Reduced G protein functions and immunoreactive levels in mononuclear leukocytes of patients with depression.

Abstract

Heterotrimeric G proteins play a pivotal role in postreceptor information transduction. These proteins were previously implicated in the biochemical mechanism underlying lithium action and in the pathophysiology of mood disorders. The present study sought to quantitatively and functionally evaluate G proteins in patients with major depression.
G proteins were measured in mononuclear leukocytes of 37 untreated patients with major depression and 31 comparison subjects. Receptor-coupled G protein function was evaluated through beta-adrenergic and muscarinic-agonist-induced increases in guanine nucleotide binding capacity, which were substantiated by quantitative measures of G proteins through immunoblot analyses that used polyclonal antibodies against stimulatory (Gs alpha) and inhibitory (Gi alpha) G proteins.
Mononuclear leukocytes of depressed patients showed significantly reduced immunoreactive quantities of Gs alpha and Gi alpha together with markedly hypofunctional Gs and Gi. The reductions in both function and quantity of Gs and Gi were significantly correlated with the severity of depressive symptoms. Moreover, simultaneous quantitative and functional measurements in a large number of patients showed significant correlations between the function and the quantity of mononuclear leukocyte Gs and Gi proteins:
These findings lend further support to the implication of G proteins in the pathophysiology of mood disorders. G protein functional and quantitative measurements in mononuclear leukocytes of patients with mood disorders may potentially serve as a biochemical marker for the affective state of these patients.

Full text

Page 1

AVISSAR, NECHAMKIN, ROITMAN, ET AL.G PROTEINS IN DEPRESSIONAm J Psychiatry 154:2, February 1997
Reduced G Protein Functions and Immunoreactive Levels
in Mononuclear Leukocytes of Patients With Depression
Sofia Avissar, Ph.D., Yakov Nechamkin, M.D.,
Gregori Roitman, M.D., and Gabriel Schreiber, M.D., Ph.D.
Objective: Heterotrimeric G proteins play a pivotal role in postreceptor information
transduction. These proteins were previously implicated in the biochemical mechanism under-
lying lithium action and in the pathophysiology of mood disorders. The present study sought
to quantitatively and functionally evaluate G proteins in patients with major depression.
Method: G proteins were measured in mononuclear leukocytes of 37 untreated patients with
major depression and 31 comparison subjects. Receptor-coupled G protein function was
evaluated through β-adrenergic and muscarinic-agonist-induced increases in guanine nucleo-
tide binding capacity, which were substantiated by quantitative measures of G proteins
through immunoblot analyses that used polyclonal antibodies against stimulatory (Gsα) and
inhibitory (Giα) G proteins. Results: Mononuclear leukocytes of depressed patients showed
significantly reduced immunoreactive quantities of Gsα and Giα together with markedly hy-
pofunctional Gs and Gi. The reductions in both function and quantity of Gs and Gi were
significantly correlated with the severity of depressive symptoms. Moreover, simultaneous
quantitative and functional measurements in a large number of patients showed significant
correlations between the function and the quantity of mononuclear leukocyte Gs and Gi pro-
teins. Conclusions: These findings lend further support to the implication of G proteins in the
pathophysiology of mood disorders. G protein functional and quantitative measurements in
mononuclear leukocytes of patients with mood disorders may potentially serve as a biochemi-
cal marker for the affective state of these patients.
(Am J Psychiatry 1997; 154:211–217)
I
bases of treatments used to combat these disorders has
focused on information transduction mechanisms be-
yond receptors. Heterotrimeric G proteins play a piv-
otal role in postreceptor information transduction (for
review see references 1 and 2). An important character-
istic of G proteins is their greater guanine nucleotide
binding following agonist stimulation, which in turn
leads to their activation. Tritiated guanine nucleotide
binding following agonist stimulation is an established
test for G protein function. This test has been used in
n recent years, research on the pathophysiological
mechanisms of mood disorders and the biochemical
membrane preparations (3, 4) and for the purification
of G proteins from crude membrane preparations (5)
and has been both kinetically and mechanistically ana-
lyzed by using various purified G proteins (6) in recon-
stituted systems (7, 8). More recently, this method was
used by Evans et al. (9) to show specific GTP binding
and hydrolysis activities in lymphocyte membranes fol-
lowing interleukin-2 stimulation.
We have adapted this method to measure early events
in signal transduction beyond receptors through acti-
vated receptor-coupled guanine nucleotide exchange on
G proteins (10, 11). Through this method the function
of the family of receptor-coupled G proteins was found
to be differentially attenuated by lithium (10–15), other
antibipolar treatments (14–16), and antidepressant
drugs (15, 16). Studies by other groups that used a va-
riety of techniques and were generally in agreement
with these results implicate the involvement of G pro-
teins in the mechanism of lithium action (17–23).
The discovery of the involvement of G proteins in the
mechanism of action of antibipolar and antidepressant
treatments suggests alterations in members of the G pro-
tein family in patients with mental disorders (24). Indeed,
lithium-sensitive hyperfunctional stimulatory (Gs) and in-
Received Jan. 19, 1996; revision received July 10, 1996; accepted
August 8, 1996. From the Departments of Clinical Pharmacology and
Psychiatry, Faculty of Health Sciences, Ben-Gurion University of the
Negev. Address reprint requests to Dr. Avissar, Department of Clini-
cal Pharmacology, Faculty of Health Sciences, Ben-Gurion University
of the Negev, P.O. Box 653, 84105 Beer-Sheva, Israel.
Supported by a Young Investigator Award from the National Alli-
ance for Research on Schizophrenia and Depression and by a fellow-
ship of the National Institute of Psychobiology in Israel in honor of
Prof. Joel Elkes to Dr. Avissar and research grants from the Stanley
Foundation USA (Drs. Avissar and Schreiber), the Chief Scientist Of-
fice of the Israel Ministry of Health (Drs. Avissar and Schreiber), and
the Zeldovich Foundation (Dr. Schreiber).
Am J Psychiatry 154:2, February 1997211

Page 2

hibitory (Gi) G proteins were detected in mononuclear
leukocytes of patients with mania (15, 25), while in a
preliminary study (26) decreases in the function and level
of Gs protein have been found in mononuclear leukocytes
of patients with major depressive disorder. In accordance
with our findings, Friedman et al. (27, 28) have shown
that lithium inhibits the association of various receptors
with membrane G proteins, and binding of labeled
GTPγS enhanced by agonists is elevated in postmortem
cerebral cortical membranes of patients with bipolar dis-
order. Although Young et al. (29) reported no changes in
the immunoreactivities of Gsα and Giα proteins in
mononuclear leukocytes of patients with unipolar de-
pression, they did report higher immunoreactive levels
and function of Gsα in postmortem cerebral cortices of
patients with bipolar disorder (30, 31). G protein quan-
titative and functional measures in human mononuclear
leukocytes were found to be age-independent (32).
In the present study we simultaneously examined, in a
larger group of patients, functional and quantitative im-
munoreactive measures of Gs and Gi proteins to test
whether, in contrast with the higher values previously re-
ported in manic patients, mononuclear leukocytes of de-
pressed patients show hypofunctional and reduced im-
munoreactive quantities of Gsα and Giα. We also sought
to determine whether the reductions in function and in
quantity of both Gs and Gi were significantly correlated
with the severity of depressive symptoms and whether
these simultaneous quantitative and functional measure-
ments showed a significant correlation between the func-
tion and the quantity of mononuclear leukocyte Gs and
Gi proteins.
METHOD
Patients
Thirty-seven patients with major depression
(26 with unipolar and 11 with bipolar depres-
sion), diagnosed according to DSM-III-R criteria
by at least two senior psychiatrists, were included
in the study. A physical examination, an ECG,
and laboratory tests for renal, hepatic, hema-
tologic, and thyroid function were administered
to all subjects, and no physical abnormalities were
reported. After complete description of the study
to the subjects, written informed consent was ob-
tained for a 60-ml blood donation. In all cases
blood was drawn between 8:00 a.m. and 10:00
a.m. The patients were given the Beck Depression
Inventory (33) before blood donation. The study
was approved by the institutional review board.
The group of 37 depressed patients consisted of
19 female and 18 male subjects whose average age
was 45.1 years (SD=15.4, range=19–78). Seven-
teen were outpatients, and 20 were hospitalized.
Before the initiation of treatment, an examination
of the patient histories confirmed that all subjects
had not received treatment for at least 1 month.
A healthy volunteer comparison group was re-
cruited from the staff of the Beer-Sheva Mental
Health Center and their families. This group con-
sisted of 31 subjects, 16 men and 15 women,
whose average age was 41.4 years (SD=15.7,
range=20–77).
Mononuclear Leukocyte Isolation
Mononuclear leukocytes were isolated from 60 ml of heparin-
ized fresh blood of adult donors by using Ficoll-Paque gradient, as
reported by Boyum (34). Cells were homogenized in 25-mM Tris
HCl and 1 mM dithiotreitol, pH=7.4. The homogenate was passed
through two layers of cheesecloth to remove debris, and mem-
branes were collected by further centrifugation at 18,000 g for 10
minutes. Membranes were then either freshly used for the func-
tional binding measures or suspended in a homogenization buffer
that contained 1 mM EGTA and 30% sucrose weight per volume,
which was frozen at –70°C until assayed by the quantitative meas-
ures. Aliquots were taken for protein concentration determination
by using Bradford’s assay.
Guanine Nucleotide Binding Assay
Binding reactions were carried out for 15 minutes at room tem-
perature in a final volume of 200 µl. The reaction buffer consisted of
25 mM Tris HCl, 1 mM ATP, 1 mM Mg2+, 1 mM EGTA, and 1 mM
dithiotreitol, pH=7.4, with varying concentrations of (3H)GppNHp
(0.05 µM–5 µM). Reactions were started by adding 50 µg of mem-
brane protein and terminated with 5 ml of ice-cold buffer (10 mM
Tris HCl and 100 mM NaCl; pH=7.4), followed by rapid filtration.
Filters were subsequently washed twice with 3 ml of cold buffer and
taken for scintillation counting. Agonist effects on (3H)GppNHp
binding were assessed by adding isoproterenol (25 µM) or carbamyl-
choline (50 µM) to the reaction mixture. These represent the minimal
concentrations that result in maximal effect of the agonists.
Immunoblot Analysis
On the day of assay, membranes were thawed, and aliquots of 10
µg membrane G proteins were separated by sodium dodecyl sulfate
polyacrylamide gel electrophoresis (10% acrylamide). The resulting
proteins were transferred to nitrocellulose paper by use of electro-
blotting apparatus. Blots were washed in Tris-buffered saline that
FIGURE 1. Gs and Gi Protein Function in Mononuclear Leukocyte Membranes of Normal
Subjects and Patients With Depressiona
aHorizontal lines represent mean values.
bSignificant difference between depressed patients and normal subjects (Mann-Whitney
U test: t=6.99, df=66, p<0.001).
cSignificant difference between depressed patients and normal subjects (Mann-Whitney
U test: t=6.58, df=60, p<0.001).
G PROTEINS IN DEPRESSION
212Am J Psychiatry 154:2, February 1997

Page 3

contained 3% Tween-20 and blocked by incubation with 5% bovine
serum albumin for 1 hour in Tris-buffered saline that contained 0.1%
Tween-20. After two washes in Tris-buffered saline that contained
0.1% Tween-20, blots were incubated overnight in antisera (NEN-
DuPont, Mississauga, Ont., Canada) that was directed specifically
against Gsα (dilution 1:2,500), Gi1,2α (dilution 1:5,000), and Gβ (di-
lution 1:1,000), followed by subsequent incubation with goat anti-
rabbit IgG labeled with horseradish peroxidase. Immunoreactivity
was detected with the Enhanced Chemiluminescence Western Blot
Detection System (Amersham, Oakville, Ont. Canada) followed by
exposure to X -ray film.
A preliminary assay that used healthy volunteer mononuclear
leukocyte membranes was carried out for the determination of the
range of linearity of the assay with respect to protein concentra-
tion. Linearity was found between 2.5 µg–15 µg membrane pro-
tein. Peak heights of immunoreactive bands were determined with
an image analysis system. The optical density of the immunoreac-
tive bands was normalized against 10-µg rat cortical membranes
run in each blot as a standard value. Although anti-Gsα detects
both 52- and 45-kDa Gsα species, only the 45-kDa species is de-
tected in leukocytes, while rat cortex membranes show predomi-
nantly the 52-kDa species. The other subunits assayed in leuko-
cytes were found to migrate at the expected molecular weights
similarly to those labeled in the rat cortex membranes.
Statistical Analysis
Correlations between functional and quantitative measures of G
proteins in human mononuclear leukocytes as well as correlations
between these G protein measures in mononuclear leukocytes of de-
pressed patients and score on the Beck Depression Inventory were
conducted by using the Spearman rank correlation. Comparisons of
each agonist-induced increase in GppNHp binding capacities, as well
as of the immunoreactivities of the various G protein subunits, be-
tween the depressed patients and the healthy comparison subjects
were conducted by using the nonparametric Mann-Whitney rank-
sum test (35).
RESULTS
G protein function was measured through agonist-in-
duced increases in GppNHp binding capacity: isopro-
terenol and carbamylcholine were used as β-adrenergic
and muscarinic receptor agonists, respectively. Both
agonists induced significant increases in GppNHp
binding capacity without substantially affecting the af-
finity of binding. Isoproterenol-induced increases in
GppNHp binding capacity were blockable by pretreat-
ment with cholera toxin, with no effect of pertussis
toxin, which indicates a specific effect through Gs pro-
tein. Carbamylcholine exerted its effects on guanine nu-
cleotide binding in a specific pertussis toxin-sensitive
manner, which suggests coupling with non-Gs proteins
(e.g., Gi) (15, 16, 25, 26).
Figure 1 shows the functional measures of Gs and Gi
following agonist stimulation for the normal and de-
pressed subjects. The group of depressed patients showed
statistically significant reductions in β-adrenergic-recep-
tor-coupled Gs protein function and in muscarinic-recep-
tor-coupled Gi protein function. These reductions for the
depressed patients in Gs and Gi protein function were
similar for unipolar and bipolar depression.
We compared depressive symptom severity as as-
sessed by the Beck Depression Inventory with receptor-
coupled G protein function measurements and found
that depression of even mild to moderate severity could
be detected in our assay system as partially inhibiting
isoproterenol- and carbamylcholine-induced increases
in GppNHp binding capacity. Patients with more se-
vere depression showed complete inhibition of recep-
tor-coupled Gs protein function. Significant negative
correlations were found between the score on the Beck
Depression Inventory and both β-adrenergic-receptor-
coupled Gs function and muscarinic-receptor-coupled
Gi protein function (figure 2).
G protein levels in human mononuclear leukocytes
were assessed through immunoblot analysis by using
polyclonal antibodies against various Gα and Gβ
subunits. Figure 3 is a representative example of im-
munoblots obtained for the quantitation of Gsα, Giα,
and Gβ subunit proteins from mononuclear leukocytes
of patients with major depression and age- and sex-
matched comparison subjects. Semiquantitative nor-
FIGURE 2. Correlation Between Gs and Gi Protein Function and Score
on the Beck Depression Inventory
aSignificant negative correlation (rs=–0.78, N=44, p<0.001).
bSignificant negative correlation (rs=–0.75, N=39, p<0.001).
AVISSAR, NECHAMKIN, ROITMAN, ET AL.
Am J Psychiatry 154:2, February 1997213

Page 4

malization of the data concerning both Gα levels ob-
tained from all subjects indicated that the mononuclear
leukocyte levels of Gsα and of Giα in depressed patients
were significantly less than that of the normal compari-
son subjects (figure 4). Moreover, as was found for Gs
and Gi protein function, mononuclear leukocyte Gsα
and Giα protein contents were negatively correlated with
depression severity as assessed by the Beck Depression
Inventory (figure 5). Immunoreactive Gβ protein levels
in depressed patients were similar to those of the com-
parison subjects (data not shown). Simultaneous G pro-
tein quantitative and functional measurements con-
ducted on the same mononuclear leukocyte membrane
preparations in patients and comparison subjects re-
sulted in significant positive correlations between β-ad-
renergic-receptor-coupled Gs protein function and Gsα
immunoreactivity as well as between muscarinic-recep-
tor-coupled Gi protein function and Giα immunoreac-
tivity (figure 6).
DISCUSSION
A role for monoaminergic and muscarinic cholinergic
mechanisms in the pathogenesis of mood
disorders has been largely suggested by
and formulated in both the catechola-
mine hypothesis and the monoaminer-
gic-cholinergic balance hypothesis (for a
review see references 36 and 37). Bio-
chemical research in studies of affective
disorders has focused on the cascade of
events involved in signal transduction:
the primary messenger level (the cate-
cholamine or acetylcholine neurotrans-
mitter), the level of the neurotransmitter
receptors, and, lately, information trans-
duction mechanisms beyond receptors.
The present study concentrated on al-
terations in early events in signal trans-
duction beyond receptors that involve G
proteins.
Signal transductions by G protein-cou-
pled receptors are regulated by various
mechanisms that act at the receptor level,
e.g., β-adrenergic receptor kinase (38) or,
at the G protein level, phosducin (39). It
is possible that the reductions observed in receptor-cou-
pled G protein function in the present study stem from
alterations at the receptor level, at G protein level, or in
regulatory factors that act at either levels. Since our
functional G protein assay system measures activated
receptor-enhanced guanine nucleotide binding and thus
receptor coupling to a G protein, the reduced function
measures may reflect reduced receptor levels, reduced
coupling, or reduced G protein levels. Indeed, reduced
levels of the β-adrenergic receptor were previously ex-
tensively described in mononuclear leukocytes of pa-
tients with unipolar and bipolar depression (40–44).
However, our functional measures, together with the
complementary immunoblot studies, suggest also di-
rect, quantitative, and functional reductions in Gs pro-
tein in mononuclear leukocytes of depressed patients at
the postreceptor level that are in accordance with func-
tional findings of reduced β-adrenergic-receptor-coupled
adenylate cyclase activity in patients with depression
(40, 41, 45–47). Most evidence suggests that musca-
rinic receptor characteristics are not altered in patients
with affective disorder (37). This information, together
with our findings of reductions in both muscarinic-re-
ceptor-coupled Gi function and Giα immunoreactivity,
supports again a direct postreceptor, quantitative, and
functional reduction in Gi protein in mononuclear leu-
kocytes of patients with depression.
Higher Gsα and Giα levels in mononuclear leukocytes
of patients with bipolar depression and normal-like lev-
els of these proteins in patients with unipolar depres-
sion were reported by Young et al. (29). These results
contrast with the present findings of reduced immuno-
reactive and functional measures of β-adrenergic- and
muscarinic-receptor-coupled Gs and Gi proteins, which
correlated with depressive symptom severity as assessed
by the Beck Depression Inventory. This discrepancy
may stem from differences in the severity of illness in
FIGURE 4. Gsα and Giα Relative Immunoreactivities in Mononuclear Leukocytes of
Normal Subjects and Depressed Patients
a Significant difference between depressed patients and normal subjects (Mann-Whitney
U test: t=5.64, df=65, p<0.001).
bSignificant difference between depressed patients and normal subjects (Mann-Whitney
U test: t=5.67, df=54, p<0.001).
FIGURE 3. Representative Immunoblots Obtained With Antisera
Against Gsα, Giα, and Gβ Subunits in Mononuclear Leukocytes of
Normal Subjects and Patients With Depression
G PROTEINS IN DEPRESSION
214Am J Psychiatry 154:2, February 1997

Page 5

the patients included in each study: almost all the de-
pressed patients in Young et al.’s study were outpa-
tients, with probably less severe depressive symptoms,
while in the present study depressed inpatients formed
more than 50% of the examined population. We have
previously described hyperfunctional receptor-coupled
Gs and Gi proteins in mononuclear leukocytes of pa-
tients with mania (15, 25). In the present study patients
with bipolar depression (as well as unipolar depression)
showed an inverse picture: hypofunctional Gs and Gi
proteins. The finding of greater and reduced function of
G proteins in mania and depression, respectively, sug-
gests that G protein activity may be a biochemical
marker indicative of the affective state rather than a
trait marker of bipolar disorder. However, one should
consider the possibility that both trait and state altera-
tions may exist in G protein levels and that these may
result in opposite changes and thus reveal different pat-
terns in different patient populations.
We are aware that the involvement of G proteins in
the pathophysiology of depression, as implicated from the
data presented here, should be taken with considerable
caution. Extrapolations from findings obtained in pe-
ripheral blood cells, which are exposed to circulatory
catecholamines and hormones, to the CNS is not straight-
forward. In this regard, it should be mentioned that ele-
vated immunoreactivities of Gs and Gi proteins were de-
scribed in postmortem brains of patients with bipolar
disorder (30). The mood state of these patients at the
time of death was not described. One possibility is that
these patients were manic at the time of death. In this case
these results confirm our findings of hyperfunctional Gs
and Gi proteins in peripheral white blood cells of pa-
tients with mania (25). However, it is far from clear that
this is the correct explanation, and an alternative expla-
nation might be that there are differences between the
brain and peripheral cells. This would reduce the sig-
nificance of the presently described findings of reduced
function and levels of G proteins in patients with bipo-
lar depression in terms of understanding the biochemi-
cal mechanisms that underlie bipolar depression.
Differences may exist in the tissue distribution of G
proteins between CNS and peripheral tissues. For ex-
ample, differences in the distribution of Gsα variants
between CNS and peripheral tissues have been de-
scribed (48). This heterogeneity might be the explana-
FIGURE 6. Correlation Between G Protein Functional and Quantita-
tive Measures in Human Mononuclear Leukocyte Membranes
aSignificant positive correlation (rs=0.83, N=39, p<0.001).
bSignificant positive correlation (rs=0.75, N=31, p<0.001).
FIGURE 5. Correlation Between Gsα and Giα Immunoreactivity and
Score on the Beck Depression Inventory
aSignificant negative correlation (rs=–0.63, N=31, p<0.001).
bSignificant negative correlation (rs=–0.69, N=31, p<0.001).
AVISSAR, NECHAMKIN, ROITMAN, ET AL.
Am J Psychiatry 154:2, February 1997 215