Note: Descriptions are shown in the official language in which they were submitted.
WO 95/07929 PCT/GB94/01992
-I- ~'1 ~ ~ .~ 31
MULTIPLE BRANCH PEPTIDE CONSTRUCTIONS
FOR USE AGAINST HIV
Background of the Invention
Human Immunodeficiency Virus (HIV) is presumably the etiologic agent of
Acquired Immunodeficiency Syndrome (AIDS). HIV establishes a persistent
infection in different cell types; many of the cells express an antigen, the
CD4
receptor, as a binding site at their surface. In such cells, the first step of
cell
infection is represented by viral attachment with binding of the external
envelope
glycoprotein of HIV to the CD4 surface antigen of the cell. This binding is
followed by a series of events involving virus envelope and target cell
membrane
fusion and internalization, through which the cell is infected. HIV infected
cells
may then infect other cells, forming syncytia (giant polynuclear cells).
Syncytia
formation between HIV-1 infected cells and uninfected CD4+ cells (cells having
the
CD4 receptor) involves an interaction between the CD4 receptor and the HIV
surface envelope glycoprotein. This process is blocked by soluble CD4, anti-
CD4
and anti-V3 antibodies.
Cell fusion processes responsible for cell-to-cell spread of the virus in vivo
make plasma neutralizing antibodies obsolete and lead to virus escape from the
immune system already damaged by lymphocyte depletion. B lymphocytes activated
by HIV produce different antibody populations. Some antibodies do not
interfere
with gp 120-CD4 interaction, but block membrane fusion, the process
responsible for
cell infection. These antibodies are principally directed against the envelope
glycoprotein V3 loop.
Due to the very high variability of V3 loop in different HIV-1 isolates,
anti-V3 antibodies generally only neutralize the isolate against which the
antibodies
were produced. Therefore neutralization is limited to that isolate and is
called
isolate specific. These antibodies are effective by cell fusion inhibition,
without any
activity on cell-virus binding.
Several attempts have been made to inhibit HIV infection by V3
loop-related peptides raising questions as to the efficacy of such peptides.
For
example, De Rossi et al. [Virology 184, 187-196 (1991)] found that some V3
peptides enhanced viral infectivity. Other synthetic peptides from the V3
loop,
cyclic or non cyclic, were ineffective in the critical step of cell fusion.
Thus,
antagonist peptides have not been developed which are capable of blocking
virus-cell
fusion and cell-to-cell fusion independently of the virus isolate.
More recently, several studies have demonstrated a CD4 independent route
WD 95/07929 PCT/GB94/01992
2171~~
of cell infection for both HIV-1 and HIV-2, suggesting the existence of at
least one
alternative viral receptor. One of these putative non-CD4 HIV receptors has
been
recently identified on CD4- brain derived cells and colon epithelial cells.
This
receptor is a neutral glycolipid, called galactosyl ceramide (GalCer). HIV
infection
of CD4-/GalCer~ cells in the brain and in the intestine may account for some
of the
HIV-associated disorders in these organs. Moreover, the presence of GalCer on
the
apical side of some mucosal epithelial cells may facilitate the entry of the
virus
during sexual intercourse. No peptide derived from the V3 loop has yet shown
ability to block the GalCer receptor.
In response to some of these problems, radially branched systems using
lysine skeletons in polymers have been used by J.P. Tam [Proc. Natl. Acad.
Sci.
USA, 85, 5409-5413 (1988)] to develop antigens without the use of carriers.
Those
antigens were designed to generate vaccines against a variety of diseases.
Specifically, antigens for generating vaccines against HIV infection are
described by
Tam in PCT application ser. W093/03766 and essentially include the sequence
IGPGR (IUPAC convention single letter nomenclature for amino acids) and are of
eleven amino acids in length to be effective in eliciting useful immune
responses.
Id. at page 13, lines 29-31. Such antigens are not, however, considered as
potential
direct therapeutic approaches to any disease, but are intended to provoke an
immunogenic response in the body.
Summary of the Invention
The present invention includes a method and peptide construction for the
therapeutic treatment of patients with HIV infections. In accordance with this
method, the multiple branch peptide construction for therapeutic
administration
comprises a core matrix attached to from 2 to 64 peptides, each comprising the
amino acid sequence GPGR preceded by from 0 to 4 amino acid residues and
succeeded by from 2 to 4 amino acid residues.
The Drawings
Figure 1 is a proposed binding mechanism for HIV and CD4+ or CD4-
cells.
Figure 2 shows the results of testing the effect of Multiple Branch Peptide
Constructions (MBPCs) on the infection of human peripheral blood lymphocytes
(PBLs) for three different strains of HIV, with the amount of p24 (pg/ml) or
Reverse Transcriptase (RT) activity plotted against three different MBPCs, 1-
none;
?- [GPGRAF]8-Multiple Lysine Core (MLC); 3- [RKSIHIGPGRAFYT].~- MLC; 4-
WO 95/U7929 2 ~ 7 ~ ~ 31 PCT/GB94/01992
-3
[PPYVEPTTTQC]4-MLC.
Figure 3 are the results of testing the MBPCs on human macrophage
infection with the MBPCs plotted against the amount of p24 (ng/ml). Five
different
MBPCs are used: 1- none; 2- [GPGRAF]8-MLC; 3- [GPGRAF]8-MLCD; 4-
[GPG(R)DAF]8 - MLC; 5- [RKSIHIGPGRAFYT]4-MLC; and 6-
[PPYVEPTTTQC]4-MLC.
Figure 4A depicts syncytia formation in infected cultures without any
MBPCs.
Figure 4B depicts the absence of syncytia formation in macrophages treated
with [GPGRAF]8-MLC.
Figure SA illustrates the effect of MBPCs on human epithelial cells
infection through the GaICer receptor.
Figure SB illustrates the ability of MBPCs to block the binding of gp120 to
the GaICer receptor.
Figures 6A, 6B, 7A, 7B and 7C all illustrate the relative non-reactivity of
rabbit sera immunized with a short MBPC, as compared to rabbit sera immunized
with a longer MBPC.
Detailed Description of the Invention
I Structure
The present invention relates to Multiple Branch Peptide Constructions
(MBPCs) which are able to inhibit HIV infection and the spread and extension
of
HIV infection. The MBPCs have a core matrix onto which peptides are covalently
bonded. The core matrix is a dendritic polymer which is branched in nature,
preferably with each of the branches thereof being identical. The core matrix
is
based on a core molecule which has at least two functional groups to which
molecular branches having terminal functional groups are covalently bonded.
Suitable core molecules include ammonia or ethylenediamine. Suitable molecular
branches include acrylic ester monomers which are polymerized onto the core
molecule. Such molecules may be created to present varying number of branches,
depending on the number of monomers branched from the core molecule.
Contemplated for use herein are MBPCs with 2 to 64 branches and preferably
between 4 and 16 branches.
Exemplary for use to form the core matrix is lysine. A central lysine
residue is bonded to two lysine residues, each through its carboxyl group, to
one of
the amino groups of the central lysine residue. This provides a molecule with
four
amino groups, which may be the core matrix for an MBPC having four peptides.
W~ 95107929 PCT/GB94/01992
-4- r
X11
Alternatively, one can provide a molecule with eight branches by bonding four
lysine residues through their carboxyl groups to one of the amino groups of
the
lysine residues which are attached to the central lysine. This molecule can
serve as
the core matrix for an MBPC having eight peptides or can alternatively receive
eight
lysine residues to form a core matrix for an MBPC having sixteen peptides. As
may
readily be seen, larger MBPCs similarly may be constructed as necessitated,
e.g.
having 32 branches.
The C-ends of peptides are covalently bonded to each of the branches of the
core matrix to form the MBPC. The peptides may be the same, which is
preferred,
or may be different from one another. The resulting molecule has a cluster of
peptides at the surface and an interior core matrix which is not presented and
is
therefore not antigenic. For examples of similar structures having peptides
different
from those contemplated herein, see J.P. Tam [Proc. Natl. Acad. Sci. USA, 85,
5409-5413 ( 1988)] and US Patent No. 5229490 to Tam and the references cited
therein.
Spacers may, if desired, be included between the peptides and the core
matrix. The carboxyl group of the first lysine residue may be left free,
amidated, or
coupled to B-alanine or another blocking compound.
Peptides can include D or L-amino acid residues. D amino acids last
longer in vivo because they are harder for peptidase to cut, but the L amino
acids
have better activity, as discussed below.
Moreover, peptide analogues, synthetic constructs using the carbon skeleton
of peptides but omitting the -CONH- peptide bonds, can be employed in place of
peptides. Thus, it should be understood that references to peptides herein may
also
be taken to include peptide analogues. It is believed that peptide analogues
will be
more resistant to peptidase and last longer in vivo.
---~ The MBPCs of the present invention include peptides derived from the V3
loop of the surface envelope glycoprotein gp120 of the HIV-1 virus, and
include the
consensus amino acid sequence GPGR (IUPAC single letter amino acid
nomenclature, i.e. Gly-Pro-Gly-Arg) therefrom. Said MBPCs are effective in
inhibiting HIV infections in human cell cultures of different cell types. The
inhibition appears to be independent from the viral strain, and even to be
effective
against HIV-2 strains. They, thus, have wholly surprising therapeutic
properties.
It should be noted that the corresponding monomer peptides do not have
any therapeutic effect. The difference is believed to occur because the core
matrix
causes the peptide to undergo a conformational change. Specifically and
unexpectedly, the two amino acids following the GPGR in the peptide have been
WO 95/07929 2171 ~~ 1 PCT/GB94/01992
-5-
found to bend when the GPGR monomer is attached to the core matrix. Thus, it
is
preferred that there are at least two amino acid residues succeeding GPGR in
the
peptides attached to the core matrix. However, because if the peptide is too
long,
the MBPC will become antigenic, it is undesirable to have more than four amino
acids following the GPGR sequence and more than four amino acids preceding
said
sequence. Hence, the peptide will have twelve or less amino acids.
Specific MBPCs showing activity on inhibition of infection by HIV-1 and
HIV-2 retroviruses include the following which are set forth using IUPAC
notation
for amino acids:
Table 1 - Exemplar My BPCs
MBPC. l: (GPGRAFY)8-(K)4-(K)~-K-(3A-OH or
(GPGRAFY)8-(K)4-(K)2-K-f3A-NHS
MBPC.3: (GPGRAF)g-(K)4-(K)2-K-13A-OH or
(GPGRAF)8-(K)4-(K)2-K-(3A-NH2
MBPC.I2: (GPGRAF)16-(K)8-(K)4-(K)2-K-BA-OH or
(GPGRAF) 16-(K)g-(K)4-(K)~-K-1iA-NHS
The OH terminal shown above on the ~-alanine indicates the carboxyl group
thereof,
with the amino group being attached to the carboxyl group of the lysine
residue.
The NHS terminal indicates a modification of the carboxyl group of the ~-
alanine, to
form a carboxamide terminal; the ~-alanine is still attached by its amino
group to
the carboxyl group of the lysine residue. In the examples reported
hereinbelow, the
OH forms were used.
Although the peptides should contain the consensus sequence GPGR to
possess the desired therapeutic effect, the inclusion of the conformational
sequence
of an isoleucine before the GPGR, such as IGPGR or IXXGPGR (where X is any
amino acid residue), also from the V3 loop, has been found to render the MBPC
poorly effective or ineffective as a therapeutical treatment when the peptides
containing the IGPGR sequence are of twelve or less amino acid residues and
thus
are not preferred for use herein. The sequence of an isoleucine (I) being
present
within three amino acids prior to GPGR is a relatively conserved one, being
found
in about 98 % of HIV genomes, but, surprisingly, it has been discovered to be
undesirable.
Despite MBPCs containing the IGPGR or IXXGPGR sequence not being
preferred, MBPCs with that peptide sequence have been found to block syncytia
WO 95107929 PCT/G~94/01992
1
formation if the replicated peptide is longer than twelve amino acid residues.
For
example, one such larger MBPC which does block syncytia formation is that in
which the peptide bonded to each branch of a four-branched MBPC is
RKSIHIGPGRAFYT. However, such large MBPCs are undesirable because such
large molecules have been shown to elicit antibodies, even at low
concentrations
(10~M), and are recognized and inhibited by HIV infected patients' antibodies,
both
of which would preclude them from being used in HIV-infected patients over a
long
term. These MBPCs also show toxicity on cell cultures at concentrations near
their
effective concentrations. Moreover, the size of these MBPCs makes them very
difficult and expensive to fabricate, and more sensitive to protease
degradation.
The invention therefore provides an MBPC comprising a core matrix to
which are bonded from 2 to 64, and preferably from 4 to 32 peptides, each of
which
comprises the sequence GPGR preceded by from 0 to 4 amino acid residues and
succeeded by from 2 to 4 amino acid residues. Preferably, the peptides do not
contain a sequence of I within three amino acids before G, such as IGPGR or
IXXGPGR, where X is an amino acid residue. Preferably, the peptides bonded to
an 8 or 16-branched core matrix are GPGRAF. However, it is believed
substitutions are possible in the AF where the substituted amino acids have
the same
characteristics, and may be desirable because the AF appears to be susceptible
to
peptidase and therefore, may be fragile in vivo.
II Preparation of MBPCs
The manufacture of the MBPC structure, that of a branched core with
peptides attached thereto, though previously called multiple antigenic
peptides
(MAPs), has been known in the art. See e.g. Tam et al, J. Immun. 148, 914-920
( 1992) and Wang et al, Science, 254, 285-288 ( 1991). Preferably, for small
quantities (under one kilogram), a solid phase method is used to obtain the
MBPCs.
Stepwise assembling of the peptide chains can be carried out automatically on
4-(oxymethyl)-phenylacetamidomethyl copoly(styrene-1 % divinyl benzene). The
Boc/benzyl strategy may be used, including a systematic double coupling scheme
with hydroxybenzotriazole active esters (Boc-amino-acid-OBt). The final
cleaving
from resin is effected with anhydrous hydrogen fluoride (1 hour at
0°C). The
MBPC is then washed with diethyl ether and solubilized in water. After
lyophilization, the MBPC may be pre-purified on a P2 or G15 type molecular
filtration column, equilibrated with O.1N acetic acid. The eluate fraction may
then
be recovered. The purification step is achieved by using C8 or C18 reversed-
phase
HPLC. The MBPC may be characterized by its amino acid content after acid
R'O 95/07929 217 i 5 31 PCT/GB94/01992
_7_
hydrolysis (6N HCI, 115°C, 24 hours) and electrospray mass
spectrometry.
III Therapeutic Effect and Use
The invention provides a method for the treatment of HIV infections in
which there is administered to a patient MBPCs comprising a core matrix to
which
are bonded from 2 to 64, and preferably from 4 to 32, peptides, each of which
comprises the sequence GPGR preceded by from 0 to 4 amino acid residues and
succeeded by from 2 to 4 amino acid residues, but preferably not the sequence
IGPGR or IXXGPGR, where X is an amino acid residue. The MBPCs should be
administered at serum levels between 10'3M and 10'~M, but preferably at about
10'6M.
The MBPCs made according to the present invention inhibit both HIV
infection and HIV-induced cytopathic effects in vitro, and may therefore
inhibit
virus multiplication and the spread of the virus in the host organism.
The MBPCs made according to the invention neutralize the virus
envelope-cell membrane fusion step, and also the infected cell membrane-
uninfected
cell membrane fusion step essential for syncytia formation, either step being
thought
to be indispensable for cell infection, virus multiplication and the spread of
virus in
the host organism. The MBPCs are able to blockade the CD4 receptor present in
cells such as lymphocytes and macrophages, including the 89.6 strain of HIV,
apparently by attaching to the CDR3 domain of the CD4 receptor. Thus, the
MBPCs of the present invention block the formation of syncytia induced by HIV-
1
and HIV-2 and inhibit infection of human peripheral blood lymphocytes (PBLs)
and
macrophages. Such blocking does not cause the cell to lose its ability to be
activated
by other antigens or mitogens, that is, the functionality of the lymphocyte is
preserved by the MBPCs.
Given the retention of the functionality of the lymphocytes, the potential
problem of artificial AIDS is avoided with the present invention. Therefore,
that the
MBPCs of the present invention attached to the fusion receptor, rather than
the
receptor for activating the T-cells, is a major advance over any previous
therapy
against HIV.
In addition to preventing HIV infection, the MBPCs of the present
invention have been shown to suppress the production of HIV in cells which
have
been infected prior to treatment.
One totally unexpected property of the MBPCs of the invention, and
especially the preferred 8 x GPGRAF and 16 x GPGRAF MBPCs, is that they have
shown an ability to bind to the GalCer receptor which is a receptor for HIV.
This
WO 95107929 PCT/GB94/01992
2 ~ ~ ~ r~ ~ -g- r1
receptor has been shown to exist in colon epithelial cells and central nervous
system
CD4- cells. This binding results in MBPCs inhibiting the infection of human
intestinal cells by distantly related isolates of both HIV-1 and HIV-2.
Another advantage of the MBPCs of the present invention is that they have
been found to be non-toxic in rodents, rabbits and rrionkeys at high doses and
thus,
when used therapeutically will not harm the patient, in contrast to many
current
AIDS therapies.
Another advantage of the MBPCs of the present invention is that in
concentrations of 10-4M or less they are not immunogenic, and thus, unlike all
prior
art attempts of using peptides derived from the V3 loop or constructions
thereof, are
not vaccines. The lack of immunogenicity of the MBPCs is an advantage because
an immune reaction would result in antibody production which would inhibit or
destroy the MBPCs. Such immunogenic MBPCs, such as the previously mentioned
RKSIHIGPGRAFYT, could only be used a few times in the same individual after
which it would become ineffective.
A Blockage of the CD4 ReceQtor
Experimental data demonstrates a specific inhibitory effect of the MBPCs
of the present invention, especially MBPC.3 and MBPC.12, on the process of HIV
infection of CD4+ cells, and a total blockage of HIV-1 and HIV-2 induced
syncytia
formation. MBPC concentrations with which inhibiting effects were observed in
vitro are around 10-6M for MBPC.3 and MBPC.12, and at higher concentrations
for
other MBPCs. The neutralization has been observed on all the HIV strains
tested,
i.e. MN, Lai, NDK, and 89.6 for HIV-1 and Rod for HIV-2. In contrast, the
individual peptide fragments used in the MBPCs are inactive on syncytia
formation
inhibition, even at very high concentrations, e.g. SxIO~M.
The MBPCs may thus interact with a cellular molecule, involved in a
post-binding event, i.e. fusion stage, needed for cellular infection, and used
by
distantly related HIV-1 and HIV-2 isolates, presenting divergent V3 loops. A
good
candidate for this postulated V3 binding site is the CDR3 domain of the V 1
region
of CD4, because: i) anti-CD4 antibodies specific for this region (e.g. 13B8-2)
block
the fusion process during HIV-1 infection, ii) CDR3-derived synthetic peptides
have some inhibitory effect in fusion assays, and iii) MBPCs recognize a
synthetic
peptide corresponding to the CDR3 domain of CD4.
The MBPCs of the present invention bind to lymphocytes without altering
the binding of the viral envelope (through the CDR2 domain), while preventing
HIV
infecting effects. The MBPCs also bind to the soluble CD4. Therefore, MBPC
WO 95107929 21715 ~ i PCT/GB94/01992
-9
binding is inhibited in the presence of soluble CD4.
The effect of MBPCs of the present invention on the interaction between
region-specific anti-CD4 antibodies and the CD4 receptor in its natural
environment
was studied. In situ binding of the anti-CDR3 monoclonal antibody (MAb) 13B8-2
to CD4 expressed on macrophages was dramatically and specifically decreased in
the
presence of MBPCs of the present invention. Binding of other anti-CD4 MAbs,
including those recognizing the gp120 binding site (Leu3a and OKT4a MAbs), was
identical in the absence and presence of MBPCs. These data demonstrate that
MBPCs bind to the CDR3 region of CD4 and thus act, at least in part, by
blocking
the interaction between the gp 120 V3 loop and this domain of CD4.
Despite the V3 loop of HIV-1 being considered to be an important
determinant in the fusion process between HIV-1 and its target cells, it has
not yet
been possible to exploit this property therapeutically until the present
invention. The
main reason is that V3 is a hypervariable region showing a high degree of
diversity
among HIV-1 isolates. Neutralizing anti-V3 antibodies are generally type-
specific
and, in the best' case, may neutralize only closely related isolates. Because
the
MBPCs of the present invention are targeted to cellular and not viral
determinants,
they by-pass the problems inherent to envelope variability. Thus, the MBPCs
used
in this invention, although derived from the consensus sequence of the HIV-1
V3
loop, are able to neutralize various HIV-1 strains, including the highly
divergent
HIV-1 (NDK), and also the unrelated HIV-2 ROD strain. Surprisingly, the MBPCs
of the present invention have also been found to inhibit infection with
primary
HIV-1 isolates, i.e. directly collected from patients, including strains that
are
resistant to AZT, J-1.
The following examples, which are not intended to limit the foregoing,
illustrate the efficacy, utility and breadth of use of MBPCs of the present
invention
with regard to HIV infection, and their binding and fusion with CD4+
lymphocytes
and macrophages.
Example 1 - S~nc"ytia Formation Blockage
Cells were cultured in RPMI 1640 supplemented with 5 % fetal calf
serum, 1 % glutamine, 1 % streptomycin-penicillin (Gibco of Irvine, Scotland)
in a
humidified atmosphere with 5 % CO~. CEM or C8166 cells were chronically
infected with HIV-1 Lai or HIV-2 Rod, or HIV-1MN, respectively. Infected cells
(1x104) were incubated for 2 hours with various concentrations of MBPCs in
96-well plates. Uninfected Molt-4 cells (4x104) were then added in 100 ~I of
culture
medium. syncytia were counted after 18 hours at 37°C. Results are shown
in the
WO 95107929 ~ ~ ~ ~ ~ PCT/GB94/01992
-10-
following Table.
Table 2 - Syncytia formation
MBPC MBPC Conc. HIV-1MN HIV-lLai HIV-2Rod
MBPC.3 SxlO'6M - - -
10'6M + + -
10'~M +++ ++++ +
MBPC.S SxlO'6M - - -
10'~M ++ + -
10'~M ++++ ++++ +
MBPC.4 SxlO'6M ++++ +++ ++++
10'~M ++++ ++++ ++++
10'~M +++ ++++ ++++
Mono.3 10'SM ++++ ++++ ++++
10-6M ++++ ++++ ++++
10''M ++++ ++++ ++++
MBPC.4 is (IGPGRAF)4-(K)~-K-!3A-OH. Mono.3 is the hexapeptide GPGRAF and
MBPC.S is the tetradecapeptide MBPC (RKSIHIGPGRAFYT)4-(K)~-K-13A-OH.
MBPC.3 is as set forth above.
The presence of syncytia is indicated by + signs, allocated in approximate
proportion to the quantities counted.
MBPC.3 exhibited an in vitro inhibiting effect on syncytia formation
induced by HIV at a concentration of around 10-6M, and MBPC.S exhibited an
inhibiting effect at a slightly higher concentration. MBPC.12 (results not
shown)
gave results similar to those of MBPC.3. By contrast, the monomeric version
Mono.3 is wholly inactive; this has also been confirmed for Mono.3 at the
higher
concentration of 10'~M.
Examt~le 2 - Prevention of Infection of PBLs
50 x .103 8E5 cells chronically infected with a RT-deficient HIV-I(IIIB)
isolate, were co-cultivated with 150 x 103 phytohemagglutinin (PHA) stimulated
human peripheral blood lymphocytes (PBLs) in the presence of the indicated
peptide
in 96-well plates. The MBPCs used are described by the peptides attached, and
the
number thereof, to a core lysine matrix, which is designated MLC. All amino
acids
used for this experiment were in dextro form.
Chemical synthesis of MBPCs was performed by the solid phase technique,
WO 95/07929 J ~ PCT/GB94/01992
-11-
as described above. The peptide chains were elongated stepwise on
4-(oxymethyl)-PAM resin using optimized t-butyloxycarbonyl/benzyl chemistry.
Amino acid analyses of the purified MBPCs agreed with the deduced amino acid
ratios.
[GPGRAF]8-MLC was further characterized by electrospray mass
spectrometry (experimental MT = 5672 Da).
The number of syncytia was determined after 24 hours of incubation in the
continued presence of the peptides and the results are presented in Table 3
below.
+++ indicates that the number of syncytia present in the well was similar to
control untreated wells; - indicates the total absence of syncytia in the
well; ~
indicates the occasional presence of few syncytia in the well. Syncytia
formation in
this test was blocked by anti-CD4 antibodies OKT4A (anti-CDR2) and 13B8-2
(anti-CDR3), consistent with previous reports. The core lysine structures
(designated "MLCs") by themselves did not induce syncytia formation in this
test.
Toxicity was evaluated by either MTT assay (described below) or trypan blue
exclusion technique.
Table 3. Inhibition of HIV-1 induced cell fusion by MBPCs
Concentration (M)
Peptide Sx 10-~ Sx 10'6 Sx 10-5
GPGRAF +++ +++ +++
[GPGRAFJg-MLC ~ _ -
[IGPGRAF)8-MLC +++ _+ -
[GPGRA]8-MLC +++ +++ +++
[GPGR]8-MLC +++ +++ +++
[GPG(R)DAFJ8-MLC +++ - Toxic
[GPGRAFJ8-MLCD +++ +++ +++
[Ac-GPGRAF]8-MLC +++ +++ -
[RKSIHIGPGRAFYT]4-MLC +++ - Toxic
[RKSIHKGPGRAFYT]4-MLC +++ - Toxic
[RKSIHTGPGRAFYT]4-MLC +++ - Toxic
RAFVTIGK +++ +++ -
Under these experimental conditions, [GPGRAF]8-MLC at a concentration
as low as SxIO-~M, induced a marked inhibition of syncytia formation, while
the
corresponding monomeric peptide GPGRAF was not active over the range of
Wo95107929 ~~~ ~ ~ PCT/GB94/01992
-12-
concentration used (up to SxlO-SM). N-terminal acetylation
([Ac-GPGRAF]8-MLC), addition of an isoleucine (I) residue ([IGPGRAF]8-MLC)
and incorporation of D-amino acids in the GPGRAF sequence resulted in a ,
significant loss of activity. MBPCs with less than 6 residues (i.e. [GPGRA]8-
MLC
or [GPGR]8-MLC), as well as MBPCs with a non-relevant sequence, did not
inhibit
cell fusion, demonstrating that the core matrix was not involved in the
biological
activity.
[RKSIHIGPGRAFYT]4-MLC was able to inhibit syncytia formation, at a
concentration of SxlO~M. Because the isoleucine residue preceding the GPGRAF
motif is highly conserved among HIV-1 isolates, two related constructions were
synthesized with a lysine or a threonine residue in place of isoleucine. No
significant differences in the anti-fusion activity of these three
constructions were
observed. However, it should be noted that [RKSIHIGPGRAFYT]4-MLC and its
derivatives induced some toxicity in 8E5 cells when used at a concentration of
SxlO-SM and are therefore undesirable for use in human therapy.
Based on these results, the more potent anti-HIV MBPC appeared to be
[GPGRAF]g-MLC. In separate tests this molecule was also able to block the
fusion
between HIV-1(LAI), HIV-1(NDK) or HIV-2(ROD) chronically infected
T-lymphblastoid CEM cells and HTLV-I-transformed MT-2 cells.
Example 3 - Effect of MBPCs on ~mphocyte FunetionalitX
a) Effect of MBPCs on Antigen and Mitogen-Induced Cell
Proliferation
Peripheral blood lymphocytes (PBL) from 3 healthy HIV sero-negative
donors were isolated from heparinized blood by the Ficoll-Hypaque technique.
The
culture medium was RPMI 1640 supplemented with 1 % glutamine, 1 % antibiotics
and 10% heat inactivated fetal calf serum. Cells (105) were incubated in the
presence of MBPCs (5x10-6M) with or without antigen (candidine, PPD) or
mitogen
(PHA). Antigen or mitogen treated cells were pulsed for eight hours with 1 mCi
of
[3H]thymidine. Then cells were harvested and [3H]thymidine incorporation in
DNA
was counted.
b) Mixed Lymphocyte Response (MLR)
In this assay, peripheral blood lymphocytes from the three healthy donors
( 105) were incubated with 105 cells from a mixture of 10 seronegative donors
in
200 ~cl final volume in the presence or absence of various concentrations of
MBPCs
in microtitre plate wells. Lymphocyte mixtures were pulsed on day 6 for 8
hours
with 1 ~,Cl of [3H]thymidine. Cells were harvested and [3H]thymidine
incorporation
WD 95!07929 ~ ~ PCT/GB94/01992
-13
into DNA was counted in a beta counter. The results are shown in Table 4 below
with MBPC.S as set forth in the First test.
Table 4 - PBL Antigen and Mitogen Induced Proliferation
[3H]thymidine incorporation (cpm)
control MBPC.3 MBPC.S
Donor 3
PHA 1 203000 162000 228000
PHA2 183000 154000 188000
PPD 1 5900 3900 4000
PPD2 3200 3100 2400
Candl 3900 3200 3700
Cand2 1700 1200 1100
Donor 2
Donor 1
PHA 1 115000 87000 115000
PHA2 112000 95000 111000
PPD 1 38000 24000 24000
PPD2 33000 27000 26000
Candl 28000 17000 15000
Candl 13000 11000 9000
PHA1
___ ---
___
_---
PHA2 _ _ _
PPD 1 1200 1300 1300
PPD2 1500 3100 2100
Candl 74000 87000 80000
Cand2 79000 100000 85000
The thymidine incorporation in cells is similar in treated and control plates.
If the
functionality of the PBLs had been negatively effected, the incorporation of
thymidine a marker of cell reproduction, would be considerably increased in
MBPC
treated plates. Thus, these results demonstrate that there is no proliferation
of
PBLs, as is seen when they have lost their ability to respond to antigens or
mitogens.
Example 4 - Effect of MBPCs on the Infection of Human PBLs
WO 95/07929 PC'T/GB94/01992
-14-
Peripheral blood lymphocytes (PBLs) were obtained from healthy donors,
stimulated with phytohemagglutinin, and cultivated in RPMI 1640 containing 10%
fetal calf serum and interleukin-2 (complete medium). Samples of 6 x 106
cells/ml
were either treated or not treated with the indicated MBPC at a concentration
of
10-SM and exposed to HIV-1 or HIV-2 (100 TCIDso) for 1 hour at 37°C in
the
continued presence of MBPCs. After thorough washing, PBLs were cultured in
complete medium with 10-SM of the corresponding MBPC. The state of infection
was assessed by determination of RT activity (for both HIV-1 and HIV-2
isolates)
and HIV-1 p24gag measurements (Coulter kit, cut-off 10 pg/ml) in the case of
HIV-1
isolates 10 days post-infection. The MBPC treatment used were as follows:
1-none; 2-jGPGRAF]8-MLC; 3-[RKSIHIGPGRAFYT]4-MLC;
4-[PPPYVEPTTTQC]4-MLC (a non-HIV related MBPC) (MLC indicates a core
lysine structure). The results shown in Figure 2 are representative of three
separate
experiments. The nature of the viral isolates used was checked by PCR (HIV-1
versus HIV-2 isolates) and RIPA for discriminating between HIV-1(LAI) and
HIV-1 (hIDK).
[GPGRAF]g-MLC and [RKSIHIGPGRAFYT]4-MLC treatment consistently
caused a delay in the production of virus progeny as assessed by the
measurement of
reverse transcriptase (RT) activity in the culture supernatants: at days 10
and 13
after infection, RT activity was more than 90 % lower in MBPC-treated samples
than
control samples. Since the peptides did not interfere with the RT assay, this
is
consistent with a MBPC induced inhibition of viral production. Similar results
were
obtained (for HIV-1 isolates) when viral production was followed by
determination
of p24gag concentration in the culture supernatant. [RKSIHIGPGRAFYT]4-MLC
was less potent than [GPGRAF]8-MLC for inhibition of HIV-2 infection. MBPCs
which did not contain a consensus sequence from the V3 loop did not inhibit
infection of PBLs.
Example 5 - MBPC Inhibition of Svncvtia Formation in Macronha~es
MBPCs were also evaluated for their ability to block the infection of human
primary macrophages by a highly cytopathic macrophage-tropic isolate,
HIV-1(89.6). Mononuclear cells were isolated from leukophoresis units enriched
for monocytes by Ficoll-Hypaque densixy separation. Macrophages were purified
by
adherence to plastic in RPMI 1640 supplemented with 10% fetal calf serum and
5%
human AB serum. The adherent cells were cultured for 5 days in the presence of
GM-CSF (1 ng/ml) and were positive for the human macrophage marker
(Boehringer Mannheim, clone 25F9). 5 x 105 cells were treated for 45 minutes
with
WO 95/07929
FCTIGB94/01992
-15-
the indicated MBPC at concentrations of Sx 10-SM and subsequently exposed to
10,000 TCIDso of the macrophage-tropic isolate HIV-1(89.6). The MBPCs used
were as follows: 1-none; 2-[GPGRAF]8-MLCD; 4-[GPG(R)DAF]g-MLC;
5-[RKSIHIGPGRAFYT]4-MLC; 6-[PPPYVEPTTTQC]4-MLC, with amino acids
being in L form unless noted as being in D form. After thorough washing, the
cells
were fed again with medium and analyzed for HIV-1 p24gag production and
syncytia
formation 4 days post-infection. The results are depicted in FIG.3 with the
MBPCs
plotted against the amount of p24 (ng/ml). As also indicated in FIG.3 the
relative
number of syncytia per well (+++, no inhibition; ++, 50% inhibition;
> 95 % inhibition) correlated with the concentration of p24gag.
Macrophages depicted in FIG.4A are untreated controls and show the
presence of numerous syncytia in infected cultures. Macrophages as depicted in
FIG.4B at 100x, which were pretreated with [GPGRAF]8-MLC, displayed very few
giant cells. [RKSIHIGPGRAFYT]4-MLC and [GPGRAF]8-MLC were able to
inhibit infection, as assessed by i) the large decrease of visible cytopathic
effects in
macrophage cultures exposed to 10,000 TCIDSO of HIV-1(89.6) in the presence of
MBPCs and, ii) the measurement of HIV-1 p24gag concentration and RT activity
in
the culture supernatants. Control MBPCs did not significantly affect the
infection
rate of macrophages by HIV-1(89.6). Interestingly, [GPGRAF]8-MLCD (i.e.
[GPGRAF]8-MLC with D-amino acid residues) and [GPG(R)DAF]8-MLC were
significantly less active than [GPGRAF]g-MLC on macrophage infection.
B GaICer Receptor
The MBPCs of the present invention, especially MBPC.3 and MBPC.12
and their derivatives, also bind to the second receptor, Galoctosyl Ceramide
receptor, (see Harouse et al, Science, 253, 320-323 (1991)), and completely
block
the infectivity of HIV on human colon cells. The corresponding monomeric
peptides are inactive on this receptor.
To obtain a total blockage of the GaICer pathway of infection in intestinal
cells using the GPGRAF peptide in the MBPC, the degree of polymerization of
the
MBPC must be at least 8. This suggests that the conformational requirement of
the
V3 loop for GaICer recognition is more stringent than for binding to the CDR3
domain of CD4. CDR3 is an acidic region (comprising a glutamic acid residue at
position 87 which has been reported to be essential for the fusion process),
which
probably interacts with basic amino acids of the V3 loop by electrostatic
interactions. In contrast, the galactose head group of GalCer, which is known
to be
involved in gp 120 binding, is neutral polar and may require a different
spatial
WO 95/07929 ~ ~ ~ PCT/GB94J01992
-16
arrangement of atoms in the V3 structure for optimal interaction. Thus, it is
surprising that a therapy for blocking CD4 route HIV infection would also
function
with the GalCer receptor.
The activity of MBPCs was evaluated on this alternative infection route as
follows: HT-29 cells (5x105 cells) were incubated with MBPCs (5x10-6M) for 45
minutes and subsequently exposed to 100 TCIDSO of HIV-1(LAI), HIV-1(NDK) or
HIV-2(ROD) for 1 hour at 37°C. The residual inoculum was inactivated
by three
successive trypsinations and infection was assessed by:
i) direct measurement of HIV-1 p24g$g (for HIV-1 isolates) and RT activity
(for HIV-1 and HIV-2 isolates) in the culture supernatants;
ii) co-cultivation with human PBLs. The cells were treated with the MBPCs
as follows: 1- none; 2- [GPGRAF]8-MLC; 3- [GPGRAF]8-MLCD; 4-
[GPG(R)DAF]$-MLC; 5- [GPGRAF]16-MLC; 6-[RKSIHIGPGRAFYT]4-MLC;
and 7- [PPPYVEPTTTQC]4-MLC, with all amino acids in L form unless noted
otherwise. The results shown in FIG.SA, which correspond to infection with
HIV-1(NDK), are representative of three separate experiments and the MBPC used
is indicated by the number on the x-axis. The concentration of p24gag (pg/ml)
was
measured during the co-culture experiment with PBLs (+++, > 10 ng/ml 10 days
post-infection; ++, > 1 ng/ml 13 days post-infection; +, > 1 ng/ml 16 days
post-infection; -, no detectable p24gag 1 month post-infection). Similar
results were
obtained with HIV-1(LAI) and HIV-2(ROD).
In another experiment with the same MBPCs, the effect of MBPCs on the
binding of gp120 to GalCer was analyzed using a high performance thin layer
chromatography (HPTLC) binding assay. MBPCs (10~M) were preincubated with
GalCer for 1 hour at room temperature. After thorough washing, the HPTLC
plates
were incubated with HIV-I recombinant gp120 (2.5 ~cg/ml) followed by rabbit
anti-gp 120 and lzsl-goat anti-rabbit IpG. The plates were washed in PBS and
exposed to Amersham hyperfilm MP for 8 hours. As shown in FIG.SB treatment of
HT-29 cells with active MBPCs protected the cells from infection by HIV-1 (NDK
isolate). This antiviral effect was correlated with its ability to block the
binding of
gp 120 to GalCer. Neither monomeric V3 peptides nor control MBPCs displayed
any antiviral activity and did not inhibit gp120 binding to GalCer.
The invention therefore additionally provides an MBPC comprising a core
matrix to which are bonded from 2 to 64 (preferably from 4 to 32) peptides,
each
derived from the V3 loop of the surface envelope glycoprotein gp120 of the HIV-
1
virus, the MBPC being capable of inhibiting the infectivity of the HIV
viruses, in
both CD4+/GalCe~ and CD4-/GalCer+ cells.
WO 95/07929 ~ ~ PCT/GB94/01992
_ 17_
C. Anti~enicity. Toxicity and Immun~enicitx
MBPCs of the present invention are not toxic to human PBLs in vitro or in
vivo to mice repeatedly injected either intraperitoneally, intravenously or
subcutaneously with MBPCs at a concentration of 10-3M. Mice receiving repeated
doses of 1 mg of MBPCs, intraperitoneally and intravenously, have not shown
any
adverse effects. Additionally, monkeys (macaca sylvana) injected intravenously
once, then subcutaneously, every day with doses of 5 mg/kg have not shown any
sign of toxic effects after thirty days. Additionally, rabbits and mice
injected with
[GPGRAFJ8-MLC did not produce significant titers of anti-GPGRAF antibodies, in
agreement with the concept that MBPCs with less than 10 amino acid residues in
each branch of the peptide are not immunogenic at concentration of less than
10-3M,
the intended concentrations of use. Moreover, in vitro, the PBLs retain the
ability
to be activated by other antigens or mitogen.
1. Anti enicitv
E_nzvme Linked Immunosorbent Assay (ELISA) of MBPC Immuno enicitv
MBPC.1, MBPC.2 and monomeric V3 consensus peptide were tested in
ELISA for their immunoreactivities against HIV-1+ sera or HIV-1- sera. For
this
test MBPC.2 is (RKSIHIGPGRAFYT)4-(K)~-K-13A-OH and MBPC.1 is
(GPGRAF)8-(K)4-(K)2-K-13A-OH. The positivity of sera was first confirmed by
Western Blot analysis using NEW LAV-BLOT kit (Diagnostic Pasteur) for the
detection of anti-HIV-1 antibodies in serum/plasma. Ninety six well microtiter
plates were coated with 500 ng/well of peptides. After saturation and washing,
50
~cl of HIV-1+ serum (1/100 dilution) were added for 2 hr at 37°C.
Staining was
performed with peroxidase-coupled swine anti-rabbit IpG. 30 HIV-1 positive
sera
and 3 HIV-1 negative sera were comparatively analyzed in ELISA assay against
V3
peptide derived from the North American/European consensus (V3 Cons) sequence
and V3 MBPCs. The results are set forth in Table 5 below.
Of the 30 HIV-1 positive sera: 26 react strongly with V3 Cons and with
MBPC.2, one reacted weakly with both peptides, one reacted selectively with
MBPC.2, and two sera did not react with any of the peptides.
Interestingly MBPC.I did not react with any of the HIV-1 positive sera.
As compared to MBPC.2, MBPC.1 has been shown to be more effective in blocking
HIV-1 and HIV-2 infection and less toxic. The fact that MBPC.1 is not
recognized
by HIV-1 positive sera may suggest that such MBPC could not interfere with the
neutralizing activity of anti-V3 antibodies. Therefore they could act in
synergy to
neutralize HIV-1 infection.
WQ 95/07929 PCT/GB94101992
-lg-
TABLE 5
ELISA
Serum No. Western Blot V3 500 ng of
HIV-I Consensus peptidc/wall
MBPC2 MBPC1
1 + +++++ +++-E+ -
2 + +++++ +++++ -
3 + +++++ +++++ -
4 + +++++ +++++ -
+ +++++ +++++ -
6 + +++++ +++++
7 + +++++ +++++ - -
8 + +++++ +++++
9 + +++++ +++++ -
+ +++++ +++++ -
I1 + +++++ +++++ -
12 + +++++ +++++ -
13 + +++++ +++++ -
14 + ++ +++ -
+ +++++ +++++ -
16 + +++++ +++++ -
17 + +++++ +++++
IS + +++++ +++++ -
19 + +++++ +++++ -
+ - - -
21 + + + -
22 + - - -
23 + - + -
24 + +++++ +++++ -
+ +++++ +++++ -
26 + +++++ +++++ -
27 + +++++ +++++ -
28 + +++++ +++++ -
29 + +++++ +++++ -
+ +++++ +++++ -
31 - - - -
32 - - - -
33 - - - -
WO 95/07929 ~ 17 ~ 5 31 PCTlGB94/01992
-19-
2. Toxicity
An MTT assay, a colorimetric cell viability assay using
3-[4,5-Dimethylthiazol-2-yl] 2,5-diphenyltetrazolium bromide, was used to
assess
the effect of MBPCs on cell viability. CEM or C8166 cells (5x104) were
cultured
continuously ( 15 days) in the presence of various concentrations of MBPCs in
the
wells of 96-well microtiter plates in 200 ~cl final volume of RPMI 1640 medium
supplemented with 5 % fetal calf serum, 1 % glutamine, 1 % streptomycin-
penicillin
(available from Gibco of Irvine, Scotland) in a humidified atmosphere with 5
CO~. Every 3 to 4 days, an MTT assay was performed on 100 ~.l of cells
suspension and MBPC were added to cell culture and the volume completed to 200
~cl. [GPGRAF]8-MLC, at concentrations of 10'~ to 10'4M, was not toxic to cells
in
this test. Other results are shown in Table 3 above. As indicated the
tetradecapeptide MLC which includes the IGPGR sequence is toxic and therefore
not
appropriate for therapeutic use as contemplated herein.
3. Anti~enicitv - Immuno enicity Studies on MBPCs
Four C57/BL6 black mice were injected daily for 36 days with 250 ~cl of
MBPC.3 at 4 mg/ml (1 mg). The injection route was intraperitoneal. The blood
samples were collected by sub-orbital eye punction. The blood samples were
left for
1 hour at 37°C, then overnight at 4°C, and the supernatants
(serum samples) thus
obtained were separated from the pellets. Sera were tested by ELISA, looking
for
specific anti-MBPC.3 antibody.
Additionally, two New Zealand rabbits were immunized with
[GPGRAF]8-MLC: 400 ~g of [GPGRAF]8-MLC in 1 ml of PBS, pH 7.4, were
mixed with an equal volume of complete Freund's adjuvant and injected
intradermally to each rabbit on day 0. This procedure was repeated
subcutaneously
with incomplete Freud's adjuvant on days 30, 60, 90 and 120. Sera were tested
by
ELISA, looking for specific anti-MBPC.3 antibodies. Data presented here is
from
the last serum samples (120 + 7 days).
96-well microtiter plates (Nunc, Rotskild, Denmark) were coated for 2 h at
37°C with various concentrations of [GPGRAF]g-MLC (500 to 25 ng of
[GPGRAF]8-MLC in 50 ~.I of PBS), pH 7.4, per well. After saturation with 400
~,I
of casein (5 g/100 ml), different serum dilutions were added for 2 hours at
37°C.
The controls were sera from non-immunized rabbits and mice. After rinsing
(Sx),
50 ~cl of 1:5000 of peroxidase-labelled swine anti-rabbit IgG or anti-mouse
IgG
(Dakopatts, Copenhagen, Denmark) were added for 1 hour at 37°C. After
further
rinsing (Sx), 100 ~c1 of ortho-phenylenediamine were added for 30 min in the
dark at
WO 95/07929 PC'1'/GB94/01992
-20-
2~,1.~~~.
room temperature. The reaction was stopped by adding 50 ~cl of 4 N sulfuric
acid,
and optical density (OD) ratios were determined at 492/620 nm.
Tables 6-10 below set forth the results, with Table 6 setting forth a
summary of the results achieved with the rabbits using optical density ratios
(x103).
Mo is the GPGRAF monomer at 10 ~,g/ml.
Table 6
serum dilution
1:10 1:100 1:1000
[GPGRAF]8-MLC (~cg/ml)
rabbit A +++ + -
rabbit B + - -
5 rabbit A +++ - -
rabbit B + - -
1 rabbit A +++ - -
rabbit B + - -
0.5 rabbit A + - -
rabbit B - - -
-, optical density (OD) < 0.2; +, OD=0.2-0.4; ++, OD=0.4-0.8; +++, OD > 0.8
SU~SiiTUTE Si~EET (MULE 26j
X171531
WO 95/07929 PCT/GB94/01992
-21-
Table 7 - microtiter plate n° 1
S10 S10 SS SS S1 S1 S.5 S.S Mo Mo
M 10-2 1162 1199 1045 933 444 419 382 381 1224 1135
M 10-3 339 291 285 236 126 111 128 135 310 287
M 10-4 172 138 121 132 50 51 56 38 104 185
M ctrl 622 550 527 S51 337 313 360 341 516 591
R ctrl 1062 1093 986 976 631 662 621 616 945 989
R 10-2 2000 2000 2000 2000 1280 1267 968 981 2000 2000
R 10-3 2000 2000 2000 2000 346 349 318 296 2000 2000
R 10-4 924 823 796 809 182 193 145 140 677 725
Legend: S 10 = 10 ~,g MBPC.3/ml; S 5 = 5 ~cg MBPC.3/ml
S 1 = 1 ~g MBPC.3/ml; S.5 = 0.5 ~cg MBPC.3/ml
Mo = GRGRAF monomer, 10 ~cg/ml
M 10-2 = [ 10-'-] Mouse; R 10-2 = [ 10-'-] Rabbit
M ctrl = [10-'-] non immun. mouse; R ctrl = [10-1] non immun. rabbit
Table 8 - microtiter plate n° 2
S S SS SS S S S.5 S.5 Mo Mo
10 10 1 1
M 10-2 1540 1457 1514 1344 381 429 315 841 1539 1575
M 10-3 334 291 263 273 54 65 47 82 342 335
M 10-4 132 188 88 97 15 17 13 83 95 126
M ctrl 490 464 385 231 241 289 235 851 482 500
R ctrl 437 429 386 345 214 185 117 153 376 395
R 10-2 2000 2()DO2000 2000 1922 87 34 716 2000 2000
R 10-3 2000 2000 2000 2000 394 87 197 181 2000 1909
R 10-4 734 689 678 626 113 84 80 74 588 625
Legend: S 10 = 10 ~cg MBPC.3/ml; S 5 = S ~,g MBPC.3/ml
S 1 = 1 ~cg MBPC.3/ml; S.5 = 0.5 ~cg MBPC.3/ml
Mo = GRGRAF monomer, 10 ~,g/ml
M 10=2 = [10-=] Mouse; R 10-2 = [10-'-] Rabbit
M ctrl = [10-Z) non immun. mouse; R ctrl = [10-1) non immun. rabbit
SUaST~gU'fE S~frET (RUCL~ 26~
Wo 95/07929 ~~ ~ PCT/GB94/01992
-22-
Table 9 - microtiter plate n° 3
S10 S10 S5 S5 S1 S1 S.5 S.5 Mo Mo ,
M 10-2 639 664 636 630 241 200 204 198
M 10-3 172 168 159 163 50 53 48 49
M 10-4 85 84 83 84 22 22 22 21
M ctrl 513 492 486 459 257 249 286 261
R ctrl 833 238 222 213 92 92 91 88
R 10-2 818 953 857 868 402 401 416 287
R 10-3 342 328 323 314 168 150 148 155
R 10-4 191 184 170 169 72 89 64 69
Legend: S 10 = 10 ~cg MBPC.3/ml; S 5 = 5 ~cg MBPC.3/ml
S 1 = 1 ~.g MBPC.3/ml; S.5 = 0.5 ~cg MBPC.3/ml
Mo = GRGRAF monomer, 10 ~,g/ml ,~
M 10=2 = [10 ] Mouse; R 10-2 = [10--] Rabbit
M ctrl = [10-~] non immun. mouse; R ctrl = [10-1] non immun. rabbit
Table 10 - microtiter plate n ° 4
S10 S10 SS S5 S1 S1 S.5 S.5 Mo Mo
M 10-2 1454 1340 1233 1336 553 439 495 468 1231 1317
M 10-3 334 318 299 297 83 88 90 95 380 386
M 10-4 133 113 124 103 27 21 28 28 123 122
M ctrl 453 381 349 400 215 235 275 278 '468 458
R ctrl 225 203 285 288 95 86 82 86 283 185
R 10-2 849 843 749 792 419 447 424 425 811 811
R 10-3 332 314 310 304 221 151 163 154 245 194
R ABU 263 279 255 251 105 101 134 128 214 218
Legend: S 10 = 10 ~cg MBPC.3/ml; S 5 = 5 ~cg MBPC.3/ml
S 1 = 1 ~cg MBPC.3/ml; S.5 = 0.5 ~cg MBPC.3/ml
Mo = GRGRAF monomer, 10 ~cg/ml
M 10-2 = [10 2] Mouse; R 10-2 = [10-'-] Rabbit
M ctrl = [10- ] non immun. mouse; R ctrl = [10-1] non immun. rabbit
R ABU = [10-2] rabbit, with peptide Abu (non-related amino acid)
SU8STITUTI: SHEET (EtUlE 26,
WO 95/07929 ~ PCT/GB94101992
-23-
In view of the high amount of background reactivity observed in some of the
results, further testing was undertaken to verify the non-reactivity of sera
immunized
with MBPC.1.
Sera from rabbits immunized with either MBPC.1 or MBPC.2 were tested in
ELISA for their ability to bind MBPC.1, MBPC.2, Mono.l (the monomeric form of
MBPC.l) and the North American/European consensus V3 sequence (V3 Cons, or
Cs).
Figure 6A shows that sera from rabbits immunized with MBPC.2 recognized
significantly MBPC.2 and V3 Cons, while no reactivity was observed against
MBPC.l
or the corresponding monomeric peptides. Figure 6B indicates that sera from
rabbits
immunized with MBPC.1 did not react against MBPC. l (except for a weak
reactivity of
the serum from rabbit A), MBPC.2, V3 Cons or Mono.l. Preimmune sera used as
negative control likewise did not react against either MBPC, V3 Cons or Mono.
1.
In another set of experiments, sera from rabbits immunized with either MBPC.1
or MBPC.2 were tested (at 1/100 and 1/1000 dilutions) against various
concentrations of
MBPC.1, MBPC.2 and V3 Cons. Figures 7A, 7B and 7C clearly indicate that sera
from
rabbits immunized with MBPC.2 reacted in a dose dependant manner against
MBPC.2
and V3 Cons. No reactivity against MBPC.1 was detected. In contrast, sera from
rabbits immunized with MBPC.1 did not react against MBPC.2, while reacting
weakly
against MBPC.1 (note the weak reactivity of the serum from rabbit A against
MBPC.l,
with no reactivity of serum from rabbit B). Interestingly, these latter sera
did not react
with the monomeric V3 Cons, a 34-amino acid linear epitope mimicking the
consensus
sequence as present on gp120.
These results indicate that MBPC.1, a short V3 MBPC, did not induce a
significant antibody response in rabbits, while MBPC.2, the longest V3 peptide
construct, did invoke such a response. These observations are in line with the
results on
the immunoreactivity of MBPC.1 with HIV-1 infected patients' sera.
In mice, repeatedly injected at 10-3M, very weak titers of antibodies could be
occasionally detected in sera at 1:100 dilution, and essentially no antibodies
could be
detected at lower serum dilutions. In rabbits, antibodies are detectable in
one animal at
1:100 serum dilution. At all other serum dilutions, the antibodies titers are
comparable
to those obtained in control animals. Such weak antigenicity demonstrates that
the
MBPCs of the present invention do not have the potential to be used as an
antigen able to
elicit an effective antibody response, which is the use of previously known
peptide
constructions. Indeed, the intended therapeutic concentrations of the MBPCs of
the
present invention are around 10~M, and no antibody response is expected at
such
concentrations. However, given the lack of antigenicity the MBPCs may be
administered
at up to 10-3M.
suesi~~u~~ sc~~~~ ~ru'~ ~~~
