- Email: [email protected]
Identification of DNA-protein interactions and enhancer activity at the 5′ end of the upstream regulatory region in human papillomavirus type 11
VIROLOGY
170, 123-l 30 (1989)
Identification of DNA-Protein Interactions and Enhancer Activity at the 5’End of the Upstream Regulatory Region in Human Papillomavirus Type 11 K. J. AUBORN,’ R. L. GALLI, T. P. DILORENZO, AND B. M. STEINBERG Department
of Otolatyngology,
Long IslandJewish
Received November
Medical Center, New Hyde Park, New York 11042
23, 1988; accepted January 25, 1989
We have examined the 5’ end of the noncoding region of the genome of a human papillomavirus, HPV-11, for regulatory elements using permissive host cells. This region of unknown function in the upstream regulatory region (URR) is known to have unusual DNA structure and frequently contains rearrangements which are associated with some more virulent isolates. This 5’ 269-bp fragment was found to exhibit both specific DNA-protein binding using laryngeal papilloma protein extracts and enhancer activity in normal and papillomatous primary laryngeal cells. The viral DNA flanks the Ll open reading frame and does not contain the viral E2 binding site. Three distinct protein binding sites are contained in a 50-bp region of the fragment. This fragment, as a whole, functions as an enhancer in primary laryngeal and papilloma cells when ligated to the SV40 promoter and SV40 T-antigen gene. We conclude that this part of the noncoding region of the papillomaviruses has elements characteristic of regulatory elements in cells permissive for infection by these viruses. o 1989Academic Press, Inc.
Inducible and constitutive enhancer domains have been identified in HPV-16 and HPV-18 (Cripe et al., 1987; Gius et a/., 1988; Gloss et al., 1987). Evidence for three separate enhancer domains is found in HPV18 (Gius et a/., 1988). An enhancer separate from the E2-inducible enhancer is found in HPV-16 which can be induced by glucocorticoid hormones (Gloss et al., 1987). An intriguing part of the URR is the 5’ end, about which little is known either in terms of function and/or regulatory elements. This part of the genome is very purine-thymidine-rich which suggests unusual DNA structure. A FYURRfragment(nt 7218-7544)from HPV6vc, an HPV 6/l l-related isolate from a vulvar carcinoma, exhibited enhancer activity while an analogous fragment from HPV-6b (from a benign lesion) showed only baseline enhancer activity (Rando et al., 198613). The HPV-6vc 5’fragment differs from HPV-6b by having two insertions (Rando et a/., 1986a). Kasher and Roman (1988a) also found insertions in the 5’ end of the URR in a different isolate from a vulvar carcinoma. Additionally, Kasher and Roman (1988b) found that cloning per se generated rearrangements at the 5’ end of the URR. The implication is that this DNA appears to be subject to rearrangements both in eukaryotes and prokaryotes which may alter expression of the virus. Clearly, this part of the genome is an enigma. We wanted to define regulatory elements for the 5’ end of the URR in the typical infection, a benign papilloma. This would be an initial step in determining the significance of this part of the URR. Using HPV-11, the etiological agent of most laryngeal papillomas and many genital condylomata, we have demonstrated that
INTRODUCTlON The human papillomaviruses (HPVs) show a very complex picture of expression. These viruses are very tissue specific; their replication cycle is dependent on the differentiation of keratinocytes, and an infection can be manifested by papillomatous disease or by latency (reviewed by Broker and Botchan, 1986). Additionally, HPVs are believed to be a factor in some malignancies (reviewed by Pfister, 1984; zur Hausen and Schneider, 1988). Understanding viral expression is hampered because no culture system has been found to be permissive for HPV growth. Little is known about the regulatory elements (host or viral) of HPVs that control viral expression. The major regulatory region of the HPVs, a noncoding region often called the upstream regulatory region (URR), was deduced from the more studied bovine papillomavirus, BPV-1, which will express a subset of genes in C-l 27 cells (Law et a/., 1981). This part of the BPV-1 genome has been shown to have sequences for the origin of replication (Waldeck et al., 1984) for plasmid maintenance (Lusky and Botchan, 1984) and enhancer and promoter elements both inducible and repressible by the viral E’2 gene products (Haugen et al., 1987; Lambert et a/., 1987; Spalholz et al., 1987, 1988). An analogous E2-inducible and -repressible enhancer has been identified in HPV-16 (Phelps and Howley, 1987; Cripe eta/., 1987) and HPV-1 1 (Hirochika eta/., 1987, 1988). Sequence data from this article have been deposited with the EMBUGenBank Data Libraries under Accession No, JO4341. ‘To whom requests for reprints should be addressed. 123
0042~6822189 $3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
124
AUBORN
the 5’ end of the URR has specific DNA binding to proteins from permissive tissue and determined the DNA sequences involved. We also found that the 5’ end of the URR is a constitutive enhancer in primary laryngeal (normal and papillomatous) cells which are permissive for infection by this virus. MATERIALS
AND METHODS
Plasmids and DNAs
The 269-bp fragment is a Ddel fragment from cloned HPV-11 (Gissmann et al., 1982) which corresponds to nt 7224 to nt 7496. The plasmids, pPT269A, pPT269B and pPT269X (detailed in text) are constructs of the 269-bp fragment cloned upstream of the SV40 promoter and SV40 T antigen in the plasmid pPT1. This enhancerless plasmid, pPT-1, contains the SV40 sequences nt 128 to nt 2533 cloned into the multiple cloning site of pGem 3 (Promega, Madison, WI). The SV40 sequences were excised from pPP-1, a plasmid with SV40 sequences nt 225 to 2533 cloned into a derivative of pBR322 (courtesy of J. Sambrook, Cold Spring Harbor, NY). Tissues and cells
Surgery discards (courtesy of A. Abramson, Long Island Jewish Medical Center) were immediately frozen in liquid nitrogen prior to their use for protein extracts. Explants used for culture of primary cells were processed immediately. The procedure of culturing of cells has been described (Steinberg et al., 1982). Protein extracts
Preparation of tissue extracts used buffers identical to those for the preparation of cell extracts (Dignam et a/., 1983). Frozen tissues were pulverized in a Braun dismembrator and suspended in extraction buffer consisting of 50 ml\/l tris(hydroxymethyl)aminomethane (Tris) at pH 7.9, 0.42 M KCI, 5 m/M MgCl*, 0.1 mMethylenediaminetetraacetic acid (EDTA), 2 mMdithiothreitol (DlT), 20% glycerol, and 10% sucrose. After centrifugation, the supernatant was precipitated with ammonium sulfate (0.33 g/ml) for greater than 30 min. The precipitate was reconstituted in 40 mM Tris, pH 7.9, 0.1 M KCI, 12.5 mlVI MgCI,, 1 rnM EDTA, 1 mM DlT, and 20% glycerol, dialyzed, and stored at -70”. Protein concentration was measured by the Pierce method of Lowry (Pierce Biochemicals, Rockford, IL).
ET AL.
mn/l DlT, and 10% glycerol at 30” for 30 min. These parameters were based on the binding of proteins to the polyoma virus enhancer (Bohnlein and Gruss, 1986). Gel retardation
Gel retention assays were performed as described by others (Fried and Crothers, 1981; Garner and Revzin, 1981) except that binding reactions were electrophoresed on 3.5% actylamide gels at 100 V in TBE buffer (0.089 MTris-borate, 0.002 M EDTA). Exonuclease
assay
The exonuclease assay was modified from the procedure described by Quinn er al. (1987). Nine units of X exonuclease (Bethesda Research Labs, Gaithersville, MD) was added to each DNA-protein binding reaction and incubated for 10 min at 30”. The reaction was stopped with an equal volume of stop reagent, 10 mM Tris at pH 7.5, 20 ml\/l EDTA, 0.5% sodium dodecyl sulfate (SDS), and deproteinized with phenol. The resulting DNA was electrophoresed at 2000 V on 8% acrylamide, 8 M urea gels and analyzed by autoradiography. Reaction products of Maxam and Gilbert sequencing (1980) of the fragment were electrophoresed alongside exonuclease reactions when mapping was desired. DNase I protection
DNase I protection was by the method described by Galas and Schmitz (1978). Binding reactions were treated with 10 ng of DNase I (Worthington Biochemicals, Fairlawn, NJ) for 5 min at 0”. Reactions were stopped, deproteinized, and electrophoresed identically to exonuclease reactions. Microinjection
of primary cells
This procedure has been described in detail (Steinberg et a/., in press). Briefly, first passage laryngeal epithelial cells were plated on 1O-mm coverslips in MCDB 153 medium (Clonetics, San Diego, CA), cultured, and then shifted to Hams F12 pls 10% fetal calf serum. Constructs of DNA (20 ng/pl or 2 nglpl) in PBS were linearized in the plasmid sequences and injected into the nuclei of cells according to the method of Graessmann and Graessman (1983) within 4 hr of the medium change. After microinjection, cells were incubated at 37” for 24 hr before staining.
DNA binding
Labeled fragment (0.25-l .O ng) and pBR322 DNA (1 pg) were incubated with protein extracts in 10 mNITris, pH 7.9, 50 mM NaCI, 5 mM MgCl*, 0.5 mM EDTA, 1
Assay for T antigen
This assay has been detailed elsewhere (Steinberg et al., in press). Cells on coverslips were fixed with
DNA BINDING AND ENHANCER ACTIVITY IN HPV-11
125
I :1 methanol:acetone and reacted with the monoclonal antibody, pAB-416 (Harlow el a/., 1981) followed by staining by the ABC method (Vector Laboratories, Burlingame, CA) using diaminobenzidine. The percentages of stained nuclei among the injected cells were scored. Measurement of intensity of staining was done using a Kinetek Compact photometer and Kinemate PH software in combination with a Leitz Laborlux microscope. RESULTS Detection
of specific DNA-protein
binding activity
Putative cis regulatory elements (DNA sequences) and corresponding trans elements (DNA binding proteins) have very specific DNA-binding characteristics (reviewed by Dynan and Tjian, 1985). To detect these types of regulatory elements, we have used the gel retardation assay to analyze protein extracts from permissive cells (laryngeal papillomas) for specific DNAprotein binding activity to the 5’end of the HPV-1 1 URR. The URR region of HPV-11 was cloned into pGem 42, excised, and further fragmented. A gel-purified fragment from the 5’ end of the URR was strongly retarded after electrophoresis on neutral gels (data not shown) after incubation with protein extracts made from laryngeal papillomas. The fragment nt 7224 to nt 7496 does not contain the motif for binding to the viral E2 proteins. This 269-bp fragment was cloned into the multiple cloning site of the pGem4Z plasmid, excised (now 301 bp with linker sequences), and tested for specific binding with the laryngeal papilloma extract (Fig. 1). We will continue to refer to this as the 269-bp fragment. Increasing concentrations of protein extract increased the binding activity to the fragment in the presence of 1OOO-foldexcess pBR322 DNA. One hundred excess copies of unlabeled fragment added simultaneously with labeled fragment competed effectively (albeit not totally) for binding with the laryngeal papilloma extract. Presumably, more copies would totally compete the binding especially if prebound to the proteins, but it is possible that some of the binding is not specific. The SV40 genome regulatory region (nt 5171 to nt 294) also competed for binding but not as well as the HPV-1 1 fragment. In retrospect, the partial competition with the SV40 regulatory sequences is not surprising since one of the binding sites was found to be identical to one in the SV40 enhancer (see below). The competition using the fragment or the SV40 DNA should be compared to lane 2 which has the same amount of extract in the binding reaction. Characterization
of the DNA binding reaction
We wanted to determine the number of binding sites and the sequences protected in the binding reaction
FIG. 1. Gel retardation of the 269-bp fragment after binding with increasing concentrations of laryngeal papilloma extracts and competing DNA. In the binding reaction, 0.25 ng of the cloned 269-bp fragment that had been labeled at the 3’ ends with 32Pand 1 pg of pBR322 DNA was bound with 0 (lane l), 1 rg (lane 2) 2 pg (lane 3) 3 pg (lane 4), 5 pg (lane 5) 1 rg (lane 6), and 1 pg (lane 7) of protein from a laryngeal papilloma extract. In addition, lane 6 contained 25 ng of unlabeled fragment, and lane 7 contained 25 ng of a 488-bp fragment containing the SV40 regulatory sequences in the binding reaction. The entire binding reaction was electrophoresed on a 3.5% neutral acrylamide gel and autoradiographed.
using proteins from the laryngeal papilloma tissue. The 269-bp fragment which had been uniquely labeled at the 3’ end of the transcriptional “sense” strand and bound with a laryngeal papilloma extract was exposed to X exonuclease, which degrades from the 5’ ends of DNA. Digested DNA pieces consisting of three sizes were the result (Fig. 2). The shortest piece was seen when low concentrations of extract were used, while the two longer fragments were seen with higher concentrations of extracts. Identical results were obtained with other laryngeal papilloma extracts. Labeling the 3’ end of the “nonsense” strand did not yield any short distinguishable fragments. All papilloma tissues contained either HPV-11 or HPV-6. Our interpretation of the exonuclease protection experiments is that at least three binding sites are present on the 269-bp ,fragment and that these binding sites are near the 3’ end of the “sense” strand. A cartoon of this exonuclease reaction is also shown in Fig. 2. In order to precisely determine which sequences in the 269-bp fragment were bound by the cellular proteins, both the X exonuclease reaction and DNase I protection were compared to the products of the Maxam and Gilbert,sequencing reactions (Fig. 3). Laryngeal papilloma protein extracts were bound to the
126
AUBORN
ET AL.
1234
A ‘A T EGCC
denaturing
c 123
gel
,“-‘, L/
B Inn
l (I*%¶ prot*in)
OR ;-: I * NW.’ /-\ X-4’
j-3
.
\
4 (mom
i-2
pmtdn)
FIG. 2. X exonuclease digestion identified three binding sites in the 269-bp fragment using laryngeal papilloma protein extracts: Protein (0, 1, 3, and 5 Pg in lanes 1, 2, 3, and 4, respectively) from a laryngeal papilloma was used in a binding reaction with 0.25 ng of the 269-bp fragment which had been uniquely labeled with 32P at the 3’ end of the sense strand. Exonuclease was added and the resultant DNA fragments were electrophoresed on a denaturing acrylamide gel and autoradiographed. The schematic interpretation of the result is shown. At low concentrations, the exonuclease reaction is blocked by protein(s) at site 1. At higher concentrations of extracts, binding factors at sites 2 or 3 block the exonuclease reaction and longer DNA fragments are seen on the gel.
fragment which had been uniquely labeled at the 3’end of the “sense” strand and subsequently treated with either exonuclease or DNase I as described under Materials and Methods. DNase I digestion showed a very strongly protected region which we called site 1. The sequence CCAGTGAC was protected, followed by a second similar sequence, CCTGlTAC, at this binding site. These CCNGTNAC sequences are at the very 3’ end of the 269-bp fragment. The sequence is repeated in a single helical turn (10 nucleotides) forming the composite binding site (1 and 1’). The protection at binding sites 2 and 3 are not as defined as those at site 1 in this assay. The binding to these sites is weaker, and this assay requires binding to a majority of the DNA molecules which is not required by the exonuclease assay. Some protection was seen at the exonuclease site 2 which corresponds to the sequence CAATAAACAAT. The actual sequence was based on the subtle increase of bands at the protected borders representing hypersensitivity to the DNase I. This sequence contains a polyadenylation signal and has some homology to the binding site of known CAAT-DNA binding proteins (Chodosh et a/., 1988). Only faint protection was seen at exonuclease site 3, and this sequence, TGTGGAAT, is present in a number of enhancers including SV40 (Johnson et a/., 1987). The sequence of the sense strand of the 269-bp fragment with these three
-1
I1
I 1’
FIG. 3. DNase I footprint analysis of the binding reaction. The 269bp fragment uniquely 3’-end-labeled with 32P (on the sense strand) was subjected to specific chemical degradation of Maxam and Gilbert (A), or binding with the laryngeal papilloma extract and subsequently treated with exonuclease (Et)or DNase I (C) as described under Materials and Methods. The base-specific reactions A, A/G, T/C, and C are shown in (A). The exonuclease reaction with 5 pg of extract is shown in (B) with binding sites 1, 2, and 3 indicated by arrows. DNase l-protected regions are shown (boxed regions) in (C) using 0, 5, and 10 pg of extract. All binding reactions were with 0.25 ng of fragment and 1 fig of pBR322 DNA.
binding sites at the 3’end is shown in Fig. 4. The 3’end of site 1 cannot be conclusively delineated because it represents the end of the fragment; however, the sequence CCNGTNAC which is repeated appears to be the key binding sequence.
269 bp FRAGMENT AAtxCrnA
Ll yuFIFl AAAAmAATA
cAGcccccAA
ACGAAAACGT
ACCAAAACCA
TAlWXlVE
-
GlTATlTATA
-A
BT
E
BATAT
tXUTATAlGT
TIWETATAT
aTVl’AI%Tl’A
‘ISTA’XlTAT
G!IlWITAlWl’
A-
GI?TAt?lWIG
lWl!ATATATl’
Tm
GTAlSTAlUl’
??mWl+&:gBA
Tl’ATGEl’GT 2
“““7
1
FIG. 4. Sequence of the sense strand of the 269-bp fragment with binding sites indicated: The high affinity binding sites (1 and 1’) are boxed with a consensus sequence for the binding site. Site 2 is boxed with a dotted line while site 3 (low affinity binding) is underlined with a dotted line.
DNA BINDING AND ENHANCER
ACTIVITY
127
IN HPV-11 TABLE 1
Constructs
T-ANTIGEN EXPRESSIONWITHTHE 269-bp FRAGMENT T- Antigen 269 bp P C...............,..... a..... .. .. ... .. .. .... . .. . .. . .. . cI. .. .. . .. .. .. .. .. .. . .. .. .. .. .. .. .. . .. ... .. .. .. . A b ’
4
, ’
................................................ B i ........... ...................................
h
'
i ...............................................
.....
r..
............................................
......
W
'
4
'
x
FIG. 5. Structure of the enhancer assay plasmids: The 269-bp fragment, first cloned into the multiple cloning site of pGem 42, was recloned into the enhancerless plasmid pPT-1 containing the promoter and T antigen of SV40 (nt 128 to nt 2533). The “A” orientation contains the 269-bp fragment in the sense orientation relative to the SV40 T antigen. The “B” construct has the fragment in the antisense orientation, The “X” construct contains three copies of the fragment in the sense, antisense, and sense orientation.
Average density positive nuclei
Tang units
11 32 1
9.0 + 3.9 16.8 + 6.9 ND
99 538 ND
9 33 0
11.8 + 5.0 20.1 * 5.1 ND
102 663 ND
Note. Expression of T antigen with the 269-bp fragment in primary laryngeal cells. Approximately 2 ng of plasmid was microinjected into the nuclei of primary cells as described under Materials and Methods. After 20-24 hr, cells were assayed forT antigen by determining the percentage of stained nuclei among injected cells and measuring the intensity of stained nuclei. T-antigen units are the product of these two values and are an expression of the total amount of T antigen expressed by the constructs. Approximately 100 cells were injected per tissue per experiment with each construct, and two tissues were used with each construct.
Enhancer activity of the 269-bp fragment To determine whether this fragment has any enhancer activity, it was cloned upstream of the SV40 promoter and SV40 T-antigen coding sequences. Constructs included the fragment in the “sense” orientation, the opposite or “antisense” orientation, and as a multiple. A cartoon of these plasmids is shown in Fig. 5. Twenty to 40 molecules of each construct were microinjected into the nuclei of cultured laryngeal papilloma cells. T antigen was expressed with all of the constructs (Fig. 6), but not with a similar plasmid, pPT1, containing only the SV40 promoter and SV40 T-antigen sequences but lacking the SV40 enhancer sequences (results not shown). T-antigen expression was better in the sense orientation (A) than in the antisense orientation (B) although expression was increased in both rela-
269A 38%
Normal laryngeal cells pPT269A pPT269X pPT1 Papilloma cells pPT269A pPT269X pPT1
% Positive nuclei
tive to the control construct which yielded only 4% Tantigen-positive cells (data not shown). Three copies of the fragment (construct X) yielded the best T-antigen expression, both in number of cells expressing T antigen and in the intensity of the T-antigen stain. Two to four copies of the constructs pPT1, pPT269A, and pPT269X were then injected into cultured normal laryngeal or cultured laryngeal papilloma keratinocytes and scored for T antigen determining both the percentage of positive cells and the average absorbance of stained nuclei containing T antigen (Table 1). In both normal and papilloma cells, a detectable
2698 30%
269x 55%
FIG. 6. Enhancer activity of constructs. Twenty nanograms of construct was microinjected into the nuclei of cultured primary laryngeal papilloma cells which were incubated with antibody to T antigen (pAB-416) and labeled second antibody and scored for the presence of T antigen 20-24 hr later. The percentage of positive nuclei out of approximately 100 injected cells is indicated. Pictures show the intensity of T-antigen staining. Essentially, no T antigen was detected with the enhancerless plasmid control.
128
AUBORN
amount of T antigen was present with construct A even with this low concentration of DNA. Staining intensities were measured on a total of 37 T-antigen-positive ceils with construct A and 177 for construct X. Considerable T antigen was present with construct X. Essentially no Tantigen was detected with the enhancerless plasmid, pPT1. Similar experiments have resulted in total T-antigen units 2-fold higher than this using either the SV40 enhancers or the entire URR of HPV-1 1 (Steinberg ef a/., in press). We estimate the enhancement due to one copy of the fragment at lo- to 20-fold. Staining densities were not determined on the enhancerless plasmid in experiments shown in the table because the result would be on the basis of a single T-antigen-expressing cell. Results with the 269-bp constructs were virtually identical in the normal and HPV-containing papilloma cells. We have previously shown that keratinocytes derived from laryngeal papilloma tissues contain HPV in copy numbers comparable to the initial tissues (Steinberg et al., 1983, 1985). The specific cells used for enhancer assays were not tested due to the limited number of available cells. Enhancer function is apparently not dependent on any viral transacting factor that may be present in laryngeal papilloma cells. The use of T antigen for the reporter gene enabled us to determine the transient expression of this gene in the limited number of primary cells from the outgrowth of tissue using microinjection. Assays (20-24 hr after injection) were at a time when biological effects of T antigen such as autoregulation of the T-antigen promoter (Hansen et a/., 1981) or possibly heterologous regulation of other promoters as seen with the HPV-18 promoter at 48 hr post-transfection (Thierry et a/., 1987) should be minimal. The use of T antigen as a reporter gene and the effect of copy number for injection of similar constructs have been described (Steinberg et a/., in press). DISCUSSION
We determined that the 5’end of the upstream regulatory region of HPV-1 1 contains two elements of regulation, namely specific DNA-protein binding interactions and enhancer function in permissive cells, although we have not yet established the relatedness of these elements, if any. Determining the existence of putative regulatory elements and a function for these noncoding sequences is a significant step in elucidating the meaning of insertions and/or deletions which appear to be important to the virulence of some variant related HPVs (Boshart and zur Hausen, 1986; Kasher and Roman, 1988a; Rando eta/., 1986a). The specific DNA-protein binding was limited to 50 bp within the 269 bp which have elements in common
ET AL.
with sequences in known enhancers and promoters. More and more, it is apparent in both viruses and eukaryotic genes that sequences in “regulatory” DNA bind to tram regulators modulating expression of DNA. As trans-acting proteins can be variant in different cell types (reviewed by Mantiatis et al., 1987) these binding proteins could serve as a source of gene regulation for these viruses. Importantly, these proteins are in cells permissive for the virus. Concomitantly, we believe that the 269-bp fragment is a constitutive enhancer in permissive cells since it functions equally well in infected and uninfected cells. Additionally, preliminary data suggest that this fragment is functional in a number of different cell types including fibroblasts (data not shown). This DNA does not overlap the E2-dependent enhancer described by Hirochika et al. (1987, 1988) the keratinocyte-dependent enhancer sequences described in HPV-16 (Cripe et a/., 1987) or the glucocorticoid-responsive element described by Gloss et a/. (1987). Interestingly, Guis et al. (1988) have evidence that the 5’ end of HPV-18 may be an enhancer. We believe the fragment is a true enhancer since both orientations greatly increased the expression of T antigen. Since the antisense orientation was not as effective as the sense orientation, we cannot rule out the presence of a promoter in this fragment. Aside from the binding sites and enhancer activity per se, this DNA fragment contains many structural anomolies that are found in regulatory regions. The sense strand is very G-T rich (Fig. 4). G-T DNA can form ZDNA(Hamada eta/., 1984) which, in turn, is in the regulatory region of a number of genes (Hamada et al,, 1982). Additionally, immediately upstream of each of the three binding sites there is a string of T’s or A’s and T’s. These can form bends in DNAs (Koo et a/., 1986) and have been implicated in increased specific binding of proteins in regulatory DNA (Wu and Crothers, 1984) including SV40 DNA (Ryder et al., 1986). We have begun to define some of the components which no doubt contribute to regulatory function in the 5’ end of HPV-1 1 (and HPVs in general) using cells which this virus can infect and thereby cause disease. With this background, we can begin to determine the significance of the insertions (and deletions) found in closely related variant strains by constructing the same types of variations and determining how these variations affect the structure and function of the regulatory elements. We can also determine the relationship, if any, between the various regulatory features of this DNA. ACKNOWLEDGMENTS This work was supported Medical Center Competitive
by grants from the Long Island Jewish Pool Award (K.A., B.S.) and 2POl-NS
DNA BINDING AND ENHANCER 19214 from the National Institute of Neurological and Communicative Disorders and Stroke of the National Institute of Health (B.S., R.G., T.D.).
REFERENCES BOHNLEIN, E., and GRUSS, P. (1986). Interaction of distinct nuclear proteins with sequences controlling the expression of polyomavirus early genes. Mol. Cell. Biol. 6, 1401-l 411. BOSHART, M., and ZUR HAUSEN, H. (1986). Human papillomaviruses in Buschke-Lowenstein tumors: Physical state of the DNA and identification of a tandem duplication in the noncoding region of a human papillomavirus 6 subtype. /. Viral. 58,963-966. BROKER,T. R., and BOTCHAN. M. (1986). Papillomaviruses retrospectives and perspectives. In “Cancer Cells: Vol. 4. DNA Tumor Viruses. Control of Gene Expression and Replication” (M. Botchan, T. Grodzicker, and P. A. Sharp, Eds.) Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. CHODOSH, L. A., BALDWIN, A. S., CARTHEW, R. W., and SHARP, P. A. (1988). Human CCAAT-binding proteins have heterologous subunits. Cell 53, 1 l-24. GRIPE,T. P., HAUGEN, T. H., TURK, J. P., TABATABAI, F., SCHMID, P. G., DURST, J., GISSMANN, L., ROMAN, A., and TUREK, L. P. (1987). Transcriptional regulation of the human papillomavirus-16 E6-E7 promoter by a keratinocyte-dependent enhancer, and by viral E2 trans-activator and repressor gene products: Implications for cervical carcinogenesis. EMBOJ. 6, 3745-3753. DIGNAM, J. D., LEBOWITZ,R. M., and ROEDER, R. G. (1983). Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mamalian nuclei. Nucleic Acids Res. 11, 1475-l 489. DYNAN, W. S., and TJIAN. R. (1985). Control of eukaryotic messenger RNA synthesis by sequence-specific DNA-binding proteins. Nature (London) 316.774-778. FRIED, M., and CROTHERS,D. M. (1981). Equilibria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9, 6505-6624. GALAS, D. J., and SCHMITZ, A. (1978). DNAse footprinting: A simple method for the detection of protein-DNA binding specificity. Nucleic Acids Res. 5, 3 157-3 170. GARNER, M. M., and REVZIN,A. (1981). A gel electrophoresis method for quantifying the binding proteins to specific DNA regions: Application to components of the Escherichia co/i lactose operon regulatory system. Nucleic Acids Res. 9, 3047-3060. GISSMANN, L., DIEHL, V., SCHULZ-C• ULON, H.-J., and ZUR HAUSEN, H. (1982). Molecular cloning and characterization of human papilloma virus DNA derived from a laryngeal papilloma. 1. Viral. 44, 393-400. GUS, D., GROSSMAN, S., BEDELL, M. A., and LAIMINS, L. A. (1988). Inducible and constitutive enhancer domains in the noncoding region of human papillomavirus type 18.1. Viral. 62, 665-672. GLOSS, B., BERNARD, H. U., SEEDORF, K., and KLOCK, G. (1987). The upstream regulatory region of the human papilloma virus-l 6 contains an E2 protein-independent enhancer which is specific for cervical carcinoma cells and regulated by glucocorticoid hormones. EMBOJ. 6,3735-3743. GRAESSMANN,M., and GRAESSMANN,A. (1983). Microinjection of tissue culture cells. In “Methods in Enzymology” (R. Wu, L. Grossman, and K. Moldave, Eds.), Vol. 101, pp. 482-492. Academic Press, New York. HAMADA, H., PETRINO, M. G., and KAKUNAGA,T. (1982). A novel repeated element with Z-DNA-forming potential is widely found in evolutionarily diverse eukaryotic genomes. Proc. Nat/. Acad. Sci. USA 79,6465-6469.
ACTIVITY
IN HPV-11
129
HAMADA, H., PETRINO, M. G., KAKUNAGA,T., SEIDMAN, M., and STOLLAR, B. D. (1984). Characterization of genomic poly(dT-dG) poly(dC-dA) sequences: Structure, organization and conformation. Mol. Cell. Biol. 4,261 O-2621. HANSEN, U., TENEN, D. G., LIVINGSTON,D. M., and SHARP, P. A. (1981). T antigen repression of SV40 early transcription from two promoters. Cell 27,603-612. HARLOW, E., CRAWFORD, L. V., PIM, D. C., and WILLIAMSON, N. M. (1981). Monoclonal antibodies specific for the SV40 tumor antigens. J. Viral. 39, 861-869. HAUGEN, T. H., GRIPE, T. P., GRINDER, G. D., KARIN, M., and TUREK, L. P. (1987). Transactivation of an upstream early gene promoter of bovine papillomavirus-1 by a product of viral E2 gene. EM60 J. 6,145-l 52. HIROCHIKA,H., BROKER,T. R., and CHOW, L. T. (1987). Enhancers and transacting E2 transcriptional factors of papillomaviruses. J. Viral. 61,2599-2606. HIROCHIKA, H., HIROCHIKA, R., BROKER,T. R., and CHOW, L. T. (1988). Functional mapping of the human papillomavirus type 11 transcriptional enhancer and its interaction with the trans-acting E2 proteins. Genes Dev. 2, 54-67. JOHNSON, P. F., LANDSCHUTZ, W. H., GRAVES, B. J., and MCKNIGHT, S. L. (1987). Identification of a rat liver nuclear protein that binds to the enhancer core element of three animal viruses. Genes Dev. 1,133-146. KASHER, M. S., and ROMAN, A. (1988a). Characterization of human papillomavirus type 6b isolated from an invasive squamous carcinoma of the vulva. Virology 165,225-233. KASHER, M. S., and ROMAN, A. (1988b). Alterations in the regulatory region of the human papillomavirus type 6 genome are generated during propagation in Escherichia co/i. J. Viral. 62,3295-3300. Koo, H. S., Wu, H. M., and CROTHERS, D. M. (1986). DNA bending adenine-thymine tracts. Nature (London) 320, 501-506. L~MBERT, P. R., SPALHOLZ, B. A., and HOWLEY, P. M. (1987). A transcriptional repressor encoded by BPV-1 shares a common carboxy-terminal domain with the E2 transactivator. Cell50, 69-78. Law, M., Lowv, D. R., DVORETZKY, I., and HOWLEY, P. M. (1981). Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. Proc. Nat/. Acad. Sci. USA 78,2727-2731. LUSKY, M., and BOTCHAN, M. R. (1984). Characterization of bovine plasmid maintenance sequences. Ce//36,391-401. MANTIATIS, T., GODDEOURN,S., and FISCHER,J. A. (1987). Regulation of inducible and tissue-specific gene expression. Science 236, 1237-l 244. MADAM, A. M., and GILBERT,W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. In “Methods in Enzymology” (L. Grossman and K. Moldave, Eds.), Vol. 65, pp 449-560. Academic Press, New York. PHELPS,W. C., and HOWLEY,P. M. (1987).TranscriptionaI transactivation by human papillomavirus type 16 E2 gene product. 1. Viral. 60, 626-634. PFISTER, H. (1984). Biology and biochemistry of papillomaviruses. Rev. Physiol. Biochem. Pharmacol. 99,ll l-l 81. QUINN, J. P., HOLBROOK,N., and LEVENS,D. (1987). Binding of a cellular protein to the gibbon ape leukemia virus enhancer. Mol. Cell. Biol. 7, 2735-2744. RANDO, R. F.. GROFF. D. E., CHIRIKJIAN,1. G., and LANCASTER,W. D. (1986a). Isolation and characterization of a novel human papillomavirus type 6 DNA from an invasive vulvar carcinoma. 1. Viral. 57, 353-356. RANDO, R. F., LANCASTER,W. D., HAN, P., and LOPEZ, C. (1986b). The noncoding region of HPV-6vc contains two distinct transcriptional enhancing elements. Virology 155, 545-556.
130
AUBORN
RYDER,K., SILVER,S., DELUCIA, A. L., FANNING, E., and TEGTMEYER,P. (1986). An altered DNA conformation in origin region I is a determinant for the binding of SV40 large T antigen. Cell 44, 7 19-725. SPALHOLZ,B. A., LAMBERT,P. F., YEE, C. L., and HOWLEY,P. M. (1987). Localization of the E2 responsive elements of the long control region./. Viral. 61, 1218-1237. SPALHOIZ, B. A., YANG, Y.-C., and HOWLEY,P. M. (1988). Transactivation of a bovine papillomavirus transcriptional element by the E2 gene product. Ce//42,83-191. STEINBERG, B. M., ABRAMSON. A. L., and HOROWITZ, B. (1985). Effects of alpha-interferon on cultured laryngeal papilloma cells. ln “Papilloma Viruses: Molecular and Clinical Aspects,” UCLA Symposia on Molecular and Cellular Biology, New Series (P. M. Howley and T. R. Broker, Eds.), Vol. 32, pp. 413-426. A. R. Liss, New York. STEINBERG,B. M., ABRAMSON.A. L., AND MEADE, R. P. (1982). Culture of human laryngeal papilloma cells in vitro. Ofolaryngol. Head Neck
Surg. 90,728-735. STEINBERG,B. M., AUBORN, K. J., BRANDSMA,J. L., and TAICHMAN, L. B.
ET AL. (1989). Tissue site specific enhancer function of the upstream regulatory region of human papillomavirus type 11 in cultured keratinocytes. J, Viral., in press. STEINBERG,B. M., TOPP, W. C., SCHNEIDER, P., and ABRAMSON, A. (1983). Laryngeal papillomavirus infection during clinical remission. N. Engl. J. Med. 308, 1262-l 264. THIERRY,F., HEARD, J. M.. DARTMANN, K., and YANIV, M. (1987). Characterization of a transcriptional promoter of human papillomavirus and modulation of its expression by simian virus 40 and adenovirus early antigens. J. Viral. 61, 134-l 42. WALDECK, S., ROSL, F., and ZENLGRAF,H. (1984). Origin of replication in episomal bovine papilloma virus type 1 DNA isolated from transformed cells. EMBOI. 3, 2173-2178. Wu, H. M., and CROTHERS,D. M. (1984). The locus of sequence directed and protein-induced DNA bending. Narure (London) 308, 509-513. ZUR HAUSEN, H., and SCHNEIDER,A. (1988). The role of papillomaviruses in human antigenital cancer. In “The Papovaviridae” (N. P. Salzman and P. M. Howley, Eds.), Vol. 2. pp. 245-263. Plenum, New York.
170, 123-l 30 (1989)
Identification of DNA-Protein Interactions and Enhancer Activity at the 5’End of the Upstream Regulatory Region in Human Papillomavirus Type 11 K. J. AUBORN,’ R. L. GALLI, T. P. DILORENZO, AND B. M. STEINBERG Department
of Otolatyngology,
Long IslandJewish
Received November
Medical Center, New Hyde Park, New York 11042
23, 1988; accepted January 25, 1989
We have examined the 5’ end of the noncoding region of the genome of a human papillomavirus, HPV-11, for regulatory elements using permissive host cells. This region of unknown function in the upstream regulatory region (URR) is known to have unusual DNA structure and frequently contains rearrangements which are associated with some more virulent isolates. This 5’ 269-bp fragment was found to exhibit both specific DNA-protein binding using laryngeal papilloma protein extracts and enhancer activity in normal and papillomatous primary laryngeal cells. The viral DNA flanks the Ll open reading frame and does not contain the viral E2 binding site. Three distinct protein binding sites are contained in a 50-bp region of the fragment. This fragment, as a whole, functions as an enhancer in primary laryngeal and papilloma cells when ligated to the SV40 promoter and SV40 T-antigen gene. We conclude that this part of the noncoding region of the papillomaviruses has elements characteristic of regulatory elements in cells permissive for infection by these viruses. o 1989Academic Press, Inc.
Inducible and constitutive enhancer domains have been identified in HPV-16 and HPV-18 (Cripe et al., 1987; Gius et a/., 1988; Gloss et al., 1987). Evidence for three separate enhancer domains is found in HPV18 (Gius et a/., 1988). An enhancer separate from the E2-inducible enhancer is found in HPV-16 which can be induced by glucocorticoid hormones (Gloss et al., 1987). An intriguing part of the URR is the 5’ end, about which little is known either in terms of function and/or regulatory elements. This part of the genome is very purine-thymidine-rich which suggests unusual DNA structure. A FYURRfragment(nt 7218-7544)from HPV6vc, an HPV 6/l l-related isolate from a vulvar carcinoma, exhibited enhancer activity while an analogous fragment from HPV-6b (from a benign lesion) showed only baseline enhancer activity (Rando et al., 198613). The HPV-6vc 5’fragment differs from HPV-6b by having two insertions (Rando et a/., 1986a). Kasher and Roman (1988a) also found insertions in the 5’ end of the URR in a different isolate from a vulvar carcinoma. Additionally, Kasher and Roman (1988b) found that cloning per se generated rearrangements at the 5’ end of the URR. The implication is that this DNA appears to be subject to rearrangements both in eukaryotes and prokaryotes which may alter expression of the virus. Clearly, this part of the genome is an enigma. We wanted to define regulatory elements for the 5’ end of the URR in the typical infection, a benign papilloma. This would be an initial step in determining the significance of this part of the URR. Using HPV-11, the etiological agent of most laryngeal papillomas and many genital condylomata, we have demonstrated that
INTRODUCTlON The human papillomaviruses (HPVs) show a very complex picture of expression. These viruses are very tissue specific; their replication cycle is dependent on the differentiation of keratinocytes, and an infection can be manifested by papillomatous disease or by latency (reviewed by Broker and Botchan, 1986). Additionally, HPVs are believed to be a factor in some malignancies (reviewed by Pfister, 1984; zur Hausen and Schneider, 1988). Understanding viral expression is hampered because no culture system has been found to be permissive for HPV growth. Little is known about the regulatory elements (host or viral) of HPVs that control viral expression. The major regulatory region of the HPVs, a noncoding region often called the upstream regulatory region (URR), was deduced from the more studied bovine papillomavirus, BPV-1, which will express a subset of genes in C-l 27 cells (Law et a/., 1981). This part of the BPV-1 genome has been shown to have sequences for the origin of replication (Waldeck et al., 1984) for plasmid maintenance (Lusky and Botchan, 1984) and enhancer and promoter elements both inducible and repressible by the viral E’2 gene products (Haugen et al., 1987; Lambert et a/., 1987; Spalholz et al., 1987, 1988). An analogous E2-inducible and -repressible enhancer has been identified in HPV-16 (Phelps and Howley, 1987; Cripe eta/., 1987) and HPV-1 1 (Hirochika eta/., 1987, 1988). Sequence data from this article have been deposited with the EMBUGenBank Data Libraries under Accession No, JO4341. ‘To whom requests for reprints should be addressed. 123
0042~6822189 $3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
124
AUBORN
the 5’ end of the URR has specific DNA binding to proteins from permissive tissue and determined the DNA sequences involved. We also found that the 5’ end of the URR is a constitutive enhancer in primary laryngeal (normal and papillomatous) cells which are permissive for infection by this virus. MATERIALS
AND METHODS
Plasmids and DNAs
The 269-bp fragment is a Ddel fragment from cloned HPV-11 (Gissmann et al., 1982) which corresponds to nt 7224 to nt 7496. The plasmids, pPT269A, pPT269B and pPT269X (detailed in text) are constructs of the 269-bp fragment cloned upstream of the SV40 promoter and SV40 T antigen in the plasmid pPT1. This enhancerless plasmid, pPT-1, contains the SV40 sequences nt 128 to nt 2533 cloned into the multiple cloning site of pGem 3 (Promega, Madison, WI). The SV40 sequences were excised from pPP-1, a plasmid with SV40 sequences nt 225 to 2533 cloned into a derivative of pBR322 (courtesy of J. Sambrook, Cold Spring Harbor, NY). Tissues and cells
Surgery discards (courtesy of A. Abramson, Long Island Jewish Medical Center) were immediately frozen in liquid nitrogen prior to their use for protein extracts. Explants used for culture of primary cells were processed immediately. The procedure of culturing of cells has been described (Steinberg et al., 1982). Protein extracts
Preparation of tissue extracts used buffers identical to those for the preparation of cell extracts (Dignam et a/., 1983). Frozen tissues were pulverized in a Braun dismembrator and suspended in extraction buffer consisting of 50 ml\/l tris(hydroxymethyl)aminomethane (Tris) at pH 7.9, 0.42 M KCI, 5 m/M MgCl*, 0.1 mMethylenediaminetetraacetic acid (EDTA), 2 mMdithiothreitol (DlT), 20% glycerol, and 10% sucrose. After centrifugation, the supernatant was precipitated with ammonium sulfate (0.33 g/ml) for greater than 30 min. The precipitate was reconstituted in 40 mM Tris, pH 7.9, 0.1 M KCI, 12.5 mlVI MgCI,, 1 rnM EDTA, 1 mM DlT, and 20% glycerol, dialyzed, and stored at -70”. Protein concentration was measured by the Pierce method of Lowry (Pierce Biochemicals, Rockford, IL).
ET AL.
mn/l DlT, and 10% glycerol at 30” for 30 min. These parameters were based on the binding of proteins to the polyoma virus enhancer (Bohnlein and Gruss, 1986). Gel retardation
Gel retention assays were performed as described by others (Fried and Crothers, 1981; Garner and Revzin, 1981) except that binding reactions were electrophoresed on 3.5% actylamide gels at 100 V in TBE buffer (0.089 MTris-borate, 0.002 M EDTA). Exonuclease
assay
The exonuclease assay was modified from the procedure described by Quinn er al. (1987). Nine units of X exonuclease (Bethesda Research Labs, Gaithersville, MD) was added to each DNA-protein binding reaction and incubated for 10 min at 30”. The reaction was stopped with an equal volume of stop reagent, 10 mM Tris at pH 7.5, 20 ml\/l EDTA, 0.5% sodium dodecyl sulfate (SDS), and deproteinized with phenol. The resulting DNA was electrophoresed at 2000 V on 8% acrylamide, 8 M urea gels and analyzed by autoradiography. Reaction products of Maxam and Gilbert sequencing (1980) of the fragment were electrophoresed alongside exonuclease reactions when mapping was desired. DNase I protection
DNase I protection was by the method described by Galas and Schmitz (1978). Binding reactions were treated with 10 ng of DNase I (Worthington Biochemicals, Fairlawn, NJ) for 5 min at 0”. Reactions were stopped, deproteinized, and electrophoresed identically to exonuclease reactions. Microinjection
of primary cells
This procedure has been described in detail (Steinberg et a/., in press). Briefly, first passage laryngeal epithelial cells were plated on 1O-mm coverslips in MCDB 153 medium (Clonetics, San Diego, CA), cultured, and then shifted to Hams F12 pls 10% fetal calf serum. Constructs of DNA (20 ng/pl or 2 nglpl) in PBS were linearized in the plasmid sequences and injected into the nuclei of cells according to the method of Graessmann and Graessman (1983) within 4 hr of the medium change. After microinjection, cells were incubated at 37” for 24 hr before staining.
DNA binding
Labeled fragment (0.25-l .O ng) and pBR322 DNA (1 pg) were incubated with protein extracts in 10 mNITris, pH 7.9, 50 mM NaCI, 5 mM MgCl*, 0.5 mM EDTA, 1
Assay for T antigen
This assay has been detailed elsewhere (Steinberg et al., in press). Cells on coverslips were fixed with
DNA BINDING AND ENHANCER ACTIVITY IN HPV-11
125
I :1 methanol:acetone and reacted with the monoclonal antibody, pAB-416 (Harlow el a/., 1981) followed by staining by the ABC method (Vector Laboratories, Burlingame, CA) using diaminobenzidine. The percentages of stained nuclei among the injected cells were scored. Measurement of intensity of staining was done using a Kinetek Compact photometer and Kinemate PH software in combination with a Leitz Laborlux microscope. RESULTS Detection
of specific DNA-protein
binding activity
Putative cis regulatory elements (DNA sequences) and corresponding trans elements (DNA binding proteins) have very specific DNA-binding characteristics (reviewed by Dynan and Tjian, 1985). To detect these types of regulatory elements, we have used the gel retardation assay to analyze protein extracts from permissive cells (laryngeal papillomas) for specific DNAprotein binding activity to the 5’end of the HPV-1 1 URR. The URR region of HPV-11 was cloned into pGem 42, excised, and further fragmented. A gel-purified fragment from the 5’ end of the URR was strongly retarded after electrophoresis on neutral gels (data not shown) after incubation with protein extracts made from laryngeal papillomas. The fragment nt 7224 to nt 7496 does not contain the motif for binding to the viral E2 proteins. This 269-bp fragment was cloned into the multiple cloning site of the pGem4Z plasmid, excised (now 301 bp with linker sequences), and tested for specific binding with the laryngeal papilloma extract (Fig. 1). We will continue to refer to this as the 269-bp fragment. Increasing concentrations of protein extract increased the binding activity to the fragment in the presence of 1OOO-foldexcess pBR322 DNA. One hundred excess copies of unlabeled fragment added simultaneously with labeled fragment competed effectively (albeit not totally) for binding with the laryngeal papilloma extract. Presumably, more copies would totally compete the binding especially if prebound to the proteins, but it is possible that some of the binding is not specific. The SV40 genome regulatory region (nt 5171 to nt 294) also competed for binding but not as well as the HPV-1 1 fragment. In retrospect, the partial competition with the SV40 regulatory sequences is not surprising since one of the binding sites was found to be identical to one in the SV40 enhancer (see below). The competition using the fragment or the SV40 DNA should be compared to lane 2 which has the same amount of extract in the binding reaction. Characterization
of the DNA binding reaction
We wanted to determine the number of binding sites and the sequences protected in the binding reaction
FIG. 1. Gel retardation of the 269-bp fragment after binding with increasing concentrations of laryngeal papilloma extracts and competing DNA. In the binding reaction, 0.25 ng of the cloned 269-bp fragment that had been labeled at the 3’ ends with 32Pand 1 pg of pBR322 DNA was bound with 0 (lane l), 1 rg (lane 2) 2 pg (lane 3) 3 pg (lane 4), 5 pg (lane 5) 1 rg (lane 6), and 1 pg (lane 7) of protein from a laryngeal papilloma extract. In addition, lane 6 contained 25 ng of unlabeled fragment, and lane 7 contained 25 ng of a 488-bp fragment containing the SV40 regulatory sequences in the binding reaction. The entire binding reaction was electrophoresed on a 3.5% neutral acrylamide gel and autoradiographed.
using proteins from the laryngeal papilloma tissue. The 269-bp fragment which had been uniquely labeled at the 3’ end of the transcriptional “sense” strand and bound with a laryngeal papilloma extract was exposed to X exonuclease, which degrades from the 5’ ends of DNA. Digested DNA pieces consisting of three sizes were the result (Fig. 2). The shortest piece was seen when low concentrations of extract were used, while the two longer fragments were seen with higher concentrations of extracts. Identical results were obtained with other laryngeal papilloma extracts. Labeling the 3’ end of the “nonsense” strand did not yield any short distinguishable fragments. All papilloma tissues contained either HPV-11 or HPV-6. Our interpretation of the exonuclease protection experiments is that at least three binding sites are present on the 269-bp ,fragment and that these binding sites are near the 3’ end of the “sense” strand. A cartoon of this exonuclease reaction is also shown in Fig. 2. In order to precisely determine which sequences in the 269-bp fragment were bound by the cellular proteins, both the X exonuclease reaction and DNase I protection were compared to the products of the Maxam and Gilbert,sequencing reactions (Fig. 3). Laryngeal papilloma protein extracts were bound to the
126
AUBORN
ET AL.
1234
A ‘A T EGCC
denaturing
c 123
gel
,“-‘, L/
B Inn
l (I*%¶ prot*in)
OR ;-: I * NW.’ /-\ X-4’
j-3
.
\
4 (mom
i-2
pmtdn)
FIG. 2. X exonuclease digestion identified three binding sites in the 269-bp fragment using laryngeal papilloma protein extracts: Protein (0, 1, 3, and 5 Pg in lanes 1, 2, 3, and 4, respectively) from a laryngeal papilloma was used in a binding reaction with 0.25 ng of the 269-bp fragment which had been uniquely labeled with 32P at the 3’ end of the sense strand. Exonuclease was added and the resultant DNA fragments were electrophoresed on a denaturing acrylamide gel and autoradiographed. The schematic interpretation of the result is shown. At low concentrations, the exonuclease reaction is blocked by protein(s) at site 1. At higher concentrations of extracts, binding factors at sites 2 or 3 block the exonuclease reaction and longer DNA fragments are seen on the gel.
fragment which had been uniquely labeled at the 3’end of the “sense” strand and subsequently treated with either exonuclease or DNase I as described under Materials and Methods. DNase I digestion showed a very strongly protected region which we called site 1. The sequence CCAGTGAC was protected, followed by a second similar sequence, CCTGlTAC, at this binding site. These CCNGTNAC sequences are at the very 3’ end of the 269-bp fragment. The sequence is repeated in a single helical turn (10 nucleotides) forming the composite binding site (1 and 1’). The protection at binding sites 2 and 3 are not as defined as those at site 1 in this assay. The binding to these sites is weaker, and this assay requires binding to a majority of the DNA molecules which is not required by the exonuclease assay. Some protection was seen at the exonuclease site 2 which corresponds to the sequence CAATAAACAAT. The actual sequence was based on the subtle increase of bands at the protected borders representing hypersensitivity to the DNase I. This sequence contains a polyadenylation signal and has some homology to the binding site of known CAAT-DNA binding proteins (Chodosh et a/., 1988). Only faint protection was seen at exonuclease site 3, and this sequence, TGTGGAAT, is present in a number of enhancers including SV40 (Johnson et a/., 1987). The sequence of the sense strand of the 269-bp fragment with these three
-1
I1
I 1’
FIG. 3. DNase I footprint analysis of the binding reaction. The 269bp fragment uniquely 3’-end-labeled with 32P (on the sense strand) was subjected to specific chemical degradation of Maxam and Gilbert (A), or binding with the laryngeal papilloma extract and subsequently treated with exonuclease (Et)or DNase I (C) as described under Materials and Methods. The base-specific reactions A, A/G, T/C, and C are shown in (A). The exonuclease reaction with 5 pg of extract is shown in (B) with binding sites 1, 2, and 3 indicated by arrows. DNase l-protected regions are shown (boxed regions) in (C) using 0, 5, and 10 pg of extract. All binding reactions were with 0.25 ng of fragment and 1 fig of pBR322 DNA.
binding sites at the 3’end is shown in Fig. 4. The 3’end of site 1 cannot be conclusively delineated because it represents the end of the fragment; however, the sequence CCNGTNAC which is repeated appears to be the key binding sequence.
269 bp FRAGMENT AAtxCrnA
Ll yuFIFl AAAAmAATA
cAGcccccAA
ACGAAAACGT
ACCAAAACCA
TAlWXlVE
-
GlTATlTATA
-A
BT
E
BATAT
tXUTATAlGT
TIWETATAT
aTVl’AI%Tl’A
‘ISTA’XlTAT
G!IlWITAlWl’
A-
GI?TAt?lWIG
lWl!ATATATl’
Tm
GTAlSTAlUl’
??mWl+&:gBA
Tl’ATGEl’GT 2
“““7
1
FIG. 4. Sequence of the sense strand of the 269-bp fragment with binding sites indicated: The high affinity binding sites (1 and 1’) are boxed with a consensus sequence for the binding site. Site 2 is boxed with a dotted line while site 3 (low affinity binding) is underlined with a dotted line.
DNA BINDING AND ENHANCER
ACTIVITY
127
IN HPV-11 TABLE 1
Constructs
T-ANTIGEN EXPRESSIONWITHTHE 269-bp FRAGMENT T- Antigen 269 bp P C...............,..... a..... .. .. ... .. .. .... . .. . .. . .. . cI. .. .. . .. .. .. .. .. .. . .. .. .. .. .. .. .. . .. ... .. .. .. . A b ’
4
, ’
................................................ B i ........... ...................................
h
'
i ...............................................
.....
r..
............................................
......
W
'
4
'
x
FIG. 5. Structure of the enhancer assay plasmids: The 269-bp fragment, first cloned into the multiple cloning site of pGem 42, was recloned into the enhancerless plasmid pPT-1 containing the promoter and T antigen of SV40 (nt 128 to nt 2533). The “A” orientation contains the 269-bp fragment in the sense orientation relative to the SV40 T antigen. The “B” construct has the fragment in the antisense orientation, The “X” construct contains three copies of the fragment in the sense, antisense, and sense orientation.
Average density positive nuclei
Tang units
11 32 1
9.0 + 3.9 16.8 + 6.9 ND
99 538 ND
9 33 0
11.8 + 5.0 20.1 * 5.1 ND
102 663 ND
Note. Expression of T antigen with the 269-bp fragment in primary laryngeal cells. Approximately 2 ng of plasmid was microinjected into the nuclei of primary cells as described under Materials and Methods. After 20-24 hr, cells were assayed forT antigen by determining the percentage of stained nuclei among injected cells and measuring the intensity of stained nuclei. T-antigen units are the product of these two values and are an expression of the total amount of T antigen expressed by the constructs. Approximately 100 cells were injected per tissue per experiment with each construct, and two tissues were used with each construct.
Enhancer activity of the 269-bp fragment To determine whether this fragment has any enhancer activity, it was cloned upstream of the SV40 promoter and SV40 T-antigen coding sequences. Constructs included the fragment in the “sense” orientation, the opposite or “antisense” orientation, and as a multiple. A cartoon of these plasmids is shown in Fig. 5. Twenty to 40 molecules of each construct were microinjected into the nuclei of cultured laryngeal papilloma cells. T antigen was expressed with all of the constructs (Fig. 6), but not with a similar plasmid, pPT1, containing only the SV40 promoter and SV40 T-antigen sequences but lacking the SV40 enhancer sequences (results not shown). T-antigen expression was better in the sense orientation (A) than in the antisense orientation (B) although expression was increased in both rela-
269A 38%
Normal laryngeal cells pPT269A pPT269X pPT1 Papilloma cells pPT269A pPT269X pPT1
% Positive nuclei
tive to the control construct which yielded only 4% Tantigen-positive cells (data not shown). Three copies of the fragment (construct X) yielded the best T-antigen expression, both in number of cells expressing T antigen and in the intensity of the T-antigen stain. Two to four copies of the constructs pPT1, pPT269A, and pPT269X were then injected into cultured normal laryngeal or cultured laryngeal papilloma keratinocytes and scored for T antigen determining both the percentage of positive cells and the average absorbance of stained nuclei containing T antigen (Table 1). In both normal and papilloma cells, a detectable
2698 30%
269x 55%
FIG. 6. Enhancer activity of constructs. Twenty nanograms of construct was microinjected into the nuclei of cultured primary laryngeal papilloma cells which were incubated with antibody to T antigen (pAB-416) and labeled second antibody and scored for the presence of T antigen 20-24 hr later. The percentage of positive nuclei out of approximately 100 injected cells is indicated. Pictures show the intensity of T-antigen staining. Essentially, no T antigen was detected with the enhancerless plasmid control.
128
AUBORN
amount of T antigen was present with construct A even with this low concentration of DNA. Staining intensities were measured on a total of 37 T-antigen-positive ceils with construct A and 177 for construct X. Considerable T antigen was present with construct X. Essentially no Tantigen was detected with the enhancerless plasmid, pPT1. Similar experiments have resulted in total T-antigen units 2-fold higher than this using either the SV40 enhancers or the entire URR of HPV-1 1 (Steinberg ef a/., in press). We estimate the enhancement due to one copy of the fragment at lo- to 20-fold. Staining densities were not determined on the enhancerless plasmid in experiments shown in the table because the result would be on the basis of a single T-antigen-expressing cell. Results with the 269-bp constructs were virtually identical in the normal and HPV-containing papilloma cells. We have previously shown that keratinocytes derived from laryngeal papilloma tissues contain HPV in copy numbers comparable to the initial tissues (Steinberg et al., 1983, 1985). The specific cells used for enhancer assays were not tested due to the limited number of available cells. Enhancer function is apparently not dependent on any viral transacting factor that may be present in laryngeal papilloma cells. The use of T antigen for the reporter gene enabled us to determine the transient expression of this gene in the limited number of primary cells from the outgrowth of tissue using microinjection. Assays (20-24 hr after injection) were at a time when biological effects of T antigen such as autoregulation of the T-antigen promoter (Hansen et a/., 1981) or possibly heterologous regulation of other promoters as seen with the HPV-18 promoter at 48 hr post-transfection (Thierry et a/., 1987) should be minimal. The use of T antigen as a reporter gene and the effect of copy number for injection of similar constructs have been described (Steinberg et a/., in press). DISCUSSION
We determined that the 5’end of the upstream regulatory region of HPV-1 1 contains two elements of regulation, namely specific DNA-protein binding interactions and enhancer function in permissive cells, although we have not yet established the relatedness of these elements, if any. Determining the existence of putative regulatory elements and a function for these noncoding sequences is a significant step in elucidating the meaning of insertions and/or deletions which appear to be important to the virulence of some variant related HPVs (Boshart and zur Hausen, 1986; Kasher and Roman, 1988a; Rando eta/., 1986a). The specific DNA-protein binding was limited to 50 bp within the 269 bp which have elements in common
ET AL.
with sequences in known enhancers and promoters. More and more, it is apparent in both viruses and eukaryotic genes that sequences in “regulatory” DNA bind to tram regulators modulating expression of DNA. As trans-acting proteins can be variant in different cell types (reviewed by Mantiatis et al., 1987) these binding proteins could serve as a source of gene regulation for these viruses. Importantly, these proteins are in cells permissive for the virus. Concomitantly, we believe that the 269-bp fragment is a constitutive enhancer in permissive cells since it functions equally well in infected and uninfected cells. Additionally, preliminary data suggest that this fragment is functional in a number of different cell types including fibroblasts (data not shown). This DNA does not overlap the E2-dependent enhancer described by Hirochika et al. (1987, 1988) the keratinocyte-dependent enhancer sequences described in HPV-16 (Cripe et a/., 1987) or the glucocorticoid-responsive element described by Gloss et a/. (1987). Interestingly, Guis et al. (1988) have evidence that the 5’ end of HPV-18 may be an enhancer. We believe the fragment is a true enhancer since both orientations greatly increased the expression of T antigen. Since the antisense orientation was not as effective as the sense orientation, we cannot rule out the presence of a promoter in this fragment. Aside from the binding sites and enhancer activity per se, this DNA fragment contains many structural anomolies that are found in regulatory regions. The sense strand is very G-T rich (Fig. 4). G-T DNA can form ZDNA(Hamada eta/., 1984) which, in turn, is in the regulatory region of a number of genes (Hamada et al,, 1982). Additionally, immediately upstream of each of the three binding sites there is a string of T’s or A’s and T’s. These can form bends in DNAs (Koo et a/., 1986) and have been implicated in increased specific binding of proteins in regulatory DNA (Wu and Crothers, 1984) including SV40 DNA (Ryder et al., 1986). We have begun to define some of the components which no doubt contribute to regulatory function in the 5’ end of HPV-1 1 (and HPVs in general) using cells which this virus can infect and thereby cause disease. With this background, we can begin to determine the significance of the insertions (and deletions) found in closely related variant strains by constructing the same types of variations and determining how these variations affect the structure and function of the regulatory elements. We can also determine the relationship, if any, between the various regulatory features of this DNA. ACKNOWLEDGMENTS This work was supported Medical Center Competitive
by grants from the Long Island Jewish Pool Award (K.A., B.S.) and 2POl-NS
DNA BINDING AND ENHANCER 19214 from the National Institute of Neurological and Communicative Disorders and Stroke of the National Institute of Health (B.S., R.G., T.D.).
REFERENCES BOHNLEIN, E., and GRUSS, P. (1986). Interaction of distinct nuclear proteins with sequences controlling the expression of polyomavirus early genes. Mol. Cell. Biol. 6, 1401-l 411. BOSHART, M., and ZUR HAUSEN, H. (1986). Human papillomaviruses in Buschke-Lowenstein tumors: Physical state of the DNA and identification of a tandem duplication in the noncoding region of a human papillomavirus 6 subtype. /. Viral. 58,963-966. BROKER,T. R., and BOTCHAN. M. (1986). Papillomaviruses retrospectives and perspectives. In “Cancer Cells: Vol. 4. DNA Tumor Viruses. Control of Gene Expression and Replication” (M. Botchan, T. Grodzicker, and P. A. Sharp, Eds.) Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. CHODOSH, L. A., BALDWIN, A. S., CARTHEW, R. W., and SHARP, P. A. (1988). Human CCAAT-binding proteins have heterologous subunits. Cell 53, 1 l-24. GRIPE,T. P., HAUGEN, T. H., TURK, J. P., TABATABAI, F., SCHMID, P. G., DURST, J., GISSMANN, L., ROMAN, A., and TUREK, L. P. (1987). Transcriptional regulation of the human papillomavirus-16 E6-E7 promoter by a keratinocyte-dependent enhancer, and by viral E2 trans-activator and repressor gene products: Implications for cervical carcinogenesis. EMBOJ. 6, 3745-3753. DIGNAM, J. D., LEBOWITZ,R. M., and ROEDER, R. G. (1983). Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mamalian nuclei. Nucleic Acids Res. 11, 1475-l 489. DYNAN, W. S., and TJIAN. R. (1985). Control of eukaryotic messenger RNA synthesis by sequence-specific DNA-binding proteins. Nature (London) 316.774-778. FRIED, M., and CROTHERS,D. M. (1981). Equilibria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9, 6505-6624. GALAS, D. J., and SCHMITZ, A. (1978). DNAse footprinting: A simple method for the detection of protein-DNA binding specificity. Nucleic Acids Res. 5, 3 157-3 170. GARNER, M. M., and REVZIN,A. (1981). A gel electrophoresis method for quantifying the binding proteins to specific DNA regions: Application to components of the Escherichia co/i lactose operon regulatory system. Nucleic Acids Res. 9, 3047-3060. GISSMANN, L., DIEHL, V., SCHULZ-C• ULON, H.-J., and ZUR HAUSEN, H. (1982). Molecular cloning and characterization of human papilloma virus DNA derived from a laryngeal papilloma. 1. Viral. 44, 393-400. GUS, D., GROSSMAN, S., BEDELL, M. A., and LAIMINS, L. A. (1988). Inducible and constitutive enhancer domains in the noncoding region of human papillomavirus type 18.1. Viral. 62, 665-672. GLOSS, B., BERNARD, H. U., SEEDORF, K., and KLOCK, G. (1987). The upstream regulatory region of the human papilloma virus-l 6 contains an E2 protein-independent enhancer which is specific for cervical carcinoma cells and regulated by glucocorticoid hormones. EMBOJ. 6,3735-3743. GRAESSMANN,M., and GRAESSMANN,A. (1983). Microinjection of tissue culture cells. In “Methods in Enzymology” (R. Wu, L. Grossman, and K. Moldave, Eds.), Vol. 101, pp. 482-492. Academic Press, New York. HAMADA, H., PETRINO, M. G., and KAKUNAGA,T. (1982). A novel repeated element with Z-DNA-forming potential is widely found in evolutionarily diverse eukaryotic genomes. Proc. Nat/. Acad. Sci. USA 79,6465-6469.
ACTIVITY
IN HPV-11
129
HAMADA, H., PETRINO, M. G., KAKUNAGA,T., SEIDMAN, M., and STOLLAR, B. D. (1984). Characterization of genomic poly(dT-dG) poly(dC-dA) sequences: Structure, organization and conformation. Mol. Cell. Biol. 4,261 O-2621. HANSEN, U., TENEN, D. G., LIVINGSTON,D. M., and SHARP, P. A. (1981). T antigen repression of SV40 early transcription from two promoters. Cell 27,603-612. HARLOW, E., CRAWFORD, L. V., PIM, D. C., and WILLIAMSON, N. M. (1981). Monoclonal antibodies specific for the SV40 tumor antigens. J. Viral. 39, 861-869. HAUGEN, T. H., GRIPE, T. P., GRINDER, G. D., KARIN, M., and TUREK, L. P. (1987). Transactivation of an upstream early gene promoter of bovine papillomavirus-1 by a product of viral E2 gene. EM60 J. 6,145-l 52. HIROCHIKA,H., BROKER,T. R., and CHOW, L. T. (1987). Enhancers and transacting E2 transcriptional factors of papillomaviruses. J. Viral. 61,2599-2606. HIROCHIKA, H., HIROCHIKA, R., BROKER,T. R., and CHOW, L. T. (1988). Functional mapping of the human papillomavirus type 11 transcriptional enhancer and its interaction with the trans-acting E2 proteins. Genes Dev. 2, 54-67. JOHNSON, P. F., LANDSCHUTZ, W. H., GRAVES, B. J., and MCKNIGHT, S. L. (1987). Identification of a rat liver nuclear protein that binds to the enhancer core element of three animal viruses. Genes Dev. 1,133-146. KASHER, M. S., and ROMAN, A. (1988a). Characterization of human papillomavirus type 6b isolated from an invasive squamous carcinoma of the vulva. Virology 165,225-233. KASHER, M. S., and ROMAN, A. (1988b). Alterations in the regulatory region of the human papillomavirus type 6 genome are generated during propagation in Escherichia co/i. J. Viral. 62,3295-3300. Koo, H. S., Wu, H. M., and CROTHERS, D. M. (1986). DNA bending adenine-thymine tracts. Nature (London) 320, 501-506. L~MBERT, P. R., SPALHOLZ, B. A., and HOWLEY, P. M. (1987). A transcriptional repressor encoded by BPV-1 shares a common carboxy-terminal domain with the E2 transactivator. Cell50, 69-78. Law, M., Lowv, D. R., DVORETZKY, I., and HOWLEY, P. M. (1981). Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. Proc. Nat/. Acad. Sci. USA 78,2727-2731. LUSKY, M., and BOTCHAN, M. R. (1984). Characterization of bovine plasmid maintenance sequences. Ce//36,391-401. MANTIATIS, T., GODDEOURN,S., and FISCHER,J. A. (1987). Regulation of inducible and tissue-specific gene expression. Science 236, 1237-l 244. MADAM, A. M., and GILBERT,W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. In “Methods in Enzymology” (L. Grossman and K. Moldave, Eds.), Vol. 65, pp 449-560. Academic Press, New York. PHELPS,W. C., and HOWLEY,P. M. (1987).TranscriptionaI transactivation by human papillomavirus type 16 E2 gene product. 1. Viral. 60, 626-634. PFISTER, H. (1984). Biology and biochemistry of papillomaviruses. Rev. Physiol. Biochem. Pharmacol. 99,ll l-l 81. QUINN, J. P., HOLBROOK,N., and LEVENS,D. (1987). Binding of a cellular protein to the gibbon ape leukemia virus enhancer. Mol. Cell. Biol. 7, 2735-2744. RANDO, R. F.. GROFF. D. E., CHIRIKJIAN,1. G., and LANCASTER,W. D. (1986a). Isolation and characterization of a novel human papillomavirus type 6 DNA from an invasive vulvar carcinoma. 1. Viral. 57, 353-356. RANDO, R. F., LANCASTER,W. D., HAN, P., and LOPEZ, C. (1986b). The noncoding region of HPV-6vc contains two distinct transcriptional enhancing elements. Virology 155, 545-556.
130
AUBORN
RYDER,K., SILVER,S., DELUCIA, A. L., FANNING, E., and TEGTMEYER,P. (1986). An altered DNA conformation in origin region I is a determinant for the binding of SV40 large T antigen. Cell 44, 7 19-725. SPALHOLZ,B. A., LAMBERT,P. F., YEE, C. L., and HOWLEY,P. M. (1987). Localization of the E2 responsive elements of the long control region./. Viral. 61, 1218-1237. SPALHOIZ, B. A., YANG, Y.-C., and HOWLEY,P. M. (1988). Transactivation of a bovine papillomavirus transcriptional element by the E2 gene product. Ce//42,83-191. STEINBERG, B. M., ABRAMSON. A. L., and HOROWITZ, B. (1985). Effects of alpha-interferon on cultured laryngeal papilloma cells. ln “Papilloma Viruses: Molecular and Clinical Aspects,” UCLA Symposia on Molecular and Cellular Biology, New Series (P. M. Howley and T. R. Broker, Eds.), Vol. 32, pp. 413-426. A. R. Liss, New York. STEINBERG,B. M., ABRAMSON.A. L., AND MEADE, R. P. (1982). Culture of human laryngeal papilloma cells in vitro. Ofolaryngol. Head Neck
Surg. 90,728-735. STEINBERG,B. M., AUBORN, K. J., BRANDSMA,J. L., and TAICHMAN, L. B.
ET AL. (1989). Tissue site specific enhancer function of the upstream regulatory region of human papillomavirus type 11 in cultured keratinocytes. J, Viral., in press. STEINBERG,B. M., TOPP, W. C., SCHNEIDER, P., and ABRAMSON, A. (1983). Laryngeal papillomavirus infection during clinical remission. N. Engl. J. Med. 308, 1262-l 264. THIERRY,F., HEARD, J. M.. DARTMANN, K., and YANIV, M. (1987). Characterization of a transcriptional promoter of human papillomavirus and modulation of its expression by simian virus 40 and adenovirus early antigens. J. Viral. 61, 134-l 42. WALDECK, S., ROSL, F., and ZENLGRAF,H. (1984). Origin of replication in episomal bovine papilloma virus type 1 DNA isolated from transformed cells. EMBOI. 3, 2173-2178. Wu, H. M., and CROTHERS,D. M. (1984). The locus of sequence directed and protein-induced DNA bending. Narure (London) 308, 509-513. ZUR HAUSEN, H., and SCHNEIDER,A. (1988). The role of papillomaviruses in human antigenital cancer. In “The Papovaviridae” (N. P. Salzman and P. M. Howley, Eds.), Vol. 2. pp. 245-263. Plenum, New York.