Interleukin (IL)-1 receptor-associated kinase (IRAK) requirement for optimal induction of multiple IL-1 signaling pathways and IL-6 production.

Interleukin (IL)-1 is a proinflammatory cytokine with pleiotropic effects in inflammation. IL-1 binding to its receptor triggers a cascade of signaling events, including activation of the stress-activated mitogen-activated protein (MAP) kinases, c-Jun NH2-terminal kinase (JNK) and p38 MAP kinase, as well as transcription factor nuclear factor kappaB (NF-kappaB). IL-1 signaling results in cellular responses through induction of inflammatory gene products such as IL-6. One of the earliest events in IL-1 signaling is the rapid interaction of IL-1 receptor-associated kinases, IRAK and IRAK-2, with the receptor complex. The relative roles of IRAK and IRAK-2 in IL-1 signaling pathways and subsequent cellular responses have not been previously determined. To evaluate the importance of IRAK in IL-1 signaling, IRAK-deficient mouse fibroblast cells were prepared and studied. Here we report that IL-1-mediated activation of JNK, p38, and NF-kappaB were all reduced in embryonic fibroblasts deficient in IRAK expression. In addition, IL-6 production in response to IL-1 was also dramatically reduced in IRAK-deficient embryonic fibroblasts and in skin fibroblasts prepared from IRAK-deficient mice. Our results demonstrate that IRAK plays an essential proximal role in coordinating multiple IL-1 signaling pathways for optimal induction of cellular responses.

The Pittsburgh insulin-dependent diabetes mellitus (IDDM) morbidity and mortality study. Mortality results.

A follow-up study of 1966 patients with insulin-dependent diabetes mellitus (IDDM) who were diagnosed at Children's Hospital of Pittsburgh (CHP) between 1950 and 1981 has been completed. The mean age of the population at follow-up was 21.2 yr with a mean duration of IDDM of 12.9 yr. Nine percent of the patients were deceased, a sevenfold excess in mortality compared with the U.S. population. The relative increase in mortality was greater for females than males and greater for blacks than whites. Before age 20, the primary excess in mortality was at onset of IDDM, or within 6 mo after onset, and was due to acute diabetic complications. After age 20, the annual mortality risk was approximately 2%, which was more than 20 times greater than for the U.S. population. Renal disease was responsible for the majority of these deaths. There was a reduced risk of dying for diabetic patients who were diagnosed between 1966 and 1971 compared with patients diagnosed during earlier years.

HLA heterogeneity of insulin-dependent diabetes mellitus at diagnosis. The Pittsburgh IDDM study.

Although some previous studies have suggested that insulin-dependent diabetes mellitus (IDDM) is a heterogeneous condition with variant forms being associated with HLA-DR types, the evidence, thus far, is conflicting. To address this issue, we have examined the presenting characteristics of a consecutive admission series of 200 newly diagnosed cases of IDDM from the Children's Hospital of Pittsburgh. Because HLA-DR frequencies vary by race, data are presented only for the 172 white cases with complete HLA-DR typing. HLA-DR3 was found more frequently among male cases and DR4 among female cases (P less than 0.005). Generally, patients with DR4 presented with a severer clinical picture, being more likely to have impaired consciousness and significant dehydration. In addition, patients with DR4 were more likely to be acidotic, ketotic, and to more frequently report a recent viral infection. This latter finding was supported by a greater frequency of antibodies to Coxsackie-B viruses in the DR4 cases at presentation. These results therefore suggest that there is considerable heterogeneity in IDDM, at least in presenting characteristics, according to HLA-DR type.

Effects of inflammatory cytokines and phorbol esters on the adhesion of U937 cells, a human monocyte-like cell line, to endothelial cell monolayers and extracellular matrix proteins.

The accumulation of mononuclear phagocytes at sites of chronic inflammation is dependent on an increase in the rate of extravasation of blood-borne monocytes through the vascular endothelium into the connective tissue. Once the monocytes have emigrated into the connective tissue, they may differentiate into tissue macrophages, presumably following interactions with extracellular matrix proteins. To study these processes, we tested the effects of cytokines and phorbol esters on the adhesion of U937 cells, a human monocyte-like cell line, to cultured endothelial cells (EC) and to matrix proteins. In the absence of cytokines, very few of the U937 cells adhered to EC (5% or less in most experiments). When EC were pretreated for optimal periods of time (4-8 hr) with recombinant interleukin-1 alpha (IL-1 alpha), IL-1 beta, tumor necrosis factor-alpha (TNF alpha), or lymphotoxin (LT; also known as TNF-beta), 35-85% of the U937 cells were able to bind. Interferon-gamma (IFN-gamma) and interleukin-2 (IL-2) did not stimulate U937-EC binding, even though IFN-gamma was shown to increase EC adhesiveness for T lymphocytes. Phorbol esters also greatly stimulated U937-EC adhesion but, in this case, the increase was due to an action on the U937 cells. A monoclonal antibody (MAb), 60.3, against the CD11/CD18 family of leukocyte adhesion molecules partially inhibited the adhesion of untreated and phorbol ester-treated U937 cells to noncytokine-treated EC. However, that MAb had no effect on U937 cell binding to TNF-alpha-treated EC. Thus U937 cells use both CD11/CD18-dependent and -independent mechanisms to adhere to EC. In the absence of stimulating agents, only a small proportion of the U937 cells (2-20%) adhered to fibronectin (FN), and almost none bound to either laminin (LN) or gelatin (denatured type I collagen). In the presence of phorbol esters, a much larger proportion of the U937 cells adhered to FN, with only slight increases in the proportion of cells which bound to LN or gelatin. Additional adhesion assays performed in the presence of a pentapeptide containing the amino acid sequence arg-gly-asp (RGD), which is part of one of the cell-binding domains of FN, demonstrated that the RGD-containing peptide almost totally blocked the phorbol ester-induced adhesion of U937 cells to FN. In contrast, the peptide had no inhibitory effect on the phorbol ester-induced binding of U937 cells to EC.

Immunomodulation and adenoviral gene transfer to the lungs of nonhuman primates.

Previous data from our laboratory and others have demonstrated a critical role for the CD4+ T lymphocyte in in vivo immune responses to recombinant adenoviral vectors. In rodent models, this subset of T cells is required for T cell proliferation, subsequent cytotoxic T cell generation, and production of anti-adenoviral antibodies by B cells. Both depleting and nondepleting anti-CD4 antibodies can attenuate these immune responses to recombinant adenovirus. On the basis of these data, we hypothesized that a nondepleting CDR-engrafted anti-human CD4 antibody (OKT4A) with cross-reactivity to rhesus macaques would attenuate both T and B cell responses to intrapulmonary administration of recombinant adenovirus and permit prolonged reporter gene expression and permit secondary gene transfer. Juvenile rhesus macaques were treated with PBS or OKT4A antibody (10 mg/kg) daily beginning 1 day prior to and up to 11 days after gene transfer. OKT4A resulted in significant attenuation of lymphocyte recruitment into the lung, lymphocyte-proliferative responses to both adenovirus capsid proteins and transgene protein, and adenovirus-induced interferon-gamma elaboration in whole blood and hilar lymph nodes. However, OKT4A was ineffective in attenuating adenovirus-induced IL-4 production in whole blood or hilar lymph nodes, generating neutralizing anti-adenoviral antibodies, or permitting secondary gene transfer. As all the monkeys in this protocol had baseline-detectable anti-adenoviral antibodies by ELISA that were nonneutralizing, analogous to most patients with cystic fibrosis, we postulate that anti-CD4 did not block the proliferation of memory B cells. Moreover, these data suggest that for transient immunomodulation to be successful, strategies need to focus specifically on B cell activation independent of CD4+ T cell help.

Lymphocyte adhesion to endothelial cells in vitro: models for the study of normal lymphocyte recirculation and lymphocyte emigration into chronic inflammatory lesions.

Lymphocytes preferentially leave the bloodstream to enter secondary lymphoid organs or sites of inflammation by first adhering to, and then migrating through, the walls of specialized postcapillary venules known as high endothelial venules. To study the initial adhesion event between lymphocytes and endothelial cells, two different in vitro assays have been developed. In the first, lymphocytes are incubated on frozen sections of lymphoid organs or other tissues. Under certain conditions, lymphocytes specifically bind to the endothelial cells of high endothelial venules in such sections. The results of monoclonal antibody inhibition studies have suggested that endothelial cells at various lymphoid organs, and at certain inflammatory lesions, may express organ-specific ligands on their surface, which bind to corresponding organ-specific "homing receptors" on the lymphocytes. The second assay measures the adhesion of lymphocytes to confluent monolayers of viable, cultured endothelial cells. Because pretreatment of such cell monolayers with a number of cytokines produced at inflammatory sites stimulates an increase in the adhesiveness of the cells for lymphocytes, it has been suggested that such cytokine-induced increases in endothelial cell adhesiveness may be important in the induction of lymphocyte traffic into such lesions. Unlike the homing receptor type of binding described above, the cytokine-induced increase in lymphocyte-endothelial cell adhesion is presumably non-tissue-specific. Results of monoclonal antibody inhibition studies in this system have suggested that at least two ligand-receptor interactions may be involved. Monoclonal antibodies to the lymphocyte function-associated antigen-1 greatly inhibited the adhesion of T cells to untreated endothelial cells, but had little or no inhibitory effect on T-cell adhesion to cytokine-treated endothelial cells.

Organ-specific and non-organ-specific lymphocyte receptors for vascular endothelium.

The recirculation of lymphocytes from blood to lymph and back to blood is necessary for the proper functioning of the immune system as it facilitates interactions between antigen-reactive clones of lymphocytes and antigen-presenting cells. The first step in the emigration of a blood-borne lymphocyte into either a secondary lymphoid organ or an inflammatory lesion is its adherence to vascular endothelial cells (EC) lining unique post-capillary venules known as high endothelial venules (HEV). Several groups have recently cloned and sequenced genes which may encode organ-specific lymphocyte receptors for the EC of such HEV. The extracellular portion of the putative murine lymphocyte homing receptor for peripheral lymph node HEV is composed of an N-terminal lectin-like domain, followed by an epidermal growth factor-like domain, and then two identical repeating domains which are homologous to a number of complement-binding proteins. A hydrophobic transmembrane domain and a cytoplasmic tail complete the structure. A very similar gene structure has been reported for a cytokine-inducible EC surface protein which is involved in neutrophil-EC adhesion in vitro. In marked contrast, the gene for a putative human lymphocyte homing receptor appears to belong to a gene family which encodes cell-surface molecules with receptor activity for extracellular matrix (ECM) proteins. Similarly, the cell-surface molecule which appears to be the murine lymphocyte receptor for Peyer's patch HEV is homologous, if not identical, to the human VLA-4 molecule, another receptor with binding activity for an ECM protein. It has also been demonstrated that lymphocyte function-associated antigen 1 (LFA-1) acts in a non-organ-specific manner to mediate lymphocyte-EC adhesion. Finally, other non-organ-specific lymphocyte adhesion molecules for EC may include CD4 and CD8 (which bind to class II and class I MHC antigens, respectively), and CD2 (which binds to LFA-3).

T cell adhesion to extracellular matrix molecules secreted by endothelial cells cultured on a substrate of type IV collagen.

T cell emigrating from the bloodstream into lymphoid organs or sites of inflammation in the connective tissue must adhere to, and traverse, the subendothelial basement membrane (BM). The goal of the current investigation was to develop a method to study the adhesion of T cells to endothelial cell (EC)-derived extracellular matrix (ECM) as a model for the interaction of T cells with the subendothelial BM in vivo. To be certain that we were truly measuring T cell adhesion to ECM molecules secreted by the EC, it was necessary to culture the EC on a substrate to which T cells could not attach. Non-tissue culture-treated microtiter plate wells which had been coated with type IV collagen (tIVC), a major constituent of BM in vivo, were found to be suitable for this purpose since EC, but very few T cells, adhered to such wells. After incubating the EC on a substrate of tIVC in non-treated wells for a period of 48 h, the EC were gently removed from their underlying ECM and T cell adhesion to that ECM was examined. Using this system, it was observed that approximately 15-40% of human peripheral blood T cells specifically adhered to ECM molecules produced by the EC. This method should be useful as a model for the interactions of T cells and other leukocytes with the vascular BM in vivo.

Pittsburgh Insulin-Dependent Diabetes Mellitus Morbidity and Mortality Study: physical activity and diabetic complications.

The long-term health consequences of chronic physical activity for patients with type I diabetes are unknown. In the current study, the association of physical activity to diabetic complications was assessed in 696 type I diabetic individuals diagnosed between 1950 and 1964. Participation in team sports in high school or college was not associated with a decreased prevalence of severe retinopathy or blindness later in life. There was, however, a suggestion of a negative association between physical activity and both cardiovascular disease and overall mortality, ie, individuals who participated in team sports were somewhat less likely to report macrovascular disease at follow-up or to have died than nonparticipants. The relationship between physical activity and diabetic complications only appeared in male subjects. The results suggest that activity early in life by patients with type I diabetes does not appear to be associated with an adverse health effect and may, in fact, be beneficial.

Interferon-gamma increases susceptibility of murine pancreatic beta cells to lysis by allogeneic cytotoxic T lymphocytes.

The selective loss of insulin-producing pancreatic beta cells which occurs in IDDM has been postulated to result from lysis by beta cell-specific cytotoxic T lymphocytes (CTL). CTL typically recognise antigen in the context of MHC class I molecules, which are normally present at low levels on beta cells. However, hyperexpression of class I antigens on islet cells has been observed in the early stages of beta cell destruction in IDDM. Since interferon-gamma (IFN-gamma) is known to increase class I expression on a number of cell types, we have investigated the responses of murine beta cells to this cytokine under various conditions. Two color immunostaining followed by FACS analysis showed that on average, only 14.9 +/- 3.1% of cultured beta cells were class I positive. However, a majority of beta cells could be induced to express class I after 24 hours of IFN-gamma treatment, and maximal induction (80-90% positive) occurred after 48 hours. Importantly, increased class I expression on beta cells could be achieved with very low concentrations of IFN-gamma (1-10 U/ml). Expression of class II MHC was never detected under any of the conditions employed to up-regulate class I. Interestingly, although islet cells were only moderately susceptible to lysis by allospecific CTL, this susceptibility was markedly enhanced by prior exposure of the islets to IFN-gamma. Taken together, these results suggest that beta cells are extremely susceptible to up-regulation of class I MHC molecules by IFN-gamma, and that this property may render these cells particularly susceptible to lysis by autologous class I-restricted CTL. Since enhanced expression of class I frequently accompanies inflammatory responses and viral infections, this property of beta cells may account in part for their selective destruction in IDDM.

Analyses on possible heterogeneity of IDDM based on presence of islet cell cytoplasmic antibody at diagnosis.

In a large, representative sample of newly-diagnosed IDDM patients, using a highly sensitive assay to detect islet cell cytoplasmic antibodies (ICA), no marked differences were found between ICA+ and ICA- patients on various clinical, genetic, immunologic, and epidemiologic characteristics. In particular, there was no evidence for associations between ICA status at diagnosis and either sex, race, family history of IDDM, HLA-DR phenotype, antibody titers to Coxsackie B viruses, immunoglobulin levels, C-peptide and glycosylated hemoglobin concentrations, or insulin requirements. The most significant relationship was between the presence of ICA and a young age at diagnosis; however, the large overlap between the distributions of the ages at onset for ICA+ and ICA- groups on this variable suggests that this association is of limited importance. These data suggest that the presence or absence of ICA at diagnosis may not be useful in defining possible subtypes of IDDM.

Familial and sporadic insulin-dependent diabetes: evidence for heterogeneous etiologies?

Heterogeneity within insulin-dependent diabetes mellitus (IDDM) has been hypothesized, but few studies have focused on differences which may exist between familial and sporadic IDDM cases. Presenting characteristics for 330 white, newly diagnosed IDDM cases were evaluated. Familial cases were older (10.2 +/- 5.1 years vs 7.9 +/- 4.2 years, P = 0.010) and had, on average, less severe metabolic disturbances at presentation, as demonstrated by lower mean hemoglobin A1 (12.6 +/- 2.4% vs 14.4 +/- 2.6%, P = 0.001) and mean insulin dose at discharge (0.62 +/- 0.35 U/kg/day vs 0.85 +/- 0.29 U/kg/day, P less than 0.001), and higher mean plasma bicarbonate concentrations (19.3 +/- 3.9 mmol/l vs 15.8 +/- 5.9 mmol/l, P = 0.023) and mean plasma C-peptide levels (0.35 +/- 0.36 pmol/ml vs 0.14 +/- 0.15 pmol/ml, P less than 0.001). Further analyses on a subset of IDDM cases (n = 100) indicated that initial differences in metabolic indices observed at diagnosis were no longer apparent at one-year post-diagnosis. These results suggest that the etiology of familial and sporadic IDDM is similar and that the less severe presentation observed at diagnosis in the familial cases may be due to earlier identification of the disease, reflecting increased parental knowledge of diabetic symptoms and/or frequent testing for diabetes.

Inhibition by IL-1 of endothelial cell activation induced by tumor necrosis factor or lymphotoxin.

Previous studies in this and other laboratories have demonstrated that IL-1, lymphotoxin (LT), and TNF rapidly stimulate a number of proinflammatory properties in cultured endothelial cells (EC) including cell-surface procoagulant activity and increased adhesivity for lymphocytes, monocytes, and polymorphonuclear leukocytes. In addition, we have demonstrated that LT and TNF, but not IL-1, stimulate increases in EC RNA synthesis, protein synthesis, and cellular volumes, changes which may correspond to the hypertrophy of EC seen at sites of inflammation in vivo. It is reported here that both human rIL-1 alpha and rIL-1 beta totally inhibit the increases in EC RNA synthesis, protein synthesis, and cell volumes induced by either TNF or LT. As little as 0.1 ng/ml of either IL-1 was sufficient to totally block the activation of EC induced by 100-fold higher concentrations (10 ng/ml) of either LT or TNF. The relevance of these findings to the regulation of inflammatory responses is discussed.

Absence of Antienterobacteriaceae and anti-HLA-B27 antibodies in mitogen stimulated cultures of lymphocytes from patients with Reiter's syndrome and ankylosing spondylitis.

Lymphocytes from HLA-B27 positive patients with ankylosing spondylitis or Reiter's syndrome and from matched controls were cultured with pokeweed mitogen and heat-killed S. aureus bacteria under conditions designed to maximize immunoglobulin production. Despite the secretion of microgram quantities of immunoglobulin, which were not significantly different in patients and controls, only negligible amounts of IgM, IgG and IgA antibodies to Campylobacter, Shigella and Yersinia were detectable. In addition, no anti-HLA-B27 antibodies were produced by any patient. Our results are not consistent with the hypothesis that HLA-B27 crossreactive antibodies to Enterobacteriaceae antigens have a role in the pathogenesis of ankylosing spondylitis or Reiter's syndrome.

Presence of complement-dependent cytotoxic activity against clonally-derived rat islet tumour cells in sera from type 1 (insulin-dependent) diabetic patients and control subjects.

Heat-inactivated sera from newly diagnosed Type 1 (insulin-dependent) diabetic patients and control subjects were tested for the presence of antibodies to islet cell surface antigens by means of a sensitive immunofluorescent, microcytotoxicity assay using two clones of a rat islet cell tumour as antigens. Complement-dependent cytotoxicity was found in 74% of diabetic patient sera and 87% of control sera, and there were no significant differences in titres between diabetic patients and control subjects. A minority of the sera from both patients and controls were cytotoxic for only one of the two clones, suggesting the presence of multiple antigen-antibody systems. Preadsorptions of the sera with rat liver powder, sheep erythrocytes, and/or protein A-conjugated agarose beads were inconsistently effective in decreasing levels of lytic activity in control sera. It is concluded that more information is required concerning the antigens of rat islet cells and islet cell cytotoxic factors present in normal sera before such cells and assays can be reliably used for the detection of islet cell surface antibodies.

Endothelial cell activation induced by tumor necrosis factor and lymphotoxin.

Alterations in the morphology and histochemistry of vascular endothelial cells (EC) have been repeatedly observed at sites of chronic inflammation and immune reactions. These changes, which are most prominent in the EC postcapillary venules present in areas with large lymphocytic infiltrates, include the acquisition of a columnar or cuboidal morphology, the development of ribonuclease-sensitive metachromasia, and an increase in intracellular organelles. Thus, EC at sites of inflammation appear to be activated and to demonstrate increased metabolic activity. This study reports that both tumor necrosis factor-alpha (TNF) and lymphotoxin (LT) can activate cultured human umbilical vein EC, as measured by: 1) increased adhesiveness for lymphocytes, 2) increased cell metabolism, as measured by RNA and protein synthesis, and 3) increased cell volume. Although gamma interferon (IFN-gamma) and interleukin-1 (IL-1) have been shown previously to stimulate EC adhesiveness for lymphocytes, these two cytokines had only marginal effects on EC RNA and protein synthesis, and both caused a decrease in EC volume. These findings suggest that TNF and LT play a role in the type of activation of EC in vivo that leads to the development of tall endothelium and increased lymphocyte emigration.

Interleukin 1 increases the binding of human B and T lymphocytes to endothelial cell monolayers.

Lymphocyte binding to specialized high-endothelial venules (HEV) in lymph nodes and Peyer's patches is the first step in normal lymphocyte emigration and recirculation. The development and maintenance of HEV in these lymphoid organs are thought to be immunologically controlled. Because postcapillary venules in chronic inflammatory tissue often resemble the HEV of lymphoid tissue and may also be a site of lymphocyte emigration, examination of the effects of immunologic and inflammatory mediators on endothelial cells (EC) may provide important information about the physiology of both normal lymphocyte recirculation and chronic inflammation. It is reported here that treatment of human umbilical vein EC monolayers in vitro with affinity-purified human interleukin 1 (IL 1) markedly enhances the binding of both B and T lymphocytes. Increased binding was observed within 1 h of treatment of EC with as little as 0.04 U/ml IL 1. This effect of IL 1 was EC-specific, because pretreatment of T cells or human skin fibroblasts with IL 1 did not increase the binding of lymphocytes. Stimulation of binding required active EC metabolism because incubation of EC with IL 1 at 4 degrees C, or prior fixation of EC, prevented enhanced binding. The action of IL 1 was not associated with EC damage. The secretion of IL 1 by macrophages and perhaps other cells in inflammatory lesions may exert a positive feedback signal on EC to enhance further emigration of lymphocytes into the inflammatory focus.

Separation and characterization of human T lymphocytes with varying adhesiveness for endothelial cells.

Previous studies in this laboratory have demonstrated that the adhesion of T lymphocytes to endothelial cell (EC) monolayers in vitro can be increased by preincubation of the EC with interferon-gamma, interleukin 1 (IL-1), tumor necrosis factor-alpha (TNF), or lipopolysaccharide (LPS), or by stimulation of the T cells with phorbol esters. In this report, we have demonstrated that three subpopulations of human peripheral blood T cells can be identified on the basis of their abilities to bind to EC: (1) a strongly binding group which binds to unstimulated EC; (2) an intermediately binding subset which adheres to EC only if these cells have been stimulated with IL-1, TNF, or LPS; and (3) a weakly binding subpopulation which adheres poorly to either unstimulated or stimulated EC. The more adhesive subgroups had larger cellular volumes than the less adhesive cells, were relatively enriched in cells bearing the OKM1 surface marker, and expressed relatively greater amounts of the lymphocyte-function-associated-1 molecule. Stimulation of the EC to bind increased numbers of T cells by IL-1, TNF, and LPS appeared to be mediated by the expression of a common adhesion molecule on the EC.