Chemical and Pharmaceutical Bulletin
Online ISSN : 1347-5223
Print ISSN : 0009-2363
ISSN-L : 0009-2363
Special Collection of Papers: Reviews
π-Delocalized Lipophilic Cations as New Candidates for Antimalarial, Antitrypanosomal and Antileishmanial Agents: Synthesis, Evaluation of Antiprotozoal Potency, and Insight into Their Action Mechanisms
Kiyosei Takasu
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML

2016 Volume 64 Issue 7 Pages 656-667

Details
Abstract

The search for new drugs that could treat tropical protozoan diseases, such as malaria or neglected tropical diseases (NTDs), motivates many medicinal chemists. New classes of antiprotozoal drugs that act through a novel mechanism of action must be developed. This review presents our efforts toward finding new candidate treatments for malaria, American trypanosomiasis, human African trypanosomiasis and leishmaniasis based on π-delocalized lipophilic cations (DLCs). DLCs, such as rhodacyanines, azarhodacyanines, β-carbolinium salts, and phenoxazinium salts, displayed strong antiprotozoal activities with highly selective indices. Several DLCs displayed moderate to excellent in vivo efficacies against Plasmodium berghei when administered intraperitoneally or orally. This review also discusses chemical biology approaches to understanding the mechanism of action underlying the antimalarial rhodacyanines.

1. Introduction

Parasitic diseases account for a significant burden of morbidity and mortality around the globe. Human infectious diseases caused by unicellular protozoa, such as malaria, Chagas disease, human African trypanosomiasis, and leishmaniasis pose the greatest threat in the tropical and subtropical regions.1,2) The latter three diseases are called “neglected tropical diseases (NTDs).” Malaria is caused by the Plasmodium protozoa, which are transmitted by female mosquitos of the genus Anopheles. Four species of parasites, including P. falciparum, P. ovale, P. vivax, and P. malaliae, are responsible for the disease in human beings. Among these, P. falciparum is the most deadly and widespread parasite. Almost half of the world’s population (about 3.2 billion people) is at risk of malaria. The WHO reported the occurrence of 214 million cases of malaria and 438000 deaths, mostly of children under 5 years of age, in 2015.1) Chagas disease (American trypanosomiasis) is caused by Trypanosoma cruzi, which is carried by triatomine bugs. Currently, around 7.5 million people are infected with T. cruzi in Latin America and another 25 million people are at risk of acquiring the disease.3) Recent surveys indicate that 41200 new cases and 12500 deaths associated with this condition are diagnosed each year. Human African trypanosomiasis (African sleeping sickness) mainly occurs in Sub-Saharan Africa and is caused by the species Trypanosoma brucei, which is carried by the tsetse flies. It is estimated that approximately 70 million people are at risk levels of African sleeping sickness.4) Leishmaniasis affects Africa, Asia, and Latin America and is caused by more than 20 Leishmania species, which are transmitted by phlebotomine sandflies. Among the various types of leishmaniasis, visceral leishmaniasis (VL), known as Kala-Azar, is the most serious form of the disease. A total of 98 countries reported endemic leishmaniasis transmission, with 0.2–0.4 million new cases of VL declared yearly.5) These diseases prevent the social and economic development of many low income countries in the tropical and subtropical regions. Recently, even inhabitants of temperate zones have become exposed to the danger of these protozoal infections due to global warming and global transportation.

Despite the tremendous toll that these diseases take, effective treatments and prevention measures for these diseases remain unavailable. Although the development of effective vaccines against the major protozoa has been investigated by the global pharmaceutical companies, the progress is still slowly.6,7) Few chemotherapeutics are available for clinical use for the treatment of the protozoal diseases (Fig. 1). These drugs suffer from certain drawbacks, such as severe side effects, low efficacy, high cost, and the occurrence of drug resistance.8,9) Consequently, development of new classes of antiprotozoal drugs that potentially act through a novel mechanism of action are required.1013) My colleagues and I have sought to develop new antiprotozoal agents under the supervision of Prof. Masataka Ihara from 2000 to 2007. In this account, I summarize these studies, including the synthesis and biological evaluation of new drug candidates for malaria and NTDs as well as the chemical biology studies conducted to understand the modes of action of these novel compounds.

Fig. 1. Clinically Available Antiprotozoal Drugsa)

a) 1–3 are used for treatment of malaria, and 46 are used for treatment of Chagas disease, Human African Trypanosomiasis and Leishmaniasis, respectively.

2. π-Delocalized Lipophilic Cationic Motif

Heterocycles, carbohydrates and aromatics synthesized in our laboratory (Ihara’s lab. in Tohoku Univ.) were screened in an effort to identify potential antimalarial lead compounds.1416) Several substances with a π-conjugated iminium moiety showed strong antimalarial activities in vitro (Table 1, entries 1–3). Particularly significant was the finding that the rhodacyanine dye (7a), MKT-077, showed strong inhibitory effects (IC50=7.0×10−8 M) against the erythrocytic stage of P. falciparum (chloroquine sensitive FCR-3 strain) with a low cytotoxicity against mammalian FM3A cells (entry 3). The properties of 7 are comparable with quinine (1) and chloroquine (2) treatment (entries 4, 5).

Table 1. Antimalarial Activities and Cytotoxicities in the Broad Screening
EntryCmpdIC50 (M)Selective indexc)
P. falciparuma)FM3Ab)
1Methylene blue1.7×10−81.1×10−665
2Rhodamine 1233.0×10−71.0×10−533
3MKT-077 (7a)7.0×10−81.5×10−5210
4Quinine (1)1.1×10−71.0×10−4910
5Chloroquine (2)1.8×10−83.2×10−51800

a) Chloroquine sensitive strain (FCR-3). b) Mouse mammary tumor FM3A cells representing a model of host. c) Selective index=IC50 value for FM3A/IC50 for P. falciparum.

MKT-077 (7a)1719) was developed as a potential antitumor agent by FUJIFILM Co. based on the DLC hypothesis, originally proposed by Chen.20) He proposed that π-delocalized lipophilic cations (DLCs) could penetrate the hydrophobic barriers of the plasma and mitochondria membrane under a cross-membrane negative potential.2123) The mitochondrial membrane potentials of carcinoma cells are more negative than those of normal cells, and this negative membrane charge induces the selective transfer of DLCs into carcinoma mitochondria. If DLCs are toxic to mitochondria at high concentrations, their selective accumulation in carcinoma mitochondria should induce selective carcinoma cell killing. Several DLC compounds reportedly exhibit selective antitumor activities. Vaidya and colleagues reported that a collapse of the mitochondrial membrane potential in malarial parasites is observed during treatment with antimalarial agents.24) We expected that DLCs could provide a new class of chemotherapeutics for the treatment of malaria.

3. Antimalarial Activities of the Rhodacyanines2529)

3.1 Synthesis of Five Types of Rhodacyanines

The skeleton of the original rhodacyanine ([0, 0] rhodacyanine), shown in Fig. 2, comprises three, linearly linked heterocycles, such that the two terminal heteroaromatic rings (A, C) are fused to a central rhodamine (4-oxothiazolidine) ring (B). The dyes are double conjugates of two different dye units, with left and right parts comprising neutral merocyanine and cationic cyanine structures. Consequently, the rhodacyanine compounds can be classified as DLCs. The reaction sequence used to prepare the [0, 0] rhodacyanine 7a is shown in Chart 1. Condensation of thiazolium salt 9a, prepared from 2-thiobenzothiazole 8a and methyl p-toluenesulfonate, with 3-rhodanine 10a in the presence of triethylamine at 0°C afforded the neutral merocyanine 11a. S-Methylation of 11a into 12a, followed by condensation with 2-methylpyridinium p-toluenesulfonate in the presence of triethylamine, provided the desired rhodacyanine as a p-toluenesulfonate salt. Treatment of the p-toluenesulfonate with the chloride ion form of ion-exchange resins, such as Amberlite IRA-400, gave the chloride salt 7a. All synthetic intermediates were readily precipitated in their pure forms in good yields after a simple work-up. We demonstrated the one-pot rapid synthesis of the [0, 0] rhodacyanines 7 from thiazolium 9, as well as their parallel combinatorial syntheses.26)

Fig. 2. General Structure of Class I [0, 0] Rhodacyanines
Chart 1. Typical Synthetic Route to [0, 0] Rhodacyanines 7a

Reagents and conditions: (a) methyl p-toluenesulfonate (TsOMe), anisole, 125°C; (b) 10, NEt3, MeCN, 0°C; (c) TsOMe, DMF, 120°C; (d) 13, NEt3, MeCN, 70°C; (e) Amberlite IRA-400 (Cl), MeOH.

We designed a series of DLCs related to the original [0, 0] rhodacyanines. In this paper, the rhodacyanines synthesized by our group were categorized into five classes in the following manner: [0, 0], [1, 0], and [0, 0, 0] rhodacyanines (classes I to III, respectively),2528) and [0, 0] and [0, 0, 0] azarhodacyanines (classes IV, V, respectively).29) Dyes in classes I, II, and IV were distinguished by the length of the methine linkage between the heterocyclic A-rings and the rhodanine B-rings. In classes I and IV, the A-ring was directly conjugated to the rhodanine through a double bond, and in class II, the A-ring was connected to the rhodanine through a bridge comprising two methine carbons. The classes III and V [0, 0, 0] (aza)rhodacyanines were unique in that they possessed two linked rhodanine moieties in the center of the skeleton. Aza-rhodacyanines (classes IV, V) possessed an imine function instead of an olefinic cyanine tether in the original rhodacyanines.

The synthetic routes to the classes II–V rhodacyanines are shown in Chart 2. The [1, 0] rhodacyanines 16 were synthesized using a modification of the procedure for synthesizing the [0, 0] rhodacyanines 7. The reaction of rhodanine 10 with N,N-diphenylformamidine, followed by acetylation, gave acetamide 14. The subsequent condensation of 14 with the benzothiazolium salt 9 furnished merocyanine 15, a vinylog of merocyanine 11. After activation of 15 through S-methylation, condensation with the picolinium salt 13 afforded 16. A possible method for synthesizing the class III [0, 0, 0] rhodacyanines 20 involves the condensation of a cyanine derivative in which the right half segment consists of two heterocycles with a merocyanine derivative (as the left half segment). The cyanine unit 19 was prepared from the thioamide 18, which was prepared from ethyl pyridylacetate (17) via transamidation and successive thioamidation. The condensation of 19 with 12 yielded the [0, 0, 0] rhodacyanine 20. The azarhodacyanines 22 (class IV) and 26 (class V) were prepared by the same procedure used to prepare 7 and 20, respectively, using an aminothiazolium moiety instead of a picolinium moiety.

Chart 2. Typical Synthetic Routes to [1, 0] and [0, 0, 0] Rhodacyanines 16a (Class II), 20b (Class III), and [0, 0] and [0, 0, 0] Azarhodacyanines 22a (Class IV) and 26a (Class V)

Reagents and conditions: (a) PhN=CHNHPh, MeCN, 70°C; (b) Ac2O, NEt3, 110°C; (c) NEt3, 9a, MeCN, 60°C; (d) TsOMe, DMF, 110°C; (e) NEt3, 1-methylpicolinium p-toluenesulfonate (13b), MeCN, 75°C; (f) EtNH2, MeOH, −78°C to r.t.; (g) Lawesson’s reagent, toluene, 115°C; (h) BrCH2COCl, CH2Cl2, r.t.; (i) MeI, acetone, 60°C; (j) 12b, NEt3, MeCN, 70°C; (k) NEt3, MeCN, 70°C; then Amberlite IRA-400 (Cl), MeOH; (l) EtNCS, benzene, 80°C; (m) ClCH2CO2H, NaOAc, EtOH, 100°C; (n) MeI, MeCN, 70°C; (o) 12c, NEt3, MeCN, 70°C; then Amberlite IRA-400 (Cl), MeOH.

3.2 In Vitro Antimalarial Activity of Classes I–V Rhodacyanines

At the outset of the structure–activity relationship study, the requirements for each heterocyclic component and the cationic character of the rhodacyanine dyes were evaluated (Table 2). The merocyanines 11a and 12a clearly displayed very weak antimalarial activity (IC50 values of less than 10−6 M, entries 2, 3). The cyanines 19b and 27, which did not include a merocyanine moiety, displayed moderate antimalarial activities, but their potencies were lower than that of MKT-077 (7a) (entries 4, 5). The antimalarial activity of the neutral analog 2830) was much lower than that of the cationic salt 7a (entry 6). These results clearly indicate that the DLC structure, in particular the cationic cyanine moiety, was essential for antimalarial activity, and the incorporation of the merocyanine unit was required to achieve a high activity.25)

Table 2. Effect of Skeletal Composition on Antimalarial Activities
EntryCmpd (R1, R2, R3)SkeletonIC50 (M) P. falciparuma)
17a (Me, Et, Et)A-B-C+7.0×10−8
211a (Me, Et, –)A-B>2.8×10−5b)
312a (Me, Et, –)A-B+3.7×10−6
419bB-C+5.2×10−7
527A-C+2.4×10−7
628A-B-C1.7×10−6

a) Chloroquine sensitive strain (FCR-3). b) IC32 Value (68% growth of P. falciparum was observed.).

The in vitro antimalarial activities of more than 200 different rhodacyanines (classes I–V; selected compounds are shown in Fig. 3) were evaluated against chloroquine (CQ)-sensitive and -resistant P. falciparum according to the reported protocols.31,32) Selected results are summarized in Table 3. The class I [0, 0] rhodacyanines showed promising in vitro antimalarial activities with IC50 values on the order of 10−7 to 10−9 M order and low cytotoxicity levels. The inhibitory activities of the [0, 0] rhodacyanines against the K-1 strain of P. falciparum (CQ-resistant) paralleled the corresponding activities against the CQ-sensitive FCR-3 strain. The incorporation of hydrophilic substituents decreased the antimalarial activity (entries 1, 2, 6 vs. 3–5). Compounds 7ce, which bore hydrophilic substituents, displayed slightly lower activities. Compounds 7g and j, which were more lipophilic than 7a, showed stronger antimalarial activities but high cytotoxicities (entries 7, 9).25,26)

Fig. 3. Representative Rhodacyanines and Azarhodacyanines Synthesized by Our Group
Table 3. Antimalarial Efficacies of Five Classes of Rhodacyanines in Vitro
EntryClassCmpdIC50 (M) against P. falciparumIC50 (M) Cytotoxicityc)
FCR-3a)K-1b)
1I7a7.0×10−82.1×10−81.1×10−4
2I7b2.6×10−8NTe)1.6×10−5f)
3I7c2.1×10−78.2×10−81.8×10−4
4I7dNTe)3.7×10−86.2×10−5
5I7e6.8×10−7NTe)>1.8×10−5
6I7f1.2×10−81.9×10−81.1×10−4
7I7g7.8×10−9NTe)6.8×10−7f)
8I7iNTe)2.7×10−7>1.7×10−4
9I7jNTe)1.0×10−91.4×10−5
10II16aNTe)1.3×10−77.6×10−5
11II16bNTe)1.2×10−83.0×10−5
12III20a3.2×10−99.7×10−94.4×10−7f)
13III20bNTe)7.1×10−77.8×10−5
14IV22aNTe)5.9×10−91.2×10−4
15IV22bNTe)4.4×10−91.1×10−5
16V26cNTe)2.2×10−81.4×10−5
17V26dNTe)1.0×10−94.6×10−5
18CQd)1.8×10−81.5×10−64.7×10−5

a) Chloroquine sensitive strain (FCR-3). b) Chloroquine resistant strain (K-1). c) Rat skeletal myoblast L-6 cells representing a model of host except for entries 2, 7 and 12. d) CQ means chloroquine. e) NT means not tested. f) Mouse mammary tumor FM3A cells.

The antimalarial activities of the class II [1, 0] rhodacyanines 16a were slightly lower, but their cytotoxicities were a factor of 10 times than those of the [0, 0] rhodacyanines (entries 10, 11). The class III [0, 0, 0] rhodacyanine 20a showed a higher antimalarial activity than 7a, possibly due to the higher lipophilicity of 20a (entry 12).28) The classes IV and V azarhodacyanines 22 and 26 exhibited strong activity (10−9 to 10−8 M) against P. falciparum with low cytotoxicities (entries 14–17).29) All classes of rhodacyanines possessed promising in vitro activities, and several showed strong activities against the CQ-resistant P. falciparum with low cytotoxicities. No obvious correlation was observed between the antimalarial efficacy and the bulkiness or length of the substituents; however, balance between the molecular hydrophilicity and hydrophobicity was important for efficacy.

3.3 In Vivo Antimalarial Potency of Rhodacyanines in Mouse Model

The high antimalarial activities and low cell cytotoxicities of several rhodacyanines were explored by conducting in vivo studies of a rodent malaria model of P. berghei NK-65 (drug-sensitive strain) in mice. The in vivo assays were performed according to Peters’ 4-d suppressive test protocol.33,34) The malaria-affected mice were followed after the end of treatment to record the mean survival in days (MSD).

The in vivo results are summarized in Table 4. All classes of the rhodacyanines tested suppressed the growth of the malaria parasites; however, despite the strong in vitro activities (1.0–21 nM against the K-1 strain), the parasite growth suppression levels of the [0, 0] rhodacyanines 7a and f (10 mg/kg/d, intraperitoneally (i.p.)) were only around 30% (entries 1, 5). Mouse death prior to the end of the drug treatment protocol was observed at higher doses. By contrast, 7c and d, which bore a hydrophilic R3 substituent moiety, showed moderate parasitemia suppression (around 60%) upon intraperitoneal or intravenous administration (5–10 mg/kg/d), although their activities in vitro were lower than the activity of 7a (see Table 3, entries 1–6). The trend in the hydrophilicity/hydrophobicity balance of 7 was opposite to that observed in the in vitro testing results. The difference may have resulted from several factors that influenced the in vivo activities, including the absorption, metabolism, distribution, and excretion (ADME).28)

Table 4. Antimalarial Efficacies of Rhodacyanines in Vivoa)
EntryClassCmpdFour days suppressive testMSDc)
Dose, methodb)% Suppression
1I7a10 mg/kg/d, i.p.30.0NDd)
2I7c5 mg/kg/d, i.v.65.5NDd)
3I7d5 mg/kg/d, i.p.60.9NDd)
4I7d10 mg/kg/d, i.p.62.6NDd)
5I7f10 mg/kg/d, i.p.27.0NDd)
6II16a10 mg/kg/d, i.p.12.6NDd)
7III20a20 mg/kg/d, i.p.28.7NDd)
8III20b10 mg/kg/d, i.p.64.0NDd)
9III20b50 mg/kg/d, i.p.68.9NDd)
10III20b100 mg/kg/d, p.o.55.0NDd)
11III20c25 mg/kg/d, i.p.88.7NDd)
12III20d25 mg/kg/d, i.p.89.0NDd)
13IV22a10 mg/kg/d, i.p.64.5NDd)
14IV22b10 mg/kg/d, i.p.54.6NDd)
15V26a25 mg/kg/d, i.p.78.45.6
16V26b25 mg/kg/d, i.p.96.422.0
17V26c25 mg/kg/d, i.p.97.125.3
18V26c100 mg/kg/d, p.o.41.76.5
19V26d25 mg/kg/d, i.p.95.718.0
20CQe)10 mg/kg/d, i.p.90.622.7

a) In vivo evaluation was carried out according as Peters’ 4-d suppressive protocol using five ICR-mice affected by P. berghei NK-65 (drug-sensitive strain). b) i.p.=intraperitoneal administration; i.v.=intravenous administration; p.o.=per os administration. c) MSD means survival days. MSD for untreated mice (control) is 5.5 d. d) ND means not determined. Most of treated mice survive only 1–2 d longer than untreated controls. e) CQ means chloroquine.

An in vivo evaluation of the class II [1, 0] rhodacyanines 16 revealed low efficacies against P. berghei (<25% parasitemia suppression) at a dosage of 10 mg/kg/d i.p. (entry 6). By contrast, and with the exception of 20a, the class III [0, 0, 0] rhodacyanines exhibited suppressed activities, similar to the [0, 0] rhodacyanines (entries 8–12). Unlike the classes I and II rhodacyanines, a lower toxicity was observed at a given dose or even at higher dosages. The properties of 20d were noteworthy. 20d includes a benzothiazole ring on the right side and displayed a high in vivo activity, providing 89% suppression at dosages of 25 mg/kg/d (entry 12). These activities were comparable to those of chloroquine (entry 20). 20b yielded a 55% suppression at a dosage of 100 mg/kg/d by oral administration; however, several signs of acute toxicity, such as diarrhea and body weight loss, were observed.28)

The classes IV and V rhodacyanines, characterized by the introduction of a nitrogen atom into the cyanine conjugation, surpassed our expectations and displayed promising antimalarial efficacy in vivo. The [0, 0] azarhodacyanines 22a and b exhibited much better parasitemia suppression (ca. 60%) than the class I [0, 0] rhodacyanines (ca. 30%) at the same dose level (10 mg/kg/d, i.p.) (entries 13, 14 vs. 1, 5). Further investigations clarified that the class V compounds 26 afforded the best in vivo results with better activities and lower toxicities. All compounds could be injected at a dose of 25 mg/kg/d (i.p.) and displayed moderate to excellent suppression levels at the dose. Compound 26c resulted in a 97% suppression (entry 17). The mice treated with 26c survived for 25 d, on average.29)

4. Improvements in the Antimalarial Activities of the Natural Alkaloids Using the DLC Hypothesis35,36)

The β-carboline alkaloids are widely found in a number of plants. Some of these alkaloids exhibit a variety of biological and pharmaceutical potencies including antimalarial activity3740) (Fig. 4). Pavanand et al. isolated 4-methoxy-1-vinyl-β-carboline (MVC, 29a)41) and related compounds as antimalarial components from Picrasma javanica, a medicinal plant used for the treatment of malaria42); however, the activities of these compounds were not strong. We envisioned that the quaternary β-carbolinium salts, which were expected to have a DLC character, could provide greater antimalarial effects than the neutral carbolines.

Fig. 4. β-Carboline Alkaloids Displaying Antimalarial Activities

We accomplished the total synthesis of MVC (29a) in addition to several related β-carbolines using Cook’s procedure for the synthesis of crenatine, a structurally related alkaloid, with some modifications4345) (Chart 3). The preparation of the β-carbolinium salts 30, which corresponded to the DLCs, was achieved by a simple quaternization with the alkyl tosylate or alkyl halide.

Chart 3. Synthesis of MVC and Its Ammonium Salt

Reagents and conditions: (a) EtO2CCHO, EtOH; (b) AcCl, DMAP (cat.), Et3N, CH2Cl2; (c) DDQ, THF–H2O, −78°C to r.t.; (d) Me2C(OMe)2, TsOH, benzene; then p-chloranil, r.t.; (e) DIBAL, CH2Cl2; (f) Ph3P=CH2, THF; (g) TsOEt, MeCN, 80°C.

The in vitro antimalarial and cytotoxic results are summarized in Table 5. The electronically neutral β-carbolines 29ad displayed weak to medium inhibitory effects against P. falciparum (10−5 M), and the cytotoxicities were comparable to their antimalarial activity levels (entries 1–4). By contrast, the N-alkyl carbolinium salts 30ad and the tetracyclic cation 30e displayed considerably enhanced antimalarial activities (entries 5–10). Quaternization of the pyridine nitrogen atom of 29 resulted in a 38-fold increase in the antimalarial activity (entry 1 vs. 5). In addition, the cytotoxicity level decreased by a factor of 3 during the transformation. A similar enhancement in the antimalarial activity was observed upon the transformation of the carbolines 29 into the corresponding carbolinium salts 30 (entry 2 vs. 6, 7, 3 vs. 8, 4 vs. 9). 30d, in particular, displayed a 67-fold increase in the antimalarial activity compared to the corresponding 29d. These results indicated that the DLC compounds provided greater antimalarial potencies within the class of the β-carbolines via a simple transformation. By contrast, quaternarization of dihydro-β-carboline had no effect on the biological activity. Both the electronically neutral molecule 35 and the cationic salt 36 displayed similarly low activities (entries 11, 12), suggesting that the broad π-delocalization of the positive charge on the expanded heteroaromatic ring should be important for improving antimalarial activity. This study demonstrated that the transformation into a DLC structure could prove to be a highly effective modified method for preparing and designing antimalarial compounds.35,36)

Table 5. Antimalarial Efficacies of β-Carbolinium Salts and Related Compounds in Vitro
EntryCmpdIC50 (M)Selective indexc)
P. falciparuma)FM3Ab)
129a5.0×10−63.8×10−60.76
229b2.2×10−51.8×10−50.82
329c2.2×10−51.8×10−50.82
429d3.1×10−53.2×10−51.0
530a1.3×10−71.0×10−577
630b3.7×10−73.0×10−581
730b3.6×10−7d)3.0×10−583
830c1.1×10−61.9×10−517
930d4.6×10−72.2×10−548
1030e3.5×10−7d)4.9×10−6e)14
11351.8×10−54.5×10−52.1
12361.6×10−5>3.8×10−5>2.4

a) Chloroquine sensitive strain (FCR-3) except for entries 7 and 10. b) Mouse mammary tumor FM3A cells representing a model of host except for entry 10. c) Selective index means IC50 value for FM3A (or L-6)/IC50 for P. falciparum. d) Chloroquine resistant strain (K-1). e) Rat skeletal myoblast L-6 cells.

5. Antimalarial Activities of the Phenoxazine Dyes46)

The antimalarial activities of heteroaromatic dyes comprising a π-conjugated cation have been investigated sporadically4749) since Guttmann and Ehrlich reported the antimalarial properties of methylene blue in 1891.50,51) Vennerstrom and co-workers reexamined the in vitro antimalarial potencies of xanthene, azine, oxazine, and thiazine dyes and reported that several of these dyes displayed strong in vitro activity against drug resistant P. falciparum.47) In connection with our working DLC hypothesis, we examined the antimalarial potencies of several readily available dyes. The phenoxazinium salts 37 were found to display promising antimalarial potencies.

Several phenoxazinium salts 37 were synthesized, as shown in Chart 4. Compounds 37ae were prepared by the reaction of m-aminophenol 38 with p-nitrosoanilline 39 in the presence of acid, followed by auto-oxidation in air52) and, if necessary, an ion-exchange process. As an alternative route, 37 may be prepared by a two-step sequence from m-anisidine 40.53) The overall chemical yield of 37a was higher in the latter method.

Chart 4. Synthesis of Phenoxazinium Salts 37

Reagents and conditions: (a) HClO4, EtOH, air, then ion exchange; (b) NaNO2, HCl; (c) 38, HClO4, EtOH, air; then ion exchange.

The in vitro and in vivo antimalarial activities against the P. falciparum K-1 strain and P. berghei in mice, respectively, are summarized in Table 6. 37a (Basic Blue 3)54) and 37c showed strong antimalarial activity and excellent selective indices in vitro (entries 1, 7). The trend observed among the rhodacyanines was observed upon the installation of an additional hydrophilic substituent on the heteroaromatic skeleton. The activities of piperazine containing the phenoxaziniums 37d and e were not strong (entries 8, 11).

Table 6. In Vitro and in Vivo Antimalarial Activities of Phenoxazinium Salts
EntryCmpdIn vitroIn vivo
IC50 (M) P. falciparuma)Selective indexb)Dose (mg/kg/d), methodc,d)% SuppressionMSDe)
137af)2.8×10−924005.0, i.p.46.3NDg)
220, i.p.95.8>14
325, p.o.82.37.6
490, p.o.100>24
537b2.8×10−922005.0, i.p.85.310
620, i.p.100>17
737c2.4×10−93505.0, i.p.83.116
837d-HCl2.8×10−823005.0, i.p.60.48.6
920, i.p.98.118
10100, p.o.10017
1137e-HCl2.0×10−74405.0, i.p.17.45.3

a) Chloroquine resistant strain (K-1). b) Selective index means IC50 value for L-6/IC50 for P. falciparum. c) In vivo evaluation was carried out according as Peters’ 4-d suppressive protocol using five ICR-mice affected by P. berghei NK-65 (drug-sensitive strain). d) i.p.=intraperitoneal administration; p.o.=per os administration. e) MSD means survival days. MSD for untreated mice (control) is 5.5 d. f) Synthetic 37a (>95% purity) was used for in vitro assay, and purchased one (ca. 60% purity) was used for in vivo evaluation. g) ND means not determined.

The phenoxaziniums 37ad,54) the R3 substituent of which was the hydrogen atom, provided good to excellent parasitemia suppression (46–85%) at a dose of 5 mg/kg/d by i.p. administration (entries 1, 5, 7, 8). Significant improvements in parasitemia suppression were observed at the higher doses of 37a, b, and d (entries 2, 6, 9). Notably, the mean survival days for the mice treated with 37b and c were significantly prolonged compared with the untreated mice. The introduction of a hydroxy group as the R3 substituent led to a significant loss in suppression and survival effects (entry 11). Unexpectedly, the phenoxazimium salts were orally bioavailable. As shown in the entries 4 and 10, 37a and d provided 100% clearance of parasitemia on day 4 at a dose of 90 and 100 mg/kg×4 d (p.o.), respectively. Preliminary toxic tests indicated that the 50% lethal doses (LD50) of both compounds exceeded 200 mg/kg/d (p.o.).

6. Chemical Biology Approach to Understanding the Modes of Action of the Rhodacyanines55)

Some of the DLCs, including the rhodacyanines 7, showed promising in vitro and in vivo antimalarial activities. The mode of action by which the DLCs functioned was unclear. Rhodacyanine itself showed weak fluorescence upon visible light irradiation. The fluorescent distributions of several rhodacyanines and their analogues57) throughout malaria-affected erythrocytes were examined using P. berghei. 7a (IC50=21 nM against P. falciparum) selectively localized in specific subcellular sites (organelles) of the parasites (Figs. 5a, b).27 The rhodacyanines that displayed strong antimalarial activities, such as 26b (IC50=22 nM) and 7i (IC50=78 nM), also exhibited specific localization. Their fluorescence appeared as localized spots (Figs. 5c–f). By contrast, the less active compound 7e (IC50=680 nM) did not show specific localization within organelles, but rather it leached away to the parasitic cytoplasm (Figs. 5g, h), although it remained in the plasmodial cells (no fluorescence was observed in the erythrocytic cytoplasm). No specific accumulation was observed for the much less active compounds 12a (IC50=4000 nM) (Figs. 5i, j). This observation indicated the presence of a good correlation between the antimalarial activity and site-specific accumulation.

Fig. 5. Antimalarial Activity-Subcellular Accumulation Relationship of 7a (a, b), 26b (c, d), 7i (e, f), 7e (g, h) and 12a (i, j) in P. bergheia)

a) The final concentration of the rhodacyanines was 5.0×10−6 M. Bright-field images (a, c, e, g, i) and fluorescent images through FITC filter (b, d, f, h, j), b) IC50 values against P. falciparum K-1 strain except for 7e, c) IC50 against P. falciparum FCR.

The weak fluorescent properties of 7a prevented further investigations into the plasmodial subcellular sites into which the rhodacyanines selectively accumulated. A better understanding of the mechanism of action was explored by synthesizing probes with stronger fluorescence. We envisaged that fixation of the flexible bonds of rhodacyanines should improve the fluorescence quantum yield by suppressing energy loss from the excited state to the ground state via vibrational transitions. We designed the novel fused rhodacyanines 42 and 43, which feature fused ring skeletons (Fig. 6). Compounds 42 and 43 displayed in vitro antimalarial activities comparable to the activities of the original 7a. Importantly, the fluorescence intensity was 70 times greater than that of 7a, and a significant red shift in the fluorescent emission was observed (excitation: λex=495 nm). The fluorescence localization of 43 among the parasitic organelles could be clearly detected, even upon treatment with a 100-fold smaller quantity of 43 (final concentration=5×10−8 M) compared to rhodacyanine 7a.

Fig. 6. Newly Designed Fused Rhodacyanines as Fluorescent Probes

Double staining experiments of the P. berghei-infected erythrocytes co-incubated with 43 and selective fluorescent markers of subcellular organelles were performed. Markers of the nucleus (DAPI) and mitochondria (Mitotracker Red CMXRos®) were used. This study indicated that rhodacyanine 43 and DAPI selectively accumulated among different organelles (Figs. 7a–c). By contrast, fluorescence localization of 43 was consistent with that of CMXRos® (Figs. 7d–f). We concluded that the rhodacyanines selectively accumulated within the plasmodial mitochondria. The selective uptake of rhodacyanines in mitochondria was consistent with the DLC hypothesis,20) our starting point for drug design. The uptake of the rhodacyanines in mitochondria may play a key role in inhibiting malaria diseases.

Fig. 7. Fluorescent Microscopic Images of the Intracellular Distribution of 43 with DAPI (a–c) and with CMXRos (d–f) in P. berghei Infected Erythrocytesa)

a) The final concentration of 43 was 5.0×10−8 M. (a, d) Through FITC filter; 43 (green spot). (b) Through DAPI filter, parasitic nucleus stained by DAPI. (e) Through Y7 filter, parasitic mitochondria (MT) stained by CMXRos (red spots). (c, f) Superimposed.

7. Antitrypanosomal and Antileishmanial Activities of the Synthetic DLCs36,5661)

The causal pathogenic parasites of NTDs, such as Trypanosoma spp. and Leishmania spp., belong to the Trypanosomatidae family in the Euglenozoa phylum. Their morphologies and life cycles are related to those of Plasmodium spp., which belongs to a different phylum (Apicomplexa). The protozoa of these two phyla bear a single mitochondrion that develops through analogous pathways, although certain aspects differ. The mitochondria are characterized by certain features that are absent in other eukaryotic organisms. Several drugs that interfere directly with mitochondrial physiology and ultrastructure in Leishmania, Trypanosoma, and Plasmodium have been reported.6264)

The anti-American-trypanosomal, anti-African-trypanosomal, and antileishmanial activities of the aforementioned classes of DLCs against T. cruzi, T. brucei, and L. donovani, respectively, were evaluated6567) and the results are summarized in Table 7. With some exceptions, almost all DLCs displayed broad-spectrum antiprotozoal activities. Several compounds showed activities comparable to that of a clinically available antiparasitic drug. It is noteworthy that the class II [1, 0] rhodacyanines 16 showed strong antileishmanial activities in vitro.

Table 7. In Vitro Antiprotozoal Activities of DLCs
EntryCmpdIC50 (M) in vitro
P. falciparuma)T. cruzic)T. bruceib)L. donovanid)L. majore)L-6f)
17a2.1×10−86.8×10−64.9×10−74.6×10−74.6×10−71.1×10−4
27f1.2×10−84.3×10−64.1×10−73.2×10−73.2×10−71.1×10−4
37j1.0×10−92.2×10−72.4×10−82.1×10−8NTg)1.4×10−5
416b1.1×10−85.1×10−61.6×10−88.3×10−9NTg)3.0×10−5
516c5.0×10−89.0×10−83.0×10−88.0×10−9NTg)1.1×10−6
616d3.8×10−82.2×10−59.3×10−81.1×10−8NTg)5.3×10−5
720a9.7×10−91.4×10−59.0×10−96.1×10−86.1×10−8>1.4×10−4
822a5.9×10−92.2×10−55.4×10−71.9×10−7NTg)1.2×10−4
926b1.2×10−87.4×10−76.3×10−81.1×10−7NTg)7.5×10−6
1026c6.0×10−9NTg)1.0×10−72.5×10−7NTg)2.7×10−5
1130e3.5×10−74.5×10−76.3×10−72.6×10−7NTg)4.9×10−6
1237b2.1×10−92.1×10−75.0×10−75.0×10−7NTg)4.7×10−6
1337c2.4×10−92.4×10−86.7×10−82.3×10−8NTg)8.4×10−7
14Control1.5×10−7h)8.7×10−7j)6.0×10−9i)2.8×10−7k)1.4×10−7l)

a) P. falciparum chloroquine resistant strain (K-1). b) T. cruzi (Tulahuen C2C4 strain). c) T. brucei rhodesiense (STIB900 strain). d) L. donovani (MHOM/67/L82 strain). e) L. major (MHOM/SU/5ASKH strain). f) Rat skeletal myoblast L-6 cells. g) NT means not tested. h) Chloroquine as a positive control. i) Benznidazole as a positive control. j) Melarsoprol as a positive control. k) Miltefosine as a positive control. l) Amphotericin B as a positive control.

Giargia lambria, which causes diarrhoeal giardiasis, is a flagellated protozoa closely related to Trypanosoma spp. and Leishmania spp. This protozoan has no mitochondrial organelles, unlike the other species. Interestingly, 7a and 30e were inactive against G. lamblia (IC50>1.0×10−4 M). These results supported the conclusion that the accumulation of the DLCs in the parasitic mitochondria played a key role in inhibiting the growth of parasites.

8. Conclusion

This review discussed the development of antiprotozoal agents based on the π-delocalized lipophilic cation (DLC) hypothesis. Several classes of DLCs, such as the rhodacyanines, azarhodacyanines, β-carbolinium salts, and phenoxazinium salts, were synthesized and evaluated for their antimalarial, antitrypanosomal, and antileishmanial activities in vitro. The DLCs showed strong antiprotozoal activities with high selective indices. The DLC structure was found to be required for achieving high activities.68) We demonstrated that some DLC compounds had promising in vivo antimalarial activities when administered orally. The antimalarial rhodacyanines were found to selectively accumulate in the erythrocytic plasmodial mitochondria, and the accumulation effects are expected to strongly influence the inhibition of plasmodial growth.

Acknowledgments

This research was carried out at Tohoku University. I am grateful to Prof. Masataka Ihara (Professor Emeritus at Tohoku University, Professor at Hoshi University), who provided the environment and opportunities for this research. I express my sincere appreciation to Prof. Yusuke Wataya, Prof. Hye-Sook Kim (Okayama University), Prof. Akira Naganuma, Prof. Yoshiteru Oshima, Prof. Hidetoshi Tokuyama (Tohoku University), Prof. Shusuke Kuge (Tohoku Medical and Pharmaceutical University), Prof. Shin-ichi Miyairi (Nihon University), Prof. Setsuko Sekita (Tokushima Bunri University, Kagawa), Dr. Tadao Shihido, Dr. Kozo Sato, Dr. Hiroshi Kitaguchi, Dr. Masayuki Kawakami (FUJIFILM Co.) and Prof. Reto Brun (Swiss Tropical Institute) for their kind help and useful suggestions. I also thank my students and colleagues, Dr. Khanitha Pudhom, Dr. Chalerm Saiin, Dr. Daiki Morisaki, Mr. Hiroshi Inoue, Mr. Hiroki Terauchi, Mr. Tsubasa Shimogama, Mr. Kazuki Kasai and Ms. Chie Satoh. This research was supported by Grants from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan, Japan Science and Technology Agency (JST), the Tokyo Biochemical Research Foundation, the Intelligent Cosmos Foundation, and the Takeda Science Foundation.

Conflict of Interest

The author declares no conflict of interest.

References and Notes
 
© 2016 The Pharmaceutical Society of Japan
feedback
Top