Paul E. Stromberg

Overexpression of Bcl-2 in the intestinal epithelium improves survival in septic mice.

Objectives The aim of this study was to determine whether decreasing intestinal epithelial apoptosis in sepsis would alter mortality rates. The roles of the antiapoptotic protein Bcl-2 and the “executioner” protease caspase-3 in sepsis-induced gut cell death also were evaluated. Design Prospective, randomized, controlled trial. Setting Animal laboratory in an academic medical center. Interventions Transgenic mice that overexpress Bcl-2 throughout the small intestinal epithelium (n = 23) and littermate controls (n = 27) were subjected to cecal ligation and puncture (CLP) and followed for 8 days to assess survival. A second group of transgenic (n = 15) and littermate animals (n = 15) were subjected to CLP and were killed between 16 and 48 hrs postoperatively to assess for intestinal apoptosis and active caspase-3 staining. Measurements and Main Results Survival of transgenic animals was 83% 8 days after CLP compared with 44% for littermate controls (p < .005). Survival curves between the two groups of animals began diverging within 24 hrs. Overexpression of Bcl-2 was associated with a significant decrease in apoptosis between 16 and 24 hrs post-CLP (p < .05) as well as decreased staining for active caspase-3. Conclusions Decreasing intestinal epithelial cell death via overexpression of Bcl-2 improves survival in septic mice. The gut may play a central role in the pathophysiology of sepsis.

Effects of aging on the immunopathologic response to sepsis.

Objective:Aging is associated with increased inflammation following sepsis. The purpose of this study was to determine whether this represents a fundamental age-based difference in the host response or is secondary to the increased mortality seen in aged hosts. Design:Prospective, randomized controlled study. Setting:Animal laboratory in a university medical center. Subjects:Young (6–12 weeks) and aged (20–24 months) FVB/N mice. Interventions:Mice were subjected to 2 × 25 or 1 × 30 cecal ligation and puncture (CLP). Measurements and Main Results:Survival was similar in young mice subjected to 2 × 25 CLP and aged mice subjected to 1 × 30 CLP (p = 0.15). Young mice subjected to 1 × 30 CLP had improved survival compared with the other groups (p < 0.05). When injury was held constant but mortality was greater, both systemic and peritoneal levels of tumor necrosis factor-α, interleukin (IL)-6, IL-10, and monocyte chemotactic protein-1 were elevated 24 hours after CLP in aged animals compared with young animals (p < 0.05). When mortality was similar but injury severity was different, there were no significant differences in systemic cytokines between aged mice and young mice. In contrast, peritoneal levels of tumor necrosis factor-α, IL-6, and IL-10 were higher in aged mice subjected to 1 × 30 CLP than young mice subjected to 2 × 25 CLP despite their similar mortalities (p < 0.05). There were no significant differences in either bacteremia or peritoneal cultures when animals of different ages sustained similar injuries or had different injuries with similar mortalities. Conclusions:Aged mice are more likely to die of sepsis than young mice when subjected to an equivalent insult, and this is associated with increases in both systemic and local inflammation. There is an exaggerated local but not systemic inflammatory response in aged mice compared with young mice when mortality is similar. This suggests that systemic processes that culminate in death may be age independent, but the local inflammatory response may be greater with aging.

Sepsis from Pseudomonas aeruginosa pneumonia decreases intestinal proliferation and induces gut epithelial cell cycle arrest

OBJECTIVES To evaluate whether the up-regulation in sepsis-induced gut epithelial apoptosis is balanced by an increase in intestinal proliferation and to assess mechanisms affecting the guts regenerative response to overwhelming infection. DESIGN Prospective, randomized, controlled study. SETTING Animal laboratory in a university medical center. INTERVENTIONS Mice were subjected to intratracheal injection of Pseudomonas aeruginosa and killed between 1.5 and 24 hrs after induction of pneumonia-induced sepsis to assess for gut epithelial proliferation and cell division and for apoptosis. Animals were compared with sham-operation controls, septic transgenic mice that overexpress Bcl-2 throughout their small intestinal epithelium, and septic p53-/- mice. MEASUREMENTS AND MAIN RESULTS Proliferation and cell division were assessed by measuring S-phase and M-phase cells in intestinal crypts. The number of S-phase cells showed a progressive decline at all time points measured, with a 5-fold decrease in proliferation between control animals and septic mice 24 hrs after intratracheal injection of pathogenic bacteria (p <.0001). In contrast, cells in M-phase remained constant for the first 12 hrs after the onset of sepsis, but increased nearly 50% at 24 hrs after instillation of P. aeruginosa (p <.005). Both the decrease in S-phase cells and the increase in M-phase cells were partially suppressible in Bcl-2 overexpressors, but cellular proliferation and division were similar between septic p53-/- and p53+/+ mice. Crypt apoptosis was increased at all time points, with maximal death occurring between 12 and 24 hrs. CONCLUSIONS Sepsis from P. aeruginosa pneumonia induces a p53-independent decrease in gut epithelial proliferation. Despite an increase in sepsis-induced intestinal apoptosis, there is no compensatory increase in intestinal epithelial proliferation, and there is evidence of a cell cycle block with an accumulation of cells in M-phase. Decreasing gut apoptosis by overexpression of Bcl-2 is associated with a partial reversal of the effect of sepsis on the cell cycle.

CD4+ lymphocytes control gut epithelial apoptosis and mediate survival in sepsis

Lymphocytes help determine whether gut epithelial cells proliferate or differentiate but are not known to affect whether they live or die. Here, we report that lymphocytes play a controlling role in mediating gut epithelial apoptosis in sepsis but not under basal conditions. Gut epithelial apoptosis is similar in unmanipulated Rag‐1–/– and wild‐type (WT) mice. However, Rag‐1–/– animals have a 5‐fold augmentation in gut epithelial apoptosis following cecal ligation and puncture (CLP) compared to septic WT mice. Reconstitution of lymphocytes inRag‐1–/– mice via adoptive transfer decreases intestinal apoptosis to levels seen in WT animals. Subset analysis indicates that CD4+ but not CD8+, γδ, or B cells are responsible for the antiapoptotic effect of lymphocytes on the gut epithelium. Gut‐specific overexpression ofBcl‐2 in transgenic mice decreases mortality following CLP. This survival benefit is lymphocyte dependent since gut‐specific overexpression ofBcl‐2 fails to alter survival when the transgene is overexpressed inRag‐1mice. Further, adoptively transferring lymphocytes toRag‐1mice that simultaneously overexpress gut‐specificBcl‐2 results in improved mortality following sepsis. Thus, sepsis unmasks CD4+ lymphocyte control of gut apoptosis that is not present under homeostatic conditions, which acts as a key determinant of both cellular survival and host mortality.—Stromberg, P.E., Wool‐sey, C.A., Clark, A.T., Clark, J.A., Turnbull, I.R., McConnell, K.W., Chang, K.C., Chung, C.‐S., Ayala, A., Buchman, T.G., Hotchkiss, R.S., Coopersmith, C.M. CD4+ lymphocytes control gut epithelial apoptosis and mediate survival in sepsis. FASEB J. 23, 1817–1825 (2009)

High-dose exogenous iron following cecal ligation and puncture increases mortality rate in mice and is associated with an increase in gut epithelial and splenic apoptosis

Objectives:Despite having dysregulated iron metabolism, critically ill patients may receive exogenous iron for the treatment of anemia. Iron is associated with increased tissue apoptosis and may facilitate bacterial growth. We hypothesized that exogenous iron administration given after the onset of sepsis would lead to increased mortality rate. To discriminate between elevated cell death and bacterial overgrowth as potential mediators of mortality, we examined gut epithelial and lymphocyte apoptosis and systemic bacterial counts in animals given iron supplementation after the onset of sepsis. Design:Prospective, randomized, controlled study. Setting:Animal laboratory in a university medical center. Subjects:Male C57BL/6 mice, 6–10 wks old. Interventions:C57BL/6 mice were subjected to cecal ligation and puncture (CLP), a well-accepted model of intra-abdominal sepsis, followed by daily subcutaneous injections of either 1 mL of iron dextran (5 mg/mL) or 0.9% NaCl for a total of five doses. Animals (n = 78) were followed for survival for 8 days. Separate cohorts (n = 76) were killed 24 or 48 hrs after cecal ligation and puncture or sham laparotomy and were assayed for gut epithelial and splenic apoptosis as well as for quantitative blood cultures. Measurements and Main Results:Eight-day survival was 7% in animals that received iron and 26% in mice that received 0.9% NaCl (p < .005). Iron supplementation after cecal ligation and puncture increased apoptosis by both active caspase 3 and hematoxylin and eosin staining in both the intestinal epithelium and spleen at 24 hrs (p < .05). Iron supplementation after sham laparotomy did not cause mortality or elevated apoptosis. Quantitative blood cultures revealed no detectable differences between septic animals that received iron and those that received 0.9% NaCl. Conclusions:High-dose iron supplementation with iron dextran after the onset of sepsis significantly increases mortality rate in this animal model. Iron-induced mortality may be mediated by an increase in gut epithelial and splenic apoptosis, whereas severity of bacteremia does not appear to play a causative role.

Antibiotics improve survival and alter the inflammatory profile in a murine model of sepsis from Pseudomonas aeruginosa pneumonia.

Differing antibiotic regimens can influence both survival and the inflammatory state in sepsis. We investigated whether the addition and/or type of antimicrobial agent could effect mortality in a murine model of Pseudomonas aeruginosa pneumonia-induced sepsis and if antibiotics altered systemic levels of cytokines. FVB/N mice were subjected to intratracheal injection of pathogenic bacteria and were given gentamicin, imipenem, or 0.9% NaCl 2 h after surgery, which continued every 12 h for a total of six doses. Survival at 7 days (n = 24 in each group) was 100% for mice given gentamicin, 88% for mice given imipenem, and 8% for sham mice treated with 0.9% NaCl (P < 0.0001). Systemic interleukin (IL) 6 levels were assayed 6 h postoperatively on all mice to see if they were predictive of outcome. Plasma IL-6 levels above 3600 pg/mL were associated with a 100% mortality, levels under 1200 pg/mL were associated with a 100% survival, and levels between 1200 and 3600 pg/mL had no utility in predicting mortality. In a separate experiment, mice were sacrificed at 3, 6, 12 or 24 h after instillation of P. aeruginosa and were assayed for levels of TNF-&agr;, IL-6, IL-10, and IL-12. Significant alterations in the proinflammatory cytokines TNF-&agr; and IL-6 were present at all time points except 3 h between mice treated with antibiotics and sham controls. In contrast, statistically significant differences in the anti-inflammatory cytokine IL-10 were present between the groups only at 6 h, and levels of IL-12 were similar at all time points. These results indicate that both gentamicin and imipenem increase survival at least 10-fold in a model of pneumonia-induced monomicrobial sepsis, and this is predominantly associated with a down-regulation of proinflammatory cytokines.

Bcl-2 inhibits gut epithelial apoptosis induced by acute lung injury in mice but has no effect on survival.

Gut epithelial apoptosis is increased in human studies and animal models of noninfectious inflammation and sepsis. Elevated intestinal cell death appears to be physiologically significant in sepsis. Previous studies demonstrate that overexpression of the antiapoptotic protein Bcl-2 in the gut epithelium of transgenic mice is associated with improved survival from Pseudomonas aeruginosa pneumonia and cecal ligation and puncture. The functional significance of elevated gut apoptosis in noninfectious inflammation has not been examined. We hypothesized that intestinal apoptosis would be detrimental to survival in noninfectious critical illness. To address this issue, acute lung injury (ALI) was induced with intratracheal injection of lipopolysaccharide (LPS, 800 &mgr;g) in wild-type (WT) FVB/N mice and transgenic mice that overexpress Bcl-2 in their intestinal epithelium. Guts were harvested at 12, 24, 48, and 72 h and assessed for apoptosis by both hematoxylin and eosin and active caspase-3 staining in 100 contiguous crypts. ALI increased gut epithelial apoptosis 12 h after LPS instillation compared with shams (P < 0.01), whereas overexpression of Bcl-2 decreased intestinal apoptosis compared with WT animals with ALI when assayed by active caspase-3 (P < 0.05). Plasma levels of tumor necrosis factor alpha, interleukin (IL)-6, and IL-10 were similar between WT and transgenic animals with ALI, both of which had elevated IL-10 levels at 12 h and elevated IL-6 levels at 24 h compared with sham animals. In a separate experiment, transgenic and WT animals with ALI were followed for mortality to determine whether gut overexpression of Bcl-2 conferred a survival advantage. Survival at 10 days was 73% in WT animals (n = 33) and 65% in Bcl-2 animals (n = 23, P = ns). These results indicate that while gut epithelial apoptosis is elevated in multiple models of critical illness, prevention of intestinal cell death by overexpression of Bcl-2 is associated with a disparate survival effect between sepsis and noninfectious inflammation.

Mechanisms of decreased intestinal epithelial proliferation and increased apoptosis in murine acute lung injury

Objectives:The aim of this study was to determine the effects of acute lung injury on the gut epithelium and examine mechanisms underlying changes in crypt proliferation and apoptosis. The relationship between severity and timing of lung injury to intestinal pathology was also examined. Design:Randomized, controlled study. Setting:University research laboratory. Subjects:Genetically inbred mice. Interventions:Following induction of acute lung injury, gut epithelial proliferation and apoptosis were assessed in a) C3H/HeN wild-type and C3H/HeJ mice, which lack functional Toll-like receptor 4 (n = 17); b) C57Bl/6 mice that received monoclonal anti-tumor necrosis factor-α or control antibody (n = 22); and c) C57Bl/6 wild-type and transgenic mice that overexpress Bcl-2 in their gut epithelium (n = 21). Intestinal epithelial proliferation and death were also examined in animals with differing degrees of lung inflammation (n = 24) as well as in a time course analysis following a fixed injury (n = 18). Measurements and Main Results:Acute lung injury caused decreased proliferation and increased apoptosis in crypt epithelial cells in all animals studied. C3H/HeJ mice had higher levels of proliferation than C3H/HeN animals without additional changes in apoptosis. Anti-tumor necrosis factor-α antibody had no effect on gut epithelial proliferation or death. Overexpression of Bcl-2 did not change proliferation despite decreasing gut apoptosis. Proliferation and apoptosis were not correlated to severity of lung injury, as gut alterations were lost in mice with more severe acute lung injury. Changes in both gut epithelial proliferation and death were apparent within 12 hrs, but proliferation was decreased 36 hrs following acute lung injury while apoptosis returned to normal. Conclusions:Acute lung injury causes disparate effects on crypt proliferation and apoptosis, which occur, at least in part, through differing mechanisms involving Toll-like receptor 4 and Bcl-2. Severity of lung injury does not correlate with perturbations in proliferation or death in the gut epithelium, and acute lung injury-induced changes in intestinal epithelial proliferation persist longer than those in apoptosis.

Sequence makes a difference: paradoxical effects of stress in vivo.

In vitro studies have shown that induction of heat shock before an inflammatory stimulus is cytoprotective, whereas induction of heat shock after an inflammatory stimulus can lead to apoptosis (the “heat shock paradox”). We sought to determine whether induction of the heat shock response in vivo caused similar, order-dependent effects on survival, and if so, by what mechanism. ND4 and C57BL/6 mice were used to calibrate the response to hyperthermia at 41.5°C via induction of inducible heat shock protein 70. Sequences of heat shock and septic stresses were studied in murine models of hyperthermia (41.5°C for 20 min) and cecal ligation and puncture (CLP), respectively. Previous heat shock to 41.5°C did not protect CLP mice when compared with control CLP animals heated to 37°C, but heat shock increased mortality when activated after CLP compared with controls. This effect of heat shock on CLP mortality was strain independent, and did not involve alterations in CLP-induced thymus, spleen, or intestinal apoptosis. We conclude that the heat shock paradox can occur in vitro and in vivo, and that the negative effects of heat shock on survival after CLP appeared to be strain independent. Furthermore, the stress of general anesthesia and warming also altered CLP mortality unexpectedly. The cellular mechanisms responsible for these “stressor” paradoxes in vivo are not known, but do not involve altered sepsis-induced apoptosis.

Myocardial transcriptional profiles in a murine model of sepsis: evidence for the importance of age.

Background: Age influences outcome of sepsis and septic shock. The mechanism of this age-dependent vulnerability to sepsis remains largely unknown. Because much of the mortality and morbidity associated with sepsis and septic shock is the result of severe derangements in the cardiovascular system, it is possible that the myocardium responds to injury in a developmentally influenced manner. We hypothesized that analysis of cardiac RNA expression profiles may differentiate between the myocardial response to sepsis in young and old mice. Methods and Results: Sixteen FVB/N male mice were stratified based on age. Young animals were 6 wks old, correlating to 4 to 6 human years, and aged animals were 20 months old correlating to 70 to 80 human years. Animals underwent either cecal ligation and puncture to produce polymicrobial sepsis or a sham operation. Both ventricles were excised after kill at 24 hrs. There were 53 genes that differed in RNA abundance between the four groups (false discovery rate of 0.005, p < 0.00001). Additionally, four genes were associated with an age-dependent response to sepsis: CYP2B2 (cytochrome P450, family 2, subfamily B, polypeptide 6), VGLL2 (vestigial like 2), and PAH (phenylalanine hydroxylase). The fourth gene is an expressed sequence tag, the function of which is related to the cytochrome P450 family. These genes play roles in phenylalanine, tyrosine, tryptophan, and fatty acid metabolism. Conclusions: This report describes the transcriptional response of the heart to sepsis. In addition, our findings suggest that these differences are in part age-dependent and serve as hypothesis generation.