Design, Synthesis and Biological Evaluation of Pyrimidine Analogues as SecA Inhibitors

SecA, a key component of the bacterial Sec-dependent secretion pathway, is an attractive target for the development of new antimicrobial agents. We have previously reported pyrimidine analogs as SecA inhibitors. Herein, we report an extension of the earlier work in the synthesis and evaluation of a series of 15 5-cyanothiouracil derivatives as SecA inhibitors. All the compounds have been evaluated for their inhibition of SecA ATPase (EcSecAN68) and for their antimicrobial activity against Escherichia coli NR698 (a leaky mutant) and Bacillus anthracis Sterne. Twelve compounds showed IC 50 of less than 6.3 µM when tested against EcSecAN68. In antimicrobial studies against E. coli NR698, six compounds showed MIC of less than 12.5 µM with three being less than 6.3 µM. Against B. anthracis Sterne, three compounds showed MIC of less than 6.3 µM.


Introduction
Infectious diseases caused by bacterial pathogens have become a major public health problem in recent years due to the widespread occurrence of drug resistance. Therefore, there is an urgent need to develop new antimicrobial agents, especially those with a new drug target and/or with the ability to overcome drug resistance. [1] Along this line, we are interested in targeting SecA, which is a critical protein secretion machinery essential for bacterial survival. [2][3][4] SecA is a key component of the bacterial protein secretion (Sec) pathways.
SecA ATPase is one of the essential components in the Sec machinery, which provides a major pathway to help protein translocation from the cytosol across or into the cytoplasmic membrane. [5][6][7][8][9][10][11] Because SecA is a conserved and essential protein in all bacteria and is absent in humans, it is considered as a promising antibacterial drug target. At present, small organic molecules that can inhibit SecA mainly include Rose Bengal, [12] bisthiouracil, [13] bistriazole [14] and their derivatives, [15] thiazolo [4,5-d]pyrimidine derivatives [16] and others. [17,18] As an extension of our earlier work, [15,19] we are working on optimizing existing SecA inhibitors of the substituted pyrimidine scaffold by exploring the chemistry space at the 2 and 4 positions of the pyrimidine core. [13] Below, we describe the results and implications in guiding future work in this area.

Results And Discussion
Chemistry Earlier work has identi ed the compound in Fig. 1 as a lead for further optimization. [19,15] To improve the potency, we focused our attention at the linker between the biphenyl rings and the para position of the benzylthio group. Speci cally, the designed compounds were synthesized by introducing (a) linkers such as thioethers (-S-, -CH 2 -S-), an inverted alkoxy (-O-CH 2 -), an alkenyl group (-CH = CH-), an alkynyl (-C ≡ C-) group, and an amido group (-CH 2 -NH-CO-) between the two phenyl rings and (b) substituents such as -N 3 , -COOMe, -CF 3 at the para position of the benzylthioether moiety. Thus, a series of pyrimidine derivatives 3a-o were synthesized. The synthetic route is shown in Scheme 1. Compounds 1a, b were prepared by nucleophilic aromatic substitution reaction of 4-nitrobenzaldehyde [20] or 4-uorobenzaldehyde [21] using the appropriate aryl thiols under basic conditions. Compound 1c was obtained by following a published procedure 20 in Scheme 2.
Palladium catalyzed Mizoroki-Hech reaction of 4-bromobenzaldehyde with styrene yielded 1d. [22] Compound 1e was prepared by palladium catalyzed Sonogoshira cross coupling of 4-bromobenzaldehyde with phenylacetylene. [23] Then 4-formylbenzoic acid was coupled with benzylamine using isobutyl chloroformate in the presence of triethylamine to provide amide 1f by following a published procedure. [24] The reaction of the appropriate aromatic aldehydes, cyano ethyl acetate and thiourea with piperidine as catalyst in absolute ethanol at re ux temperature overnight afforded thiouracil 2a-f. The target compounds 3a-o were obtained by the Sbenzylation in the presence of potassium carbonate in acetonitrile. Overall, 15 compounds have been prepared in overall yields ranging from 47-98 using readily available starting materials.

Biological evaluation
The activities of all newly synthesized compounds 3a-o were rst screened using a truncated version of E. coli SecA, EcSecAN68, at 6.25 µM. [15] As it can be seen in Fig. 2, most of these 15 new compounds showed more than 50% inhibition of the SecA ATPase activity at this concentration, suggesting potent inhibitory activity. Some of the compounds seemed to be more potent than the lead compound, SCA168, and showed more than 90% inhibition at 6.25 µM such as SCA225, 227, 230, 232, 233, 239, 245, 260, and 262. The overall results from the SecA inhibition studies seem to suggest that the "linker" part can accommodate thioethers (SCA225, 230, 239 and SCA260-262), an inverted oxo ether (SCA232, 238, 240), an alkenyl linker (SCA227, 233), and alkynyl linker (SCA245) as well as variations in the length by one methylene group (1 to 2 atoms) among those with a thioether linker. Among all these variations, it is especially interesting to see the accommodation of the linearization of the linker with an alkynyl group (SCA245). It also seems that the para-position of the thiobenzyl group of the 4-position of the pyrimidine ring can tolerated various groups. From this limited set of compounds, it is also clear that a three-atom linker in the form of an amido group is not well-tolerated.
All compounds were also evaluated for their antimicrobial activity against a "leaky" outer membrane mutant of Gram-negative E. coli NR698. [25] Among them, four compounds showed MIC at about 12.5 µM (SCA225, 232, 240, 261) and two compounds showed MIC at about 6.3 µM (SCA260, 262) against E. coli NR698 (Table 1). Compounds were also evaluated for their antimicrobial activity against Gram positive B. anthracis Sterne. Three of them showed MIC at about 6.3 µM (SCA225, 227, 232) against B. anthracis Sterne, which are comparable to some of our best compounds in this class. [15] With the limited data set, it seems that the potency against B. anthracis is higher than that of E. coli. Such results are understandable because E. coli is Gram-negative strain. Even with a compromised outer membrane, it still has an outer membrane, which presents a permeability barrier.
[26]  Overall, the results provide very useful information for those who might be interested in SecA inhibitor design for both improved potency and for avoiding the chemical space that would not be productive.

Conclusion
We have described the design, syntheses and biological evaluation of a series of 15 small-molecule SecA inhibitors with µM inhibition. The results provide information on tolerable structural features of the linker part between the two phenyl ring at the 2-position and the para-substitution of the benzyl group at the 4-position of the pyrimidine ring.

Experimental
All chemical reagents and solvents used were of reagent grade or puri ed using standard methods. TLC analyses were conducted on silica gel plates (Sorbent Silica G UV254). Column chromatography was carried out on ash silica gel (Sorbent 230-400 mesh). NMR spectra were recorded at 1 H (400 MHz) and 13 C (100 MHz) on a Bruker instrument. Coupling constants (J) and chemical shifts (δ) are given in hertz and ppm respectively, using TMS ( 1 H NMR) and solvents ( 13 C NMR) as internal standards.
General procedure for the synthesis of (2a-f) Our previously published [15] procedure was followed. Brie y, to a solution of ethanol (25 mL) and appropriate aldehyde (RCHO, 5 mmol) was added ethyl cyanoacetate (0.5 mL, 5 mmol), thiourea (0.38 g, 5 mmol) and piperidine (1.0 mL, 10 mmol). The mixture was heated at re ux overnight and then cooled to room temperature. The precipitate was dissolved in 0.5 M NaOH (20 mL) and washed with ethyl acetate (15 ml ⋅3). Then the aqueous solution was acidi ed to pH ~ 2 by slow addition of 1 M HCl. This caused the product to precipitate, which was then ltered using vacuum ltration.
General procedure for the synthesis of (3a-o) To a solution of 2-mercapto-6-oxo-4-(4-(phenylthio)phenyl)-1,6-dihydropyrimidine-5-carbonitrile derivatives (1.36 mmol) in CH 3 CN (10 ml), K 2 CO 3 (6.79 mmol) was added and the resulting mixture was stirred for 10-15 min. To this was added the appropriate (bromomethyl)phenyl derivatives (1.22 mmol) and the reaction was stirred at room temperature for 16-18h. Upon completion, the reaction mixture was cooled to ambient temperature and the solvent removed in vacuo. The dried residue was washed by H 2 O (pH = 9-10, 20 mL ⋅ 2) and brine (15 ml ⋅ 2) followed by product extraction in ethyl acetate (20 ml). The solvent was evaporated in vacuo to obtain crude product, which was puri ed using silica gel column chromatography.

ATPase assays
Inhibition on ATPase activity of EcSecAN68 was determined by malachite green colorimetric assay as previously described. [15] IC 50 is de ned as the concentration of the compound that inhibits 50% of ATPase activity.

Bacteriostatic effects
Bacteriostatic effects were evaluated at 37 o C in 96-well microtitier plates as previously described. [15] Minimum inhibitory concentration (MIC) is the lowest concentration of compounds at which bacterial cells were not able to grow at tested condition.