Cosmid clones containing the gene for human adenosine deaminase (ADA) were isolated. The gene is 32 kb long and split into 12 exons. The exact sizes and boundaries of the exon blocks including the transcription start sites were determined. The sequence upstream from this cap site lacks the TATA and CAAT boxes characteristic for eukaryotic promoters. Nevertheless, we have shown in a functional assay that a stretch of 135 bp immediately preceding the cap site has promoter activity. This 135‐bp DNA fragment is extremely rich in G/C residues (82%). It contains three inverted repeats that allow the formation of cruciform structures, a 10‐bp and a 16‐bp direct repeat and five G/C‐rich motifs (GGGCGGG) disposed in a strikingly symmetrical fashion. Some of these structural features were also found in the promoter region of other genes and we discuss their possible function. Knowledge of the exact positions of the intron‐exon boundaries allowed us to propose models for abnormal RNA processing that occurs in previously investigated ADA‐deficient cell lines.

We have cloned and sequenced an adenosine deaminase (ADA) gene from a patient with severe combined immunodeficiency (SCID) caused by inherited ADA deficiency. Two point mutations were found, resulting in amino acid substitutions at positions 80 (Lys to Arg) and 304 (Leu to Arg) of the protein. Hybridization experiments with synthetic oligonucleotide probes showed that the determined mutations are present in both DNA and RNA from the ADA‐SCID patient. In addition, wild‐type sequences could be detected at the same positions, indicating a compound heterozygosity. Studies with ADA expression clones mutagenized in vitro showed that the mutation at position 304 is responsible for ADA inactivation.

We have investigated the structural gene for adenosine deaminase (ADA) in a female infant with ADA deficiency associated severe combined immune deficiency (ADA-SCID) disease and her family by DNA restriction-fragment-length analysis. In this family a new ADA-specific restriction-fragment-length variant was detected, which involves a 3.2-kb deletion spanning the ADA promoter as well as the first exon. It was found that the patient, who was born to a consanguineous couple, was homozygous and both her parents and her brother were heterozygous for the deletion. No ADA-specific mRNA could be detected by hybridization in fibroblasts derived from this patient. Thus the patient was established to be homozygous for a true null ADA allele. In the light of the apparently normal development of most tissues except the lymphoid tissue the above finding directly questions the classification of ADA as a 'housekeeping' enzyme.

The expression of the enzyme adenosine deaminase (ADA) decreases in the course of the differentiation of the human promyelocytic leukemic cell line HL-60, dependent on the pathway chosen. Differentiation to monocytes as induced by the phorbol ester TPA leads to a 50% reduction of enzyme activity. Induction to myeloid cells as induced by DMSO has a slower and less extensive (75% remaining activity) effect. The reduction in ADA enzymatic activity is preceded by a 5-10 fold reduction in ADA-specific mRNA which is also more rapid during TPA-induced differentiation. In contrast, c-myc mRNA expression is both in TPA- and DMSO-induced differentiation reduced to less then 5% of its initial level within 4h. Nuclear run-on analysis revealed that the reduction of c-myc-mRNA expression during both TPA- and DMSO-induced differentiation could be ascribed to the abolition of transcription of the third exon, whereas no change in the transcription of the first exon could be observed. No change could be detected in the rate of transcription of either the 5' and 3' parts of the ADA gene during TPA- and DMSO-induced differentiation, indicating that the expression of the ADA gene in HL-60 is controlled at a posttranscriptional level.

No abstract