This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stange, A.
Right arrow Articles by Lindemann, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stange, A.
Right arrow Articles by Lindemann, D.

 Previous Article  |  Next Article 

Journal of Virology, October 2008, p. 9858-9869, Vol. 82, No. 20
0022-538X/08/$08.00+0     doi:10.1128/JVI.00949-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Subviral Particle Release Determinants of Prototype Foamy Virus{triangledown}

Annett Stange,1,{dagger} Daniel Lüftenegger,1,,{dagger} Juliane Reh,1 Winfried Weissenhorn,2 and Dirk Lindemann1*

Institut für Virologie, Medizinische Fakultät Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany,1 Unit for Virus Host Cell Interaction, UMR 5233 UJF-EMBL-CNRS, 6 rue Jules Horowitz, 38042 Grenoble Cedex 9, France2

Received 7 May 2008/ Accepted 1 August 2008


arrow
ABSTRACT
 
Glycoproteins of several viruses have the capacity to induce release of noninfectious, capsidless particulate structures containing only the viral glycoprotein. Such structures are often called subviral particles (SVP). Foamy viruses (FVs), a special type of retroviruses with a replication strategy combining features of both orthoretroviruses and hepadnaviruses, express a glycoprotein (Env) which has the ability to induce SVP release. However, unlike human hepatitis B virus, prototype FV (PFV) naturally secretes only small amounts of SVPs, because ubiquitination of the Env protein seems to suppress the intrinsic capacity for induction of SVP release. In this study, we characterized the structural determinants influencing PFV SVP release, examined the role of specific Env ubiquitination sites in the regulation of this process, and analyzed the requirement of the cellular vacuolar protein sorting (VPS) machinery for SVP egress. We observed that the cytoplasmic and membrane-spanning domains of both the leader peptide (LP) and the transmembrane (TM) subunit harbor essential as well as inhibitory domains. Furthermore, only ubiquitination at the most N-terminal lysine residues (K14 and K15) in LP reduced cell surface expression and suppressed SVP release to wild-type levels. This suggests that interaction of Env with cellular components required for SVP release suppression is effective only when Env is ubiquitinated at these lysine residues but not at others. Finally, SVP release was sensitive to dominant-negative mutants of late components, but not early components, of the cellular VPS machinery. PFV therefore differs from hepatitis B virus in using the same cellular pathway for egress of both virions and SVPs.


arrow
INTRODUCTION
 
Amplification and generation of new progeny viruses require the precise temporal and spatial coordination of multiple processes within the infected host cell. Not only viral but also cellular components, hijacked by the viruses, are essential for viral replication. Cellular mechanisms exploited by viruses for this purpose include cytoskeleton structures, transport machineries, and signaling pathways (reviewed in references 16 and 43). A lot has been learned in the past about the roles and functions of different viral components for these processes. However, we are only at the beginning of understanding the contributions of different cellular machineries and metabolic pathways involved in viral replication.

One of the final steps in production of membrane-enveloped progeny viruses is the release of virions from infected cells, which is accomplished by budding of capsid structures across cellular membranes. In several viral systems, different types of virus-like particles (VLPs) are produced depending on which viral proteins are expressed, and a clear synergy in virus release efficiency is observed upon coexpression of multiple viral proteins. In some cases, the viral glycoproteins are actively involved in viral particle release and either are essential for or enhance this process. In the replication cycles of some viruses, such as hepatitis B virus (HBV), vesicular stomatitis virus, Ebola virus, or foamy viruses (FVs), the release of noninfectious, so-called subviral particles (SVPs) is observed. SVPs lack viral capsid structures but contain viral glycoproteins, and expression of the individual viral glycoproteins alone is sufficient to induce SVP release (9, 36, 46, 53). However, in general little is known about the specific viral glycoprotein determinants responsible for SVP particle release, the cellular pathways involved, or the mechanisms regulating the balance between virion and SVP release.

In contrast, two cellular systems, the cellular ubiquitination and vacuolar protein sorting (VPS) machineries, are known to participate in the morphogenesis and egress of infectious virions of a large variety of different viruses (reviewed in references 5, 19, and 31). The VPS machinery naturally is involved in the sorting of ubiquitinated cargo proteins into the cell's late endocytic compartments, including the multivesicular body and lysosomes. Topological budding of cellular vesicles into the multivesicular body resembles budding of membrane-enveloped viruses at various cellular membranes. Budding is mediated by the sequential interaction of the cargo with different subcomplexes of the mammalian class E VPS machinery (termed ESCRT, or endosomal sorting complex required for transport). Recently, the interactions between certain ESCRT components and specific viral proteins have been well characterized for several different viral systems, and such studies have helped illuminate the final steps in the viral replication cycle (reviewed in references 5, 7, and 34). Generally, most viruses that hijack the cellular VPS machinery for particle release contain so-called L domains (late assembly domains) in their capsid proteins, which serve as docking sites for different factors of the ESCRT complexes or associated proteins.

How exactly the ubiquitination machinery participates in budding, how ubiquitin transfer is mediated, and how targets are recognized by ubiquitin binding proteins are still undefined. Free ubiquitin is enriched in retroviral particles, and the capsid protein (Gag) of many retroviruses is ubiquitinated to a low degree; in these studies, the extent of ubiquitination was shown to be dependent on the type of L domain found in the Gag protein (32, 37, 42). Filoviruses, rhabdoviruses, and some retroviruses, such as murine leukemia virus (MLV), even specifically recruit ubiquitin ligases through their L domain HECT (homologous to E6AP COOH terminus) during these processes (20, 32, 51). The contributions of specific individual ubiquitination sites to particle release processes have not been described, because most retroviral Gag proteins contain large numbers of lysine residues, and most of these are in fact targets for posttranslational modification. Only the simultaneous mutation of multiple ubiquitin attachment sites in human immunodeficiency virus type 1 (HIV-1) Gag resulted in a reduction of particle release (18). Therefore, it is currently not clear whether ubiquitination of Gag serves any specific function or if ubiquitination is simply a by-product of interactions with the cellular ubiquitination machinery, with modification of other participating (and yet unidentified) cellular components providing the essential mechanistic steps.

Spumaviruses, also known as FVs, comprise the only genus of the retrovirus subfamily Spumaretrovirinae. FVs display a unique replication strategy, combining some features common to other retroviruses (Orthoretrovirinae) with features resembling replication of the hepadnaviruses (reviewed in reference 45). Similarly to HBV, but unlike orthoretroviruses, FVs require simultaneous expression of viral capsid and glycoprotein (Env) for budding and release of infectious particles (2, 6, 12, 40). This suggests that both structural proteins harbor structural information important for these steps of the viral replication cycle. The release of capsidless, glycoprotein-containing SVPs from cells expressing only the FV Env protein further supports this notion and constitutes an additional similarity to HBV. However, in contrast to release of HBV, which secretes SVPs in great excess relative to infectious virions, FV SVP release is hardly detectable in wild-type, FV-infected cells (and, in fact, was discovered only a few years ago) (6, 46).

Only limited information is available about cellular machineries that FVs exploit during their replication. The best-studied and -characterized isolate to date is the so-called prototype FV (PFV; formerly known as human FV [HFV]), which was originally isolated from an African patient who presumably was infected zoonotically by a chimpanzee FV (1, 22). With respect to particle morphogenesis, it was demonstrated that PFV, similarly to other viruses, including retroviruses and HBV, uses the cellular VPS machinery for budding and release of capsids preassembled at the centrosome of the host cell (24, 38, 48, 52, 55). PFV Gag contains a PSAP L-domain motif linking the viral capsid in a TSG101-dependent manner through the ESCRT-I complex to the VPS machinery (38, 48). However, only primate FVs harbor a PSAP L-domain motif, raising the question about the nature of the L domains of nonprimate FVs. FV Gag proteins, with the exception of feline FV, contain at most two lysine residues (and in some cases none), which is highly unusual for proteins of 514 to 653 amino acids (aa) in size. So far, we and others have been unable to detect any ubiquitination of the PFV Gag protein containing a single lysine residue, and a capsid protein having this residue mutated displayed wild-type release characteristics, indicating that FVs do not require capsid ubiquitination for recruitment of the VPS machinery (56; unpublished data). Zhadina et al. actually used this special feature to demonstrate that an Env-independent budding variant of PFV Gag having the original PSAP L domain functionally replaced by a PPxY L-domain motif is dependent on catalytically active WWP1 (WW domain-containing E3 ubiquitin protein ligase 1) ubiquitin ligase recruited by the PPxY L domain, irrespective of the presence or absence of ubiquitin acceptor sites in PFV Gag (56). This suggests that ubiquitination of cellular instead of viral proteins is critical for ubiquitin-dependent particle release, at least in this experimental system.

Even though ubiquitination of PFV Gag does not seem to occur and is not required for viral particle egress, PFV interacts with the ubiquitination machinery during particle morphogenesis, as we have reported previously (49). However, it is the viral glycoprotein, not the viral capsid, which is posttranslationally modified by ubiquitin. PFV Env displays a highly unusual biosynthesis relative to other retroviral Env proteins. Initially, PFV Env is translated as a precursor protein, but it is not cotranslationally processed by the cellular signal peptidase, instead adopting a type III membrane topology (27). During its transport to the cell surface, processing by furin or furin-like proteases results in three particle-associated mature subunits, an N-terminal 18-kDa leader peptide (LP) (gp18LP) having a type II membrane topology, a central 80- to 90-kDa surface (SU) subunit (gp80SU), and a C-terminal 48-kDa transmembrane (TM) subunit (gp48TM) having a type I membrane topology (11, 14, 27, 54). The N-terminal cytoplasmic domain (CyD) of gp18LP not only contains the major interaction domain recognized by the viral capsid, which is essential for Env-dependent particle egress, but is also the target for posttranslational modification by ubiquitin (27, 49). Interestingly, glycoprotein ubiquitination seemingly has no effect on FV particle egress. However, it apparently suppresses the intrinsic ability of the PFV glycoprotein to induce release of SVPs, as a PFV Env mutant having all potential cytoplasmic ubiquitin attachment sites in gp18LP inactivated by lysine-to-arginine exchange was characterized by a massive induction of SVPs upon expression in cells (49).

The CyDs of PFV Env contain no sequences with homology to any known L-domain motifs, and therefore, it is unclear which cellular machineries are essential to PFV SVP release. In the present study, we sought to shed light on the ubiquitin-dependent regulatory mechanism of SVP release by identifying and characterizing structural determinants of the PFV Env protein essential for the induction of SVP egress and by asking whether FV SVP release is dependent on the cellular VPS machinery.


arrow
MATERIALS AND METHODS
 
Cells. The human kidney cell line 293T (10) and the human fibrosarcoma cell line HT1080 (44) were cultivated in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and antibiotics.

Expression constructs. The replication-deficient PFV Gag/Pol-expressing vector pczDWP001 and the Gag/Pol/Env-expressing proviral expression vector pDL01, containing an internal spleen focus-forming virus U3 region-driven enhanced green fluorescent protein (EGFP)-Neo fusion protein marker gene cassette, have been described previously (11, 41). All PFV envelope expression constructs generated for this study are based on either the wild type or the ubiquitination-deficient expression constructs pczHFVenvEM002 or pczHFVenvEM140, respectively (40, 49). All envelope mutants were generated using standard PCR cloning techniques and mutagenesis primers. The individual cloning strategies and mutagenesis primers are available on request. Schematic outlines of the constructs used are shown in Fig. 1A, 2A, and 3A. Based on the ubiquitination-deficient PFV Env expression construct pczHFVenvEM140 ({Delta}Ubi), various C-terminal deletion mutants (pczHFVenvEM161 [{Delta}Ubi {Delta}C981], pczHFVenvEM162 [{Delta}Ubi {Delta}C975], pczHFVenvEM160 [{Delta}Ubi {Delta}C960], pczHFVenvEM163 [{Delta}Ubi {Delta}C937], and pczHFVenvEM164 [{Delta}Ubi {Delta}C937Pi]) were generated by using mutant PFV Env expression constructs (pcHFE-1 [{Delta}C981], pcHFE-2 [{Delta}C975], pczHFVenvEM031 [{Delta}C960], pcHFE-3 [{Delta}C937], and pcHFE-3Pi [{Delta}C937Pi]) previously described (40). They exhibit successive deletions, ranging from the C-terminal CyD to the entire membrane-spanning domain (MSD) of the Env TM subunit. The {Delta}C937Pi mutant bears an additional signal sequence for a glycophospholipid anchor (40). Additionally, point mutants based on either pczHFVenvEM002 or pczHFVenvEM140 and having individual lysine residues in the PFV Env TM subunit changed to arginine were generated: pczHFVenvEM225 ({Delta}ER; K984-986R), pczHFVenvEM226 ({Delta}ER+; K959,976,984-986R), pczHFVenvEM167 ({Delta}Ubi {Delta}ER), and pczHFVenvEM168 ({Delta}Ubi {Delta}ER+).


Figure 1
View larger version (30K):
[in this window]
[in a new window]

  		FIG. 1. Analysis of SVP release of PFV Env C-terminal deletion and point mutants. 293T cells were transfected with expression con- structs for various PFV Env mutants as indicated. SVPs were harvested by ultracentrifugation through 20% sucrose from supernatants of the transfected cells. (A) Schematic organization of the PFV Env protein. The C-terminal region spanning the complete CyD and MSD of the TM subunit is enlarged. The C-terminal amino acid sequence of the wild-type envelope construct (wt), with lysine residues marked in bold, is shown at the top. The different C-terminal deletion and point mutants, generated either on the basis of a wild-type or a ubiquitination-deficient ({Delta}Ubi) LP subunit, are illustrated. Individual lysine residues of the TM subunit are indicated in the enlarged schematic illustrations of the C termini of the various constructs. For the point mutants, amino acid sequences deviating at specific positions from the wild-type sequence are shown in italic letters. h, hydrophobic region of the LP; FP, fusion peptide. (B, C) Representative Western blot analysis of cell lysates (cells) by use of a polyclonal anti-PFV Env LP ({alpha}-LP) antiserum or a monoclonal anti-{gamma}-tubulin ({alpha}-Tub) antibody (B) and of SVP preparations (particles) by use of a polyclonal anti-PFV Env LP ({alpha}-LP) antiserum (C). Cells were transfected with pczHFVenvEM002 (wt; lane 1), pczHFVenvEM140 ({Delta}Ubi; lane 2), pcDNA3.1+zeo (pcDNA; lane 3), pczHFVenvEM225 (wt{Delta}ER; lane 4), pczHFVenvEM167 ({Delta}Ubi{Delta}ER; lane 5), pczHFVenvEM226 (wt{Delta}ER+; lane 6), pczHFVenvEM168 ({Delta}ER+{Delta}Ubi; lane 7), pcHFE-1 (wt{Delta}C981; lane 8), pczHFVenvEM161 ({Delta}Ubi{Delta}C981; lane 9), pcHFE-2 (wt{Delta}C975; lane 10), pczHFVenvEM162 ({Delta}Ubi{Delta}C975; lane 11), pczHFVenvEM031 (wt{Delta}C960; lane 12), pczHFVenvEM160 ({Delta}Ubi{Delta}C960; lane 13), pcHFE-3 (wt{Delta}C937; lane 14), pczHFVenvEM163 ({Delta}Ubi{Delta}C937; lane 15), pcHFE-3Pi (wt{Delta}C937Pi; lane 16), and pczHFVenvEM164 ({Delta}Ubi{Delta}C937Pi; lane 17). (D) Quantification of PFV SVP release. Mean values and standard deviations for relative particle-associated PFV Env protein levels, corrected for intracellular expression levels (n = 3 or 4), are shown.


Figure 2
View larger version (26K):
[in this window]
[in a new window]

  		FIG. 2. Analysis of SVP release of PFV Env N-terminal deletion and point mutants. 293T cells were transfected with expression constructs for various PFV Env mutants as indicated. SVPs were harvested by ultracentrifugation through 20% sucrose from supernatants of the transfected cells. (A) Schematic outline of the PFV Env protein (as shown in Fig. 1) and the different N-terminal deletion and point mutants used. The N-terminal region spanning the complete PFV Env LP subunit is enlarged. Individual lysine residues in the CyD of the LP are indicated in the schematic illustration of the wild-type construct (wt). Amino acid sequences of individual mutant constructs deviating at specific positions from the wild-type sequence are shown in italic letters. (B) Representative Western blot analysis of cell lysates (cells) and SVP preparations (particles) by use of a mixture of anti-PFV Gag and anti-PFV Env SU ({alpha}-Gag/SU) hybridoma supernatants. 293T cells were transfected with pczHFVenvEM002 (wt; lane 2), pczHFVenvEM140 ({Delta}Ubi; lane 3), pcDNA3.1+zeo (pcDNA; lane 4), pczHFVenvEM042 (wt{Delta}N5; lane 5), pczHFVenvEM227 ({Delta}Ubi{Delta}N5; lane 6), pczHFVenvEM043 (wt{Delta}N15; lane 7), pczHFVenvEM169 ({Delta}Ubi1{Delta}N15; lane 8), pczHFVenvEM170 ({Delta}Ubi2{Delta}N15; lane 9), pczHFVenvEM070 (wt{Delta}N25; lane 10), pczHFVenvEM171 ({Delta}Ubi{Delta}N25; lane 11), pczHFVenvEM071 (wt{Delta}N40; lane 12), or pczHFVenvEM072 (wt{Delta}N50; lane 13) or cotransfected with pczHFVenvEM002 and the Gag/Pol-expressing vector pczDWP001 (wt + DWP01; lane 1). (C) Quantification of PFV SVP release. Mean values and standard deviations for relative particle-associated PFV Env protein levels, corrected for intracellular expression levels (n = 3 to 6), are shown. n.d., not detectable.


Figure 3
View larger version (59K):
[in this window]
[in a new window]

  		FIG. 3. Analysis of viral particle release regulation by PFV Env ubiquitination. (A) Schematic outline of the PFV Env protein (as shown in Fig. 1) and the examined PFV Env point mutants. The N-terminal CyD of the PFV Env LP subunit is enlarged. Individual lysine residues in the CyD of the LP are indicated in bold for the wild-type construct (wt). The lysine residues changed to arginine in the ubiquitination-deficient mutant background ({Delta}Ubi) are depicted in gray. Restored lysine residues of individual mutant constructs at natural positions are shown in bold, and those at artificial positions replacing other residues in addition are indicated by italic letters. (B) Representative Western blot analysis of cell lysates (cells) by use of a polyclonal anti-PFV Env LP ({alpha}-LP)-specific antiserum. (C) Representative Western blot analysis of corresponding SVPs (particles) purified by ultracentrifugation through 20% sucrose, using consecutive incubation with a polyclonal anti-PFV Env LP ({alpha}-LP) antiserum or a monoclonal antibody specific for ubiquitin ({alpha}-Ubi). 293T cells were transfected with pczHFVenvEM002 (wt; lane 1), pczHFVenvEM140 ({Delta}Ubi; lane 2), pczHFVenvEM145 ({Delta}Ubi K14; lane 5), pczHFVenvEM146 ({Delta}Ubi K15; lane 6), pczHFVenvEM195 ({Delta}Ubi K16; lane 7), pczHFVenvEM196 ({Delta}Ubi K17; lane 8), pczHFVenvEM147 ({Delta}Ubi K18; lane 9), pczHFVenvEM197 ({Delta}Ubi K19; lane 10), pczHFVenvEM198 ({Delta}Ubi K20; lane 11), pczHFVenvEM199 ({Delta}Ubi K27; lane 12), pczHFVenvEM148 ({Delta}Ubi K34; lane 13), or pczHFVenvEM149 ({Delta}Ubi K53; lane 14); transfected with pcDNA only (pcDNA; lane 4); or cotransfected with pczHFVenvEM002 and pczDWP001 (wt + DWP01; lane 3). (D) Quantification of PFV SVP release. Mean values and standard deviations for relative particle-associated PFV Env protein levels, corrected for intracellular expression levels (n = 4 to 8), are shown.

To define the N-terminal boundaries of the PFV Env domains involved in SVP release, we used the already described N-terminal truncation mutants: pczHFVenvEM042 ({Delta}N5), pczHFVenvEM043 ({Delta}N15), pczHFVenvEM070 ({Delta}N25), pczHFVenvEM071 ({Delta}N40), and pczHFVenvEM072 ({Delta}N50) (27). Based on these mutants, further point mutants having individual potential ubiquitination sides (lysine residues) in the LP CyD changed to arginine were generated as indicated: pczHFVenvEM227 ({Delta}N5 {Delta}Ubi; K14,15,18,34,53R), pczHFVenvEM169 ({Delta}N15 {Delta}Ubi1; K34,53R), pczHFVenvEM170 ({Delta}N15 {Delta}Ubi2; K18,34,53R), and pczHFVenvEM171 ({Delta}N25 {Delta}Ubi; R34,53R).

To analyze the regulatory mechanism of PFV SVP release by glycoprotein ubiquitination, additional mutants based on the ubiquitination-deficient PFV Env expression construct pczHFVenvEM140 ({Delta}Ubi) and having single lysine residues introduced at specific positions in the LP CyD were generated as indicated: pczHFVenvEM145 ({Delta}Ubi K14), pczHFVenvEM146 ({Delta}Ubi K15), pczHFVenvEM195 ({Delta}Ubi K16), pczHFVenvEM196 ({Delta}Ubi K17), pczHFVenvEM147 ({Delta}Ubi K18), pczHFVenvEM197 ({Delta}Ubi K19), pczHFVenvEM198 ({Delta}Ubi K20), pczHFVenvEM199 ({Delta}Ubi K27), pczHFVenvEM148 ({Delta}Ubi K34), and pczHFVenvEM149 ({Delta}Ubi K53).

For analysis of cellular proteins involved in PFV SVP release, the following previously described expression constructs were used: full-length TSG101 (TSG-F) (17), an N-terminal truncation mutant (TSG-3') (17, 50), a C-terminal truncation mutant (TSG-5') (8, 50), GFP-VPS4A, GFP-VPS4A (K173Q) (13), EGFP-PK-CHMP3 1-150, EGFP-PK-CHMP3 1-150 M3 (dimerization double mutant of M1 and M2) (35), and YFP-AIP/ALIX (33).

Generation of viral supernatants and analysis of transduction efficiency. FV supernatants containing recombinant viral or SVPs were generated essentially as described earlier (25, 28). Briefly, FV supernatants were produced by cotransfection of 293T cells with equal amounts of the Gag/Pol-expressing vector pczDWP001 and an Env expression plasmid as indicated, using polyethylenimine or CaPO4 transfection reagents. At 24 hours posttransfection, sodium butyrate (10 mM final concentration) was added to the growth medium for 8 h. Subsequently, the medium was replaced and viral supernatants were harvested an additional 16 h later. Transductions of recombinant EGFP-expressing PFV vector particles were performed by infection of 2 x 104 HT1080 cells plated 24 h in advance in 12-well plates for 4 to 6 h using 1 ml of viral supernatant or serial 10-fold dilutions thereof. The percentages of EGFP positive cells were determined by fluorescence-activated cell sorting analysis 48 to 72 h after infection and were in the range of 0.5 to 80% depending on the virus dilution used for infection. All transduction experiments were performed at least three times, and in each independent experiment, the values obtained with the ubiquitination-deficient PFV Env expression construct pczHFVenvEM140 ({Delta}Ubi) were arbitrarily set to 100%.

Antisera, Western blot expression analysis, and quantification of particle release. Western blot expression analyses of cell- and particle-associated viral proteins were performed essentially as described previously (27, 48). For analysis of the potential participation of the cellular VPS machinery in viral or SVP release, cells were cotransfected with various expression constructs for the wild type or for dominant-negative (D/N) mutants of cellular proteins, as indicated, and either the proviral vector construct pDL01 or the ubiquitination-deficient PFV Env expression construct pczHFVenvEM140 ({Delta}Ubi), respectively, at a 1:1 weight ratio. In these experiments, viral particles and SVPs were harvested at 16 h posttransfection to avoid potential general cellular toxic effects of the D/N mutant protein expression. In all other experiments, particles were harvested at 48 h posttransfection. Polyclonal antisera used were specific for PFV Gag (3), the LP of PFV Env, aa 1 to 86 (27), or bovine ubiquitin (U-5379; Sigma). Furthermore, hybridoma supernatants specific for the SU subunit (clone P3E10) of PFV Env (11) or for the PFV Gag protein (clone SGG1) (21) and commercially available monoclonal antibodies specific for ubiquitin (P4D1; Covance) or human {gamma}-tubulin (T3559; Sigma) were employed. The chemiluminescence signal was digitally recorded using a LAS-3000 imager (Fujifilm) and quantified using the Image Gauge software package (Fujifilm). Signal intensities were linear over a range of up to 2 orders of magnitude, as demonstrated previously (48). Viral particle release was determined in independent experiments by quantification of Gag protein bands, whereas SVP release was assayed by quantification of Env protein bands. Viral particle and SVP release levels are always displayed relative to the ubiquitination-deficient PFV Env mutant {Delta}Ubi. Values are the means of results from at least three independent experiments and were corrected for different intracellular Gag or Env expression levels in the individual samples.

Cell surface biotinylation. Cell surface expression of selected PFV Env proteins was analyzed essentially as described earlier (30, 40), with slight modifications. Briefly, 48 h after the addition of DNA, cell surface proteins of CaPO4-transfected 293T cells were labeled with NHS-Biotin (Calbiochem) at 1 mg/ml in phosphate-buffered saline for 30 min on ice. Subsequently, the biotinylation reaction was stopped by addition of phosphate-buffered saline containing 100 mM glycine prior to cell lysis in radioimmunoprecipitation buffer. Lysates were precipitated with a combination of FV-positive chimpanzee serum and anti-PFV Env LP rabbit serum, separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and blotted on nitrocellulose membranes (Hybond ECL; Amersham). Envelope protein expression at the cell surface was analyzed using streptavidin conjugated to horseradish peroxidase (HRP; Pierce), followed by detection with enhanced chemiluminescence (Amersham). After being stripped, the blot was consecutively reprobed using first a mixture of polyclonal rabbit antisera against PFV Gag and PFV Env LP followed by anti-rabbit HRP-coupled secondary reagent and second a mixture of mouse monoclonal antibodies specific for PFV Env SU and PFV Gag. The chemiluminescence signal of every detection reaction was digitally recorded as described above before the blot was stripped and reprobed.


arrow
RESULTS
 
C-terminal glycoprotein domains involved in PFV SVP release. We previously found that the endoplasmic reticulum (ER) retrieval signal present in the CyD of TM and the CyD itself were both dispensable for FV Env-dependent particle release (40). In contrast, the N-terminal 15 aa of the putative MSD were essential and could not be functionally replaced by heterologous viral MSDs or alternative membrane anchorage (40). In the present study, we sought to examine the role of the C-terminal glycoprotein domains in FV SVP release by using some of the previously described C-terminal truncation and point mutants (Fig. 1A). However, since we demonstrated previously (49) that posttranslational modification of the CyD of the N-terminal LP subunit has an important function in regulating the level of SVP release, we decided to generate new variants of the individual C-terminal truncation mutants in which all N-terminal lysine residues of the LP CyD, which constitute potential ubiquitin attachment sites, were replaced by arginine. These constructs are earmarked by the add-on "{Delta}Ubi" (see, for example, Fig. 1). All constructs were expressed by transient transfection of 293T cells, viral particles were then isolated from the cell culture supernatant by ultracentrifugation, and particle release efficiency was quantified by SDS-PAGE and Western blot analysis of viral particles and cell lysates (Fig. 1B to D). First, the analysis revealed that all C-terminal truncations containing a wild-type LP subunit were expressed at somewhat lower levels than the corresponding ubiquitination-deficient proteins ({Delta}Ubi) (Fig. 1B). However, the extent of the C-terminal truncations correlated well with an increase in the amount of processed gp18LP subunit and a corresponding decrease in the amount of Env precursor detected in lysates of cells transfected with wild-type LP subunit-containing constructs (Fig. 1B, lanes 1, 8, 10, 12, 14, and 16). For the {Delta}Ubi LP subunit ({Delta}Ubi)-containing constructs, a similar level of intracellular processing of the precursor was observed, although the relative levels of the different LP cleavage products seemed to vary between the mutants (Fig. 1B, lanes 2, 9, 11, 13, 15, and 17). As illustrated in Fig. 1C and quantitatively summarized in Fig. 1D, SVP release analysis (normalized for cellular expression levels) revealed the following: (i) partial or complete deletion of the CyD of the TM subunit ({Delta}C981 and {Delta}C975) in the wild-type background led to a 5- to 7-fold increase in SVP release, whereas the same mutations in the {Delta}Ubi ({Delta}Ubi) background resulted only in a minor (1.5-fold) enhancement; (ii) deletion of the C-terminal half of the putative MSD ({Delta}C960) in the {Delta}Ubi LP domain ({Delta}Ubi) context resulted in a 4-fold decrease of SVP release and only a small (3-fold) increase in the wild-type LP domain context, whereas complete deletion of the MSD ({Delta}C937) or alternative membrane anchorage ({Delta}C937Pi) in both backgrounds resulted in SVP release at levels equal to or less than those for the full-length wild-type protein; and (iii) inactivation of the dilysine ER retrieval signal by itself ({Delta}ER) or in combination with additional lysine-to-arginine substitutions in the CyD and MSD ({Delta}ER+) resulted in a minor (1.5-fold) increase in SVP release for proteins with the wild-type LP domain, whereas no difference or a slight decrease was observed in the context of the ubiquitination-deficient LP domain ({Delta}Ubi). Taken together, these data indicate that the C-terminal CyD of PFV Env not only is dispensable for SVP release but exerts a suppressive effect, with the dilysine ER retrieval contributing only partially. In contrast, the MSD domain of the TM subunit is essential for SVP release and cannot be functionally replaced by alternative membrane anchorage.

N-terminal glycoprotein domains involved in PFV SVP release. Not only does the gp18LP subunit of PFV Env target the precursor protein to the secretory pathway, but in addition, its N-terminal CyD harbors several additional structural features essential for the viral replication cycle. This includes the major interaction domain for the viral capsid, which is required during the Env-dependent particle release process, and posttranslational modification by ubiquitin, which influences the level of SVP release (27, 49). An additional potential function of subdomains within gp18LP for SVP egress was investigated here by quantifying the SVP release of various N-terminal truncation mutants, again in the wild-type or ubiquitination-deficient ({Delta}Ubi) context (Fig. 2A). Western blot analysis of cellular expression revealed that most mutants with N-terminal truncations, regardless of context (Fig. 2B, lanes 5 and 7 to 13), were expressed at somewhat lower levels than the full-length wild-type or {Delta}Ubi proteins (Fig. 2B, lanes 2 and 3), with the exception of the {Delta}Ubi {Delta}N5 mutant (Fig. 2B, lane 6). At the same time, precursor processing, indicated by the appearance of gp80SU in the cell lysate, was generally increased in the {Delta}Ubi ({Delta}Ubi) proteins relative to the full-length wild-type protein (Fig. 2B, upper panel, lanes 2, 3, 6, 9, and 11). In contrast, in the wild-type background, an increase in precursor processing was observed only for proteins with deletions of 15 or 25 aa (Fig. 2B, upper panel, lanes 7 and 10) and not for truncations comprising only the N-terminal 5 aa or more then 40 aa (Fig. 2B, upper panel, lanes 5, 12, and 13). Quantification of SVP release revealed a massive increase of SVP release when the N-terminal 15 to 25 aa were deleted, regardless of the background, whereas small deletions ({Delta}N5) led to a marginal increase, and more-extensive deletions ({Delta}N40 and {Delta}N50) abolished SVP release (Fig. 2C). Thus, the region comprising the N-terminal 15 aa, previously identified to harbor the major capsid interaction domain (27), exerts a strong inhibitory effect on SVP release. Furthermore, the region spanning aa 25 to 40 seems to be essential for this process, as mutants lacking this part of gp18LP failed to support egress of PFV SVPs.

The intracellular domains of PFV Env contain no sequences with homology to known L-domain motifs that potentially link FV particles to the cellular VPS machinery. However, the results described above imply that the region spanning aa 25 to 40 in the CyD of the LP is essential for SVP release. To further map a potential new L-domain sequence motif in the LP CyD we performed a scanning mutagenesis screen, introducing overlapping blocks of five alanines in the region spanning aa 23 to 63 and analyzing the effects on SVP release (data not shown). The SVP release characteristics of these mutants supported the findings that aa 25 to 40 constitute an important domain for SVP release, as mutations affecting residues 36 to 47 showed intermediate two- to threefold reductions in SVP egress (data not shown). However, examination of additional mutant PFV Env proteins having only individual amino acids exchanged in this region revealed wild-type SVP release levels (data not shown). Therefore, no sequence-specific motif that was essential for PFV SVP release could be identified in the PFV gp18LP CyD. This suggests that the overall structure of this region of the LP CyD, and not specific amino acids, is required for promoting SVP release.

Ubiquitination of N-terminal gp18LP lysine residues suppresses PFV SVP release. The CyD of PFV Env gp18LP harbors several lysine residues, in particular at its N terminus, which can potentially be modified by ubiquitination. A ubiquitination-deficient mutant ({Delta}Ubi) having all lysine residues mutated to arginine displayed a dramatic increase in SVP release (49). To investigate the contributions of individual ubiquitination sites to suppression of PFV Env SVP release, we reintroduced individual lysine residues at their natural positions ({Delta}Ubi K14, K15, K18, K34, and K53) in the otherwise ubiquitination-deficient ({Delta}Ubi) background but also artificially replaced individual residues at different positions throughout the LP CyD by lysine ({Delta}Ubi K16, K17, K19, K20, and K27) (Fig. 3A). Transfection of the individual mutant expression constructs in 293T cells, isolation of SVPs from the cell culture supernatant, and quantification of relative SVP release by Western blot analysis (Fig. 3D) revealed that only lysine residues near the N terminus at natural positions (K14 and K15) suppressed SVP release to wild-type levels (Fig. 3B and C, lanes 1, 2, 5, and 6, and D). Furthermore, the further from the N terminus the lysine residues were introduced, regardless of whether they corresponded to naturally occurring lysines or were artificially introduced, the weaker the observed effect was, with inhibition of SVP release reaching levels similar to those for the {Delta}Ubi mutant with {Delta}Ubi K18 (Fig. 3B and C, lanes 1, 2, and 7 to 14, and D). Introduction of a lysine at the artificial position K17 or the natural position K34 displayed a twofold reduction of SVP release in comparison with the level for the {Delta}Ubi mutant (Fig. 3B and C, lanes 8 and 13, and D). Interestingly, when released SVPs were probed with a ubiquitin-specific antibody, specific LP-derived bands were detectable for all mutants except K53, suggesting that this natural lysine residue is not modified, probably due to its close membrane proximity (Fig. 3C, lanes 5 to 14) (23). These results suggest that, although PFV Env can be ubiquitinated at different positions throughout its CyD (provided that lysine residues are present), only modifications near the N terminus are able to suppress Env-mediated SVP release to wild-type levels.

Differential PFV Env LP ubiquitination and viral release. The previous characterization of the ubiquitination-deficient PFV Env protein ({Delta}Ubi) revealed only a minor positive effect on viral particle release and infectivity (49). We wondered whether the restriction of ubiquitination to individual specific positions is compatible with viral particle morphogenesis. Therefore, we cotransfected 293T cells with expression constructs encoding the single-lysine-containing PFV Env mutants described above, together with a replication-deficient PFV transfer vector encoding Gag/Pol and an EGFP marker gene (pczDWP001). Viral particle release was analyzed by quantitative Western blotting and infectivity by a marker gene transfer assay with HT1080 target cells. The results are summarized in Fig. 4. The particle release of wild-type PFV Env was only slightly reduced compared to that of the ubiquitination-deficient mutant ({Delta}Ubi) (Fig. 4A and B, lanes 1 and 2, and C). All other single-lysine-containing mutants showed particle release at levels comparable to those for the {Delta}Ubi mutant, except {Delta}Ubi K17. In the case of {Delta}Ubi K17, particle release was reduced to about 40% of that of the {Delta}Ubi protein (Fig. 4A and C). Similar to the SVP preparations shown in Fig. 3, all FV particle samples displayed prominent higher-molecular-weight ubiquitinated variants of gp18LP, which were specifically recognized by anti-LP and anti-ubiquitin antibodies (Fig. 4B). Only for the ubiquitination-deficient mutant {Delta}Ubi and the K53 samples were these variants not detectable, and K14 Env seemed to be more efficiently ubiquitinated than the wild-type protein (Fig. 4B, lanes 2, 4, and 13). Infectivity analysis of corresponding 293T supernatants reflected the particle release characteristics of the individual mutants quite well, with two exceptions (Fig. 4D). The infectivity of the {Delta}Ubi K17 mutant was reduced to about 15% and that of the {Delta}Ubi K34 mutant to about 30% of that of the {Delta}Ubi protein, suggesting reduced specific infectivity for these two types of mutant particles in comparison to the levels for {Delta}Ubi protein-containing particles. This might reflect some steric interference with particle egress or Env functions of attached ubiquitin residues at these positions. Thus, with the exceptions of K17 and K34, individual potential ubiquitination sites within the PFV Env glycoprotein LP CyD had no major impact on viral particle release and infectivity, in clear contrast to their strong differential effects on level of SVP release.


Figure 4
View larger version (54K):
[in this window]
[in a new window]

  		FIG. 4. Analysis of viral particle release and infectivity levels of different PFV Env single-lysine revertants. Mutant PFV particles were generated by transient transfection of 293T cells with the PFV Env expression constructs as indicated. (A) Representative Western blot analysis of cell lysates (cells) by use of polyclonal anti-PFV Env LP ({alpha}-LP)- or anti-PFV Gag ({alpha}-Gag)-specific antisera. wt, wild type. (B) Representative Western blot analysis of corresponding SVPs (particles) purified by ultracentrifugation through 20% sucrose, using a mixture of polyclonal anti-PFV Env LP- and anti-PFV Gag-specific antisera ({alpha}-LP/Gag) or a monoclonal antibody specific for ubiquitin ({alpha}-Ubi). 293T cells were cotransfected with pczDWP001 and pczHFVenvEM002 (wt; lane 1), pczHFVenvEM140 ({Delta}Ubi; lane 2), pczHFVenvEM145 ({Delta}Ubi K14; lane 4), pczHFVenvEM146 ({Delta}Ubi K15; lane 5), pczHFVenvEM195 ({Delta}Ubi K16; lane 6), pczHFVenvEM196 ({Delta}Ubi K17; lane 7), pczHFVenvEM147 ({Delta}Ubi K18; lane 8), pczHFVenvEM197 ({Delta}Ubi K19; lane 9), pczHFVenvEM198 ({Delta}Ubi K20; lane 10), pczHFVenvEM199 ({Delta}Ubi K27; lane 11), pczHFVenvEM148 ({Delta}Ubi K34; lane 12), or pczHFVenvEM149 ({Delta}Ubi K53; lane 13) or transfected only with pcDNA3.1+zeo (pcDNA; lane 3). (C) Quantification of viral particle release. Mean values and standard deviations for relative particle-associated PFV Env protein levels, corrected for intracellular expression levels (n = 4 or 5), are shown. (D) Quantification of cell culture supernatant infectivity using the flow cytometric EGFP marker gene transfer assay. Mean values and standard deviations for relative infectivities (n = 7 to 11) are shown. mock, mock-infected target cells.

Ubiquitination of gp18LP N-terminal lysine residues reduces cell surface expression of Env. Intracellular trafficking and cell surface expression of PFV Env are influenced by different regions and signals throughout the glycoprotein, including the dilysine ER retrieval signal located at the cytoplasmic C terminus of the TM subunit and a conserved lysine-proline motif in the putative MSD (15, 27, 41). We next examined the contribution of LP ubiquitination to cell surface expression regulation. To do this, we performed selective cell surface biotinylation of transfected 293T cells expressing individual PFV Env ubiquitination mutants, either alone (Fig. 5A) or in combination with a Gag/Pol and EGFP marker gene expressing PFV transfer vector (Fig. 5B). PFV Env was immunoprecipitated and separated by SDS-PAGE, and after blotting, cell surface-exposed, biotinylated PFV Env was detected with streptavidin-HRP. Subsequently, after streptavidin-HRP was stripped from the membrane, total Gag and Env protein levels were determined by Western blot analysis using anti-Gag and anti-SU specific antibodies. The results shown in Fig. 5A demonstrate that wild-type PFV Env is found at the cell surface only in very small amounts when expressed by itself (Fig. 5A, lane 3), as reported previously (41). In contrast, for ubiquitination-deficient PFV Env ({Delta}Ubi), large amounts of biotinylated SU and TM subunits were detectable at the cell surface (Fig. 5A, lane 1). The cell surface biotinylation patterns of the different ubiquitination mutants analyzed were mirror images of the SVP release capacities that they showed when expressed by themselves. Mutants that secreted only small amounts of SVPs ({Delta}Ubi K14, K15, and K16) showed low cell surface Env expression, similar to the wild type (Fig. 5A, lanes 3 to 6). In contrast, mutants whose expression resulted in an enhanced release of SVPs compared to the level for wild-type PFV Env (K17, K18, K19, K27, K34, and K53) also displayed strong Env-specific cell surface biotin signals at levels similar to or even higher than those for the ubiquitination-deficient Env protein ({Delta}Ubi) (Fig. 5A, lanes 1 and 7 to 13). Upon cotransfection of a PFV Gag/Pol-expressing transfer vector (DWP01), the cell surface expression levels of all Env proteins analyzed were similar, indicating that the natural intracellular retention of PFV Env is neutralized by the specific capsid-Env interaction for all mutants equally (Fig. 5B).


Figure 5
View larger version (63K):
[in this window]
[in a new window]

  		FIG. 5. Cell surface expression levels of different PFV Env single-lysine revertants. Representative cell surface biotinylation analysis of immunoprecipitates from 293T cells transfected with indicated PFV Env expression constructs alone (A) or cotransfected with the Gag/Pol-expressing PFV vector pczDWP001 (+ DWP01) (B) as indicated. Blots were consecutively incubated with streptavidin-HRP (Strep-HRP) and, after being stripped, with a mixture of anti-PFV Env SU ({alpha}-SU) and anti-PFV Gag ({alpha}-Gag) hybridoma supernatants. 293T cells were transfected with pczHFVenvEM140 ({Delta}Ubi; lane 1), pcDNA3.1+zeo (pcDNA; lanes 2 and 16), pczHFVenvEM002 (wt; lane 3), pczHFVenvEM145 ({Delta}Ubi K14; lane 4), pczHFVenvEM146 ({Delta}Ubi K15; lane 5), pczHFVenvEM195 ({Delta}Ubi K16; lane 6), pczHFVenvEM196 ({Delta}Ubi K17; lane 7), pczHFVenvEM147 ({Delta}Ubi K18; lane 8), pczHFVenvEM197 ({Delta}Ubi K19; lane 9), pczHFVenvEM198 ({Delta}Ubi K20; lane 10), pczHFVenvEM199 ({Delta}Ubi K27; lane 11), pczHFVenvEM148 ({Delta}Ubi K34; lane 12), or pczHFVenvEM149 ({Delta}Ubi K53; lane 13); cotransfected with pczDWP001 and pczHFVenvEM140 ({Delta}Ubi; lane 15), pczHFVenvEM002 (wt; lane 17), pczHFVenvEM145 ({Delta}Ubi K14; lane 18), pczHFVenvEM146 ({Delta}Ubi K15; lane 19), pczHFVenvEM195 ({Delta}Ubi K16; lane 20), pczHFVenvEM196 ({Delta}Ubi K17; lane 21), pczHFVenvEM147 ({Delta}Ubi K18; lane 22), pczHFVenvEM197 ({Delta}Ubi K19; lane 23), pczHFVenvEM198 ({Delta}Ubi K20; lane 24), pczHFVenvEM199 ({Delta}Ubi K27; lane 25), pczHFVenvEM148 ({Delta}Ubi K34; lane 26), pczHFVenvEM149 ({Delta}Ubi K53; lane 27), or pcDNA (pcDNA; lane 28); or cotransfected with pcDNA and pczHFVenvEM140 (pcDNA + {Delta}Ubi; lane 29). wt, wild type.

PFV SVP release and the cellular VPS machinery. Like HBV and a variety of other viruses, PFV hijacks the cellular VPS machinery for viral particle release. In the case of PFV, this occurs through a PSAP L-domain-dependent interaction between the Gag protein and the ESCRT-I component TSG101 (24, 38, 48, 52). In contrast to HBV viral particle release, HBV SVP release was recently reported to be independent of the ESCRT complexes (24, 52). We examined the requirement of the VPS machinery for PFV SVP release by cotransfecting the ubiquitination-deficient PFV Env ({Delta}Ubi) expression construct with expression constructs for the wild type and for D/N mutants of various ESCRT components or associated proteins, including factors acting early or late in the VPS pathway. For comparison, the effects of coexpression of these cellular proteins on viral particle release were reevaluated in parallel by cotransfecting these constructs with a replication-deficient PFV proviral vector construct. The results of quantitative particle release analysis are summarized in Fig. 6. As reported previously (48), various forms of TSG101, known to inhibit viral particle release either in a PSAP L-domain-dependent manner (TSG 5') (8) or by interfering with the release of some viruses in a more global manner (e.g., HIV-1 and MLV) (17, 47), reduced PFV viral particle release by 3 to 10-fold, whereas SVP release was either unaffected or reduced only by 20% or less (Fig. 6A). A D/N form of ALIX (YFP-ALIX), a protein linking ESCRT complexes and interacting with YPXL L domains, enhanced viral particle and SVP release marginally (Fig. 6A). Of proteins acting late in the ESCRT machinery, viral particle and SVP release levels were inhibited fivefold by a D/N form of CHMP-3 (CHMP3 1-150) (Fig. 6B). Use of an additional, oligomerization-defective variant of this D/N mutant (CHMP-3 1-150 M3) showed that oligomerization was important for mediating the D/N interfering activity. Furthermore, overexpression of wild-type VPS4A inhibited both particle and SVP release by 30 to 50% (Fig. 6B). In contrast, a D/N form of VPS4A (VPS4A K/Q) reduced viral particle release 20-fold in comparison to mock-transfected controls, while at the same time SVP release was reduced 5-fold (Fig. 6B). Taken together, these data indicate that late but not early components or proteins associated with the cellular VPS machinery were required for efficient PFV SVP export.


Figure 6
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 6. Inhibition of viral and subviral release of PFV particles by coexpression of various cellular proteins. Viral particles and SVPs were generated by transfecting 293T cells with the replication-deficient, Gag/Pol/Env-expressing PFV proviral vector pDL01 and the ubiquitination-deficient PFV Env expression vector pczHFVenvEM140, respectively. Expression constructs of the wild type (wt) or D/N mutants for various cellular proteins, acting early (A) or late (B) in the VPS machinery, were cotransfected as indicated at the y axes of the bar graphs. Viral particles and SVPs were purified by ultracentrifugation through 20% sucrose, and particle release was quantified by Western blot analysis using polyclonal anti-PFV Env LP-specific antibodies. Mean values and standard deviations (n = 3 to 5) for relative particle-associated PFV Env protein levels in viral particle (black bars) or SVP (gray bars) preparations, corrected for intracellular expression levels, are shown. "Mock" indicates cotransfection with the respective control vector to the individual constructs (e.g., pEGFP for VPS4A).


arrow
DISCUSSION
 
In several viral systems, the viral glycoprotein is the driving force or at least an essential component for viral particle budding and release (9, 36, 53). This function is often characterized by the additional glycoprotein-induced release of capsidless particulate structures called SVPs. For orthoretroviruses, an essential role for the viral Env protein in viral particle formation has not been described, nor has the secretion of SVPs been observed. Only for FVs, which differ from other retroviruses in several aspects of their replication strategy (and are therefore grouped in a separate subfamily as Spumaretrovirinae), has SVP release been reported (46). Interestingly, SVP release is one of several unique features of the FV replication cycle that bear strong homology to that of HBV (reviewed in references 6 and 26). However, unlike HBV, which secretes SVPs in great excess relative to infectious virions, FVs naturally secrete only very small amounts. The level of FV SVP egress seems to be regulated by glycoprotein ubiquitination, a very unusual posttranslation modification for a viral glycoprotein (46, 49). In this study, we characterized viral structural requirements involved in PFV SVP release, further investigated a unique posttranslational regulatory mechanism of this process, and examined the contribution of the cellular VPS machinery to SVP egress.

The characterization of structural domains within the PFV Env protein involved in SVP egress revealed the presence of two inhibitory and two essential regions within the glycoprotein LP and TM subunits (Fig. 7). The inhibitory domains included a 10-aa region (aa 6 to 15) of the LP CyD and the C-terminal 7 to 13 aa of the TM subunit comprising its CyD. Membrane anchorage provided by the TM MSD was essential for SVP release; however, this could not be compensated for by alternative phosphatidylinositol-mediated anchorage. To a certain extent, the central portion of the LP CyD spanning aa 36 to 47 also contributed to SVP release. Some of these results are reminiscent of the requirements for FV viral particle release, for which the TM CyD is dispensable and the cognate TM MSD and LP CyD are essential (27, 39). However, deletion of as few as the five N-terminal amino acids of LP completely abolished PFV viral particle egress, but this was not sufficient to overcome the inhibitory effect on SVP release (27). Furthermore, in the context of an otherwise wild-type protein, the TM CyD even inhibited SVP release, whereas viral particle release efficiency was not affected (40). Moreover, alanine substitution of the evolutionarily conserved tryptophans W10 and W13, which are essential for the Env-Gag interaction and support viral particle release, did not diminish the levels of SVP release compared to those for the wild type (data not shown). Taken together, these results suggest that the subdomains of the PFV Env protein involved in viral particle and SVP release partially overlap but are not identical.


Figure 7
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 7. Overview of PFV Env functional domains. (A) Schematic illustration of the PFV Env precursor protein membrane topology. (B) Schematic organization of the PFV Env subunit structure. N- and C-terminal regions are enlarged, and domains which have inhibitory effects on or are essential for SVP release are indicated. N, N terminus; C, C terminus; h: hydrophobic region of the LP; FP: fusion peptide.

The inhibitory effect of the TM CyD on SVP release could be explained by loss of the dilysine ER retrieval signal present at the C terminus of the TM CyD (15). However, selective inactivation of the ER retrieval signal in the full-length wild-type glycoprotein only partially released this inhibitory effect. This suggests that either other sequences in this short CyD of the TM subunit are contributing to the inhibition or the TM CyD cooperates with other domains in the glycoprotein to achieve the inhibitory effect. One potential candidate for cooperation would be the N-terminal CyD of the LP subunit, as the N-terminal 15 aa of LP were defined in this study as a second and major suppressive domain for SVP release. One could envision that this LP subdomain is structurally required to properly expose the ER retrieval signal of the TM CyD to the COPI coat on vesicles mediating retrograde transport of dilysine signal-containing proteins from proximal Golgi compartments to the ER (reviewed in references 4 and 29). Alternatively, the inhibitory domain of the LP CyD is, by itself or in combination with the TM CyD, involved in either the interaction with the COP coat proteins or other cellular proteins, thereby mediating the suppressive effect.

We also tested the contributions of individual ubiquitin attachment sites to inhibition of the glycoprotein's intrinsic capacity to induce SVP release. Interestingly, we observed that only ubiquitination of the first two naturally occurring lysine residues (at positions 14 and 15 in the LP CyD) suppressed SVP release to wild-type levels. Naturally occurring or artificially introduced lysine residues that were more membrane proximal were also analyzed, but apparently ubiquitination at these sites did not significantly inhibit SVP egress. Furthermore, artificial N-terminal fusion of ubiquitin to an otherwise ubiquitination-deficient PFV Env protein also did not reduce PFV Env-mediated SVP release to wild-type levels (data not shown). All these results suggest that a specific, context-dependent ubiquitination is crucial for the observed suppressive effect of this posttranslational glycoprotein modification. Perhaps this is achieved through interaction with specific cellular proteins, but these remain to be identified. Such interactions might influence intracellular trafficking and/or distribution of the viral glycoprotein, which in turn might be important for SVP induction.

In this respect, it is noteworthy that there was a strict correlation between cell surface expression levels and extents of detectable SVP release. All N-terminal truncation mutants and ubiquitination site mutants that displayed high cell surface expression levels (as detected by biotinylation analysis) also secreted large quantities of SVPs. The only exception was the C-terminal truncation mutant HFE-3Pi, in which the PFV MSD of the TM subunit was replaced by an alternative phosphatidylinositol membrane anchor. This mutant, in the context of an otherwise wild-type protein, was previously shown to be expressed at high levels on the cell surface (41) but was unable to support SVP release in significant amounts, even in the context of a ubiquitination-deficient LP subunit. However, it might be that the oligomeric structure of the glycoprotein subunit complex, not just cell surface glycoprotein expression, is crucial for SVP release. In this respect, it is noteworthy that the PFV glycoprotein forms very prominent spike structures on cellular membranes and on viral particles and has been reported to form trimeric structures that are often arranged in hexagonal rings sharing adjacent Env trimers (27, 54). The domains or structures mediating PFV Env oligomerization have not been characterized yet. However, they could reside in the MSD of the TM subunit, which is absent in the HFE-3Pi mutant protein. If so, the HFE-3Pi protein might not be able to form a wild-type-like oligomeric and lattice structure that might be essential for SVP induction. This could explain the requirement of the PFV TM MSD for both viral (40) and SVP release. However, since the crystal structures of PFV Env and its subunits have not been solved, the domains responsible for PFV Env oligomerization still need to be characterized, and since no data on the oligomeric state of the HFE-3Pi protein is available, this notion is still speculative in nature.

With respect to defining cellular pathways involved in PFV particle release, we examined the potential participation of certain components of the VPS machinery. Proteins of the VPS machinery were already reported to be important for egress of various viruses, including retroviruses, and perhaps more importantly for release of FV particles; moreover, VPS proteins were recently implicated in the export of infectious HBV particles (5, 24, 38, 48). Our results suggest that late components of the VPS machinery, including CHMP3 (which participates in ESCRT-III complex formation) and the recycling AAA ATPase VPS4, affect PFV SVP release. In contrast, two components acting early in the VPS pathway, TSG101 and AIP/ALIX, seem not to be involved, consistent with the observation that the CyDs of PFV Env lack sequences with homology to known L-domain motifs. Interestingly, PFV SVP release was insensitive not only to D/N TSG101 mutants (TSG-5') that inhibit viral particle release in a PSAP-dependent manner (8, 17, 47) but also to D/N TSG101 mutants that inhibit particle egress of some viruses (e.g., HIV-1 and MLV) but not others (e.g., equine infectious anemia virus) by interfering with endosomal sorting in a more global manner (17, 47). Of course, this raises important questions: how does the FV glycoprotein use the VPS machinery for SVP release, and through which cellular proteins is the VPS machinery accessed? Furthermore, it has to be clarified whether the observed inhibitory effects of D/N mutants of late components of the VPS machinery really are specific to PFV SVP release or whether they reflect only a general requirement of the cell's endocytic machinery for this FV particle release process. In this respect, it also important to note that HBV SVP egress, but not infectious viral particle release, has been reported to be independent of the VPS machinery (24). However, unlike PFV, HBV generates three different viral glycoproteins. Of these, the small S protein is the principal and essential component of SVPs and is transported efficiently to the cell surface. In contrast, the glycoprotein essential for formation and egress of infectious HBV particles (which is VPS pathway dependent) is the large S protein. As with PFV Env, the HBV large S protein is retained in the ER. Therefore, HBV might have developed different mechanisms for the generation of viral and subviral structures, involving at least two separate glycoproteins, whereas PFV expresses a single glycoprotein, simultaneously harboring structural information essential for both particle types. This might explain the differential requirement of the VPS machinery for SVP release of HBV and PFV.

Crucial for the further understanding of the release processes of the different PFV particulate structures will be the identification of host cell factors that directly interact with the glycoprotein, in particular with its CyDs, and might be involved in intracellular trafficking and subcellular localization. In particular, it will be interesting to identify the components of the cellular-ubiquitination machinery that mediate this unusual viral glycoprotein modification, including proteins that specifically recognize posttranslationally modified PFV Env species.


arrow
ACKNOWLEDGMENTS
 
We thank Wes Sundquist, Paul Bieniasz, Heinrich Göttlinger, and Eric Freed for the generous provision of various expression constructs and Welkin Johnson for critical reading of the manuscript.

This work was supported by grants from the DFG (Li621/3-1 and Li621/4-1) and BMBF (01ZZ0102) to D.L. and from the DFG (WE 2839/1-1) to W.W.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Institut für Virologie, Medizinische Fakultät Carl Gustav Carus, Technische Universität Dresden, Fetscherstr. 74, 01307 Dresden, Germany. Phone: 49-351-458-6210. Fax: 49-351-458-6314. E-mail: dirk.lindemann{at}tu-dresden.de Back

{triangledown} Published ahead of print on 6 August 2008. Back

{dagger} A.S. and D.L. contributed equally. Back

Present address: ViroLogik GmbH, Henkestr. 91, 91052 Erlangen, Germany. Back


arrow
REFERENCES
 
    1
  1. Achong, B. G., P. W. A. Mansell, M. A. Epstein, and P. Clifford. 1971. An unusual virus in cultures from a human nasopharyngeal carcinoma. J. Natl. Cancer Inst. 46:299-307.[Medline]
    2. 2
  2. Baldwin, D. N., and M. L. Linial. 1998. The roles of Pol and Env in the assembly pathway of human foamy virus. J. Virol. 72:3658-3665.[Abstract/Free Full Text]
    3. 3
  3. Baunach, G., B. Maurer, H. Hahn, M. Kranz, and A. Rethwilm. 1993. Functional analysis of human foamy virus accessory reading frames. J. Virol. 67:5411-5418.[Abstract/Free Full Text]
    4. 4
  4. Bethune, J., F. Wieland, and J. Moelleken. 2006. COPI-mediated transport. J. Membr. Biol. 211:65-79.[CrossRef][Medline]
    5. 5
  5. Bieniasz, P. D. 2006. Late budding domains and host proteins in enveloped virus release. Virology 344:55-63.[CrossRef][Medline]
    6. 6
  6. Bruss, V. 2007. Hepatitis B virus morphogenesis. World J. Gastroenterol. 13:65-73.[Medline]
    7. 7
  7. Chen, B. J., and R. A. Lamb. 2008. Mechanisms for enveloped virus budding: can some viruses do without an ESCRT? Virology 372:221-232.[Medline]
    8. 8
  8. Demirov, D. G., A. Ono, J. M. Orenstein, and E. O. Freed. 2002. Overexpression of the N-terminal domain of TSG101 inhibits HIV-1 budding by blocking late domain function. Proc. Natl. Acad. Sci. USA 99:955-960.[Abstract/Free Full Text]
    9. 9
  9. Dubois, M. F., C. Pourcel, S. Rousset, C. Chany, and P. Tiollais. 1980. Excretion of hepatitis B surface antigen particles from mouse cells transformed with cloned viral DNA. Proc. Natl. Acad. Sci. USA 77:4549-4553.[Abstract/Free Full Text]
    10. 10
  10. DuBridge, R. B., P. Tang, H. C. Hsia, P. M. Leong, J. H. Miller, and M. P. Calos. 1987. Analysis of mutation in human cells by using an Epstein-Barr virus shuttle system. Mol. Cell. Biol. 7:379-387.[Abstract/Free Full Text]
    11. 11
  11. Duda, A., A. Stange, D. Luftenegger, N. Stanke, D. Westphal, T. Pietschmann, S. W. Eastman, M. L. Linial, A. Rethwilm, and D. Lindemann. 2004. Prototype foamy virus envelope glycoprotein leader peptide processing is mediated by a furin-like cellular protease, but cleavage is not essential for viral infectivity. J. Virol. 78:13865-13870.[Abstract/Free Full Text]
    12. 12
  12. Fischer, N., M. Heinkelein, D. Lindemann, J. Enssle, C. Baum, E. Werder, H. Zentgraf, J. G. Müller, and A. Rethwilm. 1998. Foamy virus particle formation. J. Virol. 72:1610-1615.[Abstract/Free Full Text]
    13. 13
  13. Garrus, J. E., U. K. von Schwedler, O. W. Pornillos, S. G. Morham, K. H. Zavitz, H. E. Wang, D. A. Wettstein, K. M. Stray, M. Cote, R. L. Rich, D. G. Myszka, and W. I. Sundquist. 2001. Tsg101 and the vacuolar protein sorting pathway are essential for HIV-1 budding. Cell 107:55-65.[CrossRef][Medline]
    14. 14
  14. Geiselhart, V., P. Bastone, T. Kempf, M. Schnolzer, and M. Löchelt. 2004. Furin-mediated cleavage of the feline foamy virus Env leader protein. J. Virol. 78:13573-13581.[Abstract/Free Full Text]
    15. 15
  15. Goepfert, P. A., K. L. Shaw, G. D. Ritter, Jr., and M. J. Mulligan. 1997. A sorting motif localizes the foamy virus glycoprotein to the endoplasmic reticulum. J. Virol. 71:778-784.[Abstract]
    16. 16
  16. Goff, S. P. 2007. Host factors exploited by retroviruses. Nat. Rev. Microbiol. 5:253-263.[CrossRef][Medline]
    17. 17
  17. Goila-Gaur, R., D. G. Demirov, J. M. Orenstein, A. Ono, and E. O. Freed. 2003. Defects in human immunodeficiency virus budding and endosomal sorting induced by TSG101 overexpression. J. Virol. 77:6507-6519.[Abstract/Free Full Text]
    18. 18
  18. Gottwein, E., S. Jäger, A. Habermann, and H. G. Kräusslich. 2006. Cumulative mutations of ubiquitin acceptor sites in human immunodeficiency virus type 1 Gag cause a late budding defect. J. Virol. 80:6267-6275.[Abstract/Free Full Text]
    19. 19
  19. Hartlieb, B., and W. Weissenhorn. 2006. Filovirus assembly and budding. Virology 344:64-70.[CrossRef][Medline]
    20. 20
  20. Harty, R. N., J. Paragas, M. Sudol, and P. Palese. 1999. A proline-rich motif within the matrix protein of vesicular stomatitis virus and rabies virus interacts with WW domains of cellular proteins: implications for viral budding. J. Virol. 73:2921-2929.[Abstract/Free Full Text]
    21. 21
  21. Heinkelein, M., M. Dressler, G. Jarmy, M. Rammling, H. Imrich, J. Thurow, D. Lindemann, and A. Rethwilm. 2002. Improved primate foamy virus vectors and packaging constructs. J. Virol. 76:3774-3783.[Abstract/Free Full Text]
    22. 22
  22. Herchenröder, O., R. Renne, D. Loncar, E. K. Cobb, K. K. Murthy, J. Schneider, A. Mergia, and P. A. Luciw. 1994. Isolation, cloning, and sequencing of simian foamy viruses from chimpanzees (SFVcpz): high homology to human foamy virus (HFV). Virology 201:187-199.[CrossRef][Medline]
    23. 23
  23. Hicke, L., and H. Riezman. 1996. Ubiquitination of a yeast plasma membrane receptor signals its ligand-stimulated endocytosis. Cell 84:277-287.[CrossRef][Medline]
    24. 24
  24. Lambert, C., T. Doring, and R. Prange. 2007. Hepatitis B virus maturation is sensitive to functional inhibition of ESCRT-III, Vps4, and {gamma}2-adaptin. J. Virol. 81:9050-9060.[Abstract/Free Full Text]
    25. 25
  25. Lindemann, D., M. Bock, M. Schweizer, and A. Rethwilm. 1997. Efficient pseudotyping of murine leukemia virus particles with chimeric human foamy virus envelope proteins. J. Virol. 71:4815-4820.[Abstract]
    26. 26
  26. Lindemann, D., and P. A. Goepfert. 2003. The foamy virus envelope glycoproteins. Curr. Top. Microbiol. Immunol. 277:111-129.[Medline]
    27. 27
  27. Lindemann, D., T. Pietschmann, M. Picard-Maureau, A. Berg, M. Heinkelein, J. Thurow, P. Knaus, H. Zentgraf, and A. Rethwilm. 2001. A particle-associated glycoprotein signal peptide essential for virus maturation and infectivity. J. Virol. 75:5762-5771.[Abstract/Free Full Text]
    28. 28
  28. Lindemann, D., and A. Rethwilm. 1998. Characterization of a human foamy virus 170-kilodalton Env-Bet fusion protein generated by alternative splicing. J. Virol. 72:4088-4094.[Abstract/Free Full Text]
    29. 29
  29. Lippincott-Schwartz, J., and W. Liu. 2006. Insights into COPI coat assembly and function in living cells. Trends Cell Biol. 16:e1-e4.[CrossRef][Medline]
    30. 30
  30. Lüftenegger, D., M. Picard-Maureau, N. Stanke, A. Rethwilm, and D. Lindemann. 2005. Analysis and function of prototype foamy virus envelope N glycosylation. J. Virol. 79:7664-7672.[Abstract/Free Full Text]
    31. 31
  31. Martin-Serrano, J. 2007. The role of ubiquitin in retroviral egress. Traffic 8:1297-1303.[CrossRef][Medline]
    32. 32
  32. Martin-Serrano, J., S. W. Eastman, W. Chung, and P. D. Bieniasz. 2005. HECT ubiquitin ligases link viral and cellular PPXY motifs to the vacuolar protein-sorting pathway. J. Cell Biol. 168:89-101.[Abstract/Free Full Text]
    33. 33
  33. Martin-Serrano, J., A. Yarovoy, D. Perez-Caballero, and P. D. Bieniasz. 2003. Divergent retroviral late-budding domains recruit vacuolar protein sorting factors by using alternative adaptor proteins. Proc. Natl. Acad. Sci. USA 100:12414-12419.[Abstract/Free Full Text]
    34. 34
  34. Morita, E., and W. I. Sundquist. 2004. Retrovirus budding. Annu. Rev. Cell Dev. Biol. 20:395-425.[CrossRef][Medline]
    35. 35
  35. Muziol, T., E. Pineda-Molina, R. B. Ravelli, A. Zamborlini, Y. Usami, H. Gottlinger, and W. Weissenhorn. 2006. Structural basis for budding by the ESCRT-III factor CHMP3. Dev. Cell 10:821-830.[CrossRef][Medline]
    36. 36
  36. Noda, T., H. Sagara, E. Suzuki, A. Takada, H. Kida, and Y. Kawaoka. 2002. Ebola virus VP40 drives the formation of virus-like filamentous particles along with GP. J. Virol. 76:4855-4865.[Abstract/Free Full Text]
    37. 37
  37. Ott, D. E., L. V. Coren, T. D. Copeland, B. P. Kane, D. G. Johnson, R. C. Sowder II, Y. Yoshinaka, S. Oroszlan, L. O. Arthur, and L. E. Henderson. 1998. Ubiquitin is covalently attached to the p6Gag proteins of human immunodeficiency virus type 1 and simian immunodeficiency virus and to the p12Gag protein of Moloney murine leukemia virus. J. Virol. 72:2962-2968.[Abstract/Free Full Text]
    38. 38
  38. Patton, G. S., S. A. Morris, W. Chung, P. D. Bieniasz, and M. O. McClure. 2005. Identification of domains in Gag important for prototypic foamy virus egress. J. Virol. 79:6392-6399.[Abstract/Free Full Text]
    39. 39
  39. Pfrepper, K. I., M. Löchelt, H. R. Rackwitz, M. Schnolzer, H. Heid, and R. M. Flügel. 1999. Molecular characterization of proteolytic processing of the Gag proteins of human spumavirus. J. Virol. 73:7907-7911.[Abstract/Free Full Text]
    40. 40
  40. Pietschmann, T., M. Heinkelein, M. Heldmann, H. Zentgraf, A. Rethwilm, and D. Lindemann. 1999. Foamy virus capsids require the cognate envelope protein for particle export. J. Virol. 73:2613-2621.[Abstract/Free Full Text]
    41. 41
  41. Pietschmann, T., H. Zentgraf, A. Rethwilm, and D. Lindemann. 2000. An evolutionarily conserved positively charged amino acid in the putative membrane-spanning domain of the foamy virus envelope protein controls fusion activity. J. Virol. 74:4474-4482.[Abstract/Free Full Text]
    42. 42
  42. Putterman, D., R. B. Pepinsky, and V. M. Vogt. 1990. Ubiquitin in avian leukosis virus particles. Virology 176:633-637.[CrossRef][Medline]
    43. 43
  43. Radtke, K., K. Dohner, and B. Sodeik. 2006. Viral interactions with the cytoskeleton: a hitchhiker's guide to the cell. Cell. Microbiol. 8:387-400.[CrossRef][Medline]
    44. 44
  44. Rasheed, S., W. A. Nelson-Rees, E. M. Toth, P. Arnstein, and M. B. Gardner. 1974. Characterization of a newly derived human sarcoma cell line (HT-1080). Cancer 33:1027-1033.[CrossRef][Medline]
    45. 45
  45. Rethwilm, A. 2003. The replication strategy of foamy viruses. Curr. Top. Microbiol. Immunol. 277:1-26.[Medline]
    46. 46
  46. Shaw, K. L., D. Lindemann, M. J. Mulligan, and P. A. Goepfert. 2003. Foamy virus envelope glycoprotein is sufficient for particle budding and release. J. Virol. 77:2338-2348.[Abstract/Free Full Text]
    47. 47
  47. Shehu-Xhilaga, M., S. Ablan, D. G. Demirov, C. Chen, R. C. Montelaro, and E. O. Freed. 2004. Late domain-dependent inhibition of equine infectious anemia virus budding. J. Virol. 78:724-732.[Abstract/Free Full Text]
    48. 48
  48. Stange, A., I. Mannigel, K. Peters, M. Heinkelein, N. Stanke, M. Cartellieri, H. Göttlinger, A. Rethwilm, H. Zentgraf, and D. Lindemann. 2005. Characterization of prototype foamy virus Gag late assembly domain motifs and their role in particle egress and infectivity. J. Virol. 79:5466-5476.[Abstract/Free Full Text]
    49. 49
  49. Stanke, N., A. Stange, D. Lüftenegger, H. Zentgraf, and D. Lindemann. 2005. Ubiquitination of the prototype foamy virus envelope glycoprotein leader peptide regulates subviral particle release. J. Virol. 79:15074-15083.[Abstract/Free Full Text]
    50. 50
  50. Sun, Z., J. Pan, W. X. Hope, S. N. Cohen, and S. P. Balk. 1999. Tumor susceptibility gene 101 protein represses androgen receptor transactivation and interacts with p300. Cancer 86:689-696.[CrossRef][Medline]
    51. 51
  51. Timmins, J., G. Schoehn, S. Ricard-Blum, S. Scianimanico, T. Vernet, R. W. Ruigrok, and W. Weissenhorn. 2003. Ebola virus matrix protein VP40 interaction with human cellular factors Tsg101 and Nedd4. J. Mol. Biol. 326:493-502.[CrossRef][Medline]
    52. 52
  52. Watanabe, T., E. M. Sorensen, A. Naito, M. Schott, S. Kim, and P. Ahlquist. 2007. Involvement of host cellular multivesicular body functions in hepatitis B virus budding. Proc. Natl. Acad. Sci. USA 104:10205-10210.[Abstract/Free Full Text]
    53. 53
  53. Whitt, M. A., L. Chong, and J. K. Rose. 1989. Glycoprotein cytoplasmic domain sequences required for rescue of a vesicular stomatitis virus glycoprotein mutant. J. Virol. 63:3569-3578.[Abstract/Free Full Text]
    54. 54
  54. Wilk, T., V. Geiselhart, M. Frech, S. D. Fuller, R. M. Flügel, and M. Löchelt. 2001. Specific interaction of a novel foamy virus env leader protein with the N-terminal Gag domain. J. Virol. 75:7995-8007.[Abstract/Free Full Text]
    55. 55
  55. Yu, S. F., S. W. Eastman, and M. L. Linial. 2006. Foamy virus capsid assembly occurs at a pericentriolar region through a cytoplasmic targeting/retention signal in Gag. Traffic 7:966-977.[CrossRef][Medline]
    56. 56
  56. Zhadina, M., M. O. McClure, M. C. Johnson, and P. D. Bieniasz. 2007. Ubiquitin-dependent virus particle budding without viral protein ubiquitination. Proc. Natl. Acad. Sci. USA 104:20031-20036.[Abstract/Free Full Text]

Journal of Virology, October 2008, p. 9858-9869, Vol. 82, No. 20
0022-538X/08/$08.00+0     doi:10.1128/JVI.00949-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stange, A.
Right arrow Articles by Lindemann, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stange, A.
Right arrow Articles by Lindemann, D.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»