Developmental and tissue-specific expression of human CD4 in transgenic rabbits.

A major obstacle to understanding AIDS is the lack of a suitable small animal model for studying HIV-1 infection and the subsequent development of AIDS, and for testing diagnostic, therapeutic, and preventive modalities. Our goal is to produce a rabbit model for the study of AIDS. Here we report on the generation of transgenic rabbits that express the human CD4 (hCD4) gene. The transgene, which contains the coding region for hCD4 and approximately 23 kb of sequence upstream of the translation start site, was used previously to direct hcD4 expression on the surface of CD4+ T cells of transgenic mice (Gillespie et al., 1993: Mol Cell Biol 13:2952-2958). The hCD4 transgene was detected in five males and two females derived from the microinjection in five males and two females derived from the microinjection of 271 rabbit embryos. Both hCD4 RNA and protein were expressed in peripheral blood lymphocytes (PBLs) from all five males but neither of the females. Human CD4 was expressed on PBLs from F1 offspring of all founder males. T-cell subset analysis revealed that hCD4 expression was restricted to rabbit CD4 (rCD4) expressing lymphocytes; mature rCD4- rCD8+ lymphocytes did not express hCD4. In preliminary studies, PBLs from hCD4 transgenic rabbits produced greater amounts of HIV-1 p24 core protein following HIV-1 infection in vitro than HIV-1 p24 antigen in nontransgenic rabbit infected cultures. These results extend to rabbits our previous observation that this transgene contains the sequence elements required for high-level expression in the appropriate cells of transgenic mice. Furthermore, these and previous studies demonstrating that expression of hCD4 protein enhances HIV-1 infection of rabbit T cells in vitro, coupled with reports that normal, nontransgenic rabbits are susceptible to HIV-1 infection, suggests that the hCD4 transgenic rabbits described herein will have an increased susceptibility to HIV-1 infection. In vivo HIV-1 infection studies with these rabbits are under way.

Cloning and characterization of a vertebrate cellular myosin regulatory light chain complementary DNA.

We have isolated two series of complementary DNAs (cDNAs) from a chicken gizzard cDNA library encoding two isoforms of phosphorylatable myosin regulatory light chain (RLC). One of the cDNAs encodes a previously isolated smooth muscle myosin RLC (also referred to as LC20-A); the other encodes a protein that shares 92% homology with the LC20-A isoform. The phosphorylatable threonine and serine residues at positions 18 and 19 of the two myosin RLC sequences are conserved. The two cDNAs are 81% homologous at the nucleotide level over the coding region; the 5' and 3' untranslated regions are divergent. Most of the DNA nonhomology in the coding region does not affect the protein sequence, indicating strong evolutionary conservation pressure to maintain the myosin RLC structure. Northern blot analysis using 3' untranslated region probes reveals restrictive tissue specific expression of one myosin RLC isoform (LC20-A) in smooth muscle tissue and not in other tissues examined. In contrast, the novel myosin RLC isoform messenger RNA (mRNA) is uniformly expressed in all smooth and nonmuscle tissues examined and is designated as cellular myosin RLC for this reason. Our results indicate that cellular and smooth muscle myosin RLC isoforms are distinct and are encoded by separate genes. This report describes the cloning of a novel vertebrate cellular myosin RLC mRNA that differs from previously characterized smooth muscle RLC isoform mRNAs in both primary sequence and expression pattern.

Characterization and differential expression of human vascular smooth muscle myosin light chain 2 isoform in nonmuscle cells.

The 20-kDa regulatory myosin light chain (MLC), also known as MLC-2, plays an important role in the regulation of both smooth muscle and nonmuscle cell contractile activity. Phosphorylation of MLC-2 by the enzyme MLC kinase increases the actin-activated myosin ATPase activity and thereby regulates the contractile activity. We have isolated and characterized an MLC-2 cDNA corresponding to the human vascular smooth muscle MLC-2 isoform from a cDNA library derived from umbilical artery RNA. The translation of the in vitro synthesized mRNA, corresponding to the cDNA insert, in a rabbit reticulocyte lysate results in the synthesis of a 20,000-dalton protein that is immunoreactive with antibodies raised against purified chicken gizzard MLC-2. The derived amino acid sequence of the putative human smooth muscle MLC-2 shows only three amino acid differences when compared to chicken gizzard MLC-2. However, comparison with the human cardiac isoform reveals only 48% homology. Blot hybridizations and S1 nuclease analysis indicate that the human smooth muscle MLC-2 isoform is expressed restrictively in smooth muscle tissues such as colon and uterus and in some, but not all, nonmuscle cell lines. Previously reported MLC-2 cDNA from rat aortic smooth muscle cells in culture is ubiquitously expressed in all muscle and nonmuscle cells, and it was suggested that both smooth muscle and nonmuscle MLC-2 proteins are identical and are probably encoded by the same gene. In contrast, the human smooth muscle MLC-2 cDNA that we have characterized from an intact smooth muscle tissue is not expressed in skeletal and cardiac muscles and also in a number of nonmuscle cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Minimal antigenicity of intron A in human recipients demonstrated by three analytical methods.

Antibodies to recombinant human interferon alpha 2b (Intron A) were detected in only a small number of 101 Intron A recipients. This group of cancer patients received Intron A for a mean treatment time of 4.3 months and were selected from disease categories in which subjects were expected to be immunocompetent. Three methods for the detection of antibodies were employed: (a) a bioassay measuring the neutralizing activity of the sera for the antiviral action of interferon alpha 2b, (b) a radioimmunological assay measuring the ability of the sera to prevent the detection of interferon alpha 2b by an immunoradiometric assay (IRMA), and (c) an enzyme-linked immunosorbent assay (ELISA) that measures the binding of immunoglobulins to interferon alpha 2b attached to a solid support. Three of the 101 patients developed neutralizing activity during treatment. Two of these exhibited low neutralizing titers of 1:6-1:9 and were unreactive in the IRMA and ELISA, while only one was positive by bioassay, IRMA, and ELISA. An additional seven patients were positive only in the ELISA. Six of these were borderline positive, i.e., the posttreatment:pretreatment ratio was less than or equal to 5. The results of this study confirm that Intron A is minimally antigenic in human subjects.

Alterations of the amino terminus of murine interferon-gamma: expression and biological activity.

To examine the relationship of murine interferon-gamma (MuIFN-gamma) protein structure to its biological function, several analogs (amino-terminal deletions) of MuIFN-gamma (type II IFN) have been constructed using synthetic oligonucleotide-directed primer repair. High-level expression was achieved through evaluation of such parameters as plasmid origin of replication, antibiotic resistance markers, and host background. The antiviral activity of these analogs varied depending upon the number of amino acids deleted. When the first nine amino acids of IFN-gamma were eliminated (10-136 analog), the protein product lacked detectable antiviral activity. The 10-136 analog but not the 12-136 analog was detectable by Western blots or Coomassie staining of protein gels. These results present the high-level (optimized) expression of several analogs of MuIFN-gamma and further define the role of the amino terminus of the protein with respect to antiviral activity.

Regulated secretion of MuGM-CSF in Saccharomyces cerevisiae via GAL1:MF alpha 1 prepro sequences.

Murine granulocyte-macrophage colony-stimulating factor (GM-CSF) was expressed in Saccharomyces cerevisiae using a novel regulated secretion system. This system involves the fusion of the GAL1 upstream regulatory region to the signal sequence of the alpha mating pheromone, and the integration of this GAL1:MF alpha 1 prepro:MuGM-CSF construct into the yeast chromosome. These constructs were very stable under both selective and nonselective conditions: after 30 generations of growth no plasmid loss was observed. The expression and secretion of MuGM-CSF were analyzed by biological assays and Western blots of yeast culture medium and yeast cell extracts. Expression of MuGM-CSF was regulated by galactose induction. In addition, expression levels were proportional to the number of tandem copies of the gene inserted into the yeast chromosome.

Synergistic antiproliferative effect of recombinant alpha-interferons with recombinant gamma-interferon.

Two human tumor cell lines were studied for their response to the antiproliferative effect of recombinant human interferons (IFNs) alpha 2, alpha 4, a hybrid alpha (delta 4 alpha 2 Bgl II alpha 1), and gamma, individually and in combination. Natural human alpha-IFN was used as a reference point for all experiments. RT4 (bladder carcinoma) cells were overall more sensitive to the antiproliferative effects of the IFNs than A2182 (lung adenocarcinoma) cells. Three-way analysis of variance indicated that the relative effectiveness of the alpha-IFNs was alpha 2 less than alpha 4 less than hybrid alpha less than natural alpha-IFN. On an international reference unit per milliliter basis, gamma-IFN was 50- and 75-fold more effective than natural alpha-IFN and hybrid alpha-IFN in RT4 cells and 5.6-, 12.1-, and 14.9-fold more effective than alpha 4-, hybrid alpha-, and natural alpha-IFN in A2182 cells. In contrast, when recalculated on a nanogram per milliliter basis, gamma-IFN was only threefold more effective than the hybrid alpha-IFN in RT4 and approximately twofold less effective than alpha 4 and the hybrid alpha in A2182. Combinations of alpha-IFNs gave additive or antagonistic effects. When any of the alpha-IFNs were combined with the gamma-IFN, however, a synergistic antiproliferative effect was seen. The magnitude of the synergy was dependent upon the concentration of gamma-IFN used and the type of alpha-IFN in the combination. Antagonistic effects were seen at the lowest gamma-IFN concentration studied (0.2 IRU/ml). Synergy also varied according to the potency of the alpha-IFN used.

The F plasmid may cosegregrate with either DNA strand of the Escherichia coli chromosome.

A stable association exists between the plasmid Flac and one of the polynucleotide strands of the bacterial chromosome. This polynucleotide strand was isolated and tested for uniqueness by DNA-DNA hybridization analysis. The association was found to involve either bacterial DNA strand.