Novel Antimalarial Tetrazoles and Amides Active Against the Hemoglobin Degradation Pathway in Plasmodium falciparum

Abstract

Malaria control programs continue to be threatened by drug resistance. To identify new antimalarials we conducted a phenotypic screen and identified a novel tetrazole-based series that shows fast-kill kinetics and a relatively low propensity to develop high-level resistance. Preliminary structure activity relationships (SAR) were established including identification of a sub-series of related amides with antiplasmodial activity. Assaying parasites with resistance to antimalarials led us to test whether the series had a similar mechanism of action to chloroquine (CQ). Treatment of synchronized P. falciparum parasites with active analogs revealed a pattern of intracellular inhibition of hemozoin (Hz) formation reminiscent of CQ’s action. Drug selections yielded only modest resistance that was associated with amplification of the multidrug resistance gene 1 (pfmdr1). Thus, we have identified a novel chemical series that targets the historically druggable heme polymerization pathway, and that can form the basis of future optimization efforts to develop a new malaria treatment.

Graphical Abstract

INTRODUCTION

Malaria remains a leading cause of mortality and morbidity worldwide. Global malaria deaths in 2020 were estimated at ~0.4 million, with African children under the age of 5 representing the most vulnerable demographic and accounting for 272,000 (67%) of all malaria deaths.¹ An estimated 228 million cases occurred worldwide in the same period. In high-burden countries malaria continues to place a significant strain on health care systems while leading to substantial losses in economic output, creating a cycle of poverty. The causative agent of malaria is an obligate intracellular parasite from the Plasmodium genus, of which there are five human infective species. Plasmodium falciparum and Plasmodium vivax account for most cases of malaria, with the former responsible for almost all malaria deaths.²

A number of drugs with a range of modes of action (MOA) have been used to treat malaria, including chloroquine (CQ), which became the mainstay of treatment and prevention programs in the late 1940s.³ Widespread P. falciparum resistance to CQ led to its loss of efficacy, a fate shared by sulfadoxine+pyrimethamine (which inhibit dihydropteroate synthase and dihydrofolate reductase, respectively) that was introduced in the late 1950s. Parasite resistance extends to nearly every other drug that has been used in malaria treatment programs.⁴ The introduction of artemisinin-based combination therapies (ACTs) as first-line treatment regimen restored effective malaria control and their use has led to substantial reductions in malaria disease burden in the last two decades.⁵ However, increasing ACT treatment failure rates are being reported, particularly in the Greater Mekong Subregion. P. falciparum resistance to artemisinin (ART) is associated with mutations in Kelch13 and with delayed parasite clearance times.^6–9 These reductions in ART potency have paralleled the emergence of resistance to partner compounds¹⁰, further jeopardizing the efficacy of first-line antimalarials and foreshadowing a reversal of the gains made in the last decade.⁵ New challenges in emerging resistance to current frontline therapies make the discovery of novel antimalarials a critical component to sustaining and improving malaria control programs.

Academic and commercial antimalarial high-throughput screening (HTS) campaigns have leveraged whole cell-based approaches to identify novel antimalarial candidates, some currently in clinical development.^{2, 11–12} Early phenotypic screening campaigns focused on targeting asexual blood stage (ABS) parasites, which cause clinical symptoms. This approach led to the identification of several clinical candidates for the treatment of malaria, including the spiroindolone ATP4 inhibitor KAE609 (NITD609)^13–14, the imidazolopiperazine, KAF156^15–16 and PI4 kinase inhibitors^17–18 including KDU691 and MMV390048. More recent approaches have since allowed development of screening platforms targeting multiple stages of the Plasmodium life-cycle.^19–21 These phenotypic screening efforts, combined with the use of forward genetics and whole-genome sequencing (WGS), have proven powerful in identifying the protein targets of antimalarial compounds, including for several of these current development candidates.^22–23 Target-based screens, including those against dihydroorotate dehydrogenase that led to the identification of DSM265^24–25, have also been deployed but so far have led to fewer development candidates than the phenotypic approach.

Herein, we describe the identification of a novel tetrazole-based series from a phenotypic screen against P. falciparum ABS parasites (using the 3D7 strain). This series shows a fast kill rate in vitro and life-cycle profiling reveals it to be ABS-specific. Synthesis and evaluation of additional analogs in the series has established preliminary structure activity relationships (SAR). MOA studies suggest that parasite killing is mediated by inhibition of hemozoin (Hz) formation and the toxic build-up of free heme in the parasite digestive vacuole (DV), similar to CQ. Selections for resistant parasites has yielded only low-level resistance, and WGS reveals amplification of the multidrug resistance 1 gene (pfmdr1). We have thus identified a novel class of antimalarial compounds with a fast-kill mechanism, low resistance propensity and whose MOA appears to involve inhibition of heme detoxification, a critical pathway in the parasite’s intra-erythrocytic development and a well validated antimalarial target. These compounds also have the advantage of rapid synthesis from readily available starting materials.

RESULTS

Phenotypic screen to identify novel compounds active against P. falciparum ABS parasites

A single-point (5 μM) primary screen of an 8K subset of our larger library was executed in a high throughput optimized 384-well format against the pan-susceptible P. falciparum 3D7 strain of ABS parasites, using a published SYBR Green assay.^26–27 The assay was highly robust, with an average Z’ of 0.76±0.059 (where >0.5 is considered a robust assay) across all plates (Fig. 1A). We identified a total of 79 hits based on an RZ score cut off of < −3. The RZ score represents the number of robust standard deviations (RSD) from the corrected median of a measurement for a library compound, as determined after correction for systematic errors (plate, row, column effects).^28–29 Compounds with structural alerts (e.g. PAINS compounds³⁰) or poor drug-like properties³¹ were removed and a secondary screen at three concentrations (5, 1.7, and 0.42 μM) was performed, yielding 38 confirmed hits. We prioritized compounds with EC₅₀ values < 3 μM against 3D7 parasites and selected for chemical novelty based on PubChem and SciFinder(R) searches, while deprioritizing hits that have been previously reported in other malaria screening programs. From our initial hits we selected 3 chemotypes (1, 2 and 3) for commercial resupply and evaluated them in 10-point concentration response assays against 3D7 parasites (Fig. 1B). These compounds ranged in potency on 3D7 from 0.31 – 3.5 μM, with 1 showing the best potency (EC₅₀ values of 0.31 μM (1); 0.71 μM (2); and 3.5 μM (3))(Fig. 1 legend). Cytotoxicity was assessed against HepG2 cells, with all three compounds showing a CC₅₀ > 30 μM in this assay.

Fig. 1. High-throughput screening-based identification of compounds that inhibit growth of P. falciparum 3D7 ABS parasites.

A. Range of Z’ values observed for 384-well plates covering the 8K screening library. The black bar shows the mean ± standard deviation (std dev). B. Representative concentration response curves for validated hits versus 3D7 parasites. Data for each drug concentration were collected in triplicate wells and error bars represent the std dev. Data were fitted to the log(I) vs response – four parameter variable slope equation in GraphPad Prism. Replicate data were subsequently collected, yielding the following EC₅₀’s: 1, 0.31 ± 0.10 μM; 2, 0.71 ± 0.075 μM; 3, 3.5 ± 0.71 μM, where errors are the std dev for at least three independent biological replicates. C. BRRoK rate of kill assay. Serial log-fold dilutions of compounds from 30× EC₅₀ to 0.33× EC₅₀ were incubated for 6 h with NF54-luc parasites and viability was assessed by bioluminescence at the end of the incubation. Bioluminescence signals normalized to untreated controls are plotted and represent mean ± std dev from triplicate wells. EC₅₀ values against 3D7 parasites were determined in standard 72 h SYBR Green assays prior to the start of the study and were used as indicated for compounds 1 – 3. Benchmark compounds included artemisinin, ART (fast-kill) (EC₅₀ = 0.013 μM), chloroquine, CQ (fast-kill) (EC₅₀ = 0.014 μM), atovaquone, ATQ (slow-kill) (EC₅₀ = 0.0013 μM), and DSM265 (slow-kill) (EC₅₀ = 0.0050 μM), with EC₅₀ values used to calculate compound concentrations for the study indicated in parenthesis. Similar EC₅₀ values were obtained versus NF54-luc parasites using bioluminescence as a readout in a 72 h assay (Supplemental Table S1).

Compounds that show fast kill kinetics are being sought because they provide rapid relief of symptoms, and a more rapid reduction in parasite numbers may also reduce the propensity for resistance to develop.^32–33 Kill rates also provide information about the MOA as the kill rate has been shown to be a property of the target.³⁴ To this end, compounds were profiled to estimate the rate of kill (Fig. 1C) using a rapid bioluminescence relative rate of kill (BRRoK) assay.³⁵ This method follows bioluminescence in a cytocidal 6 h exposure assay over multiple drug concentrations using a P. falciparum strain (PfNF54^luc)³⁶ that expresses luciferase. We benchmarked the assay against known fast (ART and CQ) or slow (DSM265²⁵ and atovaquone (ATQ)) acting compounds previously characterized using the standard parasite reduction rate (PRR) assay, which measures viability after treatment and regrowth over 28 days.³⁴ In the BRRoK assay both 1 and 2 showed slow kill kinetics similar to DSM265 and ATQ, while the profile for 3 was similar to CQ and thus consistent with a fast kill mechanism. Based on this desirable property, the 3-series was selected for more in-depth profiling despite lower potency compared to 1 and 2.

Chemistry

Tetrazoles were synthesized using a one-pot, four-component Ugi coupling in racemic form³⁷ (Scheme 1A). An amine, aldehyde, isocyanate and trimethylsilyl azide were combined in methanol and stirred for 12 – 36 h at room temperature (RT) to provide the corresponding coupling product in high yields. In several cases, the enantiomers were separated by preparative HPLC using a chiral stationary phase. Similarly, most amides were prepared through a three component Ugi coupling featuring phenyl phosphinic acid³⁸ (Scheme 1B). The resulting secondary amides were alkylated using methyl iodide in the presence of sodium hydride. Finally, amide 34 was prepared via a Strecker reaction involving 3-pyridine carboxaldehyde and amine 35. Nitrile hydrolysis under basic conditions and amide coupling provided product 34 in high yield (Scheme 1C).

Scheme 1.

Synthesis of antimalarial compounds

Medicinal chemistry to establish SAR of the 3-series

The 3-series is chemically attractive, with a basic amine, a stable heterocycle, and four vectors for optimization. Initial SAR of the 3-series was established through the synthesis and testing of 19 tetrazole analogs against two strains of P. falciparum. These included the 3D7 strain that is broadly drug-sensitive and the Dd2 strain that is resistant to chloroquine and partially resistant to mefloquine (MFQ) and quinine.^39–40 Compounds were also tested against human HepG2 cells to evaluate cytotoxicity. This work identified compounds with improved potency versus 3D7 of 10 to 20-fold (Table 1).

Table 1.

SAR of the Tetrazole series

Cmpd	R	R1	R2	B-ring	C-ring	P. falciparum EC50 (μM)		Cytotox CC50 (μM)
						Pf3D7	PfDd2	HepG2
3	2-OCH3	Br	CH3	Ph	o-tol	3.5±0.71 (10)	5.4±0.78 (5)	>30
(+)-3	2-OCH3	Br	CH3	Ph	o-tol	1.5, 1.9	3.0	ND
(−)-3	2-OCH3	Br	CH3	Ph	o-tol	4.6, 4.4	4.0	ND
4	2-OCH3	H	CH3	Ph	o-tol	9.7±3.0 (3)	10±1.5 (3)	ND
5	2-OCH3	CF3	CH3	Ph	o-tol	1.7±0.67 (4)	3.7±0.25 (3)	ND
6	2-OCH3	CF3	CH3	3-Py	c-tol	1.3±0.12 (3)	2.3±0.40 (3)	ND
(+)-6	2-OCH3	CF3	CH3	3-Py	c-tol	1.6, 1.5	2.5	ND
(−)-6	2-OCH3	CF3	CH3	3-Py	o-tol	1.2, 1.2	1.7	ND
7	2-OCH3	CF3	cPr	3-Py	o-tol	0.49±0.10 (3)	0.48±0.029 (3)	16
8	2-F	CF3	cPr	3-Py	o-tol	0.60±0.17 (3)	0.67±0.031 (3)	ND
9	H	CF3	cPr	3-Py	o-tol	0.38±0.071 (5)	0.54 ±0.17 (3)	15
10	H	CF3	cPr	5-Pyr	o-tol	0.40±0.070 (4)	1.8±0.34 (4)	37, 33
11	3-F	CF3	cPr	5-Pyr	o-tol	0.37±0.041 (4)	1.2±0.67 (4)	36
(+)-11	3-F	CF3	cPr	5-Pyr	o-tol	0.35	2.5	>30
(−)-11	3-F	CF3	cPr	5-Pyr	o-tol	0.42	2.6	>30
12	4-F	CF3	cPr	5-Pyr	o-tol	0.47±0.11 (3)	*EC50.1;EC50.2 0.68±0.10; 3.6±1.3 (4)	ND
13	6-F	CF3	cPr	5-Pyr	o-tol	0.47±0.18 (3)	*EC50.1;EC50.2 0.71±0.13; 4.8±1.3 (4)	ND
14	H	CF3	cPr	cPr	o-tol	4.8, 4.7	3.2, 4.2	ND
15	H	CF3	COCH3	5-Pyr	o-tol	15, 15	>30, 21	>30
16	2-F	CF3	CH3	3-Py	o-tol	1.0±0.14 (3)	2.5±0.26 (3)	ND
17	2-F	CF3	CH3	3-Py	cHex	1.7±0.56 (3)	2.6±0.49 (3)	ND
18	H	CF3	cPr	3-Py	cPr	1.5±0.29 (3)	3.0±0.55 (3)	ND
19	H	CF3	cPr	2Me- 3Pyr	cPr	>30, 14	21, 16	>30
20	H	CF3	nPr	5-Pyr	o-tol	0.31±0.056 (3)	0.52±0.11 (3)	13
21	H	CF3	nPr	3-Py	o-tol	0.23±0.025 (3)	0.41±0.083 (3)	10

Concentration response data were collected with technical duplicates or triplicates at each inhibitor concentration. Data were fitted in GraphPad Prism to the Inhibitor vs response- variable slope (four parameters) equation to determine the EC₅₀, with the exception of *, where biphasic curves were observed and data were fitted to the biphasic equation with constraints (top = 1.0, bottom >0, hill coefficients <0) to determine EC_50.1 and EC_50.2. Data are the EC₅₀ mean ± std dev with the number of independent biological replicates shown in parenthesis. Individual values are shown for studies with less than 3 biological replicates. Ph, phenyl; Py, pyridyl; Pyr, pyrimidinyl; tol, tolyl. Compounds are racemic unless otherwise indicated.

On the A-ring, the Br could be replaced with CF₃ to give 5 with a slight improvement in activity, while removing the Br decreased activity by a factor of 3 (4). Removing the A-ring altogether resulted in an inactive analog (data not shown). Replacing the B-ring phenyl with 3-pyridine (3-Py) further improved activity to ~ 1 μM (6) while also improving drug-like properties. The N-methyl at the R2 position could be replaced by cyclopropyl (cPr) (7) yielding an additional 3-fold improvement in potency against 3D7. The OMe on the A-ring could be replaced by F (8) or H (9) without loss of activity. Further modification of the B-ring showed that the 3-Py of 9 could be replaced by a 5-substituted pyrimidine (5-Pyr, 10) with similar activity against 3D7. Introduction of fluoro on the A-ring was also well tolerated at the 3 (11), 4 (12) and 6 (13) positions on the ring. The B-ring could not be replaced with a small non-aromatic cPr ring (14), and a basic amine appeared important because the R2 alkyl group could not be replaced with COCH₃ (15). In contrast, the C-ring could be replaced by either cyclohexyl (17) or cPr (18) with only a 2 or 3-fold loss of activity against 3D7 compared to compounds with matched R2 and B rings, 16 or 9, respectively. However, within the context of the cPr in the C-ring position, addition of a 2-Me to the B-ring 3-Py was not tolerated (19). The benzylic carbon between the A ring and the tertiary amine could be substituted with a cPr to block a potential site of metabolism. When combined with a propyl (Pr) at R2, these compounds yielded the most potent of the tetrazole series (20 and 21) (3D7 EC₅₀’s ~ 0.3 μM, Table 1), representing a 10-fold improvement over the initial hit 3.

Three analogs were selected for separation and evaluation of enantiomer pairs. A small 2–3-fold difference in potency was observed between the separated 3-enantiomers ((+)-3 vs. (−)-3), while the enantiomers of 6 ((+)-6 and (−)-6) and 11 ((+)-11 and (−)-11) showed equal potency. Taken together the data supports an absence of enantiomer selectivity for the series.

While the initial hit 3 showed similar activity on both 3D7 and Dd2 parasite lines, a number of the subsequent compounds in the series showed reduced activity on Dd2, with differences of up to 5–10-fold observed for the most disparate compounds (e.g. 10, 12 and 13) (Table 1). These differences will be discussed in greater detail below. Cytotoxicity was evaluated using human HepG2 cells. While 3 did not show any activity up to the highest tested concentration of 30 μM, several of the more potent analogs (9, 20 and 21) displayed cytotoxicity with CC₅₀ values in the 10–15 μM range, yielding a selectivity window of 40–50-fold in these cases (Table 1).

Molecular modeling suggested that a cis-amide could orient the A, B and C rings similarly to the tetrazole. A subseries of amides was synthesized to test this hypothesis (22-34) (Table 2). Consistently, 22 and 23 were equipotent to their tetrazole congeners 9 and 10. The NH amides were 5-fold less active, as predicted based on conformational considerations (data not shown). The SAR from the two sub-series was found to be similar, consistent with a similar MOA. This includes the finding that for the B-ring both 3-Py (22, 24, 26, 28) and 5-Pyr (23, 25, 27 and 29) showed similar activity against 3D7, while similar to the tetrazoles 9 and 10, Dd2 activity on the 5-Pyr tended to be lower than for the 3-Py. The A-ring CF₃ could be replaced with SCF₃ (24, 25) or OCF₃ (26, 27), yielding similar potency. The methyl on the C-ring could be extended to ethyl to improve activity approximately 2-fold (22 vs. 28; 23 vs. 29) whereas removing the C-ring methyl was associated with a ~2-fold loss in activity (23 vs. 30). Addition of a 2-Me to the 5-Pyr at the B-ring led to a substantial 20-fold loss in potency (31 vs 23). As observed with the tetrazole series, addition of 2-F to the A ring was well tolerated (32 vs. 22; compare to 16 in the tetrazole series), while a benzylic cyclopropyl group again improved activity to give 33 (versus 20 and 21 in the tetrazole series), currently our most potent analog (EC₅₀ = 0.21 – 0.33 μM on 3D7 and Dd2, respectively). Additionally, as with the tetrazoles, replacement of the C-ring with a saturated ring (piperidine 34) led to a 5-fold reduction in activity but showed some promise. Relative to other compounds in the series, compound 30 showed the greatest loss of potency against Dd2 parasites, requiring over 20-fold higher concentrations for complete killing (Table 2). Cytotoxicity in the amide series was improved over the tetrazoles, with no compounds showing any cytotoxic activity up to the highest tested concentration of 30 μM (Table 2).

Table 2.

SAR of the Amide sub-series

Cmpd	R	R1	R2	R3	B-ring	P. falciparum EC50 (μM)		Cytotox CC50 (μM)
						Pf3D7	PfDd2	HepG2
22	H	CF3	cPr	Me	3-Py	0.34±0.044 (3)	0.59±0.067 (3)	>30
23	H	CF3	cPr	Me	5-Pyr	0.40±0.015 (3)	0.79±0.19 (3)	>30
24	H	SCF3	cPr	Me	3-Py	0.31±0.078 (3)	0.39±0.045 (3)	>30
25	H	SCF3	cPr	Me	5-Pyr	0.26±0.046 (3)	1.0±0.55 (5)	>30
26	H	OCF3	cPr	Me	3-Py	0.32±0.035 (3)	0.52 ±0.055 (3)	>30
27	H	OCF3	cPr	Me	5-Pyr	0.24±0.044 (3)	0.59±0.17 (5)	>30
28	H	CF3	cPr	Et	3-Py	0.19±0.025 (3)	0.48±0.084 (3)	>20
29	H	CF3	cPr	Et	5-Pyr	0.24±0.042 (3)	0.57±0.14 (3)	>25
30	H	CF3	cPr	H	5-Pyr	0.72±0.021 (3)	*EC50.1;EC50.2 2.0±1.4; 16±5.4 (4)	>30
31	H	CF3	cPr	Me	2-Me, 5-Pyr	8.1, 7.5	9.8, 7.7	>30
32	F	CF3	cPr	Me	3-Py	0.31±0.026 (3)	0.46±0.046 (3)	>30
33	H	CF3	nPr	Me	3-Py	0.21±0.021 (3)	0.33±0.035 (3)	>15
34	H	CF3	cPr	C*	3-Py	1.8±0.15 (3)	4.2 ±0.84 (3)	>20

Concentration response data were collected with technical triplicates at each inhibitor concentration and data were fitted in GraphPad Prism to determine the EC₅₀ as described in the Table 1 footnote. Data are the EC₅₀ mean ± std dev with the number of independent biological replicates shown in parenthesis.

^*C-ring replaced with piperidine.

To confirm that the 3-series shows fast kill kinetics we re-evaluated the kill rate using more potent analogs from both the tetrazole (10) and amide (23) sub-series using the BRRoK assay cited above. Like 3, both compounds behaved similarly to CQ in this assay, consistent with fast kill kinetics (Fig. 2A and supplemental Fig. S1). Finally, 10 was tested in both a flow cytometry (FACS) assay⁴¹ that detects the ability of parasites to invade new RBCs after incubation with drug and in the PRR-based washout assay³⁴, with both assays confirming that 10 exhibited kill kinetics similar to CQ, though slower than ART (Fig. 2).

Fig. 2. Relative Rate of Kill of 10 versus control antimalarials.

A. BRRoK rate of kill assay. Conditions used were as described in Fig. 1. The mean EC₅₀ for 10 was 0.32 μM, as determined using 72 h SBYR Green assays. DSM265 and ART controls were collected simultaneously to the 10 data. *CQ data were replotted from Fig. 1 and are provided again in this plot as a reference. B. Kill rate determination using two-color FACS analysis. Benchmark compounds are described above. Pyrimethamine (PYR) was included as a medium rate of kill compound. Compounds were plated at 10× EC₅₀; where EC₅₀ values, determined using 48 h hypoxanthine assays, were: 10 (1.7 μM); CQ (0.021 μM); ART (0.03 μM); PYR (0.094 μM); and ATQ (0.001 μM). C. PRR washout assay. Data show % parasite survival versus h of drug treatment prior to drug washout and replating. PYR data were collected simultaneously to 10. CQ and ART data are historical control data collected in a separate study³⁴ and provided as a reference. Triplicate data were collected at each compound concentration, symbols represent the mean, and error bars represent the std dev.

ADME assessment of selected tetrazole and amide compounds

A preliminary assessment of the physicochemical properties and metabolic stability was conducted for several compounds in the tetrazole and amide subseries (Table 3). Lipophilicity in the series was high, and this property correlated with overall poor metabolic stability in human liver microsomes for all tested compounds. Compound 7 and some with lower cLogD (10 and 34) performed marginally better than other compounds in the series and overall, the tetrazoles were slightly better than the amides. Solubility in pH 6.5 PBS was poor for the tetrazoles, but was substantially better for the amides at both pH 2.0 and 6.5, with 34 showing the best overall properties in these assays. Overall, these studies identified a significant metabolic liability that would need to be addressed before these series could be expected to show in vivo efficacy.

Table 3.

Physicochemical properties and in vitro metabolism in human liver microsomes

Cmpd	MW	cLogD pH 7.4	Predicted pKa (basic)	CLint (μL/min/mg protein)	Kinetic Solubility pH 2.0 (μg/mL)	Kinetic Solubility pH 6.5 (μg/mL)
7	495	5.2	4.6, 3.3	154	12.5 – 25	1.6 – 3.1
9	465	5.3	5.1, 2.9	360	12 – 23	1.5 – 2.9
10	465	4.6	4.7	261	1.5 – 2.9	1.5 – 2.9
22	454	5.3	6.0, 3.4	781	>100	6.3 – 12.5
23	454	4.6	5.5	531	11 – 23	>45
32	472	5.4	5.0, 3.4	548	>47	1.5 – 2.9
34	418	3.9	6.1, 3.4	244	>100	25–50

Life-cycle Profiling

Representative compounds, one each from the tetrazole (9) and amide (32) series, were profiled against liver and sexual stage parasites to determine if the series would have prophylactic or transmission-blocking activity (Table 4). Liver stage activity was assessed against P. berghei to determine whether compounds could block liver stage development in HepG2 cells infected by sporozoites.^{19, 21} Modest activity was observed in this assay, but the Pb liver stage EC₅₀ for both compounds was 20 to 30-fold higher than for P. falciparum blood stages. The dual gamete formation assay (DGFA) was used to assess the ability of compounds to block maturation of both male and female stage V gametocytes, providing a readout of functional viability. The assay identifies compounds that either kill stage V gametocytes or interfere with gamete formation. No activity was observed against female gametocytes for either compound, while modest activity was seen for 32 against male gametocytes (~60% inhibition at 25 μM). Thus the 3-series compounds do not have significant liver stage or transmission-blocking activity, positioning them for use solely for treatment of ABS infections.

Table 4.

Life cycle profiling

Cmpd	Pb liver stage EC50 (μM)	HepG2 CC50 (μM)	DGFA (male/female) gametocytes EC50 (μM)
9	7.3, 12	19, >17	>25/>25
32	7.0, 16	24, 24	~60% at 25 μM/ >25

EC₅₀ values are shown for two independent replicates.

DGFA, Dual gamete formation assay.

Evaluation of cross resistance to known drugs suggests a common resistance mechanism with Hz-inhibiting antimalarials

To obtain insight into the potential MOA of the 3-series compounds we evaluated compounds against a selection of drug-sensitive and drug-resistant parasite lines. Initial work focused on comparing the effectiveness against 3D7 and Dd2 parasites. As noted above, while 3 showed similar activity against these strains (Table 1), a number of analogs in both the tetrazole and amide sub-series showed lower activity against Dd2 than 3D7 (Tables 1 and 2). The largest differences (3–5-fold) were observed for compounds with a 5-Pyr in the B-ring (10, 26), whereas the matched compounds with 3-Py in this position showed smaller (1.4 – 3.0-fold) effects (9, 27) based on comparisons of matched pairs that contained only variations in this position (Fig. 3 and Supplemental Fig. S2). Additionally, for several analogs (12, 13 and 30) containing the 5-Pyr in the B-ring we observed biphasic concentration response curves versus Dd2. In these cases, up to a 20-fold potency difference was observed between 3D7 and the second of the fitted EC₅₀ values determined for Dd2, whereas the more potent EC₅₀ values were similar to that observed for 3D7 (Fig. 4 and Tables 1 and 2). These data suggest that a second target might be responsible for the reduced Dd2 efficacy at higher concentrations for these compounds. In contrast, biphasic concentration response curves were never observed versus 3D7 for any compound. Encouragingly, our data show that it is possible to identify compounds with similar potency against 3D7 and Dd2, for example by avoiding the 5-Pyr that frequently led to biphasic concentration response curves. These results highlight the value of including multiple parasite lines in early-stage compound evaluation.

Fig. 3. Reduced efficacy of 3-analogs containing 5-Pyr against P. falciparum Dd2 compared to 3D7 parasites.

Comparison of representative concentration response curves for tetrazoles 9 (A) and 10 (B), and for amides 26 (C) and 27 (D), tested against 3D7 (green circles) and Dd2 (black squares) parasites. Results showed that compounds containing 5-Pyr have poorer activity against Dd2 than do compounds containing 3-Py in the B-ring. Compound structures are shown above the graphs. Data were fitted to the [I] vs response, variable slope (four parameter) equation in GraphPad Prism to determine the EC₅₀. Triplicate data were collected at each concentration, symbols represent the mean, and error bars represent the std dev. EC₅₀ data are reported in Tables 1 and 2, which contain average data for at least three biological replicates.

Fig. 4. Biphasic growth curves were observed for select 5-Pyr 3-series compounds assayed against Dd2 parasites.

Comparison of representative concentration response curves for tetrazoles 12 (A) and 13 (B), and for amide 30 (C) tested against 3D7 (green circles) and Dd2 (black squares) parasites. Compound structures are displayed above the graphs. Triplicate data were collected at each concentration, symbols represent the mean, and error bars represent the std dev. EC₅₀ data and curve fitting methods are reported in Tables 1 and 2.

To further probe the MOA, 9 and 10 were evaluated against two panels of resistant parasite lines that were either obtained as clinical isolates⁴² or were generated in Dd2 parasites through in vitro selections with various antimalarials. Data were compared to the drug-sensitive 3D7 and NF54 strains (Table 5A; note that 3D7 is a clone of NF54). No cross resistance was observed for 9 across the two panels. In contrast 10 showed reduced potency against the CQ-resistant Dd2 and piperaquine (PPQ)-resistant RF12 strains, although reduced activity was not observed against other tested lines with other drug-resistant phenotypes. Interestingly, both 9 and 10 were more effective against 7G8 parasites than any of the other tested lines even though this line, similar to Dd2, RF12, K1 and TM90C2B, contains point mutations in the P. falciparum CQ resistance transporter (PfCRT; Table 5B).

Table 5A.

Activity of 9 and 10 versus parasite strains with resistance to compounds from diverse structural classes and mechanisms of action.

Pf parasite line	Resistance profile	9	9	10	10
		EC50 (μM)	Fold Diff*	EC50 (μM)	Fold Diff*
	Drug-resistant field isolates42, 44, 49
3D7	Sensitive	0.38	1.0	0.40	1.0
NF54	Sensitive	0.48	1.3	0.83	2.1
K1	CQ, PYR	0.36	0.95	0.46	1.2
7G8	CQ, PYR	0.18±0.042 (3)	0.47	0.16 ±0.024 (3)	0.40
TM90C2B	CQ, ATQ, Pyr	0.46	1.2	0.66	1.7
RF12	PPQ, Pyr	0.61	1.6	1.1	2.8
Dd2	CQ, Pyr	0.54	1.4	1.8	4.5
	PfCRT gene edited isogenic lines
Dd2-GC03	Sensitive	0.89±0.12 (3)	1.0	4.8	1.0
Dd2-Dd2	CQ	0.45±0.13 (3)	0.53	2.4	0.50
Dd2-M343L	PPQ, CQ	0.38±0.10 (3)	0.41	0.90	0.19
	Lab-derived Dd2 lines resistant to development candidates
PfeEF2	DDD10749887	0.92	1.7	1.6	0.89
Pfpi4K	MMV39004818	0.75	1.4	2.0	1.1
PfDHODH	DSM26525	0.80	1.5	2.0	1.1
Pfcarl	KAF15688	0.65	1.2	2.0	1.1
PfCytB	ELQ30089	0.85	1.6	2.0	1.1

^*Fold differences were calculated relative to 3D7 for field isolates, relative to Dd2-GC03 for the isogenic gene-edited lines, or relative to Dd2 for lab-selected Dd2 resistant lines. The GC03 line (which contains the pfcrt wild-type allele, also found in 3D7 or NF54 ) was previously isolated from the Dd2×HB3 cross and is CQ sensitive. The gene edited lines were constructed using customized zinc-finger nucleases and have been previously reported.^46–48 Dd2-GC03 was derived by introducing the 3D7 pfcrt sequence into Dd2 in place of the endogenous pfcrt Dd2 locus (sequence shown in Table 5B). Dd2-Dd2 was engineered using the same method to replace the endogenous pfcrt locus with the recombinant Dd2 sequence and Dd2-M343L replaced the endogenous pfcrt locus with the recombinant Dd2 sequence plus the PPQ resistance-conferring M343L^{44, 49} mutation. Concentration response data were collected with technical duplicates at each inhibitor concentration and reported data represent the average of two biological replicates and where additional biological replicates were collected (number in parenthesis) the std dev is provided. Additional replicate data for NF54 can be found in Supplemental Table S2A. Development candidates used for selections are listed, and parasite lines are named based on the protein target of the resistance allele. CQ, chloroquine, ATQ, atovaquone, PYR, pyrimethamine, PPQ, piperaquine. Data for Dd2 and 3D7 were taken from Table 1.

Table 5B.

Documented point mutations in PfCRT across the tested strains

Strain		PfCRT position and encoded amino acid
	72	74	75	76	97	220	271	326	343	356	371
3D7	C	M	N	K	H	A	Q	N	M	I	R
NF54	C	M	N	K	H	A	Q	N	M	I	R
Dd2	C	I	E	T	H	S	E	S	M	T	I
7G8	S	M	N	T	H	S	Q	D	M	L	R
RF12	C	I	E	T	Y	S	E	S	M	T	I
K1	C	I	E	T	H	S	E	S	M	I	I
TM90C2B	C	I	E	T	H	S	E	S	M	T	I

Data were taken from.^{23, 43–44, 49, 90} Colors highlight residue differences between strains: yellow (3D7 and Dd2), green (3D7 and 7G8) and turquoise (3D7 and RF12). RF12 is also known in the literature as PH1263-C. 3D7 is a clone of NF54.⁴⁹

PfCRT in certain mutant forms mediates CQ efflux from the digestive vacuole (DV) and is causal for CQ resistance.^{23, 43–45} While each strain harbors a key set of PfCRT mutations (including K76T, Table 5B) that confer CQ resistance, the specific set of point mutations differs across strains and mediates differences in sensitivity to CQ as well as other 4-aminoquinolines such as PPQ and amodiaquine.^{44, 46–47} The 7G8 mutational background has also been reported to increase sensitivity to lumefantrine (LUM).^46–47 Thus each PfCRT isoform could display different binding affinities and transport levels of 3-series compounds as well, and thus modulate parasite susceptibility to these compounds differentially. This hypothesis was investigated by testing gene-edited isogenic lines that allow comparison between different PfCRT alleles with varied sensitivity to CQ in the same genetic background. These lines were generated previously by zinc-finger nuclease-mediated editing of Dd2 parasites to insert: 1) the wild-type pfcrt allele (Dd2-GC03); 2) the recombinant Dd2 pfcrt allele (Dd2-Dd2) (differing from wild-type PfCRT by eight point mutations Table 5B); or 3) the recombinant Dd2 pfcrt allele plus theM343L mutation (Dd2-M343L), resulting in PPQ resistance.^{44, 46–49} We found that either the Dd2 PfCRT background or the Dd2-M343L background led to a 2–3-fold increased sensitivity to both 9 and 10 (Table 5A) relative to Dd2-GC03. These data suggest that the wild-type PfCRT transporter might bind better to 9 and 10 than do the variant Dd2 or Dd2+M343L isoforms, leading to lower drug concentrations in the DV. Interestingly the addition of M343L has also been reported to reduce the degree of CQ resistance by 2 to 3-fold, compared with the Dd2 isoforms.⁴⁹ The finding that the EC₅₀ for 10 on Dd2-GC03 is higher than for Dd2 wild-type shows that the poor activity of 10 on Dd2 relative to 3D7 is not linked to differences in the pfcrt allele.

Heme polymerization is inhibited by 3-series compounds

Similarities between CQ and the 3-series compounds suggested to us the potential for a common MOA. CQ functions by interrupting heme detoxification via inhibition of the biomineralization of toxic labile heme into the chemically inert Hz in the DV.^{43, 50–51} Similarities include the killing kinetics, the life-cycle profile showing activity only on ABS and the observations that CQ-resistant (7G8 and Dd2) and genetically engineered lines containing PfCRT mutations showed differential sensitivity to compounds in the 3-series, suggesting that like CQ and other quinolines the 3-series compounds targeted a process in the DV. Additionally, the lack of enantiomer selectivity of the 3-series supports the hypothesis of a non-protein target such as heme polymerization.

To investigate the activity of 3, 9 and 10 against the heme detoxification pathway, we tested the ability of these compounds to inhibit conversion of hematin (Fe(III)PPIX) to the synthetic β-hematin (βH) form of Hz. These experiments used a pyridine-based detergent-mediated colorimetric assay that simulates the physiological micro-environment of the DV.⁴⁷ Standard antimalarials with diverse MOAs were tested in parallel to serve as controls. As expected, CQ, amodiaquine, MFQ and pyronaridine inhibited βH formation with IC₅₀ ≤100 μM, while pyrimethamine, dihydroartemisinin and doxycycline did not inhibit conversion of hematin to βH (Fig. 5). Among the tested compounds, 9 inhibited βH formation with the same potency as MFQ, while 10 showed weaker inhibition. Compound 3 was nearly inactive in the concentration range that could be studied (Fig. 5). The IC₅₀ values for inhibition of βH formation (Fig. 5) of the 3-series compounds correlated loosely with their EC₅₀ values for inhibiting NF54 proliferation (Table 5 and supplemental Table S2). 9 was the most potent in both assays, supporting the hypothesis that the mechanism of cell killing was related to inhibition of βH formation. The concentration of compound required to inhibit βH formation was ~1000-fold higher than the EC₅₀ for parasite survival not only for the 3-series but also for the 4-aminoquinolines. These differences are thought to reflect the fact that drug accumulation in the DV is in a function of pH trapping⁵², hence the extent of accumulation determines the concentration at the site of action.

Fig. 5. β-Hematin inhibition assay (βHIA) for 3-series compounds in comparison to antimalarial control compounds.

A. Representative titration data showing inhibition of β-hematin formation versus concentration profiles for 3, 9 and 10 in comparison to known antimalarials, using the β-hematin inhibition assay (βHIA).⁹¹ Positive (pyronaridine, amodiaquine, CQ and MFQ) and negative controls were assayed, with the latter representing a diverse set of MOAs (the DHODH inhibitor DSM1⁹², artesunate, sulfadoxine, and doxycycline). Data show the mean ± std dev for duplicate data at each compound concentration. Data were fitted to the Sigmoidal concentration response (variable slope) equation in GraphPad Prism to determine the IC₅₀. B. Fitted IC₅₀ values. Data represent the mean ± std dev for 4 independent studies.

Cellular Heme Fractionation

To mechanistically validate the in vitro heme polymerization results described above, we applied a heme fractionation assay to measure inhibition of Hz formation in NF54 parasites. In this assay, “% free heme”, % Hz (determined as proportions of total heme species extracted), and levels of “free” heme-Fe and Hz-Fe (absolute amounts of each species per cell calculated from total amount of Fe obtained using a heme standard curve) were quantified from isolated trophozoites after their treatment with compounds. This assay tested a range of compound concentrations at or above their EC₅₀ for antimalarial activity (0.5× − 3× EC₅₀). For the positive CQ control, both the % “free” heme and the levels of “free” heme-Fe showed a significant concentration-dependent increase as the drug concentration was increased above the EC₅₀ compared to the no-drug control (Fig. 6Ai), whereas Hz and Hz-Fe decreased over the same concentration range (Fig. 6Aii). Overlaying the parasite survival curve with the “free” heme-Fe titration curve showed that the concentration of CQ leading to 50% activity (EC₅₀ for heme-Fe release versus EC₅₀ for parasite survival) was similar in both assays (Fig. 6Aiii), consistent with CQ being a bona fide inhibitor of Hz formation.⁵³ This result contrasts with the data for the dihydrofolate reductase inhibitor pyrimethamine (Fig. 6B), included as a negative control to illustrate the robustness of this protocol in discriminating agents that do not inhibit Hz formation. No significant increase in free heme was observed over the course of the titration, regardless of drug concentration.

Fig. 6. Cellular heme fractionation and concentration response versus parasite survival.

Data are shown for A. CQ, the positive control, for B. Pyr, the negative control and for two 3-analogs C. compound 9 and D. compound 10. Panels i and ii. Heme species present as ‘free’ heme (panel i) and Hz Fe (fg) (panel ii) isolated per cell from drug-treated PfNF54 parasites. Data represent the mean ± std dev from three biological replicates each arising from two technical replicates (See supplemental Table S3) Panel iii. An overlay of parasite survival with the total amount of ‘free’ heme Fe as a function of drug concentration. Free heme data represent the mean ± std dev from three biological replicates (taken from panels i), while representative parasite survival data from one biological replicate with technical duplicates is displayed (See supplemental Table S2A for average EC₅₀’s from three biological replicates). Data were fitted to the sigmoidal concentration response (variable slope) equation in GraphPad Prism. Calculated EC₅₀ values for the drug concentration giving 50% response are as follows: CQ, 58 nM (heme) and 15 nM (survival); PYR, >100 nM (heme), 33 nM (survival); Cmpd 9, 620 nM (heme), 440 nM (survival); Cmpd 10, 1040 nM (heme), 910 nM (survival). See Supplemental Table S2A and S2B for error analysis. Significance was assessed using a two-tailed Welch’s t-test, where * <0.05, ** <0.01, *** <0.001.

Treatment with 9 and 10 led to a robust build-up of free heme (calculated as a percent of total heme species) and the level of heme-Fe, with a corresponding decrease in Hz levels (Figs. 6C and 6D). These compounds led to significantly larger amounts of free heme release (6-fold increase in free heme-Fe for 9 and 10 versus only 2-fold for CQ) and a corresponding reduction in Hz levels at 3× EC₅₀ compared to CQ (2-fold decrease in Hz heme Fe for 9 and 10 compared to 1.3-fold for CQ), demonstrating a higher amount of free heme at the compound concentrations that disrupt parasite growth. Similar to CQ, overlaying the parasite survival curve with the “free” heme-Fe titration curve showed that the concentrations of 9 and 10 at the EC₅₀ of heme-Fe release versus EC₅₀ for parasite survival were very similar in both assays. These data suggest that the MOA of the 3-series compounds is at least in part due to inhibition of Hz formation.

Selections for P. falciparum 3D7 parasites resistant to compound 9

To obtain additional insight into the potential mode of resistance to 3-series compounds, we performed single-step selection studies to obtain parasites resistant to 9. Selections were performed with a starting parasite inocula of 2×10⁹ 3D7 parasites per flask and two separate flasks were pressured with 9 at a concentration of 3× EC₅₀. Drug-treated parasites cleared in both flasks by day 5 and recrudesced on day 25. Bulk cultures showed EC₅₀ shifts of only 2.4 to 3.4-fold compared to the parental line. Three clones were then obtained from each flask by limiting dilution. EC₅₀ and ED₉₀ shifts for these 6 clones were found to be a modest 2.8 – 2.6-fold (Table 6). The high starting inocula, and the modest shifts in EC₅₀ for the selected mutant lines, suggests that it is relatively difficult to evolve resistance against 9.

Table 6.

In vitro selections for Pf3D7 parasites resistant to 9 showing EC₅₀/EC₉₀ shifts of clonal lines.

	Parental	9 resistant clonal 3D7 lines
clone	3D7-A10	F1-B6	F1-C1	F1-E4	F2-A10	F2-A11	F2-H12
EC50 μM	0.30±0.033	0.55±0.058	0.54±0.061	0.74±0.12	0.83±0.11	0.67±0.029	0.64±0.055
Fold shift	NA	1.8	1.8	2.5	2.8	2.3	2.2
EC90 μM	0.59±0.12	1.1±0.16	1.0±0.17	1.5±0.21	1.5±0.10	1.3±0.041	1.3±0.078
Fold shift	NA	1.9	1.8	2.6	2.6	2.3	2.2

Data represent the mean ± standard error of the mean for three independent replicates.

WGS performed on two clones per flask found that each clone had a copy number variation (CNVs) on chromosome 5, with flask 1 (F1) clones showing 3- to 4-fold amplification and flask 2 (F2) clones showing 5-fold amplification of the region (Tables S4 and S5 and Supplemental Fig. S3A). Clones from the two flasks differed in the chromosome 5 segment that was amplified. Nonetheless, all clones overlapped in the multidrug resistance gene 1 (pfmdr1, PF3D7_0523000) locus. This gene has previously been shown to be involved in MFQ, LUM and quinine drug resistance and can also modulate CQ potency.^54–55 This was the only change observed in all 4 clones, suggesting that it represents the major mechanism of resistance to 9 in these lines.

In addition to the CNV on chromosome 5, F2 clones A10 and A11 had another set of CNVs on chromosome 12, showing a ~2-fold amplification (Table S5 and Supplemental Fig. S3B). It is interesting to note that one of the eight genes in this amplified locus (Pf3D7_1223700) encodes a proposed vacuolar iron transporter, which could potentially be involved in modulating the consequences of disrupting heme polymerization in the DV. Finally, F1 clones F1-B6 and F1-E4, had single nucleotide polymorphisms (SNPs) in two common genes (Supplemental Table S6), a M199I substitution in PF3D7_0208200 (KRR1 small subunit processome component) and a N353K substitution in PF3D7_1316700 (unknown function), while F1-B6 also contained a D4190N mutation in PF3D7_1343800 (unknown function). The relevance of these SNPs to the resistance mechanism of 9 is unknown.

DISCUSSION

The ability of P. falciparum to develop resistance to virtually every drug that has been used clinically exemplifies the importance of working on this pathogen to further elucidate its biology and identify new drugs. Herein we describe the identification of a novel, fast-acting antimalarial compound series that has efficacy against ABS parasites. In addition to showing fast kill kinetics, selection of resistant parasites towards one of the more potent compounds 9 generated only modest levels of resistance (2–3-fold) and WGS suggested that resistance was associated with amplification of the pfmdr1 gene. We showed that 3-series compounds 9 and 10 inhibited heme polymerization both in vitro and in P. falciparum cellular assays. Importantly 9 showed equal efficacy on both CQ-sensitive and CQ-resistant strains. Thus the 3-series meets the objective of identifying fast kill compounds with a relatively low propensity of resistance for the treatment of malaria, and they provide a new scaffold that targets heme polymerization, a highly successful strategy for killing malaria parasites. Based on their lack of liver stage or sexual stage activity these compounds would be best positioned in the portfolio for treatment, and their fast kill kinetics could allow them to be suitable for development in non-ART based therapies Further work to optimize both potency and drug-like properties would however be needed for this series to advance to lead optimization and to in vivo activity.

A limited medicinal chemistry effort was undertaken for the 3-series to define the structural requirements for activity. We found that parasite killing was not dependent on the stereochemistry, and that both enantiomers were equally active for those that were tested. The A- and B-rings could be aryl or heteroaryl, but they could not be removed or replaced with aliphatic rings. Substantial substitution on the A ring was tolerated, which allowed us to replace the original Br/OMe substitution with more drug like CF₃/F groups. Similarly, we were able to improve the polarity of the series by replacing the B-ring phenyl group with heteroaryl rings including 3-pyridine (3-Py) and 5-pyrimidine (5-Pyr). This ring showed the steepest SAR profile with substantial loss in activity associated with substitution on various positions or alternative heteroaryl rings. Activity was improved by replacing the N-methyl group with larger N-alkyl groups, and the benzylic CH₂ could be replaced with a cyclopropyl group, blocking a potential metabolic liability. The tetrazole could be replaced with a conformationally restricted amide bond, and both series showed similar SAR. Overall, potency improvements of 10–20-fold were obtained relative to the initial hit 3. The amide sub-series generally showed better selectivity than the tetrazoles. Lipophilicity, indexed by LogD, was high throughout both subseries and metabolic stability was generally poor, although we were able to make some improvements in aqueous solubility.

Profiling of a panel of drug-sensitive or -resistant strains provided the first insights that the 3-series might share a common MOA with Hz inhibitors such as CQ. Direct biochemical evidence that the 3-series compounds inhibit heme polymerization was obtained both in vitro and in cell-based assays. We showed that 9 and 10 inhibited heme biomineralization in the in vitro βH inhibition assay, while in the cell-based assay treatment with these compounds caused a CQ-like signature of an accumulation of free heme with commensurate decreases in Hz levels. These effects titrated with the same dose response as parasite killing. Taken together, these data strongly implicate inhibition of Hz formation as one, but perhaps not the sole, MOA of this series. Interestingly, the levels of free heme generated in the presence of 9 and 10 are 3 – 4-fold higher than observed for CQ across the dose response range, with the larger differences observed at higher drug concentrations. These results could reflect divergent interactions of the two compound classes with heme molecules during heme-drug complex formation. Indeed atomic-force microscopy studies have shown that the various antimalarial 4-aminoquinolines affect the biomineralization of heme molecules in different ways, with CQ and quinine providing the most efficient blocking by forming interactions across the flat surface face (“step-pinning”), whereas MFQ and amodiaquine block addition of heme only to the step edge (“kink-blocking”).⁵¹

Resistance selection studies uncovered resistance mechanisms that overlapped with the CQ-related quinolines like MFQ, providing an additional link between the 3-series and quinoline compounds. First, we observed that the 3-series have a relatively low propensity to develop resistance as selections with 10⁹ parasites yielded EC₅₀ shifts of only 2 to 3-fold. Our WGS analysis found that the one genetic change common to all the resistant clones was 3–5-fold amplification of the pfmdr1 gene, which encodes the P-glycoprotein transporter Pgp1. pfmdr1 amplification is a major driver of clinical and parasite in vitro resistance to MFQ.^55–56 The level of resistance observed for 9 was less than reported for MFQ selection studies, where a lower level of amplification of the pfmdr1 locus (2-fold) yielded a 5-fold IC₅₀ increase and which could be selected with a lower starting inocula (~10⁸ parasites).⁵⁷ A greater impact of amplification was also reported for dihydroorotate dehydrogenase (DHODH) inhibitors, where 3-fold amplification of the dhodh locus led to a 5-fold EC₅₀ increase from ~10⁷ parasites⁵⁸ and 5 to 6-fold amplification arising from ~10⁸ parasites led to a 7-fold EC₅₀ increase.²⁵ Additionally for DHODH inhibitors multiple SNPs resulting in >10 to 30-fold shifts in EC₅₀ have been identified at a frequency of ~10⁸ parasites.^{25, 59} Similar findings have also been reported for other antimalarials in preclinical development where minimum inoculum of resistance of 10⁶ to 10⁸ parasites, mediated often by SNPs or CNVs, led to resistance levels that can range from 2- up to 1,000-fold based on the target, the genetic change and the compound (e.g. PfATP4 or eEF2 inhibited by SNPs in KAE609 or DDD107498, respectively⁴). In this light, our finding of low-level resistance mediated by pfmdr1 amplification is a favorable feature of the 3-series.

The interplay between PfMDR1 and PfCRT and the impact of their allelic variation on drug resistance is complex and compound-dependent. While CQ resistance is primarily associated with PfCRT mutations^{23, 43–45}, low-grade resistance to other quinolines such as MFQ and the arylaminoalcohol LUM is associated with PfMDR1 genetic changes (amplifications or point mutations).^{46–47, 60} PfCRT is localized to the DV membrane and transports compounds out of the DV. PfMDR1 also appears to be primarily localized to the Plasmodium DV membrane of the trophozoite⁶¹, where it has been associated with solute uptake into the vacuole.^{55, 60, 62} Low-level expression of PfMDR1 at the plasma membrane has also been reported⁶¹, leaving open the possibility it could be involved in efflux of compounds out of the parasitized red cell, similar to its mechanism of resistance in mammalian cells. Within this complex milieu, the potency of any given compound is dependent on the concentrations that can reach the DV, which could be influenced by binding affinity to the parasite line-specific transporter isoforms of PfCRT and PfMDR1 and by non-receptor mediated partitioning between the various cellular compartments related to the physical properties of the compound. Our studies on the 3-series suggests that like MFQ and LUM their efficacy can also be modulated by both transporters but to a lesser extent than for the quinolines. Analysis of the gene-edited lines showed that 9 and 10 were more potent on lines harboring PfCRT mutations that are observed in CQ-resistant parasites than on parasites expressing the wild-type allele. These data provide direct evidence that the PfCRT allele impacts the potency of 3-series compounds, and suggest that the wild-type isoform might bind to and transport these compounds better than isoforms found in CQ-resistant parasites. This result is similar to earlier reports for LUM.⁴⁷ Importantly these data provide evidence that the 3-series will be fully effective against CQ-resistant parasites harboring PfCRT mutations found in malaria-endemic regions.

An identified exception to the promising activity on CQ-resistant strains was identified for 3-series analogs that contained 5-Pyr in the B-ring (e.g. 10). These tended to show poorer activity against Dd2, while still being effective against 3D7 and 7G8. Some analogs also exhibited biphasic concentration response curves versus Dd2 parasites. The Dd2 line has a separate PfCRT isoform compared to the 7G8 line^43–44 for which higher potency for both 9 and 10 was observed, but data on the gene-edited lines that harbored only the PfCRT changes associated with Dd2 suggested that the poor activity of 10 on Dd2 is unrelated to PfCRT. PfMDR1 SNPs might also impact the potency in these cases. Dd2 encodes the N86Y mutation whereas 7G8 carries the S1034C/N1042D/D1246Y haplotype. Both PfMDR1 isoforms sensitize parasites to MFQ and LUM (when expressed in parasites with a single copy of pfmdr1) and impact multiple antimalarials that engage with the heme degradation pathway, either by causing increased sensitivity or by mediating a degree of resistance.^{60, 63–64} These data suggest that the N86Y mutation could be associated with the poor activity of some 3-series analogs against Dd2, and could support a model whereby the N86Y PfMDR1 mutation leads to reduced levels of 5-Pyr analogs like 10 in the DV while not impacting localization of the 3-Py analogs like 9. The interplay between the two transporters might also contribute to the biphasic curves observed with the 5-Pyr 3-series analogs (e.g. 10). Dd2 parasites also express 2 to 3 copies of pfmdr1, contrasting with the single copy of pfmdr1 expressed by 3D7 and 7G8 parasites, which may further contribute to the observed differences between Dd2 and other parasite lines. These differences in potency provide a path forward for future medicinal chemistry as compounds with 5-Pyr in the B-ring would be disfavored going forward.

An open question that remains is how the 3-series analogs interact with heme. The crystal structure of halofantrine-ferriprotoporphyrin IX shows that the basic amine of halofantrine (related to quinine) interacts with a carboxylate from heme while the hydroxyl coordinates to the heme iron.⁶⁵ Indeed, a characteristic of nearly all heme-interacting antimalarials is the presence of a protonatable amine. These groups appear to be important for interaction with heme and for accumulation in the low pH environment of the DV. All of the 3-series compounds with EC₅₀ <1 μM have predicted pKa values within reasonable physiological pH ranges, with the central tertiary nitrogen present in all molecules and calculated pKa values around 5–6, suggesting that 3-series compounds may also interact with heme through this basic amine (Table 3). For example, the acetamide 15 lacks a basic amine and is essentially inactive. Similar to the crystal structure of halofantrine, the structure of MFQ bound to ferriprotoporphyrin IX shows an alkoxide bound to Fe and a salt bridge between a basic amine and the porphyrin carboxylate.⁶⁶ In this context, we note that the amine-tetrazole or amine-amide distance in the 3-series is similar to the amine-alcohol distance in MFQ. This observation prompts the speculation that the tetrazole or amide might bind Fe while orienting the basic amine in a position to interact with the heme carboxylate.^67–68

CONCLUSION

Herein we have described a novel tetrazole-based series with antimalarial activity against ABS parasites. The compound class exhibits a fast kill rate and a relatively low propensity to develop resistance, both desirable properties for new antimalarial drugs. The MOA of the series appears to be inhibition of heme polymerization in the parasite DV, similar to 4-aminoquinolines such as CQ. The series thus provides a new chemical scaffold that functions at least in part by this highly effective mechanism for killing malaria parasites. Novel compounds that function by blocking heme biomineralization have been sought through secondary screening of HTS hits from phenotypic ABS screens, and while some new chemical entities that target this mechanism have been discovered, none have yet progressed to preclinical development.^{50, 69–70} Our initial SAR exploration has revealed areas of the scaffold that are subject to modification to optimize activity and physicochemical properties and has shown that substantial improvements are possible. Further efforts will be required to identify compounds with improved potency and ADME properties that will support in vivo efficacy and eventual development. The 3-series also provide new tools for studying the heme biomineralization mechanism and the impact of inhibiting this process on the parasite.

EXPERIMENTAL SECTION

Human samples

Human biological samples were sourced ethically and their research use was in accord with the terms of the informed consents under approved IRB/EC protocols for each site.

P. falciparum culture

P. falciparum cultures were maintained at 2% hematocrit using human blood type O+ RBCs (Valley Biomedical) employing a culturing method adapted from Trager and Jensen⁷¹. Culture media contained RPMI 1640 medium (Gibco) + 25 mM HEPES and 0.5% Albumax-I, supplemented with 23 mM sodium bicarbonate and 92 μM Sigma-hypoxanthine.

Determination of compound potency against P. falciparum in vitro

Antiplasmodial activity was evaluated using a parasite growth SYBR Green-based inhibition assay adapted from published methods.^26–27 For the primary screen (plated at 5 μM) and for hit validation 3-point cherry pick concentration response curves (plated at 5, 1.7 and 0.42 μM), we acoustically dispensed compounds (30 nL of 5 mM stocks in DMSO) into wells using a Labcyte-Echo-555 dispenser. Concentration response curves on repurchased or synthetized compounds were generated from serial dilutions prepared in DMSO in triplicate and plated using a Tecan D300e digital dispenser. The DHODH inhibitor DSM265²⁵ served as a positive control and 0.5% DMSO was plated as a negative control, with all wells normalized to a final concentration of 0.5% DMSO. Culture suspensions were prepared to a 0.5% starting parasitemia at 2% hematocrit and were dispensed (60 μL) using a BioTek- MultiFlo dispenser into 384-well plates containing pre-spotted library compounds. Prepared plates were then incubated for 72 h at 37°C in 5% CO₂, 80% humidity. Reference thick film smears from select control wells were made prior to sealing each plate in aluminum foil and storing at −80°C for 24 h. Plates were thawed at RT and all ensuing steps were conducted under reduced light. SYBR Green stock solution was prepared by mixing Sigma-SYBR® Green I nucleic acid gel stock (1.2 μL) into 1 ml of lysis buffer (80 mM Tris-HCl pH 7.52, 20 mM EDTA, 0.032% w/v Saponin, 0.32% v/v 100% Triton X-100). SYBR Green stock solution (15 μL) was dispensed into each well of the assay plates and plates were sealed in aluminum foil and stored in dark for 4 h at RT. Fluorescence was measured by detecting emission at 535 nm after excitation at 485 nm using a BioTek Synergy H1 Hybrid Reader. Data were fitted to the log(inhibitor) vs. response -- Variable slope (four parameters) model in Graph Pad Prism to determine the effective concentration that led to a 50% reduction in parasitemia (EC₅₀).

Phenotypic screen chemical library

The chemical library used for the screen is an 8K subset of the UT Southwestern compound collection assembled with support from computational chemists at Chemical Diversity Inc (Chem Div) to represent a plate-based diversity subset of our 200K chemical library. Compounds in this subset are compliant with a relaxed set of Lipinski’s rules, with 99% having a molecular weight less than 550. The above SYBR Green assay was optimized on our HTS core platforms as a readout for P. falciparum growth using 384-well plate format. Z’ was evaluated as a measure of assay robustness where a Z’ >0.5 represents a robust assay. Hits were identified based on an RZ score cut off of < −3, where the RZ score represents the number of robust standard deviations (RSD) that a measurement for a library compound is from the corrected robust median, which is determined after correction for systematic errors (plate, row, column effects) as described.^72–73

Pf drug-resistant laboratory strains and cross-resistance testing.

Drug sensitivity testing was performed using the modified [³H]-hypoxanthine incorporation assay, as previously reported.74

HepG2 in vitro cytotoxicity evaluation

Cytotoxicity was evaluated against HepG2 human hepatoma cell suspensions in Sigma-EMEM medium supplemented with 5% heat inactivated fetal bovine serum (Gibco), and 2 mM Glutamine. Compounds were dispensed into 384-well plates using a Labcyte Echo®555 acoustic dispenser. Methotrexate (Sigma) and 0.5% DMSO were used as positive and negative controls. All wells were normalized to a upper limit of 0.5% DMSO. A volume of 60 μL/well was added using a BioTek- MultiFlo dispenser from a 10,000 cells/mL cell suspension preparation of HepG2 cells for a final seeding density of 600 cells/well. Plates were then incubated for 72 h in a humidity chamber at 37°C, 5% CO₂, and 80% humidity. Cell viability was established by quantifying intracellular ATP using a luciferase-coupled ATP quantification assay (Promega-CellTiter Glo®) following the manufacturer’s instructions. Data were fitted to the log(inhibitor) vs. response -- Variable slope (four parameters) model in Graph Pad Prism to determine the cytotoxic concentration that led to 50% reduction in signal (CC₅₀).

P. falciparum relative rate of kill bioluminescence assay (BRRoK)

Parasite kill rate was estimated as previously described³⁵ using a transgenic luciferase-expressing P. falciparum strain, NF54-luc³⁶ (BEI Resources) that had the advantage of expressing luciferase throughout the asexual life cycle. Cultures were maintained in WR99210 (5 nM) but drug was removed by washing and replating in drug-free media for 2 days prior to the kill rate studies. Briefly, NF54-luc cultures were sequentially synchronized to ring stages using 5% D-sorbitol as described.⁷⁵ The assay was then conducted with trophozoite stage cultures (20–26 h post-invasion) at 2% parasitaemia and 4% haematocrit. NF54-luc parasites (200 μL/well) were added to 96-well plates at 2% hematocrit and 2% parasitemia. Triplicate concentrations (30× EC₅₀ − 0.33× EC₅₀) of benchmark antimalarials and experimental compounds were added to each well and then incubated for 6 h at 37°C, 5% CO₂, and 80% humidity. Culture samples (40 μL/well) were then transferred to a white-clear bottom 96-well plate and parasite viability determined by luciferase quantification using a luciferase bioluminescence assay (Promega-Luciferase Assay system).

Two-color flow cytometry kill rate assay

Double-colorimetric FACS analysis was used to quantify invasion of pre-stained human RBCs by drug-treated parasites as described.⁴¹ Briefly, killing profiles were estimated by culturing unlabeled RBCs infected with P. falciparum 3D7 parasites in the presence of compounds at 10× EC₅₀ for 24 or 48 h. The initial % parasitemia was 0.5% with a hematocrit of 2%. Compound EC₅₀ was determined prior to the start of the study using the 48h in vitro ³H-hypoxanthine incorporation assay as described³⁴ for parasites cultured in standard RPMI 1640 media supplemented with 25 mM HEPES and 0.225% NaHCO₃ supplemented with 2% D-sucrose, 0.3% L-glutamine, 0.005 mM hypoxanthine and 0.5% AlbuMax II. Culture media for the kill rate study was identical but contained 0.15 mM hypoxanthine. After drug treatment, compounds were removed and infected RBCs were diluted using fresh RBCs previously labeled with carboxylfluorescein diacetate succinimidyl ester (10 μM for 30 min at 37°C). Following a further 48 h incubation in standard conditions, the ability of treated parasites to establish infection in fresh labeled RBCs was detected by two-color flow cytometry after labeling parasite DNA with Hoechst 33342 solution (2 μM). Parasite viability was measured based on the percentage of infected CFDA-SE-stained RBCs in drug-treated samples versus untreated samples of the initial inocula after 48 h incubations with labeled RBCs.

PRR kill rate assay

PRR was assessed using the standard drug washout assay that has been previously described.³⁴ Briefly, parasites were treated with compound for either 24 or 48 h at 10× EC₅₀ (determined as described above) with compound renewed daily over the treatment period. Parasite samples were removed at the determined time points (0, for the control of initial number of parasites, 24 and 48 h), drug was washed out, and drug-free parasites were cultured in 96-well plates after the addition of fresh RBCs and new media in microtiter plates over a 3-fold serial dilution. Cultures were maintained for up to 28 days to enable parasites to recrudesce. Four independent serial dilutions were done with each sample to correct for experimental variation. The number of viable parasites was back-calculated based on Xⁿ⁻¹ where n was the number of wells where growth was observed and X was the dilution factor.³⁴

Liver stage and gametocidal assays

P. berghei liver stage assays were performed in the Winzeler lab (UCSD) as described.^{19, 21} P. berghei-ANKA-GFP-Luc-SMCON (Pb-Luc) sporozoites were used to infect HepG2-A16-CD81EGFP cells for this assay, which monitors the ability of compounds to block liver stage development of luciferase expressing parasites. Assay media (DMEM without Phenol Red (Life Technology, CA) supplemented with 5% FBS, 1.45 mg/mL glutamine, 500 units of penicillin, and 500μg/mL streptomycin) was used for the Pbluc and HepG2tox assays.

Gametocyte assays measuring the viability of mature stage V gametocytes were conducted to determine the ability of compounds to block formation of male and female gametes in the Baum lab (Imperial College, London) as described.^76–78 The assay identifies compounds that either kill stage V gametocytes, cause sterilization or interfere with the process of gamete formation.

In vitro heme polymerization assay

Lipid-mediated Hz formation was quantified by measuring the formation of βH from hematin, in a detergent-mediated assay that substitutes neutral lipids for the commercially available lipophilic detergent Nonidet P-40 (NP-40).⁷⁹ Unreacted hematin was detected through the formation of bis-pyridyl-Fe(III)PPIX complex which absorbs at a wavelength of 405 nm. Compound stocks (20 mM) of 3-derivatives (3, 9 and 10) and pyrimethamine were prepared in DMSO, while CQ was prepared in water. These stocks were then diluted to 2 mM with a water/NP40 detergent solution, resulting in a final solution of compound in 61.1 mM NP40/10% DMSO. Solubility of 3, 9 and 10 to 2 mM in this buffer was confirmed (see in vitro ADME methods section). Serial dilutions into this same buffer were then made for each compound. A 25 mM hematin stock solution was prepared by sonicating hemin in DMSO for one minute and then suspending 179 μL of this stock in a 1 M acetate buffer (20 mL, pH 4.8). The homogenous suspension was then added to the wells to give final buffer and hematin concentrations of 0.5 M and 100 mM, respectively. Plates were covered and incubated at 37 °C for 5 h. A solution of 50% (v/v) pyridine, 30% (v/v) H₂O, 20% (v/v) acetone and 2 M HEPES buffer (pH 7.4) was prepared and 32 μL added to each well to give a final pyridine concentration of 5% (v/v). Acetone (60 μL) was then added to each well to assist with hematin dispersion. The UV-vis absorbance of the plate wells was read on a SpectraMax P340 plate reader. Sigmoidal concentration-response curves were fitted to the absorbance data using GraphPad Prism version 8 to obtain a 50% inhibitory concentration (IC₅₀) for each compound.

Cellular heme polymerization assay

Inhibition of Hz formation was measured in cultured asexual blood stage PfNF54 parasites through a heme fractionation assay^{53, 80}, where both % “free” heme and levels of “free” heme-Fe were quantified in femtogram per cell (fg/cell) using a heme standard curve from the mass of each heme Fe species per trophozoite. Percent free heme and amounts of free heme-Fe were quantified over a range of drug concentrations (0.5–3× EC₅₀) and compared to the vehicle control. Briefly, early ring-stage NF54 parasites were treated with a combination of successive sorbitol and Percoll treatments to prepare young rings synchronized to within 3 h post-invasion.⁸¹ Young rings were incubated with the test drugs at various multiples of their IC₅₀ values with a no-drug control included (culture conditions and media as described above). After 32 h, late trophozoites/early schizonts were harvested by lysis of the RBCs with 0.05% saponin followed by multiple washes with 1× PBS (pH 7.5) to remove traces of the RBC hemoglobin. Pellets were then resuspended in 1× PBS (pH 7.5). An aliquot of the trophozoite suspension was used to quantify, using flow cytometry, the total number of trophozoites isolated. Contents of the remaining trophozoite pellet were then released by hypotonic lysis and sonication. Following centrifugation, treatment with HEPES buffer (pH 7.4), SDS, pyridine, and NaOH, the fractions corresponding to digested hemoglobin, “free” heme and Hz were carefully recovered. The UV–visible spectrum of each heme fraction as an Fe(III)heme–pyridine complex was measured using a multi-well plate reader (Spectramax 340PC; Molecular Devices). The total amount of each heme species was quantified using a heme standard curve whereby the mass of each heme-Fe species per trophozoite (fg/cell) was calculated by dividing the total amount of each heme species by the corresponding number of parasites in that fraction as determined by flow cytometry. Statistical comparisons and analyses for trends were made on GraphPad Prism version 8 using Students’ t-test (GraphPad Software Inc., La Jolla, CA, USA).

In vitro selection of drug-resistant P. falciparum lines

P. falciparum 3D7 parasites (clone A10) resistant to 9 were selected using previously described methods for single-step drug selections.⁸² Briefly, the EC₅₀ of 9 versus 3D7-A10 (genetically homogenous clonal line) ring-stage parasites (0.2% parasitemia and 1% hematocrit, cultured in RPMI-1640 media supplemented with 0.5% Albumax), was determined in a 72 h assay for a 2-fold dilution series (10 points) of 9 (final DMSO concentration was <0.5%) in duplicate along with DMSO solvent controls. Parasite survival was assessed by flow cytometry on an Accuri C6 (BD Biosciences) using SYBR Green and MitoTracker Deep Red FM (Life Technologies) as nuclear stain and vital dyes respectively. A single-step selection with 2×10⁹ parasites in duplicate was set up at 2% parasitemia and 5% hematocrit using 3× EC₅₀ of 9 (0.78 μM). Wells were monitored daily by blood smears until the culture was cleared of live parasites. Drug-containing media was replaced daily until cultures cleared and then every other day thereafter. Cultures were passaged once a week by replacing a fourth of the culture with media and fresh RBCs, with monitoring by blood smears twice a week for up to 60 days to allow for parasite recrudescence.

WGS and analyses of drug-resistant P. falciparum clones

For whole-genome sequencing, genomic DNA was used to prepare libraries using the Illumina Nextera DNA Flex library kit with dual indices, as previously reported.⁸³ The samples were multiplexed and sequenced on an Illumina MiSeq to obtain 300 bp paired end reads at 19–44× depth of coverage across the samples. Sequence reads were aligned to the P. falciparum 3D7 genome (PlasmoDB version 36) using BWA (Burrow-Wheeler Alignment). PCR duplicates and unmapped reads were filtered out using Samtools and Picard. The reads were realigned around indels using GATK RealignerTargetCreator and base quality scores were recalibrated using GATK BaseRecalibrator. GATK HaplotypeCaller (version 4.1.6) was used to identify any SNVs in clones. These were filtered based on quality scores (variant quality as function of depth QD > 1.5, mapping quality > 40, min base quality score > 18), read depth (depth of read > 5) to obtain high quality SNPs that were annotated using snpEFF. Comparative SNP analysis between the resistant clones and parent were performed to generate a final list of SNPs that were present exclusively in the resistant clones. BIC-Seq was used to discover CNVs in the resistant mutants against the parent, using the Bayesian statistical model.⁸⁴ Integrated genome viewer was used to visually verify the presence of these SNPs and CNVs in the clones.

ADME: Solubility measurements

Methods used to assess kinetic solubility in phosphate buffered saline (PBS) pH 6.5, and pH 2.0 buffers have been previously described.^{25, 85} Solubility of 3, 9 and 10 in the buffer used for the in vitro heme biomineralization studies was conducted by spiking a 40 μL aliquot of test compound stock solution into 80 μL of NP-40 solution followed by 280 μL Milli-Q water. The resulting mixtures contained a final compound concentration of 2 mM in 10% (v/v) DMSO and 61.1 mM NP-40. Samples were maintained in a 37°C incubator for the duration of the study. After 1, 2, 4 and 24 h incubations at 37°C, the bulk samples were centrifuged (3 min at 10,000 rpm) and a single aliquot of the supernatant taken and diluted 20-fold in 50% aqueous acetonitrile for quantitative analysis by HPLC with detection by UV/Vis absorbance at 244 nm.

ADME: In vitro metabolism and calculation of physical chemical properties

Compounds (1 μM) were incubated (up to 60 min at 37°C) with human liver microsomes (Xenotech LLC, Kansas City, KS) at a protein concentration 0.4 mg/mL with an NADPH regenerating buffer system as described previously.⁸⁶ Degradation half-life and in vitro intrinsic clearance were calculated from the apparent first-order degradation rate constant. Physical chemical properties were calculated using the ChemAxon chemistry cartridge via JChem for Excel software (version16.4.11).

Chemical methods

General.

All tested compounds have purity of >95% as judged by HPLC analysis (UV detection at 210 nM). Chemical shifts δ are in ppm and spectra were referenced using the residual solvent peak. The following abbreviations are used: singlet (s), doublet (d), triplet (t), quartet (q), double doublet (dd), quintet (quin), multiplet (m), broad signal (bs). Mass spectra (m/z) were recorded on an Agilent LC-MS 1290 Infinity using ESI ionization. All chemicals were used as received unless otherwise noted.

HPLC/MS Analysis

An Agilent 1290 Infinity HPLC system using an Eclipse XDB-C18 column (4.6 × 150 mm, 5 μm; Agilent) that was coupled to an Agilent 6130 quadrupole ESI mass spectrometer run in the positive mode with a scan range of 100 to 1100 m/z was used. Liquid chromatography was carried out at a flow rate of 0.5 ml/min at 20 °C with a 5 μL injection volume, using a gradient elution with aqueous acetonitrile containing 0.1% formic acid, 0 min: 30% acetonitrile/water; 0 → 6 min: gradient to 95% acetonitrile/water; 6 → 12 min 95% acetonitrile/water.

N-(3-Chlorophenyl)-2-(3-methyl-2-oxo-5-phenyl-2,3-dihydro-1H-benzo[e][1,4]diazepin-1-yl)butanamide (1).

t_R = 9.0 min. ESI-MS (m/z): 446.1 [M+H]⁺.

N-(4-((4-Acetylphenyl)carbamoyl)phenyl)-2,3-dihydrobenzo[b][1,4]dioxine-2-carboxamide (2).

t_R = 4.24 min. ESI-MS (m/z): 417.0 [M+H]⁺.

General syntheses of tetrazole derivatives37

Isocyanide (1.0 mmol), aromatic aldehyde (1.0 mmol), N-benzylamine (1.0 mmol), and trimethylsilyl azide (1.0 mmol) were dissolved in MeOH (2.0 mL, 2 M) under N₂ atmosphere. The resulting solution was stirred at rt for 2–3 days. MeOH was evaporated and the residue was purified by flash column chromatography.

N-(5-bromo-2-methoxybenzyl)-N-methyl-1-phenyl-1-(1-(o-tolyl)-1H-tetrazol-5-yl)methenamine (3).

The target compound was obtained in 73% yield as colourless thick liquid. ¹H NMR (400 MHz, CDCl₃) δ 7.48−7.42 (m, 2H), 7.33−7.19 (m, 8H), 7.02 (d, J = 8.4Hz, 1H), 6.64 (d, J = 8.6Hz, 1H), 4.94 (s, 1H), 3.69−3.61 (m, 4H), 3.46 (d, J = 14.8Hz, 1H), 2.23 (s, 3H), 1.69 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 156.7, 155.2, 136.0, 135.5, 132.6, 132.1, 131.6, 131.3, 130.6, 129.2, 128.6(d J = 20.8Hz), 127.3, 127.0, 113.0, 112.0, 63.3, 55.5, 51.7, 39.6, 17.1. ESI-MS (m/z): 478.1 [M+H]⁺. The enantiomers were separated by chiral HPLC (Chiracel OD-H column, 1% isopropanol in hexane, 1.0 mL/min, t_R= 25, 32 min). Enantiomer 1, [α]_D²⁵ +7.99 (c 0.15, EtOH); Enantiomer 2, [α]_D²⁵ −7.99 (c 0.15, EtOH).

N-(2-methoxybenzyl)-N-methyl-1-phenyl-1-(1-(o-tolyl)-1H-tetrazol-5-yl)methanamine (4).

The target compound was obtained in 78% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 7.45 (t, J = 7.6Hz, 1H), 7.38−7.25 (m, 8H), 7.20 (t, J = 7.8Hz, 1H), 7.01 (d, J = 7.9Hz, 1H), 6.90 (t, J = 7.4Hz, 1H), 6.80 (d, J = 8.2Hz, 1H), 4.94 (s, 1H), 3.74−3.66 (m, 4H), 3.50 (d, J = 14.1 Hz, 1H), 2.26 (s, 3H), 1.73 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 157.6, 155.3, 135.9, 135.7, 132.6, 131.5, 131.1, 129.9, 129.2, 128.4(d, J = 32.0Hz), 128.1, 127.2, 126.9, 126.5, 120.4, 110.3, 62.9, 55.2, 52.2, 39.2, 17.1. ESI-MS (m/z): 400.2 [M+H]⁺.

N-(2-methoxy-5-(trifluoromethyl)benzyl)-N-methyl-1-phenyl-1-(1-(o-tolyl)-1H-tetrazol-5-yl)methanamine (5).

The target compound was obtained in 70% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 7.59 (s, 1H), 7.48−7.39 (m, 2H), 7.20−7.31 (m, 7H), 7.00 (d, J = 7.7Hz, 1H), 6.82 (d, J = 8.5Hz, 1H), 4.98 (s, 1H), 3.78−3.70 (m, 4H), 3.52 (d, J = 14.8 Hz, 1H), 2.24 (s, 3H), 1.69 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 159.9, 155.0, 135.9, 135.5, 132.5, 131.6, 131.2, 129.1, 128.6 (d, J = 16.0Hz), 127.6, 127.1, 126.9, 126.4 (q, J = 15.0Hz), 125.4 (q, J = 16.0Hz), 122.7 (q, J = 131.9Hz), 109.9, 63.2, 55.5, 51.6, 39.6, 17.1. ESI-MS (m/z): 468.2[M+H]⁺.

N-(2-methoxy-5-(trifluoromethyl)benzyl)-N-methyl-1-(pyridin-3-yl)-1-(1-(o-tolyl)-1H-tetrazol-5-yl)methenamine (6).

The target compound was obtained in 63% yield as a white solid. ¹H NMR (400 MHz, CDCl₃) δ 8.56 (dd, J = 4.8, 1.6Hz, 1H), 8.49 (d, J = 2.4Hz, 1H ), 7.91 (td, J = 8.0, 2.0Hz, 1H), 7.52−7.44 (m, 3H), 7.38−7.27 (m, 3H), 7.06 (d, J = 8.0Hz, 1H), 6.85 (d, J = 8.0Hz, 1H), 5.03 (s, 1H), 3.73 (s, 3H), 3.71 (d, J = 14.8Hz, 1H), 3.51 (d, J=14.8Hz, 1H), 2.24 (s, 3H), 1.80 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 159.9, 154.0, 150.3, 149.8, 137.0, 135.6, 132.3, 131.9, 131.5, 131.1, 127.2, 126.9 (d, J = 8.68Hz), 126.6 (q, J = 14.4Hz), 125.8 (q, J = 16.0Hz), 123.5, 122.6 (q, J = 134.0Hz), 110.2, 60.5, 55.5, 51.7, 39.1, 17.2. ESI-MS (m/z): 469.3 [M+H]⁺. The enantiomers were seperated by chiral HPLC (Chiracel OD-H column, 10% isopropanol in hexane, 1.0 mL/min, t_R= 15, 17 min). Enantiomer 1, [α]_D²⁵ +5.16 (c 0.15, EtOH); Enantiomer 2, [α]_D²⁵ −5.16 (c 0.15, EtOH).

N-(2-methoxy-5-(trifluoromethyl)benzyl)-N-(pyridin-3-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)cyclopropanamine (7).

The target compound was obtained in 61% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.55 (m, 2H), 8.08 (d, J = 8Hz, 1H), 7.48−7.20 (m, 6H), 7.00 (d, J = 8.0Hz, 1H), 6.76 (d, J = 8.4Hz, 1H), 5.12 (s, 1H), 4.20 (d, J = 12Hz, 1H), 3.69 (d, J = 13.2Hz, 1H), 3.64 (s, 3H), 1.89 (s, 4H), 0.24−0.15 (m, 2H), −0.05− −0.11 (m, 1H), −0.29− −0.39 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 159.9, 153.8, 150.9, 149.8, 137.4, 135.5, 132.3, 131.7, 131.4, 130.2, 129.1 (q J = 14.6Hz), 127.6, 127.1 (d, J = 7.0Hz), 125.7 (q J = 15.0Hz), 123.2, 122.2 (q, J = 130Hz), 109.9, 58.5, 55.4, 49.5, 35.1, 17.2, 8.1, 7.5. ESI-MS (m/z): 495.1 [M+H]⁺.

N-(2-fluoro-5-(trifluoromethyl)benzyl)-N-(pyridin-3-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)cyclopropanamine (8).

The target compound was obtained in 61% yield as a colorless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.57 (td, J = 4.8, 1.8 Hz, 1H), 8.52 (bs, 1H), 8.01−7.92 (m, 1H), 7.52−7.22 (m, 6H), 7.00 (t, J = 8.8 Hz, 2H), 5.19 (s, 1H), 4.27 (d, J = 13.8 Hz, 1H), 3.87 (d, J = 13.8 Hz, 1H), 1.91 (bs, 4H), 0.35−0.16 (m, 2H), 0.00− −0.07 (m, 1H), −0.16− −0.27 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 153.5, 150.7, 150.0, 137.2, 135.5, 132.2, 131.9, 131.5, 130.3, 129.2−129.0 (m), 127.2, 127.0, 123.3, 115.9, 115.7, 58.7, 48.4, 35.2, 17.3, 8.1, 7.7. ESI-MS (m/z): 483.1 [M+H]⁺.

N-(Pyridin-3-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-N-(3-(trifluoromethyl)benzyl) cyclopropanamine (9).

The target compound was obtained in 62% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.54 (d, J = 4.4Hz, 1H), 8.45 (bs, 1H), 7.83 (d, J = 7.2Hz, 1H), 7.47−7.37 (m, 2H), 7.35−7.18 (m, 6H), 6.88 (d, J = 7.2Hz, 1H), 5.16 (s, 1H), 4.10 (d, J = 14.4Hz, 1H), 3.93 (d, J = 14.0Hz, 1H), 2.05−1.98 (m, 1H), 1.85 (s, 3H), 0.31−0.18 (m, 2H), 0.06− −0.01 (m, 1H), −0.14− −0.21 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 153.7, 150.7, 149.9, 140.3, 137.2, 135.5, 132.2 (d, J = 20.7Hz), 131.9, 131.5, 130.9, 130.2 (q, J = 135.0Hz) 128.7, 127.3, 126.9, 125.6 (q, J = 15.1Hz), 123.9 (q, J = 14.8Hz), 123.4, 58.4, 55.6, 35.1, 17.4, 8.9, 7.5. ESI-MS (m/z): 465.1.1 [M+H]⁺.

N-(Pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-N-(3-(trifluoromethyl) benzyl)cyclopropanamine (10).

The target compound was obtained in 53% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 9.17 (s, 1H), 8.80 (s, 2H), 7.52−7.43 (m, 2H), 7.40−7.27 (m, 4H), 7.21 (d, J = 7.6Hz, 1H), 6.93 (d, J = 7.2 Hz, 1H), 5.16 (s, 1H), 4.07 (d, J = 14.0 Hz, 1H), 3.91 (d, J = 13.6 Hz, 1H), 2.09−2.01 (m, 1H), 1.90 (s, 3H), 0.39−0.24 (m, 2H), 0.09−0.01 (m, 1H), −0.12− −0.20 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 158.7, 157.7, 152.5, 139.3, 135.3, 132.2, 132.1, 132.0, 131.8, 130.8 (q, J = 130.0 Hz), 128.9, 128.7, 127.4, 126.8, 125.6 (q, J = 15.2 Hz), 124.3 (q, J = 17.6Hz), 56.6, 55.8, 35.0, 17.4, 8.9, 7.7. ESI-MS (m/z): 466.1 [M+H]⁺.

N-(3-Fluoro-5-(trifluoromethyl)benzyl)-N-(pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl) methyl)cyclopropanamine (11).

The target compound was obtained in 51% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 9.18 (s, 1H), 8.81 (s, 2H), 7.50 (t, J = 7.2Hz, 1H), 7.40 (d, J = 7.6Hz, 1H), 7.31 (t, J = 7.6Hz, 1H), 7.15 (d, J = 8.0Hz, 1H), 7.07 (s, 1H), 6.99−6.91 (m, 2H), 5.17 (s, 1H), 4.09 (d, J = 14.4Hz, 1H), 3.91 (d, J = 14.4Hz, 1H), 2.07−1.99 (m, 1H), 1.92 (s, 3H), 0.41−0.26 (m, 2H), 0.12−0.04 (m, 1H), −0.05− −0.15 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 163.6, 161.2, 158.8, 157.7, 152.4, 142.6 (d, J = 25.2Hz), 135.3, 132.2, 131.9, 131.8, 128.6, 127.5, 126.7, 121.3−121.1 (m), 119.2, 118.9, 112.0 (q, J = 18.7Hz), 111.8 (q, J = 15.2Hz), 56.7, 55.4, 35.3, 17.5, 8.9, 7.7. ESI-MS (m/z): 484.0 [M+H]⁺. The enantiomers were separated by chiral HPLC (Chiracel OD-H column, 10% isopropanol in hexane, 1.0 mL/min, t_R= 10, 12 min). Enantiomer 1, [α]_D²⁵ +24.609 (c 0.195, EtOH); Enantiomer 2, [α]_D²⁵ −21.558 (c 0.195, EtOH).

N-(4-Fluoro-3-(trifluoromethyl)benzyl)-N-(pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl) methyl)cyclopropanamine (12).

The target compound was obtained in 51% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 9.18 (s, 1H), 8.80 (s, 2H), 7.51 (dt, J = 8.0, 1.6Hz, 1H), 7.40 (d, J = 7.6Hz, 1H), 7.31 (t, J = 7.6Hz, 1H), 7.24−7.18 (m, 2H), 7.03 (t, J = 10Hz, 1H), 6.97 (d, J = 8.0Hz, 1H), 5.16 (s, 1H), 4.06 (d, J = 14.0Hz, 1H), 3.85 (d, J = 14.0Hz, 1H), 2.00−1.88 (m, 4H), 0.39−0.23 (m, 2H), 0.06−0.01 (m, 1H), −0.11− −0.19 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 158.8, 157.7, 152.4, 135.3, 134.7, 134.6, 134.2, 134.1, 132.1, 131.9, 131.8, 128.7, 127.5, 126.8, 117.0, 116.8, 56.9, 54.9, 35.1, 17.4, 8.8, 7.8. ESI-MS (m/z): 484.1 [M+H]⁺.

N-(2-Fluoro-3-(trifluoromethyl)benzyl)-N-(pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl) methyl)cyclopropanamine (13).

The target compound was obtained in 50% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 9.20 (s, 1H), 8.88 (s, 2H), 7.52−7.44 (m, 2H), 7.39 (d, J = 7.6Hz, 1H), 7.34−7.27 (m, 2H), 7.11−7.02 (m, 2H), 5.15 (s, 1H), 4.21 (d, J = 13.2Hz, 1H), 3.79 (d, J = 13.6Hz, 1H), 1.97−1.86 (m, 4H), 0.36−0.19 (m, 2H), −0.02− −0.09 (m, 1H), −0.20− −0.27 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 158.8, 157.7, 152.5, 135.6 (d, J = 11.5Hz), 135.4, 132.0 (d, J = 10.1Hz), 134.6, 131.7, 128.4, 127.4, 126.9, 126.5 (q. J = 11.6Hz), 123.7 (d, J = 11.6Hz), 57.0, 48.5, 35.0, 17.4, 8.2, 7.9. ESI-MS (m/z): 484.1 [M+H]⁺.

N-(Cyclopropyl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-N-(3- (trifluoromethyl) benzyl) cyclo propanamine (14).

The target compound was obtained in 71% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 7.50−7.42 (m, 2H), 7.40−7.36 (m, 2H), 7.34−7.27 (m, 3H), 7.08 (d, J = 6.8Hz, 1H), 4.17 (d, J = 14.4Hz, 1H), 4.07 (d, J = 14.4Hz, 1H), 3.18 (d, J = 10.0Hz, 1H), 2.13−2.00 (m, 4H), 1.65−1.56 (m, 1H), 0.72−0.64 (m, 1H), 0.61−0.53 (m, 1H), 0.41−0.31 (m, 1H), 0.22−0.09 (m, 2H), −0.03− −0.12 (m, 1H), −0.21− −0.37 (m, 2H). ¹³C NMR (101 MHz, CDCl₃) δ 155.9, 141.0, 135.8, 132.8, 132.3, 131.7, 131.2, 130.3 (q, J = 127.1Hz), 128.5, 127.2, 127.0, 125.8 (q, J = 14.6Hz), 123.7 (q, J = 14.9Hz), 62.0, 55.3, 35.4, 17.5, 11.4, 8.5, 6.8, 5.6, 3.9. ESI-MS (m/z): 428.1 [M+H]⁺.

N-(Pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-N-(3-(trifluoromethyl) benzyl)acetamide (15).

1-(pyrimidin-5-yl)-1-(1-(o-tolyl)-1H-tetrazol-5-yl)-N-(3-(trifluoromethyl) benzyl) methanamine was synthesized using (3-(trifluoromethyl)phenyl)methanamine via general procedure. To a suspension of the above amine (1.0 mmol) in CH₂Cl₂ was added Et₃N (2.0 mmol) at 0 °C. Acetyl chloride (2.0 mmol) was then added dropwise over 5 min. After stirring at RT for 1 h, the mixture was diluted with CH₂Cl₂ and washed with water. The organic layer was dried over anhydrous Na₂SO₄, and the solvent was evaporated under vacuum. The residue obtained was purified by flash column chromatography (EtOAc/hexane) to furnish the target compound in 51% yield. ¹H NMR (400 MHz, CDCl₃) δ 8.97 (bs, 1H), 8.55 (bs, 2H), 7.55−7.27 (m, 5H), 7.20−6.81 (m, 3H), 5.02 (s, 2H), 2.14 (s, 3H), 1.95 (s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 172.2, 159.0, 157.9, 153.2, 137.2, 135.7, 132.2, 131.9, 131.6, 129.7, 129.3−129.2 (m), 127.5, 127.0, 124.6 (q, J = 12.5Hz), 122.5 (q, J = 16.3Hz), 49.5, 47.5, 22.2, 17.4. ESI-MS (m/z): 468.0 [M+H]⁺.

N-(2-Fluoro-5-(trifluoromethyl)benzyl)-N-methyl-1-(pyridin-3-yl)-1-(1-(o-tolyl)-1H-tetrazol-5-yl)methenamine (16).

The target compound was obtained in 61% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.57 (dt, J = 4.8, 1.3Hz, 1H), 8.45 (d, J = 2.3Hz, 1H), 7.89 (dt, J = 8.1, 2.0Hz, 1H), 7.55 (dd, J = 6.6, 2.4 Hz, 1H), 7.53−7.43 (m, 2H), 7.38−7.27 (m, 3H), 7.10−7.00 (m, 2H), 4.98 (s, 1H), 3.74 (d, J = 14.0 Hz, 1H), 3.58 (d, J = 14.1 Hz, 1H), 2.22 (s, 3H), 1.81(s, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 164.2, 161.6, 153.9, 150.3 (d, J = 22.7Hz), 136.9, 135.6, 132.2, 131.9, 131.6, 130.7, 128.1−127.9 (m), 127.3, 126.9, 126.5−126.3 (m), 126.1, 125.2, 125.1, 123.6, 122.4, 116.2, 115.9, 60.5, 50.5 (d, J = 8.8Hz), 39.0, 17.2. ESI-MS (m/z): 457.1 [M+H]⁺.

1-(1-Cyclohexyl-1H-tetrazol-5-yl)-N-(2-fluoro-5-(trifluoromethyl)benzyl)-N-methyl-1-(pyridin-3-yl)methanamine (17).

The target compound was obtained in 62% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.69 (d, J = 2.4Hz, 1H), 8.62 (dd, J = 4.8, 1.2Hz, 1H), 7.94 (td, J = 8.4, 4.0, 1.6Hz, 1H), 7.62 (dd, J = 6.4, 2.0Hz, 1H), 7.57−7.52 (m, 1H), 7.39−7.35 (m, 1H), 7.15 (t, J = 18, 9.6Hz, 1H), 5.26 (s, 1H), 4.44−4.35 (m, 1H), 3.70 (t, J = 16.0Hz, 2H), 2.23 (s, 3H), 2.07−1.70 (m, 7H), 1.42−1.23 (m, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 164.4, 161.9, 152.1, 150.3, 137.2, 130.1, 128.2−128.0 (m), 126.9−126.7 (m) 126.2, 126.1, 125.1, 123.7, 122.4, 116.5, 116.2, 61.2, 58.4, 51.4 (d, J = 8.0Hz), 38.8, 33.3, 32.9, 25.3 (d, J = 22.1Hz), 24.8. ESI-MS (m/z): 449.1 [M+H]⁺.

N-((1-Cyclopropyl-1H-tetrazol-5-yl)(pyridin-3-yl)methyl)-N-(3-(trifluoromethyl) benzyl)cyclopropanamine (18).

The target compound was obtained in 59% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.44 (d, J = 2.4Hz, 1H), 8.41 (dd, J = 5.2, 1.6Hz, 1H), 7.73 (td, J = 7.6, 2.0Hz, 1H), 7.30 (d, J = 7.6Hz, 1H), 7.23−7.12 (m, 3.5H), 7.07 (s, 0.5H), 5.34 (s, 1H), 3.86 (d, J = 14.0Hz, 1H), 3.73 (d, J = 14.0Hz, 1H), 3.12−3.05 (m, 1H), 2.04−1.97 (m, 1H), 1.15−1.05 (m, 1H), 1.00−0.76 (m, 3H), 0.31−0.20 (m, 2H), 0.04− −0.09 (m, 2H). ¹³C NMR (101 MHz, CDCl₃) δ 154.3, 150.3, 149.4, 139.9, 138.1, 132.4, 130.7, 128.9, 125.7 (q, J = 14.6Hz), 124.2 (q, J = 15.2Hz), 123.6, 58.1, 56.1, 34.7, 28.5, 8.8, 7.7, 7.3, 6.9. ESI-MS (m/z): 415.1 [M+H]⁺.

N-cyclopropyl-N-((1-cyclopropyl-1H-tetrazol-5-yl)(2-methylpyridin-3-yl)methyl)-1-(3-(trifluoromethyl)phenyl)cyclopropan-1-amine (19).

Step 1: Synthesis of 1-(3-(trifluoromethyl)phenyl)cyclopropan-1-amine.

Ethylmagnesium bromide (0.7 mL, 2.2 mmol, 3 M in ether) was added to a solution of 3-trifluoromethyl benzonitrile (171 mg, 1.0 mmol) and Ti(Oi-Pr)₄ (312 mg, 1.1 mmol) in Et₂O (5 mL) at −70 °C. The yellow solution was stirred for 10 min. After the solution was warmed to rt (1 h), BF₃‚OEt₂ (292 mg, 2.2 mmol) was added. After the mixture was stirred for 1 h, 1N HCl (3 mL) and ether (15 mL) were added. NaOH (10% aq, 10 mL) was added to the resulting two clear phases, and the mixture was extracted with ether. The combined ether layers were dried over Na₂SO₄ and concentrated under reduced pressure. The residue was purified by flash column chromatography to furnish the target compound in 63% yield as a colorless oil.

Step 2: Synthesis of N-cyclopropyl-1-(3-(trifluoromethyl)phenyl)cyclopropan-1-amine.

This was synthesized via general procedure used for synthesis of N-benzylamine.

Step 3.

The target compound was obtained in 52% yield as a colourless liquid using the general procedure for synthesis of tetrazoles. ¹H NMR (400 MHz, CDCl₃) δ 8.39 (dd, J = 4.8, 1.2Hz, 1H), 7.47−7.28 (m, 5H), 7.02 (dd, J = 8.0, 4.8Hz, 1H), 5.92 (s, 1H), 3.18−3.11(m, 1H), 2.80−2.73 (m, 1H), 2.47 (s, 3H), 1.49−1.41 (m, 1H), 1.37−1.27 (m, 1H), 1.25−1.16 (m, 2H), 1.06−0.85 (m, 4H), 0.68−0.50 (m, 2H), 0.32−0.23 (m, 1H), −0.24− −0.33 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 156.4, 156.0, 148.0, 142.1, 139.0, 132.7, 130.7, 130.2 (q, J = 130Hz), 128.7, 125.4 (q, J = 14.8Hz), 124.2 (q, J = 15.6Hz), 121.3, 56.6, 48.8, 30.5, 28.1, 21.7, 16.7, 13.0, 10.5, 6.8, 6.7, 6.6. ESI-MS (m/z): 455.1 [M+H]⁺.

N-Propyl-N-(pyrimidin-5-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-1-(3-(trifluoromethyl) phenyl)cyclopropan-1-amine (20).

The target compound was obtained in 49% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 9.09 (s, 1H), 8.48 (s, 2H), 7.49 (td, J = 7.6, 1.3Hz, 1H), 7.39 (dd, J = 10.2, 7.8Hz, 2H), 7.33−7.24 (m, 3H), 7.16 (s, 1H), 6.87 (d, J = 7.9Hz, 1H), 5.18 (s, 1H), 2.88 (m, 2H), 1.90 (s, 3H), 1.52−1.34 (m, 2H), 1.18−1.08 (m, 1H), 0.95−0.86 (m, 1H), 0.85−0.78 (m, 1H), 0.71 (t, J = 7.3Hz, 3H), 0.55−0.48 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 158.4, 157.2, 154.3, 143.6, 135.1, 132.2, 131.9, 131.8, 131.7, 131.0, 129.0, 127.4, 126.6, 124.3 (q, J = 13.1Hz), 124.0 (q, J = 14.4Hz), 55.4, 48.3, 45.8, 23.6, 17.6, 17.2, 11.4. ESI-MS (m/z): 494.2 [M+H]⁺.

N-Propyl-N-(pyridin-3-yl(1-(o-tolyl)-1H-tetrazol-5-yl)methyl)-1-(3-(trifluoromethyl) phenyl)cyclopropan-1-amine (21).

The target compound was obtained in 56% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃) δ 8.48 (dd, J = 4.8, 1.6Hz, 1H), 8.21 (d, J = 2.4Hz, 1H), 7.52−7.36 (m, 3H), 7.34−7.26 (m, 3H), 7.24−7.14 (m, 3H), 6.77 (d, J = 7.2Hz, 1H), 5.14 (s, 1H), 2.94−2.88 (m, 2H), 1.83 (s, 3H), 1.50−1.27 (m, 2H), 1.20−1.12 (m, 1H), 0.95−0.87 (m, 1H), 0.84−0.77 (m, 1H), 0.70−0.60 (m, 4H). ¹³C NMR (101 MHz, CDCl₃) δ 155.4, 150.4, 149.6, 144.4, 136.5, 135.2, 133.1, 132.0, 131.4, 130.4 (q, J = 128Hz), 128.8, 127.2, 126.5, 124.1 (q, J = 14.8Hz), 123.6 (q, J = 15.2Hz), 123.2, 57.1, 48.3, 45.7, 23.6, 17.5, 17.0, 11.3. ESI-MS (m/z): 493.1 [M+H]⁺.

General Synthesis of amide derivatives³⁸

Isocyanide (1 mmol), aromatic aldehyde (1 mmol), N-benzylamine (1 mmol), and phosphinic acid (0.1 mmol) were dissolved in anhydrous toluene (1 mL) under N₂ atmosphere. The resulting solution was heated at 80 °C for 12h. Toluene was evaporated under reduced pressure and the residue was purified by flash chromatography.

To a solution of above amide (0.5 mmol) in dry THF was added NaH (0.6 mmol) at 0 °C in one portion, then after 5 min, MeI (1.0 mmol) was added to the reaction mixture at the same temperature. The resulting mixture was allowed to warm at RT and stirred for 30 min. After completion of reaction monitored by LCMS, the reaction mixture was quenched with cold H₂O, extracted with ethyl acetate, and dried over anhydrous Na₂SO₄. The organic layer was then concentrated under reduced pressure and purified by flash chromatography.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-methyl-2-(pyridin-3-yl)-N-(o-tolyl)acetamide (22).

The target compound was obtained in 61% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.49−8.44 (m, 2H, mixture of rotamers), 8.26 (d, J = 2.3 Hz, 1H, one rotamer), 8.15 (d, J = 2.4Hz, 1H, one rotamer), 7.63 (dt, J = 7.9, 2.0Hz, 1H, one rotamer), 7.58 (dt, J = 7.9, 2.0Hz, 1H, one rotamer), 7.45−7.40 (m, 1H, one rotamer), 7.39−7.31 (m, 4H, mixture of rotamers), 7.25−7.13 (m, 8H, mixture of rotamers), 7.09−7.01 (m, 2H, mixture of rotamers), 6.85 (m, 2H, mixture of rotamers), 6.29 (dd, J = 7.9, 1.3 Hz, 1H, one rotamer), 4.54 (s, 1H, one rotamer), 4.39 (s, 1H, one rotamer), 4.30 (d, J = 14.9 Hz, 1H, one rotamer), 4.24 (d, J = 14.6 Hz, 1H, one rotamer), 3.97 (d, J = 14.7Hz, 1H, one rotamer), 3.95 (d, J = 14.7Hz, 1H, one rotamer), 3.23 (s, 3H, one rotamer), 3.19 (s, 3H, one rotamer), 2.36−2.28 (m, 2H, mixture of rotamers), 2.27 (s, 3H, one rotamer), 1.50 (s, 3H, one rotamer), 0.36−0.18 (m, 4H, mixture of rotamers), 0.07− −0.01 (m, 2H, mixture of rotamers), −0.13− −0.26(m, 2H, mixture of rotamers). ¹³C NMR (101 MHz, CDCl_3, 1:1 mixture of rotamers) δ 171.5, 171.3, 151.2, 151.0, 149.2, 149.1, 142.5, 141.9, 141.2, 141.1, 137.5, 137.1, 135.8, 135.1, 132.7, 132.0, 131.7, 131.6, 131.5, 131.4, 128.7 (d, J = 13.6Hz), 128.5 (d, J = 12.0Hz), 128.2, 128.1, 127.2, 126.9, 125.3 (q, J = 14.4Hz), 125.1 (q, J = 16.0Hz), 123.5 (q, J = 16.0Hz), 123.2 (q, J = 14.4Hz), 123.1, 123.0, 65.0, 63.5, 56.2, 55.5, 36.2, 35.7, 35.0, 17.5, 16.8, 8.9, 8.7, 7.6, 7.3. ESI-MS (m/z): 454.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-methyl-2-(pyrimidin-5-yl)-N-(o-tolyl)acetamide (23).

The target compound was obtained in 45% yield as a colourless liquid. ¹H NMR (400 MHz, MeOD, 1:1 mixture of rotamers) δ 9.04−9.00 (m, 2H, mixture of rotamers), 8.52−8.46 (m, 4H, mixture of rotamers), 7.53−7.22 (m, 11H, mixture of rotamers), 7.18−7.11 (m, 2H, mixture of rotamers), 7.02−6.91 (m, 2H, mixture of rotamers), 6.45 (d, J = 7.8Hz, 1H, one rotamer), 4.57 (s, 1H, one rotamer), 4.50 (s, 1H, one rotamer), 4.24 (d, J = 14.5Hz, 1H, one rotamer), 4.18 (d, J = 14.5Hz, 1H, one rotamer), 4.05−3.98 (m, 2H, mixture of rotamers), 3.25 (s, 3H, one rotamer), 3.21 (s, 3H, one rotamer), 2.38−2.25 (m, 5H, mixture of rotamers), 1.61 (s, 3H, one rotamer), 0.41−0.26 (m, 4H, mixture of rotamers), 0.16−0.05 (m, 2H, mixture of rotamers), −0.07− − 0.15 (m, 1H, one rotamer), −0.16− −0.24 (m, 1H, one rotamer). ¹³C NMR (101 MHz, MeOD, 1:1 mixture of rotamers) δ 171.9, 171.6, 159.1, 158.9, 158.6, 158.5, 143.2, 142.6, 142.1, 136.5 (d, J = 22.0Hz), 133.5, 133.2, 132.9, 132.8, 132.3, 131.5, 130.3 (d, J = 21.6Hz), 130.1, 129.8, 129.5, 129.4, 128.6, 128.0, 126.2 (q, J = 15.2Hz), 125.9 (q, J = 15.2Hz), 124.8 (q, J = 16.4Hz), 124.5 (q, J = 16.4Hz), 64.4, 63.2, 57.4, 56.9, 36.9, 36.8, 36.7, 36.0, 17.7, 17.0, 9.6, 9.3, 8.5, 8.3. ESI-MS (m/z): 455.2 [M+H]⁺.

2-(Cyclopropyl(3-((trifluoromethyl)thio)benzyl)amino)-N-methyl-2-(pyridin-3-yl)-N-(o-tolyl)acetamide (24).

The target compound was obtained in 59% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.50−8.45 (m, 2H, mixture of rotamers), 8.27 (d, J = 2.3Hz, 1H, one rotamer), 8.17 (d, J = 2.4Hz, 1H, one rotamer), 7.64 (dt, J = 7.9, 2.0Hz, 1H, one rotamer), 7.57 (dt, J = 7.9, 2.0Hz, 1H, one rotamer), 7.50−7.30 (m, 4H, mixture of rotamers), 7.25−7.22 (m, 4H, mixture of rotamers), 7.20−7.11 (m, 5H, mixture of rotamers), 7.10−7.00 (m, 2H, mixture of rotamers), 6.93−6.80 (m, 2H, mixture of rotamers), 6.29 (d, J = 7.8Hz, 1H, one rotamer), 4.55 (s, 1H, one rotamer), 4.39 (s, 1H, one rotamer), 4.27 (d, J = 14.9Hz, 1H, one rotamer), 4.22 (d, J = 14.5Hz, 1H, one rotamer), 3.93 (d, J = 14.7Hz, 1H, one rotamer), 3.91 (d, J = 14.7Hz, 1H, one rotamer), 3.24 (s, 3H, one rotamer), 3.20 (s, 3H, one rotamer), 2.36−2.29 (m, 2H, mixture of rotamers), 2.26 (s, 3H, one rotamer), 1.52 (s, 3H, one rotamer), 0.34−0.25 (m, 2H, one rotamer), 0.24−0.16 (m, 2H, one rotamer), 0.06− −0.02 (m, 2H, mixture of rotamer), −0.14− − 0.28 (m, 2H, mixture of rotamer). ¹³C NMR (151 MHz, CDCl₃, 1:1 mixture of rotamers) δ 171.3, 150.4, 150.2, 148.4, 141.1, 138.3, 137.9, 136.5, 136.3, 135.7, 135.1, 134.8, 134.5, 131.7, 131.5, 131.2, 131.0, 130.8, 129.2, 129.0, 128.8, 128.7, 128.4, 128.1, 127.2, 126.9, 123.9, 123.7, 123.3, 123.2, 64.9, 63.4, 56.0, 55.5, 36.3 (d, J = 26.4Hz), 35.6, 35.0, 17.5, 16.9, 8.8 (d, J = 55.0Hz), 7.6, 7.2. ESI-MS (m/z): 486.1 [M+H]⁺.

2-(Cyclopropyl(3-((trifluoromethyl)thio)benzyl)amino)-N-methyl-2-(pyrimidin-5-yl)-N-(o-tolyl)acetamide, (25).

The target compound was obtained in 46% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 9.11−9.08 (m, 2H, mixture of rotamers), 8.50 (s, 2H, one rotamer), 8.48 (s, 2H, one rotamer), 7.51 (dt, J = 7.5, 1.8Hz, 1H, one rotamer), 7.48−7.38 (m, 2H, mixture of rotamers), 7.36−7.27 (m, 3H, mixture of rotamers), 7.25−7.14 (m, 5H, mixture of rotamers), 7.09−7.00 (m, 2H, mixture of rotamers), 6.89 (td, J = 7.6, 1.7Hz, 1H, one rotamer), 6.82 (dd, J = 7.8, 1.4Hz, 1H, one rotamer), 6.33 (dd, J = 7.9, 1.3Hz, 1H, one rotamer), 4.49 (s, 1H, one rotamer), 4.38 (s, 1H, one rotamer), 4.29 (d, J = 14.8Hz, 1H, one rotamer), 4.22 (d, J = 14.3Hz, 1H, one rotamer), 3.92 (d, J = 14.6Hz, 1H, one rotamer), 3.90 (d, J = 14.6Hz, 1H, one rotamer), 3.24 (s, 3H, one rotamer), 3.21 (s, 3H, one rotamer), 2.39−2.28 (m, 2H, mixture of rotamers), 2.22 (s, 3H, one rotamer), 1.59 (s, 3H, one rotamer), 0.37−0.27 (m, 2H, one rotamer), 0.27−0.18 (m, 2H, one rotamer), 0.05− −0.03 (m, 2H, mixture of rotamers), 0.19− −0.26 (m, 1H, one rotamer), −0.32− −0.39 (m, 1H, one rotamer). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 170.3, 170.1, 158.2−157.8 (m), 157.7, 142.4, 141.8, 140.8, 136.4, 136.2, 135.3, 135.2, 135.0, 134.7, 131.7 (d, J = 12.0Hz), 131.2, 130.9, 130.6, 129.6, 129.4, 129.1 (d, J = 26.0 Hz), 129.0, 128.3, 128.1, 127.3, 127.0, 62.9, 61.6, 56.2, 55.7, 36.3 (d, J = 7.2 Hz), 35.3, 34.5, 17.4, 16.9, 9.1, 8.9, 7.8, 7.5. ESI-MS (m/z): 487.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethoxy)benzyl)amino)-N-methyl-2-(pyridin-3-yl)-N-(o-tolyl)acetamide (26).

The target compound was obtained in 56% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.50−8.45 (m, 2H, mixture of rotamers), 8.26 (bs, 1H, one rotamer), 8.15 (bs, 1H, one rotamer), 7.62 (dt, J = 7.9, 2.0Hz, 1H, one rotamer), 7.56 (dt, J = 7.8, 2.0Hz, 1H, one rotamer), 7.25−7.22 (m, 2H, mixture of rotamers), 7.22−7.11 (m, 5H, mixture of rotamers), 7.10−6.98 (m, 5H, mixture of rotamers), 6.98−6.81 (m, 5H, mixture of rotamers), 6.32−6.25 (m, 1H, one rotamer), 4.54 (s, 1H, one rotamer), 4.38 (s, 1H, one rotamer), 4.26 (d, J = 15.0Hz, 1H, one rotamer), 4.20 (d, J = 14.7Hz, 1H, one rotamer), 3.96 (d, J = 14.8Hz, 1H, one rotamer), 3.94 (d, J = 14.8Hz, 1H, one rotamer), 3.23 (s, 3H, one rotamer), 3.19 (s, 3H, one rotamer), 2.40−2.29 (m, 2H, mixture of rotamers), 2.25 (s, 3H, one rotamer), 1.49 (s, 3H, one rotamer), 0.36−0.17 (m, 4H, mixture of rotamers), 0.08− −0.00 (m, 2H, mixture of rotamers), −0.09− −0.27 (m, 2H, mixture of rotamers). ¹³C NMR (151 MHz, CDCl₃, 1:1 mixture of rotamers) δ 149.1, 148.9, 140.9, 135.5, 135.0, 131.5, 131.4, 129.3 (m), 129.1, 128.8, 128.3, 128.0 (m), 127.2, 126.6 (m), 123.3 (m), 120.7, 119.5, 118.9, 64.6, 63.1, 56.0, 55.4, 36.2, 36.1, 35.5, 34.8, 17.3, 16.7, 8.8, 8.6, 7.1. ESI-MS (m/z): 470.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethoxy)benzyl)amino)-N-methyl-2-(pyrimidin-5-yl)-N-(o-tolyl)acetamide, (27).

The target compound was obtained in 43% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 9.09 (bs, 2H, mixture of rotamers), 8.48 (s, 2H, one rotamer), 8.47 (s, 2H, one rotamer), 7.30−7.27 (m, 1H, one rotamer), 7.25−7.23 (m, 1H, one rotamer), 7.22−7.14 (m, 3H, mixture of rotamers), 7.12−6.95 (m, 7H, mixture of rotamers), 6.94−6.84 (m, 2H, mixture of rotamers), 6.83−6.76 (m, 1H, one rotamer), 6.32 (dd, J = 7.8, 1.3Hz, 1H, one rotamer), 4.48 (s, 1H, one rotamer), 4.37 (s, 1H, one rotamer), 4.27 (d, J = 15.0Hz, 1H, one rotamer), 4.20 (d, J = 14.6Hz, 1H, one rotamer), 3.96 (d, J = 14.8Hz, 1H, one rotamer), 3.94 (d, J = 14.8Hz, 1H, one rotamer), 3.24 (s, 3H, one rotamer), 3.20 (s, 3H, one rotamer), 2.45−2.39 (m, 1H, one rotamer), 2.37−2.32 (m, 1H, one rotamer), 2.22 (s, 3H, one rotamer), 1.57 (s, 3H, one rotamer), 0.41−0.19 (m, 4H, mixture of rotamers), 0.10− −0.01 (m, 2H, mixture of rotamers), −0.15− −0.19 (m, 1H, one rotamer), −0.28− −0.32 (m, 1H, one rotamer). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 170.4, 170.1, 158.0 (m), 157.7, 143.2, 142.6, 140.9, 135.3, 135.2, 131.7, 130.7, 129.6 (d), 129.3, 129.1, 129.0, 128.3, 128.1, 127.3, 127.0, 126.9, 126.6, 121.0, 120.7, 119.6, 119.2, 62.8, 61.5, 56.3, 55.7, 36.3, 35.4, 34.5, 17.4, 16.9, 9.3, 9.0, 7.8, 7.5. ESI-MS (m/z): 471.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-(2-ethylphenyl)-N-methyl-2-(pyridin-3-yl)acetamide (28).

The target compound was obtained in 53% yield as a colourless liquid. ¹H NMR (600 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.48 (d, J = 4.8Hz, 2H, mixture of rotamers ), 8.24 (s, 1H, one rotamer), 8.19 (s, 1H, one rotamer), 7.73 (s, 1H, one rotamer), 7.63 (d, J = 8.2Hz, 1H, one rotamer), 7.48−7.27 (m, 7H, mixture of rotamers), 7.27−7.19 (m, 6H, mixture of rotamers), 7.16−7.03 (m, 2H, mixture of rotamers), 6.94−6.81 (m, 2H, mixture of rotamers), 6.28 (d, J = 7.8Hz, 1H, one rotamer), 4.54 (s, 1H, one rotamer), 4.40 (s, 1H, one rotamer), 4.34 (d, J = 14.8Hz, 1H, one rotamer), 4.25 (d, J = 14.3Hz, 1H, one rotamer), 3.95 (s, 1H, one rotamer), 3.93 (s, 1H, one rotamer), 3.25 (s, 3H, one rotamer), 3.22 (s, 3H, one rotamer), 2.69−2.61 (m, 1H, one rotamer), 2.57−2.50 (m, 1H, one rotamer), 2.34−2.24 (m, 2H, mixture of rotamers), 2.05−1.95 (m, 1H, one rotamer), 1.25−1.21 (m, 4H), 0.82 (t, J = 7.6, 1.5Hz, 3H, one rotamer), 0.38−0.25 (m, 2H, one rotamer), 0.24−0.21 (m, 2H, one rotamer), 0.01 (bs, 2H, mixture of rotamers), −0.189− −0.24 (m, 2H, mixture of rotamers). ¹³C NMR (151 MHz, CDCl₃) δ 150.1, 150.01, 148.2, 141.4, 140.8, 140.5, 138.5, 138.1, 132.1, 131.7, 129.5, 129.3, 129.1 (d, J = 22.3Hz), 128.5(d,), 128.3, 128.0, 127.1, 126.7, 125.4, 125.2, 123.4 (d, J = 28.8Hz), 64.9, 63.3, 56.1, 55.6, 37.1, 37.0, 35.7, 35.1, 29.8, 23.3, 22.8, 14.4, 14.1, 8.7, 8.5, 7.7, 7.5. ESI-MS (m/z): 468.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-(2-ethylphenyl)-N-methyl-2-(pyrimidin-5-yl)acetamide (29).

The target compound was obtained in 42% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 9.08 (bs, 2H, mixture of rotamers), 8.48 (s, 2H, one rotamer), 8.45 (s, 2H, one rotamer), 7.48 (d, J = 7.3 Hz, 1H, one rotamer), 7.43−7.25 (m, 10H mixture of rotamers), 7.13 (d, J = 7.7 Hz, 1H, one rotamer), 7.05 (t, J = 7.6, 1H, one rotamer), 6.88 (t, J = 7.6Hz, 1H, one rotamer), 6.83 (d, J = 7.9Hz, 1H, one rotamer), 6.32 (d, J = 7.8Hz, 1H, one rotamer), 4.46 (s, 1H, one rotamer), 4.40−4.32 (m, 2H, mixture of rotamers), 4.25 (d, J = 14.4 Hz, 1H, one rotamer), 3.94 (d, J = 14.6Hz, 1H, one rotamer), 3.92 (d, J = 14.6Hz, 1H, one rotamer), 3.25 (s, 3H, one rotamer), 3.22 (s, 3H, one rotamer), 2.68−2.57 (m, 1H, one rotamer), 2.52−2.43 (m, 1H, one rotamer), 2.37−2.27 (m, 2H, mixture of rotamers), 2.09−1.99 (m, 1H, one rotamer), 1.36−1.15 (m, 4H, one rotamer), 0.88 (t, J = 7.6Hz, 3H, one rotamer), 0.36−0.20 (m, 4H, mixture of rotamers), 0.09 − −0.01 (m, 2H, mixture of rotamers), −0.20− −0.27 (m, 1H, one rotamer), −0.29 − −0.39 (m, 1H, one rotamer). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 170.4, 170.1, 158.1, 158.0 (d, J = 18.0Hz), 157.6, 141.7, 141.1 (d, J = 6.8Hz), 140.9, 140.3, 132.0, 131.7, 130.5, 129.7, 129.5 (d, J = 18.0Hz), 128.7, 128.5, 128.4, 128.0, 127.20, 126.9, 125.3 (q, J = 14.8Hz), 125.1 (q, J = 14.8Hz), 123.8 (q, J = 16.0Hz), 123.5 (q, J = 17.6Hz), 62.9, 61.5, 56.3, 55.8, 37.2, 37.1, 35.4, 34.7, 23.3, 22.9, 14.4, 14.3, 9.03, 8.7, 7.9, 7.7. ESI-MS (m/z): 469.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-methyl-N-phenyl-2-(pyrimidin-5-yl)acetamide (30).

The target compound was obtained in 45% yield as a colourless liquid. ¹H NMR (500 MHz, CDCl₃) δ 9.10 (s, 1H), 8.50 (s, 2H), 7.46 (d, J = 7.5 Hz, 1H), 7.38 (bs, 1H), 7.36−7.29 (m, 2H), 7.26−7.23 (m, 1H), 7.19 (t, J = 8.1Hz, 2H), 6.72 (s, 2H), 4.53 (s, 1H), 4.16 (d, J = 14.4 Hz, 1H), 3.97 (d, J = 14.5 Hz, 1H), 3.30 (s, 3H), 2.45−2.40 (m, 1H), 0.37−0.31 (m, 1H), 0.28−0.22 (m, 1H), 0.06−0.0 (m, 1H), −0.24− −0.30 (m, 1H). ¹³C NMR (101 MHz, CDCl₃) δ 169.8, 158.1, 157.7, 142.5, 141.1, 132.0, 130.6, 129.8, 128.7, 128.6, 127.2, 125.3 (q, J = 16.8Hz), 123.8 (q, J = 16.4Hz), 61.7, 56.5, 37.6, 34.6, 9.6, 7.2. ESI-MS (m/z): 441.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-N-methyl-2-(2-methyl pyrimidin-5-yl)-N-(o-tolyl)acetamide (31).

The target compound was obtained in 41% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.37 (s, 2H, one rotamer), 8.36 (s, 2H, one rotamer),7.48−7.44 (m, 1H, one rotamer), 7.41−7.31 (m, 4H, mixture of rotamers), 7.31−7.15 (m, 6H, mixture of rotamers), 7.11−6.99 (m, 2H, mixture of rotamers), 6.92 (td, J = 7.7, 1.4Hz, 1H, one rotamer), 6.84 (d, J = 8.0Hz, 1H, one rotamer), 6.37 (d, J = 7.8Hz, 1H, one rotamer), 4.44 (s, 1H, one rotamer), 4.35 (s, 1H, one rotamer), 4.29 (d, J = 14.8Hz, 1H, one rotamer), 4.23 (d, J = 14.5Hz, 1H, one rotamer), 3.96 (d, J = 14.6Hz, 1H, one rotamer), 3.93 (d, J = 14.6Hz, 1H, one rotamer), 3.23 (s, 3H, one rotamer), 3.20 (s, 3H, one rotamer), 2.70 (d, 6H, mixture of rotamers), 2.37−2.24 (m, 5H, mixture of rotamers), 1.60 (s, 3H, one rotamer), 0.40−0.19 (m, 4H, mixture of rotamers), 0.09− −0.04 (m, 2H, mixture of rotamers), −0.14− −0.21 (m, 1H, one rotamer), −0.24− −0.31 (m, 1H, one rotamer). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 170.6, 170.3, 167.6 (d, J = 14.8Hz), 158.1, 157.8, 141.9, 141.3, 141.0 (d, J = 14.4Hz), 135.4, 135.2, 132.0, 131.7 (d, J = 27.2Hz), 129.0, 128.9, 128.7, 128.4 (d, J = 28.4Hz), 128.1, 127.3, 127.1 (d, J = 19.2Hz), 126.1, 125.3 (q, J = 13.6Hz), 125.1 (q, J = 15.2Hz), 123.7 (q, J = 15.2Hz), 123.4 (d, J = 16.4Hz), 62.7, 61.5, 56.2, 55.7, 36.3 (d, J = 13.2Hz), 35.5, 34.8, 25.8, 17.4, 17.0, 8.9, 8.7, 7.9, 7.6. ESI-MS (m/z): 469.1 [M+H]⁺.

2-(Cyclopropyl(2-fluoro-5-(trifluoromethyl)benzyl)amino)-N-methyl-2-(pyridin-3-yl)-N-(o-tolyl)acetamide (32).

The target compound was obtained in 46% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.48 (m, 2H, mixture of rotamers), 8.32−8.26 (bs, 1H, one rotamer), 8.26−8.21 (bs, 1H, one rotamer), 7.72−7.68 (m, 1H, one rotamer), 7.61−7.57 (m, 1H, one rotamer), 7.45−7.38 (m, 2H, mixture of rotamers), 7.36−7.31 (m, 2H, mixture of rotamers), 7.28−7.26 (m, 1H, one rotamer), 7.24−7.12 (m, 5H, mixture of rotamers), 7.10−7.05 (m, 2H, mixture of rotamers), 7.01 (t, J = 8.8 Hz, 1H, one rotamer), 6.93 (t, J = 8.8 Hz, 1H, one rotamer), 6.86 (t, J = 7.6 Hz, 1H, one rotamer), 6.33 (d, J = 8.0Hz, 1H), 4.56 (s, 1H, one rotamer), 4.45−4.34 (m, 3H, mixture of rotamers), 4.01 (d, J = 15.1Hz, 1H, one rotamer), 3.92 (d, J = 14.7Hz, 1H, one rotamer), 3.24 (d, J = 1.2Hz, 3H, one rotamer), 3.22 (d, J = 1.2Hz, 3H, one rotamer), 2.33−2.18 (m, 5H, mixture of rotamers), 1.59 (s, 3H, one rotamer), 0.40−0.18 (m, 4H, mixture of rotamers), 0.10− −0.06 (m, 2H, mixture of rotamers), 0.11− −0.19 (m, 1H, one rotamer), −0.19 − −0.30 (m, 1H, one rotamer). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 171.3, 171.0, 151.2, 150.9, 149.3, 149.2, 141.4, 141.2, 137.5, 137.2, 135.8, 135.2, 132.3, 131.7, 131.5, 131.2, 128.8 (d, J = 6.4Hz), 128.5, 128.2, 127.3, 126.9, 125.7−125.3 (m), 123.0 (d, J = 21.9Hz), 115.7, 115.5, 115.5, 115.3, 65.0, 63.8, 48.6 (d, J = 8.0Hz), 48.2 (d J = 9.6Hz), 36.3, 36.2, 35.8, 35.4, 17.5, 16.9, 8.3, 8.0, 7.8, 7.3. ESI-MS (m/z): 472.1 [M+H]⁺.

N-Methyl-2-(propyl(1-(3-(trifluoromethyl)phenyl)cyclopropyl)amino)-2-(pyridin-3-yl)-N-(o-tolyl)acetamide (33).

The target compound was obtained in 42% yield as a colourless liquid. ¹H NMR (400 MHz, CDCl₃, 1:1 mixture of rotamers) δ 8.46−8.42 (m, 2H, mixture of rotamers), 8.04 (bs, 1H, one rotamer), 8.01 (bs, 1H, one rotamer), 7.51−7.27 (m, 10H, mixture of rotamers), 7.20−6.90 (m, 8H, mixture of rotamers), 6.75 (t, J = 7.6 Hz, 1H, one rotamer), 6.12 (d, J = 7.8 Hz, 1H, one rotamer), 4.57 (s, 1H, one rotamer), 4.29 (s, 1H, one rotamer), 3.16 (s, 3H, one rotamer), 3.12 (s, 3H, one rotamer), 3.01−2.88 (m, 2H, one rotamer), 2.85−2.68 (m, 2H, one rotamer), 2.22 (s, 3H, one rotamer), 1.46−1.34 (m, 4H), 1.41−1.12 (m, 7H, mixture of rotamers), 1.11−0.99 (m, 2H, one rotamer), 0.89−0.81 (m, 2H, one rotamer), 0.65−0.56 (m, 6H, mixture of rotamers). ¹³C NMR (101 MHz, CDCl₃, 1:1 mixture of rotamers) δ 173.1, 172.7, 151.1, 150.9, 148.9, 148.8, 146.4, 146.0, 141.2, 140.9, 137.2, 136.9, 135.9, 134.8, 134.2, 133.1, 131.7, 131.4, 130.6 (d, J = 29.6Hz), 128.83−128.54 (m), 128.1, 127.2, 126.6, 123.6 (m), 123.0−122.8(m), 64.0, 63.0, 49.0, 45.6, 45.3, 36.5, 36.3, 31.1, 29.8, 23.7 (d, J = 21.2Hz), 18.2, 17.6, 16.8, 15.4, 11.4 (d, J = 29.2Hz). ESI-MS (m/z): 482.1 [M+H]⁺.

2-(Cyclopropyl(3-(trifluoromethyl)benzyl)amino)-1-(piperidin-1-yl)-2-(pyridin-3-yl)ethan-1-one (34).

To a solution of N-(3-(trifluoromethyl)benzyl)cyclopropanamine (200 mg, 0.92 mmol) in trifluoroethanol (2 mL) was added nicotinaldehyde (90 μL, 0.92 mmol) and the solution was stirred for 30 min at 60 °C. The solution was cooled to RT and TMSCN (0.11 mL, 0.92 mmol) was added. The mixture was stirred at 60 °C for 30 min. The solvent was evaporated under reduced pressure, and the residue was purified by silica gel column chromatography using EtOAc/hexane as eluent to furnish 2-(cyclopropyl(3-(trifluoromethyl)benzyl)amino)-2-(pyridin-3-yl)acetonitrile as a colorless liquid in 82% yield.

To a solution of above 2-(cyclopropyl(3-(trifluoromethyl)benzyl)amino)-2-(pyridin-3-yl)acetonitrile (100 mg, 0.3 mmol) in EtOH : H₂O (3:3) was added NaOH (600 mg, 50 mmol), and the mixture and stirred at 95 °C for 6 h. Progress of the reaction was monitored by HPLC/MS. After completion of the reaction 6N HCl was added to pH 7, and the aqueous layer was extracted with EtOAc, dried over anhydrous Na₂SO₄ and concentrated under reduced pressure. The compound was used for next reaction without any further purification.

To a solution of the above acid derivative (40 mg, 0.11 mmol) in dry DMF (2 mL) at 0 °C were added piperidine (12 mg, 13 μL, 0.13 mmol), HATU (52 mg, 0.13 mmol) and i-Pr₂NEt (39μL, 0.22 mmol). The reaction mixture was stirred for 15 min at RT, then diluted with ethyl acetate, washed with water, and brine. The organic layer was dried over anhydrous Na₂SO₄, and the solvent was evaporated under vacuum. The residue obtained was purified by flash column chromatography (MeOH/DCM 1 : 20) to furnish the desired product in 73% yield as a colourless thick liquid.

¹H NMR (600 MHz, MeOD) δ 8.53 (d, J = 2.2Hz, 1H), 8.46 (dd, J = 4.9, 1.6Hz, 1H), 7.86 (dt, J = 7.9, 1.9Hz, 1H), 7.48 (m, 1H), 7.45−7.39 (m, 4H), 5.07 (s, 1H), 4.13 (d, J = 14.5Hz, 1H), 3.97 (d, J = 14.5Hz, 1H), 3.69−3.63 (m, 1H), 3.59−3.54 (m, 1H), 3.39−3.32 (m, 1H), 3.24 (m, 1H), 2.37 (m, 1H), 1.70−1.53 (m, 4H), 1.39 (m, 1H), 1.20−1.11 (m, 1H), 0.44−0.37 (m, 1H), 0.36−0.27 (m, 2H), 0.06− −0.04 (m, 1H). ¹³C NMR (151 MHz, MeOD) δ 170.6, 151.5, 149.5, 143.1, 139.7, 134.5, 133.5, 131.2 (q, J = 128Hz), 129.8, 126.3 (q, J = 15.6Hz), 125.0, 124.5 (d, J = 13.2Hz), 65.1, 57.3, 47.8, 44.1, 35.9, 27.2, 26.8, 25.2, 9.2, 7.6. ESI-MS (m/z): 418.1 [M+H]⁺.

Abbreviations

MOA

modes of action

ABS

asexual blood stage parasites

ACTs

artemisinin-based combination therapy

CQ

chloroquine

ART

artemisinin

ATQ

atovaquone

MFQ

mefloquine

LUM

lumefantrine

PPQ

piperaquine

WGS

whole genome sequencing

SAR

structure activity relationships

Hz

hemozoin

βH

β-hematin

DV

digestive vacuole

BRRoK

bioluminescence relative rate of kill

DGFA

dual gamete formation assay

PfCRT

P. falciparum CQ resistance transporter

CNV

copy number variations

SNPs

single nucleotide polymorphisms