Search

H1t is synthesized in mid to late pachytene spermatocytes of the male germ line and is the only tissue-specific member of the mammalian H1 histone family. As a step toward identifying DNA sequences that confer its tissue-specific expression, we have produced transgenic mice containing the intact rat H1t gene as well as a H1t-lacZ fusion gene. Transgenic mice carrying a 6.8-kilobase fragment of rat genomic DNA encompassing the H1t gene expressed rat H1t at high levels in the testis and in no other organ examined. H1t fragments truncated to within 141 base pairs (bp) of the gene in the 5' direction or within 837 bp in the 3' direction retained testis specificity. Expression of rat H1t protein was also evident in the testes of the transgenic mice, and in some lines the level of rat H1t exceeded that of the mouse protein. The stage of spermatogenesis of transgene expression was assessed by following appearance of transgenic mRNA in developing mice and by immunohistochemistry using an antiserum to rat H1t. In lines from three different constructs, expression was restricted to germinal cells, although in two strongly expressing lines the transgenes were expressed somewhat prematurely in preleptotene spermatocytes. An H1t(-948/+71)-lacZ fusion was also expressed specifically in the spermatocytes and round spermatids of a transgenic line, confirming that sequences sufficient for correct tissue and developmental expression lie within this 1,019-bp segment of the gene.

A rapid and highly sensitive HPLC assay method was developed to measure small amounts of imipramine and its major metabolite, desipramine. The assay involved simple extraction procedures using clomipramine as the internal standard. The mobile phase consisted of acetonitrile (60%) and 0.01 M triethylamine in distilled water (40%) with the pH adjusted to 3.0. Separations were achieved on a C18 column and the effluent measured for UV absorption at 260 nm. The chromatographic separation was excellent, with no interference from endogenous serum constituents. This assay was suitable for measuring drug concentrations in the range of 10-1000 ng/ml using a 0.1-ml serum sample. The method was applied to a drug disposition study in transgenic mice with increased plasma alpha 1-acid glycoprotein.

In laboratory mice we previously defined a gene, rol (resistance to osmotic lysis), based on its effect on erythrocyte osmotic fragility. Here we report a physiological characterization of rol gene action utilizing congenic strains developed for the purpose; these two strains have a common genetic background and differ only by the two alleles of rol, susceptible (rols) or resistant (rolr). In comparison to rols/s erythrocytes, rolr/r cells have a reduced mean cell volume, a higher mean corpuscular hemoglobin concentration and hemolytic volume, and respond differently to swelling induced by ion influx. Rolr/r erythrocytes also have reduced cell water and K, which are associated with a threefold higher activity of the Na-K-Cl cotransporter (measured as ouabain-resistant, bumetanide-sensitive 86Rb influx) and 30% higher Na pump activity. Apart from differences in ion transport and water content, the content of 2,3-diphosphoglycerate (2,3-DPG) in rolr/r cells is 15% lower than in rols/s ones. Analyses of membrane structural components revealed no rol-associated differences in their phospholipid or fatty acid content, nor were strain differences evident among the membrane and cytoskeletal proteins and their posttranslational modifications (phosphorylation and fatty acylation). Rol is not the structural gene for either the alpha- or the beta-chain of hemoglobin and has no effect on erythrocyte production or destruction. The concerted effect of rol variation on erythrocyte volume, water and cation content, cation cotransport, and 2,3-DPG levels is similar in many ways to the variation observed among individual humans for the same characteristics.(ABSTRACT TRUNCATED AT 250 WORDS)

We investigated the role of the kallikrein-kinin system in blood pressure control by developing transgenic mice overexpressing human tissue kallikrein. Two lines of transgenic mice carrying the human tissue kallikrein gene under the control of the mouse metallothionein metal-responsive promoter were established. Human tissue kallikrein was identified in pancreas, salivary gland, kidney, liver, and spleen of the transgenic mice by a specific radioimmunoassay for human tissue kallikrein. The immunoreactive human tissue kallikrein reached high levels in the circulation. The linear displacement curves for the transgenic product were parallel with the human tissue kallikrein standard curve, indicating their immunologic identity. The expression of human tissue kallikrein transcript in the transgenic mice was further confirmed by Northern blot analysis and by reverse transcription-polymerase chain reaction followed by Southern blot. Both lines of transgenic mice had significantly lowered blood pressure (86.4 +/- 13.5 mm Hg [mean +/- SD], n = 8 and 78.9 +/- 12.4 mm Hg, n = 8) compared with control mice (100.9 +/- 5.0 mm Hg, n = 8). Induction with zinc did not lower the blood pressure further despite elevated expression of the transgene. Administration of aprotinin, a potent tissue kallikrein inhibitor, restored the blood pressure of the transgenic mice but had no significant effect on control littermates. Our findings raise the possibility of tissue kallikrein being a powerful modulator of blood pressure and provide a new animal model for the study of blood pressure regulation.

Congenital ocular defects, including microphthalmia, anophthalmia and cataracts appear in a small proportion (5-15%) of the C57BL/6 inbred strain of laboratory mice. Previous work from this laboratory utilizing C57BL/6 in equilibrium with DBA/2 allophenic (mosaic) mice suggested that the underlying cause of these defects may be a deficiency in C57BL/6 lens development. In these C57BL/6 in equilibrium with DBA/2 allophenic animals, the major proportion of the lens cells were of the DBA/2 strain, in contrast to the other ocular and non-ocular tissues examined where no strain-specific competition was evident. Results presented here reinforce the notion that C57BL/6 lens cells are developmentally retarded. In allophenic combination with another strain, A/J, cells of the C57BL/6 strain were, again, markedly underrepresented in the mosaic lens, whereas no such genotypic specificity was evident in any of the other ocular and non-ocular tissues. Thus, the previous observation of a C57BL/6 deficiency in mosaic lens is more likely a consequence of low C57BL/6 competitiveness rather than some peculiar hyperactivity of the cells of the partner strains. Histological examination of C57BL/6, DBA/2 and A/J pure strain embryos indicated marked differences in the lens development of C57BL/6 embryos compared to the other two strains. For the first 3 days after lens invagination, the cross-sectional areas of DBA/2 lenses were two- to three-fold larger than those of C57BL/6 lenses at comparable stages of development, whereas no comparable difference was noted in the optic cup development in these two strains. This leads to a much higher lens/optic cup size ratio in DBA/2 embryos than C57BL/6 embryos. Scanning electron microscopy also revealed that lens invagination is much less extensive in the C57BL/6 embryos than the A/J embryos. The smaller C57BL/6 lens vesicle also remains deeper in the optic cup than those of the DBA/2 or A/J embryos of comparable developmental age. This is likely to result in the occasional complete envelopment of the entire lens vesicle by the optic cup and lead to arrested ocular development.

Myelin basic protein (MBP) is the second most abundant protein in CNS myelin. We have used transgenic mice to investigate the ability of the 5' flanking sequence of the mouse MBP gene to regulate the cell-type-specific- and temporal expression of a heterologous gene under its control. Transgenic mice were produced with a construct containing the bacterial chloramphenicol acetyltransferase (CAT) gene down-stream of the MBP 5' flanking sequence and CAT expression was monitored both enzymatically and histochemically. The results indicate that 1323 bp of 5' flanking sequence is sufficient to direct CAT expression specifically to the tissue and cell-type, in which MBP is normally synthesized. Additionally, this length of sequence also retains the ability to temporally regulate CAT levels in a manner analogous to endogenous MBP levels.

Transgenic mouse lines carrying the gene for rat alpha 1-acid glycoprotein (AGP) express the protein in the plasma at concentrations equal to or exceeding that of acute phase rats. Owing to the high basal level, these transgenic mice represent a unique experimental system for defining the largely unknown function of AGP. Since the carbohydrate moiety of AGP has been found to be changed during acute phase and the oligosaccharide structure to be important for immunomodulating activity of the protein, the rat AGP in transgenic mice was characterized by lectin-affinity immuno-electrophoresis. Unlike in the rat, the AGP in the transgenic mouse plasma consisted primarily of strongly concanavalin A-reactive forms. Acute phase mediated a several-fold increase in the total plasma concentration of AGP concomitant with a shift toward moderately concanavalin A-reactive forms. A similar change in concanavalin A-reactive forms was observed for the endogenous acute phase plasma protein haptoglobin. To define the role of inflammatory factors in AGP production, primary cultures of hepatocytes were prepared. In contrast to in vivo, the AGP recovered from tissue culture medium represented primarily the concanavalin A-non-reactive form. Treatment of the cells with recombinant human interleukin-1, interleukin-6 and dexamethasone stimulated the production of concanavalin A-reactive AGP forms. The data indicate that the glycosylation pattern of plasma-resident AGP is modulated by acute phase, but that the profile of AGP forms does not coincide with that secreted by hepatocytes in tissue culture. This finding demands an assessment of which of the possible glycosylated forms of AGP is functionally significant in vivo.

Cells derived from cell-cell fusion events were clonally isolated from primary methylcholanthrene (MCA)-induced tumors in allophenic mice. Compared with non-fused cells isolated from the same cultures, the fused cells had markedly greater experimental metastasizing (lung colonizing) activity, but only slightly greater tumorigenicity and the same cloning efficiency in soft agar. Cell-cell fusion may thus contribute to the generation of tumor heterogeneity that underlies the process of tumor progression.

alpha-1-Acid glycoprotein (AGP), which is produced in the mammalian liver and secreted into the blood-stream, is regulated by steroid hormones and by mediators of the acute phase response. In vitro transfection studies have shown that the response to glucocorticoids requires a cis-acting regulatory element, termed the glucocorticoid response element, that is located within 120 bp of the transcriptional start-site of the gene; induction by the acute phase mediators requires a different element, termed the distal regulatory element (DRE), that is located about 5 kb upstream of the start-site. To determine if these elements function in vivo, we have produced and characterized transgenic mice containing rat AGP gene constructs with and without the DRE. Five transgenic lines were produced from a 9.5-kb genomic AGP containing 4.7 kb of the 5' flanking region; this construct lacks the DRE. Another transgenic line was derived from a 10.7-kb clone that contains 5.3 kb of 5' flanking sequences including the DRE. All transgenic mice produced high levels of immunologically detectable rat AGP in the circulation, comparable to or in excess of that found in normal rats. There were correspondingly high concentrations of rat AGP transcripts in the liver. Transgene expression in all lines was induced in response to dexamethasone and during acute inflammation resulting from LPS treatment. The DRE-containing transgene underwent a greater induction in response to LPS than to dexamethasone; the transgene lacking the DRE responded similarly to both treatments. In cultured primary hepatocytes, the DRE-containing transgene was induced by the acute phase cytokines IL-1 and IL-6, and by dexamethasone, administered individually or in combination; the transgene lacking the DRE responded only to dexamethasone, and was not affected by the peptide hormones. Together, these results provide in vivo evidence supporting the notion that a minimum of two upstream sequences are responsible for the inflammatory induction of rat AGP. One element, which is located within the smaller 9.7-kb restriction fragment, is responsive to glucocorticoids and is likely to be the glucocorticoid response element located close to the transcriptional start site. The other element, the DRE, is located much further upstream and is responsible for imparting responsiveness to the acute phase cytokines.

The pluripotential hematopoietic stem cell is influenced by at least one gene that differs between DBA/2 and C3H/He, and C57BL/6 inbred mouse strains. This gene(s) manifests itself by its effect on susceptibility to killing of spleen colony forming cells (CFU-S) caused by hydroxyurea (HU). In strains DBA/2 and C3H/He 20% of the CFU-S population is normally in S phase whereas practically none from strain C57BL/6 are synthesizing DNA. On the other hand, in C57BL/6 in equilibrium DBA/2 allophenic mice we observed that the proportion of DBA/2 erythrocytes was higher than the proportion of DBA/2 lymphocytes; the fraction of platelets and neutrophils with the DBA/2 genotype fell between the values for erythrocytes and lymphocytes. Control experiments using mice congenic at the Fv-2 locus confirm that in both situations we are examining effects of a gene(s) other than Fv-2. For the effect on the S phase fraction of CFU-S, we refer to the gene(s) as Stk (stem cell kinetics). We suggest that the observed skewing in composition among the various mature blood cell types in C57BL/6 in equilibrium DBA/2 allophenic mice is caused by allelic variants of the Stk gene. Such variation would favor the formation of DBA/2 erythrocytes, platelets, and neutrophils over those of the C57BL/6 genotype.

Random shifts in blood cell genotypic composition are commonly observed in allophenic mice. This phenomenon was studied in 16 mosaic mice produced from very closely related strains, and no such changes were observed in the mosaic composition of erythrocytes, platelets, and lymphocytes over a period of 14 weeks. Furthermore, the mosaic distribution of a large group (66) of these mosaic mice was markedly biased in favor of those animals containing major contributions of both strains. This contrasts with what is normally found in collections of allophenic mice, in which the mosaic distribution curves are usually much flatter. While most allophenic mice have been produced from inbred strains with many genetic differences our results were obtained with congenic strains. This suggests that both properties, the unstable mosaic composition of blood cells and the flat mosaic distribution curves, are caused by specific genetic differences between cells of the two strains and are not inherent properties of allophenic mice. We propose that genetic differences cause these phenomena by inhibiting the mixing of cells of the two strains. Such might occur for example if, throughout development, cells of the same H-2 haplotype had greater affinity for each other than for ones of disparate H-2 haplotypes.

Male specific repeated sequences compose approximately 10% of the mouse Y chromosome. This was deduced from studies of genomic phage clones that contain male specific DNA, and three subcloned EcoRI fragments pBC10-0.6, pBC15-1.1, and pBA33-1.8. Southern analyses and in situ hybridization of metaphase chromosomes show these three sequences to be present in 100-200 copies on the Y chromosome and in female DNA as single copies. One of these, pBC10-0.6, was directly demonstrated to be on the X chromosome. All three exhibited male specific transcription. These sequences are found in two distinct organizations. Subclone pBC10-0.6 is uniformly found in a long HpaI repeat unit of 9.7 Kbp. Subclones pBC15-1.1 and pBA33-1.8 are distributed throughout several EcoR1 repeat families in which the male specific fragments are interspersed. These male specific fragments also differ among themselves with regard to the types and frequency of evolutionary changes.

One difficulty in the production of allophenic mice by aggregation of preimplantation embryos is that they frequently roll apart before the bonds between the blastomeres have had time to form. One solution to the problem, described here, is to pretreat one of the embryos with rabbit anti-mouse serum just prior to pushing them together. Blastocyst formation is unhampered by antibody treatment, and numerous allophenic mice have already been produced with this new procedure.

Allophenic mice composed of cells from a strain (DBA/2) susceptible to sound-induced seizures and cells from a strain (C57BL/6) resistant to them were produced by embryo aggregation techniques. Twenty-eight allophenic mice were tested for audiogenic seizure susceptibility. The results were compared with the genotypic composition of the coat melanocytes. For those animals with a predominance of one genotype or the other in the coat, their seizure phenotype was the same as that of the strain most represented in the coat. In contrast, those animals with major contributions of both genotypes in the coat demonstrated the entire spectrum of susceptibility phenotypes. Such results are likely to be a manifestation of a relatively small target tissue for the genes influencing the development of audiogenic seizure susceptibility.

Previous reports indicate that in laboratory strains of mice, males are distinct from females in possession of repetitive DNA, notably devoid of Eco RI and Hae III sites and rich in the simple tetranucleotides GATA/GACA. We report here that such sequences originated in an ancestor common to laboratory mice, Mus hortulanus, M. spretus, and possibly also M. cookii. Interestingly, other male-specific satellite sequences were detected in M. caroli, M. cookii, M. saxicola, and M. minutoides. This novel satellite is also likely to be composed of simple repetitious sequences, but does not contain GATA and GACA. Thus, the Y chromosome appears to contain a disproportionately large amount of simple repetitious DNA. An attractive explanation for these results is that long tandem arrays of simple repeated sequences are generated at high frequency throughout the genome and that they are retained for a longer time on the Y chromosome due to the absence of homologous pairing at meiosis.

Definitive evidence for the occurrence of cell fusion in tumorigenesis was sought in methylcholanthrene-induced sarcomas. This was approached by using allophenic mice generated from strains differing for electrophoretic variants of the ubiquitous, dimeric enzyme glucose phosphate isomerase, with fusion assessed by heterodimer formation. Eight-three carefully trimmed primary tumor samples (from 23 individual tumors in allophenic mice) were analyzed, as were 1,140 clones derived from them. In all primary tumor samples, zymograms exhibited one GPI homopolymeric band. Expression of a hybrid band (indicative of a fusion event) was not observed in these samples. However, 9 (0.8%) of the tumor clones demonstrated a distinct and reproducible hybrid band which was uniformly lost upon recloning. Our data suggest that cell fusion, although uncommon, occurs in the clonogenic cell fraction during primary MCA tumorigenesis and is followed rapidly by chromosome segregation.

We report here results suggesting that cells of the megakaryocytic lineage or uncommitted precursor cells may be targets for Friend-virus-induced proliferation, and that genetic differences (other than Fv-2) between strains C57BL/6 and DBA/2 affect the susceptibility of these cells to Friend virus. The evidence suggesting this was derived from experiments with C57BL/6 in equilibrium DBA/2 allophenic mice. Within the first few weeks following infection of these mice with the polycythemic NB-tropic strain of Friend virus (FV-P), we observed a rapid shift in the genotypic composition of both red cells and platelets in favor of those of the DBA/2 genotype. Infection with the anemia-inducing strain of Friend virus (FV-A) also resulted in preferential production of DBA/2 strain erythrocytes, but its effect on platelet kinetics was nil. The FV-P- and FV-A-induced change in red cell composition is consistent with the view that erythroid precursors are target cells for Friend virus and that viral infection preferentially stimulates proliferation of susceptible strain (DBA/2) erythroid precursors. As for the platelet shifts induced by FV-P (and not FV-A), we believe the changes in platelet mosaicism also could be caused by viral-induced proliferation of DBA/2 platelet precursors, or more primitive progenitors, over the C57BL/6 ones. Thus, these results implicate the existence of nonerythroid target cells for FV-P-induced proliferation, as well as the existence of genetic differences between strains C57BL/6 and DBA/2 that modulate the responsiveness of such cells to infection.

The study reported here is an examination of the organization and evolution of three Y chromosomal repeated sequences, designated pBC10-0.6, pBC15-1.1, and pBA33-1.8, in five closely related species of the genus Mus. The species distributions of major restriction fragment length polymorphisms produced with a panel of restriction enzymes is used to develop the phylogenetic relationships between the five species studied. However, the apparent degree of relatedness among these species varied a great deal with each of the three probes and was also highly dependent on the particular restriction enzyme used. The usefulness for phylogenetic studies of closely associated sequences varying in evolutionary stability is discussed.

Fv-2ss marrow cells are protected in vivo from Friend virus-induced erythroleukemia by admixture with a preponderance of Fv-2rr marrow cells. This was demonstrated both in allophenic (mosaic) mice and in bone marrow chimeras constructed from C57BL/6 strains differing at Fv-2 and an enzyme marker (glucose phosphate isomerase). The bone marrow chimeras were constructed by injection of marrow cells from Fv-2ss mice into unirradiated Fv-2rr mice. Bone marrow chimeras derived from this procedure produced 1%-2% donor (Fv-2ss) erythrocytes; this level of chimerism was maintained indefinitely. All the bone marrow chimeras as well as allophenic mice with less than 20% Fv-2ss red cells failed to develop any of the symptoms of Friend disease after infection with the polycythemic strain of Friend virus. The Fv-2rr-mediated protection of Fv-2ss marrow cells could be reversed by pretreatment of the two types of chimeras with either monoclonal or polyclonal antithymocyte antisera. Chimeras treated with either reagent and infected with Friend virus developed symptoms of Friend disease and experienced a shift in their erythrocyte mosaic composition favoring cells of the susceptible genotype. These results are consistent with the notion that a functioning immune system plays a role in the Fv-2rr-mediated protection of Fv-2ss Friend virus target cells. Furthermore, these studies establish conditions whereby a small population of sensitive strain cells in an overwhelming background of resistant strain cells can be selectively expanded. Such conditions could be useful in efforts to clone the Fv-2 gene.

Among normal mouse strains, natural genetic variation offers the potential to investigate the structure and function of cell membranes. One such polymorphism between C57BL/6J and DBA/2J is a difference in erythrocyte sensitivity to osmotic lysis. The genetic basis for erythrocyte osmotic fragility differences between mouse strains C57BL/6 and DBA/2 was examined through analyses of their serial backcross progeny, recombinant inbred (ri) strains (BXD), and congenic C57BL/6 strains with allelic differences at Hbb or Fv-2. The data indicate that the fragility difference between C57BL/6 and DBA/2 is the result of allelic differences at a minimum of two segregating loci. One of these might be linked to, but is not identical with the gene encoding the beta chain of hemoglobin (Hbb). Allelic differences at Fv-2, a gene known to control the proportion of erythroid precursors in the S phase, and at Hba, the structural locus of hemoglobin alpha chain also appear to exert no major influence on red cell osmotic fragility. Furthermore, the fact that red cells from one of the RI strans (BXD-31) are strikingly more resistant than those from the resistant parental strain DBA/2 leads to the conclusion that the degree of resistance/susceptibility for either strain is determined by the combined contributions of gene effects not all of which act in the same direction. We also found that red cells from strans C57BL/6 and DBA/2 differ in their uptake of 51Cr. This result suggests the possibility that red cell osmotic fragility differences may be due in part to differences in ion metabolism or membrane transport.

The immediate purpose of the experiment was to establish a means of introducing specific mitochondrially encoded mutant genes into mice. Mouse teratocarcinoma stem cells were used as vehicles for the cytoplasmic markers because of their known capacity for normal somatic and germinal differentiation after injection into blastocysts. The mutation of choice, chloramphenicol resistance (CAP(R)), was first produced in a melanoma cell line by mutagenesis and selection. The CAP(R) trait was then transferred from a resistant melanoma cell to a sensitive (CAP(S)) teratocarcinoma cell by fusing to the latter only the cytoplasmic portion of the CAP(R) donor. This indirect route demonstrated the cytoplasmic provenance of the mutation. Protein synthesis in mitochondria isolated from the cybrid, or cytoplasmic hybrid, cells was barely affected by chloramphenicol, in contrast to the inhibitory influence of the drug on mitochondria of the parent teratocarcinoma line. Cells of the cybrid clone resembled teratocarcinoma cells and retained their ability to form diverse tissues in solid tumors produced from subcutaneous grafts. Cells from the tumors were retransplanted and were tested periodically by culture in chloramphenicol; they were found to be stably CAP(R) even after 16 weeks in vivo in the absence of the selective agent. The CAP(R) cybrids were microinjected into blastocysts of another inbred strain and, after transfer to foster mothers, mosaic mice were obtained. They comprised both cybrid- and blastocyst-derived cells in various tissues, as indicated by strain-specific nuclear markers. These results demonstrate successful normal differentiation of the CAP(R) lineage in vivo. Teratocarcinoma cybrids thus offer a practical portal of entry of preselected mitochondrial genes into mice. This will ultimately permit in vivo investigation of maternally transmitted traits, of mitochondrial genetic influences in specialized cells, and of possible roles of cytoplasmic genes in clinical and disease states.

Infection of susceptible strain mice with the oncogenic Friend erythroleukemia virus initially results in fulminant erythroid hyperplasia. Several weeks later a frank erythroid leukemia develops. At the earliest stages of Friend disease there is extensive cell fusion involving erythroid cells but not platelets and granulocytes. Fusion was detected in experiments with allophenic (chimeric) mice whose component strains express electrophoretically distinct forms of the dimeric enzyme glucose phosphate isomerase (GPI). Infection of such animals with the polycythemic strain of Friend virus resulted in the rapid development of Friend disease. Concomitant with the appearance of early disease symptoms was the appearance in the red cells of the heterodimeric form of GPI, an unequivocal consequence of cell fusion. Platelet and granulocyte samples from the same infected animals failed to exhibit the hybrid GPI form. Furthermore, no hybrid dimer was evident in red cells from chimeric mice in which blood formation had been stimulated by phenylhydrazine treatment. These observations suggest that the occurrence of cell fusion early after infection by Friend virus is a significant aspect in the rapid development of neoplastic disease.

Infection of mice with Friend erythroleukemia virus initially causes massive proliferation of erythroid precursors accompanied by splenomegaly and reticulocytosis. Strains of mice differ among themselves in susceptibility to Friend virus and one of the major genes affecting the early response to viral infection is Fv-2. Allophenic mice compounded from a resistant strain C57BL/6 (Fv-2rr) and a susceptible one DBA/2 (Fv-2ss) were infected with the polycythemic strain of Friend virus to determine whether susceptibility/resistance was limited to cells of the respective genotypes or if there was an influence across the genotypic barriers. The manifestations of viral pathogenesis monitored were splenomegaly, reticulocytosis and leukocytosis. In addition, the proportion of red cells of the two genotypes in each animal was monitored before and after viral infection by analyses for strain specific electrophoretic variants of hemoglobin and glucose phosphate isomerase. The group of allophenic mice with 25% or more susceptible-strain red blood cells all developed symptoms of virus-induced disease and also revealed dramatic increases in the number of red cells of the susceptible-strain genotype. Thus, no evidence for protection of susceptible-strain cells by ones of the resistant strain could be observed and the disease developed primarily in susceptible strain cells. On the other hand infected animals with 15% or less DBA/2 red cells were severely retarded in the development of Friend disease. Under these circumstances susceptible strain target cells might fail to undergo viral induced replication as a result of direct protection by resistant strain cells. Alternatively, other more complex mechanisms might be involved such as protective anti-viral immune reactions.

The psychological well-being of forty-four breast cancer patients was assessed shortly before and after surgery, and at three and twelve months post-operatively. Regression analysis revealed type of treatment, control over treatment and concern for appearance as significant predictors of nature of subsequent mood state. Results are discussed with reference to the theoretical position that type of severe life event can predict the nature of ensuing psychological disorder.

Mutagenized stem cells of a cultured mouse teratocarcinoma cell line were selected for resistance to the purine base analog 6-thioguanine. Cells of a resistant clone were completely deficient in activity of the enzyme hypoxanthine phosphoribosyltransferase (HPRT, IMP:pyrophosphate phosphoribosyltransferase, EC 2.4.2.8), the same X-linked lesion as occurs in human Lesch-Nyhan disease. After microinjection into blastocysts of another genetic strain, the previously malignant cells successfully participated in normal embryogenesis and tumor-free, viable mosaic mice were obtained. Cells of tumor lineage were identified by strain markers in virtually all tissues of some individuals. Mature function of those cells was evident from their tissue-specific products (e.g., melanins, liver proteins). These mutagenized teratocarcinoma cells are therefore developmentally totipotent. Retention of the severe HPRT deficiency in the differentiated state was documented in extracts of mosaic tissues by depressed specific activity of the enzyme, and also by presence of unlabeled clones in autoradiographs of explanted cells incubated in [(3)H]hypoxanthine. Some mosaic individuals had mutant-strain cells in only one or a few tissues. Such animals may provide unique opportunities to identify the tissue sources of particular aspects of the complex disease syndrome. The tissue distribution of HPRT-deficient cells suggests that selection against them is particularly strong in blood of the mosaic mice, as is already known to be the case in human heterozygotes. This phenotypic parallelism supports the expectation that afflicted F(1) male mice that might be obtained from mutant germ cells can serve as a model of the human disease.

The mammalian Y chromosome is an isolated piece of genetic material that directs sexual determination and gametogenesis. Very little is understood about the mechanism whereby the Y chromosome carries out these functions. Also, there is a severe lack of genetic markers on this chromosome. In order to understand the structure and function of the Y chromosome at the level of its DNA sequences and to provide genetic markers, we are isolating clones of DNA whose sequences are found primarily in DNA from male mice. To this end, we have developed a procedure for the identification of such clones. Application of this screening procedure to a lambda library derived from mouse sperm DNA has yielded 12 distinct clones, part of whose sequences are present predominantly in male DNA. Besides this DNA, they also contain other sequences that are shared with female DNA. These clones are either derived from the Y chromosome or they represent autosomal sequences specifically amplified during male development.

The Tennessee Child Passenger Protection Act, mandating the use of child restraint devices for children younger than 4 years, took effect in 1978. In the years 1978 through 1983, eighty-one children younger than 4 years died in Tennessee traffic accidents; only two were in child restraint devices. During this period, as child restraint device use rose from 8% to more than 30%, the number of deaths among children younger than 4 years declined more than 50%. Analysis of supplemental accident reports filed in investigations of motor vehicle accidents involving children younger than 4 years during 1982 and 1983 showed that child restraint devices are highly effective in preventing death and in preventing or reducing injury. Children not in child restraint devices were 11 times more likely to die in an accident than children in child restraint devices. Children traveling in the arms of an adult were exposed to a risk of injury or death comparable to that of children left entirely unrestrained.

Previous work has indicated that inbred mouse strains C57BL/6 and DBA/2 produce red cells differing in their sensitivity to osmotic lysis and that the trait is under multigene control. A recombinant inbred strain (BXD-31), produced from C57BL/6 and DBA/2, has red cells manifesting resistance to osmotic lysis far greater than that of either progenitor. We demonstrate here that the fragility difference between BXD-31 and DBA/2 is the consequence of allelic variation at a single autosomal locus, termed rol. The resistance allele (rol') is almost completely recessive to the sensitive one (rols). Results of bone marrow chimera analyses indicate that (1) the mode of rol gene action is by a direct influence on the properties of the red cells rather than an indirect influence on their extracellular milieu, and (2) rol does not affect erythrocyte production and turnover. The fragility difference caused by rol variation is likely to involve the erythrocyte membrane or underlying cytoskeleton, since various red cell properties sensitive to ion metabolism differences are unaffected by the gene.

Cholesterol is a major risk factor for coronary heart disease. Of 117 employees seen consecutively in the Tennessee State Employee Health Service voluntary screening program, 86 (74%) had cholesterol levels above the reference range reported by a commercial clinical laboratory. This was three times greater than the calculated expected number of 29 (25%, 17 at moderate risk, 12 at high risk for coronary disease). Age and sex adjustment using Lipid Research Clinic guidelines reduced the number with elevated cholesterol to 55 (47%, 24 at moderate risk, 31 at high risk). Split sample cholesterol assays run independently by the commercial laboratory and a university laboratory showed excellent correlation (r = 0.99, commercial laboratory = 1.0 (university laboratory) + 10.8), but a systematic difference of 12.4 mg/dL (SD = 6.4 mg/dL, paired-t = 9.63, p less than 0.00001) between the two laboratories. Further adjustment for this difference reduced the number with elevated cholesterol to 41 (36%, 26 at moderate risk, 15 at high risk). This experience illustrates how small systematic laboratory errors in cholesterol determination can greatly exaggerate the number of persons reported to have clinically important cholesterol elevations. Clinical laboratories should report age and sex adjusted cholesterol reference ranges and provide clients periodic quality assurance reports that their measurements of cholesterol levels are accurate.

Motor vehicle accidents are the leading cause of death in children. In 1977, Tennessee adopted the nation's first law requiring the use of child restraint devices (CRDs), but despite extensive promotional efforts, a majority of young children still travel unrestrained. We surveyed all acute-care hospitals in Tennessee to determine their policies regarding CRDs. Of 109 hospitals with obstetric services, 28 (26%) had a policy calling for discharged newborns to be transported in CRDs; only seven (5%) of 128 pediatric services had such a policy. It is time for hospitals and professional organizations to adopt policies to ensure that the parents of every child discharged from an obstetric or pediatric unit are educated concerning CRD use laws and are able to comply with them. Pediatricians should consider incorporating "discharge in child restraint device" into their routine discharge orders.

Teratocarcinoma (TCC) stem cells provide unique prospects for the introduction of specific genes into mice, by virtue of their dual capacity for propagation in vitro and for normal differentiation in embryos. In this study, we have demonstrated that foreign genes amenable to selection in culture can be transferred into the stem cells and expressed. These cells maintain expression of the gene for long periods during differentiation in tumors in vivo in the absence of selective pressure. The cells also integrate an unlinked nonselectable gene at high frequency. Addition of the cloned herpes simplex virus (HSV) thymidine kinase (tk; ATP:thymidine 5'-phosphotransferase, EC 2.7.1.21) gene to cultures of tk(-)TCC cells yielded tk(+) colonies at a frequency of one colony per 4 mug of plasmid DNA. This transformation efficiency, although appreciably lower than for mouse L tk(-) cells, permits the isolation of many transformants. The HSV provenance of the transformed phenotype was verified by the characteristic electrophoretic mobility of the tk protein and by neutralization of the tk activity with specific antiserum. Moreover, blot hybridization tests revealed at least one intact copy of the viral tk gene integrated into the DNA of transformed cells. When injected into syngeneic mice, the cells formed solid tumors with various differentiating tissues. From blot hybridization comparisons with their cell lines of origin, seven of nine tumors examined had maintained the HSV tk gene without significant loss or rearrangement. Viral tk enzyme activity could also be demonstrated in at least some of the tumors. Cotransfer of the cloned human beta-globin gene along with the unlinked HSV tk gene was successful in 2 of 10 tk(+) transformants. Thus, defined genes can be stably introduced into TCC cells in culture and maintained in vivo in a form in which they are transcribed and translated to produce a functional protein.

Studies on the genetic regulation of growth are confounded by the multigenic basis of growth. There is a need to isolate simplified genetic systems for the study of growth. We compared two closely related mouse strains and their F1 hybrids with regard to birth and weaning weights. The strains we used were C57BL/6 (B6) and a congenic derivative of B6 (HW54) that contains a short segment of BALB/c chromosome 7 spanning the H-24 and Gpi-1 loci. Despite the genetic similarity of these strains, they differed significantly in both birth and weaning weights. At birth, B6 pups were on average as much as 6.6% heavier than were pups from HW54. By the time of weaning, this trend was reversed; HW54 pups were as much as 13.8% heavier than were B6 pups. (B6 x HW54)F1 hybrids were intermediate between the parental strains in birth weight but were identical to B6 animals at weaning. An analysis of the F2 generation suggested that postnatal growth differences between B6 and HW54 are probably dependent on the maternal genotype. These strain-specific growth rates result from polymorphism at a restricted portion of the genome and represent a highly simplified system for the study of the genetics of growth.

C57BL/6 mice exhibit impaired lens development as evidenced by competition studies with allophenic (chimeric) mice. Strain-specific differences in electrophoretic forms of glucose phosphate isomerase were exploited to determine the strain composition of various ocular and nonocular tissues in allophenic mice constructed from C57BL/6 and DBA/2 strains. As expected, there was variation from one animal to the next in overall mosaic composition, as well as variation among most ocular and nonocular tissues within individual animals. In contrast, when individual lenses from the same animals were assayed there was a marked and consistent deficiency of the C57BL/6 component. Control experiments indicated that these results are not the consequence of peculiarities of marker enzyme expression in the lens. Since lens formation occurs early in embryogenesis any anomaly at this stage could influence the normal induction of later ocular structures. We believe that the impaired lens-forming capacity of C57BL/6 mice may be the underlying cause of the 5-10% frequency of overt ocular defects in this strain.

The Fv-2 genotype of erythroid progenitors directly determines whether they will undergo viral-induced transformation. This conclusion was reached from studies of allophenic mice compounded from congenic C57BL/6 strains differing at Fv-2 and an enzyme marker (GPI). Infection of these Fv-2ss in equilibrium Fv-2rr mosaic animals with the polycythemic strain of Friend virus results in the development of Friend disease. Concomitant with disease symptoms is a shift in the mosaic composition of the erythrocytes in favor of those of the susceptible strain. The observed viral-induced shift in the erythrocyte composition is paralleled by a similar change in the mosaic composition of the CFU-E pool but not the primitive (d8) BFU-E pool. Thus, with regard to this particular Fv-2 phenotype (susceptibility to FV-P-induced cellular hyperplasia), Fv-2 manifests itself specifically in the erythroid lineage, either in mature (d3) BFU-E or CFU-E.