A Drug Repurposing Approach Reveals Targetable Epigenetic Pathways in Plasmodium vivax Hypnozoites

Abstract

Radical cure of Plasmodium vivax malaria must include elimination of quiescent ‘hypnozoite’ forms in the liver; however, the only FDA-approved treatments are contraindicated in many vulnerable populations. To identify new drugs and drug targets for hypnozoites, we screened the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library and a collection of epigenetic inhibitors against P. vivax liver stages. From both libraries, we identified inhibitors targeting epigenetics pathways as selectively active against P. vivax and P. cynomolgi hypnozoites. These include DNA methyltransferase (DNMT) inhibitors as well as several inhibitors targeting histone post-translational modifications. Immunofluorescence staining of Plasmodium liver forms showed strong nuclear 5-methylcystosine signal, indicating liver stage parasite DNA is methylated. Using bisulfite sequencing, we mapped genomic DNA methylation in sporozoites, revealing DNA methylation signals in most coding genes. We also demonstrated that methylation level in proximal promoter regions as well as in the first exon of the genes may affect, at least partially, gene expression in P. vivax. The importance of selective inhibitors targeting epigenetic features on hypnozoites was validated using MMV019721, an acetyl-CoA synthetase inhibitor that affects histone acetylation and was previously reported as active against P. falciparum blood stages. In summary, our data indicate that several epigenetic mechanisms are likely modulating hypnozoite formation or persistence and provide an avenue for the discovery and development of improved radical cure antimalarials.

One-Sentence Summary:

Drug repurposing screens reveal several epigenetic inhibitors as active against P. vivax hypnozoites demonstrating that epigenetic pathways play a central role in hypnozoite quiescence.

Introduction

Of the six species of Plasmodium that cause malaria in humans¹, Plasmodium vivax is the most globally widespread². Vivax malaria now accounts for the most malaria episodes in countries with successful falciparum malaria control programs³. Controlling vivax malaria is complicated by the ability of P. vivax sporozoites, the infectious stage inoculated by mosquitoes, to invade hepatocytes and become quiescent^4,5. These quiescent ‘hypnozoites’ persist, undetectable, for months or even years before resuming growth and initiating a ‘relapse’ blood stage infection, leading to subsequent transmission back to mosquitoes⁶. New evidence suggests this transmission is expedited and silent as P. vivax liver merozoites can immediately form gametocytes instead of first having to establish an asexual stage blood infection, such as is the case for P. falciparum^7–10. Clinically, a compound with radical cure efficacy is one that removes all parasites from the patient, including hypnozoites in the liver¹¹.

Hypnozoites are refractory to all antimalarials except the 8-aminoquinolines, which were first identified over 70 years ago using low-throughput screening in avian malaria models¹². Primaquine was the first 8-aminoquinoline widely-used for radical cure, however, efficacy is contingent on a large total dose administered in a 7–14 regimen, leading to adherence problems and infrequent use in malaria control programs of endemic countries¹³. Tafenoquine-chloroquine was developed from primaquine as an improved single-dose for radical cure¹⁴, but a recent clinical trial shows tafenoquine lacks efficacy when co-administered with the common antimalarial dihydroartemisinin-piperaquine, calling into question tafenoquine’s suitability in areas of high chloroquine resistance¹⁵. Furthermore, 8-aminoquinolines cannot be administered to pregnant women or glucose-6-phosphate dehydrogenase-deficient individuals and are ineffective in persons with specific cytochrome P450 genotypes¹⁶. For these reasons, the discovery and development of new chemical classes with radical cure activity are needed¹⁷.

Modern drug discovery typically relies on phenotypic screening and protein target identification¹⁸. For malaria, this approach ensures hits are acting on parasite targets and enables rational drug design, leading to several promising novel classes of antimalarials^19,20. However, due to lower cost and higher feasibility, current high throughput screening for new antimalarials focuses almost entirely on blood or liver schizonts^21,22. High throughput antimalarial screening with a target chemo-profile for killing hypnozoites has only recently been made possible with the introduction of cell-based phenotypic screening platforms featuring a monolayer of hepatocytes infected with sporozoites, a portion of which go on to form hypnozoites²³. While the first hypnozonticidal hits from these platforms are just now being reported²⁴, protein target identification approaches for hypnozonticidal drug discovery are in their infancy as the transcriptome of hypnozoites has only recently been reported and robust methods for genetic manipulation of P. vivax are still underdeveloped^25,26.

To address the lack of radical cure drug leads and targets, we used our advanced P. vivax liver stage platform to first screen the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library⁷. This library consists of approximately 12,000 developmental, approved, and discontinued drugs with the expectation that the repurposing of compounds with some optimization or regulatory success could expedite the decade-long path drugs typically progress through from discovery to licensure²⁷. To accomplish this screen, we assembled an international collaboration with laboratories in malaria-endemic countries whereby vivax-malaria patient blood was collected and fed to mosquitoes to produce sporozoites for infecting primary human hepatocytes (PHH) in screening assays performed on-site. Interestingly, two structurally related compounds used for treating hypertension, hydralazine and cadralazine, were found effective at killing hypnozoites. Because these inhibitors have been shown to modulate DNA methylation^28,29, we pursued and confirmed the existence of methyl-cytosine modifications in P. vivax sporozoite and liver stages. Having found in the ReFRAME screen a class of hits targeting an epigenetic pathway, we decided to confirm the importance of epigenetics in P. vivax hypnozoites and screened an additional commercial epigenetic inhibitor library using an improved version of our screening platform. Hypnozoites were found to be susceptible to several classes of epigenetic inhibitors, including six distinct histone deacetylase inhibitors and two inhibitors targeting histone methylation. To further assess the importance of histone acetylation in P. vivax liver stages, we tested inhibitors previously reported to be directly acting on P. falciparum acetyl-CoA synthetase, thereby modulating the pool of acetyl-CoA available for histone acetylation³⁰. We found MMV019721 selectively kills P. vivax and P. cynomolgi hypnozoites, implicating acetyl-CoA synthetase is an additional hypnozonticidal drug target. This work demonstrates that in lieu of traditional molecular biology methods, our screening platforms identify multiple, druggable epigenetic pathways in hypnozoites and adds to the growing body of evidence that epigenetic features underpin biology in P. vivax and P. cynomolgi sporozoite and liver stages^25,31–33.

Results

ReFRAME library screening cascade, hit identification, and confirmation

Chemical biology approaches have shown that hypnozoites become insensitive to most legacy antimalarials after 5 days in culture, indicating they must mature following hepatocyte infection^24,34. Hypnozoite maturation was also noted in recent single-cell transcriptomic analyses of P. vivax liver stages, which demonstrate distinct population clusters of maturing and quiescent hypnozoites^10,25. Importantly, discovery and development of hit compounds with radical cure activity in vivo, which includes elimination of hypnozoites in the liver of malaria patients¹¹, requires screening against mature hypnozoites in vitro³⁵. While our 8-day P. vivax liver stage platform, in which sporozoites are infected into primary human hepatocytes (PHH) and then allowed to mature for 5 days before being treated with test compound³⁶, has been used for screening small libraries against mature hypnozoites²⁴, the size of the ReFRAME library (12,823 compounds tested at 10 μM) presented a logistical challenge. We anticipated that dozens of P. vivax cases, each with a unique genetic background, would be needed to produce the sporozoites required to screen the 40 microtiter plates containing the library. To preclude the complex process of regular international shipments of infected mosquitoes, the P. vivax liver stage platform was successfully adapted and set up in research labs in two distinct malaria endemic areas, the Shoklo Malaria Research Unit (SMRU) in Thailand and the Institute Pasteur of Cambodia (IPC). The screening library was also divided between both sites to enable concurrent progress; ultimately 36 P. vivax cases from either site were needed to complete the primary screen over the course of 18 months (Figs. 1A, S1).

Fig. 1. Hypnozonticidal hit detection and confirmation.

(A) Index chart depicting the primary screen of the ReFRAME library against P. vivax hypnozoites in an 8-day assay. Hypnozoite counts were normalized by mean quantity per well for each plate (Z factor). Teal: library, black: DMSO, red: 1 μM monensin. (B) Primary screen hits were confirmed by dose-response in 8-day P. vivax liver stage assays and counterscreened against P. berghei liver schizonts, P. falciparum asexual blood stages (strain Dd2), HEK293T, and HepG2. Values represent pEC₅₀ or pCC₅₀ ± SD of all independent experiments (n=2–6) for which a pEC₅₀ or pCC₅₀ was obtained. (C) Dose-response curves for cadralazine against P. vivax and P. cynomolgi liver forms in 8-day assays at the IPC, UGA, and NITD. (B,C) Heat maps represent red as more potent and blue as inactive at highest dose tested. Asterisk (*) indicates only one independent experiment resulted in a calculated pEC₅₀ or pCC₅₀ (pEC₅₀ is the inverse log of potency in M concentration, e.g. pEC₅₀ 3 = 1 mM, pEC₅₀ 6 = 1 μM, and pEC₅₀ 9 = 1 nM). (C) All replicate wells were plotted together from all independent experiments (n=3 for P. vivax at IPC, n=1 for P. vivax at NITD, n=2 for P. cynomolgi at UGA, and n=4 for P. cynomolgi at NITD), bars represent SEM.

Some hits exhibited moderate selectivity and potency, with pEC₅₀’s ranging from 5.42–7.07 (pEC₅₀ is the inverse log of potency in M concentration, e.g. pEC₅₀ 3 = 1 mM, pEC₅₀ 6 = 1 μM, and pEC₅₀ 9 = 1 nM) (Figs. 1B, S2). Colforsin daropate, rhodamine 123, and poziotinib are used to treat cancer and have known human targets, indicating that the targeted host pathways may be critical for hypnozoite persistence. As an example, poziotinib inhibits HER2, a tyrosine protein kinase associated with the downregulation of apoptosis and metastasis³⁷. We recently reported that host apoptotic pathways are downregulated in P. vivax-infected hepatocytes²⁵. Poziotinib could therefore act by upregulating apoptotic pathways in infected host cells. MS-0735, an analog of our previously reported hypnozonticidal hit, MMV018983²⁴, is a ribonucleotide-reductase (RNR) inhibitor and used as an antiviral. The apparent need for nonreplicating hypnozoites to produce deoxyribonucleosides for DNA synthesis is peculiar. However, it has been reported that RNR is also critical for DNA damage repair³⁸, is important for maintaining cancel cell dormancy³⁹, and is expressed in P. vivax liver schizonts and hypnozoites²⁵. We also rediscovered previously-reported hypnozonticidal compounds included in the library, including the ionophore narasin²⁴ and the 8-aminoquinoline plasmocid⁴⁰.

From our analysis of primary screen activity, we noted several hydrazinophthalazine vasodilators were potentially active (Fig. S1C). We selected 72 compounds for confirmation of activity against hypnozoites in a dose-response format, including 10 hydrazinophthalazine analogs. These compounds were counter-screened for additional antimalarial activity against P. falciparum blood stages and P. berghei liver schizonts. They were also tested for cytotoxicity against HEK293T and HepG2 human cell lines (Figs. 1B–C, S2). We confirmed that three hydrazinophthalazines analogs--cadralazine, pildralazine, and hydralazine--were active against mature hypnozoites, with cadralazine displaying the best combination of potency (pEC₅₀ = 6.33 ± 0.33), maximal inhibition near 100%, and selectivity over PHH (> 21 fold), HEK293T (> 85 fold), and HepG2 (> 79 fold) cells (Figs. 1, B–C, S2A). Hydralazine, which was FDA-approved in 1953, is currently one of the world’s most-prescribed antihypertensives, and on the WHO list of essential medicines⁴¹. Cadralazine, which was developed in the 1980’s as an improvement over hydralazine, was abandoned due to side effects and only licensed in Italy and Japan⁴². Hydrazinophthalazines have been shown to inhibit human DNA methyltransferases (DNMT)^28,29 and hydralazine has also been recently used to study potential DNA methylation patterns in the P. falciparum asexual blood stages⁴³. Similar to our previous report⁴³, these hydrazinophthalazines were inactive when tested against P. berghei liver schizonts and P. falciparum asexual blood stages, suggesting that hypnozoite quiescence may be biologically distinct from developing schizonts²⁴. While hydrazinophthalazines may act on infected hepatocytes and not directly on the parasite, their distinct selectivity suggests that their effect is likely on a host or parasite pathways and not simply due to cytotoxicity in the host cell. Hydralazine and cadralazine were not identified as potential hits in any of the 112 bioassay screens of the ReFRAME published to date⁴⁴, suggesting these compounds specifically target P. vivax liver stages and not promiscuously active compounds.

Methods for the robust culture of P. vivax hypnozoites were only recently reported, leading to several new reports on hypnozoite biology and radical cure drug discovery^7,45. Consequentially, some hypnozoite-specific discoveries appear to be platform-specific^10,25. Select hits were shared with the Novartis Institute for Tropical Diseases (NITD), where the activity and potency of cadralazine (pEC₅₀ = 6.09 ± 0.45), hydralazine (pEC₅₀ = 6.20), and poziotinib (pEC₅₀ = 6.17) were independently confirmed in a similar 8-day P. vivax screening platform using a P. vivax case from southern Thailand. (Figs. 1C, S3A). Independent confirmation of these hits indicates their activities are not merely platform-specific and are, rather, more broadly descriptive of hypnozoite chemo-sensitivity.

Following our screening and hit confirmation, we investigated the potency, in vivo stability, and tolerability profile of our confirmed hits and chose cadralazine and hydralazine for repurposing as radical cure antimalarials. Currently, the gold-standard model for preclinical assessment of in vivo anti-relapse efficacy is rhesus macaques infected with Plasmodium cynomolgi, a zoonotic, relapsing species closely related to P. vivax⁴⁶. Because we found cadralazine substantially more potent against hypnozoites than hydralazine, it was selected for a rhesus macaque pharmacokinetic study in which plasma levels were measured over 24 h following an oral dose of 1 mg/kg, which was calculated to be well-tolerated, and 30 mg/kg, which was calculated to likely cause drug-induced hypotension^47–49. The 30 mg/kg dose resulted in maximum plasma concentration of 13.7 μg/mL (or 48.2 μM) and half-life of 2.19 ± 0.24 h, which was sufficient to cover the in vitro EC₉₀ for several hours without noticeable side effects. (Fig. S4). As another prerequisite for in vivo validation, we next sought to confirm and measure the potency of cadralazine and other ReFRAME hits against P. cynomolgi B strain hypnozoites in vitro using an 8-day assay featuring primary simian hepatocytes (PSH) at NITD. While poziotinib was active against P. cynomolgi hypnozoites when tested two of three different PSH donor lots (pEC₅₀ = 5.67 and 5.95), hydralazine and cadralazine were found inactive when tested in all three different PSH donor lots (Figs. 1C, S3B). This negative result was later confirmed in an 8-day, simianized version of the platform at the University of Georgia (UGA) using the P. cynomolgi Rossan strain infected into two different PSH lots (Fig. 1C). Altogether these data highlight potential difference between P. vivax and P. cynomolgi and challenge the gold-standard model for preclinical assessment of in vivo anti-relapse efficacy is rhesus macaques.

Synergy between cadralazine and 5-azacytidine

As molecular tools to validate drug target in P. vivax are limited, we further confirmed the possible mechanism of action of hydrazinophthalazines using drug combination studies to assess synergy, additivity or antagonsim³⁰. We used 5-azacytidine, a known DNA methyltransferase inhibitor⁵⁰, to investigate its effects on cadralazine treatment. When tested alone in dose-response from 50 μM, 5-azacytidine had no effect on hypnozoites. However, when added to cadralazine in fixed ratio combinations ranging from 8:1 to 1:8, 5-azacytdine increased the potency of cadralazine by ~2 fold across several combinations in two independent experiments (Figs. 2, S5). The most potent effect was detected using a 2:1 fixed ratio of cadralazine:5-azacytidine, resulting to an equivalent EC₅₀ decrease from 470 nM to 216 nM.

Fig. 2. Synergistic effect of cadralazine and 5-azacytidine in P. vivax liver stage assays.

(A) Isobologram of cadralazine and 5-azacytidine activity against hypnozoites in fixed ratios of 1:0, 8:1, 6:1, 4:1, 2:1, 1:1, 1:2, 1:4, 1:6, 1:8, and 0:1, bars represent SD of FICs from two independent experiments. (B) Dose-response curves for cadralazine at the most synergistic fixed ratios (2:1, 4:1, and 8:1) against hypnozoites. Cadralazine alone is represented as 1:0, 5-azacytidine alone is represented as 0:1 and plotted on the cadralazine chart for comparison. Left and right charts represent two independent experiments, bars represent replicate wells at each dose.

Immunofluorescent detection of DNA methylation in P. vivax and P. cynomolgi liver stages

To further confirmed that cadralazine interacts with P. vivax target(s), we aimed to detect and quantify DNA methylation in the P. vivax and P. cynomolgi genomes. Previous studies had identified the presence of low level 5-methylcytosine (5mC), 5-hydroxmethylcystosine (5hmC), and 5hmC-like marks throughout the genome^43,51,52. We first conducted an immunofluorescence staining assay using commercially available anti-5mC and anti-5hmC monoclonal antibodies to identify evidence of DNA methylation in P. vivax liver stages at 6 days post-infection. We found clear evidence of 5mC, but not 5hmC, in both schizonts and hypnozoites, morphologically consistent with the presence of 5mC in the parasite’s nucleus⁷ (Figs. 3A, S6–8). To segregate signals coming from the host hepatic nuclei, we used automated high content imaging analysis on hundreds of individual P. vivax liver stage parasites as an unbiased approach for quantifying 5mC signal within parasites. Image masks were generated to quantify the area of 5mC or 5hmC stain within each parasite (Fig. S9). The values were then plotted as stain area per hypnozoite or per schizont (Fig. 4B). While some evidence of 5hmC-positive forms did appear from this analysis, the net 5hmC area per parasite was found significantly lower when compared to 5mC signals (Kurskal-Wallis tests, for hypnozoites H(7) = 194.3, p <.0001, for schizonts H(7) = 88.66, p <.0001). Similar results on the ratio of 5hmC to 5mC were also recently reported in P. falciparum blood stages⁵³, confirming that 5mC marks are the predominate DNA methylation marks in both species.

Fig. 3. Cytosine modifications in P. vivax liver forms.

(A) Immunofluorescent imaging of a 5mC-positive (left) or 5hmC-negative (right) P. vivax hypnozoite (top) and schizont (bottom) at day 6 post-infection. White arrows indicate hepatocyte nuclei positive for 5mC or 5hmC. Bars represent 10 μm. (B) High-content quantification of 5mC or 5hmC stain area within hypnozoites or schizonts from sporozoites generated from three different P. vivax cases. Significance determined using Kruskal-Wallis tests, for hypnozoites H(7) = 194.3, p <.0001, for schizonts H(7) = 88.66, p <.0001, with Dunn’s multiple comparisons, *p <.05, ***p <.001, ****p <.0001, ns = not significant. Line, box and whiskers represent median, upper and lower quartiles, and minimum-to-maximum values, respectively, of all hypnozoites (177 ≤ n ≤ 257) or all schizonts (30 ≤ n ≤ 142) in culture for each case, 2’ indicates a secondary stain only control.

Fig. 4. Density of cytosine and methylated cytosine (5mC) in sporozoites.

(A) CG content of chromosome 1 for P. vivax and P. cynomolgi. The total number of cytosines was quantified on each strand using 1 kbp long non-overlapping windows. (B) The total number of methylated cytosines was quantified on each strand using 1 kbp long non-overlapping windows. (C) The number of 5mC present in all possible contexts (CG, CHG, and CHH) quantified throughout the genome of P. vivax and P. cynomolgi. (D) Repartitioned 5mC quantity within different compartments of the genome in P. vivax and P. cynomolgi. (E) Strand-specificity of 5mC for all genes in P. vivax and P. cynomolgi. Flanking regions and gene bodies were divided into five bins and the methylation level of each bin was averaged among all genes. Red: template strand, blue: non-template strand. (F) The previously reported mRNA abundance of P. vivax sporozoites was retrieved²² and genes ranked. The 5mC levels in 5’ flanking regions, gene bodies, and 3’ flanking regions were placed into five bins and are shown for highly expressed (90th percentile, left) and weakly expressed (10th percentile, right) genes. Red: template strand, blue: non template strand.

Given the different susceptibility of P. cynomolgi hypnozoites to hydrazinophthalazines as compared to P. vivax, we performed automated high content analysis of 5mC- and 5hmC-stained P. cynomolgi M/B-strain liver schizonts and hypnozoites at 8 and 12 days post-infection. Like P. vivax, we found both P. cynomolgi liver schizonts and hypnozoites are positive for 5mC, but not 5hmC. However, the 5mC stain morphology and intensity were relatively lower in P. cynomolgi hypnozoites versus P. vivax hypnozoites, suggesting potential divergence of DNA methylation pathways in these two species (Fig. S10).

Detection of cytosine modifications in P. vivax and P. cynomolgi sporozoites using liquid chromatography-tandem mass spectrometry and bisulfite sequencing

We next sought to confirm the presence of cytosine methylation in the P. vivax and P. cynomolgi genomes using mass spectrometry and bisulfite sequencing. We initially assessed that without an available single cell sequencing approach, sequencing coverage of the parasite’s genome would be overwhelmed by the genomic material from the host cell as well as neighboring uninfected hepatocytes²⁵. We therefore collected sufficient genomic material from P. vivax and P. cynomolgi sporozoites to analyze the nucleoside mixture arising from the enzymatic digestion of genomic DNA by liquid chromatography-tandem mass spectrometry as well as for detection of DNMT activity using a commercial in vitro DNA methylation assay⁴³. While we detected 5mC and DNMT activity in Plasmodium-enriched samples with these approaches, possible contamination by the mosquito’s microbiota could not be excluded (Fig. S15A–B). We next analyzed DNA methylation loci at single-nucleotide resolution using bisulfite sequencing of 3×10⁷ P. vivax sporozoites, generated from three different cases, as well as 3×10⁷ P. cynomolgi sporozoites (Fig. 4, A–B). A total of 161 and 147 million high-quality reads were sequenced for P. vivax and P. cynomolgi samples, respectively (Fig. S11C). The average 5mC level detected across all cytosines was 0.49% and 0.39% for P. vivax and P. cynomolgi, respectively. These percentages are comparable to the 0.58% methylation level detected in P. falciparum blood stages⁴³, but likely underestimate methylated loci considering the coverage we achieved (see methods).

We then monitored the distribution of detected 5mC along the P. vivax and P. cynomolgi chromosomes (Figs. S12–13) and observed a stable methylation level throughout the genomes, including in telomeric and sub-telomeric regions. We further examined the context of genome-wide methylations and, similar to what we previously observed in P. falciparum⁴³, methylation was detected as asymmetrical, with CHH (where H can be any nucleotide but G) at 69.5% and 70.5%, CG at 16% and 15.7%, and CHG at 14.3% and 13.64%, for P. vivax and P. cynomolgi, respectively (Fig. 4C). We then measured the proportion of 5mC in the various compartments of gene bodies (exons, the introns, promoters, and terminators) as well as strand-specificity (Fig. 4, D–E). We observed a slightly increased distribution of 5mC in promoters and exons compared to the intronic region, as well as in the template versus non-template strand, in P. vivax and P. cynomolgi. These results were consistent with previous data obtained in P. falciparum and in plants^43,52. Such a strand specificity of DNA methylation patterns can affect the affinity of the RNA polymerase II and impact transcription, thus we compared methylation levels to previously-report transcriptomic data from P. vivax sporozoites³². The 5mC levels in 5’ flanking regions, gene bodies, and 3’flanking regions were placed into five bins and compared to mRNA abundance, revealing an inverse relationship between methylation and mRNA abundance in the proximal promoter regions and the beginning of the gene bodies, with highly-expressed genes appearing hypomethylated and weakly-expressed genes hypermethylated (Fig. 4F). These results suggest that methylation level in proximal promoter regions as well as in the first exon of the genes may affect, at least partially, gene expression in malaria parasites. While these data will need to be further validated and linked to hypnozoite formation at a single-cell level, we have determined that 5mC is present at a low level in P. vivax and P. cynomolgi sporozoites and could control liver stage development and hypnozoite quiescence.

Assay improvements and epigenetic inhibitor library screen

The success of the original screening platform protocol and secondary confirmation of several of our initial hits provided us an invaluable opportunity to develop an improved radical cure screening assay. The current iterations of our screening platform rely on high-content analysis of parasitophorous vacuole staining of the forms that persist up to the assay endpoint^7,54. During the course of the ReFRAME primary screen we found the day 8 endpoint was sufficient for some hit compounds to act. However, other compounds like the 8-aminoquinolines exhibit a ‘delayed death’ phenotype, which leads to a false-negative result²⁴. We therefore extended the assay by 4 days to allow attenuated forms to be cleared from the culture²⁴. Also, as our screening assays were performed with multiple lots of PHH and PSH, we detected some lot-specific results, including small differences in activity of poziotinib and our monensin control (Figs. 1B, S3, S14A), possibly due to compound instability in the presence of hepatic metabolism. We therefore tested the metabolism inhibitor 1-aminobenzotriazole (1-ABT) in culture media to minimize the effect of lot-specific hepatic metabolism⁵⁵. We used a cytochrome P450 functional assay specific to CYP3A4 and determined that 100 μM of 1-ABT was sufficient to completely reduce CYP3A4 activity in both basal and rifampicin-induced PHH (Fig. S15A–B). This effect was further confirmed and quantified by mass spectrometry after 1 h of treatment at 100 μM 1-ABT. We not only detected a 75% decrease in CYP3A4 activity, but also a more than 60% reduction of CYP2B6 and CYP2E1 activity along with lesser effects on CYP2C9, CYP1A2, and CYP2D6 (Fig. S15C). These changes were incorporated into our original 8-day protocol to design an improved 12-day assay³⁶ that we then validated by re-testing 12 ReFRAME hits. The modified assay did not drastically affect the potency of most hits (Fig. S16A), but confirmed hypnozonticidal activity for poziotinib (pEC₅₀ = 6.05), which had only been previously confirmed in P. vivax and P. cynomolgi assays performed at NITD only (Figs. 1B, S3, S16B). This assay was then use in all followup experiments.

To further confirm the importance of epigenetics in hypnozoites biology,³¹ we obtained a commercially-available library containing 773 compounds targeting various inhibitors of epigenetic enzymes or pathways. These compounds were tested at 10 μM against P. vivax liver stages at both SMRU and IPC sites (Fig. S17A). We confirmed our initial hits in dose-response assays resulting in selective hypnozonticidal potency for 11 compounds targeting five different epigenetic mechanisms (Tab. 1, Fig. S17B). This includes the histone deacetylase inhibitors panobinostat (pEC₅₀ = 6.98 ± 0.18), AR42 (pEC₅₀ = 6.11 ± 0.24), abexinostat (pEC₅₀ = 5.48 ± 0.00), givinostat (pEC₅₀ = 5.35 ± 0.45), practinostat (pEC₅₀ = 5.32 ± 0.13), and raddeanin A (pEC₅₀ = 5.95 ± 0.00). Histone methyltransferase inhibitor hits included MI2 (pEC₅₀ = 5.48 ± 0.00), a compound that targets the interaction between menin (a global regulator of gene expression), and MLL (a DNA-binding protein that methylates histone H3 lysine 4⁵⁶), and cyproheptadine (pEC₅₀ = 5.24 ± 0.34), which targets the SET-domain-containing lysine methyltransferase⁵⁷. These results corroborate our hypothesis that epigenetic pathways regulate hypnozoites^31,32. Other hits, including 666–15 (pEC₅₀ = 5.88 ± 0.12), an inhibitor of the transcription factor cAMP response element-binding protein⁵⁸, and cerdulatinib (pEC₅₀ = 5.33 ± 0.20), a kinase inhibitor, suggest that signaling pathways may also be important for quiescence⁵⁹.

Table 1.

Additional epigenetic inhibitors with activity against P. vivax liver stages.

Epigenetic Inhibitor	Target(s)	Hypnozoite pEC50 ± SD	Liver Schizont pEC50 ± SD	PHH Nuclei pCC50 ± SD
Panobinostat	HDAC	6.98 ± 0.18	7.00 ± 0.15	5.68 ± 0.18
AR42	HDAC	6.11 ± 0.24	6.30 ± 0.20	5.29 ± 0.27
Raddeanin A	HDAC	5.95 ± 0.00	5.38 ± 0.13	5.49 ± 0.02
666–15	CREB	5.88 ± 0.12	5.79 ± 0.03	5.46 ± 0.03
Abexinostat	HDAC	5.48 ± 0.00	5.26 ± 0.33	< 5.00
MI2	Menin-MLL	5.48 ± 0.00	5.48 ± 0.00	< 5.00
Givinostat	HDAC	5.35 ± 0.45	5.35 ± 0.18	< 5.00
MMV019721	P. falciparum ACS	5.31 ± 0.03	5.25 ± 0.45	< 5.00
Cerdulatinib	SYK / JAK	5.33 ± 0.20	5.26 ± 0.31	< 5.00
Pracinostat	HDAC	5.32 ± 0.13	5.72 ± 0.20	< 5.00
CCT241736	FLT3 / Aurora Kinase	5.24 ± 0.33	5.24 ± 0.34	< 5.00
Cyproheptadine	SETD	5.24 ± 0.34	5.46 ± 0.03	< 5.00

HDAC: histone deacetylase. CREB: cAMP response element-binding protein. FLT3: fms-like tyrosine kinase 3. P. falciparum ACS: P. falciparum acetyl CoA synthetase. SYK: spleen tyrosine kinase. JAK: Janus kinase. SETD: SET domain containing histone lysine methyltransferase. Mean and standard deviation are from two or more independent experiments.

Having identified several histone deacetylase inhibitors as directly or indirectly active on hypnozoites, we screened compounds previously reported as inhibitors of P. falciparum acetyl-CoA synthetase (ACS), with downstream effects on histone acetylation³⁰. We found that one compound, MMV019721, was selectively active on mature P. vivax hypnozoites (Tab. 1). Given the evidence MMV019721 is directly targeting P. falciparum ACS, this result suggests ACS also is a hypnozonticidal drug target. While the molecular techniques needed to confirm the direct interaction of MMV019721 and ACS in P. vivax are currently underdeveloped, our data supplement recent reports describing epigenetics as important regulators in P. vivax and P. cynomolgi at different stage of the parasite life cycle^25,32,33.

Discussion

Herein we demonstrate several significant advances that progress radical cure antimalarial drug discovery and development, including the first report of screening a medium-sized (>10,000) compound library against mature hypnozoites as well as detection of novel hits with mechanisms unrelated to that of 8-aminoquinolines. Identification of these hits was made possible following the establishment of a complex logistical operation in which the sporozoites used for screening were produced by feeding P. vivax-infected blood from malaria patient isolates to mosquito colonies at malaria research institutes in two countries in Southeast Asia. Our international collaboration overcame several logistical hurdles to obtain positive Z’ scores for most screening plates. Hits were also confirmed via dose-response, indicating that expanded screening directed against P. vivax liver stages is likely to produce more hypnozoite-specific hits (Figs. 1, 5, S1A–B, S17).

The only class of FDA-approved compounds for radical cure, the 8-aminoquinolines, were not discovered from in vitro drug screening. Instead, they were discovered using animal models, including the P. cynomolgi-infected rhesus macaque system¹². The 8-aminoquionlines function through generation of reactive oxygen species affecting both the host and parasite, and lack a distinct parasite target^60–63. As such, this work represents one of the first applications of a radical cure development pipeline to begin with in vitro screening against P. vivax hypnozoites and end with attempted confirmation using P. cynomolgi radical cure models. While our screen generated positive results against P. vivax, we found mixed results against P. cynomolgi hypnozoites in vitro (Figs. 1C, S3B). While further studies will be needed to confirm that target(s) of our hits are parasite- or host-directed and may influence parasite survival, our data show there is sufficient diversity in gene expression, structural biology, or mechanisms of hepatic quiescence between P. cynomolgi and P. vivax hypnozoites that some newly identified hits may be species-specific. While this result could also be attributed to differential metabolism in human and monkey hepatocytes⁶⁴, the rhesus macaque radical cure model is currently considered as an important prerequisite for continued drug development, including efficacy testing in controlled human infections. The role of this model in the radical cure drug development cascade may need to be reevaluated as some compounds identified as promising for the radical cure of P. vivax may be abandoned too quickly due to the lack of activity against P. cynomolgi. This result highlights the need for further development and validation of P. vivax-specific animal models⁶⁵. As a whole this report adds to the broader discussion surrounding the successes and challenges of drug repurposing⁶⁶. While direct repositioning of a known drug as a safe treatment for a new indication is the ideal outcome, it can serve as advanced starting points for further optimization and has still the potential for reducing the time and cost involved in developing an efficacious therapy.

In addition to the identification of promising new hits and direction, we also confirmed that epigenetic control of pathogenic dormancy via DNA methylation is a pathway that could be potentially targeted by future antimalarials. This pathway has already been described for several disease agents capable of dormancy, including cancer cells⁶⁷ and tuberculosis⁶⁸. DNA methylation has also been validated as controlling critical processes in plants, which share evolutionary traits with Plasmodium⁶⁹. DNA methylation in the genus Plasmodium was first described in P. falciparum blood stages⁴³, and has been associated with gene expression, transcriptional elongation and parasite growth^51,52,70. Previous experiment have shown that hydralazine can directly inhibits DNA methylation in nuclear extracts of blood stage parasites but also inhibit a recombinant functional fragment of the P. falciparum DNMT⁴³. We pursued several biomolecular approaches to confirm that cadralazine may also interact with P. vivax DNMT in liver stage parasites. Due to technical limitations, we used a two-drug combination study in which the known DNMT inhibitor 5-azacytidine potentiated cadralazine against P. vivax hypnozoites (Figs. 2, S5). While we continue to develop new protocols and confirm the direct interaction of cadralazine with P. vivax, we successfully confirmed 5mC marks in P. vivax and P. cynomolgi liver stage parasites using both immunofluorescence and whole genome bisulfite sequencing assays (Figs. 3–4).

The current model of hypnozoite quiescence suggests RNA binding proteins (RBPs) drive hypnozoite formation by preventing translation of target mRNAs associated with schizogony³³. In this model, histone acetylation results in euchromatin at the loci of RBPs, resulting in their expression and ongoing quiescence. Hypothetically, HDAC inhibitors would favor quiescence while a treatment that decreases histone acetylation would favor schizongony. This model somewhat contrasts with our present finding that HDAC inhibitors and the ACS inhibitor MMV019721 successfully kill hypnozoites in vitro (Tab. 1). It is, however, likely that the identified RBPs are part of broader gene networks which, when perturbed by sudden modulation of epigenetic feature such as DNA methylation and histone acetylation, results in a lethal level of dysregulation. While we still need to develop P. vivax transgenic lines to successful study hypnozoite biology and further validate potential drug targets^71,72, the chemical probes that we described in this report could be used in combination with single-cell technology to more precisely perturb hypnozoites and refine our understanding of epigenetic pathways regulating hypnozoite formation and survival.

Funding Statement

Funding support was provided by the Bill & Melinda Gates Foundation (#OPP1107194 to Calibr, INV-031788 to CJJ, and #OPP1023601 to DEK), Medicines for Malaria Venture (RD/17/0042 and RD/15/0022 to BW and AV and RD/15/0022 to SPM and DEK), the National Institutes of Allergy and Infectious Diseases of the National Institutes of Health (#HHSN272201200031C to MRG and #1R01 AI136511 to KGLR) and the University of California, Riverside (#NIFA-Hatch-225935 to KGLR)

Data and materials availability:

For the purpose of Open Access, the authors have applied a CC BY public copyright license to any Author Accept Manuscript version arising from this submission. All bisulfite sequencing data generated in this study can be found in the Sequence Read Archive (SRA) at the NCBI National Library of Medicine (https://www.ncbi.nlm.nih.gov/sra) under the BioProject code PRJNA925570.