Infection of human cells with murine amphotropic replication-competent retroviruses.

Replication of the murine wild-type 4070A amphotropic retrovirus and a recombinant amphotropic replication-competent retrovirus arising from the PA12 packaging cell line varied considerably among the primate cell types tested. Medium from infected primate fibroblasts and endothelial cells contained the highest viral titers [10(4)-10(5) focus-forming units (ffu)/ml], while most hematopoietic cell lines, such as K562 and MOLT4, were associated with viral titers in the range of 10(3)-10(4) ffu/ml. Interestingly, HTLV-1-transformed T cell lines (TJF-2 and HM) and primary tumor infiltrating lymphocytes (TIL) had very low viral titer (0-10(1) ffu/ml). The low production of virus was not due to low infectivity and, in contrast to the virus, retroviral vectors were expressed without difficulty. Because screening for replication-competent retrovirus (RCR) is an important component of human retroviral-mediated gene therapy clinical protocols, a variety of assays were tested for their ability to detect RCR in virus-exposed cell lines. A biologic assay (3T3 amplification) and polymerase chain reaction (PCR) for the 4070A viral envelope are effective screening methods for RCR, even in cell lines associated with low virus production.

Analysis of the functional and host range-determining regions of the murine ectropic and amphotropic retrovirus envelope proteins.

A series of Moloney murine leukemia virus (Mo-MuLV) envelope gene constructs were analyzed for biological activity. Three classes of recombinant envelopes were examined: insertions, deletions, and chimeras. Insertion (4 to 5 amino acids) and deletion (31 to 62 amino acids) mutants spanned most of the SU (gp70)-coding region and were all biologically inactive. Radioimmunoprecipitation demonstrated that the mutant envelope proteins were incorrectly processed. The Pr80env envelope precursor proteins failed to obtain the proper posttranslational modifications and were not cleaved into SU (gp70) and TM (p15E), suggesting that disruption of Pr80env structure prevents intracellular transport and processing. To analyze the functional domains of the SU portion of the Env protein, we assembled several chimeric constructs. In these constructs, portions of the ecotropic Mo-MuLV envelope gene were replaced with corresponding sequences from the 4070A amphotropic MuLV envelope. Using a retroviral vector pseudotyping assay, 5 of 12 chimeric envelope proteins were shown to be biologically active. Host range was determined by retroviral vector transduction of the appropriate cell, by viral interference studies, and by the productive infection of Chinese hamster ovary cells expressing the murine ecotropic receptor. These results permit assignment of the amino acids responsible for host range determination. Ecotropic host range is determined by the first 88 amino acids of the Mo-MuLV SU, while the amphotropic host range-determining region spans the first 157 amino acids of the 4070A SU.

Sequences present in a small region of the AKV virus envelope gene determine the efficiency with which pseudotyped spleen focus-forming virus infects erythroid target cells.

Induction of erythroleukemia in mice by the replication-defective spleen focus-forming virus (SFFV) relies on the presence of a helper virus to deliver the SFFV genome to erythroid target cells. Pseudotyping studies with different ecotropic murine leukemia viruses (MuLV) have shown that SFFV pseudotyped with Akv, the endogenous ecotropic virus of AKR mice, inefficiently gives rise to virus-induced erythroid bursts (vBFU-E) in vitro and fails to cause erythroleukemia in mice when compared to SFFV pseudotyped with Friend or Moloney MuLV. In order to locate the region(s) of the Akv genome responsible for its inability to act as a helper for SFFV, six different Moloney MuLV chimeras containing Akv envelope sequence substitutions were constructed. Virions with the chimeric envelopes were used to pseudotype SFFV and the complexes were analyzed for their ability to induce vBFU-E in vitro and erythroleukemia in mice. SFFV preparations pseudotyped with three of the constructs containing chimeric envelope genes efficiently gave rise to vBFU-E as did SFFV pseudotyped with Moloney MuLV. SFFV pseudotypes generated from the other three constructs, which all share a common 304-bp region located near the center of the Akv gp70 coding region, and Akv gave rise to very few vBFU-E. However, all SFFV preparations, with the exception of SFFV pseudotyped with Akv, induced erythroleukemia in mice. The results suggest that specific sequences present in the envelope gene of Akv are responsible for the inefficiency of the virus to infect erythroid target cells for SFFV, but additional Akv sequences outside those used in this study affect the ability of the Akv/SFFV virus complex to cause erythroleukemia in mice.

Analysis of retroviral packaging lines for generation of replication-competent virus.

Amplification of retroviral vector sequences occurs in cocultures of ecotropic and amphotropic packaging cell lines. Mixed packaging line cocultures were used to determine both the host range and time of appearance of replication-competent virus after introduction of a retroviral vector. Replication-competent virus was generated at a characteristic time for a given ecotropic and amphotropic packaging line combination. The time required to generate replication-competent virus in a packaging line varied with the number of recombination events necessary to generate replication-competent virus from the retroviral sequences present in the line. The psi 2 packaging line generated replication-competent virus within 10 days after transfection of the N2 vector into a mixture of psi 2 (ecotropic) and PA317 (amphotropic) packaging cells. Under the same conditions, it took only 3 days to develop replication-competent virus in psi 2/PA12 cocultures. The host range of replication-competent virus was used to identify the packaging line that initially generates virus. Each packaging line combination generated replication-competent virus at a characteristic time and this time period can be used as a measure of the "safety" of the packaging line and vector combination.

Retroviral vectors expressing soluble CD4: a potential gene therapy for AIDS.

Retroviral vectors have been developed which produce a secreted form of the helper/inducer T-cell antigen, CD4. Amphotropically packaged vectors were used to transduce cells, and these cells were shown to express the secreted CD4 (sCD4) gene product. The sCD4 produced by the viral vectors is immunoprecipitated by monoclonal antibodies against CD4, which specifically block human immunodeficiency virus (HIV) infection of helper/inducer T cells. A direct physical interaction of vector-produced sCD4 and HIV-1 gp120 was demonstrated by coprecipitation of sCD4/gp120 with antiserum directed against HIV gp120. Furthermore, transduced cells producing sCD4 can protect HIV-susceptible cells from infection by HIV. These data suggest that gene therapy is a potential approach for the treatment of AIDS.