Bicistronic and two-gene retroviral vectors for using MDR1 as a selectable marker and a therapeutic gene.

We describe a series of two-gene and bicistronic retroviral vectors that use the human MDR1 gene as a selectable marker for the overexpression of a second heterologous gene in transduced cells. The vectors use Harvey murine sarcoma virus sequences for viral expression and packaging functions and include sites for cloning foreign genes of interest under the control of either an internal promoter (two-gene vectors) or an internal ribosome entry site (bicistronic vectors). To characterize these vectors, we used neo as a reporter gene for foreign gene expression and as an independently selectable marker for comparison with MDR1. Each of the vector constructions supported high-titer retrovirus production and transduction of mouse and human cell lines. Using MDR1-neo virus supernatants in parallel titering assays, we found that titers based on colchicine resistance were 10- to 20-fold lower than titers based on G418 resistance, suggesting that MDR1 is a more stringent selectable marker than neo in NIH 3T3 and KB-3-1 cell lines. Whereas neo gene expression with the two-gene vectors was subject to host-specific limitations on internal promoter activity, the bicistronic vectors were highly active in three cell lines tested. In K562 cells, using the bicistronic vector, selection with colchicine led to at least 20-fold higher expression of the MDR1 gene product than did selection with G418, suggesting that the stringent MDR1 selection system is very efficient for obtaining overexpression of foreign genes. Retroviral vectors carrying MDR1 as a selectable marker plus a second, heterologous gene of interest could have widespread utility for in vitro and in vivo applications of gene transfer technology, including gene therapy.

Transduction of NIH 3T3 cells with a retrovirus carrying both human MDR1 and glutathione S-transferase pi produces broad-range multidrug resistance.

In the experiments, we examined the ability of a retroviral vector, pHaMASV, to encode two potential chemoprotective genes on separate transcription units. We previously described the pHaMSV vector, which includes the human MDR1 gene as a selectable marker and chemoprotective gene, plus an internal SV40 promoter for expressing a second heterologous gene along with MDR1 [M. E. Metz, D. M. Best, and S. E. Kane. Virology, 208: 634-643, 1995]. To test the ability of this vector to deliver two therapeutic genes simultaneously, the cDNA for human glutathione S-transferase pi (GST pi, the most abundant member of the glutathione S-transferase family in human tumor cells) was inserted into pHaMASV, and this plasmid was transfected into ecotropic packaging cells. The resulting pHaMASV.GST pi ecotropic retrovirus, which was produced at a titer of 2 x 10(6) colony-forming units/ml, was used to transduce NIH 3T3 cells. After initial selection in 60 ng/ml colchicine, a population of transduced cells was exposed to stepwise increasing colchicine concentrations to select for amplified expression of MDR1. As MDR1 expression increased, the expression of GST pi increased in concert, as demonstrated by Northern analysis, Western analysis, and measurement of glutathione S-transferase activity. Transduced cells growing in 1280 ng/ml colchicine had about 3-fold higher total glutathione S-transferase activity than nontransduced cells and 2.5-fold higher activity than transduced cells growing in 60 ng/ml colchicine. Northern hybridizations demonstrated a 3-5-fold increase in both the full-length retroviral message encoding MDR1 and the subgenomic mRNA encoding GST pi after amplification of resistance from 60 to 1280 ng/ml colchicine. The cytotoxic effects of several xenobiotics were evaluated in NIH 3T3 cells transfected with MDR1 (3T3.MDR) or transduced with the MDR1-GST pi retrovirus (3T3.GST640 or 3T3.GST1280) to evaluate the ability of our vector to produce a spectrum of drug resistances specific for the genes expressed. 3T3.MDR and 3T3.GST1280 cells expressing equivalent levels of MDR1 had identical levels of resistance to doxorubicin or colchicine. These results suggest that GST pi expression did not contribute to doxorubicin resistance in this model system. However, 3T3.GST640 cells were about 4-fold resistant to ethacrynic acid and 1-chloro-2,4-dinitrobenzene compared to cells expressing MDR1 alone, consistent with the ability of GST pi to conjugate both of these cytotoxins. Increases in drug resistance paralleled increases in gene-specific mRNA and recombinant protein levels in all cases.4+ chemotherapy.

Expression of helix-loop-helix regulatory genes during differentiation of mouse osteoblastic cells.

Although much is known about the hormonal regulation of osteoblastic cell differentiation, much less is known about the nuclear regulatory molecules that affect this process. We analyzed the expression of several regulatory molecules of the helix-loop-helix (H-L-H) group in primary mouse calvarial cells and in MC3T3-E1 mouse osteoblastic cells in situations representing different degrees of cellular differentiation. H-L-H class regulators are known to participate directly in directing cell fate and differentiation decisions in other mesodermal lineages. Two of the molecules that we studied, Id and E12, have well-established roles in this process. The other, mTwi, the murine homolog of the Drosophila twist gene, is a newly cloned mammalian H-L-H gene. Levels of E12 RNA remained unchanged during differentiation. On the other hand, in both primary osteoblastic cells and MC3T3-E1 cells, the abundance of Id and mTwi declined with cell maturation; mTwi less dramatically than Id. That Id expression is causally related to differentiation is suggested by the finding that MC3T3-E1 cells transfected with an Id-expression plasmid fail to undergo differentiation. We conclude that helix-loop-helix regulatory genes are expressed in mouse osteoblastic cells, where they are likely to participate in differentiation. The E12 gene product is likely to function as a positive modulating factor. In contrast, Id inhibits differentiation, probably by sequestering other H-L-H gene regulators, including E12, in inactive complexes. The precise role of mTwi is more speculative at this time, but the observed pattern of expression is consistent with a role in early and midmesodermal specification that is terminated as cells differentiate.

Immunoglobulin class response to canine distemper virus in gnotobiotic dogs.

Serial serum samples from 27 gnotobiotic dogs infected with R252-canine distemper virus (CDV) were tested for anti-viral IgG, IgM and IgA immunoglobulins using an enzyme-linked immunosorbent assay (ELISA). The results were compared retrospectively to clinicopathological course of disease and to previously reported patterns of complement-fixing and virus neutralizing antibody titers determined in these same sera. Virus-specific IgA was never detected in the sera. High levels of IgG correlated with recovery from disease, whereas the antiviral IgM levels were equivalent in both persistently infected animals and those animals which recovered from disease. The inability to sustain a significant antiviral antibody response in either IgM or IgG classes was characteristic of dogs with fatal encephalitis. The data suggests that IgG is the most important Ig class for recovery from disease.

Comparison of systemic and local immunity in dogs with canine parvovirus gastroenteritis.

To determine whether resistance to canine parvovirus (CPV) gastroenteritis is mediated by local or systemic immunity or both, an enzyme-linked immunospecific antibody assay (ELISA) was developed that quantitated different classes of antibody to CPV. Antibody levels in serum and feces of dogs with CPV-associated gastroenteritis were compared with their clinical signs and viral hemagglutination (HA) titers. Dogs with high levels of CPV coproantibody had a favorable clinical prognosis, high serum antibody levels (hemagglutination inhibition [HI] and ELISA), and low viral HA titers in feces. Conversely, dogs with little or no detectable CPV coproantibody had severe clinical signs and associated mortality rates and high viral HA titers in feces. Many of these dogs had high HI antibody titers. Statistical analysis revealed that only coproantibody level correlated (inversely) with HA titer; serum antibody, whether measured by HI or ELISA, did not. These data suggest that local intestinal immunity is more important than humoral immunity in developing immunological resistance to CPV gastroenteritis.