Published online before print November 16, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
- and nitric oxide-mediated pathways
,3
* The Surgical Services/Center for Engineering in Medicine, Massachusetts General Hospital, Harvard Medical School, and Shriners Hospitals for Children, Boston, Massachusetts, USA; and
Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts, USA
3 Correspondence: Department of Pathology, Brigham and Women’s Hospital, 77 Avenue Louis Pasteur/NRB 730D, Boston, MA 02115, USA. E-mail: rmitchell{at}rics.bwh.harvard.edu
ABSTRACT
Burn injury results in immunosuppression; previous work implicated a combination of altered T lymphocyte subpopulations and the elaboration of macrophage-derived mediators. However, the conclusions were based on T cell stimulations in the setting of high-dose polyclonal mitogenic stimuli and a single kinetic time-point. In this study, splenocytes from burned animals were used to examine lymphocyte responses over a multi-day time course following saturating and subsaturating anti-CD3, as well as mixed lymphocyte response (MLR) stimulation. Burn injury resulted in suppressed splenocyte-proliferative responses to high-dose anti-CD3 (2 µg/ml) at all culture time-points (Days 2–5); this inhibition was eliminated by removing macrophages from the splenocyte cultures, by blocking NO production, or by using splenocytes from burned animals congenitally deficient in IFN-
(IFN-–/–). The results are consistent with immunosuppression attributable to burn-induced IFN- production, which in turn, drives macrophage NO synthesis (NOS). In MLR cultures, lymphocyte proliferation and IFN- production were depressed at later time-points (Days 3–5). APC from burned animals showed no defects as MLR stimulators; T cells from burned animals showed defective, proliferative responses, regardless of the stimulator population. Removing macrophages, adding a NOS inhibitor, or using IFN-–/– splenocytes did not restore the MLR response of burned splenocytes. T cells from burned IFN-–/– animals also showed depressed proliferation with subsaturating levels of anti-CD3 (0.1 µg/ml); anti-CD-28 augmented the proliferative response. We conclude that burn-induced immunosuppression to authentic antigenic stimulation is related at least in part to defective CD3 signaling pathways and not simply to increased IFN- or NO production.
Key Words: immune suppression • T lymphocytes • antigen-presenting cells • costimulation
INTRODUCTION
Sepsis syndrome occurs frequently in burned patients and is a major cause of morbidity and mortality in this population [1 ]. Depressed T cell function is postulated to underlie burn injury-induced immunosuppression, as several T cell functional parameters correlate with injury severity, as well as susceptibility to sepsis [1 2 3 ]. Numerous factors have been implicated in T cell dysfunction, including a lipid protein complex derived from burned skin [4 ], alteration of lymphocyte subpopulations (e.g., decreased CD4+/CD8+ ratio) [5 , 6 ], shift to Th2-type immune responses [7 , 8 ], and macrophage-derived mediators (e.g., PGE2 and NO) [3 , 9 10 11 ]. Recent work suggests that macrophage-derived NO, and not PGE2, plays the most important role in burn-induced immunosuppression [12 13 14 15 16 17 ]. These conclusions come from the observations that depressed mitogen responses (i.e., to anti-CD3 or Con A) in burned splenocytes are ameliorated by NO synthase (NOS) inhibitors but not by cyclooxygenase inhibitors. NO is proposed to suppress mitogen-stimulated T cell responses by inducing cell death [16 ]. The findings are in agreement with other immunosuppression models [18 , 19 ].
All previous in vitro studies regarding the mechanisms underlying burn-induced T cell defects involve polyclonal T cell stimulators at maximal concentrations (e.g., surface-bound anti-CD3, Con A, and/or PHA); the conclusions are also typically based on the data from one culture time-point. A significant caveat is that saturating concentrations of T cell mitogens can conceivably, artifactually obscure signaling defects that would be uncovered with a more physiologic stimulation model. Moreover, data from one time-point may lead to different conclusions, as peak responses may not occur concurrently for different cell population (or from different animals).
In this study, we addressed these issues by using primary mixed lymphocyte responses (MLR) and monitoring T cell proliferation and cytokine production daily over 5 days of culture using T cells and APC derived from sham or burned mice. Primary MLR cultures have the advantage of a large polyclonal response signal (such as mitogens) but more accurately recapitulate normal T cell activation pathways, as they are antigen-specific and require costimulator engagement to drive naïve T cell responses. For comparison, polyclonal stimulation by anti-CD3 was also assessed. The roles of IFN- and NO in the proliferative responses were evaluated using purified T cells, NOS inhibitor, and IFN--deficient (IFN-–/–) mice.
MATERIALS AND METHODS
Reagents
Indomethacin and NG-methyl-L-arginine acetate (NMMA) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Tritiated thymidine was from Perkin-Elmer Life and Analytical Sciences (Wellesley, MA, USA). Culture medium is referred to as complete-10 (C-10) and was prepared by supplementing RPMI 1640 with 10% heat-inactivated FBS, 1 mM pyruvate, 2 mM glutamine, 100 µM nonessential amino acids, 0.075% (w/v) sodium bicarbonate, 100 U/ml penicillin, 100 µg/ml streptomycin, and 50 µM β-ME; all culture medium constituents were from Gibco/Invitrogen (Carlsbad, CA, USA), except for β-ME (Sigma-Aldrich).
Animals and burn injury
Male C57BL/6J, BALB/c, and B6 IFN- knockout (KO) mice (B6.129S7-IFN- /J KO; 6–7 weeks old, 20–24 g) were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and were acclimatized for 
1 week under ambient conditions (12 h light/dark cycle and 25°C) with water and standard chow ad libitium prior to experiments. Animal care was provided in accordance with the National Institutes of Health guidelines, and all procedures were carried out with approval of the Massachusetts General Hospital Subcommittee on Research Animals.
Animals were randomly assigned to an unburned control group (sham) or a burned group and anesthetized by i.p. injection with a mixture of 120 mg/kg ketamine (Hospira, Lake Forest, IL, USA) and 12 mg/kg xylazine (Sigma-Aldrich). Once anesthetized, the dorsum and abdomen were shaved, and the mice were subjected to two separate burns. The first burn resulted from pressing outstretched abdominal skin between two boiling water-heated brass blocks for 10 s. In the second burn, animals were placed in a custom mold that exposed the dorsum and immersed in 80°C water for 7 s. The combined burn protocol produced a full-thickness burn on a total body surface area of 25%. Mice were resuscitated by an i.p. injection of 1.5 ml sterile, normal saline solution. Sham mice were similarly manipulated with omission of the burn steps.
Preparation of splenocytes
Seven days after burn, mice were anesthetized with halothane and killed by cardiac puncture. Spleens were aseptically removed and teased apart in sterile, plastic 100 µm strainers (BD Discovery Labware, Bedford, MA, USA) immersed in cold, sterile C-10 media. The spleen cell suspension was centrifuged for 15 min at 250 g, and erythrocytes in the resulting pellet were lysed with 4 mL Tris-buffered ammonium chloride (17 mM Tris, 144 mM NH4Cl) for 5 min at 37°C. A twofold excess of C-10 was added, and the cells were recentrifuged; the resulting pellet was suspended in C-10 (usually 10 mL) and enumerated. Cell viability was determined by trypan blue exclusion and ranged from 85% to 95%.
T cell separation
Mouse spleen T cells were isolated by positive selection using anti-CD90 (anti-Thy-1) magnetic beads and the MACS separating system (Miltenyi Biotec, Auburn, CA, USA), according to the manufacturer’s recommendations. Purified T cells and T cell-depleted splenocytes (non-T) were used for the experiments, testing the effect of sham versus burned non-T cells in stimulation cultures by anti-CD3 or MLR.
Flow cytometry
Cells were centrifuged and adjusted to 2 x 106 cells/mL in PBS containing 2% FBS. PE-conjugated rat anti-mouse CD3 and FITC-conjugated rat anti-mouse CD11B (BD PharMingen, San Diego, CA, USA) were added at final concentrations of 1 µg/mL and 0.25 µg/mL, respectively. Following 30 min incubation on ice, the stained cells were washed and resuspended in PBS. Fluorescent emission data were collected and processed on a Beckman Coulter (Fullerton, CA, USA) Epics Altra flow cytometer.
In vitro proliferation assays
T cell proliferation to anti-CD3 or allogeneic splenocyte stimulation (MLR) was determined using [3H] thymidine (3H-dT) incorporation.
For anti-CD3 stimulation, anti-CD-coated, flat-bottom, 96-well plates were prepared by adding 50 µL anti-CD3
(BD PharMingen; 2 µg/ml or 0.1 µg/ml concentrations representing maximal and submaximal anti-CD3 ligation conditions, respectively) solution per well and incubating the plates at 37°C for 2 h; splenocytes were then cultured at 5 x 105 cells/well in a final volume of 200 µl C-10 in quadruplicate or quintuplicate. For anti-CD3 + anti-CD28 stimulation, the 96-well plates were precoated with 0.1 µg/ml anti-CD3 and 2 µg/ml anti-CD28 before incubation with 5 x 105 cells/well.
For MLR, stimulators were treated with mitomycin C to block proliferation; mitomycin C-treated splenocytes were prepared by incubating cells with 25 µg/mL mitomycin C (final concentration) for 30 min at 37°C, followed by four washes with C-10. Cells/well (5x105) mitomycin C-treated splenocytes (from BALB/c or C57BL/6 allogeneic stimulators) were cocultured with equal numbers of C57BL/6 or BALB/c responder splenocytes in C-10 media (final volume, 200 µL). For experiments involving separated T cells and APC, 2.5 x 105 cells/well T cells were incubated with equal numbers of T cell-depleted APC.
In some anti-CD3 stimulation and MLR experiments, NMMA was added to a final concentration of 500 µM, and/or indomethacin was added to a concentration of 5 µM.
All proliferation experiments were conducted over a 5-day period. Individual cultures were pulsed daily with 1.25 µCi 3H-dT for the final 6 h of culture. The cells were harvested onto filter plates (Unifilter-96, GF/C plates, Perkin-Elmer Life and Analytical Sciences) using a Packard Filtermate Harvester (Perkin Elmer Life and Analytical Sciences) and counted with a TopCount microplate scintillation counter (Perkin Elmer Life and Analytical Sciences).
IFN- and IL-10 measurement
The concentration of IFN- in the cultures was determined by a sandwich ELISA. Ninety-six-well plates (Costar, Corning, NY, USA) were coated with anti-IFN- antibody (rat anti-mouse IFN- mAb, clone R4-6A2, BD PharMingen) overnight at 4°C and then blocked for 2 h at 37°C with 2% BSA in borate-buffered saline (0.17 M borate, 0.12 M NaCl), pH 8.0. After washing the plates with PBS containing 0.1% Triton X-100, sample supernatants from cultures harvested prior to 3H-dT labeling or recombinant IFN- (R&D Systems, Minneapolis, MN, USA) standard dilutions were added and incubated for 3 h at 37°C. After washing, biotinylated anti-IFN- antibody (1 ug/ml) was incubated for 45 min at room temperature (biotinylated rat anti-mouse IFN- mAb, clone XMG1.2, BD PharMingen), followed by incubation with avidin HRP (1:1000 dilution) for 30 min at room temperature (BD PharMingen). Wells were developed with H2O2 (0.2% of a 30% solution) and ABTS dye (0.054%, Roche Applied Sciences, Indianapolis, IN, USA) in citrate buffer (100 mM citrate, 0.1% Triton X-100, pH 4.5). Plates were read at 410 nm on a VersaMax microplate reader (Molecular Devices, Sunnyvale, CA, USA), and concentrations of IFN- were determined by comparing absorptions against the known standards. The concentration of IL-10 in the culture media was determined by a commercial ELISA kit, according to the manufacturer’s instructions (BD PharMingen).
RESULTS
Globally decreased T cell responses to anti-CD3 and in MLR at 7 days after burn injury
Prior studies showed significant inhibition of T cell responses to various stimuli beginning 3 days after burn injury, with maximal suppression of proliferation occurring 7–14 days in a 25% total body surface area (TBSA) murine burn model [2
, 17
]. Thus, we chose Day 7 post-burn as the time to harvest splenocytes to assess T cell responses. Measuring T cell responses to two different stimuli, anti-CD3 and foreign MHC (MLR), provides distinct pieces of information about T cell functions. Anti-CD3 stimulation addresses intrinsic, direct signaling pathways inside T cells, and MLR requires cell–cell interactions and costimulation.
As shown in Figure 1A
and 1B
, and Table 1
, dramatic differences in T cell proliferation of splenocytes from burn and sham mice were observed following anti-CD3 stimulation and MLR, respectively. The predominant cytokine (by more than an order of magnitude) in these cultures is IFN- [17
, 20
]; ELISA assays showed diminished IFN- commensurate with the diminished lymphocyte proliferation (Fig. 1C
and 1D
, and Table 1
). Overall, these two assays of T cell proliferation and cytokine productions demonstrated that burn injury impaired T cell responses to saturating anti-CD3 as well as alloantigen stimulation.
 View larger version (13K): [in a new window] |
Figure 1. Five-day time course of proliferation (A, B) and IFN- production (C, D) of C57BL/6J mouse splenocytes prepared 7 days after burn or sham injury and stimulated by saturating dose (2 µg/ml)-bound anti-CD3 (A, C) or MLR (using BALB/c splenocytes as stimulators; B, D). All four plots are representative of results from several mice (three to four). Results from one representative experiment (one mouse) are shown. Lymphocyte responses from individual mice were performed in quadruplicate or quintuplicate, and each data-point represents mean ± SD.
|
|
View this table: [in a new window] |
Table 1. Suppressed, Proliferative and Cytokine Responses to Anti-CD3 Stimulation and MLR following Burn Injurya
|
 View larger version (19K): [in a new window] |
Figure 2. Relative cellular composition of mouse splenocytes from sham (upper) or burn animal (lower) mice. Splenocytes were isolated at 7 days after sham or burn procedure and stained with PE-labeled anti-CD3 antibody or FITC-labeled anti-CD11b antibody. The percentage of total (CD3) T lymphocytes decreased, whereas the percentage of the number of macrophages (CD11b) increased significantly after burn injury. Total 1 x 106 cells were counted for each animal, and the numeric values (mean±SD) embedded in each panel were from three different animals in each group.
|
produced in MLR using stimulators harvested 7 days after burn injury compared with sham APC. This could be attributed to increased MHC or costimulation expression on "burned" APC or to a relatively larger number of functional APC recovered from spleens in burned animals. FACS data show that the CD11b+ subpopulation of macrophages in spleen was increased significantly (4% in sham vs. 9% in burned mice; Fig. 2
). However, APC from burned animals do not exhibit increased CD40 or B7 expression (data not shown). Most significantly, burn injury did not compromise APC-stimulating ability, suggesting that diminished responses were instead attributable to a T cell signaling defect.
 View larger version (21K): [in a new window] |
Figure 3. Five-day MLR time course of proliferation (A) and IFN- production (B) of BALB/c mouse splenocytes stimulated by C57BL/6J splenocytes prepared 7 days after burn or sham injury. Both plots are representative of results from four separate mice, and data points with error bar are mean ± SD from one typical animal performed in quadruplicate or quintuplicate.
|
|
View this table: [in a new window] |
Table 2. No Attenuated Function of APC following Burn Injurya
|
production are shown in Figure 4A
4B
4C
4D
, with the averaged peak values from multiple animals listed in Table 3
.
 View larger version (41K): [in a new window] |
Figure 4. Time course of the anti-CD3 or MLR-proliferative response (A, B) and IFN- production (C, D) of isolated T cells from sham or burned C57BL/6J mice, compared with two different T cell plus non-T cell combinations. Untreated BALB/c splenocytes were stimulators in the MLR. Plots are representative of results from four separate mice, and data points with error bar are mean ± SD from one typical animal performed in quadruplicate or quintuplicate.
|
|
View this table: [in a new window] |
Table 3. Anti-CD3 Stimulation but not MLR Stimulation after Burn Injury is Reduced as a Result of Non-T Cell Populationsa
|
production of the sham-treated T cells. As expected, the presence of sham-treated non-T cells did not alter the responses of burned T lymphocytes to anti-CD3 stimulation. These results are in agreement with previous work showing that T cell-proliferative responses to Con A stimulation after burn injury were improved by macrophage depletion [17
]. In MLR experiments, however, the purified, burned T cells responded poorly at all time-points. Moreover, purified, burned T cells actually showed relatively poorer MLR responses than did unfractionated, burned splenocytes (compare time course data in Fig. 4B
with Fig. 1B
or the peak values in Table 3
with those in Table 1
).
Adding burned non-T cells to purified sham T cells substantially suppressed the T cell responses for anti-CD3 stimulation and in MLR cultures, suggesting a suppressive effect produced by cells in the burned non-T populations. Indeed, numerous, previous studies demonstrated that NO produced by burned macrophages is responsible for this suppression [11
, 13
, 14
, 16
, 17
]. Conversely, adding sham APC to burned T cells shows little effect as compared with burned T cells alone. Changes in IFN- production in these cultures largely mirrored the proliferative response, although cytokine synthesis was never completely absent.
Previous studies indicated that the increased production of Th2 cytokines, especially IL-10, is involved with the defects in cellular immune responses after burn injury [7 , 8 ]. Here, we measured IL-10 concentration in the culture media to determine whether this cytokine could be correlated with a diminished, proliferative response of purified, burned T cells in MLR. Consistent with previous reports [7 , 8 , 21 ], purified, burned T cells secreted cumulatively more IL-10 over 4 days of culture in response to anti-CD3 stimulation (5.35±1.35 ng/ml for sham T cells vs. 12.53±4.05 ng/ml for burned T cell mean±SD derived from "area under the curve" with three animals in each group; P=0.042). Interestingly, sham and burned, purified T cells cumulatively produced comparable low amounts of IL-10 in MLR experiments (0.28±0.08 ng/ml for sham and 0.24±0.09 ng/ml for burn; n=3, P=0.514) over 4 days of culture. Therefore, the suppressed, proliferative response of purified, burned T cells in MLR cannot be explained by changes in IL-10 production.
Impaired T cell responses after burn are corrected by NOS inhibition in high-dose anti-CD3 stimulation but not in MLR culture
To assess whether the suppressive effects of burn injury were entirely attributable to NO production, splenocytes harvested 7 days after burn injury were subjected to MLR and anti-CD3 stimulation in the presence or absence of NOS inhibitor, NMMA. Consistent with previous results from experiments using anti-CD3, Con A, or PHA mitogen stimulation [13
, 14
, 16
, 17
], NMMA almost completely restored the proliferative capacity of burned splenocytes stimulated by anti-CD3 (Fig. 5A
or Table 4
). Indomethacin at concentrations sufficient to completely inhibit cyclooxygenase activity [22
] did not restore the proliferative capacity of these burned splenocytes, excluding PGE2 as a major mediator in the impaired T cell responses. Although adding NMMA to MLR cultures increased the proliferative response of burned splenocytes significantly, it also improved the MLR response of sham splenocytes in a similar manner. The peak value of burned splenocytes listed in Table 4
is about half of that of sham splenocytes, similar to the data shown in Table 1
. Therefore, adding NMMR did not correct the suppressed T cell response after burn injury.
 View larger version (33K): [in a new window] |
Figure 5. Time course of anti-CD3 or MLR stimulation with or without the NO inhibitor NMMA (500 µM), indomethacin (5 µM), or combinations of both. Plots are representative of results from three independent mice, and data points with error bar are mean ± SD from one typical animal performed in quadruplicate or quintuplicate.
|
|
View this table: [in a new window] |
Table 4. Impaired Splenocyte Proliferations after Burn is Corrected by NOS Inhibition in High-Dose Anti-CD3 Stimulation but Not in MLRa
|
, a potent inducer of NOS, was likely involved [14
, 17
]. To demonstrate that burn injury increased NO via an IFN--related pathway, lymphocytes from mice congenitally IFN-–/– were examined after sham or burn injury.
Time course data (Fig. 6
) from individual animals or the peak values from multiple animals (Table 5
) show that the anti-CD3-proliferative responses of splenocytes from sham and burned IFN-–/– mice were similar. In contrast, the peak of the MLR-proliferative response of splenocytes from burned IFN-–/– mice was approximately half of the peak response from sham IFN-–/– mice. Taken together, these results again suggest that the suppressive effects of burn injury on maximal anti-CD3-driven responses are predominantly attributable to NO expression by IFN--stimulated non-T cells. However, such effects explain at best only roughly half of the immunosuppression seen in the antigen-driven MLR.
 View larger version (15K): [in a new window] |
Figure 6. Time course of T cell proliferation from sham and burned IFN-–/– mice stimulated by anti-CD3 or MLR. Plots are representative of results from three independent mice, and data points with error bar are mean ± SD from one typical animal performed in quadruplicate or quintuplicate.
|
|
View this table: [in a new window] |
Table 5. NO-Mediated Inhibition of Lymphocyte Responses to High-Dose Anti-CD3 Stimulation after Burn Injury Is Driven by IFN-a
|
–/– mice suggest that NO produced by burned non-T populations is primarily responsible for burn-induced, impaired, anti-CD3-driven proliferation using saturating concentrations of anti-CD3 (2 µg/ml). However, the MLR results indicate that NO production by burned non-T populations is not the only pathway by which antigen-driven stimulation is affected following burn trauma. We therefore hypothesized that weakened primary signaling pathways masked by maximal stimulation of anti-CD3 and/or defective costimulatory pathways could also contribute to post-burn T cell dysfunction.
To test this hypothesis, a submaximal concentration of anti-CD3 (0.1 µg/ml), alone or added with anti-CD28, was used to coat microtitor wells, and splenocytes from sham or burned IFN-–/– mice were used to evaluate proliferation to avoid the confounding variable of IFN--induced NO production. The results (Fig. 7
and Table 6
) show an approximate 50% reduction in anti-CD3-induced proliferation following burn injury; this is comparable with that observed in MLR using burned T cells in the setting of NOS inhibition (Fig. 5B)
or IFN- blockage (Fig. 6B)
. The data suggest that saturating concentrations of anti-CD3 obscure the defects in post-burn anti-CD3-driven pathways. When anti-CD28 was added to submaximal anti-CD3, the proliferative responses of sham and burned splenocytes increased in a similar manner. The measured proliferation peak values of burned splenocytes were still 60% less. These data suggest that the CD28 costimulatory pathway is intact following burn injury and that the impaired, CD3-driven pathway cannot be rescued by the addition of anti-CD28.
 View larger version (40K): [in a new window] |
Figure 7. Time course of T cell proliferation from sham and burned IFN-–/– mice stimulated by submaximal anti-CD3 (wells precoated with 0.1 µg/ml anti-CD3) and anti-CD3 + anti-CD28 (precoated with 0.1 µg/ml anti-CD3 and 2 µg/ml anti-CD28). Plots are representative of results from three independent mice, and data points with error bar are mean ± SD from one typical animal performed in quadruplicate or quintuplicate.
|
|
View this table: [in a new window] |
Table 6. Defective Anti-CD3 Pathways after Burn Uncovered by Submaximal Anti-CD3 Stimulation and the Effect of Anti-CD28 Additiona
|
In the present study, two different in vitro stimulation assays were used to determine the proliferation and IFN- cytokine production of T cells following burn trauma. As demonstrated previously [2
, 14
, 17
, 23
], markedly depressed, proliferative responses and depressed IFN- production was observed at all time-points following anti-CD3 stimulation. Notably, burn-induced effects on proliferation and cytokine production were only apparent at later time-points (Days 3–5) in MLR (Figs. 1B
and 1C
4C
and 5B
). These results illustrate the need to conduct a kinetic analysis to obtain the true burn effects on T cell response for any given stimulus.
The time-course data offer an explanation for the conflicting results in the literature about the effects of burn on IFN- production. Schwacha and Somers [17
] found an increase in IFN- production in burned splenocytes during Con A stimulation in one study but later reported no significant difference in IFN- production from splenocytes of sham and burned mice in response to Con A or anti-CD3 [11
, 21
, 24
]. Results from another group [25
] showed that a 15% TBSA injury increased IFN- production from splenocyte during Con A stimulation. In contrast, two other papers reported suppressed IFN- production from PBMC or splenocytes from burn patients or mice subjected to 20% or 40% TBSA burn following stimulation with Con A or bacterial mitogen [7
, 23
]. These discrepancies may be attributable to the timing of when IFN- was assayed from the cultures.
Data in Figure 1
indicate similar IFN- production at culture time Day 2 and suppressed IFN- release from activated T cells starting at Day 3 in MLR culture. At earlier culture time-points (Days 1 and 2), T cells from burned animals may actually release equal or even greater amounts of IFN-. At the later culture time-points, there is more accumulated NO produced from burn-primed macrophages, and the affected T cells would lose the ability to make IFN-.
In anti-CD3 stimulation experiments, removing macrophages (Fig. 4
or Table 3
), adding NOS inhibitor (Fig. 5
or Table 4
), and using mice congenitally IFN-–/– (Fig. 6
or Table 5
) all largely or completely restored T cell-proliferative responses. In contrast, similar interventions in MLR cultures failed to restore proliferative and cytokine responses. These results suggest that other pathways are affected. We hypothesize that the anti-CD3 stimulation experiments mask a more general defect in costimulation and/or CD3 signaling pathways as a result of the use of saturating concentrations of anti-CD3 mAb. To specifically test this hypothesis, anti-CD3 stimulation was performed at submaximal levels, with or without costimulation provided by anti-CD28 mAb. Splenocytes from IFN-–/– animals were used to eliminate the confounding variables related to IFN-- and NO-induced inhibition. In the absence of anti-CD28, the proliferative response with the submaximal anti-CD3 closely mirrored those seen for MLR. These data are consistent with the hypothesis that impaired T cell responses in MLR are attributable to diminished anti-CD3 signaling pathways following burn injury, which can be at least partially compensated by anti-CD28 costimulation. Indeed, the phosphorylated levels of Erk1/2 and p38 (MAPK signaling pathways) in T cells have been shown to be down-regulated in anti-CD3-stimulated T cells from burned animals [26
]. The similar elevation of sham and burned T cell-proliferative response by anti-CD28 also implies that CD28 costimulation pathways are not affected significantly by burn injury.
In addition, we evaluated whether IL-10 production was involved with the diminished proliferative response of purified, burned T cells in MLR experiments. The measured data show purified, burned T cells secreted more IL-10 in response to anti-CD3 stimulation, with proliferative responses similar to sham T cells. However, comparable low levels of IL-10 were observed in MLR experiments using purified sham or burned T cells, with the latter showing less proliferation. Therefore, the poor performance of T cells from burned animals in MLR cannot be explained by IL-10 production or a Th2 shift. In fact, an in vitro antigen stimulation study from Lederer’s group has shown that burn injury causes a similar suppression of lymphocyte responses (decreased proliferation and less production of IL-2 and IFN-) in wild-type and IL-10 KO mice [27
]. It should be noted that the measured concentration of IFN- in the culture media is 20 times more than that of IL-10, and IFN- would be the dominant cytokine during antigen stimulation.
We found that NO produced during lymphocyte activation by the saturating concentrations of anti-CD3 does not necessarily affect the proliferation of T cells from sham (normal) animals; the maximal proliferative value for unfractionated splenocytes in Table 1
is similar to that for purified T cells in Figure 4A
, to splenocytes with NMMA in Table 4
, and to IFN-–/– splenocytes in Table 5
. In contrast to the substantial elevation of burn T cell proliferation to the anti-CD3 stimulation by adding NMMA, such treatment had little effect on the sham T cells (Table 4)
. All these data suggest that NO generated from sham splenocytes during anti-CD3 stimulation is below the threshold of suppressing proliferation, whereas the amount of NO from burned splenocytes is greatly above the threshold. In comparison, the amount of NO generated using sham splenocytes as responders is high enough to suppress T cell MLR responses; the maximal, proliferative values in Tables 3
4
5
are six to nine times of that in Table 1
.
In summary, we have demonstrated—in agreement with other work—that burned splenocytes have diminished responses to direct CD3 ligation, involving pathways driven through IFN- and NO. However, burn-induced inhibition of T cell responses to authentic antigenic stimulation (in this case, a MLR) cannot be entirely attributed to IFN- and NO-mediated signals. In addition, diminished MLR (antigen-driven signaling) is also not a result of defective APC function. Indeed, the observation that suboptimal anti-CD3 proliferation in burned splenocytes can be augmented with anti-CD28 ligation suggests that burn injury attenuates signaling pathways downstream of CD3 cross-linking. Thus, the isolated IFN- or NO blockade may not correct the immunocompromised state occurring in burn injury; rather, effective therapy may require modifying signaling pathways distal to CD3 ligation.
ACKNOWLEDGEMENTS
This work was supported by the Shriners Hospitals for Children (Grant 8700) and a special shared Core Facility for Genomics and Proteomics at the Boston Shriners Burns Hospital.
FOOTNOTES
1 These authors contributed equally to this work. 
2 Current address: Biochemistry Department, Temple University School of Medicine, Philadelphia, PA 19140, USA.
Received April 17, 2007; revised September 10, 2007; accepted September 21, 2007.
REFERENCES
in the induction of the nitric oxide-synthesizing pathway J. Immunol. 147,144-148[Abstract] production in thermally injured mice Cytokine 12,1669-1675[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
