TEXT

Top Abstract Text References

Foamy viruses (FV) are grouped together as the spumavirus family of complex retroviruses and are ubiquitous in several mammalian species (7). The genomes of these complex retroviruses are comprised of the three typical retrovirus genes (i.e., gag, pol, and env) in addition to the accessory bel genes. The Env of FV is a type 1 membrane-spanning protein that possesses the general structural features observed in all retroviral glycoproteins (20). Two forms of the HFV envelope proteins have been reported (9, 10, 16). The predominant form of HFV Env is synthesized as a polyprotein precursor (gp130) that is proteolytically cleaved to yield two mature subunits, SU (gp80) and TM (gp47) (10). The second form, expressed at 30 to 50% of the level of gp130, is a 170-kDa Env-Bet fusion protein. This is generated by an alternative splicing mechanism and currently its biological role is unknown (9, 16). One feature unique to FV Env relative to all other retroviral glycoproteins is the presence of the dilysine motif, or KKXX, an endoplasmic reticulum (ER) retrieval signal at the TM C terminus (13). It was demonstrated that this protein sorting signal is responsible for ER localization of the HFV glycoprotein (12) and the partitioning of HFV budding to intracellular membranes (11). A putative proteolytic processing site within Env (20) was conserved among all six FV sequences available for analysis (Table 1). This consensus tetrabasic motif, R/K-X-R/K-R, was previously identified as a putative proteolytic cleavage site for retroviral envelope glycoproteins (4). Processing of the polyprotein at this basic amino acid cluster is performed by subtilisin-like cellular endoproteases in late Golgi compartments (15). The polyprotein becomes cleaved at the carboxy-terminal side of the final arginine of the tetrabasic cluster, resulting in the mature SU and TM subunits. Mutations that ablate the ER retrieval motif of the HFV Env protein (13) resulted in significantly enhanced proteolytic processing of the polyprotein (12). Since HFV particles contain cleaved Env, viral maturation presumably disrupts ER localization of Env and results in Env proteolysis upon exposure to the Golgi endoproteases (11). In the present study, the objectives were (i) to experimentally establish the site of proteolytic cleavage in FV Env and (ii) to investigate the importance of FV Env polyprotein cleavage for other aspects of Env processing and function, for Env-driven virion budding, and for HFV infectivity.

A single-amino-acid substitution in the putative tetrabasic cleavage site of HFV Env abolished polyprotein cleavage. The C-terminal arginine (underlined) of the putative HFV cleavage sequence (RKRR), which was absolutely conserved among six FV sequences (20), was changed to threonine (i.e., the R572T mutation) by using an oligonucleotide-directed site-specific mutagenesis method as described previously (12) (Table 1). Our laboratory had earlier demonstrated that disabling mutations to the ER retrieval signal of HFV Env (e.g., KKK to SSS) resulted in increased efficiency of recombinant Env polyprotein cleavage due to increased transport through the Golgi stacks (12). For example, the SSS ER retrieval mutant polyprotein was 42% cleaved, as opposed to the wild-type KKK polyprotein, which was only 3% cleaved after a 90-min chase (12). Therefore, in the present study, the mutation to the putative cleavage site, R572T, was engineered in both an ER retrieval mutant env gene (i.e., SSS) and ER-localized wild-type env (i.e., KKK) in pSVL expression plasmids (Promega, Inc.) to better analyze the effect of R572T on Env processing. The pSVL promoter is the simian virus 40 late promoter.

View larger version (56K): [in this window] [in a new window]   FIG. 1.   Expression, proteolysis, and oligosaccharide modifications for the wild-type and R572T Env proteins. COS-1 cells were transfected with pSVL plasmids (Promega, Inc.) expressing KKK (ER retrieval competent) or SSS (ER retrieval mutant) FV glycoproteins with (KKK/clv or SSS/clv) or without the R572T cleavage site mutation. The cloning of HFV sequences into pSVL was described previously (12). Forty-eight hours posttransfection, the cells were labeled with 35S-protein labeling mix (NEN, Inc.) for 20 min and then chased for 1.5 h. Cells were lysed and immunoprecipitated with FV-positive chimpanzee plasma (kindly provided by P. Fultz). The samples were resuspended in 0.1 M sodium acetate (pH 5.5) and aliquoted. Samples were then incubated at 37°C for 16 h in the presence (+) or absence () of Endo H (50 mU/ml; Boehringer-Mannheim, Inc.) and resolved by sodium dodecyl sulfate (SDS)-PAGE. The position of molecular mass markers (in kilodaltons) is shown on the left (MW).

Uncleaved R572T Env polyproteins are exposed to the Golgi environment. Endoglycosidase H (Endo H) is an enzyme that specifically cleaves the high-mannose oligosaccharide side chains that are added to a peptide backbone in the ER. The conversion of Endo H-sensitive high-mannose N-linked glycans into complex-type sugars occurs in the medial or trans-Golgi compartments and confers Endo H resistance. In order to be certain that uncleaved R572T Env proteins were transported to Golgi compartments where the cellular endoproteases reside, we assessed the proteins' oligosaccharide processing by utilizing Endo H treatment. Following a 20-min labeling period with no chase, all four Env polyproteins (KKK, KKK/clv, SSS, and SSS/clv) were fully sensitive to Endo H (not shown). Following a 1.5-h chase, Endo H digestion of either the KKK or KKK/clv polyproteins produced a new major band with an apparent molecular mass of ~100 kDa compared to the untreated 130-kDa polyprotein, indicating Endo H sensitivity (Fig. 1, compare lane 4 to lane 3 for KKK and compare lane 6 to lane 5 for KKK/clv). Only a small fraction of the KKK or KKK/clv polyproteins were partially resistant to Endo H (lanes 4 and 6, faint bands between ~100 and 130 kDa). The Endo H sensitivity indicated that the KKK and KKK/clv polyproteins carried mainly high-mannose oligosaccharide side chains that hadn't been extensively modified by Golgi enzymes.

Uncleaved R572T Env polyproteins are transported to the plasma membrane. To analyze the effect of the R572T cleavage site mutation on the plasma membrane targeting of FV glycoproteins, the cell surface proteins of transfected COS-1 cells were biotinylated (Fig. 2). The total cellular immunoreactive proteins (lanes 1 to 5) or the subset of biotinylated (and hence surface) immunoreactive proteins (lanes 6 to 10) were visualized. As expected, in cell lysates, the KKK (lane 2) and SSS (lane 4) polyproteins were proteolytically processedmuch more for SSSwhereas, the KKK/clv and SSS/clv polyproteins containing the R572T substitution remained uncleaved (lanes 3 and 5, respectively). For ER retrieval-deficient SSS, the precursor polyprotein, SU, and TM were detected on the cell surface (lane 9). For ER-localized KKK, only gp130 polyprotein was detected at the plasma membrane (lane 7). Prolonged exposure of the gel revealed very low levels of SU and TM on the surface of cells that expressed Env with the wild-type cleavage site and an intact ER retrieval signal (data not shown). For the cleavage site mutants KKK/clv and SSS/clv (lanes 8 and 10), the Env polyprotein alone was present at the plasma membrane. No cleavage products were observed even after prolonged exposure of the gel. The enhanced cell surface expression of SSS/clv relative to KKK/clv (compare lane 10 to 8) and the greater cleavage of gp130^pre into SU and TM for SSS relative to KKK (compare lane 9 to lane 7) nicely demonstrated that mutation of the dilysine ER retrieval motif (in SSS) resulted in enhanced protein transport through the Golgi apparatus to the plasma membrane, as observed before (12). These results indicated that the R572T mutant Env polyproteins were competent for transport to the plasma membrane despite the absence of proteolytic processing.

View larger version (50K): [in this window] [in a new window]   FIG. 2.   Transport of the wild-type or R572T Env proteins to the plasma membrane. Transfected COS-1 cells were metabolically labeled and chased as described in the legend to Fig. 1. Cells were then washed with phosphate-buffered saline and incubated with biotin (EZ-link sulpho-NHS-SS-biotin; Pierce, Inc.) for 30 min at 4°C. The cells were then lysed and immunoprecipitated as before, and the samples were aliquoted. One aliquot was boiled for 5 min in the presence of 10% sodium dodecyl sulfate (SDS) and then reimmunoprecipitated with Streptavidin beads (Pierce, Inc.). The other aliquot was left untreated. Both aliquots were then boiled in sample reducing buffer and resolved by SDS-PAGE (12% polyacrylamide).

The uncleaved FV envelope glycoprotein was deficient in syncytium formation. We next investigated the effect of HFV Env cleavage site mutation on syncytium formation (Fig. 3). Cells expressing the ER-localized KKK Env demonstrated a low level of cell fusion, whereas there was clearly an increase in production of multinucleated giant cells for SSS. The two R572T cleavage mutants (KKK/clv and SSS/clv) produced no cell fusion (Fig. 3). The results of qualitative scoring of syncytium formation by three blinded, independent observers were consistent: SSS (+++), KKK (++), KKK/clv (), and SSS/clv (). Wells containing mock-transfected cells did not contain any multinucleated cells (data not shown); i.e., they were identical to KKK/clv and SSS/clv, as shown. Taken together with the Endo H and surface biotinylation experiments, these results indicated that the uncleaved R572T Env polyproteins were normally expressed and transported through the Golgi apparatus and were further transported to the cell surface, but were nonfunctional in cell-to-cell fusion. The normal intracellular transport, oligosaccharide processing, and plasma membrane targeting of the HFV Env polyprotein in the absence of cleavage were similar to those reported for other retroviruses (17, 21).

View larger version (133K): [in this window] [in a new window]   FIG. 3.   Syncytium formation by the wild-type or R572T Env proteins. Forty-eight hours after transfection, COS-1 cells were washed with phosphate-buffered saline and then fixed in 2% paraformaldehyde containing 0.1% Triton X-100 for 15 min at room temperature. Cells were then washed in phosphate-buffered saline before being stained with nuclear fast red for 30 min at room temperature. Stained cells were observed by light microscopy and photographed.

The R572T cleavage site mutation reduced HFV infectivity. We next analyzed the importance of Env polyprotein cleavage for protein expression and infectivity in the context of the entire virus. The KKK/clv or SSS/clv env genes were substituted into an infectious HFV DNA clone, pHSRV1 (mod), which utilizes the HFV long terminal repeat promoter (kindly provided by Axel Rethwilm) (19). BHK-21 cells were transfected with the wild-type or R572T mutant proviral clones. Extensive vacuolization of the cells was observed at day 3 or 5 after transfection for HFV-KKK and HFV-SSS (data not shown). However, no cytopathic effect was observed for the R572T mutants (HFV-KKK/clv and HFV-SSS/clv). At day 5, the cells transfected with HFV-KKK or HFV-SSS DNA produced detectable HFV Env, Gag, and Bet proteins, as expected (Fig. 4, lanes 2 and 3). However, the predominant viral protein seen in cells for the R572T mutants HFV-KKK/clv and HFV-SSS/clv 22 was Bet (p60); only low levels of expression of the Gag polyprotein p74/70 were observed (Fig. 4, lanes 4 and 5). A similar pattern of gene expression was observed for the env-defective proviral DNA (described in the legend to Fig. 4 [lane 8]). However, significant increases in cellular expression of all HFV proteins by the R572T mutant clones were observed when cleaved Env proteins were provided in trans by cotransfection of pCE-K or pCE-S, which expressed KKK or SSS env, respectively, under the control of a cytomegalovirus promoter (compare lanes 6 and 7 to lanes 4 and 5). The transcomplementation results were even more pronounced when pCE-K was cotransfected with an env-defective provirus clone (compare lane 9 to lane 8).

View larger version (93K): [in this window] [in a new window]   FIG. 4.   Expression of HFV proteins by cells transfected with wild-type or mutant R572T proviral DNA clones. BHK-21 cells were transfected with proviral DNA and labeled at 96 h with 35S-protein labeling mix for 16 h. The culture media were filtered through 0.45-µm-pore-diameter filters. Cell lysate supernatants or media were immunoprecipitated with chimpanzee plasma, resolved by sodium dodecyl sulfate (SDS)-PAGE, and visualized by autoradiography as before. HFV-Env def, the env-defective proviral clone, was created by site-directed mutagenesis with pALTER-1 (Promega, Inc.) as follows. Nucleotides A and G at positions 59 and 61 in env were changed to C and T, respectively, creating an early in-frame termination codon, as well as a unique Bsu36I site to screen for mutants. Also, a 971-bp internal deletion (nucleotides 164 to 1135) was created with EcoRV, followed by religation, which altered the reading frame.

View larger version (29K): [in this window] [in a new window]   FIG. 5.   FAB cell infectivity assay. Five days posttransfection with DNA constructs, the cells and media were harvested and freeze thawed three times to release virus. Virus titrations were then performed by infection of BHK/LTRlacZ indicator cells with 10-fold serial dilutions of the released virus preparations. Two days postinfection, the cells were washed in phosphate-buffered saline (PBS) and fixed with 0.5% glutaraldehyde in phosphate-buffered saline. The blue foci were counted after the cells had been stained with chromogen. The means ± standard deviations for three independent experiments are shown.

View larger version (57K): [in this window] [in a new window]   FIG. 6.   Glycoprotein-driven budding of virus-like particles. Results of sodium dodecyl sulfate (SDS)-PAGE of HFV proteins from 293T cells transfected with pC-gp expressing Gag plus Pol of HFV or with pC-gp plus Env expression plasmids. Following metabolic labeling at 36 h posttransfection, cells were lysed and immunoprecipitated with FV-specific chimpanzee plasma. Released virus particles were extracted from precleared cell culture supernatant by sedimentation through a 20% sucrose cushion (27,000 rpm for 3 h at 18°C in an SW41 rotor). The resulting pellet was lysed and immunoprecipitated as before.