Identification and characterization of the protective hepatocyte erythrocyte protein 17 kDa gene of Plasmodium yoelii, homolog of Plasmodium falciparum exported protein 1.

We recently reported the discovery of a 17-kDa Plasmodium yoelii protein expressed in infected hepatocytes and erythrocytes, P. yoelii hepatocyte erythrocyte protein 17 (PyHEP17), and have demonstrated that this protein is a target of protective antibodies and T cells. Here, we report the identification and characterization of the gene encoding this protein and reveal that it is composed of two exons. Immunization of mice with PyHEP17 plasmid DNA induces antibodies, cytotoxic T lymphocytes, and protective immunity directed against the infected hepatocyte. Based on extensive sequence homology, expression pattern, and antigenic cross-reactivity, the Plasmodium falciparum homolog of PyHEP17 is identified as the protein known as exported protein-1 (PfExp-1), also called antigen 5.1, circumsporozoite related antigen, or QF116. Identity between PyHEP17 and PfExp-1 is 37% at the amino acid level (60/161 residues), mapping primarily to two regions within the second exon of 73% (16/22 residues) and 71% (25/35 residues) identity. On this basis, PfExp-1 is proposed as an important component of pre-erythrocytic human malaria vaccines.

antibody (mAb) 1 Navy yoelii liver stage 3 (NYLS3) which specifically eliminates liver-stage parasites from in vitro culture and delays the onset and reduces the density of blood-stage parasitemia in vivo (1). A DNA vaccine containing the cDNA encoding 67.5% of PyHEP17 (complete exon 1 and 57% of exon 2) protects three of five inbred strains of mice differing in H-2 haplotype and genetic backgrounds, as well as 40 -50% of outbred mice, against P. yoelii sporozoite but not blood-stage challenge (2). This protective immunity is directed against the parasite developing within hepatocytes and is completely abrogated by in vivo depletion of CD8 ϩ T cells, treatment with anti-interferon-␥ mAb, or inhibition of nitric oxide production (2). These data commend the potential of the Plasmodium falciparum homolog of PyHEP17 as a target antigen for malaria vaccine development.
Here, we describe the cloning and characterization of the gene encoding PyHEP17 and identify the P. falciparum homolog of PyHEP17 as the antigen known as P. falciparum exported protein-1 (PfExp-1) (3), also named antigen 5.1 (4), circumsporozoite related antigen (5), or QF116 antigen (6). In addition, we present data pertaining to the immunogenicity of both PyHEP17 and PfExp-1.

EXPERIMENTAL PROCEDURES
Preparation of Parasite DNA-P. yoelii 17X(NL) (nonlethal strain, clone 1.1) was maintained by alternating passage of the parasites in Anopheles stephensi mosquitoes and CD-1 mice (Charles River Laboratories, Wilmington, MA). P. yoelii parasite genomic DNA was prepared from BALB/c mouse parasitized erythrocytes, collected when the parasitemia was approximately 25%, by saponin lysis and cesium chloride gradient centrifugation as described by Wortman et al. (7).
Construction of Genomic Expression Library and Immunoselection-Genomic DNA fragments sized 0.5-2.0 kilobase pairs were generated by partial DNase I (Boehringer Mannheim) digestion and a P. yoelii gt11 genomic expression library was constructed using commercial EcoRI linkers, gt11 arms, and packing extracts (Promega, Madison, WI). The library was screened for antigen-expressing clones using the NYLS3 mAb (1) as described previously (8).
DNA Sequencing of Genomic DNA-Plasmodial DNA from gt11 phage immunoreactive clones was prepared using standard methods (9). The derived genomic DNA sequence of the plasmodial DNA was determined from both strands with single-stranded templates generated via M13 clones, using the Sequenase version 2.0 DNA Sequencing kit (Sequenase, U. S. Biochemical Corp.) under conditions described by the manufacturer, with both forward and reverse M13 primers (Stratagene, La Jolla, CA).
Northern Blotting-Parasite RNA extracted from blood-stage para-sites was separated on a denaturing formaldehyde gel, along with ribosomal RNA molecular weight standards, and transferred to nitrocellulose (Schleicher & Schuell) using standard procedures (9). Hybridization using a 32 P nick-translated (9) partial DNA fragment (LisaEx1/ Ex2, nucleotides 769 -993) was carried out at 42°C in the presence of formamide. Filters were washed twice in 1 ϫ SSC and 0.5% SDS and twice in 0.25 ϫ SSC and 0.1% SDS, each for 15 min at room temperature, before autoradiography. Southern Blotting-Genomic DNA was digested with restriction endonucleases, or not digested, and fractionated by electrophoresis on a 0.8% agarose gel. The DNA was transferred to nitrocellulose (Schleicher & Schuell) using standard procedures (9) and prepared for hybridization by the method of Southern (10). Hybridization to a 32 P nicktranslated (9) partial DNA fragment (HEP17Ex1.2, nucleotides 378 -993) was performed at 42°C overnight. Filters were then washed twice in 0.1 ϫ SSC and 0.1% (w/v) SDS for 45 min at 50°C before autoradiography.
Preparation of Parasite RNA for Reverse Transcription-Polymerase Chain Reaction (RT-PCR)-P. yoelii 17X(NL) blood-stage parasite total RNA was prepared from BALB/C mouse parasitized erythrocytes using the guanidine isothiocyanate-based lysis, cesium chloride gradient technique (11). Oligo(dT)-primed first-strand cDNA was synthesized from DNase I-treated (Boehringer Mannheim) parasite RNA using the SuperScript Preamplification System for first-strand cDNA synthesis (Life Technologies, Inc., Grand Island, NY) with an oligo(dT) primer.
RT-PCR Amplification of PyHEP17 cDNA-Oligonucleotide primers predicted to amplify a partial and a complete PyHEP17 cDNA sequence were based on the derived genomic DNA sequence. RT-PCR amplification was performed using oligo(dT)-primed first-strand cDNA with the following oligonucleotide primers (Integrated DNA Technologies, Inc., Coralville, IA): forward HEP5.1, 5Ј-GGG TCG ACA TGA AAA TCA ATA  TAG CT-3Ј or forward Lisa1, 5Ј-GGG AAT TCA TGA AAA TCA ATA  TAG CT-3Ј; reverse HEP2.1, 5Ј-GGA GAT CTT TAT ACT TCT CTG  AAC TTA GAA GC-3Ј or reverse HEP3.1, 5Ј-GGA GAT CTC ATA GTA  TAT TTC TAA G-3Ј. The forward and reverse primers incorporated SalI or EcoRI and BglII restriction enzyme sites, respectively, in order to facilitate directional cloning. Amplifications were for 30 cycles with 200 ng of cDNA, 20 pmol each primer, 0.2 mM each deoxynucleotide, and 0.2 units of Taq polymerase (Boehringer Mannheim). Reaction buffer was as described by Boehringer Mannheim. PCR conditions were established as: 94°C, 60 s; 50°C, 90 s; 72°C, 90 s, using a DNA Thermal Cycler (Perkin Elmer, Norwalk, CT). Aerosol resistant pipette tips (Molecular BioProducts, San Diego, CA) were used to measure and aliquot all PCR reagents, and a negative control reaction without DNA template was run in parallel for each application.
DNA Sequencing of RT-PCR and cDNA Clones-DNA was purified by standard alkaline lysis followed by cesium chloride gradient centrifugation (9) or by using the QIAGEN Plasmid Mini Kit (QIAGEN Inc., Chatsworth, CA). The cDNA sequence of the RT-PCR and cDNA clones was determined from both strands using alkaline-denatured doublestranded templates (9) and the Sequenase version 2.0 DNA Sequencing kit under conditions described by the manufacturer. Sequencing primers used included commercially available (Stratagene) M13 and pUC sequencing primers, primers mapping to the vector DNA sequence adjacent to cloning sites, and PyHEP17-specific primers.
Construction of a P. yoelii cDNA Library and Isolation of PyHEP17 cDNA-P. yoelii 17X(NL) parasite total RNA was prepared from parasites isolated by acid lysis of BALB/c mouse parasitized erythrocytes (12), followed by TRIzol reagent (Life Technologies, Inc.) for total RNA isolation, after removal of leukocytes with the RCXL1 High Efficiency Leukocyte Removal Filter (Pall Biomedical Products Co., East Hills, NY). Then, mRNA was prepared from total RNA using the PolyATtract mRNA Isolation System (Promega Corp.) with biotinylated oligo(dT). cDNA was synthesized using the SuperScript Lambda System for cDNA Synthesis and Cloning (Life Technologies, Inc.) with gt22A NotI-SalI arms. All procedures were carried out as described by the manufacturer. Recombinant clones were screened for the presence of the insert by plaque hybridization to a 32 P-labeled HEP17Ex1.2 DNA probe (nucleotides 378 -993).
In Vitro Expression of PyHEP17 cDNA-Expression of the PyHEP17Ex1.2 and PyHEP17(1ϩ2) plasmid DNA constructs was assessed in vitro following transfection of COS-7 SV40 transformed cells (American Type Culture Collection, CRL 1651) with 10 g of plasmid DNA using the calcium phosphate coprecipitation technique (15). mRNA was extracted from the transfected COS cells using the Micro FastTrack mRNA isolation kit (In Vitrogen, San Diego, CA). RT-PCR was then performed using each of oligo(dT), random hexamer, and specific oligonucleotide primed first-strand cDNA as the template (Su-perScript Preamplification System for first-strand cDNA synthesis, Life Technologies, Inc.), after pre-treatment with DNase (Boehringer Mannheim). In addition, sera from mice immunized with these plasmid DNA constructs recognized the transfected COS cells, as demonstrated by IFAT (data not presented).
In Vivo Protection-Female 6 -8-week-old BALB/cByJ, A/J, B10.BR, B10.Q, and C57BL/6 mice (The Jackson Laboratory) were immunized 3 times at 3-week intervals in each tibialis anterior muscle (intramuscularly) or intradermally in the tail with a total of 100 g of PyHEP17Ex1.2 or PyHEP(1ϩ2) DNA in 100 l of saline, or unmodified nkCMVint and VR1012 plasmids. P. yoelii sporozoites were obtained 14 days after an infectious blood meal by hand-dissection of P. yoelii 17X NL (clone 1.1) infected mosquito glands in M199 medium containing 5% normal mouse serum (Rockland, Gilbertsville, PA) and diluted to a final concentration of 100 infectious sporozoites per 0.2-ml volume. Two weeks after the third immunization, mice were challenged by tail-vein injection with 100 infectious sporozoites (approximately 50 ID 50 values). Giemsa-stained blood smears were examined on days 5-14, up to 50 oil-immersion fields being scanned for parasites. Protection was defined as the complete absence of blood-stage parasitemia.

RESULTS AND DISCUSSION
Isolation and Characterization of PyHEP17 Gene-A number of immunoreactive clones in a gt11 genomic DNA expression library were recognized by the mAb NYLS3. The antigen recognized by this mAb has been designated PyHEP17 (1). The gt11 clone with the longest insert was selected for further characterization. The genomic DNA sequence (Fig. 1) revealed two major open reading frames (nucleotides 378 -498 and 774-1160) separated by a putative intron (nucleotides 499 -773) which included the appropriate consensus splicing site signal AG/gtaag. . . . tag/NN (19).
Striking amino acid sequence similarity was found between one of these opening reading frames and a region of the peptide sequence of the antigen known as P. falciparum exported protein-1 (PfExp-1) (3-6). PfExp-1 is a 23-kDa liver-stage (20) and blood-stage antigen secreted by the parasite into the host cell, accumulating at the parasitophorous vacuole membrane and within vesicles in the infected red cell cytoplasm. This pattern of intracellular localization is identical to that of PyHEP17 (1). PfExp-1 is characterized by intragenic non-coding regions and consists of three exons with a total size of 161 amino acids (486 bp).
Between bases 779 and 994 of the PyHEP17 genomic DNA sequence, a deduced 72-amino acid stretch displayed 56% amino acid sequence identity with the corresponding region of PfExp-1 (data not presented). This region was localized to exon 2 of PfExp-1 (3). This was the first suggestion that PyHEP17 may be the P. yoelii homolog of PfExp1. No other strikingly homologous regions were observed but an open reading frame of similar size to exon 1 of PfExp-1 and distance from the putative exon 2 of PyHEP17 was detected (Fig. 1, nucleotides 378 -498). A putative initiation codon at nucleotide 378 encoded the first in-frame methionine. No other methionine codons were found upstream and the preceding sequence had stop codons in all reading frames. While we cannot formally exclude the possibility and existence of a remote 5Ј exon containing the initiation methionine codon, this is unlikely since most previously described Plasmodium introns are less than 200 -300 bp in length (19). An 11-amino acid hydrophobic region (nucleotides 387-417) following the putative initiation methionine (nucleotide 378) is consistent with a signal sequence (21), providing further support for the designation of exon 1. A deduced amino FIG. 1. Nucleotide sequence and predicted amino acid sequence of PyHEP17 genomic DNA. The nucleotide numbers are shown on the right. The first ATG codon corresponding to the probable translation initiation codon is shown at nucleotides 378 -380 and the TGA stop codon at nucleotides 1160 -1162 is represented by asterisks. The intron between exons 1 and 2 is shown in lower case letters. The predicted amino acid sequence is presented in three-letter code. acid sequence (nucleotides 874 -950) encoding a putative anchor sequence was identified. A translation terminator (TGA) was observed at nucleotide 1160.
Northern blot hybridization of P. yoelii blood-stage parasite RNA with a PyHEP17 specific probe (HEP17Ex1.2) revealed a single band of approximately 1200 bp (data not presented). Southern blot hybridization of P. yoelii genomic DNA was consistent with PyHEP17 being a single copy gene, giving single bands of various sizes upon hybridization of DNA digested with various restriction enzymes (data not presented). The restriction pattern of genomic fragments hybridizing to HEP17Ex1.2 was consistent with the restriction pattern predicted from the sequence of the genomic clone and confirms that no apparent rearrangements occurred during cloning.
RT-PCR Amplification and Characterization of PyHEP17 cDNA from Parasite RNA-To determine whether any of the open reading frames represented exons and to subsequently ascertain the PyHEP17 cDNA sequence, oligo(dT)-primed firststrand cDNA was prepared from DNase-treated total bloodstage parasite RNA. Oligonucleotide primers were designed to amplify portions of one or both of the predicted exons (including all of the putative exon 1, beginning with the initiation ATG, and that portion of putative exon 2 homologous to PfExp) (Fig.  2). Three RT-PCR analyses revealed amplification products of the predicted sizes which hybridized to a PyHEP17-specific probe (LisaEx1/Ex2) internal to the amplification primers (data not presented). DNase treatment of genomic DNA template completely eliminated any signal but did not affect the signal amplified from the parasite RNA template (data not presented) indicating that the amplified products were not derived from contaminating DNA in the RNA preparation.
Analysis of the amplified cDNA sequence confirmed the presence of the putative intron separating the first and second exons. RT-PCR analysis using sequential primers (Fig. 2) designed to amplify different exon-and intron-containing regions of the PyHEP17 gene, in parallel with PCR amplification of genomic DNA using the same primers, was consistent with the predicted cDNA sequence. In particular, amplification using HEP3.1 as one of the primer pairs was consistent with the absence of any additional coding region in the 158 bp downstream of the designated stop codon (nucleotide 1160), since the deduced mRNA sequence and the genomic sequence are identical in this region and amplification products were of identical size.
Isolation and Characterization of PyHEP17 cDNA from gt22A cDNA Library-To confirm that the derived cDNA sequence does not contain any artifacts of PCR and to delineate the 5Ј-and 3Ј-untranslated regions, a P. yoelii gt22A cDNA library was constructed. A number of clones hybridized to a PyHEP17-specific probe (HEP17Ex1.2) and one clone was selected for further study. Digestion of the phage DNA of this clone with SalI and NotI restriction enzymes revealed a 1100-bp fragment, consistent with a full-length cDNA and similar to the size obtained by Northern blot of blood-stage parasite RNA. Subsequent DNA sequencing of this insert (data not presented) demonstrated, however, that this cDNA clone was not full-length as it lacked the first eight codons of the complete PyHEP17 coding sequence (Fig. 2).
Nevertheless, three sequences derived by independent means (cDNA clone isolated from a gt22A cDNA library, RT-PCR clone of oligo(dT)-primed cDNA from blood-stage parasite RNA, and genomic DNA clone isolated from a gt11 DNA library; Fig. 2) were identical at all nucleotide positions, thereby confirming the PyHEP17 cDNA sequence presented in Fig. 3. The final sequence was determined from both strands. Not shown in Fig. 3 is a 24-bp putative polyadenylation sequence that was identified approximately 620 bp downstream of the stop codon of the gt22A clone.
PyHEP17 therefore comprises two exons: exon 1, 40 amino acids (121 bp, nucleotides 378 -498 of the genomic sequence) and exon 2, 129 amino acids (386 bp, nucleotides 774-1160 bp), with a total size of 169 amino acids (507 bp) (Fig. 3). The open reading frame is followed by a TGA stop codon and at least 161 bp of the 3Ј-untranslated sequence: the 3Ј end oligonucleotide primer used to amplify the cDNA (HEP3.1) hybridized 161 bp downstream of the stop codon. The AT content of the coding and noncoding regions, 65 and 84%, respectively, are similar to those of other Plasmodium genes (19).
The predicted molecular mass based on the predicted amino acid sequence, 19 kDa, is in good agreement with the molecular mass observed for PyHEP17 in Western blotting experiments with P. yoelii infected erythrocytes (1) and P. yoelii infected hepatocytes. 2 Homology between PyHEP17 and PfExp-1-Striking amino acid sequence identity was found between the deduced Py-HEP17 sequence and that of PfExp-1. Alignment of PyHEP17 and PfExp-1 (Fig. 4A)   over the N-terminal 281 amino acids and 56% identity over the C-terminal amino acids, overall identity 42%, 239/574 amino acids) (31,32).
Interestingly, there is only one intron in PyHEP17, but two introns in PfExp-1 (Fig. 4C). Sequence analysis of two clones derived from blood-stage parasite RNA by RT-PCR and one cDNA clone derived from the gt22A expression library (Fig. 2) as well as RT-PCR analysis of blood-stage parasite RNA using sequential primers designed to amplify different exon-and intron-containing regions of the PyHEP17 gene (Fig. 3) are consistent with PyHEP17 containing only one intron. Furthermore, the overall size of PyHEP17 and PfExp-1 is similar at both the nucleotide (507 and 486 bp, respectively) and predicted amino acid (169 and 161 residues, respectively) level (Fig. 4, A and C). The first exon of both genes are identical in size. The second exon of PyHEP17 is 65 bp larger than the corresponding exon of PfExp-1, but PfExp-1 has an additional 44 bp of coding sequence in exon 3.
Antigenic Cross-reactivity between PyHEP17 and PfExp-1-Data indicate that PyHEP17 is the P. yoelii homolog of PfExp-1: (i) there is a high degree of sequence homology between PyHEP17 and PfExp-1, 37% homology at the amino acid level; (ii) the expression pattern of both antigens is similar. Neither PyHEP17 nor PfExp-1 are present in sporozoites, but both antigens are first expressed during the liver-stage of the parasite's life cycle, are present throughout the erythrocytic stage, and are exported into the cytoplasm of infected hepatocytes and erythrocytes (1,3,20); (iii) Western blot analysis of P. yoelii infected hepatocytes 3 and erythrocytes (1) using NYLS3 mAb identifies a protein of similar size to that identified in infected P. falciparum red blood cells; and (iv) consistent with the degree of sequence identity between PyHEP17 and PfExp-1, there is antigenic cross-reactivity. A mAb specific for PfExp-1, mAbN1 (17), PfExp-1 polyclonal sera, and human sera from individuals resident in an endemic area recognize COS cells transfected with PyHEP17Ex1.2 plasmid DNA and react with P. yoelii-infected liver-stage and blood-stage parasites (data not presented). Although the minimal epitope on PfExp-1 has not been defined, mAbN1 recognizes an epitope between amino acids 23 and 75 (17). Given the antigenic cross-reactivity reported here, the mAbN1 epitope is likely to map between amino acids 44 and 65 of PfExp-1 where 16/22 amino acids are identical to the homologous region of PyHEP17 (Fig. 4A).
Immunogenicity of PyHEP17 and PfExp-1-Previously, we have reported (2) that intramuscular immunization with a plasmid DNA construct encoding the complete exon 1 plus 57% of exon 2 of PyHEP17 (PyHEP17Ex1.2) induced protective immunity which was absolutely dependent on CD8 ϩ T cells, interferon-␥, and nitric oxide. We wanted to extend this observation and determine whether immunization did in fact induce CTL against PyHEP17. As shown in Table I, MHC-restricted, antigen-specific CTL were induced by immunization of BALB/c mice with PyHEP17Ex1.2 DNA. An 8-mer epitope in the amino terminus which is recognized by CTL in a CD8 ϩ T cell-dependent, antigen-specific, and MHC-restricted manner has been recently identified by peptide immunization. 4 Similar studies with the P. falciparum homolog have demonstrated that PfExp-1 is also a target of MHC-restricted antigen-specific CTL (experiment 1: % net specific lysis ϭ 46.9% at E/T ϭ 100:1, % lysis mismatched target ϭ 6.1%; experiment 2: % net specific lysis ϭ 39.6% at E/T ϭ 50:1, % lysis mismatched target ϭ 0%).
Since our previous work (2) had shown a negative association between antibodies and protection, we also wanted to determine whether inclusion of an antibody epitope in the plasmid DNA construct would alter the DNA-induced protection. The PyHEP17Ex1.2 DNA vaccine did not include 5 the region of the PyHEP17 protein recognized by the mAb NYLS3 used to identify PyHEP17. Therefore, a second construct was designed to include the NYLS3 epitope. This PyHEP17(1ϩ2) DNA vaccine comprised the complete exon 1 plus 85% of exon 2 of PyHEP17. In addition, since previous studies indicated that intradermal immunization with P. yoelii circumsporozoite protein plasmid DNA may result in an increase in antibody titer but a decrease in protection, 6 we wanted to compare intramuscular immunization of PyHEP17 plasmid DNA with intradermal immunization.
Three of four inbred mouse strains differing in genetic backgrounds and H-2 haplotypes were protected against sporozoite challenge by intramuscular and intradermal immunization with either PyHEP17Ex1.2 or PyHEP17(1ϩ2) DNA (Table II). We have previously reported (2) that intramuscular immuni- zation with the PyHEP17Ex1.2 DNA vaccine protects 71% of A/J mice, 54% of B10.BR mice, 26% of BALB/c mice, and 17% of B10.Q mice but no C57BL/6 mice. Results presented here are consistent with these data and show, furthermore, that the inclusion of the protective B cell epitope recognized by the NYLS3 mAb 7 had no effect on protection. This is consistent with previous data demonstrating that the mechanism of protection conferred by the PyHEP17Ex1.2 DNA vaccine was absolutely dependent on CD8 ϩ T cells, interferon-␥, or nitric oxide (2). In contrast to studies with DNA vaccines encoding either the P. yoelii circumsporozoite protein or the P. yoelii sporozoite surface protein 2, 8 however, we found no significant difference between intramuscular and intradermal routes of immunization in any of the mouse strains tested (Table II). Implications for Vaccine Development-The nature and magnitude of the cellular immune response to antigens depends on the intracellular fate of the protein. Conventionally, there are two major pathways of antigen processing and presentation: "endogenous" (class I) and "exogenous" (class II). MHC class I restricted T cells recognize peptides derived from processing of endogenously synthesized proteins, whereas class II restricted T cells recognize peptide fragments derived from exogenous proteins presented by antigen presenting cells (34 -37). In the erythrocyte, Plasmodium resides within a parasitophorous vacuole which is formed during merozoite invasion. The membrane of the parasitophorous vacuole forms a barrier between the parasite plasma membrane and the cytoplasm of the host cell. Exported parasite proteins have been defined as proteins which are secreted by the parasite and which remain inside the host cell or become associated with the host cell plasma membrane (38). Substantial evidence indicates that exported molecules can affect virulence and influence immunity to infection. They may act as virulence factors, as illustrated for Yersinia enterocolitica (39), Bordetella pertussis (40), and Strongyloides stercoralis (41), or as major targets of protective immunity, as illustrated for Mycobacterium tuberculosis (42)(43)(44). This apparent high immunogenicity of exported proteins may perhaps be attributable to the increased accessibility to proteolysis and transportation associated with targeting to the cytoplasm and more efficient entry into the class I and class II pathways (34 -37). Extrapolating to the malaria system, in comparison to non-exported proteins, exported proteins such as PfExp-1 may therefore be preferential targets of protective immunity.
Based on the striking amino acid sequence homology, expression pattern, and antigenic cross-reactivity reported here between the exported protein, PfExp-1, and the PyHEP17 antigen which is a target of both antibodies and CTL and can protect against malaria sporozoite challenge in three of five inbred mouse strains differing in H-2 haplotype and genetic background as well as outbred CD-1 mice (2), PfExp-1 must be nominated as a leading candidate antigen for malaria vaccine development. Indeed, DNA vaccination studies with PyHEP17 and the P. yoelii circumsporozoite protein in the rodent model demonstrate that PyHEP17 confers more broad-ranging protection against malaria (2). Extrapolation to the human system therefore suggests that any malaria vaccine formulations should include PfExp-1.