Robenidine Analogues Are Potent Antimalarials in Drug-Resistant Plasmodium falciparum

Abstract

Robenidine is a veterinary drug used in the poultry industry to treat coccidiosis caused by parasites in the Eimeria genus. Though this compound and related aminoguanidines have recently been studied in other pathogens, the chemotype has not been systematically explored to optimize antimalarial activity despite the close genetic relationship between Eimeria and Plasmodium (both are members of the Apicomplexa phylum of unicellular, spore-forming parasites). In this study, a series of aminoguanidine robenidine analogues was prepared and tested in vitro against Plasmodium falciparum, including multidrug-resistant strains. Selected compounds were further evaluated in vivo against murine Plasmodium yoelii in mice. Iterative structure–activity relationship studies led to the discovery of 1, an aminoguanidine with excellent activity against drug-resistant malaria in vitro and impressive in vivo efficacy with an ED₅₀ value of 0.25 mg/kg/day in a standard 4-day test.

Malaria remains a devastating parasitic disease responsible for 229 million infections and 409 000 deaths in 2019.¹ Among the species of malaria parasites that can infect humans, Plasmodium falciparum is the most virulent and deadly. Pregnant women and children are the most vulnerable to mortality from malaria, and most cases and deaths occur in sub-Saharan Africa, where P. falciparum is the dominant species. Widespread resistance to existing therapies for malaria is an increasingly significant concern for global health. Drug resistance continues to spread for frontline malaria therapies, and resistance has emerged in the past decade in Cambodia and neighboring countries for current “last line of defense” drugs such as artemisinin.² The spread of artemisinin resistance outside of southeast Asia seems inevitable and threatens to reverse the recent progress that has been made in decreasing malaria deaths worldwide, i.e., from the year 2000 to 2019, deaths have fallen from 736 000 to 409 000.

Development of new malaria drug classes that can evade existing resistance mechanisms is an urgent global need if elimination and eradication goals are to be achieved.³ One frequently proven successful approach to the development of novel therapeutics is to begin with a drug for a related parasite or pathogen and refine its structure using medicinal chemistry techniques.

Robenidine (Figure 1a), formerly called robenzidene, is a veterinary drug developed in the early 1970s to treat coccidiosis caused by parasites in the Eimeria genus.⁴⁻¹⁰ Robenidine is frequently added to animal feed to prevent and treat coccidiosis in chickens,¹¹ other fowl,¹² and rabbits.¹³Eimeria and Plasmodium, the parasite responsible for malaria infections, are both members of the Apicomplexan phylum, distinguished by the presence of the apical complex and frequent presence of the apicoplast organelle.¹⁴

Figure 1.

(a) Robenidine, a commercial veterinary drug; (b) 1, an aminoguanidine with potent antimalarial activity.

Robenidine and other aminoguanidines have been evaluated for efficacy against several other protozoan parasites, including Toxoplasma gondii (in vitro IC₅₀ 0.03 μg/mL for robenidine),¹⁵⁻¹⁸Leishmania donovani (in vitro IC₅₀ 18 μM for the analogue CGP 40215A),¹⁹Babesia microti (murine in vivo nonrecrudescence dose 100 mg/kg/day for oral robenidine),^20,21 and the trypanosomes, T. brucei and T. cruzi (in vitro IC₅₀ 20 μM for the analogue CGP 40215A).²²⁻²⁴ It has also been tested against microorganisms such as Lactobacillus,²⁵E. coli,²⁶Acanthamoeba polyphaga,²⁷Goussia carpelli,²⁸ and several other Gram-positive and Gram-negative bacterial pathogens.^29,30

Recently, work by Trott and McClusky³¹⁻³⁶ has further explored the aminoguanidine chemotype of robenidine, applying a medicinal chemistry approach to repurpose the drug for various bacterial pathogens. Their ongoing research has demonstrated that this chemical scaffold is amenable to synthetic modification and can successfully be refined for activity against pathogens other than Eimeria.

Despite significant genetic similarity between Eimeria and Plasmodium, there has been relatively little research into the efficacy of robenidine against malaria in its 50-year history. Robenidine was evaluated against the murine species P. berghei during its initial development in 1970 and found to have moderate activity in vivo.⁴ An immine analogue of robenidine, CGP 40215A, was tested against the human pathogen P. falciparum in vitro with low micromolar IC₅₀ value.³⁷ To date, no medicinal chemistry efforts have focused on refining and optimizing the aminoguanidine scaffold for antimalarial activity.

The current study represents the first attempt to evaluate structure–activity relationships (SAR) of the aminoguanidine chemotype against malaria parasites using in vitro assays and an in vivo murine model. A library of 38 structurally diverse aminoguanidines was created and compared for in vitro antiplasmodial activity as well as mammalian cell cytotoxicity. Compounds with promising selective activity were further evaluated in vivo in a murine model of malaria. Multiple aminoguanidines from this library were found to have high potency in vitro, which translated to robust efficacy in vivo. One compound, 1 (Figure 1b), exhibits single-digit nanomolar IC₅₀ values against P. falciparum strains with an in vivo ED₅₀ value below 1 mg/kg/day vs murine malaria.

Results and Discussion

The Aminoguanidine Robenidine Is an Effective Antimalarial in Vitro

Robenidine is a symmetrical aminoguanidine drug originally developed for the treatment of Eimeria-derived coccidiosis in poultry and livestock.⁴ Recent investigations by Trott, McCluskey, and others have demonstrated that the aminoguanidine chemotype can be modified and reapplied to pathogens other than Eimeria.³¹ Prior to the present study, robenidine had not been tested for antimalarial activity against P. falciparum despite the close genetic relationship between Plasmodium and Eimeria.

To gain an initial assessment of robenidine as a potential antimalarial drug, it was evaluated in vitro for the ability to inhibit the growth of P. falciparum strain D6, a drug-sensitive strain. Blood stage parasite cultures were incubated in a range of concentrations of robenidine, and 72 h parasite growth was measured by SYBR green staining relative to untreated controls.³⁸ In this assessment, robenidine exhibited an average in vitro IC₅₀ of 324 nM (Figure 2). This value represents an excellent starting point for further evaluation of structure–activity relationships of robenidine analogues. Notably, the concentration–response curve was characterized by a steep slope at the IC₅₀ inflection point, similar to the control drug chloroquine.

Figure 2.

IC₅₀ curve of robenidine in D6 compared with chloroquine. Y values represent fluorescence relative to untreated controls.

Synthesis of Aminoguanidine Robenidine Analogues

Given the successful initial evaluation of robenidine in vitro, a series of 38 robenidine analogues was designed and synthesized to determine the potential for improved antimalarial activity within the aminoguanidine chemotype. Robenidine analogues were synthesized in a single step route starting from commercially available substituted benzaldehydes or acetophenones with the synthetic approach utilized in robenidine’s initial discovery.⁴ Two equivalents of each starting material were refluxed in ethanol with 1,3-diaminoguanidine hydrochloride (Scheme 1). The resulting symmetrical aryl aminoguanidine products (formed as HCl salts) were crashed out of solution using diethyl ether, filtered, and washed with diethyl ether. Compounds were purified by recrystallization from ethanol.

Scheme 1. A General Synthesis of Aminoguanidine Robenidine Analogues.

Compounds were prepared by refluxing benzaldehydes or acetophenones with 1,3-diaminoguanidine hydrochloride in ethanol.

As evidenced by the continued commercial success of robenidine as a veterinary drug, this synthetic approach is highly amenable to affordable, large scale synthesis, an important attribute given that any new antimalarial drug must be inexpensive for use in resource poor regions where the disease is endemic. It is noteworthy that the aminoguanidines can be prepared in a single step from commercially available starting materials, and no chromatographic separations are required for their purification. This is a significant advantage over other antimalarial drug candidates requiring multiple steps and/or complex separations.

Structure–Activity Relationships of Aminoguanidines in Vitro

The in vitro activity of each compound in the aminoguanidine library was assessed against asynchronous cultures of Plasmodium falciparum parasites replicating within human erythrocytes (Tables 1–3). As the chemical development of robenidine was not discussed in its original publication in 1970⁴ and limited structural information is available regarding its mechanism of action,¹⁰ profiling was guided primarily by in vitro potency against the P. falciparum D6 strain cultured in human erythrocytes as described above.

Table 1. Aminoguanidine Halogen Series and Antimalarial Drugs^a.

compound	R1	R2	IC50 D6 (nM)	IC50 Dd2 (nM)	IC50 Tm90–C2B (nM)	IC50 A6 (nM)	IC50 cytotoxicity (μM)	IVTI
2	2-F	H	2866	4593	>5000	1893	>200	>100
3	3-F	H	767	>5000	>5000	426	49	64
4	4-F	H	1016	3792	4904	369	24	24
5	2-Cl	H	334	961	1219	348	>200	>100
6	3-Cl	H	202	567	664	237	35	>100
robenidine	4-Cl	H	324	814	1317	410	30	93
7	2-Br	H	280	715	974	876	94	>100
8	3-Br	H	572	1516	1838	502	62	>100
9	4-Br	H	277	876	899	269	38	>100
atovaquone	N/A	N/A	<1	<1	>5000	>5000	72	>100
chloroquine	N/A	N/A	10	70	297	16	181	>100

^aSee Scheme 1 for the aminoguanidine scaffold and sites of modification. P. falciparum IC₅₀ values are the average of two to four determinations, each carried out in quadruplicate (a more granular view of this data is provided in the Supporting Information). D6, P. falciparum pan-sensitive strain; Dd2, multidrug-resistant P. falciparum strain; Tm90–C2B, multidrug-resistant P. falciparum clinical isolate that is also resistant to atovaquone; A6, P. falciparum in-house derived mutant line resistant to respiratory antagonists.⁴⁰ Cytotoxicity assays were carried out with human hepatoma derived HepG2 cells and performed in quadruplicate. IVTI = in vitro therapeutic index, defined as cytotoxicity/D6 IC₅₀. N/A = not applicable.

Table 3. Other Aminoguanidines^a.

compound	R1	R2	IC50 D6 (nM)	IC50 Dd2 (nM)	IC50 Tm90–C2B (nM)	IC50A6 (nM)	IC50 cytotoxicity (μM)	IVTI
24	H	H	705	4298	>5000	581	>200	>100
25	4-OH	H	143	240	286	278	39	>100
26	4-Ph	H	15	87	71	33	>200	>100
27	4-NO2	H	96	198	247	85	>200	>100
28	4-NMe2	H	118	264	213	122	20	>100
29	4-COOH	H	>5000	>5000	>5000	>5000	>200	N/A
30	4-CONH2	H	527	405	727	516	42	80
31	4-SO2NH2	H	>5000	>5000	>5000	>5000	>200	N/A
32	4-propynoxy	H	102	199	376	153	>200	>100
33	4-morpholino	H	131	329	295	116	>200	>100
34	3-tetrazole	H	1645	>5000	>5000	4790	107	65
35	4-tetrazole	H	838	>5000	>5000	>5000	>200	N/A
36	(2,2-difluoro) 2,3-dioxazole	H	250	800	905	235	>200	>100
37	2,4-OMe, 5-Cl	CH3	360	1247	1180	216	3	8
38	mono 2-Fb	H	1028	2202	>5000	833	>200	>100

^aSee Table 1 legend and Methods.

^bCompound 38 has a 2-fluoro substituent at only one R₁ site. The other R₁ site is unsubstituted (4-H).

Robenidine is structured as a diarylaminoguanidine with symmetrical 4-chloro (para relative to the aminoguanidine moiety) substituents on its two phenyl rings. Our SAR studies primarily focused on the potential to improve antimalarial activity by exchanging these 4-chloro substituents for different functional groups at the same position. All compounds are symmetrical and take the form shown in Scheme 1 unless otherwise noted. A halogen series (Table 1) was prepared to examine the effects of both halide type and position on antiplasmodial activity. Within this series, chloro and bromo substituted aminoguanidines were more active than fluoro substituted compounds, with position effects varying among the halides. The ortho-fluoro compound 2 displayed much lower activity than robenidine, and the para-bromo compound 9 showed moderately improved activity.

Other replacement functional groups for robenidine’s 4-chloro substituents were varied widely for size, lipophilicity, and electron withdrawing vs donating effects on the adjacent phenyl ring. Compounds with promising activity were also prepared as their ortho and meta isomers to investigate positional effects for these functional groups. Similarly, promising compounds were prepared with additional methyl groups on the benzyl carbons of the aminoguanidine moieties as shown in Scheme 1 (the R₂ position) by starting from the analogous acetophenones rather than benzaldehydes.

The 4-methoxy robenidine derivative 11 was among the earliest compounds to show improved potency over robenidine. To further pursue this activity, the methyl (10), trifluoromethyl (12), trifluoromethoxy (16), and difluoromethoxy (18) derivatives were also prepared along with their R₂-methyl analogues (Table 2). Among these compounds 16 was the most potent, quickly becoming the frontrunner with a nearly 10-fold reduction in IC₅₀ value (39 nM vs D6) relative to robenidine (324 nM vs D6). Conversion of the R₂ moiety from H to methyl did not have a consistent effect, improving activity for the trifluoromethyl derivative (12 vs 13) while reducing activities of the trifluoromethoxy and difluoromethoxy derivatives (16 vs 17 and 18 vs 19, respectively). The ortho (14) and meta (15) variants of 16 were also prepared and demonstrated significantly reduced antimalarial activity in comparison with the para isomer.

Table 2. Aminoguanidine Methyl, Methoxy, Fluoromethyl, Fluoromethoxy, and Nitrile Series^a.

compound	R1	R2	IC50 D6 (nM)	IC50 Dd2 (nM)	IC50 Tm90–C2B (nM)	IC50 A6 (nM)	IC50 cytotoxicity (μM)	IVTI
10	4-CH3	H	905	2427	2529	>5000	43	48
11	4-OCH3	H	263	1602	>5000	289	>200	>100
12	4-CF3	H	614	1536	2742	567	40	65
13	4-CF3	CH3	211	414	451	283	114	>100
14	2-OCF3	H	1080	2328	>5000	391	156	>100
15	3-OCF3	H	670	1571	2988	689	61	91
16	4-OCF3	H	39	83	114	52	11	>100
17	4-OCF3	CH3	140	526	458	182	9	64
18	4-OCHF2	H	71	191	193	87	19	>100
19	4-OCHF2	CH3	99	403	315	171	3	30
20	2-CN	H	1024	2097	1990	969	31	30
21	3-CN	H	7	20	31	24	7	>100
22	4-CN	H	58	166	312	76	>200	>100
23	3-CN	CH3	47	70	104	17	9	>100
1	3-CN, 4-F	H	4	12	16	14	8	>100

^aSee Table 1 legend and Methods.

The early success of 16 led to an interest in exploring other electron withdrawing groups at the para position. The nitro (27) and cyano (22) derivatives were significantly more active than robenidine. 22 in particular had activity in the same range as the early hit compound 16, so the ortho (20) and meta (21) analogues were prepared to explore the position effects of the nitrile group. This series was expected to show a similar pattern to the trifluoromethoxy compounds, and so it came as quite a surprise when the 3-CN analogue 21 was found to have IC₅₀ value of 7 nM, much lower than any other aminoguanidine evaluated up to that point. The R₂-methyl analogue (23) of this compound was prepared and found to be less active than the R₂-H original.

A few general trends were observed for in vitro antimalarial activity. In general, substitution at the ortho position dramatically reduced antimalarial activity, possibly by sterically constraining rotation of the aryl rings. The carbonate 29 and tetrazoles 34 and 35 were inactive against all tested P. falciparum strains. Electron withdrawing substituents appear to have a positive effect on antimalarial activity, though the biphenyl analogue 26 is unusually potent and the dimethylamino (28) and propynoxy (32) compounds also exhibit respectable antiparasitic activity.

All aminoguanidines synthesized were also evaluated against three multidrug-resistant P. falciparum strains (Dd2, Tm90–C2B, and A6) and for mammalian cell cytotoxicity. Robenidine and the lead compounds 16, 21, and 22 were further evaluated in vivo to guide optimization, resulting in the design of 1, the overall series lead. The results of these experiments are listed in Tables 1–3 and discussed below.

Aminoguanidines Retain in Vitro Activity in Drug-Resistant Strains of Malaria

In addition to the drug-sensitive P. falciparum D6 strain, the aminoguanidines were assessed against three drug-resistant strains (Tables 1–3). P. falciparum Dd2 is a multidrug-resistant strain sensitive to atovaquone but resistant to chloroquine as well as the antifolate combination of pyrimethamine + sulfadoxine. P. falciparum Tm90–C2B is a multidrug-resistant clinical isolate including resistance to both atovaquone and chloroquine.³⁹P. falciparum A6 is derived from D6 and is resistant to respiratory antagonists such as atovaquone and antimycin A but sensitive to chloroquine.⁴⁰

The degree of cross resistance observed for the MDR strains Dd2 and C2B with the tested aminoguanidine series ranged from extensive (e.g., ∼19-fold for 11), to intermediate (e.g., 6.5-fold for 3), to modest (2–4-fold for 1 and 23) relative to the drug sensitive D6 strain of P. falciparum. The general lack of significant cross-resistance is consistent with expectations given that robenidine and other aminoguanidines are not clinically prescribed for malaria or even administered directly to humans for any indication (it is possible that trace amounts of robenidine have passed into humans via the consumption of poultry treated for coccidiosis, though these trace amounts are unlikely to drive antimalarial resistance).

Aminoguanidines Have High in Vitro Therapeutic Indices

The aminoguanidines were evaluated for cytotoxicity against human hepatoma derived HepG2 cells (Tables 1-3). In this assay, HepG2 cells were incubated with test compounds for 24 h, followed by a 24-h recovery period and subsequent staining to evaluate for cytotoxic effects with resazurin. The ratio of the resulting HepG2 IC₅₀ value to the P. falciparum D6 IC₅₀ value can be considered an “in vitro therapeutic index” or IVTI.

Cytotoxicity in Hep2G did not track proportionally with antimalarial activity in any of the four tested P. falciparum strains. Overall, substitution at the 3 (meta) position resulted in increased cytotoxicity relative to the 2 (ortho) and 4 (para) positions, though the cyano substituent was shown to be an exception to this trend. Several of the aminoguanidines, including the active compound 21, had no measurable effect on Hep2G activity at concentrations as high as 200 μM. Most aminoguanidines in the series had HepG2 IC₅₀ values above 10 μM, and nearly all active aminoguanidines had an in vitro therapeutic index of over 1000-fold (this value cannot be calculated for those aminoguanidines having no antimalarial activity). It is noteworthy that 1, with the greatest antiplasmodial activity among compounds in this series, exhibits an IVTI value of 2000, which is indicative of its highly selective antiparasitic action.

Aminoguanidine Activity Is Concentration Dependent but Not Time Dependent

Many drugs have activity dependent on their exposure time in addition to concentration, and antimalarial potency is frequently stage-specific. To determine whether the aminoguanidines acted by a time-dependent mechanism against malaria parasites, the SYBR Green activity assay was adapted to include additional incubation intervals. Two potent aminoguanidines, 16 and 21, were incubated with P. falciparum Dd2 infected erythrocytes for 48, 72, or 96 h (Figure 3). Notably, this assay measures drug-treated parasite growth during the incubation time relative to untreated parasites. A shorter incubation time allows for less parasite growth for all drug conditions, producing data with proportionally greater variability or “noise”. Conversely, longer incubation times reduce noise in the resulting data.

Figure 3.

Inhibition of P. falciparum Dd2 growth by 16 (green) and 21 (purple) at 48 h (dotted curves), 72 h (dashed curves), and 96 h (solid curves) time points.

IC₅₀ values for both 16 and 21 were somewhat higher at 48 h than at longer drug incubation times (Table 4), though this may be the result of noise in the data associated with the short incubation time. This finding may also stem from the use of asynchronous parasite cultures, wherein one life-cycle stage may be more impacted than another (16 in particular appears to exhibit a biphasic concentration–response curve, indicative of stage specificity). IC₅₀ values for both compounds were virtually identical between 72-h and 96-h incubation time points. These results indicate that a 72-h incubation may be required to attain full in vitro activity, but that no additional benefit is construed with longer incubation times. Aminoguanidine activity appears to be driven primarily by drug concentration rather than by lengthening drug incubation time. Additional experiments are planned to explore for possible stage-specific activity of the most active compounds in this series however our results combine to suggest that the molecules are not acting in a manner consistent with a “delayed death” phenotype as shown for other drugs such as doxycycline and azithromycin.⁴⁵

Table 4. P. falciparum Dd2 IC₅₀ of 16 and 21 vs Drug Incubation Length^a.

	IC50 Dd2 (nM)
compound	48 h	72 h	96 h
16	410	195	192
21	31	21	22

^aSee Table 1 legend and Methods.

Lead Aminoguanidines Have High in Vivo Efficacy in a Mouse Model of Malaria

Aminoguanidines exhibiting high in vitro antimalarial potency and robenidine were assessed for in vivo efficacy in a murine model of malaria (Plasmodium yoelii, Table 5). In this modified 4-day Peters test,⁴¹ mice were inoculated with parasites from a donor mouse (day 0), and then dosed orally with drugs on each of the subsequent 4 days (days 1–4). On day 5 of the experiment, the parasitemia for each mouse was determined microscopically by examining methanol-fixed and Giemsa-stained blood smears. The ED₅₀ represents the interpolated dose of a compound at which parasitemia was suppressed to one-half that of untreated controls. Similarly, the ED₉₀ represents the dose at which parasitemia is suppressed 10-fold. Mice were considered cured of their infection if no parasites were detected in the blood 30 days from the first drug administration, and the nonrecrudescence dose (NRD) represents the lowest dose to achieve a cure.

Table 5. In Vivo ED₅₀ and ED₉₀ of Lead Aminoguanidines^a.

compound	P. yoelii ED50 (mg/kg/day)	P. yoelii ED90 (mg/kg/day)	P. yoelii NRD (mg/kg/day)
robenidine	1.6	4.9	>25
16	1.2	3.7	12.5
22	2.7	7.7	>25
21	7.1	9.4	>25
1	0.25	0.28	>25

^aIn vivo activity values were determined from a modified 4-day Peters test. Compounds were administered by oral gavage up to 25 mg/kg/day, near the solubility limit of the PEG delivery vehicle. NRD = nonrecrudescence dose (cure dose).

Robenidine, 16, 22, and 21 were evaluated for in vivo antimalarial efficacy. That robenidine exhibited respectable in vivo antimalarial activity was somewhat surprising given that it is poorly absorbed and known to accumulate in the gastrointestinal tract.⁴⁴ The early hit compound 16 (the 4-OCF₃ analogue, ED₅₀ 1.2 mg/kg/day) showed improved in vivo activity over robenidine (ED₅₀ = 1.6 mg/kg/day), while 22 (the 4-CN analogue, ED₅₀ = 2.7 mg/kg/day) did not. Unexpectedly, 21 (the 3-CN analogue, ED₅₀ = 7.1 mg/kg/day) was 6-fold less efficacious than 16in vivo, despite being 6-fold more active than 16in vitro. For comparative purposes, consider that the ED₅₀ of chloroquine in this system is 1.5 mg/kg/day.⁴²

The only aminoguanidine which produced a cure in this model was 16 with NRD of 12.5 mg/kg/day, while other compounds were not curative (including compound 1 described below). It is important to note that failure to produce a cure in this model and by this dosing regimen is not predictive of clinical failure. Indeed, several approved clinical drugs such as chloroquine are not curative in this model at any dose level.

1 is a Highly Potent Antimalarial in Vitro and in Vivo

The discrepancy between the in vitro success (Table 2) and in vivo mediocrity (Table 5) of 21 remained a mystery which we later explored. From previous in vivo SAR studies on other scaffolds, we had noted that aryl groups without protective substitutions at the para positions were biologically unstable. While 16 was substituted at the para position, 21 was not, potentially leaving this position vulnerable to hepatic microsomal degradation.

To interrogate this hypothesis, an analogue of 21 was prepared with an additional para-fluoro substituent (1, 3-CN, 4-F). Given the low activity of 4 (4-F), this substitution was expected to have little effect on in vitro potency while potentially improving upon the in vivo activity of 21. Unexpectedly, 1 had excellent in vitro activity, becoming the new series lead in potency. 1 had a single-digit nanomolar IC₅₀ value against D6 (4 nM), IC₅₀ values in the low double digits for the drug-resistant strains, and an in vitro therapeutic index of 2000-fold.

The in vivo efficacy of 1 was even more pronounced with an ED₅₀ value of 0.25 mg/kg/day, 5-fold lower than its nearest competitor 16. The single atom difference between 21 (3-CN, 4-H) and 1 (3-CN, 4-F) resulted in a nearly 30-fold improvement in in vivo efficacy.

Murine Microsomal Stability of Aminoguanidines Correlates with in Vivo Activity

To investigate the relationship between the aminoguanidines’ in vivo efficacy and their metabolic properties, the murine microsomal stability of a selection of aminoguanidines was evaluated (Table 6). Robenidine, 16, 21, 1, and the control compound ketanserin were incubated with pooled murine liver microsomes and monitored for degradation by LC/MS/MS for 1 h. The concentration vs time plot for each compound was used to determine its biological half-life (t_1/2) and predicted intrinsic clearance (Cl_int).

Table 6. Murine Microsomal Stability of Aminoguanidines^a.

compound	murine microsomal stability, t1/2 (min)	predicted Clint (mL/min/kg)
ketanserin	14.47	377.20
robenidine	158.65	34.40
16	172.16	31.70
21	58.18	93.80
1	186.15	29.32

^aData from ChemPartner Co. Ltd., Shanghai, P.R. China. See Methods for full details.

For all of the aminoguanidines evaluated, murine microsomal stability correlated with in vivo efficacy. Robenidine, 16, and 1 were all biologically stable with half-lives above 150 min (with the same rank order for in vivo activity and stability). 21 was metabolically unstable in the presence of murine microsomes, with a half-life of only 58.18 min.

This data supports the notion that 1 is more efficacious in vivo than 21 due in part to improved stability. Substituting the 4-position H for F resulted in a threefold increase in metabolic stability. Presumably the prolonged presence of 1 in the bloodstream contributes to its excellent performance in vivo.

Conclusions

1 is a robenidine derivative with excellent in vitro potency, virtually no cross-resistance in multidrug-resistant strains, and a high in vitro therapeutic index. In a murine model of malaria, 1 displayed robust in vivo antimalarial activity propelled by a combination of high intrinsic potency and biological stability. Although speculative at this time, it is also possible that the nitrogen atoms in the aminoguanidine bridge of 1 exhibit diminished basicity (i.e., ionizability) due to the presence of two strongly electron withdrawing groups (F and CN) at the para and meta positions of the flanking aromatic rings, which may in turn enhance oral bioavailability.

Further exploration of the aminoguanidine scaffold is certainly warranted, as is developing more knowledge of its antimalarial properties, including the mode of action. Assessing the in vivo activity of these compounds in humanized mice may refine predictions of clinical success. Assessing the compounds against synchronous parasites will elucidate potential stage-specific potency effects. Beyond the blood stage, evaluating the aminoguanidines against other stages of the malaria life cycle such as the liver stage will provide valuable information useful for their potential development for use in humans. Systematically exploring the mechanism of action, including chemical biology techniques and resistance studies, can further guide SAR for this chemotype, and we are currently engaged in this work. Potential synergy with other antimalarial compounds such as artemisinin, atovaquone, and ELQ-300 will also be assessed.

As we look for new entries in the antimalarial pipeline, it may be useful to reexamine drugs and chemotypes effective in related parasites and pathogens. This appears to be the case with robenidine, a drug discovered in 1970 but which has not been methodically explored in malaria using a medicinal chemistry approach until this point. Though robenidine itself has reasonably good antimalarial activity in vitro and in vivo, it did not take long to improve upon this activity in both settings. In this study, a second look at an old drug efficiently produced a new and promising chemical lead.

Methods

General Chemistry: Materials and Instruments

All solvents, starting materials, and reagents were acquired from commercial sources (Sigma-Aldrich and Combi-Blocks). Robenidine was obtained from Santa Cruz Biotechnology (Santa Cruz, California). All materials were used without further purification. ¹H and ¹³C NMR spectra were taken on a Bruker 400 MHz instrument, and chemical shifts are reported relative to TMS (0.0 ppm). Fluorescence measurements were recorded using a Molecular Devices Spectramax iD3 equipped with Softmax Pro 7 software. Final compounds were measured to be >95% pure by high performance liquid chromatography (HPLC) using an Agilent Technologies 1260 Infinity II system (unless otherwise noted). High-resolution mass spectrometry (HRMS) using electrospray ionization was performed by the Portland State University BioAnalytical Mass Spectrometry Facility. Melting points were measured using a Stanford Research Systems OptiMelt Automated Melting Point System (model MPA100).

1: 2,2′-Bis[(3-cyano-4-fluorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-cyano-4-fluorobenzaldehyde (1.31 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.54 g, 99%. ¹H NMR (400 MHz, DMSO-d6) δ 12.56 (s, 2H), 8.76 (s, 2H), 8.64 (d, J = 5.63 Hz, 2H), 8.46 (s, 2H), 8.33–8.32 (m, 2H), 7.68 (t, J = 8.92, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 164.98, 162.40, 153.57, 146.25, 136.94, 135.95, 133.22, 131.61, 131.57, 117.73, 117.53, 114.22, 101.56, 101.40. HRMS found 352.1111, M+H. MP = 317–319 °C.

2: 2,2′-Bis[(2-fluorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 2-fluorobenzaldehyde (1.1 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.34 g, 99%. ¹H NMR (400 MHz, DMSO-d6) δ 12.54 (s, 2H), 8.72 (s, 2H), 8.64 (s, 2H), 8.36 (dt, J = 7.9, 1.72, 2H), 7.59–7.53 (m, 2H), 7.34 (t, J = 7.9 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 162.21, 159.71, 152.74, 141.62, 132.93, 127.31, 124.75, 120.82, 116.07, 115.86. HRMS found 302.1206, M+H. MP = 293–295 °C.

3: 2,2′-Bis[(3-fluorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-fluorobenzaldehyde (1.09 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.31 g, 97%. ¹H NMR (400 MHz, DMSO-d6) δ 12.56 (s, 2H), 8.69 (s, 2H), 8.48 (s, 2H), 7.98 (d, J = 10.3 Hz, 2H), 7.70(d, 7.6 Hz, 2H), 7.53 (q, J = 7.3, 2H), 7.33 (t, J = 8.3, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 164.14, 161.72, 153.47, 148.02, 136.37, 136.28, 131.34, 131,26, 125.37, 118.15, 117.94, 113.89, 113.66. HRMS found 302.1206, M+H. MP = 281–283 °C.

4: 2,2′-Bis[(4-fluorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-fluorobenzaldehyde (1.09 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.33 g, 99%. ¹H NMR (400 MHz, DMSO-d6) δ 12.32 (s, 2H), 8.56 (s, 2H), 8.44 (s, 2H), 8.03 (t, J = 6.5 Hz, 4H), 7.34 (t, J = 8.2, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 165.24, 162.77, 153.41, 148.14, 130.71, 130.62, 130.48, 130.45, 116.40, 116.19. HRMS found 302.1205, M+H. MP = 295–297 °C.

5: 2,2′-Bis[(2-chlorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 2-chlorobenzaldehyde (1.23 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 g, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.42 g, 96%. ¹H NMR (400 MHz, DMSO-d6) δ 12.62 (s, 2H), 8.88 (s, 2H), 8.69 (s, 2H), 8.43 (d, J = 7.15 Hz, 2H), 7.58–7.46 (m, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.19, 145.53, 134.06, 132.77, 131.03, 130.39, 128.51, 127.95. HRMS found 334.0620, M+H. MP = 295–297 °C.

6: 2,2′-Bis[(3-chlorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-chlorobenzaldehyde (1.23 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 g, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.48 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.39 (s, 2H), 8.69 (s, 2H), 8.43 (s, 2H), 8.17 (s, 2H), 7.83 (d, J = 6.49 Hz, 2H), 7.59–7.50 (m, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.40, 147.93, 135.91, 134.27, 131.11. 130.92, 127.71, 127.06. HRMS found 334.0619, M+H. MP = 268–270 °C.

7: 2,2′-Bis[(2-bromophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 2-bromobenzaldehyde (1.23 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 g, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.83 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.52 (s, 2H), 8.80 (s, 2H), 8.66 (s, 2H), 8.39 (d, J = 7.64 Hz, 2H), 7.73 (d, J = 7.94, 2H), 7.51 (t, J = 7.43 Hz, 2H), 7.43 (t, J = 7.56, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.21, 133.66, 133.00, 132.53, 128.87, 128.44, 124.44. HRMS found 423.9586, M+H. MP = 279–281 °C.

8: 2,2′-Bis[(3-bromophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-bromobenzaldehyde (1.63 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.80 g, 98%. ¹H NMR (400 MHz, DMSO-d6) δ 12.28 (s, 2H), 8.66 (s, 2H), 8.39 (s, 2H), 8.29 (t, J = 3.08 Hz, 2H), 7.87 (d, J = 7.80, 2H), 7.68 (dd, J = 8.00, 2.79, 2H), 7.45 (t, J = 7.86, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 136.13, 133.80, 131.34, 129.95, 128.05, 122.83. HRMS found 423.9589, M+H. MP = 258–260 °C.

9: 2,2′-Bis[(4-bromophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-bromobenzaldehyde (1.63 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 g, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.73 g, 94%. ¹H NMR (400 MHz, DMSO-d6) δ 12.32 (s, 2H), 8.61 (s, 2H), 8.41 (s, 2H), 7.91 (d, J = 8.24 Hz, 4H), 7.70 (d, J = 8.21, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.32, 148.29, 133.05, 132.23, 130.23, 124.70. HRMS found 423.9587, M+H. MP = 295–297 °C.

10: 2,2′-Bis[(4-methylphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-methylbenzaldehyde (1.06 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.11 g, 84%. ¹H NMR (400 MHz, DMSO-d6) δ 12.18 (s, 2H), 8.46 (s, 2H), 8.39 (s, 2H), 7.83 (d, J = 7.8 Hz, 4H), 7.30 (d, J = 7.8 Hz, 4H), 2.37 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 152.08, 148.22, 140.13, 130.00, 128.74, 127.24, 20.51. HRMS found 294.1708, M+H. MP = 246–248 °C.

11: 2,2′-Bis[(4-methoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-methoxybenzaldehyde (1.2 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.37 g, 94%. ¹H NMR (400 MHz, DMSO-d6) δ 12.00 (s, 2H), 8.37 (s, 2H), 8.35 (s, 2H), 7.88 (d, J = 9.1 Hz, 4H), 7.04 (d, J = 9.1 Hz, 4H), 3.83 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 181.34, 152.55, 129.56, 125.90, 114.22, 55.38. HRMS found 326.1608, M+H. MP = 218–220 °C.

12: 2,2′-Bis[(4-trifluoromethylphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of trifluoromethylbenzaldehyde (1.53 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.75 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.55 (s, 2H), 8.78 (s, 2H), 8.55 (s, 2H), 8.19 (d, J = 7.8 Hz, 4H), 7.85 (d, J = 8.1, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.12, 147.44, 137.21, 130.44, 130.12, 128.51, 125.53, 125.39, 122.68. HRMS found 402.1140, M+H. MP = 273–275 °C.

13: 2,2′-Bis[(4-trifluoromethylphenyl)ethylidene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-trifluoromethylacetophenone (1.65 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.78 g, 96%. ¹H NMR (400 MHz, DMSO-d6) δ 12.04 (s, 2H), 8.89 (s, 2H), 8.28 (d, J = 7.6 Hz, 4H), 7.81 (d, J = 8.4, 4H), 2.5 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 154.75, 141.00, 128.26, 125.63, 123.28. HRMS found 430.1455, M+H. MP = 334–336 °C.

14: 2,2′-Bis[(2-trifluoromethoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 2-trifluoromethoxybenzaldehyde (1.67 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.87 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.54 (s, 2H), 8.75 (s, 2H), 8.71 (s, 2H), 8.48 (d, J = 7.8 Hz, 2H), 7.67–7.50 (m, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 152.21, 146.36, 131.98, 127.28, 127.14, 125.67, 121.03, 120.74, 118.19. HRMS found 434.1039, M+H. MP = 194–196 °C.

15: 2,2′-Bis[(3-trifluoromethoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-trifluoromethoxybenzaldehyde (835 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 660 mg, 70%. ¹H NMR (400 MHz, DMSO-d6) δ 12.43, (s, 2H), 8.71 (s, 2H), 8.47 (s, 2H), 8.09 (s, 2H), 7.93 (d, J = 7.6 Hz, 2H), 7.63 (t, J = 7.9, 2H), 7.49 (d, J = 8.3 Hz, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.48, 149.29, 147.93, 136.22, 131.31, 128.06, 123.45, 121.84, 120.13. HRMS found 434.1039, M+H. MP = 241–243 °C.

16: 2,2′-Bis[(4-trifluoromethoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-trifluoromethoxybenzaldehyde (1.67 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.03 g, 55%. ¹H NMR (400 MHz, DMSO-d6) δ 12.24 (s, 2H), 8.61 (s, 2H), 8.45 (s, 2H), 8.10 (d, J = 8.9 Hz, 4H), 7.50 (d, J = 8.1 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 150.26, 133.07, 130.34, 121.70. HRMS found 434.1041, M+H. MP = 278–280 °C.

17: 2,2′-Bis[(4-trifluoromethoxyphenyl)ethylidene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-trifluoromethoxyacetophenone (1.79 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.88 g, 95%. ¹H NMR (400 MHz, DMSO-d6) δ 11.91 (s, 2H), 8.78 (s, 2H), 8.19 (d, J = 8.8 Hz, 4H), 7.44 (d, J = 8.1, 4H), 2.46 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 154.66, 149.85, 136.38, 129.60, 121.79, 121.17, 119.24. HRMS found 462.1352, M+H. MP = 323–325 °C.

18: 2,2′-Bis[(4-difluoromethoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-difluoromethoxybenzaldehyde (757 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 g, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 0.63 g, 73%. ¹H NMR (400 MHz, DMSO-d6) δ 12.28 (s, 2H), 8.56 (s, 2H), 8.43 (s, 2H), 8.02 (d, J = 8.6, 4H), 7.38 (t, J = 74, 2H), 7.29 (d, J = 8.6 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.17, 130.67, 130.22, 119.01, 116.58. HRMS found 398.1231, M+H. MP = 245–247 °C.

19: 2,2′-Bis[(4-difluoromethoxyphenyl)ethylidene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-difluoromethoxyacetophenone (818 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 990 mg, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 11.78 (s, 2H), 8.72 (s, 2H), 8.13 (d, J = 8.8 Hz, 4H), 7.36 (t, J = 73.56, 2H), 7.24 (d, J = 8.6, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 154.53, 152.90, 152.66, 133.97, 129.39, 119.19, 118.54, 116.62, 114.06, 15.33. HRMS found 426.1538, M+H. MP = 273–275 °C.

20: 2,2′-Bis[(2-cyanophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 2-cyanobenzaldehyde (1.15 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.23 g, 88%. ¹H NMR (400 MHz, DMSO-d6) δ 12.88 (s, 2H), 8.8 (s, 2H), 8.74 (s, 2H), 8.52 (d, J = 7.8 Hz, 2H), 7.96 (d, J = 7.8 Hz, 2H), 7.84 (t, J = 7.7 Hz, 2H), 7.68 (t, J = 7.7, 2H), 7.34 (t, J = 50.59, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 145.01. 136.32, 133.91, 133.88, 131.56, 127.55, 117.67, 111.51, 56.48, 19.03. HRMS found 316.1301, M+H. MP = 213-215 °C.

21: 2,2′-Bis[(3-cyanophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-cyanobenzaldehyde (1.15 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.40 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.44 (s, 2H), 8.74 (s, 2H), 8.56 (s, 2H), 8.46 (s, 2H), 8.22 (d, J = 8.1 Hz, 2H), 7.95 (d, J = 7.8 Hz, 2H), 7.71 (t, J = 7.6 Hz, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.60, 147.27, 135.10, 134.21, 133.25, 131.24, 130.50, 118.95, 112.49. HRMS found 316.1300, M+H. MP = 278–280 °C.

22: 2,2′-Bis[(4-cyanophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-cyanobenzaldehyde (1.15 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.40 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.69 (s, 2H), 8.79 (s, 2H), 8.52 (s, 2H), 8.17 (d, J = 8.2 Hz, 4H), 7.97 (d, J = 8.3 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.64, 147.70, 138.14, 133.11, 128.92, 119.09, 113.05. HRMS found 316.1303, M+H. MP = 303–305 °C.

23: 2,2′-Bis[(3-cyanophenyl)ethylidene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-cyanoacetophenone (1.28 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.40 g, 92%. ¹H NMR (400 MHz, DMSO-d6) δ 11.89 (s, 2H), 8.90 (s, 2H), 8.63 (t, J = 1.44 Hz, 2H), 8.33 (dt, J = 7.75, 1.39, 2H), 7.92 (dt, J = 7.76, 2.44, 2H), 7.67 (t, J = 7.88, 2H), 2.48 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 154.70, 152.16, 138.21, 133.59, 132.06, 131.04, 130.09, 119.21, 112.16, 15.29. HRMS found 344.1612, M+H. MP = 304–306 °C.

24: 2,2′-Bis(phenylmethylene)carbonimidic Dihydrazide Hydrochloride

A solution of benzaldehyde (933 mg, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.19 g, 99%. ¹H NMR (400 MHz, DMSO-d6) δ 12.23 (s, 2H), 8.53 (s, 2H), 8.44 (s, 2H), 7.96–7.94 (m, 4H), 7.50–7.49 (m, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 133.27, 130.77, 128.73, 127.86. HRMS found 266.1398, M+H. MP = 245–247 °C.

25: 2,2′-Bis[(4-phenol)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-hydroxybenzaldehyde (1.1 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.33 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 11.95 (s, 2H), 10.15 (s, 2H), 8.28 (s, 2H), 7.75 (d, J = 8.1 Hz, 4H), 6.87 (d, J = 8.3, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 160.56, 130.15, 124.76, 116.10. HRMS found 298.1296, M+H. MP = 185–187 °C.

26: 2,2′-Bis[(4-diphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-phenylbenzaldehyde (1.60 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.81 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.32 (s, 2H), 8.59 (s, 2H), 8.49 (s, 2H), 8.05 (d, J = 7.3 Hz, 2H), 7.81 (d, J = 7.3 Hz, 2H), 7.77 (d, J = 7.6 Hz, 2H), 7.51 (t, J = 7.3 Hz, 2H), 7.42 (t, J = 7.6 Hz, 1H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.28, 142.71, 139.72, 132.89, 129.52, 128.99, 128.51, 127.42, 127.28. HRMS found 418.2022, M+H. MP = 286–288 °C.

27: 2,2′-Bis[(4-nitrophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-nitrobenzaldehyde (1.33 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.56 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.68 (s, 2H), 8.82 (s, 2H), 8.57 (s, 2H), 8.33 (d, J = 8.3 Hz, 4H), 8.24 (d, J = 8.3 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.79, 148.78, 147.27, 139.98, 129.35, 124.35. HRMS found 356.1098, M+H. MP = 285–287 °C.

28: 2,2′-Bis[(4-dimethylaminophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of N,N-dimethylaminobenzaldehyde (1.31 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 90 mg, 6%. This compound was chemically unstable in chromatography solvents but was greater than 80% pure when used in in vitro assays. ¹H NMR (400 MHz, DMSO-d6) δ 11.74 (s, 2H), 8.22 (s, 2H), 8.14 (s, 2H), 7.71 (d, J = 7.8 Hz, 4H), 6.75 (d, J = 7.8 Hz, 4H), 3.00 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 152.50, 152.35, 129.69, 129.41, 121.22, 120.96, 112.01. HRMS found 352.2237, M+H. MP = 132–134 °C.

29: 2,2′-Bis[(phenyl-4-carbonate)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-formylphenylcarbonate (1.32 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 600 mg, 39%. ¹H NMR (400 MHz, DMSO-d6) δ 13.13 (s, 2H), 12,57 (s, 2H), 8,70 (s, 2H), 8.52 (s, 2H), 8.11 (d, J = 7.7 Hz, 4H), 8.03 (d, J = 7.7 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 193.48, 167.31, 167.02, 153.47, 148.46. 139.35, 137.72, 136.12, 132.79, 130.39, 130.03, 128.40. HRMS found 354.1195, M+H. MP = 336–338 °C.

30: 2,2′-Bis[(4-carbonamidophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-carbonamidobenzaldehyde (655 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 770 mg, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.43 (s, 2H), 8.85 (s, 2H), 8.50 (s, 2H), 8.13 (s, 2H), 8.04 (d, J = 7.8 Hz, 4H), 7.98 (d, J = 7.8 Hz, 4H), 7.50 (s, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 167.70, 153.41, 148.59, 136.32, 136.26, 128.30, 128.15. HRMS found 352.5100, M+H. MP = 309–311 °C.

31: 2,2′-Bis[(4-sulphonamidophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-sulphonamidobenzaldehyde (814 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 920 mg, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.41, (s, 2H), 8.69 (s, 2H), 8.50 (s, 2H), 8.14 (6, J = 7.6 Hz, 4H), 7.91 (d, J = 7.6, 4H), 7.49 (s, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.49, 146.00, 136.81, 128.72, 125.43. HRMS found 424.0852, M+H. MP = 285–287 °C.

32: 2,2′-Bis[(4-propynoxyphenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-propynoxybenzaldehyde (704 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 g, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 0.82 g, 100%. ¹H NMR (400 MHz, DMSO-d6) δ 12.19 (s, 2H), 8.42 (s, 2H), 8.37 (s, 2H), 7.91 (d, J = 8.4 Hz, 4H). ¹³C NMR (400 MHz, DMSO-d6) δ 159.67, 153.09, 129.93, 127.06, 115.56, 79.38, 79.06, 56.09. HRMS found 374.1606, M+H. MP = 217–219 °C.

33: 2,2′-Bis[(4-morpholinophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-morpholinobenzaldehyde (840 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 850 mg, 90%. ¹H NMR (400 MHz, DMSO-d6) δ 11.84 (s, 2H), 8.26 (s, 4H), 7.77 (d, J = 8.1 Hz, 4H), 7.01 (d, J = 8.1 Hz, 4H), 3.75 (s, 8H), 3.24 (s, 8H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.09, 152.75, 149.18, 129.58, 123.87, 114.53, 66.40, 47.78. HRMS found 436.2449, M+H. MP = 280–282 °C.

34: 2,2′-Bis[(3-tetrazole-1-phenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 3-tetrazole-1-benzaldehyde (766 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 g, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 870 mg, 100%. This compound was not sufficiently soluble in chromatography solvents to obtain a quantitative purity by HPLC. ¹H NMR (400 MHz, DMSO-d6) δ 8.84 (s, 2H), 8.75 (s, 2H), 8.66 (s, 2H), 8.20 (t, J = 9.0 Hz, 4H), 7.76 (t, J = 7.78, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 152.27, 147.44, 133.85, 129.83, 129.27, 128.30, 125.70, 124.25. HRMS found 402.1641, M+H. MP = 289–291 °C.

35: 2,2′-Bis[(4-tetrazole-1-phenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-tetrazole-1-benzaldehyde (1.53 g, 8.8 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 1.75 g, 100%. This compound was not sufficiently soluble in chromatography solvents to obtain a quantitative purity by HPLC. ¹H NMR (400 MHz, DMSO-d6) δ 8.70 (s, 2H), 8.52 (s, 2H), 8.20 (s, 8H), 3.47–3.42 (m, 2H), 1.06 (t, J = 7.0 Hz, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 153.39, 148.45, 136.23, 129.19, 127.77. HRMS found 402.1641, M+H. MP = 277–279 °C.

36: 2,2′-Bis{[4-(2,2-difluoro-2,3-dioxazolo)]phenyl}methylene]carbonimidic Dihydrazide Hydrochloride

A solution of 4-(2,2-difluoro-2,3-dioxazolo)benzaldehyde (820 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 800 mg, 87%. ¹H NMR (400 MHz, DMSO-d6) δ 12.58 (s, 2H), 8.66 (s, 2H), 8.60 (s, 2H), 8.04 (d, J = 7.74 Hz, 2H), 7.52 (dd, J = 7.97, 0.93, 2H), 7.32 (t, J = 8.12, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 142.55, 140.90, 140.27, 130.60, 123.91, 120.71, 116.23, 111.11. HRMS found 426.0814, M+H. MP = 276–278 °C.

37: 2,2′-Bis[(2,4-dimethoxy-5-chlorophenyl)ethylidene]carbonimidic Dihydrazide Hydrochloride

A solution of 2,4-dimethoxy-5-chloroacetophenone (655 mg, 4.4 mmol, 2.2 equiv) and 1,3-diaminoguanidine hydrochloride (250 mg, 2 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. Yield: 400 mg, 39%. ¹H NMR (400 MHz, DMSO-d6) δ 11.64 (s, 2H), 8.55 (s, 2H), 7.68 (s, 2H), 6.85 (s, 2H), 3.94 (s, 6H), 3.91 (s, 6H), 2.31 (s, 6H). ¹³C NMR (400 MHz, DMSO-d6) δ 157.97, 156.84, 154.64, 154.53, 130.97, 130.65, 120.76, 112.75, 98.42, 98.30, 57.21, 56.96, 56.79, 32.07, 19.31. HRMS found 482.1350, M+H. MP = 229–231 °C.

38: 2-(Phenylmethylene)-2′-[(2-Fluorophenyl)methylene]carbonimidic Dihydrazide Hydrochloride

A solution of benzaldehyde (466 mg, 4.4 mmol, 1.1 equiv), 2-fluorobenzaldehyde (546 mg, 4.4 mmol, 1.1 equiv), and 1,3-diaminoguanidine hydrochloride (500 mg, 4 mmol, 1 equiv) in ethanol (5 mL) was refluxed for 16 h. Diethyl ether (10 mL) was added, and the product carbonimidic dihydrazide crashed out of solution as a white solid. The product was filtered, washed with diethyl ether, and recrystallized from methanol as a hydrochloride salt. This product was isolated as a mixture also containing 24 and 2. 38 was the dominant product accounting for greater than 50% of the total material in the mixture. Yield: 1.27 g mixture, 50%. ¹H NMR (400 MHz, DMSO-d6) δ 12.30 (s, 3H), 8.69 (s, 1H), 8.62 (s, 1H), 8.57 (s, 1H), 8.53 (s, 1H), 8.44 (s, 1H), 8.35 (t, J = 6.5 Hz, 1H), 7.96–7.94 (m, 2H), 7.56 (q, J = 8.1 Hz, 1H), 7.51–7.49 (m, 2H), 7.34 (t, J = 8.1, 2H). ¹³C NMR (400 MHz, DMSO-d6) δ 133.27, 130.78, 128.74, 127.89, 127.28, 124.80, 115.88. HRMS found 284.1301, M+H. MP = 281–283 °C.

General Biology: Parasite Culture and Drug Sensitivity

The following parasite strains were used in this study and obtained through BEI Resources, NIAID, NIH. Plasmodium falciparum, Strain D6 (MRA-285, originally from Sierra Leone, has modest resistance to mefloquine).⁴⁶ Strain Dd2 (MRA-150, originated from Indochina; derived from W2-mef and is resistant to chloroquine, pyrimethamine, and mefloquine). P. falciparum strain Tm90–C2B (Thailand; resistant to mefloquine, chloroquine, atovaquone, and pyrimethamine) was obtained directly from the Division of Experimental Therapeutics of Walter Reed Army Institute of Research (WRAIR) in Silver Spring, Maryland, United States.³⁹ Strain SB1-A6 (MRA-1002, Sierra Leone was derived from D6 clone and is resistant to Atovaquone and ELQ-300.⁴⁰

P. falciparum parasites were thawed from frozen stock and cultured in suspended human erythrocytes (Lampire Biological Laboratories, Pipersville, PA) not more than 28 days old at 2% hematocrit. The culture medium used was RPMI-1640, supplemented with 25 mM HEPES buffer, 25 mg/L gentamicin sulfate, 45 mg/L hypoxanthine, 10 mM glucose, 2 mM glutamine, and 0.5% Albumax II (complete medium).⁴³ Cultures were maintained in a standard low oxygen atmosphere (5% O₂, 5% CO₂, 90% N₂) in an environmental chamber and incubated at 37 °C. Cultures were subpassaged every 3–4 days into a fresh culture flask containing complete media and erythrocytes.

IC₅₀ Determination by the Fluorescence-Based SYBR Green Assay

The aminoguanidine series was assessed for in vitro antiplasmodial activity using the fluorescence-based SYBR Green assay described previously by Smilkstein and co-workers.³⁸ Compounds were evaluated in quadruplicate in flat-bottomed Costar clear 96-well plates (Model #3585). A 2-fold serial dilution of each compound was performed across the columns of the test plates starting with 20 μM and ending with a final untreated column to serve as control wells. Asynchronous parasite infected erythrocytes in growth media were added to each well for a total volume of 100 μL, final hematocrit of 2%, and initial parasitemia of 0.2%. The commercial malaria drugs atovaquone and chloroquine were used as control drugs. Test plates were incubated for 72 h at 37 °C in an environmental chamber with a controlled low oxygen atmosphere (5% O₂, 5% CO₂, 90% N₂). After the incubation period, 100 μL SYBR Green I dye-detergent solution was added to each well, and the plates were incubated at ambient temperature and atmosphere in the dark for at least 1 h. Fluorescence was read at 497 nm excitation and 520 nm emission bands using a Spectramax iD3 plate reader. Fluorescence readings were normalized with respect to the untreated control wells representing normal parasite growth and plotted against the logarithm of drug concentration. An IC₅₀ was determined for each compound by fitting this data to a variable slope nonlinear regression curve using Graphpad Prism software (v. 8).

HepG2 Cytotoxicity Assay

Compounds were prepared as 10 mM stock solutions in DMSO. Human hepatocarcinoma (HepG2) cells were maintained in culture at 37 °C in a humidified 5% CO₂ atmosphere in RPMI-1640 medium containing 10% fetal bovine serum. HepG2 cells were added to each well of flat bottom 96-well tissue culture plates at an initial density of 2 × 10⁴ cells per well and an initial volume of 160 μL complete medium per well. After an overnight incubation at 37 °C to adhere the cells to the culture plates, 40 μL drug solutions in complete medium were applied to each well at a final concentration range of 0 to 200 μM across each plate. Drugs were tested in triplicate or quadruplicate. The cells were incubated for 24 h at 37 °C and 5% CO₂ with the drug solutions, which were then aspirated and replaced with 200 μL per well of complete medium for an additional 24 h incubation under the same conditions. To each well was added 20 μL of resazurin (Alamar Blue) in PBS buffer to a final concentration of 10 μM, and the plates were incubated for 3 h. Fluorescence was measured at 560 nm excitation and 590 nm emission bands using a Spectramax iD3 plate reader. Fluorescence readings were normalized with respect to the untreated control wells and plotted against the logarithm of drug concentration. An IC₅₀ was determined for each compound by fitting this data to a variable slope nonlinear regression curve using Graphpad Prism software (v. 8).

In Vivo Efficacy against Murine Malaria

The in vivo ED₅₀ and ED₉₀ of selected aminoguanidines was measured using a modified 4-day Peters test. Female CF1 mice from Charles River Laboratories were inoculated intravenously with approximately 2.5–5.0 × 10⁴ parasitized erythrocytes (murine malaria P. yoelii, Kenya strain MR4 MRA-428) from a donor mouse (experiment day zero). On the following 4 days (experiment days 1–4), solutions of the test compounds in PEG-300 (or PEG-300 only for control mice) were administered by oral gavage once daily. Aminoguanidines were initially assessed at 2.5, 5, and 10 mg/kg/day, and experiments were repeated to adjust the dose range as needed to obtain an interpolated ED₅₀ and ED₉₀ value (1 required a dosing down to 0.1 mg/kg/day to attain this result). Experiments were repeated with doses up to 25 mg/kg/day to obtain a nonrecrudescence dose, though only 16 was found to produce a cure in this model. Parasitemia of each mouse was determined by microscopic examination of Giemsa stained blood smears on day 5. ED₅₀ and ED₉₀ values were assessed by generating dose–response curves relative to untreated controls using Graphpad Prism (v. 8). Mice were considered cured of malarial infection if they maintained 0% parasitemia at experiment day 30. The procedures involved, together with all matters relating to the care, handling, and housing of the animals used in this study, were approved by the Portland VA Medical Center Institutional Animal Care and Use Committee.

Murine Microsomal Stability

Metabolic stability studies of selected aminoguanidines were performed at ChemPartner, Shanghai, China. Compounds were incubated at 37 °C and 1 μM concentration in murine liver microsomes (Corning) for 1 h at a protein concentration of 0.5 mg/mL in potassium phosphate buffer at pH 7.4 containing 1.0 mM EDTA. The metabolic reaction was initiated by NADPH and quenched with ice-cold acetonitrile at 15 min increments up to 1 h. The progress of compound metabolism was followed by LC-MS/MS (ESI positive ion, LC-MS/MS-034(API-6500+) using a C18 stationary phase (ACQUITY UPLC BEH C18 (2.1 × 50 mm, 1.7 μm)) and a MeOH/water mobile phase containing 0.25% FA and 1 mM NH₄OAc. Imipramine or Osalmid were used as internal standards, and ketanserin was used as a metabolically unstable control compound. Concentration versus time data for each compound were fitted to an exponential decay function to determine the first-order rate constant for substrate depletion, which was then used to calculate the degradation half-life (t_1/2) and predicted intrinsic clearance value (Cl_int) from an assumed murine hepatic blood flow of 90 mL/min/kg.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsinfecdis.1c00001.

• Calculated error values for antiplasmodial IC₅₀ and cytotoxicity assays (PDF)

The authors declare the following competing financial interest(s): Coauthors M.R. and A.K. have filed an invention disclosure relating to selected compounds described in this paper to the United States Department of Veterans Affairs and to Oregon Health & Science University as required by law.

Dedication

This manuscript is dedicated to Dr. Jonathan Vennerstrom in honor of his 65th birthday and for his outstanding contributions to the field of antimalarial drug research and development.