Previous Article | Next Article ![]()
Journal of Virology, January 2005, p. 791-799, Vol. 79, No. 2
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.2.791-799.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Institute of Molecular Biology, Friedrich-Loeffler-Institutes, Federal Research Centre for Virus Diseases of Animals, Greifswald-Insel Riems, Germany
Received 9 June 2004/ Accepted 30 August 2004
|
|
|---|
) as an intervening polypeptide. The resulting gB variants rescued gB-negative BHV-1 mutants, the resulting BHV-1 recombinants were fully infectious, and infected cells secreted biologically active GFP and boIFN-
, respectively. In contrast to the gB2Fu and gB2FuGFP precursor molecules, which were efficiently cleaved at both furin sites, the majority of pgB2FuIFN-
was not cleaved at the site between the amino-terminal (NH2) subunit and boIFN-
, whereas cleavage at the newly introduced site was normal. This resulted in virus particles that also contain the NH2-subunit/boIFN-
fusion protein within their envelopes. Our results demonstrate that BHV-1 gB can be used as a transporter for peptides and proteins which could be important for development of novel vaccines. In addition, the general principle might be useful for other applications, e.g., in gene therapy and also in nonviral systems. |
|
|---|
Among the furin-cleaved viral fusion proteins, the F proteins of bovine RSV (BRSV) and human RSV share a so-far unique feature. The respective precursor molecules are cleaved at two furin recognition sites, resulting in release of a 27-amino-acid intervening peptide that is glycosylated (27, 28). For BRSV it has been shown that the intervening peptide is secreted as a bioactive peptide of the tachykinin family after further modification (29). We demonstrated recently that the intervening peptide of the BRSV F protein can be replaced by bovine interleukin 2 (boIL-2) and boIL-4, which are cleaved out of the precursor molecules and are secreted into the culture medium of cells infected with the respective BRSV recombinants (14).
We report here the introduction of a second furin cleavage site and intervening polypeptides into gB of BHV-1 and the isolation of viable BHV-1 recombinants that express the green fluorescent protein (GFP) and bovine alpha interferon (boIFN-
) as furin-excisable proteins which are secreted as biologically active proteins from infected cells.
|
|
|---|
![]() View larger version (27K): [in a new window] |
FIG. 1. Construction of the gB mutants. (A) Schematic representation of the BHV-1 genome. Unique long (UL) and unique short (US) segments are indicated, as are the internal and terminal repeat sequences (IR and TR, respectively). Arrows indicate the isomerization of the US during replication. (B) Position of BamHI fragments (3) and localization of the gB gene (hatched area) according to the complete genome sequence (GenBank accession number AJ004801). (C) Scheme of the gB ORF. Sequences encoding the signal peptide (sig), the amino-terminal subunit, the carboxy-terminal subunit which contains the membrane anchor domain, and the furin cleavage site (FCS1) are indicated. The double-headed arrow indicates the sequence deleted in gB lacZ+ BHV-1. The fragment between the ApaI cleavage sites (underlined) which encompass the codons for FCS1 (boldface) is expanded, and the nucleotide sequence is given. The amino acid sequence for FCS1 is indicated in three-letter code. (D) Nucleotide sequence introduced after the codon for Arg504 within gB2Fu. The deduced amino acid sequence is given in three-letter code, and the position of the second furin cleavage site (FCS2) is marked. (E) Amino acid changes within mutants gB2FuGFP and gB2FuIFN- . Additional amino acids are in boldface.
|
For insertion of the coding sequence for mature boIFN-
, the respective 167 codons were isolated by PCR from plasmid pieIFN
syn, which contains an artificial ORF encoding boIFN-
subtype C (25) that has been assembled from synthetic oligonucleotides by using the codon preference of glycoprotein B of bovine herpesvirus 1 (unpublished; accession number AJ784928). This ORF had been integrated into plasmid pie1cas, a derivative of plasmid pROMie (16). PCR amplification of the fragment encoding mature boIFNa was performed with Pfx polymerase as recommended by the supplier. The primers, which also provided adapter sequences for in-frame insertion into gB, were matIFNa+ (TAAGGGCCCGCGACGTGCGCGCCGATGCCACCTGCCGCACACGCACAGC) and matIFNa (TAAGAGCTCGTCCTTGCGGCGGAAGCTCTCCTGCAG). The amplification product purified after 1 cycle of 95°C for 2 min and 33 cycles of 95°C for 30 s and 55 and 72°C for 1 min each was blunt ended with Klenow polymerase, cleaved with SacI, and integrated into the EcoRI-cleaved, blunt-ended, and SacI-cleaved plasmid pSP2Fu, resulting in plasmid pSP2FuIFN
. Plasmid pSP2FuGFP was constructed accordingly by using gB2FuGFP+ (TAAGGGCCCGCGACGTGCGCGCCGAATGGTGAGCAAGGGCGAGGAGCTG) and gB2FuGFP (TAGAAGCTTGAGCTCCTTGTACAGCTCGTCCATGCC) as primers and pEGFP-N1 (Clontech, Palo Alto, Calif.) as the template. Primer sequences were derived from the sequence provided with the plasmid.
The ApaI fragments contained in pSP2Fu, pSP2FuGFP, and pSP2FuIFN
were subsequently used to replace the ApaI fragment encompassing the coding sequence for the furin cleavage site (Fig. 1) within pSPgB, which contains a 3.8-kbp NotI fragment encompassing the entire gB gene. The resulting plasmids were named pSPgB2Fu, pSPgB2FuGFP, and pSPgB2FuIFN-
.
Construction of BHV-1 recombinants. MDBK cells were cotransfected with 5 µg of recombinant plasmid and 1 µg of purified gB lacZ+ BHV-1 or gB BHV-1/GKD DNA as described previously (8). Virus progeny from the culture supernatants was titrated on MDBK cells. Cultures were incubated under a 0.6% agarose overlay containing Bluo-Gal (8) until plaques appeared. Infected cells were isolated by aspiration, resuspended in culture medium, frozen at 70°C, and further plaque purified to homogeneity.
Immunoprecipitation. Cells were infected and proteins were metabolically labeled with [35S]methionine and [35S]cysteine as indicated. Immunoprecipitation of proteins from cell lysates were performed as described previously (8), using a monospecific rabbit serum directed against a synthetic peptide representing amino acids Leu910 to Asn928 of the carboxy terminus of BHV-1 gB (anti-gB serum). Labeled proteins were visualized after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) by fluorography or with a Fuji FLA3000 fluorescence scanner and Aida two-dimensional (2-D) gel evaluation software.
Western blotting. Proteins were separated by SDS-10% PAGE, transferred to nitrocellulose, and probed with anti-GFP monoclonal antibodies (MAbs) (Roche, Mannheim, Germany), using the Supersignal West Pico chemiluminescence kit (Pierce, Rockford, Ill.) as recommended by the supplier.
Determination of boIFN-
activity.
Secretion of biologically active boIFN-
into the cell culture medium was analyzed by use of a vesicular stomatitis virus (VSV) plaque reduction assay. Supernatants from KOP/R cells infected with BHV-1/gB2FuIFN-
or BHV-1/GKD as a control were harvested 24 h after infection. The supernatants were serially diluted in normal cell culture medium after ultracentrifugation and sterilization by UV light and were added to KOP/R cells in six-well plates. Cultures were incubated for 24 h at 37°C and then infected with approximately 100 PFU of VSV. Supernatants were removed at 1 h postinfection (p.i.), and semisolid methylcellulose-containing medium was added. Plaques were counted after 24 h of incubation at 37°C. Comparison of boIFN-
activities with or without UV light treatment revealed no differences between the samples (not shown).
Indirect immunofluorescence assay. Cells were fixed with 3% paraformaldehyde in phosphate-buffered saline for 20 min, subjected to membrane permeabilization with 0.2% Triton X-100, and sequentially incubated with gB-specific MAb 42/18/7 (8) and tetramethyl rhodamine isocyanate-conjugated rabbit anti-mouse immunoglobulin G (Dianova).
Analyses of cell culture characteristics. For single-step growth curves, MDBK cultures were infected with 10 PFU per cell. At 2 h p.i., cells were incubated for 2 min with low-pH citrate buffer (40 mM citric acid, 10 mM KCl, 135 mM NaCl [pH 3.0]) to inactivate extracellular virions (8). Cells were washed twice with cell culture medium and incubated until the times indicated, when supernatants and cells were harvested and stored at 70°C. Cells were incubated for 2 min with low-pH citrate buffer before harvest. Serial dilutions were titrated on MDBK cells, and cultures were incubated under semisolid medium containing methylcellulose. Plaques were counted after 2 days.
(i) Determination of plaque diameters. MDBK cells were infected with diluted virus stocks and incubated under semisolid medium containing methylcellulose for 2 days. Diameters of 100 randomly selected plaques were determined under a microscope by using a graduated ocular.
(ii) Penetration kinetics. MDBK cells were precooled at 4°C for 30 min and further incubated at 4°C for 2 h after addition of about 200 PFU to allow adsorption. Cultures were then shifted to 37°C, and extracellular virions were inactivated at the indicated times by incubation of the monolayers with low-pH citrate buffer for 2 min. Cells were washed twice with cell culture medium and incubated under semisolid medium containing methylcellulose. Plaques were counted after 2 days. The plaque count of untreated cultures was set as 100% penetration.
|
|
|---|
![]() View larger version (51K): [in a new window] |
FIG. 2. Maturation of wild-type gB and gB2Fu. KOP/R cells were infected with BHV-1/gBrev and BHV-1/gB2Fu at 10 PFU/cell in the absence (A) or presence (B) of the furin inhibitor dec-RVKR-CMK (100 µM), and proteins were labeled with [35S]methionine-[35S]cysteine at 6 h p.i. for 30 min, washed, and further incubated with normal cell culture medium as indicated. Cell lysates were incubated with anti-gB serum, and immunoprecipitated proteins were separated by SDS-7.5% PAGE. The positions of the precursor form (pgB), the uncleaved form (gB), and the NH2 and COOH subunits of wild-type gB are shown on the left. The corresponding apparent molecular masses are indicated on the right.
|
![]() View larger version (53K): [in a new window] |
FIG. 3. Expression of GFP by BHV-1/gB2FuGFP. MDBK cells were infected with diluted stocks of BHV-1/gBrev and BHV-1/2FuGFP and incubated under methylcellulose-containing medium for 2 days. Cells were fixed and stained with gB-specific MAb 42/18/7 and tetramethyl rhodamine isocyanate-conjugated anti-mouse secondary antibody. GFP autofluorescence and indirect immunofluorescence were visualized by using appropriate filters and photographed with an Olympus digital camera.
|
![]() View larger version (32K): [in a new window] |
FIG. 4. Maturation of gB2FuGFP. Pulse-chase analysis was performed with BHV-1/gBrev and BHV-1/gB2FuGFP as described in the legend to Fig. 2A, with the exception that the sample marked 90* was incubated in presence of 100 µM dec-RVKR-CMK during the labeling and chase period to inhibit furin cleavage.
|
![]() View larger version (36K): [in a new window] |
FIG. 5. Secretion of GFP from BHV-1/gB2FuGFP-infected cells. (A) KOP/R cells were infected with BHV-1/gB2FuGFP (lanes 1) and BHV-1/gBrev (lanes 2). Proteins from cells (left panel) and culture media (right panel), harvested at 20 h p.i., were separated by SDS-12.5% PAGE and analyzed by Western blotting with anti-GFP MAbs. Positions of marker proteins are indicated on the left. (B) Aliquots of the samples analyzed in panel A were transferred to 96-well cell culture plates, and autofluorescence was visualized with a Fuji FLA-3000 fluorescence scanner. Relative fluorescence intensities (RFI) were determined by using the AIDA 2D evaluation software.
|
as a furin-excisable protein into gB (Fig. 1) to test whether this approach is also applicable for transport and export of biologically active secreted glycoproteins. Generation and isolation of BHV-1/gB2FuIFN-
was done as described above with plasmid pSPgB2FuIFN-
and purified DNA of gB BHV-1/GKD. BHV-1/GKD was chosen as the progenitor strain to provide the same genetic background for animal experiments as for the previously tested bovine cytokine-expressing BHV-1 recombinants (13). Comparison of the intracellular transport of BHV-1/GKD gB and gB2FuIFN-
by pulse-chase experiments (Fig. 6) showed that transport of the two precursor molecules into the TGN was comparable, indicating that the IFN-
sequence did not negatively influence modification and processing within the endoplasmic reticulum and the Golgi. The kinetics of the generation of the 55-kDa subunit, which was visible already after the pulse period in cells infected with BHV-1/GKD and BHV-1/gB2FuIFN-
, also were similar (Fig. 6), indicating that furin cleavage at the introduced FCS2 was not significantly affected. In contrast, the 72-kDa amino-terminal gB subunit of gB2FuIFN-
was not detected even after 90 min of chase. A slower-migrating form with an apparent molecular mass of about 90 kDa was generated instead, with kinetics comparable to those for the formation of the 72-kDa subunit of BHV-1/GKD. Additional proteins with apparent molecular masses of about 140 and 150 kDa and with constant intensities probably represent nonspecific coprecipitates.
![]() View larger version (62K): [in a new window] |
FIG. 6. Maturation of gB2FuIFN- . Pulse-chase analysis was performed with BHV-1/GKD and BHV-1/gB2FuIFN- as described in the legend to Fig. 2A.
|
contained the correct sequence for FCS1, mature boIFN-
, and FCS2, we conclude that furin fails to cleave gB2FuIFN-
at the authentic FCS1 and that the 90-kDa protein, which is also found in purified virions (Fig. 7), is a fusion protein between the NH2 subunit and boIFN-
. Labeling of infected-cell proteins for 18 h followed by immunoprecipitation revealed the presence of a small amount of the 72-kDa NH2 subunit of gB2FuIFN-
, indicating that cleavage at FCS1 is not completely abolished (data not shown). Analysis of the gB forms associated with virus particles by immunoprecipitation showed that BHV-1/gB2FuIFN-
virions contain the uncleaved gB, the COOH subunit, the NH2-subunit/IFN-
fusion protein, and small amounts of the 72-kDa NH2 subunit (Fig. 7, lane 2). Comparison of the intensities of wild-type gB and gB2FuIFN-
obtained from virions shows that noticeably less gB was precipitated from BHV-1/gB2FuIFN-
virions than from BHV-1/GKD virions (Fig. 7, lanes 1 and 2). A comparable result was obtained when MAb 21/3/3 (8) was used to precipitate the 72-kDa BHV-1 glycoprotein D (Fig. 7, lanes 3 and 4), indicating that fewer virus particles were released from BHV-1/gB2FuIFN-
infected cells (see also Fig. 8A).
![]() View larger version (60K): [in a new window] |
FIG. 7. Incorporation of the gB-NH2/IFN- fusion protein into virions. KOP/R cells were infected with BHV-1/GKD (lanes 1 and 3) and BHV-1/gB2FuIFN- (lanes 2 and 4) at 5 PFU/cell and incubated with [35S]methionine-[35S]cysteine from 4 to 40 h p.i. Cell culture supernatants were clarified by low-speed centrifugation, and virions were pelleted by ultracentrifugation. Labeled proteins from virions were immunoprecipitated with anti-gB serum (lanes 1 and 2) or BHV-1 gD-specific MAb 21/3/3 (lanes 3 and 4) and analyzed by SDS-10% PAGE.
|
![]() View larger version (15K): [in a new window] |
FIG. 8. Cell culture characteristics of BHV-1/gB2FuIFN- . A) Growth curves. MDBK cells were infected with BHV-1/GKD (stars) and BHV-1/gB2FuIFN- (squares) at 10 PFU per cell. Cells were treated with low-pH buffer to inactivate nonpentrated virions at 2 h p.i. Cells (closed symbols) and supernatants (open symbols) were harvested at the times indicated and titrated on MDBK cells. Cultures were overlaid with methylcellulose-containing medium, and plaques were counted 2 days later. Arrows indicate the time point when extracellular BHV-1/GKD (open arrow) or BHV-1/gB2FuIFN- (solid arrow) infectivity surpasses the respective intracellular infectivity. B) Plaque diameter determination. MDBK cells (solid bars) and KOP/R cells (open bars) were infected with appropriate dilutions of the indicated viruses and incubated for 2 days under methylcellulose-containing medium. Diameters of 100 plaques from each virus were measured by using a graduated ocular. Average diameters are shown in arbitrary units.
|
, single-step growth was analyzed on MDBK cells. Cultures were infected with 10 PFU per cell, nonpenetrated virions were inactivated by low-pH treatment at 2 h p.i., and cells and culture supernatants were harvested at the times indicated in Fig. 8A. In contrast to cell-associated infectivity, which showed kinetics comparable to those for BHV-1/GKD, the start of release of infectious BHV-1/gB2FuIFN-
was delayed for about 4 h. In addition, it took roughly 22 h until extracellular infectivity surpassed the titer of intracellular virions, whereas the parental strain BHV-1/GKD needed only approximately 12 h to reach this point. In addition, the final titers reached by BHV-1/gB2FuIFN-
were about one order of magnitude lower than those released from cells infected with BHV-1/GKD or wild-type BHV-1 strains (8, 16).
Direct cell-to-cell spread of BHV-1/gB2FuIFN-
also appeared to be hindered, because the sizes of plaques formed under methylcellulose-containing semisolid medium on MDBK and PT cells were only around 50% of those achieved by BHV-1/GKD (Fig. 8B). These results might indicate that the incomplete cleavage of the gB2FuIFN-
precursor interferes with release and the gB function for direct spreading.
To assess whether the cleavage at both furin cleavage sites results in release of biologically active boIFN-
, the medium of cells infected for 24 h with BHV-1/GKD or BHV-1/gB2FuIFN-
was tested by a VSV plaque reduction assay for the presence of antiviral activity in comparison to a recombinant IFN-
standard (11) containing 2 x 106 U/ml (kindly provided by Alfred Metzler, Zürich, Switzerland) (19). The results are shown in Fig. 9. In contrast to the medium of BHV-1/GKD-infected cells, which did not contain detectable antiviral activity, the medium of BHV-1/gB2FuIFN-
-infected cells inhibited VSV plaque formation in a dose-dependent manner and contained approximately 5 x 104 U of boIFN-
per ml. Thus, significant boIFN-
activity was released from BHV-1/gB2FuIFN-
-infected cells although only a minor fraction of the precursor molecules were correctly cleaved, demonstrating that biologically active glycoproteins also can be utilized in this novel approach for the expression of secreted polypeptides.
![]() View larger version (13K): [in a new window] |
FIG. 9. Secretion of biologically active IFN- from cells infected with BHV-1/gB2FuIFN- . Culture media from KOP/R cells infected with BHV-1/GKD (squares) or BHV-1/gB2FuIFN- (stars) were clarified by ultracentrifugation and sterilized by UV irradiation. The supernatants and a standard containing 2 x 106 U of recombinant IFN- per ml (circles) were serially diluted and added to confluent KOP/R cells. Approximately 100 PFU of VSV was added to each culture 24 h later. At 1 h p.i. the inoculum was replaced by semisolid methylcellulose-containing medium, and plaques were counted at 24 h p.i.
|
|
|
|---|
That the amino acid sequence between the two furin cleavage sites may influence intracellular transport was indicated by analysis of the intracellular transport of gB2FuGFP, which was transported to the Golgi clearly less efficiently than wild-type pgB or pgB2Fu. The reduced transport consequently resulted in a delayed and less abundant appearance of the NH2 and COOH subunits. Furin cleavage of pgB2FuGFP appeared to be largely unimpaired. Glycosylation of gB also seemed to be generally unaffected. Although not analyzed in detail, the identical migration behaviors of the NH2 and COOH subunits of gB expressed by BHV-1/gBrev and BHV-1/gB2FuGFP suggest that N and O glycosylations are comparably realized. In addition, coprecipitation of the NH2 and COOH subunits by the anti-gB serum demonstrated that even with the uncleaved IFN-
present, gB is still expressed as a disulfide-linked dimer. The retarded transport of pgB2FuGFP, however, had no detectable effect on the biological properties of BHV-1/gB2FuGFP in cell culture. The recombinant grew to the same final titers with kinetics comparable to those of BHV-1/gBrev and entered the cells and spread directly from cell to cell as efficiently as BHV-1/gBrev. Analysis of the fate of the intervening GFP revealed that biologically active GFP accumulated in the medium of BHV-1/gB2FuGFP-infected cells, indicating that it was efficiently secreted after removal from uncleaved gB. Thus, although the insertion of 246 amino acids into pgB hinders intracellular transport, it apparently does not interfere with the overall processing leading to functional gB.
An influence of the intervening polypeptide on furin cleavage was observed for gB2FuIFN-
, which was transported to the trans-Golgi network as efficiently as wild-type gB. Furin cleavage, however, occurred predominantly only at FCS2, resulting in an NH2 subunit the size of a fusion protein between the authentic NH2 subunit and mature IFN-
. Long-term labeling and analysis of the gB content of virions showed that cleavage also occurred to a low extent at FCS1. We assume that the amino acid sequence downstream from FCS1 interferes with the cleavage activity of furin at this site. The structural requirements to achieve full cleavage need to be identified. Despite the inefficient removal of IFN-
, we found a substantial amount of IFN-
activity in the culture medium of cells infected with BHV-1/gB2FuIFN-
, which was close to the activity reached with IFN-
-expressing replication-defective adenoviruses (4). Thus, these results demonstrate that the engineered gB can also be used to transport secreted glycoproteins, and we speculate that if full cleavage at FCS1 can be achieved, significantly larger amounts of IFN-
will be produced by cells with accordingly engineered BHV-1/gB2FuIFN-
. This assumption is supported by preliminary results obtained with a BHV-1 recombinant that secretes approximately fivefold-higher levels of mature boIL-2 as furin-excisable protein than does BHV-1/GKD-boIL-2 (13), which expresses boIL-2 classically via an expression cassette (to be published elsewhere). Analysis of the cell culture properties of BHV-1/gB2FuIFN-
revealed that the size of the plaques formed by the recombinant was about 50% reduced, that single-step growth was delayed, and that the virus yield was reduced about 90%, which might be caused by structural alterations of gB2FuIFN-
due to boIFN-
insertion and/or reduced cleavage at FCS1. However, these effects cannot be unequivocally attributed to a possible misfolding of gB2FuIFN-
, since preincubation of MDBK cells with 104 U of recombinant IFN-
results in a similar inhibition of wild-type BHV-1 (1, unpublished results).
In summary, our results have demonstrated that BHV-1 gB can be engineered to function as a transporter for secretion of heterologous proteins and glycoproteins into the extracellular space in biologically active forms. The examples presented here show that the amino acid sequence of an intervening polypeptide may influence transport or furin cleavage but in no case destroyed gB function for BHV-1 replication, proving that the integrity of the furin cleavage site domain of BHV-1 gB and the distance between the subsequent NH2 and COOH subunits are not important for overall folding and dimerization of gB within the endoplasmic reticulum. We assume that this novel protein expression approach may be of particular interest not only for development of new vaccines but also for gene therapy applications, especially when biologically active oligopeptides need to be secreted. We speculate that this strategy will be applicable for other furin-cleaved proteins and hypothesize that it can be also applied to proteins without furin cleavage sites.
, respectively, Axel Karger and Christian Sauer for 2-D evaluation and quantitation of GFP fluorescence, and Anne Brandenburg for excellent technical assistance. |
|
|---|
1 interferon on bovine herpesvirus type 1-induced respiratory disease. J. Gen. Virol. 66:2383-2394.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»