Characterization of a novel hemoglobin-glutathione adduct that is elevated in diabetic patients.

BACKGROUND: Typically, a diagnosis of diabetes mellitus is based on elevated circulating blood glucose levels. In an attempt to discover additional markers for the disease and predictors of prognosis, we undertook the characterization of HbA1d3 in diabetic and normal patients. MATERIAL AND METHODS: PolyCAT A cation exchange chromatography and liquid chromatography-mass spectroscopy was utilized to separate the alpha- and beta-globin chains of HbA1d3 and characterize their presence in normal and diabetic patients. RESULTS: We report the characterization of HbA1d3 as a glutathionylated, minor hemoglobin subfraction that occurs in higher levels in diabetic patients (2.26 +/- 0.29%) than in normal individuals (1.21 +/- 0.14%, p < 0.001). The alpha-chain spectrum displayed a molecular ion of m/z 15126 Da, which is consistent with the predicted native mass of the HbA0 alpha-globin chain. By contrast, the mass spectrum of the beta-chain showed a mass excess of 307 Da (m/z = 16173 Da) versus that of the native HbA0 beta-globin chain (m/z = 15866 Da). The native molecular weight of the modified beta-globin chain HbA0 was regenerated by treatment of HbA1d3 with dithiothreitol, consistent with a glutathionylated adduct. CONCLUSIONS: We propose that HbA1d3 (HbSSG) forms normally in vivo, and may provide a useful marker of oxidative stress in diabetes mellitus and potentially other pathologic situations.

HMG-1 as a late mediator of endotoxin lethality in mice.

Endotoxin, a constituent of Gram-negative bacteria, stimulates macrophages to release large quantities of tumor necrosis factor (TNF) and interleukin-1 (IL-1), which can precipitate tissue injury and lethal shock (endotoxemia). Antagonists of TNF and IL-1 have shown limited efficacy in clinical trials, possibly because these cytokines are early mediators in pathogenesis. Here a potential late mediator of lethality is identified and characterized in a mouse model. High mobility group-1 (HMG-1) protein was found to be released by cultured macrophages more than 8 hours after stimulation with endotoxin, TNF, or IL-1. Mice showed increased serum levels of HMG-1 from 8 to 32 hours after endotoxin exposure. Delayed administration of antibodies to HMG-1 attenuated endotoxin lethality in mice, and administration of HMG-1 itself was lethal. Septic patients who succumbed to infection had increased serum HMG-1 levels, suggesting that this protein warrants investigation as a therapeutic target.

Induction of the chemokine beta peptides, MIP-1 alpha and MIP-1 beta, by lipopolysaccharide is differentially regulated by immunomodulatory cytokines gamma-IFN, IL-10, IL-4, and TGF-beta.

The macrophage occupies a central role in the host response to invasion, exerting its control over the developing inflammatory response largely through the elaboration of an assortment of endogenous mediators including many cytokines. The beta chemokine peptides, macrophage inflammatory protein [MIP]-1 alpha and MIP-1 beta, are two such effectors markedly up-regulated in macrophages following exposure to bacterial lipopolysaccharide (LPS). These highly homologous peptides, like the other members of the beta chemokine family, exhibit diverse but partially overlapping biological activity profiles, suggesting that the cellular participants and intensity of an inflammatory response may in part be regulated by selective expression of these chemokines. Studies reported here demonstrate that, in contrast to the "balanced" MIP-1 alpha/MIP-1 beta chemokine responses of LPS-stimulated macrophage cultures in vitro, circulating levels of MIP-1 beta are significantly higher than those of MIP-1 alpha following LPS administration in vivo. Further studies have revealed that several immunomodulatory cytokines known to be up-regulated in vivo as a consequence of exposure to an invasive stimulus (gamma-IFN, IL-10, IL-4, and transforming growth factor [TGF]-beta) down-regulated the LPS-induced release of MIP-1 alpha by macrophages in vitro, but spared the MIP-1 beta response. This altered pattern of secretion may explain, at least in part, the high circulating levels of MIP-1 beta relative to MIP-1 alpha observed in vivo in response to LPS challenge.

Nuclear localization signal of HIV-1 as a novel target for therapeutic intervention.

BACKGROUND: Human immunodeficiency virus type 1 (HIV-1) is a lentivirus and shares with other members of this retroviral subfamily the ability to replicate in nondividing cells, in particular, cells of the monocyte/macrophage lineage. This feature relies on the presence of a specific nuclear localization signal (NLS) within the viral matrix protein (MA p17), which to some degree can be complemented by the activity of the viral vpr gene product. The MA p17 NLS ensures efficient transportation of the viral preintegration complex into the nucleus of an infected macrophage and confers persistence of HIV-1 in quiescent T cells, and therefore presents an attractive target for therapeutic intervention. MATERIALS AND METHODS: Nuclear localization signals (NLS) in general and the HIV-1 MA p17 NLS in particular are characterized by a stretch of positively charged amino acids including one or more lysine residues. A series of compounds potentially capable of binding and reacting with lysine by forming Schiff base adducts was synthesized. Our special consideration was to make compounds that would preferentially bind to two closely contiguous amino functions, as opposed to isolated single lysine residues. We assumed that this approach might specifically target the compound to NLS while affecting other regions less, thus reducing nonspecific cytotoxicity. Antiviral activity was assessed in primary monocytes and in peripheral blood lymphocytes (PBL) infected with HIV-1ADA strain. Viral replication was monitored by reverse transcriptase (RT) activity in the supernatant. Efficiency of nuclear importation of the viral preintegration complex was estimated by the formation of 2-LTR circle forms of HIV-1 DNA and also by in situ PCR techniques. RESULTS: Arylene bis(methyl ketone) compounds with a nitrogenous third subsituent, especially a pyrimidinic side-chain, inhibited HIV-1 replication in human monocytes at an IC50 as low as 1 nM. These compounds did not block HIV-1 replication in peripheral blood lymphocyte cultures. The inhibitory effect observed in monocyte cultures appeared in the context of markedly reduced nuclear importation of viral DNA in the presence of the drug. No cytotoxic effects of the compounds was observed in vitro at concentrations as high as 10 microM. An amidinohydrazone derivative of the most active compound was about 100 times less active than the parent, indicating that carbonyl groups were instrumental in the antiviral effect. CONCLUSIONS: These early results suggest that retroviral replication in nondividing cells is susceptible to pharmaceutical intervention targeted against the NLS activity of HIV-1 proteins in the viral preintegration complex. The compounds described efficiently block translocation of viral DNA to the nuclei of infected primary monocytes, and inhibit viral replication. This inhibition is effective only in nondividing cells and is not seen in proliferating cultures, such as activated PBLs. Thus, drugs that target HIV-1 NLS may be useful to specifically block the macrophage arm of HIV infection and could thereby be of value in treating macrophage-specific manifestations of HIV disease, such as HIV-1 dementia. In combination with other drugs, potential therapeutics exploiting this target may also help to control the progression of HIV-1 infection and disease.

Reactive glycosylation endproducts in diabetic uraemia and treatment of renal failure.

In diabetes and ageing, glucose-derived advanced glycosylation endproducts (AGEs) cross-link proteins and cause vascular tissue damage. Elimination of circulating low-molecular weight AGE-modified molecules (LMW-AGEs) by the kidney is impaired in diabetic patients with end-stage renal disease, a group subject to accelerated atherosclerosis. We determined the effectiveness of current renal replacement treatments on elimination of serum LMW-AGEs in diabetic and non-diabetic patients with end-stage renal disease. Although diabetic patients receiving high-flux haemodialysis achieved 33% lower steady-state serum LMW-AGE than did those in conventional haemodialysis (p < 0.005), LMW-AGE concentrations remained 3.5-6 fold above normal, whether high-flux dialysis, conventional haemodialysis, or chronic ambulatory peritoneal dialysis were used. High-flux haemodialysis markedly reduced AGE during each treatment session (47.9% in the diabetic, p < 0.001 and 60.6% in the non-diabetic group, p < 0.001) but concentrations returned to pre-treatment range within 3 hours. In contrast, normal LMW-AGE concentrations were maintained in patients with functioning renal transplants. We found that LMW-AGEs with an apparent molecular weight of 2000-6000 circulate and retain strong inherent chemical reactivity--when exposed to collagen in vitro, up to 77% attached covalently to form AGE-collagen, and the AGE-crosslink inhibitor aminoguanidine completely inhibited this reaction. The results suggest that LMW-AGEs comprise a set of chemically-reactive molecules that are refractory to removal by current dialysis treatments. Through covalent reattachment onto vascular matrix or serum components, LMW-AGEs may exacerbate vascular pathology associated with end-stage renal disease.

Advanced glycation end products contribute to amyloidosis in Alzheimer disease.

Alzheimer disease (AD) is characterized by deposits of an aggregated 42-amino-acid beta-amyloid peptide (beta AP) in the brain and cerebrovasculature. After a concentration-dependent lag period during in vitro incubations, soluble preparations of synthetic beta AP slowly form fibrillar aggregates that resemble natural amyloid and are measurable by sedimentation and thioflavin T-based fluorescence. Aggregation of soluble beta AP in these in vitro assays is enhanced by addition of small amounts of pre-aggregated beta-amyloid "seed" material. We also have prepared these seeds by using a naturally occurring reaction between glucose and protein amino groups resulting in the formation of advanced "glycosylation" end products (AGEs) which chemically crosslink proteins. AGE-modified beta AP-nucleation seeds further accelerated aggregation of soluble beta AP compared to non-modified "seed" material. Over time, nonenzymatic advanced glycation also results in the gradual accumulation of a set of posttranslational covalent adducts on long-lived proteins in vivo. In a standardized competitive ELISA, plaque fractions of AD brains were found to contain about 3-fold more AGE adducts per mg of protein than preparations from healthy, age-matched controls. These results suggest that the in vivo half-life of beta-amyloid is prolonged in AD, resulting in greater accumulation of AGE modifications which in turn may act to promote accumulation of additional amyloid.

MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia.

Cytokines are critical in the often fatal cascade of events that cause septic shock. One regulatory system that is likely to be important in controlling inflammatory responses is the neuroendocrine axis. The pituitary, for example, is ideally situated to integrate central and peripheral stimuli, and initiates the increase in systemic glucocorticoids that accompanies host stress responses. To assess further the contribution of the pituitary to systemic inflammatory processes, we examined the secretory profile of cultured pituitary cells and whole pituitaries in vivo after stimulation with bacterial lipopolysaccharide (LPS). Here we identify macrophage migration inhibitory factor (MIF) as a major secreted protein release by anterior pituitary cells in response to LPS stimulation. Serum analysis of control, hypophysectomized and T-cell-deficient (nude) mice suggests that pituitary-derived MIF contributes to circulating MIF present in the post-acute phase of endotoxaemia. Recombinant murine MIF greatly enhances lethality when co-injected with LPS and anti-MIF antibody confers full protection against lethal endotoxaemia. We conclude that MIF plays a central role in the toxic response to endotoxaemia and possibly septic shock.

Cachectin/TNF-mediated lactate production in cultured myocytes is linked to activation of a futile substrate cycle.

The cytokine cachectin/TNF induces a rapid increase in lactate production and in glucose metabolism in L6 myocytes in culture; glucose uptake was maximal after 17 h, while elevated glucose utilization and lactate production persisted for up to 32 h. These increases are suggestive of increased glycolytic activity, and were associated with a 10% decrease in cellular oxygen consumption and a comparable decrease in the production of 14C-labelled CO2 from 14C-labelled glucose. This decrease in aerobic metabolism, however, could account for only a small fraction of the energetic requirement for increased glycolytic activity. Furthermore, maximal stimulation of pyruvate dehydrogenase (PDH) by dichloroacetate (DCA) treatment in conjunction with cachectin/TNF abolished lactate production, but increased glucose uptake persisted. Taken together, this suggests that the primary effect of cachectin/TNF on myocyte carbohydrate metabolism is to increase glycolysis. Correspondingly, we postulated that cachectin/TNF must activate one or more ATP-depleting cellular processes to account for the lack of feed-back inhibition on glycolysis by the ATP produced. This led to the identification of a futile substrate cycle between fructose 6-phosphate and fructose 1,6-bisphosphate as a novel energy sink that is activated by cachectin/TNF. Cachectin/TNF treatment led to increased activity of both phosphofructokinase (PFK) and fructose bisphosphate phosphatase (FBP) in myocytes in culture, detectable after 1 h of incubation and persisting for up to 16 h. The possible role of cachectin/TNF-mediated futile substrate cycling in increased glycolytic activity, increased energy expenditure, heat production and tissue wasting during bacterial infections is discussed.

Genomic cloning and promoter analysis of macrophage inflammatory protein (MIP)-2, MIP-1 alpha, and MIP-1 beta, members of the chemokine superfamily of proinflammatory cytokines.

Macrophage inflammatory protein (MIP)-2, MIP-1 alpha, and MIP-1 beta all belong to the newly recognized "chemokine" superfamily of structurally related, activation-inducible cytokines with inflammatory and growth regulatory activities. We report the isolation and sequencing of genomic clones for murine MIP-2 and murine MIP-1 beta, and analyze their regulatory sequences in comparison with each other and with several other members of the chemokine family. The murine (mu)MIP-2 genomic clone displays the canonical four exon/three intron structure typical of other genes in the chemokine alpha subfamily (e.g., IL-8). Potential cis regulatory elements in the proximal promoter region were highly conserved between muMIP-2 and its three most closely related human homologs: human (hu)GRO-alpha, huGRO-beta, and huGRO-gamma. A mouse macrophage cell line, RAW 264.7, was transfected with a growth hormone reporter construct driven by a proximal fragment of the muMIP-2 5' promoter, and nested deletion mutant analysis localized the LPS responsive element to a region that contains a conserved NF kappa B consensus motif and lies 51 to 70 bp 5' from the transcription start site. In contrast to that of MIP-2, the muMIP-1 beta genomic clone exhibited the three exon/two intron structure characteristic of the chemokine beta family members (e.g., MCP-1). A comparison of the promoters for muMIP-1 beta and muMIP-1 alpha reveals a conserved CK-1 element, but transient expression studies in RAW 264.7 macrophages with proximal fragments of either the muMIP-1 beta or the muMIP-1 alpha 5' promoter fused to a human growth hormone reporter gene link LPS-inducibility in both to promoter segments near to, but not identical with, the consensus CK-1 sequence. Proximal 5' promoter fragments cloned from both the MIP-1 alpha and MIP-1 beta genes unexpectedly conferred constitutive expression on the fused reporter gene sequences in macrophage-like cells, but initial 5' deletion analysis did not link this responsiveness to known sequence motifs. The muMIP-1 beta promoter, but not the muMIP-1 alpha promoter, was constitutively active in B16 mouse melanoma cells, and both promoters were active in the myelomonocytic cell line WEHI 3B(A)d-, the muMIP-1 alpha promoter being three times stronger.

Human decidua: a source of cachectin-tumor necrosis factor.

Cachectin-tumor necrosis factor (TNF-alpha) has been implicated as a possible signal for the initiation of human parturition in the setting of infection. These studies were conducted to determine whether human decidua can produce TNF-alpha in response to bacterial lipopolysaccharide (LPS). Decidual explants from women undergoing elective cesarean sections were incubated with and without Escherichia coli LPS (25 ng/ml) for 20 h. TNF-alpha concentration in the conditioned media was measured with an enzyme-linked immunoassay and bioassay (L929 bioassay). While conditioned media from unstimulated decidual explants contained either undetectable or low levels of TNF-alpha, conditioned media from LPS stimulated decidua contained TNF-alpha (mean = 2.6 pmol/mg protein per 20 hours, SEM +/- 1.03). There was a strong correlation between the immunoreactive and bioactive TNF-alpha (Spearman rank correlation r = 0.76, P less than 0.001). We conclude that human decidua in vitro can produce TNF-alpha in response to LPS.

Genomic structure of murine macrophage inflammatory protein-1 alpha and conservation of potential regulatory sequences with a human homolog, LD78.

The gene for a murine macrophage inflammatory cytokine, MIP-1 alpha, belongs to a newly recognized superfamily encoding small, inducible peptides shown to be up-regulated in association with cellular activation or transformation (tentatively designated the scy, or small cytokine, gene family). Secreted scy family peptides as a group, and MIP-1 alpha in particular, have inflammatory and mitogenic activities, and the family has been divided into CXC and CC subfamilies according to the spacing of conserved cysteine residues in the primary amino acid sequences. We have isolated and characterized a genomic clone encoding the CC subfamily member MIP-1 alpha. The organization of the murine MIP-1 alpha gene into three exons interrupted by two introns is identical to that found for other members of the CC subfamily (e.g., huLD78, muJE, huJE/MCP-1, muTCA3, and hul-309), which has been taken as evidence of evolution from a common ancestral gene. With the exception of the ratPF4 gene, which shares the two-intron/three-exon pattern typical of the CC subfamily, sequenced genes encoding CXC subfamily peptides (e.g., hulL-8 and hulP-10) include an additional intervening sequence that creates a fourth exon. Genomic nucleotide sequences 5' of the MIP-1 alpha cap site are highly homologous to corresponding regions of the human gene encoding a CC peptide variously designated as LD78/GOS19/pAT464, including consensus regulatory motifs in common, reinforcing the contention that MIP-1 alpha and LD78 may be interspecies homologs.

The role of cachectin/TNF and other cytokines in sepsis.

A traditional view has been that bacterial products, such as endotoxins from gram negative bacteria, have a direct deleterious effect on the host, resulting in fever, hypermetabolism, anorexia, and tissue damage. In recent years, however, it has been shown that endogenous products of the host, secreted by macrophages and other cellular elements of the immune system, act as mediators in activating the metabolic and other physiological changes characteristic of the sepsis syndrome. We will review in depth various aspects of the major, central mediator, i.e., tissue necrosis factor (TNF)/cachectin, and also briefly discuss the interleukins IL-6 and IL-1.

Metabolic effects of cachectin/tumor necrosis factor are modified by site of production. Cachectin/tumor necrosis factor-secreting tumor in skeletal muscle induces chronic cachexia, while implantation in brain induces predominantly acute anorexia.

We have developed a murine model of wasting by injecting intracerebrally cells which continuously secrete h-cachectin/TNF (CHO-TNF) to: (a) determine the effects of cachectin/TNF produced continuously in the central nervous system (CNS), and (b) compare the metabolic effects of cachectin/TNF-secreting tumor in the brain to the cachexia caused by CHO-TNF tumor in peripheral tissue (IM). Intracerebral CHO-TNF tumors produced increased serum h-cachectin/TNF levels with lethal hypophagia and weight loss (mean survival time of 11 d); these changes were not observed in association with nonsecretory control brain tumors. The metabolic consequences of intracerebral cachectin/TNF production were indistinguishable from acute, lethal starvation: whole-body lipid content was decreased significantly but protein was conserved. Although intramuscular cachectin/TNF-secreting tumors caused similar increases of serum h-cachectin/TNF levels, profound anorexia did not develop; wasting developed after a longer period of tumor burden (50 d) with classical signs of cachexia (i.e., anemia and depletion of both protein and lipid). These studies provide a reproducible animal model of site-specific cytokine production and suggest that, regardless of serum levels, cachectin/TNF produced locally in brain influences both the rate of development of wasting and its net metabolic effects.

TNF concentration in fatal cerebral, non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria.

Plasma levels of tumour necrosis factor (TNF) were significantly higher in 178 Gambian children with uncomplicated malaria due to Plasmodium falciparum than in 178 children with other illnesses. 110 children with cerebral malaria were studied shortly after admission to hospital; 28 subsequently died. Compared with the children with uncomplicated malaria, mean plasma TNF levels were twice as high in cerebral malaria survivors and ten times as high in the fatal cases. Although high TNF levels were associated with high parasitaemia and with hypoglycaemia, they predicted fatal outcome in cerebral malaria independently of parasitaemia and glucose concentrations. Concentrations of interleukin-1 alpha, but not interferon gamma, were also related to the severity of malaria. We conclude that increased TNF production is a normal host response to P falciparum infection, but that excessive levels of production may predispose to cerebral malaria and a fatal outcome.

Monocyte function in IDDM patients and healthy individuals.

Interleukin 1 beta (IL-1 beta) and tumour necrosis factor alpha (TNF-alpha) may be pathogenetically important in insulin-dependent diabetes mellitus (IDDM), which is associated with genes of the HLA region. Since a regulatory role of HLA region genes on monokine production may exist, we looked for an association between the monokine and prostaglandin E2 (PGE2) responses of monocytes (Mo) from 20 healthy males (18-50 years) with HLA-DR types relevant for IDDM susceptibility and resistance (DR1,2, DR1,3, DR1,4, DR3,4). Monokine assays were established and evaluated and the secretions of IL-1 beta, TNF-alpha, and PGE2 measured in Mo cultures (2h, 6h, 20h) prepared by endotoxin-free techniques and stimulated by low-dose E. coli lipopolysaccharides (LPS). There were no significant associations between Mo responses and HLA-DR phenotype. Likewise, Mo from DR2 (n = 5) and DR4 (n = 5) homozygous healthy males demonstrated no significant differences in monokine and PGE2 responses of Mo. In the HLA class III region a diallelic TNF-beta gene NcoI polymorphism consisting of alleles of 5.5 kb and 10.5 kb was recently described and associated with susceptibility to autoimmune diseases including IDDM. We report that IL-1 beta and TNF-alpha responses of Mo from TNF-beta 10.5 kb homozygous healthy individuals were significantly higher than for TNF-beta 5.5/10.5 kb heterozygotes. IL-1 beta and TNF-alpha responses of Mo from males (18-35 years) with newly diagnosed (n = 10) and long-standing IDDM (n = 10) and from age- and HLA-DR-matched healthy males (n = 10) were studied. LPS, gamma interferon (IFN), and TNF-alpha-stimulated Mo cultures were investigated. No significant differences were found between Mo responses of IDDM patients and controls. IFN (1000 U/ml) in the presence of LPS significantly potentiated LPS-stimulated Mo TNF-alpha secretion and reduced the levels of IL-1 beta immunoreactivity in Mo lysates. IFN and TNF-alpha did not have any effects on LPS-stimulated Mo secretion of IL-1 beta immunoreactivity. We conclude that Mo IL-1 beta and TNF-alpha production is normal in patients with recent-onset and long-standing IDDM. The interindividual differences in monokine responses may be accounted for by the diallelic human TNF-beta gene polymorphism rather than by HLA class II genes. This observation may be important for understanding the association of certain HLA haplotypes with autoimmune phenomena and disease.

Processing of newly synthesized cachectin/tumor necrosis factor in endotoxin-stimulated macrophages.

The biosynthesis and processing of cachetin/tumor necrosis-factor (TNF) were examined in the murine macrophage-like cell line RAW 264.7. Lipopolysaccharide-stimulated cells secreted both glycosylated and nonglycosylated 17-kilodalton (kDa) mature cachectin/TNF into the culture medium. Secreted cachectin/TNF was derived from membrane-associated precursors that were precipitated by polyclonal antisera raised against either the mature protein or synthetic peptide fragments of the 79 amino acid cachectin/TNF prohormone sequence. About half of the precursors were N-glycosylated, apparently cotranslationally. The cachectin/TNF precursors were then proteolytically cleaved to release soluble mature cytokine into the medium, while the membrane-bound 14-kDa presequence remained cell associated. During the period of LPS stimulation, the amount of macrophage cell surface cachectin/TNF remained at a low level, suggesting that both nonglycosylated and glycosylated precursors of cachectin/TNF are efficiently cleaved by these cells. These findings suggest the presence of a unique mechanism for the secretion of cachectin/TNF.

Cerebrospinal fluid cachectin/tumor necrosis factor-alpha and platelet-activating factor concentrations and severity of bacterial meningitis in children.

In prospective studies, tumor necrosis factor (TNF alpha) was detected in cerebrospinal fluid (CSF) of 33 of 38 children with bacterial meningitis (BM) but in none of 15 with viral meningitis/encephalitis (P less than .001). BM CSF TNF alpha (less than 35 to greater than 25,500 pg/ml) correlated with CSF bacterial density (P less than .01), CSF protein (P less than .001), endotoxin (LPS) in gram-negative disease (P less than .01), and consecutive febrile hospital days (P less than .001); initial CSF TNF alpha greater than 1000 pg/ml was associated with seizures (P less than .05). Only 5 children with BM (13%) had detectable plasma TNF alpha activity on admission. A higher proportion who died had detectable plasma TNF alpha activity compared with survivors (3/4 vs. 2/34, P less than .005). Platelet-activating factor (PAF) in CSF was higher in 19 children with Haemophilus influenzae meningitis than in 17 controls (P less than .01) and correlated with bacterial density (P less than .01), CSF LPS (P less than .01), CSF TNF alpha levels (P less than .01), and the Herson-Todd severity score (P less than .01). Elevated CSF TNF alpha and PAF are often present in children with BM and are associated with seizures and severity of disease. Detectable CSF TNF alpha appears to distinguish BM from viral meningitis.

Serum cachectin/tumor necrosis factor in critically ill patients with burns correlates with infection and mortality.

Serum cachectin/tumor necrosis factor (TNF), a cytokine implicated in the pathogenesis of septic shock, may appear in the circulation during serious infection, but the frequency of detection of elevated serum levels during protracted critical burn injury is unknown. Serial serum samples taken from 43 critically ill patients with burns with and without sepsis were analyzed for TNF using an enzyme-linked immunosorbent assay (ELISA). TNF was detectable (greater than 34 picograms per milliliter) at one or more time points in 69 per cent of the patients with sepsis versus 33 per cent of those without sepsis, in 71 per cent of the patients who died versus only 31 per cent of the survivors and in only one healthy normal control patient (n = 21). The frequency of the appearance of TNF correlated with both infection and mortality rate. Moreover, all three patients with TNF levels greater than 540 picograms per milliliter died. Neither the size of the burn nor injury from inhalation correlated with detection of TNF. A subset of 16 patients was studied longitudinally from admission until resolution of injury and these data demonstrate that TNF appears transiently and repetitively in the circulation of patients during infection and protracted critical burn injury. Also, serum cortisol levels were significantly higher during sepsis and death in the absence of serum TNF, compared with sepsis and death with detectable cachectin, suggesting that cortisol may interact with the production or detection of this cytokine during ongoing infection and lethal injury. In this study, we have demonstrated that serum TNF is detectable with greater frequency and in higher concentration in patients with sepsis and in those who ultimately succumb to the burn injury, that serum TNF appears transiently and repetitively in the circulation during injury and that higher serum cortisol levels are correlated with the absence of serum TNF during sepsis and lethal injury.