Plasmodium vivax malaria is a neglected tropical disease, despite being more geographically widespread than other forms of malaria (Organization, 2018) and causes 132–391 million clinical infections each year worldwide (Price et al., 2007). Plasmodium vivax infections are most common in Southeast Asia and South America, with the Southeast Asian isolates being the most genetically diverse (Rougeron et al., 2020). Compared to P. falciparum, P. vivax has a broader temperature tolerance and an earlier onset of gametocyte development, enabling the parasites to spread through diverse climates (Elgoraish et al., 2019) and making them more difficult to control and eliminate (Lo et al., 2017). Although P. vivax infection is relatively benign, patients may experience similar clinical symptoms as P. falciparum, such as inflammatory responses, fever, and chills (Dayananda et al., 2018). The epidemiology of P. vivax malaria is further complicated by the parasite’s unique ability to form dormant-stage hypnozoites in the host liver cells, resulting in recurrent relapse infections from weeks or months to years later (White, 2011; Chu and White, 2016). Hypnozoites, which remain arrested in the liver for weeks to years, can be activated to cause new blood-stage infections and these features have substantially impacted progress in malaria control, especially in countries that are approaching elimination by providing parasite reservoirs for transmission at any time (Robinson et al., 2015; Lawpoolsri et al., 2019; Taylor et al., 2019).

A critical stage in P. vivax transmission is the development of gametocytes, known as gametocytogenesis, from trophozoites in the erythrocytic cycle. This allows Plasmodium species to be taken up by a mosquito vector, undergo sexual reproduction in the mosquito gut, and transmit to a new human host. Indeed, in the mosquito midgut, parasites can differentiate into their sexual forms, the female macrogametes and male microgametes. Gametocyte commitment is largely based on stress factors including high parasitemia, parasite strains, red blood cell density, anemia, drug treatments, and host immune responses (Smalley and Brown, 1981; Trager and Gill, 1992; Gautret et al., 1996; Talman et al., 2004; Drakeley C et al., 2006; Koepfli et al., 2015). Changes in temperature, pH, and host age can also stimulate gametogenesis in the human host (Bousema and Drakeley, 2011). When gametocytes are taken up during a mosquito’s blood meal, a number of factors including temperature, oxygen and carbon dioxide concentration, and pH can contribute to the maturation of gametocytes inside the mosquito midgut and salivary glands (Sinden, 1997). Other mosquito-derived factors such as xanthurenic acid can together activate transformation of gametocytes to male microgamete and female macrogamete within 5-10 minutes in the mosquito midgut. Although no clinical symptoms are experienced in the human hosts during gametocytogenesis, this developmental stage is critical for sexual reproduction in the mosquitoes and subsequent development of sporozoites that can infect other new human hosts. There is considerable variation in the development time of gametocytes amongst different human Plasmodium species, ranging from 7-10 days after the initial establishment of asexual parasites for P. falciparum (Gardiner and Trenholme, 2015) and 7-15 days for P. vivax (Bousema and Drakeley, 2011), indicating the need for reliable biomarkers for early-stage gametocyte detection.

A previous study has shown approximately 10% of P. falciparum and 60% of P. vivax infections have concurrent detectable low-density gametocytemia (Tadesse et al., 2017). Molecular tests for diminutive amounts of gametocytes rely on reverse-transcription polymerase chain reaction (RT-PCR) to amplify RNA transcripts of gametocyte-specifically expressed genes. Quantitative RT-PCR of targeted RNA transcripts reveals high sensitivity in detecting gametocytes of considerably low densities. For example, there are more than 106 copies of 18S rRNA transcripts per cell but only 5 copies of 18S rRNA gene per genome (Nishimoto et al., 2008). The production of high transcript copies in the parasite cells allow for greater detection limits. For P. vivax, Pvs25 and Pvs16, genes specific to the female gametocytes, are the two conventional gene markers for gametocyte detection (Wampfler et al., 2013). One gametocyte roughly corresponds to four Pvs25 transcripts per cell (Koepfli et al., 2015), and Pvs25 can detect from approximately 0.34 gametocytes per microliter of blood from P. vivax patients in Papua New Guinea (Wampfler et al., 2013) to 2 gametocytes per microliter of blood from P. vivax patients in Ethiopia (Tadesse et al., 2017). The number of Pvs25 gene transcript copies detected by qRT-PCR directly correlated with the number of mature gametocytes as well as the overall parasite densities (Bharti et al., 2006; Bousema and Drakeley, 2011) and showed a nearly normal distribution with a mean of 1.2×10⁷ copies/µL (ranging from 1.1 to 4.8×10⁸ copies/µL) blood among symptomatic P. vivax patients in northwestern Brazil (Lima et al., 2012). Such low gametocyte densities make them extremely difficult to be detected by microscopy and highlight the importance of sensitive molecular biomarkers in malaria-endemic regions.

Gametocytes are generally detected in ~20% of the infections among adults (Bousema and Drakeley, 2011), but at much higher proportions in children under the age of 12 (Nacher et al., 2004; Olliaro et al., 2016). Yet, gametocytemia in adults is up to 20-fold higher than in children (Dixon et al., 2008; Reece et al., 2009). In areas with low transmission, submicroscopic gametocytes could be hidden reservoirs for parasites with high proportions of infectious gametocytes (Hofmann et al., 2018). In Ethiopia, symptomatic P. vivax infections are nearly four times more infectious than asymptomatic ones (Tadesse et al., 2018). A recent study of 26 P. vivax samples from Cambodian patients indicated that the expression profile of 21 predicted gametocyte genes were clustered in two distinct groups (Kim et al., 2019). One group includes Pvs25, ULG8, gametocyte developmental protein 1, guanylate kinase, HMGB1, and five CPW-WPC proteins that associate with intracellular trafficking and histone remodeling in the female gametocytes. The other group includes Pvs47, Pvs48/45. Hap2, the gamete egress and sporozoite traversal protein, s16, and three CPW-WPC proteins that associate with microtubular development in the male gametocytes. It remains unclear if these male and female gametocyte genes show higher expression than the conventional marker Pvs25 and offer high detectability of total gametocyte densities (i.e., both male and female gametocytes). The detection of total gametocytes allows for robust gametocyte sex-ratio estimates in field studies given their stability under suboptimal conditions (Meerstein-Kessel et al., 2018). Gametocytes in low density infections can initiate transmission at any time, increasing the need to utilize reliable biomarkers for their detection and control. At the genomic level, polymorphisms in gametocyte-specific genes may provide information of parasite reservoirs that are transmitted from infected humans to mosquitoes, offering new insights into transmission bottlenecks in vectors. Genetic polymorphisms in gametocyte genes including Pvs25, Pvs28, Pvs48/45, and Pvs230 will also have important implications on their potential use and effectiveness as transmission blocking vaccine (TBV) candidates. Therefore, there are two key objectives in this study. First, genetic variations of 17 and 11 genes that represent female and male gametocytes, respectively, were compared among P. vivax isolates from Southeast Asia, East Africa, and South America based on whole genome sequence data. We constructed phylogenetic trees to determine clustering patterns with the goal of identifying novel DNA biomarkers with high differentiative power for gametocyte clones. Second, the expression levels of several female and male gametocytes were compared among a subset of Ethiopian P. vivax isolates based on transcriptomic data with the goal of identifying highly sensitive RNA biomarkers for both sexes of P. vivax gametocytes.

The transmission of P. vivax relies on the development of gametocytes from committed schizonts in the human hosts and uptake by the Anopheles mosquitoes. Gametocytes especially in low-density and/or submicroscopic infections are hidden parasite reservoirs that can initiate transmission at any time. Thus, it is critical to utilize reliable biomarkers for their detection and effective malaria control intervention. Pvs25 and Pvs16 are two commonly used biomarkers for P. vivax gametocyte detection (Wampfler et al., 2013). However, they represent only female gametocytes and could underestimate the total gametocytes present in an infection. The proportions of infectious gametocytes as well as the ratio of male to female gametocytes can vary across infections, and in turn, determine parasitemia in the mosquito and the transmission potential (Tadesse et al., 2019). Gametocyte sex ratio positively correlates with gametocyte density with generally higher proportions of microgametocytes especially in low-density infections, which could result in optimal fertilization and higher gametocyte densities (Schall, 2000; Talman et al., 2004). Additionally, the ability to accurately detect male and female gametocytes will enable us to understand how anti-malarial drugs affect gametocyte production and transmission in natural infections (Henry et al., 2019) and develop transmission blocking therapies. In this study, we identify one male gametocyte CPW-WPC protein gene PVP01_1119500 and two female gametocyte genes Pvs230 (PVP01_0415800) and CPW-WPC gene PVP01_0904300 that showed high expression relative to Pvs25. Both PVP01_0904300 and PVP01_1119500 belong to the CPW-WPC gene family, and Pvs230 (PVP01_0415800) is a surface protein of P. vivax female gametocytes.

Better understanding of genetic polymorphisms in gametocyte genes have important implications on their potential in the development of transmission blocking vaccine candidates. Pvs25 is a highly expressed but conserved female gametocyte gene. It is one of the leading candidates for malaria TBV based on its high immunogenicity observed in animal model studies and high conservation among P. vivax isolates from endemic regions (Feng et al., 2011; Chaurio et al., 2016). In endemic regions of the Amazon in South America, Pvs25 was shown to be conserved among P. vivax isolates (Chaves et al., 2019). Pvs25 has been reported with positive selection in a previous study (Chaurio et al., 2016). The present study indicated that Pvs25 is less polymorphic than other female and male gametocyte genes and is not neutral. P. vivax isolates from the same country were more closely related or genetically similar than from different countries. These findings suggested that selection may have contributed to the conserved nature of this gene.

Pvs48/45, a male gamete fertilization factor, is another main transmission blocking vaccine candidate (Feng et al., 2015). Our analyses indicated a relatively conserved nature of this gene among global isolates, agrees with a previous study that indicated low levels of genetic diversity in Pvs48/45 among 200 P. vivax isolates from temperate and subtropical populations in China (Feng et al., 2015). Of the 14 SNPs, 11 were nonsynonymous with an overall nucleotide diversity of 0.0012 across the isolates. At the continental level, F _ST values based on Pvs48/45 widely ranged from 0.34 to 0.90 among samples from several countries in Asia, America, and Oceania, with an overall F _ST of 0.665 (Feng et al., 2015). Though earlier studies of isolates from China showed Pvs48/45 undergoing potential positive selection with the distribution of polymorphic sites concentrated in domain II of the gene (Feng et al., 2015), the present study did not detect significant positive or non-neutral selection in this gene. High levels of antibodies against Pfs48/45 and Pfs47 were previously reported in P. vivax endemic regions (Anthony et al., 2007). Immuno-epidemiological studies of the parasite sexual stage antigens including Pvs48/45 and Pvs230 showed that antibodies against these antigens are present in people living in P. vivax endemic areas and associated with transmission blocking activities (Bousema et al., 2006; Drakeley CJ et al., 2006; Bousema et al., 2010; Feng et al., 2015; Jones et al., 2015). It is possible that these gene antigens illicit host antibody responses and impose selection, leading to scarce genetic polymorphisms in the parasite populations.

Gametocyte production is critical for sexual reproduction of the parasite and subsequent transmission to a new host. It is essential to identify gametocyte clones and the origin of transmission to effectively identify transmission reservoirs in both symptomatic and asymptomatic infections. Prior studies have demonstrated that low density P. vivax gametocytes in asymptomatic carriers significantly contribute to transmission and genetic diversity (Bousema et al., 2014; Abdelraheem et al., 2018). Lower genetic diversity in P. vivax populations is correlated with lower transmission frequencies and low amounts of imported parasites, resulting in fewer clones (Nguitragool et al., 2017; Kepple et al., 2021). Our findings showed that two female gametocyte genes Pvs230 (PVP01_0415800) and ULG8 (PVP01_1425800) were highly polymorphic compared to Pvs25, which are potential DNA markers for determining the source of P. vivax gametocytes and differentiating gametocyte clones within and between hosts. Previous studies indicated limited polymorphism in Pvs230, making this gene a potential candidate of malaria transmission blocking vaccine (Doi et al., 2011). In a study involving 112 full length Pvs230 sequences of P. vivax isolates from regions in South America, Southeast Asia, Indo-West Pacific, and Madagascar, polymorphism of this gene was much lower than the bloodstage antigen genes (Doi et al., 2011), likely due to the fact that bloodstage antigens are targets of host antibody responses and high genetic diversity in the antigen genes would facilitate host immune response evasion (Doi et al., 2011). Our findings did not support the conservative nature of Pvs230 among different geographical isolates. Instead, we detected high genetic differentiation of Pvs230 among P. vivax from the same geographic region. Our transmission network based on Pvs230 further showed as high as 13 transitions between Cambodia and Thailand, a high level of gene flow within Southeast Asia. On the other hand, the limited gene flow among the continental isolates may explain the differentiation of Pvs230. No significant positive selection was detected in Pvs230 in this study, likely because this protein located on the surface of gametocytes is involved in gamete recognition and fertilization and could be highly divergent among parasite populations (Tsaur et al., 2001; Palumbi, 2009; Doi et al., 2011). Nevertheless, the high polymorphisms observed in Pvs230 makes this gene a less viable candidate for TBV.

The CPW-WPC proteins were previously shown to be expressed on the surface of P. falciparum and P. berghei ookinetes and may play a role in transmission by the mosquito (Kangwanrangsan et al., 2013; Rao et al., 2016). In Thailand, CPW-WPC genes were detected with significant positive selection likely due to their role in mosquito transmission (Diez Benavente et al., 2017), which may raise concerns on the reliability of using CPW-WPC protein genes as biomarkers. PvULG8 (PVP01_1452800) was detected with positive selection in the Cambodian, Thailand, and Ethiopian P. vivax. The ortholog PfULG8 is a member of the P. falciparum gene family encoding the CPW-WPC proteins. An integrated transcriptomic and proteomic analysis of P. falciparum gene expression showed that transcripts from nine CPW-WPC genes predominantly accumulate in female gametocytes and are subjected to translational repression (Lasonder et al., 2016; Siciliano et al., 2017). Though these transcripts are produced in gametocytes, they are translated only in the mosquito parasite stages (Siciliano et al., 2017). Further studies examining diversity of CPW-WPC genes, especially PVP01_0904300 and PVP01_1119500, are needed to understand the full potential of these two genes for gametocyte detection. In this study, the transcriptomic data were based on a small number of Ethiopian isolates. In the future, it is important to correspond SNPs obtained from gDNA with SNPs on expressed transcripts in gametocytes and other epidemiological features such as the infectiousness and host immune responses of broader samples in order to understand better the functional significance of polymorphisms detected at the DNA level.

Conventional methods of detecting P. vivax gametocytes may underestimate the level of transmission reservoirs. A new approach is needed to better detect and monitor the transmission potential especially in countries approaching malaria elimination. The present study examined polymorphisms and resolving power of 28 male and female gametocyte genes through phylogenetic and selection testing. Given high genetic differentiation and clear clustering of gametocyte clones by Pvs230 (PVP01_0415800) and PvULG8 (PVP01_1452800), these two genes could provide a better detection method of identifying transmission reservoirs and source of transmission for symptomatic and asymptomatic malarial infections. Transcriptomic data of a few Ethiopian P. vivax samples showed a high expression of Pvs230 (PVP01_0415800) as compared to Pvs25, suggesting its potential as a sensitive RNA marker for detecting low-gametocytemia infections. It is noted that PvULG8 was detected with positive selection in the Southeast Asian and African P. vivax, and the underlying reason warrants further investigation. Future studies on gametocyte reservoirs especially in asymptomatic and/or low-density infections using Pvs230 and PvULG8 may shed light on the importance of these genes as biomarkers in the detection of gametocytes.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

The studies involving human participants were reviewed and approved by the Scientific and ethical clearance was obtained from the institutional review boards of Jimma University, Ethiopia, and The University of North Carolina, Charlotte, USA. Written informed consent/assent for study participation was obtained from all consenting heads of households, parents/guardians (for minors under 18 years old), and from individuals who were willing to participate in the study. The patients/participants provided their written informed consent to participate in this study.

Conceptualization, LG, DY, and EL. Resources, DJ, DY, and EL. Conduct experiment and data analyses, AF, DK, JW, GK, CF, and AA. Writing—original draft preparation, AF, DK, JW, GK, and EL. Writing—review and editing, DJ, LG, DY, and EL. Funding acquisition, EL. All authors have read and agreed to the published version of the manuscript.

This research was funded by National Institutes of Health, grant number R15 AI138002.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.