Changes in memory and naive CD4+ lymphocytes in lymph nodes and spleen after thermal injury.

T cells expressing CD4 can be further subdivided by their expression of CD45R. In humans, CD45RO+ cells function as memory T cells, and CD4+CD45RA+ function as naive T cells, i.e., cells that have never previously encountered antigen. In the rat, similar functional division of CD4 cells can be made with antibody OX-22 (anti-CD45RC) with CD4+CD45RC- and CD4+CD45RC+ identifying the memory and naive phenotypes respectively. With use of the rat model, we studied the effect of burn injury on CD4 cell subpopulations and memory and naive T cells in lymph nodes draining the burn wound (NDB) and spleen. In the NDB we found that numbers of memory and naive CD4 cells increased significantly as a function of time after burn injury, and that this increase was greater for naive cells (5.7-fold) than for memory cells (4.1-fold). However, in the spleen, we found memory T-cell numbers decreased significantly on postburn days 12 and 18, whereas naive CD4 cells in the spleen manifested no changes at any time after burn. These results may be explained by the different trafficking patterns of memory and naive cells. Based on the cell surface marker, L-selectin, naive cells preferentially migrate into the NDB where they undergo differentiation into memory cells. Memory cells selectively migrate into inflamed burned tissue, which may account for their lesser-fold increase in NDB when compared to naive cells, and their decreased numbers in the spleen after burn injury.

Changes in polymorphonuclear leukocyte surface and plasma bactericidal/permeability-increasing protein and plasma lipopolysaccharide binding protein during endotoxemia or sepsis.

OBJECTIVE: To evaluate changes in levels of polymorphonuclear leukocyte surface bactericidal/permeability-increasing protein (BPI), plasma BPI, and plasma lipopolysaccharide (LPS) binding protein (LBP) in normal human volunteers administered Escherichia coli LPS and in patients with sepsis and gram-negative infections. DESIGN: Survey; case series. SETTING: Clinical research center and surgical intensive care unit of a medical school and an associated tertiary care hospital. PATIENTS OR OTHER PARTICIPANTS: Volunteers (n = 10) screened prior to study by history and physical examination to exclude those with underlying diseases or hematologic abnormalities. Consecutive sample of surgical intensive care unit patients (n = 10) meeting criteria for sepsis syndrome with gram-negative infection. An additional patient with systemic inflammatory response syndrome but no gram-negative infection. All patients were studied on meeting the criteria. Three of the patients with sepsis syndrome and the patient with systemic inflammatory response syndrome were evaluated on recovery (approximately 25 days after initial study). Because these studies in volunteers and patients overlapped temporally, the control values were those of volunteers evaluated prior to LPS administration. No matching was employed. MEASUREMENTS AND RESULTS: Compared with controls, LPS-challenged volunteers and patients with sepsis both exhibited significant granulocytosis (P < .01) and increased concentrations of polymorphonuclear leukocyte surface BPI (P < .01) and of plasma LBP (P < .01). Plasma BPI concentrations were increased (P < .01) in volunteers following LPS administration. There was a trend toward increased concentrations of plasma BPI in patients, but this was not significant relative to controls. Maximum concentrations of plasma LBP were approximately 250- and 3000-fold higher than plasma BPI concentrations in endotoxemic volunteers and in patients, respectively. CONCLUSIONS: Circulating polymorphonuclear leukocytes increase expression of BPI in response to LPS or gram-negative sepsis. Subsequently, concentrations of plasma BPI and LBP increase. Because both LBP and BPI bind to LPS, it is suggested that endogenously derived plasma levels of BPI are likely to be inadequate to compete for LPS binding to the much more abundant LBP in the circulation.

Effect of combined cortisol-endotoxin administration on peripheral blood leukocyte counts and phenotype in normal humans.

We studied the role of lipopolysaccharide and the associated hypercortisolemic response as mediators of leukocyte changes associated with endotoxemia. Normal human subjects were given continuous, 12-hour, intravenous infusions of cortisol. After 6 hours of cortisol infusion, lipopolysaccharide (20 U/kg) was administered in an intravenous bolus. Plasma cortisol and blood leukocyte counts and lymphocyte subset proportions were evaluated every hour throughout the 12-hour study period. After 6 hours of cortisol infusion, lymphocyte counts and proportions of CD4+ helper/inducer T cells had declined significantly. The fact that these cells did not decline further in response to lipopolysaccharide and continued cortisol infusion suggests that lipopolysaccharide-induced lymphocyte changes are cortisol dependent. In contrast, the granulocytosis normally observed after lipopolysaccharide administration was unaffected by cortisol infusion. Finally, the monocyte counts and proportions of B cells (HLA-DR+ or CD20+ cells) responded to cortisol infusion and LPS in a pattern distinct from that of lipopolysaccharide alone. These results indicate that lipopolysaccharide-induced hypercortisolemia plays a role in immune modulation during endotoxemia.

Changes in lymphocyte number and phenotype in seven lymphoid compartments after thermal injury.

Thermal injury is associated with dysfunction of host defense systems. The present study used flow cytometric immunofluorescence analyses to investigate changes in number and phenotype of lymphocytes in seven different lymphoid compartments at 2, 6, 12, 24, 48, and 60 days after 50% total body-surface area thermal injury in the rat. Relative to sham-injured control rats, at postburn day 2, significant lymphopenia was observed in the peripheral blood along with depletion of lymphocytes from the spleen and thymus. By day 6 after injury, lymphocytes in the bone marrow and cervical lymph nodes decreased significantly while numbers in the spleen and thymus remained depressed. Splenic and cervical node lymphocyte numbers normalized by day 12, the bone marrow and thymus numbers still were significantly lower than control, and a 6.5-fold increase in number of lymphocytes was observed in the nodes draining the burn wound, pooled axillary, brachial, inguinal, and lumbar lymph nodes. At day 24 after injury, the thymus and bone marrow virtually were depleted of lymphocytes, the mesenteric lymph nodes manifested a significant decrease, and lymphocytes in the nodes draining the burn wound continued to increase in number. This same pattern was maintained on day 48, but numbers of lymphocytes in the mesenteric nodes normalized. At day 60 after injury, lymphocyte numbers in all tissues were normalized, but the spleen and nodes draining the burn wound where increased numbers compared to control persisted. Cell-surface phenotyping was performed on all lymphoid tissues at all time intervals to determine the percentages of lymphocytes comprising the following subsets: Ia+ cells (B cells and activated T cells), T cells, T-Helper/Inducer cells (T-H/I), and T-Suppressor/Cytotoxic (T-S/C) cells. Although changes in lymphocyte subset percentages were complex, they could be divided grossly into two phases. First, all compartments showed significant phenotypic changes in the first six days after burn. With the exception of the nodes draining the burn wound and the blood, this was followed by a return towards normal on day 12. The second phase then ensued with significant phenotypic changes again occurring in most tissues from days 24 to 60 after injury. These studies demonstrate that burn injury results in dramatic alterations in lymphocyte numbers and subset percentages in different lymphoid compartments. Immune alterations observed following thermal injury may be due, in part, to a redistribution of the cellular elements responsible for generation of the immune response.

Flow cytometric analysis of lymphocyte subpopulations after thermal injury in human beings.

Several theories have been advanced in an effort to explain immunologic suppression after thermal injury, that is monocyte production of immunoregulatory prostaglandins, activation of suppressor cells, production of suppressive serum factors and alteration in helper cell function. In the current study, cytofluorometric analysis was performed on peripheral blood mononuclear cells isolated from 30 severely burned individuals using a FACS IV cell sorter. Fluorescein labeled monoclonal antibodies were used to phenotype total T cells (OKT3+), helper cells (OKT-4+), suppressor cells (OKT-8+), monocytes (antimono.2+) and B cells (anti-Ia+). After burn injury, the most striking phenotypic alterations observed were a marked decrease in the number and percentage of total OKT3+ T cells and OKT4+ helper cells. No significant increases were observed in the OKT8+ suppressor cell subpopulation. Monocytes exhibited a transitory increase during the first 48 hours postburn which returned to normal by postburn day 7. The percentage of Ia+ cells were either normal or decreased in number during the course of the injury. An OKT-4 to OKT-8 ratio of less than 1.00 at 24 to 48 hours postburn may represent a reliable predictive index for death by sepsis. These data suggest that the syndrome of burn induced immunologic suppression may be better described as a "burn induced immunodeficiency syndrome," that is characterized by decreased numbers or function of Interleukin-2 producing helper cells, or both.