AXONAL NA+/K+ ATPASE:
LOCALIZATION, LOSS, AND LESSONS LEARNED
by
ELIZABETH ANN YOUNG
Submitted in partial fulfillment of the requirements
For the degree of Doctor of Philosophy
Dissertation Adviser: Dr. Bruce D. Trapp
Department of Neurosciences
CASE WESTERN RESERVE UNIVERSITY
August, 2010
ii
TABLE OF CONTENTS
Page
LIST OF TABLES
v
LIST OF FIGURES
vi
ABSTRACT
vii
CHAPTER ONE:
GENERAL INTRODUCTION
1
Statement of the Problem
2
The axon: Structure and Fuction
3
NaV channels
4
K channels
7
Neurofilaments
9
Mitochondria
10
Sodium/Potassium Adenosine Triphosphatase
11
Multiple Sclerosis
14
CHAPTER TWO:
IMAGING CORRELATES OF DECREASED AXONAL
24
NA/K ATPASE IN CHRONIC MS LESIONS
Introduction
25
Subjects and Methods
28
Tissue and Lesions
28
MRI and Image-guided Tissue Sampling
28
Immunocytochemistry
29
iii
Antibodies
30
Confocal Microscopy
30
Quantification of Pathology
30
Results
32
NKA Subunits Located in Internodal Axolemma
32
Distribution of NKA subunits in MS Lesions
33
Imaging Correlates of Axonal NKA
35
Discussion
36
CHAPTER THREE:
GENERAL DISCUSSION
51
Internodal Na/K ATPase: A new model…
53
Functions
54
Dysfunctions
55
Loss of Na/K ATPase: Lessons from other cell types
56
Why is the Na/K ATPase missing?
56
How is the Na/K ATPase removed?
57
Axonal Na/K ATPase: A PURED candidate?
58
Na/K ATPase: A new therapeutic target?
59
MRI/Pathology correlations
60
Summary
61
REFERENCES
64
iv
List of Tables
Page
CHAPTER TWO:
Table 1.
Characteristics of Patients and Number of Lesions Studied
v
40
List of Figures
Page
CHAPTER ONE:
Figure 1.
Structural and functional properties of central nervous system axons
Figure 2.
A graphic representation of disease progression and disability by disease
subtype
20
22
CHAPTER TWO:
Figure 1.
41
Na/K ATPase is enriched in the internodal axolemma of myelinated axons
in the adult human brain.
Figure 2.
Demyelinated axons in some chronic multiple sclerosis (MS) lesions lack
Na/K ATPase.
43
Figure 3.
45
Quantification of Na/K ATPase-positive axons in control human brain and
multiple sclerosis (MS) lesions.
Figure 4.
Percentages of Na/K ATPase-positive axons are predicted by Magnetic
Resonance Imaging (MRI).
47
Figure 5.
Magnetization transfer ratios (MTRs) and T1 contrast ratios linearly
correlate with the percentage of Na/K ATPase-positive axons in chronic
MS lesions.
49
CHAPTER THREE:
Figure 1.
A new model for axonal function and dysfunction in MS.
vi
62
Axonal Na+/K+ ATPase:
Localization, Loss, and Lessons Learned
Abstract
by
ELIZABETH ANN YOUNG
Degeneration of chronically demyelinated axons is the leading causative factor in
the progression of symptoms experienced by patients with multiple sclerosis (MS),
despite the fact that MS has defied classification as a primary neurodegenerative disease.
While there are many potential causes of neuronal damage immediately following
demyelination, it is within the context of chronic demyelination that the fundamental and
intrinsic mechanisms of axonal function can, themselves, become detrimental to axonal
integrity. In the wake of demyelination, extensive submembranous cytoskeletal
derangement occurs, allowing the once-nodal voltage-gated sodium channels (NaV) to
diffuse along the denuded length of the axon. The immediate impact on the axon is twofold; first is a spatially unrestricted influx of sodium and an associated increase in
axoplasmic sodium ([Na]i), second is abrogation of saltatory conduction, which relies on
focal sodium influx. In response to a global increase in [Na]i, typically caused by an
action potential, the sodium/potassium adenosine triphosphatase (Na/K ATPase), or the
sodium pump, hydrolyzes ATP to power the active transport of three intracellular sodium
ions for two extracellular potassium ions and, thereby, repolarization of the cell. The cell
is driven into an energy debt, causing stress to mitochondria and, thereby, decreasing
vii
ATP production. Without sufficient axonal ATP, the Na/K ATPase cannot transport ions
across the axolemma. With the Na/K ATPases unable to transport additional sodium
across the axolemma, [Na]i increases and the sodium/calcium exchanger (NCX) is driven
to reverse its function and exchange intracellular sodium for extracellular calcium. An
unchecked rise in axoplasmic calcium can lead to the induction of many deleterious
calcium-dependent degradative pathways and the eventual degeneration of the
demyelinated axon. In these studies, the “secondary neurodegeneration” of MS is
explored; specifically, how the energetic properties of the chronically demyelinated axon
lead to a series of events, beginning, most notably, with the loss of the Na/K ATPase, that
lead to a run-down of the axolemmal membrane potential and, eventually, axonal
degeneration.
viii
CHAPTER 1. GENERAL INTRODUCTION
1
STATEMENT OF THE PROBLEM
Degeneration of chronically demyelinated axons is the leading causative factor in
the progression of symptoms experienced by patients with multiple sclerosis (MS),
despite the fact that MS has defied classification as a primary neurodegenerative disease.
While there are many potential causes of neuronal damage immediately following
demyelination, it is within the context of chronic demyelination that the fundamental and
intrinsic mechanisms of axonal function can, themselves, become detrimental to axonal
integrity. In recent years, research has begun to explore the connection between the
energetic status of the demyelinated axon and its functional and structural integrity. The
myelinated axon is an example of an efficient electrical conduit; a well-insulated and
energetically favorable “cable” that extends from the cell body to, up to meters away, its
synapse. Should the cable’s insulation be removed, the cable is no longer capable of
transmitting electricity efficiently. The same, then, is true of a demyelinated axon. In
the wake of demyelination, extensive submembranous cytoskeletal derangement occurs,
allowing the once-nodal voltage-gated sodium channels (NaV) to diffuse along the
denuded length of the axon. The immediate impact on the axon is two-fold; first is a
spatially unrestricted influx of sodium and an associated increase in axoplasmic sodium
([Na]i), second is abrogation of saltatory conduction, which relies on focal sodium influx.
In response to a global increase in [Na]i, typically caused by an action potential, the
sodium/potassium adenosine triphosphatase (Na/K ATPase), or the sodium pump,
hydrolyzes ATP to power the active transport of three intracellular sodium ions for two
extracellular potassium ions and, thereby, repolarization of the cell. The cell is driven
2
into an energy debt, causing stress to mitochondria and, thereby, decreasing ATP
production. Without sufficient axonal ATP, the Na/K ATPase cannot transport ions
across the axolemma. With the Na/K ATPases unable to transport additional sodium
across the axolemma, [Na]i increases and the sodium/calcium exchanger (NCX) is driven
to reverse its function and exchange intracellular sodium for extracellular calcium. An
unchecked rise in axoplasmic calcium can lead to the induction of many deleterious
calcium-dependent degradative pathways and the eventual degeneration of the
demyelinated axon. In these studies, the “secondary neurodegeneration” of MS is
explored; specifically, how the energetic properties of the chronically demyelinated axon
lead to a series of events, beginning, most notably, with the loss of the Na/K ATPase, that
lead to a run-down of the axolemmal membrane potential and, eventually, axonal
degeneration.
THE AXON: STRUCTURE AND FUNCTION
The axon is a long cylinder of axoplasm surrounded by a magnificently organized
plasma membrane, extending from soma to synapse. While it provides a physical
connection between these two distant compartments allowing organelle and protein
transport, the axon’s most critical role is as an electrically active, ionically privileged
compartment facilitating the propagation of action potentials. The ionic gradient, which
is generated by the disparity between charge across the axolemma, is set between -55mV
and -80mV (Hodgkin and Katz, 1949) and serves as the key driving force behind the
3
action potential. The specialized domains of the axon are crucial for proper axonal
function (Figure 1). In the central nervous system, these domains are generated by axoglial signaling between oligodendrocytes and axonal cytoskeletal elements during
myelination. Produced by oligodendrocytes, central nervous system myelin wraps axons
(Figure 1) and, in so doing, induces three major changes to the axon: 1. internodal
resistance is decreased, 2. NaV channels are concentrated at nodes of Ranvier, and 3.
neurofilaments become phosphorylated, increasing the diameter of the axon. Both of
these membrane changes are critical for the onset of saltatory conduction; an increase of
conduction velocity along myelinated internodes, allowing electrical currents to
effectively “jump” from one active zone (node of Ranvier) to the next (Figure 1) (Huxley
and Stampfli, 1949). The precise composition of axonal ion channels, cytoskeletal
elements, and organelles in the axon are orchestrated by myelination and crucial for
optimal action potential propagation. It follows, then, that demyelination would derange
this meticulously organized system, perturbing saltatory conduction. Just how the axonal
components behave following demyelination is paramount to understanding the damage
that ensues.
NaV channels
Since Hodgkin and Huxley’s first experiments on squid giant axons (Hodgkin and
Huxley, 1945; Hodgkin and Huxley, 1952), the role of the voltage-gated sodium channel
(NaV) as the central mediator of action potential propagation has been well-appreciated.
In normally myelinated axons, NaV channels are clustered both at the axon initial
segment (AIS, or axon hillock) and at the node of Ranvier. This localization is of
4
paramount importance to the high-fidelity, high-frequency propagation of action
potentials along the axon. It has been estimated that approximately 1000/µm2 (Waxman,
1977) to 12000/µm2 (Ritchie and Rogart, 1977) NaV channels are embedded in the nodal
axolemma. Given the fact that there are no apparent diffusion barriers within the
axoplasm (Arhem, 1976), it is reasonable to assume that the number of NaV channels
clustered at any one node is tightly controlled to ensure the optimal influx of a precise
number of sodium ions to effect immediate local depolarization without any off-target
effects. The exact mechanism by which NaV channels are trafficked to and retained at the
node is still unclear, but there is extensive evidence that the nodal architecture and, in
particular, NaV clustering relies on interactions with Ankyrin G, a submembranous
cytoskeletal component (Jenkins and Bennett, 2001; Bouzidi et al., 2002;Lemaillet et al.,
2003; Pan et al., 2006). Ankryin G is thought to associate with the node via an
interaction with betaIV spectrin (Berghs et al., 2000; Komada and Soriano, 2002). The
exquisitely tuned action potential relies not only on the precise localization of NaV
channels, but also on the correct isoform.
Early in development, the AIS and nodes are comprised mainly of NaV1.2, a
rapidly-inactivating subunit (Zhou and Goldin, 2004). Upon maturation, NaV1.6, a highfidelity (Zhou et al., 2004) and rapidly re-activating subunit (Herzog et al., 2003),
replaces NaV1.2 in the majority of CNS nodes (Boiko et al., 2001; Kaplan et al., 2001).
This subunit switch is thought to occur to ensure the continued high-fidelity, rapid
propagation of action potentials along the axon (Waxman et al., 2004). In a subset of
CNS nodes, NaV1.2 persists and is co-expressed with NaV1.8 (Arroyo et al., 2002), but
the benefit to maintaining these slower, lower-fidelity nodes is unclear.
5
Upon demyelination, there is an immediate interruption of action potential
propagation, likely due to the decreased resistance of the internodal axolemma and the
increased active distance between clusters of NaV channels. This phenomenon, known as
conduction block, can cause clinical symptoms which present as MS exacerbations
(Waxman, 1998). The effects of conduction block can be severe but are, most often,
temporary; typically lasting from two weeks to two months. As the nodal and paranodal
cytoskeletons are so closely aligned with the paranodal loops, demyelination effectively
abolishes myelin’s instructive effect on axonal organization (Scherer and Arroyo, 2002).
Derangement of the submembranous scaffolding allows for NaV channel diffusion away
from the quondam nodes, redistributing NaV channels along the length of the denuded
axon. It is unclear whether the redistribution is caused entirely by the lateral diffusion of
nodal NaV or an increased expression and membrane insertion. This phenomenon,
described originally by electrophysiological phenomena (Bostock et al., 1983; Schwarz et
al., 1991) and subsequently confirmed with immunohistochemical techniques (England et
al., 1990;Moll et al., 1991), is responsible for recovery from conduction block, albeit
with slower and less-efficient action potential propagation. In some demyelinated axons,
small clusters of NaV channels have been described (Coman et al., 2006), perhaps
representing the last vestiges of former nodes tethered together by rogue Ankryin G, or
serving as the nidus for remyelination. Axons with these remaining clusters may have
improved conduction as compared to those with completely redistributed NaV channels.
Further, it is of note that the redistributed channels are not the developmental subtype
NaV1.2, but the mature subtype NaV1.6, which has a persistant sodium current (Craner et
al., 2004). While this redistribution serves to restore action potential propagation, there
6
are consequences – specifically, the unchecked global influx of sodium along the entire
demyelinated length of the axon.
K channels
The great diversity in potassium channel subtype and function allows for fine
tuning of axonal properties such as resting membrane potential, action potential
threshold, and hyperpolarization. Not unlike NaV channels (see above), the precise
location of the potassium channels is critical for action potential firing and maintenance,
and this precise location is controlled by the potassium channels’ interaction with
Ankyrin G (Pan et al., 2006). Two main subtypes of potassium channels segregate into
discrete domains along the axon and exert their influence on the active properties of the
axon. A subset of the channels reside nodally, whereas a second population reside
juxtaparanodally. Despite all being considered “fast voltage-gated potassium channels”
(fast KV), due to their location, they have varied modulatory effects on action potentials
(Burke et al., 2001). The role of nodal KV channels, such as KV3.1 (Devaux et al., 2003)
and KCNQ2 (Binah and Palti, 1981; Devaux et al., 2004), in action potential modulation
has been debated. While some studies suggest that nodal KV channels have only a
negligible impact on the action potential (Schwarz et al., 1995), many other studies have
pointed to dramatic modulatory roles for KV3.1 and KCNQ2. For instance, when KV3.1
channels are blocked with tetraethylammonium (TEA), the action potential is lengthened,
which leads to increased neurotransmitter release (Wang et al., 1998; Ishikawa et al.,
2003). KCNQ2 channels are responsible for the M-current; so named because it is
7
antagonized by muscarinic acetylcholine receptor agonists (Brown and Adams, 1980).
The M-current can affect action potentials by decreasing NaV inactivation, thereby
strengthening and increasing action potential propagation (Vervaeke et al., 2006). The
juxtaparanodal KV channels include KV1.1 and KV1.2 (Wang et al., 1998; Rasband et al.,
1998; Trimmer and Rhodes, 2004). KV3.1b channels are, while of primary importance in
the node, also found in the juxtaparanode (Lai and Jan, 2006). KV1.1 and KV1.2 are the
canonical potassium channels responsible for ensuring high-fidelity firing by
hyperpolarizing the membrane after depolarization (Wu and Barish, 1992). Both nodal
and juxtaparanodal KV channels are responsible for shaping the action potential and
ensuring continued propagation.
Upon demyelination, KV channels diffuse away from the erstwhile juxtaparanode
(Rasband et al., 1998; Arroyo et al., 2004; Coman et al., 2006), likely due to a similar
Ankyrin G disruption mechanism as NaV channels. While redistribution of nodal and
paranodal KV channels certainly disrupts membrane potential, it is the sudden uncloaking
of the scores of internodal (Chiu and Ritchie, 1980; Chiu and Schwarz, 1987),
juxtaparanodal (Wang et al., 1993), and paranodal (Chiu and Ritchie, 1981) potassium
channels that is of the greatest concern. The fast outward potassium current carried by
the juxtaparanodal KV1.1, KV1.2, and KV3.1b channels (Lai et al., 2006), no longer
obscured by the myelin sheath, begins to impact the axon’s resting potential – drawing it
closer to the reversal potential for potassium – thereby dampening all attempts at
depolarization.
8
Neurofilaments
Directly underneath the axolemma is an intricate network of interconnecting
cytoskeletal elements called the cytoskeletal cortex (Ichimura and Ellisman, 1991).
Historically, the submembranous cytoskeleton of the axon was thought to be a dense
protoplasm of cross-linked proteins and scaffolding, then called the microtrabecular
matrix (Ellisman and Porter, 1980). Early high voltage electron micrographs (HVEM)
suggested that the axoplasm directly apposed to the axolemma was a nearly-crystalline
mesh of proteins. These results, however, have been called into question as more recent
studies (Heuser, 2002;Sardet, 2002) have demonstrated that some, if not all, of the
HVEM-visible microtrabeculae are artifacts of fixation and cell permeablization. It has
been suggested that the microtrabecular matrix first described by Porter (Ellisman et al.,
1980) is, in fact, the crystallization of soluble, cytoskeleton-associated proteins (Heuser,
2002). Regardless of the technical constraints of the early work, Porter’s theory of the
microtrabecular matrix has led to an appreciation for the complexity of the
submembranous cytoskeleton.
The main structural cytoskeletal components of the axon are microtubules and
neurofilaments. Axonal transport relies on the microtubule network and a vast array of
molecular motors that carry cargo. For the purposes of this document, however, we will
focus on the role of neurofilaments in axons. Neurofilament, a Type IV intermediate
filament, provides structural and tensile support to the axon. The three isoforms of
neurofilament are named for their molecular weight (Neurofilament light chain, NF-L,
~70kD; Neurofilament medium chain, NF-M, ~160kD; Neurofilament heavy chain, NFH, 200kD). These subunits co-assemble to form stable polymers. The light chain (NF-L)
9
is the obligate isoform which is required for polymerization, acting as the core of the
nascent neurofilament polymer. NF-M or NF-H then binds to the central NF-L. Protein
cross-bridges formed by the c-terminal domain of either NF-M or NF-H connect adjacent
neurofilament polymers and f-actin (Hirokawa, 1982). The number and phosphorylation
state of neurofilaments directly impacts axonal diameter which is, in turn, critical for
rapid axonal conduction. Myelination is responsible for the initiation and maintenance of
neurofilament phosphorylation and, by extension, an increase in axonal diameter(Sanchez
et al., 1996; Yin et al., 1998).
Upon demyelination, the axo-glial signaling responsible for maintenance of
neurofilament phosphorylation is disrupted. Myelin-associated glycoprotein (MAG) has
been identified as the myelin protein responsible for modulation of neurofilament
phosphorylation. In MAG knockout mice, axons that were myelinated were thinner and
had significantly less phosphorylated neurofilament than did wild-type animals(Yin et al.,
1998). While still functional, axons composed of dephosphorylated neurofilament are
unable to conduct action potentials at rapid velocities. Additionally, following
demyelination and NCX reversal, calcium-mediated degradative processes (such as
calpains) begin to depolymerize and destabilize the neurofilament bundles, leading to
axons that are weaker and more prone to damage.
Mitochondria
The mitochondrion plays many essential roles in the axon: adenosine triphosphate
(ATP) production, calcium buffering, and reactive oxygen species production. The
mitochondrion generates ATP during cellular respiration, during which acetyl coA is
10
oxidized to carbon dioxide. The process by which this is accomplished is known as the
citric acid cycle, or Krebs cycle, and requires the mitochondrion’s electron transport
chain.
While physically separated from the acute disruption of demyelination, the
axoplasmic mitochondrion, nonetheless, undergoes dramatic changes. To begin, the loss
of axoplasmic ion homeostasis that occurs following demyelination impacts the ionic
gradient of the inner and outer mitochondrial membranes. The increased axoplasmic Na
induces Ca release from mitochondrial stores. Additionally, gene expression studies of
mitochondrial electron transport chain complex I, III (Dutta et al., 2006), and IV (Mahad
et al., 2009) indicate decreased mRNA levels of these critical components of the ATP
production machinery. Finally, the generation of nitric oxide (NO), which occurs during
acute inflammation, can further inhibit the mitochondrial electron transport chain –
effectively reducing ATP production to zero. Given the delicate balance between
mitochondrial ATP production and axonal energy consumption, it is of considerable
importance to look to the one molecule responsible for hydrolyzing more ATP than any
other in the CNS: the Na/K ATPase (Ames, III, 2000).
SODIUM/POTASSIUM ADENOSINE TRIPHOSPHATASE
The sodium/potassium adenosine triphosphatase (Na/K ATPase), or sodium
pump, is a ubiquitous, electrogenic ion pump that resides in the plasma membranes of
cells in virtually all members of the kingdom Animalia. The Na/K ATPase is absolutely
essential for cellular function. The Na/K ATPase is critical for maintaining the relative
negative charge carried by cells’ membranes. It does this by exchanging 3 intracellular
11
Na ions for 2 extracellular K ions, which requires the hydrolysis of one molecule of ATP.
The pump exists as a heterodimer of one α subunit and one β subunit, which exist in the
cell in equimolar amounts.
The α subunit is the catalytic subunit of the holoenzyme. It binds both the ions
and the ATP. To date, there are four α subunits known: α1, α2, α3, and α4 (Shull et al.,
1986; Herrera et al., 1987; Takeyasu et al., 1990; Woo et al., 2000). Expression of the
four α subunits varies widely, from the α1 subunit which is the most common to the α4
subunit which is expressed only on the flagella of sperm (Woo et al., 2000; Blanco et al.,
2000). The protein is large; spanning the membrane ten times – both the amino and
carboxy tails of the protein are intracellular (Antolovic et al., 1991). All four α subunits
are the products of different genes (Shull and Lingrel, 1987; Sverdlov et al., 1987; Woo
et al., 2000). Displaying an impressive degree of structural conservation of the 23 exons,
each α subunit gene is thought to be the result of genetic duplication from a common
ancestor (Levenson, 1994).
The β subunit is responsible for escorting the catalytic subunit into the membrane
(Geering et al., 1985) and hydrolyzing ATP. The β subunit is the binding site for many
pump inhibitors, so it also exerts a regulatory effect on the catalytic subunit. While the
catalytic subunit remains embedded in the membrane, the β subunit is constantly turned
over through its internalization and degradation (Yoshimura et al., 2008). There are four
β subunits currently known: β1, β2, β3, and β4 (Mercer et al., 1986; Martin-Vasallo et
al., 1989; Malik et al., 1996; Zhao et al., 2004). Beta-4, the most recently described
subunit, is developmentally regulated and is only expressed in the sarcoplasmic reticula
and nuclear envelopes of skeletal muscles in adult animals (Zhao et al., 2004). The β
12
subunit is a single-pass integral membrane protein that is highly glycosylated – leading
many to speculate that it may serve additional functions, such as adhesion (Gloor et al.,
1990). Like the α subunits, the β subunits are encoded by 4 separate genes (Lane et al.,
1989; Shyjan et al., 1991; Malik et al., 1996; Zhao et al., 2004).
Ionic homeostasis is essential for the optimal function of every cell and is
absolutely of the utmost importance in electrochemically active cells, such as neurons.
The human central nervous system contains α1, α2, α3, β1, and β3 (Sweadner, 1979;
Cameron et al., 1994; Moseley et al., 2003) Na/K ATPase subunits. Neurons, in
particular, have been shown to possess isoforms of both α1/ β1 and α3/ β1 varieties.
Considering the continued need for a neuron to maintain its membrane potential, coexpression of both α1/ β1 and α3/ β1 isoforms is the perfect way to manage the task. The
α1 isoform is known as the “housekeeping” isoform because its affinity for Na is high,
while its affinity for ATP is low. This means that the α1 isoform will continue pumping
excess intracellular Na just as long as there is sufficient cellular ATP. The α3 isoform,
on the other hand, has low affinity for Na, but a high affinity for ATP. The α3-containing
isoforms have such a high affinity for ATP that they will “steal” ATP from other cellular
functions (Castro et al., 2006) in order to pump. Both of these functions are essential to
the proper function of a neuron, but they must be properly balanced.
To best understand the role of the Na/K ATPase in the axon, we must first
understand where it is located along the axon. Many studies have attempted to address
the issue of axonal localization of the Na/K ATPase, but the results have been mixed. A
large cohort of researchers has claimed to have identified Na/K ATPase in the node of
Ranvier (Wood et al., 1977; Ariyasu et al., 1985;Waxman and Ritchie, 1993;Kordeli et
13
al., 1995;Kanoh and Sakagami, 1996), where it would be spatially coupled with Na
influx. One group has described finding the Na/K ATPase on the internodal axolemma
(Mata et al., 1991), where the evolutionary benefit is unclear. Answering the question of
axonal localization is critical to deciphering the role of the Na/K ATPase during normal
function, such as action potential propagation and repolarization, and during axonal
degeneration, such as occurs following chronic demyelination.
MULTIPLE SCLEROSIS
Multiple sclerosis (MS), a chronic demyelinating disease of the central nervous
system, is the leading cause of non-traumatic neurological disability in young adults (20to 30-years old) in North America and Europe where, current estimates suggest, upwards
of 2.5 million people are living with the disease (Noseworthy et al., 2000; Hauser and
Oksenberg, 2006). Despite disease onset that occurs relatively early in life, a patient’s
lifespan is only modestly shortened; on average, seven to eight years. Thus, patients
suffer for decades with this debilitating disease. Attempts at better understanding the
etiology and progression of MS are imperative to improving these patients’ quality of
life. For decades, MS was thought to be a strictly immune-mediated disease arising from
auto-immunity to components of central nervous system myelin. While the immune
component of the disease is the primary mechanism of demyelination, research has now
begun to focus on the sequelae of demyelination that are believed to contribute to the
permanent and progressive symptoms of MS. This “secondary neurodegeneration”
involves neuronal loss due to lack of trophic support from myelin and the damage to and
14
degeneration of demyelinated axons. Specifically, the loss of chronically demyelinated
axons is the largest contributor to the cognitive decline experienced by MS patients
(Trapp and Nave, 2008).
Multiple sclerosis, as a disease, typically takes one of two trajectories (Figure 2).
The first disease course, experienced by up to 85% of patients, is one characterized by
acute exacerbations (lasting days to weeks) followed by abrogation of symptoms (lasting
weeks to decades). This disease course is known as Relapsing-Remitting Multiple
Sclerosis (RRMS) for obvious reasons. The exacerbations, which are symptomatic of
conduction block (see below), can manifest as any of a number of deficits, including:
limb weakness, gait disturbance, ataxia, double-vision, or vision loss. Upon resolution of
early symptoms, the patients typically return to their original neurological baseline. With
time and repeated relapses, a patient’s neurological status, as measured by EDSS, begins
to worsen between relapses. Approximately 8-20 years after symptom onset, RRMS
patients transition into a second phase of the disease: Secondary Progressive Multiple
Sclerosis (SPMS). In this phase, symptoms continue to progress without abating. This
transition, from RRMS to SPMS, represents a major gap in our understanding of MS.
The other 15% of patients present with an even less-well-understood form of MS, known
as Primary Progressive Multiple Sclerosis (PPMS). PPMS, as its name implies, has as its
hallmark a worsening symptom profile without relapses (Hauser et al., 2006). At present,
none of the therapeutic agents used to treat MS are approved for use in SPMS or PPMS
patients (Ebers et al., 2008).
One of the best tools for evaluating the progression of MS is magnetic resonance
imaging (MRI), which allows for prospective study of lesion development and resolution.
15
While a patient may only experience one or two exacerbations per year, an MRI is able to
detect, on average, as many as 20 inflammatory events per year; a 10:1 ratio of
demyelinating episodes to symptom exacerbation episodes (Barkhof et al., 2000;Filippi
and Rocca, 2006). This disparity between plainly discernable relapses and discrete,
“imaging-only” discernable lesions is why the majority of MS disease progression is
considered to be “clinically silent”(Trapp et al., 2008). The progression of symptoms
during the later stages of disease (following the RRMS to SPMS transition) is likely due
to degeneration of chronically demyelinated axons. A tool that would allow clinicians to
examine the progression of axonal degeneration would provide valuable insight into the
disease process. Further, it would allow clinicians to identify which brain regions are still
functional and, therefore, still treatable. Recently, MRI was used to identify certain types
of demyelinated-associated pathology within human MS brains (Fisher et al., 2007).
Using different imaging modalities (see below), MRI was able to predict which
demyelinated lesions would contain swollen and dystrophic axons. Another important
finding in the study was that a large percentage (~55%) of MRI-predicted regions (using
standard MRI protocols) were not demyelinated. This further highlights the need for
improved correlation between MRI outcomes and MS pathology. Without the use of
MRI, much of the natural history of the progression of MS would still be a mystery.
Briefly, MRI uses powerful magnets to align the protons of the water molecules in
the body. When the magnet is turned off, the protons return to their original orientations.
It is this “relaxation” that is measured by the MRI and processed as an image. The basic
measurements of proton relaxation are T1 and T2, which are time constants. Clinically,
dark spots on T1 scans have earned the sobriquet “black holes” because they are an
16
indicator of tissue damage and loss. Bright spots on T2 scans are regions of
inflammation and edema. The typical progression of a chronic MS lesion might be a
brightly enhancing T2 region for 3-6 weeks, followed in the same location by a persistant
“black hole” on a T1 scan. Using the standard MRI protocols can provide a glimpse into
the progression of the disease, but specialized scanning protocols can help to characterize
disease activity further. For instance, magnetization transfer ratio (MTR) studies how
much energy is transferred from protons bound to myelin to their surrounding water
molecules and pathologically correlated with severe tissue damage (Fisher et al., 2007).
Beyond the acute exacerbations experienced early in disease, the major cause of
progressive disability in MS is axonal degeneration. In this setting, axons are susceptible
to being injured in two distinct ways: axonal transection during inflammation and
degeneration following chronic demyelination. In the setting of acute inflammatory
demyelination, axons are transected, but the exact mechanism by which this transection
occurs is unknown. That the amount of inflammatory activity within a lesion is directly
correlated with the number of transected axons (Hohlfeld, 1997) does suggest that the
mediators of the inflammatory response (ie. macrophages, activated microglia) may be
involved. Also telling is the effect of NBQX, an AMPA/kainate receptor antagonist, in
an animal model of MS – experimental autoimmune encephalitis (EAE). In these
animals, NBQX treatment improved oligodendrocyte survival and minimized the number
of transected axons (Pitt et al., 2000), implicating a role for glutamate in axonal
transection. Striking numbers of axons are transected – up to 11,000/mm3 (Trapp et al.,
1998) – during acute inflammatory demyelination, likely as a bystander effect(Trapp and
Stys, 2009). Despite the massive loss of axons, much of this transection can be clinically
17
silent, depending on the location of the lesion. While extensive axonal transection may
occur during RRMS, it is the setting of chronic demyelination, more commonly found in
SPMS, that is permissive for axonal degeneration. Axonal degeneration, as distinct from
axonal transection, is temporally removed from the acute phase of the disease. A
chronically demyelinated axon, for reasons already discussed (see above), is functionally
impaired. Following demyelination, extensive submembranous cytoskeletal remodeling
occurs, allowing the lateral diffusion of the previously-tethered NaV and KV channels
along the entire demyelinated length of the axon. The extensive remodeling of the
demyelinated axolemma perturbs the ionic gradient that allows for rapid and repeated
action potential conduction. In the absence of the myelin sheath, the axon is fully
exposed to the extracellular compartment, including the production of nitric oxide (NO)
by nearby microglia and astrocytes. NO, a highly soluble gas, diffuses into the axon and
disrupts adenosine triphosphate (ATP) production by mitochondria. Without ATP, the
Na/K ATPase cannot maintain efforts to repolarize the axolemma – by pumping 3 Na
ions out of the cell in exchange for pumping 2 K ions in, while hydrolyzing one molecule
of ATP to ADP. Failure of the Na/K ATPase, perhaps the final straw, leads to an
unchecked rise in intracellular Na, causing the reversal of the sodium-calcium exchanger
(NCX) which, in turn, ushers in Ca ions to replace the Na ions that it extrudes. This
increase in Ca leads to the activation of destructive processes, including the activation of
calpains (for Review, see (Trapp et al., 2009)).
The failure of the Na/K ATPase is one of the final steps that lead to axonal
degeneration. A better understanding of the pump and its role in disease would help to
give insight into whether or not the sodium pump is a viable target for therapeutic
18
intervention. In an effort to address these issues, this dissertation describes the normal
location of the Na/K ATPase in myelinated human CNS axons and the distribution of
Na/K ATPase after both acute and chronic demyelination. Further, MRI techniques have
been used to identify MS lesions based on their Na/K ATPase content. These findings
represent a significant advance in understanding the degeneration of chronically
demyelinated axons.
19
FIG1
20
Figure 1. Structural and functional properties of central nervous system axons. Central
nervous system (CNS) axons are periodically wrapped by myelin, a high-resistance
multi-lamellar membrane outgrowth from oligodendrocytes (left). Nodes of Ranvier are
formed as gaps are left between myelin sheaths (left & lower right). These nodes are
essential for the rapid and repetitive conduction of action potentials along the axon, by a
property known as saltatory conduction (upper right). The internode is the highresistance, low-capacitance region, limited on both ends by nodes, along which action
potentials may travel without attenuation between the low-resistance, high-capacitance
Nodes of Ranvier. This specialization allows the action potential to appear,
electrophysiologically, as though it is hopping from node to node. Where the myelin
contacts the axon, there are other regional specializations – namely, the paranode and the
juxtaparanode (JXP, lower right). These regions, so named for their proximity to the
nodes, have unique membrane properties and are critical to the maintenance of axo-glial
signaling. Finally, the periaxonal space (lower right, pink) is a 12-20nm space that is
formed by the periaxonal membrane of the inner tongue process of the myelin sheath and
the axolemma. It is limited on both ends by the paranodal loops.
21
FIG 2
22
Figure 2. A graphic representation of disease progression and disability by disease
subtype. Relapsing-Remitting Multiple Sclerosis (RRMS, green) presents as a disease
course with occasional clinical exacerbations followed by remissions. Each remission,
however, is scored progressively higher on the Expanded Disability Status Scale (EDSS),
an indicator of neurological disability. For each clinical exacerbation, up to ten
inflammatory events may be identified by magnetic resonance imaging (MRI, arrows).
The remissions of RRMS eventually cross an imaginary line (Clinical Threshold, red) at
which point the disease converts from RRMS to Secondary Progressive Multiple
Sclerosis (SPMS, yellow). After the conversion to SPMS, clinical exacerbations are
fewer, as are MRI-events, however symptom progression continues – as denoted by a
steady increase in EDSS. Primary Progressive Multiple Sclerosis (PPMS, blue) is a less
common variant of MS. Presenting in healthy patients, PPMS has as its hallmark a
steady progression of symptoms without acute exacerbations. The X-axis, denoting
time, provides the average number of years that one disease state may persist.
23
CHAPTER 2. IMAGING CORRELATES OF DECREASED AXONAL
NA/K ATPASE IN CHRONIC MULTIPLE SCLEROSIS LESIONS
(Young, et al., 2008)
24
INTRODUCTION
Multiple sclerosis (MS) is a demyelinating disease of the central nervous system
(CNS) that leads to irreversible neurological decline. MS afflicts more than 1.5 million
people in North America and Europe, where it is the leading cause of non-traumatic
neurological disability in young adults (Noseworthy et al., 2000). Degeneration of
chronically demyelinated axons is now considered to be a major contributor to the
permanent neurological disability that most MS patients endure (Bjartmar et al., 2000;
Waxman, 2006). The preservation of axons within chronic lesions of MS presents a
unique therapeutic challenge. Although we are beginning to elucidate potential
mechanisms for this axonal degeneration (Trapp et al., 1998), we have no means by
which to measure this axonal pathology in a clinical setting.
Myelin is a tightly compacted membrane spiral that surrounds axons in the central
and peripheral nervous systems. During formation of the myelin sheath, voltage-gated
sodium (Nav) channels are concentrated at the nodes of Ranvier; small, unmyelinated
axonal segments that separate adjacent myelin internodes. Because axonal membrane
depolarization only occurs at the nodes, conduction velocities of myelinated axons are
approximately 100 times faster than those of unmyelinated axons where Nav channels are
diffusely distributed(Waxman, 1977). After each depolarization, the Na/K ATPase
rapidly exchanges axonal Na for extracellular K in an energy-dependent manner. Rapid
repolarization permits rapid and repetitive axonal firing, a necessary condition for proper
25
neuronal function. When axons are demyelinated, the Nav channels diffuse away from
the nodes(Bostock et al., 1983;Waxman et al., 2004). This Nav channel redistribution
increases both Na influx during impulse conduction and the demand for ATP during
repolarization of the axolemma. Reduced ATP production and Na/K ATPase dysfunction
are thought to initiate a cascade of ionic imbalances that lead to degeneration of
chronically demyelinated axons(Dutta and Trapp, 2007). Specifically, as axonal Na
levels increase, the Na/Ca exchanger operates in reverse and exchanges axoplasmic Na
for extracellular Ca(Baker and McNaughton, 1976;Stys et al., 1992;Stys et al., 1993).
Increased axonal Ca will activate proteolytic enzymes and eventually lead to
degeneration of chronically demyelinated axons, and by extension, to progressive
neurological decline during the latter stages of MS. Despite the potentially important role
of axonal Na/K ATPase during axonal degeneration in MS, and a report detailing a loss
of Na/K ATPase enzymatic activity in chronic MS lesions(Hirsch and Parks, 1983),
axonal Na/K ATPase distribution has not been compared in myelinated and demyelinated
axons.
Currently, magnetic resonance imaging (MRI) is the most commonly used tool for
diagnosis and monitoring of MS. Because of its high sensitivity, MRI is an invaluable
method for following the subclinical progression of the disease. Unfortunately,
conventional MRI is limited by its low pathological specificity. Postmortem imaging
studies have shown that classification of MS lesions, based on multiple imaging
modalities, can improve correlations between MRI and specific aspects of
pathology(Bruck et al., 1997;Schmierer et al., 2004;Fisher et al., 2007). For example,
hypointensity on T1-weighted images correlates with axonal loss(van Walderveen et al.,
26
1998), and decreased magnetization transfer ratio (MTR) is correlated with
demyelination(van Waesberghe et al., 1999). Recently, we have utilized a composite of
T2, T1 contrast ratio, and MTR intensity characteristics to identify chronic lesions of MS
with axonal loss and increased demyelinated axon diameter(Fisher et al., 2007).
Techniques such as diffusion tensor imaging and magnetic resonance spectroscopy also
provide valuable information on underlying pathology(Bammer et al., 2000), particularly
for axonal pathology, but these studies are more difficult to perform in a typical clinical
setting. As yet, there are no reliable MRI markers that are both specific for axonal
pathology and feasible to measure on a routine basis.
It is well established that much of the disease process in MS patients is clinically
silent. As such, brain imaging has become the major surrogate marker of disease activity
and the best predictor of disease progression. In this study, we determined the
distribution of Na/K ATPase subunits in normal human brain and in demyelinated lesions
from brains of patients with MS. Subunits α1, α3, and β1 were detected on the internodal
axolemma and absent from the nodal axolemma of myelinated fibers. In acutely
demyelinated brain tissue, all three subunits were retained on demyelinated axolemmas.
In contrast, 58% of chronically demyelinated lesions of MS contained less than 50%
Na/K ATPase-positive axons. We then compared Na/K ATPase measurements and
postmortem MRI in a subset of chronic lesions, and found a linear correlation between
the percentages of demyelinated axons without Na/K ATPase and decreased MTR and T1
contrast ratios.
27
SUBJECTS AND METHODS
Tissue and Lesions
Brains were obtained from 13 patients with MS (Table 1). Postmortem, in situ
MRI was collected on 11 brains, as described previously(Fisher et al., 2007). Brains
were removed and placed in fixative after an average postmortem interval of 5.9 hours.
Left and right hemispheres were separated and processed differently. In brains with
postmortem MRI, one hemisphere was fixed in 4% paraformaldehyde (PFA) for 4 weeks,
and then rescanned and sliced to ensure coregistration of lesion location. Lesions with
MRI data came from this hemisphere (n = 7 hemispheres). The remaining brains or
hemispheres were cut into 1cm-thick slices and placed in either 4% PFA of rapidly
frozen. Lesions without MRI data came from this hemisphere (n = 8 hemispheres).
Lesions were identified macroscopically in fixed slices, blocked, cryoprotected, frozen,
and sectioned (30µm thick) on a freezing-sliding microtome. Sections were stained for
myelin proteins and major histocompatibility complex class II, and classified as acute,
chronic active, or chronic inactive, as described previously. Three acute and 36 chronic
lesions were identified. Control brains (n = 4) were collected and processed as described
earlier, but no MRI was obtained.
Magnetic Resonance Imaging and Image-Guided Tissue Sampling
Details of the methods used for MRI-pathology correlations and image-to-tissue
coregistration have been described previously(Fisher et al., 2007). In brief, MRIs were
obtained postmortem, using a standardized protocol that included a T2-weighted fluid
28
attenuated inversion recovery image (FLAIR), a T1-weighted spin-echo image, a pair of
images to calculate MTR, and a high resolution magnetization prepared rapid gradient
echo (MPRAGE) image for coregistration purposes. The images were analyzed to
segment T2 lesions (hyperintense on FLAIR), and each T2 lesion was further classified
based on T1 and MTR characteristics. T1 and MTR contrast ratios were calculated for
each region as the mean intensity within the region divided by the mean intensity of the
normal-appearing (nonlesional) white matter (NAWM) in the same slice. Based on MTR
and T1 contrast ratios, 20 demyelinated lesions from 7 brains were selected for Na/K
ATPase immunostaining.
Immunocytochemistry
Fixed tissue was cryoprotected, frozen, and sectioned with a freezing-sliding
microtome (30µm thick). Sections were rinsed in phosphate-buffered saline (PBS),
microwaved in 10mM citric acid buffer (pH 6.0), and incubated in 3% hydrogen peroxide
and 1% Triton X-100 (Sigma, St. Louis, MO) in PBS (pH 7.4) for 30 minutes. Doubleand triple-labeled sections were pretreated as described above (but without hydrogen
peroxide), immunostained for 3-5 days at 4ºC, and then incubated with fluorescently
conjugated secondary antibodies, as described previously(Chang et al., 2002). Sections
were mounted and coverslipped with Vectashield (Vector Labs, Burlingame, CA).
29
Antibodies
The antibodies used in this study are well-characterized and include rabbit antimyelin basic protein (1:500 dilution; Dako, Glostrup, Denmark); mouse anti-human
major histocompatibility complex class II (1:250 dilution; Dako); mouse anti-chicken
Na/K ATPase α1 (clone A6F; 1:50 dilution; Developmental Studies Hybridoma Bank,
University of Iowa, Iowa City, IA); goat anti-Na/K ATPase α3 (clone C-16; 1:100
dilution; Santa Cruz Biotechnology, Santa Cruz, CA); mouse anti-Na/K ATPase β1
(clone M17-P5-F11; 1:100 dilution; Affinity BioReagents, Golden, CO); rabbit antineurofilament light, medium, and heavy chains (1:2000 dilution for light and medium and
1:1000 dilution for heavy; Serotec, Raleigh, NC); and rabbit anti-Caspr (1:250 dilution;
generous gift from James S. Trimmer, University of California at Davis, Davis, CA).
Confocal Microscopy
Fluorescently labeled sections were scanned with a Leica SP5 confocal
microscope (Leica, Heidelberg, Germany). Laser intensities were adjusted to eliminate
channel cross talk. Single optical slices were used for quantitation (Figs. 1D-G, 2, and 4;
Figs. 1A-C represent projected z-stacks of three to five slices covering 1-3µm in depth).
Quantification of Pathology
Percentages of Na/K ATPase-labeled neurofilament-positive axons were
quantified in control brain, acute and chronic demyelinated lesions, and normal-appearing
white matter (NAWM) from individuals with MS from images obtained by confocal
microscopy. Six fields (one single optical section was obtained using a 100X NA 1.3
30
objective lens and 1.96X zoom, resulting in an area of 79.4µm2) were counted for each
lesion or area of NAWM (an average of approximately 108 axons for each field), and
data are presented as the mean of 6 fields. An axon was considered positive for Na/K
ATPase if it had a neurofilament-positive core and a Na/K ATPase-positive region
surrounding the core. An axon was considered negative if it had a neurofilament-positive
core and no detectable (signal above background) Na/K ATPase labeling. Differences
between groups were determined using a Kruskall-Wallis one-way analysis of variance
with Dunn’s multiple-comparisons post-test. Standard deviation from mean was
calculated for each lesion. Fields were scored by two individuals blind to the lesion stage
and corresponding MRI characteristics. Images were taken from lesion core, an area
equidistant from all myelinated borders.
To assess the relation between Na/K ATPase and MTR or T1 contrast ratios, we
used a linear mixed model. The model also accounted for correlated data, to allow for
grouping of multiple regions of a single brain.
31
RESULTS
Na/K ATPase Subunits alpha-1, alpha-3, and beta-1 are Located in Internodal Axolemma
To determine the normal distribution of Na/K ATPases, we colocalized subunits
alpha-1, alpha-3, and beta-1 with Caspr, a protein enriched in paranodal axolemma. As
described previously(Einheber et al., 1997;Rios et al., 2000), Caspr antibodies intensely
stained paranodal regions of central nervous system myelinated fibers (Figure 1, panels
A-C, red). Both alpha-1 and alpha-3 subunit antibodies displayed specific and relatively
uniform labeling of the internodal axolemma, with little or no detectable
immunoreactivity at the nodal (arrows) or paranodal axolemma (Figure 1, panels A & B
respectively, green). Na/K ATPase internodal immunoreactivity did not appear to differ
between large or small axons. In sections double-labeled with alpha-1 or alpha-3 Na/K
ATPase and myelin basic protein antibodies, Na/K ATPase immunoreactivity was
associated with the internodal axolemma and did not appear to be present in the
surrounding myelin sheath (Figure 1, panel D). When sections were double-labeled with
alpha-1 and alpha-3 antibodies, most axons contained both alpha-1 and alpha-3
immunoreactivity, although the relative alpha-1 and alpha-3 axolemmal staining intensity
appeared to vary between some axons (Figure 1, panels E & F respectively). The Na/K
ATPase beta-1 subunit had similar distribution to alpha-1 and alpha-3 subunits, labeling
the internodal but not the nodal or paranodal axolemma (Figure 1, panel G).
32
Distribution of Na/K ATPase Subunits in Multiple Sclerosis Lesions
The distribution of Na/K ATPase alpha-1 and alpha3 subunits was determined in
39 MS lesions (3 acute and 36 chronic). Lesion stage was based on the distribution and
density of major histocompatibility complex class II-positive cells, as described
previously. Sections were stained with Na/K ATPase, myelin basic protein, and
neurofilament antibodies, and the percentage of Na/K ATPase-positive demyelinated
axons were quantified in sections double-labeled for Na/K ATPase (a cocktail of both
alpha-1 and alpha-3 subunit-specific antibodies) and neurofilament. In the three acute
and 15 of the 36 chronic lesions (42%), Na/K ATPase alpha-1 and alpha-3 antibodies
stained demyelinated axolemmas with a similar distribution and intensity to that found in
myelinated axons in brain sections from control and MS patients (Figure 2, panels A & B
respectively). In contrast, axons in 21 of the 36 chronic MS lesions (58%) had less than
50% Na/K ATPase alpha-1 and alpha-3-positive axons (Figure 2, panel C). In lesions
containing Na/K ATPase-negative axons, axonal Na/K ATPase staining was present in
myelinated fibers at the lesion edge, and the transition between Na/K ATPase-positive
and –negative staining correlated with the presence and absence of myelin (Figure 2,
panel C). These patterns of labeling were consistent in both diaminobenzidine- and
fluorescence-labeled sections. In one brain slice, one chronic lesion contained intense
Na/K ATPase labeling, whereas another chronic lesion had little to no labeling (as
demonstrated in Figure 5, panels A & B). Na/K ATPase immunoreactivity was either
consistently present (Figure 2, panel D) or consistently absent (Figure 2, panel E) along
33
the length of demyelinated axons within the lesion core. The percentage of Na/K
ATPase-positive axons was similar between control white matter, normal-appearing
white matter from MS brains, and acute demyelinated lesions from MS brains (Figure 3).
In contrast, in chronic MS lesions, the proportion of Na/K ATPase-positive axons ranged
from 100% to 0% (Figure 3). Both control (p < 0.0001) and normal-appearing white
matter (p < 0.001) groups were significantly different from the chronic group. The
significance of the difference between acute and chronic groups could not be tested
because of the small sample number of acute lesions. The average standard deviation (6
fields/lesion) for the 39 lesions studied was 4.8. The bimodal distribution of Na/K
ATPase-positive axon percentage in chronic lesions did not correlate with their
immunohistochemical classification (i.e., lesions with greater percent Na/K ATPasepositive axons were not necessarily chronic active, lesions with lower percent Na/K
ATPase-positive axons were not necessarily chronic inactive). Concordance between
measurements made by two observers was high (0.993), indicating excellent
reproducibility.
Lesions were also double-labeled with alpha-1 or alpha-3 Na/K ATPase and
neurofilament antibodies to determine whether demyelinated axons preferentially express
either subunit. Demyelinated axons expressed both or neither subunit (Figure 1, panels
E-G), with rare exceptions (data not shown). In all stained sections included in this
study, normal-appearing white matter served as an internal control for Na/K ATPase
labeling. Thus, the differential pattern of Na/K ATPase labeling in lesion subtypes
appears to be a valid observation and not due to staining artifact or tissue processing.
Because only a small proportion of a brain’s total lesion load was studied, it was not
34
possible to correlate patient sex, duration of disease, or Expanded Disability Status Scale
score with percentage of Na/K ATPase-negative demyelinated axons.
Imaging Correlates of Axonal Na/K ATPase
One possible explanation of the data described earlier is that chronic
demyelination leads to loss of axonal Na/K ATPase. We recently correlated axonal loss
and increased demyelinated axon diameter with reduced T1 contrast ratios and reduced
MTR in postmortem MS brains. In this study, we investigated whether postmortem MRI
measurements, selected to represent opposite ends of a scale of pathological severity,
could identify MS lesions with (Figure 5, panel A) and without (Figure 5, panel B)
axonal Na/K ATPase. Twenty of the 39 demyelinated lesions had postmortem MRI data
available. We compared T1 and MTR contrast ratios with the percentage of axons
without Na/K ATPase. As demonstrated in Figure 5, panel C, demyelinated lesions with
MTR contrast ratios greater than 0.8 had normal Na/K ATPase distribution and lesions
with ratios less than 0.7 had less than 20% of axons with Na/K ATPase. Likewise,
demyelinated lesions with T1 contrast ratios greater than 0.85 had normal Na/K ATPase
distribution and lesion with T1 contrast ratios less than 0.75 had less than 20% of axons
with Na/K ATPase (Figure 5, panel D). To ensure that lesions were not clustered with
other same-brain lesions, we identified lesions from each brain by a unique color symbol
in Figure 5, panels C & D. There was a statistically significant linear relationship
between percentage axons with Na/K ATPase and both MTR (p <0.0001) and T1 contrast
35
ratio (p < 0.0006). These observations are the first to correlate MRI characteristics with
molecular properties of axons.
DISCUSSION
These data impact our understanding of the pathogenesis of permanent
neurological disability during chronic stages of the disease process of MS in two ways.
Most importantly, axons in 58% of the 36 chronically demyelinated lesions of MS
examined in this study contained less than 50% Na/K ATPase-positive axons. Secondly,
quantitative MRI of a subset of these lesions (20) was able to differentiate chronic lesions
with axonal Na/K ATPase from those without. Axons lacking Na/K ATPase cannot
efficiently transmit action potentials. Reduced exchange of axonal Na/ for extracellular
K will also increase axonal Na concentrations, which will, in turn, reverse the Na/Ca
exchanger. Although this will increase axonal Ca and contribute to Ca-mediated axonal
degeneration, our data support the concept that many chronically demyelinated axons are
nonfunctional before degeneration. We propose that loss of axonal Na/K ATPase is a
contributor to the progressive neurological decline that most MS patients eventually face,
and that quantitative MRI may provide a valuable predictor of this process in longitudinal
studies of MS patients.
We used well-characterized antibodies to localize the neuronal Na/K ATPase
subunits to internodal axolemma. The internodal distribution of Na/K ATPase spatially
uncouples Na influx at the nodes from Na efflux at internodes and identifies the
periaxonal space as an important regulator of membrane potentials during nerve
36
conduction. The periaxonal space is a cloistered, 12- to 14nm-wide, extracellular space
defined by the axolemma, periaxonal membrane of the myelin internode, and the septate
junctions, which tether the paranodal loops to the axon(Rios et al., 2000). Na and K
exchange through the periaxonal space implicates the myelin/oligodendrocyte unit as an
active player in nerve conduction, rather than a passive insulating participant that renders
the internodal axolemma inert to the dynamics of ion exchange and nerve transmission.
Such a function, however, should require the presence of ion channels or pumps on both
the axolemmal and myelin sides of the periaxonal space. To date, such ion channels or
pumps have not been localized to the periaxonal membrane of the central nervous system
myelin internode. If present, they would likely be enriched in the membranes of the inner
tongue process, the small cytoplasm-containing tube that extends the length of the myelin
internode and is contiguous with the cytoplasm of the paranodal loops. The inner tongue
process occupies less than 20% of the periaxonal surface in mature central nervous
system internodes; the remainder of the periaxonal membrane is “fused” with the
cytoplasmic surface of the first compact myelin lamella.
The percentage of axons without Na/K ATPase varied from lesion to lesion within
and between postmortem brains (Figures 2 & 3). Most demyelinated axons in acute MS
lesions contained Na/K ATPase distribution similar to myelinated areas, whereas a subset
of chronic lesions contained a large percentage of axons without Na/K ATPase.
Pathologically, chronic lesions are categorized as active or inactive based on the presence
of absence of activated immune cells at the lesion border(Lassmann et al., 1998). In this
study, the loss of Na/K ATPase was not preferentially associated with either chronic
active or chronic inactive lesions. Based on postmortem MRI analysis, we recently
37
classified pathological characteristics of MS lesions(Fisher et al., 2007). Compared with
T2-only demyelinated lesions, lesions that were abnormal by T2, T1, and MTR
(“T2T1MTR lesions”) contained fewer axons, and these axons were swollen (increased
axonal diameter). We show here that the percentage of Na/K ATPase-positive axons in
demyelinated T2-only lesions (Figure 4) was similar to that of myelinated axons. In
addition, quantitative MTR and T1 contrast ratios delineated chronic MS lesions with and
without detectable axonal Na/K ATPase, and those axons without Na/K ATPase were
swollen (compare axons in Figure 5, panels A & B). A decrease in the MTR or T1
contrast ratio characteristics, therefore, in indicative of the presence of swollen axons and
a loss of axonal Na/K ATPase. Chronic loss of Na/K ATPase is, then, likely to
contribute to a persistent ion imbalance across the axolemma, which eventually leads to
water influx, axoplasmic swelling, and possibly, degeneration.
Although the acute symptoms of MS are not likely to be directly correlated with a
depletion of axonal Na/K ATPase, loss of the pump on chronically demyelinated axons
will contribute to the continuous neurological decline experienced by most MS patients.
Based on a dystrophic(Chang et al., 2002) and/or swollen(Yin et al., 2004) appearance
and reduced NaV channels(Black et al., 2007), it has been proposed that many
demyelinated axons in chronic MS lesions may be functionally compromised before
degeneration. The loss of Na/K ATPase on chronically demyelinated and swollen axons
described here provides additional support to this hypothesis. In addition, the ability of
quantitative MTR and T1 contrast ratios to identify MS lesions with little or no axonal
Na/K ATPase raises the possibility that noninvasive brain imaging techniques may
38
monitor and predict neurological decline and efficacy of neuroprotective therapies in
patients with MS.
39
Table 1
40
Figure 1.
41
Figure 1. Na/K ATPase is enriched in the internodal axolemma of myelinated axons in
the adult human brain. Na/K ATPase subunits α1 (A, green), α3 (B, green), and β1 (C,
green) are located at the axolemma of myelinated axons. As demarcated by the
paranodal marker Caspr (A-C, red), the Na/K ATPase is enriched in the internodal axon
and not detected in paranodal regions or nodal axolemma (A-C, arrows). Na/K ATPase
(D, green) was clearly expressed along the axolemma and below myelin sheaths stained
with myelin basic protein antibodies (D, red). The α subunits displayed differential
labeling in some axons (E, α1, green; F, α3, red). In the majority of axons, the α1 (red)
and α3 (green) subunits colocalized (G). Scale bars = 5µm.
42
Figure 2.
43
Figure 2. Demyelinated axons in some chronic multiple sclerosis (MS) lesions lack Na/K
ATPase. Myelin basic protein (MBP) immunoreactivity (white in A-C; blue in A”-C”)
identifies myelinated axons and the lesion border. Neurofilament (NF) staining (red)
labels myelinated and demyelinated axons. Na/K ATPase (green, A-E) distribution is
continuous along demyelinated axons in acute lesions (A’, A”) and a subset of chronic
lesions (B’, B”). In contrast, Na/K ATPase immunostaining is absent from other chronic
lesions (C’, C”) but present along myelinated axons at the border of the lesion (C’, C”).
Staining for Na/K ATPase was consistently present (D, from lesion in A) or absent (E,
from lesion in C) along individual axons within the core of a demyelinated lesion. Scale
bars = 40µm (A-C); 5µm (D, E).
44
Figure 3.
45
Figure 3. Quantification of Na/K ATPase-positive axons in control human brain and
multiple sclerosis (MS) lesions. The percentages of Na/K ATPase-positive axons in
control, normal-appearing white matter (NAWM), and acute MS lesions were closely
grouped; all containing high percentages of Na/K ATPase-positive axons. In contrast, in
chronic MS lesions, the percentage of Na/K ATPase-positive axons varied, with 58% (21
of 36) having Na/K ATPase-positive axon percentages of less than 50%. Statistical
analysis (analysis of variance with Dunn’s multiple-comparisons posttest) confirmed a
significant difference between both the control (p < 0.0001) and NAWM (p < 0.001)
groups and the chronic lesion group. Statistical analysis of the difference between the
acute lesion group and the chronic lesion group was impossible due to small group sizes.
Horizontal lines indicate the mean value for each group. Control tissue, n = 4; NAWM, n
= 7; acute lesion, n = 3; chronic lesion, n = 16.
46
Figure 4.
47
Figure 4. Percentages of Na/K ATPase-positive axons are predicted by Magnetic
Resonance Imaging (MRI). Selected to represent opposite ends of a severity scale,
demyelinated T2-Only lesions (n = 5) have a higher percentage of axons with Na/K
ATPase, as compared to T2T1MTR lesions (n = 8) which have a low percentage of axons
with Na/K ATPase.
48
Figure 5.
49
Fig 5. Magnetization transfer ratios (MTRs) and T1 contrast ratios linearly correlate with
the percentage of Na/K ATPase-positive axons in chronic MS lesions. The percentage of
Na/K ATPase-positive axons in chronically demyelinated MS lesions were correlated
with quantitative postmortem MTR and T1 contrast ratios. (A, B) Chronically
demyelinated lesions stained for Na/K ATPase (green) and neurofilament (red). The
percentage of Na/K ATPase-positive axons varied from near 100 (A; MTR =
0.9) to 0% (B; MTR = 0.5). Many axons without Na/K ATPase had increased diameters
(B). A comparison of the percentage of Na/K ATPase-positive axons in chronically
demyelinated MS lesions were correlated with quantitative postmortem MTR (p <
0.0001; C) and T1 contrast ratios (p < 0.0006; D). Dashed lines denote a fit curve to the
regression coefficient. Each data point is from a single lesion and each unique colorsymbol combination denotes one of the brains studied. Scales bars = 5µm.
50
CHAPTER 3. GENERAL DISCUSSION
51
This dissertation has made several seminal contributions to the field of
Neuroscience and, specifically, neurodegeneration as it relates to multiple sclerosis. The
first major finding is the identification of the Na/K ATPase as an internodal protein of the
axolemma. Commonly thought to have been located in the node, the presence of the
Na/K ATPase in the internode suggests a variety of new, non-pumping functions such as
structural maintenance of the internode (Young et al., 2008) or as a barrier to lateral
diffusion of nodal or paranodal structures (Doi, 2009). Being the first large (10-pass)
transmembrane protein in the internodal axolemma, the Na/K ATPase also provides an
excellent substrate for immunohistochemical detection of the axolemma. The second
major finding of this dissertation is the loss of Na/K ATPase from a subset of chronically
demyelinated axons. The Na/K ATPase is required for the optimal function of all cells.
Without the Na/K ATPase, axonal ion homeostasis is lost – and, perhaps most
importantly – a cascade of destructive processes is begun. The presence of axons lacking
Na/K ATPase in chronically demyelinated axons suggests that the axonal degeneration
that occurs does so after the axon is already functionally dead. Finally, the finding that
MTR and T1 contrast ratio can be used to identify demyelinated lesions lacking Na/K
ATPase highlights the power of MRI-pathology correlations as well as the immense
value that such studies might have for the clinicians and pharmacologists.
52
Internodal Na/K ATPase: A new model of axonal function/dysfunction
The decades-old assumption that the Na/K ATPase is located in the Node of
Ranvier seemed to make sense; to spatially couple sodium influx through Na(v) channels
during action potentials with sodium efflux through Na/K ATPase following action
potentials. It made so much sense that, despite having only questionable evidence to
support a nodal location, the Na/K ATPase has been drawn into countless figures (both in
peer-reviewed articles and textbooks) as being a major component of the nodal
axolemma. Using electron microscopy and immunocytochemistry, one group published
evidence to contradict the nodal location (Mata et al., 1991), but the evidence was largely
ignored owing to difficult-to-interpret electron microscopy images. The data presented in
this dissertation (Young et al., 2008) provide the first substantive and reproducible
evidence that the axonal Na/K ATPase is an integral internodal membrane protein in
human CNS axons. Further, this study identified methods to locate lesions in MS brains
by MRI that have lost Na/K ATPase from their axons. Taken together, these two
findings represent a major advance towards understanding how demyelinated axons
malfunction and degenerate and a potential method by which to identify chronic lesions
in patients and provide a prognosis for treatment outcomes and disease progression.
Additionally, the discovery of the Na/K ATPase as a major internodal membrane protein
of the axolemma has provided a useful biochemical marker for the axonal membrane; a
resource that will undoubtedly aid neuroscientists in the future.
Given the internodal location of the axonal Na/K ATPase(Young et al., 2008), a
new understanding of how the axon both functions and dysfunctions must be gleaned
(Figure 1).
53
Functions
Specifically, the spatial uncoupling of sodium influx and efflux raises many
questions about the axoplasmic milieu during and after an action potential. Having
already established the lack of diffusion barriers in the axoplasm(Arhem, 1976), the
sodium influx that occurs at the node likely diffuses in both directions away from the
node. The presence of Na/K ATPase along the entire length of the internode, in spite of
the high ATP cost to the cell, must be required for axonal homeostasis. Either large
quantities of sodium flow into the cell during action potential propagation, requiring
hundreds to thousands of Na/K ATPases at the ready, or the internodal Na/K ATPase has
additional functions. Perhaps, as mentioned in Chapter 2, the large, 10-transmembrane
domain α-subunit protein provides stability to an otherwise structurally poor region of the
axon. At present, no other large integral membrane proteins have been identified in the
internodal axolemma. It has been suggested a large percentage of surface-expressed
Na/K ATPase is, in fact, a non-pumping pool that is responsible for other cellular
functions such as acting as receptors for endogenous cardiotonic steroids (Liang et al.,
2007). It is also possible that the Na/K ATPase β-subunit is dynamically regulating
cellular events, as certain signaling cascades rely on the intracellular domains of the β
subunit and the β2 subunit has been identified as a cellular adhesion molecule on
Schwann cells (Gloor et al., 1990). Additionally, it has been found that, while the αsubunit remains embedded in the membrane, the β-subunit is only required for membrane
insertion and can be rapidly internalized and degraded (Yoshimura et al., 2008). The
functional impact that this β-subunit degredation has on the catalytic α-subunit is unclear.
54
Dysfunctions
In the face of demyelination, it isn’t the internodal location of the axonal Na/K
ATPase that raises concern, but the fact that the Na/K ATPase will pump sodium and
hydrolyze ATP until one or the other runs out – and in a denuded axon, the limiting factor
is far more likely to be ATP than sodium. The diffusion of Na(v) channels from the
nodes along the entire length of the demyelinated axon leads to a global increase in
sodium influx and, therefore, intracellular sodium. Since most axons in the human CNS
have comparable amounts of both α1 and α3 subunits, sodium is pumped from the cell no
matter what the cellular ATP conditions are – if the cellular ATP is high and intracellular
sodium is low, such as it is during times of relative cellular quietude, the α1 subunit is
available to pump; if the cellular ATP is low but intracellular sodium is high, such as it is
during times of cellular stress, the α3 subunit will utilize all of the ATP it can access.
This high-efficiency, high-demand sodium extrusion system, the main function of which
is to protect the axon by maintaining ionic homeostasis, can itself be one of the main
factors contributing to ATP-debt and axonal degeneration.
If presence of the Na/K ATPase can be destructive, then it stands to reason that
absence of the sodium pump might be protective – and it is, in many cell types (see
below). However, in an axon, where ionic fluxes are relentless, loss of the Na/K ATPase
may be the final straw before the commencement of axonal degeneration. Almost as a
built-in fail-safe, the sodium-calcium exchanger (NCX) is able to reverse its normal
function (which is to import sodium ions and export calcium ions) to allow for the
extrusion of excess sodium ions. Perhaps unintentionally deleterious, the importing of
55
calcium ions is almost certainly the second step in the march towards axonal
degeneration. The accumulation of intracellular calcium, due to both NCX reversal and
release from intracellular stores, begins a cascade that destroys the axon from the inside
out (Figure 1 (Trapp et al., 2009)).
Loss of Na/K ATPase: Lessons from other cell types
Two major questions left unanswered by this study are: 1) Why is the Na/K
ATPase missing from the axolemma of axons within approximately 50% of chronically
demyelinated axons? and 2) How is the Na/K ATPase removed from the axolemma of
these axons? To answer these two questions, it is best to look to other cell types in which
similar phenomena have been described and then apply the lessons learned to
demyelinated axons.
Why is the Na/K ATPase missing?
In lung and kidney cells, both osmotically and ionically active cells, there are
descriptions of cell membrane Na/K ATPase being actively removed from the membrane
during times of cellular stress. In the lung, Na/K ATPase enzymatic function is
decreased in the alveolar epithelium under conditions of hypoxia (Campbell et al.,
1999;Azzam et al., 2001;Dada et al., 2003;Chen et al., 2006;Dada et al., 2007). In the
kidney, a similar phenomenon has been described in the proximal and distal tubules in
the presence of dopamine (Done et al., 2002). It appears that Na/K ATPase removal
from the membrane has evolved as a protective strategy to avoid over-taxing an alreadystressed cell. The ATP expenditure of pumping even a temporary surge in intracellular
56
sodium may be too great for some cells to bear, and so removing the Na/K ATPase from
the membrane will prevent a cellular energy crisis.
How is the Na/K ATPase removed from the membrane?
In the aforementioned cell types (see above), Na/K ATPase is actively removed
from the membrane to avoid leading to ATP-debt during times of cellular stress. In both
alveolar epithelium and renal tubule endothelium, the Na/K ATPase is endocytosed using
a stereotyped pathway. Nicknamed the PURED pathway(Lecuona et al., 2007), the
pathway involves phosphorylation, ubiquitination, recognition, endocytosis, and
(occasionally) degradation.
Phosphorylation While the sequence of the catalytic α-subunit has many consensus
phosphorylation sites (Beguin et al., 1994), both alveolar epithelial and kidney tubule
epithelial Na/K ATPase has been demonstrated to require phosphorylation at Ser-18 for
induction of the endocytic pathway (Done et al., 2002;Dada et al., 2003;Chen et al.,
2006;Dada et al., 2007). Phosphorylation at this site does not impact Na/K ATPase
activity, but it does lead to a conformational change in the N-terminus that is required for
endocytosis (Chibalin et al., 1999). This conformational change is likely required to
allow for the activation of a PI 3-kinase (Yudowski et al., 2000).
Ubiquitination In lung alveolar epithelial cells (and not renal tubule epithelial cells),
following Ser-18 phosphorylation, the Na/K ATPase is ubiquitinated (Comellas et al.,
2006;Dada et al., 2007). In these cells, it appears that ubiquitination is a required secondstep in the sequential process of Na/K ATPase endocytosis.
57
Recognition & Endocytosis Na/K ATPase endocytosis is a clathrin-dependent
mechanism in both lung alveolar epithelial cells and renal tubule epithelial cells. The
activated PI 3-kinase and phosphorylated Ser-18 work in concert to recruit the adaptor
molecule AP-2 (Done et al., 2002). Once bound to AP-2, the Na/K ATPase is
endocytosed.
Degradation Degradation of the Na/K ATPase is another cell-specific phenomenon.
Only in lung alveolar epithelial cells do the endocytosed Na/K ATPase molecules get
degraded – via a lysosomal pathway(Comellas et al., 2006). Renal tubule epithelial cells
tend to recycle endocytosed Na/K ATPase to the membrane once the threat of cellular
stress has passed(Efendiev et al., 2007).
Axonal Na/K ATPase: a PURED candidate?
While a precedent has been set for Na/K ATPase endocytosis in other osmotically
and ionically active cells, the question remains: Could this sort of pathway exist in
neurons? One group has begun to answer this question by studying the Na/K ATPase
enzymatic activity on cultured cerebellar granule cells(Petrushanko et al., 2007). Under
conditions of hypoxia, they report a decrease in the cell surface Na/K ATPase enzyme
activity. While enzyme activity assays are a helpful first step, visualization of Na/K
ATPase endocytosis in neurons will be critical to the confirmation of a PURED (or
similar) pathway in neurons. Attempts at identifying the PURED pathway machinery in
axons with and without Na/K ATPase have, thus far, proven unsuccessful (data not
shown). One word of caution, however – this study, as well as the studies described
above, focused on the modulation and endocytosis of only the α1 subunit. While it is
58
possible that appreciable changes may occur in neurons/axons during α1 endocytosis, the
presence of the α 3 subunit may make biochemical discrimination exceedingly difficult.
It is also important to remember that, at least in the case at hand – Na/K ATPase loss
from chronically demyelinated axons – both α-subunits were lost from the membrane in
apparently equal amounts and neither was preferentially spared. This democratized loss
of both catalytic subunit isoforms suggests that, perhaps, there is a different, non-PURED
pathway responsible for the removal of the Na/K ATPase from the membranes of
chronically demyelinated axons. One remaining possibility is that the axonal Na/K
ATPase is not proactively removed from the membrane to mitigate ATP losses, but rather
it is removed because it has been damaged by a harsh intracellular milieu and is being
targeted for degradation. Unfortunately, at present, the exact cause of axonal Na/K
ATPase loss and the apparent 50/50 selectivity of axons with and without Na/K ATPase
remains elusive.
Na/K ATPase: A new therapeutic target?
Targeted treatment options for the progressive phases of MS (SPMS and PPMS,
see Chapter 1, Figure 2) are currently unavailable(Ebers et al., 2008). Given the chronic
nature of the lesions from which Na/K ATPase is lost, it is reasonable to expect that an
accumulation of Na/K ATPase-negative lesions might be a harbinger of the silent
transition between RRMS and SPMS. In fact, all of the chronic lesions studied (both
Na/K ATPase-positive and –negative) were from SPMS patients (see Chapter 1, Table 1).
Absence of the Na/K ATPase from the axonal membrane may hasten the reversal of the
NCX and deleterious calcium influx. If Na/K ATPase is endocytosed to prevent ATP
59
debt during times of cellular stress, perhaps early preventive modulation of the Na/K
ATPase might keep ATP levels high. Recent evidence points to the ability of typically
well-tolerated, orally-administered cardiotonic steroids, such as ouabain, to modulate the
activity of neuronal Na/K ATPases – at low concentrations (sub-IC50 doses) of ouabain,
Na/K ATPase pump efficiency is increased, while at higher concentrations (IC50), the
Na/K ATPase is halted(Oselkin et al., 2009). Another potential therapeutic target might
be the Ser-18 phosphorylation site that is required for Na/K ATPase endocytosis. An
agent designed to block phosphorylation of Ser-18 would inhibit AP-2 binding and stop
clathrin-dependent endocytosis. It is critical to the development of any therapeutic aimed
at this patient population that it be well-tolerated with few, if any, side effects – digoxin,
along with the other cardiotonic steroids, have been in clinical use for decades for the
treatment of congestive heart failure and atrial fibrillation(Marcus, 1989).
While there are risks for digoxin toxicity, treatment can be managed within a very safe
and tolerable window.
MRI/Pathology correlations
The finding that MRI, using MTR and T1 contrast ratio sequences, can identify
lesions with and without Na/K ATPase is a huge advance. Having the capability of
identifying, in a living patient, these Na/K ATPase-negative lesions will certainly change
the way that clinicians will be able to make diagnostic and prognostic decisions about
patients. Being able to watch in real time the progression of the silent phases of the
disease might allow doctors to predict the otherwise imperceptible shift from RRMS to
SPMS. Treating patients with sodium channel blockers, a common front-line therapy,
60
may, in a patient with many Na/K ATPase-negative lesions, be ineffective. Finally, with
a means to assess the silent progression of neurodegeneration, therapies can be assessed
based a new endpoint: maintenance of Na/K ATPase in chronically demyelinated lesions.
Summary
The data presented in this dissertation irrefutably confirm an internodal
localization for the axonal Na/K ATPase and selective loss of Na/K ATPase from a
subset of chronically demyelinated MS lesions. Additionally, a statistically significant
correlation between MRI measures and Na/K ATPase pathology has been identified.
Each of these represent a substantial addition to the knowledge base and suggest new
directions for both laboratory research into the degenerative cascade of chronically
demyelinated axons and new therapeutic targets for treating SPMS and PPMS, a
currently unmet need.
61
Figure 1.
62
Figure 1. A new model for axonal function and dysfunction in MS. 1.) In the myelinated
axon, the Na/K ATPase is enriched in the internodal axolemma. Upon action potential
conduction, sodium influx occurs at the Node of Ranvier. Intracellular sodium diffuses
through the axoplasm to the Na/K ATPase, where it is pumped from the axoplasm into
the periaxonal space. In exchange, potassium is pumped into the cell. The net reaction
is: 3 Na ions out, 2 K ions in, 1ATP. The ATP is produced by the mitochondria.
Calcium is removed from the cell by the sodium-calcium exchanger (NCX) which is,
putatively, in the node. Additionally, the neurofilaments are phosphorylated and have
extensive cross-bridging between them, thereby increasing axonal diameter. 2.) Upon
demyelination, Na(V) channels diffuse away from the erstwhile node and redistribute
along the denuded axon, causing an increased influx of sodium into the axoplasm. This
increased sodium increases Na/K ATPase function, thereby utilizing a large proportion of
the cell’s ATP stores, creating an ATP debt. Further, the once cloistered K(V) channels
are now uncloaked and produce a large outward current. Finally, the neurofilaments,
lacking the axo-glial signaling provided by myelin-associated glycoprotein (MAG), are
dephosphorylated and cross-bridges are collapsed. 3.) In a subset of chronic lesions,
Na/K ATPase is lost from the axolemma. Without control over the ion gradient, the
NCX is reversed and begins to extrude Na from the cell. This reversal causes calcium
influx, leading to destructive, calcium-mediated processes and the onset of axonal
degeneration.
63
REFERENCES
1. Ames A, III (2000) CNS energy metabolism as related to function. Brain
Res Brain Res Rev, 34, 42-68.
2. Antolovic R, Bruller HJ, Bunk S, Linder D, and Schoner W (1991) Epitope
mapping by amino-acid-sequence-specific antibodies reveals that both
ends of the alpha subunit of Na+/K(+)-ATPase are located on the
cytoplasmic side of the membrane. Eur J Biochem, 199, 195-202.
3. Arhem P (1976) Diffusion of sodium in axoplasm of myelinated nerve fibre.
Potential clamp analysis. Acta Physiol Scand, 97, 415-425.
4. Ariyasu RG, Nichol JA, and Ellisman MH (1985) Localization of
sodium/potassium adenosine triphosphatase in multiple cell types of the
murine nervous system with antibodies raised against the enzyme from
kidney. J Neurosci, 5, 2581-2596.
5. Arroyo EJ, Sirkowski EE, Chitale R, and Scherer SS (2004) Acute
demyelination disrupts the molecular organization of peripheral nervous
system nodes. J Comp Neurol, 479, 424-434.
6. Arroyo EJ, Xu T, Grinspan J, Lambert S, Levinson SR, Brophy PJ, Peles
E, and Scherer SS (2002) Genetic dysmyelination alters the molecular
architecture of the nodal region. J Neurosci, 22, 1726-1737.
7. Azzam ZS, Saldias FJ, Comellas A, Ridge KM, Rutschman DH, and
Sznajder JI (2001) Catecholamines increase lung edema clearance in rats
with increased left atrial pressure. J Appl Physiol, 90, 1088-1094.
8. Baker PF and McNaughton PA (1976) Kinetics and energetics of calcium
efflux from intact squid giant axons. J Physiol, 259, 103-144.
9. Bammer R, Augustin M, Strasser-Fuchs S, Seifert T, Kapeller P,
Stollberger R, Ebner F, Hartung HP, and Fazekas F (2000) Magnetic
resonance diffusion tensor imaging for characterizing diffuse and focal
white matter abnormalities in multiple sclerosis. Magn Reson Med, 44,
583-591.
10. Barkhof F, Karas GB, and van Walderveen MA (2000) T1 hypointensities
and axonal loss. Neuroimaging Clin N Am, 10, 739-752.
11. Beguin P, Beggah AT, Chibalin AV, Burgener-Kairuz P, Jaisser F,
Mathews PM, Rossier BC, Cotecchia S, and Geering K (1994)
Phosphorylation of the Na,K-ATPase alpha-subunit by protein kinase A
and C in vitro and in intact cells. Identification of a novel motif for PKCmediated phosphorylation. J Biol Chem, 269, 24437-24445.
64
12. Berghs S, Aggujaro D, Dirkx R, Jr., Maksimova E, Stabach P, Hermel JM,
Zhang JP, Philbrick W, Slepnev V, Ort T, and Solimena M (2000) betaIV
spectrin, a new spectrin localized at axon initial segments and nodes of
ranvier in the central and peripheral nervous system. J Cell Biol, 151, 9851002.
13. Binah O and Palti Y (1981) Potassium channels in the nodal membrane of
rat myelinated fibres. Nature, 290, 598-600.
14. Bjartmar C, Kidd G, Mork S, Rudick R, and Trapp BD (2000) Neurological
disability correlates with spinal cord axonal loss and reduced N-acetyl
aspartate in chronic multiple sclerosis patients. Ann Neurol, 48, 893-901.
15. Black JA, Newcombe J, Trapp BD, and Waxman SG (2007) Sodium
channel expression within chronic multiple sclerosis plaques. J
Neuropathol Exp Neurol, 66, 828-837.
16. Blanco G, Sanchez G, Melton RJ, Tourtellotte WG, and Mercer RW (2000)
The alpha4 isoform of the Na,K-ATPase is expressed in the germ cells of
the testes. J Histochem Cytochem, 48, 1023-1032.
17. Boiko T, Rasband MN, Levinson SR, Caldwell JH, Mandel G, Trimmer JS,
and Matthews G (2001) Compact myelin dictates the differential targeting
of two sodium channel isoforms in the same axon. Neuron, 30, 91-104.
18. Bostock H, Sears TA, and Sherratt RM (1983) The spatial distribution of
excitability and membrane current in normal and demyelinated
mammalian nerve fibres. J Physiol, 341, 41-58.
19. Bouzidi M, Tricaud N, Giraud P, Kordeli E, Caillol G, Deleuze C, Couraud
F, and Alcaraz G (2002) Interaction of the Nav1.2a subunit of the voltagedependent sodium channel with nodal ankyrinG. In vitro mapping of the
interacting domains and association in synaptosomes. J Biol Chem, 277,
28996-29004.
20. Brown DA and Adams PR (1980) Muscarinic suppression of a novel
voltage-sensitive K+ current in a vertebrate neurone. Nature, 283, 673676.
21. Bruck W, Bitsch A, Kolenda H, Bruck Y, Stiefel M, and Lassmann H
(1997) Inflammatory central nervous system demyelination: correlation of
magnetic resonance imaging findings with lesion pathology. Ann Neurol,
42, 783-793.
22. Burke D, Kiernan MC, and Bostock H (2001) Excitability of human axons.
Clin Neurophysiol, 112, 1575-1585.
65
23. Cameron R, Klein L, Shyjan AW, Rakic P, and Levenson R (1994)
Neurons and astroglia express distinct subsets of Na,K-ATPase alpha and
beta subunits. Brain Res Mol Brain Res, 21, 333-343.
24. Campbell AR, Folkesson HG, Berthiaume Y, Gutkowska J, Suzuki S, and
Matthay MA (1999) Alveolar epithelial fluid clearance persists in the
presence of moderate left atrial hypertension in sheep. J Appl Physiol, 86,
139-151.
25. Castro J, Ruminot I, Porras OH, Flores CM, Hermosilla T, Verdugo E,
Venegas F, Hartel S, Michea L, and Barros LF (2006) ATP steal between
cation pumps: a mechanism linking Na+ influx to the onset of necrotic
Ca2+ overload. Cell Death Differ, 13, 1675-1685.
26. Chang A, Tourtellotte WW, Rudick R, and Trapp BD (2002)
Premyelinating oligodendrocytes in chronic lesions of multiple sclerosis. N
Engl J Med, 346, 165-173.
27. Chen Z, Krmar RT, Dada L, Efendiev R, Leibiger IB, Pedemonte CH, Katz
AI, Sznajder JI, and Bertorello AM (2006) Phosphorylation of adaptor
protein-2 mu2 is essential for Na+,K+-ATPase endocytosis in response to
either G protein-coupled receptor or reactive oxygen species. Am J Respir
Cell Mol Biol, 35, 127-132.
28. Chibalin AV, Ogimoto G, Pedemonte CH, Pressley TA, Katz AI, Feraille E,
Berggren PO, and Bertorello AM (1999) Dopamine-induced endocytosis of
Na+,K+-ATPase is initiated by phosphorylation of Ser-18 in the rat alpha
subunit and Is responsible for the decreased activity in epithelial cells. J
Biol Chem, 274, 1920-1927.
29. Chiu SY and Ritchie JM (1980) Potassium channels in nodal and
internodal axonal membrane of mammalian myelinated fibres. Nature,
284, 170-171.
30. Chiu SY and Ritchie JM (1981) Evidence for the presence of potassium
channels in the paranodal region of acutely demyelinated mammalian
single nerve fibres. J Physiol, 313, 415-437.
31. Chiu SY and Schwarz W (1987) Sodium and potassium currents in acutely
demyelinated internodes of rabbit sciatic nerves. J Physiol, 391, 631-649.
32. Coman I, Aigrot MS, Seilhean D, Reynolds R, Girault JA, Zalc B, and
Lubetzki C (2006) Nodal, paranodal and juxtaparanodal axonal proteins
during demyelination and remyelination in multiple sclerosis. Brain, 129,
3186-3195.
33. Comellas AP, Dada LA, Lecuona E, Pesce LM, Chandel NS, Quesada N,
Budinger GR, Strous GJ, Ciechanover A, and Sznajder JI (2006) Hypoxia66
mediated degradation of Na,K-ATPase via mitochondrial reactive oxygen
species and the ubiquitin-conjugating system. Circ Res, 98, 1314-1322.
34. Craner MJ, Newcombe J, Black JA, Hartle C, Cuzner ML, and Waxman
SG (2004) Molecular changes in neurons in multiple sclerosis: altered
axonal expression of Nav1.2 and Nav1.6 sodium channels and Na+/Ca2+
exchanger. Proc Natl Acad Sci U S A, 101, 8168-8173.
35. Dada LA, Chandel NS, Ridge KM, Pedemonte C, Bertorello AM, and
Sznajder JI (2003) Hypoxia-induced endocytosis of Na,K-ATPase in
alveolar epithelial cells is mediated by mitochondrial reactive oxygen
species and PKC-zeta. J Clin Invest, 111, 1057-1064.
36. Dada LA, Welch LC, Zhou G, Ben-Saadon R, Ciechanover A, and
Sznajder JI (2007) Phosphorylation and ubiquitination are necessary for
Na,K-ATPase endocytosis during hypoxia. Cell Signal, 19, 1893-1898.
37. Devaux J, Alcaraz G, Grinspan J, Bennett V, Joho R, Crest M, and
Scherer SS (2003) Kv3.1b is a novel component of CNS nodes. J
Neurosci, 23, 4509-4518.
38. Devaux JJ, Kleopa KA, Cooper EC, and Scherer SS (2004) KCNQ2 is a
nodal K+ channel. J Neurosci, 24, 1236-1244.
39. Doi M (2009) [Pump activity-independent novel functions of Na+/K+
ATPase in the nervous systems]. Seikagaku, 81, 797-801.
40. Done SC, Leibiger IB, Efendiev R, Katz AI, Leibiger B, Berggren PO,
Pedemonte CH, and Bertorello AM (2002) Tyrosine 537 within the
Na+,K+-ATPase alpha-subunit is essential for AP-2 binding and clathrindependent endocytosis. J Biol Chem, 277, 17108-17111.
41. Dutta R, McDonough J, Yin X, Peterson J, Chang A, Torres T, Gudz T,
Macklin WB, Lewis DA, Fox RJ, Rudick R, Mirnics K, and Trapp BD
(2006) Mitochondrial dysfunction as a cause of axonal degeneration in
multiple sclerosis patients. Ann Neurol, 59, 478-489.
42. Dutta R and Trapp BD (2007) Pathogenesis of axonal and neuronal
damage in multiple sclerosis. Neurology, 68, S22-S31.
43. Ebers GC, Heigenhauser L, Daumer M, Lederer C, and Noseworthy JH
(2008) Disability as an outcome in MS clinical trials. Neurology, 71, 624631.
44. Efendiev R, Das-Panja K, Cinelli AR, Bertorello AM, and Pedemonte CH
(2007) Localization of intracellular compartments that exchange Na,KATPase molecules with the plasma membrane in a hormone-dependent
manner. Br J Pharmacol, 151, 1006-1013.
67
45. Einheber S, Zanazzi G, Ching W, Scherer S, Milner TA, Peles E, and
Salzer JL (1997) The axonal membrane protein Caspr, a homologue of
neurexin IV, is a component of the septate-like paranodal junctions that
assemble during myelination. J Cell Biol, 139, 1495-1506.
46. Ellisman MH and Porter KR (1980) Microtrabecular structure of the
axoplasmic matrix: visualization of cross-linking structures and their
distribution. J Cell Biol, 87, 464-479.
47. England JD, Gamboni F, Levinson SR, and Finger TE (1990) Changed
distribution of sodium channels along demyelinated axons. Proc Natl Acad
Sci U S A, 87, 6777-6780.
48. Filippi M and Rocca MA (2006) The use of modern magnetic resonance
techniques to monitor disease evolution in multiple sclerosis. Adv Neurol,
98, 167-183.
49. Fisher E, Chang A, Fox RJ, Tkach JA, Svarovsky T, Nakamura K, Rudick
RA, and Trapp BD (2007) Imaging correlates of axonal swelling in chronic
multiple sclerosis brains. Ann Neurol, 62, 219-228.
50. Geering K, Meyer DI, Paccolat MP, Kraehenbuhl JP, and Rossier BC
(1985) Membrane insertion of alpha- and beta-subunits of Na+,K+ATPase. J Biol Chem, 260, 5154-5160.
51. Gloor S, Antonicek H, Sweadner KJ, Pagliusi S, Frank R, Moos M, and
Schachner M (1990) The adhesion molecule on glia (AMOG) is a
homologue of the beta subunit of the Na,K-ATPase. J Cell Biol, 110, 165174.
52. Hauser SL and Oksenberg JR (2006) The neurobiology of multiple
sclerosis: genes, inflammation, and neurodegeneration. Neuron, 52, 6176.
53. Herrera VL, Emanuel JR, Ruiz-Opazo N, Levenson R, and Nadal-Ginard
B (1987) Three differentially expressed Na,K-ATPase alpha subunit
isoforms: structural and functional implications. J Cell Biol, 105, 18551865.
54. Herzog RI, Cummins TR, Ghassemi F, Dib-Hajj SD, and Waxman SG
(2003) Distinct repriming and closed-state inactivation kinetics of Nav1.6
and Nav1.7 sodium channels in mouse spinal sensory neurons. J Physiol,
551, 741-750.
55. Heuser J (2002) Whatever happened to the 'microtrabecular concept'?
Biol Cell, 94, 561-596.
68
56. Hirokawa N (1982) Cross-linker system between neurofilaments,
microtubules, and membranous organelles in frog axons revealed by the
quick-freeze, deep-etching method. J Cell Biol, 94, 129-142.
57. Hirsch HE and Parks ME (1983) Na+- and K+-dependent adenosine
triphosphatase changes in multiple sclerosis plaques. Ann Neurol, 13,
658-663.
58. Hodgkin AL and Huxley AF (1945) Resting and action potentials in single
nerve fibres. J Physiol, 104, 176-195.
59. Hodgkin AL and Huxley AF (1952) Currents carried by sodium and
potassium ions through the membrane of the giant axon of Loligo. J
Physiol, 116, 449-472.
60. Hodgkin AL and Katz B (1949) The effect of sodium ions on the electrical
activity of the giant axon of the squid. J Physiol, 108, 37-77.
61. Hohlfeld R (1997) Biotechnological agents for the immunotherapy of
multiple sclerosis. Principles, problems and perspectives. Brain, 120 ( Pt
5), 865-916.
62. Huxley AF and Stampfli R (1949) Evidence for saltatory conduction in
peripheral myelinated nerve fibres. J Physiol, 108, 315-339.
63. Ichimura T and Ellisman MH (1991) Three-dimensional fine structure of
cytoskeletal-membrane interactions at nodes of Ranvier. J Neurocytol, 20,
667-681.
64. Ishikawa T, Nakamura Y, Saitoh N, Li WB, Iwasaki S, and Takahashi T
(2003) Distinct roles of Kv1 and Kv3 potassium channels at the calyx of
Held presynaptic terminal. J Neurosci, 23, 10445-10453.
65. Jenkins SM and Bennett V (2001) Ankyrin-G coordinates assembly of the
spectrin-based membrane skeleton, voltage-gated sodium channels, and
L1 CAMs at Purkinje neuron initial segments. J Cell Biol, 155, 739-746.
66. Kanoh N and Sakagami M (1996) Impaired Ranvier node sodiumpotassium adenosine triphosphatase may induce facial palsy.
Laryngoscope, 106, 1180-1183.
67. Kaplan MR, Cho MH, Ullian EM, Isom LL, Levinson SR, and Barres BA
(2001) Differential control of clustering of the sodium channels Na(v)1.2
and Na(v)1.6 at developing CNS nodes of Ranvier. Neuron, 30, 105-119.
68. Komada M and Soriano P (2002) [Beta]IV-spectrin regulates sodium
channel clustering through ankyrin-G at axon initial segments and nodes
of Ranvier. J Cell Biol, 156, 337-348.
69
69. Kordeli E, Lambert S, and Bennett V (1995) AnkyrinG. A new ankyrin
gene with neural-specific isoforms localized at the axonal initial segment
and node of Ranvier. J Biol Chem, 270, 2352-2359.
70. Lai HC and Jan LY (2006) The distribution and targeting of neuronal
voltage-gated ion channels. Nat Rev Neurosci, 7, 548-562.
71. Lane LK, Shull MM, Whitmer KR, and Lingrel JB (1989) Characterization
of two genes for the human Na,K-ATPase beta subunit. Genomics, 5, 445453.
72. Lassmann H, Raine CS, Antel J, and Prineas JW (1998)
Immunopathology of multiple sclerosis: report on an international meeting
held at the Institute of Neurology of the University of Vienna. J
Neuroimmunol, 86, 213-217.
73. Lecuona E, Trejo HE, and Sznajder JI (2007) Regulation of Na,K-ATPase
during acute lung injury. J Bioenerg Biomembr, 39, 391-395.
74. Lemaillet G, Walker B, and Lambert S (2003) Identification of a conserved
ankyrin-binding motif in the family of sodium channel alpha subunits. J Biol
Chem, 278, 27333-27339.
75. Levenson R (1994) Isoforms of the Na,K-ATPase: family members in
search of function. Rev Physiol Biochem Pharmacol, 123, 1-45.
76. Liang M, Tian J, Liu L, Pierre S, Liu J, Shapiro J, and Xie ZJ (2007)
Identification of a pool of non-pumping Na/K-ATPase. J Biol Chem, 282,
10585-10593.
77. Mahad DJ, Ziabreva I, Campbell G, Lax N, White K, Hanson PS,
Lassmann H, and Turnbull DM (2009) Mitochondrial changes within axons
in multiple sclerosis. Brain, 132, 1161-1174.
78. Malik N, Canfield VA, Beckers MC, Gros P, and Levenson R (1996)
Identification of the mammalian Na,K-ATPase 3 subunit. J Biol Chem,
271, 22754-22758.
79. Marcus FI (1989) The use of digitalis for the treatment of congestive heart
failure: a tale of its decline and resurrection. Cardiovasc Drugs Ther, 3,
473-476.
80. Martin-Vasallo P, Dackowski W, Emanuel JR, and Levenson R (1989)
Identification of a putative isoform of the Na,K-ATPase beta subunit.
Primary structure and tissue-specific expression. J Biol Chem, 264, 46134618.
70
81. Mata M, Fink DJ, Ernst SA, and Siegel GJ (1991) Immunocytochemical
demonstration of Na+,K(+)-ATPase in internodal axolemma of myelinated
fibers of rat sciatic and optic nerves. J Neurochem, 57, 184-192.
82. Mercer RW, Schneider JW, Savitz A, Emanuel J, Benz EJ, Jr., and
Levenson R (1986) Rat-brain Na,K-ATPase beta-chain gene: primary
structure, tissue-specific expression, and amplification in ouabain-resistant
HeLa C+ cells. Mol Cell Biol, 6, 3884-3890.
83. Moll C, Mourre C, Lazdunski M, and Ulrich J (1991) Increase of sodium
channels in demyelinated lesions of multiple sclerosis. Brain Res, 556,
311-316.
84. Moseley AE, Lieske SP, Wetzel RK, James PF, He S, Shelly DA, Paul RJ,
Boivin GP, Witte DP, Ramirez JM, Sweadner KJ, and Lingrel JB (2003)
The Na,K-ATPase alpha 2 isoform is expressed in neurons, and its
absence disrupts neuronal activity in newborn mice. J Biol Chem, 278,
5317-5324.
85. Noseworthy JH, Lucchinetti C, Rodriguez M, and Weinshenker BG (2000)
Multiple sclerosis. N Engl J Med, 343, 938-952.
86. Oselkin M, Tian D, and Bergold PJ (2009) Low-dose cardiotonic steroids
increase sodium-potassium ATPase activity that protects hippocampal
slice cultures from experimental ischemia. Neurosci Lett.
87. Pan Z, Kao T, Horvath Z, Lemos J, Sul JY, Cranstoun SD, Bennett V,
Scherer SS, and Cooper EC (2006) A common ankyrin-G-based
mechanism retains KCNQ and NaV channels at electrically active
domains of the axon. J Neurosci, 26, 2599-2613.
88. Petrushanko IY, Bogdanov NB, Lapina N, Boldyrev AA, Gassmann M, and
Bogdanova AY (2007) Oxygen-induced Regulation of Na/K ATPase in
cerebellar granule cells. J Gen Physiol, 130, 389-398.
89. Pitt D, Werner P, and Raine CS (2000) Glutamate excitotoxicity in a model
of multiple sclerosis. Nat Med, 6, 67-70.
90. Rasband MN, Trimmer JS, Schwarz TL, Levinson SR, Ellisman MH,
Schachner M, and Shrager P (1998) Potassium channel distribution,
clustering, and function in remyelinating rat axons. J Neurosci, 18, 36-47.
91. Rios JC, Melendez-Vasquez CV, Einheber S, Lustig M, Grumet M,
Hemperly J, Peles E, and Salzer JL (2000) Contactin-associated protein
(Caspr) and contactin form a complex that is targeted to the paranodal
junctions during myelination. J Neurosci, 20, 8354-8364.
71
92. Ritchie JM and Rogart RB (1977) Density of sodium channels in
mammalian myelinated nerve fibers and nature of the axonal membrane
under the myelin sheath. Proc Natl Acad Sci U S A, 74, 211-215.
93. Sanchez I, Hassinger L, Paskevich PA, Shine HD, and Nixon RA (1996)
Oligodendroglia regulate the regional expansion of axon caliber and local
accumulation of neurofilaments during development independently of
myelin formation. J Neurosci, 16, 5095-5105.
94. Sardet C (2002) Special issue: microtrabecular concept of the cytoplasm
revisited. Biol Cell, 94, 553.
95. Scherer SS and Arroyo EJ (2002) Recent progress on the molecular
organization of myelinated axons. J Peripher Nerv Syst, 7, 1-12.
96. Schmierer K, Scaravilli F, Altmann DR, Barker GJ, and Miller DH (2004)
Magnetization transfer ratio and myelin in postmortem multiple sclerosis
brain. Ann Neurol, 56, 407-415.
97. Schwarz JR, Corrette BJ, Mann K, and Wietholter H (1991) Changes of
ionic channel distribution in myelinated nerve fibres from rats with
experimental allergic neuritis. Neurosci Lett, 122, 205-209.
98. Schwarz JR, Reid G, and Bostock H (1995) Action potentials and
membrane currents in the human node of Ranvier. Pflugers Arch, 430,
283-292.
99. Shull GE, Greeb J, and Lingrel JB (1986) Molecular cloning of three
distinct forms of the Na+,K+-ATPase alpha-subunit from rat brain.
Biochemistry, 25, 8125-8132.
100. Shull MM and Lingrel JB (1987) Multiple genes encode the human
Na+,K+-ATPase catalytic subunit. Proc Natl Acad Sci U S A, 84, 40394043.
101. Shyjan AW, Canfield VA, and Levenson R (1991) Evolution of the Na,Kand H,K-ATPase beta subunit gene family: structure of the murine Na,KATPase beta 2 subunit gene. Genomics, 11, 435-442.
102. Stys PK, Sontheimer H, Ransom BR, and Waxman SG (1993)
Noninactivating, tetrodotoxin-sensitive Na+ conductance in rat optic nerve
axons. Proc Natl Acad Sci U S A, 90, 6976-6980.
103. Stys PK, Waxman SG, and Ransom BR (1992) Ionic mechanisms of
anoxic injury in mammalian CNS white matter: role of Na+ channels and
Na(+)-Ca2+ exchanger. J Neurosci, 12, 430-439.
72
104. Sverdlov ED, Monastyrskaya GS, Broude NE, Ushkaryov Y, Allikmets RL,
Melkov AM, Smirnov Y, Malyshev IV, Dulobova IE, Petrukhin KE, and .
(1987) The family of human Na+,K+-ATPase genes. No less than five
genes and/or pseudogenes related to the alpha-subunit. FEBS Lett, 217,
275-278.
105. Sweadner KJ (1979) Two molecular forms of (Na+ + K+)-stimulated
ATPase in brain. Separation, and difference in affinity for strophanthidin. J
Biol Chem, 254, 6060-6067.
106. Takeyasu K, Lemas V, and Fambrough DM (1990) Stability of Na(+)-K(+)ATPase alpha-subunit isoforms in evolution. Am J Physiol, 259, C619C630.
107. Trapp BD and Nave KA (2008) Multiple sclerosis: an immune or
neurodegenerative disorder? Annu Rev Neurosci, 31, 247-269.
108. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, and Bo L (1998)
Axonal transection in the lesions of multiple sclerosis. N Engl J Med, 338,
278-285.
109. Trapp BD and Stys PK (2009) Virtual hypoxia and chronic necrosis of
demyelinated axons in multiple sclerosis. Lancet Neurol, 8, 280-291.
110. Trimmer JS and Rhodes KJ (2004) Localization of voltage-gated ion
channels in mammalian brain. Annu Rev Physiol, 66, 477-519.
111. van Waesberghe JH, Kamphorst W, De Groot CJ, van Walderveen MA,
Castelijns JA, Ravid R, Nijeholt GJ, van d, V, Polman CH, Thompson AJ,
and Barkhof F (1999) Axonal loss in multiple sclerosis lesions: magnetic
resonance imaging insights into substrates of disability. Ann Neurol, 46,
747-754.
112. van Walderveen MA, Kamphorst W, Scheltens P, van Waesberghe JH,
Ravid R, Valk J, Polman CH, and Barkhof F (1998) Histopathologic
correlate of hypointense lesions on T1-weighted spin-echo MRI in multiple
sclerosis. Neurology, 50, 1282-1288.
113. Vervaeke K, Gu N, Agdestein C, Hu H, and Storm JF (2006)
Kv7/KCNQ/M-channels in rat glutamatergic hippocampal axons and their
role in regulation of excitability and transmitter release. J Physiol, 576,
235-256.
114. Wang H, Kunkel DD, Martin TM, Schwartzkroin PA, and Tempel BL (1993)
Heteromultimeric K+ channels in terminal and juxtaparanodal regions of
neurons. Nature, 365, 75-79.
73
115. Wang LY, Gan L, Forsythe ID, and Kaczmarek LK (1998) Contribution of
the Kv3.1 potassium channel to high-frequency firing in mouse auditory
neurones. J Physiol, 509 ( Pt 1), 183-194.
116. Waxman SG (1977) Conduction in myelinated, unmyelinated, and
demyelinated fibers. Arch Neurol, 34, 585-589.
117. Waxman SG (1998) Demyelinating diseases--new pathological insights,
new therapeutic targets. N Engl J Med, 338, 323-325.
118. Waxman SG (2006) Axonal conduction and injury in multiple sclerosis: the
role of sodium channels. Nat Rev Neurosci, 7, 932-941.
119. Waxman SG, Craner MJ, and Black JA (2004) Na+ channel expression
along axons in multiple sclerosis and its models. Trends Pharmacol Sci,
25, 584-591.
120. Waxman SG and Ritchie JM (1993) Molecular dissection of the myelinated
axon. Ann Neurol, 33, 121-136.
121. Woo AL, James PF, and Lingrel JB (2000) Sperm motility is dependent on
a unique isoform of the Na,K-ATPase. J Biol Chem, 275, 20693-20699.
122. Wu RL and Barish ME (1992) Two pharmacologically and kinetically
distinct transient potassium currents in cultured embryonic mouse
hippocampal neurons. J Neurosci, 12, 2235-2246.
123. Yin X, Crawford TO, Griffin JW, Tu P, Lee VM, Li C, Roder J, and Trapp
BD (1998) Myelin-associated glycoprotein is a myelin signal that
modulates the caliber of myelinated axons. J Neurosci, 18, 1953-1962.
124. Yin X, Kidd GJ, Pioro EP, McDonough J, Dutta R, Feltri ML, Wrabetz L,
Messing A, Wyatt RM, Balice-Gordon RJ, and Trapp BD (2004)
Dysmyelinated lower motor neurons retract and regenerate dysfunctional
synaptic terminals. J Neurosci, 24, 3890-3898.
125. Yoshimura SH, Iwasaka S, Schwarz W, and Takeyasu K (2008) Fast
degradation of the auxiliary subunit of Na+/K+-ATPase in the plasma
membrane of HeLa cells. J Cell Sci, 121, 2159-2168.
126. Young EA, Fowler CD, Kidd GJ, Chang A, Rudick R, Fisher E, and Trapp
BD (2008) Imaging correlates of decreased axonal Na+/K+ ATPase in
chronic multiple sclerosis lesions. Ann Neurol, 63, 428-435.
127. Yudowski GA, Efendiev R, Pedemonte CH, Katz AI, Berggren PO, and
Bertorello AM (2000) Phosphoinositide-3 kinase binds to a proline-rich
motif in the Na+, K+-ATPase alpha subunit and regulates its trafficking.
Proc Natl Acad Sci U S A, 97, 6556-6561.
74
128. Zhao H, Pestov NB, Korneenko TV, Shakhparonov MI, and Modyanov NN
(2004) Accumulation of beta (m), a structural member of X,K-ATPase
beta-subunit family, in nuclear envelopes of perinatal myocytes. Am J
Physiol Cell Physiol, 286, C757-C767.
129. Zhou W and Goldin AL (2004) Use-dependent potentiation of the Nav1.6
sodium channel. Biophys J, 87, 3862-3872.
75