Esther Sobel

The use of a single serological marker underestimates the prevalence of celiac disease in Israel: a study of blood donors.

OBJECTIVES:Recent studies suggest that celiac disease was previously underdiagnosed. To find out whether antiendomysial antibodies underestimate the prevalence of celiac disease, we elected to use a strategy combining multiple serological markers to explore the prevalence of celiac disease in Israel and the usefulness of the various antibodies in screening for celiac disease.METHODS:Serum samples from 1571 healthy blood donors were tested. A small intestinal biopsy was offered to all patients who tested positive for either human tissue transglutaminase antibodies, an ELISA kit based on antiendomysium (EMA-ELISA), immunoglobulin A antigliadin verified by antiendomysial immunofluorescence antibodies, and to patients who were IgA deficient with elevated antigliadin IgG.RESULTS:A total of 59 subjects (3.8% of study population) were offered an intestinal biopsy based on serological findings, and 30 of 59 patients agreed to undergo intestinal biopsy (1.9% of study population). Celiac disease was diagnosed in 10 patients, establishing a prevalence of at least 1:157 in the general population (0.6%, CI = 0.3–1.1%). Using any serological marker alone would have underestimated the prevalence of celiac disease, as it was diagnosed in only two patients who tested positive for endomysial immunofluorescence antibodies (prevalence of 1:786, 0.1%, CI = 0.02–0.5%), six patients positive for tissue transglutaminase (prevalence of 1:262, 0.4%, CI = 0.1–0.9%), and seven patients positive for ELISA-EMA (prevalence of 1:224, 0.45%, CI = 0.2–0.9%).CONCLUSIONS:The prevalence of celiac disease in Israel is at least 1:157 in the general population, confirming its underdiagnosis in previous studies. The disparity between the various serological markers suggest that the use of one serological marker is insufficient for establishing the true prevalence of celiac disease.

Intravenous gamma globulin inhibits the production of matrix metalloproteinase‐9 in macrophages

Degradation of the extracellular matrix (ECM) is essential for progression and metastasis of cancer cells. The ECM‐degrading enzymes, matrix metalloproteinases (MMPs), are produced mainly by intratumor monocytes/macrophages. MMPs, particularly MMP‐9, are reported to be of crucial significance for both growth and tumor invasiveness. Inhibition of the expression of MMP‐9 may prevent tumor development. High‐dose intravenous gamma globulins (IVIG) effectively inhibit metastatic spread of tumors in mice and humans and a variety of mechanisms have been suggested to explain this effect.

Expression of matrix metalloproteinases, sICAM-1 and IL-8 in CSF from children with meningitis.

The combined expression of the inflammatory mediators, matrix metalloproteinases (MMPs), soluble form of intracellular adhesion molecule ICAM-1 (sICAM-1) and interleukin (IL)-8, was evaluated in children infected with bacterial or viral meningitis. MMP-2 and IL-8 were detected in all CSF samples and were enhanced in both bacterial and viral infected samples, compared to those from control children. The expression of MMP-9 as well as sICAM-1 was not detected in control CSF while observed in viral infected and further elevated in bacterial infected samples. This pilot study supports a role for MMPs, IL-8 and sICAM in infectious meningitis and suggests further research to determine their possible use as biomarkers for various forms of meningeal infection as well as the use of their specific antagonists as potential therapeutic agents for central nervous system (CNS) inflammatory processes.

The prevalence of coeliac disease antibodies in patients with the antiphospholipid syndrome

High prevalence of coeliac disease (CD) has been reported in various autoimmune disorders, but has not been studied in the antiphospholipidsyndrome (APS). We aimed to establishthe prevalenceof CD antibodiesin a cohort of APS patients, and to examine whether CD may be responsiblefor some of the manifestationsof APS. Fifty-seven patients (47 females, 10 males) with APS were studied for clinical manifestations and serological markers of the disease, as well as the presence of anti-endomysial antibodies using an ELISA assay (EMA-ELISA). Control subjects were 171 healthy individuals, ageand sex-matched (141 females). Eight patients with APS (14%, six females) were found to have EMA-ELISA antibodies, compared with 2/141 (1.1%) of controls (P 0.0003). Antibodies against b2-glycoprotein-I (b2GPI) epitopes (GRTCPKPDDLP) were more prevalent in EMA-positive patients than in EMA-negative patients (P 0.006). Vasculitic skin lesions were significantly more common in EMA-ELISA-positive compared with EMA-ELISA-negative patients (62.5 versus 16.3%, P 0.01). Among the skin manifestations, superficial cutaneous necrosis (37.5 versus 2%, P 0.007) was more prevalent in EMA-ELISA-positive than in EMA-ELISA-negative patients. EMA-ELISA antibodies are common in APS, and their presence is associated with high prevalence of antibodies recognizing certain b2-glycoprotein epitopes, and with cutaneous manifestations of APS.

Divergent effects of cytokines on human leukocyte antigen‐DR antigen expression of neoplastic and non‐neoplastic human thyroid cells

Apparently complex modulatory effects of alpha‐interferon (α ‐IFN), tumor necrosis factor (TNF), and epidermal growth factor (EGF) have been found in neoplastic human thyroid cells, which could possibly affect the final outcome in neoplastic disease. This was achieved by examining the influence of α‐IFN, TNF, and EGF alone and in combination, on human leukocyte antigen‐DR (DR) antigen expression and viability of neoplastic and non‐neoplastic human thyroid cells in culture. α‐IFN‐induced DR antigen expression on non‐neoplastic human thyroid cells, whereas TNF‐α or EGF alone were ineffective. The addition of the same TNF‐α concentrations (10 to 100 ng/ml) to α‐IFN enhanced the expression of DR antigens compared with the effect of α‐IFN alone. However, EGF inhibited α‐IFN‐induced DR on the same cells and at the same concentrations (10 to 500 ng/ml) at which the growth factor alone was ineffective. In contrast to the common pattern of cytokine effects on DR expression of all nonmalignant thyroid cell lines, neoplastic thyroid cell lines showed divergent responses to α‐IFN, TNF‐α, and EGF. In three malignant thyroid cell lines that were DR negative (follicular carcinoma WRO 82‐1 and NRO 87‐1 cell lines, and anaplastic carcinoma ARO 81‐1), DR antigen could be induced by α‐IFN and enhanced by TNF‐α, whereas EGF was ineffective. In a fourth cell line (an anaplastic carcinoma SW1736) α‐IFN, TNF‐α, and EGF alone were capable of inducing DR, and a combination of either TNF‐α and EGF with α‐IFN potentiated DR induction. In a fifth neoplastic cell line (papillary carcinoma, NPA) that constitutively expressed surface DR, its expression was inhibited by both α‐IFN and TNF‐α and was not affected by EGF.

Concomitant effect of 2'-deoxycoformycin on natural killer cell activity and tumour cell sensitivity to lysis in hairy cell leukaemia--discordant effects of alpha interferon.

The effects of 2′‐deoxycoformycin (dCF) and alpha interferon (IFN‐a) on natural killer (NK) cell‐enriched fractions and hairy cell (HC) targets from three patients with HC leukaemia (HCL) were investigated. There was no significant increase in NK activity when either the HC targets or NK‐enriched cells were preincubated with dCF. However, preincubation of both HC and NK cells with dCF resulted in increased NK activity. Culture of enriched NK cells with IFN‐α enhanced their activity. However, preincubation of HC targets with this drug in the presence or absence of dCF resulted in a protective effect. Maximal NK activity towards HCL was obtained when the target tumour cells were separately precultured with dCF and the NK‐enriched effectors precultured with dCF + IFN‐α. The effect of dCF and IFN‐α was also measured using the standard K562 cells as targets for NK activity. dCF enhanced NK activity following preculture of both effector and target K562 cells, but IFN‐a did not reduce K562 cell susceptibility to NK lysis as it did for HC cells. Our findings suggest that (a) dCF and IFN‐α, which are used to treat HC, could function via activation of NK cells, (b) effects on both effector and tumour target cells should be taken into account, and (c) caution should be exercised in extrapolating the effects of NK‐cell activity against K562 cells to those on HC targets.