ABSTRACT
Blastocystis is an enteric parasite with extensive global prevalence. Studies have linked infection with this protist with a variety of gastrointestinal disorders, including irritable bowel syndrome. Due to the polymorphic nature of Blastocystis, studies on the parasite could be complicated, as results can be easily misinterpreted. Metronidazole is the commonly prescribed drug for Blastocystis infection, although there have been increasing reports of drug resistance. Hence, there is a need to identify alternative drugs to eliminate Blastocystis infection. In this study, LOPAC1280 was screened and drugs that can decrease the viability of three Blastocystis isolates in cultures were identified. Using apoptosis assay and imaging flow cytometry, phenotypic changes in Blastocystis cells after treatment were also analyzed to obtain insights into the possible mechanism of action of these drugs. Three drugs—diphenyleneiodonium chloride, auranofin, and BIX 01294 trihydrochloride hydrate—were effective against all three isolates tested. Repurposing of these drugs for Blastocystis treatment could be a way of combating metronidazole resistance relatively quickly and at a lower cost.
INTRODUCTION
Blastocystis is a common enteric protistan isolated in humans and many animals. Its pathogenic status remains to be established, although there are reports linking the organism to pathological conditions with gastrointestinal and cutaneous symptoms in the absence of other known causes (1). Blastocystis spp. are classified into different subtypes (STs) based on their genetic characteristics. These STs exhibit variations in morphology, growth rate, host immune response, and drug sensitivity (2–5). While the majority of Blastocystis infections are asymptomatic, it is recommended that cases with chronic symptoms should be treated (6). The most widely used drug is metronidazole (Mz), although there are reported cases where this was not effective (3). For example, Blastocystis ST7 isolates have been found to be resistant to this compound in a drug-screening study (7). Treatment of human Blastocystis infections may not be necessary for all cases, but approaches to eliminate the parasite in immunocompromised individuals should be explored. Along with metronidazole, other drugs such as paromomycin, trimethroprim-sulfamethoxazole, nitazoxanide, emetine, furazolidone, iodochlorhydroxyquin, and pentamidine have also been used and found to have various efficacies (3). No single drug is effective across all isolates of different subtypes. Finding alternative drugs with broader coverage to treat all Blastocystis STs should therefore be pursued.
In this study, an assay to screen 1,280 drugs included in the library of pharmaceutically active compounds (LOPAC1280) was performed against three isolates of Blastocystis representing three subtypes. A resazurin-based assay was used as a high-throughput screen to measure the parasite's viability (7). After the drug hits were identified, inhibitory concentrations were determined. An apoptosis assay and nuclear staining coupled with conventional flow cytometry were done to study the effects of the drugs. An imaging flow cytometry assay was also used to observe morphological changes in Blastocystis in response to the different drugs. These provided insights into the drugs' mechanisms of action against different isolates of the parasite.
RESULTS
High-throughput screening revealed various compounds that can decrease the viability of Blastocystis cells.Resazurin is a dye commonly used to test for bacterial or yeast contamination. The oxidized, nonfluorescent dye is reduced by cellular enzymes into resorufin, a pink fluorescent dye (8); hence, resazurin can be used to determine the metabolic activity of cells. Mirza et al. (7) had previously optimized a protocol using resazurin to determine the viability of Blastocystis cells. Using the resazurin-based viability assay, the efficacy of drugs at 10 μM concentrations against Blastocystis was tested. A cutoff mark of 20% viability was chosen to screen for positive hit drugs that appeared effective in killing Blastocystis. Figure 1 shows an overview of the LOPAC1280 drug library screen results. In total, there were 12 hits for isolate ST1-NUH9, 14 hits for isolate ST4-WR1, and 4 hits for isolate ST7-B, of which three drugs—diphenyleneiodonium chloride (DPI), auranofin (AU), and BIX 01294 trihydrochloride hydrate (BIX)—appeared to be effective against all three isolates (Fig. 2 and Table 1; see also Fig. S1 in the supplemental material). AU was the most effective, which brought the viability of all three isolates to below 10% (Fig. 2). Three of the four positive hits for isolate ST7-B were effective against both ST4-WR1 and ST1-NUH9, with NSC348884 hydrate being effective against ST7-B and ST4-WR1. Three drugs—SR 57227 A, diphenhydramine hydrochloride, and (−)-scopolamine hydrobromide—were effective against both ST4-WR1 and ST1-NUH9. The results suggest that ST1-NUH9 was more sensitive to drugs that affect nitric oxide and phosphorylation, whereas ST4-WR1 was more sensitive to drugs that affect apoptosis (Table 1). Isolate ST7-B did not appear to be particularly susceptible to drugs that affect a specific target class of molecules. In general, isolate ST4-WR1 appeared to be more susceptible to drug therapy, which gave lower viability percentages throughout the screen compared to ST1-NUH9 and ST7-B (Fig. 1). In addition, there were some drugs that appeared to increase Blastocystis viability above 125% (Fig. 1). Since increases in fluorescence readings for the resazurin viability microassay could either be from real increases in cell number or increases in cellular enzymatic activity due to the addition of drugs, direct light microscopy counting was performed to validate a real increase in cell numbers. It was found that only one of the six drugs, ZM 39923 hydrochloride, resulted in a real increase in isolate ST1-NUH9 viability over 10% (data not shown). For isolate ST4-WR1, five drugs appeared to increase viability by over 10%: mianserin hydrochloride, (−)-naproxen sodium, 1-phenyl-3-(2-thiazolyl)-2-thiourea, pyridoxal-5′-phosphate-6-(2′-naphthylazo-6′-nitro-4′,8′-disulfonate) (PPNDS) tetrasodium, and papaverine hydrochloride. The remaining drugs that did not increase viability using direct microscopy may have induced increased enzymatic activity in the parasite, which caused the increase in resazurin fluorescence readings.
Screening of LOPAC1280 for the ability to decrease viability of Blastocystis. The cutoff was set at less than or equal to 20% viability relative fluorescence units.
Relative fluorescence units (RFU) of Blastocystis cultures treated with compounds from LOPAC1280. Each compound was compared to Mz of the same isolate, and we determined whether the RFU was significantly lower. **, P > 0.01; ***, P > 0.001. DPI, diphenyleneiodonium chloride; AUR, auranofin; BIX, BIX 01294 trihydrochloride hydrate; Mz, metronidazole.
Drugs that inhibit growth of Blastocystis isolates and their target class
IC50 determination indicated auranofin to be most potent drug that can lower the viability of Blastocystis.To compare the compounds with each other as well as with metronidazole, the 50% inhibitory concentrations (IC50s) of these drugs that are effective against all isolates were calculated and compared to Mz (Table 2). Blastocystis ST1-NUH9 and ST4-WR1 had 50% lower viability at concentrations of 5.26 and 5.74 μg/ml, respectively. On ST7-B, Mz had a higher IC50 of 31.5 μg/ml (Table 2). The IC50 results for isolates ST4-WR1 and ST7-B were consistent with those published by Mirza et al. (7, 9), with the former being Mz sensitive and the latter being not susceptible. Although ST1-NUH9 does not have any published IC50s, the results suggested that this isolate may also be a susceptible isolate. Among the positive hit drugs tested, DPI, AU, and BIX appeared to be very effective in killing Blastocystis, with IC50s that were much lower than that for Mz. DPI had IC50s of 97.2, 105, and 75.9 ng/ml against Blastocystis ST1-NUH9, ST4-WR1, and ST7-B, respectively (Table 2). For AU, the IC50s were similar: 62.2, 122, and 113 ng/ml for isolates ST1-NUH9, ST4-WR1, and ST7-B, respectively (Table 2). The IC50s of BIX against the three isolates were slightly higher compared to DPI and AU: 467 ng/ml, 648 ng/ml, and 1.39 μg/ml for isolates ST1-NUH9, ST4-WR1, and ST7-B, respectively (Table 2). Overall, the three compounds were more effective than Mz in terms of the concentration needed to kill Blastocystis. AU was the most effective against all three Blastocystis isolates by molar concentration. By weight, AU was most effective against isolate ST1-NUH9, while DPI was the most effective against isolates ST4-WR1 and ST7-B. Notably, all three isolates from three different subtypes displayed various susceptibilities against the tested drugs. Although isolate ST7-B was not affected by Mz, the isolate was the most susceptible to DPI.
IC50s of drugs that inhibit growth of Blastocystis isolates
Blastocystis exhibited apoptotic and necrotic features when incubated with the selected compounds from LOPAC1280.DPI and AU caused Blastocystis ST1-NUH9 and ST7-B isolates to undergo externalization of plasma membrane phosphatidylserine (PS) residues, as indicated by the increase in the proportion of cells with annexin V-fluorescein isothiocyanate (FITC) staining (Fig. 3). BIX also caused PS flipping but only in ST7-B isolates. All of the compounds tested caused considerable increases in annexin V-FITC staining in ST4-WR1 isolate, but these were coupled with increases in PI staining as well, indicating necrosis of the cells. On one hand, this shows the relative fragility of this isolate compared to the other two (as observed in the untreated cultures). On the other hand, the drugs' action could have happened at higher rates in ST4-WR1 compared to ST1-NUH9 and ST7-B cultures. Mz-treated cultures presented apoptotic and necrotic features on ST4-WR1 but not on ST1-NUH9 and ST7-B isolates. In addition to ST4-WR1, the assay also showed the relative fragility of ST1-NUH9 as the untreated cultures already showed more than 10% of the cells with annexin V-FITC staining.
Blastocystis cultures show different apoptotic profiles when treated with drugs from LOPAC1280 for 6 h. In general, ST4-WR1 showed the least proportion of viable cells compared to the other isolates after treatment. Diphenyleneiodonium chloride caused the lowest viability on ST4-WR1 cultures. ST7-B cultures are relatively resistant to the drugs. ST1-NUH9 and ST4-WR1 cultures had lower viability even without treatment, suggesting their fragility when exposed to aerobic conditions. DPI, diphenyleneiodonium chloride; AUR, auranofin; BIX, BIX 01294 trihydrochloride hydrate; Mz, metronidazole.
BIX-treated Blastocystis cultures showed higher cellular DNA content.Blastocystis did not exhibit changes in Hoechst staining when treated with AU, DPI, and Mz for 6 h. BIX-treated Blastocystis cultures, however, indicated an increase in cellular DNA content compared to untreated controls (Fig. 4). This was observed among all three Blastocystis isolates.
Flow cytometry histograms of Hoechst-stained Blastocystis cells showed no changes in nucleic acid content when treated with selected compounds from LOPAC1280 except with BIX. The latter compound increased nucleic acid content in three Blastocystis isolates. DPI, diphenyleneiodonium chloride; AUR, auranofin; BIX, BIX 01294 trihydrochloride hydrate; Mz, metronidazole.
High-content imaging flow cytometry presented various morphological phenotypes of Blastocystis.Running each sample in imaging flow cytometer yielded 4,000 individual cell images per run. Morphological forms reported previously (5) were identified. The typical round and vacuolar cells (Fig. 5A) were found in the majority of populations of all the Blastocystis isolates. Granular cells (Fig. 5B) were selected using side-scatter channel. Aspect ratios were used to separate the typical round cells from the irregularly shaped cells (Fig. 5C). The spot-counting wizard in Hoechst-stained cells at the EDF setting made it possible to differentiate cells based on their nuclear arrangement (Fig. 5D and E). Hoechst 33342 is a fluorescent stain that can cross the plasma membrane and bind to all nucleic acid (10), and its localization in the parasite indicates the number and location of the nuclei.
Image gallery showing various forms of Blastocystis obtained and identified using an imaging flow cytometer. (A) Round and vacuolar cells are typically found in cultures. (B) Granular cells were identified as having high intensity values in the side-scatter channel. (C) The irregularly shaped cells have lower aspect-ratio values in the brightfield channel. (D and E) The EDF setting made it possible to perform spot counting in Hoechst-stained cells. These spots could indicate nuclear and other nucleic acid-containing structures in the cells.
Compounds that decreased Blastocystis viability caused morphological changes in the parasite.The effects on Blastocystis of the four common positive hits that affected all three isolates were analyzed using imaging flow cytometry (IFC). This was also done for Mz-treated cultures to provide comparisons. The aim was to deduce possible drug's mechanism of action and to observe the cell death effects on the parasites. All the samples were compared to untreated samples prepared at the same time to ensure that the only differences observed would be due to drug treatment. Although the three isolates were of the same genus of the parasite, their cell death effects in response to drug treatment varied greatly. All treatments resulted in changes to cell shape, with all three compounds from LOPAC1280 and Mz significantly increasing the proportion of Blastocystis cells with irregular shapes by at least 2-fold (Fig. 6A). The greatest increase in ST7-B was with DPI treatment, while Mz treatment gave the highest fold changes for isolates ST1-NUH9 and ST4-WR1. Generally, the drug treatments resulted in slight increases in granularity for all three Blastocystis isolates. However, the increase in granularity for isolate ST7-B when treated with AU was the only one found to be significant (Fig. 6B). Mz was also found to cause a significant increase in granular cells in ST4-WR1 and ST7-B isolates. Regarding cell size as determined by cell diameter, most of the drugs caused cell shrinkage across all isolates. The Blastocystis isolates that recorded significant changes in cell size are expectedly the normally bigger ones: ST1-NUH9 and ST7-B (Fig. 6C). BIX, among all selected drugs from LOPAC1280, caused the biggest reduction in cell diameter. AU caused a significant decrease in cell size but only on ST1-NUH9 cultures. All drug treatments resulted in an increase in the proportion of multinucleated cells in isolate ST7-B, although the changes in DPI-treated cultures were not significant. The greatest increase was seen in BIX treatment, at 5-fold (Fig. 6D). Once again, there were differences in the effects observed between isolates.
Changes in phenotypic features of Blastocystis cultures treated with drug hits from LOPAC1280 for 24 h as determined using imaging flow cytometry. (A) The proportion of irregular-shaped cells increased after treatment. (B) There was an increase in the proportion of granular cells of Blastocystis upon treatment compared to untreated controls. (C) Most drugs caused Blastocystis to decrease in size. (D) Multinucleated cells determined by Hoechst staining increased in Blastocystis ST7-B isolates when treated with the drugs. Each isolate treated with the compounds was compared to untreated cultures of the same isolate. *, P < 0.05; **, P < 0.01; ***, P < 0.001. DPI, diphenyleneiodonium chloride; AUR, auranofin; BIX, BIX 01294 trihydrochloride hydrate; Mz, metronidazole.
NAC modulates the effect of auranofin on Blastocystis.Auranofin enhances the production of oxygen free radicals by inhibiting redox enzymes (11, 12). These radicals are toxic to many cell types, including the gut protistan parasites Giardia, Cryptosporidium, and Entamoeba. In the latter, it was suggested that auranofin's target is the organism's thioredoxin reductase. This enzyme protects the parasite from a toxic environment, including flux of oxygen free radicals (13). On the other hand, N-acetyl cysteine (NAC) is a known antioxidant. One of its mechanisms is by acting directly on oxygen radicals as a nucleophile (14). This compound was used to further investigate the activity of auranofin on Blastocystis cultures. There was a dose-dependent rescue of Blastocystis from the effect of auranofin, as shown in a higher fluorescence reading when NAC is added to auranofin-treated cultures (Fig. 7). This observation was observed in all isolates used in this study. In addition, NAC alone induced a higher Blastocystis viability. This could indicate that supplementation with NAC made the medium more suited for anaerobiosis.
N-Acetylcysteine (NAC) reduced the effect of AU in Blastocystis. A resazurin-based assay showed lower metabolic activities in the three Blastocystis isolates when treated with AU, as indicated by the lower RFU emitted by Blastocystis cultures. There was a dose-dependent rescue when NAC was added to the cultures, as shown by significantly higher RFU values. ***, P < 0.001 (comparing the RFU values of each culture conditions with the culture treated with 10 μM auranofin [AU] of the same isolate).
DISCUSSION
Despite Blastocystis being one of the most common human parasites, knowledge of this protist and its treatment methods are incomplete. Although Blastocystis has been implicated in gastrointestinal discomfort and irritable bowel syndrome, it is also commonly isolated in healthy individuals (15). Blastocystis infection is generally self-limiting, with some patients clearing the parasite without treatment (16). However, in extreme cases, chemotherapy may be needed to eradicate the parasite (6). To date, metronidazole appears to be the most effective antibiotic in treating Blastocystis infections, although other drugs, such as sulfamethoxazole and nitazoxanide, are also used in therapy (17). Mz treatment allows a high proportion of patients to achieve clinical remission and 6-month fecal clearance (17). Although it is considered the first line drug, its parasite clearance rate has been reported to be anywhere between 0 and 100% (17). One study further showed that although Mz could clear the parasites 1 month after therapy, there were higher incidences of parasitological relapses in both symptomatic and asymptomatic patients 6 months following therapy (18). Due to these observations, it is suggested that Mz may not be able to achieve complete eradication of the parasite due to drug resistance (17). It has also been observed that Blastocystis isolates from differing geographical regions exhibit different degrees of Mz resistance; thus, different dosages and drug combinations are required to eliminate the parasite from human hosts (19). It is also important to note that drug resistance can arise due to patient noncompliance, as the drug may not have been taken for a sufficient period to fully exert its effects. In addition, other mechanisms, such as inactivation by normal gut microbiota and the differing pharmacokinetic properties of the drug, could play a role in the ineffectiveness of Mz (19). Based on all these factors, there is a need to find new avenues for treating symptomatic cases of human Blastocystis infections.
In this study, a commercially available library of drugs was screened for activity against Blastocystis. A modified high-throughput method that measured Blastocystis activity in cultures was used to identify the drugs that can decrease the viability of Blastocystis from three isolates representing three different STs. Among the isolates used in this study was ST7-B, which has a relative resistance against metronidazole treatment (9, 19). Another isolate, ST1-NUH9, is one of the most prevalent ST globally (20), while the last is ST4-WR1, which is reported to be dominant in Europe (21). The LOPAC1280 screen identified diphenyleneiodonium chloride, BIX 01294 trihydrochloride hydrate, and auranofin as hit drugs that can kill all Blastocystis isolates used in this study. The IC50s of these drugs on Blastocystis were then determined. Further assays included finding out the ability of these drugs to induce apoptosis in Blastocystis, as well as their effects on the parasite's genetic material. Lastly, morphological changes caused by the drugs were analyzed using imaging flow cytometry. In all of these drug assays, Mz was included to provide comparisons between a standard antiprotozoal drug and novel candidates against Blastocystis. While Mz also caused morphological changes in Blastocystis even in ST7-B isolate (Fig. 6), IC50s and apoptosis assay data still confirmed its relatively low levels of activity against this particular isolate.
The first drug identified that can lower the viability of all three isolates of Blastocystis was DPI. This compound inhibits flavin-containing enzymes such as NADPH oxidase, which is involved in apoptosis in gastric cells (22) and is also thought to be an epithelial nitric oxide synthase inhibitor (23). A previous study had shown that nitric oxide (NO) induced apoptosis-like programmed cell death (PCD) in Blastocystis (24), and a Mirza et al. (9) study was consistent with this. However, DPI's inhibition of the same enzymes that catalyze the reduction of resazurin presents a caveat on the viability assay used in this study. The lower fluorescence of treated cultures compared to untreated control may be partially due to the failure to reduce resazurin and not completely related to the viability of the cells. Nevertheless, the results of the conventional and imaging flow cytometry analysis indicated that DPI affected Blastocystis cells via apoptosis-like PCD (Fig. 3 and 6). For isolate ST4-WR1, the increase in PI staining suggested a higher rate of necrosis compared to ST1 and ST7 isolates. Imaging flow cytometry also revealed that DPI treatment resulted in cell shrinkage in all three isolates, another hallmark of apoptosis-like PCD (25). In addition, DPI, as well as the rest of the selected compounds, increased the proportion of irregularly shaped cells (Fig. 6A). This particular morphological change is a good measure of the poor health of cells (5), and these data supported the finding in the resazurin and apoptosis assays. In a study involving another protistan parasite, DPI's inhibition of flavin-dependent pathways was accompanied by inhibition of pyruvate:ferredoxin oxidoreductase (PFOR) and hydrogenosomal malate dehydrogenase activities (26). PFOR is also found in Blastocystis' mitochondrion-like organelles which act as anaerobic mitochondria (27, 28). DPI's action on Blastocystis could therefore lead to a significant, if not total, termination of the organism's metabolic pathways. A study by Mirza et al. (9) also showed that isolate ST7-B is more susceptible to NO stress compared to isolate ST4-WR1, with a lower IC50. This trend was also seen in Blastocystis treatment with DPI, with isolate ST7-B having the lowest IC50. Thus, DPI may exert effects similar to those by NO. Interestingly, Mirza et al. (9) showed that Blastocystis was able to downregulate epithelial inducible NO synthase as a survival mechanism. However, DPI is an epithelial NO synthase inhibitor. Hence, it is possible that in in vivo models, the drugs may not be effective, since it inhibits NO formation, to which Blastocystis spp. are susceptible. Nonetheless, knowledge of the drug chemical structure may provide useful information for the creation of a novel drug that can exert its effect only on Blastocystis.
Another compound found to be effective against all three Blastocystis isolates was BIX 01294 trihydrochloride hydrate. BIX acts as a G9a histone methyltransferase inhibitor by occupying the histone binding site (29). Methyltransferases, along with deacetylases, are essential epigenetic regulatory enzymes. Because of this, BIX can induce autophagy-associated cell death via G9a dysfunction and intracellular reactive oxygen species production (30). Aside from studies involving human cell lines, BIX has also been studied against other parasites, such as Plasmodium falciparum (31, 32). The compound could inhibit the parasite, including its sexual stage. This was associated with significant inhibition of the Plasmodium's histone methyltransferase. In this study, BIX was the only drug among the selected compounds that caused shifts in Hoechst staining in all three Blastocystis isolates (Fig. 5). As the cell cycle of many protistan parasites still has to be elucidated (33), this increase in cellular DNA content caused by BIX is interesting. It could be that the compound causes a cell cycle arrest in the parasite before cell division could occur and after DNA synthesis has been completed. Although BIX has been found to inhibit proliferation in mammalian cells (34, 35), more information is needed to determine whether this effect is also found in Blastocystis. Regarding morphological changes, BIX affected isolates ST1-NUH9 and ST7-B by decreasing the parasite's cell diameter (Fig. 6C). In addition, ST7-B showed a significant increase in the proportion of multinucleated cells when the cultures are treated by BIX (Fig. 6D). Similar to AU, there was little change in PI staining (except in ST4-WR1). Increase in annexin V-FITC-stained cells was only observed in ST7-B isolates (Fig. 3). These varied effects of BIX on different Blastocystis STs could probably be explained by the genetic heterogeneity present in these isolates and closely associated with the drug's effect on the parasite's genetic machinery.
Among the common hits, auranofin was the most potent among the drug hits based on IC50s (Table 2). AU is a gold-containing compound that is commonly used to treat rheumatoid arthritis (36) and has been found to inhibit human leukocytes, including neutrophils (37, 38). AU is able to inhibit chemotaxis and the 5-lipoxygenase pathway in human neutrophils, thus inhibiting neutrophil motility and cellular synthesis of proinflammatory compounds (38). Against Blastocystis, AU did not appear to kill the parasite via pathways that directly destroys the membrane, as the exclusive PI staining across all three isolates remained low (Fig. 3). Furthermore, the different morphological changes among the isolates indicated that AUR affected each in a different manner (Fig. 6A). AU-treated isolate ST1-NUH9 had the biggest decrease in terms of cell diameter (Fig. 6C). The drug caused greater effects on granularity and possible nuclear fragmentation in isolate ST7-B (Fig. 6B and D). Although AU is no longer the drug of choice for rheumatoid arthritis treatment, it has been investigated for therapeutic applications in other diseases, as well as bacterial and parasitic infections (11). The main mechanism of action of AU is inhibition of enzymes necessary in oxidoreductive processes. For any anaerobic organism susceptible to oxidative injury like Blastocystis, these enzymes are necessary for controlling the level of the reactive oxygen species (ROS) in the environment. The first anaerobic parasite found to be sensitive to AU was Entamoeba histolytica. Transcriptional profiling indicated that AU affects the parasite's thioredoxin reductase, making the trophozoites more sensitive to reactive oxygen-mediated killing (39). Another study done on Giardia lamblia showed that AU was effective against multiple isolates of the parasite (40). It was thought that this was due to the inhibition of giardial thioredoxin oxidoreductase. However, a more recent study that involved overexpression of the enzyme in Giardia has suggested that thioredoxin oxidoreductase is not a critical target of AU (41). At this point, AU's target in Blastocystis is yet to be established. But it is more likely that this would be similar as in the case of microaerophilic or anaerobic protistan parasites. The addition of N-acetyl cysteine to the cultures attenuated the effect of AU on Blastocystis (Fig. 7). This was also observed in microaerophilic parasites E. histolytica, Trichomonas vaginalis, and G. lamblia (42). Cysteine is a known antioxidant which could potentially scavenge ROS produced because of AU treatment. However, a study has implied that cysteine not only reacts with ROS but also with other compounds, including AU (42). The addition of cysteine could therefore effectively lower the intended drug concentration in the medium.
Overall, the analyses of the morphological phenotypes, together with the differences in IC50s and apoptotic features, indicated that the three Blastocystis isolates respond very differently to drug treatments. This study therefore highlights the diversity within Blastocystis. Biological differences such as size, dominant shapes, and growth rates have already been described, particularly in ST1, ST4, and ST7 (4, 5). Genetic heterogeneity was also reported. In a study comparing the genome sequences of isolates from the three STs mentioned, the data showed differences in DNA base composition, genome size, number of genes, and number of introns (43). These differences then could contribute to the variations in drug susceptibilities (7, 44), as well as effects on the host (2, 45) and host immune responses (9, 46–48). Geographical variations of Blastocystis incidence also occur. For example, ST1 and ST3 are the most prevalent globally, but ST4 is particularly dominant in Europe (21). In South America, ST2 is also detected more frequently compared to the rest of the world (49).
The positive hit drugs elucidated from the screen can potentially provide new and broad therapeutic options for the treatment of Blastocystis infections considering the high prevalence of the parasite, as well as its heterogeneity. Drug repurposing refers to the identification of new therapeutic purposes for known and approved drugs (11). Drug discovery is an expansive and time-consuming process with low success rates (50); thus, drug repurposing is a way to make treatments more affordable and achievable. Furthermore, the drug can be quickly brought in for therapeutic use at a fraction of the cost of a novel drug (11). Three of the four drugs in this study—DPI, AUR, and BIX—showed promising results in treating Blastocystis, especially the metronidazole-insensitive isolate ST7-B. As these drugs are already on the market, repurposing them for Blastocystis treatment is a way to quickly bring in drugs for therapeutic use at a lesser cost.
MATERIALS AND METHODS
Parasite cultivation.Three Blastocystis isolates were used in this study: NUH9, WR1, and B, representing ST1, ST4, and ST7, respectively (5, 51, 52). The axenized parasite cultures were maintained in 8 ml prereduced Iscove's modified Dulbecco medium (IMDM; Gibco) supplemented with 10% horse serum (Gibco) (53). These were incubated in anaerobic jars (Oxoid) with Anaerogen gas packs (Oxoid) at 37°C. ST1-NUH9 was subcultured every 7 days, while ST4-WR1 and ST7-B isolates were subcultured every 3 to 4 days, alternately. The differences in harvesting time correspond to the variation in the growth rate of Blastocystis isolate ST1-NUH9 from the other two.
Drug preparation.LOPAC1280 (Sigma) is a collection of 1,280 biologically active small molecules that are commonly used for drug discovery. These compounds are classified into target classes, which indicate the pathways by which the drug affects cells. Stock solutions were diluted in dimethyl sulfoxide (MP Biomedicals) at 10 mM and stored at −20°C until use.
Resazurin-based viability assay.Primary screening was done to identify the drugs in LOPAC1280 that can decrease the viability of Blastocystis. Compounds were prepared to reach a 10 μM final concentration in IMDM. These were aliquoted in 96-well flat bottom plates (Greiner) and prereduced for 4 h before addition of 5 × 105 Blastocystis cells per well. After 24 h of incubation at 37°C in anaerobic conditions, resazurin was added at a 5% final concentration. After 3 h of incubation at 37°C in anaerobic conditions, the plates were read using Infinite M200 reader (Tecan) at 561-nm excitation and 592-nm emission wavelengths. The assay was done in triplicates. Compounds in LOPAC1280 which decreased Blastocystis growth to less than 20% of healthy control were selected and obtained commercially (Sigma). The drug stocks were serially diluted to concentrations ranging from 0 to 100 μg/ml. The viability assay mentioned above was repeated to determine the relative fluorescence units (RFU) emitted by Blastocystis cultures upon incubation with different concentrations of the compounds. Parasite controls were prepared to determine the percentage viability of the samples. Parasite dilutions of 0, 25, 50, and 100% were made, and the results were plotted into a straight-line graph, from which the RFU could be converted into viability percentages. These values were used to perform nonlinear regression analyses in Prism version 5.0 (GraphPad) to determine the half-maximal inhibitory concentration (IC50) of the compound against each of the Blastocystis isolates used.
Apoptosis assay and Hoechst staining.To determine whether the selected compounds from LOPAC1280 cause apoptotic features in Blastocystis, cultures with 5 × 105 cells were treated for 6 h at 37°C in anaerobic conditions at a 10 μM drug concentration. The cells were then washed and assayed using an apoptosis kit assay with annexin V-FITC and propidium iodide (PI) stains (BioVision) according to the manufacturer's instructions. The cells were then analyzed using Attune NxT flow cytometer (Life Technologies). Analysis and generation of graphs were done using FlowJo software version 10. For the PI-positive control, the samples were heated at 80°C for 15 min before staining with PI. For the FITC-positive control, the cells were treated with 10 μM staurosporine (Sigma-Aldrich). To analyze the effect of the compounds on Blastocystis cellular DNA content, cultures were treated for 24 h with the other conditions the same as described above. Both drug-treated and untreated Blastocystis cultures were then stained with Hoechst 33342 (Life Technologies) and analyzed using flow cytometry.
High-content IFC.To analyze morphological changes in Blastocystis caused by the selected compounds from LOPAC1280, cultures with 5 × 105 cells were treated at a 10 μM drug concentration for 24 h at 37°C in anaerobic conditions. The cells were then washed once in 1× phosphate-buffered saline (PBS) and stained with carboxyfluorescein succinimidyl ester (CFSE) and Hoechst 33342 (Life Technologies) for 15 mins. CFSE stains the vacuole of the parasite, while Hoechst stains the DNA inside the cell whether they are viable or not. After staining, the cells were fixed in 2% formaldehyde for 30 min. The samples were then centrifuged at 1,000 rpm for 5 min and washed once with PBS. Single stained samples of each Blastocystis isolate were also prepared for the creation of the compensation matrix. Images of the stained Blastocystis cells were captured using the Amnis ImageStream Mark II imaging flow cytometer under normal and extended depth of field (EDF) settings. The EDF function uses deconvolution to improve the spatial resolution of the image capture. The cells were run in the flow cytometer at minimum flow speed and ×60 magnification, with the 375-, 488-, and 561-nm lasers. The side-scatter channel was also used to determine cell granularity. A total of 4,000 events were collected for each sample in each run. The images captured were analyzed using IDEAS statistical image analysis software version 6.1. Gating strategies (Fig. S2) include selection of focused and single cells based on gradient root mean square (RMS) of the rate of change of the image intensity profile and aspect ratio values, respectively, from brightfield channel (Fig. S2A and B). Cell viability was determined using PI staining characteristics (Fig. S2C). Cell shape was determined via gating using the aspect ratios of brightfield and CFSE staining characteristics (Fig. S2D). The proportion of granular cells in each sample was gated using the intensity values from the side scatter channel (SSC) (Fig. S2E). The gate was positioned based on the brightfield images. The proportions of granular cells in the cultures were further verified using brightfield microscopy. Each sample was diluted 10-fold and placed onto glass slides. For each sample, Blastocystis cells (granular and nongranular) were then counted in three high-power objective fields (×400). Only fields that show more than 100 cells were counted to ensure statistical reliability. The sizes of the Blastocystis cells were calculated using the width of the brightfield mask (Fig. S2F). The spot wizard was run using the EDF file to determine the number of nuclei present in each cell. Based on the number of spots, the cells were gated into mononucleated and multinucleated cells.
Statistical analysis.All experiments in this study were performed at least three times. Statistical significance was determined using Student t test. For data from more than two groups with two independent variables (subtype and proportion of a specific morphological phenotype), two-way analysis of variance was performed with the Bonferonni test done post hoc. All statistical analyses were performed using Prism version 5.0 (GraphPad software).
ACKNOWLEDGMENT
This project was supported by a generous grant from the Ministry of Education (MOE)—Singapore (R-571-000-037-114).
FOOTNOTES
- Received 1 February 2018.
- Returned for modification 12 March 2018.
- Accepted 26 May 2018.
- Accepted manuscript posted online 4 June 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00208-18.
- Copyright © 2018 American Society for Microbiology.