Analogs of Marinopyrrole A Show Enhancement to Observed In Vitro Potency against Acute Toxoplasma gondii Infection

ABSTRACT

The apicomplexan parasite Toxoplasma gondii is the causative agent of toxoplasmosis, a globally distributed infection with severe clinical consequences for immunocompromised individuals and developing fetuses. There are few available treatments, and these are associated with potentially severe adverse effects. Marinopyrrole A, a compound discovered in a marine Streptomyces species, has previously been found to exhibit potent antimicrobial activity, prompting our interest in exploring efficacy against Toxoplasma gondii. We found that marinopyrrole A was a highly potent anti-Toxoplasma molecule, with an in vitro 50% maximal inhibitory concentration (IC₅₀) of 0.31 μM, corresponding to a higher potency than that of the current standard of care (pyrimethamine); however, addition of 20% serum led to abrogation of potency, and toxicity to human cell lines was observed. Yet, application of marinopyrrole A to an in vivo lethal acute infection model facilitated significantly enhanced survival at doses of 5, 10, and 20 mg/kg. We then tested a series of marinopyrrole A analogs (RL002, RL003, and RL125) and demonstrated significantly increased potency in vitro, with IC₅₀ values ranging from 0.09 to 0.17 μM (3.6- to 6.8-fold increase relative to pyrimethamine). No detectable cytotoxicity was observed up to 50 μM in human foreskin fibroblasts, with cytotoxicity in HepG2 cells ranging from ∼28 to 50 μM, corresponding to >200-fold selectivity for parasites over host cells. All analogs additionally showed reduced sensitivity to serum. Further, RL003 potently inhibited in vitro-generated bradyzoites at 0.245 μM. Taken together, these data support further development of marinopyrrole A analogs as promising anti-Toxoplasma molecules to further combat this prevalent infection.

INTRODUCTION

Toxoplasma gondii is a globally distributed apicomplexan parasite infecting up to one-third of the human population (1–3). While most individuals experience mild symptoms, rapid proliferation of the parasite in brain tissue of immunocompromised patients (including HIV/AIDS patients and solid-organ transplant recipients) leads to toxoplasmic encephalitis, which is often fatal without treatment (4, 5). Further, crossing of the placenta by parasites following exposure during pregnancy yields prenatal infection culminating in crippling or fatal birth defects, including macrocephaly and hydrocephaly (6–8). Complicating matters, few existing treatments effectively clear acute infection, with a combinatorial pyrimethamine/sulfadiazine treatment regimen being the most common pharmaceutical intervention (9, 10); however, extended treatment with high-dose pyrimethamine causes suppression of bone marrow, even when compensatory folinic acid supplementation is provided (11), while sulfadiazine exhibits a high incidence of allergic reactions in patients (12). In addition, there are no FDA-approved treatments that effectively or potently clear chronic infection (13), leaving the potential for previously healthy patients who become immunocompromised after initial T. gondii infection to experience severe reactivation of latent cysts (4, 12, 14). Taken together, these findings emphasize a clear need for improvements to existing therapeutic options against this proliferative and prevalent parasite.

In attempting to alleviate this shortage of treatments against T. gondii, marine natural products represent a wealth of potential antimicrobial compounds, capitalizing on the existing competition between organisms to generate potent, complex organic molecules that have already been found to range in activity from antibacterial (15, 16) to anticancer (17, 18). Previous work has provided compelling evidence identifying marinopyrrole A ([4,5-dichloro-1-[4,5-dichloro-2-(2-hydroxybenzoyl)-1H-pyrrol-3-yl]pyrrol-2-yl]-(2-hydroxyphenyl)methanone), a metabolite of marine Streptomyces species CNQ-418, as a potent bactericidal agent against methicillin-resistant Staphylococcus aureus (MRSA) in vitro (19), although the mechanism of action remains largely undefined. At higher concentrations, this compound is an inducer of apoptosis in a variety of cancer cell lines both in vitro and in vivo (15). Antibacterial molecules, including doxycycline (20), trimethoprim (21), and spiramycin (22), have been shown previously to inhibit T. gondii proliferation, and so, based on the high potency of marinopyrrole A, we sought to evaluate its potential antiparasitic activity in vitro and in vivo (10).

RESULTS

Marinopyrrole A inhibits T. gondii RH-dimerized Tomato tachyzoite growth in vitro.

Based on the observed broad antimicrobial effects of marinopyrrole A, and given the historical capacity of antibacterial agents to serve as potent antiparasitic molecules (10, 13), we sought to elucidate the potential anti-Toxoplasma activity of marinopyrrole A in vitro. To this end, we performed an in vitro identification of the 50% inhibitory concentration (IC₅₀) for marinopyrrole A in technical triplicate against type I RH strain T. gondii tachyzoites stably expressing the fluorescent dimerized Tomato protein (RH-dTom), as discussed in our previous work (23). As shown in Fig. 1, marinopyrrole A demonstrated dose-dependent toxicity and high potency against T. gondii parasites, with a submicromolar IC₅₀ of 0.31 μM (Fig. 1).

FIG 1.

Marinopyrrole A demonstrates submicromolar efficacy against T. gondii RH-dTom tachyzoites in vitro. Determination of marinopyrrole A efficacy was completed by infecting confluent HFF host monolayers in a 96-well plate with T. gondii RH strain tachyzoites stably expressing the dimerized Tomato protein fluorescent reporter (dTom) at 2,000 parasites/well, containing 10% heat-inactivated serum. Marinopyrrole A was added at increasing concentrations in triplicate across infected wells, with a volume-matched DMSO solvent control included for comparison. Following 5 days of treatment, relative fluorescence of infected wells was quantified, and background host fluorescence was removed, after which the 50% inhibitory concentration (IC₅₀) was calculated by comparing each infected well to the corresponding solvent control. Results demonstrated that marinopyrrole A is effective against RH-dTom tachyzoites in vitro, with an IC₅₀ of 0.31 μM. Error bars represent standard error across triplicate samples.

The anti-Toxoplasma activity of marinopyrrole A is inhibited by the presence of ≥20% bovine calf serum.

Previous characterization of the antimicrobial properties of marinopyrrole A identified strong reduction in anti-MRSA activity when liquid medium was supplemented with 20% serum (19). While we conducted our initial analysis of in vitro T. gondii efficacy in the recommended 10% serum for tachyzoite and host cell growth (24), we sought to identify potential dose-dependent effects of increasing serum concentrations on the anti-Toxoplasma activity of marinopyrrole A. We therefore repeated the parasite inhibition assay to determine the IC₅₀ values using the same method as described above, although the medium in this case was supplemented with 1%, 10%, 20%, 30%, or 50% bovine calf serum (BCS) in conjunction with the observed tolerance of our parasites to these conditions. Dilutions of marinopyrrole A in each serum concentration were completed in technical triplicate. A slight, but not significant, 1.2-fold increase in IC₅₀ between 1% BCS and 10% BCS was present (Table 1), but the IC₅₀ could no longer be reached (NR) up to 10 μM in the presence of 20 to 50% BCS, corresponding to a >38-fold decrease in potency relative to 1% BCS (Table 1). These findings suggest that, as demonstrated previously in other organisms, serum concentrations at or exceeding 20% inhibit the anti-T. gondii activity of marinopyrrole A in vitro.

TABLE 1.

Marinopyrrole A demonstrates a dose-dependent decrease in potency following the addition of serum, culminating in loss of efficacy against RH-dTom tachyzoites at and beyond 20% serum

Concentration of serum (%)	Marinopyrrole A IC50 (μM)a	Fold increase in IC50 relative to 1% BCS
1%	0.26
10%	0.31	1.2
20%	NR	>38
30%	NR	>38
50%	NR	>38

^aIC₅₀ analyses were conducted in the presence of 1%, 10%, 20%, 30%, or 50% bovine calf serum (BCS) in the base medium. We noted little change in IC₅₀ values between 1% and 10% BCS (∼1.2-fold increase). Yet, marinopyrrole A efficacy against T. gondii was completely eliminated in the presence of 20% BCS, with the IC₅₀ not being reached (NR) up to 10 μM. This corresponded to a >38-fold decrease in potency.

Marinopyrrole A shows efficacy against acute Toxoplasma infection in a murine model.

Marinopyrrole A was assessed for its capacity to allow for significant survival in Swiss-Webster mice following a lethal challenge of 5,000 T. gondii ME49 tachyzoites (Fig. 2), representing wild-type-like infection. Following infection, mice were treated daily with various dosages of marinopyrrole A (5, 10, and 20 mg/kg) through intraperitoneal injection for 10 days postinfection (n = 7). Pyrimethamine-treated mice were included as a positive control and exhibited 71.4% survival. Mice treated with only solvent exhibited 14.3% survival (n = 7). Marinopyrrole A at both 5 mg/kg (P = 0.01, n = 7) and 10 mg/kg (P = 0.04, n = 7) exhibited 71.4% survival. When marinopyrrole A was increased to 20 mg/kg, 85.7% of mice survived (P = 0.004, n = 7). All P values shown are respective to solvent-treated mice. Mice treated with 5 mg/kg and 10 mg/kg showed minimal side effects, while mice treated with 20 mg/kg showed a slight increase in irritability and lethargy, as described in (25).

FIG 2.

Survival of mice lethally infected with T. gondii treated with various marinopyrrole A doses. Swiss-Webster mice were infected with 5,000 ME49 Toxoplasma gondii tachyzoites. Following infection, marinopyrrole A treatment began via intraperitoneal injection at 5, 10, or 20 mg/kg given every day for a total of 10 days (n = 7 for each dosage). Pyrimethamine was given at 5 mg/kg via intraperitoneal injection once per day for a total of 10 days as a positive control for mouse survival (n = 7). Solvent-treated mice (intraperitoneal injection, 10 days of treatment) were also included as a negative control (n = 7); 71.4% of mice survived when treated with either marinopyrrole A at 5 mg/kg (P = 0.01) and 10 mg/kg (P = 0.04), while 85.7% of mice survived when treated with 20 mg/kg of marinopyrrole A (P = 0.004). Additionally, 71.4% of pyrimethamine-treated mice survived (P = 0.02) while only 14.3% of solvent-treated mice survived infection. All P values are respective to solvent-treated mice; *P < 0.05 by log rank test.

Design of marinopyrrole A analogs generated novel compounds RL002, RL003, and RL125.

We have previously generated a series of drug-like molecules derived from marinopyrrole A (15). We further designed a marinopyrrole-based small-molecule library with clog (P) values of >5 (26–28). A majority of these library compounds have poor water solubility. Using a fragment-based approach, we recently designed a natural product small-molecule library with lower clog (P) values ranging from 2.0 to 5.0 (29–40). Our design improved the physicochemical and drug-like properties as reported in references 29, 30, and 31–40, producing RL002, RL003, and RL125 (Fig. 3). RL002 and RL003 have clog (P) values of 4.1 and 4.7, respectively. Ligand efficiency (LE ≅ − ΔG/HAC, defined as the free energy of binding divided by the number of nonhydrogen atoms [HAC]) values of RL002 and RL003 are 0.54 and 0.56, respectively, doubling that of marinopyrrole A (LE = 0.27).

FIG 3.

Structural diagrams of marinopyrrole A analogs. Marinopyrrole A served as a parent molecule for the generation of derivative RL125, which was subsequently used in a fragment-based design process yielding analogs RL002 and RL003 as described in the text.

Analogs of marinopyrrole A improve potency against T. gondii RH-dTom parasites without a concomitant increase in toxicity to key host cell lines.

Given the importance of specificity to parasites over host cells, we next evaluated the toxicity of each synthesized analog to human foreskin fibroblast (HFF) and HepG2 (human hepatocellular carcinoma) cell lines using the resazurin assay (38). Compounds were serially diluted in triplicate, and fluorescence (correlating with viability) was measured relative to the solvent control to determine the IC₅₀ values. Results demonstrated that RL002, RL003, and RL125 showed no detectable toxicity in HFF cells, as reflected by the absence of a quantifiable IC₅₀ up to 50 μM shown in Table 2. Analogs did show some toxicity to HepG2 cells, varying from 28.0 to 49.7 μM (Table 2), which was not unexpected due to marinopyrrole A’s activity against cancerous cell types.

TABLE 2.

Marinopyrrole A derivatives RL002, RL003, and RL125 show no or limited toxicity in human foreskin fibroblast (HFF) and human hepatocarcinoma (HepG2) cell lines in vitro

Compound identifiera	IC50 against HFF (μM)	IC50 against HepG2 (μM)
Marinopyrrole A	>50	5.3
RL002	>50	28.0
RL003	>50	49.7
RL125	>50	46.5

^aConfluent HFF or HepG2 monolayers in a 96-well plate were treated in technical triplicate with increasing concentrations of each compound of interest or a volume-matched solvent (DMSO) control for 5 days. Results showed that neither marinopyrrole A nor any of the marinopyrrole derivatives were toxic to the HFF cell line up to 50 μM. Conversely, marinopyrrole A was toxic to HepG2 cells, with an IC₅₀ of 5.3 μM; while seemingly low, this represents a >50-fold higher IC₅₀ than that required for T. gondii RH-dTom parasites. The marinopyrrole A analogs had much lower toxicity against HepG2 cells.

To determine the activity of RL002, RL003, and RL125 against RH-dTom tachyzoites, we repeated the in vitro parasite inhibition assay as described above for marinopyrrole A, performing each dilution series in technical triplicate. All analogs showed dose-dependent submicromolar efficacy against tachyzoites, ranging from 0.09 to 0.37 μM (corresponding to a statistically significant 3.6- to 6.8-fold increase in potency relative to our in-house-generated 0.60 μM pyrimethamine IC₅₀) (Fig. 4). These analogs demonstrated up to 3.3-fold increases in potency compared to the marinopyrrole A parent (Fig. 4), suggesting improved anti-Toxoplasma activity as well as selectivity for parasites in vitro. Parasite IC₅₀ values represented ∼291- to 517-fold increased potencies compared to the HepG2 IC₅₀ values (Table 2), suggesting high selectivity for parasites over the host.

FIG 4.

In vitro antiparasitic activity for marinopyrrole analogs demonstrates substantial dose-dependent efficacy against T. gondii RH-dTom tachyzoites with enhanced potency compared to pyrimethamine. Relative viability compared to the solvent (DMSO) control demonstrated a dose-dependent antiparasitic effect across all derivatives, with pyrimethamine included as a positive control. Derivatives RL002, RL003, and RL125 showed higher potency than the marinopyrrole A parent (included for reference), and RL003 was identified to have the highest potency, with an IC₅₀ of 0.092 μM. Assays for each analog were run in technical triplicate and are treated as independent experiments (with the exception of pyrimethamine, which was run on each plate). Error bars represent standard error across triplicate samples; *P < 0.001 by Student’s paired t test assuming unequal variance.

Marinopyrrole A analogs RL002, RL003, and RL125 show reduced sensitivity to serum.

As with marinopyrrole A, analogs RL002, RL003, and RL125 were evaluated in vitro for sensitivity to serum in technical triplicate from 1% to 50% serum. RL002 showed a substantial 29.4-fold decrease in potency between 1% BCS and 50% BCS, yielding an IC₅₀ of 5.0 μM at 50% BCS (Table 3). RL003 exhibited only a 5.9-fold increase in IC₅₀ at 50% serum and, further, maintained its submicromolar efficacy and overall high potency with an IC₅₀ of 0.15 μM (Table 3). RL125 had a 38-fold decrease in potency, although it still maintained an IC₅₀ of 0.38 μM at 50% serum (Table 3). These data suggest that marinopyrrole A derivatives RL002, RL003, and RL125 are less sensitive to serum than the parent compound, with RL003 and RL125 retaining higher potency than pyrimethamine up to 50% serum.

TABLE 3.

Marinopyrrole A derivatives RL002, RL003, and RL125 demonstrate modest decreases in potency in the presence of increasing serum concentrations

Concentration of serum (%)a	Marinopyrrole A IC50 (μM)	RL002 IC50 (μM)	RL003 IC50 (μM)	RL125 IC50 (μM)
1%	0.26	0.17	0.03	0.01
10%	0.31	0.17	0.09	0.16
20%	NR	3.4	0.10	0.18
30%	NR	4.4	0.14	0.37
50%	NR	5.0	0.15	0.38

^aAll marinopyrrole A derivatives exhibited decreased sensitivity in serum compared to marinopyrrole A. These data suggest that marinopyrrole A derivatives RL002, RL003, and RL125 have reduced susceptibility to the serum-mediated loss of potency shown in the parent compound, with RL003 and RL125 maintaining high potency at 50% serum.

RL003 is a potent inhibitor of in vitro-generated T. gondii cysts.

Due to its potency against RH-dTom tachyzoites, we aimed to evaluate the efficacy of RL003 against in vitro-generated cysts. RH-dTom parasites are not generally considered to form mature cysts in culture due to their high virulence (41), and so we performed our analyses with PruΔKU80-GFP, a type II cyst-forming strain expressing green fluorescent protein (GFP) fused to the bradyzoite-specific lactose dehydrogenase (LDH2) (42). We therefore proceeded to evaluate the activity of RL003 against encysted bradyzoites of this strain.

To this end, PruΔKU80-GFP tachyzoites were used to infect HFF cells in 96-well plates and permitted to invade for 24 h before differentiation was induced using an alkaline induction medium (pH 8.1). Induction was permitted to occur for 9 days in the absence of added CO₂. The presence of cysts was validated by fluorescence microscopy for GFP and light microscopy. RL003 was then added at increasing concentrations in triplicate to the wells and incubated for 5 days before counting via automated microscopy. Results showed a dose-dependent decrease in GFP-positive cysts following exposure to increasing concentrations of RL003, reaching a submicromolar IC₅₀ of 0.245 μM (Fig. 5). This suggests that RL003 is a potent inhibitor of T. gondii bradyzoites in vitro even when present in cysts.

FIG 5.

RL003 demonstrates submicromolar efficacy against in vitro-generated PruKU80 T. gondii cysts. Evaluation of the capacity for RL003 to act against tissue cysts of T. gondii was conducted in vitro. Ninety-six-well plates containing PruKU80 parasites constitutively expressing GFP in the bradyzoite stage were treated with alkaline induction medium (pH 8.1) for 9 days without CO₂ to induce differentiation into bradyzoites. Following successful generation of GFP-positive cysts, increasing concentrations of RL003 were added, along with a volume-matched DMSO control. Cysts were counted using an automated microscope, and IC₅₀ values were calculated relative to the solvent control. All concentrations were evaluated in triplicate. Results showed a dose-dependent response of cysts to increasing RL003 concentrations, reaching an IC₅₀ of 0.245 μM. This suggests that RL003 is a potent antibradyzoite agent able to act against tissue cysts. Error bars represent standard error from the mean.

DISCUSSION

Toxoplasma gondii infection constitutes a substantial burden to patient health, especially for immunocompromised individuals and prenatally infected infants (7). Limitations to existing treatments (13) emphasize the importance of developing improved therapeutic interventions against T. gondii. To this end, the marine microbial product marinopyrrole A has been previously found to show potent antibacterial activity against Gram-positive Staphylococcus species (19) and to elicit proapoptotic stimulation across a variety of cancer cell lines (39), indicating broad antiproliferative properties of this compound across multiple divergent species (both prokaryotic and eukaryotic); therefore, we aimed to address the scarcity of viable T. gondii treatments by evaluating the antiparasitic activity of marinopyrrole A in vitro and in vivo. In this work, we describe the previously unreported inhibition of T. gondii RH-dTom parasites by marinopyrrole A at a significantly higher potency than pyrimethamine and, further, report less serum-sensitive analogs with even higher potencies up to ∼7 times that of pyrimethamine.

Our in vitro analysis of marinopyrrole A efficacy against T. gondii tachyzoites shows a remarkably low IC₅₀ (0.31 μM) (Fig. 1), demonstrating critical inhibition of parasite proliferation during the acute-stage lytic cycle. Given our previously reported in vitro IC₅₀ of 0.60 μM for pyrimethamine (23), marinopyrrole A presents a substantial and significant improvement in potency, adding to its existing repertoire of antimicrobial activity (26). While previous literature reports some toxicity to human cells via activation of proapoptotic cell death pathways in a wide variety of cancer cell lines (39), we demonstrate that toxicity of this compound to the noncancerous, terminally differentiated HFF cell line remains undetected. However, marinopyrrole A did show substantial toxicity to HepG2 cells, with an IC₅₀ of 5.3 μM (Table 2); while ostensibly concerning, this value still represents >50-fold selectivity for parasites over HepG2 cells, supporting our continued analysis of this compound against Toxoplasma. Additionally, given that HepG2 cells are a hepatocellular carcinoma line and based on the aforementioned toxicity of marinopyrrole A to cancer cell lines (39), this toxicity against HepG2 is to be expected, and, given the continued maintenance of high selectivity, this supports the in vitro viability of using this compound in acute T. gondii models.

However, despite these promising data, we found that the submicromolar efficacy of marinopyrrole A was completely lost by the addition of serum at concentrations greater than or equal to 20% of the media composition, at which point the IC₅₀ could no longer be reached at up to ∼38-fold higher concentrations of marinopyrrole A than the previously determined IC₅₀ value; that is, up to 10 μM, it was not possible to determine a concentration at which 50% of tachyzoites had been killed (Table 1). Such strong inhibition by serum is not unprecedented, as Haste and others previously reported abrogation of marinopyrrole A bactericidal activity against methicillin-resistant S. aureus in the presence of 20% human serum (19), although, intriguingly, this effect was less pronounced in in vivo examinations of anticancer properties for neuroblastomas (30). Nevertheless, loss of anti-Toxoplasma activity at 20% serum represents a potential physiological barrier to the translational application of this compound in mammalian models, prompting our progression to in vivo models.

Despite the significant loss of marinopyrrole A potency following the addition of serum, we performed an in vivo analysis of anti-Toxoplasma activity in a lethal acute murine infection model using outbred Swiss-Webster mice infected with ME49 strain tachyzoites. Surprisingly, we show that dosages at 5 and 10 mg/kg marinopyrrole A facilitated 71.4% survival, equivalent to pyrimethamine controls. At 20 mg/kg, marinopyrrole A allowed for an increase in survival at 85.7%, suggesting that the observed in vitro serum-mediated loss of potency was abrogated, at least to some extent, in vivo (Fig. 2).

The high potency of marinopyrrole A prompted us to investigate several synthesized analogs of this compound, here noted as RL002, RL003, and RL125 (Fig. 3), for potential improvements to the antiparasitic activity and serum sensitivity of the parent. Our results showed that RL002, RL003, and RL125 exhibit increased parasiticidal activity, with IC₅₀ values ranging from 0.092 μM to 0.160 μM, corresponding to significant ∼2- to 7-fold increases in potency relative to pyrimethamine (Fig. 4). These analogs also demonstrate comparatively low toxicity to human cell lines up to 50 μM (Table 2), with no observed toxicity to HFF cells and HepG2 IC₅₀ values greater than 250-fold higher than those reached in T. gondii (Fig. 4). This represents an improvement over the toxicity of the parent to HepG2 cells. In sum, these data demonstrate that, even in a cancer cell line likely to be impacted by a marinopyrrole, these analogs have extremely high (>500-fold) selectivity for T. gondii tachyzoites over either human cell line, a desirable trait for a potential therapeutic compound. Expanded screening with additional cell lines against RL002, RL003, and RL125 would be beneficial in further determining possible tissue-dependent toxic effects.

Critically, we also report a concomitant decrease in sensitivity to serum across all analogs, with each showing an IC₅₀ of ≤5 μM up to 50% serum; this is in stark contrast to marinopyrrole A, which showed dramatic loss of potency (Table 1). While RL002 demonstrated an ∼8-fold lower potency than pyrimethamine at 50% serum, RL003 sand RL125 maintained submicromolar (and subpyrimethamine) IC₅₀ values despite the ∼40-fold loss of potency for RL125 from 1 to 50% BCS (Table 3). Notably, RL003 showed the lowest loss of potency across serum concentrations at ∼5.9-fold, maintaining an ∼4-fold lower IC₅₀ than pyrimethamine despite this minor reduction in parasiticidal activity (Table 3). This suggests that, while present, serum sensitivity of these analogs is substantially lower than that of the parental marinopyrrole A molecule and, importantly, that these analogs maintain high potency up to 50% serum. Particularly, the maintenance of such high potency in RL003 and RL125 even in higher serum concentrations is a desirable trait for continued analysis in future work. Additionally, with these improvements in vitro suggesting therapeutic potential, it will be crucial to evaluate in vivo toxicity and efficacy against acute toxoplasmosis. This will require further optimization of our current synthesis techniques to facilitate proper scaling of compound to sufficient levels for in vivo testing.

Perhaps most surprisingly, RL003 showed potent inhibition of in vitro-generated cysts. Not only was the IC₅₀ reached, which itself suggests promise, but the value of 0.245 μM indicated a remarkably high potency (Fig. 5). This is particularly critical for addressing the gaps in treatment options for T. gondii infections; given that the cyst stage has no current FDA-approved treatments capable of penetrating the cyst wall and effectively reducing parasite burden (10), this may point to an additional therapeutic application for RL003.

Further in vivo chronic studies will be needed to more thoroughly evaluate the potential antiparasitic activities of RL003 in an animal host. While little in vivo data are currently available for pyrrolomycins, one study by Yang et al. examined the pharmacokinetic properties of a similar pyrrolomycin in an in vivo murine model (43). Results of this analysis indicated that the maximum plasma concentration occurred at 0.25 h posttreatment, suggesting rapid absorption of the compound (43). The half-life of intravenous-administered compound was 6.04 h, while that of orally administered compound was 6.75 h; however, oral bioavailability was limited to 35% (43). Another study by Aldhafiri et al. characterized the in vivo pharmacokinetic properties of pyrrolomycin MP1, which shows a half-life of ∼9 h and reached maximum plasma concentration in 0.6 h when administered by oral gavage (44). In comparison, pyrimethamine has been shown in mice to reach a maximum plasma concentration in ∼1 to 2 h and to have a half-life of ∼5 to 6 h (45). These comparable pharmacokinetic properties, particularly given the structural similarities between RL003 and other pyrrolomycins, provide promising support for the favorable in vivo activity of RL003, although experimental validation will be essential to address these data.

An additional consideration of future utilization is the cost of synthesis for relatively complex natural products such as marinopyrrole A or its analogs. Initial synthetic generation of marinopyrrole A was described as a 9-step procedure with 30% yield (15). The value of generating and evaluating analogs of this parent compound, as described here, may yet yield a more facile synthetic process in the future.

Taken together, our data point to marinopyrrole A and its analogs as previously unreported anti-Toxoplasma molecules, providing a potential avenue for improved therapeutic interventions against this infection. Continued analysis of these or related marinopyrrole analogs may provide additional insight into structural components contributing to in vitro and in vivo potency to parasites as well as sensitivity (or lack thereof) to serum, facilitating continued development of further-improved compounds.

Conclusion.

In this work, we showed that marinopyrrole A is a high-potency anti-Toxoplasma molecule demonstrating submicromolar efficacy at 1 to 10% serum concentrations in vitro. We also demonstrated a dose-dependent reduction in potency following increasing serum concentrations, with potency being largely lost at serum levels of 20% and above. However, our in vivo data demonstrate that, despite this apparent obstacle, marinopyrrole A at 5, 10, and 20 mg/kg facilitated significantly enhanced survival in mice compared to the vehicle control. Improving upon the existing advantageous features of marinopyrrole A, our synthesized analogs RL002, RL003, and RL125 showed low toxicity to HFF and HepG2 cells, significantly increased potency against T. gondii infection in vitro, and reduced sensitivity to serum. Further, RL003 showed potent inhibition of encysted bradyzoites in vitro. These analogs therefore represent a series of promising novel anti-Toxoplasma compounds for future evaluation in vivo for safety, efficacy, and ease of use, with the ultimate aim of addressing the severe limitations to existing therapeutic options against this infection. Given the enhanced serum stability of the analogs and our demonstration of their in vitro potency, future goals include improved synthesis as well as analysis of these analogs with in vivo assays.

MATERIALS AND METHODS

Commercial antimicrobial compounds.

Marinopyrrole A and pyrimethamine were obtained from Sigma-Aldrich (St. Louis, MO). These compounds were dissolved in dimethyl sulfoxide (DMSO) before completion of relevant assays.

Synthesis of marinopyrrole A analogs.

We reported the synthesis of marinopyrrole analogs RL002, RL003, and RL125 previously (16, 31–40).

Cell lines and maintenance.

HepG2 (human hepatocellular carcinoma) cells were obtained through CH3 Biosystems (Buffalo, NY), and HFF (ATCC CRL-4001) cells were acquired from ATCC (Manassas, VA). Both cell lines were maintained under serial passage at 37°C and 5% CO₂. Type I Toxoplasma gondii RH parasites constitutively expressing an integrated stable fluorescent dimerized Tomato protein (dTom) construct were serially passaged in HFF cells at 37°C and 5% CO₂ as reported previously (23). Both host cells and parasites were maintained in D10 medium with the following formulation: Dulbecco’s modified Eagle’s medium (DMEM) (Lonza, Walkersville, MD) supplemented with 20% Medium-199 (Corning, Manassas, VA), 10% heat-inactivated bovine calf serum (GE Healthcare Life Sciences, Logan, UT), 2 mM l-alanyl-l-glutamine (Corning, Manassas, VA), 100 μg/ml penicillin/streptomycin (Corning, Manassas, VA), and 20 μg/ml gentamicin sulfate (Corning, Manassas, VA).

Parasite growth assays.

Determination of the half-maximal inhibitory concentration (IC₅₀) was completed through the procedure outlined in references 23 and 46. Briefly, 96-well plates containing confluent HFF monolayers were infected with 2,000 tachyzoites/well of RH-dTom parasites and were subsequently incubated for 12 h to allow for invasion. Compounds of interest were then added at increasing dilutions (each dilution in technical triplicate) as indicated in the text, with a volume-matched, solvent-only control, and plates were incubated at 37°C and 5% CO₂ for 5 days. Fluorescence at 554/581 nm (dTom) was quantified for each well through the use of a BioTek Synergy HT multimode plate reader, and IC₅₀ values were calculated by determining the compound concentration at which wells exhibited 50% fluorescence relative to the solvent control.

For growth assays in the presence of increasing serum concentrations, heat-inactivated bovine calf serum (BCS) at 1 to 50% was added to the base medium (i.e., D10 without 10% BCS) 48 h before the addition of compound to allow parasites and host cells to acclimate. IC₅₀ assays were then completed as above.

Cell viability assays.

Calculation of the IC₅₀ for HFF and HepG2 host cells was conducted through the use of a resazurin assay as described in reference 25. To complete this, confluent monolayers of HFF or HepG2 cells were treated with various concentrations of each compound of interest, including a volume-matched solvent (DMSO) control, in technical triplicate. Following incubation for 5 days, resazurin salt (MP Biomedicals, Solon, OH) was added at a final concentration of 10 μM, and plates were allowed to incubate at 37°C and 5% CO₂ for 4 h protected from light. Fluorescence was then measured through the use of a BioTek Synergy HT multimode plate reader as above, and IC₅₀ values were calculated as the concentration at which wells exhibited 50% fluorescence relative to the solvent control.

In vivo analysis of compound efficacy against acute infection.

Two-month-old female Swiss-Webster mice were obtained from Charles River. Mice were infected with 5,000 T. gondii ME49 tachyzoites via intraperitoneal injection. Subsequently, after infection, marinopyrrole A treatment began through intraperitoneal injection at 5, 10, and 20 mg/kg (n = 7 for each dosage) given once a day for a total of 10 days. Intraperitoneally injected pyrimethamine at 5 mg/kg was included as a positive control (n = 7). Solvent-treated mice were included as a control for mouse death (n = 7). Marinopyrrole A and pyrimethamine were dissolved in PTD solvent (41): 30% propylene glycol (Sigma-Aldrich), 5% Tween 80 (Sigma-Aldrich), 65% D5W (5% dextrose in water), and 3% DMSO (Sigma-Aldrich) at a final pH of 4 to 5. Mice were monitored and weighed daily to monitor overall health. Kaplan-Meier survival curves and P value analyses were obtained using IBM SPSS Statistics software. All in vivo studies were institutionally approved and performed under IACUC number 18-075-07.

Cyst viability assay.

PruΔKU80 parasites tagged with GFP (PruΔKU80-GFP) on the LDH2 protein, which is expressed exclusively in bradyzoites (47), were used to infect a confluent HFF monolayer at a multiplicity of infection (MOI) of 10, and invasion was allowed to proceed for 24 h. Induction of differentiation was then completed largely as described in references 48 and 49; tachyzoite-preferential D10 medium was replaced with alkaline induction medium, pH 8.1 (RPMI 1640, 1% heat-inactivated fetal bovine serum [FBS], 50 mM HEPES) and grown without CO₂ at 37°C for 9 days, with replacement of medium every 2 days and concomitant monitoring of wells for cyst formation. RL003 was then added at increasing concentrations in triplicate across wells, including a volume-matched solvent-only control. After 5 days, GFP-expressing cysts (excitation/emission: 469/525-nm filter cube) were counted through the use of a BioTek Lionheart automated microscope and compared to the solvent control.

Statistical analyses.

All statistical tests for variation in IC₅₀ values were completed using a Student’s t test assuming unequal variance, with a P value of <0.01 unless otherwise indicated. All statistical tests for significant increases in mouse survival were completed using IBM SPSS Statistics software. Kaplan-Meier curves were calculated, and the log rank test was implemented to determine overall significance of marinopyrrole A-treated versus solvent-treated mice.