Anti-Toxoplasma gondii Activities and Structure-Activity Relationships of Novel Fluoroquinolones Related to Trovafloxacin

Abstract

Eleven novel fluoroquinolones closely related to trovafloxacin were evaluated for their in vitro activity against Toxoplasma gondii, and their structure-activity relationships were examined. The 50% inhibitory concentration (IC₅₀) of trovafloxacin against T. gondii was 2.93 μM; the IC₅₀ of the 11 analogs ranged from 0.53 to 14.09 μM. Six analogs had IC₅₀s lower than that of trovafloxacin. Examination of the structure-activity relationships of the compounds revealed that addition of a -CH₃ at C-5 of the 1,8-naphthyridone ring, at C-2 of the azabicyclohexane ring, or on the -NH₂ at the 6 position of the azabicyclohexane ring resulted in a four- to sixfold increase in activity. Moreover, replacement of 2,4-difluorophenyl by cyclopropyl at N-1 of the 1,8-naphthyridone ring increased activity twofold, and moving the -NH₂ one atom further away from the azabicyclohexane ring decreased activity. There was no difference between the naphthyridone and quinolone analogs. These results indicate that structure-activity studies of compounds related to drugs active against T. gondii may be useful in producing compounds with more potent activities against the parasite.

Trovafloxacin, a novel fluoronaphthyridone with a unique 6-amino-3-azabicyclo[3.1.0]hexyl side chain (Fig. 1), is active against Toxoplasma gondii both in vitro and in murine models of acute toxoplasmosis either when administered alone (6) or in combination with other drugs (5). We examined 11 structurally related analogs of trovafloxacin for their in vitro activity against T. gondii. In addition, we examined structure-activity relationships in an effort to determine if structural features can be identified that may help in understanding the antiparasitic activity of the fluoroquinolones and possibly lead to development of compounds with greater potency. The structure-activity relationship study was conducted because of previous observations (6) showing that tosufloxacin, a compound structurally very similar to trovafloxacin, was not active against T. gondii either in vitro or in vivo. Unlike trovafloxacin, which has a 3-azabicyclo[3.1.0]hexane substituent at position C-7 on the naphthyridone nucleus, tosufloxacin has a 3-aminopyrrolidine substituent. The loss in activity due to the single differing substituent in tosufloxacin indicated that derivatization at the C-7 position in the naphthyridone ring is critical for activity against T. gondii. Other compounds that had substituents other than a 3-azabicyclo[3.1.0]hexane at the C-7 position, such as a piperazine in ciprofloxacin or a methylpiperazine in fleroxacin, ofloxacin, and temafloxacin, did not have in vitro or in vivo activities against T. gondii in our earlier study (6). Therefore, it was considered of interest to evaluate compounds with substitutions for the 3-azabicyclo[3.1.0]hexane groups at C-7 and additional changes at N-1 and C-5 in either the naphthyridone or quinolone ring for in vitro activity against T. gondii.

FIG. 1.

Chemical structure of trovafloxacin.

Tachyzoites of the RH strain of T. gondii were obtained from the peritoneal cavities of Swiss Webster mice 2 days postinfection, as previously described (1). Human foreskin fibroblast (HFF) cells were grown in Dulbecco’s modified Eagle’s medium (Gibco BRL, Grand Island, N.Y.) containing 100 U of penicillin, 1 μg of streptomycin/ml, and 10% heat-inactivated bovine serum free of antibodies to T. gondii (HyClone Laboratories, Logan, Utah). Trovafloxacin (CP-99,219-27) and the 11 related analogs, 1 to 11, were obtained from Pfizer Inc., Groton, Conn.

In vitro studies.

Trovafloxacin and the analogs were examined at concentrations ranging from 0 to 60 μM. Each drug was dissolved in a small volume of dimethyl sulfoxide and brought to the required volume for testing with Dulbecco’s modified Eagle’s medium. All solutions were made and stored in the dark. Inhibition of intracellular replication of T. gondii was determined by the [³H]uracil incorporation technique (6). Briefly, confluent monolayers of HFF cells were infected with 6 × 10⁴ tachyzoites. Four hours later, the monolayers were washed to remove extracellular parasites and different concentrations of the drugs were added to triplicate wells. Addition of the drugs marked the starting time point. Incorporation of radioactivity was determined at 48 h by the addition of [³H]uracil (1 μCi/well) 4 h prior to harvesting the cells. The cells were collected with a cell harvester, and radioactivity was counted with a scintillation counter. Infected monolayers treated with medium that contained only 1% dimethyl sulfoxide served as controls. Fifty percent inhibitory concentrations (IC₅₀) were calculated by CalcuSyn, a dose-effect analysis computer program (Biosoft, Ferguson, Mo.).

Toxicity of the drugs for the HFF cells was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell proliferation assay (Cell Titer 96 kit; Promega Corp., Madison, Wis.) as previously described (6).

Results.

Each of the 11 analogs of trovafloxacin demonstrated potent in vitro activity against T. gondii. IC₅₀ ranged from 0.53 to 14.09 μM (Table 1). Six of the analogs had IC₅₀ lower than that of trovafloxacin, indicating that their in vitro activities against T. gondii were significantly higher than that of the parent compound. The activity against T. gondii of one representative analog, 1, and trovafloxacin as the control drug is shown in Fig. 2. Both analog 1 and trovafloxacin demonstrated toxicity for HFF cells (34 to 49% inhibition) only at the highest concentration (60 μM) (data not shown). However, analog 1 and trovafloxacin had parasite inhibition values of 98 and 99%, respectively, at a concentration that was fivefold lower than the concentration showing toxicity. Similar results were obtained with other analogs.

TABLE 1.

Activity of trovafloxacin and its analogs against T. gondii

No.	Compound	X	R1	R2          R3	IC50 (μM)a
1	1	N	H		0.53
2	2	N	H		0.53
3	3	N	CH3		0.79
4	4	N	H		1.68
5	5	N	H		2.16
6	6	C	H		2.65
7	CP-99,219-27 (trovafloxacin)	N	H		2.93
8	7	N	H		3.41
9	8	N	H		7.21
10	9	N	H		11.95
11	10	N	H		12.47
12	11	N	H		>4.97
					<14.09

^aFollowing a 48-h exposure.  

FIG. 2.

Effect of exposure to trovafloxacin and a representative analog (1) on the growth of intracellular tachyzoites of T. gondii in vitro measured by [³H]uracil incorporation.

Several structure-activity relationships were observed (Table 1). Thus, naphthyridone and quinolone analogs did not have a significant difference in activity (Fig. 3). Movement of the -NH₂ moiety further from the 6 position of the azabicyclohexane ring resulted in a two- to sevenfold decrease in activity (Fig. 4). Replacement of the 2,4-difluorophenyl moiety with a cyclopropyl at N-1 of the 1,8-naphthyridone ring resulted in a twofold increase in activity (Fig. 5). Interestingly, addition of a -CH₃ group at C-5 of the 1,8-naphthyridone ring, at C-2 of the azabicyclohexane ring, or on the -NH₂ group at the 6 position of the azabicyclohexane ring resulted in a four- to sixfold increase in activity compared to that of trovafloxacin (Fig. 6). Additionally, the positioning of the -NH₂ group on the cyclopropane ring also affected the activities of the compounds (Fig. 7).

FIG. 3.

Chemical structures of trovafloxacin and analog 6.

FIG. 4.

Chemical structures of trovafloxacin and analogs 11, 4, and 9.

FIG. 5.

Chemical structures of trovafloxacin and analog 4.

FIG. 6.

Chemical structures of analogs 1, 2, and 3.

FIG. 7.

Chemical structures of trovafloxacin and analogs 5 and 8.

Discussion.

The results described above reveal that each of the 11 novel fluoroquinolone compounds that are structurally related to trovafloxacin has potent in vitro activity against T. gondii. Six of these compounds had greater activity than the parent compound, trovafloxacin. Favorable pharmacokinetic parameters of trovafloxacin in humans, such as long half-life of elimination, rapid oral absorption, a linear increase in serum levels with increasing dose, and achievement of well-tolerated concentrations in serum (8), combined with excellent in vitro and in vivo activities against T. gondii (5, 6) make it an attractive candidate for further development for treatment of human toxoplasmosis. In addition, an understanding of changes in activity with structural modification may help the search for better drug candidates.

It has been reported that methyl substitutions on the piperazine ring at position C-7 in fluoroquinolones (noted in pefloxacin, ofloxacin, amifloxacin, and sparfloxacin) generally result in improved absorption and oral bioavailability (4). Since we also observed an enhancement in activity following methyl substitutions at different positions on either the naphthyridone ring or the C-7 substituent, it is possible that these modifications enhanced permeation of these drugs across membranes in our in vitro model and contributed to the observed higher activity. It has also been reported (4) that a cyclopropyl group at the N-1 position of the quinolone ring (noted in ciprofloxacin and sparfloxacin) enhances antibacterial potency. We observed similar results against T. gondii when the difluorophenyl in trovafloxacin was replaced with a cyclopropyl in analog 4.

Inhibition of the DNA gyrases or the topoisomerases as the mechanism of action of fluoroquinolone against bacteria has been reviewed by Drlica and Zhao (2). It has also been shown recently that the molecular structure of the quinolone determines the target preference between the DNA gyrase and topoisomerase IV in different bacteria (7). Fichera and Roos have reported that replication of the apicomplexan plastid (apicoplast) genome in T. gondii tachyzoites can be specifically inhibited by ciprofloxacin and that this inhibition blocks parasite replication (3). The mechanism of action of trovafloxacin against T. gondii is not known, and it may be a key factor in further studies of the structure-activity relationships to develop more potent fluoroquinolones against the organism.

In conclusion, our results indicate that structurally related analogs of trovafloxacin have potent in vitro antitoxoplasma activities. The structure-activity relationships that we observed may be useful in developing compounds with more potent activities against the parasite.