Objective To determine the expression of Pfg377 ortholog gene in Plasmodium vivax, and examine its correlation with mosquito infection.
Methods Seventy clinical blood samples positive for P. vivax by microscopy, were used for the mosquito infectivity assay. Infectivity to female Anopheles dirus was determined from oocyst counts. The transcripts of Pfg377 ortholog gene of P. vivax from blood samples infective and non-infective to mosquitoes were examined using quantitative real-time reverse transcriptase-polymerase chain reaction (RT-PCR).
Results Of 70 P. vivax positive blood samples, 50 (71.4%) samples were mosquito-infective and 20 (28.6%) were not. In infective samples, the expression level of Pfg377 ortholog gene was significantly higher than in the non-infective group (P <0.05). In infective samples, the expression level of Pfg377 ortholog gene at ≥100 copies/ml of blood cut-off point correlated with ≥10 oocysts/mosquito cut-off point of average oocyst numbers and with ≥50% cut-off point of per cent infected mosquitoes (Pearson’s chi-square correlation, P =0.014 and P =0.026, respectively).
Conclusion The cut-off point of the expression level of Pfg377 ortholog gene could be used to predict the infectiousness of P. vivax gametocytes leading to mosquito infection and parasite transmission in the field.
Objectif: Déterminer l’expression du gène orthologue Pfg377 de Plasmodium vivax et analyser sa corrélation avec l’infection des moustiques.
Méthodes: 70 échantillons cliniques de sang, positifs pour P. vivax par la microscopie, ont été utilisés pour le test d’infectivité des moustiques. L’infectivité des femelles d’Anopheles dirus a été déterminée à partir du dénombrement des oocystes. Les transcrits du gène orthologue Pfg377 de P. vivaxà partir des échantillons de sang infectieux et non infectieux pour les moustiques ont été examinés par la réaction en chaîne de la polymérase quantitative, couplée à la transcriptase inverse en temps réel (RT-PCR).
Résultats: Parmi les 70 échantillons sanguins positifs pour P. vivax, 71,43%étaient infectieux pour les moustiques et 28,57% ne l’étaient pas. Parmi les échantillons infectieux, le niveau d’expression du gène orthologue Pfg377était significativement plus élevé que dans les échantillons non-infectieux (p < 0,05). Dans les échantillons infectieux, le niveau d’expression du gène orthologue Pfg377 au seuil de ≥ 100 copies/ml de sang corrélait avec un seuil de ≥ 10 oocystes/moustique du nombre moyen d’oocystes et avec un seuil de ≥ 50% pour le pourcentage de moustiques infectés (corrélation chi-carré de Pearson, p = 0,014 et 0,026, respectivement).
Conclusion: Le seuil du niveau d’expression du gène orthologue Pfg377 pourrait être utilisé pour prédire l’infectivité des gamétocytes de P. vivax conduisant à l’infection des moustiques et à la transmission du parasite sur le terrain.
Objetivo: Determinar, en Plasmodium vivax, la expresión del gen ortólogo al Pfg377, y examinar su correlación con la infectividad en el mosquito.
Métodos: Para realizar el ensayo de infectividad en el mosquito, se utilizaron 70 muestras clínicas de sangre, positivas para P. vivax por microscopía. La infectividad de la hembra Anopheles dirus se determinó mediante el conteo de ooquistes. Mediante una RT-PCR cuantitativa se examinaron las copias del gen de P. vivax, ortólogo del Pfg377, provenientes de muestras de sangre tanto infecciosas como no infecciosas para los mosquitos,
Resultados: De 70 muestras de sangre positivas para P. vivax, un 71.43% eran infectivas para los mosquitos y un 28.57% no lo era. Entre las muestras infectivas, el nivel de expresión del gen ortólogo al Pfg377 era significativamente mayor que en el grupo no-infectivo (p <0.05). En muestras infectivas, el nivel de expresión del gen ortólogo al Pfg377 en el punto de corte de ≥100 copias/ml de sangre, se correlacionaba con el punto de corte de número promedio de oosquistes de ≥10 ooquistes/mosquito y con ≥50% del punto de corte de porcentaje de mosquitos infectados, p =0.014 y p =0.026, respectivamente).
Conclusión: El punto de corte del nivel de expresión del gen ortólogo de Pfg377 podría utilizarse para predecir la infectividad de los gametocitos de P. Vivax, resultando en la infección de mosquitos y la transmisión del parásito en el campo.
Plasmodium vivax causes more than half of malaria infections outside Africa, with an annual estimation of 130–435 million new cases of which 75 million have acute clinical symptoms (Mendis et al. 2001; Hay et al. 2004). Plasmodium vivax is one of the most difficult species to be eradicated due to its hypnozoite or dormancy stage in the liver, lack of a vaccine, parasite drug resistance and vector resistance to insecticides (Sina 2002).
Plasmodium vivax asexual stages cause the clinical symptoms while the sexual stages (gametocytes) are transmitted from human hosts to mosquito vectors to complete the life cycle. Transmission of P. vivax from humans to mosquitoes depends on the existence of infectious gametocytes, which are taken up by female mosquitoes during a blood meal. The presence of infectious mature gametocytes in human blood circulation determines the success of transmission. In a malaria metric survey, it is not practical to determine the infectiousness of mature gametocytes to mosquitoes by feeding mosquitoes directly on infected people or infected blood. Therefore, an indicator of infectious mature gametocytes is based solely on an analysis of gametocyte rates from Giemsa-stained blood smears for epidemiological study of malaria transmission. The probability of mosquito infection can be expected to increase proportionally with increasing gametocyte density. However, the correlation of gametocyte density with mosquito infection remains unclear. Several studies have demonstrated that there was no correlation between gametocyte and oocyst densities, as the mosquitoes could be infected with submicroscopic gametocyte densities (Graves 1980; Sattabongkot et al. 1991; Boudin et al. 1993; Coleman et al. 2004; Drakeley et al. 2006; Schneider et al. 2007). Thus, successful malaria transmission to mosquitoes not only depends on the quantity but also on quality of gametocytes. The gametocyte quality is regulated by specific genes and proteins.
To date, although mechanisms and functions driving gametocyte development are still not fully understood, a number of stage-specific genes and proteins of various sexual stages have been identified, among which is Pfg377 (Young et al. 2005). The Pfg377 is a female-specific protein involved in the formation of osmiophilic body, which is essential for the emergence of female gametocytes from red blood cells (RBCs) to allow fertilization with the short-lived male gamete in the mosquito midgut (Severini et al. 1999; de Koning-Ward et al. 2008). A recent study on P. vivax gene expression from clinical blood samples using microarray analysis revealed that Pfg377 ortholog gene expression was upregulated in zygotes (Westenberger et al. 2010). This indicates that the Pfg377 ortholog gene is involved in the parasite development in mosquitoes.
On the basis of important functions of Pfg377 in P. falciparum gametogenesis and the essential nature of Pfg377 ortholog gene on mosquito stage development, we aimed to identify the expression of Pfg377 ortholog gene in P. vivax gametocytes using quantitative real-time RT-PCR, and examine its correlation with mosquito infection. The expression of this gene may be used as an indicator of P. vivax gametocyte maturation, and thus allow prediction of mosquito infection and parasite transmission in the field.
Materials and methods
Clinical blood sample collection
Seventy heparin-treated blood samples from patients positive for P. vivax by microscopy were collected from malaria clinics in Mae Sot, Tak, Northwestern Thailand during 2002–2004. Blood samples were collected following the protocol approved by the Institutional Ethics Committee of Thai Ministry of Public Health and the Human Subject Research Review Board of The United States Army. Only patients infected with a single infection of P. vivax (as confirmed by nested PCR), with no complication, were recruited in this study. None of the patients were undergoing chemoprophylaxis when they visited the clinics. All patients signed an informed consent form before enrolment. Twenty millilitre of blood was collected by venous puncture and prepared for thick and thin blood films. White blood cells (WBCs) of the rest of the blood sample were removed using a Plasmodipur filter (Euro Diagnostica, the Netherlands). Blood samples were centrifuged at 600 g for 5 min at room temperature to remove plasma, and the RBC pellet was collected. One hundred and fifty microlitre of packed RBC was used for the mosquito infectivity assay, and the remainder was subjected to total RNA isolation followed by quantitative real-time RT-PCR.
Mosquito infectivity assay
The mosquito infectivity assay was performed using Anopheles dirus (Bangkok strain), maintained in a colony at the Armed Forces Research Institute of Medical Science (AFRIMS) in Bangkok, Thailand (Sucharit et al. 1983). In brief, 150 μl of RBC pellet from blood samples was suspended in pooled normal AB serum to a packed cell volume (PVC) of 50%. Then, 300 μl of the suspension was fed for 30 min to 100 female mosquitoes (5–7 days old) via an artificial membrane attached to a water-jacketed glass feeder maintained at 37 °C. Next, unfed mosquitoes were removed, and engorged mosquitoes were kept on 10% sucrose solution and maintained at 26 °C and 80% room humidity for 7 days. On day 7 after feeding, 10 mosquitoes were randomly selected and dissected. Their midguts were stained with 10% mercurochrome. Oocyst numbers in each midgut were counted under a stereo microscope (Arevalo-Herrera et al. 2005).
Thick blood smears were examined under 1000× magnification by an expert microscopist to identify malaria parasites. Parasite density per 500 WBCs was counted and calculated as the number of parasites/microlitre of blood using a standard value for a WBC count (8000 WBCs/μl of blood) (Moody 2002). The initial thick film was considered negative if no parasites were seen after 500 WBCs were counted. The result from conventional microscopy was later confirmed by nested PCR using primers specific for 18S rRNA genes of P. falciparum, P. vivax, P. malariae and P. ovale as described previously (Kimura et al. 1997).
Quantitative real-time RT-PCR
Total P. vivax RNA was extracted from 1 ml of RBC pellet of P. vivax-infected blood using Trizol reagent (Invitrogen, USA) according to the manufacturer’s instructions. To eliminate the parasite genomic DNA contamination, total RNA preparation was treated with RNase-free DNase I enzyme (Invitrogen) and kept at −80 °C until use.
The mRNA was converted to cDNA using SuperScript™ III First-Strand Synthesis System for Reverse Transcription-PCR (Invitrogen) as recommended by the manufacturer. The expression Pfg377 ortholog gene was determined by quantitative real-time RT-PCR using Platinum® SYBR® Green qPCR SuperMix-UDG kit (Invitrogen). Primer sets were designed based on the sequences of Pfg377 ortholog (GenBank accession no. Pvx_101400) and 18S rRNA (conserved sequences among GenBank accession no. U07367, U93095 and U07368) of P. vivax obtained from Plasmodium genome resource (http://www.plasmodb.org) and using Primer Analysis Software (http://www.genefisher.com) (Table 1). Quantitative real-time RT-PCR was carried out in 25 μl of mixture containing 3 μl of cDNA sample, 12.5 μl of Platinum® SYBR® Green qPCR SuperMix-UDG, 0.5 μl of 10 μm a gene specific reverse primer, 0.5 μl of 10 μm a gene specific forward primer, 0.5 μl of ROX reference dye and 8 μl of DEPC water. Quantitative real-time RT-PCR was performed in triplicate using Real Time Thermal Cycler, Chlomo4™ Four-Color Real-Time PCR System (Bio-Rad, USA). Amplification condition was 95 °C for 2 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. After PCR cycling, samples were heated to 95 °C for 30 s, then at 55 °C for 15 s followed by heating to 95 °C at a linear rate of 0.2 °C/s. Melting curve analysis was performed to confirm specificity of PCR product. To quantify the amount of cDNA in samples, a standard curve prepared from 10-fold serial dilutions (105, 104, 103, 102 and 10 copies of DNA) of plasmid containing the sequences of Pfg377 ortholog was used. The 18S rRNA was used to confirm the presence of P. vivax in the sample.
Table 1. Primers used for amplification of Pfg377 ortholog gene and 18S rRNA using quantitative real-time RT-PCR
Forward primers (5′–3′)
Reverse primers (5′–3′)
*Size = 320 bp.
†Size = 154 bp.
Construction of a plasmid used for construction of quantitative real-time RT-PCR standard curves
To construct a standard curve for quantitative real-time RT-PCR, a target DNA sequence of Pfg377 ortholog gene of P. vivax was amplified using PCR, and was then cloned into pCR®2.1-TOPO® TA cloning vector (Invitrogen) according to the manufacturer’s instruction. The presence of inserted DNA was confirmed using DNA sequencing (Cooperative Research Station Southeast Asia International Center for Biotechnology, Osaka University at Faculty of Science, Mahidol University, Bangkok, Thailand).
SPSS software version 12.0 was used for statistical analysis. Pearson’s chi-square correlation analysis was utilized to analyse the relationship between the expression level of Pfg377 ortholog gene and numbers of total oocysts in An. dirus. Mann–Whitney U test was employed for significance determination of differences of levels of (i) Pfg377 ortholog gene expression and (ii) parasitemia and gametocytemia between samples infective and non-infective to mosquitoes. A P-value <0.05 is considered significant for both tests.
Determination of mosquito infectivity
Of 70 clinical blood samples positive for P. vivax gametocytes by microscopy, 50 (71.4%) samples were infective to mosquitoes and 20 (28.6%) were not. The number of average oocyst in each batch of mosquitoes fed on 50 infective blood samples ranged from 0.1 to 350.8 oocysts/mosquito (median, 21.1 oocysts/mosquito) (Table 2). No oocyst was detected in 20 non-infective blood samples (Table 3). There were no significant differences in parasitemia, numbers of asexual stages and numbers of male and female gametocytes between blood samples infective and non-infective to mosquitoes (Table 4).
Table 2. Pfg377 mRNA copy numbers and oocyst numbers in 50 blood samples infective to Anopheles dirus mosquitoes
Table 4. Comparison between the parasite numbers in blood samples infective and non-infective to Anopheles dirus mosquitoes
Infective blood samples
Non-infective blood samples
Asexual stages (parasites/μl)
Male gametocytes (parasites/μl)
Female gametocytes (parasites/μl)
Total gametocytes (parasites/μl)
Detection of Pfg377 ortholog gene expression in clinical blood samples infective and non-infective to mosquitoes
Fifty mosquito-infective and 20 non-infective blood samples were examined for expression of a P. vivax gametocytes-specific gene, Pfg377 ortholog gene, using quantitative real-time RT-PCR. The level of 18S rRNA was also analysed and used as an internal control to confirm the presence of P. vivax in the samples. As there are no accepted housekeeping genes of P. vivax gametocytes that can be used as an internal control for quantitative real-time RT-PCR, a standard curve generated with a plasmid harbouring Pfg377 ortholog gene was used to quantify the expression of this gene. To prevent bias during the experimental process, the samples were treated blindly and the transcript of Pfg377 ortholog gene was determined twice. The data shown here are the average value of two independent experiments. The standard deviation of the result from each sample was close to zero. As there were no significant differences in parasitemia, numbers of asexual stages and numbers of male and female gametocytes between mosquito-infective and non-infective samples, differences in the gene expression of Pfg377 ortholog between the two groups of samples could be compared directly.
Among 50 infective samples, the transcript of Pfg377 ortholog gene was detected in 44 (88%) samples, and in the remaining, six (12%) samples were negative (Table 2). Four of these latter six samples were positive for the presence of both male and female gametocytes by conventional microscopy, and two samples were positive for only female gametocytes. Transcripts of P. vivax 18S rRNA were detected in all six samples negative for Pfg377 ortholog transcripts. This indicated that RNA extraction of these six samples had been successful and that P. vivax parasites were present in the samples. Fourteen (70%) of 20 non-infective blood samples were positive for Pfg377 ortholog transcripts (Table 3). The remaining six (30%) samples were negative for Pfg377 ortholog transcripts and were positive for 18S rRNA transcripts, indicating the presence of P. vivax in the samples.
Of 44 samples positive for Pfg377 ortholog mRNA expression, the copy number of Pfg377 ortholog mRNA varied markedly ranging from 1 to 54 349 copies (mean ± SD, 3980 ± 1552 copies). Among 20 mosquito non-infective samples, the transcript of Pfg377 ortholog gene was detected in 14 (70%) samples. Copy numbers of the Pfg377 ortholog mRNA ranged from 1 to 2190 copies (mean ± SD, 283 ± 169 copies).
There was no significant difference in total gametocyte numbers between samples infective and non-infective to mosquitoes (Figure 1a). The expression levels of Pfg377 ortholog gene in samples infective to mosquitoes are significantly higher than those of samples non-infective to mosquitoes (Mann–Whitney U test, P <0.05; Figure 1b). No correlation was observed between total gametocyte numbers and Pfg377 ortholog gene expression (Figure 2).
Correlation between Pfg377 ortholog gene expression and mosquito infection
In all 50 infective samples, 100 copies/ml of blood cut-off point of Pfg377 ortholog gene expression, 10 oocysts/mosquito cut-off point of average oocysts and 50% cut-off point of per cent infected mosquitoes were used for two-by-two tables correlation analysis. These cut-off points were chosen as the correlation of expression levels with oocyst numbers and with per cent infected mosquitoes was initially observed at these points. No correlation was observed between average oocyst numbers and parasitemia (Figure 3a). There was a weak association between average oocyst numbers and total gametocyte numbers (Pearson Correlation, P =0.009; r = 0.311) (Figure 3b). There was a correlation of Pfg377 ortholog gene with the expression level at ≥100 copies/ml of blood and average oocysts ≥10 oocysts/mosquito (Pearson’s chi-square correlation, P =0.014; Figure 3c; Table 5). A correlation was also not observed between numbers of per cent infected An. dirus and parasitemia (Figure 4a) or total gametocytes numbers (Figure 4b). A correlation was also seen in the samples with the expression level of Pfg377 ortholog gene at ≥100 copies/ml of blood and per cent infected mosquitoes ≥50% (Pearson’s chi-square correlation, P =0.026; Figure 4c; Table 5).
Table 5. Correlation analysis of blood samples with Pfg377 ortholog gene transcripts at ≥100 copies/ml of blood cut-off point and average oocyst numbers at ≥10 oocysts/mosquito cut-off point or per cent infected mosquitoes at ≥50% cut-off point
An index of malaria transmission is usually measured by an analysis of gametocyte rates from blood films during a malariometric survey. However, this method is inaccurate for determining parasite infectiousness, as shown by the lack of the correlation between P. vivax gametocyte and oocyst densities in An. dirus mosquitoes (Sattabongkot et al. 1991). Consistently, this study clearly showed that there was no significant difference in a gametocyte density of P. vivax between samples infective and non-infective to mosquitoes, and there was a weak correlation between a gametocyte density from blood films and an oocyst density. These findings strongly suggest that P. vivax gametocyte numbers determined by microscopy cannot be used for accurately predicting the successful transmission of malaria parasites from humans to mosquitoes. This may be due, in part, to submicroscopic levels of gametocytes (Schneider et al. 2007). Moreover, no correlation was found between parasitemia and an oocyst density and between parasitemia and the percentage of infected mosquitoes. This implies that the level of parasitemia has no effect on mosquito infection. In this study, we evaluated the feasibility of using the detection of Pfg377 ortholog gene transcripts to evaluate mosquito infectivity of blood samples obtained from patients in a P. vivax-endemic area of Thailand.
In P. falciparum, the expression of Pfg377 gene is known to be predominately in mature stages of the gametocyte, which causes parasite transmission to mosquitoes (Severini et al. 1999; de Koning-Ward et al. 2008). Our findings demonstrated that the mRNA expression level of Pfg377 ortholog gene in P. vivax gametocytes was higher in infective than in non-infective blood samples. This suggests that elevated transcription of this gene is an indicative of the presence of fully mature gametocytes, which are capable of forming gametes and undergoing fertilization and further developing to oocysts in mosquitoes.
In infective blood samples, there was a significant correlation of the expression level of Pfg377 ortholog gene at ≥ 100 copies/ml of blood cut-off point with ≥10 oocysts/mosquito cut-off point of average oocysts and with ≥ 50% cut-off point of per cent infected mosquitoes. Our study also revealed an absence of mRNA expression of Pfg377 ortholog gene in six (12%) infective blood samples. The correlation of the expression level of Pfg377 ortholog gene with oocyst numbers in mosquitoes and with per cent infected mosquitoes in infective blood samples is not quite straightforward. This may be due to a genome diversity of P. vivax in natural infection (Feng et al. 2003), which could result in variations of the target sequence. Thus, new primer sets for this gene would be required to amplify regions that are universally conserved among all P. vivax genotypes to obtain more reliable results. Moreover, more mosquitoes should be dissected for each blood sample to have more precise numbers of the oocysts.
In summary, this is the first study to demonstrate that Pfg377 ortholog gene is highly expressed in P. vivax-blood samples infective to mosquitoes. The cut-off point of the expression level of this gene from infective samples could be used to predict the infectiousness of P. vivax gametocytes in infectious reservoirs of a community. The findings in this study should be useful for implementing malaria surveillance measures.
We thank staffs at the Entomology Department, AFRIMS and the malaria clinic in Mae Sot, Tak province, Thailand for technical support in blood sample collection and microscopic examination of malaria parasites. We also thank Professor Prapon Wilairat for the proof reading of this manuscript. This work was supported by Joint SEARO-TDR Small Grants Program, World Health Organization and US Military Infectious Diseases Research Program. The view of the authors does not purport to reflect the position of the US Department of the Army or Department of Defense.