[Virus and gastrointestinal infections]. / Virus et infections gastro-intestinales.

Numerous viruses found in the gut are not associated with primary infection or disease of the gastrointestinal tract. Other groups or viruses are not classically associated with infection of the gut but can infect the gastrointestinal tract in immunocompromised individuals (herpes simplex virus, cytomegalovirus, papillomavirus ....). The viruses associated with gastroenteritis represent a large number of taxonomic group. Because these viruses have in general been difficult to cultivate, most members of this group were firstly detected by electron microscopic examination (adenovirus, astrovirus, calicivirus, coronavirus, rotavirus ....). The most widely used diagnostic techniques for adenovirus (40/41), rotavirus and astrovirus detection in faecal samples include immunoassays such as Elisa and latex agglutination. Nucleic acid hybridization techniques have generally not proven to be substantially sensitive and the more sensitive techniques recently developed use the polymerase chain reaction (adenovirus) or the reverse transcription/polymerase chain reaction (astrovirus, calicivirus, coronavirus, rotavirus). Special efforts have been made in the search for efficient procedures to extract viral nucleic acids, and to establish the optimal conditions for the amplification and identification of PCR products but the candidate viruses were very different, consensus procedures were not determined, and amplification kits were not actually commercialized.

Evaluation of rapid culture centrifugation method for adenovirus detection in stools.

Routine laboratory testing for adenovirus (Adv) requires a procedure that is rapid and reliable, especially for samples from children and immunosuppressed patients, when diarrhea may signal the onset of severe gastrointestinal disorders. An improved culture technique for Adv isolation, using centrifugation step of 24-well plates and needing only 48 h incubation, was evaluated for 382 stool samples. This technique was compared with conventional tube cell culture and a commercial enzyme-linked immunosorbent assay (ELISA) kit. Adv was isolated in 36 samples (9.4%) by rapid culture, in 32 (8.4%) by conventional culture, and in 42 samples (11%) using genus-specific ELISA. A total of 30 isolates were found to be Adv positive in both rapid and conventional cultures, and half of the Adv-positive rapid culture isolates were identified as serotypes 40/41 using a type-specific ELISA. The improved culture method considerably reduces incubation time and also offers a slightly enhanced sensitivity to Adv serotypes. Combined with appropriate cell lines adapted to the isolation of enteric adenoviruses, it therefore constitutes a valuable laboratory test particularly useful in the diagnosis of gastroenteritis.

[Adenoviruses from stool samples in hospital units. Comparison with main pathogens in gastroenteritis (rotavirus, Campylobacter, Salmonella)]. / Les adénovirus dans les prélèvements de selles en milieu hospitalier. Comparison avec les principaux agents des gastroentérites (rotavirus, Campylobacter, Salmonella).

During a six-years period (1988-1993), a total of 14,644 stool samples from in-patients of Limoges University Hospital were examined for the presence of principal enteric pathogens, such as adenovirus, rotavirus, Campylobacter, Salmonella, Shigella and others. Stools were processed for identification of bacteria by standard methods and viruses were detected in fecal specimens using antigen detection methods: ELISA (Enzyme Linked Immunosorbent Assay) and latex agglutination test. The decreasing rates of presence of enteric agents were respectively 6% for rotavirus, 3.2% for Salmonella, 2% for adenovirus, 1.6% for Campylobacter and 0.2% for Shigella, but according to the lack of sensibility of latex agglutination test, adenovirus prevalence was probably underestimated. Concerning the distribution of enteric pathogens throughout the year, our data demonstrate that rotavirus were rather shed during the months from January to April, adenovirus between April and August, Campylobacter during summer and Salmonella from July to October. The two thirds of Campylobacter and rotavirus infections and the half of adenovirus and Salmonella infections were identified during the ten first years of life. The highest prevalence occurs before 5 years old, during the 2nd year of life for adenovirus (4.4%) and rotavirus (22.3%) and during the 3rd year of life for Campylobacter (6.84%) and Salmonella (8.6%).

Enteric prevalence of adenovirus in human immunodeficiency virus seropositive patients.

To evaluate the prevalence of adenovirus strains in human immunodeficiency virus (HIV)-positive patients and to investigate their possible role in the onset of diarrhea, a total of 103 stools from HIV-seropositive patients at various stages of infection and 200 stools from sex and age cross-matched control subjects were examined. Adenovirus prevalence was measured by ELISA as well as conventional and rapid cell culture techniques. Results were compared between patients suffering from diarrhea and those without diarrhea. Adenovirus prevalence was statistically greater in HIV-seropositive cases than controls (8.7%, 2.5%, respectively). No significant difference was found between HIV-positive patients with diarrhea and those without gastrointestinal complications (P > 0.05). However, a significant difference in adenovirus prevalence was found between HIV-positive patients with diarrhea and control subjects with diarrhea (P = 0.02). Although viral prevalence varied with the different stages of HIV infection, differences were not statistically significant. In conclusion, although current opinion considers adenoviruses to be no more than opportunistic pathogens, the results of this large-scale study do not exclude a potential reactivation of latent adenovirus in HIV infection and suggest that further effort should be directed to elucidating such a mechanism if it exists as well as investigating the specific role of certain adenovirus serotypes in provoking diarrhea during later stages of HIV infection.

The major cicatricial pemphigoid antigen is a 180-kD protein that shows immunologic cross-reactivities with the bullous pemphigoid antigen.

Recent studies have shown that sera from patients with cicatricial pemphigoid (CP) contained autoantibodies against epidermal antigens of molecular weight 230 kD and/or 180 kD by immunoblotting, similar to those recognized by bullous pemphigoid (BP) sera. Previous immunoprecipitation studies have shown that BP sera only precipitated the 230-kD antigen. To characterize the CP antigen(s) we tested 10 CP sera, 10 BP sera, and four controls by both immunoprecipitation of radiolabeled cells and immunoblotting of epidermal extracts. For immunoprecipitation, we used 0.5% NP-40 extracts of both normal human keratinocytes and Pam cells. All CP sera precipitated a 180-kD protein that co-migrated with the BP180 antigen precipitated by some individual BP sera. Two of these CP sera also faintly bound a 230-kD protein of similar molecular weight as the major BP230 antigen. CP and BP sera with an immunoblotting pattern of 180 kD immunoprecipitated a co-migrating 180-kD protein. CP sera reacting by immunoblotting with the 230-kD antigen precipitated the 180-kD and/or the 230-kD antigen. In contrast, BP sera reacting with the 230-kD antigen only precipitated this antigen. In further experiments, labeled 0.5% NP-40 extracts from Pam cells were first preabsorbed with a reference BP serum and then immunoprecipitated with CP sera. Under these conditions, CP sera that immunoprecipitated both 180-kD and 230-kD proteins with the standard procedure no longer precipitated these proteins. Our results suggest that a 180-kD protein is the major CP target-antigen that demonstrated immunologic cross-reactivities with the BP180 and the BP230 antigens.

The subclass distribution of IgG autoantibodies in cicatricial pemphigoid and epidermolysis bullosa acquisita.

To study the subclass distribution of autoantibodies and their complement-fixing capacity in cicatricial pemphigoid (CP) and epidermolysis bullosa acquisita (EBA) we studied the sera from 23 patients by both indirect immunofluorescence (IIF) on 4-microns cryostat sections of normal human skin and immunoblotting of epidermal or dermal extracts. Monoclonal antibodies of strict specificity for human IgG subclasses were used. Sera from 20 patients with BP served as controls. In addition, total IgG subclass levels were determined by indirect competitive ELISA in all sera. Complement binding capacity was studied by IIF using antibodies to C3 after incubation of skin section with autoantibodies and source of fresh complement. CP autoantibodies reacting with the 230-240 kD and/or the 180-kD epidermal bands showed an IgG4/IgG1 subclass restriction, with a predominance of IgG4 in 10 cases, of IgG1 in four. In BP sera, IgG4 and IgG1 autoantibodies were detected with a similar frequency (100% and 83%, respectively). In EBA sera, autoantibodies reacting with the 290 kD and 145 kD dermal bands also showed an IgG1/IgG4 restriction. Concordant results were obtained by IIF. However, the IIF method had a lower sensitivity for the detection of IgG4 CP antibodies and IgG1 EBA antibodies than immunoblotting. Finally, when CP antibodies were analyzed for their complement-binding activity, it was found that sera containing IgG4 autoantibodies alone never fixed complement whereas all complement-fixing CP sera had IgG1 autoantibodies, suggesting that only this subclass of antibodies is capable of fixing complement.