Constitutive "light" adaptation in rods from G90D rhodopsin: a mechanism for human congenital nightblindness without rod cell loss.

A dominant form of human congenital nightblindness is caused by a gly90-->asp (G90D) mutation in rhodopsin. G90D has been shown to activate the phototransduction cascade in the absence of light in vitro. Such constitutive activity of G90D rhodopsin in vivo would desensitize rod photoreceptors and lead to nightblindness. In contrast, other rhodopsin mutations typically give rise to nightblindness by causing rod cell death. Thus, the proposed desensitization without rod degeneration would be a novel mechanism for this disorder. To explore this possibility, we induced mice to express G90D opsin in their rods and then examined rod function and morphology, after first crossing the transgenic animals with rhodopsin knock-out mice to obtain appropriate levels of opsin expression. The G90D mouse opsin bound the chromophore and formed a bleachable visual pigment with lambda(max) of 492 nm that supported rod photoresponses. (G+/-, R+/-) retinas, heterozygous for both G90D and wild-type (WT) rhodopsin, possessed normal numbers of photoreceptors and had a normal rhodopsin complement but exhibited considerable loss of rod sensitivity as measured electroretinographically. The rod photoresponses were desensitized, and the response time to peak was faster than in (R+/-) animals. An equivalent desensitization resulted by exposing WT retinas to a background light producing 82 photoisomerizations rod(-1) sec(-1), suggesting that G90D rods in darkness act as if they are partially "light-adapted." Adding a second G90D allele gave (G+/+, R+/-) animals that exhibited a further increase of equivalent background light level but had no rod cell loss by 24 weeks of age. (G+/+, R-/-) retinas that express only the mutant rhodopsin develop normal rod outer segments and show minimal rod cell loss even at 1 year of age. We conclude that G90D is constitutively active in mouse rods in vivo but that it does not cause significant rod degeneration. Instead, G90D desensitizes rods by a process equivalent to light adaptation.

Absence of Ki-ras mutations in exocrine pancreatic tumors from male rats chronically exposed to gabapentin.

Human pancreatic malignancies originating from duct cells most frequently demonstrate activation of Ki-ras gene by G-to-A transition at codons 12 and 13. Rat pancreatic exocrine tumors more frequently and almost exclusively derive from acinar cells and thus differ morphologically from human pancreatic neoplasms. Male Wistar rats fed with 2% gabapentin (1-(aminomethyl)cyclohexane acetic acid) in diet for 2 years developed pancreatic exocrine adenomas and adenocarcinomas. To study the mutations in Ki-ras gene, rat pancreatic proliferative lesions induced by gabapentin were retrospectively analyzed by PCR amplification of DNA isolated from paraffin sections of formalin-fixed rat pancreatic adenomas and adenocarcinomas, using specific primers for regions encoding exon 1 (codon 12/13) and exon 2 (codon 61). The amplified 110-bp fragments of exon 1 and exon 2 were analyzed for mutations at codon 12/13 and 61. The results showed Ki-ras mutations at codon 12 in human pancreatic carcinomas. Novel mutations GGT-to-TGT and GGT-to-CGT were detected at codon 12 in 1/5 and 2/5 human pancreatic tumors. Rat adenomas or carcinomas induced by gabapentin expressed wild type sequences at codons 12, 13 and 61. These findings were confirmed by allele-specific oligonucleotide hybridization, single-strand confirmation polymorphism of exon 1 and direct sequencing of exon 1 and exon 2. The absence of mutations in these rat pancreatic tumors suggests that these tumors do not correspond to the human tumors, and that the pathogenesis of this rodent tumor formation may follow different molecular mechanisms.

Correction of mucolipidosis III in vitro by gene transfer.

Mucolipidosis II (ML II, I-cell disease) and mucolipidosis III (ML III, pseudo-Hurler polydystrophy) are human autosomal recessive genetic disorders resulting from deficient UDP-N-acetylglucosamine:lysosomal enzyme precursor N-acetylglucosamine-1-phosphotransferase (GNPT) activity. Normally, this enzyme is involved in the processing of most lysosomal enzymes. Cultured fibroblasts from individuals with either disorder are deficient in a broad array of lysosomal enzymes as a result of the diminished GNPT activity. We report the correction of this phenotype by fusing transformed ML III cells generated for this study to lethally irradiated rodent cells. This method of gene transfer does not require selection for the gene of interest, animal models, nor any knowledge of the gene product except a screening method for its presence. It has generated corrected cell hybrids that contain approximately 1% hamster-derived sequences. These cell lines, which contain the hamster analogue to the human phosphotransferase gene, are useful for the molecular cloning of the gene defective in ML II and ML III.

Molecular biology of the alpha-L-fucosidase gene and fucosidosis.

Human alpha-L-fucosidase, a lysosomal enzyme, hydrolyzes alpha-L-fucose from glycolipids and glycoproteins. Its activity is deficient in human fucosidosis an autosomal recessive disease. In order to understand the molecular basis of this lysosomal storage disorder we have cloned several cDNAs coding for human alpha-L-fucosidase from a human hepatoma and a human liver cDNA library constructed in lambda gt11. Compiling the cDNA sequences of these clones we have identified 1,829 base pairs (bp) encoding human alpha-L-fucosidase. This includes an open reading frame of 1,172 bp, a consensus polyadenylation signal AAT AAA and a poly(A)+ tail. The sequence is incomplete at the 5'-end, and clones encoding the amino terminus of the native protein, the propeptide and leader signal have not yet been isolated. The open reading frame encodes for 390 amino acids with a calculated Mr of 45,557. This represents 78-95% of the mature processed alpha-L-fucosidase. The availability of these cDNA clones has enabled us to map the structural gene for alpha-L-fucosidase to chromosome 1p34.1-1p36.1 by Southern blot analysis of DNA from human-rodent somatic cell hybrids and by in situ hybridization. Furthermore, a Pvu II restriction fragment length polymorphism (RFLP) has been identified at the human alpha-L-fucosidase gene locus. Analysis of mRNA by Northern blotting gives a major species of 2.25 kb. In 4 patients with fucosidosis no mRNA signal was detected and Western blots gave no immunoreactive enzyme. Southern blotting after Eco RI digestion in two fucosidosis families revealed a banding abnormality (extra 6-kb band).(ABSTRACT TRUNCATED AT 250 WORDS)

Chromosome 1 localization of the human alpha-L-fucosidase structural gene with a homologous site on chromosome 2.

Two cDNA clones coding for human alpha-L-fucosidase, one from the coding region and the other primarily from the 3' untranslated region, were used to map the location of the alpha-L-fucosidase gene. Southern filter analysis of somatic cell hybrid lines mapped the structural gene to the short arm of human chromosome 1, and in situ hybridization to chromosomes of human leukocytes further localized the homologous area to the 1p36.1----p34.1 region, with the most likely location being the distal region of 1p34. Further Southern filter analysis detected a second site of homology on chromosome 2. This alpha-L-fucosidase-like site has been designated FUCA1L.