LTR with BamHI in ACH2 and 8E5 cells and with Earl inUl cells. DNA probes. ...... Jones, K. A., J. T. Kadonaga, P. A. Luciw, and R. Tjian. 1986. Activation of the ...
JOURNAL OF VIROLOGY, Dec. 1991, p. 6790-6799 0022-538X/91/126790-10$02.00/0 Copyright © 1991, American Society for Microbiology
Vol. 65, No. 12
DNase I-Hypersensitive Sites Are Associated with Both Long Terminal Repeats and with the Intragenic Enhancer of Integrated Human Immunodeficiency Virus Type 1 ERIC VERDIN
Laboratory of Viral and Molecular Pathogenesis, National Institute of Neurological Disorders and Stroke, Building 36, Room 5C22, Bethesda, Maryland 20892 Received 12 July 1991/Accepted 16 September 1991
After reverse transcription and integration of the genome of human immunodeficiency virus (HIV) in a target cell, the viral DNA becomes packaged into chromatin. Regions of chromatin associated with regulatory functions in eukaryotes can generally be distinguished from the bulk of chromatin by an increased accessibility of the DNA to nucleases (nuclease-hypersensitive sites). In this report, the chromatin structure of the complete HIV-1 genome has been analyzed in three chronically infected cell lines of monocyte/macrophage and lymphoid origins. Digestion of purified nuclei from these cells with DNase I followed by restriction digestion and Southern blotting identified several DNase I-hypersensitive regions throughout the viral genome. Two constitutive sites were associated with the U3 region of the 5' long terminal repeat (LTR) in which the viral promoter and enhancer are located. An additional site in the R region of the 5' LTR was present only after activation of viral transcription by phorbol ester or tumor necrosis factor alpha. A fourth site was identified in all cell lines downstream of the 5' LTR (nucleotides [nt] 656 to 720), and the band corresponding to this site decreased in intensity upon activation of transcription. In the 3' LTR, a constitutive hypersensitive site was identified in all cell lines (nt 9322 to 9489). A major site (nt 4534 to 4733) was present only in a cell line of macrophage/monocyte origin in a region of the genome in which an intragenic enhancer was recently identified (E. Verdin, N. Becker, F. Bex, L. Droogmans, and A. Burny, Proc. Natl. Acad. Sci. USA 87:4874-4878, 1990). This study defines regions of the HIV genome associated with an open chromatin configuration and points to the potential regulatory role of these elements in the HIV life cycle. After entering a host cell, the RNA genome of human immunodeficiency virus type 1 (HIV-1) is transcribed into double-stranded DNA by the viral reverse transcriptase and then becomes integrated into the cell genome. Once integrated, the expression of the HIV genome is under the combined influence of trans-acting cellular and viral regulatory factors and the local chromatin environment at the site of integration. These factors exert their regulatory functions by interacting with viral cis-acting elements, both at the DNA and RNA levels (for reviews, see references 5, 6, 17, 21, 30, 36, 38, and 49). At the transcriptional level, viral long terminal repeats (LTRs), which are present at both the 5' and 3' extremities of the viral genome, contain all cis-acting elements necessary for transcription initiation and termination. Transcription is initiated in the 5' LTR, and several transcription factors have been shown to bind in vitro to each of the four functionally defined regions in this element. From the 5' end to the 3' end, the four regions are as follows: first, the negative regulatory element, a silencer (16, 37, 41, 58) which contains binding sites for at least four distinct factors (15, 16, 29, 40, 42); second, the enhancer (16, 37, 41, 58) has been shown to interact with at least three distinct proteins: NF-KB (27), HIVEN86A (14), and EBP-1 (56, 57); third, the promoter contains three Spl binding sites (22), a TATA box, and additional sites close to the site of initiation of transcription (21); fourth, downstream of the site of initiation of transcription is the Tat-responsive element (37) and binding sites for three factors: LBP-1 or UBP-1, TCF-1, and CTF (21, 23, 56, 57). In addition to these elements, we have recently identified a new element in the pol gene of HIV presenting the characteristics of a transcriptional enhancer (47). This element is tetradecanoylphorbol acetate (TPA)
inducible in HeLa cells and composed of two distinct subdomains (47). Despite rapid growth in our knowledge of viral and cellular factors involved in HIV gene regulation, little is known on the relevance in vivo of the interactions between transcription factors and viral elements observed in vitro. Indeed, since the DNA of eukaryotes is organized in nucleosomes, in which DNA is wrapped around histone octamer cores, it is probable that the accessibility of DNA to soluble regulatory factors is restricted. Consequently, factors identified by virtue of their ability to bind in vitro to viral cis-acting elements might never gain access to their DNA target in vivo. For this reason, it has been suggested that cis-acting DNA elements of eukaryotes might either be nucleosomefree or that the DNA-nucleosome complex is in a more open configuration. As a consequence, DNA present in those regions should become more sensitive to a variety of chemical or enzymatic probes. Pioneering studies demonstrated that discrete regions of the simian virus 40 and Drosophila genome were hypersensitive to digestion with DNase I in vivo (39, 46, 55). Several studies, encompassing numerous viral and cellular genes, have since generalized these observations: DNase I-hypersensitive sites have been found in association with a large variety of cis-acting elements including promoters, enhancers, upstream activating sequences, silencers, terminators, recombination loci, telomeres, and centromeres (for a review, see reference 19). Moreover, variation in the sensitivity of a particular hypersensitive site has been found to precede or to accompany gene activation or silencing in a number of systems, thus strengthening the correlation between these sites and gene regulation (19). We have initiated studies aimed at defining in molecular 6790
NUCLEASE-HYPERSENSITIVE SITES IN HIV CHROMATIN
VOL. 65, 1991
taking place in vivo between cellular and viral regulatory factors and the viral DNA elements. Our first step consisted in defining chromatin boundaries and examining the HIV genome for possible nuclease-hypersensitive sites. We have used three well-characterized chronically infected cell lines, ACH2 and 8E5 (both derived from the CEM cell line, a CD4+ lymphoid cell line [3, 13]) and Ul (derived from the U937 cell line, a monocyte/macrophage cell line [12]). Two of these cell lines (ACH2 and Ul) produce very low amounts of virus under basal conditions and can be induced to produce virus when stimulated with specific cytokines, phorbol esters, and other agents (3, 10-12, 18, 24, 31-34, 43, 44). The 8E5 cell line constitutively produces virus (13). These cells were used to examine the virus chromatin in a variety of cellular environments and under different levels of transcription rates. These experiments revealed the presence of at least five major hypersensitive sites in the Ul cell line and four major sites in the ACH2 and 8E5 cell lines. These sites were associated with previously recognized cis-acting elements located in the 5' and 3' LTRs and in the intragenic enhancer (47). In addition, other regions of the HIV genome were also identified downstream of the 5' LTR and in the 5' part of the pol gene, pointing to the presence of potential new regulatory elements in these regions.
terms the interactions
MATERIALS AND METHODS Cell lines. All cell lines (ACH2, 8E5, and Ul) were obtained from the AIDS Research and Reference Reagent Program (National Institute of Allergy and Infectious Diseases, Bethesda, Md.). Cells were grown in RPMI medium (GIBCO/Bethesda Research Laboratories) containing 10% fetal calf serum (HyClone) supplemented with 50 U of penicillin per ml, 50 ,g of streptomycin per ml, and 2 mM glutamine at 37°C in a 95% air-5% CO2 atmosphere. Cells were routinely maintained at a density of 0.25 x 10' to 1 x 106 cells per ml by diluting them with fresh medium. When indicated, cells at a density of less than 5 x 105 cells per ml were treated with 10 nM TPA dissolved in dimethyl sulfoxide (final concentration, 0.01%) or with dimethyl sulfoxide alone (final concentration, 0.01%) as a control. In one experiment, the ACH2 cell line was treated with tumor
necrosis factor alpha (TNF-oa) (Amgen) at a concentration of 100 U/ml for 12 h. Antigen p24 assays. Supernatants from treated and untreated cells were collected after low-speed centrifugation (1,000 rpm in a GPR tabletop centrifuge [Beckman]) of the cell cultures and kept frozen at -70°C until processed. Antigen p24 content was measured by SRA Technologies (Rockville, Md.) using the Coulter HIV p24 antigen assay kit (Coulter Immunology, Hialeah, Fla.). DNase I treatment of nuclei. Our experimental protocol was a modification of the method described by Bushel et al. (2). Exponentially growing cell were harvested by centrifugation at 1,000 rpm for 10 min at 4°C and washed twice with ice-cold phosphate-buffered saline. All subsequent operations were performed on ice with precooled buffers. Cells
were
counted and resuspended at 25
x
106 cells
per
ml in
RSB (10 mM Tris [pH 7.4], 10 mM NaCl, 3 mM MgCl2) and allowed to swell for 5 min. An equal volume of RSB-0.2% Nonidet P-40 was added, and the cells were incubated for another 5 min with intermittent mixing. Nuclei were centrirpm for 10 min, resuspended, washed in 50 volumes of RSB, and centrifuged again at 2,000 rpm for 10 min. The nuclear pellet was thoroughly resuspended in RSB
fuged at 2,000
6791
at 25 x 106 nuclei per ml by 15 strokes in a Dounce homogenizer (pestle B). Nuclei were examined and counted after staining with trypan blue, and aliquots were put into cooled 15-ml polypropylene tubes (5 x 107 to 7.5 x 107 nuclei per tube). Various concentrations of DNase I (Sigma) (as indicated) were added for 10 min on ice. The digestion reaction was stopped by adding 3 volumes of proteinase K buffer (50 mM Tris [pH 7.5], 100 mM NaCl, 1 mM EDTA, 0.5% sodium dodecyl sulfate [SDS]) and mixing vigorously. Samples were solubilized for 1 h at 55°C and treated for 1 h at 37°C with 50 ,ug of DNase-free RNase A per ml. Proteinase K was added at 200 ,ug/ml, and the digestion was allowed to continue overnight at 55°C. Samples were extracted three times with phenol and three times with chloroform-isoamyl alcohol (24:1) and precipitated with ethanol. DNA was resuspended in sterile water, and the DNA concentration was estimated by measuring the A26. Three independent DNase I digestions of nuclei were performed for each cell line and found to generate similar results. DNase I treatment of naked DNA. DNA from exponentially growing untreated cells was purified after an overnight digestion with 200 ,ug of proteinase K per ml in proteinase K buffer. After three phenol extractions and three chloroformisoamyl alcohol extractions, DNA was ethanol precipitated and resuspended in sterile water. DNA was digested for 10 min on ice with increasing concentrations of DNase I in RSB buffer at a DNA concentration of 0.3 mg/ml. Preliminary experiments were performed to determine the concentration range of DNase I necessary to generate a similar level of digestion, as observed after digestion of intact nuclei. Reactions were stopped by the addition of a fourfold volume excess of proteinase K buffer and processed as described for the nuclei after DNase I treatment. Southern blotting. Purified DNA (30 ,ug) was digested with restriction enzymes, and the fragments generated were separated by electrophoresis in 0.8 or 1.5% agarose gels in Tris-acetate buffer at 1.5 V/cm. Size markers were electrophoresed along with the samples. Each size marker was generated by digesting HIV-1 DNA (cloned in plasmid pBru2, a gift from Simon Wain-Hobson) with two restriction enzymes: the same enzyme used to digest the sample and another enzyme chosen to generate a fragment of defined size and location in the region under study. Several of these markers were mixed together, added to 30 ,ug of cellular DNA, and coelectrophoresed with the samples. Agarose gels were incubated twice for 20 min (each time) in denaturing solution (1.5 M NaCl, 0.5 M NaOH) and twice for 20 min (each time) in neutralizing solution (1.5 M NaCl, 0.5 M Tris [pH 7.2], 1 mM EDTA) and transferred overnight by capillarity in 20x SSPE (20x SSPE is 3 M NaCl, 0.2 M NaH2PO4, 20 mM EDTA, pH 7.4) to nylon membranes (N-Hybond; Amersham). DNA was cross-linked to nylon membranes by exposure to UV light (UV Stratalinker 1800; Stratagene), washed for 20 min in 2x SSPE and prehybridized for 1 to 2 h at 42°C in hybridization buffer (50% formamide, 3.6x SSPE, 1% SDS, 10% dextran sulfate, Sx Denhardt's solution [0.1% Ficoll, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin], 0.1 mg of sonicated herring sperm DNA per ml). Denatured DNA probes were added to the prehybridization buffer and allowed to hybridize for at least 16 h at 42°C. Membranes were washed twice for 20 min (each time) in 2x SSPE-0.1% SDS, twice for 20 min (each time) in 0.2x SSPE-0.1% SDS at room temperature, and once for 30 min at 65°C in 0.2x SSPE-0.1% SDS. Autoradiographic exposures with two intensifying screens were carried out for 1 to 5 days at -70°C.
6792
VERDIN
A 5'_
J. VIROL.
Earl
SphI PstI
BamHI
B
Lc
LTR
DNase l
LTnrR _^
-N
0
PROBE A +TPA
2
4
6
8
--
-TPA ~ 1
-TPA __ 2 __.. f~~~~~~~6m,
PROBE C + TPA _
-
.,_
s% 3'
10 Kb
I I I I I L FIG. 1. Probes used to map hypersensitive sites in the HIV genome. Three sets of probes and restriction enzymes were used to map the complete genome of HIV. Probe A spanned nt 643 to 1415 and was used to examine the 5' LTR with PstI in all cell lines. Probe B spanned nt 1420 to 2272 and was used to examine the intragenic region with PstI in ACH2 and 8E5 cells and with SphI in Ul cells. Probe C contained nt 8523 to 9113 and was used to examine the 3' LTR with BamHI in ACH2 and 8E5 cells and with Earl in Ul cells.
DNA probes. Three probes, identified as A, B, and C in Fig. 1, were used in these studies. Each probe was synthesized by 15 cycles (95°C for 2 min, 55°C for 2 min, and 72°C for 3 min) of polymerase chain reaction, using 1 ng of a plasmid containing a complete clone of HIV-1 (pBru2) as a template and primers EV1 and EV2 for probe A, primers EV5 and EV6 for probe B, and primers EV3 and EV4 for probe C. Polymerase chain reactions were conducted by the protocols provided with the AmpliTaq DNA polymerase (Perkin-Elmer Cetus), using a Perkin-Elmer Cetus Thermal Cycler. Probe A spans nucleotides (nt) 643 to 1415 (where +1 is the first nucleotide in the 5' LTR U3 region), probe B spans nt 1420 to 2272, and probe C spans nt 8523 to 9113. The sequences of the oligonucleotides were as follows: EV1, 5' (nt 1415) GCTTCCTCATTGATGGTCTC 3'; EV2, 5' (nt 643) CGAACAGGGACTTGAAAGCG 3'; EV3, 5' (nt 8523) GATCCTTAGCACTTATCTGGG 3'; EV4, 5' (nt 9113) AAAGTGGCTAAGATCTACAGC 3'; EV5, 5' (nt 1420) GAATGGGATAGAGTGCATCC 3'; and EV6, 5' (nt 2272) GTTCCTTGTCTATCGGCTCC 3'. The amplified products were separated from the template on a 1.5% agarose gel and purified by using the Geneclean procedure (Bio 101, La Jolla, Calif.). DNA fragments were labeled by the random primer reaction (8) and purified on a G-50 Sephadex column. RESULTS
Preliminary studies. Since the cell lines used in these studies were generated by infection with a viral strain (LAVBRU) composed of several distinct quasispecies, it was necessary to establish detailed restriction maps of the viruses integrated in these cell lines prior to Southern blotting. DNA from these infected cells was extracted, digested with a number of restriction enzymes on the basis of the sequence of Bru previously published by Wain-Hobson et al. (50), and examined by Southern blotting. Preliminary experiments demonstrated that a single copy of HIV had integrated in the ACH2 and 8E5 cell lines and that the Ul cell line contained two integrated copies. None of the cell lines contained detectable amounts of unintegrated HIV. Unique restriction sites were identified in the 5' and 3' portions of the HIV genome (Fig. 1). The PstI site at nt 1415 was a unique site in 8E5 and ACH2 and was used to study the complete HIV genome with either probe A or B (Fig. 1). This PstI site was also used to examine the 5' LTR in Ul but could not be used to examine the intragenic region, since an additional PstI site was present downstream. To examine the intragenic region in Ul, a unique SphI site (nt 1443) was used with probe B. To study the 3' LTR, a unique BamHI site (nt
a-
b-
c-
de-
i| : ii,X t
I-
g.
9
b
~~~~~~~~~e-.1-
FIG. 2. Digestion of ACH2 nuclei with increasing doses of DNase I. Purified nuclei from TPA-treated and untreated ACH2 cells were incubated with the following doses of DNase I: 0, 0.5, 1, 2, 4, 6, 8, 10, 12, 14, 16, and 20 U/ml. After extraction and purification, DNA was either digested with PstI and hybridized with probe A to examine the 5' LTR or digested with BamHI and hybridized with probe C to examine the 3' LTR. Markers are double digests of pBru2 with PstI (nt 1415) and HindlIl (nt 1085 [marker a]), AccI (nt 959 [marker b]), HgaI (nt 792 [marker c]), Sacl (nt 678 [marker d]), AflII (nt 517 [marker e]), and HpaII (nt 309 [marker fl). Hypersensitive sites are indicated by circles and correspond to sites II, III, and IV in Fig. 3.
8522) was used in ACH2 and 8E5 and the unique Earl site (nt 8564) was used in Ul with probe C. Exponentially growing ACH2 cells (density ranging from 2 x 10' to 5 x 105 cells per ml) were either treated for 12 h with 10 nM TPA to induce viral expression or not treated and used as a control. The cells were harvested, and the nuclei were purified by centrifugation after lysis with Nonidet P-40. Purified nuclei were treated with increasing concentrations of DNase I (from 0 to 20 U/ml) for 10 min at 4°C to maintain the chromatin architecture in a state as close as possible to the state in vivo. After phenol extraction and purification, genomic DNA was cut with either PstI (for hybridization with probe A) or BamHI (with probe C) and analyzed by Southern blotting, using the indirect end labelling technique (28, 54). This technique uses a small labelled probe abutting the restriction site and consequently allows the direct mapping of the hypersensitive site by determining the size of the fragment generated by the double digestion (DNase I and restriction enzyme). Digestion with as little as 6 U of DNase I per ml resulted in the appearance of three new smaller bands (indicated by circles in Fig. 2) mapping to the 5' LTR and the region downstream of the 5' LTR (probe A). When the 3' LTR was examined with probe C, two new smaller bands mapping to the 3' LTR were visualized in the absence and presence of TPA with increasing DNase I concentrations (probe C). These new bands were dependent on a double digestion by both DNase I and the restriction enzyme, since they were absent when DNase I was absent (Fig. 2) or when the DNA was not cut by a restriction enzyme (PstI or BamHI in this case) (not shown). As the concentration of DNase I was further increased, progressive digestion of the bulk of genomic DNA was observed, resulting in the disappearance of the primary band of viral DNA (Fig. 2). Preliminary titrations of DNase I digestion were performed for the other cell lines and probes (data not shown), and only the appropriate concentrations of DNase I will be shown hereafter (indicated in the figure legends). Mapping of hypersensitive sites in the 5' region of the genome. By using the approach outlined above, the hyper-
TABLE 1. Induction of HIV expression in chronically infected cells Cell line
ACH2 Ul 8E5
6793
NUCLEASE-HYPERSENSITIVE SITES IN HIV CHROMATIN
VOL. 65, 1991
Amt (pg/ml) of p24 antigen
Without TPA
With TPA
132 64 14,564
7,976 950 NDa
r-IN VIVO-1 34
1 2
IN VITRO-5 6 7 8 91011121314
Induction (fold)
60 15
a ND, not determined.
lb *a mu sensitive sites in ACH2, Ul, and 8E5 cell lines were compared. ACH2 and Ul were examined both with and without TPA treatment, since this agent was previously shown to induce viral expression in these cells (31). The induction of viral expression was confirmed by measuring the release of the viral antigen p24 in the culture medium in the absence and presence of TPA (Table 1). Small but detectable amounts of p24 antigen was found in untreated ACH2 and Ul, indicating that the viruses in these cell lines are not truly latent but that these cells are better defined as chronic low producers. In contrast, the 8E5 cell constitutively secreted a large amount of p24 antigen (Table 1). TPA treatment of Ul and ACH2 for 12 h resulted in a 15- and 60-fold increase in p24 release in the medium, respectively (Table 1). Three major hypersensitive sites were observed in all three cell lines (sites II, III, and IV; Fig. 3, lanes 2 for ACH2 and Ul cells and lane 4 for 8E5 cells) in the absence of TPA. To determine their exact position in the viral genome, a double digestion of a molecular clone of HIV (pBru2) by PstI and by several other enzymes scattered throughout the region examined was run along the samples. By using these markers as references, site II mapped to nt 223 to 325, site III mapped to nt 390 to 449, and site IV mapped to nt 656 to 720 (values averaged from three independent experiments, with a standard deviation of 12 bp). Two changes were noted in ACH2 and Ul after TPA induction: first, site III became larger at the expense of its 3' boundary, which moved from nt 449 to nt 583, its 5' boundary being unchanged at nt 390 (compare the space between sites III and IV on lanes 2 and 4 for ACH2 and Ul); second, the intensity of the site IV band was decreased by at least 50% (this is most visible on Fig. 3 for Ul cells [compare lanes 2 and 4]). In Ul cells, TPA induction also resulted in the appearance of a new minor site, site I at nt 56 to 114 (Fig. 3, Ul cells [compare lanes 2 and 4]). The pattern of hypersensitive sites in untreated 8E5 cells was similar to that observed after TPA induction in ACH2 and Ul cells (Fig. 3, 8E5). To prove that the hypersensitive sites observed were the consequence of chromatin organization and not secondary to sequence-dependent cleavage preference by DNase I, the following experiment was performed: DNA was extracted and purified from each cell line and submitted in vitro to digestion with increasing doses of DNase I, and DNA was then digested with PstI and examined by indirect end labelling as described above. The range of DNase I concentrations was chosen to generate similar levels of digestion under in vitro and in vivo conditions. This experiment indicated that although DNase I exhibited some sequence preference for cutting (Fig. 3, lanes 5 to 14), no comparable hypersensitive sites were noted (compare lanes 2 or 4 and 14 for ACH2 and Ul cells and lanes 4 and 13 for 8E5). To unequivocally prove that each band observed (hereafter referred to as a hypersensitive site) was dependent on digestion by both DNase I on isolated nuclei and PstI
5 6 7 8 9 1011121314
34
1 2
B# 11W F r 4." i
.
"M-W
-F
0
U1
..4
c-
d
w1" |v
a11
t
l
4
I
ilt
e-
8E5
FIG. 3. Mapping of DNase I-hypersensitive sites in the 5' portion of the HIV genome. The pattern of digestion by DNase I of purified nuclei from ACH2 and Ul cells (without TPA [lanes 1 and 2] and with TPA [lanes 3 and 4]) and 8E5 cells (without TPA [lanes 1 to 4a]) is compared with that by DNase I of naked DNA in vitro (lanes 5 to 14). For ACH2 and Ul cells, either no DNase I (lanes 1, 3, and 5) or 15 U of DNase I (lanes 2 and 4) was used. For 8E5 cells, lanes 1, 2, 3, 4, and 4a show digestion by 0, 3.75, 7.5, 15 and 30 U of DNase I per ml, respectively. For all cell lines, lanes 5, 6, 7, 8, 9, 10, 11, 12, 13, and 14 show digestion by 0, 0.17, 0.24, 0.33, 0.47, 0.65, 0.91, 1.28, 1.79, and 2.5 U of DNase I per ml, respectively. Markers are described in the legend to Fig. 2. Hypersensitive sites are indicated by roman numbers.
digestion of purified DNA, DNA samples purified after DNase I digestion of nuclei were examined by indirect end labelling with and without PstI digestion (Fig. 4). Samples digested solely with DNase I showed high-molecular-weight bands (indicated by open circles on Fig. 4) and no bands with sizes similar to those observed after double digestion with DNase I and PstI (Fig. 4, sites II, III, and IV). This experiment indicates that the fragments visualized (II, III, and IV) depend on both DNase I and PstI digestions. The
6794
VERDIN ACH2 +Pst I --I'- Pst I-_
m
1 2 3 4 5 6 7 8 9 10
J. VIROL.
u1 r-+Pst lI -l Pst I 1 2 3 4 5 6 7 8 9 10
r-Control -
r-ITpa
m 1 23 45
1 2 3 45
r-Tnf- X I 1 2 3 4 5m
0
*0~~~~~
o*1ma
-,W_.I i"" Im
a_ bd_
C-. HI .0 dgv IV e-l f*
FIG. 4. Hypersensitive site visualization depends on digestion by both DNase I and PstI. DNA samples extracted after DNase I digestion of purified nuclei from ACH2 (treated with TPA) or Ul cells (untreated) were digested or not with PstI and examined by using indirect end labelling with probe A. Size markers (lane m) are described in the legend to Fig. 2. For ACH2 cells, DNase I concentrations (in units per milliliter) were 0 (lanes 1 and 6), 9 (lanes 2 and 7), 12 (lanes 3 and 8), 15 (lanes 4 and 9), and 18 (lanes 5 and 10). For Ul cells, DNase I concentrations (in units per milliliter) were 0 (lanes 1 and 6), 7.5 (lanes 2 and 7), 10 (lanes 3 and 8), 15 (lanes 4 and 9), and 20 (lanes 5 and 10). Hypersensitive sites are indicated by roman numerals.
high-molecular-weight band observed in the absence of PstI digestion probably represents the complete HIV genome resulting from DNase I digestion in hypersensitive sites located in both the 5' and 3' LTRs, as will be seen later. Since TNF-ot has been shown to induce viral expression in ACH2 cells and to mediate in part the induction observed after TPA treatment through an autocrine loop (32), the chromatin of the 5' LTR was examined after TNF-ot treatment (100 U/ml for 12 h) and compared with TPA treatment in these cells (Fig. 5). TNF-c. treatment was associated with changes in the 5' LTR indistinguishable, at this level of resolution, from those observed after TPA treatment (Fig. 5). In Fig. 6, the hypersensitive sites described above have been aligned with an illustration of the 5' portion of the HIV genome (the LTR and the beginning of the gag gene). Binding sites for some of the transcription factors known to interact with the LTR are indicated as landmarks. Site I, which is present only in Ul cells and slightly increases in intensity after TPA treatment, corresponds to a region of the LTR in which binding sites for transcription factors have been identified including AP-1 (15) (Fig. 6). Site II partially overlaps a silencer (negative regulatory element) containing a binding site for transcription factor USF (16, 37, 41, 58) and contains a region of the viral promoter in which several proteins have been shown to bind in vitro (16, 40, 42, 51). Site III is separated from site II by a 65-nt space which probably indicates the protection of DNA from DNase I digestion in purified nuclei by a DNA-bound factor. Interestingly, the protected region maps to the enhancer (16, 37, 41, 58), where at least three distinct pro-
so
_
SI i II-b
4.0I IV B
g-e e-~~~~~~ f--f FIG. 5. TPA and TNF-a treatment induce the same modifications in DNase I hypersensitivity in the 5' LTR. Exponentially growing ACH2 cells were treated for 12 h with TPA (10 nM) or TNF-a (100 U/ml). Purified nuclei were treated with DNase I at the following concentrations (in units per milliliter): 0 (lanes 1), 9 (lanes 2), 12 (lanes 3), 15 (lanes 4), and 18 (lanes 5). Markers (lanes m) are described in the legend to Fig. 2. Purified DNA were digested with PstI and examined with probe A. Hypersensitive sites are indicated by roman numbers.
teins have been shown to bind, NF-KB (Fig. 4) (27), HIVEN86A (14), and EBP-1 (56, 57). However, we cannot completely exclude the possibility that this protection is secondary to a phased nucleosome present in this region. Site III, in the absence of TPA, covers a region containing three Spl binding sites (22), the TATA box where TFIId binds, and additional sites close to the site of initiation of transcription, at the U3-R junction (21). TPA induction results in an extension of site III downstream to include the Tat-responsive region (36). Finally, site IV, which is most apparent in uninduced Ul cells (Fig. 6), maps 3' of the LTR in a region for which a regulatory function has not been discovered yet. Mapping of hypersensitive sites in the 3' LTR. The 5' and 3' LTRs in retroviruses exert different functions despite an identical nucleotide sequence. The 5' LTR acts as a promoter: transcription is initiated at the U3-R junction. The 3' LTR functions as a site of polyadenylation of the viral transcripts. Different mechanisms have been invoked to explain why transcripts are not polyadenylated in the 5' LTR at the R-US junction, a region located downstream of the initiation site (20, 52) and why transcription is not or poorly initiated in the 3' LTR (7). It was therefore of interest to determine whether these functional differences might be accompanied by structural differences in chromatin organization. DNase I digestion of purified nuclei resulted in the appearance of one major site (VIII) and two minor sites (Vllla and VlIIb) in the 3' LTR in all three cell lines (Fig. 7). Contrary to what had been observed in the 5' LTR, no change was observed in ACH2 and Ul cells after TPA treatment. Using markers, as described for the 5' LTR, these sites were mapped as follows: VIIIa, nt 9178 to 9221 (nt 94 to 137 in the 5' LTR); VIII, nt 9322 to 9489 (nt 238 to 405 in the 5' LTR); and VlIIb, nt 9535 to 9586 (nt 451
6795
NUCLEASE-HYPERSENSITIVE SITES IN HIV CHROMATIN
VOL. 65, 1991
U3
_0
U 5_--
o-R--o
-N.
.N GA" Ap- I
USF
:11: +
kB
SplTF.ld
I
III
TAR
IV
I
'l"N.i\. ..
..
v
.:
+
N
u
FIG. 6. Alignment of hypersensitive sites in the 5' LTR with cis-acting regulatory element. An enlargement of hypersensitive sites in Fig. 2 from ACH2 and Ul cells in the presence (+) and the absence (-) of TPA are aligned to scale with the 5' LTR and its cis-acting regulatory elements. TAR, Tat-responsive element; kB, NF-KB.
TPA
VIII approximately covers a region encompassing both site II and III and the intervening space. However, even shorter exposure times showed a single site with no intervening space, indicating that the protein(s) that protected this area in the 5' LTR is not bound in the homologous region in the 3' LTR. The intragenic region contains a hypersensitive site mapping to the pol gene. Since we had previously identified an intragenic enhancer in the pol gene of HIV-1 (47) and since enhancer elements are frequently associated with hypersensitive sites (19), the possibility of DNase I hypersensitivity in this region was particularly intriguing. All three cell lines were tested as described above with probe B (Fig. 1). In all cell lines, a major hypersensitive site was observed in both the presence and absence of TPA at nt 9400 (Fig. 8). This corresponds approximately, given the uncertainty of size determination of such a large fragment, to site VIII, which was described above as the major site in the 3' LTR. No other major site was observed in ACH2 or 8E5 cells (Fig. 8). In Ul cells, however, three additional sites (V, VI, and VII) were apparent in the absence of TPA (Fig. 8, compare lanes 7 with 8 and 9). Site VII mapped to nt 4534 to 4733. This region is precisely located between the two functional domains of the intragenic enhancer identified in HeLa cells (47). Two additional sites (V and VI [Fig. 8]) were identified in Ul and mapped to nt 2849 to 2956 for site V and nt 3073 to 3187 for site VI. Both sites V and VI were also faintly visible in ACH2 upon longer exposure times of the autoradiograms (data not shown).
a-
-
+
0
o dd-0 0
-
-
+
-
7 8 91011 12
1 2 3 4 5 6
M
e-~
r
8E5
Ul
ACH2
to 502 in the 5' LTR). If one compares the patterns of hypersensitivity in the 5' and 3' LTRs, it is apparent that site
o*v:
'illao m
13141516
i
Aftb t oVillb; oVli~~~~~~~~~~~~li. ! 1
oVilibo
f
|
FIG. 7. The 3' LTR contains a single major DNase I-hypersensitive site. DNA samples from TPA-treated (+) and untreated (-) cells were digested with DNase I at the following concentrations (in units per milliliter): 0 (lanes 1, 4, 7, 10, and 13), 12 (lanes 2, 5, 8, 11, and 15), and 18 (lanes 3, 6, 9, 12, and 16). DNA was purified and digested with BamHI (for ACH2 and 8E5 cells) or with Earl (for Ul cells) and analyzed by electrophoresis on a 1.5% agarose gel. Markers (lane M) are double digests of pBru2 with BamHI (for ACH2 and 8E5) or EarI (for Ul) and AflII (nt 9649 [marker a]), BanII (nt 9541 [marker b]), HpaII (nt 9442 [marker c]), DraIll (nt 9339 [marker d]), and EcoRV (nt 9165 [marker e]). Hypersensitive sites are indicated by roman numbers.
J. VIROL.
VERDIN
6796
M
r,
8E5
Ul
ACH2 TPA
+
-
+ T I-I-
131415
7 8 9101112
1 2 3 4 5 6
M
^
a-
aF
'r
I
.vIII
b-v I-, h-@
VII. 9.
*..Vil
VIl
-
-_
j_w
* 0
V°
*VI oV
v
b. FIG. 8. Colocalization of a hypersensitive site and an enhancer in the intragenic region. DNA from nuclei that had been treated with DNase I (same concentrations as in the legend to Fig. 5) were digested with PstI (for ACH2 and 8E5 cells) or Sp7hI (for Ul cells), analyzed on a 0.8% agarose gel, and hybridized wvith probe B after transfer to nylon membranes. Markers (lanes M) aire double digests of pBru2 with PstI (for ACH2 and 8E5) or SphI (for Ul) and (nt 9619 [marker a]), XhoI (nt 8944 [marker b]), BamH [ (nt 8522 [marker c]), HhaI (nt 7863 [marker d]), DraIll (nt 6627 [m; 4684 [marker 5821 [marker fl), AlwnI (nt 5423 [marker g]), EcoR] and HindII h]), KpnI (nt 3862 [marker i]), PvuII (nt 3335 [mark (nt 2532 [marker k]).
Sak
Ik(nt Ierj]),
DISCUSSION In this report, I have used three HIV-1 chronically infected cell lines to study the chromatin stru4cture of HIV-1. By using three sets of probes and restriction,,sites, five major sites and five minor sites for DNase I cut ting have been identified in isolated nuclei. These results, summarized in Fig. 9 indicate that three major sites are piresent in the 5' portion of the HIV genome (sites II, III, and IV) in all three cell lines. The three cell lines also conttain one major hypersensitive site in the 3' LTR (site VIII). In addition, the Ul cell line contains one major (VII) and twto minor (V and ACH2 Ul
+
8E5 I
+
+
+
+
+
+
+
+
II
III
IV
+
VI) sites located in the pol gene. TPA treatment, which is associated with the induction of viral expression in ACH2 and Ul cells, results in a decrease in intensity of three sites, IV, VI, and VII, and more remarkably, in a displacement of the 3' boundary of site II into the R region of the 5' LTR. Some of these sites are associated with previously identified cis-acting regulatory elements, such as the HIV promoter in the 5' LTR (sites II and III), or with an intragenic enhancer (site VII). Other sites (IV, V, and VI), however, cover regions of the HIV genome with no known regulatory role. These findings raise several interesting questions which will be discussed below. Our data on the pattern of digestion of the 5' LTR is consistent with a large nucleosome-free region spanning nt 223 to 449 in the absence of TPA and nt 223 to 583 after TPA treatment. We suspect that the minor hypersensitive site I observed in Ul represents a region of internucleosomal linker DNA, since it is located 138 nt from the 5' limit of site II, which corresponds approximately to the length of DNA included in one nucleosome. Preliminary observations with micrococcal nuclease, which preferentially cleaves chromatin DNA in the internucleosomal linker region, corroborate this hypothesis (47a). Binding sites for AP-1 transcription factor have been identified in this region (15), and more recently, a site with homology to steroid or thyroid hor-
mone-responsive elements was identified in the same region
(29) and found to bind a factor called COUP-TF (4). Several transcription factors have been shown to bind specifically to the region spanning sites II and III (reviewed in Results and shown in Fig. 4). The region separating site II from III (65 nt) contains an enhancer (16, 37, 41, 58) and several factors have been shown to bind within this short domain (NF-KB, EBP-1, and HIVEN86A). Since this region is too small to accommodate a nucleosome, although this possibility cannot completely be excluded, the absence of digestion indicates that the DNA is protected from DNase I digestion by DNA-bound factor(s). These results could indicate, with the resolution afforded by this method (±12 nt), that DNAbinding proteins are bound to the enhancer even before activation by TPA. Since NF-KB activity is dependent on cellular activation (by TPA, for example), it is probable that another factor, such as EBP-1 (56, 57), binds to the enhancer before cellular activation. After TPA or TNF-ot induction, site III increases markedly in size, largely by moving its 3'
+
+ -
V
VI
VII
IS//X80%) in the intensity of this site after induction, suggesting that this site could also function as a silencer (Fig. 3 and 6). The 5' LTR and 3' LTR have the same nucleotide sequence but different functional properties as discussed in Results. The molecular mechanism(s) underlying these differences is still not fully understood. A major functional difference between the two LTRs is the apparent low rate of transcription starting in the 3' LTR in comparison with the 5' LTR. Studies by Cullen et al. (7) have established a model of promoter occlusion in which transcripts initiating in the 5' LTR and continuing through the U3 region of the 3' LTR prevent the assembly of a stable preinitiation complex, thereby inhibiting transcription initiation. The present study demonstrates a significant structural difference between the two LTRs in terms of DNase I sensitivity, specifically, the absence of a footprint in a region of U3 that corresponds to the enhancer in the 3' LTR, whereas the same site is protected from digestion in the 5' LTR. The absence of protein(s) bound to this region could represent the molecular basis of promoter occlusion. These results would suggest that even low levels of transcription, as observed before TPA induction, are able to inhibit binding of proteins to DNA in this region of the promoter. A major hypersensitive site is located in the part of the pol gene encoding the integrase protein, in close proximity to a region of HIV-1 in which we previously identified an enhancer (47). The enhancer contains two domains in HeLa cells, fragment 5103 (nt 4079 to 4342) and fragment 5105 (nt 4781 to 6026), and site VII maps to nt 4534 to 4733, precisely between the two functional domains. These results provide evidence that the intragenic enhancer is an element participating in the control of HIV transcription in vivo. TPA induction resulted in a threefold reduction in the intensity of this site (by densitometry scanning), whereas the activity of the enhancer was increased by TPA in HeLa cells. The reason for this discrepancy is unclear but could be a result of cell-type differences between HeLa cells and Ul cells. The fact that this hypersensitive site is observed only in Ul cells, which are of monocytic origin, and not in ACH2 and 8E5 cells, both of lymphoid origin, could point to a cellular specificity associated with this intragenic element. It is also possible that the nature of chromatin at the site of viral integration in the cellular genome determines the appearance or the nonappearance of this site. Studies in other viral systems have indeed shown that hypersensitive sites can induce or suppress the appearance of other hypersensitive sites, even over long distances, and retroviruses frequently become integrated near DNase I-hypersensitive sites in
6797
chromatin (35, 48). Consequently, depending on the site of proviral integration in the cellular genome, the provirus could encounter an environment favorable or unfavorable to the establishment of a given hypersensitive site. A last possibility is that the nucleotide sequence of viruses integrated in the different cell lines differs in the region of the intragenic enhancer and that these differences account for the binding or the absence of binding of specific factors necessary for the establishment of a hypersensitive site. These possibilities are currently under investigation. This study has identified several regions of the HIV genome that probably play a role in the control of HIV transcription by virtue of their accessibility in chromatin. In order to generalize these observations and to draw conclusions in terms of cell specificity, these studies have to be extended to a larger number of cell lines, particularly to primary cultures, and to several sites of integration per cell line. These studies along with the high-resolution mapping of protein binding sites within these regions by genomic footprinting will undoubtedly increase our understanding of HIV regulation in vivo. ACKNOWLEDGMENTS I thank Peter Paras, Jr., for superb technical assistance and Monique Dubois-Dalcq for support and encouragement. I thank Tom Folks, Guido Poli, Anthony Fauci, and the AIDS Reagent Reference Program at NIAID for the generous gift of the cell lines used in these studies; Simon Wain-Hobson for the gift of pBru2 plasmid; and the members of the Laboratory of Viral and Molecular Pathogenesis for comments on the manuscript. The initial part of this project was conducted in the laboratories of Arsene Burny (Universite Libre de Bruxelles, Brussels, Belgium) and Susie Sprecher (Institut Pasteur du Brabant, Brussels, Belgium) and I thank them for encouragement and support. REFERENCES 1. Barklis, E., R. C. Mulligan, and R. Jaenish. 1986. Chromosomal position or virus mutation permits retrovirus expression in embryonal carcinoma cells. Cell 47:391-399. 2. Bushel, P., K. Rego, L. Mendelsohn, and M. Allan. 1990. Correlation between patterns of DNase I-hypersensitive sites and upstream promoter activity of the human e-globin gene at different stages of erythroid development. Mol. Cell. Biol. 10:1199-1208. 3. Clouse, K. A., D. Powell, I. Washington, G. Poli, K. Strebel, W. Farrar, P. Barstad, J. Kovacs, A. S. Fauci, and T. M. Folks. 1989. Monokine regulation of human immunodeficiency virus type 1 expression in a chronically infected human T cell clone. J. Immunol. 142:431-438. 4. Cooney, A. J., S. Y. Tsai, B. W. O'Malley, and M.-J. Tsai. 1991. Chicken ovalbumin upstream promoter transcription factor binds to a negative regulatory region in the human immunodeficiency virus type 1 long terminal repeat. J. Virol. 65:2853-
2860.
5. Culien, B. R. 1990. The HIV-tat protein: an RNA sequence-
specific processivity factor? Cell 63:655-657.
6. Cullen, B. R., and W. C. Greene. 1989. Regulatory pathways
governing HIV-1 replication. Cell 58:423-426.
7. Cullen, B. R., P. T. Lomedico, and G. Ju. 1984. Transcriptional interference in avian retroviruses-Implications for promoter insertion model of leukaemogenesis. Nature (London) 307:241-
245.
8. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13. 9. Feuer, G., M. Taketo, R. C. Hanecak, and H. Fan. 1989. Two blocks in Moloney murine leukemia virus expression in undifferentiated F9 embryonal carcinoma cells as determined by transient expression assays. J. Virol. 63:2317-2324. 10. Folks, T. M., K. A. Clouse, J. Justement, A. Rabson, E. Duh,
6798
11.
12.
13.
14.
15.
16.
17. 18.
19. 20.
21. 22. 23.
24.
25. 26.
27. 28.
29.
VERDIN
J. H. Kehrl, and A. S. Fauci. 1989. Tumor necrosis factor-a induces expression of human immunodeficiency virus in a chronically infected T-cell clone. Proc. Natl. Acad. Sci. USA 86:2365-2368. Folks, T. M., J. S. Justement, A. Kinter, C. A. Dinarello, and A. S. Fauci. 1987. Cytokine-induced expression of HIV-1 expression in a chronically infected prom nocytic cell line. Science 238:800-802. Folks, T. M., J. S. Justement, A. Kinter, S. Schnittman, J. Orenstein, G. Poli, and A. S. Fauci. 1988. Characterization of a promonocytic clone chronically infected with HIV and inducible by 13-phorbol-12-myristate acetate. J. Immunol. 140:11171122. Folks, T. M., D. Powell, M. Lightfoote, S. Koenig, A. S. Fauci, S. Benn, A. Rabson, D. Daugherty, H. G. Gendelman, M. D. Hoggan, S. Venkatesan, and M. A. Martin. 1986. Biological and biochemical characterization of a cloned leu-3- cell surviving infection with the acquired immune deficiency syndrome virus. J. Exp. Med. 164:280-290. Franza, B. R., S. F. Josephs, M. Z. Gilman, W. Ryan, and B. Clarkson. 1987. Characterization of cellular proteins that recognize the human immunodeficiency virus enhancer using a microscale DNA-affinity precipitation assay. Nature (London) 330:391-395. Franza, B. R., F. J. Rauscher, S. F. Josephs, and T. Curran. 1988. The fos complex and fos-related antigens recognize sequence elements that contain AP-1 binding sites. Science 239: 1150-1153. Garcia, J. A., F. K. Wu, R. Mitsuyasu, and R. B. Gaynor. 1987. Interactions of cellular proteins involved in the transcriptional regulation of the human immunodeficiency virus. EMBO J. 6:3761-3770. Greene, W. C. 1991. The molecular biology of human immunodeficiency virus type 1 infection. N. Engl. J. Med. 324:308-317. Griffin, G. E., K. Leung, T. M. Folks, S. Kunkel, and G. J. Nabel. 1989. Activation of HIV gene expression during monocyte differentiation by induction of NFKB. Nature (London) 339:70-73. Gross, D. S., and W. T. Garrard. 1988. Nuclease hypersensitive sites in chromatin. Annu. Rev. Biochem. 57:159-197. Iwasaki, K., and H. M. Temin. 1991. The efficiency of RNA 3'-end formation is determined by the distance between the cap site and the poly(A) site in spleen necrosis virus. Genes Dev. 4:2299-2307. Jones, K. A. 1989. HIV trans-activation and transcription control mechanisms. The New Biol. 1:127-135. Jones, K. A., J. T. Kadonaga, P. A. Luciw, and R. Tjian. 1986. Activation of the AIDS retrovirus promoter by the cellular transcription factor Spl. Science 232:755-759. Jones, K. A., P. A. Luciw, and N. Duchange. 1988. Structural arrangements of transcription control domains within the 5'untranslated leader regions of the HIV-1 and HIV-2 promoters. Genes Dev. 2:1101-1114. Kinter, A. L., G. Poli, W. Maury, T. M. Folks, and A. S. Fauci. 1990. Direct and cytokine-mediated activation of protein kinase C induces human immunodeficiency virus expression in chronically infected promonocytic cells. J. Virol. 64:4306-4312. Loh, T. P., L. L. Silvert, and R. Scott. 1987. Proviral sequences that restrict retroviral expression in mouse embryonal carcinoma cells. Mol. Cell. Biol. 7:3775-3784. Loh, T. P., L. L. Silvert, and R. Scott. 1988. Negative regulation of retrovirus expression in embryonal carcinoma cells mediated by an intragenic domain. J. Virol. 62:4086-4095. Nabel, G., and D. Baltimore. 1987. An inducible transcription factor activates expression of human immunodeficiency virus in T cells. Nature (London) 326:711-713. Nedospasov, S. A., and G. P. Georgiev. 1980. Non-random cleavage of SV40 DNA in the compact minichromosome and free in solution by micrococcal nuclease. Biochem. Biophys. Res. Commun. 92:532-539. Orchard, K., N. Perkins, C. Chapman, J. Harris, V. Emery, G. Goodwin, D. Latchman, and M. Collins. 1990. A novel T-cell protein which recognizes a palindromic sequence in the negative
J. VIROL.
regulatory element of the human immunodeficiency virus long terminal repeat. J. Virol. 64:3234-3239. 30. Pavlakis, G. N., and B. K. Felber. 1990. Regulation of expression of human immunodeficiency virus. The New Biol. 2:20-31. 31. Poli, G., P. Bressler, A. Kinter, E. Duh, W. C. Timmer, A. Rabson, J. S. Justement, S. Stanley, and A. S. Fauci. 1990. Interleukin 6 induces human immunodeficiency virus expression in infected promonocytic cells alone and in synergy with tumor necrosis factor-a by transcriptional and post-transcriptional mechanisms. J. Exp. Med. 172:151-158. 32. Poli, G., A. Kinter, J. S. Justement, J. H. Kehrl, P. Bressler, S. Stanley, and A. S. Fauci. 1989. Tumor necrosis factor-a( functions in an autocrine manner in the induction of HIV expression. Proc. Natl. Acad. Sci. USA 87:782-785. 33. Poli, G., J. M. Orenstein, A. Kinter, T. M. Folks, and A. S. Fauci. 1989. Interferon-a but not AZT suppresses HIV expression in chronically infected cell lines. Science 244:575-577. 34. Pomerantz, R. J., M. B. Feinberg, D. Trono, and D. Baltimore. 1990. Lipopolysaccharide is a potent monocyte/macrophagespecific stimulator of human immunodeficiency virus type 1 expression. J. Exp. Med. 172:253-261. 35. Rohdewohld, H., H. Weiher, W. Reik, R. Jaenish, and M. Breindl. 1987. Retrovirus integration and chromatin structure: Moloney murine leukemia proviral integration sites map near DNase I-hypersensitive sites. J. Virol. 61:336-343. 36. Rosen, C. A. 1991. Regulation of HIV gene expression by RNA-protein interactions. Trends Genet. 7:9-14. 37. Rosen, C. A., J. G. Sodroski, and W. A. Haseltine. 1985. The location of cis-acting regulatory sequences in the human T cell lymphotropic virus type III (HTLV-III/LAV) long terminal repeat. Cell 41:813-823. 38. Rosenberg, Z. F., and A. S. Fauci. 1990. Immunopathogenic mechanisms of HIV infection: cytokine induction of HIV expression. Immunol. Today 11:176-180. 39. Scott, W. A., and D. J. Wigmore. 1978. Sites in simian virus 40 chromatin which are preferentially cleaved by endonucleases. Cell 15:1511-1518. 40. Shaw, J. P., P. J. Utz, D. B. Durand, J. J. Toole, E. A. Emmel, and G. R. Crabtree. 1988. Identification of a putative regulator of early T cell activation genes. Science 241:202-205. 41. Siekevitz, M., S. F. Josephs, M. Dukovich, N. Peffer, F. WongStaal, and W. C. Greene. 1987. Activation of the HIV-1 LTR by T cell mitogens and the trans-activator protein of HTLV-1. Science 238:1575-1578. 42. Smith, M. R., and W. C. Greene. 1989. The same 50-kDa cellular protein binds to the negative regulatory elements of the interleukin-2 receptor a-chain gene and the HIV-1 LTR. Proc. Natl. Acad. Sci. USA 86:8526-8530. 43. Stanley, S. K., P. B. Bresler, G. Poli, and A. S. Fauci. 1990. Heat shock induction of HIV production from chronically infected promonocytic and T cell lines. J. Immunol. 145:1120-1126. 44. Stanley, S. K., T. M. Folks, and A. S. Fauci. 1989. Induction of expression of human immunodeficiency virus in a chronically infected human promonocytic cell line by ultraviolet irradiation. AIDS Res. Hum. Retroviruses 5:375-384. 45. Tsukiyama, T., N. Ohtsura, and K. Yokoro. 1990. Characterization of the negative regulatory element of the 5' noncoding region of Moloney murine leukemia virus in mouse embryonal carcinoma cells. Virology 177:772-776. 46. Varshavsky, A. J., 0. H. Sundin, and M. J. Bohn. 1978. SV40 minichromosome: preferential exposure of the origin of replication as probed by restriction endonucleases. Nucleic Acids Res. 5:3469-3479. 47. Verdin, E., N. Becker, F. Bex, L. Droogmans, and A. Burny. 1990. Identification and characterization of an enhancer in the coding region of the genome of human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 87:4874-4878. 47a.Verdin, E., and P. Paras. Unpublished data. 48. Vijaya, S., D. L. Steffen, and H. L. Robinson. 1986. Acceptor sites for retroviral integration map near DNase I-hypersensitive sites in chromatin. J. Virol. 60:683-692. 49. Virelizier, J. L. 1990. Cellular activation and human immunodeficiency virus infection. Curr. Opin. Immunol. 2:409-413.
VOL. 65, 1991
NUCLEASE-HYPERSENSITIVE SITES IN HIV CHROMATIN
50. Wain-Hobson, S., P. Sonigo, 0. Danos, S. Cole, and M. Alizon. 1985. Nucleotide sequence of the AIDS virus, LAV. Cell 40:9-17.
51. Waterman, M. L., and K. A. Jones. 1990. Purification of TCF-1 a, a T-cell specific transcription factor that activates the T-cell receptor C a gene enhancer in a context-dependent manner. The New Biol. 2:621-636. 52. Weichs an der Glon, C., J. Moonks, and N. J. Proudfoot. 1991. Occlusion of the HIV poly(A) site. Genes Dev. 5:244-253. 53. Weiher, H., E. Barklis, W. Ostertag, and R. Jaenish. 1987. Two distinct sequence elements mediate retroviral gene expression in embryonal carcinoma cells. J. Virol. 61:2742-2746. 54. Wu, C. 1980. The 5' ends of Drosophila heat shock genes in chromatin are hypersensitive to DNase I. Nature (London) 286:854-860.
6799
55. Wu, C., P. M. Bingham, K. J. Livak, R. Holmgren, and S. C. R. Elgin. 1979. The chromatin structure of specific genes. 1. Evidence for higher order domains of defined DNA sequence. Cell 16:797-806. 56. Wu, F. K., J. A. Garcia, D. Harrich, and R. B. Gaynor. 1988. Purification of the human immunodeficiency virus type 1 enhancer and TAR binding proteins EBP-1 and UBP-1. EMBO J. 7:2117-2129. 57. Wu, F., J. Garcia, R. Mitsuyasu, and R. Gaynor. 1988. Alterations in binding characteristics of the human immunodeficiency virus enhancer factor. J. Virol. 62:218-225. 58. Zeichner, S. L., J. Y. H. Kim, and J. C. Alwine. 1991. Linikerscanning mutational analysis of the transcriptional activity of the human immunodeficiency virus type 1 long terminal repeat. J. Virol. 65:2436-2444.
