Synthesis and in vitro/in vivo Evaluation of the Antitrypanosomal Activity of 3-Bromoacivicin, a Potent CTP Synthetase Inhibitor

Abstract The first convenient synthesis of enantiomerically pure (αS,5S)-α-amino-3-bromo-4,5-dihydroisoxazol-5-yl acetic acid (3-bromoacivicin) is described. We demonstrate that 3-bromoacivicin is a CTP synthetase inhibitor three times as potent as its 3-chloro analogue, the natural antibiotic acivicin. Because CTP synthetase was suggested to be a potential drug target in African trypanosomes, the in vitro/in vivo antitrypanosomal activity of 3-bromoacivicin was assessed in comparison with acivicin. Beyond expectation, we observed a 12-fold enhancement in the in vitro antitrypanosomal activity, while toxicity against mammalian cells remained unaffected. Despite its good in vitro activity and selectivity, 3-bromoacivicin proved to be trypanostatic and failed to completely eradicate the infection when tested in vivo at its maximum tolerable dose.


Introduction
Human African trypanosomiasis (HAT) is caused by protozoan parasites of the Trypanosoma brucei subgroup. [1,2] Two subspecies are responsible for causing the disease in humans: T. b. gambiense and T. b. rhodesiense. T. b. gambiense is endemic in western and central Africa, causing a chronic form of the sickness, whereas T. b. rhodesiense is restricted to eastern and southern Africa, causing an acute illness that leads to death within a few weeks of infection. The trypanosomes are transmitted by tsetse flies to their mammalian hosts, where they first establish an infection in the blood and lymph. In the second stage of the disease, the parasites invade the central nervous system (CNS). Chemotherapy is currently the main way to control this disease, as there are no effective vaccines. However, currently available treatments suffer numerous drawbacks including poor efficacy, difficulties to administer, insecure supplies, cost, and an increasing threat of resistance. Therefore, there is a great need for new drugs, and efforts have been made to identify new molecular targets and innovative therapeutic tools.
Cytidine triphosphate synthetase (CTPS), a glutamine amidotransferase (GAT) responsible for the de novo synthesis of CTP, has been proposed and recently genetically validated as a potential drug target for the treatment of HAT. [3] In fact, despite CTP being essential for cell survival, T. brucei has low pools of CTP relative to mammalian cells and completely lacks the ability to salvage cytosine or cytidine. [3] Acivicin (Figure 1), an antibiotic isolated from the fermentation broths of Streptomyces sviceus, previously tested as an antitumor agent, has long been known to behave as an inhibitor of several GATs including those which play important roles in purine and pyrimidine metabolism. [4] More recently, its CTPS inhibitory activity has been correlated to the observed antitrypanosomal activity against bloodstream T. brucei in culture and in a mouse model. [3] Acivicin binds to the glutaminase domain of CTPS and mimics the natural substrate l-glutamine ( Figure 1). The enzyme is irreversibly inactivated through formation of a covalent adduct produced by nucleophilic attack by a Cys residue thiol group to C3 of the isoxazoline nucleus, with displacement of the chlorine atom. [5] This mechanism of action indicates that the inhibitory activity should be influenced by the nature of the leaving group. In a previous work, we reported that 3-bromoacivicin ( Figure 1) has a profile similar to that of acivicin in tests against a panel of tumor cell lines. [6] However, a direct The first convenient synthesis of enantiomerically pure (aS,5S)a-amino-3-bromo-4,5-dihydroisoxazol-5-yl acetic acid (3-bromoacivicin) is described. We demonstrate that 3-bromoacivicin is a CTP synthetase inhibitor three times as potent as its 3chloro analogue, the natural antibiotic acivicin. Because CTP synthetase was suggested to be a potential drug target in African trypanosomes, the in vitro/in vivo antitrypanosomal activi-ty of 3-bromoacivicin was assessed in comparison with acivicin. Beyond expectation, we observed a 12-fold enhancement in the in vitro antitrypanosomal activity, while toxicity against mammalian cells remained unaffected. Despite its good in vitro activity and selectivity, 3-bromoacivicin proved to be trypanostatic and failed to completely eradicate the infection when tested in vivo at its maximum tolerable dose.  comparative study between acivicin and 3-bromoacivicin has not yet been made, and 3-bromoacivicin has only been previously evaluated as a racemate.
Herein we compare the CTPS inhibitory activity and the antitrypanosomal activity of acivicin with that of its 3-bromo analogue. We describe the development of a convenient synthesis of enantiomerically pure (aS,5S)-a-amino-3-bromo-4,5-dihydroisoxazol-5-yl acetic acid (3-bromoacivicin) allowing it to be tested, in parallel with acivicin, both for CTPS inhibition and as an antitrypanosomal agent.

Results and Discussion
The synthesis of 3-bromoacivicin in a racemic form was previously described. [7] However, a survey of the literature shows that an efficient preparation of enantiomerically pure (aS,5S)-3bromoacivicin has never been described; the only reported methods require multistep procedures and rely on resolution of racemic mixtures through the formation of diastereomeric salts followed by crystallization. [8] Herein we describe an efficient synthetic procedure to prepare (aS,5S)-3-bromoacivicin in a few synthetic steps with good overall yield. We took advantage of our previously described method for the preparation of (aS,5S)-acivicin. [9] As detailed in references [9], the 1,3-dipolar cycloaddition of bromonitrile oxide to (S)-3-(tert-butoxycarbonyl)-2,2-dimethyl-4-vinyloxazolidine [(S)-3, Scheme 1], yields the two pairs of diastereomeric cycloadducts erythro-(+)-4/threo-(À)-5 in a 65:35 ratio and in 88 % overall yield. The two isomers can be easily separated by silica gel column chromatography. We now used the erythro cycloadduct (+)-4 as a key intermediate to prepare (aS,5S)-3-bromoacivicin. The acetonide function was removed upon treatment with an aqueous solution of acetic acid, and the N-Boc-protected amino alcohol (+)-6 was oxidized to the corresponding carboxylic acid (+)-7 by using ruthenium(IV) oxide and sodium periodate, followed by removal of the amino protecting group by treatment with a 33 % solution of hydrogen bromide in acetic acid. After cation-exchange chromatography, (aS,5S)-3-bromoacivicin (+)-2 was obtained in its zwitterionic form. The overall yield after five steps starting from alkene (S)-3 was 25 %. The starting material (S)-3 can be purchased or easily prepared by starting from the commonly commercially available (R)-Garner's aldehyde, following a published procedure. [10] Acivicin (1) and 3-bromoacivicin (2) were assayed for inhibitory activity against purified recombinant T. b. brucei CTPS ( Figure 2). Acivicin displayed an IC 50 value of 320 AE 25 nm, whereas 3-bromoacivicin was threefold more potent, with an IC 50 value of 98 AE 10 nm, indicating the nature of the leaving group plays a role in the interaction with the enzyme.
We then evaluated the antitrypanosomal activity against the bloodstream form T. brucei (s427) (Figure 3 a). 3-Bromoacivicin displayed a 12-fold greater antitrypanosomal potency (ED 50 : 38 AE 13 versus 450 AE 10 nm), whilst the activities of 3-bromoacivicin and acivicin against mammalian HEK cells (Figure 3 b) were similar (ED 50 : 12.91 AE 0.10 and 15.65 AE 0.21 mm, respectively). Therefore, 3-bromoacivicin appeared to be a good drug candidate with antitrypanosomal activity in the low nanomolar range and a selectivity index of~300-fold relative to human HEK cells.   Given that pharmacokinetics and toxicity issues mitigated against the development of acivicin as a new trypanocide, [3b] we considered the superior activity of 3-bromoacivicin to offer the potential of lower dosing in mammals. We therefore evaluated the efficacy and primary toxicity profile of 3-bromoacivicin in vivo, using T. b. brucei-infected mice. Administration of a single dose of 3-bromoacivicin did not cause signs of acute toxicity (i.e., acute stress, difficulty in breathing, malaise, lack of appetite) at 10 and 50 mg kg À1 . The individual animals did lose 9.4 and 11.9 % of their body weight at the respective doses of 10 and 50 mg kg À1 by day 14 post-treatment. However, this was deemed acceptable according to safety protocol measures, by which a compound is considered toxic if weight loss is > 25 %. We then tested the ability of a 50 mg kg À1 dose to treat trypanosomes in the mouse model ( Figure 4). Treatment of mice infected with T. b. brucei (s427) commenced on day 3, that is, two days post-infection, where the parasitemia level was at~2.5 10 5 cells mL À1 for each animal. Parasites were not detectable throughout treatment days (i.e., days 3-6), indicating in vivo activity of the compound. However, parasites started to reappear on day 8, that is, two days after treatment was stopped, and proliferated progressively. The experiment had to be terminated on day 11 due to high parasitemia. Pentamidine as the positive control was effective at ensuring 100 % cure throughout the experiment (i.e., '0' parasitic count).
The parasitemia level peaked with the untreated group as expected, and these animals were culled on day 8 due to high parasitemia levels. This in vivo result with 3-bromoacicivin is similar to a previous observation for acivicin. [3b] Interestingly, in culture, acivicin and 3-bromoacivicin are trypanocidal after 72 h above 10 mm and at 1 mm after 96 h. However, they are only trypanostatic for 24 h at 100 nm; beyond this time point or below this concentration (10 nm), the trypanosomes continue to multiply, albeit at a decreased growth rate, depending on the concentration of inhibitor.

Conclusions
In summary, we show that substitution of the 3-chloro group of acivicin with a bromo group to give 3-bromoacivicin increases the inhibitory potency against the target enzyme CTPS threefold. More interestingly, this translates into a 12-fold increase in the in vitro antitrypanosomal activity, while leaving the toxicity against mammalian cells unaffected. This could be due to an enhanced ability of the bromo analogue to enter trypanosomes, or due to its ability to inhibit targets in addition to CTPS in a manner that enhances its antitrypanosomal activity. Unfortunately, in vivo testing of 3-bromoacivicin at its maximum tolerable dose appeared to eradicate the trypanosomes initially, but was found to be ineffective at complete cure of the infection, as the parasites re-emerged upon removal of the treatment and proliferated progressively thereafter.
Future efforts will be devoted to the optimization of the lead compound 3-bromoacivicin through the design of new ligands with an increased inhibitory activity toward CTPS, which, according to the results presented herein, may lead to substantial increases in antitrypanosomal activity, while having minimal impact on the toxicity profile.  . Mice in groups of three were treated daily for four days (represented by arrows "fl") with 3-bromoacivicin 50 mg kg À1 (&) and pentamidine 4 mgkg À1 (^) given via i.p. injections (100 mL per injection). Untreated mice (~) acted as negative control. Parasitemia levels were monitored daily by venepunctures, and for cases in which parasitemia reached, or was about to reach~10 8 cells mL À1 , the animal(s) were euthanized using Schedule 1, or at the end of procedure. Error bars represent standard deviations between the three animals within each group.

Experimental Section
Materials and methods. All reagents were purchased from Sigma, and rodents were supplied by Harlan Laboratories UK Ltd. Acivicin was prepared according to a published procedure. [9a] Compound (+)-4 [9] was prepared according to published procedures by starting from (S)-3. [10] 1 H NMR and 13 C NMR spectra were recorded with a Varian Mercury 300 (300 MHz) spectrometer. Chemical shifts (d) are expressed in ppm and coupling constants (J) in Hz. MS analyses were performed on a Varian 320-MS triple quadrupole mass spectrometer with ESI source. Rotary power determinations were carried out using a Jasco P-1010 spectropolarimeter, coupled with a Haake N3-B thermostat. TLC analyses were performed on commercial silica gel 60 F 254 aluminum sheets; spots were further evidenced by spraying with a dilute alkaline solution of KMnO 4 or with ninhydrin. Melting points were determined on a model B 540 Büchi apparatus and are uncorrected. (aS,5S)-a-tert-Butoxycarbonylamino-3-bromo-4,5-dihydroisoxazol-5-yl acetic acid (+)-7. Compound (+)-6 (0.80 g, 2.6 mmol) was dissolved in a 10:7:7 mixture of H 2 O, CH 3 CN and CCl 4 (24 mL). NaIO 4 (2.22 g, 10.4 mmol) and a catalytic amount of RuO 2 ·H 2 O (6.6 mg, 0.05 mmol) were added and the suspension was vigorously stirred at room temperature for 2 h. After disappearance of the starting material, CH 2 Cl 2 (50 mL) and H 2 O (50 mL) were added. The organic phase was separated and evaporated under reduced pressure. The residue was dissolved in EtOAc (10 mL), the organic layer was extracted with a 10 % aqueous solution of K 2 CO 3 (3 10 mL) and the aqueous phase was made acidic with 2 n HCl and newly extracted with EtOAc (3 10 mL). The organic extracts were washed with brine, dried over anhydrous Na 2 SO 4 , and the solvent evaporated to give (+)-7 as a white foam (0.55 g, 65 % yield); (aS,5S)-a-Amino-3-bromo-4,5-dihydroisoxazol-5-yl acetic acid (+)-2. Compound (+)-7 (0.55 g, 1.7 mmol) was treated with a 33 % solution of HBr in AcOH (20 mL) at room temperature for 5 min. The volatiles were removed under vacuum, the residue was dissolved in H 2 O and submitted to cation-exchange chromatography using Amberlite IR-120H. The acidic solution was slowly eluted onto the resin, and then the column was washed with H 2 O until the pH was neutral. The compound was eluted off the resin with 0.5 n aqueous ammonia, and the product-containing fractions (detected with ninhydrin stain on a TLC plate) were combined. The solvent was freeze-dried to give (+)-2 as a white powder (0.36 g, CTPS enzyme assays. T. brucei CTPS was recombinantly expressed and purified as described elsewhere. [3c] A coupled spectrophotometric assay was used consisting of a final volume of 150 mL containing: 70 mm MOPS pH 7.6, 150 mm KCl, 12 mm MgCl 2 , 500 mm ATP, 1.0 mm DTT, 1 mm phosphoenolpyruvate, 0.5 mm NADH, 1.5 units of pre-mixed pyruvate kinase and lactate dehydrogenase, 655 mm GTP, 625 mm UTP, 1.25 mm l-Gln, and 13 mg purified T. brucei CTPS. [3c] Various concentrations of acivicin and 3-bromoacivicin were pre-incubated with T. brucei CTPS while on ice for 5 min, prior to assaying. In vitro toxicity assays for T. brucei (s427) and HEK mammalian cells. T. brucei (s427) was cultured to the optimum density of 1-2 10 6 cells mL À1 in HMI-9 supplemented with 10 % fetal calf serum (FCS) under environmental conditions of 37 8C and 5 % CO 2 . Solutions of test compounds were prepared in culture media at stock concentration of 200 mm and diluted serially (1:2) across the 96well, flat-bottom solid white plates to give a total of 11 decreasing concentrations (100 mL per well). The last well of each series was left blank (drug free, negative control). Cells were prepared at the concentration of 4 10 4 cells mL À1 and added to each well of the respective compound series (100 mL well À1 ). Plates were incubated at 37 8C and 5 % CO 2 for 48 h prior to the addition of a solution of Alamar Blue (20 mL per well, 0.49 mm in 1 PBS, pH 7.4) followed by a further 24 h. Assay end points were measured fluorimetrically with a fluorescence spectrometer (FluoStar, BMG LabTech, Germany) and Optima program set at l excitation = 544 nm and l emission = 590 nm. Data were analyzed using Prism 5.0 software to obtain ED 50 values. Experiments were performed in duplicate and repeated three times. A similar Alamar Blue assay was carried out with human embryonic kidney (HEK) cells, cultured in DMEM supplemented with 10 % FCS and 2 mm l-Gln. HEK cells were plated at an initial concentration of 3 10 5 cells mL À1 (100 mL per well) and incubated with test compounds for 16 h prior to the addition of Alamar Blue solution.
Trypanocidal or trypanostatic effects of acivicin and 3-bromoacivicin. Cultured bloodstream T. brucei parasites (1 10 5 cells mL À1 ) were incubated in replicates of six with various concentrations (1 nm, 10 nm, 100 nm, 1 mm and 10 mm) of either acivicin or 3-bromoacivicin. The parasites were counted every 24 h. At 72 and 96 h, triplicate samples of the cells treated with 1 and 10 mm acivicin or 3-bromoacivicin were centrifuged (500 g, 5 min), washed with fresh HMI-9 medium, and suspended in fresh medium in the absence of acivicin or 3-bromoacivicin and monitored for a further 10 days.
In vivo acute toxicity and efficacy of 3-bromoacivicin. Female adult ICR mice were tested with 3-bromoacivicin at progressive increasing doses (10 and 50 mg kg À1 ) for signs of acute toxicity. Compound was dissolved in H 2 O and administered via intraperitoneal (i.p.) injection (100 mL per dose). The treated rodents were observed and weighed daily for 14 days. Euthanasia via Schedule 1 methods (see below) were carried out at signs of acute toxicity and/or body weight loss of > 25 % or at the end of procedure. In the efficacy assay, mice were infected with T. brucei (s427) (2 10 4 parasites per mouse, 100 mL i.p. per injection) in groups of three. Once infection was established (normally 48 h post-infection), mice were treated with pentamidine (4 mg kg À1 , 100 mL i.p. per injection, positive control) and 3-bromoacivicin (50 mg kg À1 , 100 mL i.p. per injection). Untreated mice acted as negative control. Treatments were prepared fresh daily and repeated daily for a further 3 days (i.e., four doses in total). Daily monitoring of parasitemia levels was carried out via venepuncture. For instances in which the parasitemia level reached or was about to reach~10 8 cells mL À1 , or at the end of the procedure, mice were euthanized using Schedule 1 methods according to legislation of the United Kingdom Animals (Scientific Procedures) Act 1986, the national and the University of Glasgow maintenance and care guidelines.