Cross-reactive anti-PfCLAG9 antibodies in the sera of asymptomatic parasite carriers of Plasmodium vivax.

The PfCLAG9 has been extensively studied because their immunogenicity. Thereby, the gene product is important for therapeutics interventions and a potential vaccine candidate. Antibodies against synthetic peptides corresponding to selected sequences of the Plasmodium falciparum antigen PfCLAG9 were found in sera of falciparum malaria patients from Rondônia, in the Brazilian Amazon. Much higher antibody titres were found in semi-immune and immune asymptomatic parasite carriers than in subjects suffering clinical infections, corroborating original findings in Papua Guinea. However, sera of Plasmodium vivax patients from the same Amazon area, in particular from asymptomatic vivax parasite carriers, reacted strongly with the same peptides. Bioinformatic analyses revealed regions of similarity between P. falciparum Pfclag9 and the P. vivax ortholog Pvclag7. Indirect fluorescent microscopy analysis showed that antibodies against PfCLAG9 peptides elicited in BALB/c mice react with human red blood cells (RBCs) infected with both P. falciparum and P. vivax parasites. The patterns of reactivity on the surface of the parasitised RBCs are very similar. The present observations support previous findings that PfCLAG9 may be a target of protective immune responses and raises the possibility that the cross reactive antibodies to PvCLAG7 in mixed infections play a role in regulate the fate of Plasmodium mixed infections.

The cytoadherence-linked asexual gene (clag) was initially identified in parasite lines with a deletion in chromosome 9 (Shirley et al. 1990) that lost the ability of binding to C32 melanoma cells bearing surface CD36 (Trenholme et al. 2000). Further sequencing of Plasmodium falciparum genome (Gardner et al. 2002), established that the chromosome 9 deletion (D10 deletion) affected a subtelomeric region containing 20 coding sequences. Additional experiments led to the identification of the gene encoding the CLAG protein containing nine exons (Holt et al. 1998, Gardiner et al. 2000). At the same time, a family of clag genes homologous to Pfclag9 was described (Holt et al. 2001). Clag family members were found on chromosome 2 (clag2), 3 (clag3.1 and clag3.2) and 8 (clag8). To date, three CLAG proteins were located in the rhoptries and they were associated to the high molecular weight protein complex of the RhopH family (Kaneko et al. 2001, Gardiner et al. 2004).

More recently, the functional and structural roles of other components of the clag gene family have been elucidated. The PfRhopH complex containing CLAG proteins is composed of three subunits named RhopH1, RhopH2 and RhopH3 (Kaneko et al. 2005). In the mature schizont, the subunits are localised in the merozoites' rhoptries, whose contents are discharged at the moment of contact with the erythrocyte membrane, concomitantly with the formation of the moving junction and the parasitophorous vacuole (PV). The three protein known, components of the PfRhopH1 subunit (CLAG2, CLAG3.1 and CLAG9), are then discharged into the PV (Ling et al. 2004, Kaneko 2007, Iriko et al. 2008).

The rhoptry neck protein 2 is associated in erythrocyte invasion (Cao et al. 2009). It is expressed in the apical portion of the rhoptry in association with the RHopH1 complex that includes CLAG9. It is known that Plasmodium proteins can be exported to the host cell cytosol (Richard et al. 2010) via the translocon export complex of proteins that include the CLAG family. Thus merozoites can secrete directly products from the apical organelles into the PV and enter the PV membrane, or via the Plasmodium export element (de Koning-Ward et al. 2009, Mayer et al. 2009) they reach the erythrocyte plasma membrane. Therefore, the RhopH/CLAG complex discharged by the merozoites will participate in remodelling the infected red blood cells (RBCs).

Recent genetic experiments (Nguitragool et al. 2011) using clones obtained from the cross of HB3 and Dd2 P. falciparum strains showed that PfCLAG3 participates in the Plasmodial Surface Anion Channel formation. In addition, the traffic of PfCLAG3 after its injection into the cytosol and entry into the PV membrane (PVM) has been followed up to its final destination in the infected erythrocyte membrane. Goel et al. (2010) proposed that the exported PfCLAG9 also traffics to the erythrocyte membrane PfCSA variant antigen.

On the other hand, in recent studies, surprising conclusion concerning the functional role of CLAG9 has been reached (Nacer et al. 2011). In convincing detailed experiments using atomic force microscopy and knockout disruption of the Pfclag9 gene, it was shown that CLAG9 does not contribute to cytoadherence to CD36. Thus the non-adherent phenotype in the original D10 deletion of chromosome 9 (Shirley et al. 1990) must be dependent on another gene(s) encoded in the D10 deletion (Nacer et al. 2011). The authors conclude that CLAG9 function, like that of CLAG3 (Nguitragool et al. 2011), is associated with the metabolic requirements of the parasite.

Considering the important roles of the proteins encoded by the clag gene family in the life cycle of P. falciparum at the asexual blood stages, including the erythrocyte invasion step, the participation of PfCLAG9 in the development of immunity to falciparum malaria was investigated in Papua New Guinea. A direct correlation with high antibodies titres against peptides representing linear epitopes of PfCLAG9 and immunity in semi-immune children and adults was found (Trenholme et al. 2005). In the present study we prepared synthetic peptides corresponding to different segments of PfCLAG9 and analysed their antigenicity in individuals from the Brazilian Amazon infected with falciparum or vivax parasites. Two groups were analysed: (i) individuals presenting clinical symptoms and (ii) asymptomatic parasite carriers.

SUBJECTS, MATERIALS AND METHODS

Study site and population - The study was performed with patients from suburban and rural riverside areas of the Rio Madeira in Porto Velho, capital of the state of Rondônia (RO), an area with a high incidence of malaria located in the Brazilian Amazon.

The population presents a profile of several previous episodes of malaria, as described (Tada et al. 2007). This characteristic of the population due to a high density of the Anopheles darlingi vector (Gil et al. 2003, 2007). Annual parasite index levels found for residents of these localities were 200-800, in association with the development of natural immunity among long time residents and a prevalence of asymptomatic malaria parasite carriers in the adult population varying from 15-30% for Plasmodium vivax and 5-10% for P. falciparum (Katsuragawa et al. 2009, 2010). Control measures developed in recent years, based on new formulations of drugs against P. falciparum malaria, have reduced the risk of P. falciparum infections in the studied areas to less than 10% (Epidemiologic Surveillance of Information System/Brazilian Health Ministry) (portalweb04.saude.gov.br/sivep_malaria/default.asp). The county of Porto Velho still accounts for 10% of the total P. vivax and P. falciparum malaria cases in Brazil.

Diagnostic, sera collection and processing - Sample collection of parasites, as well as blood sample collection and processing, have been previously described (Tada et al. 2007). Malaria infection diagnostic procedures were performed by microscopy examination and nested polymerase chain reaction (PCR) methods. The protocol used was based in Snounou (1996) and used genus-specific (rPLUS5 and rPLUS6) and species-specific (rFAL1,rFAL2, rMAL1, rMAL2, rVIV1 and rVIV2) primers. Briefly, the 20 µL of reaction volume per tube consisted of 250 nm of each primer, 125 µM dNTPs, 2 mM MgCl2, 50 mM KCL, 10 mM Tris pH 8.3, 0.4U Taq polymerase (Invitrogen) and 1 µL of genomic DNA. In the first PCR, the samples were subjected to initial denaturation at 95ºC for 5 min, a step at 58ºC for 2 min, a step at 72ºC for 2 min that was followed by 30 cycles at 94ºC for 1 min, 50ºC for 2 min, 72ºC for 2 min and a final step at 72ºC for 5 min in Gene Amp(r) PCR System 9700.

The second reaction was performed with 35 amplification cycles. All PCR assays included positive control (P. vivax, P. falciparum) and negative controls (ultrapure water). The fragments were separated by electrophoresis on a 1.5% agarose gel in tri-borate-ethylenediamine tetraacetic acid buffer and were visualised with ethidium bromide (Invitrogen) or Blue Green (LCG Biotechnologies, Brazil) under ultraviolet light. Asymptomatic malaria parasite carriers were defined by PCR detection of parasites in the absence of clinical symptoms and signs of malaria in the 30 days of surveillance following the positive examination.

Human serum samples - Serum samples were obtained from two groups of malaria patients: (i) Those who actively sought malaria treatment and who showed symptoms of acute malaria and positive blood smears for P. falciparum or P. vivax. These sera were collected between 2006-2008 in the Malaria Post of the Centre for Research in Tropical Medicine (CEPEM). (ii) Another group of sera was obtained from asymptomatic individuals patients living in the malaria endemic localities in the suburban and rural localities of Vila Candelária, Cachoeira do Teotônio, Santo Antônio and Vila Amazonas of Porto Velho (Tada et al. 2012). Asymptomatic sera were thus classified as those from individuals in the same localities, with positive PCR for P. vivax and/or P. falciparum and without recorded clinical symptoms for at least 30 days period after the positive PCR diagnosis.

CLAG9 peptides - The selection of synthetic peptides representing possible epitopes of CLAG9 (GenBank 167963009) (ncbi.nlm.nih.gov/genbank) was based the following programs: Flexibility Prediction (Karplus & Schulz. 1985), Hydrophilicity Prediction (Parker et al. 1986), Antigenic Propensity (Kolaskar & Tongaonkar 1990), Bepipred Linear Epitope Prediction (Larsen et al. 2006) and Surface Accessibility Prediction (Emini et al. 1985). Calculations in each method were made using the software Antibody Epitope Prediction available from the Immune Epitope Database (immuneepitope.org/tools/bcell/iedb_input). The three selected peptides derived from PfCLAG9 were named A, B and C (Fig. 1) and were synthesised at GenScript (Piscataway, NJ, USA). The comparison between PfCLAG9 and PvCLAG7 (GenBank 156081674) and PvCLAG8 (GenBank 156096581) proteins was done using the CLUSTALW software (Thompson et al. 1994).

Fig. 1

localization of the three synthetic peptide (Pep) in PfCLAG9 protein. Peptide A reproduces the 32 amino acid sequence from the residue 728-760, peptide B reproduces the 43 amino acid (aa) sequence from residue 901-944 and peptide C reproduces the 56 amino acid sequence from the residue 1284-1340.

Antibody reactivity to synthetic peptides in sera of malaria patients - All serum samples from malaria patients as well as negative controls were tested in duplicate by the enzyme-linked-immunosorbent-assay (ELISA) as described by Braga et al. (2002). This test was performed thrice. The cut-off value was defined as the mean plus two standard deviations of the absorbance values obtained with 64 negative control sera from malaria-naïve blood donors living in Porto Velho and São Paulo. The cut-off absorbance values used to define positive results were 0.216, 0.226 and 0.236 for peptides A, B and C, respectively. The results were expressed as reactivity index (RI) defined by optical density (OD)450nm values of tested samples divided by the value of the cut-off. Values of RI > 1.0 were considered positive.

Immunisation of BALB/C mice with synthetic PfCLAG9 peptides - Female BALB/c mice (4-6 weeks old) were immunised intramuscularly with three doses of 50 µg of peptide for each synthetic peptides A, B and C (5 mice per group). The first dose of peptide was emulsified in Freud's complete adjuvant (Sigma) and the two subsequent peptide boosters 20 and 40 days later were emulsified in Freud's incomplete adjuvant (Sigma). Sera were collected prior to the first immunisation and 20 days after administering the last booster dose. The sera of each group of mice were pooled and stored at minus 20ºC until use.

Parasitised RBCs (PRBCs) preparation and immunofluorescence antibody test (IFAT) - P. falciparum 3D7 (Walliker et al. 1987) PRBC were obtained from parasite cultures as described by Trager and Jensen (1976). Parasite cultures were synchronised by sorbitol lysis (Lambros & Vanderberg 1979) and trophozoites and schizonts obtained as in Lelievre et al. (2005). For P. vivax infected RBCs isolation, peripheral blood samples from donor's patients with P. vivax malaria were collected into 20 mL of heparinised containing vacuntainer tubes. PRBCs in stage of trophozoites and schizonts were concentrated using a discontinuous Percoll gradient (GE Healthcare) (Andrysiak et al. 1986). Thin smears were prepared and placed on multisport slides at room temperature and stored at -80ºC. Slides were fixed with formaldehyde for 2 min, washed twice with phosphate buffered saline (PBS) according to a previously established protocol (Nacer et al. 2011). Slides were incubated in humid chamber for 1 h at 37ºC with primary antibody (mouse anti-peptides A, B, C and negative control - polyclonal antibodies) diluted to 1:40 in blocking buffer (0,5% gelatin, 1,5% bovine serum albumin, 0,02% Tween 20 in PBS). Sequentially, slides were washed three times with PBS and incubated with goat anti-mouse IgG Alexa Fluor 488 (1:600) (Life Technologies) 45 min. Nuclei was stained with 0.2 mg/mL 4'6-diamindino-2-phenylidole (Sigma) for 5 min and finally mounted and examined using fluorescent microscope (Nikon eclipse 80i).

Statistical analysis - Statistical analysis was evaluated using GraphPad Prism 5 for Windows(r). All comparisons were between groups defined as asymptomatic and symptomatic malaria parasite carriers. For quantitative data, the non-parametric data, the Mann-Whitney U test was applied (RI comparisons considering only positive values). For qualitative data (positive and negative sera) Fisher's exact test applied. Confidence intervals of 95% were also calculated.

Ethical statements - The research protocols for human subjects were approved by the Ethical Research Committee (CEP) of the CEPEM in accordance with ethical principles of conduct. Blood samples of individuals in the study population were collected after informed consent and written agreement of each individual of the population sample (CEP-CEPEM decision 25/2004, Candelária; 49/2006, Porto Velho CEPEM; 70/2008, Sector - Santo Antônio - Teotônio, Cachoeira do Teotônio - Santo Antônio). All procedures for mice immunisation and preparation of anti-PfCLAG9 antigens sera were evaluated and approved by the Ethical Committee on Animal Use of the Research Institute of Tropical Pathology-Oswaldo Cruz Foundation (2009/2).

RESULTS

Antibodies against PfCLAG9 A, B and C peptides present in malaria patients - The first ELISA test were performed in sera of falciparum malaria patients (n = 115), independent of or not with clinical symptoms at the time of blood sampling. As negative controls (n = 64) and sera from vivax malaria patients (n = 139). Surprisingly, positive values were obtained in the sera of vivax patients, with stronger signals in asymptomatic parasite carriers (Table).

TABLE

Enzyme-linked-immunosorbent-assay test comparing sera reactivity of Plasmodium falciparum and Plasmodium vivax clinical patients and asymptomatic (Asymp) parasite carriers against PfCLAG9 peptides A, B and C

Non parametric Mann-Whitney test		P. falciparum					P. vivax
				95% CI					95% CI
Peptides PfCLAG9	Group	n	Mean	Lower	Upper	p	n	Mean	Lower	Upper	p
A	Symp	42	5.494	4.630	6.357	0.0004	33	1.978	1.527	2.430	0.0001
	Asymp	64	3.747	3.101	4.392		52	3.872	3.368	4.375
	Total	106	4.439	3.903	4.975		85	3.137	2.735	3.538
B	Symp	41	6.055	4.982	7.128	0.0094	46	1.963	1.672	2.255	0.0001
	Asymp	68	7.940	7.052	8.829		56	5.745	4.647	4.752
	Total	109	7.231	6.534	7.928		102	4.039	3.327	4.752
C	Symp	44	2.667	2.165	3.169	0.0141	32	1.384	1.154	1.614	0.0001
	Asymp	66	3.718	3.113	4.324		57	2.687	2.276	3.099
	Total	110	3.298	2.877	3.729		89	2.219	1.916	2.522
PfMSP1-19	Symp	-	-	-	-	-	16	2.598	1.819	3.376	0.0579
	Asymp	-	-	-	-	-	9	3.604	2.957	4.252
	Total	-	-	-	-	-	25	2.96	2.404	3.516
PvMSP1-19	Symp	17	2.287	1.586	2.988	0.1361	56	3.068	2.591	3.545	0.2513
	Asymp	29	2.767	2.271	3.262		59	2.556	2.231	2.880
	Total	46	2.589	2.194	2.985		115	2.805	2.520	3.090

the p values shown are from the non-parametric Mann-Whitney U test. Only positive (reactivity index ≥ 1) results were considered for descriptive and tests statistics. CI: confidence interval; MSP1-19: merozoite surface protein of P. falciparum; PvMSP1-19: MSP of P. vivax; Symp: symptomatic.

The results presented in Table confirm previous observations with P. falciparum patients in Papua New Guinea (Trenholme et al. 2005), showing higher ELISA RI values for peptides A, B and C in asymptomatic carriers than in patients with clinical symptoms. However, a high proportion of sera from P. vivax patients also recognised the A, B and C peptides of PfCLAG9 and the intensity of the responses correlated with clinical immunity to vivax infections.

Nevertheless, positive values were found in 90-100% of all P. falciparum patient samples, while the frequency was 40-60% of symptomatic P. vivax patients and 80-90% of asymptomatic subjects. When considering only the reactive sera among clinical patients samples, the average RI value observed was two-three fold higher in P. falciparum than in P. vivax patients. However, for asymptomatic carriers of both parasite species the observed RI values were equivalent (Table).

Antibodies against merozoite surface protein (MSP)1- antigens of P. falciparum or P. vivax origin in sera of symptomatic and asymptomatic vivax malaria patients - Our results suggest that some symptomatic or asymptomatic P. vivax patients had been previously exposed to P. falciparum infections. To study this possibility we tested all sera against recombinant MSP1-19 antigens of P. vivax and of P. falciparum origin (PvMSP1-19 and Pf MSP1-19).

Fig. 2 and Table compare ELISA titres performed with PfCLAG peptides A, B and C and those with those using PfMSP1-19 and PvMSP1-19 antigens. The selected samples corresponded to sera from patients with positive PCR for P. vivax and negative PCR for P. falcipa-rum and with antibody reactivity with at least one of the PfCLAG9 peptides. Only 9 (15.2%) out of 59 sera from asymptomatic patients reacted against the heterologous PfMSP119, indicating that the large majority of these patients had not been recently exposed to P. falciparum infections. Among patients with clinical symptoms, only 16 (28.6%) out of 59 sera from P. vivax patients presented reactivity against PfMSP1-19 antigen while 100% reacted against the homologous PvMSP1-19.

Fig. 2

comparison of frequency and antibody serum samples levels of each recombinant protein merozoite surface protein (MSP)1-19 of Plas modium. falciparum and Plasmodium vivax antigens carries positive PfCLAG9 with P. vivax asymptomatic (Asymp) and symptomatic (Symp). Negative (Neg) [reactivity index (RI) < 1] and positive (Pos) (RI ≥ 1) sera counts where done using Fisher's exact test.

Molecular basis of the cross reactivity - In order to understand the molecular basis for the observed CLAG9 cross reactivity we investigated possible sequence identities and similarities between PfCLAG9 synthetic peptides A, B and C and the P. vivax CLAG orthologs encoded on chromosomes 7, PvCLAG7 (Moreno-Perez et al. 2011) and 8, PvCLAG8 (Carlton et al. 2008). The other hypothetical genes deposited, such as PvCLAG14, showed no significant similarities for comparison.

As shown in Fig. 3A, significant similarity (84.44%) and identity (64.4%) was observed between peptide B (PfCLAG9 - residues 901-944) with amino acids residues 888-932 of PvCLAG7. Peptide B presents also significant similarity (86.66%) and identity (55.6%) with P. vivax CLAG 8, but the similarity amino acids are found at a slightly shifted position (897-941) of PvCLAG8 (Fig. 3B). In addition, a lower but still significant degree of similarity was observed for peptides A and peptide C (Fig. 3A, B).

Fig. 3

sequence similarity between synthetic PfCLAG9 peptides A, B and C with the corresponding regions of PvCLAG7 encoded on chromosome 7 (A) and CLAG8 encoded on chromosome 8 (B).

The highest identity/similarity observed with peptide B (residues 901-944 of PfCLAG9 and residue 888-932 of PvCLAG7) coincided with the highest ELISA RI values, or 7.9 and 6.0 for P. falciparum asymptomatic and symptomatic infections, respectively and 5.7 and 1.9 for P. vivax. Similar pattern was shown for peptide A (Table). These results show that cross reactivity is likely due to similarity between the CLAG proteins of both parasite species, supporting the notion that PvCLAG7 and PvCLAG8 are orthologs of PfCLAG9.

Cross reactive antibodies recognise epitopes in RBCs infected by P. falciparum or P. vivax. Fig. 4 represents the result of indirect immunofluorescence assays of trophozoite and schizonts stages of P. vivax and P. falciparum using mouse antibodies to synthetic peptide B of PfCLAG9 antigen. As shown, the pattern of parasites labelling in the PRBCs is very similar. Comparable profiles were observed in schizonts incubated with antisera to PfCLAG A, PfCLAG B and PfCLAG C (data not shown). No positive labelling was observed in ring stages.

Fig. 4

indirect immunofluorescence pattern of mouse polyclonal anti-serum against synthetic peptides of PfCLAG9 in parasitised red blood cells (PRBC) with schizonts of Plasmodium vivax (A) and Plasmodium falciparum (B) and in PRBC with trophozoites of P. vivax (C) and P. falciparum (D), respectively, by goat anti-mouse IgG Alexa Fluor 488.

DISCUSSION

Results in the present study show that antibodies with high reactivity to PfCLAG9 are found in asymptomatic P. falciparum parasite carriers patients correlating with clinical immunity and supporting previous epidemiological observations in Papua New Guinea (Trenholme et al. 2005). Interestingly, high reactive cross-reactive antibodies against PfCLAG9 peptides were also found in sera of asymptomatic P. vivax malaria parasite carriers patients and this is the main focus of the discussion.

The first point to considerer is if the observed cross reactivity of antibodies corresponds to some immune-physiological process, it could be related to interactions between malaria parasites in the vertebrate host. Previous studies with semi-immune children in Papua Guinea, harbouring mixed infections by both Plasmodium species, have shown that parasite density of Plasmodium species in mixed infections oscillates around a threshold and that peaks of infection with each species do not coincide (Bruce et al. 2000, Bruce & Day 2003). The authors proposed that malaria parasitaemia is controlled in a density-dependent manner in these semi-immune children by a cross-species parasite regulatory mechanism involving variant parasite antigens.

Existence of regulatory interactions between parasites in mixed infections has been previously also suggested by Bouharoun-Tayoun et al. (1990). In their experiments, passive transfer of IgG collected from adult immune donors to P. falciparum malaria infections, from West Africa, to young Thai patient receivers, with active P. falciparum infections, resulted in a significant decrease of the P. falciparum parasitaemia and improvement of clinical symptoms. However, in three of the receivers, following administration of the anti-falciparum immune IgG, the development of a new malaria infection by P. vivax was observed with an increasing parasitaemia that needed chloroquine treatment.

Malaria surveillance performed by our group in RO resulted in similar effects. After treating all identified symptomatic and asymptomatic P. falciparum parasite carriers in the riverside locality of Vila Candelária, no more clinical cases of falciparum malaria were observed in the following year. However, the number of clinical P. vivax cases doubled in this locality, from 53 in the previous year to 105 in the following year (Tada 2008, Tada et al. 2012). Both observations might be explained if clearing of the P. falciparum parasites results in the elimination of an "inhibitory signal", affecting the multiplication of P. vivax parasite cells, either of hypnozoite origin or from a quiescent infection.

These and other observations (Boyd & Kitchen 1938, Maitland et al. 1997) indicate that interactions between parasites affecting population densities in mixed infections do exist, but no convincing mechanisms have been found to explain the nature of the interaction process. It can be speculated that a negative inhibitory factor produced by a dominant species would have an advantageous effect to avoid competition for nutrients obtained from the host and to avoid compromising deleterious effects, which would indirectly compromise the dominant parasite survival. Such a hypothetical inhibitor could be of immunological origin. The authors of studies with semi-immune children in Papua Guinea proposed that parasitaemia are controlled in a density-dependent manner by variant antigens from one of the partners (Bruce & Day 2003). Our results suggest a possible role for PfCLAG9 antigen in which anti-PfCLAG9 antibodies originated during P. falciparum infection would recognised a vivax parasite target, possibly the PvCLAG7 ortholog. The following elements in favour of this hypothesis may be presented: (i) PfCLAG9 product seems to participate in metabolic remodelling of PRBC like it was demonstrated for PfCLAG3 (Goel et al. 2010, Nacer et al. 2011, Nguitragool et al. 2011) and suggested for PfClag9, (ii) PvCLAG7 by its similarities with PfCLAG9 seems to represent orthologs molecules (Moreno-Perez et al. 2011) and CLAG8 and (iii) search of similarities between members of the clag families of P. falciparum and P. vivax indicate that similarities between protein is good candidate for being assessed in cellular, immunological and functional studies aimed at establishing its role during invasion (Cowman et al. 2012)

While the clinical immunity status in vivax malaria correlates with repetitive infections (Alves et al. 2002, Gil et al. 2007) we hypothesise that the higher RI anti-PfCLAG9 values observed among asymptomatic P. vivax patients (Fig. 2, Table) are originate from a boosting effect by the ortholog PvCLAG7 and eight homologous sequences or, eventually, in mixed infections, directly by PfCLAG9. Immunoclinical follow up of selected mixed infected patients and immuno-parasitological studies are being undertaken in our laboratory to test this hypothesis.