Structure Activity Relationship of N-Substituted Phenyldihydropyrazolones Against Trypanosoma cruzi Amastigotes

Current drugs for Chagas disease have long treatment regimens with occurrence of adverse drug effects leading to poor treatment compliance. Novel and efficacious medications are therefore highly needed. We previously reported on the discovery of NPD-0227 (2-isopropyl-5-(4-methoxy-3-(pyridin-3-yl)phenyl)-4,4-dimethyl-2,4-dihydro-3H-pyrazol-3-one) as a potent in vitro inhibitor of Trypanosoma cruzi (pIC50 = 6.4) with 100-fold selectivity over human MRC-5 cells. The present work describes a SAR study on the exploration of substituents on the phenylpyrazolone nitrogen. Modifications were either done directly onto this pyrazolone nitrogen or alternatively by introducing a piperidine linker. Attention was pointed toward the selection of substituents with a cLogP preferably below NPD-0227s cLogP of 3.5. Generally the more apolar compounds showed better activities then molecules with cLogPs <2.0. Several new compounds were identified with potencies that are in the same range as NPD-0227 (pIC50 = 6.4) and promising selectivities. While the potency could not be improved, valuable SAR was obtained. Furthermore the introduction of a piperidine linker offers new opportunities for derivatization as valuable novel starting points for future T. cruzi drug discovery.


INTRODUCTION
The protozoan parasite Trypanosoma cruzi is the causative agent of Chagas disease. It is estimated that over six million people are infected worldwide, the majority in Latin-America where the parasite is endemic. (Lidani et al., 2019;WHO, 2021). T. cruzi is spread by several insect vectors, most of which belong to the triatomine family (Kollien and Schaub, 2000;Martinez et al., 2019). Other ways of transmission are via blood transfusions, laboratory accidents or parental transmission from mother to infant (Schmunis, 1999;Torrico et al., 2004;Rassi et al., 2010). The life cycle of T. cruzi consists of several developmental stages in the mammalian host and insect vector (Rassi et al., 2010).
While the vector appears unaffected by the parasite, infected humans can develop life-threatening symptoms and pathologies. The acute stage of the disease starts shortly after the parasite enters the body but a strong immune response will cause a large reduction of the initial peak of parasitemia, however, without full elimination of the parasite leading to persistent infection of certain tissues (Dias, 1984). The acute symptoms are non-specific, such as fever, mild splenomegaly and edema, making diagnosis diffucult (Prata, 2001). After 2 months, the disease enters the chronic phase in which the parasite becomes dormant and no symptoms are observed. This dormant phase can last for over 10 years up to lifelong (Prata, 2001). Estimates of further evolution of chronic infection toward clinical symptoms, such as cardiopathy or megaesophagus and megacolon, ranges between 10-60% of infected people and varies widely in different regions (Coura and Viñas, 2010;Coura and Borges-Pereira, 2011). Patients can become life-long asymptomatic carriers thus representing a parasite reservoir.
Vector control, early diagnosis and effective chemotherapy are essential to combat Chagas disease. Two drugs are currently marketed: the nitro-imidazole benznidazole (1, Figure 1) and the nitro-furan nifurtimox (2) (Rassi et al., 2010). Both drugs contain a nitro-aromatic functionality which is commonly associated with toxic side effects. Furthermore, the treatment regimens of both vary between 60-90 days and are known to cause various side effects which cause patients to discontinue their treatment (Castro et al., 2006;Pinazo et al., 2013). Benznidazole (1) is currently used as first-line treatment as it has a better overall safety profile (Alpern et al., 2017). While the efficacy of both drugs is well established in the acute phase, much debate is ongoing on about their efficacy in the chronic phase (Sgambatti de Andrade et al., 1996;Zhang and Tarleton, 1999;Bern, 2011). The recent BENEFIT trial investigated the efficacy of Benznidazole (1) on chronic Chagas' heart disease. While reduced parasitemia was observed by PCR, this did not result in significant reduction of clinical deterioration of cardiac function after 5 years (Morillo et al., 2015). Afterward several concerns were raised in literature with regard to the PCR analysis and benznidazole dosage, precluding firm conclusions on the efficacy potential in the chronic phase of infection (Hamers et al., 2016).
The overall need for new and improved chemotherapies for Chagas disease is high, both for current and future patients. New medication is also needed as contingency measure for upcoming resistance against the available current drugs (Coura and Viñas, 2010;Mejia et al., 2012). Unfortunately, the pipeline toward new drugs is largely empty. Most research on T. cruzi is performed in academia though the last few years several public-private collaborations have been initiated. In this paper, we describe part of the research of the EU-funded, public-private consortium PDE4NPD, that focuses on 3′5′cyclic nucleotide phosphodiesterases (PDEs) as targets against several neglected tropical diseases. The present work further elaborates the SAR of the previously reported T. cruzi inhibitor NPD-0227 (3, Figure 1) and investigates the role of different substituents on the pyrazolone nitrogen (Sijm et al., 2019). Besides the SAR aiming at improving potency, compounds were also specifically designed to improve physicochemical properties, such as cLogP and solubility. As T. cruzi ultimately proceeds via a dormant intracellular form, a possible drug needs to pass several cell membranes before reaching the parasite. During this process the drug is transferred multiple times between hydrophobic and hydrophilic phases, as such an optimal cLogP is thought to be between 0 and 2 and the drug having no charges (Basore et al., 2015;Bennion et al., 2017).

Chemistry
Phenylpyrazolone analogues with a substituent directly on the pyrazolone nitrogen atom were synthesized using two different routes. The first route started with a Suzuki reaction to obtain dihydropyrazolone 5 (Route A, Scheme 1) which was then functionalized using sodium hydride and various acyl chlorides and sulfonylchlorides to obtain compounds 6-10 ( Table 1). The second route (Route B) started with alkylation of the dihydropyrazolone 4, either using potassium carbonate to yield intermediates 11-12, or using sodium hydride to generate intermediates 13-18. These intermediates were then converted to the 3-pyridinyl derivatives via a Suzuki cross coupling, yielding final compounds 20-26 (Scheme 1, Table 1).
Introduction of methylene-oxadiazoles on the pyrazolone head group started with the previously synthesized ester analogues 13 and 14 (Scheme 2). Subsequent hydrazination yielded the corresponding hydrazides (27 and 28) which were ring-closed with either triethylorthoformate or cyanuric bromide to deliver the unsubstituted oxadiazoles (29 and 31) and the amine substituted oxadiazoles (30 and 32) respectively. These oxadiazoles were then converted to the  (Sijm et al., 2019). This keto-ester was condensed with 4-hydrazinylpiperidine to yield piperidine substituted dihydropyrazolone 38. The piperidine nitrogen atom was protected using boc-anhydride to give intermediate (39) which could be used in a Suzuki cross-coupling to install the 3-pyridine moiety. Subsequent deprotection of 40 with 4 M HCl in dioxane yielded the free amine building block (41) which was used to install the desired electrophiles (42-83, Tables 2,3) via various methodologies.

RESULTS AND DISCUSSION
In the previous work by Sijm et al., the pyrazolone nitrogen was substituted with various (cyclo)-alkyl moieties resulting in the identification of NPD-0227 (3) as a potent (pIC 50 6.4) T. cruzi inhibitor with 100-fold selectivity over human MRC-5 cells ( Table 1) (Sijm et al., 2019;Sijm et al., 2020). To further explore SAR and to investigate molecules with varying cLogPs, more polar and diverse substituents was explored.
Introduction of a carbonyl-alkyl moieties directly onto the pyrazolone nitrogen (6-8, Table 1) resulted in weak inhibitors with pIC 50 's around 4.5. Similarly, introduction of sulphonylalkyl moieties such as dimethylsulfonamide (9) or sulphonylcyclopentyl (10) did not lead to inhibitors with pIC 50 values above 5.0. Epoxides were used to introduce β-alcohols on the pyrazolone ring resulting in compounds 19 and 20. These alcohols also showed a large decrease in activity compared to NPD-0227 (pIC 50 6.4) with pIC 50 's of 4.4 and 5.1, respectively. Similarly, the introduction of ethyl-esters (21, 22) resulted in decreased potencies compared to NPD-0227 with both compounds showing pIC 50 values of only 5.2.
While the substituents with a polar moiety (6-10, 19-22, Table 1) showed decreased activities, as expected, introduction of more apolar moieties such as methyl-cyclohexyl (23) and pyrazole (24) showed moderate activity with pIC 50 values of 5.5. Introduction of more polar substituents attached to an aliphatic linker, such as imidazole 25 and morpholine 26 with cLogPs around 2.3 resulted in lower potencies (pIC 50 < 4.6). While the introduction of various oxadiazoles (33-36) on this position yielded compounds with a more desirable cLogP's, this did not lead to very active compounds, with 1-propan-oxazole (34) showed the highest potency (pIC 50 5.2). As activities were generally close to the lower detection limit of the assay, selectivity indexes were quite low or defined as "bigger then".
The initial attempts to introduce polar functional groups such as carbonyls, sulphonyls and alcohols close to the pyrazolone ring had a negative effect on their potency. One of the most potent compounds so far was methylcyclohexyl 23, a bulky and apolar substituent. It was thought that introduction of a piperidine would introduce a similar sized moiety which would improve options to introduce polarity as a new handle to further modify.
Overall, the introduction of polar moieties on the phenyldihydropyrazolones generally led to low activities against T. cruzi.
Especially the introduction of a polar moiety directly next to the pyrazolone nitogen such as all carbonyl, sulphonyl and β-alcohol linked moieties (6-10, 19, 20) resulted in analogues with low potencies. Introduction of aliphatic substituents or aromatic moieties resulted in interesting new anti-T. cruzi inhibitors with the best compounds 54 and 76 reaching sub micromolar potency. Installing a piperidine linker allowed for the introduction of a variety of substituents and it became apparent that aromatic moieties, either directly, with a methylene bridge or with a 2-atom linker performed best, showing pIC 50 's around 6.0.

CONCLUSION
Our study results in valuable SAR data that has been obtained by introducing a variety of substituents on the dihydropyrazolone nitrogen atom. While decorating this position is synthetically very efficient and a wide variety of chemical functionalities are allowed, our studies did not lead to compounds that have a better activity than NPD-0227. Especially the piperidine linker opens a whole new range of options to introduce substituents and analogues reached low micromolar inhibitory activities against T. cruzi. The most active compounds were obtained if the piperidine is substituted with apolar moieties such as aryl or benzyl rings. Introduction of more polar substituents, such as imidazole 25 (cLogP 2.3), succinimide 46 (cLogP 0.9) and morpholine 73 (cLogP 1.6) generally lead to compounds with high micromolar activities. The structural moiety performing best is a 2-atom linker followed by an aromatic moiety, of which sulphonamide 54 (cLogP 3.8), urea 59 (cLogP 4.0) and β-alcohols 65 and 66 (cLogP 3.8/3.9) performed best. These molecules all have activities around 6.0 (pIC 50 ) and especially sulphonamide 54 is promising with a pIC50 of 6.2 and >100-fold selectivity over MRC-5 cells. Also other aromatic substituents performed well, such as 4-fluorobenzyl (76), 3-cyanophenyl (79) and 5-methylnicotinate (82) show promising activities around 6.0, with 79 and 82 showing best selectivies (>32-fold) toward T. cruzi. While these results of this study are important in the design of novel 'drug leads' for Chagas disease, the original goal, of identifying high potency molecules with a lower cLogP did not succeed. Most potent compounds with a cLogP below 2.0 were oxadiazole 33 and sulfamide 51, which both showed a pIC 50 of 5.1.

Experimental Section Biology
Trypanosoma cruzi in vitro assay. Bloodstream trypomastigotes of the Y strain of T. cruzi were obtained by cardiac puncture of infected Swiss Webster mice on the parasitaemia peak (Meirelles et al., 1986;Batista et al., 2010). For the standard in vitro susceptibility assay on intracellular amastigotes, T. cruzi Tulahuen CL2, β-galactosidase strain (nifurtimox-sensitive) was used. The strain is maintained on MRC-5 SV2 (human lung fibroblast) cells in MEM medium, supplemented with 200 mM L-glutamine, 16.5 mM NaHCO 3 and 5% inactivated fetal calf serum (FCSi). All cultures and assays were conducted at 37°C under an atmosphere of 5% CO 2 . Benznidazole was included in the assays as a positive control.
MRC-5 cytotoxicity in vitro assay. MRC-5-SV2 cells, originally from a human diploid lung cell line, were cultivated in MEM supplemented with L-glutamine (20 mM), 16.5 mM sodium hydrogen carbonate and 5% FCSi. For the assay, 10 4 cells/well were seeded onto the test plates containing the pre-diluted sample and incubated at 37°C and 5% CO 2 for 72 h. Cell viability was assessed fluorometrically 4 h after addition of resazurin (excitation 550 nm, emission 590 nm). The results are expressed as percentage reduction in cell viability compared to untreated controls. Tamoxifen was included in the assays as a positive control.
The use of laboratory rodents was carried out in strict accordance to all mandatory guidelines (EU directives, including the Revised Directive 2010/63/EU on the Protection of Animals used for Scientific Purposes that came into force on January 1, 2013, and the Declaration of Helsinki in its latest version).

Experimental Section Chemistry
Chemicals and reagents were obtained from commercial suppliers and were used without further purification. Anhydrous DMF, THF, and DCM were obtained by passing them through an activated alumina column prior to use. Microwave reactions were executed using a Biotage ®  (Hz). Chemical shifts are reported in ppm with the natural abundance of deuterium in the solvent as the internal reference [CDCl 3 : δ 7.26, (CD 3 ) 2 SO: δ 2.50]. 13 C NMR spectra were recorded on a Bruker Avance 500 (126 MHz) or Bruker Avance 600 (150 MHz). Chemical shifts are reported in ppm with the solvent resonance resulting from incomplete deuteration as the internal reference [CDCl 3 : δ 77.16 or (CD 3 ) 2 SO: δ 39.52]. Systematic names for molecules according to IUPAC rules were generated using the Chemdraw AutoName program. LC-MS data was gathered using a Shimadzu HPLC/MS workstation with a LC-20AD pump system, SPD-M20A diode array detection, and a LCMS-2010 EV mass spectrometer. The column used is an Xbridge C18 5 μm column (100 mm × 4.6 mm). Solvents used were the following: solvent B ACN, 0.1% formic Acid; solvent A water, 0.1% formic acid. The analysis was conducted using a x?ow rate of 1.0 ml/min, start 5% B, linear gradient to 90% B in 4.5 min, then 1.5 min at 90% B, linear gradient to 5% B in 0.5 min and then 1.5 min at 5% B, total run time of 8 min. All reported compounds have purities >95%, measured at 254 nm, unless otherwise mentioned. All HRMS spectra were recorded on a Bruker micrOTOF mass spectrometer using ESI in positive-ion mode. In case of bromine containing molecules, the lowest peak of 79 Br was reported. Column puri?cations were either carried out automatically using Biotage equipment or manually, using 60-200 mesh silica. TLC analyses were performed with Merck F254 alumina silica plates using UV visualization. All reactions were done under N 2 atmosphere, unless Frontiers in Chemistry | www.frontiersin.org April 2021 | Volume 9 | Article 608438 8 specifically mentioned. The cLogPs were calculated using CDD vault, CDD vault uses the ionic logP algorithm. Compound 5,6,15,23,27,29,33,[38][39][40][41][42] and 83 are described below, other compounds can be found in the supporting information.