Mary L. Rodrick

The effects of injury on the adaptive immune response

For more than thirty years it has been apparent that serious injury in humans and experimental animals is associated with a decrease in immune functions dependent upon T cells, the principal cells involved in initiating adaptive immune responses. This review focuses on more recent evidence that T helper cell function is altered after serious injury with loss of T helper 1 function and cytokine production and with preservation of T helper 2 function and an increased production of T helper 2 cytokines. Emphasis is placed on the importance of interactions between the innate and adaptive immune systems in the perturbed immune responses seen following injury. Immunomodulatory strategies are mentioned that have had success in animal models in ameliorating the diminished resistance to infection commonly seen after major traumatic or thermal injury. Finally, it is emphasized that immunomodulatory treatments that are successful in preventing infection may be contraindicated once infection is manifest.

The immunologic response to injury.

Research on the immune consequences of shock and trauma by multiple laboratories over more than 20 years has resulted in the following paradigm, which is currently accepted by most investigators in this field: serious traumatic or thermal injury is quickly followed, after initial resuscitation, by the systemic inflammatory response syndrome (SIRS) which, in a sizeable minority of patients, will lead inexorably to multiple organ dysfunction syndrome (early MODS), with an attendant high mortality. The majority of seriously injured patients survive the initial SIRS response without developing early MODS, and after a period of relative clinical stability, manifest a compensatory antiinflammatory response syndrome (CARS) with suppressed immunity and diminished resistance to infection. Resultant infection and its attendant inflammation in turn can lead to multiple organ dysfunction (late MODS) and death (Fig. 1). This paradigm has several implications of potential importance in interpreting the sometimes conflicting results of research in this area:

Major injury induces increased production of interleukin-10 by cells of the immune system with a negative impact on resistance to infection.

OBJECTIVEnThe purpose of this study was to compare the production of interleukin-10 (IL-10) by peripheral blood mononuclear cells (PBMC) from injured patients and control subjects to determine the responsible cell types and to relate IL-10 production to the occurrence of sepsis. A mouse model of burn injury was used to confirm the human findings and to assess the importance of IL-10 in the lowered resistance to infection after injury.nnnSUMMARY BACKGROUND DATAnSevere injury is associated with depressed immune responses. Although IL-10 is known to inhibit several aspects of immune reactivity, the role of IL-10 in postinjury immune suppression remains controversial.nnnMETHODSnPeripheral blood mononuclear cells from 14 burn and 12 trauma patients and 16 healthy individuals were studied at serial intervals for IL-10 production stimulated by a T-cell mitogen, phytohemagglutinin, and by bacterial lipopolysaccharide. To determine the source of IL-10, CD4+ and CD8+ lymphocyte subsets were obtained by selective depletion of PBMC with antibody-coated magnetic beads and were stimulated by anti-CD3 antibody to induce IL-10 secretion. In addition, IL-10 production by patients PBMC in the first 10 days after injury was assessed for correlation with subsequent septic events. Anti-CD3-stimulated IL-10 production also was determined for CD4- and CD8-enriched lymphocyte subsets obtained by antibody and complement depletion of splenocytes harvested from groups of burn and sham burn mice at day 10 after injury, the time of maximal susceptibility to a septic challenge, cecal ligation and puncture (CLP). Finally, to test the importance of IL-10 in immune suppression in vivo, groups of burn and sham burn mice were treated with anti-IL-10 monoclonal antibody or control immunoglobulin G (IgG) on days 1 and 3 postinjury and were observed for survival after CLP on day 10.nnnRESULTSnPatients PBMC produced significantly more IL-10 than did controls PBMC 7 to 14 days after injury. Patients CD4+ (T-helper) but not CD8+ (T-cytotoxic) lymphocytes also showed increased IL-10 production versus those of control subjects early after injury. Increased PBMC IL-10 production in the first 10 days postinjury correlated significantly (p < 0.05) with subsequent septic events. Burn mouse CD4-enriched but not CD8-enriched splenocytes produced more IL-10 than did sham burn splenocyte subsets on day 10 after injury. Burn mice treated with anti-IL-10 antibody but not with control IgG had significantly increased survival after CLP.nnnCONCLUSIONnSerious injury in humans and in a mouse burn model is followed by increased stimulated production of IL-10 by cells of the immune system. The CD4+ T-helper cells appear to be a major source of IL-10 after injury. In injured patients, increased IL-10 production is correlated with subsequent septic events, and in the burn mouse, IL-10 appears to induce decreased resistance to infection.

Depression of cellular immunity after multiple trauma in the absence of sepsis

We have previously reported that severe burn injury was regularly accompanied by impaired lymphocyte responses to T cell mitogens, circulating suppressor lymphocytes, and serum factors suppressive of lymphocyte activation. However, in burned patients it was difficult to determine whether these manifestations of suppressed immunity were predictive of, or the result of, sepsis which was ubiquitous in this population. In an attempt to clarify this issue, we have studied 31 patients with multiple trauma (without burns) mean age, 31 years; average injury severity score, 22; range, 9-56; in whom sepsis was less common. Patients were tested for lymphocyte response to the T cell mitogens PHA and Con A, the percentage of circulating putative suppressor (OKT8) and helper (OKT4) T cells using monoclonal antibodies, circulating suppressor cell activity as revealed by functional assays, and serum suppression of lymphocyte activation. Patients were compared with ten normal volunteers (mean age, 32) studied simultaneously. Significant suppression (greater than 50% compared with controls) in lymphocyte responses to mitogens 1 to 5 days after injury was seen in 12 patients, was accompanied by a shift in the ratio of helper (OKT4) to suppressor (OKT8) T cells (patients, 0.96:1; normals, 1.82:1; p less than 0.01), and was followed by the appearance of significant (greater than 50%) serum suppressive activity in six of the 12 patients. Circulating suppressor cell activity as revealed by functional assays was also seen early after injury in three of 12 patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Neutrophil Activation in Thermal Injury as Assessed by Increased Expression of Complement Receptors

We studied neutrophil activation in patients with burns by serial immunofluorescent measurement of neutrophil expression of the complement opsonin receptors CR1 and CR3. CR1-dependent fluorescence was initially (days 0 through 5 after the burn) elevated (mean +/- SEM, 294 +/- 42 vs. 63 +/- 6 in the controls; P less than 0.001) and gradually returned to normal (days 6 through 8, 270 +/- 62, P less than 0.001; days 9 through 13, 185 +/- 38, P less than 0.001; days 14 through 19, 143 +/- 27, P less than 0.001; and days 20 through 50, 93 +/- 5, P less than 0.04). CR3-dependent fluorescence paralleled that of CR1. Neutrophil chemotaxis in response to zymosan-activated serum, a source of C5a, was depressed (days 0 through 5, 77 +/- 4 percent of control, P less than 0.001; days 6 through 8, 70 +/- 4 percent, P less than 0.001; days 9 through 13, 74 +/- 3 percent, P less than 0.001; days 14 through 19, 90 +/- 4 percent, P less than 0.01; and days 20 through 50, 97 +/- 3 percent, P not significant) and inversely correlated with CR1- and CR3-dependent fluorescence (r = -0.559, P less than 0.001; and r = -0.709, P less than 0.001, respectively). Plasma C3a desArg levels were above normal (100 +/- 5 ng per milliliter) throughout (days 0 through 5, 305 +/- 42; days 6 through 8, 546 +/- 69; days 9 through 13, 490 +/- 72; days 14 through 19, 409 +/- 54; and days 20 through 50, 260 +/- 36; all P less than 0.005). Thus, neutrophils in burned patients were activated as indicated by increased expression of complement receptors. The correlation between this increase and the depression of chemotaxis in response to zymosan-activated serum suggests that C5a is responsible for systemic neutrophil activation, which may contribute to the increased susceptibility to infection of patients with burns.

Injury Induces Deficient Interleukin-12 Production, But Interleukin-12 Therapy After Injury Restores Resistance to Infection

OBJECTIVEnTo assess at serial intervals the production of interleukin-12 (IL-12) by monocytes/macrophages from the peripheral blood of injured patients and control subjects, and using a mouse model to confirm human findings and explore the effectiveness of low-dose IL-12 therapy in restoring resistance to infection after injury.nnnSUMMARY BACKGROUND DATAnSerious injury is associated with loss of function of the T helper 1 lymphocyte phenotype, but little is known about IL-12 production in injured patients. The authors previously reported that early, moderate-dose IL-12 therapy in a mouse model of burn injury restored resistance to a later infectious challenge (cecal ligation and puncture, CLP). However, the efficacy of clinically relevant low-dose IL-12 therapy carried out to or beyond the time of septic challenge remains to be tested.nnnMETHODSnPeripheral blood mononuclear cells (PBMCs) and adherent cells were obtained from 27 patients with major burns or traumatic injury and 18 healthy persons and were studied at serial intervals for IL-12 production stimulated by bacterial lipopolysacharide (LPS). PBMCs from 18 of the same patients were studied for IL-10 production as well. IL-12 production by adherent cells from the spleens of burn or sham burn mice was studied at serial intervals after injury to confirm the human findings. Low-dose IL-12 or vehicle was given every other day to groups of burn and sham burn mice, which were then challenged with CLP on day 10, and survival was determined. Finally, spleens were harvested from burn or sham burn animals receiving low-dose IL-12 or vehicle after CLP. After splenic cellularity was determined by hemocytometer, splenocytes were cultured and production of tumor necrosis factor-alpha, interferon-gamma, and IL-10 were assessed by immunoassay.nnnRESULTSnAdherent cells from patients PBMCs produced significantly less IL-12 than normal PBMCs after injury, reaching a nadir 8 to 14 days after injury. Stimulation of whole PBMCs by LPS indicated that at 8 to 14 days after injury, IL-12 production by PBMCs was significantly lower and IL-10 production was significantly higher than that of PBMCs from healthy persons. Low-dose IL-12 therapy significantly increased survival after CLP. Splenocytes from burn mice treated with IL-12 had significantly increased production of TNF-alpha and IF-beta, both before and after CLP, when compared with vehicle-treated burn animals. IL-10 production by bum splenocytes remained high after IL-12 treatment. Splenic cellularity increased after IL-12 treatment in burn mice.nnnCONCLUSIONnThe capacity to produce IL-12 by adherent cells of the monocyte/macrophage lineage is significantly reduced after serious injury in humans and in a mouse burn model. In humans, there is a reciprocal relation between diminished IL-12 production and increased IL-10 production at approximately 1 week after injury. Low-dose IL-12 therapy in the mouse burn model markedly increased survival after a septic challenge, even when treatment was carried beyond the onset of sepsis. Low-dose IL-12 treatment in the mouse increased production of proinflammatory mediators important in host defense and at the same time maintained or increased production of IL-10, an important antiinflammatory cytokine.

Interleukin-12 treatment restores normal resistance to bacterial challenge after burn injury

BACKGROUNDnPreliminary studies in this laboratory have shown that treatment with interleukin-12 (IL-12), a cytokine that induces expression of the T-helper-1 lymphocyte phenotype, in an animal model of burn injury increased survival after a septic challenge. The purpose of this study was to define the efficacy of IL-12 therapy and to explore its mechanism of action.nnnMETHODSnAdult male A/J mice were subjected to 25% full-thickness scald or sham burn. Starting on day 3 after burn, groups of mice received five daily injections of IL-12, interferon-gamma (IFN-gamma), or saline solution. Some animals received anti-IFN-gamma monoclonal antibody. At day 10 most animals underwent cecal ligation and puncture (CLP) and were observed for survival. Some animals were killed at day 10, and CD4-enriched splenocytes were stimulated with anti-CD3 antibody or concanavalin A and were studied for cytokine production and mRNA expression.nnnRESULTSnIL-12 treatment, 25 ng daily for 5 days, increased survival of the burn group after CLP to that of the sham burn control group. Anti-IFN-gamma antibody, 500 micrograms, administered 1 day before IL-12 treatment, reduced the efficacy of IL-12. IFN-gamma treatment, 7000 units, moderately increased survival. IL-12 had no effect on survival of the sham burn group. At the time of CLP IL-12 therapy had induced a marked decrease in CD4+ lymphocyte IL-4 and a moderate increase in IFN-gamma production and mRNA expression without affecting IL-2.nnnCONCLUSIONSnIL-12 is the most effective therapy so far tested in this burn plus CLP model. It acts at least in part through IFN-gamma. However, IFN-gamma therapy was not as effective as IL-12.

The role of prostaglandin E2 in immune suppression following injury.

It has been thought for some time that prostaglandin E2 (PGE2) released from activated monocytes/macrophages may contribute to the suppression of immunity seen after burns and major injury because PGE2 inhibits the activation of T lymphocytes. To clarify this issue, we studied 15 patients with total body surface area burns of 20% to 90% (mean, 48%). Peripheral blood mononuclear cells (PBMC) were obtained from these patients one to two times each week for 1 month after burn and were stimulated with the T-cell mitogen phytohemagglutinin (PHA). On 14 occasions the PBMCs from eight patients were significantly suppressed (30% or more) in their response to PHA (suppressed [sup] burn) as compared with PBMCs from normal controls. In 38 instances PBMCs from 12 patients were not significantly suppressed in PHA (nonsuppressed [nonsup] burn). Sup burn PBMCs and control PBMCs were cultured with or without the addition of the cyclooxygenase (CO) inhibitor indomethacin (Indo, 1 microgram/mL) and studied for PHA response and the production of the stimulatory cytokine interleukin-2 (IL-2). Indo partially restored the PHA response of sup burn PBMCs to normal. Sup burn PBMCs also were deficient in production of IL-2. Indo increased IL-2 production by sup burn PBMCs significantly more (160% +/- 20%, p less than 0.005) than control (57% +/- 5%) and nonsup PBMCs (67% +/- 8%). Next inhibition of the PHA response of PBMCs from 12 burn patients and 17 controls was studied by exogenous PGE2. At all time periods after burn injury, patients PBMCs were significantly more sensitive to inhibition by PGE2 (50% inhibition at 10(-8) mol/L [molar] PGE2) than PBMCs from normal controls (50% inhibition at 10(-6) mol/L PGE2) with maximum sensitivity occurring 8 to 14 days after injury. Peripheral blood mononuclear cells from patients with more than 40% burns were significantly (p less than 0.05) more sensitive to PGE2 than those from patients with lesser burns. Interleukin-2 was added to cultures of sup burn PBMC, nonsup burn PBMC, and controls containing 10(-7) mol/L PGE2. Interleukin-2 totally reversed PGE2 inhibition of the PHA response in PBMC from both controls and burn patients. Because endotoxin leak from the gut has been implicated as a trigger for a number of the metabolic and immunologic abnormalities following injury, the authors looked for the effect of a bolus infusion of Escherichia coli endotoxin (Endo, 4 ng/kg) in seven normal healthy volunteers on the response of PBMC to PHA and on the production of PGE2 and IL-2.(ABSTRACT TRUNCATED AT 400 WORDS)

Is circulating endotoxin the trigger for the systemic inflammatory response syndrome seen after injury

OBJECTIVEnPatients with severe traumatic or burn injury and a mouse model of burn injury were studied early after injury to determine the relation of plasma endotoxin (lipopolysaccharide [LPS]) to the production of proinflammatory cytokines and subsequent resistance to infection.nnnSUMMARY BACKGROUND DATAnElevated levels of plasma LPS have been reported in patients after serious injury. It has been suggested that circulating LPS may be a trigger for increased proinflammatory cytokine production and may play a role in the septic syndromes seen in a substantial portion of such patients. Yet, despite multiple reports of leakage of LPS from the gut and bacterial translocation after injury in animal models, there is little direct evidence linking circulating LPS with production of inflammatory mediators.nnnMETHODSnThe authors studied serial samples of peripheral blood from 10 patients with 25% to 50% surface area burns and 8 trauma patients (injury Severity Score, 25-57). Patients were compared with 18 healthy volunteers. The study was focused on the first 10 days after injury before the onset of sepsis or the systemic inflammatory response syndrome. Plasma samples were assayed for LPS, and adherent cells from the blood were studied for basal and LPS-stimulated production of tumor necrosis factor-alpha (TNF-alpha), interleukin-1 beta (IL-1 beta), and interleukin-6 (IL-6). The correlation of increased plasma LPS with TNF-alpha production was studied as was the association of increased plasma LPS and increased TNF-alpha production with subsequent septic complications. We also studied a mouse model of 25% burn injury. Burn mice were compared with sham burn control subjects. Plasma samples were assayed at serial intervals for LPS, and adherent cells from the spleens were studied for basal- and LPS-stimulated production of TNF-alpha, IL-1 beta, and IL-6. Expression of the messenger RNAs for IL-1 beta and TNF-alpha also was measured. The relation of increased TNF-alpha production with mortality from a septic challenge, cecal ligation and puncture (CLP), was determined. Finally, the effect of administration of LPS to normal mice on subsequent mortality after CLP and on TNF-alpha production was studied.nnnRESULTSnElevated plasma LPS (> 1 pg/mL) was seen in 11 of the 18 patients within 10 days of injury and in no normal control subjects. In this period, patients as compared with control subjects showed increased stimulated production of TNF-alpha, IL-1 beta, and IL-6. Increased TNF-alpha production was not correlated with elevated plasma LPS in the same patients. Neither increased plasma LPS nor increased TNF-alpha production early after injury was correlated with subsequent development of systemic inflammatory response syndrome or sepsis in the patients. Burn mice, as compared with sham burn control subjects, showed elevated plasma LPS levels chiefly in the first 3 days after injury. Increased stimulated production of proinflammatory cytokines by adherent splenocytes from the burn mice also was seen at multiple intervals after injury and did not correlate with mortality from CLP. Increased production of TNF-alpha and IL-1 beta was associated with increased expression of messenger RNAs for these cytokines. Finally, two doses of 1 ng LPS administered 24 hours apart to normal mice had no effect on mortality from CLP performed 7 days later nor on the production of TNF-alpha at the time of CLP.nnnCONCLUSIONSnThese findings call into question the idea that circulating LPS is the trigger for increased proinflammatory cytokine production, systemic inflammatory response syndrome, and septic complications in injured patients.

Suppression of natural killer-cell function in humans following thermal and traumatic injury.

Depressed cell-mediated and humoral immune functions have been reported to occur following severe thermal and traumatic injury. In this study we have questioned whether another immune function, natural killing (NK), is also disturbed in these injured patients. Twenty-two thermally injured patients with burns ranging from 5 to 75% of the total body surface area and 15 traumatically injured patients with injury severity scores ranging from 9 to 56 were followed postinjury and compared to 29 age-matched controls. NK activity was measured as the percentage cytotoxicity in chromium-51 release assays with K562 target cells. The more severely burned patients had significantly depressed NK activity for the 40-day period following injury that remained reduced for the duration of the study. Patients with lesser burns had reduced NK-cell function for the initial 10-day period postburn that returned slowly to the normal range. Traumatically injured patients had depressed NK-cell function during the 3- to 6-day period postinjury. The percentage of cells bearing phenotypic markers for the groups in which Nk cells are found was either normal or elevated in these patients. A correlation was found between NK activity and interleukin 2 generation by mononuclear cells from these patients. In order to investigate the mechanism of NK suppression in these patients, NK-cell function was studied following the infusion of cortisol, epinephrine, and glucagon into volunteer subjects in amounts known to reproduce serum levels seen following injury of moderate severity. NK-cell function was reduced an average of 66% following infusion, suggesting that the inhibition of NK-cell function seen in patients may be mediated by the stress response to injury.