|Home | About | Journals | Submit | Contact Us | Français|
Metronidazole, the U.S. Food and Drug Administration-approved drug against trichomoniasis, is nonspermicidal and thus cannot offer pregnancy protection when used vaginally. Furthermore, increasing resistance of Trichomonas vaginalis to 5-nitro-imidazoles is a cause for serious concern. On the other hand, the vaginal spermicide nonoxynol-9 (N-9) does not protect against sexually transmitted diseases and HIV in clinical situations but may in fact increase their incidence due to its nonspecific, surfactant action. We therefore designed dually active, nonsurfactant molecules that were capable of killing Trichomonas vaginalis (both metronidazole-susceptible and -resistant strains) and irreversibly inactivating 100% human sperm at doses that were noncytotoxic to human cervical epithelial (HeLa) cells and vaginal microflora (lactobacilli) in vitro. Anaerobic energy metabolism, cell motility, and defense against reactive oxygen species, which are key to survival of both sperm and Trichomonas in the host after intravaginal inoculation, depend crucially on availability of free thiols. Consequently, molecules were designed with carbodithioic acid moiety as the major pharmacophore, and chemical variations were incorporated to provide high excess of reactive thiols for interacting with accessible thiols on sperm and Trichomonas. We report here the in vitro activities, structure-activity relationships, and safety profiles of these spermicidal antitrichomonas agents, the most promising of which was more effective than N-9 (the OTC spermicide) in inactivating human sperm and more efficacious than metronidazole in killing Trichomonas vaginalis (including metronidazole-resistant strain). It also significantly reduced the available free thiols on human sperm and inhibited the cytoadherence of Trichomonas on HeLa cells. Experimentally in vitro, the new compounds appeared to be safer than N-9 for vaginal use.
Sexually transmitted infections (STIs) and unintended pregnancies are the leading causes of morbidity and mortality among women of child bearing age (8). An estimated 340 million cases of curable STIs occur annually worldwide, with trichomoniasis having the highest incidence of ~51.2% (43). On the other hand, of 208 million pregnancies that occurred globally in 2008, ~41% were unintended (33). It has been seen that the primary responsibility of pregnancy and sexually transmitted disease (STD) protection lies with the female partner during most of the heterosexual contacts, including the “most vulnerable” contacts among adolescents and promiscuous adults (4). Often under such circumstances condom (the only dually protective contraceptive) is either inaccessible or its use is practically not negotiable. Consequently, woman-controlled, dually protective contraceptives that are safe, effective, and virtually imperceptible are highly desirable to curb the STD/HIV epidemic.
Trichomoniasis, the most-prevalent nonviral STD, predisposes women to viral STDs, including HIV (23). On the other hand, HIV-infected men with symptomatic trichomonal urethritis have nearly 6-fold-higher HIV concentrations in semen than men without urethritis. Therefore, it would be quite rational to conclude that controlling trichomoniasis alone can significantly reduce the incidence of new HIV infections. Moreover, trichomoniasis in women is also associated with vaginitis, endometritis, adnexitis, pyosalpinx, infertility, preterm delivery, low birth weight, bacterial vaginosis, and increased risk of cervical cancer (10). Antitrichomonas agents offering simultaneous pregnancy protection would have the advantage of providing the required impetus for use compliance since women prefer dual-protection contraceptives over pure microbicides (28). Metronidazole, the U.S. Food and Drug Administration-approved drug against trichomoniasis (5), is nonspermicidal and thus does not offer contraceptive protection when used vaginally. On the other hand, dually active molecules such as nonoxynol-9 (N-9) and cellulose sulfate failed to provide twin protection in clinical situations (13, 29) plausibly due to their general, nonspecific toxicity against sperm and cervicovaginal cells. Hence, there is a need for designed synthesis of potent, more specifically acting spermicidal microbicides for prophylactic contraception. Our enduring attempts to target human sperm through differently guided molecules against various functional moieties yielded some very promising structures (7, 12, 14–20, 22), but the two most promising spermicides (14, 16) were weakly microbicidal against Trichomonas vaginalis (15).
We therefore took up designing novel, dually capable molecules for killing human spermatozoa almost immediately after ejaculation in the vagina, as well T. vaginalis, the causative agent of trichomoniasis and a prominent HIV risk factor. Trichomoniasis, although curable, is often asymptomatic (especially in males) and therefore goes unreported, resulting in its persistent spread through heterosexual contacts. On the other hand, growing resistance of T. vaginalis to 5-nitroimidazole drugs (45) such as metronidazole is another serious concern and calls for the identification of newer antitrichomonal agents. Recently, some nonimidazoles have shown exceptional intravaginal efficacy against Trichomonas infection, including firmly metronidazole-resistant strains (36), but these are devoid of contraceptive activity.
Free thiol groups play an important role in the survival of predominantly anaerobic cells such as spermatozoa (40) and T. vaginalis (11). The strongly nucleophilic character of the sulfur atom and the unique redox properties of the thiol group make it a key residue for enzyme catalysis, protein folding, and redox signaling and regulation (42), which are important for the cellular energy metabolism, motility, and subsistence of sperm and Trichomonas. We hypothesized that a topical agent capable of targeting free thiols would arrest both sperm and Trichomonas in semen quite specifically since free thiols are unlikely to be available elsewhere in normal vaginal environment due to the low pH. Although agents independently targeting sperm and Trichomonas are available, in this major study we have attempted to combine the two intense capabilities together through rational drug design. We report here the in vitro activity and safety profile of some very potent spermicidal microbicides, the most promising of which were more effective than N-9 (the OTC spermicide) in killing sperm and more efficacious than metronidazole in killing susceptible and resistant T. vaginalis. Experimentally in vitro, these compounds appeared to be safer than N-9 for intravaginal use.
All of the chemicals, culture media, and other reagents were procured from Sigma-Aldrich (St. Louis, MO) unless stated otherwise. N-9 was purchased from Spectrum Chemical Manufacturing Corp. (New Brunswick, NJ). The 15 new compounds were synthesized according to the synthetic schemes provided as supporting information. The complete chemistry details of the new compounds, including characterization, purity, and instrumentation, are provided in the supplemental material.
Clinical isolates of metronidazole-susceptible T. vaginalis collected at Post Graduate Institute of Medical Research and Education, Chandigarh, India, were obtained from the laboratory of Divya Singh (37), and a metronidazole-resistant strain of T. vaginalis (CDC085 [ATCC 50143]) was procured from the American Type Culture Collection (ATCC). Both strains were cultured under partial anaerobic condition in TYM medium (0.1% K2HPO4, 0.06% KH2PO4, 0.5% NaCl, 0.5% glucose, 2.0% yeast extract, 0.2% l-lysine, 0.15% tryptone [pH 6.8]) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum, 2% vitamin mixture, 100 U of penicillin/ml, and a 100-μg/ml streptomycin solution in 15-ml screw-cap sterile tubes, followed by incubation at 37 ± 0.5°C. The vitamin mixture contained niacin (41.6 μg/ml), 4-aminobenzoic acid (83.3 μg/ml), pyridoxine hydrochloride (41.6 μg/ml), thymine hydrochloride (16.6 μg/ml), d-pentothenic acid (16.6 μg/ml), myoinositol (83.3 μg/ml), riboflavin (16.6 μg/ml), niacinamide (41.6 μg/ml), D-biotin (20.0 μg/ml), folic acid (20.0 μg/ml), calciferol (40 μg/ml), retinyl palmitate (40 μg/ml), vitamin K (menadione sodium bisulfite, 8.0 μg/ml), and α-tocopherol acetate (16.6 μg/ml) in MilliQ water. Organisms in the logarithmic phase of growth and exhibiting motility and normal morphology were harvested, centrifuged, and resuspended in fresh TYM medium for the experiments.
Freshly ejaculated human semen samples were obtained by masturbation from healthy and fertile volunteers, collected directly into sterile plastic tubes, and transported immediately into the laboratory. The samples were allowed to liquefy at 37°C for 45 min. Semen samples were analyzed according to World Health Organization guidelines (44). Sperm count and motility was analyzed manually, as well as in a Computer Automated Semen Analyzer (CASA; Hamilton Thorne) using a small drop of liquefied semen placed on a Makler counting chamber (Sefý Medica, Hafia, Israel) prewarmed to 37°C. Semen samples with a sperm count of >65 million per ml, >70% motility, and normal sperm morphology were used. This study was approved by the Institutional Ethics Committee.
The T. vaginalis parasites to be used in drug susceptibility assays were grown in TYM medium for 1 day following regular subculturing and were in the log phase of growth. In vitro drug susceptibility assays were carried out according to the standard procedure (38), and the metronidazole susceptibility criteria of Sobel et al. (34) was used to determine the resistance of T. vaginalis strains to metronidazole. Accordingly, the clinical isolate was categorized as susceptible, and the ATCC strain was categorized firmly resistant. Stock solutions (100 mM) of test compounds were prepared in dimethyl sulfoxide (DMSO). These stock solutions were diluted with TYM medium to obtain a concentration of 400 μM and then further serially diluted with the same medium to 1.0 μM in a 48-well plate. DMSO (0.05%) in TYM was used as the vehicle in control wells. Parasites (5 × 103 trophozoites/well) were added to these wells and incubated anaerobically at 37°C. Trophozoite growth and viability in drug-containing wells were monitored by trypan blue staining and cell number score on a daily basis, in comparison to the control. Assay results were clearly defined after 48 h in terms of the MIC (the lowest concentration of compound at which all trophozoites were nonviable). Using this MIC value as a reference, finer dilutions of the active compounds were prepared, and the drug susceptibility assay was repeated to obtain the final MIC precisely through a concentration versus viable-trophozoite-number curve. Viability was determined by trypan blue exclusion, and 100% eradication was confirmed by transferring 100 μl of the suspension to a 15-ml tube with fresh medium and recording the growth at 37°C for 14 days (1). Metronidazole (1 to 39 μM susceptible strain, 40 to 400 μM resistant strain; Sigma-Aldrich) was used as reference standard. Three separate experiments were performed for each strain to confirm the MIC.
The minimum effective (spermicidal) concentration (MEC) was determined by the modified Sander-Cramer assay as detailed earlier (12, 15). Briefly, the test compounds were dissolved in a minimum volume of DMSO and diluted with physiological saline (0.85% NaCl in distilled water) to make a 10.0 mM solution and serially diluted to 0.125 mM with saline. A spermicidal test was performed with each compound solution starting from 10 mM until the MEC was achieved. For this purpose, 0.05 ml of liquefied human semen was added to 0.25 ml of test solution and vortexed for 10 s at a low speed. A drop of the mixture was immediately placed on a microscope slide, covered with a cover glass, and immediately examined under a phase-contrast microscope in five fields of vision. The results were scored positive if 100% spermatozoa became immotile in ~20 s and remained immotile even after dilution with 1.0 ml of Krebs Ringer bicarbonate buffer for another 30 min at 37°C. The MEC was determined in three individual semen samples from different donors. The minimum concentration of compound capable of killing 100% sperm in ~30 s in all three semen samples was denoted as the MEC.
We used HeLa cell monolayers as an in vitro model of cervicovaginal epithelium (31) for testing the cytotoxicity of the new compounds. HeLa cells procured from National Centre for Cell Sciences, Pune, India, were grown in Dulbecco modified Eagle medium (DMEM; Sigma-Aldrich) supplemented with fetal bovine serum (10%), and antibiotics (a penicillin-streptomycin mixture [100 U/ml]). Cells at 80 to 90% confluence were split by trypsin (0.25% in phosphate-buffered saline [PBS]; pH 7.4), and the medium was changed at 24-h intervals. The cultures were maintained in a CO2 incubator at 37°C in a 5% CO2–95% air atmosphere. An MTT (3-[4,5-dimethyl thiazol-2-yl]-2,5-diphenyltetrazolium bromide)-based colorimetric assay for evaluation of the cytotoxicity of drug formulations against the human cervical cell line (HeLa) was used (6). Cells seeded at a density of 5 × 104 per well in 96-well plates were incubated in culture medium (DMEM with 10% fetal calf serum) for 24 h at 37°C in a 5% CO2–95% air atmosphere. After 24 h, the culture medium was replaced with fresh medium containing dilutions of test compounds in experimental wells and 0.05% DMSO in culture medium in control wells. After an incubation for another 24 h, 5 μl of MTT solution (5 mg/ml in PBS [pH 7.4]) was added to each well. The formazan crystals formed inside the viable cells were solubilized in DMSO, and the optical density at 540 nm (OD540) was recorded in a microplate reader (Microquant; BioTek).
Spores of Lactobacillus jensenii (ATCC 25258, strain 62G) were procured from the ATCC and grown in 6% Rogosa SL broth medium (Hi Media, India) containing 0.132% acetic acid at 37°C. The effect of test compounds on L. jensenii was determined by a method published earlier with slight modification (15). Briefly, Rogosa SL broth medium was prepared in MilliQ water, boiled for 2 to 3 min, and distributed in a 48-well plate (500 μl/well). Serial dilutions of test compounds were added to experimental wells, and vehicle was added to control wells in triplicate. Approximately 1,000 CFU of L. jensenii were inoculated into each well. The plates were incubated at 37°C in a humidified atmosphere containing 5% CO2 for 24 h. At the end of the experiment, the cultures were mixed thoroughly, 100 μl from each well was transferred to the corresponding well of a 96-well plate, and the numbers of lactobacilli were estimated by measuring the turbidity (OD610) in a microplate reader.
The effect of the most promising compound, compound 2, on T. vaginalis cytoadherence to human cervical (HeLa) cells was tested in an in vitro model of Trichomonas infection and was adapted from a published procedure (2) (for a listing of the compounds, see Table 1). Briefly, the parasites were labeled with tritium by incubation overnight in 1.0 μCi of [3H]thymidine/ml of complete TYM medium. After incubation, the parasites were thoroughly washed with normal TYM medium to remove unincorporated [3H]thymidine and then resuspended in interaction medium (TYM-DMEM [1:2]). To evaluate the cytoadherence, 106 [3H]thymidine-labeled trichomonads were allowed to interact with the monolayer of HeLa cells at a ratio of 5:1 for 3 h at 37°C in the presence of different concentrations (10, 25, 50, and 100 μg/ml) of compound 2. 3H-labeled trophozoites treated with 0.05% DMSO served as a control. After incubation for a period of 3 h, the cells were thoroughly washed with DMEM-TYM (2:1) for the removal of nonadherent T. vaginalis. HeLa cells were finally treated with trypsin (0.25% trypsin plus 1 mM EDTA) and pelleted, along with adherent Trichomonas, and suspended in aqueous scintillation counting fluid (ACS II; Amersham). Radioactivity was measured on an LS Analyzer 6500 (Beckman Instruments, Inc.).
Free thiols remaining on human sperm after treatment with new compound were measured using the colorimetric method with DTNB [5,5′-dithiobis(2-nitrobenzoic acid)] (9) by making suitable adaptations. Briefly, 1.0 ml of human semen was treated with 5.0 ml of the most promising, dually active compound (compound 2 [pyrrolidinium pyrrolidine-1-carbodithioate]) at the spermicidal MEC (stock solution diluted in saline) for 10 min. In parallel, 1.0 ml of semen from the same sample was treated with 5.0 ml of 0.15% DMSO in saline and used as a control. Thereafter, 1.0 ml of 100 mM β-mercaptoethanol was added to experimental and control tubes, followed by incubation at 4°C for 2 h to regenerate all of the unreacted thiols. After incubation, the sperm samples were pelleted in a refrigerated centrifuge, washed four to five times with PBS to completely remove the mercaptoethanol, and finally suspended in 1.0 ml of PBS. To various aliquots of this sample (finally measuring 120 μl), 5 μl of DTNB (100 mM) and 25 μl of Tris buffer (1.0 M; pH 8.0) were added, followed by mixing and incubation for 15 min. Thereafter, the multiwell plates were centrifuged, and 80 μl of supernatant was carefully aspirated and transferred into the corresponding wells of a new 96-well plate. The OD412 of the supernatant was measured in a microplate reader (Microquant) and corrected for a 1.0-cm light path using KC Junior software (Bio-Tek). A molar extinction coefficient of 14,150 cm−1 M−1 of DTNB (30) was used to calculate the number of free thiols.
Since the new molecules are intended for intravaginal use that normally has a low pH, a stock solution of the most promising compound (compound 2, 100 mM in DMSO) was serially diluted in a simulated vaginal fluid (pH 4.2) (27) up to 1.0 mM. After 60 min of incubation of this solution at 37°C, it was further diluted with TYM or PBS, and its antitrichomonal and spermicidal activities were determined. Given that the drug resistance of Trichomonas is better under aerobic growth conditions, the most active compound (compound 2) was also tested against trophozoites grown under aerobic conditions.
Each experiment was performed in triplicate and repeated three times. A semen sample from a different donor was used in every experiment. The data were analyzed by one-way analysis of variance, and P values of <0.05 were considered significant. The 50% inhibitory concentrations (IC50s) indicating cytotoxicity toward HeLa and Lactobacillus cells were calculated by computer-based curve fitting using the CompuSyn software.
Compared to standard drugs, 10 new molecules (compounds 1 to 3 and compounds 7 to 13) of the 15 compounds synthesized exhibited dual activity and 9 of these (compound 7 being the exception) were more effective than metronidazole (MET) in inhibiting the growth of MET-resistant Trichomonas strain (Table 1). Five compounds were found to be more active than nonoxynol-9 in spermicidal potential with an activity hierarchy as follows: compound 6 > compound 2 = compound 9 > compound 3 > compound 1 > N-9, while two compounds exceeded metronidazole in antitrichomonal activity (compound 2 > compound 3 > MET, susceptible strain). Three compounds appeared promising due to their superior dual activity compared to N-9 (compound 2 > compound 3 > compound 1), but the most promising compound (pyrrolidinium pyrrolidine-1-carbodithioate, compound 2) was ~1.6 times more potent than N-9 as a spermicide and killed MET-susceptible and -resistant strains of T. vaginalis approximately 15 and 6.5 times more efficiently than N-9 and approximately 2 and 29 times more effectively than metronidazole, respectively. Compound 2, after preincubation in simulated vaginal fluid (pH 4.2), became marginally more active against Trichomonas (both susceptible and resistant strains), whereas its spermicidal activity remained practically unchanged (Table 2). On the other hand, Trichomonas grown under aerobic conditions were moderately more resistant to both metronidazole and compound 2 (Table 2).
The 15 new compounds synthesized (compounds 1 to 15, Table 1) included secondary amine salts of dialkyldithiocarbamic acid (compounds 1 to 11), tertiary amine salts of dialkyldithiocarbamic acid (compounds 12 and 13), and tertiary amine salts of cycloalkyldithiocarbamic acid (compounds 14 and 15). Twelve compounds (compounds 1 to 5 and compounds 7 to 13) showed appreciable antitrichomonal activity against a MET-susceptible strain with MICs ranging from 4.8 to 243.3 μM (MET = 10.7 μM). Of the compounds with identical amine residues in both the dithiocarbamic acid and the ammonium portions (compounds 1 to 11), pyrrolidinium pyrrolidine-1-carbodithioate (compound 2) and piperidinium piperidine-1-carbodithioate (compound 3) were, respectively, 2.2 and 1.3 times more effective against susceptible Trichomonas than MET. Even while open-chain amine residues (compound 1) showed a marginal decrease in activity, a successive enlargement of cyclic ring of amine by one methylene group (compounds 2, 3, and 9) also exhibited a decreasing trend in activity. On the other hand, substitution of the methyl group (compounds 4 to 6) in the piperidine ring from position 4 to position 2 resulted in a remarkable enhancement in activity (MIC >400 to 30.41 μM), whereas a hydroxyl (compound 7) or a 1,4-dioxospiro group (compound 8) at position 4 of the piperidine ring retained the antitrichomonal activity at 37.47 and 20.14 μM, respectively. An additional hetero atom in the amine (compounds 10 and 11) resulted in compounds with activity comparable with MET. Alternatively, in compounds with different amine residues in dithiocarbamic acid and ammonium parts (compounds 12 to 15), a secondary-amine/tertiary-amine combination (compounds 12 and 13) gave compounds with activity comparable to MET, while a primary-amine/tertiary-amine arrangement (compounds 14 and 15) resulted in a total loss of activity. It seemed that an unsubstituted five or six membered secondary amine salts of dithiocarbamic acid were the desirable scaffolds for antitrichomonal activity.
Evaluation of all 15 new compounds against MET-resistant Trichomonas identified 11 compounds (compounds 1 to 5 and compounds 8 to 13) that exhibited very potent to moderate activity with MIC ranging from 11.94 to 332.13 μM (MET = 340.8 μM). However, when matched against N-9, four compounds (compounds 2 [6.3-fold], 8 [3.2-fold], 11 [1.8-fold], and 13 [1.9-fold]) were more potent while three compounds (compounds 1, 3, and 12) showed comparable efficacy. Among compounds with same amine residue in dithiocarbamic acid and ammonium portions, compound 2 was most effective, while compounds with ring enlargement (compounds 3 and 9), an acyclic residue (compound 1), or methyl substitution at position 2 or 3 in the piperidine ring (compounds 4 and 5) were marginally less active. A methyl (compound 6) or a hydroxyl (compound 7) group at position 4 of the piperidine ring resulted in a loss of activity, whereas a dioxospiro group (compound 8) increased the activity by 14.8- and 3.28-fold over MET and N-9, respectively. For the compounds with dissimilar amine residues in both portions (compounds 12 to 15), the results suggested the same pattern of structure-activity relationship (SAR) as seen with the susceptible strain.
The spermicidal activity of new compounds (compounds 1 to 15, Table 1) was evaluated in comparison to MET and N-9. Eleven compounds (compounds 1 to 3 and compounds 6 to 13) irreversibly immobilized 100% human sperm at concentrations 121.65 to 8,393.3 μM in ~30 s. MET showed no effect even at 50 mM, and N-9 exhibited a spermicidal MEC of 243.18 μM. Among secondary amine salts of dialkyldithiocarbamic acid (compounds 1 to 11), five compounds were found to be more potent spermicide than N-9 (compounds 6 [2-fold], 2 and 9 [1.6-fold], 3 [1.4-fold], and 1 [1.3-fold]). Introduction of a methyl substituent at position 4 (compound 6) of the piperidine ring increased the spermicidal activity in comparison to piperidine compound (compound 3), while that at position 2 (compound 4) or position 3 (compound 5) resulted in loss of activity. On the other hand, substituting a hydroxyl group (compound 7), 1,4-dioxaspiro group (compound 8), oxygen (compound 11), or N-methyl group (compound 10) at position 4 in piperidine led to mild spermicidal activity, suggesting that the presence of a methyl substituent at position 4 of the piperidine ring (compound 6) was more important for spermicidal activity. Nevertheless, spermicidal activity was retained either by increasing (compound 9) or decreasing (compound 2) the ring size or by incorporating an open-chain analogue (compound 1). Substitution of secondary amine either in ammonium ion (compounds 12 to 15) or in dithiocarbamate ion (compounds 14 and 15) by tertiary and primary amines reduced the activity. The overall results showed that a secondary amine residue with a methyl group at position 4 of the piperidine ring (compound 6) was the most desirable for spermicidal activity.
Based on the safety of the new compounds (compounds 1 to 15) toward HeLa cells, we have compounds with IC50s of >4,000 μM (compounds 10, 14, and 15) and compounds with IC50s in the range of approximately 1,500 to 2,650 μM (compounds 1 to 9 and compounds 11 to 13). Among the safest structures, only compound 10 exhibited appreciable antitrichomonal and spermicidal activities, which could be attributed to its better tendency (i.e., than compounds 14 and 15) to release thiocarbamic acid for interactions with sulfhydryl groups. Among the rest, the secondary amine salts of dialkyldithiocarbamic acid (compounds 1 to 9 and 11) exhibited enhanced safety toward HeLa cells by cyclization of carbon chain (compounds 1 and 2). Moreover, an increment in the ring size further improved safety (compounds 2, 3, and 9). On the other hand, a substitution in the six-membered cyclic framework compromised the safety (compounds 3, 4, and 6 to 8), whereas a methyl group at position 3 of the piperidine ring did not affect the safety profile (compounds 3 and 5). Among tertiary amine salts of morpholino-substituted dithiocarbamic acid, enhancement in the carbon chain of ammonium ion marginally forfeited safety (compounds 12 and 13).
However, all of the compounds (compounds 1 to 15) exhibited greater safety toward HeLa cells (IC50, 1,447 to >4,000 μM) and much better compatibility with Lactobacillus (IC50, 591.2 to >4,000 μM) than N-9 (IC50, 53.8 and 52.7 μM, respectively) and therefore appeared to be much safer for vaginal use.
The most promising compound (compound 2) significantly inhibited the cytoadherence of T. vaginalis on HeLa cell monolayers at a concentration of 10 to 100 μg/ml during a 3-h incubation period (Fig. 1). No significant change in the viability of Trichomonas was observed at concentrations of compound 2 of 10 to 50 μg/ml during this treatment period. On the other hand, when incubated with human sperm at the spermicidal MEC for 10 min, compound 2 significantly (P < 0.001) reduced the number of available free thiols on the sperm cell (Fig. 2).
The new, nonsurfactant molecules exerted a more potent and specific dual action compared to the general nonspecific effect of N-9, which killed both sperm and Trichomonas by its universal ability to destabilize cell membranes through surfactant action. The nonspecificity of N-9 action was also demonstrated by its potential to disrupt cervical epithelial (HeLa) cells and eliminate useful microflora (Lactobacillus) at its spermicidal and microbicidal concentrations. On the contrary, all of the new molecules, although designed to provide reactive chemical groups for inactivating vital thiols on target cells, did not universally inhibit both sperm and Trichomonas. They were very active against either sperm or Trichomonas or both, while a few were negligibly active against either cell type. Only three structures exhibited potent dual activity. This indicates that the promising new structures not only provide reactive groups for incapacitating thiols but also present a unique molecular design(s) capable of gaining access and targeting crucial thiols on both cell types. Both sperm and Trichomonas possess very well developed mechanisms for survival in the host; these mechanisms include cell motility, anaerobic energy metabolism, and defense against reactive oxygen species (ROS). Free thiol groups on the cell surface play an important role in the functional survival of human spermatozoa (40) and T. vaginalis (11). Both Trichomonas (39) and human sperm (25) are essentially anaerobes and depend on glycolysis for energy, which is also required for motility. The inhibition of glycolytic enzymes in T. vaginalis by thiol disruption is known to result in severe depletion of its intracellular ATP (35). On the other hand, ROS pose a serious threat to the survival of both sperm and Trichomonas, wherein thiols play an essential role in protection against the oxidative stress (3, 26). Although these cells do not essentially require oxygen for energy metabolism, they are exposed to oxygen in their natural environment inside the host. T. vaginalis lacks glutathione (the intracellular redox buffer), glutathione-dependent peroxidase, and catalase, and therefore it relies heavily on cysteine for protection against oxidative stress, which constitute >70% of cell's total thiol pool (42). The thioredoxin/thioredoxin-reductase system maintains the thiol status of Trichomonas for deactivating ROS (26). Thiol-inactivating agents are thus capable of seriously disrupting multiple fundamental processes in Trichomonas, as well as sperm. A very important feature of the new molecules is their extreme potency against metronidazole-resistant strains of Trichomonas. Metronidazole, which itself is nontrichomonacidal, gets activated in the hydrogenosomes of T. vaginalis by the anaerobic reduction of its nitro group, resulting in the formation of cytotoxic nitro radical-ion intermediates. This activation step is nonoperational in resistant pathogens (21). On the other hand, the most promising new molecule (compound 2) killed metronidazole-resistant Trichomonas almost 30 times more efficiently than metronidazole. This may indicate that a mechanism different than metronidazole is operational in the case of the new compounds, which makes them almost equally effective against both metronidazole-susceptible and resistant strains of the parasite.
Surface thiols are equally important for virulence of T. vaginalis, which requires the activity of surface cysteine proteinases to adhere to the host cell during infection (2, 24). Cysteine proteases, also known as thiol proteinases, have nucleophilic thiols in their catalytic domain for imparting proteolytic activity, which is involved in the cytotoxicity, hemolysis, and immune evasion (24) of T. vaginalis. Adherence of T. vaginalis to the epithelial cells of the urogenital tract is an essential step in pathogenesis. Since the new molecules were designed to suppress thiols' action on cell surface, we expected a marked inhibition of the virulence of T. vaginalis before cell death. In the in vitro model for Trichomonas infection of the HeLa cell monolayer, the cytoadherence capability of Trichomonas to the host cell surface was significantly inhibited by the most promising compound (pyrrolidinium pyrrolidine-1-carbodithioate [compound 2]) in the first 3-h incubation period, during which the viability of protozoa did not change significantly. This indicates that the new compounds can seriously impair the pathogenic potential of Trichomonas before killing the parasite.
The normal human vagina is naturally protected against STDs by its low pH, which is growth inhibitory for several pathogenic organisms. At this pH, free thiols are liable to be protonated to sulfenic acid (41), and therefore the likelihood of available free functional thiols in the normal vaginal environment is quite negligible. However, it has been seen that infections normally occur when the vaginal pH is disturbed, especially during the deposition of alkaline semen containing sperm and STD pathogen(s). The ability of new compounds to target thiols on sperm and Trichomonas at a seminal pH may add considerably to their activity and safety.
All of the new compounds synthesized were apparently much safer than N-9 toward human cervical epithelial (HeLa) cells and vaginal microflora (Lactobacillus). Since N-9 is a mixture of oligomers (32), its activity and toxicity depends on the ratio of these oligomers in the final preparation (46). The standard N-9 sourced from a chemical company for the present study exhibited better activity and a lower toxicity profile than the gift chemical used earlier (7); however, its lowest active concentration against either of the cell type (Trichomonas/sperm) was toxic enough to completely annihilate HeLa cells and lactobacilli in vitro. On the other hand, the most promising new compound, compound 2, killed sperm and Trichomonas at concentrations that were about 15 and 450 times lower than its IC50 against HeLa cells and about 8.5 and 280 times lower than its IC50 against Lactobacillus. This clearly indicates a highly specific action and a much better safety index for the new compound compared to N-9 and warrants further investigation for the development of a safe microbicidal contraceptive.
This study was supported by a grant from the Ministry of Health and Family Welfare, Government of India. We thank the Indian Council of Medical Research (A.J., V.V., R.K., and V.S.), the University Grants Commission (N.L., L.K., and L.K.), and the Council of Scientific and Industrial Research (R.K.M. and A.S.), New Delhi, India, for research fellowships.
We thank the SAIF Division of CDRI for the spectral data and Tara Rawat and C. Yadav for their technical assistance.
†This is CDRI communication no. 8091.
‡Supplemental material for this article may be found at http://aac.asm.org/.
Published ahead of print on 27 June 2011.