Ap-1 recognizes sequence elements on hiv-1 LTR in human epithelial tumor-cell lines.

TLDR: These results demonstrate the presence of three novel AP-1 binding sites on HIV-1 LTR, one of which was found within the TAR element and in the Tat protein binding region and suggest thatAP-1 could be contributing to HIV- 1 transcriptional regulation through its interaction with the AP- 1 binding sites of HIV-2 LTR.

Abstract: Investigation of the nucleotide sequence of the HIV-1 LTR showed the presence of four novel short DNA regions which are homologous to the recognition site for the cellular transcription factor AP-1. Four short oligonucleotide hybrids containing these potential AP-1 sites were constructed and used in gel retardation assays and in competition experiments in order to determine the role of the AP-L protein in the regulation of HIV-1 expression. The breast MDA MB 468 and cervical HeLa turner cell lines, which are known to overexpress the AP-1 protein were used in a gel retardation assay as a control to study the affinity of the AP-1 to synthesized oligonucleotide sequences. We have observed specific binding of nuclear factor AP-1 to three of these oligonucleotide hybrids. These results demonstrate the presence of three novel AP-1 binding sites on HIV-1 LTR, one of which was found within the TAR element and in the Tat protein binding region. Moreover, they suggest that AP-1 could be contributing to HIV-1 transcriptional regulation through its interaction with the AP-1 binding sites of HIV-1 LTR.

Full text

ONCOLOGY
REPORTS 1: 397401, 1994
AP-1
recognizes
sequence
elements
on HIV-1 LTR in
human
epithelial
tumor
cell
lines
V.
ZOUMPOURLIS
1
·
2
,
M. ERGAZAKI
1
·
2
and D.A. SPANDIDOS
1
'
2
'institute
of Biological Research and Biotechnology, National Hellenic Research
Foundation,
Athens;
Medical School, University of
Crete,
Heraklion, Creece
Received November 17, 1993; Accepted December 23, 1993
Abstract.
Investigation of the nucleotide sequence of the
HIV1 LTR showed the presence of four novel short DNA
regions which are homologous to the recognition site for the
cellular transcription factor AP1. Four short oligonucleotide
hybrids containing these potential AP1 sites were
constructed
and used in gel retardation
assays
and in
competition
experiments in order to determine the role of the
AP1 protein in the regulation of HIV1 expression. The
breast MDA MB 468 and cervical HeLa tumor cell lines,
which are known to overexpress the AP1 protein were used
in
a gel retardation
assay
as a control to study the affinity of
the
AP1 to synthesized oligonucleotide sequences. We have
observed specific binding of nuclear factor AP1 to three of
these oligonucleotide hybrids. These results demonstrate the
presence of three novel AP1 binding sites on HIV1 LTR,
one
of which was found within the TAR element and in the
Tat
protein binding region. Moreover, they
suggest
that AP1
could be contributing to HIV1 transcriptional regulation
through
its interaction with the AP1 binding sites of HIV1
LTR.
Introduction
The
human immunodeficiency
virus
type 1 (HIV1) is the
etiologic agent and the primary cause of AIDS (13). The
expression of
the
virus
is regulated both at transcriptional and
posttranscriptional
levels
by
several
human and
viral
proteins
(46). Control of HIV1 transcription is mediated by
c/sacting elements located in the
viral
long terminal repeats
(LTRs),
by the
viral
transregulatory protein Tat, and by
cellular transcription factors which are constitutively
expressed in most cells (e.g.
NFκΒ
and NFAT1) (7). eis
elements include the negative regulatory element (NRE,
Correspondence
to: Professor Demetrios A. Spandidos, Institute
of Biological Research and Biotechnology, National Hellenic
Research Foundation, 48 Vas. Constantinou Avenue, 116 35
Athens, Greece
Key
words:
epithelial tumor cells, HIV1 LTR sequences, AP1
binding
located in the region between nucleotide positions 357 to 185
relative to the transcription initiation site
+1),
the enhancer
(103 to 81), the Spi element (75 to 47), and the TATA
box (28 to 24). The iransregulatory element TAR is
located between residues 17 to +80. Various cellular
proteins
have been found to interact with the eis elements,
such as AP1 (8) and USF (9) with NRE; EBP1 with the
enhancer
(10); Spi with the sequence motif Spi (11) and
TFIID
with the TATA box (12). The
viral
protein Tat
interacts
with the fransregulatory element TAR (5,13,14).
The
AP1 binding site was initially described in the enhancer
elements of the Simian
Virus
40 promoter and the human
metallothionein
IIA promoter (15). A group of polypeptides
originally purified from HeLa cells was designated as
transcription
factor AP1 on the
basis
of DNA binding
specificity and in
vitro
transcriptional analysis (16). The
proteins
named AP1 in fact represent a family of
transcription
factors encoded by the members of the jun and
fos multigene families able to bind as homo and/or
heterodimers to the AP1 consensus (17,18).
Two binding sites for transcription factor AP1 have been
mapped
within the NRE of HIV1 LTR. Sequences between
nucleotides 348 to 343 and 336 to 331 are similar to HeLa
cell AP1 binding sites and have been shown to interact with
the
FOScomplex and FOSrelated antigens (8). In this study
we have further examined the HIV1 LTR, and we have
identified three additional AP1 binding sites one of which
was found within the TAR element. According to this result,
AP1 could be contributing to the HIV1 transcriptional
regulation through its interaction with AP1 binding sites of
HIV1 LTR.
Materials
and
methods
Cells
and
culture
conditions.
Human MDA MB 468 breast
and
HeLa cervical epithelial tumor cells were grown
exponentially in Ham's medium containing 10% fetal calf
serum and used for the preparation of nuclear extracts.
Preparation
of
cell
extracts.
The tumor cell lines were
homogenized in 2 ml hypotonic buffer (25 mM TrisHCl pH
7.5 KCl, 0.5 mM MgCl
2
, 0.5 mM
DTT,
0.5 mM PMSF) at 5
10 mg/ml. The nuclei were pelleted at
2500
rpm in a Sorvall
SS34 rotor for 10 min at 4°C. The pellets were washed 3

398
ZOUMPOURLIS
and SPANDIDOS: AP1
RECOGNIZES
SEQUENCE
ELEMENTS
ON HIV1 LTR
times with 2 ml isotonic buffer (25 mM TrisCl pH 7.5, 5
mM
KCl, 0.5 mM MgCl
2
, 0.1 M sucrose, 0.5 mM DTT, 1
mM
PMSF),
resuspended in nuclei extraction buffer (25 mM
TrisCl pH 7.5, 1 mM EDTA, 0.1% NP40, 0.5 mM
DTT,
0.5
mM
PMSF) and were further clarified after centrifugation at
25000
rpm in a Beckman Ti 50 rotor for 60 min at 4°C.
Supernatant
was removed and the extracts were stored at 70°C.
Protein
estimation was performed as described by Bradford
(19).
Preparation
of
double
stranded
oligonucleotide
hybrids.
Eight
single
stranded DNA oligonucleotides were made on
an
Applied Biosystems 381A DNA synthesizer. These were:
1
a
AGCTT
A AG ACC A ATGACTT AC
AAGGC
AGC
A
1
b ATTCTGGTTACTGAATGTTCCGTCGTTCGA
2a AGCTTCTAGTACCAGTTGAGCCAGAGAAGTTA
2b AGATCATGGTCAACTCGGTCTCTTCAATTCGA
3a AGCTTCATGGAATGGATGACCCGGAGAGAGAA
3b AGTACCTTACCTACTGGGCCTCTCTCTTTCGA
4a AGCTTAGACCAGATCTGAGCCTGGGAGCTCTCTTA
4b ATCTGGTCTAGACTCGGACCCTCGAGAGAATTCGA
The
oligonucleotides were removed from the synthesis
column
by elution with 3x1 ml of ammonia. This solution
was incubated at
55°C
overnight to deprotect the
oligonucleotides. To further purify the oligonucleotides, an
Applied Biosystems oligonucleotide purification cartridge
(OPC)
was used. To anneal complementary
single
stranded
oligonucleotides (i.e. la to lb) they were both incubated at a
concentration
of 0.05 M in TE. The solution was then heated
to
90°C
and allowed to cool
slowly
to
less
than 30°C. This
results in the formation of double stranded oligonucleotide
hybrids. To check the succession of the annealing the double
stranded oligonucleotide hybrids were run on an 8%
Polyacrylamide gel alongside the singlestranded oligo-
nucleotides.
Double
stranded oligonucleotides were 5' endlabelled
using γ
32
ΡΑΤΡ and T4 polynucleotide kinase and end
filled
using the Klenow fragment of DNA polymerase according to
Maniatis
et al (20).
Oligonucleotide
labelling.
The above described oligo-
nucleotide
hybrids and the oligonucleotide E
3
AP1
representing the region between nucleotides 81 and 103 at
the
ElAinducible E
3
promoter (21) were labelled with
γ
32
Ρ
ATP using T4 polynucleotide kinase from Boehringer. The
oligos
were incubated sequentially at
37°C
for 30 min, at
70°C
for 5 min, at 37° for 10 min, at RT for 5 min and on ice
for
5
min.
Gel
retardation
assays.
DNA binding reactions were carried
out
as
follows:
2000
cpm Y
32
Poligo were mixed with nuclear
proteins
(20 µg) in binding buffer (50 mM Hepes pH 8.0,
500 mM NaCl, 0.5 mM
PMSF,
0.5 mg/ml BSA, 20%
glycerol, 1 mM EDTA) plus 1 mM DTT and 150 µg/ml
poly(dldC).
The reaction mixture was
left
for 30 min at 0°C.
Samples were subjected to electrophoresis on 5%
Polyacrylamide
gels,
dried and exposed to
Xray
film (RX
Fuji,
Japan).
A rabbit polyclonal antibody to the human JUN protein
(a
gift
from Dr D. Gillespie, The Beatson Institute, Glasgow,
UK)
was employed in gel mobility shift
assays.
Results
The
HIV1 LTR contains four short DNA regions (lab to
4ab) which are homologous to the recognition site for the
cellular transcription factor AP1 (Fig. 1). From these four
sequences, the AP1
like
sequence in the region 2ab
shows
the
highest homology (85.8%) to the AP1 consensus and
therefore the corresponding oligonucleotide hybrid 2ab was
initially used to test the affinity of the AP1 protein in MDA
MB
468 and HeLa tumor cell lines. This was achieved by
preparing nuclear extracts, mixing them with y
32
Pend
labelled double stranded oligonucleotides E
3
AP1 and 2ab
and
analysing the formation of DNAprotein complexes by
gel retardation
assays.
In addition, we also examined whether
the
remaining nontested AP1like sites on HIV1 LTR are
functional.
In Fig. 2 is shown a panel of lanes with
competition
experiments between 2ab endlabelled
oligonucleotide and the representative set of AP1like sites
in
HIV1 LTR (lab or 2ab or 3ab or 4ab) oligonucleotides.
These competition reactions were performed on the MDA
MB
468 tumor cell line. The lab, 2ab and 4ab
oligonucleotides compete to a different extent depending on
their
homology to AP1 consensus (see Fig. 1) for the
labelled 2ab oligonucleotide. This is not surprising since lab
and
4ab are highly homologous (71.5% to the E
3
AP1 which
contains
the AP1 binding consensus from the E
3
promoter)
(21),
(Fig. 1). However, the nonlabelled oligonucleotide
hybrid 3ab did not compete for AP1 binding activity with
the
labelled oligonucleotide hybrid 2ab, presumably because
it shares only a limited homology (57.1%) to the AP1
consensus.
To
test further whether these putative AP1 binding sites
are functional, the control endlabelled oligonucleotide
hybrid E
3
AP1 was used in competition reactions with lab,
2ab 3ab, 4ab or nonlabelled E
3
AP1 oligonucleotide in
nuclear
extracts from the MDA MB 468 tumor cell line. As
shown in Fig. 3 nonlabelled oligonucleotides E
3
AP1, 2ab
and
4ab compete
well,
lab compete
weakly
and 3ab does not
compete
to the labelled oligonucleotide E
3
AP1 for DNΑ-
ΑΡ
1
protein complex formation. The oligonucleotide hybrid
2ab was then used to test the presence of AP1 activity in
HeLa
cells. Competition experiments (Fig. 4, lanes 3 to 5)
showed that in all cases nonlabelled lab, 2ab and 4ab
oligonucleotides compete for endlabelled 2ab oligo-
nucleotide.
We have shown the effect of JUN antibody on the
formation
of the DNAprotein complex between E
3
AP1 and
the
AP1 protein from HeLa nuclear extracts. In Fig. 5 is
shown the effect of JUN antibody on the formation of the
DNAprotein
complex between E
3
AP1 and the AP1 protein
in
HeLa nuclear extracts (lane 6).
Also
in the same Figure is
shown that the nonlabelled oligonucleotide hybrids lab or,
2ab and 4ab competed with the labelled oligonucleotide
hybrid E
3
AP1 for DNAAP1 protein complex formation
(lanes 35), therefore
suggesting
that these binding sites are
likely
to be functional.

ONCOLOGY
REPORTS
1:
397401,
1994
399
-
300
400
300
_l_
200
I
100
_L_
357
HIV-1
LTR

165  103
 81
 17
'fflfflffiffiffl
1
£££82£
/yy//yyyyyyy//y/
W/ps/Pfa
516
495
NRE
2
ao
3
ab
I
1 I 1
310
 287

231

208
16
· 42
5'-T
G
A
G
Τ
C
A
-3'
AP-1
consensus
5'-T G
A
C
Τ
Τ
A
-3'
1
ab
(-509
to
-503)
5'-T GAGÇCA-3'
2
ab
(-300
to
-294)
5'-T
G
A
C
Ç.
C
G-3'
3
ab
(-221 to-215)
5'-TGAGÇCT-3'
4ab
(+25 to+31)
Figure
1.
Schematic representation
of
the HIV-1 LTR.
The
locations
of
NRE, enhancer
and
TAR
sequences
are
shown.
The
nucleotide sequence
of
AP-1
consensus
is
shown
as
well
as the
nucleotide sequences
of
putative
AP-1
sites located
in the
HIV-1
LTR
alongside their nucleotide position. Nucleotide
differences between AP-1 consensus
and
HIV-1
LTR
AP-1 sequences
are
underlined.
MDA
MB 468
CELLS
CELLS
MDA MB
468
COMPETITOR
: -
PROBE
:
1
mr
\%
1
lw
_Q XI
CO
CM
2ab
r>
CÖ
CO
-a
CO
•if
|
1
COMPETITOR
PROBE:
f
1
i
i
ο
ω
.ο
m
,—
η
Cfl
CM
.a
co
•if
j
J3
1
C0
en
E3AP1
AP-1-»
« 4
V
AP-1
-*
Ì-A
-B
-C
FREE
DNA
LANE
Figure
2.
Effect
of
competitor sequences
on gel
electrophoretic mobility
shift. Nuclear extracts from
MDA MB 468
cells were incubated with
γ
32
Ρ
end
labelled
E
3
AP1
and 2abAPl
oligonucleotide
hybrid (lanes 1 and
26,
respectively).
In
competition
assays, 200fold excess
of
cold
competitor
oligonucleotides
1
ab,
2ab,
4ab
and
3ab
(lanes 36), were
incubated
with
the
same
nuclear
extracts.
The
DNAAP1
protein
complex
is
indicated
by
the
arrow.
FREE
DNA
LANE
Figure
3.
Effect
of
competitor
sequences
on gel
electrophoretic
mobility
shift.
Nuclear
extracts from MDA MB
468
tumor
cells were incubated with
Y
32
Pend
labelled E
3
AP1 oligonucleotide hybrid.
In
competition
assays,
200fold excess
of
cold
competitor
oligonucleotides E3API
and
lab, 2ab,
4ab
and
3ab
(lanes
2
and
36,
respectively), were
incubated
with the nuclear
extracts
before adding
the
probe.
The
DNAAP1
protein
complex
is

400
ZOUMPOURLIS
and SPANDIDOS: API
RECOGNIZES
SEQUENCE ELEMENTS ON HIV1 LTR
CELLS
COMPETITOR
:
HeLa
η η η
co
co cu
τ-
C\J ·**•
PROBE
:
JE
3
API
I 2ab }
AP-1
•*
•A
-Β
C
D
LANE
Figure 4. Effect of competitor sequences on gel electrophoretic mobility
shift. Nuclear extracts from HeLa cells,
were
incubated with y
32
Pend
labelled Ε,ΑΡ1 and 2abAPl oligonucleotide hybrids (lanes 1 and 25,
respectively). In competition
assays,
200fold
excess
of cold competitor
oligonucleotides lab, 2ab and 4ab (lanes 35),
were
incubated with the
nuclear extracts before adding the probe. The
DN
AAP1 protein complex is
indicated by the arrow.
Discussion
CELLS
:
Γ
ANTIBODY
:
Γ
COMPETITOR
:
PROBE:
[
a.
<
_
m
HeLa
a
a η
CO
CS 03
y- c\i *3-
63AP'
,1
1
1
AP-1
-
FREE
DNA
LANE
Figure 5. Effect of competitor sequences and JUN antibody on the AP1/
oligonucleotide complexes analyzed by gel electrophoretic mobility shift.
Nuclear
extracts from HeLa
cells
incubated with γ
32
Ρεηα labelled E
3
AP1
oligonucleotide hybrid. In competition
assays,
200fold
excess
of cold
competitor oligonucleotides lab, 2ab, 4ab and E,AP1 (lanes 24 and 5,
respectively),
were
incubated with the nuclear extracts before adding the
probe. IUN antibody and nuclear extracts from HeLa
cells
were
incubated
with Ε,ΑΡ1 oligonucleotide (lane 6). The
shift
in the mobility of the JUN
complex is indicated by the arrow.
In
this study we found that the HIV1 LTR contains three
novel AP1 binding sites. One of the three novel AP1 sites
was found, within the TAR region, one within NRE and the
other
further upstream. All three sites are functional since
their
corresponding oligonucleotide hybrids compete one
another
as
well
as the oligonucleotide hybrid E
3
AP1 which
contains
the AP1 binding site of the E
3
inducible promoter.
The
potential transcriptional regulation of HIV1 gene
expression by the AP1 binding site within NRE is further
supported
by the fact that Curran and his colleagues have
identified two additional AP1 sites in this LTR region (8).
However, the novel AP1 site we found within the TAR
region appears to be of even greater interest. Maximal
expression of proviral HIV1 DNA is attributed mainly to the
Tat
protein,
which is thought to function at the transcriptional
level through a nascent RNA copy of the TAR region.
According to Berkhout and his colleagues the Tat protein
could
target the LTR transcriptional unit directly bypassing
the
use of TAR RNA (22). In this instance it is possible that
AP1 protein could participate in this mechanism through
TAR DNA sequence.
Further
experiments are needed to test
this
hypothesis.
Although our study demonstrated novel AP1 binding
sites in the HIV1 LTR sequences, further studies would be
required to establish a functional relationship of these HIV
LTR
AP1 binding sites and the elevated AP1
levels
in the
HIV
life
cycle.
Acknowledgements
We would like to thank Dr E. Gonos for critical reading of
the
manuscript.
References
1. BarreSinousi F, Cherman JC, Rey F, Huyeyre MT, Chamaret S,
Gruest J, Dauguet C,
AlxerBlin
C, VenizetBrun F, Rouzioux C,
Rosenbaum W and Montagnier L: Isolation of a Tlymphotropic
retrovirus from a patient at
risk
for acquired immunodeficiency
syndrome (AIDS). Science 220: 868871, 1983.
2. Gallo RC, Salahuddin SZ, Popovic M, Shearer GM, Kaplan M,
Haynes BF, Palker TJ, Redfield R, Oleske J, Safai B, White G,
Foster Ρ and Markhan PD: Frequent detection and isolation of
cytopathic retrovirus (HTLVII) from patients with AIDS and at
risk
for AIDS. Science 224:
500503,1984.

ONCOLOGY
REPORTS
1: 397401, 1994
401
3.
Levy
SA, Hoffman AD, Kramer
SM,
Landis JA, Shimabukuro JM
and
Oshiro LS: Isolation of lymphocytopathic retrovirus from
San Francisco patients with AIDS. Science 225: 840842, 1984.
4. Varmus H: Regulation of HIV and HTLV gene expression.
Gene
Develop 2:
10551062,
1988.
5. Rosen CA: Regulation of HIV gene expression by RNAprotein
interactions.
Trends Genet 7: 914, 1991.
6.
Vaishnav
VN and WongStaal F: The biochemistry of AIDS.
Ann Rev Biochem 60:
577630,
1991.
7. Greene WC: Regulation of HIV1 gene expression. Annu Rev
Immunol
8: 453475, 1990.
8. Franza BR, Rouscher FJ, Josephs SF and Curran T: The fos
complex and /osrelated antigens recognize sequence elements
that
contain AP1 binding sites. Science 239: 11501153, 1988.
9. Garcia JA, Wu FK, Mistuyasu R and Gaynor RB: Interactions
of cellular proteins
involved
in the transcriptional regulation of
the
human immunodeficiency
virus.
EMBO J 6:
37613770,
1987.
10. Wu FK, Garcia JA, Harrich D and Gaynor RB: Purification of
the
human immunodeficiency
virus
type 1 enhancer and TAR
binding proteins EBP1 and
UBP1.
EMBO J 7:
21172129,
1988.
11.
Jones KA, Kadonaya JT, Luciw PA and Tjian R: Activation of
the
AIDS retrovirus promoter by the cellular transcription factor
Spi. Science 232: 755759, 1986.
12. Sawadoyo M and Sentenac A: RNA polymerase Β (II) and
general transcription factors. Annu Rev Biochem 59:
711754,
1990.
13.
Berkhout B, Silverman RH and Jeang KT: Tat rransactivates
the
human immunodeficiency
virus
through a nascent RNA
target. Cell 59: 273282, 1989.
14. Sharp PA and Marciniak RA: HIV TAR: An RNA enchancer?
Cell 59:
229230,
1989.
15.
Lee W, Haslinger A, Karin M and Tjian R: Activation of
transcription by two factors that bind promoter and enhancer
sequences of the human metallothionein gene and
SV40.
Nature
325: 368372, 1987.
16. Lee W, Haslinger A, Karin M and Tjian R: Two factors that
bind and activate the human metallothionein IIA gene in
vitro
also recognize the
SV40
promoter and enchancer regions.
Nature
325: 368372, 1987.
17. Curran Τ and Franza JR: Fos and Jun: The AP1 connection.
Cell 55: 395397, 1988.
18. Gillespie DAF:/os
ana
jun: oncogenes and transcription factors.
Semin
Virol
2: 329339, 1991.
19. Bradford
MM:
A rapid and
sensitive
method for the quantitation
of microgram quantities of protein
using
the principle of
proteindye binding. Anal Biochem 72:
248254,
1976.
20. Maniatis T, Fritsch EF and Sambrook J: Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory, Cold
Spring Harbor, NY, 1989.
21.
Hurst HC and Jones NC: Identification of factors that interact
with the ElAinducible adenovirus E
3
. Gene Develop 1: 1132
1146, 1987.
22. Berkhout B, Gatignol A, Rabson AB and Jeang KT: TAR
independent
activation of the HIV1 LTR: evidence that Tat
reguires
specific
regions of the promoter. Cell 62: 757767,
1990.