2021 Volume 69 Issue 12 Pages 1195-1199
A series of quinone derivatives with a variety of side chains were synthesized. These synthetic quinone compounds were evaluated for in vitro antitrypanosomal activity against trypomastigotes and amastigotes of Trypanosoma cruzi, the causative agent of Chagas disease. Measurement of solubility of quinones and their ability to permeate cell membranes were assessed to address their possible use as oral drugs. Some synthesized compounds exhibited potent antitrypanosomal activity. However, most compounds with a promising activity showed poor solubility that did not seem suitable for oral usage. Meanwhile, compound 5a, an N-tert-butoxycarbonylpiperidine derivative, exhibited good antitrypanosomal activity, ability to permeate membranes, and good solubility.
Chagas disease is a neglected tropical disease caused by the protozoan parasite, Trypanosoma cruzi (T. cruzi).1) The parasite is primarily transmitted via bites of the carrier vector, the blood-sucking “kissing bug.” Blood transfusion, organ transplants, congenital transmission, and ingestion of contaminated food have been documented as other routes of infection.2–4) In rural areas of 21 Latin American countries where T. cruzi is endemic, Chagas disease is an important health problem. The disease affects 5–8 million people and causes approximately 10000 deaths per year. Almost 70 million people are at risk of infection.5) Recently, the disease has become a public health concern in nonendemic countries in North America, Europe, Australia, and Japan as a result of immigration from Latin America.6–9)
Chagas disease has two distinct phases. The acute phase is often asymptomatic, or patients develop nonspecific symptoms, such as fever and general malaise. In most cases, these symptoms decline spontaneously after 1–2 months.10,11) After activation of the host immune system, parasitemia fades and becomes undetectable. This indeterminate state can last for years and even decades. The majority of infected individuals will continue their life unaffected. However, between 30 and 40% of infected individuals will progress to chronic disease. Such patients may develop severe cardiac and digestive pathologies leading to severe disability or death.12)
Only two drugs, benznidazole and nifurtimox developed more than 40 years ago, are currently in use.13,14) Both drugs cause significant side effects, require long courses of treatment—up to 60 d, and are effective only in the acute phase of infection. Efficacy is limited in the chronic phase. The emergence of drug resistance is also problematic. CYP51 inhibitors, such as posaconazole and E1224 (prodrug of ravuconazole), recently demonstrated encouraging results in in vitro and in vivo models and were tested in clinical trials.15,16) Unfortunately, these azoles were ineffective as compared with benznidazole. Thus, an urgent need exists for safe drugs with high clinical efficacy in both phases of the disease.
Komaroviquinone, a quinone-type diterpene isolated from Dracocephalum komarovii, inhibits the growth of trypomastigotes of T. cruzi.17,18) We developed an asymmetric total synthesis of komaroviquinone and studied the importance of the quinone moiety and fused cyclic structure for activity against T. cruzi. We thus evaluated the antitrypanosomal activity of synthetic intermediates of komaroviquinone and related quinone compounds.19,20) The quinone moiety of komaroviquinone was crucial for expression of antitrypanosomal activity but the complex structure with chiral centers was not. Further, we identified that some quinones with simplified organic syntheses showed antitrypanosomal activity similar to komaroviquinone. A representative is compound 2 in Fig. 1. We present a structure–activity relationship study of compound 2. We synthesized compound 2-related quinone compounds and evaluated their in vitro antitrypanosomal activity and assessed their solubility and ability to permeate cell membranes.
An overview of analogs is provided in Fig. 2. We synthesized compounds 3a–k where the other substituent was replaced by an isopropyl group in compound 2. These analogs were used to examine whether lipophilicity, size, and electronic character of substituents (electron donating or accepting) on the quinone structure affected antitrypanosomal activity. Compounds 4a–h, 5a, and 5b with acyl groups consisting of a substituted phenyl ring or N-substituted piperidyl group instead of the cyclohexyl group were synthesized to investigate the importance of the acyl group for antiparasitic activity.
A series of quinone derivatives was synthesized (Chart 1 or 2). Aldehydes 6a–k were reacted with hydroxylamine to form corresponding oximes, which were reduced with zinc to amines 7a–k under acidic conditions. These amines were further reacted with cyclohexanecarbonyl chloride and final oxidation using cerium ammonium nitrate (CAN) to produce quinone analogs 3a–k.
Reagents and Conditions: (a) (1) HONH2·HCl, C2H5OH, room temperature (r.t.), (2) AcCl, C2H5OH, 0 °C, (3) Zn, C2H5OH, 50 °C; (b) cHexCOCl, iPr2NEt, CH2Cl2, 0 °C; (c) CAN, CH3CN/H2O, 0 °C.
For the synthesis of quinone derivatives 4a–h, 5a, and 5b we first prepared compound 9 under previously reported conditions20) (Chart 2). Carboxylic acids used as starting materials were commercially available. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDCI·HCl) mediated coupling of amine 9, and various carboxylic acids gave amides 10a–h, 11a, and 11b. Quinone derivatives 4a–h, 5a, and 5b were obtained by oxidation of amides 10a–h, 11a, and 11b with CAN.
Reagents and Conditions: (a) Carboxylic Acid, EDCI·HCl, CH2Cl2, r.t.; (b) CAN, CH3CN/H2O, 0 °C.
We used the above compounds to evaluate antitrypanosomal activity. Three forms of T. cruzi are known: epimastigotes, trypomastigotes, and amastigotes. Trypomastigotes and amastigotes are present in the human body during infection. Trypomastigotes are the form of the parasite found in the human bloodstream. T. cruzi trypomastigotes invade tissues and undergo a morphological change to amastigotes. This form proliferates within cells. E1224 and posaconazole both showed inadequate efficacy in Phase 2 studies, with a high rate of protozoan relapse after the end of the treatment period.15,16) Posaconazole was reported to exhibit potent antiprotozoal activity against amastigotes, except for a few strains, but not against trypomastigotes.21) Posaconazole activity may require the parasite to undergo a certain number of replications. We assessed antiprotozoal activity of the quinone derivatives against both amastigotes and trypomastigotes to confirm activity throughout the life cycle of the parasite. We evaluated the viability inhibitory activity of quinone derivatives against trypomastigotes, their inhibitory activity against trypomastigotes to invade host cells, and their viability inhibitory activity against amastigotes. IC50 values against trypomastigotes and amastigotes and cytotoxicity of compounds are provided in Table 1.
| Compound | T. cruzi IC50 (µM) | Swiss 3T3 CC50 | ||
|---|---|---|---|---|
| Trypomastigotes | Amastigotes | |||
| Infection rate | Amastigotes/infected cell | |||
| Benznidazole | 6.4 | 5.6 | 1.9 | >100 |
| 2 | 6.8 | >1 | 0.9 | >100 |
| 3a | 2.2 | 0.9 | >1 | >100 |
| 3b | 2.9 | 0.34 | 0.009 | >100 |
| 3c | 0.1 | 0.5 | 0.17 | >100 |
| 3d | 0.3 | 4.6 | 2.8 | >100 |
| 3e | 0.38 | 0.72 | 0.08 | >100 |
| 3f | N.D. | 0.6 | 0.08 | >100 |
| 3g | 0.14 | 3 | 2.9 | >100 |
| 3h | 0.16 | 3.4 | 3.2 | >100 |
| 3i | 1.5 | >1 | 0.32 | >100 |
| 3j | 1.2 | >1 | >1 | >100 |
| 3k | 1.8 | >1 | >1 | >100 |
| 4a | 0.35 | N.D. | N.D. | >100 |
| 4b | 1.2 | 3.9 | 1.8 | >100 |
| 4c | 0.15 | 2.7 | 1.4 | >100 |
| 4d | 0.2 | 3.8 | 1.6 | >100 |
| 4e | 0.17 | 7.1 | 3.6 | >100 |
| 4f | 0.14 | 3.8 | 3.2 | >100 |
| 4g | 0.12 | 1.1 | 2.4 | >100 |
| 4h | 0.1 | 0.17 | 0.13 | >100 |
| 5a | 2.3 | 0.54 | 0.068 | >100 |
| 5b | 1.6 | 0.25 | 0.058 | >100 |
“Infection rate” is the concentration of a drug at which the percentage of host cells infected with T. cruzi is halved. “Amastigotes/infected cell” is the concentration of drug at which the number of parasites in the host cell is halved compared with the absence of the drug. Antiprotozoal activity of most compounds against trypomastigotes is greater than the antiprotozoal activity of original compound 2. The antiprotozoal activity of compounds 3b, 3c, 3e, 3f, 3i, 4h, 5a, and 5b against amastigotes was also greater than the activity of compound 2. Replacement of the isopropyl group on the quinone with a linear alkyl group (compounds 3b and 3c) or with ring structures (compounds 3e and 3f) showed high antiprotozoal activity against amastigotes. Replacement of the isopropyl group with a methoxy group (compound 3i) showed higher antiprotozoal activity against trypomastigotes and amastigotes than observed for compound 2. In contrast, replacement with halogens (compounds 3j and 3k) did not show any improvement in activity against amastigotes. All compounds with a monosubstituted phenyl group instead of the cyclohexyl group on the amide side of compound 2 showed good antiprotozoal activity against trypomastigotes. However, except for compound 4h, with a methoxy group, all compounds showed lower antiprotozoal activity against amastigotes.
Compounds with an N-alkoxycarbonylpiperidine instead of the cyclohexyl group (compounds 5a and 5b) showed high antiprotozoal activity against amastigotes, with IC50 values among the highest for compounds synthesized for this study. Antiprotozoal activity against trypomastigotes was modest compared with other compounds, but 3- to 4-fold greater compared with compound 2. No clear correlation was found between the antiprotozoal activity against trypomastigotes and amastigotes. For example, compound 4e showed seven times higher antiprotozoal activity against trypomastigotes compared with compound 4b; however, the activity against amastigotes (as amastigotes/infection cell) was half that of compound 4b. Similar results were observed between compounds 3b, 3g, and 3h. High antiprotozoal activity against one form of T. cruzi does not necessarily indicate high activity against another form. Compounds with low activity in inhibition of survival against trypomastigotes but high activity in inhibition of infection, such as compounds 3b, 5a, and 5b, may inhibit the adhesion of trypomastigotes to host cells and the invasion process. On the other hand, compounds with high activity in inhibition of survival against trypomastigotes but low inhibition of infection, such as compounds 3d, 3g, 3h, and 4c–4f, may take longer to inhibit the activity of trypomastigotes and allow them to invade and proliferate in host cells. In order to evaluate the overall antiprotozoal activity of a compound, the antiprotozoal activity against both forms should be evaluated from several perspectives.
Drugs for Neglected Diseases initiative (DNDi) indicates in its target product profile for Chagas disease that an ideal drug formulation is compatible with oral administration.22) Absorption of orally administered compounds depends on dissolution in gastrointestinal fluid and permeation through the small intestinal mucosa during gastrointestinal transit. For a drug to be absorbed orally, it must have adequate solubility and the ability to cross cell membranes.23) We evaluated these properties by measuring solubility at pH 6.8 and membrane permeability using Parallel Artificial Membrane Permeability Assay (PAMPA)24) (Table 2).
| Solubilitya) | Permeabilityb) (nm/s) | Solubilitya) | Permeabilityb) (nm/s) | ||||
|---|---|---|---|---|---|---|---|
| Compound | (µM) | (µg/mL) | Compound | (µM) | (µg/mL) | ||
| 2 | 107 | 34 | 354 | 4a | 121 | 38 | 320 |
| 3a | 462 | 135 | 264 | 4b | 70 | 23 | 290 |
| 3b | 33 | 10 | 307 | 4c | 8 | 3 | N. D. |
| 3c | <1 | <1 | N. D. | 4d | 3 | 1 | N. D. |
| 3d | 6 | 2 | N. D. | 4e | 59 | 21 | 210 |
| 3e | <1 | <1 | N. D. | 4f | 6 | 2 | N. D. |
| 3f | 12 | 4 | 276 | 4g | 9 | 3 | N. D. |
| 3g | 3 | 1 | N. D. | 4h | 19 | 5 | 275 |
| 3h | 11 | 4 | N. D. | 5a | 373 | 177 | 217 |
| 3i | 1132 | 348 | 129 | 5b | 38 | 17 | 186 |
| 3j | Decomp. | Decomp | N. D. | Caffeine | N. D. | N. D. | 120 |
| 3k | 920 | 272 | N. D. | Sulfasalazine | N. D. | N. D. | 1.1 |
a) Solubility at pH 6.8. b) Corning® BioCoat™ Pre-coated PAMPA Plate System was used.
Compound 2 showed higher membrane permeability than caffeine. Caffeine displays high membrane permeability. The solubility of compound 2 at pH 6.8 was 34 µg/mL. Compounds 3a and 3i, with the isopropyl group replaced by the methyl and methoxy groups, respectively, showed good permeability and improved solubility. Compounds 3a and 3i exhibited approximately 4- and 10-fold increased solubility, respectively, compared with compound 2. However, compound 3b, with an n-propyl group instead of the isopropyl group, showed approximately a three-fold decrease in solubility. Further, compounds 3c–h had longer alkyl chains or cyclic alkyl groups and displayed poor solubility, showing that oral administration would be impractical. Compounds 4a–h, with a phenyl ring substituted for the cyclohexyl ring, displayed lower solubility than compound 2. Compound 4h, with promising antitrypanosomal activity, also showed problematic aqueous solubility as low as compounds with monosubstituted phenyl rings. Compounds 5a and 5b, both with an N-alkoxycarbonylpiperidine group, showed good permeability but varied substantially in solubility. Compound 5a, with a tert-butoxycarbonyl group at the nitrogen of piperidine ring, showed 4.6-fold increased solubility compared with compound 2; however, a benzyloxycarbonyl group in compound 5b yielded half this solubility. Adequate antitrypanosomal activity and permeability combined with contrasting aqueous solubility of compounds 5a and 5b suggest that solubility can be improved while maintaining desirable biological activity and membrane permeability by exploring the alkoxy group of compounds with a carbamate structure as contained in compounds 5a and 5b.
Compound 5a was predicted to be efficiently absorbed by the gastrointestinal tract using Swiss ADME25) to estimate various parameters of the compound. This prediction probably reflected very high membrane permeability and good solubility of compound 5a. The calculated log P value of compound 5a was 2.17, suggesting an adequate balance between lipophilicity and hydrophilicity. IC50 values of compound 5a against trypomastigotes and amastigotes were 2.3 and 0.068 µM, respectively. Lipophilic ligand efficacies (LLE)26) calculated from the log P value were 3.46 and 5.01. An LLE value should be greater than 5. Compound 5a meets this requirement on the basis of its antiprotozoal activity against amastigotes.
Metabolic stability in mouse and human liver microsomes, plasma protein binding (PPB), and photostability of compound 5a are summarized in Table 3.
| Metabolic stabilitya) (mL/min/kg) | Plasma protein binding (%) | Light stabilityb) (%) | |||
|---|---|---|---|---|---|
| Human | Mouse | Human | Mouse | Light, 0 °C | Dark, 37 °C |
| 343 | 1849 | 96 | 93 | 45 | >90 |
a) Clearance in human and mouse liver microsomes. b) % of remaining after 3 h.
The rate of metabolism of compound 5a was high in both mouse and human liver microsomes, as were PPB values. These properties may pose a problem for the evaluation of efficacy and safety using model animals and are issues that must be resolved in the future. Instability in light was also identified as a problem for compound 5a. This compound was stable at 37 °C in a buffer solution in the dark, but considerable decomposition was observed even at 0 °C under fluorescent light irradiation. This problem is not critical for evaluating drug efficacy at the laboratory level, but care must be taken when handling solutions. Komaroviquinone, the starting point for quinone synthesis is converted to komarovispirone by light irradiation,27) and photodegradation of compound 5a may occur by a similar mechanism. Photoisomerization of komaroviquinone likely involves radical dehydrogenation of the hydroxyl group at position 7 adjacent to the quinone structure as a starting point for subsequent structural transformation. Therefore, adjusting chemical moieties around the quinone structure of compound 5a may improve its stability in light.
We synthesized quinones in which the isopropyl group on the quinone ring of compound 2 was replaced by other groups. Also, we synthesized quinones in which the cyclohexyl ring on the amide side of compound 2 was replaced by other cyclic structures. These compounds were evaluated for their antiprotozoal activity against trypomastigotes and amastigotes of T. cruzi. In addition, solubility and membrane permeability were measured to confirm whether compounds with promising antiprotozoal activity had physicochemical properties compatible with oral administration. Data were also used to assess relationships between chemical structure and physicochemical properties. Most compounds with good antiprotozoal activity showed poor solubility; however, compound 5a showed adequate antiprotozoal activity, solubility, and high membrane permeability. We are planning to evaluate the in vivo efficacy of compound 5a using acute and chronic phase mouse models. Compound 5a has issues with metabolic stability and photostability. We hope to solve these problems with an additional study of structure–activity and structure–property relationships around the amide nitrogen and carbamoyl groups.
The present work was supported by a Grant Science and Technology Research Partnership for Sustainable Development (SATREPS) from Japan Agency for Medical Research and Development (AMED). This research was partially supported by Platform Project for Supporting Drug Discovery and Life Science Research (basis for Supporting Innovative Drug Discovery and Life Science Research (BINDS)) from AMED under support number 2729.
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
