Lowe's oculocerebrorenal syndrome (OCRL) is a human X-linked developmental disorder of unknown pathogenesis and has a pleiotropic phenotype affecting the lens, brain and kidneys. The OCRL locus has been mapped to Xq25-q26 by linkage and by finding de novo X; autosome translocations at Xq25-q26 in two unrelated females with OCRL. Here we use yeast artificial chromosomes with inserts that span the X chromosomal breakpoint from a female OCRL patient in order to isolate complementary DNAs for a gene that is interrupted by the translocation. We show that the transcript is absent in both female OCRL patients with X; autosome translocations and that it is absent or abnormally sized in 9 of 13 unrelated male OCRL patients with no detectable genomic rearrangement. The open reading frame encodes a new protein with 71% similarity to human inositol polyphosphate-5-phosphatase. Our results suggest that OCRL may be an inborn error of inositol phosphate metabolism.

Pathogenic Yersinia species and Pseudomonas aeruginosa share a similar type III secretion/translocation system. The translocation system consists of 3 secreted proteins, YopB/PopB, YopD/PopD, and LcrV/PcrV; the latter is known to be a protective antigen. In an in vitro assay, the translocation system causes the lysis of erythrocytes infected with wild-type (wt) P. aeruginosa. wt Y. enterocolitica is not hemolytic, but a multiknockout mutant deprived of all the effectors and of YopN ( Delta HOPEMN) is hemolytic. In the presence of antibodies against PcrV and Y. pestis LcrV, the hemolytic activity of P. aeruginosa was inhibited. Similarly, the hemolytic activity of Delta HOPEMN was inhibited in the presence of anti-LcrV antibodies. The assembly of the translocon, composed of PopB/D and YopB/D proteins, was disturbed in immunoprotected erythrocyte membranes, mimicking the phenotypes of V knockout mutants. Thus, protective antibodies against the V antigens of Yersinia species and P. aeruginosa act at the level of the formation of the translocon pore in membranes of infected host cells by blocking the function of LcrV/PcrV. The hemolysis assay could be adapted for high-throughput screening of anti-infectious compounds that specifically target the type III translocon.

Laron dwarfism is associated with resistance to growth hormone (GH). To investigate its genetic basis, we used genetic linkage to test whether the disorder results from a defect in the gene for the human GH receptor. Denaturing gradient gel electrophoresis and sequencing of specific GH-receptor-gene fragments allowed us to characterize specific intragenic DNA markers in 35 control subjects of Mediterranean descent, for use in linkage studies. In two Mediterranean families in which the parents were consanguineous and some of the children had Laron dwarfism, the disease trait and DNA polymorphisms were inherited together. Moreover, an analysis of the GH-receptor-gene RNA transcripts in lymphocytes from one of these families allowed us to identify a thymidine-to-cytosine substitution that generated a serine in place of a phenylalanine at position 96 in the extracellular coding domain of the mature protein. This defect probably affects the receptor adversely and is probably responsible for the lack of plasma GH-binding activity in the patients. This mutation was not found in the GH-receptor genomic sequences of seven unrelated subjects with Laron dwarfism who belonged to different population groups. An analysis of the GH-receptor markers in these patients indicated that different gene frameworks (polymorphic sites within the single gene) were associated with the mutant alleles. We conclude that Laron dwarfism is due to abnormalities in the gene for GH receptor, which may differ from family to family.

We performed positional cloning of genes carried on yeast artificial chromosomes that span a human translocation breakpoint associated with a human disease and isolated by chance human and bovine genes with strong homology to the S. cerevisiae genes, SNF2/SWI2 and STH1, and the D. melanogaster gene brahma. We report here sequence analysis, expression data, and functional studies for this human SNF2-like gene (hSNF2L) and its bovine homolog (bovSNF2L). Despite strong homology at the amino acid level, hSNF2L is not capable of complementing the yeast mutations snf2 or sth1 in S. cerevisiae. Furthermore, in contrast to SNF2 itself, a fusion protein consisting of the DNA binding domain of LexA and hSNF2L did not transactivate a reporter gene downstream of LexA binding sites in a yeast expression system. The strong similarity between hSNF2L and these yeast and drosophila genes suggest that the mammalian genes are part of an evolutionarily conserved family that has been implicated as global activators of transcription in yeast and fruitflies but whose function in mammals remains unknown.