Contrasting effects of ENU induced embryonic lethal mutations of the quaking gene.

Multiple alleles of the quaking (qk) gene have a variety of phenotypes ranging in severity from early embryonic death to viable dysmyelination. A previous study identified a candidate gene, QKI, that contains an RNA-binding domain and encodes at least three protein isoforms (QKI-5, -6 and -7). We have determined the genomic structure of QKI, identifying an additional alternative end in cDNAs. Further we have examined the exons and splice sites for mutations in the lethal alleles qkl-1, qkkt1, qkk2, and qkkt3. The mutation in qkl-1 creates a splice site in the terminal exon of the QKI-6 isoform. Missense mutations in the KH domain and the QUA1 domains in qkk2 and qkkt3, respectively, indicate that these domains are of critical functional importance. Although homozygotes for each ENU induced allele die as embryos, their phenotypes as viable compound heterozygotes with qkv differ. Compound heterozygous qkv animals carrying qkkt1, qkk2, and qkkt3 all exhibit a permanent quaking phenotype similar to that of qkv/qkv animals, whereas qkv/qkl-1 animals exhibit only a transient quaking phenotype. The qkl-1 mutation eliminates the QKI-5 isoform, showing that this isoform plays a crucial role in embryonic survival. The transient quaking phenotype observed in qkv/qkl-1 mice indicates that the QKI-6 and QKI-7 isoforms function primarily during myelination, but that QKI-5 may have a concentration-dependent role in early myelination. This mutational analysis demonstrates the power of series of alleles to examine the function of complex loci and suggests that additional mutant alleles of quaking could reveal additional functions of this complex gene.

Genetic mapping of hph2, a mutation affecting amino acid transport in the mouse.

We describe the genetic mapping of hyperphenylal-aninemia 2 (hph2), a recessive mutation in the mouse that causes deficient amino acid transport similar to Hartnup disorder, a human genetic amino acid transport disorder. The hph2 locus was mapped in three separate crosses to identify candidate genes for hph2 and a region of homology in the human genome where we propose the Hartnup Disorder gene might lie. The mutation maps to mouse Chromosome (Chr) 7 distal of the simple sequence length polymorphism (SSLP) marker D7Mit140 and does not recombine with D7Nds4, an SSLP marker in the fibroblast growth factor 3 (Fgf3) gene. Unexpectedly, the mutant chromosome affects recombination frequency in the D7Mit12 to D7Nds4 interval.

A candidate mouse model for Hartnup disorder deficient in neutral amino acid transport.

The mutant mouse strain HPH2 (hyperphenylalaninemia) was isolated after N-ethyl-N-nitrosourea (ENU) mutagenesis on the basis of delayed plasma clearance of an injected load of phenylalanine. Animals homozygous for the recessive hph2 mutation excrete elevated concentrations of many of the neutral amino acids in the urine, while plasma concentrations of these amino acids are normal. In contrast, mutant homozygotes excrete normal levels of glucose and phosphorus. These data suggest an amino acid transport defect in the mutant, confirmed in a small reduction in normalized values of 14C-labeled glutamine uptake by kidney cortex brush border membrane vesicles (BBMV). The hyperaminoaciduria pattern is very similar to that of Hartnup Disorder cases also show niacin deficiency symptoms, of Hartnup Disorder cases also show niacin deficiency symptoms, which are thought to be multifactorially determined. Similarly, the HPH2 mouse exhibits a niacin-reversible syndrome that is modified by diet and by genetic background. Thus, HPH2 provides a candidate mouse model for the study of Hartnup Disorder, an amino acid transport deficiency and a multifactorial disease in the human.

Genetic evaluation of candidate genes for the Mom1 modifier of intestinal neoplasia in mice.

As genetic mapping of quantitative trait loci (QTL) becomes routine, the challenge is to identify the underlying genes. This paper develops rigorous genetic tests for evaluation of candidate genes for a QTL, involving determination of allelic status in inbred strains and fine-structure genetic mapping. For the Mom1 modifier of intestinal adenomas caused by ApcMin, these tests are used to evaluate two candidate genes: Pla2g2a, a secretory phospholipase, and Rap1GAP, a GTPase activating protein. Rap1GAP passes the first test but is excluded by a single fine-structure recombinant. Pla2g2a passes both tests and is a strong candidate for Mom1. Significantly, we also find that ApcMin-induced adenomas remain heterozygous for the Mom1 region, consistent with Mom1 acting outside the tumor lineage and encoding a secreted product.

Gene therapy for phenylketonuria: phenotypic correction in a genetically deficient mouse model by adenovirus-mediated hepatic gene transfer.

Classical phenylketonuria (PKU), which predisposes affected individuals to severe mental retardation, is caused by a deficiency of hepatic phenylalanine hydroxylase (PAH). A recombinant adenoviral vector containing the human PAH cDNA was constructed and administered to PAH-deficient mice (strain PAHenu2). The hyperphenylalaninemic phenotype of these animals was completely normalized within 1 week of treatment. Although this therapeutic effect did not persist, analysis of the relationship between hepatic PAH activity and serum phenylalanine levels indicated that only 10-20% of normal enzymatic activity in the mouse liver is sufficient to restore normal serum phenylalanine levels. These results demonstrate that PKU and other metabolic disorders secondary to hepatic deficiencies can be completely corrected by gene therapy when more persistent vector systems are developed.

Near-microscopic magnetic resonance imaging of the brains of phenylalanine hydroxylase-deficient mice, normal littermates, and of normal BALB/c mice at 9.4 Tesla.

The near-microscopic resolution of the mouse brain, by magnetic resonance imaging (MRI) at 9.4 T, permits in situ examination of the entire brain and longitudinal studies of neural development. MRI can be utilized to reveal brain structure at a resolution of 100 microns in the X, Y, and Z planes of brain, to differentiate the gray from white (myelin-rich) matter, and to reveal the ventricular compartments. The present report describes the structure of normal BALB/c mouse brain as revealed by imaging at 9.4 T and by histological stains; the structure of normal brain is compared with that from a phenylalanine hydroxylase-deficient mouse mutant line (Pah(enu2)) and those from normal littermates. The brains of patients with phenylketonuria (PKU) were reported to have demyelination and other structural abnormalities revealed by magnetic resonance imaging (MRI). Therefore, high-resolution MRI was used to examine the brain of this mutant, an animal model for the study of human phenylketonuria. Our study revealed no evidence of demyelination or other abnormalities in the brains of Pah(enu2) mice. Histologically, the mutant and normal mouse brains appear similar. This is consistent with a recent study from our laboratory which demonstrated that the histology of the brain of an untreated male patient, who died with PKU at the age of 29, was similar to control brain with the exception of changes directly related to visual blindness and seizures experienced by the patient.

A 1.2-Mb YAC contig spans the quaking region.

We describe here a 1.2-Mb yeast artificial chromosome (YAC) contig within the region of mouse chromosome 17 between Brachyury (T) and D17Rp17e, and spanning the quaking (qk) region. We describe six new probes distributed across 1.2 Mb: D17Leh502, D17Leh503, D17Leh504, D17Leh505, D17Leh506, and D17Leh507. Probes D17Leh502 and D17Leh507 are at the extreme ends of the YAC contig. With the exception of D17Leh507, all of these probes are within a deletion associated with the quaking(viable) (qkv) allele of quaking. We have positioned these probes on a detailed YAC physical map together with two previously published probes, D17Leh508 and D17Aus119. We show here that D17Leh508 is also within the qkv deletion. Genetic mapping of D17Leh504 and D17Leh507 on two high-resolution genetic crosses carrying qkv and quaking(lethal-1) (qkl-1) alleles shows that these probes do not recombine with quaking and are therefore within 0.04 cM of qkv and 0.05 cM of qkl-1 mutations. The deletion breakpoint contained within the YAC contig has been positioned to within 90 kb by restriction mapping of wildtype and mutant DNA. This contig will form the basis for identification and mapping of expressed sequences and for an investigation of genome organization.

Detailed physical and genetic mapping in the region of plasminogen, D17Rp17e, and quaking.

We present here a detailed physical map encompassing over 600 kb of mouse Chromosome (Chr) 17 in the region of plasminogen, D17Rp17e, and quaking. This region is cloned in yeast artificial chromosomes (YACs). We have identified several CpG islands within this region from pulsed field gel mapping of mouse genomic DNA and YAC DNA. Five new DNA probes have been generated. One, D17Leh514, is a minimum of about 90 kb distal to plasminogen. Four, D17Leh513, D17Leh512, D17Leh511, and D17Leh510, are proximal to D17Rp17e, the closest previously described genetic marker to quaking viable and quaking lethal-1 mutations. We have genetically mapped D17Leh511 to within 0.15 cM of these mutations. The genetic distance to D17Rp17e from D17Leh511 is also 0.15 cM; the physical distance of less than 360 kb (minimum 200 kb) is consistent with an approximation of 2 Mbp per cM.

Mouse models of human phenylketonuria.

Phenylketonuria (PKU) results from a deficiency in phenylalanine hydroxylase, the enzyme catalyzing the conversion of phenylalanine (PHE) to tyrosine. Although this inborn error of metabolism was among the first in humans to be understood biochemically and genetically, little is known of the mechanism(s) involved in the pathology of PKU. We have combined mouse germline mutagenesis with screens for hyperphenylalaninemia to isolate three mutants deficient in phenylalanine hydroxylase (PAH) activity and cross-reactive protein. Two of these have reduced PAH mRNA and display characteristics of untreated human PKU patients. A low PHE diet partially reverses these abnormalities. Our success in using high frequency random germline point mutagenesis to obtain appropriate disease models illustrates how such mutagenesis can complement the emergent power of targeted mutagenesis in the mouse. The mutants now can be used as models in studying both maternal PKU and somatic gene therapy.

Meiotic mapping of murine chromosome 17: the string of loci around l(17)-2Pas.

We describe a genetic analysis of l(17)-2Pas, an embryonic lethal mutation on murine chromosome 17. Males transmitted the l(17)-2 allele to only 38% of their offspring, whereas females transmitted this allele at 50%. Two-point crosses revealed tight linkage between l(17)-2 and brachyury (T), and deletion mapping placed l(17)-2 outside of the hairpin-tail deletion (Thp). To map this mutation more precisely, we intercrossed hybrid mice that carry distinct alleles at many classical and DNA loci on chromosome 17 and obtained 172 animals recombinant in the T to H-2 region. Strong positive interference was observed over the 14 cM interval from T to H-2K. Thus, a single recombinant can be informative; one such recombinant places l(17)-2 distal of the molecular marker D17Leh66D. Robust genetic maps can be constructed with multilocus crosses that share anchor loci. DNA markers can be interpolated onto these maps retrospectively.

Pahhph-5: a mouse mutant deficient in phenylalanine hydroxylase.

Mutant mice exhibiting heritable hyperphenylalaninemia have been isolated after ethylnitrosourea mutagenesis of the germ line. We describe one mutant pedigree in which phenylalanine hydroxylase activity is severely deficient in homozygotes and reduced in heterozygotes while other biochemical components of phenylalanine catabolism are normal. In homozygotes, injection of phenylalanine causes severe hyperphenylalaninemia and urinary excretion of phenylketones but not hypertyrosinemia. Severe chronic hyperphenylalaninemia can be produced when mutant homozygotes are given phenylalanine in their drinking water. Genetic mapping has localized the mutation to murine chromosome 10 at or near the Pah locus, the structural gene for phenylalanine hydroxylase. This mutant provides a useful genetic animal model affected in the same enzyme as in human phenylketonuria.

Mapping to molecular resolution in the T to H-2 region of the mouse genome with a nested set of meiotic recombinants.

We describe a meiotic fine-structure mapping strategy for achieving molecular access to developmental mutations in the mouse. The induction of lethal point mutations with the potent germ-line mutagen N-ethyl-N-nitrosourea has been reported. One lethal mutation of prime interest is an allele at the quaking locus on chromosome 17. To map this mutation, quaking(lethal-1), we have intercrossed hybrid mice that carry distinct alleles at many classical and DNA marker loci on proximal chromosome 17. From this cross we have obtained 337 animals recombinant in the T to H-2 region. This number of crossovers provides a mapping resolution in the size range of single mammalian genes if recombinational hot spots are absent. DNA samples obtained from these recombinant animals can be used retrospectively to map any restriction fragment length polymorphism in the region. This set of DNA samples has been used to map the molecular marker D17RP17 just distal of quaking(lethal-1). With the nested set of crossover DNA samples and appropriate cloning techniques, this tightly linked marker can be used to clone the quaking locus.

Saturation germ line mutagenesis of the murine t region including a lethal allele at the quaking locus.

The proximal region of mouse chromosome 17 contains many genes affecting embryonic development, germ cell differentiation, and the immune system. Although the study of natural variation, including t haplotypes, has yielded some information about the function of these genes, spontaneous variants often exhibit manifold genetic effects and are generally not carried on inbred backgrounds. To clearly connect phenotypes with the actions of individual genes, mutants in which genes are altered singly are needed. Therefore, we used a highly efficient point mutagen, N-ethyl-N-nitrosourea, in combination with classical breeding schemes to induce and identify recessive lethal mutations in the t region. Of 350 mutagenized gametes examined, at least 10 independent recessive embryonic lethal mutations have been identified; an additional two are perinatal lethals. A spontaneous brachyury mutation, TWis, arose on a genetic background that permits high-resolution mapping of the induced recessive mutations against cloned DNA sequences from the t region. One lethal mutation is an allele at the quaking locus. The multiple alleles of quaking and the feasibility of high-resolution mapping permit investigation of the pleiotropic action of this locus in mammalian development.

Teratocarcinoma transplantation antigens are encoded in the H-2 region.

Evidence is presented for the existence of teratocarcinoma transplantation antigens (Gt) encoded within the H-2 complex and present also on adult tissues. It has not been possible to separate these Gt loci from H-2 by recombination, and Gt factors map to each end of the H-2 complex. Previous reports indicating separation of all Gt loci from H-2 are reinterpreted. One class of such apparent recombinants has been shown to result from the outgrowth of tumor variants in mice of resistant genotype.

Strong teratocarcinoma transplantation loci, Gt-1 and Gt-2, flank H-2.

Evidence is presented for the existence of two strong murine teratocarcinoma transplantation antigens (Gt) on the cell line PCC3. It is shown that the loci governing expression of these antigens are linked to the H-2 complex. These loci have been further mapped with respect to the brachyury marker (T) and H-2: Gt-1 lies 5 +/- 2 crossover units proximal to H-2 and 12 +/- 2 crossover units distal to T; Gt-2 lies 21 +/- 4 crossover units distal to H-2. It is possible that these strong transplantation antigens provide an embryonic analogue to the adult major histocompatibility system.

The genetics of teratocarcinoma transplantation: tumor formation in allogeneic hosts by the embryonal carcinoma cell lines F9 and PCC3.

Two cultured lines of murine embryonal carcinoma, F9 and PCC3, have been grafted to a variety of allogeneic hosts. The host strains have been classified by their resistance or sensitivity to these carcinomas. Resistance seems to be immunological in nature.Allograft rejection does not correlate withH-2 haplotype, and seems to be controlled by a limited number of recessive factors, presumably histocompatibility genes. We infer that these factors have limited polymorphism in the mouse species. Recombinational analysis of strain A/He has revealed the presence of a recessive factor linked to theH-2 locus. Tumor resistance of strains C57BL/6 and AKR appears to result from the interaction of dominant or semi-dominant factors in theH-2 region with other recessive elements in the genetic background.Though F(1) hybrids between resistant mouse strains and the syngeneic strain 129 are largely tumor-sensitive, a low level of hybrid resistance to F9 has been observed and shown to be eliminated by X-irradiation.

Guanosine 3',5'-cyclic monophosphate and the in vitro physiology of frog photoreceptor membranes.

Frog rod outer segments freshly detached from dark-adapted retinas contain approximately 1-2 molecules of guanosine 3',5'-cyclic monophosphate (cyclic GMP) for every 100 molecules of visual pigment present. This cyclic GMP decays to 5'-GMP, and the conversion is accelerated upon illumination of the outer segments. Bleaching one rhodopsin molecule can lead to the hydrolysis of 1,000-2,000 molecules of cyclic GMP within 100-300 ms. The decline in cyclic GMP concentration becomes larger as illumination increases, and varies with the logarithm of light intensity at levels which bleach between 5 X 10(2) and 5 X 10(5) rhodopsin molecules per outer segment-second. Light suppression of plasma membrane permeability, assayed in vitro as light suppression of outer segment swelling in a modified Ringer's solution, occurs over this same range of light intensity. The correlation between cyclic GMP and permeability or swelling is maintained in the presence of two pharmacological perturbations: papaverine, a phosphodiesterase inhibitor, increases both cyclic GMP levels and the dark permeability of the plasma membrane; and beta,gamma-methylene ATP increases the effectiveness of light in suppressing both permeability and cyclic GMP levels.