- Email: [email protected]
Virus-like Particles and E1AE4 Protein Expressed from the Human Papillomavirus Type 11 Bicistronic E1AE4AL1 Transcript
VIROLOGY
222, 43–50 (1996) 0396
ARTICLE NO.
Virus-like Particles and E1 O E4 Protein Expressed from the Human Papillomavirus Type 11 Bicistronic E1 O E4 O L1 Transcript DARRON R. BROWN,*,†,1 LINDA PRATT,* JANINE T. BRYAN,‡ KENNETH H. FIFE,*,‡,§ and KATHRIN JANSENØ Departments of *Medicine, ‡Microbiology and Immunology, and §Pathology, and †Richard L. Roudebush Veterans Administration Medical Center, Indiana University School of Medicine, Indianapolis, Indiana 46202; and ØDepartment of Virus and Cell Biology, Merck Research Laboratories, West Point, Pennsylvania 19486 Received March 5, 1996; accepted June 5, 1996 Detection of E1 O E4 protein in human papillomavirus (HPV 11)-infected tissue is tightly linked to detection of L1 major capsid protein. The only L1-containing transcript identified in HPV 11-infected tissue is the bicistronic E1 O E4 O L1 mRNA, potentially encoding both the E1 O E4 and the L1 proteins. It has not been established that these proteins can be expressed from the E1 O E4 O L1 transcript. The HPV 11 E1 O E4 O L1 sequence was cloned by reverse transcriptase polymerase chain reaction into the p1393 vector to produce recombinant baculoviruses. Immunoblots of recombinant baculovirus-infected Sf9 cell lysates demonstrated both the E1 O E4 and the L1 proteins. An ELISA was performed on infected Sf9 cells using a monoclonal antibody specific for nondenatured L1, demonstrating that 10 ng of native L1 protein was present per microgram of total nuclear protein. Electron microscopic analysis revealed 50- to 60-nm icosahedral virus-like particles. In vitro transcription/translation was performed using pSPORT constructs containing the E1 O E4 O L1 sequence or, as controls, monocistronic pSPORT-E1 O E4 or L1 constructs. The pSPORT-E1 O E4 O L1 construct produced the E1 O E4 and L1 proteins at a ratio of 17:1. For E1 O E4 protein, expression was greater from the pSPORT-E1 O E4 O L1 construct than from the monocistronic pSPORT-E1 O E4 construct. In contrast, more L1 protein was expressed from pSPORT-L1 than from pSPORT-E1 O E4 O L1. A mutant E1 O E4 O L1 construct containing no E1 O E4 start codon expressed L1 protein in amounts nearly equal to that expressed from the pSPORT-L1 construct. Addition of an antisense oligonucleotide directed at the E1 O E4 start codon region to in vitro reactions using pSPORT-E1 O E4 O L1 was associated with inhibition of E1 O E4 protein synthesis and increased translation of L1 protein. q 1996 Academic Press, Inc.
INTRODUCTION
of several spliced HPV 11 mRNAs. Rotenberg et al. (1989) provided direct evidence for the existence of several spliced HPV mRNAs, including the E1 O E4 O L1 bicistronic mRNA, by isolation of RNA from genital lesions and employment of reverse transcriptase polymerase chain reaction (RT PCR). It has not been established that the E1 O E4 O L1 transcript is a functional message. The goals of this study were (1) to clone the HPV 11 E1 O E4 O L1 transcript from an HPV 11-infected human tissue, (2) to verify that both potential ORFs can be expressed in eukaryotic cells, (3) to compare in vitro transcription and translation of the E1 O E4 O L1 sequence to that of control constructs containing monocistronic E1 O E4 and L1 sequences, and (4) to examine the mechanism of translation of the E1 O E4 O L1 transcript using antisense oligonucleotides. To accomplish these goals, the HPV 11 E1 O E4 O L1 sequence was cloned by RT PCR and expressed in Sf9 cells by a recombinant baculovirus. In addition, the E1 O E4 O L1 sequence was cloned into the pSPORT vector and expressed in a wheat germ extract coupled-transcription/translation system.
Human papillomaviruses (HPVs) potentially encode six or more early (E) and two late (L) gene products. Overall, the most abundant viral mRNA species detected in HPV 11-infected tissue is a spliced message encoding an 11kDa gene product containing the first 5 amino acids of the E1 open reading frame (ORF) fused to the last 85 amino acids of the E4 ORF (Nasseri et al., 1987). The resulting E1 O E4 gene product has been identified in HPV-infected cutaneous and genital tissues but the function of this protein has not been determined (Brown et al., 1988, 1991; Doorbar et al., 1986; Doorbar, 1991; Neary et al., 1987). Because the E1 O E4 protein is detected late in HPV 11 infection, and exclusively in differentiated cells that contain the L1 major capsid protein, the E1 O E4 protein might be considered a late gene product, although the genomic sequences reside in the early region. A viral transcript identified as the doubly spliced E1 O E4 O L1 mRNA potentially encodes both the E1 O E4 and the L1 proteins (Chow et al., 1987; Rotenberg et al., 1989). Chow et al. (1987) first demonstrated the existence
MATERIALS AND METHODS Cloning and expression of the E1 O E4 O L1 sequence An HPV 11-infected human foreskin implant grown in an athymic mouse for 12 weeks was used to isolate total
1 To whom correspondence and reprint requests should be addressed. Fax: (317) 274-1587.
0042-6822/96 $18.00
43
AID
VY 8042
/
6a1b$$$181
07-07-96 19:41:36
Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
vira
AP: Virology
44
BROWN ET AL.
RNA as previously described (Brown et al., 1994b). Firststrand cDNA synthesis for the RT PCR was performed for 50 min at 427 using SuperScript II (Gibco BRL, Gaithersburg, MD) and oligo(dT) as primer. PCR was performed with a 5* primer that spanned the E1/E4 splice (ATGGCGGACGATTCAG/CACTG) and a 3* primer (GTGTAGTGGGGTCAAACAGGGATG) in the L1 ORF beginning at nucleotide (nt) 6047, which is 3* to the unique XbaI site in HPV 11. The resulting 580-bp DNA fragment was cloned into the pCR3 vector (Invitrogen, San Diego, CA) to create pCR3-E1 O E4 O L1-6047. The complete L1 ORF was created by digesting a clone containing the XbaI/ SphI L1 fragment of HPV 11 (pATH-L1-11) (Brown et al., 1994b) with EcoRI and XbaI, and ligating this linear plasmid with the E1 O E4 O L1 fragment that had been purified from the pCR3-E1 O E4 O L1-6047 clone by digesting with EcoRI and XbaI. The resulting plasmid, pATH-E1 O E4 O L1, contained the E1 O E4 and L1 ORFs separated by 12 nt and the 3* untranslated region through nt 7928. The E1 O E4 O L1 cDNA fragment was then subcloned into the baculovirus expression vector p1393 (PharMingen, San Diego, CA). To test expression in a cell-free coupled-transcription/ translation system, the E1 O E4 O L1 sequence was removed from p1393 and subcloned into pSPORT (Gibco BRL) at the EcoRI site. To create the monocistronic control plasmids, the E1 O E4 and L1 ORFs were amplified by PCR. For the E1 O E4 ORF, the pATH-E1 O E4 plasmid was used as template (Brown et al., 1994b). The primers used were the 5* primer described above and the 3* primer sequence CTATAGGCGTAGCTGCACTGTGACAG. For L1, the 5* primer sequence was ATGTGGCGGCCTAGCGACAG, and the 3* primer was GGGCCCAGATCTAAGAAGGAAATATGTAGGGTGTGGG, which began at nt 7928 and included a 3* BglII site extension. DNA extracted from a condylomata acuminata lesion containing HPV 11 was used as template for the L1 PCR (Brown et al., 1994a). After PCR, the E1 O E4 and L1 DNA fragments were cloned into pCR3, then subcloned into pSPORT. As a control, a construct was made (pSPORT-XE1 O E4 O L1) lacking an E1 O E4 start codon. The pSPORT-XE1 O E4 O L1 construct was prepared by amplifying the cloned E1 O E4 O L1 sequence by PCR using a 5* primer containing an AAG in place of the ATG start codon. The resulting PCR product was cloned into pCR3, then subcloned into pSPORT. The pSPORT-XE1 O E4 O L1 construct was used in in vitro reactions as described below. The distance between the T7 promoter and the first start codon was the same in each of the pSPORT constructs. Expression of E1 O E4 O L1 in Sf9 cells
and used to infect Sf9 cells. Cells were harvested after 6 days of growth and pelleted by centrifugation at 1000 g. Proteins were separated by 15% SDS–PAGE then transferred to nitrocellulose (Towbin et al., 1979). Immunoblots were performed as previously described (Brown et al., 1991). Nitrocellulose membranes were probed with one of two polyclonal antisera: an antiserum raised against a bacterial-expressed HPV 11 trpE/E1 O E4 or against a trpE/L1 fusion protein (Brown et al., 1994b). Detection of virus-like particles (VLPs) Sf9 cell pellets were lysed with PBS plus 2% Triton X, and nuclei were isolated and concentrated by centrifugation at 1000 g, then 5000 g, followed by two washes in cold PBS. Nuclei were sonicated in PBS and clarified by centrifugation at 1000 g. The clarified supernate was assayed by ELISA and by electron microscopy. For the ELISA, the supernate was incubated in wells of Nunc Maxisorb plates (Nalge, Rochester, NY) for 12 hr at 47. Plates were washed once with TTBS (20 mM Tris–HCl, pH 7.6; 150 mM NaCl; 0.1% Tween 20), then incubated with TTBS plus 10% nonfat dry milk for 2 hr at room temperature. Plates were washed three times with TTBS before adding a monoclonal antibody to HPV 11 previously shown to specifically recognize nondenatured L1 protein (generously provided by Neil Christensen) in TTBS plus 10% nonfat dry milk (Christensen et al., 1994). After a 1-hr incubation at room temperature, plates were washed three times in TTBS. Goat anti-mouse IgG–alkaline phosphatase conjugate was added as secondary antibody at 1:1000 dilution in TTBS plus 10% nonfat dry milk and incubated for 1 hr at room temperature. Plates were washed three times, and p-nitrophenyl phosphate in diethanolamine buffer (Kirkegaard and Perry, Gaithersburg, MD) was added as substrate. Plates were incubated for 30 min at room temperature before stopping the color reactions with 3 M NaOH. The absorbance of each well was measured at 405 nm. The amount of nondenatured L1 protein was determined by endpoint dilution compared to mean absorbance readings of control wells containing known quantities of yeast-expressed, sucrose gradient-purified HPV 11 L1 VLPs (Hofmann et al., 1995). For electron microscopy a drop of clarified Sf9 cell nuclear supernate was placed on a 200-mesh carboncoated copper grid. A drop of 2% phosphotungstic acid, pH 7.0, was placed on the grid. After grids had air dried, microscopy was performed with a JEOL 100CX transmission electron microscope at an accelerating voltage of 100 kV. In vitro transcription and translation of monocistronic and bicistronic sequences
Sf9 insect cells were cotransfected with the p1393E1 O E4 O L1 plasmid and mutant baculovirus DNA (BaculoGold; PharMingen). Supernate containing recombinant baculoviruses was harvested 5 days after transfection
A wheat germ in vitro transcription/translation system (TNT coupled wheat germ extract system; Promega,
AID
vira
VY 8042
/
6a1b$$$182
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
Madison, WI) was used to compare transcription and translation of the E1 O E4 O L1 bicistronic sequence to control constructs (monocistronic E1 O E4 and L1 ORFs or pSPORT-XE1 O E4 O L1). Five hundred nanograms of each pSPORT construct was linearized at the unique HindIII site (in the vector, not present in the HPV DNAs) and added to in vitro reactions as instructed by the manufacturer. Coupled transcription/translation was performed in wheat germ extracts using T7 RNA polymerase in the presence of 35S-labeled L-cysteine (30 mCi per 50-ml reaction) (ICN, Costa Mesa, CA). Translated products were analyzed by 15% SDS–PAGE followed by autoradiography. The autoradiograph was scanned with a Beckman DU-64 spectrophotometer using the GelScan Soft-Pac Module (Beckman Instruments, Fullerton, CA). Band intensities were normalized by calculations based on the number of cysteine residues present in each protein (three in E1 O E4 and nine in L1), then further normalized for the amount of transcript detected in each in vitro transcription/translation reaction. RNA analysis of in vitro transcription/translation reactions To determine if differences in transcription efficiency were responsible for differences in protein expression, total RNA was extracted using Trizol (Gibco BRL) from 50-ml in vitro reactions of pSPORT constructs containing E1 O E4, L1, E1 O E4 O L1, or XE1 O E4 O L1 sequences. Equal amounts of total RNA from each of the reactions were applied to 1.5% agarose gels containing 5% formaldehyde and 11 MOPS. Following electrophoresis in 11 MOPS, RNA was transferred to nitrocellulose in 101 SSC. Membranes were prehybridized at 657 for 6 hr in 51 SSPE, 51 Denhardt solution (Denhardt, 1966), and 0.5% SDS, then hybridized with a 32P-end-labeled E4 O L1 DNA probe. The probe was prepared by digesting the pCR3-E1 O E4 O L1 clone with AhdI and XbaI, and isolating a 286-bp DNA fragment that spanned the 3* end of the E4 ORF, the 12 uncoded intracistronic nucleotides, and the 5* end of L1. Hybridization was performed for 16 hr at 657. After three washes at 507 in solutions of decreasing salt concentrations, autoradiography was performed. Two Northern blots were performed: one to compare the transcripts from the pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs, and a second to compare transcripts from the three L1-encoding constructs (pSPORT-L1, E1 O E4 O L1, and XE1 O E4 O L1). Scanning densitometry was performed on autoradiograms as described above for SDS–PAGE autoradiograms. Antisense oligonucleotides Two antisense oligonucleotides (24-mers) were synthesized to examine the relationship of translation of the two ORFs from the E1 O E4 O L1 transcript. The positions of the oligonucleotides in relation to E1 O E4 O L1 se-
AID
VY 8042
/
6a1b$$$182
07-07-96 19:41:36
45
quences are shown in Fig. 1. The A4E oligonucleotide is antisense to the E1 O E4 start codon region, and A1L is antisense to the L1 start codon region. Oligonucleotides were dissolved in sterile water and added to in vitro reactions at concentrations from 0.05 to 2.0 mM. Reactions were analyzed by 15% SDS–PAGE and autoradiographs performed. RESULTS Cloning and expression of the E1 O E4 O L1 sequence DNA sequencing of the pCR3-E1 O E4 O L1 clone verified that both the E1 O E4 and E4 O L1 splice sites and the 12nucleotide untranslated sequence between the two ORFs were present exactly as described by Rotenberg et al. (1989). An illustration of the cloned E1 O E4 O L1 sequence is shown in Fig. 1. Immunoblot analysis of Sf9 cells infected with recombinant E1 O E4 O L1 baculoviruses was performed using antisera raised against either HPV 11 trpE/E1 O E4 or trpE/ L1 fusion proteins (Brown et al., 1994b). Both the E1 O E4 and the L1 proteins were detected in the immunoblots (Fig. 2). The E1 O E4 protein was detected as a strong 11-kDa band and a faint 10-kDa band (Fig. 2A). For L1, a 54-kDa protein was detected, with an additional faint band observed at approximately 28 kDa (not shown). The relative intensity of expressed E1 O E4 protein was significantly greater than that of the L1 protein in the immunoblots developed to the same degree, but specific quantification was not attempted because two different primary antibodies were used. The ELISA demonstrated that 10 ng of nondenatured L1 protein was present per microgram of Sf9 cell nuclear protein. Icosahedral 50- to 60-nm VLPs were detected in electron micrographs (Fig. 3). These VLPs were similar in appearance to those produced in Sf9 cells by other investigators, using recombinant baculoviruses expressing monocistronic L1 ORFs (Hagensee et al., 1993; Rose et al., 1993). In vitro transcription/translation of the E1 O E4 O L1 and control sequences The E1 O E4 O L1 sequence was cloned into the pSPORT vector and expressed in a wheat germ extract in vitro coupled-transcription/translation system. In addition, several control pSPORT constructs were made. Northern blots were performed to determine if differences in transcriptional efficiency were responsible for differences in protein expression. Because an end-labeled E4 O L1 probe was used in the Northern blot analysis, signal intensity was indicative of the relative amount of each transcript. In the in vitro coupled-transcription/translation reactions, transcripts expressed from the pSPORT-E1 O E4 construct were 9.5 times more abundant than those de-
vira
AP: Virology
46
BROWN ET AL.
FIG. 1. A partial representation of the HPV 11 E1 O E4 O L1 DNA sequence (at top) and pSPORT constructs used in the in vitro transcription/ translation reactions. For the E1 O E4 O L1 sequence, start codons are underlined; E4 ORF sequences are shown in bold. Numbers below the sequence represent the nucleotides involved in splicing. Twelve nucleotides are present between the E4 stop codon and the splice donor site. These untranslated nucleotides are a continuation of the sequence 3* to the E4 stop codon. For the pSPORT constructs: 1, the E1 O E4 O L1 construct, with the positions of the A4E and A1L antisense oligonucleotides shown below the ORFs; 2, the pSPORT-XE1 O E4 O L1 construct lacking an E1 O E4 start codon; 3, the monocistronic E1 O E4 construct; and 4, the monocistronic L1 construct.
FIG. 2. Immunoblot blot analysis of Sf9 cell extracts for (A) E1 O E4 or (B) L1 protein. Molecular weight marker positions are shown on the left. The antisera used were rabbit polyclonal sera raised against either an HPV 11 trpE/E1 O E4 or a trpE/L1 fusion protein. Approximately 5 mg of total protein was applied to each lane. Lane 1, uninfected Sf9 cells; lane 2, wild-type baculovirus-infected Sf9 cells; lane 3, E1 O E4 O L1 recombinant baculovirus-infected Sf9 cells. The immunoblot in B was allowed to develop longer than the blot in A.
Proteins expressed from the pSPORT-E1 O E4 and L1 monocistronic control constructs were of expected size (Fig. 4B). The E1 O E4 protein was detected as a strong band at 11 kDa and a weaker band at 10 kDa (Fig. 4B, lane 1). No higher molecular weight gene products were seen. For L1, a 54-kDa band and a weaker 28-kDa band were visualized (Fig. 4B, lane 2). The predicted molecular weight of the L1 protein is 54 kDa. The 28-kDa protein may represent a breakdown product of the 54-kDa protein, as it was also seen in the immunoblots of Sf9 cells expressing L1 protein. Alternatively, the 28-kDa protein may be a result of internal initiation within the L1 ORF. The bicistronic E1 O E4 O L1 construct produced both the E1 O E4 and the L1 proteins, but not in equal amounts (Fig. 4B, lane 3). More E1 O E4 protein was expressed than L1 protein from the pSPORT-E1 O E4 O L1 construct. After normalizing signal intensities for cysteine content (three in the E1 O E4 protein and nine in the L1 protein), the ratio of E1 O E4 to L1 proteins expressed from the bicistronic construct was 17:1. When expression from bicistronic and monocistronic constructs was compared, scanning densitometry showed that the pSPORT-E1 O E4 O L1 construct expressed 6.8 times more E1 O E4 protein than the monocistronic pSPORT-E1 O E4 construct. This observation was in contrast to the RNA studies, in which the pSPORT-E1 O E4 construct produced 9.5 times more mRNA than the
AID
vira
rived from either the pSPORT-L1 or the pSPORTE1 O E4 O L1 constructs (Fig. 4A). Transcripts from the pSPORT-L1 and pSPORTE1 O E4 O L1 constructs (Fig. 4A) were equally abundant (within 5% of each other by scanning densitometry).
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
47
FIG. 5. Transcription and translation from in vitro reactions using the pSPORT-L1, pSPORT-E1 O E4 O L1, and pSPORT-XE1 O E4 O L1 constructs. (A) RNA analysis. Lane 1, RNA transcribed from pSPORT-L1; lane 2, RNA from pSPORT-E1 O E4 O L1; lane 3, RNA from pSPORTXE1 O E4 O L1. Size of RNA fragments (nt) is indicated on the left. (B) Autoradiogram of in vitro transcription/translation of pSPORT-L1, E1 O E4 O L1, and XE1 O E4 O L1 constructs. Molecular weight marker positions (kDa) are shown on the left. Lane 1, expression from the pSPORTL1; lane 2, expression from the pSPORT-E1 O E4 O L1 construct; and lane 3, expression from pSPORT-XE1 O E4 O L1.
FIG. 3. Electron microscopy of VLPs expressed in E1 O E4 O L1 recombinant baculovirus-infected Sf9 cells. Original magnification 100,0001. Solid bar represents 100 nm at plane of photograph.
pSPORT-E1 O E4 O L1 construct (Fig. 4A). When protein expression levels were further normalized for mRNA, the pSPORT-E1 O E4 O L1 construct expressed 64 times more E1 O E4 protein than the monocistronic pSPORT-E1 O E4 construct. For L1, the pSPORT-E1 O E4 O L1 and pSPORT-L1 constructs produced equal amounts of mRNA (Fig. 4A, lanes
2 and 3). In contrast, 2.5 times more L1 protein was expressed from the monocistronic pSPORT-L1 construct used as a control than from pSPORT-E1 O E4 O L1 (Fig. 4B, lanes 2 and 3). The pSPORT-XE1 O E4 O L1 construct was made to test the effect of the E1 O E4 ORF, lacking a start codon, on L1 protein expression. As shown in Fig. 5A, transcripts from the three L1-encoding constructs (pSPORT-L1, pSPORT-E1 O E4 O L1, and pSPORT-XE1 O E4 O L1) were of nearly equal abundance (within 5% of each other by scanning densitometry). While the pSPORT-E1 O E4 O L1 construct expressed L1 protein in amounts less than the monocistronic pSPORT-L1 construct, the pSPORTXE1 O E4 O L1 expressed a 54-kDa L1 protein in amounts slightly greater (1.2 times) than that of the monocistronic pSPORT-L1 construct (Fig. 5B). Effects of antisense oligonucleotides
FIG. 4. Transcription and translation from in vitro reactions using the pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs. (A) RNA analysis. A 286-bp E4 O L1 DNA probe was end-labeled with [g-32P]ATP. An equal amount of total RNA was added to each lane. Lane 1, RNA transcribed from pSPORT-E1 O E4; lane 2, RNA from pSPORT-L1; lane 3, RNA from pSPORT-E1 O E4 O L1. Size of RNA fragments (nt) is indicated on the left. (B) Autoradiogram of in vitro transcription/translation of pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs. Molecular weight marker positions (kDa) are shown on the left. Lane 1, expression from the pSPORT-E1 O E4 construct; lane 2, expression from pSPORT-L1; lane 3, expression from pSPORT-E1 O E4 O L1.
AID
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
The A4E oligonucleotide (antisense to the E1 O E4 start codon region) inhibited E1 O E4 protein production expressed from the pSPORT-E1 O E4 O L1, and an increase in L1 protein production was observed (Fig. 6A). This increase in L1 protein expression occurred at the lowest concentration of A4E (0.05 mM), and was observed at all higher A4E concentrations as well. The A1L oligonucleotide (antisense to the L1 start codon region) inhibited L1 protein from both pSPORT-L1 (not shown) and E1 O E4 O L1 constructs (Fig. 6B). E1 O E4 protein production was not altered by the presence of A1L when expressed from the pSPORT-E1 O E4 O L1 (Fig. 6B). DISCUSSION For HPV 11, and possibly for the other genital HPVs as well, the only L1-containing transcript detected in tissues is the bicistronic E1 O E4 O L1 mRNA, potentially encoding both the E1 O E4 and the L1 proteins (Chow et al.,
vira
AP: Virology
48
BROWN ET AL.
1987). All of the HPV 11 E1 O E4 O L1 splice sites are conserved in the closely related HPV 6 and in other genital HPV types including cancer-associated HPVs such as HPV types 16 and 18. A consensus splice donor site is identically positioned 12 nucleotides downstream from the E4 termination codons in these genital HPV types. The splice acceptor site for the third exon (the L1 ORF) of the E1 O E4 O L1 mRNAs is located at the initiation codon of the L1 ORF. The 5* ends of the HPV 11 E1 O E4 and E1 O E4 O L1 transcripts and the promoter regions that regulate their expression have not been clearly defined. The 5* untranslated sequences in the authentic mRNAs may influence translational efficiency of both transcripts, and the promoters will influence the level of the messages. For the HPV 11 E1 O E4 transcript, Nasseri et al. (1987) cloned and sequenced a cDNA, finding the 5* end originating at nucleotide 716. Primer extension assays were not performed. For HPV 6, Ward and Mounts (1989) identified heterogeneous 5* start sites by primer extension analysis
from several respiratory tract isolates. These 5* start sites ranged from nucleotide 628 to 758, but the location of the primer could not distinguish between the E1 O E4 and the E1 O E4 O L1 transcripts. For HPV 31, an HPV type associated with cervical malignancy, the major L1-encoding transcript was found to be a bicistronic species (E1 O E4 O L1), which initiated at the differentiation-specific promoter located in the middle of the E7 ORF (nt 742) (Hummel et al., 1995). Because the 5* start sites have not been clearly defined for either the HPV 11 E1 O E4 or the E1 O E4 O L1 transcript, and because it is not known if the two transcripts have the same 5* start site, we chose to compare protein expression from the E1 O E4 and E1 O E4 O L1 sequences beginning at the E1 start codon. In addition, the monocistronic pSPORT-L1 construct, used as a control for comparison of expression to the bicistronic E1 O E4 O L1 construct, was also constructed to begin at the L1 start codon. The start codons of all pSPORT constructs thus were an equal distance from the T7 promoter. These constructs therefore allowed direct comparisons of E1 O E4 and L1 translation without the confounding questions of influences of the 5* untranslated region. Admittedly our results could differ if the precise 5* ends of each cDNA were known and included in the pSPORT or baculovirus constructs. While establishing the exact 5* ends of these transcripts was not the goal of the current investigation, subsequent studies will address this issue. Because a potential splice acceptor is present in the E4 ORF at nt 3377, used to produce the E1 O E2C transcript, a splicing event utilizing this acceptor and a baculovirus splice donor could potentially produce a transcript with the L1 AUG being the first one in a favorable context. This event is not likely because no potential splice donor sequence is present in the clone between the polyhedron promoter and the start of the E1 O E4 O L1 sequence. In addition, the E1 O E4 O L1 sequence was cloned as cDNA into the pVL baculovirus expression vector, eliminating the possibility of further or aberrant splicing of the transcript at the E1 splice donor site. For the antisense oligonucleotide studies, inhibition of E1 O E4 translation with A4E resulted in an increase in L1 translation, supporting the scanning model of translation, in which ribosomes begin translation at the first available AUG in a favorable context (Kozak, 1989). Interestingly, the marked increase in L1 translation occurred at very low concentrations of A4E (0.05 mM). In fact, E1 O E4 protein expression was only modestly reduced at these A4E concentrations. Inhibition of L1 translation by the AIL oligonucleotide resulted in no change in E1 O E4 protein expression from the pSPORT-E1 O E4 O L1 construct, suggesting that L1 translation is not directly responsible for the increased level of E1 O E4 protein expression observed from the pSPORT-E1 O E4 O L1 construct compared to pSPORT-E1 O E4. It is likely, however, that contributions of the 3* untranslated region of the
AID
vira
FIG. 6. Effects of antisense oligonucleotides on in vitro transcription/ translation reactions using pSPORT-E1 O E4 O L1. (A) The A4E oligonucleotide was added to reactions. Lane 1, no A4E; lanes 2 through 6, A4E was added to reactions at a final concentration of 0.05, 0.1, 0.2, 0.5, or 1.0 mM. (B) The A1L oligonucleotide was added to reactions. Lane 1, no A1L; lanes 2 and 3, 0.5 and 2 mM A1L added to reactions. Molecular weight marker positions (in kDa) are shown on the left.
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
49
bicistronic transcript are at least partially responsible for the observed increased E1 O E4 protein expression. The greater length of the E1 O E4 O L1 transcript compared to the E1 O E4 transcript may also influence translational efficiency, perhaps by stabilizing transcripts. The pSPORT-XE1 O E4 O L1 construct, which lacks an E1 O E4 start codon, was efficiently transcribed and expressed L1 protein with the same efficiency as the monocistronic pSPORT-L1 construct, suggesting that the difference between these constructs with respect to the distance of the T7 promoter to the L1 start codon was of minimal importance. Cell-free systems such as the one in this study are useful tools in analysis of transcriptional and translational efficiency of HPV genes and the effects of antisense oligonucleotides on those genes. Our studies using the wheat germ extract in vitro transcription/translation system show that the cloned E1 O E4 O L1 sequence expresses both the E1 O E4 and the L1 proteins. This observation was also found in eukaryotic cells. We cannot prove that the relative translational efficiencies of the two ORFs in this bicistronic message would be the same in differentiating epithelium. However, we have demonstrated that while transcription of E1 O E4 message is most efficient from the monocistronic pSPORT-E1 O E4 construct, translational efficiency of E1 O E4 is greater from the bicistronic E1 O E4 O L1 construct. A recent study by Stacey et al. (1995) examined translation from the HPV 16 E6 O E7 bicistronic transcript. Using in vitro and cellular translation systems, it was shown that the splicing events leading to the E6* O E7 transcript was not required for E7 synthesis. In addition, site-directed mutagenesis experiments showed that the majority of ribosomes used to translate the E7 ORF ignored some or all of the ORFs upstream from E7. This finding is in contrast to other studies, which suggest that for most structurally bicistronic viral transcripts, generally only the 5* ORF is translated (Kozak, 1986). Although some viral transcripts are bicistronic, the phenomenon of a single mRNA species with the potential to encode both an early and a late protein has not been observed in other virus systems. As mentioned above, the E1 O E4 protein may in reality be a late gene product, even though the sequences are present in the early region of the HPV genome. We and others have shown that E1 O E4 transcripts are found early in HPV 11 infection, but E1 O E4 protein is tightly linked to detection of L1 protein, which appears late in infection, and only in the most differentiated epithelial cells (Brown et al., 1995, 1994b; Stoler and Broker, 1986; Stoler et al., 1990). Most viral mRNAs are monocistronic, but in the case of those transcripts that are bicistronic, ribosomes can reinitiate translation at the second start codon (Kozak, 1984). The efficiency of reinitiation in these cases is generally low (Kozak, 1984). Our analysis of the HPV 11 E1 O E4 O L1 transcript is consistent with the scanning
model for translation as applied to polycistronic viral transcripts (Kozak, 1989). The E1 O E4 O L1 transcript is functionally bicistronic, expressing both the E1 O E4 and the L1 proteins in an in vitro system and in eukaryotic cells.
AID
vira
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
ACKNOWLEDGMENTS This study was supported in part by the Merit Review Board of the Department of Veterans Affairs and the National Institute of Allergy and Infectious Diseases (AI 9416, Project 4). We thank Lutz Gissmann for his helpful advice, and Tonya Moss for secretarial assistance.
REFERENCES Brown, D. R., Chin, M. T., and Strike, D. G. (1988). Identification of human papillomavirus type 11 E4 gene products in human tissue implants from athymic mice. Virology 165, 262–267. Brown, D. R., Bryan, J., Rodriguez, M., Rose, R. C., and Strike, D. G. (1991). Detection of human papillomavirus types 6 and 11 E4 gene products in condylomata acuminatum. J. Med. Virol. 34, 20–28. Brown, D. R., Bryan, J. T., Cramer, H., Katz, B. P., Handy, V., and Fife, K. H. (1994a). Detection of multiple human papillomavirus types in condylomata acuminata from immunosuppressed patients. J. Infect. Dis. 170, 759–765. Brown, D. R., Fan, L., Jones, J., and Bryan, J. (1994b). Colocalization of human papillomavirus type 11 E1 O E4 and L1 proteins in human foreskin implants grown in athymic mice. Virology 201, 46–54. Brown, D. R., Bryan, J. T., Pratt, L., Handy, V., Fife, K. H., and Stoler, M. H. (1995). Human papillomavirus type 11 E1 O E4 and L1 proteins colocalize in the mouse xenograft system at multiple time points. Virology 214, 259–263. Chow, L. T., Nasseri, M., Wolinsky, S. M., and Broker, T. R. (1987). Human papillomavirus types 6 and 11 mRNAs from genital condylomata acuminata. J. Virol. 61, 2581–2588. Christensen, N. D., Ho¨pfl, R., DiAngelo, S. L., Cladel, N. M., Patrick, S. D., Welsh, P. A., Budgeon, L. R., Reed, C. A., and Kreider, J. W. (1994). Assembled baculovirus-expressed human papillomavirus type 11 L1 capsid protein virus-like particles are recognized by neutralizing monoclonal antibodies and induce high titres of neutralizing antibodies. J. Gen. Virol. 75, 2271–2276. Denhardt, D. T. (1966). A membrane-filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Commun. 23, 641– 646. Doorbar, J. (1991). An emerging function for E4. Papillomavirus Rep. 2, 145–147. Doorbar, J., Campbell, D., Grand, R. J. A., and Gallimore, P. H. (1986). Identification of the human papilloma virus-1a E4 gene products. EMBO J. 5, 355–362. Hagensee, M. E., Yaegashi, N., and Galloway, D. A. (1993). Self-assembly of human papillomavirus type 1 capsids by expression of the L1 protein alone or by coexpression of the L1 and L2 capsid proteins. J. Virol. 67, 315–322. Hofmann, K. J., Cook, J. C., Joyce, J. G., Brown, D. R., Schultz, L. D., George, H. A., Rosolowsky, M., Fife, K. H., and Jansen, K. U. (1995). Sequence determination of human papillomavirus type 6a and assembly of virus-like particles in Saccharomyces cerevisiae. Virology 209, 506–518. Hummel, M., Lim, H. B., and Laimins, L. A. (1995). Human papillomavirus type 31b late gene expression is regulated through protein kinase C-mediated changes in RNA processing. J. Virol. 69, 3381–3388. Kozak, M. (1984). Selection of initiation sites by eucaryotic ribosomes: Effect of inserting AUG triplets upstream from the coding sequence for preproinsulin. Nucleic Acids Res. 12, 3873–3893. Kozak, M. (1986). Regulation of protein synthesis in virus-infected animal cells. Adv. Virus Res. 31, 229–292.
AP: Virology
50
BROWN ET AL.
Kozak, M. (1989). The scanning model for translation: An update. J. Cell Biol. 108, 229–241. Nasseri, M., Hirochika, R., Broker, T. R., and Chow, L. T. (1987). A human papillomavirus type 11 transcript encoding an E1 O E4 protein. Virology 159, 433–439. Neary, K., Horwitz, B. H., and DiMaio, D. (1987). Mutational analysis of open reading frame E4 of bovine papillomavirus type 1. J. Virol. 61, 1248–1252. Rose, R. C., Bonnez, W., Reichman, R. C., and Garcea, R. L. (1993). Expression of human papillomavirus type 11 L1 protein in insect cells: In vivo and in vitro assembly of viruslike particles. J. Virol. 67, 1936–1944. Rotenberg, M. O., Chow, L. T., and Broker, T. R. (1989). Characterization of rare human papillomavirus type 11 mRNAs coding for regulatory and structural proteins, using the polymerase chain reaction. Virology 172, 489–497. Stacey, S. N., Jordan, D., Snijders, P. J. F., Mackett, M., Walboomers,
J. M. M., and Arrand, J. R. (1995). Translation of the human papillomavirus type 16 E7 oncoprotein from bicistronic mRNA is independent of splicing events within the E6 open reading frame. J. Virol. 69, 7023–7031. Stoler, M. H., and Broker, T. R. (1986). In situ hybridization detection of human papillomavirus DNAs and messenger RNAs in genital condylomas and a cervical carcinoma. Hum. Pathol. 17, 1250–1258. Stoler, M. H., Whitbeck, A., Wolinsky, S. M., Broker, T. R., Chow, L. T., Howett, M. K., and Kreider, J. W. (1990). Infectious cycle of human papillomavirus type 11 in human foreskin xenografts in nude mice. J. Virol. 64, 3310–3318. Towbin, H., Staehelin, T., and Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350– 4354. Ward, P., and Mounts, P. (1989). Heterogeneity in mRNA of human papillomavirus type-6 subtypes in respiratory tract lesions. Virology 168, 1–12.
AID
vira
VY 8042
/
6a1b$$$184
07-07-96 19:41:36
AP: Virology
222, 43–50 (1996) 0396
ARTICLE NO.
Virus-like Particles and E1 O E4 Protein Expressed from the Human Papillomavirus Type 11 Bicistronic E1 O E4 O L1 Transcript DARRON R. BROWN,*,†,1 LINDA PRATT,* JANINE T. BRYAN,‡ KENNETH H. FIFE,*,‡,§ and KATHRIN JANSENØ Departments of *Medicine, ‡Microbiology and Immunology, and §Pathology, and †Richard L. Roudebush Veterans Administration Medical Center, Indiana University School of Medicine, Indianapolis, Indiana 46202; and ØDepartment of Virus and Cell Biology, Merck Research Laboratories, West Point, Pennsylvania 19486 Received March 5, 1996; accepted June 5, 1996 Detection of E1 O E4 protein in human papillomavirus (HPV 11)-infected tissue is tightly linked to detection of L1 major capsid protein. The only L1-containing transcript identified in HPV 11-infected tissue is the bicistronic E1 O E4 O L1 mRNA, potentially encoding both the E1 O E4 and the L1 proteins. It has not been established that these proteins can be expressed from the E1 O E4 O L1 transcript. The HPV 11 E1 O E4 O L1 sequence was cloned by reverse transcriptase polymerase chain reaction into the p1393 vector to produce recombinant baculoviruses. Immunoblots of recombinant baculovirus-infected Sf9 cell lysates demonstrated both the E1 O E4 and the L1 proteins. An ELISA was performed on infected Sf9 cells using a monoclonal antibody specific for nondenatured L1, demonstrating that 10 ng of native L1 protein was present per microgram of total nuclear protein. Electron microscopic analysis revealed 50- to 60-nm icosahedral virus-like particles. In vitro transcription/translation was performed using pSPORT constructs containing the E1 O E4 O L1 sequence or, as controls, monocistronic pSPORT-E1 O E4 or L1 constructs. The pSPORT-E1 O E4 O L1 construct produced the E1 O E4 and L1 proteins at a ratio of 17:1. For E1 O E4 protein, expression was greater from the pSPORT-E1 O E4 O L1 construct than from the monocistronic pSPORT-E1 O E4 construct. In contrast, more L1 protein was expressed from pSPORT-L1 than from pSPORT-E1 O E4 O L1. A mutant E1 O E4 O L1 construct containing no E1 O E4 start codon expressed L1 protein in amounts nearly equal to that expressed from the pSPORT-L1 construct. Addition of an antisense oligonucleotide directed at the E1 O E4 start codon region to in vitro reactions using pSPORT-E1 O E4 O L1 was associated with inhibition of E1 O E4 protein synthesis and increased translation of L1 protein. q 1996 Academic Press, Inc.
INTRODUCTION
of several spliced HPV 11 mRNAs. Rotenberg et al. (1989) provided direct evidence for the existence of several spliced HPV mRNAs, including the E1 O E4 O L1 bicistronic mRNA, by isolation of RNA from genital lesions and employment of reverse transcriptase polymerase chain reaction (RT PCR). It has not been established that the E1 O E4 O L1 transcript is a functional message. The goals of this study were (1) to clone the HPV 11 E1 O E4 O L1 transcript from an HPV 11-infected human tissue, (2) to verify that both potential ORFs can be expressed in eukaryotic cells, (3) to compare in vitro transcription and translation of the E1 O E4 O L1 sequence to that of control constructs containing monocistronic E1 O E4 and L1 sequences, and (4) to examine the mechanism of translation of the E1 O E4 O L1 transcript using antisense oligonucleotides. To accomplish these goals, the HPV 11 E1 O E4 O L1 sequence was cloned by RT PCR and expressed in Sf9 cells by a recombinant baculovirus. In addition, the E1 O E4 O L1 sequence was cloned into the pSPORT vector and expressed in a wheat germ extract coupled-transcription/translation system.
Human papillomaviruses (HPVs) potentially encode six or more early (E) and two late (L) gene products. Overall, the most abundant viral mRNA species detected in HPV 11-infected tissue is a spliced message encoding an 11kDa gene product containing the first 5 amino acids of the E1 open reading frame (ORF) fused to the last 85 amino acids of the E4 ORF (Nasseri et al., 1987). The resulting E1 O E4 gene product has been identified in HPV-infected cutaneous and genital tissues but the function of this protein has not been determined (Brown et al., 1988, 1991; Doorbar et al., 1986; Doorbar, 1991; Neary et al., 1987). Because the E1 O E4 protein is detected late in HPV 11 infection, and exclusively in differentiated cells that contain the L1 major capsid protein, the E1 O E4 protein might be considered a late gene product, although the genomic sequences reside in the early region. A viral transcript identified as the doubly spliced E1 O E4 O L1 mRNA potentially encodes both the E1 O E4 and the L1 proteins (Chow et al., 1987; Rotenberg et al., 1989). Chow et al. (1987) first demonstrated the existence
MATERIALS AND METHODS Cloning and expression of the E1 O E4 O L1 sequence An HPV 11-infected human foreskin implant grown in an athymic mouse for 12 weeks was used to isolate total
1 To whom correspondence and reprint requests should be addressed. Fax: (317) 274-1587.
0042-6822/96 $18.00
43
AID
VY 8042
/
6a1b$$$181
07-07-96 19:41:36
Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
vira
AP: Virology
44
BROWN ET AL.
RNA as previously described (Brown et al., 1994b). Firststrand cDNA synthesis for the RT PCR was performed for 50 min at 427 using SuperScript II (Gibco BRL, Gaithersburg, MD) and oligo(dT) as primer. PCR was performed with a 5* primer that spanned the E1/E4 splice (ATGGCGGACGATTCAG/CACTG) and a 3* primer (GTGTAGTGGGGTCAAACAGGGATG) in the L1 ORF beginning at nucleotide (nt) 6047, which is 3* to the unique XbaI site in HPV 11. The resulting 580-bp DNA fragment was cloned into the pCR3 vector (Invitrogen, San Diego, CA) to create pCR3-E1 O E4 O L1-6047. The complete L1 ORF was created by digesting a clone containing the XbaI/ SphI L1 fragment of HPV 11 (pATH-L1-11) (Brown et al., 1994b) with EcoRI and XbaI, and ligating this linear plasmid with the E1 O E4 O L1 fragment that had been purified from the pCR3-E1 O E4 O L1-6047 clone by digesting with EcoRI and XbaI. The resulting plasmid, pATH-E1 O E4 O L1, contained the E1 O E4 and L1 ORFs separated by 12 nt and the 3* untranslated region through nt 7928. The E1 O E4 O L1 cDNA fragment was then subcloned into the baculovirus expression vector p1393 (PharMingen, San Diego, CA). To test expression in a cell-free coupled-transcription/ translation system, the E1 O E4 O L1 sequence was removed from p1393 and subcloned into pSPORT (Gibco BRL) at the EcoRI site. To create the monocistronic control plasmids, the E1 O E4 and L1 ORFs were amplified by PCR. For the E1 O E4 ORF, the pATH-E1 O E4 plasmid was used as template (Brown et al., 1994b). The primers used were the 5* primer described above and the 3* primer sequence CTATAGGCGTAGCTGCACTGTGACAG. For L1, the 5* primer sequence was ATGTGGCGGCCTAGCGACAG, and the 3* primer was GGGCCCAGATCTAAGAAGGAAATATGTAGGGTGTGGG, which began at nt 7928 and included a 3* BglII site extension. DNA extracted from a condylomata acuminata lesion containing HPV 11 was used as template for the L1 PCR (Brown et al., 1994a). After PCR, the E1 O E4 and L1 DNA fragments were cloned into pCR3, then subcloned into pSPORT. As a control, a construct was made (pSPORT-XE1 O E4 O L1) lacking an E1 O E4 start codon. The pSPORT-XE1 O E4 O L1 construct was prepared by amplifying the cloned E1 O E4 O L1 sequence by PCR using a 5* primer containing an AAG in place of the ATG start codon. The resulting PCR product was cloned into pCR3, then subcloned into pSPORT. The pSPORT-XE1 O E4 O L1 construct was used in in vitro reactions as described below. The distance between the T7 promoter and the first start codon was the same in each of the pSPORT constructs. Expression of E1 O E4 O L1 in Sf9 cells
and used to infect Sf9 cells. Cells were harvested after 6 days of growth and pelleted by centrifugation at 1000 g. Proteins were separated by 15% SDS–PAGE then transferred to nitrocellulose (Towbin et al., 1979). Immunoblots were performed as previously described (Brown et al., 1991). Nitrocellulose membranes were probed with one of two polyclonal antisera: an antiserum raised against a bacterial-expressed HPV 11 trpE/E1 O E4 or against a trpE/L1 fusion protein (Brown et al., 1994b). Detection of virus-like particles (VLPs) Sf9 cell pellets were lysed with PBS plus 2% Triton X, and nuclei were isolated and concentrated by centrifugation at 1000 g, then 5000 g, followed by two washes in cold PBS. Nuclei were sonicated in PBS and clarified by centrifugation at 1000 g. The clarified supernate was assayed by ELISA and by electron microscopy. For the ELISA, the supernate was incubated in wells of Nunc Maxisorb plates (Nalge, Rochester, NY) for 12 hr at 47. Plates were washed once with TTBS (20 mM Tris–HCl, pH 7.6; 150 mM NaCl; 0.1% Tween 20), then incubated with TTBS plus 10% nonfat dry milk for 2 hr at room temperature. Plates were washed three times with TTBS before adding a monoclonal antibody to HPV 11 previously shown to specifically recognize nondenatured L1 protein (generously provided by Neil Christensen) in TTBS plus 10% nonfat dry milk (Christensen et al., 1994). After a 1-hr incubation at room temperature, plates were washed three times in TTBS. Goat anti-mouse IgG–alkaline phosphatase conjugate was added as secondary antibody at 1:1000 dilution in TTBS plus 10% nonfat dry milk and incubated for 1 hr at room temperature. Plates were washed three times, and p-nitrophenyl phosphate in diethanolamine buffer (Kirkegaard and Perry, Gaithersburg, MD) was added as substrate. Plates were incubated for 30 min at room temperature before stopping the color reactions with 3 M NaOH. The absorbance of each well was measured at 405 nm. The amount of nondenatured L1 protein was determined by endpoint dilution compared to mean absorbance readings of control wells containing known quantities of yeast-expressed, sucrose gradient-purified HPV 11 L1 VLPs (Hofmann et al., 1995). For electron microscopy a drop of clarified Sf9 cell nuclear supernate was placed on a 200-mesh carboncoated copper grid. A drop of 2% phosphotungstic acid, pH 7.0, was placed on the grid. After grids had air dried, microscopy was performed with a JEOL 100CX transmission electron microscope at an accelerating voltage of 100 kV. In vitro transcription and translation of monocistronic and bicistronic sequences
Sf9 insect cells were cotransfected with the p1393E1 O E4 O L1 plasmid and mutant baculovirus DNA (BaculoGold; PharMingen). Supernate containing recombinant baculoviruses was harvested 5 days after transfection
A wheat germ in vitro transcription/translation system (TNT coupled wheat germ extract system; Promega,
AID
vira
VY 8042
/
6a1b$$$182
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
Madison, WI) was used to compare transcription and translation of the E1 O E4 O L1 bicistronic sequence to control constructs (monocistronic E1 O E4 and L1 ORFs or pSPORT-XE1 O E4 O L1). Five hundred nanograms of each pSPORT construct was linearized at the unique HindIII site (in the vector, not present in the HPV DNAs) and added to in vitro reactions as instructed by the manufacturer. Coupled transcription/translation was performed in wheat germ extracts using T7 RNA polymerase in the presence of 35S-labeled L-cysteine (30 mCi per 50-ml reaction) (ICN, Costa Mesa, CA). Translated products were analyzed by 15% SDS–PAGE followed by autoradiography. The autoradiograph was scanned with a Beckman DU-64 spectrophotometer using the GelScan Soft-Pac Module (Beckman Instruments, Fullerton, CA). Band intensities were normalized by calculations based on the number of cysteine residues present in each protein (three in E1 O E4 and nine in L1), then further normalized for the amount of transcript detected in each in vitro transcription/translation reaction. RNA analysis of in vitro transcription/translation reactions To determine if differences in transcription efficiency were responsible for differences in protein expression, total RNA was extracted using Trizol (Gibco BRL) from 50-ml in vitro reactions of pSPORT constructs containing E1 O E4, L1, E1 O E4 O L1, or XE1 O E4 O L1 sequences. Equal amounts of total RNA from each of the reactions were applied to 1.5% agarose gels containing 5% formaldehyde and 11 MOPS. Following electrophoresis in 11 MOPS, RNA was transferred to nitrocellulose in 101 SSC. Membranes were prehybridized at 657 for 6 hr in 51 SSPE, 51 Denhardt solution (Denhardt, 1966), and 0.5% SDS, then hybridized with a 32P-end-labeled E4 O L1 DNA probe. The probe was prepared by digesting the pCR3-E1 O E4 O L1 clone with AhdI and XbaI, and isolating a 286-bp DNA fragment that spanned the 3* end of the E4 ORF, the 12 uncoded intracistronic nucleotides, and the 5* end of L1. Hybridization was performed for 16 hr at 657. After three washes at 507 in solutions of decreasing salt concentrations, autoradiography was performed. Two Northern blots were performed: one to compare the transcripts from the pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs, and a second to compare transcripts from the three L1-encoding constructs (pSPORT-L1, E1 O E4 O L1, and XE1 O E4 O L1). Scanning densitometry was performed on autoradiograms as described above for SDS–PAGE autoradiograms. Antisense oligonucleotides Two antisense oligonucleotides (24-mers) were synthesized to examine the relationship of translation of the two ORFs from the E1 O E4 O L1 transcript. The positions of the oligonucleotides in relation to E1 O E4 O L1 se-
AID
VY 8042
/
6a1b$$$182
07-07-96 19:41:36
45
quences are shown in Fig. 1. The A4E oligonucleotide is antisense to the E1 O E4 start codon region, and A1L is antisense to the L1 start codon region. Oligonucleotides were dissolved in sterile water and added to in vitro reactions at concentrations from 0.05 to 2.0 mM. Reactions were analyzed by 15% SDS–PAGE and autoradiographs performed. RESULTS Cloning and expression of the E1 O E4 O L1 sequence DNA sequencing of the pCR3-E1 O E4 O L1 clone verified that both the E1 O E4 and E4 O L1 splice sites and the 12nucleotide untranslated sequence between the two ORFs were present exactly as described by Rotenberg et al. (1989). An illustration of the cloned E1 O E4 O L1 sequence is shown in Fig. 1. Immunoblot analysis of Sf9 cells infected with recombinant E1 O E4 O L1 baculoviruses was performed using antisera raised against either HPV 11 trpE/E1 O E4 or trpE/ L1 fusion proteins (Brown et al., 1994b). Both the E1 O E4 and the L1 proteins were detected in the immunoblots (Fig. 2). The E1 O E4 protein was detected as a strong 11-kDa band and a faint 10-kDa band (Fig. 2A). For L1, a 54-kDa protein was detected, with an additional faint band observed at approximately 28 kDa (not shown). The relative intensity of expressed E1 O E4 protein was significantly greater than that of the L1 protein in the immunoblots developed to the same degree, but specific quantification was not attempted because two different primary antibodies were used. The ELISA demonstrated that 10 ng of nondenatured L1 protein was present per microgram of Sf9 cell nuclear protein. Icosahedral 50- to 60-nm VLPs were detected in electron micrographs (Fig. 3). These VLPs were similar in appearance to those produced in Sf9 cells by other investigators, using recombinant baculoviruses expressing monocistronic L1 ORFs (Hagensee et al., 1993; Rose et al., 1993). In vitro transcription/translation of the E1 O E4 O L1 and control sequences The E1 O E4 O L1 sequence was cloned into the pSPORT vector and expressed in a wheat germ extract in vitro coupled-transcription/translation system. In addition, several control pSPORT constructs were made. Northern blots were performed to determine if differences in transcriptional efficiency were responsible for differences in protein expression. Because an end-labeled E4 O L1 probe was used in the Northern blot analysis, signal intensity was indicative of the relative amount of each transcript. In the in vitro coupled-transcription/translation reactions, transcripts expressed from the pSPORT-E1 O E4 construct were 9.5 times more abundant than those de-
vira
AP: Virology
46
BROWN ET AL.
FIG. 1. A partial representation of the HPV 11 E1 O E4 O L1 DNA sequence (at top) and pSPORT constructs used in the in vitro transcription/ translation reactions. For the E1 O E4 O L1 sequence, start codons are underlined; E4 ORF sequences are shown in bold. Numbers below the sequence represent the nucleotides involved in splicing. Twelve nucleotides are present between the E4 stop codon and the splice donor site. These untranslated nucleotides are a continuation of the sequence 3* to the E4 stop codon. For the pSPORT constructs: 1, the E1 O E4 O L1 construct, with the positions of the A4E and A1L antisense oligonucleotides shown below the ORFs; 2, the pSPORT-XE1 O E4 O L1 construct lacking an E1 O E4 start codon; 3, the monocistronic E1 O E4 construct; and 4, the monocistronic L1 construct.
FIG. 2. Immunoblot blot analysis of Sf9 cell extracts for (A) E1 O E4 or (B) L1 protein. Molecular weight marker positions are shown on the left. The antisera used were rabbit polyclonal sera raised against either an HPV 11 trpE/E1 O E4 or a trpE/L1 fusion protein. Approximately 5 mg of total protein was applied to each lane. Lane 1, uninfected Sf9 cells; lane 2, wild-type baculovirus-infected Sf9 cells; lane 3, E1 O E4 O L1 recombinant baculovirus-infected Sf9 cells. The immunoblot in B was allowed to develop longer than the blot in A.
Proteins expressed from the pSPORT-E1 O E4 and L1 monocistronic control constructs were of expected size (Fig. 4B). The E1 O E4 protein was detected as a strong band at 11 kDa and a weaker band at 10 kDa (Fig. 4B, lane 1). No higher molecular weight gene products were seen. For L1, a 54-kDa band and a weaker 28-kDa band were visualized (Fig. 4B, lane 2). The predicted molecular weight of the L1 protein is 54 kDa. The 28-kDa protein may represent a breakdown product of the 54-kDa protein, as it was also seen in the immunoblots of Sf9 cells expressing L1 protein. Alternatively, the 28-kDa protein may be a result of internal initiation within the L1 ORF. The bicistronic E1 O E4 O L1 construct produced both the E1 O E4 and the L1 proteins, but not in equal amounts (Fig. 4B, lane 3). More E1 O E4 protein was expressed than L1 protein from the pSPORT-E1 O E4 O L1 construct. After normalizing signal intensities for cysteine content (three in the E1 O E4 protein and nine in the L1 protein), the ratio of E1 O E4 to L1 proteins expressed from the bicistronic construct was 17:1. When expression from bicistronic and monocistronic constructs was compared, scanning densitometry showed that the pSPORT-E1 O E4 O L1 construct expressed 6.8 times more E1 O E4 protein than the monocistronic pSPORT-E1 O E4 construct. This observation was in contrast to the RNA studies, in which the pSPORT-E1 O E4 construct produced 9.5 times more mRNA than the
AID
vira
rived from either the pSPORT-L1 or the pSPORTE1 O E4 O L1 constructs (Fig. 4A). Transcripts from the pSPORT-L1 and pSPORTE1 O E4 O L1 constructs (Fig. 4A) were equally abundant (within 5% of each other by scanning densitometry).
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
47
FIG. 5. Transcription and translation from in vitro reactions using the pSPORT-L1, pSPORT-E1 O E4 O L1, and pSPORT-XE1 O E4 O L1 constructs. (A) RNA analysis. Lane 1, RNA transcribed from pSPORT-L1; lane 2, RNA from pSPORT-E1 O E4 O L1; lane 3, RNA from pSPORTXE1 O E4 O L1. Size of RNA fragments (nt) is indicated on the left. (B) Autoradiogram of in vitro transcription/translation of pSPORT-L1, E1 O E4 O L1, and XE1 O E4 O L1 constructs. Molecular weight marker positions (kDa) are shown on the left. Lane 1, expression from the pSPORTL1; lane 2, expression from the pSPORT-E1 O E4 O L1 construct; and lane 3, expression from pSPORT-XE1 O E4 O L1.
FIG. 3. Electron microscopy of VLPs expressed in E1 O E4 O L1 recombinant baculovirus-infected Sf9 cells. Original magnification 100,0001. Solid bar represents 100 nm at plane of photograph.
pSPORT-E1 O E4 O L1 construct (Fig. 4A). When protein expression levels were further normalized for mRNA, the pSPORT-E1 O E4 O L1 construct expressed 64 times more E1 O E4 protein than the monocistronic pSPORT-E1 O E4 construct. For L1, the pSPORT-E1 O E4 O L1 and pSPORT-L1 constructs produced equal amounts of mRNA (Fig. 4A, lanes
2 and 3). In contrast, 2.5 times more L1 protein was expressed from the monocistronic pSPORT-L1 construct used as a control than from pSPORT-E1 O E4 O L1 (Fig. 4B, lanes 2 and 3). The pSPORT-XE1 O E4 O L1 construct was made to test the effect of the E1 O E4 ORF, lacking a start codon, on L1 protein expression. As shown in Fig. 5A, transcripts from the three L1-encoding constructs (pSPORT-L1, pSPORT-E1 O E4 O L1, and pSPORT-XE1 O E4 O L1) were of nearly equal abundance (within 5% of each other by scanning densitometry). While the pSPORT-E1 O E4 O L1 construct expressed L1 protein in amounts less than the monocistronic pSPORT-L1 construct, the pSPORTXE1 O E4 O L1 expressed a 54-kDa L1 protein in amounts slightly greater (1.2 times) than that of the monocistronic pSPORT-L1 construct (Fig. 5B). Effects of antisense oligonucleotides
FIG. 4. Transcription and translation from in vitro reactions using the pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs. (A) RNA analysis. A 286-bp E4 O L1 DNA probe was end-labeled with [g-32P]ATP. An equal amount of total RNA was added to each lane. Lane 1, RNA transcribed from pSPORT-E1 O E4; lane 2, RNA from pSPORT-L1; lane 3, RNA from pSPORT-E1 O E4 O L1. Size of RNA fragments (nt) is indicated on the left. (B) Autoradiogram of in vitro transcription/translation of pSPORT-E1 O E4, L1, and E1 O E4 O L1 constructs. Molecular weight marker positions (kDa) are shown on the left. Lane 1, expression from the pSPORT-E1 O E4 construct; lane 2, expression from pSPORT-L1; lane 3, expression from pSPORT-E1 O E4 O L1.
AID
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
The A4E oligonucleotide (antisense to the E1 O E4 start codon region) inhibited E1 O E4 protein production expressed from the pSPORT-E1 O E4 O L1, and an increase in L1 protein production was observed (Fig. 6A). This increase in L1 protein expression occurred at the lowest concentration of A4E (0.05 mM), and was observed at all higher A4E concentrations as well. The A1L oligonucleotide (antisense to the L1 start codon region) inhibited L1 protein from both pSPORT-L1 (not shown) and E1 O E4 O L1 constructs (Fig. 6B). E1 O E4 protein production was not altered by the presence of A1L when expressed from the pSPORT-E1 O E4 O L1 (Fig. 6B). DISCUSSION For HPV 11, and possibly for the other genital HPVs as well, the only L1-containing transcript detected in tissues is the bicistronic E1 O E4 O L1 mRNA, potentially encoding both the E1 O E4 and the L1 proteins (Chow et al.,
vira
AP: Virology
48
BROWN ET AL.
1987). All of the HPV 11 E1 O E4 O L1 splice sites are conserved in the closely related HPV 6 and in other genital HPV types including cancer-associated HPVs such as HPV types 16 and 18. A consensus splice donor site is identically positioned 12 nucleotides downstream from the E4 termination codons in these genital HPV types. The splice acceptor site for the third exon (the L1 ORF) of the E1 O E4 O L1 mRNAs is located at the initiation codon of the L1 ORF. The 5* ends of the HPV 11 E1 O E4 and E1 O E4 O L1 transcripts and the promoter regions that regulate their expression have not been clearly defined. The 5* untranslated sequences in the authentic mRNAs may influence translational efficiency of both transcripts, and the promoters will influence the level of the messages. For the HPV 11 E1 O E4 transcript, Nasseri et al. (1987) cloned and sequenced a cDNA, finding the 5* end originating at nucleotide 716. Primer extension assays were not performed. For HPV 6, Ward and Mounts (1989) identified heterogeneous 5* start sites by primer extension analysis
from several respiratory tract isolates. These 5* start sites ranged from nucleotide 628 to 758, but the location of the primer could not distinguish between the E1 O E4 and the E1 O E4 O L1 transcripts. For HPV 31, an HPV type associated with cervical malignancy, the major L1-encoding transcript was found to be a bicistronic species (E1 O E4 O L1), which initiated at the differentiation-specific promoter located in the middle of the E7 ORF (nt 742) (Hummel et al., 1995). Because the 5* start sites have not been clearly defined for either the HPV 11 E1 O E4 or the E1 O E4 O L1 transcript, and because it is not known if the two transcripts have the same 5* start site, we chose to compare protein expression from the E1 O E4 and E1 O E4 O L1 sequences beginning at the E1 start codon. In addition, the monocistronic pSPORT-L1 construct, used as a control for comparison of expression to the bicistronic E1 O E4 O L1 construct, was also constructed to begin at the L1 start codon. The start codons of all pSPORT constructs thus were an equal distance from the T7 promoter. These constructs therefore allowed direct comparisons of E1 O E4 and L1 translation without the confounding questions of influences of the 5* untranslated region. Admittedly our results could differ if the precise 5* ends of each cDNA were known and included in the pSPORT or baculovirus constructs. While establishing the exact 5* ends of these transcripts was not the goal of the current investigation, subsequent studies will address this issue. Because a potential splice acceptor is present in the E4 ORF at nt 3377, used to produce the E1 O E2C transcript, a splicing event utilizing this acceptor and a baculovirus splice donor could potentially produce a transcript with the L1 AUG being the first one in a favorable context. This event is not likely because no potential splice donor sequence is present in the clone between the polyhedron promoter and the start of the E1 O E4 O L1 sequence. In addition, the E1 O E4 O L1 sequence was cloned as cDNA into the pVL baculovirus expression vector, eliminating the possibility of further or aberrant splicing of the transcript at the E1 splice donor site. For the antisense oligonucleotide studies, inhibition of E1 O E4 translation with A4E resulted in an increase in L1 translation, supporting the scanning model of translation, in which ribosomes begin translation at the first available AUG in a favorable context (Kozak, 1989). Interestingly, the marked increase in L1 translation occurred at very low concentrations of A4E (0.05 mM). In fact, E1 O E4 protein expression was only modestly reduced at these A4E concentrations. Inhibition of L1 translation by the AIL oligonucleotide resulted in no change in E1 O E4 protein expression from the pSPORT-E1 O E4 O L1 construct, suggesting that L1 translation is not directly responsible for the increased level of E1 O E4 protein expression observed from the pSPORT-E1 O E4 O L1 construct compared to pSPORT-E1 O E4. It is likely, however, that contributions of the 3* untranslated region of the
AID
vira
FIG. 6. Effects of antisense oligonucleotides on in vitro transcription/ translation reactions using pSPORT-E1 O E4 O L1. (A) The A4E oligonucleotide was added to reactions. Lane 1, no A4E; lanes 2 through 6, A4E was added to reactions at a final concentration of 0.05, 0.1, 0.2, 0.5, or 1.0 mM. (B) The A1L oligonucleotide was added to reactions. Lane 1, no A1L; lanes 2 and 3, 0.5 and 2 mM A1L added to reactions. Molecular weight marker positions (in kDa) are shown on the left.
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
AP: Virology
HUMAN PAPILLOMAVIRUS TYPE 11 E1 O E4 O L1 TRANSCRIPT
49
bicistronic transcript are at least partially responsible for the observed increased E1 O E4 protein expression. The greater length of the E1 O E4 O L1 transcript compared to the E1 O E4 transcript may also influence translational efficiency, perhaps by stabilizing transcripts. The pSPORT-XE1 O E4 O L1 construct, which lacks an E1 O E4 start codon, was efficiently transcribed and expressed L1 protein with the same efficiency as the monocistronic pSPORT-L1 construct, suggesting that the difference between these constructs with respect to the distance of the T7 promoter to the L1 start codon was of minimal importance. Cell-free systems such as the one in this study are useful tools in analysis of transcriptional and translational efficiency of HPV genes and the effects of antisense oligonucleotides on those genes. Our studies using the wheat germ extract in vitro transcription/translation system show that the cloned E1 O E4 O L1 sequence expresses both the E1 O E4 and the L1 proteins. This observation was also found in eukaryotic cells. We cannot prove that the relative translational efficiencies of the two ORFs in this bicistronic message would be the same in differentiating epithelium. However, we have demonstrated that while transcription of E1 O E4 message is most efficient from the monocistronic pSPORT-E1 O E4 construct, translational efficiency of E1 O E4 is greater from the bicistronic E1 O E4 O L1 construct. A recent study by Stacey et al. (1995) examined translation from the HPV 16 E6 O E7 bicistronic transcript. Using in vitro and cellular translation systems, it was shown that the splicing events leading to the E6* O E7 transcript was not required for E7 synthesis. In addition, site-directed mutagenesis experiments showed that the majority of ribosomes used to translate the E7 ORF ignored some or all of the ORFs upstream from E7. This finding is in contrast to other studies, which suggest that for most structurally bicistronic viral transcripts, generally only the 5* ORF is translated (Kozak, 1986). Although some viral transcripts are bicistronic, the phenomenon of a single mRNA species with the potential to encode both an early and a late protein has not been observed in other virus systems. As mentioned above, the E1 O E4 protein may in reality be a late gene product, even though the sequences are present in the early region of the HPV genome. We and others have shown that E1 O E4 transcripts are found early in HPV 11 infection, but E1 O E4 protein is tightly linked to detection of L1 protein, which appears late in infection, and only in the most differentiated epithelial cells (Brown et al., 1995, 1994b; Stoler and Broker, 1986; Stoler et al., 1990). Most viral mRNAs are monocistronic, but in the case of those transcripts that are bicistronic, ribosomes can reinitiate translation at the second start codon (Kozak, 1984). The efficiency of reinitiation in these cases is generally low (Kozak, 1984). Our analysis of the HPV 11 E1 O E4 O L1 transcript is consistent with the scanning
model for translation as applied to polycistronic viral transcripts (Kozak, 1989). The E1 O E4 O L1 transcript is functionally bicistronic, expressing both the E1 O E4 and the L1 proteins in an in vitro system and in eukaryotic cells.
AID
vira
VY 8042
/
6a1b$$$183
07-07-96 19:41:36
ACKNOWLEDGMENTS This study was supported in part by the Merit Review Board of the Department of Veterans Affairs and the National Institute of Allergy and Infectious Diseases (AI 9416, Project 4). We thank Lutz Gissmann for his helpful advice, and Tonya Moss for secretarial assistance.
REFERENCES Brown, D. R., Chin, M. T., and Strike, D. G. (1988). Identification of human papillomavirus type 11 E4 gene products in human tissue implants from athymic mice. Virology 165, 262–267. Brown, D. R., Bryan, J., Rodriguez, M., Rose, R. C., and Strike, D. G. (1991). Detection of human papillomavirus types 6 and 11 E4 gene products in condylomata acuminatum. J. Med. Virol. 34, 20–28. Brown, D. R., Bryan, J. T., Cramer, H., Katz, B. P., Handy, V., and Fife, K. H. (1994a). Detection of multiple human papillomavirus types in condylomata acuminata from immunosuppressed patients. J. Infect. Dis. 170, 759–765. Brown, D. R., Fan, L., Jones, J., and Bryan, J. (1994b). Colocalization of human papillomavirus type 11 E1 O E4 and L1 proteins in human foreskin implants grown in athymic mice. Virology 201, 46–54. Brown, D. R., Bryan, J. T., Pratt, L., Handy, V., Fife, K. H., and Stoler, M. H. (1995). Human papillomavirus type 11 E1 O E4 and L1 proteins colocalize in the mouse xenograft system at multiple time points. Virology 214, 259–263. Chow, L. T., Nasseri, M., Wolinsky, S. M., and Broker, T. R. (1987). Human papillomavirus types 6 and 11 mRNAs from genital condylomata acuminata. J. Virol. 61, 2581–2588. Christensen, N. D., Ho¨pfl, R., DiAngelo, S. L., Cladel, N. M., Patrick, S. D., Welsh, P. A., Budgeon, L. R., Reed, C. A., and Kreider, J. W. (1994). Assembled baculovirus-expressed human papillomavirus type 11 L1 capsid protein virus-like particles are recognized by neutralizing monoclonal antibodies and induce high titres of neutralizing antibodies. J. Gen. Virol. 75, 2271–2276. Denhardt, D. T. (1966). A membrane-filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Commun. 23, 641– 646. Doorbar, J. (1991). An emerging function for E4. Papillomavirus Rep. 2, 145–147. Doorbar, J., Campbell, D., Grand, R. J. A., and Gallimore, P. H. (1986). Identification of the human papilloma virus-1a E4 gene products. EMBO J. 5, 355–362. Hagensee, M. E., Yaegashi, N., and Galloway, D. A. (1993). Self-assembly of human papillomavirus type 1 capsids by expression of the L1 protein alone or by coexpression of the L1 and L2 capsid proteins. J. Virol. 67, 315–322. Hofmann, K. J., Cook, J. C., Joyce, J. G., Brown, D. R., Schultz, L. D., George, H. A., Rosolowsky, M., Fife, K. H., and Jansen, K. U. (1995). Sequence determination of human papillomavirus type 6a and assembly of virus-like particles in Saccharomyces cerevisiae. Virology 209, 506–518. Hummel, M., Lim, H. B., and Laimins, L. A. (1995). Human papillomavirus type 31b late gene expression is regulated through protein kinase C-mediated changes in RNA processing. J. Virol. 69, 3381–3388. Kozak, M. (1984). Selection of initiation sites by eucaryotic ribosomes: Effect of inserting AUG triplets upstream from the coding sequence for preproinsulin. Nucleic Acids Res. 12, 3873–3893. Kozak, M. (1986). Regulation of protein synthesis in virus-infected animal cells. Adv. Virus Res. 31, 229–292.
AP: Virology
50
BROWN ET AL.
Kozak, M. (1989). The scanning model for translation: An update. J. Cell Biol. 108, 229–241. Nasseri, M., Hirochika, R., Broker, T. R., and Chow, L. T. (1987). A human papillomavirus type 11 transcript encoding an E1 O E4 protein. Virology 159, 433–439. Neary, K., Horwitz, B. H., and DiMaio, D. (1987). Mutational analysis of open reading frame E4 of bovine papillomavirus type 1. J. Virol. 61, 1248–1252. Rose, R. C., Bonnez, W., Reichman, R. C., and Garcea, R. L. (1993). Expression of human papillomavirus type 11 L1 protein in insect cells: In vivo and in vitro assembly of viruslike particles. J. Virol. 67, 1936–1944. Rotenberg, M. O., Chow, L. T., and Broker, T. R. (1989). Characterization of rare human papillomavirus type 11 mRNAs coding for regulatory and structural proteins, using the polymerase chain reaction. Virology 172, 489–497. Stacey, S. N., Jordan, D., Snijders, P. J. F., Mackett, M., Walboomers,
J. M. M., and Arrand, J. R. (1995). Translation of the human papillomavirus type 16 E7 oncoprotein from bicistronic mRNA is independent of splicing events within the E6 open reading frame. J. Virol. 69, 7023–7031. Stoler, M. H., and Broker, T. R. (1986). In situ hybridization detection of human papillomavirus DNAs and messenger RNAs in genital condylomas and a cervical carcinoma. Hum. Pathol. 17, 1250–1258. Stoler, M. H., Whitbeck, A., Wolinsky, S. M., Broker, T. R., Chow, L. T., Howett, M. K., and Kreider, J. W. (1990). Infectious cycle of human papillomavirus type 11 in human foreskin xenografts in nude mice. J. Virol. 64, 3310–3318. Towbin, H., Staehelin, T., and Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350– 4354. Ward, P., and Mounts, P. (1989). Heterogeneity in mRNA of human papillomavirus type-6 subtypes in respiratory tract lesions. Virology 168, 1–12.
AID
vira
VY 8042
/
6a1b$$$184
07-07-96 19:41:36
AP: Virology