Clues to Evolution of the SERA Multigene Family in the Genus Plasmodium

Malaria, one of the most serious infectious diseases prevalent in the tropics, is caused by the genus Plasmodium. Despite considerable global efforts to control this parasitic disease at least 90% of deaths still occur in sub-Saharan Africa (WHO, 2010). The rising threat of drugresistant parasites, together with key interventions dependent on the use of a limited class of insecticides, underscore the fragility of malaria control. A better understanding of the parasite biology is required to gain insights for cost-effective tools or strategies including malaria vaccines and new antimalarial drugs that can be instrumental for sustained control, if not elimination, of malaria. Malaria parasites comprise a diverse group of over 200 Plasmodium species that infect mammals, birds and reptiles (Levine, 1988). Each Plasmodium species exhibits a restricted host range, such that primate parasite cannot infect rodent, bird or reptile hosts. To find genomic factors that determine host range transcends the interest of malaria researchers. It is essential for control and conservation of wildlife. Recently, genome projects on several Plasmodium species from different hosts have been completed (Gardner et al., 2002; Carlton et al., 2002, 2008; Pain et al., 2008), with gene information available in the public database (http://plasmodb.org/plasmo/). By comparing the genomes from different species we obtain basic information at the molecular level on how Plasmodium has evolved and allows us to infer the function of genes and noncoding regions in the genome. One of the prominent features of Plasmodium genomes is the presence of various unique multigene families. Multigene families are a group of related genes that are presumed to share a common ancestor and are derived from each other by duplication and subsequent divergence. One such example, the largest family identified so far in human, primate and rodent malaria, is the Plasmodium interspersed repeat, pir (Janssen et al., 2004). The pir gene family members are highly species-specific, suggesting evolution of lineage-specific immune evasion mechanisms. P. falciparum var gene family is by far the best documented multigene family of the most virulent human malaria parasite. Products of var genes appear on the surface of infected erythrocytes and are involved in antigenic variation to evade host immunity. Other species-specific gene families encode proteins involved in host cell invasion, e.g. rhoptry proteins and parasite surface antigens, merozoite surface protein-3 and -7. There are also examples of families with few gene members. In sharp contrast to several hundreds of tandem arrayed rRNA gene family members in other eukaryotes,

Plasmodium has 4-7 gene units physically separated in the genome (Nei & Rooney, 2005;Carlton et al., 2008).Thus, Plasmodium possesses unique multigene families with distinctive evolutionary conundrums.For more than 10 years after the first description of a gene family member, the existence of the Plasmodium serine repeat antigen (SERA) multigene family has been overlooked.Serine repeat antigen family proteins share homology with the papain family of cysteine proteases (Kiefer et al., 1996;Gor et al., 1998;Bourgon et al., 2004;Arisue et al., 2007Arisue et al., , 2011)).Almost all SERA genes are clustered in a head-to-tail manner and the number of SERA genes in the clustered region varies among parasite species (Bourgon et al., 2004;McCoubrie et al., 2007;Arisue et al., 2007Arisue et al., , 2011)).This leads us to infer that gene duplication occurred repeatedly during evolution.Some SERA genes were confirmed to play essential role(s) in the parasite life cycle (Miller et al., 2002;Aly & Matuschewski, 2005;McCoubrie et al., 2007).In addition, a gene family member in P. falciparum, SERA5, is a vaccine candidate now on phase Ib clinical trial (Horii et al., 2010).Two observations promise SERA5 as a vaccine candidate: (1) SERA genes are not differently expressed like other antigen encoding gene families such as var and rifin that show antigenic variation to evade host immune response (Aoki et al., 2002;Miller et al., 2002;Palacpac et al., 2006;Schmidt-Christensen et al., 2008;Putrianti et al., 2010;Arisue et al., 2011); and (2) P. falciparum SERA5 is less polymorphic (Fox et al., 1997: Morimatsu et al., 1997;Liu et al., 2000) than other vaccine candidate genes such as merozoite surface protein 1 (McBride et al., 1985) and apical membrane protein 1 (Polley et al., 2003;Cortés et al., 2003).These characteristics are indeed appealing and show the unique biological features of Plasmodium SERA.Here, we summarize current reports and our recent findings to understand the evolution of the SERA gene family.

SERA gene repertories in Plasmodium species
We refer briefly to the research history of the SERA multigene family: (i) the identification of SERA5 and the proteolytic processing of the protein; (ii) the discovery of the gene family following chromosome 2 sequencing of the P. falciparum genome; (iii) currently known SERA genes from different species; and (iv) the resulting analyses of the multigene family in various malaria parasites.

Research history of the SERA multigene family
SERA was first found in P. falciparum as an abundant, exported, soluble late-trophozite to schizont stage protein (Perrin et al., 1984).The protein, independently isolated by different groups, was described under various names as Pf140 (Perrin et al., 1984), p113 (Chulay et al., 1987), p126 (Deplace et al., 1987) or SERP (Knapp et at., 1989).All identified a gene with a long stretch of repeated serine residues in the N-terminal domain to which the family owes its name (Bzik et al., 1988).At the central domain, SERA possess a motif which align with two active site-determining regions of cysteine proteinases.The secreted protein was described to accumulate in the parasitophorous vacuole, and released into the culture medium at the time of schizont rupture.Notably, before the sequence of P. falciparum Chromosome 2 was opened, Knapp et al. (1991) discovered a SERA homolog (serp H) lacking the characteristic serine homopolymer, and subsequently Fox and Bzik (1994) reported SERA as one of three consecutive series of homologous genes.The complete genome sequence of P. falciparum revealed that SERA belongs to a multigene family (Gardner et al., 1998).The originally described SERA gene was renamed SERA5 according to the gene arrangement order in P. falciparum (Aoki et al., 2002;Miller et al., 2002).It is interesting to note that SERA5 is the only member with repeated serine residues among nine gene members.The characteristic of the family is not the richness in serine residues but motifs that generate the framework of a cysteine protease.SERA homologs were identified in other Plasmodium species.Kiefer et al. (1996) found five SERA genes from another human parasite, P. vivax and Gor et al. (1998) identified three SERA genes from the rodent parasite of P. vinckei.Completed or ongoing genome projects of eight Plasmodium species: two human malaria parasites, P. falciparum and P. vivax; chimpanzee parasite P. reichenowi; macaque parasite P. knowlesi; three rodent parasites P. berghei, P. yoelii and P. chabaudi; and avian parasite P. gallinaceum, confirmed the gene organization and allowed phylogenetic analysis of the SERA gene family (Burgon et al., 2005;Arisue et al., 2007).In addition, Arisue et al. (2011) newly identified SERA genes in 11 Plasmodium species that further elaborate the genome organization of the gene family.

Processing of P. falciparum SERA5
The in vitro observation that P. falciparum SERA5 was released into the culture supernatant at the time of schizont rupture/melozoite release corresponds to its specific processing into several polypeptides.The full-length 120 kDa precursor accumulates in the parasitophorous vacuole during late trophozoite and schizont stages.As shown in Fig. 1., during the course of schizont rupture/merozoite release, SERA is proteolytically processed into a 47 kDa Nterminal (P47), a 50 kDa central (P50), an 18 kDa C-terminal (P18) and a 6 kDa domain (Delplace et al., 1987(Delplace et al., , 1988;;Debrabant et al., 1992;Li et al., 2002a).The N-terminal P47 fragment is further processed into two 25 kDa fragments (P25n and P25c) in some allelic types (Li et al., 2002a).P47 is linked with the C-terminal P18 via disulfide bond that is localized at the merozoite surface that is localized at the merozoite surface (Delplace et al., 1987;Li et al., 2002a;Okitsu et al., 2007).The proteolytic processing is mediated by subtilisin-like serine protease subtilase 1 or SUB1 (Yeoh et al., 2007).Inhibition of SERA maturation blocks parasite egress from the host erythrocyte (Li et al., 2002b;Yeoh et al., 2007).

SERA genes in the database
The discovery of the SERA gene family in P. falciparum sparked the subsequent identification of a number of SERA genes in different Plasmodium species.Currently known SERA genes in the PlasmoDB database (http://plasmodb.org/plasmo/)are summarized in Table 1 and accession numbers of SERA genes found in the public database (NCBI, http://www.ncbi.nlm.nih.gov/) are summarized in Table 2.We opted not to list SERA genes of P. reichenowi and P. gallinaceum in either Table, but for our analysis, their SERA gene sequences were assembled from various reads in the partial genome shotgun database of Plasmodium at The Sanger Institute.Blast programs in the following web sites were used to search SERA coding reads for: P. reichenowi: http://www.sanger.ac.uk/cgi-bin/blast/submitblast/p_reichenowi; and P. gallinaceum: http://www.sanger.ac.uk/cgi-bin/blast/submitblast/p_gallinaceum).Identified SERA gene sequences could be referred to in Arisue et al. (2007).

Characteristic features of the SERA multigene family
Almost all SERA genes were aligned in a tandem cluster between the conserved hypothetical protein gene and the iron-sulfur assembly protein gene.According to genetic background, SERA genes can be categorized into Groups I to IV (Arisue et al., 2007(Arisue et al., , 2011)).Characteristic features of each Group are summarized in Fig. 2. Group I to Group III SERA genes possess the protease motif that includes an active site cysteine residue, in contrast to Group IV SERA genes where the cysteine residue is replaced by a serine (Bourgon et al., 2005;Arisue et al., 2007Arisue et al., , 2011)).The mRNA transcription and/or protein expression of Group I SERA genes were observed in the mosquito vector, while those of Group II to Group IV SERA genes were observed in the vertebrate host (Ali & Matuschewski, 2005;Arisue et al., 2007Arisue et al., , 2011;;Putrianti et al., 2010).The difference in the gene repertoire among species is due to the number of Group IV SERA genes.SERA gene repertories are summarized in Table 3.
*The gene region of PY02062 and PY00294 was re-annotated and used as P. yoelii SERA4.P. cynomolgi m a c a q u e 3 P. coatneyi m a c a q u e 3 P. fragile m a c a q u e 3 P. simiovale m a c a q u e 3 P. fieldi m a c a q u e 3 P. inui m a c a q u e 5

Primary structure of SERA molecules and genes
Schematic representation of SERA gene structure is shown in Fig. 3A.Group I SERA genes code for around 700 amino acids.They share a similar six exon and five intron structure, except for P. falciparum SERA8 and P. vivax SERA12 that both lack one intron.Group II to Group IV SERA genes code for about 1000 amino acid residues, and similar to Group I, share a common four exon/three intron structure with few exceptions.All SERA genes have the structural context of cysteine proteinases, however, it is important to note that the canonical Cys His Asn triad of the active proteinase is not present in all.The relatively small number of amino acid residues in Group I SERA resulted to a shorter N-terminal region when compared to Group II to IV SERA.Multiple amino acid sequence alignments revealed the consensus primary structure of SERA, which consists of six putative domains shown in Fig. 3B.Amino acid sequences of the putative pro-enzyme and enzyme domains are remarkably conserved, but extensive sequence variations are found in variable domains 1 and 2. In the C-terminal cysteine rich conserved domain, seven cysteine residues are perfectly conserved in all SERA genes.The pro-enzyme and enzyme domains of P. falciparum SERA5 was identified by functional genetic and structural analyses (Hodder et al., 2003(Hodder et al., , 2009)).These domains, corresponding to P50 in Fig. 1, are flanked by the reported SUB1 cleavage sites (Yeoh et al., 2007).The consensus sequence of the cleavage site is (Val/Leu/Ile)-Xaa-(Gly/Ala)-Paa, in which Xaa is any amino acid residue and Paa is any non-polar residue except for Leu (Yeoh et al., 2007).This consensus sequence is well conserved with slight modifications in all Group II to IV Plasmodium SERA genes which we have analyzed.In vitro, P. falciparum SERA4 (Group IV) and SERA6 (Group II) were cleaved by recombinant PfSUB1 (Yeoh et al., 2007).Peculiarly, Group I SERA genes lack most of the N-terminal variable domain 1 and SUB1 cleavage sites.

Phylogenetic relationships of SERA genes
The categorization of SERA genes into four groups, Group I to IV, was based on phylogenetic analysis (Arisue et al., 2007(Arisue et al., , 2011)).SERA amino acid sequences from 18 Plasmodium species were aligned using CLUSTAL W program (http://clustalw.ddbj.nig.ac.jp/top-j.html)under default options with manual corrections.Unambiguously aligned amino acid positions corresponding to the putative pro-enzyme domain, enzyme domain and cysteine rich conserved domain were selected and used for the phylogenetic analysis.Maximum likelihood tree was inferred using the PROML program in PHYLIP version 3.69 (Felsenstein, 1996).Except for the number of genes and number of amino acid sequences included in the analysis, the same method was used to infer the phylogenetic tree shown in Fig. 4 and 5.
A simplified maximum likelihood tree inferred from 134 SERA genes with 392 amino acid positions is shown in Fig. 4. Bootstrap proportion values were placed only on the common ancestor branch of each group.The monophyletic grouping of Group I SERA genes was supported with a bootstrap value of 100%.The long internal branch separating Group I from Groups II to IV suggests that the root of the tree is located on the branch leading to the common ancestor of Group I SERA genes.It is thus likely that Group I genes have appeared early in the evolution of the SERA gene family.P. gallinaceum SERA1 branches at the common ancestor of Group II to IV, suggesting that gene duplication events which produced Groups II, III and IV likely occurred after the divergence of P. gallinaceum from the common ancestral lineage of Plasmodium.Group IV SERA genes which diverged after the cysteine to serine substitution in the catalytic site were further divided to five monophyletic sub-Groups: (i) P. falciparum and P. reichenowi, (ii) P. ovale, (iii) rodent Plasmodium species, (iv) P. vivax and P. vivax-related monkey parasite species, and (v) P. malariae.These indicate that genes have duplicated independently in each of the sub-group lineages.To increase the resolution of the tree, the long branched Group I genes were excluded from the dataset and the maximum likelihood tree was re-constructed from 115 SERA genes categorized into Group II to Group IV.The resultant tree is shown in Fig. 5. Interestingly, Group IV SERA genes of P. vivax and P. vivax-related monkey parasites (10 species) were further categorized into six orthologous gene groups, namely, Clade 1 to Clade 6; and each clade has 5 (Clade 6) to 10 (Clade 5) parasite species.The number of SERA genes analyzed varied from 5 (P.fragile) to 12 (P.vivax).This does suggest that a common ancestor of P. vivax and related monkey malaria parasites had at least 6 SERA genes of Group IV; and that gene duplications and gene deletions occurred in each lineage.
Orthologous relationships between SERA gene members and their relative arrangement are shown in Fig. 6.In addition to several gene members characterizing Group IV, there are multiple SERA gene fragments and pseudogenes containing multiple stop codons.Taken together, these extensive gene duplications, gene deletions as well as pseudogenization/truncation are evident only in the serine type SERA gene (Group IV) of P. vivax and related monkey malaria parasites.

Transcription analyses of SERA genes
Transcription analyses revealed, likewise, some discordance among Plasmodium species.
Transcription profile of the SERA gene family was analyzed first by Aoki et al. (2002) in P. falciparum.Genes were most actively transcribed at the late trophozoite to schizont stages of the parasite with SERA5 predominantly transcribed among the family (Fig. 7.).Similar to P. falciparum, multiple number of SERA genes were transcribed at late trophozoite to schizont stages when transcription analysis was done for the SERA gene family in other Plasmodium parasites: human parasite P. vivax (Palacpac et al., 2006); rodent parasite P. berghei (Arisue et al., 2011); and three monkey parasites P. knowlesi, P. cynomolgi and P. coatneyi (Arisue et a.l, 2011).A representative summary of SERA gene transcription analysis is shown in Fig. 8.In malaria parasite species infecting humans, one of Group IV SERAs of P. falciparum (SERA5) and P. vivax (SERA4) showed the highest transcription level among other gene members.In the rodent malaria parasite P. berghei, Group III SERA gene (SERA3) was predominantly expressed.In three monkey malaria parasites, the abundantly expressed genes are members of Group IV Clade 3: SERA3 and SERA5, in P. cynomolgi; SERA3 in P. coatneyi; and SERA2 in P. knowlesi.These results show that SERA genes were differently expressed between rodent and primate parasites.Based on the malaria parasite mitochondrial genome, P. falciparum belongs to the primate parasite group 1 lineage whereas P. vivax and the three macaque parasites belong to primate parasite lineage 2. Phylogenetic analysis showed that these two lineages are not closely related; and the rodent parasite lineage is positioned between them (Hayakawa et al., 2008).Note, however, that both primate lineages showed similar transcription pattern of SERA gene which might suggest a possible relationship between SERA function and host specificity.

Duplication in the multigene family
In general, duplicated genes undergo either (i) concerted evolution or (ii) birth-and-death evolution (Nei & Rooney, 2005).In the concerted evolution model, all members of a gene family evolve as a unit rather than independently.When a mutation occurs in a gene, mutation spreads to all other gene members by unequal crossover or gene conversion.As a result, all members of the gene family show identical sequence to each other.The evolution of rRNA multigene families in vertebrates is a classic example of concerted evolution.Analysis of MHC genes in mammals (Hughes & Nei, 1989;Nei et al., 1997;Nei & Hughes, 1992), other immune system related genes (Hughes & Nei, 1990;Ota & Nei, 1994) and disease-related genes (Zhang et al., 2000) show a quite different evolutionary pattern.The birth-and-death evolution model was proposed to explain differential duplication/independent diversification processes that result to subsequent loss or maintenance of genes in a multigene family.Thus, some duplicated genes are maintained in the genome for a long time while others are deleted or became pseudogenes through deleterious mutations.This model applies to rRNAs of Plasmodium species in marked contrast to the concerted evolution of rRNAs in most organisms.The model aptly explains the observation that rRNA genes in Plasmodium were structurally and functionally distinct (Rooney, 2004;Nishimoto et al., 2008).The observed gene duplication and gene deletion found in the Plasmodium SERA genes are clearly in concordance with the birth-and-death model, although traits of gene conversion are detected in a few of Group IV SERA genes.The birth-and-death model has, likewise, been recently proposed for gene duplication/gene deletion of merozoite surface protein 7, an immune target parasite surface antigen gene (Garzón-Ospina et al., 2010).It, thus, seem most probable that diversification of Plasmodium SERA multigene family was also driven by the birth-and-death evolution.Inferred gene duplication events in the evolution of the Plasmodium SERA gene family are shown in Fig. 9. Theileria is the only other genus to possess a SERA homolog gene.The apicomplexan parasite is closely related to Plasmodium.Sequence similarity search against Theileria genome database at The Sanger Institute identified a single gene from both T. parva and T. annulata which has similarity with cysteine-type SERA (Arisue et al., 2007;McCoubrie et al., 2007).Based on Fig. 9, as all Plasmodium species have multiple SERA genes, we likely infer that the first duplication occurred at the common ancestor lineage of Plasmodium.Because every Plasmodium species has Group I SERA gene, the first duplication event is from Group I SERA gene.The duplication events which gave rise to Group II and IV SERA genes occurred after the divergence of P. gallinaceum from the branch leading to a common ancestral species of other Plasmodium species since P. gallinaceum has no Group II to IV SERA gene.The rest of the 17 Plasmodium species might have diverged into five lineages of (i) P. falciparum and P. reichenowi, (ii) P. ovale, (iii) rodent Plasmodium species, (iv) P. vivax and P. vivax-related monkey parasite species, and (v) P. malariae, and duplications of Group IV SERA genes occurred independently on each lineage.In addition, gene deletions as well as pseudogenization/truncation occurred frequently in P. vivax and P. vivax-related primate parasite lineage.

Conclusion and open issues in SERA study
Multigene families are believed to provide an organism with a set of related genes that allow fine tuning of its biological function with possibly different temporal or topologic expression patterns.SERA gene duplications during Plasmodium evolution generated four types of SERA genes: Group I to Group IV.The speculated function of SERA during the parasite life cycle is summarized in Fig. 10.Group I SERA gene, bearing the canonical cysteine, was shown to be transcribed in the oocyst stage and its gene product was a protease required for sporozoite egress from the oocyst (Aly & Matuschewski, 2005).Group III SERA gene was suggested to play an essential role during schizont rupture/merozoite release in the mammalian host (Yeoh et al., 2007;Arastu-Kapur et al., 2008;Putrianti et al., 2010).Group IV SERA genes bear the characteristic replacement of active-site cysteine to a serine residue.Perhaps importantly, only mammalian parasites have Group IV SERA gene; and Group IV SERA gene of primate parasite was suggested to play an essential role in schizont rupture/merozoite release together with Group III SERA gene (Yeoh et al., 2007;Arisue et al., 2011).The duplication of Group IV SERA gene occurred particularly frequent in two evolutionarily distinct primate lineages and it is intriguing to assume that duplications of SERA genes were associated with host range expansion.The study of the SERA gene family points to its unique features reinforcing the importance of investigating other uncharacterized gene families of Plasmodium to further understand the evolutionary history and biology of this harmful parasite.Many questions still remain in the analysis of SERA.SERA genes are thought to be subject to birth-and-death evolution, and thus, a pattern of interspecific gene clustering is expected to characterize the SERA family whereby functional genes are maintained in the genome for a long time and others are deleted or become non-functional.Group I and Group III SERA genes are highly conserved in Plasmodium species.For Group II SERA genes, although maintained among Plasmodium species with significant sequence similarity, no function has yet been predicted.Gene disruption studies with Group II SERA gene of P. berghei showed no apparent phenotypic change (Arisue et al., unpublished data).Group II is similar to Group I and Group III in being a cysteine-type SERA gene which has been suggested to have proteolytic activity to cleave host membrane structure (Aly & Matuschewski, 2005;Yoeh et al., 2007).The papainlike cysteine protease motif in its amino acid sequence suggests the possibility that Group II SERA act as a protease sometime in the parasite life cycle.Parasite egress from the host cell is an important process that remains poorly understood.P. falciparum SERA5 is a vaccine candidate molecule now on clinical trial in Uganda (Horii et al., 2010).Serum antibodies against the N-terminal domain of P. falciparum SERA5 in individuals living in malaria endemic area protect infants from clinical malaria and inhibit in vitro parasite growth (Okech et al., 2001(Okech et al., , 2006;;Aoki et al., 2002;Horii et al., 2010).During blood stage growth, all SERA gene family member of P. falciparum are transcribed most actively at trophozoite and schizont stages.SERA5 is the most abundantly expressed gene family member, with expression levels estimated to be approximately 0.5-1.5% of the whole mRNA at schizont stage (Aoki et al, 2002).However, sero-positivity rate against the Nterminal domain of P. falciparum SERA5 was observed to be relatively low (Fig. 11.; Aoki et al., 2002;Horii et al., 2010).Since the SERA gene family does not show antigenic variation to evade host immune response (Fig. 7.), there may possibly be another mechanism of host parasite evasion/molecular mimicry/interference or competition.The host range or host specificity of Plasmodium is believed to be restricted, although, primate malaria parasites generally infect multiple hosts.For example, it has been reported that, P. knowlesi and P. cynomolgi have the ability to infect a wide variety of macaques and human (Coatney et al., 1971); additionally, two human parasites P. malariae and P. ovale have been detected in chimpanzees (Hayakawa et al., 2009;Duval et al., 2009).It may be probable that duplications of Group IV SERA genes that occurred frequently in both primate parasite lineages may be associated with host range expansions.To date, no experimental support lends credence to this speculation.As described above, the molecular function of SERA genes in each group, the relationship of immune evasion mechanism and the SERA gene family, and the association of host range with Group IV SERA genes remain important issues that needs to be addressed.The importance of SERA genes in parasite egress and their role in host-parasite interactions serve to propel further studies in understanding this multigene family.

Acknowledgment
This work was supported by KAKENHI (18073013 and 20390120) from the Japanese Ministry of Education, Science, Sports, Culture and Technology.

Fig. 3 .
Fig. 3. Schematic representation of the SERA gene structure (A) and their putative domain organization (B).

Fig. 6 .
Fig. 6.Organization and phylogenetic relationship of SERA gene in 18 Plasmodium species.

Fig. 7 .
Fig. 7.The relative abundance of mRNA for each P. falciparum SERA gene during late trophozoite to schizont stages of the parasite.

Fig. 8 .
Fig. 8. Representation summary of SERA gene transcription analyses in primate and rodent Plasmodium spp.Abundantly transcribed SERA genes are differentiated using enlarged solid circles.

Fig. 9 .
Fig. 9. Inferred gene duplication events in the evolution of the Plasmodium SERA gene family.Asterisk in P. reichenowi denotes that the SERA gene number in this species is tentative.

Fig. 10 .
Fig. 10.The life cycle of the malaria parasite and inferred role(s) of SERA.

Table 1 .
GeneID of SERA genes in the PlasmoDB (http://plasmodb.org/plasmo/).Clues to Evolution of the SERA Multigene Family in the Genus Plasmodium 319

Table 2
(Arisue et al., 2011)f SERA genes in the public NCBI database.Fig.2.Organization of Plasmodium SERA multigene family and the characteristic features of each SERA group.Because of limited information for P. reichenowi, P. gonderi and P. vinckei, gene numbers are tentative for these species.The total number of SERA pseudogenes, truncated gene and gene fragments in Group IV SERA gene region are still undetermined in primate parasites(Arisue et al., 2011).Except for P. gallinaceum, all parasite species have one each of Group I to III SERA genes.The number of Group IV SERA genes remarkably increased in the primate parasite lineage.The bird parasite, P. gallinaceum, has the least number of SERA genes: two from Group I, and one from a branched common ancestor of Group II to IV SERA gene.

Table 3
. The number of SERA genes that belong to each group from several Plasmodium species.'Degenerate'(*) denotes defective gene copies, i.e., the total number of pseudogenes, truncated gene and gene fragments found.