The human GATA1, hGATA1 (previously called NF‐E1, GF‐1 or Eryf‐1), a major sequence‐specific DNA‐binding protein of the erythrocytic lineage, is a member of a zinc‐finger family of DNA‐binding proteins. We report here the cloning of a human cDNA for a new member of this family. This member, called hGATA3, has 85% amino acid homology with hGATA1 in the DNA‐binding domain and no homology elsewhere in the protein. Unlike hGATA1, hGATA3 is not localized on the X chromosome and we map it to the 10p15 band of the human genome. Northern blot analysis indicates that this factor is a T‐cell specific transcription factor, present before activation and up‐regulated during T‐cell activation. The encoded hGATA3 protein, made in an in vitro transcription‐translation assay, binds the WGATAR motif present in the human T‐cell receptor (TCR) delta gene enhancer and, by transfection in HeLa cells, we show that hGATA3 can transactivate this TCR delta gene enhancer. Interestingly this enhancer binds and is also transactivated by hGATA1. Conversely, the promoter of the human glycophorin B (GPB), which is erythroid‐specific and contains two WGATAR motifs, binds and is transactivated by hGATA1 and, to a lesser extent, by hGATA3. These results indicate that the activation of specific genes by hGATA1 or hGATA3 is partly governed by the lineage expression of these two factors during haematopoiesis and that, in the T‐cell lineage, hGATA3 binds the human TCR delta gene enhancer and is involved in its expression.

Erythrocyte bisphosphoglycerate mutase (BPGM) deficiency is a rare disease associated with a decrease in 2,3-diphosphoglycerate concentration. A complete BPGM deficiency was described in 1978 by Rosa et al (J Clin Invest 62:907, 1978) and was shown to be associated with 30% to 50% of an inactive enzyme detectable by specific antibodies and resulting from an 89 Arg-->Cys substitution. The propositus' three sisters exhibited the same phenotype, while his two children had an intermediate phenotype. Samples from the family were examined using polymerase chain reaction and allele-specific oligonucleotide hybridization and sequencing techniques. Amplification of erythrocyte total RNA from the propositus' sister around the 89 mutation indicated the presence of two forms of messenger RNAs, a major form with the 89 Arg-->Cys mutation and a minor form with a normal sequence. Sequence studies of the propositus' DNA samples indicated heterozygosity at locus 89 and another heterozygosity with the deletion of nucleotide C 205 or C 206. Therefore, the total BPGM deficiency results from a genetic compound with one allele coding for an inactive enzyme (mutation BPGM Creteil I) and the other bearing a frameshift mutation (mutation BPGM Creteil II). Examination of the propositus' two children indicated that they both inherited the BPGM Creteil I mutation.

The contribution of the hepatitis B virus enhancers I and II in the regulation of the activity of the core and the X promoters was assessed in transgenic mice. Surprisingly, despite the presence of heterologous promoters linked 5 of the X gene, the transgene expression is mostly due to core promoter (Cp) activity present in the X coding sequence. Moreover, the restriction of Cp activity to hepatic tissue required the combined action of both enhancers I and II, whereas the proximity of these two enhancers was insufficient to confer tissue specificity on Xp activity. Furthermore, the liver-specific activity of the Cp was developmentally regulated in an enhancer I-independent manner.

The human erythrocyte 2,3‐bisphosphoglycerate mutase (BPGM) is a multifunctional enzyme which controls the metabolism of 2,3‐diphosphoglycerate, the main allosteric effector of haemoglobin. Several cDNA banks were constructed from reticulocyte mRNA, either by conventional cloning methods in pBR322 and screening with specific mixed oligonucleotide probes, or in the expression vector lambda gt 11. The largest cDNA isolated contained 1673 bases [plus the poly(A) tail], which is slightly smaller than the size of the intact mRNA as estimated by Northern blot analysis (approximately 1800 bases). This cDNA encodes for a protein of 258 residues; the protein yielded 34 tryptic peptides which were subsequently isolated by h.p.l.c. Our nucleotide sequence data were entirely confirmed by the amino acid composition of these tryptic peptides and reveal several major differences from the published sequence; the revised amino acid sequence of human BPGM is presented. These findings represent the first step in the study of the expression and regulation of this enzyme as a specific marker of the erythroid cell line.