Immunosuppression mediated by nitric oxide
Eur. J. Immunol. 1992. 22: 2249-2254
2249
Basel K. Al-Ramadi., Joseph J. Meissler Jr., Duan Huang and Toby K. Eisenstein
Immunosuppression induced by nitric oxide and its inhibition by interleukin-4*
Department of Microbiology and Immunology, Temple University School of Medicine, Philadelphia
Mice immunized with attenuated Salmonella typhirnurium, strain SL3235, while protected against virulent challenge, are unable to mount in vivo and in vitro antibody responses to non-Salmonella antigens, such as tetanus toxoid and sheep red blood cells, and exhibit profoundly suppressed responses to B and Tcell mitogens. Suppression of antibody responses is mediated by macrophage (Ma)-released soluble factors, and is completely reversed by treatment with interleukin (1L)-4. The present report identifies the suppressor factor as nitric oxide (NO), and provides evidence for a mechanism by which IL-4 abrogates suppression. Suppressed antibody responses correlated with high levels of NO secretion by splenocytes of SL3235-immunized mice. NO production was observed only in cultures consisting of the adherent cell fraction of immune splenocytes. Further, immunosuppression was reversed by NG-monomethylL-arginine (NMLA), a competitive inhibitor of NO synthesis, and was completely blocked by the addition of excess L-arginine. Treatment with IL-4, or antiinterferon (1FN)-y monoclonal antibody (mAb), also abrogated suppression. Optimal reversal of suppression was observed only when NMLA, IL-4, or anti-IFN-y mAb, was added at day 0 of the 5-day plaque-forming cell assay. Treatment with either IL-4 or anti-IFN-y mAb also lead to a sharp inhibition of NO production by immune spleen cells. Moreover, the addition of IL-4 to splenic adherent M@ inhibited their ability to generate NO. Our data characterize an immunoregulatory pathway, involving I F N y and NO, by which M a mediate immunosuppression and identify IL-4 as a potent inhibitor of this pathway.
1 Introduction Macrophages ( M a ) have been shown to mediate immunosuppression in a number of intracellular microbial infections caused by bacterial, parasitic, and fungal agents [l-41. In general, suppression is thought to facilitate the establishment and dissemination of the pathogen in the host. In some systems, however, suppression is observed upon treatment with immunopotentiating agents, such as Mycobacterium tuberculosis BCG and aro A- mutants of Salmonella typhimurium [5-81. In these cases, nonspecific suppression appears to correlate with M@ activation [7, 9, 101. Attenuated aro A- mutants of Salmonella have recently received much attention for their potential use as vaccines against virulent Salmonella infections [11, 121 as well as carriers for cloned genes of a number of protective proteins of other pathogenic organisms [13-16]. Therefore, an understanding of the nature of the immune response invoked by these mutants is of great importance. Utilizing the Salmonella model, we have recently demonstrated that
Ma-mediated suppression of in vitro anti-SRBC plaqueforming cell (PFC) responses, induced by immunization with an aro A- mutant (strain SL3235), was completely abrogated upon treatment with IL-4, but not IL-2 [17]. Furthermore, IL-4 appeared to mediate its effect by acting at the level of M@ [17]. Recent studies demonstrated that activated M a produce reactive nitrogen intermediates (RNI), of which nitric oxide (NO) is the ultimate effector molecule [18, 191, being directly responsible for the tumoricidal and some microbicidal activities of M a (for a review, see [20]). Given the M a activating properties of SL3235 [8, 91, an assessment of the involvement of NO in the observed suppression of PFC responses was undertaken. This report describes the identification of the suppressive factor in SL3235-immunized mice as NO and provides a mechanism by which IL-4 reverses the observed suppression.
2 Materials and methods [I 106121
* This work was supported by United States Public Health Service Grant AI-15613 from the National Institutes of Health. Current address: Section of Immunobiology, Howard Hughes Medical Institute at Yale University School of Medicine, New Haven, CT 06510, USA. Correspondence: Toby K. Eisenstein, Department of Microbiology and Immunology, Temple University School of Medicine, Philadelphia, PA 19140, USA Abbreviations: NMLA: Nc-monomethyl-L-arginine NO: Nitric oxide Not-: Nitrite RNI: Reactive nitrogen intermediates
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1992
2.1 Mice Female C3HeB/FeJ mice were purchased from Jackson Laboratories (Bar Harbor, ME) and housed in sterilized plastic cages with Absorb-Dri for bedding. Purina Mouse Chow and fresh water were available ad libitum. All mice were acclimatized for at least 1week prior to being used at 8-12 weeks of age.
2.2 Bacterial strain S. typhirnuriurn SL3235, kindly provided by Dr. Bruce A. D. Stocker (Stanford University School of Medicine, 0014-2980/92/0909-2249$3.50+ .2510
2250
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eisenstein
Stanford, CA), is a smooth, avirulent strain with an LD5,, (50% lethal dose) of greater than lo7 bacteria when given i.p. For immunization, 5 x 105-10 x lo5 of log-phase bacteria, prepared as described in detail previously [S, 211, were used.
Eur. J. Immunol. 1992. 22: 2249-2254
2.6 Measurement of N02- accumulation
Spleen cells, either unfractionated or following fractionation into adherent and nonadherent populations, were cultured at 1 x lo7 cells/ml in 96-well plates in RPMI 1640 (Gibco) supplemented with 2% heat-inactivated FCS for 3-120 h. Cell-free culture supernatants were assayed for 2.3 Reagents N02-, a measure of NO synthesis [23], by the colorimetric Griess reaction [24]. Briefly, 100-pl aliquots of supernatant Indomethacin, catalase and L-arginine were obtained from were mixed with an equal volume of Griess reagent (1% Sigma Chemical Co. (St. Louis, MO). NG-monomethyl- sulfanilamide/O.1YO napht hylethylene diamine dihydroL-arginine (NMLA) was purchased from Calbiochem- chloride/2% H3P04) (Sigma) and incubated at room temBehring Corp. (La Jolla, CA). Rabbit IgG neutralizing perature for 10 min. The absorbance at 562 nm was measantibody to transforming growth factor (TGF)-Pl and pz ured in an automated microplate reader. NOz- was quanwas from R & D Systems (Minneapolis, MN). The neutral- titated using NaNOz as a standard. izing dosej" (ND5")of this antibody was 3-5 pg/ml. Control non-immune rabbit serum was kindly provided by Dr. K. Blank (Temple University School of Medicine, Philadel- 2.7 Statistics phia, PA), and was used after heat inactivation for 30 min at 56 "C. Neutralizing hamster IgG mAb to murine IFN-y was The significance of the differences observed was assessed obtained from Genzyme (Boston, MA). Murine rIL-4 was by the Student's t-test. Differences were considered significant when p values of < 0.05 were obtained. purchased from Biosource Int. (Camarillo, CA).
2.4 Cell preparation Single-spleen cell suspensions were prepared as previously described [S].Briefly, cell suspension was prepared in tissue culture medium (TCM), and RBC lysed by hypotonic shock.TCM consisted of MEM (Gibco, Grand Island, NY) supplemented with 10% defined heat-inactivated FCS (LPS concentration = 0.013 ng/ml; Hyclone Labs, Logan, UT), 2 mM L-glutamine, 50 pg/ml gentamicin, 1mM nonessential amino acids (Gibco), 1 mM sodium pyruvate (M.A. Bioproducts,Walkersville, MD), 10 pg/ml of adenosine, uridine, cytosine and guanosine, and 0.05 mM of 2-ME (Sigma).The viability of cells was routinely > 95% as determined by trypan blue dye exclusion. In some experiments, splenocytes from SL3235-injected mice were fractionated into adherent and nonadherent subpopulations by allowing them to adhere to plastic for 2 h at 37°C. The adherent cells, comprising approximately 37% of total splenocytes, were > 95% activated M@ as determined by Diff-Quik stain [S]. To determine nitrite (NOz-) accumulation, adherent cells were gently scraped off, adjusted to 1 X lo7 cells/ml and cultured as described in Sect. 2.6.
2.5 Primary in vitro PFC response Antibody-producing cells were generated in vitro by the method of Mishell and Dutton with modification, as detailed previously [S]. Briefly, 1.0 X lo7viable spleen cells from normal or SL3235-injected mice were dispensed in 1-ml aliquots into flat-bottom 24-well tissue culture plates (Costar, Cambridge, MA), and immunized with 3.5 x lo6 pre-washed SRBC (Rockland, Gilbertsville, PA) in 50 p1 volume. For each group tested, a minimum of three cultures were set up. On day 5 of the assay, cells from individual wells were harvested and tested in the PFC assay. The number of direct PFC was determined by the Cunningham modification of the Jerne hemolytic plaque assay [22]. The results are expressed as the % of normal response (calculated from the equation: mean PFC of immune cells/means PFC of normal cells x 100).
3 Results 3.1 Production of NO by spleen cells of SL3235-immunized mice Activated M@ secrete several suppressive substances, including prostaglandins [25], hydrogen peroxide [26], and TGF-P [27].The possible involvement of these compounds in the observed suppression of PFC responses was studied by testing the effect of adding indomethacin (1-5 pg/ml), catalase (1 x 103 - 5 x 103U/ml), or a neutralizing antibody specific to TGF-P (5-50 pg/ml) at culture initiation. Our results indicated that suppression was not sensitive to any of the above treatments, suggesting the lack of involvement of the above factors in the observed suppression (data not shown). 100
80
-6
60
*z.
r,
40
20
0 0
24
48
72
96
120
144
Hours of Incubation
Figure 1. Measurement of NO secretion by normal and SL3235immune spleen cells. Cells (1 x lo7)from C3HeB/FeJ mice injected 7 days previously with the indicated doses of SL323.5, or from saline-iajected controls, were cultured in 1.0 ml of medium in the absence of additional stimuli for 3-120 h. Cell-free supernatants were assayed for NOz- as described in Sect. 2.6. The data represents the means f SEM of triplicate determinations for each group, and is expressed as pM NOz-/107 cells.
Immunosuppression mediated by nitric oxide
Eur. J. Immunol. 1092. 22: 2249-2254
'To assess whether N O is involved in immunosuppression, NO: production, a breakdown product of NO and an indicator of its synthesis [ 2.31,was investigasted. Spleen cells from mice injected with various doses of SL3235, or from control mice (injected with saline), were cultured in medium without any additional stimuli for 3-120 h, after which NO, - accumulation was quantitated. The results (Fig. 1) demonstrate that SLd235 immuniLation leads to a dose-dependent induction of high levels of NO2 production. A lag period of 3 h was required before NO2 production could be detected, with the majority of NO2 synthesis occurring in the first 24 h of culture, which is in accordance with findings by others [2X]. Nitrite accumulation reached optimal levels after spleen cells were cultured for 48 h and continued thereafter at plateau levels for up to 130 h (Fig. 1). Production of NO? by splenocytes of saline-injected control mice was undetectable (Fig. 1).
2251
1 .
B
40-
k
"V A
0
1
0
2
0
3
0
4
n
5
0
M
NMLA (pM)
3.2 Adherent spleen cells are responsible for NO production To study the cell type(s) responsible for the observed NO2 production, immune spleen cells were fractionated into adherent and nonadherent subpopulations, and cultured in the absence of any additional stimuli for 48 h. The data, illustrated in Fig. 2, demonstrate that only adherent cells, which represent mature activated MQ,,were able to generate NO?-. Furthermore, the addition of KMLA, a competitive inhibitor of N O synthesis, resulted in a dosedependent inhibition of NO2 production (Fig. 2). These data clearly demonstrate that SL3235 immunization leads to M a activation and subsequent release of high levels of NO. 3.3 NMLA restores immune responsiveness With the direct demonstration of N O secretion by immune spleen cells, the ability of NMLA to reverse suppression
Pzgure 3 Effect of NMLA on PFC responses of immune spleen cells. Sormal, or immune, spleen cells were cultured with SRHC in the presence of the indicated concentration of UMLA, with or without I,-arginine, added at a final concentration of 4 8 mM Both NMLA and I,-arginine were added at culture initiation. PFC' responses wcrc enumerated 5 days later. The data represent the means zk SEM of a minimum of triplicate culture5 per group. The PFC responses of immune cell cultures are shown as the %, of normal response calculated from the equation: mean PFC of immune cells/mean PFC of normal cells X LOO. The control normal cell responses were 2777 f 95 (for cultures without I -arginine) and 2152 k 174 (for cultures with I -argmine).
was studied. As shown in Fig. 3, the addition of NMLA to immune cell cultures resulted in a dose-dependent enhancement of PFC responses. This effect was completely blocked in the presence of excess 1,-arginine,demonstrating that suppression is mediated by an L-arginine-dependent mechanism.
3.4 Effect of anti-IFN-y mAb and rIL-4 on suppression
4s
The most potent inducer of NO secretion by MQ, is IFN-y [ 181. Thus, we attempted to reverse suppression by the addition of a neutralLing mAb to murine IFN-y (Fig. 4 A). The results demonstrate that the addition of anti-IFN-y mAb blocked immunosuppression in a dose-dependent manner (Fig. 4 A), demonstrating the 1FN-ydependence of the process. Higher doses of the mAb (up to 1000 ng/ml) gave similar results (data not shown). The observed suppression in antibody responses is also exquisitely sensitive to IL-4, since treatment with 1-5 U/ml of rIL-4 led to a full reversal of suppression (Fig. 4B) [17].
40 35
n
5 u c 3
. H 3
30
25 20
15
10
5
n 4
hMLA Concentration/ FM
rzguio 2 Production of NO is a function of adherent M Q and is
inhibited by \MI A Immune splenocytes were fractionated into adherent and nonadhercnt \ubpopulation\ and 1 X lo7 cells/ml of either unfractiondted. or fractionated. populations were cultured without further stimulation in the presence o f increasing doses of \ M I A After 18 h of culture, \O:- accumulation was measured in cell-free q x r n d t a n t 5 , as described. The data repesents the m e m \ i- S t M o f triplicate determination5 for each group, and is exprewd ILM YO2 /lo7cells
3.5 Abrogation of suppression by NMLA, anti-IFN-y mAb, or rIL-4 occurs with similar time course 'Time-course experiments were carried out in which the above inhibitors were added either at culture initiation (day 0) or at different times thereafter. The results, shown in Fig. 5, demonstrate that optimal reversal of suppression occurred only when the inhibitors (NMLA, rIL-4, or anti-IFN-y mAb) were added at culture initiation, suggesting that their presence is required during the induction phase of KNI synthesis. It is noteworthy that, of the three
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eisenstein
Eur. J. Immunol. 1992. 22: 2249-2254 9
B 100
80
60
40
20 0
20
40
M)
80
100 120 140
0.0
anti-IFN-u ( ng I rnl )
0.2-
0.4
0.6
0.8
1.0
1.2
rIL-4 ( U l m l )
Figure 4. Abrogation of immunosuppression by anti-IFN-y mAb ( A ) and rIL-4 ( B ) . The indicated concentrations of either antiTFN-y mAb, or m u r k rIL-4, were added t o cell cultures as day 0. PFC responses were determined at day 5 of assay. The results are represented as % of normal ccll responses which were 1.503 k 71 ( A ) and 1035 f 135 ( B ) .
Con1rol
Ddy 0
Day I
Day 2
Day 3
Day of ‘Treatment
Figurc5. Timc course of NMLA, rIL-4, and anti-IFN-y mAb action. Normal, or immune, cell cultures were set up in the presence ol SRBC antigen. NMLA (at a final concentration of 50 ELM).rIL-4 ( 5 U/ml). or anti-IFN-y mAb (200 g/ml),were added either at culture initiation (day 0) or at the indicated times thereafter. PFC rcsponses wcrc dctermincd at day 5 of assay, and are shown as the YO of normal cell response. “Control” indicates the response of immune cells in the absence of any inhibitor. The responses of normal ccll cultures were: 2375 f 217 (in prcsence of NMLA). 933 ? 131 (rlL-4). and 1107 f 48 (anti-IFN-y mAb).
inhibitors used, only rIL-4 was effective in significantly reversing suppression when added after culture initiation (61% at day 1 and 31% at day 2; Fig. 5).
3.6 Treatment with rIL-4 or anti-IFN-y mAb directly inhibits NO secretion To define more clearly the mechanism by which IL-4 and anti-IFN-y mAb blocked suppression, the capacity of
-
......
. .......
1;’
.-
0
1
10
’
m
10’
lo3
nglml
Concentration of Inhibitor
Figure 6. Inhibition of NOz- secretion by anti-IFN-y mAb ( A ) ,or rIL-4 ( B ) . Immune spleen cells were cultured at a concentration of 1 X lO7/rnl with the indicated concentrations of each inhibitor and NO*- concentration in cell-free supernatants was determined after 72 h.The data represent the means of triplicate cultures per group and are depicted as percent inhibition of NO*- secretion, as compared with immune cells cultured under identical conditions, but in the absence of inhibitors. The means k SEM of the control cells. ~ response was 41.8 f 1.6 p ~ / l O
immune splenocytes to produce RNI in the presence of these inhibitors was investigated. Both treatments resulted in a dose-dependent inhibition of NO2- production (Fig. 6 A and B). In anti-IFN-y-treated cultures, doses of 3 x lo2 and 1 x lo4 ng/ml resulted in 20%-65% inhibition, respectively (Fig. 6A). Similarly, addition of rIL-4 at 3-100 U/ml led to 20%-80% inhibition in NOz- secretion, respectively (Fig. 6B). It is important to note that the inhibitor concentrations required to inhibit RNI synthesis were approximately 30-50-fold higher than those needed to reverse suppression of PFC responses (compare Fig. 4 with Fig. 6). We believe this difference is due to the buffering action of SRBC which were used as the antigen in the PFC assay, since oxyhemoglobin has been shown to bind NO with high affinity [29]. This is supported by our findings which showed that the addition of increasing concentrations of SRBC to immune spleen cell cultures led to decreases in N02- accumulation (data not shown). 3.7 IL-4 inhibits NO synthesis by acting directly on adherent M a
A previous report from this laboratory demonstrated that IL-4 acts directly on MQ to reverse the suppression of PFC responses of immune spleen cells [17]. We, therefore, studied whether IL-4 could also inhibit NO production in a similar fashion. The results, shown inTable 1, demonstrate that IL-4 acts directly on splenic adherent cells and inhibits their production of NO.The lower level of NO2- production observed in cultures of adherent cells, compared to that of unfractionated splenocytes, could be the result of the
Imrnunosupprcssion mediated by nitric oxide
Eur. J. Immunol. 1992. 22: 2249-2254 Table 1. IL-4 inhibits NO production by acting directly on splenic adherent cells Cclls in culturen)
Nitrite production (pM/107 ~ c l l s ) ~ J Untreated + NMLA + IL-4 (2 m i ) (100 Ulml)
Whole immune SC 42.0 f 1.4 Adherent SC: 27.9 -+ 1.2 &onadherent SC 2.8 ? 1.0
3.5 f 0.6 2.2 k 0.3 1.0 f 0.3
5.3 ? 0.5 9.7 k 1.8 2.4 f 0.4
a) Splccn cells, either unfractionated or after fractionation into adherent and nonadherent fractions, were cultured at 1 X lo7 cclldml in the presence or absence of either 1 mM NMLA or 100 U/ml rIL-4 (added at culture initiation). b) Accumulation of NOz- in cell-free supernatants was determined after 48 h of culture. Data represent the mean & SEM of triplicate cultures and are expressed as p ~ / l Ocells/48 ~ h.
depletion of immuneT cells, which are a source of IFN-y. In fact. our preliminary data indicate that the major type of Tcells activated as a result of immunization with SL323.5 is the IFN-y-producingTh1 cells (data not shown), which is in agreement with recent findings [16]. Collectively, these results strongly suggest that the mechanism by which rIL-4 abrogates suppression of PFC responses is via the inhibition of NO production by activated MQ.
4 Discussion Previous reports from this laboratory have documented the occurrence of two apparently paradoxical phenomena in mice immunized with an attenuated strain of S. typhimurium, strain SL3235, namely, excellent protection against virulent Salmonella challenge, and immunosuppression of proliferative responses to mitogens and of antibody responses to heterologous antigens [7, 8,21, 301. In addition to Salmonella-specific resistance, which persisted for at least 7 months, SL3235 also induced transient crossprotection against virulent Listeria monocytogenes infection which lasted for approximately 3-4 weeks post immunization [30].These findings were seen as being consistent with the induction of activated M@ and the development of cell-mediated immunity [31, 321. Indeed, direct evidence for the ability of SL3235 to activate M a was previously obtained by demonstrating the induction of tumoricidal and microbicidal activities in peritoneal M a isolated from SL3235-immunized mice [9]. Of significance were the findings that the SL3235-induced nonspecific suppression also occurred over the first 3-4 weeks following immunization, and that depletion of adherent MQ resulted in an alleviation of suppression [S, 21, 331. Furthermore, the finding that immune spleen cells could suppress the PFC responses of normal, MHC-incompatible splenocytes, together with the demonstration that T lymphocyte depletion from immune spleen cells had little effect on their suppressive activity, all provide additional evidence for the lack of involvement of T, cells in the observed suppression [8, 171.Thus, in the first month after immunization, during which the two phenomena seem to overlap, resistance to infection and nonspecific immunosuppression appeared to be mediated by MQ.
2253
The present study was undertaken to characterize the mechanism by which M a mediate nonspecific suppression and identify the suppressor factor(s). Several lines of evidence strongly support a role for NO in mediating the suppression. SL323.5 immunization induced the release of high levels of NO by spleen cells. Both the degree of immunosuppression [34] and level of NO secretion were directly dependent on the dose of SL3235 used. Further, only the Ma-enriched, adherent fraction of immune splenocytes was capable of NO synthesis. This is in agreement with our previous findings demonstrating that mature M a are involved in the observed immunosuppression [8, 211. Suppression of PFC responses was reversed by the addition of NMLA, an inhibitor of NO synthase [35, 361. The addition of excess L-arginine, in the presence of NMLA, completely abrogated NMLA-mediated reversal of suppression. Treatment of immune spleen cells either with an mAb specific to IFN-y, or with rIL-4, also resulted in reversal of suppression. Most importantly, these treatments were also shown to directly inhibit NO production by immune spleen cells. The addition of IL-4 to adherent M a also directly blocked their NO synthesis. Furthermore, we have previously reported that the addition of immune splenocytes to normal cell cultures as late as day 4 of the 5-day PFC assay still inhibited the co-culture response [17], which is consistent with the rapid and cytostatic action of NO on mammalian cells [ 191.Taken together, these findings strongly suggest that NO plays a major role in SL3235mediated suppression. Nitric oxide is synthesized by the oxidation of the terminal guanido nitrogen atom of L-arginine by the enzyme NO synthase [35,37-391. NO produced by M a is inducible by bacterial LPS or lymphokines, such as IFN-y alone or in combination withTNF-a or TNF-fI [18, 28, 40, 411, and is inhibited by TGF-fI and Ma-deactivating factor [42]. The present results demonstrate that the synthesis of NO is under the reciprocal regulation of IFN-y and IL-4. In addition to its general stimulatory activities on multiple cells of the immune system [43], IL-4 down-regulates the production by activated M a of several inflammatory factors such as TNF-a, IL-1, prostaglandins, and hydrogen peroxide [44, 451. Our data demonstrate the ability of IL-4 to also inhibit the IFN-y-induced release of NO by splenic adherent MQ. This result is in agreement with a recent report in which IL-4 was shown to block NO synthesis by IFN-y-activated murine peritoneal M a by down-regulating the activity of NO synthase enzyme [46]. Several reports recently described the inhibitory effects of RNI in rat splenic and peritoneal M a [47, 481. Thus, the addition of L-arginine to rat resident peritoneal M a inhibited several of their functions, including superoxide production, phagocytosis, tumoricidal activity, electron transport chain activity, and protein synthesis [47]. Similarly, the Con A-induced proliferation of normal rat, or Corynebacteriumparvum-activated mouse, spleen cells was significantly enhanced by the addition of NMLA [48]. The present data, together with our previous findings [7, 9, 301, demonstrate that resistance to infection induced by an attenuated strain of S. typhimuriurn is accompanied by splenic M a activation and subsequent release of NO. The available evidence indicates that NO is primarily responsible for the decreased immune responsiveness observed in vitro [8, 17, 211. This mechanism of immunosuppression
2254
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eis;enstein
may well have additional in vivo significance, especially in light of the demonstration that the in vivo antibody responses to SRBC and tetanus toxoid were also significantly suppressed in SL3235-immunized mice [8, 341. The possible involvement of N O in regulating antibody responses in vivo is a subject of great interest and is currently under investigation. The identification of IL-4 as an inhibitor of NO production by M4, may have application in the treatment of anergic states in disseminated infections, such as tuberculosis. Further, NO produced by endothelial cells has been identified as endothelium-derived relaxing factor, which lowers blood pressure [49]. NO has also been identified as a natural neurotransmitter in the cerebellum [50, 511. Administration of IL-4 could be tested for ability to antagonize the biological effects of NO in these diverse biological systems. We wish to thank Dr. Bruce A. D. Stocker f o r his generous gift of .struin SL323.5. Received May 7. 1992
5 References 1 Klcinhenz, M. E. and Ellner, J. J., J. Infect. Dis. 1985. 152: 171. 2 Edwards 111. C. K., Hedegaard, H . B., Zlotnik, A , , Gangadharam, P. R.. Johnston Jr, R. B. and Pabst, M. J., J. Immunol. 1986. 136: 1820. 3 Silcghem, M., Darji, A., Hamers, R.,Van De Winkel, M. and De Bactselier, P., Eur. J. Immunol. 1989. 19: 829. 4 Nickcrson, D. A , , Havens, R. A. and Bullock, W. E., Cell. Immunol. 1981. 60: 287. 5 Orme, I. M. and Collins, F. M., J. Exp. Med. 1983. 158: 74. 6 Schrier, D. J.. Allen, E. M. and Moore,V. L., Cell. Immunol. 1980. 56: 347. 7 Eisenstcin,T. K., Killar, L. M., Stocker, B. A. D. and Sultzer, B. M., J. Immunol. 1984. 133: 958. 8 Al-Ramadi, B. K., Brodkin, M. A., Mosser, D. M. and Eisenstein, T. K., .I. Immunol. 1YYl. 146: 2737. 9 Schafcr, R., Nacy, C. A. and Eisenstein, T. K., J. Immunol. 1988. 140: 1638. 10 Appelberg. R., Soares, R., Ferreira, P. and Silva, M.T., Scand. .I. Immunol. 1989. 30: 165. 11 Hoiseth, S. K. and Stocker, B. A. D., Nature 1981. 291: 238. 12 Lcvine, M. M., Herrington, D., Murphy, J. R., Morris, J. G., Losonsky, G.,Tall, B., Lindberg, A. A., Svenson, S., Baqar, S., Edwards, M. F. and Stockcr, B., J. Clin. Invest. 1987. 79: 888. 13 Dougan. G., Hormaeche, C. E. and Maskell, D. J., Parasite Immunol. 1987. 9: 151. 14 Poirer,T. P.. Kehoc, M. A . and Beachey, E. H., J. Exp. Med. 1988. 168: 25. 15 Sadoff, J. C.,Ballou,W. R..Baron,L. S.,Majarian,W. R.,Brey, R. N., Hockmeyer,W.T.,Young, J. F., Cryz, S. J., Ou, J., Lowell, G. H. and Chulay, J. D., Science 1988. 240: 336. 16 Yang, D. M., Fairweather, N., Button, L. L., McMaster,W. R., Kahl, L. P. and Liew, F. Y , J. Immunol. 1990. 145: 2281. 17 Al-Ramadi. B. K., Chen, Y-W., Meissler, J. J., Jr. and Eisenstein,T. K., J. Immunol. 1991. 147: 1954. 18 Ding, A , , Nathan, C. F. and Stuehr, D. J., J. Immunol. 1988. 141: 2407. I0 Stuchr, D. J. and Nathan, C. F., J. Exp. Med. 1989.169; 1543.
Eur. J. Immunol. 1992. 22: 2249-2254
20 Liew, F.Y. and Cox, F. E. G., Immunol. Today 1991.12: A17. 21 Lee, J.-C., Gibson, C.W. and Eisenstein,T. K., Cell. Immunol. 1985. 91: 75. 22 Cunningham, A . and Szenberg, A., Immunology 1968.14: 599. 23 Marletta, M. A., Yoon, P. S., Iyengar, R., Leaf, C. D. and Wishnok, J. S., Biochemistry 1988. 27: 8706. 24 Green, L. C., Wagner, D. A., Glogowski, J., Skipper, l? L., Wishnok, J. S. and Tannenbaum, S. R., Anal. Biochem. 1982. 126: 131. 25 Metzger, Z., Hoffeld, J. T. and Oppenheim, J. J., J. Immunol. 1980. 124: 983. 26 Boraschi, D., Ghezzi, P., Pasqualetto, E., Salmona, M., Nencioni, L., Soldateschi, D., Villa, L. and Tagliabue, A., Immunology 1983. 50: 359. 27 Assoian, R. K., Fleurdelys, B. E., Stevenson, H. C., Miller, P. J., Madtes, D. K., Raines, E.W., Ross, R. and Sporn, M. B., Proc. Natl. Acad. Sci. U S A 1987. 84: 6020. 28 Stuehr, D. J. and Marletta, M. A,, J. lmmunol. 1987.139: 518. 29 Goretski, J. and Hollocher, T. C.. J. Biol. Chem. 1988. 263: 2316. 30 Killar, L. M. and Eisenstein, T. K., Infect. Immun. 1985. 47: 605. 31 Mackaness, G. B., Blanden, R.V. and Collins, F. M., J. Exp. Med. 1966. 124: 573. 32 Mackaness, G. B., .I. Exp. Med. 1969. 129: 973. 33 Eisenstein,T. K., Dalal, N., Killar, L., Lee, J.-C. and Schafer, R. in Eisenstein, T. K., Bullock, W. E. and Hanna, N. (Eds.), Host Defences and Immunomodulation to Intracellular Pathogens, Plenum Press, New York 1988, p. 353. 34 Al-Ramadi, B. K., Greene, J. M., Meissler, J. J., Jr. and Eisenstein, T. K., Microb. Pathog. 1992, in press. 35 Hibbs, J. B., Jr., Taintor, R. R. and Vavrin, Z., Science 1987. 235: 473. 36 Hibbs, J. B., Jr.,Vavrin, Z. and Taintor, R. R., .I. Immunol. 1987. 138: 550. 37 Iyengar, R., Stuehr, D. J. and Marletta, M. A., Proc. Natl. Acad. Sci. USA 1987. 84: 6369. 38 Sakuma, I., Steuhr, D., Gross, S., Nathan, C. F. and Levi, R., Proc. Natl. Acad. Sci. U S A 1988. 85: 8664. 39 Palmer, R. M. J., Ashton, D. S. andMoncada, S., Nature 1988. 333: 664. 40 Stuehr, D. J. and Marletta, M. A,, Proc. Natl. Acad. Sci. USA 1985. 82: 7738. 41 Drapier, J. C., Wietzerbin, J. and Hibbs, J. B., Jr.. Eur. J. Immunol. 1988. 18: 1587. 42 Ding, A , , Nathan, C. F., Graycar, J., Derynck, R., Stuehr, D. J. and Srimal, S., J. Immunol. 1990. 145: 940. 43 Paul, W. E., FASEB J. 1987. I : 456. 44 Hart, P. H.,Vitti, G. F., Burgess, D. R.,Whitty, G. A., Piccoli, D. S. and Hamilton, J. A , , Proc. Natl. Acad. Sci. USA 1989.86: 3803. 45 Lehn, M., Weiser, W.Y., Engelhorn, S., Gillis, S. and Remold, H. G., J. Immunol. 1989. 143: 3020. 46 Liew, F.Y., Li,Y., Severn, A.. Millott, S., Schmidt, J., Salter, M. and Moncada, S., Eur. J. Immunol. 1991. 21: 2489. 47 Albina, J. E., Caldwell, M. D., Henry,W. L., Jr. and Mills, C. D., J. Exp. Med. 1989. 169: 1021. 48 Mills, C. D., J. Immunol. 1991. 146: 2719. 49 Palmer, R. M. J., Ferrige, A. G. and Moncada, S., Nature 1987. 327: 524. 50 Garthwaite, J., Charles, S. L. and Chess-Williams, R., Nature 1988. 336: 385. 51 Bredt, D. S. and Snyder, S. H., Proc. Natl. Acad. Sci. U S A 1989. 86: 9030.
Eur. J. Immunol. 1992. 22: 2249-2254
2249
Basel K. Al-Ramadi., Joseph J. Meissler Jr., Duan Huang and Toby K. Eisenstein
Immunosuppression induced by nitric oxide and its inhibition by interleukin-4*
Department of Microbiology and Immunology, Temple University School of Medicine, Philadelphia
Mice immunized with attenuated Salmonella typhirnurium, strain SL3235, while protected against virulent challenge, are unable to mount in vivo and in vitro antibody responses to non-Salmonella antigens, such as tetanus toxoid and sheep red blood cells, and exhibit profoundly suppressed responses to B and Tcell mitogens. Suppression of antibody responses is mediated by macrophage (Ma)-released soluble factors, and is completely reversed by treatment with interleukin (1L)-4. The present report identifies the suppressor factor as nitric oxide (NO), and provides evidence for a mechanism by which IL-4 abrogates suppression. Suppressed antibody responses correlated with high levels of NO secretion by splenocytes of SL3235-immunized mice. NO production was observed only in cultures consisting of the adherent cell fraction of immune splenocytes. Further, immunosuppression was reversed by NG-monomethylL-arginine (NMLA), a competitive inhibitor of NO synthesis, and was completely blocked by the addition of excess L-arginine. Treatment with IL-4, or antiinterferon (1FN)-y monoclonal antibody (mAb), also abrogated suppression. Optimal reversal of suppression was observed only when NMLA, IL-4, or anti-IFN-y mAb, was added at day 0 of the 5-day plaque-forming cell assay. Treatment with either IL-4 or anti-IFN-y mAb also lead to a sharp inhibition of NO production by immune spleen cells. Moreover, the addition of IL-4 to splenic adherent M@ inhibited their ability to generate NO. Our data characterize an immunoregulatory pathway, involving I F N y and NO, by which M a mediate immunosuppression and identify IL-4 as a potent inhibitor of this pathway.
1 Introduction Macrophages ( M a ) have been shown to mediate immunosuppression in a number of intracellular microbial infections caused by bacterial, parasitic, and fungal agents [l-41. In general, suppression is thought to facilitate the establishment and dissemination of the pathogen in the host. In some systems, however, suppression is observed upon treatment with immunopotentiating agents, such as Mycobacterium tuberculosis BCG and aro A- mutants of Salmonella typhimurium [5-81. In these cases, nonspecific suppression appears to correlate with M@ activation [7, 9, 101. Attenuated aro A- mutants of Salmonella have recently received much attention for their potential use as vaccines against virulent Salmonella infections [11, 121 as well as carriers for cloned genes of a number of protective proteins of other pathogenic organisms [13-16]. Therefore, an understanding of the nature of the immune response invoked by these mutants is of great importance. Utilizing the Salmonella model, we have recently demonstrated that
Ma-mediated suppression of in vitro anti-SRBC plaqueforming cell (PFC) responses, induced by immunization with an aro A- mutant (strain SL3235), was completely abrogated upon treatment with IL-4, but not IL-2 [17]. Furthermore, IL-4 appeared to mediate its effect by acting at the level of M@ [17]. Recent studies demonstrated that activated M a produce reactive nitrogen intermediates (RNI), of which nitric oxide (NO) is the ultimate effector molecule [18, 191, being directly responsible for the tumoricidal and some microbicidal activities of M a (for a review, see [20]). Given the M a activating properties of SL3235 [8, 91, an assessment of the involvement of NO in the observed suppression of PFC responses was undertaken. This report describes the identification of the suppressive factor in SL3235-immunized mice as NO and provides a mechanism by which IL-4 reverses the observed suppression.
2 Materials and methods [I 106121
* This work was supported by United States Public Health Service Grant AI-15613 from the National Institutes of Health. Current address: Section of Immunobiology, Howard Hughes Medical Institute at Yale University School of Medicine, New Haven, CT 06510, USA. Correspondence: Toby K. Eisenstein, Department of Microbiology and Immunology, Temple University School of Medicine, Philadelphia, PA 19140, USA Abbreviations: NMLA: Nc-monomethyl-L-arginine NO: Nitric oxide Not-: Nitrite RNI: Reactive nitrogen intermediates
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1992
2.1 Mice Female C3HeB/FeJ mice were purchased from Jackson Laboratories (Bar Harbor, ME) and housed in sterilized plastic cages with Absorb-Dri for bedding. Purina Mouse Chow and fresh water were available ad libitum. All mice were acclimatized for at least 1week prior to being used at 8-12 weeks of age.
2.2 Bacterial strain S. typhirnuriurn SL3235, kindly provided by Dr. Bruce A. D. Stocker (Stanford University School of Medicine, 0014-2980/92/0909-2249$3.50+ .2510
2250
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eisenstein
Stanford, CA), is a smooth, avirulent strain with an LD5,, (50% lethal dose) of greater than lo7 bacteria when given i.p. For immunization, 5 x 105-10 x lo5 of log-phase bacteria, prepared as described in detail previously [S, 211, were used.
Eur. J. Immunol. 1992. 22: 2249-2254
2.6 Measurement of N02- accumulation
Spleen cells, either unfractionated or following fractionation into adherent and nonadherent populations, were cultured at 1 x lo7 cells/ml in 96-well plates in RPMI 1640 (Gibco) supplemented with 2% heat-inactivated FCS for 3-120 h. Cell-free culture supernatants were assayed for 2.3 Reagents N02-, a measure of NO synthesis [23], by the colorimetric Griess reaction [24]. Briefly, 100-pl aliquots of supernatant Indomethacin, catalase and L-arginine were obtained from were mixed with an equal volume of Griess reagent (1% Sigma Chemical Co. (St. Louis, MO). NG-monomethyl- sulfanilamide/O.1YO napht hylethylene diamine dihydroL-arginine (NMLA) was purchased from Calbiochem- chloride/2% H3P04) (Sigma) and incubated at room temBehring Corp. (La Jolla, CA). Rabbit IgG neutralizing perature for 10 min. The absorbance at 562 nm was measantibody to transforming growth factor (TGF)-Pl and pz ured in an automated microplate reader. NOz- was quanwas from R & D Systems (Minneapolis, MN). The neutral- titated using NaNOz as a standard. izing dosej" (ND5")of this antibody was 3-5 pg/ml. Control non-immune rabbit serum was kindly provided by Dr. K. Blank (Temple University School of Medicine, Philadel- 2.7 Statistics phia, PA), and was used after heat inactivation for 30 min at 56 "C. Neutralizing hamster IgG mAb to murine IFN-y was The significance of the differences observed was assessed obtained from Genzyme (Boston, MA). Murine rIL-4 was by the Student's t-test. Differences were considered significant when p values of < 0.05 were obtained. purchased from Biosource Int. (Camarillo, CA).
2.4 Cell preparation Single-spleen cell suspensions were prepared as previously described [S].Briefly, cell suspension was prepared in tissue culture medium (TCM), and RBC lysed by hypotonic shock.TCM consisted of MEM (Gibco, Grand Island, NY) supplemented with 10% defined heat-inactivated FCS (LPS concentration = 0.013 ng/ml; Hyclone Labs, Logan, UT), 2 mM L-glutamine, 50 pg/ml gentamicin, 1mM nonessential amino acids (Gibco), 1 mM sodium pyruvate (M.A. Bioproducts,Walkersville, MD), 10 pg/ml of adenosine, uridine, cytosine and guanosine, and 0.05 mM of 2-ME (Sigma).The viability of cells was routinely > 95% as determined by trypan blue dye exclusion. In some experiments, splenocytes from SL3235-injected mice were fractionated into adherent and nonadherent subpopulations by allowing them to adhere to plastic for 2 h at 37°C. The adherent cells, comprising approximately 37% of total splenocytes, were > 95% activated M@ as determined by Diff-Quik stain [S]. To determine nitrite (NOz-) accumulation, adherent cells were gently scraped off, adjusted to 1 X lo7 cells/ml and cultured as described in Sect. 2.6.
2.5 Primary in vitro PFC response Antibody-producing cells were generated in vitro by the method of Mishell and Dutton with modification, as detailed previously [S]. Briefly, 1.0 X lo7viable spleen cells from normal or SL3235-injected mice were dispensed in 1-ml aliquots into flat-bottom 24-well tissue culture plates (Costar, Cambridge, MA), and immunized with 3.5 x lo6 pre-washed SRBC (Rockland, Gilbertsville, PA) in 50 p1 volume. For each group tested, a minimum of three cultures were set up. On day 5 of the assay, cells from individual wells were harvested and tested in the PFC assay. The number of direct PFC was determined by the Cunningham modification of the Jerne hemolytic plaque assay [22]. The results are expressed as the % of normal response (calculated from the equation: mean PFC of immune cells/means PFC of normal cells x 100).
3 Results 3.1 Production of NO by spleen cells of SL3235-immunized mice Activated M@ secrete several suppressive substances, including prostaglandins [25], hydrogen peroxide [26], and TGF-P [27].The possible involvement of these compounds in the observed suppression of PFC responses was studied by testing the effect of adding indomethacin (1-5 pg/ml), catalase (1 x 103 - 5 x 103U/ml), or a neutralizing antibody specific to TGF-P (5-50 pg/ml) at culture initiation. Our results indicated that suppression was not sensitive to any of the above treatments, suggesting the lack of involvement of the above factors in the observed suppression (data not shown). 100
80
-6
60
*z.
r,
40
20
0 0
24
48
72
96
120
144
Hours of Incubation
Figure 1. Measurement of NO secretion by normal and SL3235immune spleen cells. Cells (1 x lo7)from C3HeB/FeJ mice injected 7 days previously with the indicated doses of SL323.5, or from saline-iajected controls, were cultured in 1.0 ml of medium in the absence of additional stimuli for 3-120 h. Cell-free supernatants were assayed for NOz- as described in Sect. 2.6. The data represents the means f SEM of triplicate determinations for each group, and is expressed as pM NOz-/107 cells.
Immunosuppression mediated by nitric oxide
Eur. J. Immunol. 1092. 22: 2249-2254
'To assess whether N O is involved in immunosuppression, NO: production, a breakdown product of NO and an indicator of its synthesis [ 2.31,was investigasted. Spleen cells from mice injected with various doses of SL3235, or from control mice (injected with saline), were cultured in medium without any additional stimuli for 3-120 h, after which NO, - accumulation was quantitated. The results (Fig. 1) demonstrate that SLd235 immuniLation leads to a dose-dependent induction of high levels of NO2 production. A lag period of 3 h was required before NO2 production could be detected, with the majority of NO2 synthesis occurring in the first 24 h of culture, which is in accordance with findings by others [2X]. Nitrite accumulation reached optimal levels after spleen cells were cultured for 48 h and continued thereafter at plateau levels for up to 130 h (Fig. 1). Production of NO? by splenocytes of saline-injected control mice was undetectable (Fig. 1).
2251
1 .
B
40-
k
"V A
0
1
0
2
0
3
0
4
n
5
0
M
NMLA (pM)
3.2 Adherent spleen cells are responsible for NO production To study the cell type(s) responsible for the observed NO2 production, immune spleen cells were fractionated into adherent and nonadherent subpopulations, and cultured in the absence of any additional stimuli for 48 h. The data, illustrated in Fig. 2, demonstrate that only adherent cells, which represent mature activated MQ,,were able to generate NO?-. Furthermore, the addition of KMLA, a competitive inhibitor of N O synthesis, resulted in a dosedependent inhibition of NO2 production (Fig. 2). These data clearly demonstrate that SL3235 immunization leads to M a activation and subsequent release of high levels of NO. 3.3 NMLA restores immune responsiveness With the direct demonstration of N O secretion by immune spleen cells, the ability of NMLA to reverse suppression
Pzgure 3 Effect of NMLA on PFC responses of immune spleen cells. Sormal, or immune, spleen cells were cultured with SRHC in the presence of the indicated concentration of UMLA, with or without I,-arginine, added at a final concentration of 4 8 mM Both NMLA and I,-arginine were added at culture initiation. PFC' responses wcrc enumerated 5 days later. The data represent the means zk SEM of a minimum of triplicate culture5 per group. The PFC responses of immune cell cultures are shown as the %, of normal response calculated from the equation: mean PFC of immune cells/mean PFC of normal cells X LOO. The control normal cell responses were 2777 f 95 (for cultures without I -arginine) and 2152 k 174 (for cultures with I -argmine).
was studied. As shown in Fig. 3, the addition of NMLA to immune cell cultures resulted in a dose-dependent enhancement of PFC responses. This effect was completely blocked in the presence of excess 1,-arginine,demonstrating that suppression is mediated by an L-arginine-dependent mechanism.
3.4 Effect of anti-IFN-y mAb and rIL-4 on suppression
4s
The most potent inducer of NO secretion by MQ, is IFN-y [ 181. Thus, we attempted to reverse suppression by the addition of a neutralLing mAb to murine IFN-y (Fig. 4 A). The results demonstrate that the addition of anti-IFN-y mAb blocked immunosuppression in a dose-dependent manner (Fig. 4 A), demonstrating the 1FN-ydependence of the process. Higher doses of the mAb (up to 1000 ng/ml) gave similar results (data not shown). The observed suppression in antibody responses is also exquisitely sensitive to IL-4, since treatment with 1-5 U/ml of rIL-4 led to a full reversal of suppression (Fig. 4B) [17].
40 35
n
5 u c 3
. H 3
30
25 20
15
10
5
n 4
hMLA Concentration/ FM
rzguio 2 Production of NO is a function of adherent M Q and is
inhibited by \MI A Immune splenocytes were fractionated into adherent and nonadhercnt \ubpopulation\ and 1 X lo7 cells/ml of either unfractiondted. or fractionated. populations were cultured without further stimulation in the presence o f increasing doses of \ M I A After 18 h of culture, \O:- accumulation was measured in cell-free q x r n d t a n t 5 , as described. The data repesents the m e m \ i- S t M o f triplicate determination5 for each group, and is exprewd ILM YO2 /lo7cells
3.5 Abrogation of suppression by NMLA, anti-IFN-y mAb, or rIL-4 occurs with similar time course 'Time-course experiments were carried out in which the above inhibitors were added either at culture initiation (day 0) or at different times thereafter. The results, shown in Fig. 5, demonstrate that optimal reversal of suppression occurred only when the inhibitors (NMLA, rIL-4, or anti-IFN-y mAb) were added at culture initiation, suggesting that their presence is required during the induction phase of KNI synthesis. It is noteworthy that, of the three
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eisenstein
Eur. J. Immunol. 1992. 22: 2249-2254 9
B 100
80
60
40
20 0
20
40
M)
80
100 120 140
0.0
anti-IFN-u ( ng I rnl )
0.2-
0.4
0.6
0.8
1.0
1.2
rIL-4 ( U l m l )
Figure 4. Abrogation of immunosuppression by anti-IFN-y mAb ( A ) and rIL-4 ( B ) . The indicated concentrations of either antiTFN-y mAb, or m u r k rIL-4, were added t o cell cultures as day 0. PFC responses were determined at day 5 of assay. The results are represented as % of normal ccll responses which were 1.503 k 71 ( A ) and 1035 f 135 ( B ) .
Con1rol
Ddy 0
Day I
Day 2
Day 3
Day of ‘Treatment
Figurc5. Timc course of NMLA, rIL-4, and anti-IFN-y mAb action. Normal, or immune, cell cultures were set up in the presence ol SRBC antigen. NMLA (at a final concentration of 50 ELM).rIL-4 ( 5 U/ml). or anti-IFN-y mAb (200 g/ml),were added either at culture initiation (day 0) or at the indicated times thereafter. PFC rcsponses wcrc dctermincd at day 5 of assay, and are shown as the YO of normal cell response. “Control” indicates the response of immune cells in the absence of any inhibitor. The responses of normal ccll cultures were: 2375 f 217 (in prcsence of NMLA). 933 ? 131 (rlL-4). and 1107 f 48 (anti-IFN-y mAb).
inhibitors used, only rIL-4 was effective in significantly reversing suppression when added after culture initiation (61% at day 1 and 31% at day 2; Fig. 5).
3.6 Treatment with rIL-4 or anti-IFN-y mAb directly inhibits NO secretion To define more clearly the mechanism by which IL-4 and anti-IFN-y mAb blocked suppression, the capacity of
-
......
. .......
1;’
.-
0
1
10
’
m
10’
lo3
nglml
Concentration of Inhibitor
Figure 6. Inhibition of NOz- secretion by anti-IFN-y mAb ( A ) ,or rIL-4 ( B ) . Immune spleen cells were cultured at a concentration of 1 X lO7/rnl with the indicated concentrations of each inhibitor and NO*- concentration in cell-free supernatants was determined after 72 h.The data represent the means of triplicate cultures per group and are depicted as percent inhibition of NO*- secretion, as compared with immune cells cultured under identical conditions, but in the absence of inhibitors. The means k SEM of the control cells. ~ response was 41.8 f 1.6 p ~ / l O
immune splenocytes to produce RNI in the presence of these inhibitors was investigated. Both treatments resulted in a dose-dependent inhibition of NO2- production (Fig. 6 A and B). In anti-IFN-y-treated cultures, doses of 3 x lo2 and 1 x lo4 ng/ml resulted in 20%-65% inhibition, respectively (Fig. 6A). Similarly, addition of rIL-4 at 3-100 U/ml led to 20%-80% inhibition in NOz- secretion, respectively (Fig. 6B). It is important to note that the inhibitor concentrations required to inhibit RNI synthesis were approximately 30-50-fold higher than those needed to reverse suppression of PFC responses (compare Fig. 4 with Fig. 6). We believe this difference is due to the buffering action of SRBC which were used as the antigen in the PFC assay, since oxyhemoglobin has been shown to bind NO with high affinity [29]. This is supported by our findings which showed that the addition of increasing concentrations of SRBC to immune spleen cell cultures led to decreases in N02- accumulation (data not shown). 3.7 IL-4 inhibits NO synthesis by acting directly on adherent M a
A previous report from this laboratory demonstrated that IL-4 acts directly on MQ to reverse the suppression of PFC responses of immune spleen cells [17]. We, therefore, studied whether IL-4 could also inhibit NO production in a similar fashion. The results, shown inTable 1, demonstrate that IL-4 acts directly on splenic adherent cells and inhibits their production of NO.The lower level of NO2- production observed in cultures of adherent cells, compared to that of unfractionated splenocytes, could be the result of the
Imrnunosupprcssion mediated by nitric oxide
Eur. J. Immunol. 1992. 22: 2249-2254 Table 1. IL-4 inhibits NO production by acting directly on splenic adherent cells Cclls in culturen)
Nitrite production (pM/107 ~ c l l s ) ~ J Untreated + NMLA + IL-4 (2 m i ) (100 Ulml)
Whole immune SC 42.0 f 1.4 Adherent SC: 27.9 -+ 1.2 &onadherent SC 2.8 ? 1.0
3.5 f 0.6 2.2 k 0.3 1.0 f 0.3
5.3 ? 0.5 9.7 k 1.8 2.4 f 0.4
a) Splccn cells, either unfractionated or after fractionation into adherent and nonadherent fractions, were cultured at 1 X lo7 cclldml in the presence or absence of either 1 mM NMLA or 100 U/ml rIL-4 (added at culture initiation). b) Accumulation of NOz- in cell-free supernatants was determined after 48 h of culture. Data represent the mean & SEM of triplicate cultures and are expressed as p ~ / l Ocells/48 ~ h.
depletion of immuneT cells, which are a source of IFN-y. In fact. our preliminary data indicate that the major type of Tcells activated as a result of immunization with SL323.5 is the IFN-y-producingTh1 cells (data not shown), which is in agreement with recent findings [16]. Collectively, these results strongly suggest that the mechanism by which rIL-4 abrogates suppression of PFC responses is via the inhibition of NO production by activated MQ.
4 Discussion Previous reports from this laboratory have documented the occurrence of two apparently paradoxical phenomena in mice immunized with an attenuated strain of S. typhimurium, strain SL3235, namely, excellent protection against virulent Salmonella challenge, and immunosuppression of proliferative responses to mitogens and of antibody responses to heterologous antigens [7, 8,21, 301. In addition to Salmonella-specific resistance, which persisted for at least 7 months, SL3235 also induced transient crossprotection against virulent Listeria monocytogenes infection which lasted for approximately 3-4 weeks post immunization [30].These findings were seen as being consistent with the induction of activated M@ and the development of cell-mediated immunity [31, 321. Indeed, direct evidence for the ability of SL3235 to activate M a was previously obtained by demonstrating the induction of tumoricidal and microbicidal activities in peritoneal M a isolated from SL3235-immunized mice [9]. Of significance were the findings that the SL3235-induced nonspecific suppression also occurred over the first 3-4 weeks following immunization, and that depletion of adherent MQ resulted in an alleviation of suppression [S, 21, 331. Furthermore, the finding that immune spleen cells could suppress the PFC responses of normal, MHC-incompatible splenocytes, together with the demonstration that T lymphocyte depletion from immune spleen cells had little effect on their suppressive activity, all provide additional evidence for the lack of involvement of T, cells in the observed suppression [8, 171.Thus, in the first month after immunization, during which the two phenomena seem to overlap, resistance to infection and nonspecific immunosuppression appeared to be mediated by MQ.
2253
The present study was undertaken to characterize the mechanism by which M a mediate nonspecific suppression and identify the suppressor factor(s). Several lines of evidence strongly support a role for NO in mediating the suppression. SL323.5 immunization induced the release of high levels of NO by spleen cells. Both the degree of immunosuppression [34] and level of NO secretion were directly dependent on the dose of SL3235 used. Further, only the Ma-enriched, adherent fraction of immune splenocytes was capable of NO synthesis. This is in agreement with our previous findings demonstrating that mature M a are involved in the observed immunosuppression [8, 211. Suppression of PFC responses was reversed by the addition of NMLA, an inhibitor of NO synthase [35, 361. The addition of excess L-arginine, in the presence of NMLA, completely abrogated NMLA-mediated reversal of suppression. Treatment of immune spleen cells either with an mAb specific to IFN-y, or with rIL-4, also resulted in reversal of suppression. Most importantly, these treatments were also shown to directly inhibit NO production by immune spleen cells. The addition of IL-4 to adherent M a also directly blocked their NO synthesis. Furthermore, we have previously reported that the addition of immune splenocytes to normal cell cultures as late as day 4 of the 5-day PFC assay still inhibited the co-culture response [17], which is consistent with the rapid and cytostatic action of NO on mammalian cells [ 191.Taken together, these findings strongly suggest that NO plays a major role in SL3235mediated suppression. Nitric oxide is synthesized by the oxidation of the terminal guanido nitrogen atom of L-arginine by the enzyme NO synthase [35,37-391. NO produced by M a is inducible by bacterial LPS or lymphokines, such as IFN-y alone or in combination withTNF-a or TNF-fI [18, 28, 40, 411, and is inhibited by TGF-fI and Ma-deactivating factor [42]. The present results demonstrate that the synthesis of NO is under the reciprocal regulation of IFN-y and IL-4. In addition to its general stimulatory activities on multiple cells of the immune system [43], IL-4 down-regulates the production by activated M a of several inflammatory factors such as TNF-a, IL-1, prostaglandins, and hydrogen peroxide [44, 451. Our data demonstrate the ability of IL-4 to also inhibit the IFN-y-induced release of NO by splenic adherent MQ. This result is in agreement with a recent report in which IL-4 was shown to block NO synthesis by IFN-y-activated murine peritoneal M a by down-regulating the activity of NO synthase enzyme [46]. Several reports recently described the inhibitory effects of RNI in rat splenic and peritoneal M a [47, 481. Thus, the addition of L-arginine to rat resident peritoneal M a inhibited several of their functions, including superoxide production, phagocytosis, tumoricidal activity, electron transport chain activity, and protein synthesis [47]. Similarly, the Con A-induced proliferation of normal rat, or Corynebacteriumparvum-activated mouse, spleen cells was significantly enhanced by the addition of NMLA [48]. The present data, together with our previous findings [7, 9, 301, demonstrate that resistance to infection induced by an attenuated strain of S. typhimuriurn is accompanied by splenic M a activation and subsequent release of NO. The available evidence indicates that NO is primarily responsible for the decreased immune responsiveness observed in vitro [8, 17, 211. This mechanism of immunosuppression
2254
B. K. Al-Ramadi, J. J. Meissler, D. Huang and T. K. Eis;enstein
may well have additional in vivo significance, especially in light of the demonstration that the in vivo antibody responses to SRBC and tetanus toxoid were also significantly suppressed in SL3235-immunized mice [8, 341. The possible involvement of N O in regulating antibody responses in vivo is a subject of great interest and is currently under investigation. The identification of IL-4 as an inhibitor of NO production by M4, may have application in the treatment of anergic states in disseminated infections, such as tuberculosis. Further, NO produced by endothelial cells has been identified as endothelium-derived relaxing factor, which lowers blood pressure [49]. NO has also been identified as a natural neurotransmitter in the cerebellum [50, 511. Administration of IL-4 could be tested for ability to antagonize the biological effects of NO in these diverse biological systems. We wish to thank Dr. Bruce A. D. Stocker f o r his generous gift of .struin SL323.5. Received May 7. 1992
5 References 1 Klcinhenz, M. E. and Ellner, J. J., J. Infect. Dis. 1985. 152: 171. 2 Edwards 111. C. K., Hedegaard, H . B., Zlotnik, A , , Gangadharam, P. R.. Johnston Jr, R. B. and Pabst, M. J., J. Immunol. 1986. 136: 1820. 3 Silcghem, M., Darji, A., Hamers, R.,Van De Winkel, M. and De Bactselier, P., Eur. J. Immunol. 1989. 19: 829. 4 Nickcrson, D. A , , Havens, R. A. and Bullock, W. E., Cell. Immunol. 1981. 60: 287. 5 Orme, I. M. and Collins, F. M., J. Exp. Med. 1983. 158: 74. 6 Schrier, D. J.. Allen, E. M. and Moore,V. L., Cell. Immunol. 1980. 56: 347. 7 Eisenstcin,T. K., Killar, L. M., Stocker, B. A. D. and Sultzer, B. M., J. Immunol. 1984. 133: 958. 8 Al-Ramadi, B. K., Brodkin, M. A., Mosser, D. M. and Eisenstein, T. K., .I. Immunol. 1YYl. 146: 2737. 9 Schafcr, R., Nacy, C. A. and Eisenstein, T. K., J. Immunol. 1988. 140: 1638. 10 Appelberg. R., Soares, R., Ferreira, P. and Silva, M.T., Scand. .I. Immunol. 1989. 30: 165. 11 Hoiseth, S. K. and Stocker, B. A. D., Nature 1981. 291: 238. 12 Lcvine, M. M., Herrington, D., Murphy, J. R., Morris, J. G., Losonsky, G.,Tall, B., Lindberg, A. A., Svenson, S., Baqar, S., Edwards, M. F. and Stockcr, B., J. Clin. Invest. 1987. 79: 888. 13 Dougan. G., Hormaeche, C. E. and Maskell, D. J., Parasite Immunol. 1987. 9: 151. 14 Poirer,T. P.. Kehoc, M. A . and Beachey, E. H., J. Exp. Med. 1988. 168: 25. 15 Sadoff, J. C.,Ballou,W. R..Baron,L. S.,Majarian,W. R.,Brey, R. N., Hockmeyer,W.T.,Young, J. F., Cryz, S. J., Ou, J., Lowell, G. H. and Chulay, J. D., Science 1988. 240: 336. 16 Yang, D. M., Fairweather, N., Button, L. L., McMaster,W. R., Kahl, L. P. and Liew, F. Y , J. Immunol. 1990. 145: 2281. 17 Al-Ramadi. B. K., Chen, Y-W., Meissler, J. J., Jr. and Eisenstein,T. K., J. Immunol. 1991. 147: 1954. 18 Ding, A , , Nathan, C. F. and Stuehr, D. J., J. Immunol. 1988. 141: 2407. I0 Stuchr, D. J. and Nathan, C. F., J. Exp. Med. 1989.169; 1543.
Eur. J. Immunol. 1992. 22: 2249-2254
20 Liew, F.Y. and Cox, F. E. G., Immunol. Today 1991.12: A17. 21 Lee, J.-C., Gibson, C.W. and Eisenstein,T. K., Cell. Immunol. 1985. 91: 75. 22 Cunningham, A . and Szenberg, A., Immunology 1968.14: 599. 23 Marletta, M. A., Yoon, P. S., Iyengar, R., Leaf, C. D. and Wishnok, J. S., Biochemistry 1988. 27: 8706. 24 Green, L. C., Wagner, D. A., Glogowski, J., Skipper, l? L., Wishnok, J. S. and Tannenbaum, S. R., Anal. Biochem. 1982. 126: 131. 25 Metzger, Z., Hoffeld, J. T. and Oppenheim, J. J., J. Immunol. 1980. 124: 983. 26 Boraschi, D., Ghezzi, P., Pasqualetto, E., Salmona, M., Nencioni, L., Soldateschi, D., Villa, L. and Tagliabue, A., Immunology 1983. 50: 359. 27 Assoian, R. K., Fleurdelys, B. E., Stevenson, H. C., Miller, P. J., Madtes, D. K., Raines, E.W., Ross, R. and Sporn, M. B., Proc. Natl. Acad. Sci. U S A 1987. 84: 6020. 28 Stuehr, D. J. and Marletta, M. A,, J. lmmunol. 1987.139: 518. 29 Goretski, J. and Hollocher, T. C.. J. Biol. Chem. 1988. 263: 2316. 30 Killar, L. M. and Eisenstein, T. K., Infect. Immun. 1985. 47: 605. 31 Mackaness, G. B., Blanden, R.V. and Collins, F. M., J. Exp. Med. 1966. 124: 573. 32 Mackaness, G. B., .I. Exp. Med. 1969. 129: 973. 33 Eisenstein,T. K., Dalal, N., Killar, L., Lee, J.-C. and Schafer, R. in Eisenstein, T. K., Bullock, W. E. and Hanna, N. (Eds.), Host Defences and Immunomodulation to Intracellular Pathogens, Plenum Press, New York 1988, p. 353. 34 Al-Ramadi, B. K., Greene, J. M., Meissler, J. J., Jr. and Eisenstein, T. K., Microb. Pathog. 1992, in press. 35 Hibbs, J. B., Jr., Taintor, R. R. and Vavrin, Z., Science 1987. 235: 473. 36 Hibbs, J. B., Jr.,Vavrin, Z. and Taintor, R. R., .I. Immunol. 1987. 138: 550. 37 Iyengar, R., Stuehr, D. J. and Marletta, M. A., Proc. Natl. Acad. Sci. USA 1987. 84: 6369. 38 Sakuma, I., Steuhr, D., Gross, S., Nathan, C. F. and Levi, R., Proc. Natl. Acad. Sci. U S A 1988. 85: 8664. 39 Palmer, R. M. J., Ashton, D. S. andMoncada, S., Nature 1988. 333: 664. 40 Stuehr, D. J. and Marletta, M. A,, Proc. Natl. Acad. Sci. USA 1985. 82: 7738. 41 Drapier, J. C., Wietzerbin, J. and Hibbs, J. B., Jr.. Eur. J. Immunol. 1988. 18: 1587. 42 Ding, A , , Nathan, C. F., Graycar, J., Derynck, R., Stuehr, D. J. and Srimal, S., J. Immunol. 1990. 145: 940. 43 Paul, W. E., FASEB J. 1987. I : 456. 44 Hart, P. H.,Vitti, G. F., Burgess, D. R.,Whitty, G. A., Piccoli, D. S. and Hamilton, J. A , , Proc. Natl. Acad. Sci. USA 1989.86: 3803. 45 Lehn, M., Weiser, W.Y., Engelhorn, S., Gillis, S. and Remold, H. G., J. Immunol. 1989. 143: 3020. 46 Liew, F.Y., Li,Y., Severn, A.. Millott, S., Schmidt, J., Salter, M. and Moncada, S., Eur. J. Immunol. 1991. 21: 2489. 47 Albina, J. E., Caldwell, M. D., Henry,W. L., Jr. and Mills, C. D., J. Exp. Med. 1989. 169: 1021. 48 Mills, C. D., J. Immunol. 1991. 146: 2719. 49 Palmer, R. M. J., Ferrige, A. G. and Moncada, S., Nature 1987. 327: 524. 50 Garthwaite, J., Charles, S. L. and Chess-Williams, R., Nature 1988. 336: 385. 51 Bredt, D. S. and Snyder, S. H., Proc. Natl. Acad. Sci. U S A 1989. 86: 9030.
