Lipopolysaccharide impairs regulation under conditions microenvironment Carol
J. SwaIIow
*Department tDivision
of Surgery, of Cell
macrophage cytoplasmic pH simulating the inflammatory
Sergio Grinstein,t and On Toronto
Biology,
General
Research
Hospital,
Institute,
Hospital
Abstract: Within the acidic inflammatory milieu, rophages (m4s) must maintain their cytoplasmic (pHi) within a range conducive to optimal function. was previously concentrations mM) impairs
and
macpH It
shown that metabolism of L-arginine at present in vitro in RPM! medium (1.14 the ability of ms to regulate pHi. How-
ever, concentrations of L-arginine in vivo reportedly range from 100 sM in serum to 50 sM in wounds. To investigate the potential in vivo relevance of this inhibition, m pH1 regulation was examined following incubation with low concentrations of L-arginine that mimic the inflammatory microenvironment, in the presence or absence of lipopolysaccharide (LPS). pH1 regulation was evaluated as the ability of thioglycolate-elicited murine peritoneal m4s to recover from an imposed cytoplasmic acid load. The m4 pH1 was measured using a pHsensitive fluorescent probe. Following incubation for 2 h in the absence of LPS, the pH recovery rate was equivalent in cells incubated with and without L-arginine. Coincubation with LPS, however, resulted in marked inhibition of pH recovery at L-arginine concentrations as low as 12.5 sM. The inhibition was not due to LPS alone, since LPS without L-arginine was not inhibitory. Inhibition of pH1 recovery was observed at LPS concentrations ranging from 10 ng/ml to 10 sg/ml. The L-arginine-dependent inhibition was apparent within 60 mm of cxposure to LPS, in both freshly harvested cells and cells preincubated for 2 h in the absence of L-arginine and then exposed to both L-arginine and LPS. Under conditions mimicking the in vivo setting, LPS-stimulated Larginine metabolism impairs m4 pH1 regulation. Modulation of pH by this mechanism may compromise m4 function within the acidic microenvironment of inflammation. J. Leu.koc. Biol. 52: 395-399; 1992.
Key
Words:
L-arginine
.
vacuolar-type
H
ATPcse
leukocyte
(murine)
D. Rotstein*
Institute for
Sick
of Medical
Science,
Children,
Toronto,
University Oniario
of Toronto,
and
Canada
tam a physiologic pH1 by two mechanisms: (1) the acidic cxtracellular milieu characteristic of inflammatory settings [2, 3] induces cytoplasmic acid loading by passive diffusion of protons and (2) the numerous stimuli present in such environments result in cellular activation, generating increased amounts of metabolic acid. Predictably, it has been found that within the hypoxic, acidic milieu of inflammation, phagocytic cells have impaired ability to migrate to the infected focus and are unable to kill ingested bacteria [4-6]. Studies from our laboratory have defined a novel mechanism by which macrophages (m4s) are able to maintain their cytoplasmic pH (pH1) level close to the physiological range, even in an acidic environment, thereby preserving normal cell function [7-9]. ATP-dependent proton pumps located in the plasma membrane of m4s serve to extrude protons from the cytoplasmic space [7, 10]. These W pumps are virtually quiescent under resting conditions but are activated by cytoplasmic acidification. Our studies have also shown that incubation of m4s with L-arginine resulted in a significant and progressive decline in H pump-mediated pH1 regulation in vitro [11]. This effect appeared to be mediated via the generation of nitric oxide, a short-lived and highly reactive intermediate in the oxidative metabolism of L-arginine. One of the salient differences between the in vivo inflammatory milieu and the typical in vitro incubation conditions is the discrepancy in L-arginine levels between the two. Prepared media typically contain 0.4 mM (minimum essential medium; MEM) to 1.14 mM (RPM!) L-arginine. By contrast, Larginine concentrations in vivo range from 100 sM in serum [12] to 50 jM in wounds [13, 14]. In addition, the presence oflipopolysaccharide (LPS) is a ubiquitous and important feature of the local inflammatory microenvironment that develops in response to bacterial infection. LPS markedly enhances the nitric oxide synthase activity of m4s [15-17] and therefore may influence pH1 regulation within the inflammatory milieu. The purpose of the present studies was to define the significance of the L-arginine-mediated in-
W
hibition of m4 microenvironment
pumps under of inflammation.
conditions
that
mimic
the
INTRODUCTION The cytoplasmic pH (pH1) of mammalian cells is closely regulated by a variety ofion transport systems [1]. This tight control of pH1 is necessary because of the marked pH sensitivity of many intracellular enzymes. Maintenance of pH1 within a range conducive to optimal function is particularly problematic microenvironment their complex
in
the
case within functions
of
phagocytic leukocytes. The local which phagocytic cells must effect may threaten their ability to main-
Abbreviations:
HBSS,
BCECF,
Hanks’
minimum
essential
L-arginine.
pH1,
2’,7’-biscarboxyethyl-5(6)-carboxyfluorescein;
balanced
salt
medium; cytoplasmic
solution; m,
LPS,
lipopolysaccharide;
macrophage;
N-MMA,
pH.
Reprint requests: On D. Rotstein, Department of Surgery, General Hospital, 200 Elizabeth St., EN 9-236, Toronto, Ontario, M5G 2C4. Received November 27, 1991; accepted May 18, 1992.
Journal
of
Leukocyte
MEM,
N-monomethyl-
Biology
Volume
52,
October
1992
Toronto Canada,
395
MATERIALS
Materials
AND
slits, respectively. Calibration was done following addition of the K/H ionophore nigericin (1 sM), using aliquots of concentrated 2-(morpholino) ethanesulfonic acid or Trizma base, as described [19]. Recording of fluorescence commenced immediately upon resuspension of cells in KC1 medium, and the initial rate of pH1 recovery was measured. In order to examine pH1 regulation by H pumps, recov-
METHODS
and solutions
Calciumand magnesium-free Hanks’ (HBSS) and RPM! 1640 select-amine tamed from Gibco. Powdered brewer’s and LPS B (E. coli 0.111:B4) were obtained
balanced salt solution kit medium were obthioglycolate medium from Difco. HEPES
cry
and L-arginine were from Sigma and nigericin from Calbiochem. The acetoxymethyl ester of 2’ ,7’-biscarboxyethyl-5(6)carboxyfluorescein (BCECF) was obtained from Molecular Probes (Eugene, OR). RPM! select-amine kit medium was reconstituted without adding bicarbonate. L-Arginine was omitted or added at the concentration indicated. All other components provided were included in the reconstituted medium. HEPES-RPM! was prepared by titrating HCO#{231}-free RPM! medium with
were found to be Limulus amebocyte
Cape Cod, solubilized uniformly
Cell
Wood’s in H20 green.
isolation
Hole, and
and
Thioglycolate in the dark
medium was at 22#{176}Cuntil
characterization
Incubation
of peritoneal cells identified was consistently 80-85%. by trypan blue exclusion.
as m4s Viability
by was
conditions
Aliquots RPM! cubated gently
(1 ml) of peritoneal cells (107 cells/ml with or without L-arginine, as indicated) in a water bath at 37#{176}Cin room air. agitated every 15 mm throughout the
HEPESwere inCells were incubation
period. mm), plasmic
Following the indicated incubation period a sample of the cell suspension was obtained acid loading and subsequent measurement
(30 to 120 for cytoof pHi.
pH
measurement
and
manipulation
pH1 was measured using the pH-sensitive fluorescent probe BCECF. At the indicated time point, 2 x 106 cells were loaded with dye by a 20-mm incubation in the same medium with 2 jg/ml of the precursor acetoxymethyl ester at 37#{176}C. Cytoplasmic incubation followed NH4-free
acid loading with 40 mM by sedimentation KC1 medium
was accomplished by simultaneous NH4C1 during this 20-mm period, and resuspension in 2 ml in a fluorometer cuvette. The prin-
ciples of the NH4 “prepulse” technique of cytoplasmic acid loading are described elsewhere [1, 18]. The regimen used in these studies consistently yielded acid loading to a pH of 6.4. The NH4 prepulse technique achieved a comparable of acid loading in m4s incubated with L-arginine and with and without LPS. Fluorescence was determined using a 650-40, LS-5, or LS-50 fluorescence spectrometer tation at 495 nm and emission at 525 nm using
Perkin-Elmer with exci5- and 10-nm
396
52,
degree
Journal
of Leukocyte
Biology
Volume
and
without
October
acid
medium.
loading
was
Under
studied
in
Na-
and
these
Results are presented Statistical significance variance followed comparisons.
as means ± 1 SEM of n experiments. was established by one-way analysis by Newman-Keuls multiple intergroup
of
RESULTS Previous experiments showed that the ability of H pumps to regulate pH1 was inhibited in peritoneal m4s incubated for 4 or 6 h with L-arginine (1.14 mM) compared to m4s in. cubated in arginine-free medium [11]. This inhibition was accelerated by coincubation with LPS (10 g/m1): whereas incubation for 2 h with L-arginine alone did not affect m4 pH1 regulation compared to incubation in arginine-free medium, coincubation with LPS and L-arginine for 2 h was associated with a marked impairment of pH1 regulation [11]. This acceleration of the inhibition by LPS was consistent with the known ability of LPS to markedly stimulate the oxi-
Sixto eight-week-old female Swiss Webster mice (Charles River Breeding Laboratories, Wilmington, MA) were injected intraperitoneally with 2 ml of thioglycolate medium. After 4 days, peritoneal cells were harvested by lavage with 10 ml of HBSS. The cells were washed twice with cold HBSS (5#{176}C), and resuspended in HEPES-RPM! at 10 cells/mb. The proportion Wright’s staining >95% as assessed
KC1
Statistics
endotoxin free ( < 0.03 ng/ml) lysate assay from Associates of
MA. stored
cytoplasmic
conditions, ‘-90% of pH1 recovery is mediated by vacuolar-type H pumps, as indicated by sensitivity to nanomolar concentrations of bafilomycin A1, the potent, specific inhibitor of vacuolartype W pumps [10, 20].
20 mM Na-HEPES to pH 7.35 at 37#{176}C.KC1 medium contamed (in mM) 145 KC1, 10 glucose, 2 CaC12, and 10 HEPES, pH 7.35 at 37#{176}C.The osmolarity of KC1 medium was adjusted to 290 ± 5 mOsm with concentrated KC1. All solutions using the
from
HCO3-free
dative metabolism of L-arginine by ms, resulting in enhanced nitric oxide production [15-17]. Speculations as to the in vivo relevance of this LPSinduced, L-arginine-dependent inhibition of pH1 recovery were limited by the discrepancy in L-arginine concentrations between the media used for these in vitro experiments and the in vivo setting. Whereas prepared RPM! medium contains 1.14 mM L-arginine, levels present in vivo reportedly range from -100 sM in serum [4] to 50 jiM in wounds [13, 14]. To investigate the potential in vivo relevance of the L-arginine-dependent effect of incubation mating the reported examined. Peritoneal HEPES-RPMI with
inhibition with low levels of wound and serum cells were incubated various concentrations
the presence or absence of LPS (10 sg/ml). As Figure 1, in the absence of LPS, incubation for arginine concentrations ranging from 12.5 to 500 effect on pH1 recovery. Coincubation with LPS, resulted in marked inhibition of pH1 recovery at concentrations as low as 12.5 sM. The initial rate covery following incubation in medium containing L-arginine without LPS was 0.74 ± 0.02 pH/mm By contrast, incubation at the same L-arginine tion but with LPS was associated of 0.51 ± 0.03 pH/mm (n = 5, P This inhibitory effect was not due bation with LPS in the absence
shown in 2 h with LjiM had no however, L-arginine of pH1 re12.5 sM (n = 5). concentra-
with a pH1 recovery rate < .05 vs. without LPS). to LPS alone, since incuof arginine did not sig-
nificantly inhibit pH1 recovery. Thus, trations simulating the inflammatory LPS-stimulated L-arginine metabolism mediated pH1 regulation.
1992
of pH1 recovery, the L-arginine approxiconcentrations was at 37#{176}Cfor 2 h in of L-arginine in
at
L-arginine concenmicroenvironment, impaired W pump-
harvested medium
freshly C
RPM!
E
bring
0.8
I
them
tivity.
The
sured pH/mm
at
cells were incubated without L-arginine
for and
2 h in without
to a stable baseline level of efficient rate of pH1 recovery of acid-loaded
HEPESLPS
H
pump m4s
to
acmea-
0.
4) > 0 (5 4)
0.6
I 0. C)
0.2
0
60,
and
the
0
100
200
300
LPS
[L-Arginine]
60
P < parent
(MM)
1. Effect oflow L-arginine concentrations and LPS on pH1 recovery of acid-loaded macrophages. Thioglycolate-elicited peritoneal cells were incubated for 2 h at 37#{176}Cin HEPES-RPMI medium (l0 cells/mi) in the presence (filled circles) or absence (open circles) of LPS (10 g/ml) with the indicated concentrations of L-arginine. The pH1 recovery rate of cells acid loaded by the NH4 prepulse method was then measured. P < .05 vs. no LPS, n = 5. “P < .05 vs. no L-arginine, n = 5.
LPS
or sg/ml)
and pH1 recovery 90 mm of incubation pH1 recovery rate
rate
and
±
vs.
No
at
vs.
90
of
was
rate
had
0),
and
0.06
vs.
0.71
further mm.
to
±
0.04
this
incubated
slightly
pH1
30
maximally pH/mm
in
cells
n
recovery
rate
time
with mm reduced
by
respectively,
in
both
following 4). As cxwith LPS
over
became
L-arginine, decline
added
of cells
decreased
time without
L-arginine-containing
stable
recovery
L-arginine
with .05).
the
0.70 ± 0.05 resuspended
was assessed at 37#{176}C(Fig. cells incubated
remained
contrast,
significant mm (0.45
cubated
500
400
0 in Fig. 4) was then pelleted and
arginine-free
L-arginine
By
( not by
fresh
(time were (10
without
both 0.0
point Cells
of cells,
period.
**
0
ck:
medium.
but
*
C)
4)
in
mM)
pected,
U) 0
0
cells/mi)
(1.14 30,
E 0.
(107
groups
0.4
this time (n = 3-4).
=
was
in-
3-4, ap-
Fig.
Pathophysiologically relevant concentrations of L-arginine in conjunction with LPS (10 g/ml) were thus capable of inhibiting m pH1 regulation. The concentration of LPS required to induce inhibition in the presence of such low Larginine levels was next determined. M4s were incubated for 2 h in medium containing 12.5 tM L-arginine with varying concentrations of LPS and then tested for recovery from acid loading (Fig. 2). pH1 recovery was inhibited by incubation with LPS in doses ranging from 10 ng/ml to 10 g/ml. The LPS dose of 10 ng/ml was sufficient to induce maximal inhibition of pH1 recovery (0.45 ± 0.02 vs. 0.67 ± 0.11 pH/mm in cells incubated with 10 ng/ml LPS vs. no LPS, respectively, n = 5, P < .05). Thus, as little as 10 ng/ml LPS inhibited the H pump-mediated pH1 regulation of m4s incubated with a concentration of L-arginine mimicking that of the wound microenvironment. To define the time course of this effect, the pH1 recovery rate of acid-loaded m4s was measured immediately after harvesting from the peritoneal cavity and then after incubation for 30, 60, 90, and 120 mm in the presence or absence of L-arginine, with or without LPS (Fig. 3). The rate of pH1 recovery was significantly lower in freshly harvested cells (0.45 ± 0.04 pH/mm, n = 5) than in cells incubated for 2 h in the absence of arginine (0.67 ± 0.03 and 0.62 ± 0.06 pH/mm, without and with LPS, respectively, n = 5, P < .05 vs. time 0). In addition, freshly isolated cells displayed slower pH1 recovery than did cells incubated for 2 h with Larginine but without LPS (0.62 ± 0.02 pH/mm in the latter, n = 5, P < .05 vs. time 0). Only cells incubated with both L-arginine and LPS displayed a persistently low pH1 recovcry rate over the entire incubation period (Fig. 3). While the pH1 1 h of incubation
recovery (Fig. 3),
rates that
rose in of freshly
all other harvested
groups m4s
DISCUSSION Recent studies from our laboratory have defined the presence of a proton pump in the plasma membrane of peritoneal m4s [7, 8, 10]. The ability of this pump to maintain pH1 close to the physiological range during exposure to an acidic microenvironment has been shown to be crucial to the normal bactericidal activity of the cell [9]. In previous studies, metabolism of the amino acid L-arginine with generation of nitric oxide negatively influenced the activity of this H pump in vitro [11]. Those experiments were performed using an L-arginine concentration (1.14 mM) which greatly
I
exceeded
that
measured
Swallow
vivo
50-200
(-
*
c
Fig. phages
M)
0.8
0.001
by in-
cubated with both LPS and L-arginine remained relatively low. This suggested that LPS rapidly induced nitric oxide production sufficient to inhibit H pump-mediated pH1 regulation. However, the low pH1 recovery rates observed in all four groups at time 0 made it difficult to determine the time course of the LPS effect. To investigate this question,
in
2. Effect
of varying
incubated
with
concentrations 12.5
sM
*
.o
10.0
recovery
of macro-
0.01
0.1
[LPS]
(g/ml)
of LPS L-arginine.
on pH1
*
Thioglycolate-elicited
pen-
toneal cells were incubated for 2 h at 37#{176}C in HEPES-RPMI medium (107 cells/ml) with 12.5 tM L-arginine, without or with various concentrations of LPS. The pH recovery rate of cells acid loaded by the NH4 prepulse method was then measured. P < .05 vs. no LPS and vs. 1 ng/ml LPS, n = 5.
ci aL
Macrophage
cytoplasmic
pH
and
inflammation
397
[12-14],
raising
questions
as
to the
in vivo
relevance
of this
phenomenon. The present studies show that an L-arginine concentration as low as 12.5 sM is sufficient to inhibit W pump activity, provided that a small amount of endotoxin (as low as 10 ng/ml) is also present. This inhibition was apparent within 1 h of exposure to LPS, a finding in keeping with the recently demonstrated ability of LPS to induce expression of mRNA for nitric oxide synthase within a short time frame [21, and RB. Lorsbach et al., unpublished observations]. These studies suggest that in vivo, L-arginine-mediated impairment of pH1 regulation occurs only in the presence of infection. However, it should be noted that other inflammatory
cytokines
such
as interferon-’y,
necrosis factor independently late the oxidative metabolism tion, interferon--y is a potent nitric oxide synthase activity
interleukin-1,
and
tumor
22]. The presence of these and other cytokines within the inflammatory milieu could conceivably modulate L-arginine metabolism and perhaps in this manner affect m pH1 regulation. In keeping with this notion, we previously showed that thioglycolate-elicited m4s demonstrated Larginine-dependent inhibition of pH1 recovery whereas resident m4s did not [11].
the absence of LPS this slow recovery Whether this effect nitric oxide synthase
C
was in
for 2 h. The mechanism responsible for was not elucidated in these studies. was due to in vivo stimulation of the pathway in the inflammatory micro-
0.8
0.
4,
0.6
> 0
0 a,
0.4 0
#
E
(I)
0 0 > C)
0
a, I.-
I
0.4 E 0 0.
0.2
0
-
>‘
0 0
a,
0.0-
0
I
0
I
30
I
60
Incubation
time
90 (mm)
Fig. 4. Time course of inhibition of pH recovery by LPS and L-arginine. Thioglycolate-elicited peritoneal cells were preincubated for 2 h at 37#{176}Cin arginine-free HEPES-RPMI medium (10 cells/mI), without 125. Cells were then circles) (10 g/ml) 60, and loaded and vs.
pelleted
and
resuspended
at time
0 in fresh
arginine-free
0
30
60
Incubation
90 time
120
L-arginine-containing (1.14 mM) (filled cirdes) medium. LPS was added to both groups of cells. Following incubation for 30, 90 mm under these conditions, the pH, recovery rate of cells acid by the NH4 prepulse method was measured. P < .05 vs. time 0 arginine-free medium, n - 3-4.
of the
thioglycolate-treated
peritoneal
Biology
Volume
(mm)
52,
October
milieu may alter not only nitric oxide 16, 17, 22], but also L-arginase activity
study.
Fig. 3. pH regulation in freshly harvested macrophages. The pH1 recovery rate of freshly harvested thioglycolate-elicited peritoneal cells was measured either immediately following harvesting (time 0) or after 30, 60, 90, or 120 mm of incubation at 37#{176}Cin HEPES-RPMI medium (107 cells/mi) with (triangles) or without (cirdes) 1.14 mM L-arginine, in the presence (filled symbols) or absence (open symbols) of LPS (10 ig/ml). The pH1 recovery rate of cells acid loaded by the NH4 prepulse method was then measured. Where absent, the error bar was smaller than the symbol. P < .05 vs. time 0, n - 3-4. #P < .05 vs. all other groups at I 90 mm, n - 3-4. P < .05 vs. arginine-free medium, without LPS, n - 3-4.
of Leukocyte
(open
or
lines of evidence arginase did not
0.0
Journal
* *
0
tory [15,
0
398
> 0
cavity
(2) the arginase pathway, by which the guanidino incorporated into urea, with the other product ornithine. Exposure to factors present within the
0.2
‘4-
a) 0 c1
0.6
a,
or
medium (Fig. 3). Ms metabolize L-arginine via two major pathways: (1) the nitric oxide synthase pathway, which results in oxidation of the guanidino nitrogen, liberating nitric oxide and yielding nitrite, nitrate, and citrulline as stable end products; and
5..
0.
0.
to some other in vivo event, such as the trauma of peritoneal lavage, requires further investigation. The former explanation would appear to be unlikely, since one would expect cells incubating in L-arginine following their harvest to continue to metabolize L-arginine, resulting in continued impairment of pH1 recovery. However, m4 pH1 recovery rose to levels equivalent to that of cells incubating in arginine-free
*
I
I
I
environment
E
a,
0.8
E
U)
and/or synergistically stimuof L-arginine [22]. In addipriming agent for enhanced in response to LPS [15, 16, 17,
The present studies demonstrate that pH1 recovery significantly slower in freshly harvested thioglycolate-elicited ms than in cells incubated with or without L-arginine
C
1992
mediated showed cubated
First,
neither
suggest contribute
that
urea
nor
to
metabolism the findings L-ornithine
nitrogen being inflamma-
synthase [23, 24].
is L-
activity Several
of L-arginine reported in affects
pH1 regulation [11]. Second, previous that addition of exogenous L-arginase with LPS reversed the L-arginine-dependent
H
by this
pump-
experiments to m4s
inin-
hibition of pH1 recovery from acid loading [11]. This would not be expected if metabolism by endogenously produced Larginase were significantly limiting the supply of L-arginine for the nitric oxide synthase pathway. Finally, other investigators have reported that neither peritoneal release L-arginase in vitro [13]. The oxidative metabolism of L-arginine the most part, been studied in vitro. In this
nor
wound
by ms setting,
m4s
has, for nitrogen
oxides derived m toxicity
for
from a
27], as well documented
as for the
L-arginine variety
have been intracellular
of
tumor cells significance
vironment:
found to mediate parasites [25-
[12]. Recent of oxidative
10.
studies have L-arginine
H j
metabolism in vivo. Kilbourn et al. [30] found that treatment with the competitive inhibitor of L-arginine metabolism, N-monomethyl-L-arginine (N-MMA), ameliorated shock induced by tumor necrosis factor in dogs. Evidence that oxidative metabolism of L-arginine by ms plays a significant role in vivo comes from experiments reported by Liew et al. [31], who injected N-MMA into the mouse footpad. This inhibited the in vivo killing of Leishmania by ms, indicating that nitric oxide derived from L-arginine mcdiated the antiparasitic activity of m4s in vivo. By demonstrating inhibition of H pump-mediated m4 pH1 regulation by LPS in combination with levels of L-arginine mimicking that of the inflammatory milieu, these experiments suggest that nitric oxide may impair m4 pH, regulation in vivo. Such an inhibition of H pump activity would be particularly deleterious within the acidic extracellular environment typically prevailing in an abscess or wound. Under these conditions, H pump-mediated pH1 regulation is critical to efficient m4 function [9]. Consequently, this represents a potential mechanism by which m4 function may be modulated in vivo. We are currently investigating the impact of nitrogen oxide generation in vivo on m4 pH1 regulation using a recently described technique to inhibit nitric oxide synthase
activity
in
vivo
11.
by
nine
Medical
Research
arginine Marletta,
The
penitoneal
environment
monomicrobial
and
Surg. 213, 253. 3. Silver, IA. (1978) Tissue Exp. Med Biol. 94, 769. 4. Mandell,
G.L.
(1974)
P#{176}2 changes
O.D.,
Nasmith,
PO2
Immun.
Gninstein,
T.L.
17.
the
effect
and
pH.
and
S. (1987)
The
7. Swallow, Cj., Gninstein, regulation in macrophages dicycbohexylcarbodiimide-sensitive
ment of a plasma 19558. 8.
Swallow,
Cj.,
cytoplasmic 46, 588. 9. Swallow,
Modulation
membrane
Rotstein, by an
O.D. (1988) ATP-dependent mechanism:
23.
proton
pump.
j
Biol.
Chem.
recovery
O.D.,
Gninstein,
S. (1989)
in acid-loaded
macrophages.
Grinstein,
of the
S.,
macrophage
Sudsbury,
R.A.,
respiratory
Rotstein,
burst
O.D.
Swallow
R.,
Leaf,
CD.,
Wishnok,
J.S.
and
nitrate:
nitrate
bio-
S.,
S., Rotstein,
Ada
and
1022,
Furuya, W. Na/H antiport
O.D.
activated
(1990)
penitoneal
Regulation
of
macrophages.
203. (1986) Characterization of human neutrophils.
of
the
Am. j
amilo-
Physiol.
(M4):
Cloning
Ding, nitrogen mouse Albina,
A.H.,
and
inducibility.
FASEBJ
6, A1973.
Nathan, CF., Stuehr, Dj. (1988) Release of reactive intermediates and reactive oxygen intermediates from peritoneal macrophages. j Immunol. 141, 2407. J.E., Mills, CD., Henry, W.L., Jr., Caldwell, M.D. (1990)
expression wounds.
of different Immunol.
j
pathways of arginine 144, 3877.
metabolism
Granger,
L-arginine.
j
DL.,
Immunol.
Hibbs,
Metabolic fate of macrophages.
J.B.,
Green, vated gotes
Sj., Meltzer, MS., macrophages destroy by an L-arginine-dependent
144,
278.
28.
James, S.L., schistosomula
production
144, Jr.,
2725.
Perfect,
JR.,
Durack,
of L-arginine in relation to microbiostatic J. Clin. Invest. 85, 264.
27.
(1990)
capacity
Hibbs, J.B., Jr., Nacy, CA. (1990) Actiintracellular Leishmania major amastikilling mechanism. j Immunol.
Glaven, J. (1989) Macrophage of Schistosoma mansoni involves
of reactive
D.T.
nitrogen
intermediates.
cytotoxicity against arginine-dependent j Immunol. 143,
4208. Malkin, R., Flescher, E., Lengy, J., Keisani, Y. (1987) On the interactions between macrophages and developmental stages of Schisiosoma mansoni: the cytotoxic mechanisms involved in macrophagemediated killing of schistosomula in vitro. Immunobiology 176, 63. Kilbourn, R.G., Gross, S.S., Jubran, A Adams, J., Griffith, D.W.,
29.
30.
Levi,
R.,
Lodato,
RE
(1990)
L-N”-Methyl-argsnine
inhibits
tumor necrosis factor-induced hypotension: implications for the involvement of nitric oxide. Proc. Nail. Acad. Sci. USA 87, 3629. Liew, F.Y., Millott, S., Parkinson, C., Palmer, R.M.J., Moncada, S. (1990) Macrophage killing ofLcishmania parasite in vivo is mediated by nitric oxide from L-arginine. j Immunol. 144, 4794. Granger, DL., Hibbs, J.B., Jr., Broadnax, L.M. (1991) Urinary nitrate excretion in relation to munine macrophage activation.
(1990)
en-
Med.
26.
of Res.
by an acidic
(1989)
Exp.
25.
32. Cj.,
CD. j
Shearer, J.D., Mills, CD., Caldwell, M.D. (1992) Differential regulation of macrophage arginine metabolism. FASEBJ 6, A1719. Adams, LB., Hibbs, J.B., Jr., Taintor, R.R., Krahenbuhl, J.L. (1990) Microbiostatic effect of munine activated macrophages for Toxoplasma gondii: role for synthesis of inorganic nitrogen oxides
263,
Mechanisms j Surg.
macro-
C283.
Temporal in healing
Bacieroides
Cytoplasmic pH and N,N’possible involve-
pH
peritoneal
Bowman, Ej., Siebers, A., Altendorf, K. (1988) Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc. Nail. Acad. Sci. USA 85, 7972. Xie, Q -W., Cho, Hj., Calaycay, J., Mumford, R., Ding, A., Troso, T., Nathan, C. (1992) Nitric oxide synthase (NOS) of macrophages
31.
Rotstein,
pH
S.,
in
W.L., Jr., Mills, by L-arginine.
Iyengar,
in resident
Biophys.
Grinstein, ride-sensitive
22.
9, 337.
by-product, succinic acid, inhibits neutrophil respiratory burst by reducing intracellular pH. Infeci. Immun. 55, 864. 6. Rotstein, O.D., Fiegel, V.D., Simmons, R.L., Knighton, DR. (1988) The deleterious effect of reduced pH and hypoxia on neutrophil migration in vitro. j Surg. Res. 45, 298.
P.S.,
Gninstein, pH
Biochim. 19.
Adv. anaerobic
(1991)
wounds.
Yoon,
Cj.,
cytoplasmic
of Ann.
ATPases
O.D.
cytoplasmic
CD.,
in
M.A.,
18. Swallow,
(1991)
inflammation.
of aerobic
Infect.
P.E.,
on
in acute
activity
neutrophils.
Pruett,
infection:
bacteria
Bactericidal
polymorphonuclear 5. Rotstein,
during
polymicrobial
Rev. 61, 296.
Rotstein,
impairs
mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proc. NaiL Acad. Sci. USA 82, 7738. Stuehr, Dj., Gross, S.S., Sakuma, I., Levi, I.R., Nathan, C. (1989) Activated murine macrophages secrete a metabolite of arginine with the bioactivity of endothelium-denived relaxing factor and the chemical reactivity of nitric oxide. j Exp. Med. 169, lOll.
from Physiol.
pH. RB.,
macrophages.
synthesis:
REFERENCES (1981) Intracellular M.D., Adams,
A vacuolar-type
Macrophage oxidation of L-arginine to nitrite nitric oxide is an intermediate. Biochemisiry 27, 8706. 16. Stuehr, D.J., Marletta, MA. (1985) Mammalian
24.
1. Roos, A., Boron, W.F. 2. Sawyer, R.G., Spengler,
R.A.,
L-arginine
H 1009.
(1990)
in munine
(1988)
Council
of Canada. C.J.S. is recipient of an Ethicon-Society of University Surgeons Surgical Research Fellowship and a Medical Research Council of Canada Fellowship. S.G. is recipient of an International Research Scholars Award from the Howard Hughes Medical Institute.
O.D.
pH
by proton
Barbul, A., Caldwell, M.D. (1988) ArgiAm. j Physiol. 17, E459-E467. GA., Gyure, L., Cifuentes, L. (1979) Microenvironmental depletion by macrophages in vivo. Br j Cancer 39, 613.
14. Curnie, 15.
Mills,
pH regulation
Sudsbury,
from
by vacuolar-type Exp. Mcd. 174,
metabolism
250,
the
S.,
derived
169, 1021. 13. Albina, J.E.,
[32].
supported
Rotstein,
Albina, J.E., Caidwell, M.D., Henry, Regulation of macrophage functions
21.
was
S.,
Gninstein,
oxide
regulation phages.j
ACKNOWLEDGMENT work
Cj.,
Nitric
12.
role of cytoplasmic 108, 363.
ATPase regulates cytoplasmic Biol. Chem. 265, 7645.
Swallow,
20.
This
the critical
pumps. Surgery Cj., Gninstein,
extrusion Swallow,
j
ci al.
Immunol.
146,
Macrophage
1294.
cytoplasmic
pH
and
inflammation
399
J. SwaIIow
*Department tDivision
of Surgery, of Cell
macrophage cytoplasmic pH simulating the inflammatory
Sergio Grinstein,t and On Toronto
Biology,
General
Research
Hospital,
Institute,
Hospital
Abstract: Within the acidic inflammatory milieu, rophages (m4s) must maintain their cytoplasmic (pHi) within a range conducive to optimal function. was previously concentrations mM) impairs
and
macpH It
shown that metabolism of L-arginine at present in vitro in RPM! medium (1.14 the ability of ms to regulate pHi. How-
ever, concentrations of L-arginine in vivo reportedly range from 100 sM in serum to 50 sM in wounds. To investigate the potential in vivo relevance of this inhibition, m pH1 regulation was examined following incubation with low concentrations of L-arginine that mimic the inflammatory microenvironment, in the presence or absence of lipopolysaccharide (LPS). pH1 regulation was evaluated as the ability of thioglycolate-elicited murine peritoneal m4s to recover from an imposed cytoplasmic acid load. The m4 pH1 was measured using a pHsensitive fluorescent probe. Following incubation for 2 h in the absence of LPS, the pH recovery rate was equivalent in cells incubated with and without L-arginine. Coincubation with LPS, however, resulted in marked inhibition of pH recovery at L-arginine concentrations as low as 12.5 sM. The inhibition was not due to LPS alone, since LPS without L-arginine was not inhibitory. Inhibition of pH1 recovery was observed at LPS concentrations ranging from 10 ng/ml to 10 sg/ml. The L-arginine-dependent inhibition was apparent within 60 mm of cxposure to LPS, in both freshly harvested cells and cells preincubated for 2 h in the absence of L-arginine and then exposed to both L-arginine and LPS. Under conditions mimicking the in vivo setting, LPS-stimulated Larginine metabolism impairs m4 pH1 regulation. Modulation of pH by this mechanism may compromise m4 function within the acidic microenvironment of inflammation. J. Leu.koc. Biol. 52: 395-399; 1992.
Key
Words:
L-arginine
.
vacuolar-type
H
ATPcse
leukocyte
(murine)
D. Rotstein*
Institute for
Sick
of Medical
Science,
Children,
Toronto,
University Oniario
of Toronto,
and
Canada
tam a physiologic pH1 by two mechanisms: (1) the acidic cxtracellular milieu characteristic of inflammatory settings [2, 3] induces cytoplasmic acid loading by passive diffusion of protons and (2) the numerous stimuli present in such environments result in cellular activation, generating increased amounts of metabolic acid. Predictably, it has been found that within the hypoxic, acidic milieu of inflammation, phagocytic cells have impaired ability to migrate to the infected focus and are unable to kill ingested bacteria [4-6]. Studies from our laboratory have defined a novel mechanism by which macrophages (m4s) are able to maintain their cytoplasmic pH (pH1) level close to the physiological range, even in an acidic environment, thereby preserving normal cell function [7-9]. ATP-dependent proton pumps located in the plasma membrane of m4s serve to extrude protons from the cytoplasmic space [7, 10]. These W pumps are virtually quiescent under resting conditions but are activated by cytoplasmic acidification. Our studies have also shown that incubation of m4s with L-arginine resulted in a significant and progressive decline in H pump-mediated pH1 regulation in vitro [11]. This effect appeared to be mediated via the generation of nitric oxide, a short-lived and highly reactive intermediate in the oxidative metabolism of L-arginine. One of the salient differences between the in vivo inflammatory milieu and the typical in vitro incubation conditions is the discrepancy in L-arginine levels between the two. Prepared media typically contain 0.4 mM (minimum essential medium; MEM) to 1.14 mM (RPM!) L-arginine. By contrast, Larginine concentrations in vivo range from 100 sM in serum [12] to 50 jM in wounds [13, 14]. In addition, the presence oflipopolysaccharide (LPS) is a ubiquitous and important feature of the local inflammatory microenvironment that develops in response to bacterial infection. LPS markedly enhances the nitric oxide synthase activity of m4s [15-17] and therefore may influence pH1 regulation within the inflammatory milieu. The purpose of the present studies was to define the significance of the L-arginine-mediated in-
W
hibition of m4 microenvironment
pumps under of inflammation.
conditions
that
mimic
the
INTRODUCTION The cytoplasmic pH (pH1) of mammalian cells is closely regulated by a variety ofion transport systems [1]. This tight control of pH1 is necessary because of the marked pH sensitivity of many intracellular enzymes. Maintenance of pH1 within a range conducive to optimal function is particularly problematic microenvironment their complex
in
the
case within functions
of
phagocytic leukocytes. The local which phagocytic cells must effect may threaten their ability to main-
Abbreviations:
HBSS,
BCECF,
Hanks’
minimum
essential
L-arginine.
pH1,
2’,7’-biscarboxyethyl-5(6)-carboxyfluorescein;
balanced
salt
medium; cytoplasmic
solution; m,
LPS,
lipopolysaccharide;
macrophage;
N-MMA,
pH.
Reprint requests: On D. Rotstein, Department of Surgery, General Hospital, 200 Elizabeth St., EN 9-236, Toronto, Ontario, M5G 2C4. Received November 27, 1991; accepted May 18, 1992.
Journal
of
Leukocyte
MEM,
N-monomethyl-
Biology
Volume
52,
October
1992
Toronto Canada,
395
MATERIALS
Materials
AND
slits, respectively. Calibration was done following addition of the K/H ionophore nigericin (1 sM), using aliquots of concentrated 2-(morpholino) ethanesulfonic acid or Trizma base, as described [19]. Recording of fluorescence commenced immediately upon resuspension of cells in KC1 medium, and the initial rate of pH1 recovery was measured. In order to examine pH1 regulation by H pumps, recov-
METHODS
and solutions
Calciumand magnesium-free Hanks’ (HBSS) and RPM! 1640 select-amine tamed from Gibco. Powdered brewer’s and LPS B (E. coli 0.111:B4) were obtained
balanced salt solution kit medium were obthioglycolate medium from Difco. HEPES
cry
and L-arginine were from Sigma and nigericin from Calbiochem. The acetoxymethyl ester of 2’ ,7’-biscarboxyethyl-5(6)carboxyfluorescein (BCECF) was obtained from Molecular Probes (Eugene, OR). RPM! select-amine kit medium was reconstituted without adding bicarbonate. L-Arginine was omitted or added at the concentration indicated. All other components provided were included in the reconstituted medium. HEPES-RPM! was prepared by titrating HCO#{231}-free RPM! medium with
were found to be Limulus amebocyte
Cape Cod, solubilized uniformly
Cell
Wood’s in H20 green.
isolation
Hole, and
and
Thioglycolate in the dark
medium was at 22#{176}Cuntil
characterization
Incubation
of peritoneal cells identified was consistently 80-85%. by trypan blue exclusion.
as m4s Viability
by was
conditions
Aliquots RPM! cubated gently
(1 ml) of peritoneal cells (107 cells/ml with or without L-arginine, as indicated) in a water bath at 37#{176}Cin room air. agitated every 15 mm throughout the
HEPESwere inCells were incubation
period. mm), plasmic
Following the indicated incubation period a sample of the cell suspension was obtained acid loading and subsequent measurement
(30 to 120 for cytoof pHi.
pH
measurement
and
manipulation
pH1 was measured using the pH-sensitive fluorescent probe BCECF. At the indicated time point, 2 x 106 cells were loaded with dye by a 20-mm incubation in the same medium with 2 jg/ml of the precursor acetoxymethyl ester at 37#{176}C. Cytoplasmic incubation followed NH4-free
acid loading with 40 mM by sedimentation KC1 medium
was accomplished by simultaneous NH4C1 during this 20-mm period, and resuspension in 2 ml in a fluorometer cuvette. The prin-
ciples of the NH4 “prepulse” technique of cytoplasmic acid loading are described elsewhere [1, 18]. The regimen used in these studies consistently yielded acid loading to a pH of 6.4. The NH4 prepulse technique achieved a comparable of acid loading in m4s incubated with L-arginine and with and without LPS. Fluorescence was determined using a 650-40, LS-5, or LS-50 fluorescence spectrometer tation at 495 nm and emission at 525 nm using
Perkin-Elmer with exci5- and 10-nm
396
52,
degree
Journal
of Leukocyte
Biology
Volume
and
without
October
acid
medium.
loading
was
Under
studied
in
Na-
and
these
Results are presented Statistical significance variance followed comparisons.
as means ± 1 SEM of n experiments. was established by one-way analysis by Newman-Keuls multiple intergroup
of
RESULTS Previous experiments showed that the ability of H pumps to regulate pH1 was inhibited in peritoneal m4s incubated for 4 or 6 h with L-arginine (1.14 mM) compared to m4s in. cubated in arginine-free medium [11]. This inhibition was accelerated by coincubation with LPS (10 g/m1): whereas incubation for 2 h with L-arginine alone did not affect m4 pH1 regulation compared to incubation in arginine-free medium, coincubation with LPS and L-arginine for 2 h was associated with a marked impairment of pH1 regulation [11]. This acceleration of the inhibition by LPS was consistent with the known ability of LPS to markedly stimulate the oxi-
Sixto eight-week-old female Swiss Webster mice (Charles River Breeding Laboratories, Wilmington, MA) were injected intraperitoneally with 2 ml of thioglycolate medium. After 4 days, peritoneal cells were harvested by lavage with 10 ml of HBSS. The cells were washed twice with cold HBSS (5#{176}C), and resuspended in HEPES-RPM! at 10 cells/mb. The proportion Wright’s staining >95% as assessed
KC1
Statistics
endotoxin free ( < 0.03 ng/ml) lysate assay from Associates of
MA. stored
cytoplasmic
conditions, ‘-90% of pH1 recovery is mediated by vacuolar-type H pumps, as indicated by sensitivity to nanomolar concentrations of bafilomycin A1, the potent, specific inhibitor of vacuolartype W pumps [10, 20].
20 mM Na-HEPES to pH 7.35 at 37#{176}C.KC1 medium contamed (in mM) 145 KC1, 10 glucose, 2 CaC12, and 10 HEPES, pH 7.35 at 37#{176}C.The osmolarity of KC1 medium was adjusted to 290 ± 5 mOsm with concentrated KC1. All solutions using the
from
HCO3-free
dative metabolism of L-arginine by ms, resulting in enhanced nitric oxide production [15-17]. Speculations as to the in vivo relevance of this LPSinduced, L-arginine-dependent inhibition of pH1 recovery were limited by the discrepancy in L-arginine concentrations between the media used for these in vitro experiments and the in vivo setting. Whereas prepared RPM! medium contains 1.14 mM L-arginine, levels present in vivo reportedly range from -100 sM in serum [4] to 50 jiM in wounds [13, 14]. To investigate the potential in vivo relevance of the L-arginine-dependent effect of incubation mating the reported examined. Peritoneal HEPES-RPMI with
inhibition with low levels of wound and serum cells were incubated various concentrations
the presence or absence of LPS (10 sg/ml). As Figure 1, in the absence of LPS, incubation for arginine concentrations ranging from 12.5 to 500 effect on pH1 recovery. Coincubation with LPS, resulted in marked inhibition of pH1 recovery at concentrations as low as 12.5 sM. The initial rate covery following incubation in medium containing L-arginine without LPS was 0.74 ± 0.02 pH/mm By contrast, incubation at the same L-arginine tion but with LPS was associated of 0.51 ± 0.03 pH/mm (n = 5, P This inhibitory effect was not due bation with LPS in the absence
shown in 2 h with LjiM had no however, L-arginine of pH1 re12.5 sM (n = 5). concentra-
with a pH1 recovery rate < .05 vs. without LPS). to LPS alone, since incuof arginine did not sig-
nificantly inhibit pH1 recovery. Thus, trations simulating the inflammatory LPS-stimulated L-arginine metabolism mediated pH1 regulation.
1992
of pH1 recovery, the L-arginine approxiconcentrations was at 37#{176}Cfor 2 h in of L-arginine in
at
L-arginine concenmicroenvironment, impaired W pump-
harvested medium
freshly C
RPM!
E
bring
0.8
I
them
tivity.
The
sured pH/mm
at
cells were incubated without L-arginine
for and
2 h in without
to a stable baseline level of efficient rate of pH1 recovery of acid-loaded
HEPESLPS
H
pump m4s
to
acmea-
0.
4) > 0 (5 4)
0.6
I 0. C)
0.2
0
60,
and
the
0
100
200
300
LPS
[L-Arginine]
60
P < parent
(MM)
1. Effect oflow L-arginine concentrations and LPS on pH1 recovery of acid-loaded macrophages. Thioglycolate-elicited peritoneal cells were incubated for 2 h at 37#{176}Cin HEPES-RPMI medium (l0 cells/mi) in the presence (filled circles) or absence (open circles) of LPS (10 g/ml) with the indicated concentrations of L-arginine. The pH1 recovery rate of cells acid loaded by the NH4 prepulse method was then measured. P < .05 vs. no LPS, n = 5. “P < .05 vs. no L-arginine, n = 5.
LPS
or sg/ml)
and pH1 recovery 90 mm of incubation pH1 recovery rate
rate
and
±
vs.
No
at
vs.
90
of
was
rate
had
0),
and
0.06
vs.
0.71
further mm.
to
±
0.04
this
incubated
slightly
pH1
30
maximally pH/mm
in
cells
n
recovery
rate
time
with mm reduced
by
respectively,
in
both
following 4). As cxwith LPS
over
became
L-arginine, decline
added
of cells
decreased
time without
L-arginine-containing
stable
recovery
L-arginine
with .05).
the
0.70 ± 0.05 resuspended
was assessed at 37#{176}C(Fig. cells incubated
remained
contrast,
significant mm (0.45
cubated
500
400
0 in Fig. 4) was then pelleted and
arginine-free
L-arginine
By
( not by
fresh
(time were (10
without
both 0.0
point Cells
of cells,
period.
**
0
ck:
medium.
but
*
C)
4)
in
mM)
pected,
U) 0
0
cells/mi)
(1.14 30,
E 0.
(107
groups
0.4
this time (n = 3-4).
=
was
in-
3-4, ap-
Fig.
Pathophysiologically relevant concentrations of L-arginine in conjunction with LPS (10 g/ml) were thus capable of inhibiting m pH1 regulation. The concentration of LPS required to induce inhibition in the presence of such low Larginine levels was next determined. M4s were incubated for 2 h in medium containing 12.5 tM L-arginine with varying concentrations of LPS and then tested for recovery from acid loading (Fig. 2). pH1 recovery was inhibited by incubation with LPS in doses ranging from 10 ng/ml to 10 g/ml. The LPS dose of 10 ng/ml was sufficient to induce maximal inhibition of pH1 recovery (0.45 ± 0.02 vs. 0.67 ± 0.11 pH/mm in cells incubated with 10 ng/ml LPS vs. no LPS, respectively, n = 5, P < .05). Thus, as little as 10 ng/ml LPS inhibited the H pump-mediated pH1 regulation of m4s incubated with a concentration of L-arginine mimicking that of the wound microenvironment. To define the time course of this effect, the pH1 recovery rate of acid-loaded m4s was measured immediately after harvesting from the peritoneal cavity and then after incubation for 30, 60, 90, and 120 mm in the presence or absence of L-arginine, with or without LPS (Fig. 3). The rate of pH1 recovery was significantly lower in freshly harvested cells (0.45 ± 0.04 pH/mm, n = 5) than in cells incubated for 2 h in the absence of arginine (0.67 ± 0.03 and 0.62 ± 0.06 pH/mm, without and with LPS, respectively, n = 5, P < .05 vs. time 0). In addition, freshly isolated cells displayed slower pH1 recovery than did cells incubated for 2 h with Larginine but without LPS (0.62 ± 0.02 pH/mm in the latter, n = 5, P < .05 vs. time 0). Only cells incubated with both L-arginine and LPS displayed a persistently low pH1 recovcry rate over the entire incubation period (Fig. 3). While the pH1 1 h of incubation
recovery (Fig. 3),
rates that
rose in of freshly
all other harvested
groups m4s
DISCUSSION Recent studies from our laboratory have defined the presence of a proton pump in the plasma membrane of peritoneal m4s [7, 8, 10]. The ability of this pump to maintain pH1 close to the physiological range during exposure to an acidic microenvironment has been shown to be crucial to the normal bactericidal activity of the cell [9]. In previous studies, metabolism of the amino acid L-arginine with generation of nitric oxide negatively influenced the activity of this H pump in vitro [11]. Those experiments were performed using an L-arginine concentration (1.14 mM) which greatly
I
exceeded
that
measured
Swallow
vivo
50-200
(-
*
c
Fig. phages
M)
0.8
0.001
by in-
cubated with both LPS and L-arginine remained relatively low. This suggested that LPS rapidly induced nitric oxide production sufficient to inhibit H pump-mediated pH1 regulation. However, the low pH1 recovery rates observed in all four groups at time 0 made it difficult to determine the time course of the LPS effect. To investigate this question,
in
2. Effect
of varying
incubated
with
concentrations 12.5
sM
*
.o
10.0
recovery
of macro-
0.01
0.1
[LPS]
(g/ml)
of LPS L-arginine.
on pH1
*
Thioglycolate-elicited
pen-
toneal cells were incubated for 2 h at 37#{176}C in HEPES-RPMI medium (107 cells/ml) with 12.5 tM L-arginine, without or with various concentrations of LPS. The pH recovery rate of cells acid loaded by the NH4 prepulse method was then measured. P < .05 vs. no LPS and vs. 1 ng/ml LPS, n = 5.
ci aL
Macrophage
cytoplasmic
pH
and
inflammation
397
[12-14],
raising
questions
as
to the
in vivo
relevance
of this
phenomenon. The present studies show that an L-arginine concentration as low as 12.5 sM is sufficient to inhibit W pump activity, provided that a small amount of endotoxin (as low as 10 ng/ml) is also present. This inhibition was apparent within 1 h of exposure to LPS, a finding in keeping with the recently demonstrated ability of LPS to induce expression of mRNA for nitric oxide synthase within a short time frame [21, and RB. Lorsbach et al., unpublished observations]. These studies suggest that in vivo, L-arginine-mediated impairment of pH1 regulation occurs only in the presence of infection. However, it should be noted that other inflammatory
cytokines
such
as interferon-’y,
necrosis factor independently late the oxidative metabolism tion, interferon--y is a potent nitric oxide synthase activity
interleukin-1,
and
tumor
22]. The presence of these and other cytokines within the inflammatory milieu could conceivably modulate L-arginine metabolism and perhaps in this manner affect m pH1 regulation. In keeping with this notion, we previously showed that thioglycolate-elicited m4s demonstrated Larginine-dependent inhibition of pH1 recovery whereas resident m4s did not [11].
the absence of LPS this slow recovery Whether this effect nitric oxide synthase
C
was in
for 2 h. The mechanism responsible for was not elucidated in these studies. was due to in vivo stimulation of the pathway in the inflammatory micro-
0.8
0.
4,
0.6
> 0
0 a,
0.4 0
#
E
(I)
0 0 > C)
0
a, I.-
I
0.4 E 0 0.
0.2
0
-
>‘
0 0
a,
0.0-
0
I
0
I
30
I
60
Incubation
time
90 (mm)
Fig. 4. Time course of inhibition of pH recovery by LPS and L-arginine. Thioglycolate-elicited peritoneal cells were preincubated for 2 h at 37#{176}Cin arginine-free HEPES-RPMI medium (10 cells/mI), without 125. Cells were then circles) (10 g/ml) 60, and loaded and vs.
pelleted
and
resuspended
at time
0 in fresh
arginine-free
0
30
60
Incubation
90 time
120
L-arginine-containing (1.14 mM) (filled cirdes) medium. LPS was added to both groups of cells. Following incubation for 30, 90 mm under these conditions, the pH, recovery rate of cells acid by the NH4 prepulse method was measured. P < .05 vs. time 0 arginine-free medium, n - 3-4.
of the
thioglycolate-treated
peritoneal
Biology
Volume
(mm)
52,
October
milieu may alter not only nitric oxide 16, 17, 22], but also L-arginase activity
study.
Fig. 3. pH regulation in freshly harvested macrophages. The pH1 recovery rate of freshly harvested thioglycolate-elicited peritoneal cells was measured either immediately following harvesting (time 0) or after 30, 60, 90, or 120 mm of incubation at 37#{176}Cin HEPES-RPMI medium (107 cells/mi) with (triangles) or without (cirdes) 1.14 mM L-arginine, in the presence (filled symbols) or absence (open symbols) of LPS (10 ig/ml). The pH1 recovery rate of cells acid loaded by the NH4 prepulse method was then measured. Where absent, the error bar was smaller than the symbol. P < .05 vs. time 0, n - 3-4. #P < .05 vs. all other groups at I 90 mm, n - 3-4. P < .05 vs. arginine-free medium, without LPS, n - 3-4.
of Leukocyte
(open
or
lines of evidence arginase did not
0.0
Journal
* *
0
tory [15,
0
398
> 0
cavity
(2) the arginase pathway, by which the guanidino incorporated into urea, with the other product ornithine. Exposure to factors present within the
0.2
‘4-
a) 0 c1
0.6
a,
or
medium (Fig. 3). Ms metabolize L-arginine via two major pathways: (1) the nitric oxide synthase pathway, which results in oxidation of the guanidino nitrogen, liberating nitric oxide and yielding nitrite, nitrate, and citrulline as stable end products; and
5..
0.
0.
to some other in vivo event, such as the trauma of peritoneal lavage, requires further investigation. The former explanation would appear to be unlikely, since one would expect cells incubating in L-arginine following their harvest to continue to metabolize L-arginine, resulting in continued impairment of pH1 recovery. However, m4 pH1 recovery rose to levels equivalent to that of cells incubating in arginine-free
*
I
I
I
environment
E
a,
0.8
E
U)
and/or synergistically stimuof L-arginine [22]. In addipriming agent for enhanced in response to LPS [15, 16, 17,
The present studies demonstrate that pH1 recovery significantly slower in freshly harvested thioglycolate-elicited ms than in cells incubated with or without L-arginine
C
1992
mediated showed cubated
First,
neither
suggest contribute
that
urea
nor
to
metabolism the findings L-ornithine
nitrogen being inflamma-
synthase [23, 24].
is L-
activity Several
of L-arginine reported in affects
pH1 regulation [11]. Second, previous that addition of exogenous L-arginase with LPS reversed the L-arginine-dependent
H
by this
pump-
experiments to m4s
inin-
hibition of pH1 recovery from acid loading [11]. This would not be expected if metabolism by endogenously produced Larginase were significantly limiting the supply of L-arginine for the nitric oxide synthase pathway. Finally, other investigators have reported that neither peritoneal release L-arginase in vitro [13]. The oxidative metabolism of L-arginine the most part, been studied in vitro. In this
nor
wound
by ms setting,
m4s
has, for nitrogen
oxides derived m toxicity
for
from a
27], as well documented
as for the
L-arginine variety
have been intracellular
of
tumor cells significance
vironment:
found to mediate parasites [25-
[12]. Recent of oxidative
10.
studies have L-arginine
H j
metabolism in vivo. Kilbourn et al. [30] found that treatment with the competitive inhibitor of L-arginine metabolism, N-monomethyl-L-arginine (N-MMA), ameliorated shock induced by tumor necrosis factor in dogs. Evidence that oxidative metabolism of L-arginine by ms plays a significant role in vivo comes from experiments reported by Liew et al. [31], who injected N-MMA into the mouse footpad. This inhibited the in vivo killing of Leishmania by ms, indicating that nitric oxide derived from L-arginine mcdiated the antiparasitic activity of m4s in vivo. By demonstrating inhibition of H pump-mediated m4 pH1 regulation by LPS in combination with levels of L-arginine mimicking that of the inflammatory milieu, these experiments suggest that nitric oxide may impair m4 pH, regulation in vivo. Such an inhibition of H pump activity would be particularly deleterious within the acidic extracellular environment typically prevailing in an abscess or wound. Under these conditions, H pump-mediated pH1 regulation is critical to efficient m4 function [9]. Consequently, this represents a potential mechanism by which m4 function may be modulated in vivo. We are currently investigating the impact of nitrogen oxide generation in vivo on m4 pH1 regulation using a recently described technique to inhibit nitric oxide synthase
activity
in
vivo
11.
by
nine
Medical
Research
arginine Marletta,
The
penitoneal
environment
monomicrobial
and
Surg. 213, 253. 3. Silver, IA. (1978) Tissue Exp. Med Biol. 94, 769. 4. Mandell,
G.L.
(1974)
P#{176}2 changes
O.D.,
Nasmith,
PO2
Immun.
Gninstein,
T.L.
17.
the
effect
and
pH.
and
S. (1987)
The
7. Swallow, Cj., Gninstein, regulation in macrophages dicycbohexylcarbodiimide-sensitive
ment of a plasma 19558. 8.
Swallow,
Cj.,
cytoplasmic 46, 588. 9. Swallow,
Modulation
membrane
Rotstein, by an
O.D. (1988) ATP-dependent mechanism:
23.
proton
pump.
j
Biol.
Chem.
recovery
O.D.,
Gninstein,
S. (1989)
in acid-loaded
macrophages.
Grinstein,
of the
S.,
macrophage
Sudsbury,
R.A.,
respiratory
Rotstein,
burst
O.D.
Swallow
R.,
Leaf,
CD.,
Wishnok,
J.S.
and
nitrate:
nitrate
bio-
S.,
S., Rotstein,
Ada
and
1022,
Furuya, W. Na/H antiport
O.D.
activated
(1990)
penitoneal
Regulation
of
macrophages.
203. (1986) Characterization of human neutrophils.
of
the
Am. j
amilo-
Physiol.
(M4):
Cloning
Ding, nitrogen mouse Albina,
A.H.,
and
inducibility.
FASEBJ
6, A1973.
Nathan, CF., Stuehr, Dj. (1988) Release of reactive intermediates and reactive oxygen intermediates from peritoneal macrophages. j Immunol. 141, 2407. J.E., Mills, CD., Henry, W.L., Jr., Caldwell, M.D. (1990)
expression wounds.
of different Immunol.
j
pathways of arginine 144, 3877.
metabolism
Granger,
L-arginine.
j
DL.,
Immunol.
Hibbs,
Metabolic fate of macrophages.
J.B.,
Green, vated gotes
Sj., Meltzer, MS., macrophages destroy by an L-arginine-dependent
144,
278.
28.
James, S.L., schistosomula
production
144, Jr.,
2725.
Perfect,
JR.,
Durack,
of L-arginine in relation to microbiostatic J. Clin. Invest. 85, 264.
27.
(1990)
capacity
Hibbs, J.B., Jr., Nacy, CA. (1990) Actiintracellular Leishmania major amastikilling mechanism. j Immunol.
Glaven, J. (1989) Macrophage of Schistosoma mansoni involves
of reactive
D.T.
nitrogen
intermediates.
cytotoxicity against arginine-dependent j Immunol. 143,
4208. Malkin, R., Flescher, E., Lengy, J., Keisani, Y. (1987) On the interactions between macrophages and developmental stages of Schisiosoma mansoni: the cytotoxic mechanisms involved in macrophagemediated killing of schistosomula in vitro. Immunobiology 176, 63. Kilbourn, R.G., Gross, S.S., Jubran, A Adams, J., Griffith, D.W.,
29.
30.
Levi,
R.,
Lodato,
RE
(1990)
L-N”-Methyl-argsnine
inhibits
tumor necrosis factor-induced hypotension: implications for the involvement of nitric oxide. Proc. Nail. Acad. Sci. USA 87, 3629. Liew, F.Y., Millott, S., Parkinson, C., Palmer, R.M.J., Moncada, S. (1990) Macrophage killing ofLcishmania parasite in vivo is mediated by nitric oxide from L-arginine. j Immunol. 144, 4794. Granger, DL., Hibbs, J.B., Jr., Broadnax, L.M. (1991) Urinary nitrate excretion in relation to munine macrophage activation.
(1990)
en-
Med.
26.
of Res.
by an acidic
(1989)
Exp.
25.
32. Cj.,
CD. j
Shearer, J.D., Mills, CD., Caldwell, M.D. (1992) Differential regulation of macrophage arginine metabolism. FASEBJ 6, A1719. Adams, LB., Hibbs, J.B., Jr., Taintor, R.R., Krahenbuhl, J.L. (1990) Microbiostatic effect of munine activated macrophages for Toxoplasma gondii: role for synthesis of inorganic nitrogen oxides
263,
Mechanisms j Surg.
macro-
C283.
Temporal in healing
Bacieroides
Cytoplasmic pH and N,N’possible involve-
pH
peritoneal
Bowman, Ej., Siebers, A., Altendorf, K. (1988) Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc. Nail. Acad. Sci. USA 85, 7972. Xie, Q -W., Cho, Hj., Calaycay, J., Mumford, R., Ding, A., Troso, T., Nathan, C. (1992) Nitric oxide synthase (NOS) of macrophages
31.
Rotstein,
pH
S.,
in
W.L., Jr., Mills, by L-arginine.
Iyengar,
in resident
Biophys.
Grinstein, ride-sensitive
22.
9, 337.
by-product, succinic acid, inhibits neutrophil respiratory burst by reducing intracellular pH. Infeci. Immun. 55, 864. 6. Rotstein, O.D., Fiegel, V.D., Simmons, R.L., Knighton, DR. (1988) The deleterious effect of reduced pH and hypoxia on neutrophil migration in vitro. j Surg. Res. 45, 298.
P.S.,
Gninstein, pH
Biochim. 19.
Adv. anaerobic
(1991)
wounds.
Yoon,
Cj.,
cytoplasmic
of Ann.
ATPases
O.D.
cytoplasmic
CD.,
in
M.A.,
18. Swallow,
(1991)
inflammation.
of aerobic
Infect.
P.E.,
on
in acute
activity
neutrophils.
Pruett,
infection:
bacteria
Bactericidal
polymorphonuclear 5. Rotstein,
during
polymicrobial
Rev. 61, 296.
Rotstein,
impairs
mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proc. NaiL Acad. Sci. USA 82, 7738. Stuehr, Dj., Gross, S.S., Sakuma, I., Levi, I.R., Nathan, C. (1989) Activated murine macrophages secrete a metabolite of arginine with the bioactivity of endothelium-denived relaxing factor and the chemical reactivity of nitric oxide. j Exp. Med. 169, lOll.
from Physiol.
pH. RB.,
macrophages.
synthesis:
REFERENCES (1981) Intracellular M.D., Adams,
A vacuolar-type
Macrophage oxidation of L-arginine to nitrite nitric oxide is an intermediate. Biochemisiry 27, 8706. 16. Stuehr, D.J., Marletta, MA. (1985) Mammalian
24.
1. Roos, A., Boron, W.F. 2. Sawyer, R.G., Spengler,
R.A.,
L-arginine
H 1009.
(1990)
in munine
(1988)
Council
of Canada. C.J.S. is recipient of an Ethicon-Society of University Surgeons Surgical Research Fellowship and a Medical Research Council of Canada Fellowship. S.G. is recipient of an International Research Scholars Award from the Howard Hughes Medical Institute.
O.D.
pH
by proton
Barbul, A., Caldwell, M.D. (1988) ArgiAm. j Physiol. 17, E459-E467. GA., Gyure, L., Cifuentes, L. (1979) Microenvironmental depletion by macrophages in vivo. Br j Cancer 39, 613.
14. Curnie, 15.
Mills,
pH regulation
Sudsbury,
from
by vacuolar-type Exp. Mcd. 174,
metabolism
250,
the
S.,
derived
169, 1021. 13. Albina, J.E.,
[32].
supported
Rotstein,
Albina, J.E., Caidwell, M.D., Henry, Regulation of macrophage functions
21.
was
S.,
Gninstein,
oxide
regulation phages.j
ACKNOWLEDGMENT work
Cj.,
Nitric
12.
role of cytoplasmic 108, 363.
ATPase regulates cytoplasmic Biol. Chem. 265, 7645.
Swallow,
20.
This
the critical
pumps. Surgery Cj., Gninstein,
extrusion Swallow,
j
ci al.
Immunol.
146,
Macrophage
1294.
cytoplasmic
pH
and
inflammation
399
