Comprehensive proteomics investigation of P. vivax-infected human plasma and parasite isolates

P. vivax malaria is a public health concern in the tropical and sub-tropical regions of the world. Reflecting this, the estimated vivax malaria burden, in 2017, was around 7.5 million cases worldwide [1] with more people living at risk. The last few years have witnessed a surge in the number of severe vivax malaria cases [2, 3], posing a serious threat to health and economy. The species has unique biological features that support its survival in different climates and geographical regions. An important characteristic is the appearance of gametocytes before the onset of illness which allows transmission even before patients receive treatment [4]. Secondly, the presence of quiescent liver stage forms called hypnozoites trigger multiple episodes of relapse infection from a single infectious mosquito bite [5, 6]. Lower parasite densities and greater vectorial capacity are other features of P. vivax that make it different from P. falciparum. Therefore, it is important to design and implement interventions specifically targeting P. vivax for its control and elimination [7].

Proteomics offers a major advantage in the discovery of new target biomolecules because of its ability to simultaneously characterize proteomes and sub-proteomes [8] without any prior knowledge of the nature of proteins. It also offers great promise in studying protein expressions, protein interactions and protein modifications [9]. Mass-spectrometry (MS) qualifies as an excellent tool for the identification and quantification of differentially regulated proteins that provide information about altered pathways and deeper insights into disease biology. Global and stage-specific mass spectrometric analyses of P. falciparum in culture have revealed important features of the parasite’s biology [10] including sexual stages [11]. In 2012, Ray and coworkers applied a quantitative proteomics approach to identify differentially expressed host serum proteins in vivax and falciparum malaria patients. The same group also identified host markers of severe infection [12, 13]. Around the same time, Bachmann et al., reported potential muscle damage and microvasculature lesions during the course of cerebral malaria based on elevated muscle protein levels in the plasma of children with cerebral malaria [14]. Recently, an MS-based proteomics study comparing 9 complicated malaria (CM) and 10 uncomplicated malaria (UM) patient samples reported the association of selected P. falciparum proteins with the pathophysiology of cerebral malaria [15].

Unlike P. falciparum, scientific progress towards understanding P. vivax has been minimal, partly due to the inability to maintain continuous cultures for its propagation. Although efforts in this direction have been promising [16] a standard in vitro culture for P. vivax has not yet been established. Hence, scientists have had to rely on clinical samples and animal models [17‐21] for the study of P. vivax, further delaying the identification of new molecules and their functional characterization. Very recently, the proteomes of plasma-derived exosomes from P. vivax-infected FRG huHep mice were analyzed to study liver-stage expressed markers of infection. The study revealed human arginase-I and an uncharacterized P. vivax protein as potential markers for hypnozoite infection [22]. Proteomics technologies have also facilitated the identification of diagnostic and therapeutic targets. A recent study, investigating plasma samples from falciparum malaria patients, reported four parasite-specific enzymes namely, pHPRT (PF10_0121, parasite hypoxanthine phosphoribosyltransferase,), pPGM (PF11_0208, parasite phosphoglycerate mutase), pLDH (PF13_0141, lactate dehydrogenase) and pFBPA (PF14_0425, fructose bisphosphate aldolase) as possible diagnostic biomarkers [23]. On the contrary, there are no published reports for P. vivax malaria, despite a pressing need for new P. vivax diagnostic antigens.

Here, we describe a comprehensive proteomics investigation of P. vivax, exploring both human plasma and clinical parasite isolates from whole blood. We aim to identify secretory parasite proteins, proteins released upon erythrocyte rupture and abundant proteins expressed by the parasite during their blood-stages within the human host. Our findings provide deeper insights into parasite biology and expand our knowledge of the P. vivax clinical proteome. This work also reveals parasite proteins with diagnostic potential and represents an important step in the development of P. vivax detection tools. A method for performing validation experiments using targeted proteomics assays is also described.

Plasma mirrors the pathophysiological state of any diseased individual [24]. In case of malaria, the study of human plasma is significant because RBCs represent sites of maximum parasite activity. Parasites grow and multiply within RBCs and subsequently rupture them to enter the bloodstream and invade other uninfected RBCs. In the process, several proteins are released into circulation. Identifying the presence of these circulating parasite proteins in plasma is crucial in the context of parasite biology and disease pathogenesis. Some proteins may also be important diagnostic antigens, if they can be easily detected and measured. However, the complex nature of human plasma and presence of several abundant host proteins, mask their detection.

Recent advancements in mass spectrometry provide a sensitive platform for studying complex proteomes, but the dynamic range of plasma poses a serious challenge to MS-based proteomics. So far, only a few thousand human proteins have been discovered by LC-MS/MS, while parasite proteins are completely masked by the abundant human proteins present in concentrations that range from 50 mg/mL to as low as 5 pg/mL [25]. In this study, plasma samples from VM patients were first depleted to eliminate abundant host proteins like serum albumin, apolipoproteins, immunoglobulins, haptoglobin, fibrinogen, macroglobulin, transferrin and a few others. The number of P. vivax proteins obtained after MS analysis varied largely among the 12 patients. Nine patients showing only 1 or 2 parasite proteins in their plasma were grouped together (Group A). The other 3 patients (Group B) displayed a ratio distortion with significantly higher parasite proteins accompanied by a surprisingly lower number of human proteins. Unfortunately, these disparate observations could not be correlated to the phenotypes of health and disease as clinical information for these patients was not available. The most important data missing was the parasite density for each sample which could have explained the differences observed between group A and group B samples.

Several hypothetical proteins were identified in human plasma among other exported proteins with unknown functions. According to a comparative genome study published few years ago, almost half the genes known to have orthologs in P. falciparum, P. knowlesi and P. yoelii, encode conserved hypothetical proteins [26]. Therefore, their detection in large numbers was expected. Unfortunately, due to lack of P. vivax functional assays, hypothetical proteins along with several others remain functionally uncharacterized even today. Heat shock proteins (HSPs, molecular chaperons) constituted the third class of proteins in human plasma. In P. falciparum malaria, HSP70 was previously shown to mediate protective immunity [27]. Recently, PvHSP70 was also characterized and evaluated for its serodiagnostic applicability [28]. We speculate that its presence in plasma, as determined from this study, could be one reason for the high seroreactivity observed among malaria positive patients in previous studies. The detection of cytoskeletal, ribosomal, and nuclear proteins majorly reflects parasite lysis during infection as most of these proteins are not secretory in nature. On the contrary, several glycolytic enzymes that were found in circulation may indeed be secreted during different stages of the lifecycle or released upon invasion or erythrocyte rupture. Glycolytic proteins such as PGK (phosphoglycerate kinase), Protein disulfide isomerase and GAPDH (Glyceraldehyde 3-phosphate dehydrogenase), previously reported in the context of malaria [29‐31] were detected in plasma with high confidence. Interestingly, a recent study explored the role of GAPDH as a new malaria diagnostic biomarker for P. falciparum. The authors reported high antibody levels against epitopes specific for P. falciparum [31]. Apart from these proteins, few others like PHIST, ETRAMP (early transcribed membrane protein), ran binding protein 1, Pvstp1, skeleton-binding protein 1 and cytoplasmic and nuclear enzymes were found. More importantly, four Plasmodium exported proteins, two hypothetical proteins, Pvstp1 and one protein of the Pv-fam-d family were found to be unique to P. vivax. Although these proteins represent good leads, it is important to mention that we were unable to detect many other parasite proteins present at extremely low concentrations, despite using highly sensitive MS technologies.

In order to improve the parasite proteome coverage, alternative methods to overcome the existing challenges in plasma biomarker discovery must be explored. One strategy to greatly enhance the in-depth analysis of plasma proteomes could be the use of fractionation methods and other protein separation techniques such as SDS-PAGE prior to MS analysis, Alternate strategies involve the use of Data independent acquisition (DIA), a superior technique in MS which fragments every single peptide in a sample, unlike DDA. This technique permits an unbiased acquisition of data and provides larger number of peptides with greater reproducibility [32]. Other alternatives to improve plasma proteome coverage for biomarker discovery have been extensively described by Geyer and coworkers in their article [25].

A comprehensive analysis of 14 parasite isolates using advanced Orbitrap technology revealed many novel proteins that have never been identified previously from clinical samples. Very interestingly, ribosomal subunit proteins, integral membrane proteins, proton transporting ATPase complexes formed a major group of highly expressed parasite proteins which were not found in plasma. Of particular interest were several members of the PvTRAg (Pv-fam-a) gene family which could not be identified previously using less sensitive mass spectrometers. Consistent with previous findings, Plasmodium exported proteins and hypothetical proteins were found to be highly abundant. Surprisingly, none of the merozoite surface proteins (MSPs), except MSP 8, were detected in both parasite isolates and plasma. Twenty-six parasite proteins were identified in both plasma and parasite isolates, of which 5 were unique to P. vivax. Pvstp1 was particularly interesting because it was detected in all 14 parasite isolates as well as in plasma. Further evaluation of these proteins as potential markers for P. vivax because of their specificity and high abundance (based on frequency of occurrence in vivax patients) is highly recommended. A preliminary validation of our findings was performed by adopting a targeted-proteomics based approach (MRM assays) for selected peptides. Reliable assays were successfully generated that could confidently identify target peptides in clinical samples. While this opens up new avenues for parasite biomarker validation in clinical samples, we are still far away from understanding the effect of variable criteria on the quality of results. To address the problems associated with the interpretation of Targeted MS assays, it is important to apply strict procedures and guidelines to establish more confidence in the data [33]. To overcome some of these challenges, we intend to include synthetic peptides in our assays, using concentrated samples in future.