Sepsis-induced apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans.

Patients with sepsis have impaired host defenses that contribute to the lethality of the disorder. Recent work implicates lymphocyte apoptosis as a potential factor in the immunosuppression of sepsis. If lymphocyte apoptosis is an important mechanism, specific subsets of lymphocytes may be more vulnerable. A prospective study of lymphocyte cell typing and apoptosis was conducted in spleens from 27 patients with sepsis and 25 patients with trauma. Spleens from 16 critically ill nonseptic (3 prospective and 13 retrospective) patients were also evaluated. Immunohistochemical staining showed a caspase-9-mediated profound progressive loss of B and CD4 T helper cells in sepsis. Interestingly, sepsis did not decrease CD8 T or NK cells. Although there was no overall effect on lymphocytes from critically ill nonseptic patients (considered as a group), certain individual patients did exhibit significant loss of B and CD4 T cells. The loss of B and CD4 T cells in sepsis is especially significant because it occurs during life-threatening infection, a state in which massive lymphocyte clonal expansion should exist. Mitochondria-dependent lymphocyte apoptosis may contribute to the immunosuppression in sepsis by decreasing the number of immune effector cells. Similar loss of lymphocytes may be occurring in critically ill patients with other disorders.

Overexpression of Bcl-2 in transgenic mice decreases apoptosis and improves survival in sepsis.

In sepsis there is extensive apoptosis of lymphocytes, which may be beneficial by down-regulating the accompanying inflammation. Alternatively, apoptosis may be detrimental by impairing host defense. We studied whether Bcl-2, a potent antiapoptotic protein, could prevent lymphocyte apoptosis in a clinically relevant model of sepsis. Transgenic mice in which Bcl-2 was overexpressed in T cells had complete protection against sepsis-induced T lymphocyte apoptosis in thymus and spleen. Surprisingly, there was also a decrease in splenic B cell apoptosis in septic Bcl-2 overexpressors compared with septic HeJ and HeOuJ mice. There were marked increases in TNF-alpha, IL-1beta, and IL-10 in thymic tissue in sepsis in the three species of mice, and the increase in TNF-alpha and IL-10 in HeOuJ mice was greater than that in Bcl-2 mice. Mitotracker, a mitochondrial membrane potential indicator, demonstrated a sepsis-induced loss of membrane potential in T cells in HeJ and HeOuJ mice but not in Bcl-2 mice. Importantly, Bcl-2 overexpressors also had improved survival in sepsis. To investigate the potential impact of loss of lymphocytes on survival in sepsis, Rag-1-/- mice, which are totally deficient in mature T and B cells, were also studied. Rag-1-/- mice had decreased survival compared with immunologically normal mice with sepsis. We conclude that overexpression of Bcl-2 provides protection against cell death in sepsis. Lymphocyte death may be detrimental in sepsis by compromising host defense.

Effect of Ca2+ agonists in the perfused liver: determination via laser scanning confocal microscopy.

Ca2+ is a critical intracellular second messenger, but few studies have examined Ca2+ signaling in whole organs. The amplitude and frequency of Ca2+ oscillations encode important cellular information. Using laser scanning confocal microscopy in the indo 1 acetoxymethyl ester dye-loaded rat liver, we investigated the effect of various Ca2+ agonists that act at distinct mechanistic sites on Ca2+ signaling. Perfusion with suprathreshold doses of arginine vasopressin (AVP) (2-20 nM) caused a single Ca2+ wave that originated in the pericentral vein region and spread centrifugally to the periportal area. Lower doses of AVP (0.2-2 nM) caused multiple Ca2+ waves and Ca2+ oscillations. Perfusion with ATP (1. 4-17.5 microM) caused rapid transient elevations in intracellular free Ca2+ concentration ([Ca2+]i) occurring in isolated hepatocytes or groups of hepatocytes throughout the lobule and were of shorter duration than those due to AVP. Also in contrast to AVP, there was no specific anatomic location within the hepatic lobule that was more susceptible to ATP. Thapsigargin and cyclopiazonic acid did not cause a Ca2+ wave but rather produced a uniform and fairly simultaneous increase in [Ca2+]i in all hepatocytes in the lobule. Perfusion with 14 microM ryanodine produced a single transient spike in [Ca2+]i in a small number (<2%) of hepatocytes. Dantrolene, an inhibitor of Ca2+ release, reduced the increased [Ca2+]i occurring after AVP. Insight into the mechanism of action of these Ca2+-active compounds on Ca2+ signaling in the intact liver is provided.

Stress-induced fractal rearrangement of the endothelial cell cytoskeleton causes apoptosis.

BACKGROUND: Apoptosis, a mechanism of cell death prominent in critical illnesses including disseminated inflammation and multiorgan dysfunction syndrome, is characterized by morphologic changes including cell shrinkage, condensation of organelles, blebbing, and chromatin fragmentation. These phenomena suggest substantial changes in cytoskeletal structure. We hypothesized that stress-induced apoptosis in endothelial cells is, in part, a consequence of a critical cytoskeletal rearrangement. METHODS: Porcine aortic endothelial cells in culture, surrogates for the microvasculature in vivo, were exposed sequentially to Escherichia coli endotoxin (25 micrograms/mL; 18 hours) to induce the inflammatory response and then to sodium arsenite (160 mumol/L; 120 minutes) to induce the heat-shock response, a well-characterized model of stress-induced apoptosis. Laser confocal micrographs of fluorescein isothio-cyanate-labeled phalloidin-stained cells were analyzed to calculate the border fractal dimension of the cytoskeleton. Other cells were exposed to cytochalasin D, a fungal metabolite, which interferes with polymerization of actin from its globular to its filamentous form, and similarly were analyzed with respect to fractal dimension, viability (neutral red assay), and manner of death (annexin V fluorescence-activated cell scanning analysis). RESULTS: Induction of the inflammatory or heat-shock responses caused subtle and distinct rearrangement of the actin cytoskeleton. When these stimuli were applied in sequence, a synergistic interaction led to profound cytoskeletal collapse. Reversal of the sequence did not induce the cytoskeletal disruption. Cytochalasin D alone induced a dose-dependent cytoskeletal collapse indistinguishable from that caused by the acute phase-heat shock sequence that caused cell death by apoptosis. The effect of lower doses of Cytochalasin D could be potentiated by subsequent induction of the heat-shock response. CONCLUSIONS: Sequential stresses that mimic pathophysiologic "two-hit" stimuli induce a characteristic fractal rearrangement of the actin cytoskeleton. Because cytochalasin D-induced rearrangement of this cytoskeleton produced apoptosis indistinguishable from the stress-induced apoptosis, we conclude that the cytoskeletal rearrangement is likely a critical event in the pathway to apoptosis. This disruption of intracellular interconnections mirrors endotoxin-induced disruption in signals among organs and supports the mechanistic hypothesis that multiorgan dysfunction syndrome generally reflects disruption of signals and connections at several levels of biologic organization.

Calcium antagonists inhibit oxidative burst and nitrite formation in lipopolysaccharide-stimulated rat peritoneal macrophages.

Activated macrophages are important cell effectors in sepsis/endotoxemia. Superoxide (SO) and nitric oxide (NO) are produced by activated macrophages and are responsible for host defense against microorganisms. Using laser scanning confocal microscopy, we investigated the role of intracellular free calcium ([Ca2+]i) on SO and NO production by rat peritoneal macrophages activated by lipopolysaccharide (LPS). Calcium influx from the extracellular space versus release of calcium from intracellular stores was determined using calcium channel blockers (diltiazem [DIL], verapamil [VER], and nicardipine [NIC]) and dantrolene (DAN), respectively. Cells incubated with LPS had a 30-50 nM increase in [Ca2+]i, (p < .05) compared with non-LPS-treated cells. When stimulated with phorbol myristate acetate, both control and LPS-treated cells sustained a comparable increase in [Ca2+]i, but [Ca2+]i, remained elevated 30 min later in LPS-treated cells. Calcium channel blockers and DAN reduced phorbol myristate acetate-stimulated SO and LPS-stimulated NO production at all concentrations tested (p < .05). Although increased extracellular calcium influx and calcium from intracellular stores are important regulators of SO and NO production in macrophages, extracellular calcium influx seems to have the predominant effect. Calcium antagonists may modulate the inflammatory response via their effects on macrophages.

Alanyl-glutamine prevents muscle atrophy and glutamine synthetase induction by glucocorticoids.

The aims of this work were to establish whether glutamine infusion via alanyl-glutamine dipeptide provides effective therapy against muscle atrophy from glucorticoids and whether the glucocorticoid induction of glutamine synthetase (GS) is downregulated by dipeptide supplementation. Rats were given hydrocortisone 21-acetate or the dosing vehicle and were infused with alanine (AA) or alanyl-glutamine (AG) at the same concentrations and rates (1.15 mumol.min-1.100 g body wt-1, 0.75 ml/h) for 7 days. Compared with AA infusion in hormone-treated animals, AG infusion prevented total body and fast-twitch muscle mass losses by over 70%. Glucocorticoid treatment did not reduce muscle glutamine levels. Higher serum glutamine was found in the AG-infused (1.72 +/- 0.28 mumol/ml) compared with the AA-infused group (1.32 +/- 0.06 mumol/ml), but muscle glutamine concentrations were not elevated by AG infusion. Following glucocorticoid injections, GS enzyme activity was increased by two- to threefold in plantaris, fast-twitch white (superficial quadriceps), and fast-twitch red (deep quadriceps) muscle/fiber types of the AA group. Similarly, GS mRNA was elevated by 3.3- to 4.1-fold in these same muscles of hormone-treated, AA-infused rats. AG infusion diminished glucocorticoid effects on GS enzyme activity to 52-65% and on GS mRNA to 31-37% of the values with AA infusion. These results provide firsthand evidence of atrophy prevention from a catabolic state using glutamine in dipeptide form. Despite higher serum and muscle alanine levels with AA infusion than with AG infusion, alanine alone is not a sufficient stimulus to counteract muscle atrophy. The AG-induced muscle sparing is accompanied by diminished expression of a glucocorticoid-inducible gene in skeletal muscle. However, glutamine regulation of GS appears complex and may involve more regulators than muscle glutamine concentration alone.

Glutamine interferes with glucocorticoid-induced expression of glutamine synthetase in skeletal muscle.

Skeletal muscle atrophy from glucocorticoids is prevented by glutamine infusion. Because the gene-encoding glutamine synthetase (GS) is glucocorticoid inducible, it represented an appropriate model for resting whether glucocorticoids and glutamine exert opposing actions on the expression of specific genes related to atrophy in muscle tissue. Rats were administered hydrocortisone 21-acetate or the dosing vehicle (carboxymethyl cellulose) and were infused with saline (Sal) or glutamine (Gln, 240 mM, 0.75 ml/h) for 7 days. Hormone treatment did not significantly lower glutamine levels in fast-twitch white or red regions of the quadriceps. Despite higher serum glutamine concentrations with amino acid infusion [1.52 +/- 0.03 (Gln) vs. 1.20 +/- 0.04 (Sal) mumol/ml], muscle glutamine concentrations were not markedly increased in these fiber types. In saline-infused animals, glucocorticoid treatment produced 200-300% increases in plantaris, fast-twitch white, and fast-twitch red muscle GS enzyme activity and mRNA. Moreover, in all muscle types studied, glutamine infusion diminished glucocorticoid effects on GS enzyme activity to 131-159% and on GS mRNA to 110-200% of the values in saline-treated controls. These data demonstrate that glutamine infusion results in inhibiting GS expression, but the absence of changes in muscle glutamine concentration suggests the interplay of additional regulators of the GS gene.

Calcium antagonists decrease plasma and tissue concentrations of tumor necrosis factor-alpha, interleukin-1 beta, and interleukin-1 alpha in a mouse model of endotoxin.

Calcium plays an important role in the toxic effects of endotoxin, and calcium antagonists also have been shown to improve survival in animals challenged with endotoxin. Calcium may be involved in regulating cytokine production. Therefore, the protective effect of calcium-antagonists in endotoxin may be due to decreased cytokine formation and/or systemic release. In a mouse model of endotoxin, dantrolene (10 mg/kg) and azumolene (20 mg/kg), drugs that decrease calcium release from intracellular stores, or diltiazem (20 mg/kg), a calcium channel blocker, decreased plasma tumor necrosis factor-alpha (TNF-alpha), interleukin-1 alpha (IL-1 alpha), and IL-1 beta (47.2, 63.2, and 62.4%, respectively, p < .05) when the animals were injected intraperitoneally with endotoxin. Dantrolene and azumolene decreased IL-1 alpha by 56.6 and 65.4%, respectively, (p < .05) and IL-1 beta by 51.7 and 69.7%, respectively (p < .05). Diltiazem had no effect on IL-1 alpha or IL-1 beta. Dantrolene decreased TNF-alpha in lung (26.1%), liver (29.4%), and spleen (35.4%) (p < .05) and IL-1 alpha in lung (30.0%) and liver (25.4%) (p < .05). The present findings indicate that calcium-antagonists may be efficacious in treating cytokine mediated inflammatory disorders.

Glucose metabolism in epitrochlearis muscle of acutely exercised and trained rats.

Effects of insulin on glycogen synthesis (GS), glycolytic utilization (GU), and glucose uptake (GT) were studied in isolated epitrochlearis muscles from exercise-trained or sedentary rats during recovery from acute exercise or at rest. During the 1st h after acute exercise, the enhanced basal and insulin-stimulated GT was directed mainly toward replenishment of glycogen but basal GU was also increased. During the second through third hours after exercise, basal GS decreased but remained greater than rest and basal GU and GT returned to normal. Insulin sensitivity of these parameters was enhanced. Training alone reduced basal GS but enhanced insulin sensitivity of GT and GU. Training reduced the acute exercise-stimulated increase in basal and insulin sensitivity of GS during recovery from acute exercise, probably due to elevated glycogen stores. Thus recovery from acute exercise or training, either alone or in combination, enhances insulin stimulated GT in muscle; however, the increased glucose is primarily channeled toward GS after acute exercise, which is reduced by prior training and is directed to GU in trained animals either at rest or after acute exercise.

Muscle protein turnover: effects of exercise training and renal insufficiency.

Chronic renal failure is associated with an enhanced catabolism of muscle protein. To determine the effect of exercise training and moderate renal insufficiency on net protein catabolism and protein synthesis in isolated epitrochlearis muscles, three-fourth nephrectomized and control rats were exercise trained or remained sedentary. Net muscle protein degradation was determined by measuring the rates of release of phenylalanine and tyrosine. Protein synthesis was determined by measuring the incorporation of [U-14C]phenylalanine into muscle protein. Exercise training reduced the elevated protein degradation of uremia to control levels. In control rats, exercise training had no effect on protein degradation. Exercise training increased alanine release in control rats but did not further increase the elevated alanine release of uremia. Protein synthesis was unaffected by both uremia and exercise training. Exercise training in control and uremic rats moderately increased the responsiveness of muscle to insulin by reducing net protein degradation but did not further enhance the insulin-stimulated increase in protein synthesis. Thus exercise training ameliorates the enhanced muscle protein degradation of moderate renal insufficiency.