Glycine: A new anti-inflammatory immunonutrient

Abstract

The mechanism of the immunosuppressive effects of glycine and its pathophysiological applications are discussed in this review. Glycine has been well characterized in spinal cord as an inhibitory neurotransmitter which activates a glycine-gated chloride channel (GlyR) expressed in postsynaptic membranes. Activation of the channel allows the influx of chloride, preventing depolarization of the plasma membrane and the potentiation of excitatory signals along the axon. Glycine has recently been shown to have similar inhibitory effects on several white blood cells, including hepatic and alveolar macrophages, neutrophils, and lymphocytes. Pharmacological analysis using a GlyR antagonist strychnine, chloride-free buffer, and radiolabeled chloride has provided convincing evidence to support the hypothesis that many white blood cells contain a glycine-gated chloride channel with properties similar to the spinal cord GlyR. Molecular analysis using reverse transcription-polymerase chain reaction and Western blotting has identified the mRNA and protein for the beta subunit of the GlyR in total RNA and purified membrane protein from rat Kupffer cells. Dietary glycine is protective in rat models against endotoxemia, liver ischemia-reperfusion, and liver transplantation, most likely by inactivating the Kupffer cell via this newly identified glycine-gated chloride channel. Glycine also prevents the growth of B 16 melanomas cell in vivo. Moreover, dietary glycine is protective in the kidney against cyclosporin A toxicity and ischemia-reperfusion injury. Glycine may be useful clinically for the treatment of sepsis, adult respiratory distress syndrome, arthritis, and other diseases with an inflammatory component.

Full text

CMLS, Cell. Mol. Life Sci. 56 (1999) 843–856
1420-682X/99/100843-14 $ 1.50+0.20/0
© Birkha¨user Verlag, Basel, 1999
Review
Glycine: a new anti-inflammatory immunonutrient
M. D. Wheeler, K. Ikejema, N. Enomoto, R. F. Stacklewitz, V. Seabra, Z. Zhong, M. Yin, P. Schemmer,
M. L. Rose, I. Rusyn, B. Bradford and R. G. Thurman*
Department of Pharmacology, University of North Carolina at Chapel Hill, Chapel Hill (North Carolina,
USA)
Received 3 June 1999; received after revision 16 August 1999; accepted 17 August 1999
a glycine-gated chloride channel with properties similarAbstract. The mechanism of the immunosuppressive ef-
fects of glycine and its pathophysiological applications to the spinal cord GlyR. Molecular analysis using re-
verse transcription-polymerase chain reaction and West-are discussed in this review. Glycine has been well
ern blotting has identified the mRNA and protein forcharacterized in spinal cord as an inhibitory neurotrans-
the i subunit of the GlyR in total RNA and purifiedmitter which activates a glycine-gated chloride channel
(GlyR) expressed in postsynaptic membranes. Activa- membrane protein from rat Kupffer cells. Dietary
glycine is protective in rat models against endotoxemia,tion of the channel allows the influx of chloride, prevent-
liver ischemia-reperfusion, and liver transplantation,ing depolarization of the plasma membrane and the
most likely by inactivating the Kupffer cell via thispotentiation of excitatory signals along the axon.
Glycine has recently been shown to have similar in- newly identified glycine-gated chloride channel. Glycine
hibitory effects on several white blood cells, including also prevents the growth of B16 melanomas cell in vivo.
hepatic and alveolar macrophages, neutrophils, and Moreover, dietary glycine is protective in the kidney
lymphocytes. Pharmacological analysis using a GlyR against cyclosporin A toxicity and ischemia-reperfusion
antagonist strychnine, chloride-free buffer, and radiola- injury. Glycine may be useful clinically for the treatment
beled chloride has provided convincing evidence to sup- of sepsis, adult respiratory distress syndrome, arthritis,
and other diseases with an inflammatory component.port the hypothesis that many white blood cells contain
Key words. Glycine; immunoregulation; anti-inflammatory; glycine receptor.
General introduction
Glycine has long been known to be an inhibitory neuro-
transmitter in the spinal cord [1]. Glycine-mediated
inhibitory neurotransmission is essential for startle re-
sponses, voluntary motor control and sensory signal
processing in the spinal cord [2– 4]. Glycine exerts its
inhibitory actions by binding its receptor (GlyR) which
is largely localized in postsynaptic neuronal membranes
[5]. Inhibitory postsynaptic signals oppose the depolar-
izing action of excitatory/stimulatory neurotransmission
by increasing chloride permeability across the postsy-
naptic neuronal membrane. The identity of glycine as
an inhibitory neurotransmitter was originally proposed
by Aprison et al. [6] and Davidoff et al. [4], who
described in detail the distribution of glycine through-
out the central nervous system. Autoradiographic stud-
ies with radiolabeled glycine demonstrated that glycine
is localized in spinal cord synaptic regions [7]. Func-
tional studies later demonstrated that glycine hyperpo-
larizes postsynaptic motor neurons by increasing
* Corresponding author. Laboratory of Hepatobiology and Toxi-
cology, CB c 7365 Mary Ellen Jones Bldg., University of North
Carolina at Chapel Hill, Chapel Hill (North Carolina 27599-7365,
USA), Fax +919 966 1893, e-mail: thurman@med.unc.edu

M. D. Wheeler et al. Anti-inflammatory effects of glycine844
chloride conductance [3, 8, 9]; thus, the receptor for
glycine is often referred to as a glycine-gated chloride
channel. Inhibitory neurotransmission by glycine was
shown to be selectively blocked by strychnine, a plant
alkaloid, which enabled further characterization of
glycine action in the nervous system [10, 11]. With the
use of the high-affinity inhibitor strychnine, the GlyR
was purified from membrane fractions of the adult rat
spinal cord [12, 13]. The subunit composition and bind-
ing sites of the receptor and the amino acid sequence of
many of the subunits have been characterized (reviewed
by Rajendra et al., ref. [1]).
Recently, work from our group has demonstrated that a
wide variety of white blood cells involved in inflamma-
tion (i.e., Kupffer cells, alveolar macrophages, and neu-
trophils) also contain glycine-gated chloride channels
[14–16]. By hyperpolarizing the plasma membrane of
leukocytes, glycine makes them less sensitive to inflam-
matory stimuli such as endotoxin and possibly a wide
variety of growth factors. Since glycine is one of the
amino acids in serum that declines in shock, the im-
munoregulatory role of glycine may be very important.
Moreover, elevation of blood levels of glycine with
simple dietary administration has shown remarkable
improvement in shock [17], alcoholic liver injury [18],
some forms of cancer [19], and nephrotoxicity due to
certain drugs [20]. The purpose of this article, therefore,
is to review recent evidence on the beneficial effects of
glycine.
Role of calcium in signaling in Kupffer cells
Calcium is central in cellular regulation, and its basic
physiological and biochemical properties have been
studied for decades [21]. Only recently, with advances in
molecular characterization of calcium channels and
fluorescent indicators sensitive enough to track intracel-
lular movement of calcium, have advances become ex-
ponential. A survey of the entire area is beyond the
scope of this review; however, interested readers are
referred to Berridge [22] and Putney [23]. Briefly, the
plasma membrane of mammalian cells contains two
classes of calcium channels, which are either responsive
to agonists and antagonists (receptor-operated chan-
nels) or regulated by membrane potential (voltage-oper-
ated channels). The former involves production of
inositol 1,4,5-triphosphate (IP
3
) via G-protein-linked
mechanisms triggered by binding of the agonist (e.g.,
epinephrine) to the receptor which releases Ca
2+
from
intracellular stores (reviewed by Putney et al., ref. [24]).
The opening of the latter channel is dependent on the
membrane potential of the plasma membrane which is a
function of distribution of ions in the basal state. Col-
lectively, these channels are involved in muscle contrac-
tion, release of neurotransmitters, regulation of cell
proliferation, apoptosis, and cell differentiation.
It is known that Kupffer cells, the resident hepatic
macrophages, are activated by calcium, but conclusive
evidence that they contain voltage-dependent calcium
channels has only recently been demonstrated in this
laboratory [25]. By measuring the cytosolic free calcium
concentration ([Ca
2+
]
i
) of cultured Kupffer cells, re-
placement of extracellular Na
+
by K
+
caused an in-
crease in [Ca
2+
]
i
in a concentration-dependent manner,
most likely due to membrane depolarization. Further,
increases in intracellular calcium via influx of calcium
through voltage-dependent calcium channels can be in-
duced by stimulating Kupffer cells with endotoxin
(LPS). The exact mechanism whereby LPS causes an
increase in [Ca
2+
] is not completely understood. One
possible mechanism is that LPS generates a signal in
Kupffer cells by activating its receptor CD14. CD14 is
associated with a transmembrane protein toll-like recep-
tor (tlr 2/4), which may serve as a tyrosine kinase
similar to the intracellular region of the interleukin
(IL)-1 receptor [26]. The hypothesis is that initiation of
this signaling pathway by LPS leads to the activation of
phospholipase C and the generation of IP
3
causing the
release of calcium from intracellular stores. Simulta-
neously, depolarization of the plasma membrane causes
Ca
2+
influx, but the mechanisms for this are poorly
understood. The change in the membrane potential
activates voltage-operated calcium channels causing the
influx of extracellular calcium [27, 28]. The dramatic
increase in intracellular free calcium then serves as a
second-messenger signal for cellular signaling events,
cell mobilization, and transcription and translation of
key cytokines (see fig. 1).
L-type Ca
2+
channels in Kupffer cells
Hijioki et al. [25] showed that K
+
-induced Ca
2+
influx
in Kupffer cells was sensitive to the L-type Ca
2+
chan-
nel blocker, nisoldipine. Moreover, it was shown that
Ca
2+
influx induced by the Ca
2+
channel agonist BAY
K 8644 was also inhibited by nitrendipine. Thus it was
hypothesized that Kupffer cells contained L-type Ca
2+
channels. In support of this idea, Takei et al. [29]
showed that the L-type Ca
2+
channel blocker nisoldip-
ine improves graft survival after cold storage and trans-
plantation of rat liver, actions that were confirmed in
isolated perfused rabbit liver [30]. We also found that
nisoldipine decreased tumor necrosis factor-h (TNF-h)
release from liver after transplantation in this model
[31]. Patients with alcoholic hepatitis or cirrhosis fre-
quently exhibit endotoxemia, leading to the production
of TNF-h by Kupffer cells. Therefore, modulation of
Kupffer cell function by blocking Ca
2+
channels repre-
sents a new approach to treatment of hepatic patholog-

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 845Review Article
ical conditions such as alcoholic liver injury, and im-
provement in procedures such as liver transplantation.
Characteristics of glycine-gated chloride channels
The GlyR was first purified by Pfeiffer and Betz [12]
using high-affinity strychnine binding from rat spinal
cord. The channel is comprised of three distinct protein
subunits: a 48-kDa h subunit, a 58-kDa i subunit, and
a 93-kDa cytoplasmic anchoring protein, gephyrin [12,
13]. Three different isoforms of the h subunit have been
identified and cloned from rat: the original purified
48-kDa h subunit (h1), a 49-kDa h2 subunit [32], and a
50-kDa h3 subunit [33, 34]. Moreover, homologues of
the h1, h2, and the i subunits of the GlyR have been
identified and cloned from human and mouse spinal
cord [35 –40]. Recently, a fourth h subunit has been
identified (denoted h4) by Matzenbach et al. [41]. For
rat, mouse, and human, the h subunits share striking
sequence identity with each other and with subunits of
the nicotinic acetylcholine receptor (nAchR) and the
GABA type A receptor (GABA
A
R), as well as several
other ligand-gated chloride channels [1, 33]. The GlyR
is comprised of five subunits, formed from either h
subunits or a combination of h and i subunits, ar-
ranged in a pentameric complex which spans the cell
membrane. The cytoplasmic region of the i subunit
forms a complex with the anchoring protein gephyrin.
The functional properties of the GlyR are related to the
subunit composition of the complete pentamer. Gener-
ally, subunit composition is dictated by either develop-
mental stage or region of expression.
In functional analysis of cultured or isolated in vitro
spinal cord preparations, the EC
50
values for glycine
activation range from 20 to 150 mM [42, 43]. These
values are usually consistent with those obtained with
recombinant GlyR expressed in mammalian cells [44,
45]. However, most recombinant GlyRs have signifi-
cantly less sensitivity to glycine (EC
50
0.3–1 mM) [34,
Figure 1. Working hypothesis. Glycine activates a ligand-gated chloride channel in the plasma
membrane of Kupffer cells which causes an influx of chloride ions leading to the hyperpolarization
of the membrane. Upon an external stimulus such as endotoxin, voltage-dependent influx of
extracellular free calcium occurs through voltage-operated channels. This increase in intracellular
calcium is blunted due to the hyperpolarized state of the plasma membrane by chloride. Intracellu-
lar signaling and cytokine production which is dependent upon the increase in intracellular calcium
are blunted, preventing the cascade of inflammatory cytokines following activation of Kupffer cells
and other white blood cells which contain the glycine receptor. (DAG, diacylglycerol; Glyr, glycine
receptor; IP
3
, inositol 1,4,5-triphosphate; LPS, lipopolysaccharide; PLC, phospholipase c; tlr-2,
toll-like receptor-2; VOC, voltage-operated calcium channel; TNFh, tumor necrosis factor-h; LBP,
lipopolysaccharide binding protein).

M. D. Wheeler et al. Anti-inflammatory effects of glycine846
Figure 2. Effect of glycine on lipopolysaccharide (LPS)-induced
increases in intracellular Ca
2+
concentration ([Ca
2+
]
i
) in isolated
Kupffer cells. [Ca
2+
]
i
in a cultured Kupffer cell was measured
fluorometrically using the fluorescent Ca
2+
indicator fura-2 [14].
LPS was added to stimulate the increase in [Ca
2+
]
i
, while 1000
mM glycine was added 3 min before LPS.
Glycine (1 mM) prevented this increase nearly com-
pletely. Moreover, low concentrations of strychnine (1
mM), an antagonist to the GlyR in the central nervous
system, reversed the inhibitory effect of glycine com-
pletely. The effect of glycine was prevented when cells
were incubated in chloride-free buffer. To test the hy-
pothesis that glycine-gated chloride channels hyperpo-
larize the plasma membrane of Kupffer cells, changes in
plasma membrane potential were estimated in individ-
ual Kupffer cells using the voltage-sensitive dye, bis-ox-
onol (fig. 3). High potassium (25 mM), which
depolarizes the plasma membrane, increased fluores-
cence intensity, indicating that changes in fluorescence
indeed reflect changes in membrane potential. Glycine
(1 mM) gradually decreased fluorescence within 2 min
and blunted increases in fluorescence caused by potas-
sium. Thus, it was concluded that glycine hyperpolar-
izes the plasma membrane, making depolarization more
difficult. Further, LPS increased fluorescence intensity
transiently, an effect also blunted by glycine. This indi-
cates that LPS causes depolarization of the Kupffer cell
plasma membrane and that the effect can be prevented
by glycine. To evaluate the effect of glycine on cytokine
Figure 3. The effect of glycine on Kupffer cell membrane poten-
tial. The relative membrane potential of isolated Kupffer cells was
measured fluorometrically using the fluorescent voltage-sensitive
dye bis-oxonol. (A) LPS was added to induce membrane depolar-
ization of Kupffer cells. (B) Glycine (1 mM) was added 3 min
prior to LPS.
36, 46, 47]. This is important since blood glycine levels
can be elevated over the EC
50
values simply by dietary
treatment. Interestingly, the GlyR can also be activated
by i-alanine and taurine; however, they are less potent
than glycine [48].
Pharmacological evidence for glycine-gated chloride
channels in Kupffer cells
Strychnine, chloride-free buffer, and membrane
potential
Based on studies with strychnine, chloride-free buffer,
and measurement of radioactive chloride flux, it has
been shown that Kupffer cells contain a glycine-gated
chloride channel [14, 18]. First, the effect of glycine on
[Ca
2+
]
i
in cultured Kupffer cells stimulated with LPS
was investigated to assess whether they contain a
glycine-gated chloride channel. LPS increased [Ca
2+
]
i
rapidly with peak values reaching over 300 nM (fig. 2).

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 847Review Article
Table 1. Evidence that a wide variety of white cells contain glycine-gated chloride channels.
Reversal byIC
50
value (mM)Cell type Agonist-induced increase in Dependence of extra- Glycine-stimu-
[Ca
2+
]
i
blocked by glycine lated
36
Cl
−
strychnine cellular chloride
influx
0.3–0.6 ++Kupffer cell ++
+ 0.1 +++Alveolar
macrophage
0.3–0.5 +Neutrophil ++ +
+++Jurkat
T lymphocyte +
+Blood mono-
cyte
Blanks indicate that the particular parameter has not been studied.
production by Kupffer cells, LPS-induced TNF-h pro-
duction was measured. As expected, isolated Kupffer
cells produced large amounts of TNF-h in the presence
of LPS (1 mg/ml). However, glycine (1 mM) reduced
LPS-induced TNF-h production by about 70%. This
effect of glycine on TNF-h production was also reversed
by low-dose strychnine (1 mM). In comparison to the
GlyR in the spinal cord where the IC
50
for glycine is
30–100 mM [48], the IC
50
for glycine on the Kupffer cell
is slightly higher ( 0.3 mM). Conversely, GlyRs ex-
pressed in in vitro expression systems displayed much
lower sensitivity to glycine (IC
50
, 0.3–1 mM) [34]. It
was concluded therefore from this pharmacological evi-
dence that Kupffer cells contain a glycine-gated chlo-
ride channel similar to one described previously in the
spinal cord. Prevention of increases in [Ca
2+
]
i
due to
LPS by activation of chloride influx reduces synthesis
and release of toxic cytokines by Kupffer cells.
Glycine stimulates chloride flux in Kupffer cells
In synaptosomes, influx of radiolabeled chloride is used
to measure the flux of chloride through chloride chan-
nels [49]. Therefore, we reasoned that if Kupffer cells
contain a glycine-gated chloride channel, glycine would
stimulate movement of radiolabeled chloride [18]. In-
deed, glycine stimulated chloride movement with EC
50
values between 0.1 and 0.5 mM. This provides hard
physical evidence that Kupffer cells contain a glycine-
gated chloride channel.
Taurine
Since taurine, a ubiquitous sulfur-containing i-amino
acid, acts like glycine in neurons by causing hyperpolar-
ization, it was hypothesized that taurine would act via a
mechanism similar to that of glycine and blunt the
LPS-induced increase in [Ca
2+
]
i
in Kupffer cells. To
test this hypothesis, Kupffer cells were isolated from
rats and cultured for 24 h. LPS-induced changes in
[Ca
2+
]
i
were monitored fluorometrically in single cells,
while levels of TNF-h released by Kupffer cells after
exposure to LPS were measured by ELISA. Taurine
significantly blunted the LPS-induced increase in [Ca
2+
]
i
in a dose-dependent manner (IC
50
, 0.11 mM), the IC
50
being similar to that for its action on the spinal cord
GlyR (IC
50
, 0.05 –0.1 mM) [50]. This effect was reversed
by strychnine (1 mM) and was prevented when chloride
was removed from the extracellular medium. Moreover,
like glycine, taurine increased
36
Cl
−
uptake by Kupffer
cells in a dose-dependent manner. In contrast, other
sulfur-containing amino acids (i.e., cysteine and me-
thionine) were without effect. These results support the
hypothesis that taurine, like glycine, activates a chloride
channel in Kupffer cells. In addition, LPS-induced
TNF-h production was reduced by more than 40% by
taurine, an effect which was also reversed by strychnine.
Thus, taurine blocks the increase in [Ca
2+
]
i
due to LPS
and significantly reduces TNF-h production by mecha-
nisms involving chloride influx into the Kupffer cell.
Glycine-gated chloride channels appear to be ubiquitous
in leukocytes
One important question is whether the glycine-gated
chloride channel is unique to Kupffer cells or whether it
exists in other leukocytes. In the six types of leukocytes
we have studied to date (see table 1), glycine blunted
agonist-induced increases in [Ca
2+
]
i
, a phenomenon
dependent on extracellular chloride and reversed by
strychnine. Further, glycine and taurine stimulated radi-
olabeled chloride flux in all leukocytes studied so far,
providing good physical evidence to support the idea
that glycine-gated chloride channels are widespread in
leukocyte populations. Recently, Spittler et al. [51]
showed that glycine inhibited TNF-h and IL-1 produc-
tion and enhanced expression of IL-10 from isolated
blood monocytes, further confirming the immunosup-
pressive effects of glycine.

M. D. Wheeler et al. Anti-inflammatory effects of glycine848
Molecular evidence for glycine-gated chloride channels
in Kupffer cells
Since the GlyR in the central nervous system is a
pentameric assembly of four subtypes of ligand-binding
h subunits and a single subtype of a structural i sub-
unit, mRNA from isolated Kupffer cells was reverse
transcribed and amplified using primers specific for an
internal region of the glycine receptor i subunit. RT-
PCR amplification from Kupffer cell mRNA resulted in
a 550-base-pair fragment as predicted from the cloned
sequence from the spinal cord glycine receptor (fig. 4).
PCR amplification of the GlyR i subunit from spinal
cord cDNA also resulted in a 550-base-pair fragment.
The low quantity of the mRNA in Kupffer cells com-
pared to spinal cord may reflect the relative expression
of the receptor in the different tissues. It is likely that
Kupffer cells express much less GlyR than spinal cord.
Alternatively, the minute quantity may be due to ineffi-
cient amplification of the cDNA in Kupffer cells, result-
ing from slightly different nucleotide sequences. Since
the primers for RT-PCR were designed based on the
spinal cord GlyR sequence, differences in the nucleotide
sequence in the Kupffer cell GlyR mRNA would lower
PCR amplification efficiency.
The GlyR in the brain has been successfully detected
using the monoclonal antibody, anti-GlyR4a, which
recognizes regions on both h and i subunits of the
receptor [35]. Western blot analysis of Kupffer cell
membranes using the GlyR4a antibody yielded evidence
for both h and i subunits; however, the proteins were
slightly larger than the subunits detected in rat spinal
cord membranes [52].
Molecular evidence has been presented for both sub-
units of the GlyR in the central nervous system [36],
and Western blotting has identified the i subunit in the
kidney tubule [53]. RT-PCR and Western blotting for
the glycine-gated chloride channel in the Kupffer cell
provide molecular evidence for the existence of the
receptor in Kupffer cells. However, the molecular
weights of the proteins detected in the Kupffer cell with
the anti-GlyR4a monoclonal antibody differ from those
of the h and i subunits of the spinal cord GlyR.
However, the molecular weight of the amplified RT-
PCR product from the Kupffer cell was as predicted
(fig. 4). Therefore, it was hypothesized that the two
proteins identified in the Kupffer cell are highly similar
to the h and i subunits of cloned GlyR in specific
regions, such as ligand-binding sites or transmembrane-
spanning regions, but are somehow different in overall
structure and/or sequence, thus yielding larger proteins
than expected. Differences in molecular weights may be
due to posttranslational modifications of the subunits.
There are three putative N-glycosylation sites on both
the h and i subunits. Whether or not there are modifi-
cations which contribute to the discrepancy in molecu-
lar weights is not known. However, these data provide
the first molecular evidence that Kupffer cells indeed
possess a glycine-gated chloride channel similar to that
expressed in neuronal tissue.
Examples of beneficial effects of glycine
For many, it is difficult to fathom that beneficial effects
can be obtained in several disease states with the sim-
plest amino acid, glycine. However, evidence continues
to mount in favor of this idea. It is now clear that
dietary glycine, which increases the blood concentration
of glycine to greater than 1 mM from basal concentra-
tions ranging from 0.1–0.2 mM, protects against shock
caused either by blood loss or endotoxin. It reduces
alcohol levels in the stomach and improves recovery
from alcoholic hepatitis [54]. It also reduces fibrosis
caused by experimental drugs. It diminishes liver injury
caused by hepatotoxic drugs and blocks programmed
cell death [55]. Moreover, it reduces cancer caused by
chemicals [56]. In the kidney, it reduces the nephrotoxi-
city caused by the drug cyclosporin A and prevents
hypoxia and free radical formation [57]. However, we
predict that it will be useful in other diseases because it
(i) diminishes TNF-h production and (ii) decreases cell
signaling via a unique mechanism.
Unique concept of glycine action
The question then arises as to how glycine has such
beneficial effects. The answer is that it most likely has
an inhibitory effect on cell signaling mechanisms in cells
that contain a glycine-gated chloride channel. As men-
Figure 4. Reverse transcription-polymerase chain reaction (RT-
PCR) for the glycine receptor i subunit in Kupffer cells. RNA
isolated from Kupffer cells was subjected to RT-PCR using PCR
primers specific for the glycine receptor i subunit, with i actin as
control. Spinal cord RNA was also used as a positive control.
Lane 1, 100-base-pair DNA ladder; 2, spinal cord i actin; 3,
Kupffer cell i actin; 4, spinal cord glycine receptor i subunit; 5,
Kupffer cell glycine receptor i subunit.

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 849Review Article
tioned above, receptor- and voltage-gated calcium chan-
nels are central in elevation of calcium for intracellular
signaling in many immune cell types such as the Kupf-
fer cell. Furthermore, it is known that increases in
[Ca
2+
]
i
trigger opening of a chloride channel in the
plasma membrane leading to hyperpolarization, making
voltage-dependent calcium channels more difficult to
open. We hypothesize that glycine opens a chloride
channel in the plasma membrane of Kupffer cells and
other white blood cells, rendering calcium influx trig-
gered by a variety of agonists, drugs, and growth fac-
tors more difficult or impossible. In vitro data from
isolated Kupffer cells clearly support this hypothesis
and explain the near universal action of glycine [14].
Thus, many other disease states involving activated
immune cells, in particular macrophages, neutrophils,
and lymphocytes, should be affected by elevated levels
of glycine, according to our hypothesis.
Reperfusion injury and surgical manipulation
Because glycine prevents cell death induced by anoxia
in proximal tubules of the kidney, we studied its effect
on hypoxia-reoxygenation in the liver. We used a low-
flow, reflow liver perfusion model [58]. With this proto-
col, livers were perfused at low flow rates of  1ml/g
per minute for 75 min, which caused cells in pericentral
regions of the liver lobule to become anoxic due to
insufficient delivery of oxygen. When normal flow rates
( 4ml/g per minute) were restored for 40 min, oxy-
gen-dependent reperfusion injury occurred. Upon
reflow, lactate dehydrogenase (LDH), a cytosolic en-
zyme, and malondialdehyde (MDA), an end product of
lipid peroxidation, were released into the effluent per-
fusate. LDH increased from basal levels of  1to35
IU/g per hour in livers from control rats. Glycine
(0.06–2.00 mM) minimized enzyme release in a dose-de-
pendent manner (half-maximal decrease=133 mM),
with maximal values reaching only 5 IU/g per hour
when glycine was increased to 2 mM. Reflow for 40 min
after 75 min of low-flow hypoxia caused death in 
30% of previously anoxic parenchymal cells in pericen-
tral regions; however, infusion of glycine decreased cell
death to less than 10% (fig. 5). Trypan blue distribution
time, an indicator of hepatic microcirculation, was re-
duced significantly by glycine at 5 and 40 min after
reflow. Time for oxygen to reach steady state upon
reflow was reduced by glycine in a dose-dependent
manner, and the rates of entry and exit of a dye
confined to vascular space (fluorescein dextran) were
increased two- to threefold by glycine, respectively.
Taken together, these data indicate that reperfusion
injury that occurs in previously hypoxic pericentral re-
gions of the liver upon reintroduction of oxygen is
minimized by glycine, most likely by action on a
Figure 5. Effect of glycine on rates of cell death in periportal and
pericentral regions of the liver lobule following reperfusion. Livers
were perfused for 45 min at low-flow rates (1 ml/min). Glycine (2
mM final concentration) was infused 10 min prior to reperfusion
at normal flow rates (4 ml/min). Trypan blue was infused at the
end of 40 min of reperfusion. Nuclei of parenchymal cells in a
zone radiating five cells from either pericentral or periportal
regions were identified as trypan blue positive or negative. The
percentage of staining was calculated from the number of stained
nuclei divided by the total number of cells in a region. Values are
means9 SE (n = 4 – 5, ANOVA, P = 0.001). *P B 0.05 compared
with control group.
glycine-sensitive anion channel which improves micro-
circulation during the reperfusion period, possibly by
decreasing the production of vasoactive cytokines and
eicosanoids from Kupffer cells.
Next, we wanted to see if this information on reperfu-
sion injury and glycine could be applied to a clinically
relevant problem. For this purpose, we chose primary
nonfunction associated with liver transplantation [59,
60]. The etiology of primary graft nonfunction and
dysfunction is still unknown but most likely involves
Kupffer-cell-dependent reperfusion injury; however, re-
cent evidence indicates that the donor operation may
also be important [60]. Moreover, treatment with
gadolinium chloride, a Kupffer cell toxicant [61], de-
pletes the liver of Kupffer cells and reduces liver dys-
function following transplantation, clearly supporting
the hypothesis that Kupffer cells participate in primary
nonfunction [60]. The role of manipulation of the liver
which cannot be prevented completely during standard
harvesting techniques was assessed (fig. 6). Donor rat

M. D. Wheeler et al. Anti-inflammatory effects of glycine850
livers were harvested either with or without gentle ma-
nipulation. Subsequently, orthotopic liver transplanta-
tion was performed after 1 h of cold storage in
University of Wisconsin cold storage solution. In some
rats, Kupffer cells were treated with dietary glycine
before harvest. In the nonmanipulated group, survival
was 100%; however, gentle manipulation decreased sur-
vival to 30% after transplantation. Further, manipu-
lation elevated transaminases five-fold 8 h after
transplantation. Dietary glycine prevented the effects of
organ manipulation on all parameters studied. These
data indicate for the first time that brief, gentle manipu-
lation of the liver has a marked detrimental effect on
survival by mechanisms involving Kupffer cells.
Endotoxin shock
The effects of a glycine-containing diet (5%) on mortal-
ity and liver injury due to intravenous injection of LPS
were studied in rats in vivo (fig. 7A) [17]. Fifty percent
of the rats fed the control diet died within 24 h after an
Figure 7. Effect of dietary glycine on survival and TNF-h pro-
duction due to endotoxin shock. Data are means9 SE (n= 6,
*PB 0.05 with Mann-Whitney’s rank-sum test). (A) Rats were fed
a 5% glycine or control diet for 3 days prior to the injection of
endotoxin (5–30 mg/kg). Survival rates were monitored for 24 h.
(B) Rats were fed a 5% glycine or control diet for 3 days prior to
endotoxin injection (10 mg/kg). Serum TNF-h levels were moni-
tored for up to 310 min.
Figure 6. Effects of gentle organ manipulation on serum aspar-
tate transaminase (AST) levels following transplantation. Animals
were untreated or treated with gadolinium chloride (10 mg/kg) or
a 5% glycine diet for 3 days prior to surgery. Blood was collected
8 h after transplantation for serum AST measurement. Values are
means9 SE (n= 4 – 8, two-way ANOVA, P = 0.05).
a
PB 0.05
compared with no manipulation;
b
PB 0.05 compared with manip-
ulation without pretreatment.
intravenous injection of LPS (10 mg/kg), whereas feed-
ing rats glycine totally prevented mortality and
markedly reduced the LPS-induced elevation of serum
transaminases, hepatic necrosis, and lung injury. The
elevation in serum TNF-h due to LPS was also blunted
and delayed significantly by glycine feeding (fig. 7B). In
a two-hit model (hepatic ischemia and injection of sub-
lethal LPS), all rats fed control diet died, whereas 83%
of glycine-fed animals survived with a significant reduc-

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 851Review Article
tion in transaminases and improved liver and lung
histology. LPS elevated [Ca
2+
]
i
in cultured Kupffer
cells, an effect blocked almost completely by glycine
[14]. Glycine most likely reduces injury and mortality by
preventing the LPS-induced elevation of [Ca
2+
]
i
in
Kupffer cells, thereby minimizing toxic eicosanoid and
cytokine (e.g., LTB
4
and TNF-h) production.
Alcohol
Tsukamoto-French model. When Kupffer cells were in-
activated with gadolinium chloride (GdCl
3
) or endo-
toxin was minimized by antibiotics, early
alcohol-induced liver injury was blocked in experimen-
tal animals. Since glycine also inactivates Kupffer cells,
we explored its actions in alcohol-induced liver injury
[54]. Male Wistar rats were exposed to ethanol (10– 12
g/kg per day) continuously for up to 4 weeks via an
intragastric feeding protocol. The effect of glycine on
the first-pass metabolism of ethanol was also examined
in vivo, and the effect on alcohol metabolism was
estimated specifically in perfused liver. Glycine de-
creased ethanol concentrations precipitously in urine,
breath, peripheral blood, portal blood, feces, and stom-
ach contents. Moreover, serum aspartate aminotrans-
ferase levels were elevated to 183 U/l after 4 weeks of
ethanol treatment. In contrast, values were significantly
lower in rats given glycine along with ethanol. Hepatic
steatosis and necrosis were also reduced significantly by
glycine. Glycine dramatically increased the first-pass
elimination of ethanol in vivo but had no effect on
alcohol metabolism in the perfused liver. Thus, it is
concluded that glycine minimizes alcohol-induced liver
injury in vivo, preventing ethanol from reaching the
liver by accelerating first-pass metabolism by the stom-
ach [54, 62].
Recovery from alcoholic hepatitis. Since the effect of
glycine on ethanol metabolism in the stomach compli-
cated the evaluation of the effect of glycine in alcohol-
induced liver disease, we studied its effect on the
recovery phase [18]. When patients with alcoholic liver
disease enter hospital, alcohol is removed. Accordingly,
we induced alcoholic hepatitis with the Tsukamoto-
French protocol for 6 weeks. Then, either control or
glycine-containing diets were given and recovery from
liver injury was assessed. In as little as 1 week, liver
histology and serum transaminases were improved over
30% by a glycine-containing diet compared to a control
diet. Correspondingly, TNF-h mRNA was reduced
more in liver tissue by glycine than in controls, most
likely by increasing the Cl
−
flux into Kupffer cells,
thereby diminishing cytokine and eicosanoid produc-
tion. These experiments are very important, since they
suggest that simple dietary glycine, which should be well
tolerated by patients, could speed recovery from alco-
holic hepatitis.
Cancer
Peroxisome proliferator WY-14,643-induced hepatocyte
proliferation. Peroxisome proliferators are a group of
nongenotoxic carcinogens which include a number of
hypolipidemic drugs, solvents, and industrial plasticiz-
ers. Although the mechanism by which they cause can-
cer remains unknown, one likely possibility is that they
act as tumor promoters by increasing cell proliferation
[63]. Kupffer cells represent a rich source of mitogenic
cytokines (e.g., TNF-h) and have been shown to be
stimulated by peroxisome proliferators [64]. Since
glycine prevents activation of Kupffer cells, these exper-
iments were designed to test the hypothesis that a diet
containing glycine could prevent the mitogenic effect of
the peroxisome proliferator WY-14,643 (fig. 8) [56]. The
effects of glycine on WY-14,643-induced increases in
cell proliferation after a single dose or after feeding
WY-14,643 in the diet (0.1%) for three weeks were
assessed. As expected, 24 h after a single dose of WY-
14,643, rates of cell proliferation were increased about
Figure 8. Cell proliferation after 3 weeks of WY-14,643 (WY)
and glycine (GLY) in the diet. Cell proliferation was assessed by
BrdU incorporation. Values are means9 SE [n = 5, two-way
ANOVA, *P B 0.05, compared with control (CON) and WY+
GLY groups].

M. D. Wheeler et al. Anti-inflammatory effects of glycine852
fivefold. Glycine largely prevented the increase in cell
turnover. Acyl CoA oxidase, a marker enzyme for per-
oxisomes, increased significantly, indicating that perox-
isomes were induced about twofold in livers of
WY-14,643-treated rats after 24 h. Unlike cell prolifera-
tion, however, acyl CoA oxidase was not affected by
glycine, consistent with the hypothesis that cell and
peroxisome proliferation results from different signaling
pathways. After 3 weeks, dietary glycine reduced basal
rates of cell proliferation by about 50% and completely
prevented the sustained fivefold increase in cell prolifer-
ation caused by feeding dietary WY-14,643. Thus,
weeks of dietary exposure to WY-14,643 caused a six-
fold increase in acyl CoA oxidase activity which was
also unaffected by glycine, demonstrating that a diet
containing glycine prevents the increase in hepatocyte
proliferation caused by a potent peroxisome prolifera-
tor without affecting induction of peroxisomes. These
data support the hypothesis that dietary glycine could
be effective in preventing cancer caused by
nongenotoxic carcinogens such as WY-14,643.
These data are consistent with the hypothesis that pro-
duction of TNF-h by Kupffer cells plays a central role
in the development of peroxisome-proliferator-induced
liver cancer and raises the possibility that Kupffer cells
may also be important in the development of cancer
caused by other nongenotoxic carcinogens. The com-
plete prevention of WY-14,643-induced cell prolifera-
tion and the 50% reduction in basal levels of hepatocyte
replication with a diet containing glycine predicts that
glycine may be an effective dietary tool for the preven-
tion and possibly even the treatment of cancer.
Tumors from B-16 melanoma cells. Since dietary glycine
inhibited hepatocyte proliferation in response to WY-
14,643 [56] and cell replication is associated with hepatic
cancer caused by WY-14,643 [65], glycine may have
general anticancer properties. Therefore, the hypothesis
that glycine would inhibit the growth of tumors arising
from implanted B16 melanoma cells was tested [19].
Mice were fed a diet containing 5% glycine and 15%
casein or a control diet containing 20% casein. After
monitoring tumor volume daily for 14 days, the tumor
was removed, weighed, and sectioned for histological
analysis (fig. 9). Tumors from glycine-fed mice weighed
65% less than those from control animals after 14 days;
however, neither tumor size nor mitotic index differed 2
days after implantation when tumor growth was inde-
pendent of vascularization. Further, tumors from mice
fed glycine had fewer arteries after 14 days, suggesting
an inhibitory role of glycine on angiogenesis and tumor
vascularization. Indeed, glycine (0.01–10 mM) inhibited
the growth of endothelial cells in vitro, supporting the
hypothesis that glycine inhibits tumor growth in 6i6o
through mechanisms involving endothelial cell
proliferation.
Figure 9. Effect of dietary glycine on tumor volume. Tumor
diameter was measured using digital calipers and the volume was
calculated. Data shown are means9 SE (n = 5, repeated-measures
ANOVA on ranks, *PB 0.05).
Nephrotoxicity
Prevention of cyclosporin-A-induced nephrotoxicity with
dietary glycine. The nonessential amino acid glycine has
been used previously to prevent hypoxic and ischemic
injury to kidney tissue in vitro [53, 66, 67]. Since there is
some evidence that the immunosuppressant cyclosporin
A causes nephrotoxicity through a hypoxia-reoxygena-
tion mechanism that could involve infiltration and acti-
vation of macrophages and neutrophils, we
hypothesized that dietary glycine could prevent this
injury (fig. 10). Accordingly, rats were fed a diet con-
taining glycine (5%) or a control diet for 3 days prior to
cyclosporin A treatment. To produce nephrotoxicity,
cyclosporin A (25 mg/kg daily by gavage) was adminis-
tered for 28 days while animals were maintained on
glycine or control diets. Serum creatinine and urea,
glomerular filtration rates, and kidney histology were
evaluated. As expected, cyclosporin A caused kidney
damage in rats fed the control diet, reflected in signifi-
cantly elevated serum urea and creatinine. In addition,
cyclosporin A treatment decreased glomerular filtration
rates by nearly 70%, caused proximal tubular dilation
and necrosis as well as increased macrophage and neu-
trophil infiltration into the kidney. Dietary glycine pre-

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 853Review Article
vented or minimized kidney damage due to cyclosporin
A in all parameters studied. Furthermore, feeding
glycine for up to 1 month had no detrimental effect on
kidney function. Dietary glycine is a safe and effective
treatment to reduce the nephrotoxicity of cyclosporin
A.
One major advantage of glycine over drug therapy is
that it most likely acts at several points in the pathology
due to cyclosporin A by preventing vasoconstriction,
proximal tubular hypoxia, as well as activation of
macrophages and mesangial cells. It is hypothesized
that glycine inactivates macrophages by blunting the
increase in [Ca
2+
]
i
, thereby minimizing the release of
vasoactive and inflammatory eicosanoids and cytokines.
A second advantage is that glycine is a natural, non-
toxic amino acid circulating in the 100–200 mM range
under normal conditions. Thus, dietary supplementa-
tion of cyclosporin A patients with glycine should be of
immense benefit in preventing the major side effect of
nephrotoxicity.
Role of hypoxia and free radicals in cyclosporin A toxic-
ity. It is likely that cyclosporin A causes vasoconstric-
Figure 10. Prevention of kidney pathology due to cyclosporin A (CSA)with glycine. Photomicrographs of perfusion-fixed kidneys from
rats after 4 weeks of treatment at low (A,C,E)(×100) and high (B,D,F)(×400) magnification. (A,B) Sections from control animals.
(C,D) Animals treated for 4 weeks with cyclosporin (25 mg/kg). (E,F) Sections from animals treated for 4 weeks with cyclosporin (25
mg/kg) fed a glycine-containing (5%) diet.

M. D. Wheeler et al. Anti-inflammatory effects of glycine854
tion which leads to hypoxia-reperfusion injury; there-
fore, experiments were designed to attempt to obtain
physical evidence for hypoxia and free radical produc-
tion in kidney following cyclosporin A treatment. Rats
were treated daily with cyclosporin A (25 mg/kg, p.o.)
for 5 days, and pimonidazole, a 2-nitroimidazole hy-
poxia marker, was injected 2 h after the last dose of
cyclosporin A. h-(4-Pyridl 1-oxide)-N-tert-butylnitrone
(POBN) was injected 3 h after cyclosporin A to trap
free radicals. Cyclosporin A nearly doubled serum crea-
tinine and decreased glomerular filtration rates by 65%
as expected. Pimonidazole adduct binding in the kidney
was increased nearly threefold by cyclosporin A,
providing physical evidence for tissue hypoxia. More-
over, cyclosporin A increased POBN/radical adducts
nearly sixfold in the urine but did not alter levels in the
serum. Glycine, an amino acid which prevents cy-
closporin A toxicity, minimized cyclosporin A-induced
hypoxia, blocked free radical production, and blunted
decreases in glomerular filtration rate. These results
demonstrate clearly for the first time that cyclosporin A
causes hypoxia in renal cells and increases production
of a new free radical species exclusively in the kidney.
Therefore, it is concluded that cyclosporin A causes
renal injury by mechanisms involving hypoxia-reoxy-
genation. Moreover, these effects can be prevented ef-
fectively by dietary glycine.
Clinical considerations
Based on these exciting findings, it is reasonable to
propose that glycine would be useful in the treatment of
many inflammatory-type diseases in humans. Obvious
disease targets include sepsis and endotoxemia, experi-
enced in many patients following abdominal surgery or
trauma. It is also reasonable to think that glycine may
be useful in many respiratory diseases such as adult
respiratory distress syndrome and asthma. Moreover,
the use of glycine in the prevention and/or treatment of
certain types of cancer looks promising. As the mecha-
nisms of tumorigenesis and angiogenesis become
clearer, the role of glycine as an anticancer agent may
become more exciting. Certainly, the prospect of pre-
and posttransplant treatment with glycine in combina-
tion with standard immunosuppressive agents is an ex-
citing possibility. Since common immunosuppressive
agents exhibit many toxic effects, cotreatment with
glycine may ameliorate these effects, allowing doses of
several toxic drugs to be lowered. All the ramifications
of dietary glycine treatment have certainly not been
addressed, but the practical implications of the diet
have been surprising. A long-term 5% glycine diet (\4
weeks) still proves beneficial in some models. Addition-
ally, lowering the glycine composition to 1.25% in the
diet is sufficient to protect tissues in several models.
These considerations should stimulate further investiga-
tion into clinical applications of dietary glycine. Because
glycine can be elevated simply by dietary measures, the
feasibility of therapeutic and preventive approaches for
many diseases with this new immunonutrient is quite
promising.
1 Rajendra S., Lynch J. W. and Schofield P. R. (1997) The
glycine receptor. Pharmacol. Ther. 73: 121–146
2 Werman R., Davidoff R. A. and Aprison M. H. (1968)
Inhibitory action of glycine on spinal neurons in the cat. J.
Neurophysiol. 31: 81–95
3 Werman R., Davidoff R. A. and Aprison M. H. (1967)
Inhibition of motoneurons by ionophoresis of glycine. Nature
214: 681– 683
4 Davidoff R. A., Graham L. T. Jr, Shank R. P., Werman R.
and Aprison M. H. (1967) Changes in amino acid concentra-
tions associated with loss of spinal interneurons. J. Neu-
rochem. 14: 1025–1031
5 Araki T., Ito M. and Oscarsson O. (1961) Anion permeability
of the synaptic and non-synaptic motoneurone membrane. J.
Physiol. 159: 410–435
6 Aprison M. H. and Werman R. (1965) The distribution of
glycine in cat spinal cord and roots. Life Sci. 4: 2075–2083
7 Hokfelt T. and Ljungdahl A. (1971) Light and electron micro-
scopic autoradiograph on spinal cord slices after incubation
with labeled glycine. Brain Res. 32: 189–194
8 Curtis D. R., Hosli L., Johnston G. A. R. and Johnston I. H.
(1968) The hyperpolerization of spinal motoneurones by
glycine and related amino acids. Exp. Brain Res. 5: 235–258
9 Curtis D. R., Hosli L. and Johnston G. A. R. (1968) A
pharmacological study of the depression of spinal neurones
by glycine and related amino acids. Exp. Brain Res. 6: 1–18
10 Curtis D. R. (1962) The depression of spinal inhibition by
electrophoretically administered strychnine. Int. J. Neu-
ropharmacol. 1: 239–250
11 Young A. B. and Snyder S. H. (1973) Strychnine binding
associated with glycine receptors of the central nervous sys-
tem. Proc. Natl. Acad. Sci. USA 70: 2832–2836
12 Pfeiffer F. and Betz H. (1981) Solubilization of the glycine
receptor from the rat spinal cord. Brain Res. 226: 273–279
13 Pfeiffer F., Graham D. and Betz H. (1982) Purification by
affinity chromotography of the glycine receptor of rat spinal
cord. J. Biol. Chem. 257: 9389–9393
14 Ikejima K., Qu W., Stachlewitz R. F. and Thurman R. G.
(1997) Kupffer cells contain a glycine-gated chloride channel.
Am. J. Physiol. 272: G1581–G1586
15 Wheeler M. D. and Thurman R. G. (in press) Production of
superoxide and TNFh from alveolar macrophages is blunted
by glycine. Am. J. Physiol.
16 Stachlewitz R. F., Ikejima K. and Thurman R. G. (1995)
Increases in intracellular calcium in neutrophils (PMNs) due
to formyl-methionine-leucine-phenylalanine (FMLP) and en-
dotoxin are blocked completely by glycine. Hepatology 22:
1105
17 Ikejima K., Iimuro Y., Forman D. T. and Thurman R. G.
(1996) A diet containing glycine improves survival in endo-
toxin shock in the rat. Am. J. Physiol. 271: G97–G103
18 Yin M., Ikejima K., Arteel G. E., Seabra V., Bradford B. U.,
Kono H. et al. (1998) Glycine accelerates recovery from
alcohol-induced liver injury. J. Pharmacol. Exp. Ther. 286:
1014– 1019
19 Rose M. L., Madren J., Bunzendahl H. and Thurman R. G.
(1999) Dietary glycine inhibits the growth of B16 melanoma
tumors in mice. Carcinogenesis 20: 793–798
20 Thurman R. G., Zhong Z., Frankenberg M. v., Stachlewitz
R. F. and Bunzendahl H. (1997) Prevention of cyclosporin-in-
duced nephrotoxicity with dietary glycine. Transplantation
63: 1661– 1667

CMLS, Cell. Mol. Life Sci. Vol. 56, 1999 855Review Article
21 Clapham D. E. (1995) Calcium signaling. Cell 80: 259–268
22 Berridge M. J. (1984) Inositol trisphosphate and diacylglyc-
erol as second messengers. Biochem. J. 220: 345–360
23 Putney J. W. (1993) Excitement about calcium signaling in
inexcitable cells. Science 262: 676–678
24 Putney J. W., Takemura H., Hughes A. R., Horstman D. A.
and Thastrup O. (1989) How do inositol phosphates regulate
calcium signaling? FASEB J. 3: 1899–1905
25 Hijioka T., Rosenberg R. L., Lemasters J. J. and Thurman R.
G. (1992) Kupffer cells contain voltage-dependent calcium
channels. Mol. Pharmacol. 41: 435–440
26 Perera P-Y., Vogel S. N., Detore G. R., Haziot A. and
Goyert S. M. (1997) CD14-dependent and CD14-independent
signaling pathways in murine macrophages from normal and
CD14 knockout mice stimulated with lipopolysaccharide or
taxol. J. Immunol. 158: 4422–4429
27 Pugin J., Kravchenko V. V., Lee J. D., Kline L., Ulevitch R.
J. and Tobias P. S. (1998) Cell activation mediated by glyco-
sylphosphatidyl inositol-anchored or transmembrane forms of
CD14. Infect. Immunol. 66: 1174–1180
28 Ulevitch R. J. and Tobias P. S. (1995) Receptor-dependent
mechanisms of cell stimulation by bacterial endotoxin. Annu.
Rev. Immunol. 13: 437–457
29 Takei Y., Marzi I., Kauffman F. C., Lemasters J. J. and
Thurman R. G. (1990) Increase in survival time of liver
transplants from injury by protease inhibitors and a calcium
channel blocker, nisoldipine. Transplantation 50: 14–20
30 Marzi I., Walcher F. and Buhren V. (1993) Macrophage
activation and leukocyte adhesion after liver transplantation.
Am. J. Physiol. 265: G172–G177
31 Takei Y., Marzi I., Kauffman F. C., Cowper K. B., Lemasters
J. J. and Thurman R. G. (1990) Prevention of early graft
failure by the calcium channel blocker nisoldipine: involve-
ment of Kupffer cells. Transplant. Proc. 22: 2202–2203
32 Becker C.-M., Hoch W. and Betz H. (1988) Glycine receptor
heterogeneity in rat spinal cord during postnatal develop-
ment. EMBO J. 7: 3717–3726
33 Betz, H., Langosch, D., Hoch, W., Prior, P., Pribilla, I.,
Kuhse, J. et al. (1991) Structure and expression of inhibitory
glycine receptors. In: Neuroreceptor Mechanisms in Brain,
pp. 421–429, Kito S. (ed.), Plenum Press, New York
34 Kuhse J., Schmieden V. and Betz H. (1990) Identification and
functional expression of a novel binding subunit of the in-
hibitory glycine receptor. J. Biol. Chem. 265: 22317–22320
35 Pfeiffer F., Simler R., Grenningloh G. and Betz H. (1984)
Monoclonal antibodies and peptide mapping reveal structural
similarities between the subunits of the glycine receptor of rat
spinal chord. Proc. Natl. Acad. Sci. USA 81: 7224–7227
36 Grenningloh G., Schmieden V., Schofield P. R., Seeberg P.
H., Siddique T., Mohandas T. K. et al. (1990) Alpha subunit
variants of the human glycine receptor: primary structures,
functional expression, and chromosomal localization of the
corresponding genes. EMBO J. 9: 771–779
37 Handford C. A., Lynch J. W., Baker E., Webb G. C., Ford J.
H., Sutherland G. R. et al. (1996) The human glycine receptor
beta subunit: primary structure, functional characterization
and chromosomal localization of the human and murine
genes. Brain Res. Mol. Brain Res. 35: 211–219
38 Ryan S. G., Buckwalter M. S., Lynch J. W., Handford C. A.,
Segura L., Shiang R. et al. (1994) A missense mutation in the
gene encoding the alpha 1 subunit of the inhibitory glycine
receptor in the spasmodic mouse. Nat. Genet. 7: 131–135
39 Saul B., Schmieden V., Kling C., Mulhardt C., Gass P.,
Kuhse J. et al. (1994) Point mutation of glycine receptor
alpha 1 subunit in spasmatic mouse affects agonist responses.
FEBS Lett. 350: 71–76
40 Kingsmore S. F., Giros B., Suh D., Bieniarz M., Caron M. G.
and Seldin M. F. (1994) Glycine receptor beta-subunit gene
mutation in spastic mouse associated with LINE-1 element
insertion. Nat. Genet. 7: 136–141
41 Matzenbach B., Maulet Y., Sefton L., Courtier B., Avner P.,
Guenet J.-L. et al. (1994) Structural analysis of mouse glycine
receptor alpha subunit genes: identification and chromosomal
localization of a novel variant. J. Biol. Chem. 269: 2607– 2612
42 Krishtal O. A., Osipchuk Y. V. and Vrublevsky S. V. (1988)
Properties of glycine-activated conductances in rat brain neu-
rones. Neurosci. Lett. 84: 271–276
43 Akaike N. and Kaneda M. (1989) Glycine-gated chloride
currents in acutely isolated rat hypothalamic neurons. J.
Neurophysiol. 62: 1400–1408
44 Sontheimer H., Becker C.-M., Pritchett D. B., Schofield P. R.,
Grenningloh G., Kettenmann H. et al. (1989) Functional
chloride channels by mammalian cell expression of rat glycine
receptor subunit. Neuron 2: 1491–1497
45 Pribilla I., Takagi T., Langosch D., Bormann J. and Betz H.
(1992) The atypical M2 segment of the beta-subunit confirs
picrotoxin resistance to inhibitory glycine receptor channels.
EMBO J. 11: 4305–4311
46 Genderson C. B., Miledi R. and Parker I. (1984) Properties of
human brain glycine receptors expressed in Xenopus oocytes.
Proc. R. Soc. Lond. B 221: 235–244
47 Akagi H. and Miledi R. (1988) Heterogeneity of glycine
receptors and their messenger RNAs in rat brain and spinal
cord. Science 242: 270–273
48 Rajendra S., Lynch J. W., Pierce K. D., French C. R., Barry
P. H. and Schofield P. R. (1995) Mutation of a single amino
acid in the human glycine receptor transforms i-alanine and
taurine from agonist into competitive antagonists. Neuron 14:
169– 175
49 Morrow A. L. and Paul S. M. (1988) Benzodiazepin enhance-
ment of gamma-aminobutyric acid mediated Cl
−
ion flux in
rat brain synaptoneurosomes. J. Neurochem. 50: 302–306
50 Seabra V., Stachlewitz R. F. and Thurman R. G. (1998)
Taurine blunts LPS-induced increases in intracellular calcium
and TNF-h production by Kupffer cells. J. Leukoc. Biol. 64:
615– 621
51 Spittler A., Reissner C. M., Oehler J. R., Gornikiewicz A.,
Gruenberger T., Manhart N. et al. (1999) Immunomodula-
tory effects of glycine on LPS-treated monocytes: reduced
TNF-h production and accelerated IL-10 expression. FASEB
J. 13: 563–571
52 Wheeler, M. D., Seabra, V., and Thurman, R. G. (in press)
Molecular evidence for glycine-gated chloride channel in
Kupffer cells. In: Cells of the Hepatic Sinusoid, Wisse, E.,
Knook, D. L. and Wake, K. (eds), The Kupffer Cell Founda-
tion, Leiden, The Netherlands
53 Miller G. W., Lock E. A. and Schnellmann R. G. (1994)
Strychnine and glycine protect renal proximal tubules from
various nephrotoxicants and act in the late phase of necrotic
cell injury. Toxicol. Appl. Pharmacol. 125: 192–197
54 Iimuro Y., Bradford B. U., Forman D. T. and Thurman R.
G. (1996) Glycine prevents alcohol-induced liver injury by
decreasing alcohol in the stomach. Gastroenterology 110:
1536– 1542
55 Stachlewitz R. F., Seabra V., Bradford B. U., Bradham C. A.,
Rusyn I., Germolec D. et al. (1999) Glycine and uridine
prevent D-galactosamine hepatotoxicity in the rat: role of
Kupffer cells. Hepatology 29: 737–745
56 Rose M. L., Germolec D. R., Arteel G. E., Schoonhoven R.
and Thurman R. G. (1997) Dietary glycine prevents increases
in hepatocyte proliferation caused by the peroxisome prolifer-
ator WY-14,643. Chem. Res. Toxicol. 10: 1198–1204
57 Zhong Z., Arteel G. E., Connor H., Yin M., Frankenberg M.
v., Stachlewitz R. F. et al. (1998) Cyclosporin A increases
hypoxia and free radical production in the rat kidney: preven-
tion by dietary glycine. Am. J. Physiol. 275: F595–F604
58 Zhong Z., Jones S. and Thurman R. G. (1996) Glycine
minimizes reperfusion injury in a low-flow, reflow liver perfu-
sion model in the rat. Am. J. Physiol. 270: G332–G338
59 Schemmer P., Bradford B. U., Rose M. L., Bunzendahl H.,
Raleigh J. A., Lemasters J. J. et al. (1999) Intravenous glycine
improves survival in rat liver transplantation. Am. J. Physiol.
276: G932
60 Schemmer P., Schoonhoven R., Swenberg J. A., Bunzendahl
H. and Thurman R. G. (1998) Gentle in situ liver manipula-
tion during organ harvest decreases survival after rat liver
transplantation: role of Kupffer cells. Transplantation 65:
1015– 1020

M. D. Wheeler et al. Glycine: a new anti-inflammatory immunonutrient856
61 Hardonk M. J., Dijkhuis F. W. J., Hulstaert C. E. and
Koudstaal J. (1992) Heterogeneity of rat liver and spleen
macrophages in gadolinium chloride-induced elimination and
repopulation. J. Leukoc. Biol. 52: 296–302
62 Iimuro Y., Bradford B. U., Forman D. T. and Thurman R.
G. (1995) Role of glycine in preventing alcohol-induced liver
injury. Ann. Clin. Lab. Sci. 25: 357
63 Rao M. S. and Reddy J. K. (1987) Peroxisome proliferation
and hepatocarcinogenesis. Carcinogenesis 8: 631–636
64 Rose M. L., Germolec D. R., Schoonhoven R. and Thurman
R. G. (1997) Kupffer cells are causally responsible for the
mitogenic effect of peroxisome proliferators. Carcinogenesis
18: 1453– 1456
65 Marsman D. S., Cattley R. C., Conway J. G. and Popp J. A.
(1988) Relationship of hepatic peroxisome proliferation and
replicative DNA synthesis to the hepatocarcinogenicity of the
peroxisome proliferators di(2-ethylhexyl)phthalate and [4-
Chloro-6-(2,3-xylidino)-2-pyrimidinylthio]acetic acid (Wy-
14,643) in rats. Cancer Res. 48: 6739–6744
66 Garza-Quintero R., Weinberg J. M., Ortega-Lopez J., Davis
J. A. and Venkatachalam M. A. (1993) Conservation of
structure in ATP-depleted proximal tubules: role of calcium,
polyphosphoinositides, and glycine. Am. J. Physiol. 265:
F605– F623
67 Venkatachalam M. A., Weinberg J. M., Patel Y., Saikumar P.
and Dong Z. (1996) Cytoprotection of kidney epithelial cells
by compounds that target amino acid chloride channels.
Kidney Int. 49: 449–460
.