CIRCADIAN RHYTHMS OF REDOX STATE AND REGULATION
OF NEURONAL EXCITABILITY IN SUPRACHIASMATIC
NUCLEUS OF RODENTS
BY
TONGFEI WANG
DISSERTATION
Submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy in Molecular and Integrative Physiology
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2012
Urbana, Illinois
Doctoral Committee:
Professor Martha U. Gillette, Chair, Director of Research
Assistant Professor Hee Jung Chung
Professor Charles L. Cox
Professor Rhanor Gillette
Associate Professor Yingxiao P. Wang
ABSTRACT
Daily rhythms of mammalian physiology, metabolism, and behavior parallel the daynight cycle. They are driven by the central circadian clock in the brain, the
suprachiasmatic nucleus (SCN), where a genetic oscillator plays an essential role. Clockgene transcription/translation is sensitive to metabolic (redox) change; however, energetic
cycles manifest as circadian rhythms in protein oxidation have been reported in anucleate
cells, where no transcription occurs. Whether the brain clock expresses redox cycles and
how such metabolic oscillations might affect neuronal physiology are unknown. Here we
show a self-sustained circadian rhythm of SCN redox state that requires the molecular
clockwork. The redox oscillation determines the excitability of SCN neurons through a
non-transcriptional mechanism: alterations in redox state rapidly (< 2 min) reverse
membrane polarization of SCN neurons via changes in multiple K+ channels. The redox
regulation of neuronal excitability gates the SCN sensitivity and response to entraining
signals of light, by modulating Ca2+ signaling in response to excitatory neurotransmission,
targeting on a ryanodine receptor–dependent intracellular Ca2+ store. Our study provides
a novel pathway for the metabolic oscillator to engage in the organization with the central
circadian clock; it couples the molecular clockwork and cellular energetics with
membrane physiology, suggesting a basis for dynamic regulation of SCN excitability that
is closely tied to metabolism.
ii
ACKNOWLEDGEMENT
I would like to take this opportunity to express my gratitude to my advisor, Dr. Martha U.
Gillette, for her support and guidance on my way towards being a scientist in
neurophysiology. Martha is a great mentor, who always encourages students to develop
and pursue their own projects independently, while standing behind them when in need. It
has been a pleasure to work in the Gillette Lab during my graduate study. Very special
thanks to Drs. Sheue-Houy Tyan and Yanxun Yu, for their kindly help in the early years
when I got started with my graduate study in U.S.
I would like to express my gratitude to my close collaborator, advisory committee
member and good friend, Dr. Charles L. Cox. Lee was my interviewer when I was
applying for the Ph. D program in MIP, UIUC; after I joined the school, I did my first
rotation in his lab. From this point of view, it was Lee who brought me out of my
hometown, Beijing, where I have been living ever since my birth, to Urbana-Champaign,
where I have been making my life for the past a few years. I have been collaborating with
the Cox Lab for almost 4 years; during this time, I have gotten outstanding training in
electrophysiology and research in neuroscience. It is my fortune to work with Dr.
Govindaiah Gubbi, who has been instrumental essentially in our redox project; I could
never have made it without the help from “G” man.
Many thanks to my colleagues in both the Gillette Lab and the Cox Lab: Dr.
Jennifer Mitchell, Karen Weis, Maureen Holtz, Jennifer Arnold, James Chu, Dr. Sufang
Huang, Sam Irving, Raj Iyer, Anika Jain, Chris Liu, Harry Rosenberg, Dr. Kush Paul,
Shane Crandall, and Aurora Cruz-Torres, for their technical support, suggestion,
intriguing discussion and all the enjoyable time in the lab. I wish them all the best in their
iii
scientific careers.
I would like to thank my advisory committee members, Drs. Hee Jung Chung,
Julia George, Rhanor Gillette, and Yingxiao Peter Wang, for their advices on my Ph. D
study and supports on my future career. In particular, I would like to thank Dr. Rhanor
Gillette for his suggestion on real-time redox imaging in the inception of the redox
project, as well as many other aspects in my research. I would also like to thank Dr. Peter
Wang, for providing us a vector of FRET-based intracellular Ca2+ sensor; even though we
have not made use of it on my watch, we believe it must be fruitful in the future.
Last but not least, I would like to express my gratitude to my parents, Mr.
Huanqing Wang （王焕卿）and Mrs. Meiling Yang (杨美玲), whom I am eternally
grateful to have in my life. They have always been encouraging and supporting me in
pursuing my dream; I feel fortunate to have been raised in such a distinguished
environment of freedom and love, which greatly fostered my achievement to date. I love
you, Mom and Dad; I will always be together with you!
iv
TABLE OF CONTENTS
LIST OF ABBREVIATION ............................................................................................... vi
CHAPTER 1. ORGANIZATION OF THE CIRCADIAN CLOCK IN THE
MAMMALIAN SUPRACHIASMATIC NUCLEUS ..........................................................1
Circadian Rhythms in Mammals ......................................................................................1
Circadian Rhythm and Metabolic State ...........................................................................3
Circadian Rhythm and Excitability in SCN Neurons ......................................................5
Light/Glutamate Signaling in SCN Neurons ...................................................................6
Glutamate Signaling and Redox Regulation ....................................................................7
Aims and Significance .....................................................................................................7
Figure 1 ............................................................................................................................9
CHAPTER 2. CIRCADIAN RHYTHM OF REDOX STATE REGULATES
EXCITABILITY IN RODENT SCN NEURONS .............................................................10
Introduction ....................................................................................................................10
Materials and Methods ................................................................................................... 11
Results ............................................................................................................................19
Discussion ......................................................................................................................25
Figures 2 .........................................................................................................................28
CHAPTER 3. CALCIUM SIGNALING AND REDOX REGULATION IN
GLUTAMATE-INDUCED PHASE-SHIFTS IN RAT SCN .............................................38
Introduction ....................................................................................................................38
Materials and Methods ...................................................................................................39
Results ............................................................................................................................42
Discussion ......................................................................................................................45
Figures 3 .........................................................................................................................48
CHAPTER 4. METABOLIC CYCLE AS A CORE COMPONENT OF CIRCADIAN
CLOCK ..............................................................................................................................57
Summary ........................................................................................................................57
Perspective .....................................................................................................................58
Figure 4 ..........................................................................................................................61
REFERENCES ..................................................................................................................63
v
LIST OF ABBREVIATION
4-AP
AA
ACSF
Bupi
Ca2+i
CE-LIF
CREB
CT
DHA
DIA
DIV
DMPD
EBSS
ER
F-actin
FAD
G-actin
Glu
GSH
GSSG
I-V
L/D
NAD+/NADH
NADP+/NADPH
NMDA-R
NCX
NOS
PKG
RBC
RHT
RyR
SAP
SCN
SERCA
sGC
SUA
TEA
Tg
4-aminopyridine
ascorbic acid
artificial cerebrospinal fluid
bupivacaine
intracellular Ca2+
capillary electrophoresis with laser-induced fluorescence detection
cAMP-response element binding protein
circadian time
dehydroascorbic acid
diamide
days in vitro
4,5-dimethyl-1,2-phenylenediamine
Essential Balanced Salt Solution
endoplasmic reticulum
filamentary actin
flavin adenine dinucleotide
globular actin
glutamate
glutathione
glutathione disulfide
current-voltage
light/dark
nicotinamide adenine dinucleotide
nicotinamide adenine dinucleotide phosphate
N-methyl d-aspartate receptor
soldium-calcium exchanger
nitric oxide synthase
cGMP-dependent protein kinase
red blood cells
retinohypothalamatic tract
ryanodine receptor
spontaneous action potentials
suprachiasmatic nucleus
2+
sarco/endoplasmic reticulum Ca -ATPase
soluble guanylate cyclase
single unit activity
tetraethylammonium
thapasigargin
vi
TTL
TTX
Vm
ZT
τ
transcriptional-translational feedback loop
tetrodotoxin
membrane potential
zeitgeber time
endogenous circadian period
vii
CHAPTER 1. ORGANIZATION OF THE CIRCADIAN CLOCK IN
THE MAMMALIAN SUPRACHIASMATIC NUCLEUS
Circadian Rhythms in Mammals
The 24-hour day-night cycle on Earth shapes the living condition of organisms. The daily
oscillation of environment determines organisms’ fluctuating metabolism. An internal
clock adapts the organism to the changing external world by orchestrating the metabolism
and physiology of different organs. As a result, physiological functions in organisms are
coordinated with the “right time of the day”.
In mammals, these oscillations are controlled by a master clock, circadian clock,
in the suprachiasmatic nucleus (SCN) in the brain (Lehman et al., 1987; Ralph et al.,
1990). The SCN is a small pair of nuclei located in the anterior ventral hypothalamus,
right above the optic chiasm, and separated by the third ventricle (Moore et al., 2002). As
early as 1970s, lesion studies suggested that the SCN plays an essential role in circadian
rhythm in mammals. However, it was not until 1987, by tissue transplantation, SCN was
confirmed to be the master clock regulating the organisms’ circadian rhythms (Lehman et
al., 1987; Ralph et al., 1990).
The circadian rhythm is generated on molecular level by a negative feedback loop
of a group of clock genes and proteins; the transcriptional translational post-translational
regulation of clock genes and their products forms the core clockwork machinery
(Lowrey and Takahashi, 2004). In addition, proteins that mediate post-translational
modification, such as protein kinases (Robles et al., 2010), and small signaling molecules,
such as cAMP (Prosser and Gillette, 1991; O'Neill et al., 2008) and Ca2+ (Ikeda et al.,
1
2003; Harrisingh et al., 2007), also participate in generating the intrinsic circadian rhythm.
The endogenous clock maintains a ~24-hour period, or free-run, even without
external environmental cues. If the mice are maintained in a constant environment room,
they still show strong circadian rhythms similar to normal light/dark (L/D) conditions.
The only difference is that their intrinsic period, measured by the pattern of wheelrunning activity of the animals, is not exactly 24 hours, The endogenous free-running
circadian period (τ) of mice is less than 24 hours, while in rats τ is more than 24 hours. In
human τ is also a little longer than 24 hours for most individuals.
To align with the changing day- and night-length of the seasons, the animals have
to readjust their intrinsic clocks to the day-night cycle in nature, a process called
entrainment (Golombek and Rosenstein, 2010). Light is the most important
environmental time cue to entrain the circadian clock (Gillette and Mitchell, 2002).
Photic information is conveyed from the photosensitive, melanopsin-expressing retinal
ganglion cells to the neurons in SCN via retinohypothalamatic tract (RHT) (Berson et al.,
2002; Hattar et al., 2002). The effect of light on SCN phasing is specific to the time of
day (DeCoursey, 1960; Gillette and Mitchell, 2002). In early night (after lights-off in the
animal colonies), a light pulse signals an extended day to the SCN, so the phase of the
circadian clock is delayed, to adapt to this change. On the other hand, in late night (before
the entrained lights-on time), premature light signals an early day, so the clock responds
by advancing its phase. The clock phase is not altered by light during the day time, when
light is normally present and compatible with synchronization.
When studying the circadian clock, it is necessary to define the time within the
subject. If an animal is held under 12 h : 12 h L/D schedule, we use zeitgeber time (ZT)
2
to define the phase of the clock relative to the “time giving” signal, light. ZT 0/24 is the
time of light-on, while ZT 12 corresponds to the time of light-off. If an animal is held in a
constant environment without any time cue, circadian time (CT) is used instead, which
corresponds to ZT before the time cue is deprived. CT is used in brain slice study as well,
corresponding to the original ZT in the donor animals.
Circadian Rhythm and Metabolic State
The organism’s metabolic state, including levels of liver enzymes and energy production
/utilization, also oscillates with a circadian rhythm. Superimposed upon circadian
rhythms of metabolism are near 24-h oscillations due to the contingencies of life,
including metabolites in circulation and intracellular micro-environments, hormones
related to feeding/fasting, and ingestive behavior (Green et al., 2008; Bass and Takahashi,
2010). Genomic analysis found that 20% of rhythmic mRNAs are related to metabolism
(Akhtar et al., 2002). These include enzymes critical for energy metabolism such as
glycogen phosphorylase, cytochrome oxidase, lactate dehydrogenase, and the
monocarboxylate transporter, MCT2 (Rutter et al., 2002). Disruption of core clock genes
Clock and Bmal 1 transcription in mice leads to obesity, diabetes, and other metabolic
syndromes (Turek et al., 2005; Marcheva et al., 2010). These data support the contention
that metabolism is one of the most important reasons for the requirement of an internal
circadian clock.
Redox state, defined as the potential to receive or donate electrons of molecules in
biochemistry, is the manifestation of cellular metabolic state (Droge, 2002; Satoh and
Lipton, 2007). It describes the balance of several sets of metabolites, such as lactate and
3
pyruvate, whose interconversion depends upon reducer/oxidizer homeostasis in the cell
or organ. Reactive free radicals, such as nicotinamide adenine dinucleotide (NAD+) and
flavin adenine dinucleotide (FAD), originate from the redox reactions in metabolism, and
occur universally in all cells. They are part of the redox molecules that contribute to cell
physiology and intracellular/intercellular signaling (Droge, 2002). It follows that the
redox state could be evaluated by the ratio of redox-molecule pairs, such as
NAD+/NAD(P)H or glutathione disulfide (GSSG)/glutathione (GSH).
Circadian
and
energetic
cycles
are
coupled
through
the
transcriptional/translational modulation by core clock proteins of genes that regulate
metabolism, as well as the sensitivity of clock gene transcription to redox state (Rutter et
al., 2002; Green et al., 2008; Bass and Takahashi, 2010). Redox species such as
NAD(P)H and CO regulate the clock proteins CLOCK/BMAL1 and CRY through an
electron transport chain involving heme moieties (Rutter et al., 2001; Dioum et al., 2002).
This enables clock proteins to alter transcription in response to metabolic change.
Rhythmic expression of nuclear hormone receptors, such as REV-ERBα and RORα,
provides an additional pathway through which metabolic signals can engage the core
clockwork (Balsalobre et al., 1998; Preitner et al., 2002).
Recently, O’Neill and colleagues reported a circadian rhythm in redox state based
upon peroxiredoxin oligomerization in red blood cells (RBC) and green algae (O'Neill
and Reddy, 2011; O'Neill et al., 2011), however their relevance to cellular function and
systems physiology was not explored. Rhythmic activity of peroxiredoxin is independent
from transcription/translation in the anucleate RBC, but is disrupted in embryonic
fibroblasts from clock-mutant mice (O'Neill and Reddy, 2011). This suggests that in
4
nucleated cells, the transcriptional and redox oscillators are closely coupled.
Circadian Rhythm and Excitability in SCN Neurons
Each SCN is a neural circuit of ~10,000 neurons that undergo daily oscillation in
neuronal excitation, including fluctuation of membrane potential and the level of
intracellular Ca2+ (Ca2+i). The daily rhythm of electrical activity in the SCN is essential
for the functionality of the central pacemaker in synchronizing the body clocks to
changing environmental time cues (Brown and Piggins, 2007; Golombek and Rosenstein,
2010). Single-unit activity (SUA) in SCN neurons exhibits a daily fluctuation of
spontaneous action potentials (SAP), with higher frequency during the daytime than night
(Prosser and Gillette, 1989; Prosser et al., 1989). Patch-clamp recordings of neuronal
membrane properties from mouse SCN, over the circadian cycle, presents similar patterns
(Belle et al., 2009). Ca2+i, another indicator of neuronal excitability, also exhibits a
circadian oscillation in SCN neurons (Ikeda et al., 2003).
Recent studies have found evidence of cross-talk between the molecular clock and
ionic translocations across both plasma membrane (Na+, K+, Ca2+ flux-dependent
membrane potential fluctuations) and endoplasmic reticulum (ER) membrane (Ca2+ fluxdependent transient Ca2+i increase). In Drosophila, K+ channel-mediated electrical
silencing leads to deficits in circadian rhythms of locomotor, and rundown of the cellautonomous intracellular molecular clock (Nitabach et al., 2002). In the mammalian SCN,
a periodic transmembrane Ca2+ influx is essential for the sustained molecular clockwork
(Lundkvist et al., 2005). These findings support a more fundamental role of neuronal
excitation, not only as an output of core clock to synchronize peripheral clocks, but by
5
itself as an key element in the generation of the intrinsic rhythm.
Light/Glutamate Signaling in SCN Neurons
Glutamate (Glu) is the primary neurotransmitter that the RHT employs to communicate
with SCN neurons (Ding et al., 1994; Ebling, 1996). It activates the N-methyl d-aspartate
receptor (NMDA-R) on the neurons, which permits Ca2+ influx into the cell (Ding et al.,
1994). Ca2+ influx in turn activates nitric oxide synthase (NOS), leading to the release of
nitric oxide (NO) (Ding et al., 1994; Watanabe et al., 1994; Watanabe et al., 1995; Weber
et al., 1995), which serves as an intercellular messenger to neurons without direct
innervations from RHT (Golombek and Rosenstein, 2010). Although light acts via Glu
stimulation, NMDA-R opening, and NOS activation in both early and late night, the
signaling pathway bifurcates downstream (Ding et al., 1994; Gillette and Mitchell, 2002).
In early night, the ryanodine receptor (RyR) is activated and Ca2+ is presumably released
from ER Ca2+i store (Ding et al., 1998). The elevated Ca2+i further activates Ca2+dependent kinases, such as CamKII and MAPK (Golombek and Ralph, 1994; Golombek
and Ralph, 1995; Shibata and Moore, 1994; Tischkau et al., 2003), leading to the
phosphorylation of cAMP-response element binding protein (CREB) and subsequent
transcription of the Period 2 gene, part of the molecular feedback loop (Ding et al., 1997;
Gau et al., 2002; Tischkau et al., 2003). These are necessary elements of the signal
transduction leading to phase delay. During late night, NO, elicited by Ca2+ influxstimulating NOS, activates soluble guanylate cyclase (sGC) and cGMP is produced
(Tischkau et al., 2003; Tischkau et al., 2004); therefore, cGMP-dependent protein kinase
(PKG) pathway is activated and transcription of the Period 1 gene is induced (Tischkau
6
et al., 2003). These elements are necessary to the signal transduction leading to phase
advance. (Fig. 1.1)
Glutamate Signaling and Redox Regulation
Previous work in our lab found that SCN redox state could switch the directionality of the
Glu-induced phase-shift in early and late night (Yu, 2007). This was examined by SUA
recording of the near 24-hour rhythm of spontaneous firing rate in the SCN brain slice.
Pre-treatment of the SCN slice with a reducing reagent, glutathione (GSH), before Glustimulation at CT 14 results in a phase advance, rather than delay. On the other hand,
exposure to an oxidizing reagent, diamide (DIA), before Glu-stimulation at CT 19 inverts
the response, inducing phase-delay rather than phase-advance. These findings suggest
that redox state may participate in determining the bifurcation point of Glu signaling.
Aims and Significance
While intracellular redox state represents the metabolic requirement of circadian clock,
SCN neurons function by circadian fluctuation in neuronal excitation. We thus
hypothesize that both of these parameters should exhibit strong circadian rhythms. In
addition, we hypothesize that there exists a regulatory relationship between intracellular
redox state and neuronal excitability in SCN neurons. Furthermore, we hypothesize that
this modulation plays an essential role in signal transduction in SCN.
The aims of this study are two-fold: 1) to examine the circadian rhythm of redox
state and its regulation of membrane excitability in rodent SCN; 2) to explore the
physiological importance of redox regulation of neuronal excitability in the signal
7
transduction in clock entrainment.
We will use real-time imaging to examine the redox fluorometry of SCN neurons
over the period of 24-h. We will perform whole-cell patch-clamp recordings on SCN
neurons and examine their electrophysiological profiles at different CTs, points in the
circadian cycle. In addition, we will manipulate redox state in SCN by exogenous
pharmacological treatment and examine how membrane properties and Ca2+i respond.
We will perform two-photon microscropy to study Ca2+i signaling in SCN neurons, to
elucidate the bifurcation point, as well as the target of redox regulation of Glu signaling.
The significance of this study lies in evaluating the interdependency between
redox state and neuronal excitability, so as to bridge the gap between cellular metabolism
and system physiology in the circadian clock.
8
Figure 1
Light
Early night
Late night
RyR
Glutamate
sGC
Ca2+i release
NMDA-R
cGMP
Ca2+ influx/NO
PKG
Phase Advance
Phase Delay
Fig. 1.1. Key elements on the signal pathway of light/Glu-induced phase-shift in SCN.
9
CHAPTER 2. CIRCADIAN RHYTHM OF REDOX STATE
REGULATES EXCITABILITY IN RODENT SCN NEURONS*
Introduction
In the early 1990’s, oscillations of membrane current and Ca2+i driven by metabolic state
were reported in rodent heart cells. The metabolic fluctuation was found to originate from
the level of glycolysis, and the membrane changes were mediated by an ATP-sensitive K+
channel (O'Rourke et al., 1994). These oscillations underlie heart function (beating), not
are directly relevant to circadian rhythms. Subsequently, studies of the redox regulation
of K+ channels, synaptic transmission, and Ca2+ signaling were carried out in both the
circulatory and nervous systems, mainly focusing on the consequence of metabolic deficit
and pathology, such as ischemia, stroke and epilepsy (Aon et al., 2008). Recently,
biophysicists provided insight into the molecular mechanism of redox regulation on ionchannels, based on studies from a clone-expression system: Pan and colleagues described
a mechanism by which N-type inactivation of Kvβ1 is modulated by redox state in a
NADPH-dependent manner (Pan et al., 2011).
Considering the concept that redox state regulates excitability, and that the
metabolism is extensively influenced by the circadian system, here we hypothesize that a
circadian rhythm of redox state might exist in SCN, and that this drives the circadian
oscillation in membrane excitability of SCN neurons.
*
The content of this chapter has been accepted by Science for publication: Wang, T.A., Yu, Y.V.,
Govindaiah, G., Ye, X., Artinian, L., Coleman, T., Sweedler, J.V., Cox, C.L., Gillette, M.U. (2012).
Circadian Rhythm of Redox State Regulates Excitability in Suprachiasmatic Nucleus Neurons.
Science. (In Press). The author of this dissertation (Wang, T.A.) performed all the experiments and
data analysis, except the DHA/AA assay, which were performed by Ye, X. and Yu, Y.V.
10
To test this hypothesis, we used organotypic slices of SCN-bearing hypothalamus
and performed ratiometric redox fluorometry using two-photon laser microscopy (Huang
et al., 2002). The relative redox state is measured non-invasively from the ratio of autofluorescence emissions in response to 730 nm excitation of two co-factors of cellular
metabolism, FAD (500+ nm) and NAD(P)H (400+ nm). Only oxidized FAD and reduced
NAD(P)H exhibit intrinsic fluorescence, and these two signals respond reversely to
changes in redox state; thus, ratiometric redox fluorometry based on FAD and NAD(P)H
can be used to evaluate cellular metabolism. This strategy is suitable for long-term
imaging, because it minimizes fluctuations of interfering factors, such as light absorption
of excitation and emission, light scattering, and mitochondrial density. In addition, we
performed biochemical and analytical chemistry to examine the phasing of redox state in
SCN.
To explore the interdependency of metabolic state and membrane excitability, we
performed patch-clamp recording in rat SCN neurons at different CTs, with
pharmacological manipulation of redox state. We further analyzed the possible target
redox regulation on membrane properties by voltage-clamp recording.
Materials and Methods
Animals and brain slice preparation
LE/BluGill rats (University of Illinois) were used for real-time imaging of redox state
measurement, BioGEE assay, DHA/AA assay and patch-clamp recording; they were
maintained under a 12:12 hour L/D schedule, receiving food and water ad libitum.
C57BL6 mice from Jackson Laboratory (Bar Harbor, ME) were used for real-time
11
imaging of redox state, maintained under a 12:12 hour L/D schedule, receiving food and
water ad libitum. Bmal 1 +/- heterozygotes were obtained from Jackson Laboratory (Bar
Harbor, ME), and homozygous Bmal 1 -/- were bred from the heterozygotes. Bmal 1 -/mice were maintained under a 12:12 hour L/D schedule, receiving food and water ad
libitum, until 3 d before experiments, when they were moved to dark/dark, receiving food
and water ad libitum (Bunger et al., 2000). All the protocols were approved by IACUC at
University of Illinois, Urbana-Champaign, and fully compliant with NIH guidelines for
humane treatment of animals.
Organotypic brain slice cultures were prepared from 3-4-wk-old animals,
sacrificed between ZT 5-7. A 400-µm-thick coronal hypothalamic brain slice containing
the paired SCN was cut on mechanical chopper, then transferred to a Millipore tissue
culture insert (Millipore, Billerica, MA) with DMEM containing 0.5% B27 supplement,
1.0 mM glutamine and 25 µg/ml penicillin/streptomycin (Gibco, Carlsbad, CA).
Organotypic slices were fed the next day and ready for imaging after 2 d in vitro (DIV).
Animals for BioGEE and DHA/AA assays were 5-7-wk old. A 500-µm-thick
coronal hypothalamic brain slice containing the paired SCN was cut on mechanical
chopper between ZT 2-10, depending on the experimental time to be examined. Some
samples were duration-matched to control for time in vitro. The SCN was punched with a
2 mm-diameter corer, to remove non-SCN tissues. These reduced brain slices were
maintained in the brain slice chamber, perfused with Earle’s Essential Balanced Salt
Solution (EBSS, NaCl 116.4 mM, KCl 5.4 mM, CaCl2 1.8 mM, MgSO4 0.8 mM,
NaH2PO4 1.0 mM, glucose 24.5 mM, NaHCO3 26.2 mM, gentamicin 1 mg/L, pH 7.3, 300
mOsm/L) saturated with 95% O2/5% CO2 at 37°C for at least 2 h until collection for
12
biochemical assay at specific CTs.
Animals for patch-clamp studies were 2-wk old, and sacrificed from ZT 2-10,
depending on the experimental time to be examined. The brain was quickly removed and
placed into cold, oxygenated slicing medium: KCl 2.5 mM, MgSO4 10.0 mM, CaCl2 0.5
mM, NaH2PO4 1.2 mM, NaHCO3 26.0 mM, glucose 11.0 mM, and sucrose 234.0 mM,
saturated with 95% O2/5% CO2. A 350-µm-thick brain slice containing paired SCN was
cut using a vibrating blade microtome (Leica, Wetzlar, DE). Brain slices were then
transferred to a holding chamber containing artificial cerebrospinal fluid (ACSF): NaCl
126.0 mM, KCl 2.5 mM, CaCl2 2.0 mM, MgCl2, 2.0 mM, NaH2PO4 1.2 mM, glucose 10
mM, NaHCO3 26.0 mM, 300 mOsm/L, saturated with 95% O2/5% CO2 at room
temperature. Brain slices were incubated at least 1 h before recording commenced.
Real-time redox imaging (long-term)
An organotypic slice cultured for 2 DIV was transferred to a 37 °C chamber on the
microscope stage, where EBSS was perfused continuously. Two-photon laser-scanning
microscopy was performed with the Zeiss LSM 510 confocal laser-scanning microscope
system equipped with MaiTai laser and 20X 0.8 N.A objective (Carl Zeiss, Obercochen,
DE). Excitation wavelength was 730 nm, while 2 channels of emission at 430-500 nm
and 500-550 nm were recorded simultaneously (Huang et al., 2002). Tissue was
examined up to 15 µm from the surface. Imaging started at CT 10 and frame-scan, with
sampling rate at 4 sec/frame plus 356-sec interval, was performed for 720 frames (72 h
total). Fluorescence intensity of series scanning at 400+ and 500+ nm for each frame was
acquired by Zeiss LSM software (function of region of interest, ROI). Relative redox
13
state was calculated from the ratio of fluorescence at 500+ over 400+ (F500+/F400+).
χ2 periodogram was performed with a MATLAB toolbox, ClockLab (Actimetrics,
Wilmette, IL), to quantify the length of circadian period (τ). χ2 values were calculated
from recording data from WT rat, WT mouse, and KO mouse, and the τ was determined
from the highest value above confidence interval of 0.001. Recording datasets with no χ2
values above 0.001 confidence interval, or with calculated τ > 32 h, which is the
mathematical consequence of estimating a period that is approximately half of the length
of the recording time, were not regarded as circadian.
Real-time redox imaging (short-term)
An organotypic slice cultured for 2 DIV was incubated in EBSS for at least 2 h before
transferred to microscope stage, with the same set-up as above. A frame-scan, with
sampling rate at 4 sec/frame plus 26-sec interval, was performed for 60 frames (30 min
total). Redox reagents (diamide, DIA, 5 mM or glutathione, GSH, 1 mM, , Sigma, St.
Louis, MO) were delivered through a syringe pump from 5 – 15 min. Fluorescence
intensity of series scanning at 400+ and 500+ nm for each frame was acquired by Zeiss
LSM software (function of region of interest, ROI). Relative redox state was calculated
from the ratio of fluorescence at 500+ over 400+ (F500+/F400+). Redox changes before and
during drug treatment are based on the average ratio of F500+/F400+ between 1-2 min vs.
15-16 min, respectively.
BioGEE assay
Reduced SCN brain slices (Gillette, 1991) were incubated with 250 µM biotinylated
14
glutathione ethyl ester (BioGEE, Invitrogen, Carlsbad, CA) for 1 h (Sullivan et al., 2002)
at CT 0-1, 6-7, 11-12, 13-14, or 19-20. After freezing by dry ice, tissue was stored at 80 °C until assay. Frozen tissue samples were mixed with 30 µL Tissue Protein Extraction
Reagent (T-PER, Pierce, Rockford, IL), plus 0.2% SDS, 1 mM EDTA, and 1x Complete
protease inhibitor cocktail (CalBioChem, Darmstadt, DE) on ice, and homogenized. After
2-min incubation on ice, the samples were centrifuged at 14,000 rmp for 2 min, and the
supernatant was transferred to a clean tube. Protein content of each sample was
determined by BCA protein assay (Pierce, Rockford, IL). Total protein (25 µg/sample)
was resolved in 8% SDS-PAGE and transferred to nitrocellulose membrane (Bio-Rad,
Hercules, CA). Membranes were probed with 1:2,000 dilution of mouse anti-biotinperoxidase antibody (Cellsignaling, Danvers, MA) overnight, and developed with
SuperSignal West Femto Maximum Sensitivity Substrate (Pierce, Rockford, IL). After
scanning, the anti-biotin antibody was stripped by Restore Western Blot Stripping Buffer
(Thermo, Rockford, IL), and an anti-tubulin antibody (Cellsignaling, Danvers, MA) was
applied to probe total tubulin level. After incubation with secondary antibody, the
membrane was developed with SuperSignal Pico Chemiluminescent Substrate (Thermo,
Rockford, IL), and scanned again. The relative glutathiolation level was determined by
the ratio of overall biotin intensity of each lane over the band intensity of tubulin, and
then normalized to the maximum value on the same gel. One-Way ANOVA was used for
statistical analysis, followed by Tukey HSD Test for multiple comparisons.
DHA/AA assay
Reduced SCN brain slices were collected at CT 4, 8, 14, 20, and 23 in pre-chilled
15
microtubes on dry ice. Each slice was homogenized with 50 µL acetate buffer, and
further centrifuged at 10,000 rmp at 4 °C for 5 min; two aliquots (2.5 µL each) of
supernatant were sampled for dehydroascorbic acid (DHA) or DHA+ ascorbic acid (AA)
concentration, respectively. For DHA derivatization, the sample was mixed with 1.5 µL
of 1 mg/mL 4,5-dimethyl-1,2-phenylenediamine (DMPD) solution in phosphate buffer;
after 4-min incubation at room temperature, an extra 1 µL of phosphate buffer was added
to bring the final volume to 5 µL. DHA+AA concentration was obtained with similar
procedure, except for adding 1 µL of 17 unit of ascorbate oxidase (Roche, Basel, CH)
solution and reacting for 1 min before derivatization, to oxidize AA to DHA. DHA or
DHA+AA were measured by capillary electrophoresis with laser-induced fluorescence
(CE-LIF) detection using DMPD (Kim et al., 2002). Peak heights in standards and
samples were determined by Grams 386 software (Thermo Galactic, Salem, NH), and
sample concentration was determined by linear regression; the ratio of DHA/AA was
calculated to evaluate the relative redox state in SCN tissue. One-Way ANOVA was used
for statistical analysis, followed by Tukey HSD Test for multiple comparisons.
Patch-clamp recording
Whole-cell patch electrodes had pipette tip resistances of 4-6 MΏ, and were filled with
solution containing: K-gluconate 140 mM, KCl 5.0 mM, CaCl2 0.07 mM, MgCl2 1.0 mM,
EGTA 0.1 mM, HEPES 10 mM, Na-ATP 4 mM, Na-GTP 0.4 mM, pH 7.3, and osmolality
290-300 mOsm/L. Recordings were performed using a Multiclamp 700B amplifier
(Molecular Devices, Sunnyvale, CA). Signals were sampled at 10 kHz, low-pass filtered
at 10 kHz using a Digidata 1320 digitizer, and stored on computer for subsequent
16
analyses using pClamp software (Molecular Devices, Sunnyvale, CA).
Membrane potential (Vm) of SCN neurons was recorded under current-clamp
mode, 2 min/cell. A total of 364 SCN neurons were recorded at different CTs, with
running means calculated over 24 h. In addition, average Vm at 5 CTs (CT 1, 7, 11, 14,
and 20) were calculated. One-Way ANOVA was performed for statistical analysis,
followed by Tukey HSD Test for multiple comparisons.
Input resistance of SCN neurons was recorded from the same population as for
Vm above. During recording, a current injection of -20 pA every 17 s was applied to test
Vm changes, and input resistances were determined from the changes of Vm over testing
current (Rin = ∆V / -20 pA). A sub-population of the current-clamped SCN neurons were
challenged with a current-steps (duration 800 ms) from -100 pA to 120 pA, with a 20 pA
increment at CTs 1, 7, 11, 14, and 20, to construct the current-voltage (I-V) curve and
precisely calculate input resistance. In this case, input resistance was determined from the
slope of the I-V curve at 0 pA holding current by linear regression. One-Way ANOVA
was performed for statistical analysis, followed by Tukey HSD Test for multiple
comparisons.
Some of the current-clamped SCN neurons were treated with redox reagent (DIA
or GSH) and responses were examined at specific CTs. The initial Vm was recorded
immediately after whole-cell clamp; and then a current-steps protocol was performed to
construct the I-V curve at rest. Next, redox reagents were bath-applied and ∆Vm was
recorded. Before wash-out, another current-steps protocol was performed so that the I-V
curve under drug treatment was obtained. To prevent secondary effects from synaptic
transmission, tetrodotoxin (TTX, 1µM, Tocris, Ellisville, MO), a voltage-gated Na+
17
channel blocker, was applied during the entire process.
Changes of membrane potential (∆Vm) in response to redox reagents were
compared by paired Student’s t-Test. ∆Vm was further classified by CT and averaged.
One-Way ANOVA followed by the Tukey HSD Test was performed for statistical analysis.
Redox reagent-induced changes in the I-V curve and input resistance were compared by
paired Student’s t-Test. Percentage changes of input resistance were further calculated and
classified by CT. Averaged data were subjected to One-Way ANOVA.
Under voltage-clamp mode, a slow ramp of voltage from -50 mV to -110 mV and
back to -50 mV was applied to SCN neurons, 6 s/trial, 3 trials/min. During this process,
redox reagents were bath-perfused and the holding currents changes were recorded.
Inhibitors of voltage-gated Na+ channels (TTX, 1µM), AMPA-Rs (DNQX, 20 µM, Tocris,
Ellisville, MO) and NMDA-Rs (d-CPP, 10 µM, Tocris, Ellisville, MO) were applied
during the entire process. Holding current was plotted against the command voltage so
that the I-V curves before and during treatments were obtained. DIA-evoked currents
were calculated based on the difference between the two I-V curves. To further test the
hypothesis that K+ channel(s) is a target of redox regulation, Cs+ (140 mM) was used as
an alternative of K+-filled glass electrode. Difference between control and DIA treatment
on the holding current and conductance changes were determined from the first 100
pixels of the ramp at -50 mV, and analyzed by paired Student’s t-Test. Similar recordings
and analysis were performed with reducing reagent, GSH, and with K+ electrodes in the
presence of bupavacaine (Bupi, 100 µM, Tocris, Ellisville, MO), a leak K+-channel
blocker. One-Way ANOVA followed by the Tukey HSD Test was performed for statistical
analysis.
18
The voltage-step command recording was performed in ACSF with HEPES buffer:
NaCl 138.0 mM (118.0 mM, in the presence of TEA), KCl 2.5 mM, CaCl2 2.0 mM,
MgCl2, 2.0 mM, glucose 10.0 mM, HEPES 10.0 mM, 300 mOsm/L, saturated with 100%
O2 at 37 °C. Voltage-clamped SCN neurons were held at -60 mV, and challenged with -10
mV pulses (250 ms duration) twice with an interval of 3 s; a 400 ms pre-pulse of -90 mV
then -40 mV was delivered before each pulse. A cocktail of inhibitors of voltage-gated
Na+ channels (TTX, 1 µM), Ca2+ channels (Cd2+, 200 µM), AMPA-Rs (DNQX, 20 µM),
NMDA-Rs (d-CPP, 10 µM) and GABAA-Rs (SR95531, 10 µM, Tocris, Ellisville, MO)
was applied in ACSF during the entire process; in addition, 10 mM EGTA was included
in the recording electrode solution to block Ca2+i fluctuation. Each sweep was repeated
every 20 s, before, during and after treatment with redox reagent (DIA and GSH), as well
as K+ channels inhibitors (4-aminopyridine, 4-AP, 5 mM; tetraethylammonium, TEA, 20
mM, Sigma, St. Louis, MO). Voltage-dependent K+ current in response to -10 mV
voltage-step stimulation was calculated from the difference between the current responses
to the voltage-step command of -10 mV steps, following either -90 mV or -40 mV prepulse. The transient peak current within 10 ms and the persistent current between 230 –
250 ms were analyzed across the treatments of redox reagent (DIA, GSH) and K+
channels blockers (4-AP, TEA). One-Way ANOVA followed by the Tukey HSD Test was
performed for statistical analysis.
Results
By real-time redox imaging, we found an endogenous, near-24-h oscillation of redox
state in SCN tissue from wild-type rat and mouse (τ = 23.74 ± 0.26 h, τ = 23.75 ± 0.30 h,
19
respectively, χ2 periodogram analysis, n = 5, Fig. 2.1A, B, D, E). Application of the
oxidizing reagent, diamide (DIA, 5 mM), increases the FAD/NAD(P)H ratio within 10
min (∆F500+/F400+ = 0.022 ± 0.018, P < 0.05, paired Student’s t-Test, n = 6, Fig. 2.2A, B).
On the other hand, exposure to a reducing reagent, glutathione (GSH, 1 mM), decreases
the ratio (∆F500+/F400+ = -0.022 ± 0.010, P < 0.01, paired Student’s t-Test, n = 6, Fig.
2.2C). To determine whether the circadian oscillation of redox state requires an intact
molecular clockwork, we evaluated SCN slices from Bmal 1 -/- mice, which lack
circadian rhythms (Bunger et al., 2000; Ko et al., 2010). Compared with wild types, Bmal
1 -/- SCN exhibit stochastic, but not circadian, oscillations in relative redox state (Fig.
2.1C, F, n = 5). Thus, circadian redox oscillations in rodent SCN require a functional
molecular clockwork involving the clock gene, Bmal 1.
To determine temporal phasing of the SCN redox oscillation, we evaluated two
indicators of redox state. First, we examined points across the circadian cycle of SCN
brain slices for glutathiolation, the capacity of SCN proteins to incorporate reduced GSH,
which binds to available disulfide bonds (Sullivan et al., 2002). We found a significant
change in glutathiolation, with a higher level in early night (CT 14, 97.36 ± 2.39 %),
indicating a relatively oxidized state, and a lower glutathiolation level in mid-day (CT 7,
79.01 ± 3.30 %), reporting a relatively reduced state (Fig. 2.3A, B, P < 0.05, One-Way
ANOVA, n = 6). Second, we analyzed the ascorbic acid system, an important antioxidant
and neuroprotective buffer in the brain (Rice, 2000). We employed capillary
electrophoresis with laser-induced fluorescence detection (CE-LIF) to directly measure
the concentrations of dehydroascorbic acid (DHA) vs. its reduced counterpart, ascorbic
acid (AA) (Kim et al., 2002; Lapainis and Sweedler, 2008). We found that DHA/AA
20
oscillates with a circadian rhythm, similar in pattern to glutathiolation levels: high
DHA/AA in early night and low in midday (Fig. 2.3C, P < 0.05, One-Way ANOVA, n =
3). Parallel changes in these distinct redox systems support circadian oscillation of global
redox state in rat SCN, with a significantly oxidized state in early night vs. reduced state
during daytime.
To assess possible relationships between the circadian oscillation of redox state
and neuronal physiology, we evaluated membrane excitability in rat SCN neurons by
examining intrinsic membrane properties, including resting membrane potential (Vm),
input resistance, and spontaneous action potential (SAP). SCN neurons generate selfsustained circadian rhythms of electrical activity that communicate clock phase to other
brain and body sites (Brown and Piggins, 2007; Guo et al., 2005; Schwartz et al., 1987).
Recording from current-clamped neurons, we observed circadian oscillations of Vm (Fig.
2.4A, n = 364; Fig. 2.4B, n = 36-60 at 5 circadian times (CTs) around the free-running
clock cycle, P < 0.001, One-Way ANOVA), input resistance (Fig. 2.4C, n = 337; Fig.
2.4D, n = 15-30, P < 0.05, One-Way ANOVA), frequency of SAP (Fig. 2.4E, n = 334),
and percentage of neurons discharging SAP (Fig. 2.4F, n = 334). During subjective
midday (CT 7), SCN neuronal activity is maximal: resting Vm is relatively depolarized (55.17 ± 0.61 mV), with higher input resistance (816.6 ± 58.7 MΩ), SAP frequency (1.65
Hz, mean of active and inactive cells), and percentage of active cells (57.8%). During
subjective night, these parameters are significantly lower: Vm is hyperpolarized
(minimum at CT 14, -60.15 ± 0.69 mV), with decreased input resistance (CT 14, 583.7 ±
32.0 MΩ), SAP frequency (CT 13, 0.55 Hz), and active-cell percentage (CT 16, 20.8%).
Comparing circadian oscillations of redox state (Fig. 2.3B, C) with Vm in SCN neurons
21
(Fig. 2.4B) reveals that the daytime reduced state matches the depolarized Vm, whereas
the oxidized state of subjective night aligns with hyperpolarized Vm.
To probe potential interdependency of redox state and neuronal excitability, we
tested effects of pharmacological redox manipulation on Vm of SCN neurons at various
points in the circadian cycle. We found that the oxidizing reagent, DIA (5 mM),
hyperpolarizes Vm in a reversible manner (Fig. 2.5A, mean around circadian cycle -8.68
± 0.64 mV, n = 104); in contrast, exposure to the reducing reagent, GSH (1 mM), causes
depolarization (Fig. 2.5C, +10.67 ± 1.03 mV, n = 97). Exogenous redox regulation of Vm
in SCN depends upon CT: both oxidizing and reducing reagents cause maximal effects
(hyperpolarization or depolarization, respectively) preceding the transition from day to
night (subjective dusk, CT 10 – 12), and compared with minimal effects near subjective
dawn (CT 0 – 2, Fig. 2.5E, P < 0.01, One-Way ANOVA, n = 10-20).
Shifts in Vm due to exogenous redox reagents are associated with changes of input
resistance. Based upon the slopes of current-voltage (I-V) curves constructed before and
during redox reagent treatment, the oxidizing reagent decreases the input resistance of
SCN neurons by 364 ± 30 MΩ (48.83%, Fig. 2.5B, n = 41), while the reducing reagent
increases input resistance by 64 ± 11 MΩ (10.81%, Fig. 2.5D, n = 36). Percentage
changes in input resistance are not correlated with changes in Vm (P > 0.05, Linear
Correlation and Regression, data not presented) and are independent of CT (Fig. 2.5F, P >
0.05, One-Way ANOVA, n = 5-9). Redox-induced changes in membrane properties are
rapid, occurring in < 2 min (latency = 68.7 ± 10.6 s).
To identify potential targets for redox regulation of neuronal excitability, we
compared I-V curves before and during exposure to redox reagents (Fig. 2.5B, D). The
22
membrane current produced by DIA or GSH reverses at -90 mV, near the calculated
equilibrium potential of K+. This suggests that targets of redox regulation could be K+
channels.
To further evaluate the ionic mechanism, ramped command voltages from -50 to 110 mV (6-s duration) were performed on voltage-clamped SCN neurons between CT 913. By comparing membrane currents before and during exposure to the redox reagents,
the voltage-dependency of target ion channel(s) was isolated. Bath application of
oxidizing reagent (DIA) elicits a strong outward current (100.1 ± 22.8 pA at -50 mV, n =
6, Fig. 2.6A, B, C, G), associated with an increased conductance as indicated by the
greater slope of the current response to the ramped voltage command. The I-V
relationship before and during DIA treatment reveals a reversal potential at -81.4 ± 7.0
mV and conductance increases of 154.9 ± 22.5 % (Fig. 2.6B, H). This suggests that the
action of DIA is via an outward-rectifier K+ channel(s) (Fig. 2.6C).
We confirmed this prediction by replacing K+ with Cs+, a general K+-channel
blocker, in the recording pipette. Under this condition, the oxidizing reagent-evoked
outward current is significantly attenuated (-1.6 ± 1.2 pA at -50 mV, P < 0.01, Tukey
HSD Test, n = 6, Fig. 2.6D-G). Only small conductance changes could be detected during
DIA treatment (14.2 ± 4.7 %, Fig. 2.6E, H). In contrast to the effects of oxidizing reagent,
exposure to the reducing reagent GSH elicits an inward current in SCN neurons (-40.6 ±
8.5 pA at -50 mV, n = 6, Fig. 2.6G, sample trace not presented), with a reversal potential
at -79.9 ± 4.0 mV and a conductance change of -38.2 ± 6.3 % (Fig. 2.6H). The GSHevoked inward current is attenuated by Cs+ in the internal solution, as well (-7.9 ± 2.1 pA
at -50 mV, 34.1 ± 16.8 % conductance changes, P < 0.01, Tukey HSD Test, n = 6, Fig.
23
2.6G, H), also supporting the involvement of a K+ current.
To explore the potential role of leak K+ channels in redox regulation, a specific
blocker of this channel, bupivacaine (Bupi, 100 µM), was bath-applied with a normal
recording electrode. In the presence of Bupi, DIA induced an outward current of 31.9 ±
4.4 pA at -50 mV (Fig. 2.6G, n = 5), with conductance changes of 70.4 ± 11.2% (Fig.
2.6H, n = 5), but amplitudes were lower than ACSF only (Fig. 2.6G, H, P < 0.05, Tukey
HSD Test). Bupi attenuated GSH-induced inward current and conductance changes, as
well (-40.6 ± 8.5 pA at -50 mV, 11.0 ± 12.4%, P < 0.05, Tukey HSD Test, n=5, Fig. 2.6G,
H). These results support the leak K+ channel as a target of redox reagent-induced
currents.
Considering the voltage-dependence of redox reagent-evoked current under
voltage-ramp command, we used voltage-step commands to examine the possibility that
redox state regulates voltage-gated K+ channels (Fig. 2.7A-C). Voltage-clamped SCN
neurons were held at -60 mV, and challenged with -10 mV pulses (250 ms duration) twice
with an interval of 3 s; the first test pulse was following a pre-pulse of -90 mV (400 ms
duration), while the second was following a -40 mV pre-pulse. Current responses to the
two test pulses were recorded, and differences were calculated, so as to determine the
time-course of voltage-dependent K+ currents in response to voltage-step stimulation of 10 mV (Fig. 2.7D).
We found that oxidizing reagent, DIA, induced a significant increase in the
transient peak of the outward current (288.4 ± 36.5 pA, n = 5, Fig. 2.7D, G), which was
completely abolished by 4-aminopyridine (4-AP, 5 mM), a selective A-type K+-channel
blocker (-11.7 ± 17.2 pA, P < 0.01, Tukey HSD Test, n = 6, Fig. 2.7E, G). On the other
24
hand, the DIA-evoked transient outward current was insensitive to tetraethylammonium
(TEA, 20 mM), a delayed rectifier K+-channel blocker (282.4 ± 34.7 pA, P > 0.05, Tukey
HSD Test, n = 6, Fig. 2.7F, G). Meanwhile, the persistent outward current was insensitive
to DIA treatment with/without 4-AP or TEA (P > 0.05, One-Way ANOVA, n = 5 - 6, Fig.
2.7H). The reducing reagent, GSH, suppressed the transient peak of the outward current
induced by -10 mV steps (-149.5 ± 37.3 pA, sample trace not presented, n = 5, Fig. 2.7G);
similar to DIA, the suppression was sensitive to 4-AP (6.8 ± 14.3 pA, P < 0.01, Tukey
HSD Test, n = 5, Fig. 2.7G), but not TEA (-164.8 ± 25.4 pA, P > 0.05, Tukey HSD Test, n
= 6, Fig. 2.7G). The persistent outward current was not affected by GSH treatment
with/without 4-AP or TEA (P > 0.05, One-Way ANOVA, n = 5 - 6, Fig. 2.7H). These
results support the involvement of a 4-AP sensitive voltage-gated K+ channel in redox
regulation.
Discussion
In summary, we found that global redox state oscillates in SCN of rat and mouse. The
circadian rhythm in redox state was detected by three complementary measurements: 1)
FAD/NAD(P)H auto-fluorescence assessed non-invasively in real-time, 2) glutathiolation
analyzed by immunoblot, and 3) DHA/AA levels by analytical chemistry. Circadian
rhythms in redox state based upon peroxiredoxin oligomerization have been reported in
red blood cells (RBC) and green algae (O'Neill and Reddy, 2011; O'Neill et al., 2011),
however their relevance to cellular function and system physiology were not explored.
Rhythmic activity of peroxiredoxin is independent from transcription/translation in the
anucleate RBC, but is disrupted in embryonic fibroblasts from clock-mutant mice
25
(O'Neill and Reddy, 2011). This suggests that in nucleated cells, the transcriptional and
redox oscillators are closely coupled. Our results support and extend the coupling of
transcriptional and redox oscillators to the SCN (Fig. 4).
While redox state is influenced by the transcriptional/translational machinery, the
redox oscillator can affect clock gene expression reciprocally (Rutter et al., 2002; Green
et al., 2008; Bass and Takahashi, 2010). Redox species such as NAD(P)H and CO
regulate the clock proteins CLOCK/BMAL1 and CRY through an electron transport
chain involving heme moieties, which enables clock proteins to alter transcription in
response to metabolic change (Rutter et al., 2001; Dioum et al., 2002). Rhythmic
expression of nuclear hormone receptors, such as REV-ERBα and RORα, provides an
additional pathway through which metabolic signals can engage the core clockwork
(Balsalobre et al., 1998; Preitner et al., 2002). Here we find that redox effects on neuronal
membrane physiology are on a time-scale orders of magnitude faster than transcription,
as is appropriate for altering excitability. Our finding of a regulatory relationship between
the redox state and SCN neuronal excitability provides a novel mechanism of metabolic
signaling, one where the cellular metabolism can directly influence input and output
pathways of the central circadian clock without transcriptional regulation (Fig. 4).
The daily rhythm of electrical activity in SCN is essential for the functionality of
the central pacemaker in synchronizing the body clocks to changing environmental time
cues (Brown and Piggins, 2007; Golombek and Rosenstein, 2010). SUA in SCN neurons
exhibits a daily fluctuation of SAP, with higher frequency during the daytime than night
(Prosser and Gillette, 1989; Prosser et al., 1989). Patch-clamp recordings of membrane
properties from rat (Fig. 2.4) and mouse (Belle et al., 2009) SCN, over the circadian cycle,
26
exhibits similar patterns. At least three ionic factors, K+ and Ca2+ currents, as well as
cytosolic Ca2+ concentration underlie rhythmic oscillating Vm (Belle et al., 2009; Brown
and Piggins, 2007). Patch-clamp recordings of SCN neurons at various CTs have
identified K+ channels involved in determining the circadian oscillation of neuronal
excitability. These include an outward-rectifier K+ current (De Jeu et al., 2002), a fastdelayed-rectifier K+ current (Itri et al., 2005), and an A-type K+ current (rapid
inactivation) (Itri et al., 2010). Modulation of these currents could be on the level of
channel expression, conductance, or gating. We found that K+ conductance regulated by
redox state undergoes circadian changes in SCN neurons (Fig. 2.5), with an I-V
relationship consistent with an outward-rectifier (Fig. 2.6), and sensitive to Bupi and 4AP (Fig. 2.6 and 2.7), suggesting the involvement of both leak and A-type K+ channels.
This is the first demonstration of redox-based post-translational modulation of neuronal
excitability in the brain.
27
Figures 2
D
A 1.00
800
600
0.99
400
0.98
200
0.97
0
0
24
48
72
E
Amplitude
F500+/F400+
B 1.02
1.01
1.00
0.99
24
36
12
24
36
800
600
400
200
0
0.98
C 1.02
12
0
24
48
72
F
800
600
1.00
400
0.98
200
0.96
0
0
24
48
CT (h)
72
12
24
36
Period (h)
Fig. 2.1. Redox state undergoes circadian oscillation in rodent SCN (i). (A-C) Realtime imaging of relative redox state in SCN of wild-type rat (A), wild-type mouse (B)
and Bmal1 -/- mouse (C). (D-F) χ2 periodograms (solid line) of redox oscillations in
SCN of wild-type rat (D), wild-type mouse (E), and Bmal1 -/- mouse (F), based on
data in A-C, respectively, with the confidence interval of 0.001 (dashed line); τ =
23.74 ± 0.26 h (rats), τ = 23.75 ± 0.30 h (mice); n= 5 for each group.
.
28
F500+/F400+
A
DIA
1.00
0.95
0.90
0
10
20
30
Time (min)
B
C
*
F500+/F400+
1.10
**
1.20
1.05
1.15
1.00
1.10
0.95
0.90
1.05
DIA
GSH
Fig. 2.2. Real-time changes in SCN redox state in response to exposure to redox
reagents. (A) Sample trace of real-time redox imaging of SCN slice, during
application of the oxidizing reagent, diamide (DIA, 5 mM), 5 - 15 min. DIA induces
an increase of the ratio of F500+/F400+ in < 2 min, indicating a shift toward oxidized
state. (B) Summary of the effect of DIA treatment, which shifts the SCN towards
oxidized state; each line represent an individual slice trial (*, P < 0.05, paired
Student’s t-Test; n = 6). (C) Summary of the effect of glutathione (GSH, 1 mM)
treatment, which shifts the SCN towards reduced state; each line represent an
individual slice trial (**, P < 0.01, paired Student’s t-Test; n = 6). Amplitudes of these
redox-reagent-induced shifts are comparable to the amplitude of the endogenous
redox oscillation (Fig. 2.1).
.
29
A
B
C
Fig. 2.3. Redox state undergoes circadian oscillation in rodent SCN (ii). (A)
Glutathiolation patterns of BioGEE incorporation into rat SCN tissue over 5 points of
circadian time (CT, which has a free-running time-base driven by the endogenous
clock). (B) Relative glutathiolation levels over 5 CTs in rat SCN (P < 0.05, One-Way
ANOVA; *, P < 0.05, Tukey HSD Test; n = 6). (C) DHA/AA ratio in rat SCN over 5
CTs (P < 0.05, One-Way ANOVA; *, P < 0.05, Tukey HSD Test; n = 3).
.
30
A
12
18
24
-50
-55
-60
-60
*
**
-65
*
D
1800
1200
1200
800
600
400
0
0
SAP Frequency (Hz)
Rin (MΩ)
B
-50
-70
C
E
6
0
6
12
18
8
6
4
2
0
0
6
12
18
24
F
24
CT (h)
*
1
Active Cell %
Vm (mV)
-40
0
7 11 14 20
60
40
20
0
0
6
12
18
24
CT (h)
Fig. 2.4. Circadian oscillation of membrane excitability in rat SCN neurons. (A)
Individual neurons (grey dots, 2-min recording, n = 364) and 1-h averages (star) of
membrane potential (Vm). (B) Vm averages at 5 CTs (P < 0.001, One-Way ANOVA;
**, P < 0.01, *, P < 0.05, Tukey HSD Test; n = 36-60). (C) Individual (grey dot, 2min recording, n = 337) and 1-h averaged (star) membrane input resistance, measured
by hyperpolarizing current steps. (D) Averaged input resistance by I-V curve at 5 CTs
(P < 0.05, One-Way ANOVA; *, P < 0.05, Tukey HSD Test; n = 15-30). (E) SAP
frequencies recorded from current-clamped SCN neurons (grey dot = SAP frequency
of each individual neuron, n = 334; black star = averaged SAP frequency in each
hour). (F) Percentage of neurons discharging SAP/ 2-min recording, plotted in each
hour over the circadian cycle.
.
31
A
B
C
D
E
F
Figures 2.5
32
Fig. 2.5. Redox regulation of membrane excitability in rat SCN neurons. (A, C)
Current-clamp recording of Vm in response to oxidizing reagent (A, DIA, 5 mM) or
reducing reagent (C, GSH, 1 mM), truncated SAP. When Vm plateaued, current was
injected to clamp the Vm back to the rest to measure the input resistance changes
during drug treatment. (B, D) I-V curve constructed by current steps from -100 to
+120 pA (20 pA increments, 800 ms duration) before (filled) and during (open) DIA
(B) or GSH (D) treatment. (E) Summary of redox reagent-induced changes in Vm at 5
CTs (P < 0.01, One-Way ANOVA; **, P < 0.01, *, P < 0.05, Tukey HSD Test; n = 1020). (F) Summary of redox reagent-induced changes in input resistance at 5 CTs (P >
0.05, One-Way ANOVA; n = 5-9).
.
33
A
D
B
E
C
F
G
H
Figures 2.6
34
Fig. 2.6. K+ current mediates redox regulation of neuronal excitability (CT 9-13). (A)
Voltage-clamp recording of SCN neuron with repeating slow-ramped voltage
commands from -50 mV to -110 mV, reveals that DIA treatment produces a reversible
outward current. (B) I-V curve constructed based on the command voltage and
membrane current recorded, before (black) and during (grey) DIA treatment. (C) The
DIA-evoked current as calculated from the difference in the membrane response in
(B). (D-F) Similar voltage clamp recordings as (A-C), except that Cs+ replaced K+ in
the patch pipette. (G) Holding current changes induced by redox reagents (DIA,
black; GSH, white) with electrodes containing K+ (P < 0.01, paired Student’s t-Test
for both DIA and GSH; n = 6), Cs+ (P > 0.05 for DIA, P < 0.05 for GSH, paired
Student’s t-Test; n = 6), or K+ (in electrode) with bupivacaine (Bupi, 100 µM) in bath
(P < 0.01, paired Student’s t-Test for both DIA and GSH; n = 5); for the comparison
across groups, P < 0.01, One-Way ANOVA; **, P < 0.01, *, P < 0.05, Tukey HSD
Test. (H) Conductance changes at -50 mV caused by redox reagents (DIA, GSH)
recorded with electrode containing K+ (P < 0.01, paired Student’s t-Test for both DIA
and GSH; n = 6), Cs+ (P < 0.01 for DIA, P < 0.05 for GSH, paired Student’s t-Test; n
= 6), or K+ (in electrode) with Bupi in bath (P < 0.01 for DIA, paired Student’s t-Test;
n = 5); for the comparison across groups, P < 0.01, One-Way ANOVA; **, P < 0.01,
*, P < 0.05, Tukey HSD Test.
.
35
A
B
D
C
G
E
H
F
Figures 2.7
36
Fig. 2.7. Redox regulation of voltage-dependent K+ currents (CT 9 – 13). (A)
Recording protocol of repeating voltage-step commands to voltage-clamped SCN
neurons; in each sweep, Vm was held at -60 mV, and challenged with -10 mV steps
(250 ms duration) twice with an interval of 3 s, following a 400 ms pre-pulse of -90
mV then -40 mV; each sweep was repeated every 20 s, before, during, and after
application of redox reagents (DIA and GSH), and/or K+-channel inhibitors (4aminopyridine, 4-AP, 5 mM; tetraethylammonium, TEA, 20 mM). (B, C) Current
responses to the voltage-step commands of -10 mV pulses, following either -90 mV
(B) or -40 mV (C) pre-pulse, before (black) or during (red) DIA treatment. (D)
Voltage-dependent outward current in response to -10 mV voltage-step stimulation,
calculated from the difference between the current responses in (B) and (C); DIA (red)
enhances the transient peak of the outward current. (E, F) Outward current evoked by
DIA is abolished by bath in 4-AP (E), but not TEA (F, control/TEA overlapping). (G)
Transient current (< 10 ms) changes in response to redox treatment (DIA, black; GSH,
white), with/without 4-AP or TEA (P < 0.01, One-Way ANOVA; **, P < 0.01, Tukey
HSD Test; n = 5-6). (H) Persistent current (230-250 ms) changes in response to redox
treatment, with/without 4-AP or TEA (P > 0.05, One-Way ANOVA; n = 5-6).
.
37
CHAPTER 3. CALCIUM SIGNALING AND REDOX REGULATION
IN GLUTAMATE-INDUCED PHASE-SHIFTS IN RAT SCN*
Introduction
In the signaling cascades of light-induced phase-shift in SCN, the same incoming signal
(light/Glu) triggers distinct downstream effectors between early and late night. The
cellular bases of these paradoxical responses have been unknown, but could be the result
from differential intracellular state of SCN neurons at different time windows.
Ca2+ plays a pivotal role in cellular signaling in almost all types of cells. As one of
the most important second messengers, Ca2+ sensors detect input signals and integrate
them via transient Ca2+i fluctuations, and generate output signals in response to external
signals of change. Consequently, Ca2+i serves as a functional link between rapid signaling
from plasma membrane and long-lasting cellular adaptive response. Ca2+ participates in
numerous cellular signal transduction pathways, with the signal specificity precisely
regulated by the spatial-temporal distribution of Ca2+i patterns (Verkhratsky, 2005;
Rizzuto and Pozzan, 2006).
In SCN neurons, Glu is able to activate Ca2+i store via RyR in early night; this
pathway cannot be activated in late night by Glu-stimulation (Ding et al., 1998). Thus,
the source and/or rate of Ca2+i release might distinguish the Glu-evoked signal
transduction between early and late night. To evaluate this hypothesis, here we performed
real-time Ca2+ imaging and patch-clamp recording to study Ca2+i signaling in rat SCN
*
Data from this chapter are in preparation for submission: Wang, T.A., Yu, Y.V., Govindaiah, G., Cox,
C.L., Gillette, M.U. (2012). Calcium Signaling and Redox Regulation in Glutamate-induced Phaseshifts in Rat Suprachiasmatic Nucleus Neurons. The author of this dissertation (Wang, T.A.)
performed all the experiments and data analysis in this chapter.
38
neurons, with the aim of elucidating the bifurcation point of Glu signaling.
Previous work in our lab found that the Glu-induced phase-shift in early and late
night can be switched by redox state manipulation (Yu, 2007), suggesting a role of redox
regulation in signal transduction of the clockwork machinery. Here we hypothesize that
the Ca2+ signaling in SCN phase-shift is regulated by redox state in early and late night,
with the target as the bifurcation point of the light/Glu signal cascades.
Materials and Methods
Animals and brain slice preparation
LE/BluGill rats (University of Illinois) were used for real-time Ca2+ imaging and patchclamp recording; they were maintained under a 12:12 hour L/D schedule, receiving food
and water ad libitum. All the protocols were approved by IACUC at University of Illinois,
Urbana-Champaign, and fully compliant with NIH guidelines for humane treatment of
animals.
Animals for Ca2+ imaging were 3-4 weeks old. They were sacrificed at ZT 10, and
a 400-µm-thick coronal hypothalamic brain slice containing the paired SCN was cut on
mechanical chopper. The brain slices were maintained in chamber with the perfusion of
EBSS at least 1 h before recording commences.
Animals for patch-clamp studies were 2-wk old, and sacrificed at ZT 10. The
brain was quickly removed and placed into cold, oxygenated slicing medium. A 350-µmthick brain slice containing paired SCN was cut using a vibrating blade microtome. Brain
slices were then transferred to a holding chamber containing ACSF at room temperature.
Brain slices were incubated at least 1 h before recording commences.
39
Real-time Ca2+ imaging
Brain slice was freshly prepared as stated above. At 2 hours before the time of
investigation, the brain slice was moved to a special loading chamber (250 µl volume),
loaded with 200 µl EBSS containing the Ca2+ indicator Oregon Green 488 BAPTA-1 AM
(10 µM, Invitrogen, Carlsbad, CA) for 90 min at 37°C. Then, the slice was washed in
EBSS in regular brain slice chamber for 30 min, and subsequently set onto microscope
stage perfused with EBSS. Two-photon laser-scaning microscopy was performed with
the Zeiss LSM 510 confocal laser-scanning microscope system equipped with MaiTai
and x40 1.2 N.A. water-immersion objective. Excitation wavelength was set to be 820
nm, while emission was 500-550 nm. Tissue is examined as deep as 60 µm from surface.
For Glu, thapsigargin (Tg, Tocris, Ellisville, MO) and caffeine (Sigma, St. Louis,
MO) treatment, frame-scan with sampling rate at 125 ms/f was performed for 960 frames
(120 s in total). The drug (Glu, 10 mM, 30 s; Tg, 10 µM, 60 s; caffeine, 10 mM, 30 s)
was delivered with a perfusion needle from 30 s. Immediately, another 960 frames scan
of the same cells was performed, but the slice was treated with high K+ (NaCl 80 mM,
KCl 50 mM, CaCl2 1.8 mM, MgSO4 0.81 mM, NaH2PO4 1.0 mM, glucose 24.5 mM,
HEPES 10 mM, gentamicin 1 mg/L, pH 7.3 at 37°C) with the same set-up, and all the
scan parameters were repeated. Only those cells exhibiting a positive response to high K+
treatment were regarded as neurons (of the 289 cells studied, 196 cells were identified to
be neurons). In some cases, slice was perfused with drugs for at least 10 min before
scanning: ryanodine (50 µM, Tocris, Ellisville, MO), DIA (5 mM), or GSH (1 mM),
which were delivered through a micro-syringe pump.
For RyR sensitivity study, SCN neurons were scanned for 180 s with the sampling
40
rate at 1 s/frame. Slices were treated with Na+, Ca2+ free solution (0Na/0Ca-Tetracaine,
LiCl 84.0 mM, KCl 4.36 mM, MgSO4 0.81 mM, KH2PO4 1.0 mM, glucose 24.5 mM,
HEPES 10.0 mM, EGTA 10.0 mM, LiOH 30.0 mM, pH 7.3 at 37°C) from 30 s to 90 s.
After a 20-min recovery, another 180 s scan of the same slice is performed, and the slice
is treated with Na+, Ca2+ free solution plus tetracaine (1 mM, Sigma, St. Louis, MO,
0Na/0Ca+Tetracaine), and all the scan parameters are repeated. These solutions were
delivered through a perfusion needle.
The soma of SCN neurons were measured, averaged, and normalized to baseline
(f/f0); the baseline (f0) is determined from the first 30 s of the recording. The Ca2+i change
is represented as the relative fluorescence intensity change.
Patch-clamp recording
Whole-cell patch electrodes had pipette tip resistances of 4-6 MΏ, and were filled with a
solution containing: K-gluconate 140 mM, KCl 5.0 mM, CaCl2 0.07 mM, MgCl2 1.0 mM,
EGTA 0.1 mM, HEPES 10 mM, Na-ATP 4 mM, Na-GTP 0.4 mM, pH 7.3, and osmolality
290-300 mOsm/L. Recordings were performed using a Multiclamp 700B amplifier.
Signals were sampled at 10 kHz, low-pass filtered at 10 kHz using a Digidata 1320
digitizer, and stored on computer for subsequent analyses using pClamp software.
The SCN neurons were recorded under voltage-clamp mode, with the holding
potential at -40 mV. The current is recorded in response to puff treatment of NMDA (200
µM, Sigma, St. Louis, MO); single-pulse stimulation is delivered very 15 sec for 10
repeats. Immediately, another 10 trials of NMDA puff were applied to the same cell, but
with an inhibitor of NMDA-R, d-CPP (10 µM) in bath. The entire process is performed
41
with TTX (1 µM) in bath to avoid secondary effect via synaptic transmission.
Results
Real-time imaging was performed to examine the Ca2+ response to Glu treatment in SCN
neurons in early and late night. We found that Glu (10 mM, 30 s) applied at CT 14
induced a strong transient Ca2+ increase, with a time course of ~20 s (Fig. 3.1A, E, ∆f/f0 =
43.8 ± 3.9%, 19 cells in 13 slices). On the other hand, at CT 19 Glu treatment induced a
brief (1 – 3 s), low amplitude response in SCN neurons (Fig. 3.1C, E, ∆f/f0 = 14.3 ± 1.4%,
17 cells in 12 slices). The strong Ca2+ response in early night was abolished by the RyR
antagonist, ryanodine (50 µM, 10 min pretreatment; Fig. 3.1B, E, ∆f/f0 = 11.7 ± 2.0%, 18
neurons in 8 brain slices); but the weak response in late night was insensitive to
ryanodine (Fig. 3.1D, E, ∆f/f0 = 14.0 ± 2.0%, 16 neurons in 6 brain slices).
We next measured Ca2+ influx into SCN neurons by recording the Ca2+ current
through the NMDA-R with excitatory synaptic receptor activation. Voltage-clamp
recording was performed with SCN neurons, with Vm held at -40 mV. Local puff
application of NMDA (200 µM) was applied 10 times to a single cell every 15 s. The
NMDA current was determined from the averaged changes of current from the 10 trials;
no significant difference was detected between early and late night recordings (Fig. 3.3,
CT 14, 60.2 ± 20.1 pA vs. CT 19, 52.9 ± 27.8 pA; P > 0.05, Student’s T-Test; n = 10).
We further estimated the capacity of ER Ca2+i store by real-time Ca2+ imaging,
with the treatments of two drugs, thapsigargin (Tg) and caffeine. Tg, as a specific
inhibitor of sarco/endoplasmic reticulum Ca2+-ATPase (SERCA), could deplete the ER
Ca2+i store and lead to a transient increase in Ca2+i, which can be used to evaluate the
42
loading level of ER Ca2+i store. We found strong Ca2+ responses to Tg treatment (10 µM,
60 s) in both early and late night, which no significant difference in amplitude (Fig. 3.4AC; CT 14, ∆f/f0 = 32.3 ± 3.2%, 11 neurons in 6 brain slices vs CT 19, ∆f/f0 = 34.6 ±
10.2%, 10 neurons in 4 brain slices; P > 0.05, Student’s T-Test).
High concentration of caffeine (10 mM, 30 s) functions likewise, that depletes the
ER Ca2+i store and induce a transient increase in Ca2+i, by activating RyR. We applied this
treatment as a complementary strategy to evaluate the volume of ER Ca2+i store, and
similar results were obtained: caffeine induced strong transient Ca2+i increases in early
night of 36.0 ± 9.1% (Fig. 3.4D, F, 10 neurons in 6 slices), while in late night of 39.5 ±
12.5% (Fig. 3.4E, F, 10 neurons in 6 slices), with comparable amplitude (P > 0.05,
Student’s T-Test).
To evaluate the activity of RyR at CT 14 vs. CT 19, we measured the ER Ca2+
leak-load relationship in situ, an established strategy to examine the function of ER Ca2+i
store (Shannon et al., 2002). With Na+-, Ca2+-free solution (See Materials and Methods)
blocking the soldium-calcium exchanger (NCX) and Ca2+ leaking through all the Ca2+
channels on the plasma membrane, cytosolic Ca2+ homeostasis can be isolated from
extracellular environment, and depends solely on the balance between Ca2+i store leak
through RyR (and IP3R) and load through SERCA. The further addition of tetracaine (1
mM) blocks the Ca2+i store leak through RyR, then the cytosolic Ca2+ decline represents
the loading through SERCA. Thus, the difference of the Ca2+ shifts in the two steps of
treatment reflects the leaking rate of Ca2+i store, which highly depends on the RyR
activity. Based on this rationale, we can evaluate the RyR activity by measuring the
leaking rate of Ca2+i store through RyR in SCN neurons. As is shown in Fig. 3.5, the
43
differential slopes of the two curves during the treatment at CT 14 (Fig. 3.5A, B; P < 0.01,
paired Student’s T-Test; 7 cells x 3 SCN slices) represents a relatively higher RyR activity
than at CT 19 (Fig. 3.5C, D; P > 0.05, paired Student’s T-Test; 7 cells x 3 SCN slices).
To explore the possible relationship between redox state and Ca2+ signaling in
SCN neurons, we examined Ca2+ responses to Glu treatment in the presence of redox
reagents (DIA or GSH). We found that in early night, with reducing reagent in bath (GSH,
1 mM, 10 min), Glu treatment only induced a weak Ca2+i increase (Fig. 3.6E, G, ∆f/f0 =
21.4 ± 1.1%, 16 neurons in 10 brain slices; P < 0.01, Tukey HSD Test, compared to
normal EBSS); this response was not sensitive to RyR inhibition (data not present). On
the other hand, exposure to the oxidizing reagent (DIA, 5 mM, 10 min) in late night, led
to a strong Ca2+i increase in response to Glu, with the time course of 20 s (Fig. 3.6D, G,
∆f/f0 = 38.7 ± 3.5%, 13 neurons in 8 brain slices; P < 0.01, Tukey HSD Test, compared to
normal EBSS); this effect could be blocked by RyR antagonist (data not present).
Application of DIA in early night (Fig. 3.6C, G, ∆f/f0 = 41.5 ± 4.3%, 11 neurons in 7
brain slices), or GSH in late night (Fig. 3.6F, G, ∆f/f0 = 21.4 ± 1.0%, 19 neurons in 10
brain slices) did not alter the Ca2+ responses to Glu treatment significantly (P > 0.05,
Tukey HSD Test, compared to normal EBSS).
To test the hypothesis of RyR as a possible target of redox regulation, we
evaluated the ER Ca2+ leak-load relationship to measure RyR activity in early and late
night, with the same procedure as stated above, while exposing to redox reagents. In both
early and late night, the activity of RyR is found to be enhanced by the oxidizing reagent
(DIA; Fig. 3.7C, D; CT 14, P < 0.01, CT 19, P < 0.05, paired Student’s t-test; 7 cells x 3
SCN slices), but suppressed by the reducing reagent (GSH; Fig. 3.7E, F; P > 0.05, paired
44
Student’s t-test; 7 cells x 3 SCN slices). These findings support the hypothesis that the
Ca2+ signaling in light/Glu-induced phase-shift is regulated by redox state in early and
late night, with the target of RyR-dependent Ca2+i release.
Discussion
Light/Glu have long been known to induce phase-shifts in the SCN with opposite effects
in between early and late night. With real-time Ca2+ imaging and patch-clamp recording,
we here dissected Ca2+i signaling as a key operator of the signal cascade bifurcation,
which is modulated by redox state.
We found that RyR-dependent Ca2+i release differentiates early and late night
responses: Glu induces a strong Ca2+i increase, which can be attenuated by RyR
inhibition, at CT 14, but not CT 19 (Fig. 3.1 & 3.2). The differential Ca2+ response is not
due to changes from Ca2+ influx through NMDA-R, with the supporting evidence from
patch-clamp recording (Fig. 3.3). In addition, the differential response of Ca2+ is not due
to the alternation of ER Ca2+i store neither, because pharmacological deletion of the store
by Tg or caffeine could induce strong Ca2+i release in both early and late night (Fig. 3.4).
Our data suggest that the differential Ca2+i response to Glu treatment in early and late
night might result from the changes in RyR activity. At CT 14 when RyR is relatively
sensitive to excitatory synaptic input (Fig. 3.5A, B), the Glu-induced Ca2+ influx through
NMDA-R could activate RyR and trigger a strong Ca2+i release; conversely at CT 19,
with a relative inactive RyR (Fig. 3.5C, D), the same trigger cannot open the channel, so
as no Ca2+i released from the store. Differential pattern of the transient Ca2+i increase
could recruit and activate different downstream factors, so that the signal cascade
bifurcates and the physiological effects of light/Glu stimulation are completely different
45
between early and late night.
As a ubiquitous second messenger, Ca2+ decodes a variety of extracellular stimuli
in the context of certain cellular state, into extremely different intracellular reaction. This
capability depends upon the complex spatial-temporal distribution of Ca2+i, which further
selects downstream effectors to be recruited and activated.
We further found that the Ca2+ signaling in response to Glu stimulation is
modulated by redox state, which could control the directionality of Glu-induced phaseshift in SCN (Fig. 3.8). In early night with a relatively oxidized state, the RyRs in SCN
neurons are highly active (Fig. 3.7), so the Ca2+ influx elicited by Glu treatment is able to
activate RyR and induce a strong Ca2+i release (Fig. 3.6). In late night, the reducing
environment suppresses the activity of RyRs (Fig. 3.7), thus the Ca2+ current through
NMDA-R is not strong enough to trigger RyR-dependent Ca2+i release (Fig. 3.6). By
regulating Ca2+ signaling in SCN neurons, the redox state could influence the input
signals from the environment to the central pacemaker, so as to actively participate in the
organization of clockwork machinery.
Our study on the signal transduction of the light/Glu-induced phase-shift
demonstrates that the physiological importance of metabolic oscillator extends beyond its
modulation of membrane excitability. The metabolic state of the SCN set the tone of
signal transduction pathways, thereby determining the nature of the pathway activated
and how it engages the molecular clockwork. By driving the daily oscillation of neuronal
excitability and setting the tone of signal transduction elements, the metabolic oscillator
modulates coupling between the central pacemaker and output targets, and gates SCN
sensitivity and response to input signals (Gillette and Mitchell, 2002; Rutter et al., 2002).
46
The metabolic cycle is not only driven by the circadian clock and benefits from it, but it
is part of the clock!
47
Figures 3
A 1.6
GLU
f/f0
1.4
1.2
CT14 1.0
0.8
0.6
0
30
60
1.6
1.4
1.2
CT19 1.0
0.8
0.6
30
1.6
60
30
60
90 120
D 1.6
1.4
1.2
1.2
1.0
0.8
0.6
1.0
90 120
**
1.4
0
90 120
**
E
1.4
1.2
1.0
0.8
0.6
C
0
Bath in Ry
B 1.6
0
30
60
90 120
GLU
GLU-Ry
CT14
CT19
Time (s)
Fig. 3.1. Differential Ca2+ responses to Glu treatment of the SCN in subjective early
(CT 14) and late (CT 19) night. (A) Glu (10 mM, 30 s) induced a strong transient Ca2+i
increases in early night. (B) The strong Ca2+ response to Glu was abolished by the
RyR antagonist, ryanodine (50 µM, 10 min pretreatment) in early night. (C) Glu
application in late night induced a weak Ca2+ response. (D) The weak response in late
night was not sensitive to ryanodine. (E) Amplitudes of Ca2+ response to Glu
with/without ryanodine, in early and late night (**, P < 0.01, Student’s t-Test; n = 1619).
.
48
NCX
NMDA-R
trigger
amplifier
SERCA
RyR
VGCC
InsP3R
Fig. 3.2. Proposed model of Glu-induced, RyR-dependent, Ca2+i increases in early
night. NMDA-R, RyR, and capacity of Ca2+i store are essential modulators underling
this response.
49
60
C
40
20
0
IN 0
(p A )
CT14
-20
50 pA
-40
-60
IN 1
(m V )
-80
0
5
5.5
6
Time (s)
6.5
500 ms
100
Sweep:9 Visible:1 of 12
80
B
60
IN 0
(pA )
40
CT19
20
0
NMDA-evoked current (pA)
A
90
60
30
0
-20
IN 1
(m V )
-40
CT 14
0
3.5
4
4.5
Time (s)
CT 19
5
Sweep:12 Visible:1 of 12
Fig. 3.3. Ca2+ influx through NMDA-R into SCN neurons remains the same in early
and late night. (A, B) Puff application of NMDA (200 µM) on voltage-clamped (-40
mV) SCN neurons evokes inward currents in early (A) and late (B) night. (C)
Amplitude of current responses to NMDA treatment in early and late night (P > 0.05,
Student’s t-Test; n = 10).
.
50
A
Thapsigargin
2.0
1.5
1.5
1.0
1.0
0.5
0.5
f/f0
CT14
B
CT19
0
60
90 120
E 2.0
1.5
1.5
1.0
1.0
0.5
0.5
1.6
Amplitude
30
2.0
0
C
D 2.0
30
60
90 120
Time (s)
0
30
60
90 120
0
30
60
90 120
Time (s)
F 1.6
1.4
1.4
1.2
1.2
1.0
1.0
CT14
Caffeine
CT19
CT14
CT19
Fig. 3.4. Capacity of Ca2+i store in SCN neurons remains the same in early and late
night. (A, B) Thapsigargin treatment (Tg, 10 µM, 60 s) on SCN neurons induced
strong transient Ca2+i increases in early (A) and late (B) night. (C) Amplitude of Ca2+
responses to Tg in early and late night (P > 0.05, Student’s t-Test; n = 10-11). (D, E)
Caffeine treatment (10 mM, 30 s) on SCN neurons induced strong transient Ca2+i
increases in early (C) and late (D) night. (C) Amplitude of Ca2+ responses to caffeine
in early and late night (P > 0.05, Student’s t-Test; n = 10-11).
.
51
f/f0
A
2.0
C 2.0
CT 14
1.6
1.6
1.2
1.2
0.8
0.8
0.4
0.4
0 Na+/0 Ca2+ - Tetracaine
0 Na+/0 Ca2+ +Tetracaine
0.0
0
Slope (x10-3 s-1)
B
60
0
D
120 180
-
+
3
0
0
-3
-3
-
60
Time (s)
**
-
0 Na+/0 Ca2+ - Tetracaine
0 Na+/0 Ca2+ +Tetracaine
0.0
120 180
3
-6
CT 19
-6
+
+
-
+
Fig. 3.5. RyR is more active in early than late night. (A, C) Two-steps treatment
(black, 0Na/0Ca-Tetracaine; grey, 0Na/0Ca+Tetracaine; 30 – 90 s) was applied to
SCN neurons in early (A) and late (C) night; 0Na/0Ca-Tetracaine treatment did not
alter Ca2+i neither in early nor in late night, Na/0Ca+Tetracaine induced Ca2+i decline
in early night, but not in late night. (B, D) The slope of each curve (-, 0Na/0CaTetracaine; +, 0Na/0Ca+Tetracaine) in A and C from 40 to 90 s was calculated by
linear regression, and the average of 7 cells x 3 SCN slices in each case was
summarized in B and D, respectively (**, P < 0.01, paired Student’s t-Test).
.
52
CT14
A 1.6
CT19
B
GLU
1.4
CTRL
1.2
1.0
0.8
0.6
0
30
60
90 120
C
D
f/f0
1.6
1.4
DIA
1.2
1.0
0.8
0.6
0
30
60
F
1.6
GSH 1.4
1.2
1.0
0.8
0.6
30
60
90 120
0
30
60
90 120
0
30
60
90 120
0
30
60
90 120
1.6
1.4
1.2
1.0
0.8
0.6
90 120
E
0
1.6
1.4
1.2
1.0
0.8
0.6
1.6
1.4
1.2
1.0
0.8
0.6
Time (s)
CTRL
DIA
GSH
◆◆
G 1.6
**
**
**
**
f/f0
1.4
1.2
1.0
CT14
CT19
Figures 3.6
53
Fig. 3.6. Redox state modulates Ca2+ responses to Glu treatment in early vs. late night.
(A, B) Glu treatment on SCN neurons induced a strong transient Ca2+i increases in
early night (A), but not in late night (B). (C, D) Bath application of an oxidizing
reagent (DIA, 5 mM, 10 min pretreatment) led to strong Ca2+ responses to Glu
treatment in both early (C) and late (D) night. (E, F) Exposure to a reducing reagent
(GSH, 1 mM, 10 min pretreatment) attenuated Ca2+ responses to Glu treatment in both
early (E) and late (F) night. (G) Amplitudes of Ca2+ responses to Glu treatment in
different conditions and CTs in A-F (P < 0.01, One-Way ANOVA, for both CT 14 and
CT 19; **, P < 0.01, Tukey HSD Test; ♦♦, P < 0.05, Student’s t-Test; n = 11 - 19).
.
54
CTRL
**
A
3
C
0
DIA
**
GSH
E
3
3
0
0
-3
-3
CT14
B
Slope (x10-3 s-1)
-3
-6
-
-
-6
+
+
3
+
*
3
-6
+
-
D
0
-
F
--
++
--
++
3
0
0
-3
-3
CT19
-3
-6
-
-
+
+
-6
-
-
+
+
-6
Fig. 3.7. Redox state regulates RyR activity in SCN neurons. Experiments were
performed and data were plotted the same as Fig. 3.5, except that the slices were
bathed in EBSS (A, B), with the oxidizing reagent (DIA, C and D), or with the
reducing reagent (GSH, E and F), at CT 14 (A, C, E) and CT 19 (B, D, F). (A, B) RyR
is more active in early (A) than late (B) night (**, P < 0.01, paired Student’s t-Test; 7
cells x 3 SCN slices). (C, D) RyR exhibits higher activity in oxidizing environment in
both early (C) and late (D) night (**, P < 0.01, *, P < 0.05, paired Student’s t-Test; 7
cells x 3 SCN slices). (E, F) RyR activity is suppressed by exposure to a reducing
reagent in both early (E) and late (F) night (P > 0.05, paired Student’s t-Test; 7 cells x
3 SCN slices).
.
55
Light/Glu
Ca2+ influx
Early night
Late night
Oxidized state
Reduced state
Sensitive RyR
Insensitive RyR
Ca2+i Release
Ca2+i Release
Phase Delay
Phase Advance
Fig. 3.8. Proposed model of signal pathway bifurcation of light/Glu-induced phaseshift in SCN, with the underlying molecular mechanism of redox regulation on Ca2+i
signaling. In early night with a relatively oxidized state, the RyRs in SCN neurons are
highly active (Fig. 3.7), so the Ca2+ influx elicited by Glu treatment is able to activate
RyR and induce a strong Ca2+i release (Fig. 3.6); while in late night, the reducing
environment suppresses the activity of RyRs (Fig. 3.7), thus the Ca2+ current through
NMDA-R is not strong enough to trigger RyR-dependent Ca2+i release (Fig. 3.6).
Differential pattern of the transient Ca2+i increase could recruit and activate different
downstream factors, so that the signal pathway bifurcates and the physiological effects
of light/Glu stimulation are completely different between early and late night.
56
CHAPTER 4. METABOLIC CYCLE AS A CORE COMPONENT OF
CIRCADIAN CLOCK
Summary
We found parallel oscillations of redox state and neuronal excitability in SCN, and redox
reagent-induced switching of membrane polarization that establish a reciprocal
interaction between metabolic state and neuronal excitability (Fig. 4.1). Energetic
fluctuation in the central nervous system has been considered to be a consequence of
neuronal activity. However, our study suggests that the cellular metabolic state changes
could be the cause of, rather than result from, neuronal activity. Crosstalk between
energetic and neuronal states could serve as a negative feedback mechanism that protects
neurons from over-excitation or deficits in energy metabolism.
In the context of circadian physiology, the regulatory relationship between redox
state and membrane excitability provides a non-transcriptional pathway for the metabolic
oscillator to engage the clockwork machinery. By driving the daily oscillation of
membrane excitability, the metabolic oscillator modulates coupling between the central
pacemaker and output targets, and gates SCN sensitivity and response to input signals
(Gillette and Mitchell, 2002; Rutter et al., 2002). Our study on the signal transduction of
light/Glu-induced phase-shift identified the transient Ca2+i increase as the bifurcation
point, which is modulated by redox state in SCN through RyR (Fig. 3.8). These results
provide the first demonstration of the physiological importance of metabolic oscillator
engaged in the clockwork.
57
Perspective
The circadian rhythm is traditionally regarded as the functional consequence of a
transcriptional-translational feedback loop (TTL), consist of a group of genes, clock
genes (Lowrey and Takahashi, 2004). Along with the self-sustained genetic oscillator,
post-translational elements are driven to fluctuate under the intrinsic period generated by
the central pacemaker.
Small signaling molecules, such as cAMP (Prosser and Gillette, 1991; O'Neill et
al., 2008) and Ca2+ (Ikeda et al., 2003; Harrisingh et al., 2007), have been observed in
SCN neurons, where the mammalian circadian clock resides, and their circadian
oscillations are reported by different groups. Sophisticated analysis of the
interdependence between these post-translational oscillators and genetic oscillator
suggests the TTL is neither sufficient nor necessary to generate the circadian rhythm
(Harrisingh and Nitabach, 2008). Study on cAMP oscillation in SCN neurons provides
the first demonstration of the concept of small signaling molecules as a core component
of circadian clock (O'Neill et al., 2008): 1) pharmacological manipulation of cellular
cAMP concentration affects the daily-rhythmic expression of clock genes; 2) suppressing
the level of cAMP sustains the free-running oscillation of clock genes, and shifts the
phase of the SCN slice to an identical new time point; 3) the rate of cAMP synthesis
influences the length of the intrinsic period generated by the central pacemaker. Similar
characters have been described for Ca2+ in Drosophila system (Harrisingh et al., 2007)
and cADPR in plant (Dodd et al., 2007). These findings suggest that the interplay
between TTL and post-translational oscillators, rather than the TTL only, determines the
intrinsic rhythm.
58
The metabolic cycle is highly influenced by the circadian clock (Green et al.,
2008; Bass and Takahashi, 2010). A circadian rhythm of protein oxidization is observed
in anucleate cells (O'Neill and Reddy, 2011; O'Neill et al., 2011), which suggests that the
redox oscillation, the reflection of metabolic cycle, could be self-sustained. However, in
nucleated system, the circadian oscillation of redox state is dependent upon the TTL (Fig.
2.1; O'Neill and Reddy, 2011). Whether the metabolic cycle is an independent oscillator
from the TTL is still a paradox, and the answer might lie on the development of advanced
technique in recording redox state in live specimens, with genetic manipulation of clock
genes. The real-time redox imaging applied in our study has the potential to achieve this
goal (Fig. 2.1).
As a potential core component of central pacemaker, the redox oscillator has two
pathways to engage in the clockwork machinery (Fig. 4.1): transcriptional and nontranscriptional pathways. The transcriptional pathway was originally identified in vitro
(Rutter et al., 2001; Dioum et al., 2002), and is well-recognized in peripheral tissues, such
as liver (Green et al., 2008; Bass and Takahashi, 2010). The transcriptional regulation of
clock genes by redox state is likely to occur in SCN, the central circadian clock, but is not
yet explored, partially due to lack of evidence supporting the metabolic alteration in SCN,
until we reveal the circadian redox oscillation by real-time imaging and other
complementary strategies (Fig. 2.1-2.3). Based upon the literature report of redox
regulation of clock genes’ transcription (Rutter et al., 2001; Dioum et al., 2002), and our
finding of redox oscillation in SCN (Fig. 2.1-2.3), it is reasonable to speculate that the
redox state could modulate the amplitude, period, or phase of the genetic oscillation in
59
SCN pacemaker. This hypothesis could be examined by real-time imaging of clock gene
expression with luciferase reporter (Yoo et al., 2004).
The non-transcriptional regulation of redox state on neuronal excitability in SCN
neurons is a novel pathway to integrate the metabolic cycle into the core clockwork
machinery. We have provided the first demonstration of the importance of redox
regulation in circadian physiology, the light-induced phase-shift (Fig. 3). As we proposed,
the redox state gates the SCN sensitivity and response to input signal entraining the
central pacemaker, via changing excitability of neurons. This mechanism may not be
limited to light/Glu signaling, but may apply to other entraining signals as well. Thus, it
would be worthwhile to screen other input signals to SCN, for sensitivity to redox
regulation. Considering that the output signals from SCN is also influenced by neuronal
excitability, which is under controlled by redox state, it would be interesting to explore
the role of metabolic state in the coupling between central pacemaker in SCN and
peripheral clocks.
We have identified several candidate targets of redox regulation of neuronal
excitability, including K+ channels on plasma membrane (Fig. 2.6, 2.7) and RyR on ER
membrane (Fig. 3.7). The biophysical mechanism of redox regulation on these ion
channels is yet to be revealed. In addition, the interplay across different redox targets
might be more challenging to dissect. These studies are important because the redox
regulation of neuronal excitability is likely to be a universal regulation that occurs in
most excitable cells. Our study suggests that it might not only occur under conditions of
metabolic deficit, but play a fundamental role in normal physiological conditions.
60
Figure 4
Transcriptional/Translational
Oscillator
Redox
Oscillator
(A)
(B)
(C)
Membrane Excitability
Oscillation
Figures 4.1
61
Fig.
4.1.
Proposed
model
of
the
relative
interdependency
of
the
transcriptional/translational oscillator, redox oscillator, and membrane excitability
oscillation. (A) The circadian oscillation of redox state in SCN depends on
functionally intact transcriptional/translational machinery (Fig. 2.1; O'Neill and
Reddy, 2011), and redox state modulates clock-genes expression, reciprocally (solid
arrows) (Rutter et al., 2001; Dioum et al., 2002; Balsalobre et al., 1998; Preitner et al.,
2002). (B) Several ion channels, which underlie membrane excitability, are
rhythmically expressed under the control of clock genes (solid arrow) (Brown and
Piggins, 2007); conversely, membrane excitability in SCN neurons can gate signal
input to the transcription/translation oscillator, which in turn affects clock-gene
expression (dashed arrow) (Nitabach et al., 2002; Lundkvist et al., 2005; Gillette and
Mitchell, 2002; Golombek and Rosenstein, 2010). (C) Redox state can regulate
neuronal excitability in SCN neurons via K+ currents (solid arrow) (Fig. 2.6 and 2.7);
concomitantly, increased neuronal activity can increase blood flow, glucose uptake by
astrocytes, and energy availability, which can feedback to modulate neuronal
metabolic state (dashed arrow) (Rutter et al., 2002).
.
62
REFERENCES
Akhtar, R.A., Reddy, A.B., Maywood, E.S., Clayton, J.D., King, V.M., Smith, A.G.,
Gant, T.W., Hastings, M.H., and Kyriacou, C.P. (2002). Circadian cycling of the mouse
liver transcriptome, as revealed by cDNA microarray, is driven by the suprachiasmatic
nucleus. Curr. Biol. 12, 540-550.
Aon, M.A., Cortassa, S., and O'Rourke, B. (2008). Mitochondrial oscillations in
physiology and pathophysiology. Adv. Exp. Med. Biol. 641, 98-117.
Balsalobre, A., Damiola, F., and Schibler, U. (1998). A serum shock induces circadian
gene expression in mammalian tissue culture cells. Cell 93, 929-937.
Bass, J., and Takahashi, J.S. (2010). Circadian integration of metabolism and energetics.
Science 330, 1349-1354.
Belle, M.D., Diekman, C.O., Forger, D.B., and Piggins, H.D. (2009). Daily electrical
silencing in the mammalian circadian clock. Science 326, 281-284.
Berson, D.M., Dunn, F.A., and Takao, M. (2002). Phototransduction by retinal ganglion
cells that set the circadian clock. Science 295, 1070-1073.
Brown, T.M., and Piggins, H.D. (2007). Electrophysiology of the suprachiasmatic
circadian clock. Prog. Neurobiol. 82, 229-255.
Bunger, M.K., Wilsbacher, L.D., Moran, S.M., Clendenin, C., Radcliffe, L.A.,
Hogenesch, J.B., Simon, M.C., Takahashi, J.S., and Bradfield, C.A. (2000). Mop3 is an
essential component of the master circadian pacemaker in mammals. Cell 103, 10091017.
DeCoursey, P.J. (1960). Daily light sensitivity rhythm in a rodent. Science 131, 33-35.
De Jeu, M., Geurtsen, A., and Pennartz, C. (2002). A Ba(2+)-sensitive K(+) current
contributes to the resting membrane potential of neurons in rat suprachiasmatic nucleus. J.
Neurophysiol. 88, 869-878.
Ding, J.M., Buchanan, G.F., Tischkau, S.A., Chen, D., Kuriashkina, L., Faiman, L.E.,
Alster, J.M., McPherson, P.S., Campbell, K.P., and Gillette, M.U. (1998). A neuronal
ryanodine receptor mediates light-induced phase delays of the circadian clock. Nature
394, 381-384.
Ding, J.M., Chen, D., Weber, E.T., Faiman, L.E., Rea, M.A., and Gillette, M.U. (1994).
Resetting the biological clock: mediation of nocturnal circadian shifts by glutamate and
NO. Science 266, 1713-1717.
63
Ding, J.M., Faiman, L.E., Hurst, W.J., Kuriashkina, L.R., and Gillette, M.U. (1997).
Resetting the biological clock: mediation of nocturnal CREB phosphorylation via light,
glutamate, and nitric oxide. J. Neurosci. 17, 667-675.
Dioum, E.M., Rutter, J., Tuckerman, J.R., Gonzalez, G., Gilles-Gonzalez, M.A., and
McKnight, S.L. (2002). NPAS2: a gas-responsive transcription factor. Science 298, 23852387.
Dodd, A.N., Gardner, M.J., Hotta, C.T., Hubbard, K.E., Dalchau, N., Love, J., Assie,
J.M., Robertson, F.C., Jakobsen, M.K., Goncalves, J., Sanders, D., and Webb, A.A.
(2007). The Arabidopsis circadian clock incorporates a cADPR-based feedback loop.
Science 318, 1789-1792.
Droge, W. (2002). Free radicals in the physiological control of cell function. Physiol. Rev.
82, 47-95.
Ebling, F.J. (1996). The role of glutamate in the photic regulation of the suprachiasmatic
nucleus. Prog. Neurobiol. 50, 109-132.
Gau, D., Lemberger, T., von Gall, C., Kretz, O., Le Minh, N., Gass, P., Schmid, W.,
Schibler, U., Korf, H.W., and Schutz, G. (2002). Phosphorylation of CREB Ser142
regulates light-induced phase shifts of the circadian clock. Neuron 34, 245-253.
Gillette, M. U. (1991). SCN electrophysiology in vitro: Rhythmic activity and
endogenous clock properties, in: Suprachiasmatic Nucleus: The Mind's Clock. Klein,
D.C., Moore, R.Y. and Reppert, S.M. (eds.), Oxford University Press, New York 125-143
Gillette, M.U., and Mitchell, J.W. (2002). Signaling in the suprachiasmatic nucleus:
selectively responsive and integrative. Cell Tissue Res. 309, 99-107.
Golombek, D.A., and Ralph, M.R. (1995). Circadian responses to light: the calmodulin
connection. Neurosci. Lett. 192, 101-104.
Golombek, D.A., and Ralph, M.R. (1994). KN-62, an inhibitor of Ca2+/calmodulin
kinase II, attenuates circadian responses to light. Neuroreport 5, 1638-1640.
Golombek, D.A., and Rosenstein, R.E. (2010). Physiology of circadian entrainment.
Physiol. Rev. 90, 1063-1102.
Green, C.B., Takahashi, J.S., and Bass, J. (2008). The meter of metabolism. Cell 134,
728-742.
Guo, H., Brewer, J.M., Champhekar, A., Harris, R.B., and Bittman, E.L. (2005).
Differential control of peripheral circadian rhythms by suprachiasmatic-dependent neural
signals. Proc. Natl. Acad. Sci. U. S. A. 102, 3111-3116.
64
Harrisingh, M.C., Wu, Y., Lnenicka, G.A., and Nitabach, M.N. (2007). Intracellular
Ca2+ regulates free-running circadian clock oscillation in vivo. J. Neurosci. 27, 1248912499.
Harrisingh, M.C., and Nitabach, M.N. (2008). Circadian rhythms. Integrating circadian
timekeeping with cellular physiology. Science 320, 879-880.
Hattar, S., Liao, H.W., Takao, M., Berson, D.M., and Yau, K.W. (2002). Melanopsincontaining retinal ganglion cells: architecture, projections, and intrinsic photosensitivity.
Science 295, 1065-1070.
Huang, S., Heikal, A.A., and Webb, W.W. (2002). Two-photon fluorescence
spectroscopy and microscopy of NAD(P)H and flavoprotein. Biophys. J. 82, 2811-2825.
Ikeda, M., Sugiyama, T., Wallace, C.S., Gompf, H.S., Yoshioka, T., Miyawaki, A., and
Allen, C.N. (2003). Circadian dynamics of cytosolic and nuclear Ca2+ in single
suprachiasmatic nucleus neurons. Neuron 38, 253-263.
Itri, J.N., Michel, S., Vansteensel, M.J., Meijer, J.H., and Colwell, C.S. (2005). Fast
delayed rectifier potassium current is required for circadian neural activity. Nat. Neurosci.
8, 650-656.
Itri, J.N., Vosko, A.M., Schroeder, A., Dragich, J.M., Michel, S., and Colwell, C.S.
(2010). Circadian regulation of a-type potassium currents in the suprachiasmatic nucleus.
J. Neurophysiol. 103, 632-640.
Kim, W.S., Dahlgren, R.L., Moroz, L.L., and Sweedler, J.V. (2002). Ascorbic acid assays
of individual neurons and neuronal tissues using capillary electrophoresis with laserinduced fluorescence detection. Anal. Chem. 74, 5614-5620.
Ko, C.H., Yamada, Y.R., Welsh, D.K., Buhr, E.D., Liu, A.C., Zhang, E.E., Ralph, M.R.,
Kay, S.A., Forger, D.B., and Takahashi, J.S. (2010). Emergence of noise-induced
oscillations in the central circadian pacemaker. PLoS Biol. 8, e1000513.
Lapainis, T., and Sweedler, J.V. (2008). Contributions of capillary electrophoresis to
neuroscience. J. Chromatogr. A 1184, 144-158.
Lehman, M.N., Silver, R., Gladstone, W.R., Kahn, R.M., Gibson, M., and Bittman, E.L.
(1987). Circadian rhythmicity restored by neural transplant. Immunocytochemical
characterization of the graft and its integration with the host brain. J. Neurosci. 7, 16261638.
Lowrey, P.L., and Takahashi, J.S. (2004). Mammalian circadian biology: elucidating
genome-wide levels of temporal organization. Annu. Rev. Genomics Hum. Genet. 5,
407-441.
65
Lundkvist, G.B., Kwak, Y., Davis, E.K., Tei, H., and Block, G.D. (2005). A calcium flux
is required for circadian rhythm generation in mammalian pacemaker neurons. J.
Neurosci. 25, 7682-7686.
Marcheva, B., Ramsey, K.M., Buhr, E.D., Kobayashi, Y., Su, H., Ko, C.H., Ivanova, G.,
Omura, C., Mo, S., Vitaterna, M.H., Lopez,J.P., Philipson,L.H., Bradfield,C.A.,
Crosby,S.D., JeBailey,L., Wang,X., Takahashi,J.S., Bass,J.. (2010). Disruption of the
clock components CLOCK and BMAL1 leads to hypoinsulinaemia and diabetes. Nature
466, 627-631.
Moore, R.Y., Speh, J.C., and Leak, R.K. (2002). Suprachiasmatic nucleus organization.
Cell Tissue Res. 309, 89-98.
Nitabach, M.N., Blau, J., and Holmes, T.C. (2002). Electrical silencing of Drosophila
pacemaker neurons stops the free-running circadian clock. Cell 109, 485-495.
O'Neill, J.S., Maywood, E.S., Chesham, J.E., Takahashi, J.S., and Hastings, M.H. (2008).
cAMP-dependent signaling as a core component of the mammalian circadian pacemaker.
Science 320, 949-953.
O'Neill, J.S., and Reddy, A.B. (2011). Circadian clocks in human red blood cells. Nature
469, 498-503.
O'Neill, J.S., van Ooijen, G., Dixon, L.E., Troein, C., Corellou, F., Bouget, F.Y., Reddy,
A.B., and Millar, A.J. (2011). Circadian rhythms persist without transcription in a
eukaryote. Nature 469, 554-558.
O'Rourke, B., Ramza, B.M., and Marban, E. (1994). Oscillations of membrane current
and excitability driven by metabolic oscillations in heart cells. Science 265, 962-966.
Pan, Y., Weng, J., Levin, E.J., and Zhou, M. (2011). Oxidation of NADPH on Kvbeta1
inhibits ball-and-chain type inactivation by restraining the chain. Proc. Natl. Acad. Sci. U.
S. A. 108, 5885-5890.
Preitner, N., Damiola, F., Lopez-Molina, L., Zakany, J., Duboule, D., Albrecht, U., and
Schibler, U. (2002). The orphan nuclear receptor REV-ERBalpha controls circadian
transcription within the positive limb of the mammalian circadian oscillator. Cell 110,
251-260.
Prosser, R.A., and Gillette, M.U. (1991). Cyclic changes in cAMP concentration and
phosphodiesterase activity in a mammalian circadian clock studied in vitro. Brain Res.
568, 185-192.
Prosser, R.A., and Gillette, M.U. (1989). The mammalian circadian clock in the
suprachiasmatic nuclei is reset in vitro by cAMP. J. Neurosci. 9, 1073-1081.
66
Prosser, R.A., McArthur, A.J., and Gillette, M.U. (1989). cGMP induces phase shifts of a
mammalian circadian pacemaker at night, in antiphase to cAMP effects. Proc. Natl. Acad.
Sci. U. S. A. 86, 6812-6815.
Ralph, M.R., Foster, R.G., Davis, F.C., and Menaker, M. (1990). Transplanted
suprachiasmatic nucleus determines circadian period. Science 247, 975-978.
Rice, M.E. (2000). Ascorbate regulation and its neuroprotective role in the brain. Trends
Neurosci. 23, 209-216.
Rizzuto, R., and Pozzan, T. (2006). Microdomains of intracellular Ca2+: molecular
determinants and functional consequences. Physiol. Rev. 86, 369-408.
Robles, M.S., Boyault, C., Knutti, D., Padmanabhan, K., and Weitz, C.J. (2010).
Identification of RACK1 and protein kinase Calpha as integral components of the
mammalian circadian clock. Science 327, 463-466.
Rutter, J., Reick, M., and McKnight, S.L. (2002). Metabolism and the control of
circadian rhythms. Annu. Rev. Biochem. 71, 307-331.
Rutter, J., Reick, M., Wu, L.C., and McKnight, S.L. (2001). Regulation of clock and
NPAS2 DNA binding by the redox state of NAD cofactors. Science 293, 510-514.
Satoh, T., and Lipton, S.A. (2007). Redox regulation of neuronal survival mediated by
electrophilic compounds. Trends Neurosci. 30, 37-45.
Schwartz, W.J., Gross, R.A., and Morton, M.T. (1987). The suprachiasmatic nuclei
contain a tetrodotoxin-resistant circadian pacemaker. Proc. Natl. Acad. Sci. U. S. A. 84,
1694-1698.
Shannon, T.R., Ginsburg, K.S., and Bers, D.M. (2002). Quantitative assessment of the SR
Ca2+ leak-load relationship. Circ. Res. 91, 594-600.
Shibata, S., and Moore, R.Y. (1994). Calmodulin inhibitors produce phase shifts of
circadian rhythms in vivo and in vitro. J. Biol. Rhythms 9, 27-41.
Sullivan, D.M., Levine, R.L., and Finkel, T. (2002). Detection and affinity purification of
oxidant-sensitive proteins using biotinylated glutathione ethyl ester. Methods Enzymol.
353, 101-113.
Tischkau, S.A., Mitchell, J.W., Pace, L.A., Barnes, J.W., Barnes, J.A., and Gillette, M.U.
(2004). Protein kinase G type II is required for night-to-day progression of the
mammalian circadian clock. Neuron 43, 539-549.
67
Tischkau, S.A., Weber, E.T., Abbott, S.M., Mitchell, J.W., and Gillette, M.U. (2003).
Circadian clock-controlled regulation of cGMP-protein kinase G in the nocturnal domain.
J. Neurosci. 23, 7543-7550.
Turek, F.W., Joshu, C., Kohsaka, A., Lin, E., Ivanova, G., McDearmon, E., Laposky, A.,
Losee-Olson, S., Easton, A., Jensen, D.R., et al. (2005). Obesity and metabolic syndrome
in circadian Clock mutant mice. Science 308, 1043-1045.
Verkhratsky, A. (2005). Physiology and pathophysiology of the calcium store in the
endoplasmic reticulum of neurons. Physiol. Rev. 85, 201-279.
Watanabe, A., Hamada, T., Shibata, S., and Watanabe, S. (1994). Effects of nitric oxide
synthase inhibitors on N-methyl-D-aspartate-induced phase delay of circadian rhythm of
neuronal activity in the rat suprachiasmatic nucleus in vitro. Brain Res. 646, 161-164.
Watanabe, A., Ono, M., Shibata, S., and Watanabe, S. (1995). Effect of a nitric oxide
synthase inhibitor, N-nitro-L-arginine methylester, on light-induced phase delay of
circadian rhythm of wheel-running activity in golden hamsters. Neurosci. Lett. 192, 2528.
Weber, E.T., Gannon, R.L., Michel, A.M., Gillette, M.U., and Rea, M.A. (1995). Nitric
oxide synthase inhibitor blocks light-induced phase shifts of the circadian activity rhythm,
but not c-fos expression in the suprachiasmatic nucleus of the Syrian hamster. Brain Res.
692, 137-142.
Yoo, S.H., Yamazaki, S., Lowrey, P.L., Shimomura, K., Ko, C.H., Buhr, E.D., Siepka,
S.M., Hong, H.K., Oh, W.J., Yoo, O.J., Menaker, M., and Takahashi, J.S. (2004).
PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent
circadian oscillations in mouse peripheral tissues. Proc. Natl. Acad. Sci. U. S. A. 101,
5339-5346.
Yu, Y. (2007). Redox state as a circadian state variable: regulation of directionality of
glutamate-induced phase shifts by determining activatability of key signaling proteins in
the suprachiasmatic nucleus of rat. University of Illinois at Urbana-Champaign. Ph.D
thesis.
68