 |
Description  |
|
|
This invention relates to antigenic compounds found in malarial parasites
and to their isolation and use. In particular the invention relates to
antigenic glycoproteins present in Plasmodium falciparum merozoites. The
invention also relates to corresponding monoclonal antibodies and
hybridoma cell lines, and to vaccines comprising the antigenic
glycoproteins.
Malaria is currently the most prevalent death-threatening infectious
disease in the world. Malaria is predominantly a problem in developing
nations; thus, if a vaccine could be produced, it would be likely to have
dramatic effects on the general health and productivity of these people.
Although accurate statistics are not readily available, there are likely
to be 300-500 million cases of malaria per year. Approximately 1% result
in death the rest (297-495 million) have varying degrees of illness which
on the average claim one or more weeks of effective work time, even among
the educated who are properly diagnosed and treated. Thus, the morbidity
due to this disease is even more devastating to the economy of these
countries than is the mortality.
For the elimination of the asexual blood stages of the human malarial
parasite Plasmodium falciparum it is necessary for the immune response to
recognize antigens, including modified or unmasked antigens, found at the
surface of either the parasitized erythrocyte or the extracellular
merozoite. The importance of merozoite surface antigens has been indicated
by in vitro assays which suggest that antibodies to these antigens can
block the merozoites from binding to and penetrating erythrocytes.
Several approaches have been used to directly and indirectly investigate
the polypeptide composition of the merozoite. Metabolically labeled
parasites have been immunoprecipitated with immune sera which inhibit the
growth of the parasite in vitro. By comparing these results to those
obtained with noninhibiting sera, and assuming that the inhibition
observed in such an assay is due to antibody binding to merozoites and not
to infected erythrocyte surfaces, Reese et al., Am. J. Trop. Med. Hyg.
30:1168 (1981) showed that the ability to inhibit parasite growth in vitro
correlated with the ability to immunoprecipitate proteins of relative
molecular weights (M.sub.r) of approximately 200K, (that is, 200,000),
70-85K, and 45K. Comparable results were obtained by Perin and Dayal,
Immunological Rev. 61:245 (1982), and Perrin et al., Trans. Roy. Soc.
Trop. Med. Hyg. 75:163 (1981) except that they also identified two
additional proteins of approximate M.sub.r 140K and 110K. Conversely,
Brown et al., Nature 279:591 (1982), using a similar technique were able
to identify only one such protein (approximate M.sub.r 96K).
A second way of investigating the polypeptide composition of the merozoites
is to metabolically label parasites late in the cell cycle and to note
which labeled proteins are still present in the ring stages after
reinvasion. Using such a procedure Myler et al., Mol. Biochem. Parasitol.
9:37 (1983), identified fifteen .sup.35 S-methionine-labeled merozoite
proteins (approximate M.sub.r 240-14K). Heidrich et al., Z. Parasitenkd.
69:715 (1983), identified six merozoite "surface" antigens (approximate
M.sub.r 81K, 38K, 35K, 20K, and 12K) while Freeman and Holder J. Exp. Med.
158:1647 (1983), have identified eleven such proteins (approximate M.sub.r
160K, 105K, 83K, 73K, 70K, 65K, 48K, 42K, 41K, 38K, and 37K).
Finally, monoclonal antibodies (McAb's), which react with what appear by
immunofluorescence to be merozoites, have been used to immunoprecipitate
antigens from metabolically-labeled parasites. Using this approach Perrin
and Dayal, (1982), supra, described at least three proteins (approximate
M.sub.r 140K, 82K, and 41K), Howard et al. three proteins (approximate
M.sub.r 82K, 39K and 37K), Hall et al., Mol. Biochem. Parasitol. 7:247
(1983), two proteins (approximate M.sub.r 190K and 160K), and Holder and
Freeman J. Exp. Med. 156:1528 (1982), one protein (approximate M.sub.r
83K), which they believe to be merozoite associated.
The invention comprises a class of antigenic glycoproteins, in essentially
pure form, substantially similar to or derived from antigenic present on
the surface of Plasmodium falciparum, in particular on the surface of the
merozoite form of P. falciparum. For present purposes, a glycoprotein is
"substantially similar" to another glycoprotein where their amino acid
sequences are substantially the same and where the "glyco", or sugar
parts, are substantially the same. This class of glycoproteins includes
glycoproteins of molecular weights of approximately 185K, 88K, 56K, 46K
and 34K, as present in isolates of P. falciparum. The class also includes
an approximately 50K glycoprotein which appears in some isolates instead
of the 56K glycoprotein in other isolates. The 56K and 50K glycoproteins
are of particular importance to this invention as antigens.
The invention also comprises monoclonal antibodies (McAb's) which bind to
the glycoproteins of the invention; hybridoma cell lines which are capable
of producing these monoclonal antibodies; and vaccines and vaccine
compositions comprising these glycoproteins or epitopes substantially
similar to or cross reactive with these glycoproteins or genes or gene
fragments encoding such epitopes. Physiologically acceptable adjuvants or
carriers may also comprise part of the vaccines or vaccine compositions.
Methods of preparing parasite cultures and hybridomas for use with the
present invention and procedures for electron microscopy and
immunoprecipitation are described below, followed by the results of
experiments.
Parasites
The Plasmodium falciparum isolates FVO (Vietnam), Indochina I (Vietnam),
Honduras I/CDC (Central America), Geneva (Senegal, West Africa), Kenya
(East Africa), and Tanzania I (East Africa) were cultured using standard
procedures (Trager and Jensen, Science 193:673 (1976)). Cultures of these
isolates are available. To obtain material enriched for segmenters and
merozoites, the cultures were treated with, Physiogel (Reese et. al.,
Bull. WHO 57 (suppl. 1):53 (1979)) when most of the parasites were early
trophozoites. The infected erythrocytes (about 50-70% parasitemia) were
then returned to culture (0.5% hematocrit) for 18 hr. Segmenters and free
merozoites were harvested, washed twice with RPMI-1640 medium containing
25 mM HEPES (RPMI), and used as follows: (1) frozen at -70.degree. C. for
subsequent enzyme-linked immunosorbent assays (ELISAs), (2) smeared on
slides and acetone fixed for indirect fluorescence antibody tests (IFAT),
(3) repeatedly frozen and thawed for mouse immunizations.
Hybridomas
Balb/c ByJ mice were injected eight times at weekly intervals with FVO
parasite material corresponding to 10.sup.6 -10.sup.7 infected
erythrocytes. Three days after the final injection the splenocytes
(10.sup.8 cells) from one mouse were fused with an equal number of
P3-X63-Ag8 myeloma cells using polyethylene glycol and standard
methodology (Kohler and Milstein, Nature 56:495 (1975)). The fused cells,
in Dulbecco's medium supplemented with hypoxanthine, aminopterin and
thymidine and 10% (v/v) fetal bovine serum (FBS), were then placed in
twenty-five 96-well tissue-culture plates using Balb/c ByJ thymocytes as a
feeder layer. The cells from wells in which hybridomas grew were
subcultured and the spent medium assayed by ELISA using malarial antigen
attached to 96-well tissue culture plates and an alkaline
phosphatase-conjugated goat anti-mouse Ig (Tago, Burlingame, CA). The
media which were considered positive by ELISA were subsequently assayed by
IFAT. The hybridomas producing antibodies specific for merozoites and late
schizonts were cloned by limiting dilution and adapted to growth in
Dulbecco's medium plus 10% gamma globulin-free horse serum. Supernatants
from these cultures were then concentrated 10x using PM30 Amicon Diaflo
ultra filters.
Electron Microscopy
To obtain sufficient numbers of merozoites for immunoelectron microscopy,
the segmenter and merozoite enriched culture material was incubated with
agitation for 2 hr at a 1% packed cell volume (PCV) in RPMI plus 20% human
serum. The cultures were then centrifuged at 300 g (10 min) to remove most
of the erythrocytes and then at 1300 g (10 min) to pellet the remaining
merozoites. The merozoites were washed once with RPMI and fixed for 10 min
with 0.075% glutaraldehyde in RPMI. After washing three times with RPMI,
the cells were resuspended in the concentrated hybridoma culture
supernatants and placed on ice for 15 min. They were then washed three
times with RPMI and 1-2 ul PCVs were placed in 200 ul volumes of a
ferritin-labeled anti-mouse Ig conjugate (Cappel, West Chester, Pa.)
appropriately diluted in PBS containing 1% (w/v) BSA. The cells were left
on ice for 15 min, washed three times with RPMI, fixed with 2%
glutaraldehyde in cacodylate buffer (pH 7.3), dehydrated, and embedded in
EPON 812. Uranyl acetate stained thin sections were then examined in a
Hitachi HU-12A electron microscope.
Immunoprecipitation
Trophozoites were concentrated by Physiogel treatment and 2.times.10.sup.8
infected cells were grown approximately 18 hr in 10 ml RPMI containing 15%
human serum and I mCi .sup.3 H-glucosamine (Amershm, Arlington Heights,
Il). The infected erythrocytes and free merozoites were pelleted by
centrifugation at 1300 g for 10 min, diluted with PBS to a 20% PCV, and
frozen at -70.degree. C. Immediately prior to an immunoprecipitation assay
the cells were solubilized at a 1% PCV in radioimmunoprecipitation assay
Buffer A (150 mM NaCl, 40 mM NaF, 20 mM EDTA, 2% (v/v) Trasylol, 1% Trito
X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 10 mM Tris
HCl, pH 7.4, at 0.degree. C.) and centrifuged at 100,000 g for 1 hr at
4.degree. C. The supernatant (100 ul) was then used to resuspend 10 ul
antibody-bound protein A-Sepharose, previously prepared by the methods of
Schneider et al., J. Biol. Chem. 257:10766 (1982). Antibodies from Aotus
monkeys (karyotype VI) immune to the FVO isolate of P. falciparum had been
covalently bound directly to the protein A-Sepharose while the hybridoma
antibodies were indirectly bound using affinity purified polyvalent rabbit
anti-mouse Ig antibody (Tago, Burlingame, CA). The antibody-bound protein
A-Sepharose was incubated in the antigen solution for 1 hr (4.degree. C.).
The protein A-Sepharose-linked immune complexes were then washed twice
with 0.75 ml Buffer A, once with 0.75 ml Buffer A containing 500 mM NaCl,
and once with distilled water. The antigen was dissociated by boiling in
sample buffer. Samples were then electrophoresed in parallel with .sup.14
C-labeled molecular weight marker proteins on 9% acrylamide gels
containing sodium dodecyl sulfate.
Hybridomas were produced which produce antibodies specific for P.
falciparum merozoites. The monoclonal antibodies which reacted with, that
is, bound to or immunoprecipitated, merozoites from the FVO isolate gave
two distinctly different fluorescence patterns. The first pattern
(typified by McAb 4-10-2A), appeared as two closely associated intense
fluorescent spots localized within a region of the merozoite separate from
the nucleus. The second pattern (obtained with McAb's 4-8-5D and 4-13-4B)
is characterized by a bright fluorescence surrounding each merozoite and
would appear to be due to antibodies bound to the merozoite surface.
However, since fluorescence microscopy on acetone-fixed parasites alone is
inadequate to unequivocally localize the antigens, immunoelectron
microscopy was employed. The results demonstrate that McAb's 4-8-5D and
4-13-4B bound to the merozoite surface while McAb 4-10-2A did not bind to
the surface.
To determine the isolate specificity of McAb's 4-8-5D and 4-13-4B,
parasites from five different areas were examined by IFAT and one
(Honduras I/CDC) was also examined by immunoelectron microscopy. While
McAb 4-13-4B reacted by IFAT with the Honduras I/CDC isolate and most of
the other isolates, McAb 4-8-5D only reacted by IFAT with the FVO and
Geneva isolates. The inability of McAb 4-8-5D to bind efficiently to the
Honduras I/CDC isolate was subsequently confirmed by immunoelectron
microscopy. In this experiment McAb 4-13-4B was used as a positive control
to demonstrate that the Honduras I/CDC merozoites contained other surface
antigens even though they lacked or contained significantly lower amounts
of the antigen to which the McAb 4-8-5D binds.
Immunoprecipitation assays were conducted to characterize the molecules
with which McAb 4-8-5D reacted. When .sup.3 H-glucosamine-labeled FVO or
Geneva parasites were employed as antigen, McAb 4-8-5D immunoprecipitated
a single molecule of approximate M.sub.r 56K as determined by sodium
dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Similar
results were obtained when the parasites were metabolically labeled with
.sup.35 S-methionine. These data suggest that the molecule is a
glycoprotein (Gp). When a total cell extract of .sup.3
H-glucosamine-labeled FVO proteins are separated by SDS-PAGE, it is clear
that Gp56K is one of the major parasite glycoproteins. Resolution of the
.sup.3 H-glucosamine-labeled glycoproteins from all six isolates on
SDS-PAGE showed that the FVO and Geneva isolates both had a major
glycoprotein of 56K. The other four isolates lacked Gp56K but had major
glycoproteins of slightly lower molecular weight (approximately 50K).
To assess immunologically the relatedness of Gp50K and Gp56K, the antigens
were immunoprecipitated with sera from two Aotus monkeys made immune to P.
falciparum by repeated injections of the FVO isolate. The results using
two such sera produced virtually identical results. The Gp56K from the FVO
and Geneva isolates was clearly immunoprecipitated while Gp50K from the
other isolates apparently was not recognized by the immune monkey sera
from animals exposed to only FVO parasites. Either Gp56K or Gp50K in each
of the isolates was bound by antibodies produced by immunizing monkeys
with FVO and Honduras I/CDC parasites.
McAb (4-8-5D) produces a peripheral staining of both intracellular and
extracellular merozoites by an IFAT. This type of fluorescence has been
ascribed to antibody binding to surface antigens. However, unlike other
McAb's which bound putative surface and intracellular antigens (Hall et
al., (1983), supra,; Holder and Freeman, (1982), supra,), and which
immunoprecipitated multiple polypeptides, McAb 4-8-5D bound no other major
constituent of P. falciparum besides Gp56K. The immunoprecipitation of a
single (glyco)polypeptide by McAb 4-8-5D together with the immunoelectron
microscopic data which demonstrates the extension of this molecule into
the aqueous environment surrounding the cell indicated that Gp56K itself
must be a component of the merozoite coat and/or plasma membrane.
The glycoproteins of the invention can be labeled in various ways,
including .sup.35 S-methionine labeling and .sup.3 H-glucosamine labeling.
.sup.35 S-methionine labeling was used to identify six proteins of
approximate molecular weights of 202K, 185K, 142K, 136K, 82K and 46K, and
.sup.3 H-glucosamine labeling was used to identify five glycoproteins of
approximate molecular weights of 185K, 88K, 56K, 46K and 34K. The
glycoprotein from FVO parasites of M.sub.r 56K was labeled with .sup.3
H-glucosamine and .sup.3 H-mannose, but not detectably labeled with
.sup.35 S-methionine, .sup.3 H-fucose, or the sialic acid precursor .sup.3
H-N-acetyl mannosamine. The classification of this antigen as being a
glycoprotein was based on its sensitivity to pronase treatment and that it
could be labelled with .sup.3 H-glucosamine. The use of monoclonal
antibodies has allowed isolation of this antigen and clear demonstration
that it can be labeled with .sup.35 S-methionine. In addition, it has
allowed purified glucosamine labeled Gp56K to be treated with glycosidases
to show that the .sup.3 H-glucosamine had been incorporated into sugar
side-chains as a monosaccharide, not into amino acids. This firmly
establishes the molecule as a glycoprotein. The use of monoclonal
antibodies showed that this antigen could be labeled with .sup.35
S-methionine and, therefore, was a glycoprotein.
The fact that the isolates with which McAb 4-8-5D did not react lack the
56K glycoprotein but contain major glycoproteins of 50K suggests homology
between the 56K and 50K glycoproteins. For all isolates tested, each one
has either the 50K or the 56K glycoprotein. Further evidence of the
homology between these two glycoproteins is their isoelectric points. They
are two of the most acidic glycoproteins made by the parasite, with an
isoelectric point around 5.5. Most other glycoproteins have isoelectric
points closer to neutrality. The hybridoma which produces McAb 4-8-5-D is
on deposit with the American Type Culture Collection as ATCC HB 8938.
* * * * *
|
|
 |
|
 |
Description |
|
