Note: Descriptions are shown in the official language in which they were submitted.
W092/14755 1 ~1 0 1~19 PCT/US92/01227
A NOVEL INTEGRIN SPECIFICITY FOR
THE HIV Tat PROTEIN
This invention was made in part with Government
support under Grant Nos. CA 42507, CA 28896, CA 086~6, CA
5~412 and CA 30199 from the National Cancer Institute.
The United States Government may have certain rights in
this invention.
BACKGROUND OF THE INVENTION
The present invention generally relates to the
HIV Tat protein and more particularly to the integrin
cell surface receptor capable of binding to the HIV Tat
protein.
Human immunodeficiency virus (HIV-l) encodes a
regulatory protein, termed "Tat", that transactivates
genes expressed from the long terminal repeat of the
virus. Tat exists as either a 72 or 86 amino acid
protein, depending on whether it is expressed from one or
both coding exons through differential splicing. Both
forms are functional transactivators and contain an
acidic domain at the amino terminus, a basic region (six
arginines and two lysines out of nine contiguous
residues) and a cysteine-rich region ~seven out of
sixteen residues). The basic domain has been shown to be
responsible for nuclear/nucleolar targeting of the Tat
protein as well as its capacity to bind to its cognate
RNA sequence, TAR. Although the cysteine-rich domain is
also critical for transactivation, its function is less
defined but has been proposed to mediate metal-linked
dimerization of the protein.
The 14-amino acid sequence encoded by the
second exon has the tripeptide arginine-glycine-aspartic
acid (RGD) as its most notable feature. An RGD sequence
is required for integrin mediated cell adhesion to
wo g2,l475~ 2 ~ ~ 1 9 ~ 9 PCT/US92/01227
extracellular matrix proteins such as fibronectin,
vitronectin, and fibrinogen as reported in Pierschbacher
and Ruoslahti, Nature 309:30-33, (1984) and Ruoslahti and
Piersch~acher, Science 238:491-497, (1987).
It has recently been demonstrated that Tat may
function as an exogenous factor. Extracellular Tat is
internalized by cells and transported to the nucleus,
where it retains the ability to transactivate the HIV
promoter as discussed in Frankel and Pabo, S~ll 55:1189-
1193 (1988). In addition, Tat can be released from HIV-l
acutely infected cells or Tat-transfected cells. Similar
processes of release and uptake have been observed with
another retroviral transactivator protein, namely, Tax of
HTLV-1. Furthermore, Tat has been shown to modulate cell
proliferation, both in the suppression of proliferation
of antigen-acti~ated T-cells and in the specific
stimulation of proliferation of Kaposi's sarcoma (KS)
derived cells as reported in Ensoli et al., Nature
345:84-86 (1990). Thus, Tat may contribute to the
pathogenesis of HIV via mechanisms that go beyond the
activation of virus replication. However, the details of
these mechanisms are not well understood. For example,
it is not known how extracellular Tat stimulates KS cell
growth or how it is internalized.
One mechanism underlying the interactions of
cells with one another, with extracellular matrices and
with soluble proteins, such as the Tat protein, is
binding of cell surface receptors to a cognate ligand on
the surface of cells or in the extracellular matrix. In
many cases, such a receptor belongs to a class of
proteins known as integrins.
The integrins are a large family of cell
surface glycoproteins that mediate cell-to~cell and cell-
to-matrix adhesion as described, for example, in
W092/1475~ PCT/US92tO1227
3 2 1 01 919
Ruoslahti and Pierschbacher, supra (1987). All known
members of this family of adhesion receptors are
heterodimers consisting of an ~ and a ~ subunit
noncovalently bound to each other. Over the past few
years, the primary structures of six integrin B subunits
from mammalian cells and one from Drosophila have been
- deduced from cDNA. Eleven distinct ~ subunits have thus
far been described.
The adhesion of cells to extracellular matrices
is mediated in many cases by the binding of a cell
surface receptor to an RGD containing sequence in the
matrix protein, as reviewed in Ruoslahti and
Pierschbacher, supra (1987). The RGD sequence is a cell
attachment site at least in fibronectin, vitronectin,
fibrinogen, von Willebrand factor, thrombospondin,
osteopontin, and possibly various collagens, laminin and
tenascin. Despite the similarity of their cell
attachment sites, these proteins can be recognized
individually by their interactions with specific
receptors.
Because of the importance of the Tat protein in
the maintenance and propagation of HIV infection, a need
exists to control the effects of this protein on normal
and HIV infected cells. The present invention satisfies
this need and provides related advantages as well.
SUMMARY OF THE INVENTION
The present invention relates to methods of
inhibiting HIV Tat protein binding to a cell expressing
an HIV Tat binding integrin cell surface receptor
("integrin"). The methods are accomplished by blocking
the binding of the HIY Tat protein to the integrin by
binding the HIV Tat protein with a reagent having
reactivity with the integrin binding site of the HIV Tat
WO92/147~ 9 PCT/US92/01227
protein or by binding the integrin with a reagent having
reactivity with the HIV ~at binding site of the integrin.
The integrin can be, for example, human ~vB5 or rat ~V~8-
The The reagent can be an antibody or active fragments of
either the inte~rin or the HIV Tat protein.
The present invention further provides
fragments of the HIV Tat protein that can be used in the
claimed methods. Such fragments containing an integrin
binding site include, for example, the basic domain of
the HIV Tat protein. A secondary binding site for cell
adhesion may be an RGD-containing region of the Tat
protein. Thus, the fragments of the present invention
can also include an RGD-containing domain of the HIV Tat
protein.
Isolated nucleic acids encoding the HIV Tat
protein, a Tat binding integrin or reactive fragments of
either are also provided. The invention further relates
to vectors having a nucleic acid encoding such
polypeptides or fragments and to host cells containing
these vectors.
Methods of detecting ligands of the Tat binding
integrins in a sample are also provided. Such methods
include contacting a Tat binding integrin with the sample
and determining the binding of the integrin to the
sample. Binding of the integrin to the sample indicates
the presence of an integrin-reactive ligand. The methods
can be used to detect or purify HIV Tat proteins, active
fragments of such proteins or HIV Tat peptide mimetics.
The present invention additionally relates to
methods of increasing the binding of HIV Tat proteins to
cells expressing the integrins of the present invention
by overexpressing the integrins.
W092/14755 PCT/US92/01227
2l ~l 9l 9
BRIEF DESCRIPTION OF THE ~RAWINGS
Figure l depicts the structure of the Tat
protein and Tat-derived peptides. Shown are the cysteine
rich domain and the basic domain in exon l and the RGD
sequence in exon 2.
Fig~re 2 shows the results of cell adhesion to
Tat peptides. Figure 2A shows the results of rat L8
cells; Figure 2B shows the results of human SK-LMS cells.
Wells of microtiter dishes were coated with various
concentrations of Tat 1-86 (O), Tat 45-86 (~), or Tat 57-
86 (-). L8 cells (-105 cells per well) were added to each
well and incubated at 37-C for one hour. The attached
cells were fixed and stained with crystal violet. The
dye was eluted and the absorbance at 600 nm measured in
an ELISA reader. SK-LMS cells were tested for attachment
to the same peptides, including a 12 amino acid peptide
consisting of the basic domain, Tat 45-57 (~).
Figure 3 shows the results of studies relating
to the inhibition of cell attachment to Tat with anti-~vB3
VNR antibodies. Wells of microtiter dishes were coated
with either fibronectin, vitronectin or Tat peptide 45-86
at 5 ~g/ml. SK-LMS or L8 cells (-lOs cells per well) were
added to each well in the presence of anti-~vB3 (VNR) or
anti-5~1 (FNR) integrin antibodies. The attached cells
were fixed and stained with crystal violet. The dye was
eluted and the absorbance was measured in an ELISA
reader.
Figure 4 shows the affinity chromatography on
Tat peptide. An extract of surface iodinated L8 cells
was fractionated on Tat peptide 45-86 coupled to
Sepharose. After washing, the column was eluted with
either l mg/ml of the Tat peptide or 200 ~g/ml of the
f1~ll-length Tat protein. The fractions were analyzed by
WO92/147ss PCT/US92/01227
2101919 6
SDS-polyacrylamide gel (7.5~) electrophoresis under non-
reducing conditions.
Figure 5 shows the results of
immunoprecipitation of material affinity-purified on Tat
peptide. An extract of surface-iodinated SK-LMS cells
was fractionated on Tat 45-88 coupled to Sepharose as in
Figure 4. After sequential elution with GRGDSP, Tat 57-
86 and Tat 45-57 ~basic 12-mer), peak fractions of each
eluate were immunoprecipitated with antibodies to the
B1, ~ or B5 subunit cytoplasmic domain, and analyzed by
SDS-polyacrylamide gel (7.5%) electrophoresis under non-
reducing conditions. Shown are the immunoprecipitations
of the flow-through (unbound), GRGDSP eluate and Tat 45-
57 eluate (basic 12-mer).
Figure 6 shows the results of
immunoprecipitation o~ material affinity-purified on
GRGDSPK-Sepharose. An extract of surface-iodinated SK-
LMS cells was fractionated on GRGDS~K-Sepharose. After
elution with GRGD~P (1 mg/ml), the flow-through (unbound)
and eluate (bound) was immunoprecipitated with antibodies
to the ~v~ B1, ~3 or B5 subunit cytoplasmic domains, and
analyzed by SDS-polyacrylamide gel (7.5%) electrophoresis
under non-reducing conditions.
Figure 7 shows the elution of integrin from Tat
column with EDTA or NaCl. Extracts of surface-iodinated
L8 or SK-L~S cells were fractionated on Tat 45-86 coupled
to Sepharose. Sequential elution of the columns with 10
mM EDTA (in 100 mM NaCl; lanes 2,6), 250 mM NaCl (lanes
3,7) and Tat 45-86 (1 mg/ml; lanes 4,8), followed by
immunopr~cipiTation with a polyclonal anti-vitronectin
receptor antibody and analysis by SDS-polyacrylamide gel
(7.5~ electrophoresis under ~on-reducing conditions are
sh~wn. Lanes 1 and 5 are immunoprecipitates of the flow-
through material.
W092/1475~ PCT/US92/01227
7 2101919
Figure 8 shows the results of inhibition of SK-
LMS cell binding to vitronectin and Tat peptide. SK-LMS
cells were added to microtiter wells coated with Tat 45-
86 or vitronectin (VN) in the presence or absence of
inhibitory antibodies. P3G2 is a monoclonal antibody
shown previously to inhibit the interaction f ~VBs with
vitronectin. LM 609 inhibits the interaction f ~VB3 with
vitronectin. Anti-VNR is a polyclonal antibody raised
against the ~vB3 integrin purified from a placental
extract on a GRGDSPK-Sepharose column.
Figure 9 shows the effect of antibodies and
peptide on Tat transactivation. LTR-CAT was transfected
into L8 cells and transactivated by exogenous GST-Tat
fusion protein in the presence of anti-~vB3 polyclonal
antibody (anti-VNR), rabbit anti-~sB1 (anti-FNR), normal
rabbit serum, or ~at basic peptide. Lane l, LTR-CAT
alone; lane 2, LTR-CAT plus GST-Tat fusion protein; lane
3, LTR-CAT plus GST protein; lanes 4-8, LTR-CAT plus GST-
Tat fusion protein; and lane 4, 300 ~g basic peptide;
lane 5, 80 ~l anti-~vB3; lane 6, 8 ~l anti-~vB3; lane 7, 80
~l anti-a5B1 serum; lane 8, normal rabbit serum.
DETAILED DESCRIPTION OF THE INVENTION
The present invention relates to integrin cell
surface receptors, referred to herein as "integrins,"
that bind to the HIV Tat protein. In human cells, such
integrins are the known ~vB5~ while rat cells may contain
a previously unidentified B subunit, designated herein as
B8, and the known subunit ~v The ~v subunit is the most
versatile of the integrin ~ subunits. It has been known
to combine with three, possibly four, different B
subunits. Thus, the rat B8 subunit of the present
invention may add at least o~e additional B subunit to
this group. The human B5 subunit was previously known to
combine with ~v~ although the Tat binding activity of
WO92/147S5 ~101919 PCT/US92/01227
human ~vB5 was not previously known.
The present invention particularly relates to a
novel interaction between the HIV Tat protein and the
integrins to which the Tat protein bind. Although
previous reports suggest an RGD-dependent mechanism in
the binding of Tat to rat L8 cells, as described in Brake
et al., J. Cell Biol. 111:1275-1281 (1990), it has now
been discovered from cell adhesion and affinity
chromatography data that the basic region of the Tat
protein mediates this interaction. Immunoprecipitation
of the protein purified on the Tat columns identified the
Tat binding proteins as components of the integrin ~vB
This result was consistent with data showing that cell
adhesion to Tat could be blocked by polyclonal antibodies
to the ~vB3 integrin. Antibodies to this integrin would
be expected to bind to the ~v subunit shared by ~V~5 and
B3 and inhibit the function of both heterodimers.
The ~VBs integrin has been shown to bind to
vitronectin through the RGD sequence. Although the Tat
protein contains an RGD sequence, it has been shown in
the studies relating to the present invention that the
integrin recognition sequence is the basic domain of Tat.
This conclusion is based on the complete correlation
found between the ability of Tat-derived peptides to
support cell attachment and bind the ~VBs integrin in
affinity chromatography with the presence of the basic
sequence in the peptide. In contrast, the presence of
absence of the RGD sequence had no influence in either
type of assay. Apparently the RGD sequence is present in
a context not suitable for integrin binding because even
the ~vB3 integrin, which has the greatest binding activity
with short RGD-containing peptides as shown in Figure 6
failed to bind appreciably to the RGD-containing Tat
peptides.
WO92/147S~ 21 0 1 91 9 PCT/US92/01227
Most integrins require divalent cations for
their activity and are dissociated from their ligands in
the presence of EDTA. In addition to utilizing the basic
domain, the interaction between ~VB5 and Tat has the
unusual feature that it is stable in the presence of 10
mM EDTA and is therefore not divalent cation-dependent.
The interaction of the ~5 integrin with Tat was,
however, inhibited by high NaCl concentrations. Although
this salt sensitivity is unusual among integrins, it is
not unprecedented. For example, the ~3B1 integrin also
demonstrates a salt-labile, RGD-independent binding to
collagen and laminin as discussed in Elices et al., J.
Ce~ Biol. 112:169-181 (1991). It could be that ~3~1 also
binds to a basic region within these proteins.
Interestingly, ~3B1 also has been found to bind
to fibronectin in an RGD-dependent manner, suggesting
that a3B1 integrin may have two functionally distinct
ligand binding sites. The a4B1 integrin has two distinct
ligand binding sites, one for the endothelial cell
ligand, V-CAM, and the other for an alternatively spliced
segment of fibronectin as reported in Elices et al., Cell
60:577-584 (1990). The inability to inhibit the binding
of L8 and SK-LMS cells to Tat with a monoc}onal antibody
that inhibits their interaction with vitronectin suggests
that ~VB5 may have a second ligand binding site for a
basic extracellular protein yet to be identified. The
integrin IIb/IIIa may also share some of the basic
peptide binding properties of ~yBs. Peptides containing
both an RGD and a basic segment bind more avidly to
IIb/IIIa than peptides containing RGD alone as reported
in Savage et al., J. Biol. Chem. 265:11766-11772 (1990).
Both rat L8 and human SK-LMS cells remained
round when plated on a surface coated with Tat and Tat-
derived peptides containing the basic domain. This
behavior is consistent with the finding that cells
W092/14755 2 1 0 1 9 1 9 PCT/US92/01227
expressing ~5 attach to vitronectin, but do not spread.
According to réports in Wayner et al., J. Cell Biol.
113:919-929 (1991), spreading on vitronectin is dependent
on the presence of ~vB3
Based on early studies with rat L8 cells, the
Tat was found to bind an integrin containing a B subunit
designated herein as B8 and the known ~v In further
studies wit~ human cells, the Tat protein was found to
bind the known human integrin, ~VBs. However, the rat
integrin may also be ~VBs because it behaved identically
to the human ~VBs integrin in the affinity chromatography
experiments and migrated similarly in SDS-PAGE. The
antibodies prepared against the cytoplasmic peptide of
the human Bs subunit were weakly reactive against a number
of rat cell lines. Poor reactivity of the antibodies
with the rat Bs subunit may therefore explain the lack of
immunoprecipitation of the L8 integrin observed. It is
also possible that the B subunit of the Tat binding
integrin from the L8 cells may be an alternatively
spliced Bs variant or a completely different B subunit
altogether.
The rat B8 subunit is distinguishable from known
B subunits. First, direct comparisons show that the
apparent molecular weight of the B8 subunit is lower than
many of the previously characterized B subunit. In
addition, although the ~v subunit has been shown to
associate with the B1, B3, or Bs subunits and may also
combine with B6, antibodieæ to the cytoplasmic tails of
each o~ these known subunits failed to immunoprecipitate
the Tat binding receptor. The possibility that the B
subunit differs from a previously characterized B subunit
by alternative splicing or proteolytic processing cannot
be excluded. However, it is clear that the L8 cells have
a B3 subunit that is functional, but does not bind to Tat.
The ~v subunit is the most versatile of the integrin ~
W092/147SS -21dl`919 PCT/US92/01227
subunits in that it is known to combine with three,
possibly four, different B subunits. It now appears that
at least one additional ~ subunit can be added to this
group.
The rat ~8 integrin is likewise
distinguishable from the known integrins in its binding
specificity. For example, the ~4~ integrin binds to
sequences that do not contain the RGD seque~ces, yet the
Tat protein does not contain any sequences significantly
homologous to the ~4B1 target sequence.
An important characteristic of the B subunit is
that in combination with the ~v subunit it forms the
integrins of the present invention that binds to the HIV
Tat protein. The specificity of the integrins of the
present invention is such that it can bind the intact Tat
protein or a truncated form lacking the RGD-containing
règion and, therefore, can be used to control the
activities of the Tat protein.
The B component of the Tat binding integrins
gives a double band in a non-reduced SDS-gel
electrophoresis in the molecular weight range of about
75-95 kD, representing two forms of the same B subunit or
two different subunits. The present invention is
contemplated to include both.
The full length Tat protein contajns the
following amino acid sequence:
MEPVDPRLEPWKHPGSQPKTACTNCYCKKCCFHCQVCFITXALGISYGRRKRRQRRR
AHQNSQTHQASLSKQPTSQSRGDPT~PKE. A basic domain of the Tat
protein has been found to be the dominant binding site
for the integrin of the present invention. This basic
region contains nine amino acid residues of which six are
arginine and two are lysine. The amino acid sequence of
the basic region is RKXRRQRRR in the HI~ strain HXB2.
WO92/14755 2 1 0 1 9 1 9 PCT/US92/0122,
12
In studies to determine the role of RGD in
binding the Tat protein, it was found that a variant
peptide in which RGD was replaced with KGE was still
capable of binding to the Tat protein with only a small
r~duction in the binding activity. Whereas Brake et
al., sU~ra (1990) found that the cell attachment to the
Tat protein was inhibited by short RGD-containing
peptides, such peptides did not have a significant effect
on the binding of the Tat protein to the integrin in
affinity chromatography. This result was supported by
later studies in which RGD was replaced by KGE. Thus, it
appears that the RGD-containing region is a secondary
binding site of the Tat protein. This secondary binding
region contains the amino acid sequence: PTSQSRGDPTGPKE
in the HIV strain HXB2.
Other ligands which bind to tne Tat binding
integrins can contain both a basic domain and a nearby
RGD sequence. While the basic region appears to be the
primary binding site of these integrins than the RGD-
containing region, at least some of the RGD-directed
integrins recognize basic regions in addition to the RGD-
containing sequences. This conclusion is supported in
part by studies in which peptides that contain both an
RGD and a basic segment bound more avidly to the ~I~JB3
integrin than peptides containing RGD alone as described
in Savage et al., J. Biol. Chem. 265:11766-11772 (199O).
Accordingly, the present invention also
provides isolated fragments specific to the Tat binding
integrins, particularly to the B subunmits such as, for
example, human B5 and rat B8. Such fragments can be
regions obtained from the native Tat protein or synthetic
polypeptides having amino acid sequences of the Tat
protein regions. These fragments are necessarily of
sufficient length to be distinguishable from fragments of
other known Bs and, therefore, are "specific to" or
W092/14755 PCT/US92/01227
13 210~9~9
"unique to" the B subunits of the Tat binding integrins,
particularly human B5 and rat B8, for example. Such
fragments specific to B8 can be determined using methods
disclosed herein and known in the art. These fragments
also retain the Tat binding function and can therefore be
used, for example, as inhibitors of the binding of the
Tat binding integrins to the Tat protein, or as an
indicator to detect the Tat binding integrins of the
present invention. One skilled in the art can determine
other uses for such fragments.
The terms "substantially purified" or
"isolated" mean that the material, either polypeptide or
nucleic acid, is substantially free of contaminants
normally associated with the native or natural
environment. A reactive fragment of the HIV Tat protein
refers to a polypeptide having substantially the amino
acid sequence o~ a portion of the ~IV Tat protein that
retains the integrin binding site so as to remain
substantially reactive with the Tat binding integrins.
Similarly, a reactive fragment of a Tat binding integrin
refers to a polypeptide having substantially the amino
acid sequence of a portion of the integrin that retains
the Tat binding function. Thus, modifications of the
amino acid sequence that do not substantially destroy the
functions and that retain the essential sequences of rat
B8 or human B5 are included within the definition of these
B subunits. Amino acid sequences, such as those for B1,
B2 and B3, having less than 50% homology with the sequence
of B or B5 are not substantially the same sequence and,
therefore, do not fall within the definition of these B
subunits. Given the amino acid sequences set forth
herein, additions, deletions or substitutions can be made
and tested to determine their effect on the function of
the B subunits. In addition, one skilled in the art
would recognize that certain amino acids can be modified
to alter integrin binding function.
WO 92tl475~ PCT/US92/0122~
` ~10~9t 9 14
The invention also provides reagents having
specificity for the integrins of the present invention
and particularly the rat B8 and human Bs subunits. One
skilled in the art could readily make reagents, such as
antibodies, that are specifically reactive with the B
subunits of the present invention. Such reagents can be
used to immunologically distinguish these B subunits from
other molecules. Various methods of raising such
antibodies are well established and are described, for
example, in Antibodies, A k~boratoEy_~anual, E. Harlow
and D. Lane, pp. 139-283 (Cold Spring Harbor Laboratory,
1988), incorporated herein by reference.
The invention further provides isolated nucleic
acids that encode the Tat protein or reactive fragments
thereof having binding sites or regions recognized by the
B-containing integrins of the present invention.
Similarly, isolat~d nucleic acids encoding the Tat
binding integrins or reactive fragments thereof are also
provided. Following standard methods as described, for
example, in Maniatis et al., Molecular Clonina, Cold
Spring Harbor (1989), these nucleic acid sequences can be
identified, isolated and then cloned into an appropriate
expression vector. The vector can then be inserted into
a host, which will then be capable of expressing the Tat
recombinant proteins, Tat binding integrins or reactive
fragments of each. Thus, the invention also relates to
vectors containing nucleic acids encoding such sequences
and to hosts containing these vectors.
The present invention further relates to
nucleic acids that can be used as probes for diagnostic
purposes. Such nucleic acid probes can hybridize with a
nucleic acid having a nucleotide sequence specific to the
Tat binding integrins or to the Tat protein but do not
hybridize with nucleic acids encoding non-Tat binding
integrins or Tat proteins, particularly other cell
WO92/1475~ PCT/US92/0122~
21~1919
surface receptors or non-Tat proteins, respectively.
Nucleic acids are also provided which specifically
hybridize to either the coding or non-coding DNA of the
Tat binding integrins or the Tat protein. Such nucleic
acids can be identified and prepared by methods known in
the art, such as a standard nucleic acid synthesizer.
In the methods of inhibiting the binding of Tat
proteins to the Tat binding integrins provided by the
present invention, the binding of the Tat binding
integrins, such as ~8 or avBs, to the Tat protein can be
blocked by various means. For example, the binding of a
B8- or Bs-containing integrin can be blocked by a reagent
that binds the B subunit or the B-containing integrin.
Examples of such reagents include, for example, peptides
and polypeptides containing the basic region of the Tat
protein or antibodies specifically reactive with the B
~ubunit or a B-containing integrin. Thus, the binding of
the Tat protein ~o the Tat binding integrins can be
blocked by binding the integrins with Tat-derived
peptides having an amino acid sequence that recognizes
the Tat binding site of the integrins. Alternatively,
blocking can be carried out by binding the Tat protein or
a reactive fragment thereof with a reagent specific for
the Tat protein at a site that inhibits the Tat protein
from binding with the integrin, such as an anti-Tat
antibody or a reactive fragment of the integrin.
The ability to block the binding of the Tat
protein by the Tat-binding integrins can be used to
control HIV mediated conditions. Thus, the activities of
the Tat protein as a growth factor for Kaposi's sarcoma
cells and as a transactivator of the HIV promoter when
internalized by cells can be inhibited or otherwise
controlled.
WO92/1475~ %i~9~9 16 PCT/US92/01227
Since the binding of the Tat binding integrins
to the Tat protein can mediate cell adhesion, preventing
this binding can prevent the adhesion of cells to the
endogenous ligand of the integrin. Other activities of
the integrin can be similarly inhibited by the use of the
Tat mimicking compounds of the present invention.
Alternatively, cell adhesion can be promoted by
increasing the expression of these integrins by a cell.
The activities of the endogenous ligand or ligands can be
mimicked with the compounds of this invention.
.
As a method of controlling the binding of Tat
proteins to cells expressing the Tat binding integrins,
the binding can be enhanced by methods in which the Tat
binding integrin is overexpressed in such cells. Methods
of enhancing the overexpression of such integrins can be
accomplished, for example, by introducing a nucleic acid
encoding for the integrin into the genome of a cell by
methods known to those skilléd in the art.
The present invention further provides methods
of detecting ligands that bind the Tat binding integrins.
The methods include contacting the integrin or binding
fragment thereof with a solution containing ligands known
to or suspected of binding the Tat binding integrins.
Ligands that bind such integrins are then detected.
Assays useful to carry out these methods are well known
in the art and are described, for example, in Hautanen et
al., J. Biol. Chem. 264:1437-1442 (1989) and Smith et
al., . Biol. Chem. 265: 11008-11013 (1990), both of
which are incorporated herein by reference.
These methods can be used to identify or screen
for additional compounds that are bound by the Tat
binding site on the integrins of the present invention.
Such compounds can be naturally occurring ligands, such
as derivatives of the Tat binding site or other compounds
WO92/14755 PCT/US92/012
17 2 1 ~ 19 19
capable of being bound by the integrin Tat binding site.
Also included are synthetic compounds designed to mimic
the desired binding activity of the Tat protein or other
Tat binding integrin binding ligands. Such compounds
having the desired binding function can be peptides,
peptide derivatives or mimetics, or other compounds, so
long as they are capable of being bound by the Tat
protein's integrin binding site.
The following examples are intended to
illustrate but not limit the invention.
EXAMPLE I
Peptide Svnthesis
The intact Tat protein was synthesized using t-
butoxycarbonyl (BOC)-protected amino acids for stepwise
synthesis on an Applied Biosystems 431A (Foster City, CA)
solid phase automated peptide synthesizer. Amino acids
were added as hydroxybenzotriazole (HOBt) esters using n-
methylpyrrolidone as the coupling solvent. The synthesis
was accomplished starting with 0.5 mmol of Boc-Glu(OBzl)-
O-phenylacetamidomethyl resin (0.69 g substituted at 0.72
mmol) with a minimum of two couplings for each amino
acid. The average repetitive coupling efficiency was
99.32% as determined by a qua~itative ninhydrin assay.
The cysteine sulfhydryls were protected with a p-
methylbenzyl group to yield a fully reduced form afterlow/high HF cleavage using suitable scavengers. All
other peptides used were synthesized with an Applied
Biosystems Model 43OA synthesizer using similar
chemistry.
EXAMPLE II
Isolation of Tat-binding Proteins
To identify the integrins or other cell surface
W092t147~; PCT/US92/Ot22/
2~0~9~9 18
molecules capable of binding to Tat, affinity
chromatography according to Example VII was performed on
the 86-amino acid Tat protein and the Tat derived
p~ptides. L8 rat myoblast cells (ATCC # CRL-1769) were
chosen for thi experiment since they had previously been
shown to bind to Tat and vitronectin but not to laminin,
collagen, or fibronectin as described in Brake et al.,
supra, (1990). In SDS-PAGE under non-reducing
conditions, a band of 150 kD and a doublet of 75-95 kD
were identified that bound to a full length Tat protein
coupled to a Sepharose column and eluted with 1 mg/ml of
a Tat peptide (amino acid residues 4S-86) containing the
basic region and the RGD region but lacking the cysteine
rich region of the protein. A second peptide
representing the cell adhesion site of fibronectin,
GRGDSP, was not effective at eiuting the Tat binding
proteins.
In another study, the peptide covering;residues
45-86 coupled to Sepharose bound the same proteins.
Identical bands were eluted with the truncated peptide
(residues 45-86) or the full length Tat protein,
demonstrating that the peptide and protein had similar
binding sites. An additional band of approximately 120
kD was eluted from the full length Tat protein column,
but not from the shorter Tat peptides. This band has not
been further identified, but may represent a subunit of a
second binding integrin.
To investigate the role of the RGD sequence
further, the peptide that contained residues 45-86 was
synthesized with the KGE sequence substituted for RGD.
The same proteins were bound to the XGE variant peptide
and were eluted from the Tat column with this variant
peptide indicating that the role of RGD in binding Tat is
not a primary one.
W092/147s~ PCT/US92/0122
19 21~1919
To determine whether these cells contained RGD
binding proteins that did not bind to Tat, affinity
chromatography on GRGDSP-Sepharose was performed with the
unbound material from the Tat peptide column. A band
with a similar electrophoretic mobility to the 150 kD Tat
binding protein together with a band distinct from the 85
and 95 kD bands obtained from the Tat column were eluted
from the GRGDSP-Sepharose column. These bands were
identified as the subunits of the ~vB3 integrin.
To identify the integrins or other cell surface
molecules capable of binding to Tat, affinity
chromatography was performed using the 86-amino acid Tat
protein and the Tat peptides. As in the cell attachment
experiments shown in Figure 2, all of the peptides that
~ontained the basic domain of Tat gave similar results
with rat L8 and human leiomyosarcoma SK-LMS (ATCC No. HTB
88) cells. Shown in Figure 4 are the proteins from an
iodinated L8 cell extract bound to the Tat 45-86 peptide
and eluted with this peptide or with full-length Tat. In
all experiments with peptides containing the Tat basic
region, bands of 150 kD and a doublet at approximately 90
kD were eluted from the column. Changing the RGD
sequence to KGE or deleting the second exon entirely had
no discernible effect on the identity of the proteins
eluted from the column. Even the 12 amino acids
comprising the basic domain were sufficient to bind and
elute these bands (see Figure 5). However, peptides
lacking the basic domain did not bind or el~te
significant amounts of iodinated cell surface proteins.
Figure 5 shows an immunoprecipitation of
proteins eluted from the Tat 45-86 peptide column. The
column was eluted sequentially with the peptide GRGDSP
followed by the Tat 57-86 peptide and finally the 12
amino acid peptide containing the basic domain of Tat.
Shown is the material eluted from the column with the
W0 92/1475~ 9 19 PCr/US92/01227
indicated peptide after immunoprecipitation with
antibodies to the specified integrin subunits. A small
amount of material eluted with the peptide GRGDSP could
be immunoprecipitated with B1 and B3 antibodies. However,
5 the maj ority of the B1 and B3 subunit-associated material
did not bind to the column and was detected in the
unbound fraction. This suggests that the peptide GRGDSP
and other peptides lacking the basic domain were
relatively ineffective at eluting the proteins from the
lO Tat peptide columns. In contrast, the majority of the
material eluted with the basic peptide was detectable
with the anti-B5 subunit antibody ( Figure 5 ) . The
heterogeneity of the Bs subunit may have been due to
partial proteolysis resulting from harvesting the cells
15 for the chromatography with trypsin. No labeled material
could be immunoprecipitated from the fractions eluted
with the Tat 57-86 peptide (not shown). These results,
together with the inhibition of cell attachment by the
antl-VNR serum ~Figure 3), indicate that the ~vB5 integrin
2 0 binds to the Tat protein and that this binding occurs at
the basic domain of Tat.
In contrast to the results obtained with the
basic peptide, affinity chromatography with the peptide
GRGDSPK coupled to Sepharose revealed a strikingly
different pattern (Figure 6). Whereas the predominant
integrin binding to Tat-Sepharose was ~VBs, the ~vB3
integrin was the only integrin substantially enriched in
the GRGDSPK-bound fraction (Figure 6). Although some ~VBs
was observed in the bound fraction, the majority of this
30 integrin was in the unbound fraction. This is in
agreement with reports that demonstrated a relatively
weak affinity f ~VBs for the peptide GRGDSPK as described
in Freed et al ., EMBO J. 8:2955-2965 (1989). Together
these data suggest that although these cells contain a
functional ~vB3 integrin that is capable of ~inding the
WO92/1475~ PCT/US92/01227
21 2101919
peptide GRGDSPK, the RGD sequence in Tat is present in a
context unfavorable to integrin binding.
Since the ~vB5 integrin binds to the basic
domain of Tat and not the RGD sequence, it appears to be
an unusual interaction of an integrin with its ligand.
To test this further, the EDTA-sensitivity of the
integrin association with Tat was determined. It is
known that integrins typically require divalent cations
to bind their ligands and can be eluted from ligand
affinity columns with EDTA. The interaction between ~V~5
and Tat was, however, insensitive to elution with lO mM
EDTA in affinity chromatography experiments (Figure 7).
The NaCl concentration in the solution containing lO mM
EDTA was lowered from 150 mM to lO0 mM to insure elution
would result from divalent cation chelation and not an
overall increase in the salt concentration of the
solution. Immunoprecipitations of the peak fractions
eluted with lO ~M EDTA, 250 mM NaCl, or the Tat 46-86
peptide revealed that the receptor was eluted from the
Tat column with high salt or with the peptide, but
chelating the divalent cations which are normally
required for integrin function was not effective in
disrupting this unusual integrin-ligand interaction.
Since the interaction between Tat and ~vB5 was
not a previously defined integrin-ligand binding,
experiments were undertaken to determine if Tat bound to
the same site on ~vB5 as vitronectin, the principal ~vB5
ligand. The monoclonal antibody P3G2 (Bristol Myers-
Squibb) had previously been shown to bind to ~vB5 and to
inhibit its interaction with vitronectin. This result
was reproduced using SK-LMS cells as shown in Figure ~.
However, the monoclonal antibody did not inhibit the
binding of the cells to Tat. As shown in Figure 3, a
polyclonal anti-~vB3 antibody did inhibit the interaction
of both Tat and vitronectin with the SK-LMS cells. A
WO9~/14755 2 1 0 1 9 1 9 ~2 PCT/US9~ 27 ~,
monoclonal antibody recognizing ~vB3 (LM 609) did not
inhibit the binding of the cells to either substrate.
The results suggest that ~VBs mediates the attachment of
SK-LMS cells to both vitronectin and Tat and that
distinct regions of the receptor may be utilized for the
binding to each ligand.
EXAMPLE III
Identification of the Tat-bindinq Proteins
To determine the identity of the Tat binding
proteins in L8 cells, immunoprecipitations were performed
with polyclonal antibodies to the cytoplasmic domains of
~v~ B1, B5, and B6. Antibodies against the ~v cytoplasmic
domain precipiTated all three bands eluted from the Tat
peptide column. However none of the reagents against
various integrin B subunits known or suspected to be
associated with the ~v subunit, B1, B3, B5, and B6,
precipitated the complex. Control experiments with
rodent cell lines and tissues demonstrated that the
reagents were capable of reacting with the relevant
integrins. These results indicate that the protein
eluted from the Tat columns is an integrin comprised of
the ~v subunit and possibly one or two previously unknown
B subunits.
The unbound fraction from the Tat column was
rechromatographed on a GRGDSPK-Sepharose column. The
material purified on the GRGDSP-Sepharose column was
immunoprecipitated by antibodies to both the ~v and B3
subunits. These cells synthesize a functional
vitronectin receptor (~vB3)~ that did not bind to the Tat
column.
WO92/1475~ PCT/US92/01227
23 ~1 0
EXAMPLE IV
Cell Adhesion Assays
Cell attachment assays were performed
essentially as described in Ruoslahti et al., Meth.
Enzymol. 82 Pt A:803-831 (1982), incorporated herein by
reference. Microtiter plates (96 wells) were coated with
the full length peptide (Tat-86), a shorter Tat peptide
of residues 45-86 (Tat-41) or a second truncated Tat
peptide of residues 56-86 (Tat-30) as the substrate for
one hour in the presence of 0.25% glutaraldehyde. The
plates were washed and then treated with 1 M ethanolamine
and 2.5 mg/ml bovine serum albumin. L8 rat cells were
detached from their substrate with 0.5 mg/ml trypsin as
described in Brake et al., supra, washed three times with
0.5 mg/ml soybean trypsin inhibitor and resuspended in
DMEM at lo6 cells/ml. 100 ~1 of cell suspension was added
to each well in the presence or absence of inhibitory
antibody. After a one hour incubation, the attached
cells were fixed in 3% paraformaldehyde and stained with
0.5% crystal violet. The dye was eluted from the stained
cells with 100 ~1 of 50% ethanol containing 100 mM sodium
citrate (pH 4.2). Attachment was quantified by reading
the absorbance at 600 nm. The approximate O.D. values
obtained in this study are summarized in Table 1.
Table 1
Pe~tide O.D. (600 nm~
Tat-86 0.92
Tat-41 0.93
Tat-30 0.06
To confirm that the cell adhesion to Tat was
mediated by an integrin, various antibodies were used to
W092/1475~ ~ PCT/US92tO1227
21019i9` 24
inhibit cell adhesion to Tat. The above method was
generally followed with 100 ~1 of either vitronectin or
Tat peptide of residues 45-86 (Tat-41) as the substrate.
In addition, approximately 105 L8 cells per well were
added to each well in the presence of anti-vitronectin
receptor (anti-VNR) or anti-fibronectin receptor (anti-
FNR) antibodies. No antibodies were used in the
controls. The attached cells were fixed and stained with
crystal violet. The dye was eluted and the O.D. was
measured in an ELISA reader. The approximate O.D. values
obtained in this study are summarized in Table 2.
Table 2
Substrate O.D. (600 nm)
Tat-41:
Control 1.03
Anti-FNR 0.88
Anti-VNR 0.14
Vitronectin:
Control 0.94
Anti-FNR O.87
Anti-VNR 0.14
The anti-VNR polyclonal antibodies against the
vitronectin receptor inhibited the adhesion of these
cells to both vitronectin and to the Tat peptide. Anti-
FNR antibodies were not effective in preventing theadhesion of these cells to either substrate. These
results indicate that a receptor related to the
vitronectin receptor is responsible for the adhesion of
the cells to Tat.
The cell attachment assays with a variety of
Tat derived peptides (Figure 1) were used to investigate
the mechanism of cell interaction with the Tat protein.
WO92/147~ 2 ~ O 1 9 1 9 PCT/US92/01227
L8 rat skeletal muscle cells were chosen for these
experiments becaùse they had been shown to bind to Tat in
a previous study described in Brake et al., suDra. In
agreement with this earlier study, L8 cells readily
attached to the Tat protein (Figure 2A). L8 cells also
attached to the Tat 45-86 peptide that contained the RGD
sequence and the basic domain, but not the cysteine rich
domain. An unexpected result was that the cells did not
attach to a shorter peptide in which the basic regions
was deleted (residues 57-86) with residue 57 changed from
arginine to lysine), even though this peptide contained
the cell attachment sequence RGD (Figure 2A). Similar
results were obtained with the human leiomyosarcoma cell
line, SK-LMS (Figure 2B). The cells bound only to those
peptides that contained the basic region. Peptides with
the RGD sequence changed to XGE or deleted entirely could
still support cell attachment, provided the basic region
was retained (Figure 2B). In fact, a peptide containing
only the basic domain and three flanking amino acids
(residues 45-57) supported cell attachment as well as
full-length Tat, on a molar basis. These results,
together with the result that the attachment of cells to
Tat was inhibited by heparin, suggest that the basic
region of Tat, in addition to its function in
transactivation, is required for cell adhesion, and that
the RGD containing region of ~at by itself is incapable
of supporting cell adhesion.
EXAMPLE V
Antibodies Aaainst Various ~ and ~ Subunits
Polyclonal antibodies against the cytoplasmic
tails f ~v~ ~3~ and B5 were raised in rabbits by
i~munization with synthetic peptides coupled to keyhole
limpet hemocyanin. The peptides used were
KRVRPPQEEQEREQLQPHENGEGNSET from the C-terminus of ~v as
described in Suzuki et al., Proc. Natl. Acad. Sci. USA
WO92/1475~ PCT/US92/01227
~ 9 ~9 26
83:8614-8618 (1986), XFEEERARAKWDTANNPLYKEATSTFTNITYRGT
from the C-terminus of B3 as described in Fitzgerald, J.
Biol. Chem. 262:3936-3939 tl987), and
KXPISTHTVDFTFNKSYNGTVD from the C-terminus of Bs as
described in Suzùki et al., Proc. Natl. Acad. Sci. USA
87:5354-5358 (1990), all incorporated herein by
reference. All peptides were synthesized with an Applied
Biosystems Model 430A (Foster City, CA). Additional
details on the preparation of some of these antibodies
are described in Freed et al., EMB0_~. 8:2955-2965
(1989). The anti-B1 subunit antiserum has also been
described in Giancotti and Ruoslahti, Çell 60:849-859
(1990). Each of the antibodies was shown to bind to the
appropriate integrin subunit in immunoblotting and to
immunoprecipitate integrins containing these subunits
from various surface labeled cell lines.
Po}yclonal antibodies prepared against the ~vB3
and ~5B1 integrins have been described in Argraves et al.,
Cell ~iol. 105:1183-1190 (1987) and Suzuki et al.,
Proc. Natl. Acad. Sci. USA 83:8614-8618 (1986).
Monoclonal antibodies (LM 609) were a gift of Dr. David
Cheresh and are described in Cheresh and Spiro, J. Biol.
Chem. 262:17703-17711 (1987) or Bristol Myers-Squibb
Pharmaceutical Research Institute (P3G2) which is
described in Wayner et al., J. Cell Biol. 113:919-929
( 1991)
To determine whether adhesion to Tat was
mediated by an integrin, various antibodies were used to
inhibit cell adhesion to Tat. Polyclonal antibodies
against the vitronectin receptor (~vB3 integrin) inhibited
the adhesion of the L8 and SK-LMS cells to both
vitronectin and to the Tat 45-86 peptide (Figure 3).
Anti-fibronectin receptor (hsB1 integrin) antibodies did
not prevent the adhesion of these cells to either
substrate. Control experi~ents showed that the anti-~vB3
WO92/1475~ PCT/US92/01227
210~919' .,
antiserum did not inhibit the attachment of cells to
fibronectin, whereas the anti-~5Bl antiserum did (Figure
3). A receptor related to the ~VB3 integrin, therefore,
appeared to be responsible for the adhesion of the cells
to Tat.
EXAMPLE VI
Immunoprecipitation Assays
Immunoprecipitations were performed by
incubating material in the presence of 5 ~l of immunized
rabbit serum and 50 ~l protein A-Sepharose for 1 hour.
The receptor-antibody-protein A complex was spun down and
washed three times with 0.5% Triton X-100, 150 ~M NaCl,
50 ~M Tris, pH 7.4. The complex was then boiled in
electrophoresis sample buffer and loaded on ?.5% SDS-
polyacrylamide gels.
EXAMPLE_yII
~ffini~y_Chromatoaraphy
The Tat-binding proteins were isolated from
surface-iodinated cells essentially as described in
Pytela et al., PNAS ~USA~ 82:5766-5770 (19~5) and Pytela
et al., Cell 40:191-198 (1985), both incorporated herein
by reference. Cells were detached from culture plates in
100 ~g/ml trypsin (Sigma) as in the cell adhesion assays,
and washed three times in 500 ~g/ml soybean trypsin
inhibitor (Sigma). Cells were surface iodinated and
extracted with a buffer containing 150 ~m octyl
glucoside, 1 mM CaC12, 1 mM MgC12, 1 ~g/ml aprotinin, 1
~g/ml leupeptin, 0.4 ~g/ml pepsTatin, 150 mM NaCl and 50
mM Tris, pH 7.4. The extracts were clarified at 15,000 x
g znd passed over a column containing various peptides
coupled to cyanogen bromide activated Sepharose 4B
(Pharmacia). After an incubation of 2 hours, the column
was washed with several volumes of extraction buffer
WO92/1475~ 21 ~19 19 PCT/US92/0122-
~ 28
containing 50 mM octyl glucoside. The bound receptor was
then eluted with the appropriate peptide at a
concentration of 1 mg/ml in the buffer used to wash the
column. ~liquots were boiled in electrophoresis sample
buffer and run on 7.5% SDS-polyacrylamide gels or used
~or immunoprecipitation.
EXAMPLE VIII
Production of Polyclonal Antibodies
Polyclonal antibodies can be prepared by any
method known in the art using as the immunogen an
appropriate protein or a synthetic peptide derived
therefrom. In addition, the integrin of the present
invention, the Tat protein or equivalent compounds can be
used as the immunogen. Such methods as described, for
example, in Argraves et al., J. Cell Biol. 105:1163-1173
can be used. For example, a synthetic peptide containing
at least one lysine residue is coupled to Keyho~e Limpet
Hemocyanin (KLH) as follows: by stirring 5.6 ml KLH (10
mg/ml in phosphate buffered saline) with 0.5 ml m-
maleimidobenzoyl-N-hydroxysuccinimide ester (25 mg/ml in
dimethyl formamide) for 30 minutes at room temperature.
The mixture is filtered and 25 mg of the peptide is added
followed by three hours of stirring at room temperature.
After dialysis against phosphate buffered saline, the
mixture containing the coupled peptide is emulsified in
Freund's complete adjuvant, mixed and injected into a New
Zealand White female rabbit. After one month, the animal
is reinjected with the mixture emulsified in incomplete
adjuvant. The animal is bled approximately two weeks
after the second injection. The blood is allowed to clot
and the resulting serum is used as a source of polyclonal
antibodies.
WO92/1475~ PCT/US92/012'
EXAMPLE ~ 1919
Inhibition of Transactivation Assay
Cells were seeded in 10 cm dishes (lo6
cells/dish). The following day, cells were transfected
with 5 ~g LTR-CAT and/or 5 ~g LTR-Tat plasmids by DEAE
dextran precipitation as described in David et al., Basic
Methods in Mo~ecular Bioloay, p. 388 (Elsevier 1986),
incorporated herein by reference. Two days
posttransfection, cells were incubated with 80 ~g or 8 ~1
rabbit anti-human vitronectin receptor (anti-~vB3), anti-
fibronectin receptor (anti-~sB~) or normal rabbit serum in
3 ml of complete DMEM for 30 minutes before the addition
of 0.5 ~1 of E. coli extract containing glutathione
transferase-Tat fusion protein. Basic peptide (300 ~g)
of Tat protein was also included for the inhibition assay
of Tat transactivation. Sixteen hours later, cells were
haFvested and disrupted by 3 cycles of freezing and
thawing to harvest the cell lysate, which was then used
for CAT assay as described in Gorman et al., Mol. Cell.
Biol. 2:1044-1051 (1982), incorporated herein by
reference.
To determine whether ~VBs serves as the receptor
for Tat internalization, the effects of basic Tat
peptides and antibodies to ~V~3 on transactivation by
extracellular Tat were studied. Antibodies to ~5B~ were
used as a control. As shown in Figure 9 (lanes 1-3), a
recombinant GST-Tat fusion protein added to L8 cells was
able to efficiently and specifically transactivate an LTR
reporter gene transfected into these same cells.
However, CAT activity was not diminished in the presence
of anti-~vB3 antibodies at concentrations sufficient to
inhibit cell adhesion (lane 5) nor was it inhibited in
the presence of 300 ~g of basic pep*ide (lane 4). These
results suggest that the entry of functional Tat protein
into L8 cells is not dependent on its interaction with
W092/l4755 2 10 1 9~1 9 PCT/US92/01227
the avB5 integrin, consistent with an earlier suggestion
that Tat inte~nalization was not receptor-mediated as
described in Mann and Frankel, EMBo J. 10:1733-1739
(1991). .
Although the invention has been described with
reference to the presently preferred embodiment, it
should be understood that various modifications can be
made without departing from the spirit of the invention.
Accordingly, the invention is limited only by the
following claims.
