Removal of proteoglycans increases efficiency of retroviral gene transfer

We have previously shown that medium conditioned by virus producer cells inhibits retrovirus transduction, and that a portion of the inhibitory activity is sensitive to chondroitinase ABC. In this study, we have quantitatively evaluated the fraction of the inhibitory activity that is due to chondroitinase ABC-sensitive material and partially characterized the inhibitors. The kinetics of chondroitinase ABC digestion of glycosaminoglycans and virus inhibitory activity in cell culture medium were measured, and the results used to estimate the amount of the chondroitinase ABC-sensitive virus inhibitory activity that was initially in the medium. We found that up to 76% of the inhibitory activity of medium conditioned by packaging cells derived from NIH 3T3 cells is sensitive to chondroitinase ABC. The remainder of the inhibitory activity is not sensitive to other glycosaminoglycan lyases (heparitinase I or heparinase I), which suggests that substances other than glycosaminoglycans or proteoglycans are present in virus stocks and inhibit transduction. To further characterize the inhibitors, proteoglycans from conditioned medium were purified by batch anion exchange and size exclusion chromatography. Two major size groups (100 kDa and 950 kDa) of proteoglycans were isolated. Transduction was inhibited 50% by 0.6 microg/mL of the high-molecular-weight proteoglycan or by 1.7 microg/mL of the low-molecular-weight proteoglycan. Significantly, the proteoglycans, because of their large size and poor sieving properties, coconcentrated with virus particles concentrated by ultrafiltration and prevented any significant increases in transduction efficiency. Transduction efficiencies of virus stocks were increased more than tenfold by ultrafiltration, but only when the concentrated virus was treated with chondroitinase ABC.     

Introduction of human genomic sequences renders CHO-K1 cells susceptible to infection by amphotropic retroviruses.

To learn more about the nature of the block to infection by amphotropic retroviruses exhibited by Chinese hamster cells (CHO-K1), CHO-K1 cells were made susceptible to amphotropic retrovirus infection by introducing genomic DNA from infectable human cells. A clone, designated CHO18, was obtained and shown to be infected as efficiently as NIH 3T3 fibroblasts. Susceptibility of CHO18 cells to infection was specific to retroviruses and vectors bearing an amphotropic envelope. By comparison to CHO-K1 cells, CHO18 cells may provide a useful model for analysis of the molecular events involved in the retrovirus-receptor interaction.     

Complexation with chondroitin sulfate C and Polybrene rapidly purifies retrovirus from inhibitors of transduction and substantially enhances gene transfer

Using amphotropic retrovirus stocks produced by TELCeB6-A cells that encode the Escherichia coli lacZ gene, we found that complexation with chondroitin sulfate C (CSC) and Polybrene (PB) is an effective means to purify retrovirus. Virus stocks contained high levels of inhibitory activity that blocked amphotropic, but not ecotropic, retrovirus transduction. When virus stocks were brought to 80 microg/mL each of CSC and PB, complexes of CSC and PB formed. These complexes incorporated more than 70% of the virus particles but less than 0.4% of all other proteins and no detectable inhibitory activity. Purified virus transduced NIH 3T3 murine fibroblasts 21 to 186-fold more efficiently than virus that was not purified. In addition, virus purification significantly altered the dose response of transduction. When virus that had not been purified was used to transduce cells, the relationship between transduction and virus concentration was highly non-linear. In contrast, when purified virus was used, transduction increased monotonically and was linearly proportional to virus concentration, except when high doses of virus were used. Interestingly, when high doses of virus were used gene transfer reached a maximum plateau level, most likely because particle-associated amphotropic envelope proteins had saturated the cellular receptors for the virus. Our findings illustrate that retrovirus purification increases the maximum number of genes that can be transferred, reduces the amount of virus required to achieve a given level of gene transfer, and reduces uncertainties about the relationship between the amount of virus used and the number of genes transferred.     

Heparin binds to murine leukemia virus and inhibits Env-independent attachment and infection

Certain glycosaminoglycans (GAGs), including heparin, inhibit infection by murine leukemia virus (MLV). We now show that this is due to inhibition of virus attachment independent of the interaction between viral envelope proteins (Env) and their cellular receptors. Heparin blocked the binding of both Env-deficient and amphotropic MLV (MLV-A) particles to NIH 3T3 fibroblasts, CHO cells which lack the amphotropic retroviral receptor Pit-2, and CHO cells transfected with Pit-2 (CHO-Pit-2). Heparin also inhibited the transduction of NIH 3T3 cells by MLV-A over a similar concentration range. This effect was observed within 15 min of exposure to retrovirus. Preloading target cells with heparin had no effect on transduction and both MLV-A and Env-deficient retrovirus bound efficiently to heparin-coated agarose beads, suggesting that heparin interacts with the virus rather than the target cell. This requires both a strong negative charge and a specific structure since GAGs with different charge and carbohydrate composition inhibited virus infection variably. The specificity of GAG-virus interaction also depends on the producer cells, since virus packaged by murine GP+EnvAM12 cells was 1,000-fold more sensitive to inhibition by chondroitin sulfate A than was virus packaged by human FLYA13 packaging cells. No evidence for an interaction between MLV and cell surface proteoglycans was found, however, since the attachment of MLV-A and envelope-defective virus to proteoglycan-deficient CHOpgsA-745 cells was similar to that seen with both wild-type and CHO-Pit-2 cells. Although the molecular mechanism is unclear, this study presents evidence that Env receptor-independent attachment is an important step in MLV infection.     

Retrovirus infection: effect of time and target cell number.

Using a model amphotropic recombinant retrovirus encoding the Escherichia coli lacZ gene and quantitative assays to measure virus infection, we have determined the effects of time and target cell number on infectivity. Infection of various numbers of NIH 3T3 fibroblasts showed that the extent of lacZ virus infection was dependent on virus concentration and independent of target cell number. These results demonstrate that multiplicity of infection is not an accurate predictor of the efficiency of retroviral infection. Varying the time of viral infection revealed that maximal infection occurred after greater than 24 h of exposure of the cells to the lacZ virus. Half-maximal infection occurred after 5 h of exposure. After 2 h of adsorption at 37 degrees C, the majority of infectious virus was not adsorbed to cells but was unbound and able to infect other cells. These results are discussed in terms of both their relevance to the fundamental biology of retrovirus infection and the use of recombinant retroviruses for retrovirus-mediated gene transfer with purposes of gene therapy.     

Subgenomic fragment of molecular cloned Friend murine leukemia virus DNA contains the gene(s) responsible for Friend murine leukemia virus-induced disease.

Friend murine leukemia virus (G-MuLV) is a helper-independent, type C retrovirus isolated from stocks of Friend virus complex (spleen focus-forming virus plus MuLV). In cell culture, F-MuLV has an ecotropic and NB-tropic host range and causes XC cells to fuse. When injected into newborn NIH Swiss mice, F-MuLV produces hepatosplenomegaly, severe anemia, and numerous circulating hematopoietic precursors in the peripheral blood with normal thymus and lymph nodes after 3 to 6 weeks. Recently, we molecularly cloned an 8.5-kilobase pair (kbp) form of F-MuLV DNA from which we could recover the pathogenic F-MuLV virus by DNA transfection of NIH 3T3 cells. From this molecularly cloned F-MuLV DNA, we have now subcloned in pBR322 a 4.1-kbp HindIII fragment which contains in continuity 3.0 kbp from the 3' terminus (env and c region), 0.6 kbp of the terminal repeat sequences, and 0.5 kbp from the 5'terminus of the viral RNA (genome). NIH 3T3 fibroblasts were transfected with this DNA fragment an then infected with the wild mouse amphotropic retrovirus (cl 1504-A). In cell culture, 1504-A is a helper-independent type C virus which has an N-tropic host range and does not cause fusion of XC cells. When injected into newborn NIH Swiss mice, 1504-A does not produce splenomegaly or thymic enlargement in mice held for up to 8 months. The transfection with the F-MuLV fragment and the infection with 1504-A consistently yielded virus preparations that were XC positive. From such virus stocks we were able to isolate both helper-independent and replication-defective XC-positive viruses. The helper-independent virus was shown to be a recombinant virus since it contains a gp70 molecule derived at least in part from F-MuLV and a specific gag precursor derived from 1504-A as determined by radioactive immune precipitation assays. When injected into newborn Swiss mice, the recombinant helper-independent virus caused hepatosplenomegaly in approximately 50% of the mice in 6 to 8 weeks. The histology of the diseased splenic tissue was indistinguishable from that seen in the disease caused by the whole F-MuLV. The replication-defective virus could be pseudotyped with new 1504-A virus, and this viral complex also caused the F-MuLV disease picture when the complex was injected into newborn Swiss mice. We conclude that the genetic information responsible for the pathogenicity of F-MuLV is contained within the 4.1-kbp DNA fragment, which includes env gene sequences, the terminal repeat sequences, and the c region sequences of the F-MuLV genome.     

Caveola-dependent endocytic entry of amphotropic murine leukemia virus

Early results suggested that the amphotropic murine leukemia virus (A-MLV) does not enter cells via endocytosis through clathrin-coated pits and this gammaretrovirus has therefore been anticipated to fuse directly with the plasma membrane. However, here we present data implicating a caveola-mediated endocytic entry route for A-MLV via its receptor Pit2. Caveolae belong to the cholesterol-rich microdomains characterized by resistance to nonionic detergents such as Triton X-100. Extraction of murine fibroblastic NIH 3T3 cells in cold Triton X-100 showed the presence of the A-MLV receptor Pit2 in detergent-insoluble microdomains. Using coimmunoprecipitation of cell extracts, we were able to demonstrate direct association of Pit2 with caveolin-1, the structural protein of caveolae. Other investigations revealed that A-MLV infection in contrast to vesicular stomatitis virus infection is a slow process (t(1/2) approximately 5 h), which is dependent on plasma membrane cholesterol but independent of NH4Cl treatment of cells; NH4Cl impairs entry via clathrin-coated pits. Furthermore, expression of dominant-negative caveolin-1 decreased the susceptibility to infection via Pit2 by approximately 70%. These results show that A-MLV can enter cells via a caveola-dependent entry route. Moreover, increase in A-MLV infection by treatment with okadaic acid as well as entry of fusion-defective fluorescent A-MLV virions in NIH 3T3 cells further confirmed our findings and show that A-MLV can enter mouse fibroblasts via an endocytic entry route involving caveolae. Finally, we also found colocalization of fusion-defective fluorescent A-MLV virions with caveolin-1 in NIH 3T3 cells. This is the first time substantial evidence has been presented implicating the existence of a caveola-dependent endocytic entry pathway for a retrovirus.     

Detection of receptor-specific murine leukemia virus binding to cells by immunofluorescence analysis.

Four classes of murine leukemia virus (MuLV) which display distinct cellular tropisms and bind to different retrovirus receptors to initiate virus infection have been described. In the present study, we describe a rapid, sensitive immunofluorescence assay useful for characterizing the initial binding of MuLV to cells. By using the rat monoclonal antibody 83A25 (L. H. Evans, R. P. Morrison, F. G. Malik, J. Portis, and W. J. Britt, J. Virol. 64:6176-6183, 1990), which recognizes an epitope of the envelope gp70 molecule common to the different classes of MuLV, it is possible to analyse the binding of ecotropic, amphotropic, or xenotropic MuLV by using only a single combination of primary and secondary antibodies. The MuLV binding detected by this assay is envelope receptor specific and matches the susceptibility to infection determined for cells from a variety of species. The binding of amphotropic MuLV to NIH 3T3 cells was shown to be rapid, saturable, and temperature dependent. Chinese hamster ovary (CHO-K1) cells normally lack the ability to bind ecotropic virus and are not infectible by ecotropic vectors. Expression of the cloned ecotropic retrovirus receptor gene (Rec) in CHO-K1 cells confers high levels of ecotropic virus-specific binding and confers susceptibility to infection. Characterization of MuLV binding to primary cells may provide insight into the infectibility of cells by retroviruses and aid in the selection of appropriate vectors for gene transfer experiments.     

Identification of envelope protein residues required for the expanded host range of 10A1 murine leukemia virus.

The 10A1 murine leukemia virus (MuLV) is a recombinant type C retrovirus isolated from a mouse infected with amphotropic MuLV (A-MuLV). 10A1 and A-MuLV have 91% amino acid identity in their envelope proteins yet display different host ranges. For example, CHO-K1 cells are resistant to A-MuLV but susceptible to infection by 10A1. We have now determined that retroviral vectors bearing altered A-MuLV envelope proteins containing 10A1-derived residues at positions 71 (A71G), 74 (Q74K), and 139 (V139M) transduce CHO-K1 cells at efficiencies similar to those achieved with 10A1 enveloped vectors. A-MuLV enveloped retroviral vectors with these three 10A1 residues were also able to transduce A-MuLV-infected NIH 3T3 cells. This observation is consistent with the ability of vectors bearing this altered A-MuLV envelope protein to recognize the 10A1-specific receptor present on NIH 3T3 cells and supports the possibility that residues at positions 71, 74, and 139 of the 10A1 envelope SU protein account for the expanded host range of 10A1.     

A recombinant murine retrovirus for simian virus 40 large T cDNA transforms mouse fibroblasts to anchorage-independent growth.

A recombinant murine retrovirus containing the intact cDNA sequence for the simian virus 40 (SV40) large T antigen (T) was constructed by using the pZIPNeo SV(X)1 vector. Psi 2 packaging cells were then transfected, and G418-resistant clones were used to generate helper-free viral stocks. NIH 3T3 mouse fibroblasts infected by the recombinant T cDNA retrovirus were selected fro G418 resistance. Such cultures synthesized authentic SV40 T and were transformed to anchorage-independent growth at high efficiency. Therefore, this vector has allowed the study of the transformation properties of T under conditions of neutral drug selection and in the absence of SV40 small t antigen.     

Retroviral infection and expression of cationic amino acid transporters in rodent hepatocytes.

The susceptibility of rodent hepatocytes to infection by mouse type C retroviruses was examined in vivo and in vitro and compared with the expression of two membrane proteins that function as transporters for the cationic amino acids CAT-1 and CAT-2. CAT-1 expression in rodents determines susceptibility to ecotropic retrovirus infection by serving as the virus receptor. Recently, it has been suggested that CAT-2 may be a receptor for amphotropic murine leukemia virus. In the present study, CAT-1 expression was observed in Hepa1, a cell line derived from a murine hepatoma, and in rat hepatocytes propagated on collagen monolayers in vitro but not in intact or regenerating rat liver in vivo. The expression of CAT-1 correlated with susceptibility to infection by an ecotropic retrovirus encoding beta-galactosidase. CAT-2 expression was observed in hepatocytes in vitro and in vivo, consistent with reports of infection of regenerating and cultured hepatocytes by amphotropic retroviruses. However, introduction of murine CAT-2 into nonpermissive Chinese hamster cells was not sufficient to confer susceptibility to amphotropic retrovirus infection, using a protocol that could easily demonstrate CAT-1-dependent infection by an ecotropic virus. Our data establish CAT-1 as a major determinant of ecotropic retrovirus infection in rodent hepatocytes and suggest that CAT-2 is not a receptor for viruses in the amphotropic subgroup.     

Matrix fibronectin binds gammaretrovirus and assists in entry: new light on viral infections

A major entry route for the gammaretrovirus amphotropic murine leukemia virus (A-MLV) into NIH 3T3 fibroblasts is via caveola-dependent endocytosis. However, during the infection time, few viral particles can be observed intracellularly. Analyzing the dynamics of the A-MLV infection process by using total internal reflection fluorescence microscopy, we show that the majority of viruses are extracellular and bound to the fibronectin matrix. Moreover, the amounts of bound virus and of fibronectin correlated. Using confocal microscopy, nanoparticles targeted to fibronectin by a III1C-fibronectin fragment or anti-fibronectin antibody were detected intracellularly in NIH 3T3 cells; unconjugated nanoparticles neither bound to cells nor were detectable intracellularly. Furthermore, A-MLV colocalized intracellularly with the fibronectin-targeted nanoparticles, suggesting that they were taken up by the same cellular pathway. Both A-MLV entry and fibronectin turnover depend on caveolar endocytosis, and we found that inhibiting viral binding to the extracellular NIH 3T3 fibronectin-matrix dramatically reduced A-MLV infection, indeed, showing an active role of fibronectin in infection. We suggest that binding to the cellular fibronectin matrix provides a new mechanism by which viruses can enter cells.     

Ovarian carcinoma cells synthesize both chondroitin sulfate and heparan sulfate cell surface proteoglycans that mediate cell adhesion to interstitial matrix

Metastatic ovarian carcinoma metastasizes by intra-peritoneal, non-hematogenous dissemination. The adhesion of the ovarian carcinoma cells to extracellular matrix components, such as types I and III collagen and cellular fibronectin, is essential for intra-peritoneal dissemination. The purpose of this study was to determine whether cell surface proteoglycans (a class of matrix receptors) are produced by ovarian carcinoma cells, and whether these proteoglycans have a role in the adhesion of ovarian carcinoma cells to types I and III collagen and fibronectin. Proteoglycans were metabolically labeled for biochemical studies. Both phosphatidylinositol-anchored and integral membrane-type cell surface proteoglycans were found to be present on the SK-OV-3 and NIH:OVCAR-3 cell lines. Three proteoglycan populations of differing hydrodynamic size were detected in both SK-OV-3 and NIH:OVCAR-3 cells. Digestions with heparitinase and chondroitinase ABC showed that cell surface proteoglycans of SK-OV-3 cells had higher proportion of chondroitin sulfate proteoglycans (75:25 of chondroitin sulfate:heparan sulfate ratio), while NIH:OVCAR-3 cells had higher proportion of heparan sulfate proteoglycans (10:90 of chondroitin sulfate:heparan sulfate ratio). RT-PCR indicated the synthesis of a unique assortment of syndecans, glypicans, and CD44 by the two cell lines. In adhesion assays performed on matrix-coated titer plates both cell lines adhered to types I and III collagen and cellular fibronectin, and cell adhesion was inhibited by preincubation of the matrix with heparin, heparan sulfate, chondroitin sulfate, dermatan sulfate, or chondroitin glycosaminoglycans. Treatment of the cells with heparitinase, chondroitinase ABC, or methylumbelliferyl xyloside also interfered with adhesion confirming the role of both heparan sulfate and chondroitin sulfate cell surface proteoglycans as matrix receptors on ovarian carcinoma cells.     

Isolation and characterization of NIH 3T3 cells expressing polyomavirus small T antigen.

The polyomavirus small T-antigen gene, together with the polyomavirus promoter, was inserted into a retrovirus vector pGV16 which contains the Moloney sarcoma virus long terminal repeat and neomycin resistance gene driven by the simian virus 40 promoter. This expression vector, pGVST, was packaged into retrovirus particles by transfection of psi 2 cells which harbor packaging-defective murine retrovirus genome. NIH 3T3 cells were infected by this replication-defective retrovirus containing pGVST. Of the 15 G418-resistant cell clones, 8 express small T antigen at various levels as revealed by immunoprecipitation. A cellular protein with an apparent molecular weight of about 32,000 coprecipitates with small T antigen. Immunofluorescent staining shows that small T antigen is mainly present in the nuclei. Morphologically, cells expressing small T antigen are indistinguishable from parental NIH 3T3 cells and have a microfilament pattern similar to that in parental NIH 3T3 cells. Cells expressing small T antigen form a flat monolayer but continue to grow beyond the saturation density observed for parental NIH 3T3 cells and eventually come off the culture plate as a result of overconfluency. There is some correlation between the level of expression of small T antigen and the growth rate of the cells. Small T-antigen-expressing cells form small colonies in soft agar. However, the proportion of cells which form these small colonies is rather small. A clone of these cells tested did not form tumors in nude mice within 3 months after inoculation of 10(6) cells per animal. Thus, present studies establish that the small T antigen of polyomavirus is a second nucleus-localized transforming gene product of the virus (the first one being large T antigen) and by itself has a function which is to stimulate the growth of NIH 3T3 cells beyond their saturation density in monolayer culture.     

Enhancement of cell growth by intracellularly expressed herpes simplex virus thymidine kinase

A recombinant cell line (NIH3T3:pLtkSN) was made by infecting parental cells (NIH3T3) with a recombinant retrovirus (pLtkSN) encoding herpes simplex virus thymidine kinase (HSVtk) gene. The cells expressing HSVtk (NIH3T3:pLtkSN) grew 2.3 times more than the parental cells (NIH3T3) in Dulbecco's Modified Eagles Media containing 10% (v/v) horse serum. The NIH3T3:pLtkSN cells also showed a significant enhancement in the maximal cell concentration and the specific growth rate even at 2.5% serum concentration. The specific O2 uptake rate of NIH3T3 was 2.1 times greater than that of NIH3T3:pLtkSN. Under both O2-limited and O2-unlimited conditions, it appears that HSVtk plays an important role in enhancing the growth characteristics of animal cells.     

Construction and isolation of a transmissible retrovirus containing the src gene of Harvey murine sarcoma virus and the thymidine kinase gene of herpes simplex virus type 1

We constructed lambda recombinants containing the Harvey murine sarcoma virus genome and the thymidine kinase (tk) gene of herpes simplex virus type 1 linked to each other. The tk gene was located in a position downstream from both the long terminal repeat and the src gene of Harvey murine sarcoma virus. The DNAs of the lambda recombinants were used to transfect NIH3T3 mouse fibroblasts in order to obtain Harvey murine sarcoma virus DNA-induced foci of transformed cells. The transformed foci were superinfected with a helper-independent retrovirus, and new individual retrovirus were isolated from the superinfected foci. The new viruses could induce focus formation on NIH3T3 cells and could convert NIH3T3(TK-) cells into TK+ cells by carrying the herpes simplex virus type 1 tk gene into the TK- cells. From virus-infected cells, we isolated nonproducer foci on NIH3T3 cells and TK+ transformants on NIH3T3(TK-) cells containing one such new viral genome coding for the dual properties. The new retroviral sequence in the nonproducer cells could be rescued into virus particles at high titers by superinfection with a helper-independent retrovirus. A hybridization analysis indicated that the recombinant virus contained both the Harvey murine sarcoma virus src sequence and the tk gene sequence in a single RNA species approximately 4.9 kilobases long. We concluded that retroviruses can be used as true vectors for genes other than genes that lead to oncogenesis.     

Functional characterization of a U5 ribozyme: intracellular suppression of human immunodeficiency virus type 1 expression.

We have designed a ribozyme that cleaves human immunodeficiency virus type 1 (HIV-1) RNA in U5 (at nucleotide +115). This ribozyme was tested in vitro and was found to give efficient and specific digestion of RNA containing the HIV-1 U5 sequence. When the U5 ribozyme was placed into the HIV-1 genome, virus replication was suppressed in tissue culture. Introduction of this ribozyme into cells by using an amphotropic retrovirus vector significantly reduced expression of U5-containing RNA in cells chronically infected with HIV-1. Naive T cells were cocultivated with packaging cells that produce defective amphotropic retroviruses containing the U5 ribozyme. These lymphocytes were found to be partially protected from HIV-1 infection.     

Human immunodeficiency virus pseudotypes with expanded cellular and species tropism.

One mechanism for expanding the cellular tropism of a virus is through the formation of phenotypically mixed particles or pseudotypes, a process commonly occurring during viral assembly in cells infected with two or more viruses. We report here that dual infection of cells with human immunodeficiency virus (HIV) and a murine amphotropic retrovirus leads to the production of HIV pseudotypes that have acquired the host range of the amphotropic retrovirus and are capable of infecting not only CD4- human cells but also mouse cells. The replication of the HIV pseudotypes in the various CD4- cells was determined by measuring the appearance of HIV antigens in the supernatants, by cocultivation of CD4+ CEM cells with the infected CD4- cells, and in some cases by assaying the culture supernatants directly for infectious virus. Of the cells tested, human foreskin fibroblasts were the best host cells, and by in situ cytohybridization, we were able to document that all cells in the culture were infected. In addition, the temporal appearance of HIV-specific proteins in the HIV pseudotype-infected fibroblasts was similar to that seen in CD4+ CEM cells. If the human fibroblasts were first infected with the amphotropic retrovirus, they demonstrated the property of superinfection exclusion and were resistant to subsequent infection by the HIV pseudotype. In other cell lines, including the human glioblastoma-derived cell line U373MG, HeLa cells, BALB/c mouse embryo cells, and SC-1 wild mouse cells, although the HIV pseudotype infection appeared to be less efficient, substantial amounts of HIV were nevertheless produced. These results indicate that the HIV (amphotropic retrovirus) pseudotypes may be useful for studying the molecular biology of HIV infections in a wide range of cells.     

Comparison of Rous sarcoma virus RNA processing in chicken and mouse fibroblasts: evidence for double-spliced RNA in nonpermissive mouse cells.

Rous sarcoma virus, an avian retrovirus, transforms but does not replicate in mammalian cells. To determine to what extent differences in RNA splicing might contribute to this lack of productive infection, cloned proviral DNA derived from the Prague A strain of Rous sarcoma virus was transfected into mouse NIH 3T3 cells, and the viral RNA was compared by RNase protection with viral RNA from transfected chicken embryo fibroblasts by using a tandem antisense riboprobe spanning the three major splice sites. The levels of viral RNA in NIH 3T3 cells compared with those in chicken embryo fibroblasts were lower, but the RNA was spliced at increased efficiency. The difference in the ratio of unspliced to spliced RNA levels was not due to the increased lability of unspliced RNA in NIH 3T3 cells. Although chicken embryo fibroblasts contained equal levels of src and env mRNAs, spliced viral mRNAs in NIH 3T3 cells were almost exclusively src. In NIH 3T3 cells the env mRNA was further processed by using a cryptic 5' splice site located within the env coding sequences and the normal src 3' splice site to form a double-spliced mRNA. This mRNA was identical to the src mRNA, except that a 159-nucleotide sequence from the 5' end of the env gene was inserted at the src splice junction. Smaller amounts of single-spliced RNA were also present in which only the region between the cryptic 5' and src 3' splice sites was spliced out. The aberrant processing of the viral env mRNA in NIH 3T3 cells may in part explain the nonpermissiveness of these cells to productive Rous sarcoma virus infection.