Distribution of TNF alpha-like proteins correlates with some regions of programmed cell death in the chick embryo.

Early chick embryos have previously been shown to express tumor necrosis factor-alpha-cross-reactive proteins (TNF alpha-CRPs) in a developmentally regulated manner, thus implicating these proteins in programmed cell death and in tissue remodeling. In this study, cells undergoing DNA fragmentation have been identified, using terminal deoxynucleotide transferase (TdT) mediated dUTP-biotin nick-end-labeling (TUNEL), during the embryonic development of the chick, between stages 18 and 29. DNA fragmentation is indicative of cells undergoing programmed cell death. TUNEL-positive cells were identified in several well documented areas of programmed cell death, including the limb buds, the heart, spinal motoneurons, dorsal root ganglia, and the ventral horn of the neural tube. In addition, other areas of cell death were identified including the floor plate and the mesonephros. In several locations, a close correlation was noted between the presence of TUNEL-positive cells and regions of TNF alpha-immunoreactivity. These regions included the ventral horn and marginal zone of the neural tube, spinal motoneurons, paravertebral ganglia, parts of the myotome, mesenchyme of the body wall, and the mesonephros. In addition, using the TNF alpha-sensitive L929-8 bioassay it was shown that homogenate of stage 18 chick embryos is cytotoxic to L929-8 cells and that this toxicity can be reduced using neutralizing antibodies to mouse TNF alpha. This bioassay allowed us to estimate the mean concentration of TNF alpha-like activity in embryo homogenate, which is within the range of physiological (pg/ml) levels of TNF alpha found in other systems. These results suggest that proteins with TNF alpha-like activity may have a role in programmed cell death in some tissues during early chick embryo development.

A membrane TNF-alpha/TNFR ratio correlates to MODS score and mortality.

This study hypothesizes that post-trauma elevated membrane-associated tumor necrosis factor-alpha (mTNF) and decreased TNF receptor shedding may be more related to development of multiple organ dysfunction syndrome (MODS) than elevated secreted TNF-alpha. We also address several of the possible reasons for the previous conflicting reports in studies correlating trauma patients sera TNF-alpha levels to their clinical outcome. These are 1) the lack of an objective quantitative score of clinical illness severity, 2) the lack of multiple TNF-alpha measurements in one patient to allow for trend analysis, 3) the lack of analysis of membrane-associated as well as secreted TNF-alpha levels, 4) the lack of concomitant analysis of soluble TNF-alpha receptors which may bind TNF-alpha in the serum, and 5) the possible requirement for more than one dysfunction in monocyte (M phi) TNF-alpha production and regulation to initiate pathology. Here, the MODS score was used to quantitate patients' illness severity over the length of their intensive care unit (ICU) stay. Patients' and normals' monocytes (stimulated and unstimulated) were assessed for production of secreted as well as membrane-associated TNF-alpha (sTNF and mTNF) and for shed p75 TNF-alpha receptor (TNFR) levels. These parameters of M phi TNF-alpha production and regulation were correlated to the MODS score as an indicator of clinical outcome. There was no correlation between sTNF and MODS score (p = .9025). There was a correlation between increased mTNF (p = .057) or decreased TNFR shedding (p = .0021) to increased MODS, but this lacked specificity. However, when the stimulated M phi production of mTNF and TNFR are expressed as the mTNF/TNFR ratio, an increased ratio correlates with high specificity to development of organ failure (p = .0002). These data indicate that a dual deregulation in M phi TNF-alpha production reflects increasing mTNF-alpha levels concomitant to decreased M phi shedding of neutralizing TNFR and correlates with the development of MODS.

Tumor necrosis factor production is deficient in diabetes-prone BB rats and can be corrected by complete Freund's adjuvant: a possible immunoregulatory role of tumor necrosis factor in the prevention of diabetes.

The Bio-Breeding (BB) rat develops spontaneous insulin-dependent diabetes mellitus (IDDM) and provides a useful animal model to study this human autoimmune disease. Treatment of BB rats with tumor necrosis factor (TNF) has been reported to prevent the development of IDDM. This suggests that deficient TNF production may be involved in the immunopathogenesis of autoimmune diabetes. In this study, we evaluated TNF production in diabetes-resistant (DR) BB rats, diabetes-prone (DP) BB rats, and DP BB rats protected from diabetes by the immunoadjuvant, complete Freund's adjuvant (CFA). TNF production in short-term cultures of peritoneal macrophages from DP rats was significantly less than that from control DR rats, both in the basal state and after stimulation with either interferon-gamma (IFN-gamma) or lipopolysaccharide (LPS) in vivo and in vitro. In contrast, TNF production by macrophages from CFA-injected DP rats (basal and IFN-gamma or LPS-stimulated) was equal to or greater than that by macrophages from DP rats and similar to TNF production by macrophages from CFA-injected DR rats. These results suggest that development of autoimmune diabetes in BB rats may be causally related to deficient macrophage production of TNF, and that upregulation of TNF production may protect against diabetes development.

Suppression by thymidine-requiring mutants of Escherichia coli K-12.

Thymidine-requiring strains of Escherichia coli isolated by trimethoprim selection often simultaneously acquire the ability to suppress bacteriophage T4 nonsense mutations. Suppression is lost in Thy+ revertants and recombinants, but is sometimes retained in thyA plasmid-bearing transformants. Suppression is restricted in Strr derivatives of the Thy- mutants, indicating that suppression occurs at the level of translation.

Tumor necrosis factor mediates the protective effect of Freund's adjuvant against autoimmune diabetes in BB rats.

In previous studies we reported that a single injection of complete Freund's adjuvant (CFA) in early life inhibits the development of autoimmune diabetes in BB diabetes-prone (DP) rats, and that CFA upregulates tumor necrosis factor-alpha (TNF alpha) production by macrophages of DP rats. In this study, we asked if CFA-induced prevention of diabetes in DP rats might be mediated by TNF alpha. Chronic intraperitoneal injection of TNF alpha, 20 micrograms daily from age 50 to 110 days, significantly decreased diabetes incidence in DP rats from 83% (15 of 18) to 37% (seven of 19) at age 110 days. Similarly, a single intraperitoneal injection of CFA, 100 microliters at age 26 days, significantly decreased diabetes incidence from 80% (16 of 20) to 50% (10 of 20) at age 110 days. Prevention by either TNF alpha or CFA was associated with decreased insulitis and preservation of islet beta-cells. By contrast, treatment of DP rats with CFA, plus antiserum to TNF alpha three times a week from age 26 to 110 days, obviated CFA-induced protection against beta-cell destructive insulitis and diabetes. Thus, diabetes incidence was 75% (18 of 24) in DP rats treated with CFA plus anti-TNF alpha antiserum, similar to that in untreated DP rats (80%), but only 44% (eight of 18) in DP rats treated with CFA plus control serum, similar to that in DP rats with CFA alone (50%). These results suggest that TNF alpha may be a mediator of the protective effects of CFA against autoimmune diabetes development in BB rats.

Autoimmunity-prone BB rats lack functional cytotoxic T cells.

BB rats are prone to develop an autoimmune form of insulin-dependent diabetes mellitus (IDDM) and thyroiditis. Development of autoimmunity is thymus dependent. Previous studies have shown that BB rats lack a population of T cells bearing the RT6 antigen and have very low numbers of suppressor/cytotoxic T cells. In this study, we confirm that BB rats have decreased numbers of phenotypic T suppressor/cytotoxic (Ts/c) cells (OX19+, OX8+ cells) in their lymphoid organs. Moreover, we find that the phenotypic Ts/c cells of BB rats lack apparent cytotoxic activity. These T cells fail to kill allogeneic target cells in a cell-mediated lympholysis assay and fail to generate lectin-dependent cytotoxicity. The addition of interleukin 2, gamma-interferon, and other lymphokines to cultures of BB T cells does not induce functional cytotoxic T lymphocytes. We find that the activated T cells of newly diabetic rats are incapable of killing major-histocompatibility-complex-matched islet cells, despite the ability of these cells to cause IDDM in passive transfer experiments. We conclude that autoimmune disease occurs in BB rats in the absence of functional cytotoxic T cells.

Class I and II major histocompatibility complex gene product expression by a rat insulinoma cell line in vitro following exposure to gamma interferon.

A study of Class I and II major histocompatibility complex gene product expression by a rat insulinoma cell line (RINm5F) was performed using monoclonal antibodies and immunoperoxidase techniques. RINm5F cells were incubated with different concentrations of gamma interferon. RINm5F cells exhibit low levels of Class I molecules and are normally devoid of Class II gene products. Upon exposure to gamma interferon, RINm5F cells showed a dramatic increase in Class I expression. This expression was homogenous and could be detected on all cells after 18 h of incubation with as little as 1 unit/ml of interferon. In contrast, de novo Class II expression was not homogeneous and required 36 h of incubation with 10 units/ml of interferon. The number of RINm5F cells expressing Class II antigens was dose- and time-dependent. Interferon treatment did not affect the morphology of RINm5F cells as determined by ultrastructural analysis. Withdrawal of interferon from the culture medium for as long as 78 h diminished but did not abolish the expression of Class I and Class II molecules already induced. The ability of interferon to enhance expression of Class I gene products and induce de novo expression of Class II molecules on B-cell-derived RINm5F cells supports the hypothesis that aberrant expression of major histocompatibility complex gene products on pancreatic B cells may be an important factor in triggering the immune response in Type 1 (insulin dependent) diabetes mellitus.

Characterization of a granule-independent lytic mechanism used by CTL hybridomas.

The mechanism(s) by which CTL induce target cell lysis have not been clearly elucidated. Perforin and the cytotoxic cell proteinases (granzymes) contained within the granules of CTL and NK, have been implicated, but abundant evidence for the existence of alternate lytic pathways has accumulated. In this report we characterize the mechanism of killing used by two cytolytic hybridomas (PMM-1 and MD90) that express neither perforin nor the granzymes. These characteristics are compared with results obtained by using a representative Ag-dependent, granule-containing T cell clone in cytolysis assays. The major differences were that the granule-negative hybridomas could lyse a variety of target cells in the presence of cyclosporin and the absence of calcium. All the effectors could kill in the presence of protein synthesis inhibitors (cycloheximide and emetine) and induced DNA fragmentation in the target cells. The cytolytic hybridomas had to be stimulated to be cytolytic and this activation required the presence of calcium, was dependent on protein synthesis, and inhibited by the addition of cyclosporin. Although TNF was shown not be involved, the sensitivity of the target cells to lysis by the granule-negative killers correlated with the level of expression of Fas Ag. With the use of L1210 and an L1210 cell line transfected with Fas cDNA we demonstrated that these MD90 and PMM-1 kill the latter much more effectively and that this increase was effectively inhibited with anti-Fas Ab. Furthermore the lack of sensitivity to cyclosporin, cycloheximide, emetine, and EGTA was confirmed with these targets. We conclude that these two cytolytic hybridomas use the Fas lytic pathway to induce lysis in target cells.