SUMMARY Molecular hybridisation using cloned hepatitis B virus DNA (HBV DNA) was applied to liver and serum samples from 46 children (39 with liver diseases and seven controls) for detection of HBV DNA sequences, free and integrated into the liver cell genome. HBV DNA integration was observed in 10 children. The young age of some of these cases indicates that such integration can occur early in liver disease and is not related to the duration of viral infection. Thirteen children exhibited serological evidence of active viral multiplication. All but one had free HBV DNA in liver tissue and integrated HBV DNA sequences were found in four cases. Integrated HBV DNA sequences alone were also detected in three children with neither HBV-antigens nor HBV DNA in serum. One had inactive cirrhosis, and the two others, chronic active hepatitis. Consequently DNA hybridisation may be useful for diagnosis, in the absence of serological signs of HBV infection; its specificity was enhanced in the present investigation by negative results in six children with autoimmune chronic active hepatitis. Taken together, the above results imply that HBV-DNA

Different impacts of alleles aLEPRA and a LELY as assessed versus a novel, virtually null allele of the SPTA1 gene in trans A subset of hereditary spherocytosis (HS) is associated with a reduction in red cell membrane spectrin (Agre et al, 1985). The SPTA1 gene, related to recessively inherited spherocytosis belongs to this subset. It is characterized by a pronounced and isolated decrease in spectrin. Under normal conditions, there is an overproduction of spectrin a-chains, however its extent has not been accurately assessed in humans. Wichterle et al (1996) elucidated the first mutations in a compound heterozygote. One mutation was null (mutation Prague). The other was polymorphic allele a LEPRA , carrying a c fi t transition at position )99 of intron 30, or mutation a LEPRA . An acceptor splice site at position )70 of intron 30 was activated, generating a frameshift. However, 16% of the transcripts escaped the abnormal splicing and allowed low a-chain production from allele a LEPRA (Wichterle et al, 1996). Allele a LEPRA accounts for nearly 5% of SPTA1 alleles among white people (Wichterle et al, 1996).Polymorphic allele a LELY is another low expression, more frequent, allele of the SPTA1 gene known to be involved with hereditary elliptocytosis (HE) (Alloisio et al, 1991;Wilmotte et al, 1993). A c fi t substitution at position )12 of intron 45 caused exon 46 (18 bp) to be skipped in 50% of the transcripts. The a-chains missing the corresponding 6 amino-acids were unsuitable for dimerisation (Wilmotte et al, 1997). The unaffected half of the transcripts ensured a 50% production from allele a LELY . The haploid set of a-chains in trans, carrying the HE mutation at or near the tetramerisation site, took over. More ab HE dimers than normal ab dimers were formed. The tetramerisation defect was enhanced and HE was aggravated. An effect of allele a LELY on the HS alleles of the SPTA1 gene has never been documented. Allele a LELY accounts for 20-30% of SPTA1 alleles, in all ethnic groups studied thus far (Maréchal et al, 1995;Basseres et al, 1998 Généthon,

Family history and blood cell morphology analysis in experienced hands are the first steps in discriminating AITP from inherited thrombocytopenia in children with isolated chronic thrombocytopenia. In contrast, bone marrow examination and search for specific autoantibodies using the MAIPA test are of little help. An isotopic platelet life span study, when available, should be performed before considering splenectomy to exclude the diagnosis of inherited thrombocytopenia, especially when steroids and/or IgG IV administration failed to raise the platelet count.

A simplified spot method for determination in serum of hepatitis B virus DNA (HBV DNA) by molecular hybridization is proposed. For simultaneous testing of 30 serum samples, it reduced to about 1 hr the duration of the steps preceding hybridization proper. The method also greatly reduced the loss of DNA during these steps and allowed more sensitive detection in samples of only 25 or 50 pl.HBV DNA was determined in 181 serum samples by this method, and the results were pooled with 67 previous determinations by the Southern blot technique. Results for the pool were then compared to those obtained with radioimmunoassay for serological HBV markers.Ninety-six of the 248 samples were HBV DNA positive. Eleven others gave variable or inconclusive results, probably due to low viral particle titers.Seventy-two HBsAg-and HBeAg-positive sera contained HBV DNA, confirming that HBeAg is a marker of active viral replication. Fourteen other HBsAg-and HBeAg-positive sera, obtained from eight patients, were either HBV DNA negative or oscillated between negative and positive, or, again, were weakly positive; serological follow-up in 7 patients showed seroconversion to antiHBe in 5 , 3 of which became HBsAg negative. Eight of the HBsAg-positive sera were negative or borderline for HBeAg but contained HBV DNA and may, therefore, have been infective; seven of these sera had anti-HBe. Six HBsAg-negative sera contained HBV DNA and may also have been infective; five of these exhibited HBV antibodies.These results indicate that molecular hybridization not only provides a more sensitive and direct method for detecting hepatitis B virus in serum but also defines additional serological patterns with predictive or epidemiological value.Molecular hybridization is a powerful tool for detecting complementary nucleic acid sequences. We have already used this technique to determine the presence of hepatitis B virus DNA (HBV DNA) in the liver and serum of patients with liver diseases. In this previous study, we used the Southern blot method (1) for both liver and