Cloning and expression of the Campylobacter jejuni glyA gene in Escherichia coli.

Genetic studies of Campylobacter jejuni are greatly hampered by the lack of genetic markers and an established classical gene transfer mechanism between strains of this species. To facilitate future genetic studies and to provide a recombinant DNA approach for analyzing genes of C. jejuni, we constructed an extensive genomic library of a pathogenic C. jejuni strain TGH9011 (serotype 0:3) using pBR322. We report the isolation of a number of recombinant plasmids containing the complete structural gene of glyA, that encodes serine hydroxymethyltransferase (SHMT) of C. jejuni. Escherichia coli cells containing this multicopy recombinant plasmid with the glyA gene produce high levels of SHMT. The SHMT-encoding fragment was identified by subcloning and functional complementation. The expression of the C. jejuni glyA gene was probably via transcription initiated from its own promoter.

Catalytic facilitation by diffusion of adsorbed substrate on membrane surface.

Membrane-alkaline phosphatase shows greater velocity of reaction than solubilized enzyme at low substrate concentration, whereas at saturation-concentration the opposite is true. The catalytic rate enhancement with the membrane-enzyme, when substrate availability is limiting, is attributed to non-specific adsorption of substrate to the membrane followed by its surface-diffusion to the active site resulting in an enhanced collision rate for the substrate with the enzyme. Experimental evidence for the adsorption-diffusion is provided by the dynamic quenching of 1-anilino-naphthalene-8-sulphonate, a membrane-bound probe's fluorescence by the substrate, 4-nitrophenyl-phosphate.

Desialylated alkaline phosphatase: activation by 4-nitrophenol.

Mouse ileal alkaline phosphatase is a sialyl enzyme (12-14 moles per mole of enzyme). When partially desialylated by treatment with neuraminidase, the enzyme loses most of its activity, associated with reduced apparent Vmax and Km. Part of that loss, however, is recovered as the product 4-nitrophenol's concentration builds up in the cuvette. Experimental results are presented to demonstrate that the activation is due to the binding of 4-nitrophenol as a ligand by the partially desialylated enzyme and that both the loss of activity by sialic acid removal and activation by ligand-binding are correlated with changes in protein conformation.

Isoantibodies to epididymal sperm antigens in BALB/c strain of mice.

Female mice of the BALB/c isogenic strain were immunized with epididymal sperm of the male mice. Four antigenic preparations were used: whole sperm (WS), homogenized sperm (HS), protein fraction (PF) and lipid fraction (LF). Antisera were assayed for antibody activity by sperm immobilization and indirect immunofluorescence tests. The differential results between the WS and HS antisera obtained in the two tests suggest that sperm membrane surface antigen(s) may be distinct from intracellular antigen(s). PF-antiserum showed lower but significant activity with the sperm immobilization test but weak activity with the indirect immunofluorescence test. The lipid fractions showed extremely weak activities with both tests. Indirect immunofluorescent staining was carried out on 10 micrometer sections of unfertilized and fertilized eggs and 2-cell embryos. Unfertilized eggs showed no antibody binding and fertilized eggs showed patchy fluorescence of the plasma membrane and cytoplasm. In the 2-cell embryo the fluorescence was diffuse, suggesting spreading of the sperm antigens. Comparative electrophoresis of aliquots of HS antiserum with and without incubation with washed sperm showed a clear reduction, due to adsorption by sperm, of IgG class of antibodies, on the basis of molecular weight. The differences in antigenicity between the four sperm preparations are discussed in relation to (i) possible localization of the antigens in the spermatozoa and (ii) the conformational differences of the antigens manifested by organic solvent denaturation.

Glycoproteins of the brush border membrane of mouse duodenum.

Qualitative and quantitative chemical analyses have been made of the carbohydrate and protein moieties of the glycoproteins in purified mouse duodenal brush border membranes. Solubilized membrane protein was electrophoresed on S.D.S.--polyacrylamide gels and glycoproteins were subsequently identified by periodate-Schiff staining. A parallel experiment showed that these glycoprotein bands also incorporated 14C counts from intraluminally-administered 14C-glucosamine. Positively-staining bands were collected from several identically run gels and subjected to digestion with pronase. The sugar and amino acid compositions of the digested glycoproteins were determined. Five of the membrane proteins appeared to be complexed with sugars, and further analyses showed that there were differences between the glycoproteins with respect to sialic acid, uronic acid, amino acids, amino sugar and monosaccharide contents.

Studies on the brush border membrane on mouse duodenum: lipids.

Lipids were extracted from purified mouse duodenal brush border membranes. Lipid: protein ratios in different membrane preparations varied from 0.58 to 0.68. Both the chloroform and non-chloroform phases were quantitatively analysed for lipids. Chloroform extracts were composed of cholesterol, triglycerides and phospholipids. The major neutral lipid was cholesterol. Rechromatrography of the phospholipid spot showed sphingomyelin (7.9%), phosphatides of ethanolamine (61.7%), inositol (14.2%) and serine (16.2%). The average molar ratio of cholesterol to phospholipid was 1.39. Lipids in the non-chloroform phase were all glycosylated, being cerebrosides (69.3%), cerebroside sulphates (28.8%) and galgliosides (1.9%). Overal membrane lipid composition was neutral lipid 24%, phospholipid 33%, and glycolipid 43%.

Glucose-binding membrane proteins.

To identify glucose-binding proteins amongst the polypeptides of the mouse duodenal brush border membrane, three types of experiments are reported. The first involved the introduction of labelled glucose and its analogue phlorizin into the lumen of separate groups of ligatured duodenal segments. Several proteins were shown to have bound both labelled species in situ by liquid scintillation counting of slices from polyacrylamide gels on which solubilised membrane protein had been electrophoretically separated. The second type of experiment was designed to determine the competitive nature of the binding of both labelled and cold phlorizin to proteins which had already bound glucose. Only three bands could competitively bind phlorizin. Finally, gels on which solubilised protein from in situ glucose-binding experiments had been run were placed in solutions containing labelled phlorizin. The binding of phlorizin to proteins in the same three bands as above suggested a confirmation of the conclusion that there were three membrane protein types which appeared to be involved in phlorizin-sensitive glucose-binding.

beta-Galactosidases in mouse duodenal mucosa.

beta-Galactosidase activity, in fetal mice, first appears at 16 days of gestation and has a pH optimum of 4. In postnatal development the enzyme activity of cell homogenates tends to show bimodal pH at 4 and 5.6. There are two molecular forms of the enzyme, separable both by molecular-sieve chromatography and electrophoresis. One of the molecular forms of the enzyme is active over a wider range of pH (3.2-6.2) and has half as much activity at 5.6 as it does at 4. This isoenzyme is continuously present in both fetal and postnatal stages. The second isoenzyme first appears at birth, is active over a narrower range of pH (4.6-6.2) and inactive at PH 4. The bimodal pH optima observed in postnatal stages in the cell homogenates, appears to be due to the combined activity of the two molecular forms. In isolated brush border membranes, isoenzyme 2 is the only one present. The other organelles (mitochondria, microsomes, lysosomes, nuclei and cytoplasm) have variable proportions of both isoenzymes, as indicated by the activity ratio at pH 4/5.6.

Studies on the brush border membrane of mouse duodenum. III. Iodination of membrane proteins.

Microvillus membranes were iodinated from luminally administered lactoperoxidase, H2O2 and 125I before and after neuraminidase treatment. Membranes were isolated, solubilized in sodium dodecyl sulphate buffer and electrophoresed on gels. Gels were stained for protein, and then sliced for liquid scintillation counting. When membranes were not treated with neuraminidase, the nonpermeating iodination probe attached only to a band containing protein of 150,000 daltons approximate molecular weight. This size class of protein may reside on the luminal side of the brush border membrane as opposed to the serosol side. Qualifications of this statement are discussed with reference to the location of tyrosyl residues and to the possibilities of masking molecules. Membranes treated from the luminal side with neuraminidase to remove possibly masking carbohydrate and then iodinated, appeared to contain an additional protein of estimated molecular weight 220,000 daltons which was accessible to 125I. Thus, a 220,000 dalton protein may also be on the luminal side of the membrane. An explanation is attempted for the discrepancy between the very few proteins labelled and the many proteins involved in terminal digestion and transport which would all be expected to be available to luminally administered iodination probe. Membranes were isolated, exposed to iodination and then solubilized for electrophoresis. Nearly all proteins were labelled, which indicated that there is an asymmetric distribution of proteins in the plane of the membrane. The two smallest molecular weight polypeptides which were not iodinated were proposed to be so disposed in the membrane that they were inaccessible to the probe, from either side.

Studies on the brush border membrane of mouse duodenum. II. Membrane protein metabolism.

Mouse duodenal microvillus membrane protein metabolism was measured using radioactive labelling techniques. Labelled amino acids were introduced into the lumen of ligatured duodena. Following exposure to label, brush border membranes were isolated and analyzed. Experiments measuring the specific activity of protein labelled with a single amino acid revealed that total membrane protein appeared to turnover in about 14 hr. Protein in the mucosal homogenate had a faster turnover rate. Turnover rates of individual proteins were measured with single and dual isotope experiments. Membrane protein was solubilized with sodium dodecyl sulphate (SDS) buffer. Single isotope experiments showed that all polypeptides separated on SDS-gels were maximally labelled at 6 hr after injection. Bands did not incorporate label linearly. Rates of loss (degradation) of label from membrane proteins in the seventeen bands appeared to be related to the estimated molecular size of the proteins. Rates were highest for larger polypeptides. A double isotope technique, in which proteins were allowed to incorporate the same amino acid in two isotopic forms, delivered with a set time interval intervening, revealed that the ratios of the second label to the first in the SDS-separated polypeptides were highest for larger proteins and lowest for smaller polypeptides. Certain assumptions were outlined and the ratios taken as measures of turnover of proteins. Loss of label due to cell sloughing is discussed. A mixture of labelled amino acids (excluding leucine) was used to show that differences in leucine contents of different proteins was not an explanation for the variation in level of leucine radioactivity in different bands. For specific activity measurements throughout, protein in gels was quantitated with reference to the uptake of Coomassie stain. The use of this stain was validated by the finding that, at low protein concentration, the amount of stain taken up was proportional to the amount of bovine serum albumin or membrane protein loaded.

Studies on the brush border membrane of the mouse duodenum. I. Membrane isolation and analysis of protein components.

Brush border membranes have been isolated from villus epithelial cells of the adult Swiss mouse duodenum. Preparations of these membranes are not contaminated by other organelles as judged from electron-micrographs of sectioned pellets of brush borders. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of proteins from brush borders solubilized in Tris-sodium dodecyl sulfate buffer reveals a reproducible Coomassie Brilliant Blue pattern of 17 bands. By comparing the brush border protein band positions with those of standard proteins run concurrently on sodium dodecyl sulfate-polyacrylamide gel slabs it is estimated that the 17 brush border proteins and subunits have molecular weights ranging from over 250,000 to around 16,000. Periodate-fuchsin sulfite staining shows that the five more slowly migrating, high molecular weight proteins are glycoproteins. The two proteins of smallest molecular size react positively with Oil Red O but have very small amounts of lipophilic amino acid residues, which indicates that the lipid extractable from the gels in these areas is a contaminant and is not bound to the proteins.

Molecular heterogeneity of mouse duodenal alkaline phosphatase. Association of lipids and peptides.

Polyacrylamide-gel electrophoresis and Bio-Gel P-300 molecular-sieve chromatography of mouse duodenal alkaline phosphatase demonstrates its molecular heterogeneity, which, in a kinetic sense, is manifest also in the differential relative velocities of the heterogeneous forms of the enzyme with two substrates, phenylphosphate and beta-glycerophosphate. Different treatments that eliminate most of the electrophoretic and chromatographic variability of the enzyme also decrease the velocities with both substrates so that the molar ratio of hydrolysis of one substrate relative to the other is also altered to a low but stable value. Concomitant with these changes, lipids and peptides are dissociated from the enzyme. The lipids are tentatively identified as a sterol and phospholipids. The peptides have an average composition of four to six amino acids and appear to be strongly electropositive. The conditions of dissociation suggest that their binding to the enzyme is non-covalent and predominantly based on hydrophobic and ionic bonding. The concept of lipid and peptide association would suggest prima facie differential molecular weights as a factor in the observed electrophoretic and chromatographic heterogeneity. However, the molecular forms of the enzyme with differences in elution volume equivalent to more than one-half the void volume of the Bio-Gel P-300 column, or even enzyme fractions dissociated from the lipids and peptides compared with undissociated portions, do not show any differences in sedimentation on sucrose-density-gradient centrifugation. This may be because the alterations in molecular weight owing to binding of small molecules are too small to be detected by this method. Alternatively, since lipids are involved, the binding may alter the partial specific volume in such a way that the buoyant density is not significantly altered.

Intestinal alkaline phosphatase: regulation by a strain-specific factor in mouse milk.

Two inbred strains of mice show a threefold difference in duodenal phosphatase activity at 11 days of age. When half-litters of the two strains are interchanged between the two mothers on the day of birth, enzyme activity in young of the low-activity strain is unaffected at 11 days by the source of milk, but is significantly reduced in high-activity young nursed by a low-activity mother.