US20070036818A1

US20070036818A1 - Recombinant protein containing a C-terminal fragment of Plasmodium MSP-1

Info

Publication number: US20070036818A1
Application number: US11/476,865
Authority: US
Inventors: Shirley Longacre-Andre; Charles Roth; Faridabano Nato; John Barnwell; Kamini Mendis
Original assignee: Institut Pasteur de Lille; New York University NYU
Current assignee: Institut Pasteur de Lille; New York University NYU
Priority date: 1996-02-14
Filing date: 2006-06-29
Publication date: 2007-02-15
Also published as: EP2194132A1; CA2241802A1; US7696308B2; EP0880588B1; KR20000065266A; DK0880588T3; WO1997030158A3; JP2000516083A; EP0880588A2; US20100291133A1; CN1276975C; AU1884197A; ATE456658T1; WO1997030158A2; CN1222936A; CN101230104A; US6958235B1; WO1997030158A9; PT880588E; FR2744724B1

Abstract

The present invention provides a recombinant protein comprising a 19 kDa C-terminal fragment of the surface protein 1 of the merozoite form of a Plasmodium type parasite other than Plasmodium vivax which is infectious in man.

Description

CONTINUATION DATA

This application is a Continuation of U.S. application Ser. No. 09/134,333, filed Aug. 14, 1998, pending, which is a CIP of PCT/FR97/00290, filed Feb. 14, 1997, now patented as U.S. Pat. No. 6,958,235.
In any event, what the real vaccination rate would be which could possibly be obtained with such recombinant proteins is also questionable, bearing in mind the discovery—reported below—of the presence in p42s from Plasmodiums of the same species, and more particularly in the corresponding p33s of hypervariable regions which would in many cases render uncertain the immunoprotective efficacy of antibodies induced in individuals vaccinated with a p42 from a Plasmodium strain against an infection by other strains of the same species (14).
MSP-1 has already been the subject of a number of studies. It is synthesized in the schizont stage of Plasmodium type parasites, in particular Plasmodium falciparum, and is expressed in the form of one of the major surface constituents of merozoites both in the hepatic stage and in the erythrocytic stage of malaria (1, 2, 3, 4). Because of the protein's predominant character and conservation in all known Plasmodium species, it has been suggested that it could be a candidate for constituting anti-malarial vaccines (5, 6).
The same is true for fragments of that protein, particularly the natural cleavage products which are observed to form, for example during invasion by the parasite into erythrocytes of the infected host. Among such cleavage products are the C-terminal fragment with a molecular weight of 42 kDa (7, 8) which is itself cleaved once more into an N-terminal fragment with a conventional apparent molecular weight of 33 kDa and into a C-terminal fragment with a conventional apparent molecular weight of 19 kDa (9) which remains normally fixed to the parasite membrane after the modifications carried out on it, via glycosylphosphatidylinositol (GPI) groups (10, 11).
It is also found at the early ring stage of the intraerythrocytic development-cycle (15, 16), whereby the observation was made that the 19 kDa fragment could play a role which is not yet known, but which is doubtless essential in re-invasive processes. This formed the basis for hypotheses formed in the past that that protein could constitute a particularly effective target for possible vaccines.
It should be understood that the references frequently made below to the p42 and p19 proteins from a certain type of Plasmodium are understood to refer to the corresponding C-terminal cleavage products of the MSP-1 protein of that Plasmodium or, by extension, to products containing substantially the same amino acid sequences, obtained by genetic recombination or by chemical synthesis using conventional techniques, for example using the “Applied System” synthesizer, or by “Merrifield” type solid phase synthesis. For convenience, references to “recombinant p42” and “recombinant p19” refer to “p42” and “p19” obtained by techniques comprising at least one genetic engineering step.
Faced with the difficulty of obtaining large quantities of parasites for P. falciparum and the impossibility of cultivating P. vivax in vitro, it has become clear that the only means of producing an anti-malaria vaccine is to resort to techniques which use recombinant proteins or peptides. However, MSP-1 is very difficult to produce whole because of it large size of about 200 kDa, a fact which has led researchers to study the C-terninal portion, the (still unknown) function of which is probably the more important. In addition, the extensive polymorphism in the N-terminal portion of MSP-1 has negative implications for the use of these parts of the molecule in vaccine preparation.
Recombinant proteins concerning the C-terminal portion of the P. falciparum MSP-1 which have been produced and tested in the monkey (12, 40, 41) are:

- a p19 fused with a glutathione-S-transferase produced in E. coli (40);
- a p42 fused with a glutathione-S-transferase produced in E. coli (12);
- a p19 fused with a polypeptide from a tetanic anatoxin and carrying auxiliary T cell epitopes produced in S. cerevisiae (12);
- a p42 produced in a baculovirus system (41).

A composition containing a p19 protein fused with a glutathione-S-transferase produced in E. coli combined with alum or liposomes did not exhibit a protective effect in any of six vaccinated Aotus nancymai monkeys (40).
A composition containing a p42 protein fused with a glutathione-S-transferase produced in E. coli combined with Freunds complete adjuvant did not exhibit a protective effect in two types of Aotus monkeys (A. nancymai and A. vociferans) when administered to them. The p19 protein produced in S. cerevisiae exhibited a protective effect in two A. nancymai type Aotus monkeys (12). In contrast, there was no protective effect in two A. vociferans type Aotus monkeys.
Some researchers (18) have also reported immunization tests carried out in the rabbit using a recombinant p42 protein produced in a baculovirus system and containing one amino acid sequence in common with P. falciparum (18). Thus these latter authors indicate that in the rabbit that recombinant p42 behaves substantially in the same way as the entire recombinant MSP-1 protein (gp195). This p42 protein in combination with Freunds complete adjuvant has been the subject matter of a vaccination test in a non-human primate susceptible to infection by P. falciparum, Aotus, lemurinus grisemembra (40). The results showed that 2 of 3 animals were completely protected and the third, while exhibiting a parasitemia which resembled that of the controls, had a longer latent period. It is nevertheless risky to conclude to a protective nature in man of the antibodies thus induced against the parasites themselves. It should be remembered that there are currently no very satisfactory experimental models in the primate for P. vivax and P. falciparum, The Saimiri model, developed for P. falciparum and P. vivax, and the Aotus model for P. falciparum, are artificial systems requiring the parasite strains to be adapted and often requiring splenectomy of the animals to obtain significant parasitemia. As a result, the vaccination results from such models can only have a limited predictive value for man.
In any event, what the real vaccination rate would be which could possibly be obtained with such recombinant proteins is also questionable, bearing in mind the discovery—reported below—of the presence in p42s from Plasmodiums of the same species, and more particularly in the corresponding p33s, of hypervariable regions which would in many cases render uncertain the immunoprotective efficacy of antibodies induced in individuals vaccinated with a p42 from a Plasmodlum strain against an infection by other strains of the same species (13).
It can even be assumed that the high polymorphism of the N-terminal portion of p42 plays a significant role in immune evasion, often observed for that type of parasite.
The aim of the present invention is to produce vaccinating recombinant proteins which can escape these difficulties, the protective effect of which is verifiable in genuinely significant experimental models or is even directly in man.

BRIEF DESCRIPTION OF THE DRAWING

FIG. 1A illustrates the nucleotide and amino acid sequences of the synthetic gene (Bac 19) and the “native gene” (PF19) of P. falciparum described by Chang et al.
FIG. 1B illustrates the nucleotide and amino acid sequences of the synthetic gene (Bac 19) and the “native gene” (PF19) of the Uganda Palo Alto isolate of P. falciparum.
FIG. 1C illustrates the PfMSP1_P19A recombinant protein sequence before cutting out the signal.
FIG. 1D illustrates the PfMSP1_P19A recombinant protein after cutting out the signal sequence.
FIG. 2A is an immunoblot using SDS-PAGE of the soluble recombinant PfMSP1_P19A antigen purified by immunoaffinity in the presence (reduced) or absence (non-reduced) of β-mercaptoethanol.
FIG. 2B is an immunoblot with human antiserum of recombinant purified MSP-1 P19 from P. vivax and P. cynomolgi under non-reduced (NR), reduced only in the charging medium (R) and irreversibly reduced (IR) conditions.
FIG. 3A is an immunoblot of the soluble PvMSP1_p42recombinant antigen in the presence of protein fractions derived from merzoites of P. faliciparum and separately isoelectric focusing in the presence (reduced) or absence (nonreduced) of β-mercaptoethanol.
FIG. 3B is a graph illustrating the results of an ELISA inhibition technique of P. vivax MSP-1 P42 and P19 antigens by the antiserum of individuals with an acquired immunity to P. vivax.
FIG. 4 recites nucleotide sequences. The underlined oligonucleotides originate from P. vivax and are used as primers in a PCR reaction. The lower portion of FIG. 4 illustrates the percent identity between two isolates of P. vivax and P. cynomolgi.
FIG. 5 shows curves illustrating the variation in the measured parasitemia as the number of parasited red blood cells per microliter of blood as the ftmction of time passed after infection. Curve A corresponds to the average values observed in three vaccinated monkeys and curve B corresponds to the average values in five controls.
FIG. 6A is a graph illustrating the parasitemia observed in non-vaccinated control animals as a function of time after injection.
FIG. 6BA is a graph illustrating the parasitemia observed in control animals which contained a saline solution also contain Freunds adjuvant as a function of time after injection.
FIG. 6C is a superposition of FIGS. 6A and 6B.
FIG. 6D is a graph illustrating parasitemia at the end of vaccination with p42 as a function of time.
FIG. 6E is a graph illustrating parasitemia in animals vaccinated with p19 alone as a function of time.
FIG. 6F is a graph illustrating parasitemia in animals with a mixture of P42 and P19 as a function of time.
FIG. 6G is the data obtained to produce the graphs in FIGS. 6A to 6F.
FIG. 7A is an immunoblot illustrating the in vivo response of monkeys to injections of p19 with Freunds adjuvant (1), with alum (2) and in the form of liposomes (3).
FIG. 7B is an immunoblot illustrating the in vivo response of a squirrel monkey after three injections with p19 with Freunds adjuvant, with alum and in the form of liposomes.
FIG. 8A is a graph illustrating the percent parasitemia versus days post infection of six monkeys, which were immunized with recombinant MSP-1 (p19) six months earlier.
FIG. 8B is a graph illustrating the percent parasitemia versus days post infection of six monkeys that were immunized with normal saline and an adjuvant.
FIG. 8C is a graph illustrating the percent parasitemia versus days post infection of monkeys that were used as controls.
FIG. 8D is the data obtained to produce the graphs in FIGS. 8A to 8C.
FIG. 9A is a graph illustrating the percent parasitemia versus days post infection of 2 macaques immunized with recombinant p19 and alum.
FIG. 9B is a graph illustrating the percent parasitemia versus days post infection of 2 macaques immunized with recombinant p19 and alum.
FIG. 9C is a graph illustrating the percent parasitemia versus days post infection of a macaque immunized with p19.
FIG. 9D is a graph illustrating the percent parasitemia versus days post infection of 3 control macaques immunized with physiological water and alum.
FIG. 9E is the data obtained to generate the graphs in FIGS. 9A to 9D.
FIG. 10A is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with MSP-1 p19 and alum.
FIG. 10B is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with MSP-1 p19 and Freunds.
FIG. 10C is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with MSP-1 p19 with liposomes.
FIG. 10D is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with alum as the control.
FIG. 10E is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with Freunds as the control.
FIG. 10F is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with liposomes as the control.
FIG. 10G is a graph illustrating the percent parasitemia versus days post infection in a squirrel monkey immunized with physiological water as the control.
FIG. 11A is a drawing of the backbone of MSP1₁₉from P. cynomolgi showing disulfide bridges in bold line.
FIG. 11B is a drawing of the backbone of MSP1₁₉showing positions of sequence differences between P. cynomolgi and P. vivax.
FIG. 11C is a drawing of the backbone of homology-modeled MSP1₁₉of P. falciparum showing positions of sequence differences with P. cynomolgi.
FIG. 12D is a NOESY spectrum of P. vivax MSP1₁₉.
FIG. 12E is a NOESY spectrum of P. vivax MSP1₁₉.
FIG. 12F is a NOESY spectrum of P. vivax MSP1₁₉.
FIG. 12.0 a is a NOESY spectrum of P. cynomolgi MSP1₁₉.
FIG. 12.0 b is a NOESY spectrum of P. cynomolgi MSP1₁₉.
FIG. 12.0 c is a TOCSY spectrum of P. cynomolgi MSP1₁₉.
FIG. 12.1 a is a NOESY spectrum of P. vivax MSP1₁₉.
FIG. 12.1 b is a NOESY spectrum of P. vivax MSP1₁₉.
FIG. 12.1 c is a TOCSY spectrum of P. vivax MSP1₁₉.
FIG. 12.2 a is a NOESY spectrum of P. falciparum MSP1₁₉.
FIG. 12.2 b is a NOESY spectrum of P. falciparum MSP1₁₉.
FIG. 12.2 c is a TOCSY spectrum of P. falciparum MSP1₁₉.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

More particularly, the invention provides vaccinating compositions against a Plasmodium type parasite which is infectious for man, containing as an active principle a recombinant protein which may or may not be glycosylated, whose essential constituent polypeptide sequence is:

- either that of a 19 kilodalton (p19) C-terminal fragment of the surface protein 1 of the merozoite form {MSP-1 protein) of a Plasmodium type parasite which is infectious for man, said C-terminal fragment remaining normally anchored to the parasite surface at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;
- or that of a portion of that fragment which is also capable of inducing an immune response which can inhibit in vivo parasitemia due to the corresponding parasite;
- or that of an immunologically equivalent peptide of said p19 fragment or said portion of that fragment; and
- said recombinant protein further comprises conformational epitopes which are unstable in a reducing medium and which constitute the majority of the epitopes recognized by human antiserums formed against the corresponding Plasmodium.

The presence of such conformational epitopes plays an important role in the protective efficacy of the active principle of the vaccines. They are particularly found in the active principles which exhibit the other characteristics defined above, when they are produced in a baculovirus system. If need be, it is mentioned below that the expression “baculovirus vector system” means the ensemble constituted by the baculovirus type vector itself and the cell lines, in particular cells of insects transfectable by a baculovirus modified by a sequence to be transferred to these cell lines resulting in expression of that transferred sequence. Preferred examples of these two partners in the baculovirus system have been described in the article by Longacre et al. (14). The same system was used in the examples below. It goes without saying, of course, that variations in the baculovirus and in the cells which can be infected by the baculovirus can be used in place of those selected.
In particular, the recombinant protein is recognized by human antiserums formed against the corresponding Plasmodium or against a homologous Plasmodium when it is in its non reduced state or in a reduced non irreversible state, but is not recognized or is only recognized to a slight extent by these same antiserums when it is irreversibly reduced.
The unstable character of these conformational epitopes in a reducing medium can be demonstrated by the test described below in the examples, in particular in the presence of β-mercaptoethanol. Similarly, the examples below describe the experimental conditions applicable to obtain irreversible reduction of the proteins of the invention.
From this viewpoint, the recombinant protein produrced by Longacre et al. (14) can be used in such compositions. It should be remembered that S. Lorigacre et al. succeeded in producing a recombinant p19 from the MSP-1 of P. vivax in a baculovirus vector system containing a nucleotide sequence coding for the p19 of Plasmodium vivax, in particular by transfecting cultures of insect cells [Spodoptera frugipeida (Sf9) line] with baculovirus vectors containing, under the control of the polyhedrin promoter, a sequence coding for the peptide sequences defined below, with the sequences being placed in the following order in the baculovirus vector used:

- a 35 base pair 5′ terminal fragment of the polyhedrin signal sequence, in which the methionine codon for initiating expression of this protein had been mutated (to ATT);
- a 5′-terminal nucleotide fragment coding for a 32 amino acid peptide corresponding to the N-terminal portion of MSP-1, including the MSP-1 signal peptide;
- either a nucleotide sequence coding for P19, or a sequence coding for the p42 of the MSP-1 protein of Plasmodium vivax, depending on the case, these sequences also being provided with (“anchored” forms) or deprived of (soluble forms) 3′ end regions of these nucleotide sequences, whose end C-terminal expression products are reputed to play an essential role in anchoring the final p19 protein to the parasite membrane;
- 2 TAA stop codons.

For p42, the sequences derived from the C-terminal region of MSP-1 extend consequently from amino acid Asp 1325 to amino acid Leu 1726 (anchored form) or to amino acid Ser 1705 (soluble form) and for p19, the sequences extend from amino acid Ile 1602 to amino acid Leu 1726 (anchored form) or to amino acid Ser 1705 (soluble form) it being understood that the complete amino acid sequences of p42 and p19, whose initial and terminal amino acids have been indicated above follow from the gene of the Belem isolate of P. vivax which has been sequenced (20).
Similar results were obtained using, in the same vector systems, nucleotide sequences coding for the p42 and p19 of Plasmodium cynomolgi. The interest in P. cynomolgi is twofold: it is a parasitic species very close to P. vivax which is infectious for the macaque. It can also infect man. Furthermore, a natural host of P. cynomolgi, the toque macaque, is accessible for testing the efficacy of the protection of MSP-1 from P. cynomolgi in a natural system. In addition, the rhesus monkey which is considered to be one of the most representative species for immune reactions in man, can also be infected by P. cynomolgi.
In particular, excellent results have been obtained in vaccination tests carried out using the toque macaque with two recombinant polypeptides: soluble p42 and, in particular, soluble p19 derived from P. cynomolgi, respectively produced in a baculovirus system and purified on an affinity column with monoclonal antibodies recognizing the corresponding regions of the native MSP-1 protein. The following observations were made: six monkeys immunized with only p19 (three monkeys) and the p19 and p42 together (three monkeys) all exhibited practically sterile immunity after challenge infection. The results obtained in the three monkeys immunized with p42 were less significant. Two of them were as above, but since the third exhibited a lower parasitemia than the controls immunized with a PBS buffer in the presence of Freunds adjuvant (3 monkeys) or not immunized (3 monkeys), it was less clear.
A second challenge infection showed that the monkeys which had received p19 alone were protected for at least six months. A second vaccination test with p19 in combination with alum in this system (toque macaque P. cynomolgi) exhibited significant protection for 2 of the 3 monkeys. This is the first time that MSP-1 or another recombinant antigen has demonstrated a protective effect in the presence of alum (42).
The particularly effective test results carried out with the macaque with recombinant polypeptides produced in a baculovirus system using a recombinant p19 from P. cynomolgi showed that recombinant polypeptides respectively containing recombinant p19s from other Plasmodiums must behave in the same manner. They are more meaningful for malaria in man than the results from tests carried out with P. vivax or P. falciparum in their “artificial hosts”.
Baculovirus recombinant proteins derived from a C-terminal MSP-1 portion (p19) have a very significant antimalarial protective effect in a natural system, which constitutes the most representative model for evaluating the protective effect of MSP-1 for man.
The protective effect obtained can be further improved if the p19 form is deprived of the hypervariable region of the N-terminal portion of p42, the effect of which can be deleterious in natural situations in which the vaccinated subject is confronted by a great deal of polymorphism. Further, p19 appears to possess specific epitopes which are not present in p42 (Holm et al, 1997).
The 19 kDa C-terminal fragment, the sequence of which is present in the active principle of the vaccine, can be limited to the sequence for the p19 itself, in the absence of any polypeptide sequence normally upstream of the p19 sequence in the corresponding MSP-1 protein. Clearly, though, the essential constiuent polypeptide sequence for the C-terminal side belonging to the 33 kDa (p33) N-terminal fragment still associated with the p19 in the corresponding p42, before natural cleavage of the latter, if the presence of this fragment does not modify the immunological properties of the active principle of the vaccine. As will be seen below, in particular in the description of the examples, the C-terminal sequences of the p33 in various strains of the same species of Plasmodium (see the C-terminal portion of the peptide sequences of “region III” in FIG. 4 (SEQ ID NOS:11-14)) also have a degree of homology or substantial conservation of the sequence, for example on the order of at least 80%, in different varieties of Plasmodiums which are infectious for man, such that they do not fundamentally modify the vaccinating properties of the active principle (the sequence of which corresponds to region IV in FIG. 4 (SEQ ID NOS:11-14)), in particular using the hypothesis which follows from this figure; that the presumed cleavage site between the p19 and region III of the p33 is located between the leucine and asparagines residues in a particularly well conserved region (LNVQTQ-SEQID NO:15).
Normally the C-terminal polypeptide sequence of the p33, when it is present, comprises less than 50 amino acid residues, or even less than 35, preferably less than 10 amino acid residues.
The essential constituent polypeptide sequence of the active principle of the vaccine need not comprise all of the sequence coding for p19, naturally providing that the latter retains the ability to induce antibodies which protect against the parasite. Preferably, this polypeptide fragment portion contains the two EGF (Epidermal Growth Factor) regions.
Clearly, the skilled person could distinguish between active fragments and those which would no longer be so, in particular experimentally by producing modified vectors containing inserts with different lengths originating from the p19, respectively isolated from the fragments obtained from the sequence coding for p19, by reaction with appropriate restriction enzymes, or by exonucleolytic enzymes which would be kept in contact with the fragment coding for p19 for differing periods; the capacity of the expression products from these inserts in the corresponding eukaryotic cells, in particular in insect cells, transformed by the corresponding modified vectors, to exert a protective effect can then be tested, in particular under the experimental conditions which are described below in the examples. In particular, the expression products of these inserts must be able to inhibit a parasitemia induced in vivo by the corresponding whole parasite.
Thus, the invention includes all vaccinating compositions in which the essential constituent polypeptide sequence of the active principle is constituted by a peptide which can induce a cellular and/or humoral type immunological response equivalent to that produced by p19 or a fragment as defined above, provided that the addition, deletion or substitution in the sequence of certain amino acids by others would not cause a large modification of the capacity of the modified peptide—hereinafter termed the “immunologically equivalent peptide”—to inhibit said parasitemia.
The p19 fragment can naturally also be associated at the N-terminal side or the C-terminal side or via a peptide bond to a further plasmoidal protein fragment having a vaccinating potential (such as Duffy binding protein from P. vivax (29) or EBA-175 from P. falciparum (30) and (31), one region of which is specifically rich in cysteine), provided that its capacity to inhibit parasitemia normally introduced in vivo by the corresponding parasite is not altered but is amplified.
Upstream of the N-terminal end of p19, the fragment coding for p19 or a portion thereof can also contain a peptide sequence which is different again, for example a C-terminal fragment of the signal peptide used, such as that for the MSP-1 protein. This sequence preferably comprises less than 50 amino acids, for example 10 to 40 amino acids.
These observations pertain in similar fashion to the p19 from other Plasmodia, in particular P. falciparum, the dominant species of the parasite, responsible for the most serious forms of malaria.
However, the techniques summarized above for producing a recombinant p19 from P. vivax or P. cynomolgi in a baculovirus system are difficult to transpose unchanged to producing a recombinant p19 of P. falciparum in a satisfactory yield, if only to obtain appreciable quantities which will allow immunoprotective tests to be carried out.
The invention also provides a process which overcomes this problem to a large extent. It also becomes possible to obtain much higher yields of P. falciparum p19—and other Plasmodiums where similar difficulties are encountered—using a synthetic nucleotide sequence substituting the natural nucleotide sequence coding for the p19 of Plasmodium falciparum in an expression vector of a baculovirus system, this synthetic nucleotide sequence coding for the same p19, but being characterized by a higher proportion of G and C nucleotides than in the natural nucleotide sequence.
In other words, the invention follows from the discovery that expression of a nucleotide sequence coding for a p19 in a baculovirus system is apparently linked to an improved compatibility of successive codons in the nucleotide sequence to express with the “cellular machinery” of the host cells transformable by the baculovirus, in the manner of that observed for the natural nucleotide sequences normally contained in these baculovirus and expressed in the infected host cells; hence the poor expression, or even total absence of expression of a native P. falciparum nucleotide sequence; hence also a possible explanation of the more effective expression observed by Longacre et al. (14) for the p19 of P. vivax in a baculovirus system and, as the inventors have also shown, of the P. cynomolgi sequence from corresponding native p19 nucleotide sequences, because of their relatively much higher amounts of G and C nucleotides than those of the native nucleotide sequences coding for the p19 of P. falciparum.
The invention thus more generally provides a recombinant baculovirus type modified vector containing, under the control of a promoter contained in said vector and able to be recognized by cells transfectable by said vector, a first nucleotide sequence coding for a signal peptide exploitable by a baculovirus system, characterized by a second nucleotide sequence downstream of the first, also under the control of the promoter and coding for the peptide sequence:

- either of a 19 kilodalton (p19) C-terminal fragment of the surface protein 1 of the merozoite form (MSP-1 protein) of a Plasmodium type parasite other than Plasmodium vivax which is infectious for man, said C-terminal fragment remaining normally anchored to the parasite surface at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;
- or of a portion of that peptide fragment provided that the expression product from the second sequence in a baculovirus system is alsocapable of inducing an immune response which can inhibit in vivo parasitemia due to the corresponding parasite;
- or of an immunologically equivalent peptide of said C-terminal peptide fragment (p19) or said peptide fragment portion by addition, deletion or substitution of amino acids not resulting in a large modification of the capacity of said immunologically equivalent peptide to induce a cellular and/or humoral type immunological response similar to that produced by said p19 peptide fragment or said portion of said fragment; and

said nucleotide sequence having, if necessary, a G and C nucleotide content in the range 40% to 60%, preferably at least 50%, of the totality of the nucleotides from which it is constituted. This sequence can be obtained by constructing a synthetic gene in which the natural codons have been changed for codons which are rich in G/C without modifying their translation (maintaining the peptide sequence).
The nucleotide sequence, provided by a synthetic DNA, may have at least 10% of modified codons with respect to the natural gene sequence or cDNA while retaining the characteristics of the natural translated sequence, i.e., maintaining the amino acid sequence.
It is not excluded that this G and C nucleotide content could be further increased provided that the modifications resulting therefrom as to the amino acid sequence of the recombinant peptide—or immunologically equivalent peptide—produced do not result in a loss of immunological properties, or protective properties, of the recombinant proteins formed, in particular in the tests which will be described below.
These observations naturally apply to other Plasmodium which are infectious for man, in particular those where the native nucleotide sequences coding for corresponding p-19s would have T and A nucleotide contents which are poorly compatible with effective expression in a baculovirus system.
The sequence coding for the signal used can be that normally associated with the native sequence of the Plasmodium concerned. But it can also originate from another Plasmodium, for example P. vivax or P. cynomolgi or another organism if it can be recognized as a signal in a baculovirus system.
The sequence coding for p19 or a fragment thereof in the vector under consideration is, in one case, deprived of the anchoring sequence of the native protein to the parasite from which it originates, in which case the expressed protein is generally excreted into the culture medium (soluble form). It is also remarkable in this respect that under the conditions of the invention, the soluble and anchored forms of the recombinant proteins produced, in particular when they are from P. falciparum or P. cynomolgi or P. vivax, tend to form oligomers, this property possibly being at the origin of the increased immunogenicity of the recombinant proteins formed.
The invention also concerns vectors in which the coding sequence contains the terminal 3′ end sequence coding for the hydrophobic C′-terminal end sequence of the p19 which is normally implicated in the induction of anchoring the native protein to the cell membrane of the host in which it is expressed. This 3′-terminal end region can also be heterologous as regards the sequence coding for the soluble p19 portion, for example corresponding to the 3′-terminal sequence from P. vivax or from another organism when it codes for a sequence which anchors the whole of the recombinant protein produced to the cell membrane of the host of the baculovirus system used. An example of such anchoring sequences is the GPI of the CD59 antigen which can be expressed in the cells of 10 Spodoptera frugiperda (32) type insects or the GPI of a CD14 human protein (33).
The invention also, naturally, concerns recombinant proteins, these proteins comprising conformational epitopes recognized by human serums formed against the corresponding Plasmodium.
In general, the invention also concerns any recombinant protein of the type indicated above, provided that it comprises conformational epitopes such as those produced in the baculovirus system, in particular those which are unstable in a reducing medium.
The invention also, naturally, concerns said recombinant proteins, whether they are in their soluble form or in the form provided with an anchoring region, in particular to cellular hosts used in the baculovirus system.
The invention also encompasses oligomers spontaneously produced in the baculovirus systems used or produced a posterori, using conventional protein oligomerisation techniques. The most commonly used technique involves glutaraldehyde. However, any conventional system for bridging between the respective amine and carboxyl functions in proteins can be used. As an example, any of the techniques described in European patent application EP-A-0 602 079 can be used.
The term “oligomer” means a molecule containing 2 to 50 monomer units, each of the monomer units containing p19 or a fragment thereof, as defined above, capable of forming an aggregate. The invention also encompasses any conjugation product between a p19 or a p19 fragment as defined above, and a carrier molecule—for example a polylysine-alanine—for use in producing vaccines, via bonds which are covalent or otherwise. The vaccinating compositions using them also form part of the invention.
The invention still fiurther concerns vaccine compositions using these oligomeric or conjugated recombinant proteins, including proteins from Plasmodium vivax, these observations also extending to oligomers of these recombinant proteins.
The invention also encompasses compositions in which the recombinant proteins defined above are associated with an adjuvant, for example an alum. Recombinant proteins containing the C-terminal end region allowing them to anchor to the membrane of the cells in which they are produced are advantageously used in combination with lipids which can form liposomes appropriate to the production of vaccines. Without being limiting, lipids described, for example, in the publication entitled “Les liposomes aspects technologique, biologique et pharmacologique” [Liposomes: technological, biological and pharmacological aspects] by J. Delattre et al., INSERM, 1993, can be used.
The presence of the anchoring region in the recombinant protein, whether it is a homologous or heterologous anchoring region as regards the vaccinating portion proper, encourages the production of cytophilic antibodies, in particular IgG_2aand IgG_2btype in the mouse which could have a particularly high protective activity, so that associating the active principles of the vaccines so constituted with adjuvants other than the lipids used to constitute the liposome forms could be dispensed with. This amounts to a major advantage, since liposomes can be lyophilized under conditions which enable them to be stored and transported, without the need for chains of cold storage means.
Other characteristics of the invention will become clear from the following description of examples of recombinant proteins of the invention and the conditions under which they can be produced. These examples are not intended to limit the scope of the invention.

Description of the Construction of PfMSP1_p19S (Soluble) (Soluble p19 from P. falciparum)

The recombinant construction PfMSP1_p19S contains the DNA corresponding to 8 base pairs of the leader sequence and the first 32 amino acids of the MSP-1 of Plasmodium vivax from Met1 to Asp₃₂(Belem isolate; Del Portillo et al., 1991, P. N. A. S., 88, 4030) followed by GluPhe due to the EcoR1 site connecting the two fragments. This is followed by the synthetic gene described in FIG. 1, coding the Plasmodium falciparum MSP1₁₉from Asn₁₆₁₃to Ser₁₇₀₅(Uganda-Palo Alto isolate; Chang et al., 1988, Exp. Parasitol., 67, 1). The construction is terminated by two TAA stop codons. This construction gave rise to a recombinant protein which was secreted in the culture supernatant from infected cells.
In the same manner and for comparison, a recombinant construction was produced under conditions which were similar to those used to produce the p19 above, but working with a coding sequence consisting of a direct copy of the corresponding DNA of the P. falciparum strain (FUP) described by Chang et al., Exp. Parasit. 67, 1; 1989. The natural gene copy (from asparagine 1613 to serine 1705) was formed from the native gene by PCR.
FIG. 1A shows the sequences of both the synthetic gene (Bac19) and the “native gene” (PF19).
It can be seen that 57 codons of the 93 codons of the native sequence coding for the p19 from P. falciparum were modified (the third nucleotide in 55 of them and the first and third nucleotides in the other 2 codons). New codons were added to the 5′ end to introduce the peptide signal under the conditions indicated above and to introduce an EcoRI site for cloning, and similarly two stop codons were added which were not present in the P. falciparum p19 to obtain expression termination signals. The individual letters placed above successive codons correspond to the respective successive amino acids. Asterisks (*) show the stop codons. Vertical lines indicate the nucleotides which are the same in the two sequences

Description of the PfMSP1_p19A Construction (Anchored GPI) (Anchored p19 of P. falciparum)

The PfMSP1_p19A construction had the characteristics of that above except that the synthetic sequence (FIG. 1B) codes for the MSP1_p19of Plasmodium falciparum (Uganda-Palo Alto isolate) from Asn₁₆₁₃to ILe₁₇₂₆followed by two TAA stop codons. This construction gave rise to a recombinant protein which was anchored in the plasma membrane of infected cells by a glycosyl phosphatidyl inositol (GPI) type structure.
FIG. 1C represents the PfMSP1_p19S recombinant protein sequence before cutting out the signal sequence.
FIG. 1D represents the PfMSP1_p19S recombinant protein sequence after cutting out the signal sequence.
The amino acids underlined in FIGS. 1C and 1D originate from the EcoR1 site used to join the nucleotide sequences derived from the N-terminal portion of the MSP-1 of P. vivax (with signal sequence) and the MSP-1_p19of P. falciparum.
FIG. 2—The soluble recombinant PfMSP1_p19antigen purified by immunoaffinity was analyzed by immunoblot using SDS-PAGE in the presence (reduced) or absence (non reduced) of β-mercaptoethanol. Samples were charged onto gel after heating to 95° C. in the presence of 2% SDS. Under these conditions only covalent type bonds (disulphide bridges) can resist disaggregation. The left hand blot was revealed with a monoclonal antibody which reacted with a linear epitope of natural p19. The right hand blot was revealed with a mixture of 13 human antisera originating from subjects with acquired immunity to malaria due to Plasmodium falciparum. These results show that the recombinant baculovirus molecule can reproduce conformational epitopes in the form of a polymer the majority of which are recognized by human antiserum.

- FIG. 2B: Immunoblot analysis with human antiserum of recombinant purified MSP-1 p19 from P. vivax and P. cynomolgi under non reduced (NR), reduced only in the charging medium (R) and irreversibly reduced (IR) conditions:

This work was based on the idea that the baculovirus expression system correctly reproduced the conformational epitopes present in vivo on the C-terminal portion of MSP-1 in large amounts. The best means of measuring this property (which may be the only possible means in the absence of native purified proteins corresponding to p19) was to study the reactivity of the recombinant proteins with the antiserum of individuals exposed to malaria, this reflecting the native proteins as “seen” by the human immune system.
Thus soluble recombinant PvMSP-1 p19 and PcMSP-1 p19 antigens purified by immunoaffinity were analyzed by immunoblot using SDS-PAGE (15%) in the presence (reduced) or absence (non reduced) of DTT. Samples were loaded onto the gel after heating to 95° C. in the presence of 2% SDS. The irreversible reduction was carried out as follows: the protein was resuspended in 0.2 M Tris-HCl, pH 8.4, 100 mM DTT, 1.0% SDS and heated for 30 minutes at 70° C. After diluting with water, acrylamide was added to a final concentration of 2 M and the mixture was incubated under nitrogen in the dark for 1 hour at 37° C. The immunoblot was revealed with a mixture of 25 human antisera originating from subjects with an acquired immunity to malaria due to Plasmodium vivax. V and C respectively designate proteins derived from the MSP-1 of P. vivax and P. cynomolgi. It should be noted that irreversibly reduced recombinant proteins exhibited no reactivity with human antiserum while non irreversibly reduced proteins or non reduced proteins exhibited good reactivity. (The non reduced Pv MSP-1 p19 was a little weak since in its glycosylated state it does not bind well to nitro-cellulose paper). These results show that recognition of baculovirus MSP-1 p19 molecules by human antiserum is largely if not completely dependent on conformational epitopes sensitive to reduction which are reproduced in this system.
FIG. 3—The soluble PvMSP1_P42recombinant antigen (Longacre et al., 1994, op. Cit.) was incubated for 5 hours at 37° C. in the presence of protein fractions derived from merozoites of P. falciparum and separated by isolectrofocussing. The samples were then analyzed by immunoblot in the presence (reduced) or absence (non reduced) of β-mercaptoethanol. Isolectrofocussing fractions 5 to 12, and two total merozoite extracts made in the presence (Tex) or absence (T) of detergent, were analyzed. The immunoblot was revealed with monoclonal antibodies specific for MSP1_p42and _p19of P. vivax . The results suggest that there is a proteolytic activity in the P. falciparum merozoites which can be extracted with detergent. Digestion of p42 in certain fractions appear to cause polymerization of the digestion products (p19); this polymerization is probably linked to the formation of disulphide bridges since in the presence of β-mercaptoethanol, the high molecular weight forms disappear in favor of a molecule of about 19 kDa (Tex-R). The p19 polymerization observed in these experiments could thus be an intrinsic property of this molecule in vivo.

FIG. 3B

The Differential Contribution of p42 and p19 Antigens to the P. vivax anti-MSP-1 Human Response

Recognition of P. vivax MSP-1 p42 and p19 antigens by the antiserum of individuals with an acquired immunity to P. vivax was compared using the ELISA inhibition technique as follows: a mixture of 25 human antisera originating from subjects with an acquired immunity to malaria due to P. vivax was diluted to 1:5000 and incubated for 4 hours at ambient temperature either alone, or in the presence of a 1 mM purified P. vivax recombinant p42 or p19. This mixture was transferred to a microtiter well which had been coated for 18 hours at 4° C. with 500 ng.ml⁻¹of purified absorbed recombinant p42 or p19, and incubated for 30 minutes at ambient temperature. After washing with PBS containing 0.1% of Tween 20, a goat anti-mouse IgG conjugated with peroxidase was added and the mixture was incubated for 1 hour at 37° C. The enzymatic activity was revealed by reading the optical density at 492 nm. The percentage inhibition was calculated based on values of 100% of antiserum activity with the coated antiserum on the microtiter plate in the absence of a competing antigen. Statistical data were calculated using a Statview program. Each bar represents the average percentage inhibition of a pair of competing/absorbed antigens based on 4 to 12 determinations; the vertical lines correspond to a 95% confidence interval. Asterisks (*) designate the antigens produced in the presence of tunicamycin, thus with no N-glycosylation. The important parameters of these measurements were the dilution of the antiserum by 1:5000 which is in the region which is sensitive to ELISA curves and the competing antigen concentrations of 1 mM which includes competition by low affinity epitopes. Thus these data reflect the maximum resemblance between the two compared antigens. The results show that the majority, if not all of the p42 epitopes recognized by the human antisenim are present on the p19 since in the presence of the latter, the reactivity of the human antiserum against p42 is inhibited as much as by the p42 antigen itself. In contrast, however, about 20% of the p19 epitopes recognized by human antiserum were not or were not accessible on the p42, since the reactivity of the human antiserum against the p19 was much less inhibited by p42 than by p19 itself. Such specific epitopes of p19 could be constituted or revealed only after cleaving the p42 into p19 and p33. These results were not affected by glycosylation showing that the effect is really due to a difference between the peptide components of p19 and p42 and not to a difference in glycosylation. These results underline the fact that p19 has a distinct immunological identity compared to p42.

Description of the PcMSP1_p19S (Soluble) Construction (Soluble p19 of P. cynomolgi)

The DNA used for the above construction was obtained from a clone of the Plasmodium cynomolgi ceylonesis strain (22-23). This strain had been maintained by successive passages through its natural host (Macaca sinica) and cyclic transmissions via mosquitoes (27).
Blood parasites in the mature schizont stage were obtained from infected monkeys when the parasitemia had attained a level of 5%. They were then purified using the methods described in (25). The DNA was then extracted as described in (26).
A 1200 base pair fragment was produced using a PCR reaction using the oligonucleotides underlined in FIG. 4 originating from P. vivax (see amino acids 1-6 and 373-380 of SEQ ID NO:13). The 5′ oligonucleotide comprised an EcoRI restriction site and the 3′ oligonucleotide comprises two synthetic TAA stop codons followed by a BglII restriction site. This fragment was introduced by ligation and via these EcoRI and BglII sites into the pVLLSV₂₀₀plasmid already containing the signal sequence for the MSP-1 protein of P. vivax (19). The new plasmid (pVLSV₂₀₀C₄₂) was used to analyze the DNA sequences.
The P cynomolgi (SEQ ID NO:11) and the corresponding P. vivax (SEQ ID NOS:12 and 13) sequences were aligned. The black arrows designate the presumed primary and secondary cleavage sites. They were determined by analogy with known sites in P. falciparum (27, 28). The vertical lines and horizontal arrows localize the limits of the four regions which were studied. Region 4 corresponded to the sequence coding for the P. cynomolgi p19. Glycosylation sites are boxed and the preserved cysteines are underlined. The lower portion of FIG. 4 shows the percentage identify between the two isolates of P. vivax and P. cynomolgi.
The recombinant construction PcMSP1_p19S contains the DNA corresponding to 8 base pairs of the leader sequence and the first 32 amino acids of the MSP-1 of Plasmodium vivax from Met₁to Asp₃₂(Belem isolate; Del Portillo et al., 1991, P. N. A. S., 88, 4030) followed by GluPhe, due to the EcoR1 site, connecting the two fragments. This is followed by the sequence coding for the Plasmodium cynomolgiMSP1_p19from Lys₂₇₆to Ser₃₈₀(Ceylon strain). The construction was terminated by two TAA stop codons. This construction gave rise to a recombinant protein which was secreted in the culture supernatant of infected cells.

Purification of Recombinant PfMSP1p19 Protein by Immunoaffinity Chromatography with a Monoclonal Antibody Specifically Recognizing the p19 of Plasmodium falciparum

The chromatographic resin was prepared by binding 70 mg of a monoclonal antibody (obtained from a G17.12 hybridoma deposited at the CNCM [National Collection of Microorganism Cultures] (Paris, France) on the 14 Feb. 1997, registration number 1-1846; this G17.12 hybridoma was constructed from X63 Ag8 653 myeloma producing IgG 2a/k recognizing the P. falciparum p19) to 3 g of activated CNBr-Sepharose 4B (Pharmacia) using standard methods detailed in the procedure employed by Pharmacia. The culture supernatants containing the soluble PfMSP1p19 were batch incubated with the chromatographic resin for 16 hours at 4° C. The column was washed once with 20 volumes of 0.05% NP40, 0.5 M of NaCl, PBS; once with 5 volumes of PBS and once with 2 volumes of 10 mM of sodium phosphate, pH 6.8. Elution was carried out with 30 ml of 0.2 M glycine, pH 2.2. The eluate was neutralized with 1 M sodium phosphate, pH 7.7 then concentrated by ultrafiltration and dialyzed against PBS. To purify the anchored PfMSP1p19, all of the washing and elution solutions contained a supplemental 0.1% of 3-(dimethyl-dodecylammonio)-propane sulphonate (Fluka).

Recombinant Plasmodium vivax (p42 and p19) MSP1 Vaccination Test in the Squirrel Monkey Saimiri sciureus

This vaccination test was carried out on male non splenectomized 2 to 3 year old Saimiri sciureus boliviensis monkeys. Three monkeys were injected 3 times intramuscularly at 3 week intervals with a mixture of about 50 to 100 μg each of recombinant soluble PvMSP1_p42and _p19(19), purified by immunoaffinity. Complete and incomplete Freunds adjuvant was used as follows: 1^stinjection: 1:1 FCA/FIA; 2^ndinjection: 1:4 FCA/FIA; 3^rdinjection: FIA. These adjuvant compositions were then mixed 1:1 with the antigen in PBS. Five control monkeys received the glutathione-S-transferase (GST) antigen produced in E. coli using the same protocol. The challenge infection was carried out by injecting 2×10⁶red blood cells infected with an adapted Plasmodium vivax strain (Belem) 2.5 weeks after the final injection. The protection was evaluated by determining parasitemia daily in all animals by examining smears stained with Giemsa.
The curves in FIG. 5 show the variation in the measured parasitemia as the number of parasitic red blood cells per microliter of blood (logarithmic scale on the ordinate) as a function of the time passed after infection (in days). Curve A corresponds to the average values observed in the three vaccinated monkeys; curve B corresponds to the average values in the five control monkeys.
An examination of the Figure shows that the effect of the vaccination was to greatly reduce the parasitmisa.

Recombinant Plasmodium cynomolgi (p42 and p19) MSP1 Vaccination Test in the Toque Macaque Macaca sinica

Fifteen captured monkeys were used as follows: (1) 3 animals injected with 100 μg of soluble PcMSP1p₄₂; (2) 3 animals injected with 35 μg (1^stinjection) or 50 μg (2^ndand 3^rdinjections) of soluble PcMSP1_p42; (3) 3 animals injected with a mixture of PcMSP1_p42and _p19; (4) 3 animals injected with adjuvant plus PBS; (5) 3 animals not injected. Complete and incomplete Freunds adjuvant was used in the protocol described above. Injections were intramuscular at 4 week intervals. The challenge infection was made by injecting 2×10⁵red blood cells infected with Plasmodium cynomolgi 4 weeks after the last injection. Protection was evaluated by determining parasitemia daily in all animals by examining the parasitemia with Giemsa. Parasitemia were classified as negative only after counting 400 smear fields. The parasitemia were expressed as a percentage of parasitised red blood cells.
FIGS. 6A-6G show the results obtained. Each of them shows parasitemia (expressed as the percentage of parasitised red blood cells along the ordinate on a logarithmic scale) observed in the challenge animals as a function of the time after infection {in days along the abscissa).
The results relate to:

- in FIG. 6A; non vaccinated control animals;
- FIG. 6B relates to animals which received a saline solution also containing Freunds adjuvant;
- FIG. 6C is a superposition of FIGS. 6A and 6B, with the aim of highlighting the relative results resulting from administration of Freunds adjuvant to the animals (the variations are clearly not significant);
- FIG. 6D provides the results obtained after vaccination with p42;
- FIG. 6E concerns animals vaccinated with p19 alone;
- finally, FIG. 6F concerns animals vaccinated with a mixture of p19 and p42.

The p42 certainly induced a certain level of protection. However, as shown in FIGS. 6E and 6F, the protection conferred by the recombinant p19 of the invention was considerably better.
The numbers used to produce graphs (6A-6F) are given in FIG. 6G.

P. cynomolgi Toque Macaque Vaccination Test; Second Challenge Infection of Monkeys Vaccinated with p19 Alone and Controls (FIGS. 8)

Six months later, with no other vaccination, the 3 macaques which received the p19 MSP-1 alone with FCA/FIA (FIG. 6E) and the 3 macaques which received a saline solution containing Freunds adjuvant (FIG. 6B) and 2 new unvaccinated monkeys underwent a new challenge infection by injecting 1×10⁶red blood cells infected with Plasmodium cynomolgi. Protection was evaluated by determining parasitemia daily in all animals by examining Giemsa smears. The parasitemia were classified negative only after counting 400 smear fields. The parasitemia were expressed as the percentage of parasitised red blood cells (the figures used to produce graphs 8A-C are given in FIG. 8D). The six immunized animals which underwent challenge infection six months earlier had no detectable parasitemia except for 1 animal in each group which exhibited a parasitemia of 0.008% for 1 day (FIGS. 8A and 8B). The two unaffected controls exhibited a conventional parasitemia with a maximum of 0.8% and for 21 days (FIG. 8C). Thus the 3 animals vaccinated with the MMSP-1 p19 were as well protected six months later as the 3 controls which exhibited a complete conventional infection after the first challenge infection, despite the absence of or a very slight parasitemia after the first challenge infection, (it is likely that protection against a homologous strain of P. vivax in humans can also be induced by a single blood infection provided that the infection is allowed to run its natural course without treatment since untreated P. vivax infection in humans gave rise to complete clinical protection against the subsequent challenge with the identical strain but not with a different strain of the same species). Together these results suggest either that the protection period for p19 is at least six months or that the immunity induced can be effectively boosted by minimal exposure to parasitic infection.

Vaccination Test with p19 in Association with Alum in the P. cynomolgi Toque Macaque System (FIG. 9)

The previous positive protection results were obtained using complete (FCA) or incomplete (FIA) Freunds adjuvant. However, the only adjuvant which is currently allowed in man is alum. For this reason, we carried out a vaccination test with P. cynomolgi MSP-1 p19 in the toque macaque in the presence of alum as the adjuvant. Six captured macaques were used as follows: (1) 3 animals injected with 4 doses of 50 mg of recombinant P. cynomolgi MSP-1 p19 with 10 mg of alum; (2) 3 animals injected 4 times with physiological water and 10 mg of alum. The injections were intramuscular at 4 week intervals. The challenge infection was made by injecting 2×10⁵red blood cells infected with P. cynomolgi 4 weeks after the last injection. Protection was evaluated by daily determination of parasitemia in all animals by examining Giemsa smears. The parasitemia were classified negative only after counting 400 smear fields. Parasitemia were expressed as the percentage of parasitised red blood cells. The results of this experiment were as follows. 2 of the 3 macaques immunized with recombinant p19 with alum had about 30 times less total parasitemia during the infection period (FIGS. 9A and 9B) than the 3 control macaques immunized with physiological water and alum (FIG. 9D) after the challenge infection. The third macaque immunized with p19 (FIG. 9C) was not very different from the controls. For the vaccination test using Plasmodium cynomolgi p19 in the toque macaque, macaca sinica, described in FIG. 9, the data used to produce the graphs (9A-9D) are given in (FIG. 9E). While the results are a little less spectacular than the preceding results (FIGS. 6, 8), this is the first time that significant protection has been observed for recombinant MSP-1 with alum.

FIG. 10

Vaccination Test with a Recombinant Plasmodium falciparum p19 in the Squirrel Monkey

Twenty Saimiri sciureus guyanensis (squirrel monkeys) of about 3 years old raised in captivity were used as follows: (1) 4 animals injected with 50 μg of soluble Pf MSP-1 p19 in the presence of Freunds adjuvant as follows: 1^stinjection: 1:1 FCA/FIA; 2_ndinjection: 1:4 FCA/FIA; 3^rdinjection: FIA. These adjuvant compositions were then mixed with 1:1 antigen in PBS; (2) 2 control animals received Freunds adjuvant as described for (1) with only PBS; (3) 4 animals injected with 50 μg of soluble Pf MSP-1 p19 in the presence of 10 mg of alum (Alu-Gel-S, Serva); (4) 2 control animals received 10 mg of alum with only PBS; (5) 4 animals injected with about 50-100 mg of GPI anchored PfMSP-1 p19 reconstituted into liposomes as follows: 300 mmoles of cholesterol and 300 mmoles of phosphatidyl choline were vacuum dried and resuspended in 330 mM of N-octylglucoside in PBS with 1.4 mg of PfMSP-1 p19, GPI. This solution had been dialyzed against PBS with adsorbent Bio-Beads SM-2 (Bio-Rad) and the liposomes formed were concentrated by centrifuging and resuspended in PBS The 1^stinjection was made with fresh liposomes kept at 4° C. and the 2_ndand 3_rdinjections were made with liposomes which had been frozen for preservation; (6) 2 animals injected with control liposomes made in the same way, in the absence of the p19, GPI antigen as described for (5); (7) 2 animals injected with physiological water. Three intramuscular injections were made at 4 week intervals. The challenge infection was made by injecting 1×10⁶red blood cells infected with Plasmodium falciparum. Protection was evaluated by determining parasitemia daily in all animals by examining the Giemsa smears. Parasitemia were expressed as the percentage of parasitised red blood cells. The results of this vaccination test are shown in FIGS. 10, A-G.
The groups immunized with p19 in Freunds adjuvant or liposome demonstrated similar parasitemia to the control groups after a challenge infection (one animal (number 29) vaccinated with p19 in Freunds adjuvant died several days after challenge infection for reasons independent of vaccination (cardiac arrest). Irregularities in administration of the antigen in these 2 groups (poor Freunds emulsion, congealed liposomes) did not allow the significance of these results to be completely evaluated. In the alum group, 2 animals showed total parasitemia for the duration of the infection about 8 times less than the controls, 1 animal about 3 times less and 1 animal was similar to the controls. This experiment was a little difficult to interpret due to the variability in the controls, probably due to the strain of parasite used for the challenge infection which would not have been quite adapted to the non splenectomized Saimiri model developed only recently in Cayenne. However, the real effect with alum, although imperfect, is encouraging in that our antigens seem to be the only recombinant P. falciparum MSP-1 versions which currently have shown a certain effectiveness in combination with alum.

FIG. 7

Specific Reactivity with P. falciparum MSP1 p19 Monkey Anti-sera with High Molecular Weight Aggregates (Same Antisera as for FIG. 10)

Monkeys bred in captivity were injected intramuscularly with 1 ml of inoculum twice at 4 week intervals as follows: (1) 4 animals injected with 50 μg of soluble PfMSP1p19 in the presence of Freunds adjuvant as follows: 1^stinjection: 1:1 FCA/FIA; 2^ndinjection: 1:4 FCA/FIA; and mixed then 1:1 with the antigen in PBS; (2) 4 animals injected with 50 μg of soluble PfMSPp19 in the presence of 10 mg of alum; (3) 4 animals injected with about 50 μg of GPI anchored PfMSP1p19 reconstituted into liposomes composed of 1:1 molar cholesterol and phosphatidyl choline. The animals were bled 17 days after the second injection.
Red cells from a squirrel monkey with 30% parasitemia due to P. falciparum (with the mature forms in the majority) were washed with PBS and the residue was diluted 8 times in the presence of 2% SDS and 2% dithiothreitol and heated to 95° before being charged onto a polyacrylamide gel of 7.5% (separation gel) and 4% (stacking gel). After transfer to nitrocellulose, immunoblot analysis was carried out with antisera as follows: (1) pool of antisera of 4 monkeys vaccinated with soluble PfMSP1p19 in Freunds adjuvant, 1:20; dilution; (2) pool of antisera of 4 monkeys vaccinated with soluble PfMSP1p19 in alum adjuvant, 1:20 dilution; (3) pool of antisera of 4 monkeys vaccinated with anchored PfMSP1p19 in liposomes, 1:20 dilution; (4) monoclonal antibody, which reacts with a linear epitope of PfMSP1p19, 50 μg/ml; (5) SHI90 antisera pool originating from about twenty monkeys repeatedly infected with P. falciparum and which had become resistant to any subsequent infection with P. falciparum, 1:500 dilution; (6) antiserum pool of unaffected monkeys (never exposed to P. falciparum ), 1:20: dilution.
The results in FIG. 7 show that the 3 antisera pools of monkeys vaccinated with PfMSP1p19 either with Freunds adjuvant (track 1), alum (track 2) or in liposomes (track 3) reacted specifically with very high molecular weight complexes (diffuse in the stacking gel) and present in parasite extracts containing more mature forms. These results support the hypothesis that a specific aggregate of PfMSP1p19 is present in vivo comprising epitopes which are reproduced in recombinant PfMSP1p19 molecules synthesized in the baculovirus system, in particular oligomeric forms thereof.
In FIG. 7B: The techniques and methods were the same as for FIG. 7 except that the individual antiserum for each monkey was tested after three injections the day of the challenge injection and the SHI antiserum was diluted by 1:250. The numbers correspond to the individual monkeys noted in FIG. 10. The results show that the anti serum for 4 monkeys vaccinated with p19 and alum reacted strongly and specifically with very high molecular weight complexes while the monkeys of other groups vaccinated with p19 and Freunds adjuvant or liposomes showed only a little reactivity as in the control reactivity with these complexes. Since the monkeys vaccinated with p19 and alum were also the best protected, this reactivity with the high molecular weight complexes appeared to indicate a protective effect, despite one monkey in the group not being protected with respect to the controls and that another was only partially protected.
The invention also, of course, concerns other applications, for example those described below with respect to certain of the examples, although these are not limiting in character.

Therapy

The recombinant molecule PfMSP1p19 can be used to produce specific antibodies which can possibly be used by passive transfer for therapeutics for severe malaria due to P. falciparum when there is a risk of death.

Diagnostics

The recombinant molecules PvMSP1p42, PvMSP1p19 and PfMSPp19 derived from baculovirus can and have been used to produce specific murine monoclonal antibodies. These antibodies, in combination with polyclonal anti-MSP1p19 antisera originating from another species such as the rabbit or goat can form the basis of a semi-quantitative diagnostic test for malaria which can distinguish between malaria due to P. falciparum, which can be fatal, and malaria due to P vivax, which is generally not fatal. The principle of this test is to trap and quantify any MSP-1 molecule containing the p19 portion in the blood.
In this context, the advantages of the MSP1p19 molecule are as follows:

- (i) it is both extremely well conserved in the same species and sufficiently divergent between different species to enable specific species reactants to be produced. No cross reaction has been observed between antibodies derived from PfMSP1p19 and PvMSP1p19;
- (ii) the function of MSP1p19, while not known with precision, seems to be sufficiently important that this molecule does not vary significantly or is deleted without lethal effect for the parasite;
- (iii) it is a major antigen found in all merozoites and thus it must in principle be detectable even at low parasitemia and proportionally to the parasitemia;
- (iv) since the recombinant MSP1p19 molecules derived from baculovirus appear to reproduce more of the native structure of MSP1p19, the antibodies produced against these proteins will be well adapted to diagnostic use.

The microorganisms identified below have been deposited at the Collection Nationale de Culture de Microorganisms de L'Institut Pasteur (CNCM) under Rule 6.1 of the Treaty of Budapest on 1 Feb. 1996. under the following registration numbers:

Identification reference Registration numbers

PvMSP1p19A 1-1659

PvMSP1p19S 1-1660

PfMSP1p19A 1-1661

PfMSP1p19S 1-1662

PcMSP1p19S 1-1663
The microorganism identified below has been deposited at the CNCM under Rule 6.1 of the Treaty of Budapest on Jul. 18, 1998 under the following registration number:

Identification reference Registration number

P1MSP1p19S; HisExt; SigPf 1-2041
The invention also concerns the use of these antibodies, preferably fixed to a solid support (for example for affinity chromatography) for the purification of type p19 peptides initially contained in a mixture.
Purification means bringing this mixture into contact with an antibody, dissociating the antigen-antibody complex and recovering the purified p19 type peptide.
The invention also concerns vaccine compositions, also comprising mixtures of proteins or fragment, in particular mixtures of the type:

- P. falciparum p19 and P. vivax p19;
- P. falciparum p19 and P. falciparum p42, the latter if necessary being deprived of its most hypervariable regions;
- P. vivax p19 and P. vivax p42, the latter if necessary being deprived of its most hypervariable regions;
- P. falciparum p19 and P. falciparum p42, the latter if necessary being deprived of its most hypervariable regions, and P. vivax p19 and P. vivax p42, the latter if necessary being deprived of its most hypervariable regions.

In the present case, the most hypervariable regions are defined as region II or region II and all or part of region III, the portion of region II which is preferably deleted being that which is juxtaposed to region II (the conserved portion being located to the side of the C-terminal of p33, close to the p19). Regions II and III are illustrated in FIG. 4 (SEQ ID NOS: 11-14).
The invention is not limited to the production of human vaccine, it is also applicable to the production of veterinary vaccine compositions using the corresponding proteins or antigens derived from parasites which are infectious for mammals and products under the same conditions. It is known that infections of the same type, babesiosis, also appear in cattle, dogs and horses. One of the antigens of the Babesia species has a high conformational homology (in particular in the two EFG-like and cysteine- rich domains) and functional homology with a protein portion of MSP-1 [(36), (37) and (38)].
Examples of veterinary vaccines using a soluble antigen against such parasites have been described (39).
It goes without saying that the p19s used in these mixtures can also be modified as described in the foregoing when considered in isolation.
The invention also concerns hybridomas secreting specific antibodies selectively recognizing the p19 of a MSP-1 protein in the merozoite form of a Plasmodium type parasite which is infectious for man other than Plasmodium vivax and which does not recognize Plasmodium vivax.
In particular, these hybridomas secrete monoclonal antibodies which do not recognize Plasmodium vivax and which specifically recognize Plasmodium falciparum p19.
The invention also concerns a hybridoma characterized in that it produces a specific antibody which specifically recognizes the p19 of P. vivax and the p19 of P. cynomolgi. A F10-3 hybridoma has been constructed from the X63 Ag8 653 myeloma producing IgG 2b/k recognizing the p42 glycoprotein of Plasmodium vivax. The F10-3 hybridoma has been deposited at the European Collection of Cell Cultures (ECACC) under Rule 6.1 of the Treaty of Budapest on Aug. 6, 1998 under the Accession Number: 98080510.
It should be noted that in the text which follows as well as in the Figures whose numerals are headed by numbers 11, 12, 13, 14 and 15 respectively the references to MSP1₁₉and MSP1₄₂stand for p19 and p42 as they have been defined hereabove.
Of particular significance is the high degree of purity of the recombinant p19 proteins as obtainable by expression in baculovirus systems and immunoaffinity or metallo-affinity chromatography.
Particularly, the recombinant p19 protein is crystallizable implicating an absolute purity. In addition, mass spectrometry measurements also show a very high degree of purity estimated at.greater than 95%.
Of particular significance also are the reduction-sensitive conformational epitopes in the two EGF domains of the preferred p19 recombinant proteins and the long term memory response directed in a substantially specific manner against said conformational epitopes, which said p19 recombinant proteins are capable of eliciting in laboratory animals.
These properties as well as other characteristics of preferred purified p19 recombinant proteins are further illustrated by the results provided by further studies which are reported hereafter.
The invention relates also particularly to recombinant proteins, as obtainable in a baculovirus vector system:

- in a pure state
- substantially free of any other form of recombinant protein which, has the same peptide sequences, but which contains alternate conformations in the two EGF regions. This alternate conformation is different from the conformational form as defined by:
- (a) the atomic coordinates as defined in Annexes I, II or III obtained by crystallography (the Annexes 1, II or III include respectively the atomic coordinates which define the P. cynomolgi MSP1₁₉ , P. vivax MSP1₁₉and P. falciparum MSP1₁₉three-dimensional molecular structure); and
- (b) the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

The following data emphasize the importance of reduction-sensitive conformational epitopes in the two EGF domains comprising the MSP119 antigens. The quantitative reproduction of a single conformation likely to resemble the native parasitic protein in Plasmodium C-terminal MSP119 recombinant proteins derived from a higher order expression system such as baculovirus (indicated by crystal formation as described below), is thought to be one of the essential active principles necessary for the protective effect of these antigens in the vaccination trials described previously. Three types of experimental data are included:
(i) The spatial organization of the baculovirus recombinant MSP1₁₉molecule from Plasmodium cynomolgi as defined by the determination of its crystal structure at 1.8 Å resolution. The probable structures of the corresponding MSP1₁₉antigens derived from Plasmodium vivax and Plasmodium falciparum were determined by molecular modeling by the replacement of alternative residues present in the latter two species as compared to the P. cynomolgi MSP1₁₉.
(ii) The nuclear magnetic resonance (NMR) spectra for each of the 3 MSP1₁₉recombinant proteins derived from P. cynomolgi, P. vivax and P. falciparum. These spectra represent precise, defined “fingerprints” of the conformation of the corresponding proteins in solution.
(iii) ELISA data indicating that the immune response to the recombinant MSP1₁₉molecules is almost entirely directed against reduction-sensitive conformational epitopes both in humans exposed to the malaria parasite and in the highly protected toque monkeys inoculated with the P. cynomolgi p19 recombinant protein (FIG. 6E). These results underline the importance of accurate and quantitative reproduction of these epitopes in the recombinant proteins designed to elicit a protective anti-malaria immune response. In addition it is shown that vaccination with the MSP1₄₂recombinant protein favors a long term memory response which is preferentially directed against non-conformational epitopes present in the MSP1₄₂in contrast to the MSP1₁₉whose long term memory response is also directed against the conformational epitopes associated with protection.
Reference is made hereafter to the sets of drawings headed by numerals 11, 12, 13, 14 and 15, and to which the following legends respectively correspond.
FIG. 11A: Backbone of MSP1₁₉from P. cynomolgi showing disulfide bridges in bold line.
FIG. 11B: Backbone of MSP1₁₉showing positions of sequence differences between P. cynomolgi and P. vivax.
FIG. 11C: Backbone of homology-modeled MSP1₁₉of P. falciparum with positions of sequence differences with P. cynomolgi.
FIG. 12: Reconstructed mass spectra and m/z spectra respectively of metalloaffinity purified P. cynomolgi (A,B), P. falciparum (C,D) and P. vivax (E,F) MSP1₁₉.
FIG. 12.0 a to 12.0 c: Regions of the NOESY (a and b) and TOCSY (c) spectra of P. cynomolgi MSP1₁₉. The chemical shifts in the F1 and F2 dimensions are expressed in parts per million (ppm). (a) and (b): crosspeaks outside the diagonal correspond to dipolar interactions between two hydrogen nuclei of the protein. Signals at 4.67 ppm in (a) correspond to the residual water signal, and probably to dipolar interactions of H nuclei of the protein resonating at the same frequency as the water protons. (c): crosspeaks correspond to through-bond interactions between hydrogen nuclei in an amino acid residue. As in the NOESY spectrum, the peaks at 4.67 ppm arise from the residual water signal and probably from interactions of two nuclei in the same residue. Twenty positive contours are plotted at a vertical scale of 5000 (a, b) or 15000 (c) and a threshold level of 7 (a, b) or 6 (c) using the VNMR 5.3 software.
FIG. 12.1: Regions of the NOESY (a and b) and TOCSY (c) spectra of P. vivax MSP1₁₉. Twenty positive contours are plotted at a vertical scale of 5000 (a), 4000 (b) or 30000 (c) and a threshold level of 7 (a, b) or 6 (c) using the VNMR 5.3 software.
FIG. 12.2: Regions of the NOESY (a and b) and TOCSY (c) spectra of P. falciparum MSP1 ₁₉. Twenty positive contours are plotted at a vertical scale of 5000 (a), 4000 (b) or 10000 (c) and a threshold level of 7 (a, b) or 6 (c) using the VNMR 5.3 software.
FIG. 13: ELISA titration of tertiary monkey anti P. cynomolgi recombinant MSP1 antisera. Plates were coated either with native (N) or reduced, denatured (D) P. cynomolgi MSP1₁₉or MSP1₄₂. OD: Optical density at 492 nm. FIGS. A to F represent respectively, monkeys 426-427-429 (anti-MSP1₁₉) and 428-434-435 (anti-MSP1₄₂) (Perera et al. 1998).
FIG. 14: ELISA analysis of human P. vivax infected donors under reducing and non reducing conditions using immunoaffinity (A) or metallo affinity purified (B) MSP1₁₉coating antigen.
FIG. 15: ELISA titration of murine anti P. cynomolgi MSP1₁₉and MSP1₄₂antisera.
(I) Crystallization and Structure Determination
The expression of recombinant MSP1₁₉in baculovirus is described by Holm et al., 1997 which is hereby incorporated by reference. Modified versions of the constructs described were produced by including a carboxyterminal hexahistidine tag to facilitate purification by metalloaffinity chromatography. Culture supernatants from spinner cultures were harvested and dialyzed extensively against 5 mM Tris-HCI, pH 8.0 at 4° C. The dialysate was adjusted to 20 mM Tris, pH 8.0, 0.1 M NaCI (loading buffer) and passed on columns of “Talon” metalloaffinity resin (Clontech, Palo Alto, U.S.A.) previously equilibrated with loading buffer. The charged resin was washed with loading buffer and eluted with the same buffer supplemented with 100 mM imidazole, pH 8.0. Eluted fractions were pooled and dialyzed against 10 mM potassium phosphate, pH 7.2.
Crystals were grown from a buffer containing 30% PEG 4000 (wt./vol.), 0.2 M ammonium acetate and 0.1 M sodium citrate at pH 5.6. The crystals belong to the trigonal space group P3₂21 (the enantiomorph P3 ₁21 was excluded during the structure analysis; see below) with unit cell dimensions a=b=43.77 Å, c=92.04 Å and with one molecule of MSP1₁₉accommodated in the asymmetric unit.
All X-ray data were recorded with a MAR-Research image plate at the synchrotron beam line DW32, LURE, Orsay. Native diffraction data were obtained at 18° C. from one crystal with dimensions 0.4×0.2×0.2 mm using an X-ray wavelength of 0.970 Å. A total of 139,331 Bragg intensities were measured between 15.0 to 1.8 Å resolution, which reduced to 9980 unique reflections upon merging equivalent reflections, corresponding to a complete data set within this resolution range. A single heavy-atom derivative was prepared by soaking a crystal overnight in a solution of p-chloromercuryphenyl sulphonic acid dissolved at a concentration of 6 mM in the crystallization buffer. Diffraction data were recorded at 18° C. from one crystal, similar in size to that used as native, with an X-ray wavelength of 0.997 Å. The 57,824 individual intensity measurements reduced to a unique data set of 3905 reflections that was complete between the resolution limits of 20.0 and 2.5 Å. The unit cell dimensions of the derivative remained close to those of the native: a=b=43.93 Å, c=92.57 Å.
Isomorphous and anomalous difference Pattersons both gave peaks consistent with a single heavy-atom site. The mercury parameters were refined with the program SHARP (La Fortelle and Bricogne, 1997) using isomorphous and anomalous structure amplitude differences between the resolution limits of 13.0 and 2.5 Å. The figure of merit and phasing power was 0.56 and 1.7, respectively, for the acentric data, and 0.41 and 1.4, respectively, for the centric data. The correct space group enantiomorph was distinguished during phase refinement by solvent flattening with the program SOLOMON (Abrahams and Leslie, 1996); the final overall R-factor was 0.316 for P3 ₁21 and 0.223 for P3 ₂21, thus clearly indicating the latter space group.
The polypeptide chain could be readily traced for the whole of the first domain. By contrast, the second domain presented certain difficulties owing to lack of continuity of the electron density for some regions of the polypeptide chain. An electron density map offering a more facile interpretation of the second domain was subsequently obtained by combining the probability distributions of calculated phases from the traced structure of the first domain with those of SIRAS phases (using the program SIGMMA (Read, 1986)), followed by density modification of the resulting Fourier synthesis (using the program DM (Cowtan, 1994; Bayley, The CCP4 suite, 1994)). The atomic parameters were refined with the program REFMAC using the maximum-likelihood option. The final R-factor and free R-factor were 0.212 and 0.279, respectively, for all reflections included within the resolution limits of 15.0 to 1.8 Å.
The final model of MSP1₁₉includes all residues from the amino terminus, Met1, to His91, the second residue of the carboxyterminal hexahistidine tag, with the exception of the segment from Asp66 to Asn68 located at the extremity of a β-hairpin turn. In spite of continuous electron density being present in the region from Asp66 to Asn68, it was not sufficiently well defined to propose a conformation for the main chain of this segment, suggesting that this part of the structure is flexible. Although there was no unambiguous indication for multiple conformers for the side chains of MSP1₁₉in the final electron density maps, the side chains of Lys50, Lys63 and Glu80 could not be modeled beyond their Cβ atoms because of weak or absent density in these regions. The final model includes 65 solvent molecules. Annex I includes the atomic coordinates which define the P. cynomolgi MSP1₁₉3-dimensional molecular structure and Annexes II and III respectively are those of P. vivax and P. falciparum as determined by homology modeling.
FIG. 11A shows the trace of the alpha-carbon backbone and disulfide bridges (bold lines) of the MSP119 molecule from Plasmodium cynomolgi determined by the solution of its crystal structure. The two epidermal growth factor (EGF)-like domains predicted from the primary structure are clearly visible as well as the characteristic disulfide bonds which are essential for the maintenance of this complex structure. It is important to note that although only 2 of the 3 “classic” EGF disulfide bonds are present in the first EGF domain for the P. cynomolgi and P. vivax moieties, the conformation of this domain does not differ significantly from either the second MSP119 EGF domain or other examples of EGF domains with 3 disulfide bridges.
One unexpected feature of the molecular structure of MSP1₁₉is that the 2 EGF-like domains have close and extensive contacts at their interface which include a number of hydrophobic and polar interactions. An important implication of this observation is that the two EGF domains together form a very defined entity which is likely to be essential for the nature of the active protective principle in recombinant analogs of this molecule. These data suggest that the single EGF domains either alone or combined separately would not display the same required conformation.
The probable structures of the corresponding MSP1₁₉antigens derived from Plasmodium vivax and Plasmodium falciparum (FIGS. 11B and 11C respectively) were determined by molecular modeling based on the replacement of alternative residues present in the latter two species (11/89 for P. vivax and 48/89+ a 4-residue insertion for P. falciparum). It is important to note that these replacements do not put any noticeable steric strains on the experimentally determined P. cynomolgi MSP1₁₉structure and that the interactions at the EGF-domain interface are maintained by invariant residues or conservative substitutions, in spite of several replacements seen particularly for the P. falciparum protein.
It should be emphasized here that, apart from allowing the precise determination of its molecular structure, the crystallization of the MSP119 recombinant protein is prima facie evidence of its absolute purityi its quantitatively reproducible, defined conformation as produced in the baculovirus expression system, and its stability. The fact that an identical preparation of P. cynomolgi MSP119 recombinant protein (as determined by mass spectrometry) was recently shown to confer excellent protection in the P. cynomolgi-toque monkey system described previously (Perera et al., 1998; S. Longacre, I. Holm, L. Perera and S. Handunetti, unpublished data) suggests that this particular conformation is important for protective efficacy.
(II) Nuclear Magnetic Resonance (NMR) “Fingerprint” Spectra
The three baculovirus MSP119 recombinant proteins derived from P. cynomolgi, P. vivax and P. falciparum, each with a carboxyterminal hexahistidine tag, were purified by metalloaffinity chromatography as described above, dialyzed extensively against 10 mM ammonium bicarbonate, and lyophilized. Five mg of lyophilized protein was dissolved in 380 ml of 20 mM deuterated sodium acetate, 10% D2O, pH 4.0, centrifuged 30 min at 20° C. and loaded into Shigemi (Shigemi Inc., Allison Park, Pa.) tubes.
In order to obtain an NMR “fingerprint” of a protein, the later must be pure to at least ca. 95%, as the signals of any proton-containing contaminant present at substantial concentrations would be observed in ¹H NMR spectra. The purity and sequence homogeneity of the MSP1₁₉samples from P. cynomolgi, P. falciparum and P. vivax were assessed by electrospray mass spectrometry. The lyophilized protein was dissolved in water:methanol:formic acid (50:50:10). The sample was introduced in an API 365 triple-quadrupole mass spectrometer (Perkin Elmer-Sciex, Thornill, Canada) at 5 ml/min by means of a syringe pump (Harvard Apparatus, South Natick, Mass.). The device was equipped with an atmospheric pressure ion source used to sample positive ions which were produced by a pneumatically-assisted electrospray interface. The ion spray probe tip was held at 4.5 kV and the orifice voltage was set at 45 V. The mass spectrometer was scanned continuously from m/z 950 to 1600 (P. cynomolgi sample) or from 1050 to 1900 (P. falciparum sample) or from 1150 to 1800 (P. vivax sample) with a scan step of 0.1 and a dwell time per step of 2.0 ms. This resulted in a scan duration of 13.0 s (P. cynomolgi and P. vivax samples) and 17.0 s (P. falciparum sample). Ten scans were averaged for each experiment. Mass calibration of the instrument was accomplished by matching propylene glycol ions to their known reference masses which are stored in the mass calibration table of the spectrometer. Data were collected on a Power Macintosh 8600/200 and processed with the Biotoolbox 2.2 software from Sciex. The reconstructed mass spectra and m/z spectra of MSP1₁₉from P. cynomolgi, P. falciparum and P. vivax are shown in FIGS. 12A and B, 12C and D, and 12E and F respectively. The spectra of all three proteins over a wide range of m/z values indicate that there is mainly a single species with no major contaminants, as all the major peaks correspond to different m/z values of a single species or its adducts. The average mass calculated from the experiment for the P. cynomolgi (10767±1.1 Da), P. vivax (10524.70±0.48) and P. falciparum (11041.1±0.6 Da) proteins correspond, within experimental error, to the expected mass of the oxidized proteins with an N-terminus at MSS (P. cynomolgi and P. vivax ) and ISQ (P. falciparum). The minor peak in the P. vivax sample with an average mass of 10624.54±O0.70 corresponds to the expected mass of the same P. vivax protein with an N-terminus at TMSS. Hence, the protein samples used for NMR experiments are greater than 95% pure as required to obtain an NMR “fingerprint” of the proteins and their conformation.
NMR experiments were carried out on a Varian Unity 500 spectrometer (11.7 T) operating at a proton frequency of 500 MHz and equipped with a 5 mm triple-resonance, z-gradient detection probe. Data were processed on a Sun workstation using the software VNMR 5.3 (Varian Inc., Palo Alto). All 2D proton NMR experiments were acquired in the phase-sensitive mode using the hypercomplex scheme (States et al., 1982). ¹H chemical shifts are referred to DSS (sodium 4,4-dimethyl-silapentane sulfonate) used as an external reference.
NOESY (States et al., 1982) and TOCSY (Griesinger et al., 1988) spectra were acquired at 35.0° C. with 2048 data points in the direct dimension and 256 t₁increments. Forty (P. cynomolgi NOESY), 48 (P. cynomolgi TOCSY), 64 (P. vivax ) or 48 (P. falciparum) transients per t₁increment were accumulated. Solvent suppression was achieved by using the WATERGATE pulse scheme (Piotto et al., 1992). Mixing times in NOESY and TOCSY were 120 and 70 ms, respectively. The spin lock during the mixing time of the TOCSY experiment was obtained by applying the MLEV-17 pulse sequence (Bax and Davis, 1985). Spectra were recorded using a spectral window of 6000 Hz and a recycling delay of 2.2 s (NOESY) or 2.0 s (TOCSY).
2D experiments were transformed using a low pass-filter centered at the transmitter frequency to reduce the intensity of the residual water signal. The width of the filter was 10 Hz and the number of coefficients to calculate the filter shape were 41. Spectra were apodized with shifted square sine bell functions in both dimensions, and baseline corrected in the direct dimension. The first point of every free induction decay (FID) was multiplied by 0.5 prior to Fourier transformation to reduce drift corrections. The resolution in the indirect dimension (F1) was increased by forward linear prediction to 512 points, except for the TOCSY spectrum of the P. vivax sample. Zero filling to 4096 points in the direct dimension (F2) and to 2048 points in the indirect dimension was performed for each spectrum.
The chemical shift of a signal in an NMR spectrum depends on the chemical environment of the hydrogen nucleus that produces the signal. The chemical shift, and thereby the topology or peak pattern of an NMR spectrum, is highly dependent on molecular conformation. The TOCSY experiment gives through-bond connectivities within each amino acid residue of a protein while the NOESY experiment provides dipolar (through-space) connectivities between atoms that are less than 5 Å apart. The intensity of a signal strongly depends on the distance between two atoms, as well as on the dynamics of the molecule. The topology of a NOESY spectrum is thus particularly sensitive to differences in molecular conformation. Together, the TOCSY and NOESY spectra provide a detailed fingerprint of the structure of a given protein in solution.
The relevant regions of the NOESY (a and b) and TOCSY (c) spectra of the recombinant MSP1₁₉from P. cynomolgi are shown in FIG. 12.0. The good dispersion of the signals in both spectra is indicative of a folded protein. The down-field shift of many H resonances (FIG. 12.0 a and c) relative to the random coil range (˜4.1-˜4.8 ppm) and the presence of strong sequential Hi-NHi+1 and weak NH—NH correlations in the NOESY spectrum (FIG. 12.0 b and a, respectively) are all indicative of β-sheet structures and are consistent with the predominantly β-sheet structure determined for MSP1₁₉by crystallography. The corresponding regions of the NOESY and TOCSY spectra of the recombinant MSP1₁₉proteins from P. vivax and P. falciparum are displayed in FIGS. 12.1 and 12.2. As in the case of the P. cynomolgi protein, the spectra of both proteins show good dispersion of chemical shifts, down-field shifted H signals, strong sequential Hi-NHi+1 and weak NH—NH correlations indicating that these proteins also have a substantial amount of β structures. As the proteins from P. cynomolgi and P. falciparum have ca. 50% different residues, a comparison of the NMR spectra of both proteins is not possible. However, the presence of β-sheet structures in P. falciparum MSP1₁₉as evidenced by the NMR spectra, and the relatively high sequence identity of the proteins (ca. 48% for P. cynomolgi and P. falciparum) support their structural homology. The NOESY and TOCSY spectra of the P. cynomolgi. and P. vivax proteins display many crosspeaks as well as crosspeak patterns at the same (or similar) chemical shifts. The similarities that can be observed in the spectra of both proteins, in spite of the existence of 11 different residues, suggest that both proteins have a similar structure. Thus, the NMR data, as well as the high sequence identity of both proteins (ca. 86%), strongly support the similitude of the MSP1₁₉structures from P. cynomolgi and P. vivax species as postulated by homology modeling based on the crystallographic data.
Unlike X-ray crystallography, NMR technology provides a relatively accessible means of assessing if a protein preparation displays a given conformation. Thus, once having defined in detail a conformation which constitutes the active principle of a protective antigen, one can verify by NMR whether competing preparations contain conformationally similar species and/or whether these represent a mixture of more than one conformer (with an estimated 5-10% limit of detection). It should be noted that some variation in the peak position can be expected for independent experiments performed under the same conditions and with the same protein, with an estimated ±0.02 ppm variation in each dimension. Also, as the contour plots presented in FIGS. 12.0, 12.1 and 12.2 are cross-sections of a spectrum which contains three-dimensional peaks as well as noise, and in which peak intensity depends on many experimental and data-processing factors (such as the sensitivity of the probe, the suppression of the water-signal, the phasing of the spectrum and the functions used to apodize the spectrum and correct the baseline) in addition to the nuclei interactions, some variations in the intensity and thereby in the pattern of crosspeaks at a given cross-section level must be expected,. especially for the less intense peaks, the peaks near the diagonal and those close to the water signal (4.67 ppm). Bearing this in mind, the NMR spectra can serve as an identifying “fingerprint” which can be used not only for product quality control, but also to establish distinguishing characteristics of different preparations of the same or similar polypeptides.
(III) ELISA Analysis of Anti-MSP1₁₉and Anti-MSP1₄₂Antibody Reactivity with Native and Reduced Antigen
Recombinant baculovirus Plasmodium vivax and Plasmodium cynomolgi MSP1₁₉and MSP1₄₂was produced and purified by immunoaffinity or metalloaffinity chromatography as described (Holm et al., 1997; see above). The antisera from Macaca sinica (toque) monkeys vaccinated with baculovirus recombinant P. cynomolgi MSP1₁₉or MSP1₄₂was obtained as described by Perera et al. (1998). Anti-P. vivax human antisera were obtained from endemically exposed donors in Thailand.
For the ELISA assay, microtiter plates (NUNC) were coated with the purified MSP1₁₉or MSP1₄₂recombinant protein at 200 ng ml⁻¹overnight at 4° C. After 3 washes with PBS containing 0.1% Tween 20 (PBS-T), the polyclonal human, monkey and mouse antisera were incubated for 2 h at 37° C. Plates were washed and incubated for 1 h at 37° C. with horseradish peroxidase conjugated anti-human or anti-mouse IgG. After 5 washes, 100 μl of freshly prepared 0.2% orthophenylenediamine containing 0.03% hydrogen peroxidase in 0.1 M citrate buffer, pH 5.2 were added to each well. The reaction was stopped by the addition of 50 μl of 3 N HCI and the optical density was measured at 492 nm. For ELISA assays done with reduced coating antigen, the antigen was reduced in situ by incubation in the microtiter plates with 0.2 M Tris-HCI (pH 8.5), 5 mM EDTA, 20 mM dithiothreitol for 2 h at room temperature followed by alkylation for 20 min in the dark with 60 mM iodoacetamide added to the reducing reaction mixture.
FIGS. 13A-F show ELISA titration of tertiary antisera from 6 individual toque monkeys corresponding to 2 groups of 3 animals each vaccinated with either recombinant P. cynomolgi MSP1₁₉or MSP1₄₂(Perera et al. 1998). In each figure a given individual antisera is reacted with either the MSP119 or MSP142 antigen in native (N) or reduced, denatured (D) form. In the 3 antisera from MSP119 vaccinated animals (A-C) virtually all the epitopes in the MSP119 or MSP142 (also containing the MSP119 polypeptide) coating antigens are reduction sensitive since no significant ELISA reactivity is observed when the coating antigens are reduced. These results indicate clearly that the macaque humoral immune response to the recombinant MSP119 polypeptide, whether presented alone or in the context of the larger MSP142 protein, is primarily directed against reduction sensitive conformational epitopes. Since these MSP119 vaccinated monkeys were extremely well protected against a challenge infection of the P. cynomolgi blood stage parasite (Perera et al., 1998), it is thus highly likely that a conformationally intact recombinant MSP119 antigen is crucial for a relevant, protective immune response. The crystallographic results described above indicate that the MSP119 recombinant antigen obtained from this higher order expression system represents a single conformational species which must constitute, at least in part, the active principle in these vaccine preparations.
The antisera from MSP1₄₂vaccinated monkeys (FIGS. 13D-F) show that although all the MSP1₄₂epitopes are not reduction sensitive, those corresponding to the MSP1₁₉antigen are completely sensitive. From the above observations regarding the MSP1₁₉vaccinated animals, it is likely that antibodies specific for the conformationally sensitive epitopes were important for the good protection conferred by the MSP1₄₂antigen in 2 of the 3 vaccinated monkeys (n° 434 and 428; Perera et al., 1998). Nevertheless, the third animal (n° 435), which also had reduction sensitive anti-MSP119 antibodies, was not protected and it is possible that in some cases antibodies specific for non-conformational epitopes in the MSP1₄₂immunogen could interfere with protection, arguing against the use of the MSP1₄₂antigen in vaccine preparations.
FIG. 14 shows the recombinant MSP119 reactivity of antisera (1:100 dilution) from 22 P. vivax infected Thai donors as determined by ELISA analysis using native or reduced antigen. The data from FIGS. 14A and 14B were obtained using immunoaffinity or metalloaffinity purified antigens respectively. In all cases the native antigen is much more reactive than the reduced moiety, indicating that the large majority of MSP1₁₉epitopes recognized by human anti-parasite sera are reduction sensitive. Thus any protective effect, as well as natural boosting, of an MSP119 based malaria vaccine would likely depend on the quantitative presence of such conformational epitopes in the vaccinating immunogen.
For FIG. 15, 50 and 25 mg respectively of recombinant P. cynomolgi MSP1₄₂and MSP1₁₉purified by immunoaffinity chromatography (Holm et al., 1997) were inoculated subcutaneously three times into mice at 10 to 14 day intervals in the absence of adjuvant. A fourth and final boost was administered 8 months after the third injection. The dependence of the humoral immune response on reduction sensitive conformational epitopes present in the recombinant antigens was analyzed with pooled anti-MSP1₄₂or anti-MSP1₁₉sera using the ELISA technique with native or reduced antigen as described above.
The results show that while the anti-MSP1₄₂response is partially dependent on reduction sensitive conformational epitopes in the p42 moiety after 3 closely spaced injections (Third bleed), this dependence appears to be significantly reduced in the antisera obtained by boosting after an 8 month interval (Fourth bleed). Since the MSP1₄₂antisera reactivity directed against MSP1₁₉is completely dependent on reduction sensitive epitopes, the long term memory response to the MSP1₄₂antigen in the absence of adjuvant clearly favors non-conformational, non-MSP1₁₉epitopes. In contrast, the anti-MSP1₁₉sera are completely dependent on reduction sensitive conformational epitopes regardless of the short or long term memory tontext. These observations, although preliminary, suggest that the N-terminal, non-p19 epitopes in the MSP1₄₂might be favored by natural boosting to the detriment of the conformational MSP119 epitopes which appear to be important for protection (see above). This would also discourage the use of the MSP1₄₂for vaccination under conditions where natural boosting by antigens containing MSP1₄₂specific epitopes is expected.

REFERENCES

(1) Holder, J. A. et al. (1982) Biosynthesis and processing of a Plasmodium falciparum schizont antigen recognized by immune serum and a monoclonal antibody”. J. Exp. Med. 156:1528-1538.
(2) Howard, R. et al. (1984) “Localization ofthe major Plasmodium falciparum glycoprotein on the surface of mature intracellulartrophozoites and schizonts”. Mol. Biochem. Parasitol. 11: 349-362,
(3) Pirson, P. et al. (1985) “Characterization with monoclonal antibodies of a surface antigen of Plasmodium falciparum merozoites”. J. Immunol. 134 1946-1951.
(4) Aley, S. B. et al. (1987) “Plasmodium vivax: Exoerythrocytic schizonts recognized by monoclonal antibodies against blood-stage schizonts”. Exp. Parasitol. 64:188-194.
(5) Holder, A. A. (1988) “The precursor to major merozoite surface antigen: structure and role in immunity”. Prog. Allergy 41: 72-97.
(6) Cooper, J. A. (1993) “Merozoite surface antigen-1 of Plasmodium”. Parasitol. Today 9: 50-54.
(7) Holder, A. A., et al. (1987) “Processing of the precursor to the major merozoite antigens of Plasmodium falciparum” Parasitology 94:199-208.
(8) Lyon, J. A. et al. (1986) “Epitope map and processing scheme for the 195 000-dalton surface glycoprotein of Plasmodium falciparum merozoites deduced from cloned overlapping segments of the gene”. Proc. Natl. Acad. Sci. USA 83: 2989-2993.
(9) Blackman, M. J. et al. (1992) “Secondary processing of the Plasmodium falciparum merozoite surface protein-1 (MSP1) by calcium-dependent membrane-bound serine protease: shedding of MSP133 as a noncovalently associated complex with other fragments of the MSOP1”. Mol, Biochem. Parasitol. 50: 307-316.
(10) Haldar, K., et al. (1985) “Acylation of a Plasmodium falciparum merozoite surface antigen via sn-1,2-diacyl glycerol.”. J. Biol. Chem. 260: 4969-4974.
(11) Braun Breton, C. et al. (1990) “Glycolipid anchorage of Plasmodium falciparum surface antigens”. Res. Immunol. 141: 743-755.
(12) Kumar, S. et al. (1995)“Immunogenicity and in vivo Efficacy of Recombinant Plasmodium falciparum Merozoite surface protein-1 in Aotus Monkeys”. Molecular Medicine, Vol. 1, 3: 325-332.
(13) Longacre, S. et al. (1994) “Plasmodium vivax merozoite surface protein 1 C-terminal recombinant proteins in baculovirus”. Mol. Biochem. Parasitol. 64:191-205.
(14) McBride, J. S. et al. (1987) “Fragments of the polymorphic Mr 185 000 glycoprotein from the surface of isolated Plasmodium falciparum merozoites form an antigenic complex”. Mol. Biochem. Parasitol. 23: 71-84.
(15) Blackman, M. J. et al. (1990) “A single fragment of a malaria merozoite surface protein remains on the parasite during red cell invasion and is the target of invasion-inhibiting antibodies”. J. Exp. Med. 172: 379-382.
(16) Kaslow, D. C. et al. (1994) “Expression and antigenicity of Plasmodium falciparum major merozoite surface protein (MSP119) variants secreted from Saccharomyces cerevisiae”. Mol. biochem. Parasitol. 63; 283-289.
(17) Chang, S. P., et al. (1992) “A carboxyl-terminal fragment of Plasmodium falciparum gp195 expressed by a recombinant baculovirus induces antibodies that completely inhibit parasite growth”. J. Immunol. 149: 548-555.
(18) Longacre, S. (1995) “The Plasmodium cynomolgi merozoite surface protein 1 C-terminal sequence and its homologies with other Plasmodium species”. Mol. Biochem. Parasitol. 74:105-111.
(19) Del Portillo, H. A., et al. (1990) “Primary structure of the merozoite surface antigen 1 of Plasmodium vivax reveals sequences conserved between different Plasmodium species”. Proc. Natl. Acad. Sci. USA 88 :4030-4034.
(20) Gibson, H. L., et al. (1992) “Structure and expression of the gene for Pv200, a major blood-stage surface antigen of Plasmodium vivax”. Mol. biochem. Parasitol. 50: 325-334.
(21) Dissanaike, A. S. et al. (1965) “Two new malaria parasites, Plasmodium cynomolgi ceylonensis sub sp. nov. and Plasmodium fragile sp. nov. from monkeys in Ceylon”. Ceylon Journal of Medical Science 14:1-9.
(22) Cochrane, A. H., et al. (1986) “Further studies on the antigenic diversity of the circumsporozoite proteins of the Plasmodium cynomolgi complex”. Am. J. Trop. Med. Hyg. 35: 479-487.
(23) Naotunne, T. de S., et al. (1990) “Plasmodium cynomolgi: serum-mediated blocking and enhancement of infectivity to mosquitos during infections in the natural host, Macaca sinica”. Exp. Parasitol, 71, 305-313.
(24) Ihalamulla, R. L. et al. (1987) “Plasmodium vivax: isolation of mature asexual stages and gametocytes from infected human blood by colloidal silica (Percoll) gradient centrifugation”. Trans. R. Soc. Trop. Med. Hyg. 81:25-28.
(25) Kimura, E., et al. (1990) “Genetic diversity in the major merozoite surface antigen of Plasmodium falciparum: high prevalence of a third polymorphic form detected in strains derived in strains derived from malaria patients”. Gene 91: 57-62.
(26) Heidrich, H.-G., et al. (1989) “The N-terminal amino acid sequences of the Plasmodium falciparum (FCB1) merozoite surface antigen of 42 and 36 kilodalton, both derived from the 185-195 kilodalton precursor”. Mol. Biochem. Parasitol. 34:147-154.
(27) Blackman, M. J., et al. (1991) “Proteolytic processing of the Plasmodium falciparum merozoite surface protein-1 produces a membrane-bound fragment containing two epidermal growth factor-like domains”. Mol. Biochem. Parasitol. 49: 29-34.
(28) Adams, J. M. et al. (1992) “A family of erythrocyte binding proteins of malaria parasites”. Proc. Natl. Acad. Sci, 89:7085-7089.
(29) Sim B. K. L (1995) “EBA-175: An erythrocyte-binding ligand of Plasmodium falciparum”. Parasitology Today, vol.II, no 6:213-217.
(30) Sim B. K. L. (1994) “Receptor and ligand domains for invasion of erythrocytes by Plasmodium falciparum”. Science, 264:1941-1944.
(31) Davies, A. et al. (1993), Biochem J. 295 (Pt3): 889-896. “Expression of the glycosylphosphatidylinositol-linked complement-inhibiting protein CD59 antigen in insect cells using a baculovirus vector”.
(32) Haziot A. et al. (1994) J. Immunol. 152: 5868. “Recombinant soluble CD14 Inhibits LPS-lnduced Tumor Necrosis Factor & Production by Cells in Whole Blood”.
(33) Chang, S. P. et al., (1988) “Plasmodium falciparum: gene structure and hydropathy profile of the major merozoite surface antigen (gp195) of the Uganda-Palo Alto isolate. Exp. Parasitol. 67:1-11.
(34) Holder, A. S. et al. (1985) “Primary structure of the precursor to the three major surface antigens of Plasmodium falciparum merozoites”, Nature 317: 270-273
(35) G. Bourdoiseau et al. (May 1995) “Les babesioses bovines”, Le point vétérinaire, vol. 27, n-168.
(36) P. Bourdeau et al. (May 1995) “La babésiose canine á Babesia canis”, Le point vétérinaire, vol. 27, n˜168.
(37) C. Soule (May 1995) “Les babésioses équines”, Le point vétérinaire, vol. 27 n˜168.
(38) T. P. M. Schetters et al. (1995) “Vaccines against Babesiosis using Soluble Parasite Antigens”, Parasitology Today, vol. 11, no 12.
(39) P. A. Burghaus et al. (1996) “Immunization of Aotus nancymai with Recombinant C Terminus of Plasmodlum falciparum Merozoite Surface Protein 1 in Liposomes and Alum Adjuvant Does Not Induce Protection against a Challenge Infection”, Infection and Immunity, 64:3614-3619.
(40) S. P. Chang, et al. (1996) A Recombinant Baculovirus 42-Kilodalton C-Terminal Fragment of Plasmodium falciparum Merozoite Surface Protein 1 Protects Aotus Monkeys against Malaria”, Infection and Immunity, 64: 253-261.
(41) L. H. Miller et al. (1997) “The Need for Assays Predictive of Protection in Development of Malaria Bloodstage Vaccines”, Parasitology Today, vol. 13, no 2:46-47
(42) Abrahams, J. P. and Leslie, A. G. W. (1996) Methods used in the structure determination of bovine mitochondrial F1 ATPase. Acta Crystallogr. D52, 30-42.
(43) Bax, A. and Davis, D. G. (1985). MLEV-17-based two-dimensional homonuclear magnetization transfer spectroscopy. J. Magn. Reson. 65, 355-360.
(44) Cowtan, K. (1994) ‘dm’, An automated procedure for phase improvement by density modification. Joint CCP4 and ESF-EACBM newsletter on protein crystallography, 31, 34-38.
(45) De La Fortelle, E. and Bricogne, G. (1997) Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methods. Methods in Enzvmol. 276, 272-492.
(46) Griesinger, C., Otting, G., Wüthrich., K. and Ernst, R. R. (1988). Clean Tocsy for 1H spin system identification in macromolecules. J. Am. Chem. Soc. 110, 7870-7872.
(47) Holm, I., Nato, N., Mendis, K. N. and Longacre, S. (1997) Characterization of C-terminal merozoite surface protein-1 baculovirus recombinant proteins from Plasmodium vivax and Plasmodium cynomolgi as recognized by the natural anti-parasite immune response. Molec. Biochem. Parasitol. 89, 313-319.
(48) Longacre, S. (1995) The Plasmodium cynomolgi merozoite surface protein 1 C-terminal sequence and its homologies with other Plasmodium species. Molec. Biochem. Parasitol. 74, 105-111.
(49) Perera, K., Handunnetti, S., Holm, I., Longacre, S. and Mendis, K. (1998) Baculovirus merozoite surface protein 1 C-terminal recombinant antigens are highly protective in a natural primate model for human Plasmodium vivax malaria. Infect. Immun. 66, 1500-1506.
(50) Piotto, M., Saudek, V. and Sklenár, V. (1992). Gradient-tailored excitation for single-quantum NMR spectroscopy of aqueous solutions. J. Biomol. NMR. 2, 661-665.
(51) Read, R. J. (1986) Improved Fourier coefficients for maps using phases from partial structures with errors. Acta Crystallogr. A42, 140-149.
(52) Bayley, S. (1994). The CCP4 Suite: Programs for protein crystallography. Acta Crvstallogr. D50, 760-763.
(53) States, D. J. Haberkorn, R. A. and Ruben, D. J. (1982) A two dimensional nuclear Overhauser experiment with pure absorption phase in four quadrants. J. Magn. Reson. 48, 286-292.

Claims

1. A recombinant protein whose essential constituent polypeptide sequence comprises:

a 19 kilodalton (p19) C-terminal fragment of the surface protein 1 of the merozoite form (MSP-1 protein) of a Plasmodium type parasite other than Plasmodium vivax which is infectious in man, the C-terminal fragment remaining normally anchored to the surface of the surface of the parasite at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;

or that of a portion of that fragment which is also capable of inducing an immune response which can inhibit an in vivo parasitemia in a host infected with such parasite;

or that of a peptide which is capable of inducing a cellular and/or humoral immunological response equivalent to that produced by said p19 fragment or said portion of that fragment; and

wherein said recombinant protein comprises conformational epitopes which are unstable in a reducing medium and which constitute the majority of the epitopes recognized by human antisera formed against the corresponding Plasmodium.

2. The recombinant protein of claim 1 which is not recognized by said human antisera when it is in the reduced form.

3. The recombinant protein of claim 2 which is substantially free of any form of a recombinant protein having the same sequence of amino acids, yet in which the conformational epitopes which are unstable in a reducing medium and which constitute the majority of the epitopes recognized by human antisera formed against the corresponding Plasmodium are not in a conformational form as defined by:

(a) the atomic coordinates as detailed in Annexes I, II or II, obtained by crystallography; and

(b) the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

4. The recombinant protein of claim 2 in which at least 20% contains an epitope in a conformational form as defined by the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

5. The recombinant protein of claim 2 or claim 3 which is in a pure state.

6. A recombinant protein of claim 1, which is recognized by human antisera formed against the corresponding Plasmodium or against a homologous Plasmodium when it is in its non reduced state or in a reduced non irreversible state, but is not recognized or is only recognized to a slight extent by these same antisera when it is irreversibly reduced.

7. The recombinant protein of claim 5 which is defined by the crystallized form with the atomic coordinates as detailed in Annexes I, II or III, and the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

8. The recombinant protein of claim 5 which has a very high degree of purity as determined by electrospray mass spectrometry.

9. The recombinant protein of claim 5 which elicits a long term memory response directed in a substantially specific manner against said conformational epitopes in animals to which they are administered.

10. The protein of claim 1 which inhibits the reactivity of an immune antiserum against p42 produced from the same MSP-1 protein and is itself only partially inhibited by an immune antiserum produced against p42.

11. The recombinant of claim 1 which is essentially free of any polypeptide having a sequence of amino acids in the C-terminal region of p33 (33 kDa N-terminal fragment) resulting from the natural cleavage of the p42 protein of the same MSP1 protein.

12. The recombinant protein of claim 1 which comprises, upstream of the polypeptide sequence of p19, a polypeptide region whose sequence is the C-terminal region of p33 resulting from the cleavage of p42 of the same MSP-1 protein, wherein said polypeptide region contains less than 50 amino acid residues.

13. The recombinant protein of claim 11 wherein said polypeptide region contains less than 10 amino acid residues.

14. The recombinant protein of claim 1 which comprises, upstream of the polypeptide sequence of p19, a polypeptide region whose sequence is the C-terminal region of the p33 resulting from the cleavage of p42 of the same MSP-1 protein, wherein said C-terminal portion is restricted to that part which is substantially conserved in P. falciparum and P. vivax.

15. The recombinant protein of claim 1 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

16. The recombinant protein of claim 2 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

17. The recombinant protein of claim 3 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

18. The recombinant protein of claim 4 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

19. The recombinant protein of claim 5 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

20. The recombinant protein of claim 6 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

21. The recombinant protein of claim 7 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

22. The recombinant protein of claim 8 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

23. The recombinant protein of claim 9 wherein the sequence of amino acids of said essential constituent polypeptide encompasses the two sequences of the two EGF regions of the p19 protein.

24. The recombinant protein according to claim 1, wherein the constituent polypeptide carries a glycosylphosphatidylinositol (GPI) group of the type which enables the p19 fragment to anchor to the membrane of a eukaryotic cell infected with the MSP-1 protein.

25. The recombinant protein of claim 1, wherein the constituent polypeptide is free of the sequence of amino acids in the hydrophobic C-terminal portion of the p19 which intervenes in the induction of an anchoring of said p19 to the cell membrane of a host infected with a Plasmodium type parasite.

26. The recombinant protein according to claim 24, which is hydrosoluble.

27. The recombinant protein of claim 1 which comprises the sequences of amino acids of the p19 of the MSP-1 protein of Plasmodium falciparum.

28. The recombinant protein of claim 1 which comprises the sequence of amino acids of the MSP-1 protein of Plasmodium cynomolgi.

29. An oligomer of the recombinant protein of claim 1.

30. The oligomer of claim 29, which comprises from 2 to 50 monomer units of the sequence of said recombinant protein.

31. The recombinant protein of claim 1, which is conjugated to a carrier molecule for use in the production of vaccines.

32. A vaccination composition against a Plasmodium type parasite which is infectious in man, containing as an active principle a recombinant protein whose essential constituent polypeptide sequence comprises:

a 19 kilodalton (p19) C-terminal fragment of the surface protein 1 of the merozoite form (MSP-1 protein) of a Plasmodium type parasite which is infectious in man, said C-terminal fragment remaining normally anchored to the surface of the parasite at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;

said recombinant protein further comprising conformational epitopes which are unstable in a reducing medium and which constitute the majority of the epitopes recognized by human antisera formed against the corresponding Plasmodium.

33. The vaccinating composition of claim 32, wherein said recombinant protein is not recognized by said human antisera when it is in the reduced form.

34. The vaccinating composition of claim 32, wherein said recombinant protein is substantially free of any form of said recombinant protein having the same sequence of amino acids, yet in which the conformational epitopes which are unstable in a reducing medium and which constitute the majority of the epitopes recognized by human antisera formed against the corresponding Plasmodium are not in the conformational form as defined by:

(b) the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

35. The vaccinating composition of claim 32, wherein said recombinant protein is in a pure state.

36. The vaccinating composition of claim 32, wherein said recombinant protein is recognized by human antisera formed against the corresponding Plasmodium or against a homologous Plasmodium when it is in its non reduced state or in a reduced non irreversible state, but is not recognized or is only recognized to a slight extent by these same antiserums when it is irreversibly reduced.

37. The vaccinating composition of claim 32, wherein said recombinant protein is defined by the crystallized form with the atomic coordinates as detailed in Annexes I, II or Ill, and the NMR fingerprints as illustrated in FIGS. 12.0 a to 12.2 c.

38. The vaccinating composition of claim 32, wherein said recombinant protein has a very high degree of purity as determined by electrospray mass spectroscopy.

39. The vaccinating composition of claim 32, wherein said recombinant protein elicits a long term memory response directed in a substantially specific manner against said conformational epitopes in animals to which they are administered.

40. The vaccinating composition of claim 32, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

41. The vaccinating composition of claim 33, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

42. The vaccinating composition of claim 34, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

43. The vaccinating composition of claim 35, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

44. The vaccinating composition of claim 36, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

45. The vaccinating composition of claim 37, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

46. The vaccinating composition of claim 38, wherein the sequence of amino acids of said essential constituent polypeptide encompasses the sequences of the two EGF regions of the p19 protein.

47. The vaccinating composition of claim 32 wherein the essential constituent polypeptide of said recombinant protein comprises the sequence of the p19 Plasmodium falciparum.

48. The vaccinating composition of claim 32 wherein the essential constituent polypeptide of said recombinant protein comprises the sequence of the p19 Plasmodium vivax.

49. An antibody which specifically recognizing the p19 of a MSP-1 protein of the merozoite form of a Plasmodium type parasite which is infectious in man other than Plasmodium vivax and which does not recognize Plasmodium vivax.

50. The antibody of claim 49, which is monoclonal.

51. The monoclonal antibody of claim 50, which specifically recognizes the p19 of P. falciparum.

52. The monoclonal antibody of claim 50, which specifically recognizes the p19 of P. vivax.

53. A recombinant baculovirus type modified vector containing, under the control of a promoter contained in the vector and able to be recognized by cells transfectable by said vector, a first nucleotide sequence coding for a signal peptide and a second nucleotide sequence downstream of the first nucleotide sequence, also under the control of said promoter, wherein said first nucleotide sequence authorizes the expression of said second nucleotide sequence in a baculovirus system and wherein said second nucleotide sequence codes for one of the following peptide sequences:

a 19 kilodalton (P19) C-terminal fragment of the surface protein 1 of the merozoite form (MSP-1 protein) of a Plasmodium type parasite which is infectious in man, the C-terminal fragment remaining normally anchored to the surface of the parasite at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;

or of a portion of that peptide fragment provided that the expression product from the second sequence in a baculovirus system is also capable of inducing an immune response which can inhibit an in vivo parasitemia in a host infected with said parasite;

or of a peptide which is capable of inducing a cellular or humoral immunological response equivalent to that produced by said peptide fragment P19 or said peptide fragment portion; and

wherein said second nucleotide sequence has a G and C content in the range of from 40% to 60% of the totality of nucleotides from which said second nucleotide sequence is constituted.

54. The vector of claim 53, wherein said second nucleotide sequence has a G and C content in the range of at least 50% of the totality of nucleotides from which said second nucleotide sequence is constituted.

55. The vector of claim 53, wherein the second nucleotide sequence is a synthetic sequence.

56. The vector of claim 53, wherein the first nucleotide sequence codes for a signal peptide from Plasmodium vivax or Plasmodium falciparum normally associated with the Plasmodium MSP-1 protein.

57. The vector of claim 53, which consists of a modified baculovirus.

58. An organism transfected by the vector of claim 53.

59. A synthetic DNA containing a first nucleotide sequence including a portion which codes for the peptide sequence:

of a 19 kilodalton (p19) C-terminal fragment of the surface protein 1 of the merozoite form (MSP-1 protein) of Plasmodium falciparum, said C-terminal fragment remaining normally anchored to the surface of the parasite at the end of its penetration phase into human erythrocytes in the event of an infectious cycle;

or of a portion of that peptide fragment provided that the expression product of said DNA in a baculovirus system is also capable of inducing an immune response which can inhibit an in vivo parasitemia in a host infected with such parasite;

or of a peptide capable of inducing a cellular or humoral type immunological response equivalent to that produced by said p19 peptide fragment or said portion of that fragment or both; and

wherein said nucleotide sequence has a G and C nucleotide content in the range of from 40% to 60% of the totality of nucleotides from which said synthetic DNA is constituted.

60. The synthetic DNA of claim 59, wherein said nucleotide sequence has a G and C content of at least 50% of the totality of nucleotides from which said synthetic DNA is constituted.

61. The synthetic DNA sequence of claim 59, which further comprises upstream from said first nucleotide sequence, a second nucleotide sequence coding for a signal peptide normally associated with a Plasmodium MSP-1 protein which is homologous or heterologous relative to said first nucleotide sequence.

62. The synthetic DNA sequence of claim 61, wherein the signal sequence originates from P. vivax.

63. A baculovirus type selected from the group comprising:

a virus deposited at the CNCM [Collection Nationale de Cultures de Microorganismes; National Collection of Microorganism Cultures] with registration number I-1659;

a virus deposited at the CNCM with registration number I-1660;

a virus deposited at the CNCM with registration number I-1661;

a virus deposited at the CNCM with registration number I-1662;

a virus deposited at the CNCM with registration number I-1663; and

a virus deposited at the CNCM with registration number I-2041.

64. A hybridoma secreting the monoclonal antibodies of claim 50.

65. A hybridoma secreting the monoclonal antibodies of claim 51.

66. A hybridoma which secretes the monoclonal antibody of claim 50, which has been deposited at the CNCM, (Paris, France) with registration number I-1846, on the 14^thof Feb. 1997.

67. A vaccine composition according to claim 32, which comprises a mixture of nucleotide sequences selected from the mixtures of recombinant proteins comprising the sequences of:

P. falciparum p19 and P. vivax p19;

P. falciparum p19 and P. falciparum p42;

P. vivax p19 and P. vivax p42;

P. falciparum p19 and P. faiciparum p42; and

P. vivax p19 and P. vivax p42.