1987). "Identification of interferon-modulated
proliferation-related cDNA sequences." Proc Natl Acad Sci U S A 84(23):
8453-7.
Le Beau, J. M., B. Tedeschi et al. (1991). "Increased expression of pp60c-src proteintyrosine kinase during peripheral nerve regeneration." J Neurosci Res 28(2):
299-309.
Leaver, S. G., Q. Cui et al. (2006). "AAV-mediated expression of CNTF promotes longterm survival and regeneration of adult rat retinal ganglion cells." Gene Ther
13(18): 1328-41.
Lee, J. M., P. Yan et al. (2008). "Methylprednisolone protects oligodendrocytes but not
neurons after spinal cord injury." J Neurosci 28(12): 3141-9.
Lee, N., K. L. Neitzel et al. (2004). "STAT3 phosphorylation in injured axons before
sensory and motor neuron nuclei: potential role for STAT3 as a retrograde
signaling transcription factor." J Comp Neurol 474(4): 535-45.
193
Lehmann, M., A. Fournier et al. (1999). "Inactivation of Rho signaling pathway promotes
CNS axon regeneration." J Neurosci 19(17): 7537-47.
Levy, D. E. and J. E. Darnell, Jr. (2002). "Stats: transcriptional control and biological
impact." Nat Rev Mol Cell Biol 3(9): 651-62.
Levy, D. E. and C. K. Lee (2002). "What does Stat3 do?" J Clin Invest 109(9): 1143-8.
Lewis, S. E., M. S. Rao et al. (1994). "Coordinate regulation of choline acetyltransferase,
tyrosine hydroxylase, and neuropeptide mRNAs by ciliary neurotrophic factor
and leukemia inhibitory factor in cultured sympathetic neurons." J Neurochem
63(2): 429-38.
Li, F., G. Ambrosini et al. (1998). "Control of apoptosis and mitotic spindle checkpoint
by survivin." Nature 396(6711): 580-4.
Li, H. Y., E. H. Say et al. (2007). "Isolation and characterization of neural crest
progenitors from adult dorsal root ganglia." Stem Cells 25(8): 2053-65.
Li, S. and S. M. Strittmatter (2003). "Delayed systemic Nogo-66 receptor antagonist
promotes recovery from spinal cord injury." J Neurosci 23(10): 4219-27.
Liu, B. P., A. Fournier et al. (2002). "Myelin-associated glycoprotein as a functional
ligand for the Nogo-66 receptor." Science 297(5584): 1190-3.
Liu, L., K. M. McBride et al. (2005). "STAT3 nuclear import is independent of tyrosine
phosphorylation and mediated by importin-alpha3." Proc Natl Acad Sci U S A
102(23): 8150-5.
Liu, M. and A. Grigoriev (2005). "Fast parsers for Entrez Gene." Bioinformatics 21(14):
3189-90.
Livak, K. J. (1997). "Comparative Ct method, ABI Prism 7700 Sequence Detection
System." PE Applied Biosystems(User Bulletin no. 2.).
Livesey, F. J., J. A. O'Brien et al. (1997). "A Schwann cell mitogen accompanying
regeneration of motor neurons." Nature 390(6660): 614-8.
Lonze, B. E., A. Riccio et al. (2002). "Apoptosis, axonal growth defects, and
degeneration of peripheral neurons in mice lacking CREB." Neuron 34(3):
371-85.
Lund, L. M. and I. G. McQuarrie (1997). "Calcium/calmodulin-dependent protein kinase
II expression in motor neurons: effect of axotomy." J Neurobiol 33(6): 796810.
194
Ma, J. and X. Cao (2006). "Regulation of Stat3 nuclear import by importin alpha5 and
importin alpha7 via two different functional sequence elements." Cell Signal
18(8): 1117-26.
Ma, Z., T. Liu et al. (2006). "Identification of Up-Regulated Genes After Complete
Spinal Cord Transection in Adult Rats." Cell Mol Neurobiol.
Maglott, D., J. Ostell et al. (2005). "Entrez Gene: gene-centered information at NCBI."
Nucleic Acids Res 33(Database issue): D54-8.
Malik, J. M., Z. Shevtsova et al. (2005). "Long-term in vivo inhibition of CNS
neurodegeneration by Bcl-XL gene transfer." Mol Ther 11(3): 373-81.
Mao, C., J. C. Cushman et al. (2003). "ESTAP--an automated system for the analysis of
EST data." Bioinformatics 19(13): 1720-2.
Maritano, D., M. L. Sugrue et al. (2004). "The STAT3 isoforms alpha and beta have
unique and specific functions." Nat Immunol 5(4): 401-9.
Marklund, N., C. T. Fulp et al. (2006). "Selective temporal and regional alterations of
Nogo-A and small proline-rich repeat protein 1A (SPRR1A) but not Nogo-66
receptor (NgR) occur following traumatic brain injury in the rat." Exp Neurol
197(1): 70-83.
Mason, M. R., A. R. Lieberman et al. (2002). "Transcriptional upregulation of SCG10
and CAP-23 is correlated with regeneration of the axons of peripheral and
central neurons in vivo." Mol Cell Neurosci 20(4): 595-615.
Mayford, M. and E. R. Kandel (1999). "Genetic approaches to memory storage." Trends
Genet 15(11): 463-70.
McKeon, R. J., R. C. Schreiber et al. (1991). "Reduction of neurite outgrowth in a model
of glial scarring following CNS injury is correlated with the expression of
inhibitory molecules on reactive astrocytes." J Neurosci 11(11): 3398-411.
McKerracher, L., S. David et al. (1994). "Identification of myelin-associated glycoprotein
as a major myelin-derived inhibitor of neurite growth." Neuron 13(4): 805-11.
Meiri, K. F., J. L. Saffell et al. (1998). "Neurite outgrowth stimulated by neural cell
adhesion molecules requires growth-associated protein-43 (GAP-43) function
and is associated with GAP-43 phosphorylation in growth cones." J Neurosci
18(24): 10429-37.
195
Menet, V., M. Prieto et al. (2003). "Axonal plasticity and functional recovery after spinal
cord injury in mice deficient in both glial fibrillary acidic protein and vimentin
genes." Proc Natl Acad Sci U S A 100(15): 8999-9004.
Merry, D. E., D. J. Veis et al. (1994). "bcl-2 protein expression is widespread in the
developing nervous system and retained in the adult PNS." Development
120(2): 301-11.
Mi, S., X. Lee et al. (2004). "LINGO-1 is a component of the Nogo-66 receptor/p75
signaling complex." Nat Neurosci 7(3): 221-8.
Miyamoto, M., T. Fujita et al. (1988). "Regulated expression of a gene encoding a
nuclear factor, IRF-1, that specifically binds to IFN-beta gene regulatory
elements." Cell 54(6): 903-13.
Mladinic, M., M. Wintzer et al. (2005). "Differential expression of genes at stages when
regeneration can and cannot occur after injury to immature mammalian spinal
cord." Cell Mol Neurobiol 25(2): 407-26.
Moller, K., M. Reimer et al. (1997). "The effects of axotomy and preganglionic
denervation on the expression of pituitary adenylate cyclase activating peptide
(PACAP), galanin and PACAP type 1 receptors in the rat superior cervical
ganglion." Brain Res 775(1-2): 166-82.
Molliver, D. C., D. E. Wright et al. (1997). "IB4-binding DRG neurons switch from NGF
to GDNF dependence in early postnatal life." Neuron 19(4): 849-61.
Montminy, M. R. and L. M. Bilezikjian (1987). "Binding of a nuclear protein to the
cyclic-AMP response element of the somatostatin gene." Nature 328(6126):
175-8.
Moreau-Fauvarque, C., A. Kumanogoh et al. (2003). "The transmembrane semaphorin
Sema4D/CD100, an inhibitor of axonal growth, is expressed on
oligodendrocytes and upregulated after CNS lesion." J Neurosci 23(27): 922939.
Morgenstern, D. A., R. A. Asher et al. (2002). "Chondroitin sulphate proteoglycans in the
CNS injury response." Prog Brain Res 137: 313-32.
Motoyama, N., F. Wang et al. (1995). "Massive cell death of immature hematopoietic
cells and neurons in Bcl-x-deficient mice." Science 267(5203): 1506-10.
Mukhopadhyay, G., P. Doherty et al. (1994). "A novel role for myelin-associated
glycoprotein as an inhibitor of axonal regeneration." Neuron 13(3): 757-67.
196
Murphy, P. G., L. S. Borthwick et al. (1999). "Nature of the retrograde signal from
injured nerves that induces interleukin-6 mRNA in neurons." J Neurosci
19(10): 3791-800.
Murphy, P. G., J. Grondin et al. (1995). "Induction of interleukin-6 in axotomized
sensory neurons." J Neurosci 15(7 Pt 2): 5130-8.
Mynard, V., O. Latchoumanin et al. (2004). "Synergistic signaling by corticotropinreleasing hormone and leukemia inhibitory factor bridged by phosphorylated
3',5'-cyclic adenosine monophosphate response element binding protein at the
Nur response element (NurRE)-signal transducers and activators of
transcription (STAT) element of the proopiomelanocortin promoter." Mol
Endocrinol 18(12): 2997-3010.
Namikawa, K., M. Honma et al. (2000). "Akt/protein kinase B prevents injury-induced
motoneuron death and accelerates axonal regeneration." J Neurosci 20(8):
2875-86.
Neumann, S., F. Bradke et al. (2002). "Regeneration of sensory axons within the injured
spinal cord induced by intraganglionic cAMP elevation." Neuron 34(6): 88593.
Neumann, S. and C. J. Woolf (1999). "Regeneration of dorsal column fibers into and
beyond the lesion site following adult spinal cord injury." Neuron 23(1): 83-91.
Ng, D. C., B. H. Lin et al. (2006). "Stat3 regulates microtubules by antagonizing the
depolymerization activity of stathmin." J Cell Biol 172(2): 245-57.
Oblinger, M. M., J. Wong et al. (1989). "Axotomy-induced changes in the expression of a
type III neuronal intermediate filament gene." J Neurosci 9(11): 3766-75.
O'Brien, C. A. and S. C. Manolagas (1997). "Isolation and characterization of the human
gp130 promoter. Regulation by STATS." J Biol Chem 272(23): 15003-10.
Park, J. B., G. Yiu et al. (2005). "A TNF receptor family member, TROY, is a coreceptor
with Nogo receptor in mediating the inhibitory activity of myelin inhibitors."
Neuron 45(3): 345-51.
Paschen, W., C. Gissel et al. (1998). "Changes in interferon-regulatory factor-1 mRNA
levels after transient ischemia in rat brain." Neuroreport 9(14): 3147-51.
Peeters, P. J., J. Aerssens et al. (2006). "Molecular profiling of murine sensory neurons in
the nodose and dorsal root ganglia labeled from the peritoneal cavity." Physiol
Genomics 24(3): 252-63.
197
Pocock, R. and O. Hobert (2008). "Oxygen levels affect axon guidance and neuronal
migration in Caenorhabditis elegans." Nat Neurosci 11(8): 894-900.
Pradervand, S., H. Yasukawa et al. (2004). "Small proline-rich protein 1A is a gp130
pathway- and stress-inducible cardioprotective protein." Embo J 23(22): 451725.
Qiu, J., W. B. Cafferty et al. (2005). "Conditioning injury-induced spinal axon
regeneration requires signal transducer and activator of transcription 3
activation." J Neurosci 25(7): 1645-53.
Qiu, J., D. Cai et al. (2002). "Spinal axon regeneration induced by elevation of cyclic
AMP." Neuron 34(6): 895-903.
Quelle, F. W., W. Thierfelder et al. (1995). "Phosphorylation and activation of the DNA
binding activity of purified Stat1 by the Janus protein-tyrosine kinases and the
epidermal growth factor receptor." J Biol Chem 270(35): 20775-80.
Raivich, G., M. Bohatschek et al. (2004). "The AP-1 transcription factor c-Jun is required
for efficient axonal regeneration." Neuron 43(1): 57-67.
Ramer, M. S., T. Bishop et al. (2002). "Neurotrophin-3-mediated regeneration and
recovery of proprioception following dorsal rhizotomy." Mol Cell Neurosci
19(2): 239-49.
Richardson, P. M. and V. M. Issa (1984). "Peripheral injury enhances central
regeneration of primary sensory neurones." Nature 309(5971): 791-3.
Rosenfeld, J., S. Cook et al. (1997). "Expression of superoxide dismutase following
axotomy." Exp Neurol 147(1): 37-47.
Rudge, J. S. and J. Silver (1990). "Inhibition of neurite outgrowth on astroglial scars in
vitro." J Neurosci 10(11): 3594-603.
Sayer, F. T., E. Kronvall et al. (2006). "Methylprednisolone treatment in acute spinal
cord injury: the myth challenged through a structured analysis of published
literature." Spine J 6(3): 335-43.
Schaden, H., C. A. Stuermer et al. (1994). "GAP-43 immunoreactivity and axon
regeneration in retinal ganglion cells of the rat." J Neurobiol 25(12): 1570-8.
Schena, M., D. Shalon et al. (1995). "Quantitative monitoring of gene expression patterns
with a complementary DNA microarray." Science 270(5235): 467-70.
198
Schindler, C., K. Shuai et al. (1992). "Interferon-dependent tyrosine phosphorylation of a
latent cytoplasmic transcription factor." Science 257(5071): 809-13.
Schmitt, A. B., S. Breuer et al. (1999). "GAP-43 (B-50) and C-Jun are up-regulated in
axotomized neurons of Clarke's nucleus after spinal cord injury in the adult
rat." Neurobiol Dis 6(2): 122-30.
Schnell, L. and M. E. Schwab (1990). "Axonal regeneration in the rat spinal cord
produced by an antibody against myelin-associated neurite growth inhibitors."
Nature 343(6255): 269-72.
Schobitz, B., F. Holsboer et al. (1992). "Peripheral and central regulation of IL-6 gene
expression in endotoxin-treated rats." Endocr Regul 26(3): 103-9.
Schwaiger, F. W., G. Hager et al. (2000). "Peripheral but not central axotomy induces
changes in Janus kinases (JAK) and signal transducers and activators of
transcription (STAT)." Eur J Neurosci 12(4): 1165-76.
Schweizer, U., J. Gunnersen et al. (2002). "Conditional gene ablation of Stat3 reveals
differential signaling requirements for survival of motoneurons during
development and after nerve injury in the adult." J Cell Biol 156(2): 287-97.
Seijffers, R., C. D. Mills et al. (2007). "ATF3 increases the intrinsic growth state of DRG
neurons to enhance peripheral nerve regeneration." J Neurosci 27(30): 791120.
Shao, Z., J. L. Browning et al. (2005). "TAJ/TROY, an orphan TNF receptor family
member, binds Nogo-66 receptor 1 and regulates axonal regeneration." Neuron
45(3): 353-9.
Shen, Y., K. Schlessinger et al. (2004). "Essential role of STAT3 in postnatal survival
and growth revealed by mice lacking STAT3 serine 727 phosphorylation." Mol
Cell Biol 24(1): 407-19.
Shuai, K. (2000). "Modulation of STAT signaling by STAT-interacting proteins."
Oncogene 19(21): 2638-44.
Shuai, K. and B. Liu (2003). "Regulation of JAK-STAT signalling in the immune
system." Nat Rev Immunol 3(11): 900-11.
Silver, J. and J. H. Miller (2004). "Regeneration beyond the glial scar." Nat Rev Neurosci
5(2): 146-56.
Simpson, C. S. and B. J. Morris (2000). "Regulation of neuronal cell adhesion molecule
expression by NF-kappa B." J Biol Chem 275(22): 16879-84.
199
Sivasankaran, R., J. Pei et al. (2004). "PKC mediates inhibitory effects of myelin and
chondroitin sulfate proteoglycans on axonal regeneration." Nat Neurosci 7(3):
261-8.
Skene, J. H. (1989). "Axonal growth-associated proteins." Annu Rev Neurosci 12: 12756.
Skene, J. H. and M. Willard (1981). "Axonally transported proteins associated with axon
growth in rabbit central and peripheral nervous systems." J Cell Biol 89(1): 96103.
Skene, J. H. and M. Willard (1981). "Changes in axonally transported proteins during
axon regeneration in toad retinal ganglion cells." J Cell Biol 89(1): 86-95.
Skold, M. K., H. H. Marti et al. (2004). "Induction of HIF1alpha but not HIF2alpha in
motoneurons after ventral funiculus axotomy-implication in neuronal survival
strategies." Exp Neurol 188(1): 20-32.
Sobel, A., M. C. Boutterin et al. (1989). "Intracellular substrates for extracellular
signaling. Characterization of a ubiquitous, neuron-enriched phosphoprotein
(stathmin)." J Biol Chem 264(7): 3765-72.
Sole, C., X. Dolcet et al. (2004). "The death receptor antagonist FAIM promotes neurite
outgrowth by a mechanism that depends on ERK and NF-kapp B signaling." J
Cell Biol 167(3): 479-92.
Song, G., C. Cechvala et al. (2001). "GeneChip analysis after acute spinal cord injury in
rat." J Neurochem 79(4): 804-15.
Song, X. Y., J. H. Zhong et al. (2004). "Suppression of p75NTR does not promote
regeneration of injured spinal cord in mice." J Neurosci 24(2): 542-6.
Song, Y., L. Qian et al. (2008). "Fra-1 and Stat3 synergistically regulate activation of
human MMP-9 gene." Mol Immunol 45(1): 137-43.
Spillmann, A. A., C. E. Bandtlow et al. (1998). "Identification and characterization of a
bovine neurite growth inhibitor (bNI-220)." J Biol Chem 273(30): 19283-93.
Stephanou, A., B. K. Brar et al. (2000). "Opposing actions of STAT-1 and STAT-3 on
the Bcl-2 and Bcl-x promoters." Cell Death Differ 7(3): 329-30.
Steward, O., B. Zheng et al. (2003). "False resurrections: distinguishing regenerated from
spared axons in the injured central nervous system." J Comp Neurol 459(1): 18.
200
Strittmatter, S. M., C. Fankhauser et al. (1995). "Neuronal pathfinding is abnormal in
mice lacking the neuronal growth cone protein GAP-43." Cell 80(3): 445-52.
Stromberg, H., S. P. Svensson et al. (2000). "Distribution of the transcription factor
signal transducer and activator of transcription 3 in the rat central nervous
system and dorsal root ganglia." Brain Res 853(1): 105-14.
Suzuki, S., K. Tanaka et al. (2001). "Phosphorylation of signal transducer and activator of
transcription-3 (Stat3) after focal cerebral ischemia in rats." Exp Neurol
170(1): 63-71.
Tachibana, T., K. Noguchi et al. (2002). "Analysis of gene expression following spinal
cord injury in rat using complementary DNA microarray." Neurosci Lett
327(2): 133-7.
Takahashi, K. and S. Yamanaka (2006). "Induction of pluripotent stem cells from mouse
embryonic and adult fibroblast cultures by defined factors." Cell 126(4): 66376.
Takeda, K., K. Noguchi et al. (1997). "Targeted disruption of the mouse Stat3 gene leads
to early embryonic lethality." Proc Natl Acad Sci U S A 94(8): 3801-4.
Tamm, I., Y. Wang et al. (1998). "IAP-family protein survivin inhibits caspase activity
and apoptosis induced by Fas (CD95), Bax, caspases, and anticancer drugs."
Cancer Res 58(23): 5315-20.
Tanaka, H., T. Yamashita et al. (2002). "Cytoplasmic p21(Cip1/WAF1) regulates neurite
remodeling by inhibiting Rho-kinase activity." J Cell Biol 158(2): 321-9.
Tharakaraman, K., O. Bodenreider et al. (2008). "The biological function of some human
transcription factor binding motifs varies with position relative to the
transcription start site." Nucleic Acids Res 36(8): 2777-86.
Thomas, P. D., M. J. Campbell et al. (2003). "PANTHER: a library of protein families
and subfamilies indexed by function." Genome Res 13(9): 2129-41.
Tom, V. J., C. M. Doller et al. (2004). "Astrocyte-associated fibronectin is critical for
axonal regeneration in adult white matter." J Neurosci 24(42): 9282-90.
Tonra, J. R., R. Curtis et al. (1998). "Axotomy upregulates the anterograde transport and
expression of brain-derived neurotrophic factor by sensory neurons." J
Neurosci 18(11): 4374-83.
201
Tucker, B. A., M. Rahimtula et al. (2006). "Laminin and growth factor receptor
activation stimulates differential growth responses in subpopulations of adult
DRG neurons." Eur J Neurosci 24(3): 676-90.
Turnley, A. M. and P. F. Bartlett (1998). "MAG and MOG enhance neurite outgrowth of
embryonic mouse spinal cord neurons." Neuroreport 9(9): 1987-90.
Veldman, M. B., M. A. Bemben et al. (2007). "Gene expression analysis of zebrafish
retinal ganglion cells during optic nerve regeneration identifies KLF6a and
KLF7a as important regulators of axon regeneration." Dev Biol 312(2): 596612.
Villar, M. J., R. Cortes et al. (1989). "Neuropeptide expression in rat dorsal root ganglion
cells and spinal cord after peripheral nerve injury with special reference to
galanin." Neuroscience 33(3): 587-604.
Wakisaka, S., K. C. Kajander et al. (1991). "Increased neuropeptide Y (NPY)-like
immunoreactivity in rat sensory neurons following peripheral axotomy."
Neurosci Lett 124(2): 200-3.
Wang, G. L., B. H. Jiang et al. (1995). "Hypoxia-inducible factor 1 is a basic-helix-loophelix-PAS heterodimer regulated by cellular O2 tension." Proc Natl Acad Sci
U S A 92(12): 5510-4.
Wang, J. H., B. J. Manning et al. (2003). "Endotoxin/lipopolysaccharide activates NFkappa B and enhances tumor cell adhesion and invasion through a beta 1
integrin-dependent mechanism." J Immunol 170(2): 795-804.
Wang, J. T., N. J. Kunzevitzky et al. (2007). "Disease gene candidates revealed by
expression profiling of retinal ganglion cell development." J Neurosci 27(32):
8593-603.
Wang, K. C., V. Koprivica et al. (2002). "Oligodendrocyte-myelin glycoprotein is a Nogo
receptor ligand that inhibits neurite outgrowth." Nature 417(6892): 941-4.
Wen, Z. and J. E. Darnell, Jr. (1997). "Mapping of Stat3 serine phosphorylation to a
single residue (727) and evidence that serine phosphorylation has no influence
on DNA binding of Stat1 and Stat3." Nucleic Acids Res 25(11): 2062-7.
Werner, A., M. Willem et al. (2000). "Impaired axonal regeneration in alpha7 integrindeficient mice." J Neurosci 20(5): 1822-30.
Winton, M. J., C. I. Dubreuil et al. (2002). "Characterization of new cell permeable C3like proteins that inactivate Rho and stimulate neurite outgrowth on inhibitory
substrates." J Biol Chem 277(36): 32820-9.
202
Wong, S. T., J. R. Henley et al. (2002). "A p75(NTR) and Nogo receptor complex
mediates repulsive signaling by myelin-associated glycoprotein." Nat Neurosci
5(12): 1302-8.
Wynick, D., S. W. Thompson et al. (2001). "The role of galanin as a multi-functional
neuropeptide in the nervous system." Curr Opin Pharmacol 1(1): 73-7.
Xia, X. G., H. D. Hofmann et al. (2002). "Induction of STAT3 signaling in activated
astrocytes and sprouting septal neurons following entorhinal cortex lesion in
adult rats." Mol Cell Neurosci 21(3): 379-92.
Yamamoto, J., T. Kawamata et al. (2008). "Down-regulation of mu opioid receptor
expression within distinct subpopulations of dorsal root ganglion neurons in a
murine model of bone cancer pain." Neuroscience 151(3): 843-53.
Yamauchi, K., K. Osuka et al. (2006). "Activation of JAK/STAT signalling in neurons
following spinal cord injury in mice." J Neurochem 96(4): 1060-70.
Yiu, G. and Z. He (2006). "Glial inhibition of CNS axon regeneration." Nat Rev Neurosci
7(8): 617-27.
Yoo, J. Y., D. L. Huso et al. (2002). "Specific ablation of Stat3beta distorts the pattern of
Stat3-responsive gene expression and impairs recovery from endotoxic shock."
Cell 108(3): 331-44.
Yu, Z., W. Zhang et al. (2002). "Signal transducers and activators of transcription 3
(STAT3) inhibits transcription of the inducible nitric oxide synthase gene by
interacting with nuclear factor kappaB." Biochem J 367(Pt 1): 97-105.
Zhang, X., J. Zhu et al. (2007). "Dishevelled promotes axon differentiation by regulating
atypical protein kinase C." Nat Cell Biol 9(7): 743-54.
Zhang, Y., X. Bo et al. (2005). "Growth-associated protein GAP-43 and L1 act
synergistically to promote regenerative growth of Purkinje cell axons in vivo."
Proc Natl Acad Sci U S A 102(41): 14883-8.
Zheng, B., J. Atwal et al. (2005). "Genetic deletion of the Nogo receptor does not reduce
neurite inhibition in vitro or promote corticospinal tract regeneration in vivo."
Proc Natl Acad Sci U S A 102(4): 1205-10.
Zheng, B., C. Ho et al. (2003). "Lack of enhanced spinal regeneration in Nogo-deficient
mice." Neuron 38(2): 213-24.
203
Zhong, J., I. D. Dietzel et al. (1999). "Sensory impairments and delayed regeneration of
sensory axons in interleukin-6-deficient mice." J Neurosci 19(11): 4305-13.
Zhong, Z., Z. Wen et al. (1994). "Stat3: a STAT family member activated by tyrosine
phosphorylation in response to epidermal growth factor and interleukin-6."
Science 264(5155): 95-8.
Zuo, J., D. Neubauer et al. (1998). "Degradation of chondroitin sulfate proteoglycan
enhances the neurite-promoting potential of spinal cord tissue." Exp Neurol
154(2): 654-62.