Correction of a genetically caused enzyme defect by somatic cell hybridization.

Liver cells obtained from newborn mice homozygous for any one of several overlapping deletions in chromosome 7 fail to express a number of liver-specific differentiated traits. Among these is the activity of the membrane-bound liver-specific enzyme glucose-6-phosphatase (Glc-6-Pase; D-glucose-6-phosphate phosphohydrolase, EC 3.1.3.9). Previous studies have led to the suggestion that the region of the genome covered by these deletions includes genes that normally regulate the expression of structural genes encoding liver-specific enzymes and proteins mapping elsewhere in the genome. To find out whether the deficiency of Glc-6-Pase may be caused by the deletion of the relevant structural gene, mouse liver cells homozygous for the deletion c14CoS were hybridized with 2S Faza rat hepatoma cells, and the hybrid cell cultures were analyzed for mouse and rat Glc-6-Pase activity. Hybrids showed expression of mouse Glc-6-Pase activity, proving that the structural gene for this enzyme is not included in the deletion c14CoS in chromosome 7. In the hybrid cells the rat hepatoma genome apparently contributes a factor that activates the structural gene of the mouse and corrects its failure of expression, which most likely resulted from the deletion of an essential regulatory or processing gene. By using as a marker glucose-6-phosphate isomerase (Glc-6-PIase; glucosephosphate isomerase, D-glucose-6-phosphate ketolisomerase, EC 5.3.1.9), known to map on chromosome 7, this entire chromosome could be excluded as a possible carrier of the Glc-6-Pase structural gene. In addition, the structural genes for Glc-6-Pase and for tyrosine aminotransferase (TyrATase; L-tyrosine:2-oxoglutarate aminotransferase, EC 2.6.1.5), another enzyme deficient in lethal deletion homozygotes, were shown to map on two different chromosomes. Together with our previous studies of TyrATase gene regulation, the present experiments suggest that the region of the mouse genome defined by the deletions includes one or more genes regulating the expression of several structural genes that map on different chromosomes and that encode liver-cell-type specific traits.

Protein synthesis with membrane-bound polysomes and albumin messenger RNA from livers of mutant mice.

Investigation of deficiencies in serum protein synthesis resulting from deletion-mutations at the albino locus in mice was continued using in vitro conditions. Previous work showed that although total protein synthesis was only slightly lower in livers from albinos, newly synthesized protein appearing in plasma was 22% of that in controls. It was thought that the disorganized endoplasmic reticulum and Golgi apparatus, characteristic for the liver (and kidney) of these mutants, might be responsible for the observed deficiencies. In the present study membrane-bound polysomes isolated from the livers of newborn albinos were 55% (c3H/c3H strain) and 62% (c14CoS strain) as efficient as those from normal littermates in incorporating radioactive leucine into protein in a cell-free system. These differences could not be eliminated by the addition of excess liver mRNA, exogenous soluble factors or by the exchange of cell sap between albino and control polysomes. In an earlier study albino liver slices synthesized only 13% (or 17% per mg of total protein synthesized) as much albumin as controls. We have now found that the level of albumin poly(A)+-RNA isolated from albino livers assayed with a reticulocyte lysate, was almost as high (85%) as in controls. It was concluded that the very low level of albumin in albino livers did not result from a deficiency of albumin mRNA. Whether the rate-limiting step in synthesis of albumin in mutant livers is at the level of translation or processing for secretion requires further investigation.

Complementation of gene deletions by cell hybridization.

Overlapping deletions in chromosome 7 of the mouse are responsible for activity deficiencies of various liver-specific enzymes, including tyrosine aminotransferase (TAT). In an effort to elucidate the nature and type of action of the deleted genes, somatic cell hybridization experiments were carried out. Enzyme-deficient liver cells of homozygous mutant mice or normal liver cells of control newborn mice were hybridized with 2S Faza rat hepatoma cells and the hybrid cell colonies were analyzed for TAT activity, The results show the presence of inducible mouse TAT activity in mutant-2S Faza hybrid cells, thereby excluding the possibility that the structural gene for TAT is included in the gene sequences deleted in the mutants. Furthermore, determinations of mouse glucose-6-phosphate isomerase 1 as a marker eliminate chromosome 7 as the possible carrier of the TAT structural gene, which therefore appears to map on a different chromosome. The deletions interfering with normal enzyme activities apparently include genes other than the respective structural genes, namely those with essential functions in controlling the expression of the differentiated state of the liver cell.

Deficiency in plasma protein synthesis caused by x-ray-incuded lethal albino alleles in mouse.

Plasma protein synthesis was studied in mice bearing x-ray induced lethal mutations at the albino locus. Newborn albino mutants showed a decrease in each of the three principal plasma proteins, albumin, alpha-fetoprotein, and transferrin, when compared with colored littermate controls. Incorporation of [14C]leucine into plasma proteins of the newborn albinos 30 min after injection was only 1/5 that of the controls, but incorporation into total liver protein was only slightly diminished. Incorporation of [14C]leucine into an albumin fraction obtained by immunoprecipitation from livers incubated in vitro in an amino acid mixture was also strongly diminished. Thus, the liver of 18-day-old albino fetuses incorporated into this fraction 1/3 and that of newborn albinos 1/8 as much as the controls, but in both cases the incorporation into total liver protein was only 25% less than in the respective controls. These results indicate that the rather severe structural abnormalities observed in the mutants in the endoplasmic reticulum and the Golgi apparatus are not associated with a general deficiency of hepatic protein synthesis. Instead the data from this and previous work show that the progressive deficiency from fetal life to birth involves certain specific proteins represented by several perinatally developing enzymes and by plasma proteins. It is suggested that the mutational effects observed in these mice are due to deletions involving regulatory rather than structural genes at or near the albino locus.

The effect of p-hydroxymercuribenzoate and congeners on microsomal glucose-6-phosphatase.

Iodoacetamide, N-ethylmaleimide, p-hydroxymercuribenzoate (p-MB) and HgCl2 were tested as inhibitors of microsomal glucose-6-phosphatase. Iodoacetamide had no effect at 2 mM. N-ethylmaleimide inhibited only crude, but not purified microsomal preparations (M2) or crude microsomes exposed to deoxycholate. 14C-labelled N-ethylmaleimide was not bound by the M2 protein fraction. p-MB inhibited all types of preparations and the inhibition was not counteracted by detergent. A more detailed study was carried out with the purified M2 fraction (specific activity: 2-4 mumoles Pi/min/mg protein). Glucose-6-phosphate hydrolysis was inhibited 50% by 5 X 10(-5) M p-MB. The inhibition was completely reversible by dithiothreitol except when the enzyme was pre-incubated with p-MB in the absence of substrate. Then p-MB accelerated the temperature-dependent inactivation of glucose-6-phosphatase. Binding studies showed that around 3 mumoles 14C-p-MB were incorporated into 100 mg M2 protein regardless of the concentration of mercurial in the incubation mixture. That is, over a 25 fold range of p-MB concentration, causing up to 80% inhibition of enzyme activity, no difference was seen in the amount of labelled p-MB which was irreversibly bound to M2 protein. Kinetically p-MB behaved like a reversible inhibitor and this was confirmed by dilution experiments. Several compounds, including some amino acids, antagonized the inhibition by p-MB. The order of effectiveness was EDTA greater than barbital greater than tryptophan greater than histidine greater than lysine greater than other amino acids. Glycine, Tris and urea were ineffective competitors of p-MB inhibition. Double reciprocal plots showed that the Km for glucose-6-phosphate was increased and the Vmax reduced in the presence of p-MB. HgCl2 was a more effective inhibitor than p-MB with a Ki of 6 X 10(-6) M. We conclude that a reaction of p-MB with M2 sulfhydryls does not play a part in the inhibition of enzyme activity. It is suggested that p-MB may interact with one or more amino acid side chains in such a way that enzyme conformation is altered.

Relipidation of phospholipid-depleted microsomal particles with high glucose 6-phosphatase activity.

Microsomal particles enriched up to 20-fold in glucose 6-phosphatase activity (as compared to crude microsomal fractions) were prepared from livers of phenobarbital-treated, normal, and diabetic rats by a method involving sucrose-density gradient centrifugation through a layer containing deoxycholate, followed by flotation of the delipidated particles in a phospholipid-detergent mixture. The flotation with phospholipid resulted in the extraction of additional protein, in a corresponding increase in specific activity, and in relipidation of the particles to their original phospholipid content. Detergents alone did not extract any additional protein. Relipidation caused a change in the properties of the microsomal membrane, as indicated by a 4-fold decrease in the K(m) for glucose 6-phosphate and mannose 6-phosphate. The maximal rate was not affected. The crude homogenate of diabetic rat liver contained more latent enzyme than that of normal rats. The diabetic rats also yielded purified microsomal particles of specific glucose 6-phosphatase activity twice that of normal and five times that of phenobarbital-treated rats, indicating that some regulatory mechanism exists for the incorporation of different amounts of the enzyme into the endoplasmic reticulum.

Complementation studies of lethal alleles in the mouse causing deficiencies of glucose-6-phosphatase, tyrosine aminotransferase, and serine dehydratase.

The pleiotropic effects of six radiation-induced lethal alleles at the albino locus in the mouse include deficiencies of various enzymes and of serum proteins. Closely correlated with the biochemical deficiencies are severe ultrastructural abnormalities of endoplasmic reticulum membranes and the Golgi apparatus in liver and kidney cells. Complementation studies led to the distinction of four different complementation groups among the six lethal alleles. The results are compatible with the interpretation that the six lethal alleles represent overlapping deletions of various sizes at and around the albino locus in the mouse, in a region concerned with the control of biochemical and structural differentiation.