key: cord-0312717-o0nbkoi1 authors: Madrigal-Aguilar, Damián A.; Gonzalez-Silva, Adilene; Rosales-Acosta, Blanca; Bautista-Crescencio, Celia; Ortiz-Álvarez, Jossué; Escalante, Carlos H.; Sánchez-Navarrete, Jaime; Hernández-Rodríguez, César; Chamorro-Cevallos, Germán; Tamariz, Joaquín; Villa-Tanaca, Lourdes title: Antifungal activity of fibrate-based compounds and substituted pyrroles inhibiting the enzyme 3-hydroxy-methyl-glutaryl-CoA reductase of Candida glabrata (CgHMGR), and decreasing yeast viability and ergosterol synthesis date: 2021-09-15 journal: bioRxiv DOI: 10.1101/2021.09.14.460412 sha: 30acd8818714887bd23ad1418bf10f8d2e42e3fe doc_id: 312717 cord_uid: o0nbkoi1 Due to the emergence of multi-drug resistant strains of yeasts belonging to the Candida genus, there is an urgent need to discover antifungal agents directed at alternative molecular targets. The aim of the current study was to evaluate the capacity of synthetic compounds to inhibit the Candida glabrata enzyme denominated 3-hydroxy-methyl-glutaryl-CoA reductase (CgHMGR), and thus affect ergosterol synthesis and yeast viability. One series of synthetic antifungal compounds were analogues to fibrates, a second series had substituted 1,2-dihydroquinolines and the third series included substituted pyrroles. α-asarone-related compounds 1c and 5b with a pyrrolic core were selected as the best antifungal candidates. Both inhibited the growth of fluconazole-resistant C. glabrata 43 and fluconazole-susceptible C. glabrata CBS 138. A yeast growth rescue experiment based on the addition of exogenous ergosterol showed that the compounds act by inhibiting the mevalonate synthesis pathway. A greater recovery of yeast growth occurred for the C. glabrata 43 strain and after the 1c (versus 5b) treatment. Given that the compounds decreased the ergosterol concentration in the yeast strains, they probably target the ergosterol synthesis. According to the docking analysis, the inhibitory effect of the 1c and 5b could possibly be mediated by their interaction with the amino acid residues of the catalytic site of CgHMGR. Since 1c displayed higher binding energy than α-asarone and 5b, it is a good candidate for further research, which should include structural modifications to increase its specificity and potency as well as in vivo studies on its effectiveness at a therapeutic dose. HIGHLIGHTS Fibrate-based and pyrrole-containing compounds were tested as C. glabrata inhibitors. The best inhibitor from fibrate was 1c and from pyrroles was 5b. These agents inhibited C. glabrata growth better than the reference antifungals. They also inhibited ergosterol synthesis by the two C. glabrata strains tested. Experimental research, which should include structural modifications to increase its specificity and potency 47 as well as in vivo studies on its effectiveness at a therapeutic dose. 1) Fibrate-based and pyrrole-containing compounds were tested as C. glabrata inhibitors. 56 2) The best inhibitor from fibrate was 1c and from pyrroles was 5b. 57 3) These agents inhibited C. glabrata growth better than the reference antifungals. 58 4) They also inhibited ergosterol synthesis by the two C. glabrata strains tested. 100 + 0.4 *** 95.5 + 1.1 *** 100 + 0 *** 81.8 + 3.4 *** 300 100 + 0 *** 98.06 + 1.9 *** 100 + 0 *** 90.1 + 11.3 *** 600 100 + 0 *** 99.51 + 0.8 *** 100 + 0 *** 100 + 0 *** 139 a The relative growth was calculated as a percentage of the growth detected in the absence of 140 any inhibitor (considered as 100%). Data are expressed as the average of three replicates + 141 SD. Significant differences were analyzed with two-way ANOVA. ***P<0.001. 142 8 144 145 146 147 148 149 150 151 Table 2 152 Effect of 1c, 5b, ⍺-asarone and atorvastatin on the growth of Candida glabrata 43. 87.2 + 8.9 *** 84.3 + 11.5 *** 100 + 0 *** 53.7 + 12.6 *** 300 96.5 + 3.0 *** 100 + 0 *** 100 + 0 *** 93.9 + 6.2 *** 600 100 + 0 *** 100 + 0 *** 100 + 0 *** 100 + 0 *** 155 a The relative growth was calculated as a percentage of the growth detected in the absence of 156 any inhibitor (considered as 100%). Data are expressed as the average of three replicates + 157 SD. Significant differences were analyzed with two-way ANOVA. ***P<0.001. A yeast growth rescue experiment was carried out to verify that the inhibition of the HMGR 173 enzyme affects the levels of ergosterol, the final product of the biosynthesis pathway ( Figure 174 2). The compounds were applied at the sublethal concentrations estimated in the previous 175 experiment (MIC70-90). When exogenous ergosterol was subsequently added to the culture 176 medium, yeast growth did indeed occur, in contrast to the lack of growth caused by the 177 inhibitor. In some cases, such as with compound 1c applied to C. glabrata 43, the recovery of yeast growth reached an even higher level than the control (the yeast cultured in the 179 absence of an inhibitor). Thus, this finding confirmed that the compound derived from α-180 asarone altered the pathway for the production of ergosterol in C. glabrata, and more 181 specifically that it targeted the synthesis of the HMGR enzyme. 182 183 Figure 2 . Yeast growth rescue experiment of Candida glabrata with ergosterol. After yeast 184 growth was stopped by treatment with HMGR inhibitors (antifungal reference compounds 185 and test compounds used at their IC70-90), the addition of exogenous ergosterol led to a 186 recovery of the growth of C. glabrata CBS 138 and C. glabrata 43. The strains were also 187 grown without any inhibitor as a control (considered as 100% growth). + represents addition 188 of the inhibitor or ergosterol to the medium;indicates the absence of the same. The optical 189 density was determined in a Thermo Scientific TM Multiskan™ FC microplate photometer at 190 620 nm. Growth rate values (As600) are expressed as the average of three independent assays 191 + SD. ***P<0.001 compared to the assay without any inhibitor, based on the Student's t- test. 192 193 To explore the possible association between the loss of viability of C. glabrata and the 196 inhibition of the production of ergosterol, the level of ergosterol in the yeasts was measured 197 after 18 h of treatment with 1c, 5b, simvastatin or α-asarone (the latter two as reference 198 compounds; data not shown). The corresponding absorption spectra (Figure 3 The absorption peak corresponding to 281.5 nm was used to quantify the concentration 213 of ergosterol, allowing for the calculation of the percentage of inhibition of its synthesis 214 (Table 4 ). In general, residual ergosterol levels were higher in the C. glabrata 43 versus C. 215 glabrata CBS 138 strain. In both strains, a greater decrease in ergosterol was caused by 1c 216 than 5b. Simvastatin and ⍺-asarone served as positive controls for the inhibition of 217 CgHMGR, since previous studies demonstrated their capability of inhibiting the recombinant 218 HMGR of C. glabrata [8] . It is observed that the higher the concentration of the inhibitor, 219 the greater the percentage of inhibition of ergosterol synthesis (Table 4) . The level of ergosterol was calculated based on the absorbance obtained at 281.5 nm, 240 expressing it as a percentage of the wet weight of the cells, as described by Skaggs et al. [15] . C. glabrata was grown in YPD medium treated with different 242 concentrations (50, 150, 300 and 600 μM) of the inhibitors: simvastatin, α-asarone, 1c and 243 5b. For the controls, the yeast was grown in YPD medium without any treatment or with 244 DMSO only. The data represent the average of the three independent assays for each 245 treatment. The previous results allowed for the calculation of the IC50, the concentration of 246 the inhibitor that causes 50% inhibition of ergosterol synthesis in C. glabrata (Supplementary 247 CgHMGR. The related values for 1c and 5b are shown in Table 5 . 1c has the highest binding 254 energy in silico, which correlates with the in vitro results (Table 1) . Atorvastatin had the 255 lowest binding energy (Table 5 ). The interaction of compounds 1c and 5b with the amino 256 acid residues in the catalytic site is depicted in Figure 4 . 1c exhibited hydrogen bonds with a 257 length of 2.58-2.99 Å between the hydroxyl groups at C-5 and C-8 and Glu93 and Asn192, 258 respectively, as well as an electrostatic interaction of the O11 methoxy group with Met191. 259 For 5b, there were hydrogen bonds 2.19 and 19.7 Å in length between the hydroxyl group at 260 C-5 and Met191, and between the carboxyl group at C-7 and Asp303, respectively. The 261 interaction between atorvastatin and the HMGR catalytic site revealed that van der Waals 262 interactions are predominant, although two hydrogen bonds are detected (19.7 and 22.7 Å) 263 between the carboxyl group at C-17 and Gly341. Additionally, Asp303 interacted by 264 hydrogen bonds with the carboxyl group at C-17 and the hydroxyl group at C-15 ( Figure 4) . 265 The calculated binding energies of 1c and 5b (-5.99 and -5.71 kcal/mol, respectively) were 266 better than those found for α-asarone and atorvastatin (4.53 and -2.13 kcal/mol, respectively) 267 (Table 5) . 268 The problem of drug-resistant strains will always exist due to the process of natural 2 evolution and selection of yeasts and bacteria [25] . Therefore, the probability of applying an 3 effective treatment to patients would be increased by having a broad battery of antifungal 4 agents from which to choose as well as distinct molecular targets among such drugs. 5 The HMGR enzyme (particularly CgHMGR) has for some time been proposed as a 6 possible target, leading to the study of some cholesterol-lowering drugs (e.g., simvastatin and 7 atorvastatin) as inhibitors of the growth of pathogenic yeasts [10,26]. According to in vitro 8 evolutionary experiments, treatment of C. glabrata with some statins may allow for the 9 selection of mutants. However, gene sequencing has not detected any changes in the catalytic 10 domain of CgHMGR, indicating no effect on HMGR activity. C. glabrata is a useful model 11 for examining resistance to statins and the precise molecular mechanisms of resistance to 12 compounds that inhibit the CgHMGR enzyme [5] . 13 In the current effort, three series of compounds were evaluated as inhibitors of C. 14 glabrata. Two of the best derivatives were selected to determine their effect on yeast growth 15 and ergosterol synthesis. Complementary studies were carried out with yeast growth rescue 16 assays and docking simulations. 17 The compounds presently investigated were originally designed as lipid-lowering [11] 18 and anti-inflammatory agents [12] . Their chemical structure could plausibly enable them to 19 inhibit the activity of the CgHMGR enzyme. In fact, substituted pyrroles have been 20 considered as antifungals [13, 23] and their fungicidal activity is reported. However, the 21 possible molecular target has not been previously explored in an in-depth manner. 22 Compounds such as statins (e.g., simvastatin and atorvastatin) and fibrates that inhibit 23 HMGR have been administered to lower cholesterol levels in humans [27] . Additionally, 24 they have been assessed as growth inhibitors of Candida spp., Aspergillus spp. and Ustilago 25 maydis [7-10,14,26]. Based on its hypercholesterolemic activity, α-asarone underwent initial 26 studies [27, 28] that resulted in a finding of high toxicity. Thus, new derivative compounds 27 have been designed and synthesized, and these have produced good activity against different 28 fungi, such as C. glabrata and Ustilago maydis [9, 14] . 29 When the test compounds were examined in vitro, the growth inhibition of both strains 30 of C. glabrata was better for 1c than 5b and α-asarone. On the other hand, 5b did not induce 31 a greater growth inhibition than its reference compound, atorvastatin. The latter statin, 32 bearing a substituted pyrrolic ring, has already been proposed as an antifungal agent to inhibit 33 the growth of Candida spp. [26] . Although the antifungal activity of 1c has already been 34 studied [11] , this is the first evaluation, to our knowledge, of its effect on an opportunistic 35 pathogenic yeast. Furthermore, the current investigation constitutes the first in-depth 36 exploration of the mechanism of action and molecular target of the inhibitors. 37 According to the yeast growth rescue experiment, the test compounds likely inhibited 38 the pathway for sterol biosynthesis [9,26]. The addition of ergosterol to C. glabrata CBS 138 39 resulted in a recovery of growth at a level below that of the control (without treatment with 40 an inhibitor), while its addition to C. glabrata 43 led to growth that overcame the control 41 level. This behavior can be explained by what is observed in the fluconazole-resistant C. consumption and production of ergosterol. Hence, the present test compounds probably 46 inhibit the pathway for sterol biosynthesis, as fluconazole does [30, 31] . Since 1c and 5b inhibited ergosterol synthesis, they may reduce the activity of the 48 CgHMGR enzyme [26] . A better inhibition of the production of ergosterol was found for 1c 49 in both strains of C. glabrata compared to its control (α-asarone) and 5b. Of these 50 compounds, 1c had the lowest IC50. Previous publications have documented the capability of 51 simvastatin, α-asarone, and derivatives of the latter to inhibit recombinant CgHMGR [8, 9] . 52 A correlation has been detected in C. albicans strains between their sensitivity to azoles 53 and their total ergosterol concentration [15] . Therefore, it was important to demonstrate that 54 the test compounds were capable of inhibiting the synthesis of ergosterol in both strains of 55 C. glabrata (the fluconazole-susceptibility and -resistant strains). 56 The experimental results from the assays on yeast growth inhibition and the inhibition 57 of ergosterol synthesis were complemented by docking simulations based on molecular 58 coupling between the test compounds and CgHMGR. The binding energy values calculated 59 for 1c and 5b were congruent with the in vitro findings for these two compounds. 1c exhibited 60 the lowest binding energies and the best in vitro inhibition of yeast growth. Better binding to 61 the active site of CgHMGR was displayed by 1c and 5b than α-asarone and its derivates, To verify that inhibitors affect yeast viability by inhibiting ergosterol synthesis, a growth 157 rescue experiment was conducted. Growth was first stopped by subjecting yeasts to the 158 sublethal concentration (IC70-90) of one of the inhibitors, determined by the CLSI M27-A3 protocol (see section 2.3), and then ergosterol was added. Briefly, to each well of 96-well 160 microplates were added 100 µL of one of the antifungal solutions (2x) prepared in RPMI-161 1640 medium (Sigma-Aldrich), followed by 80 µL of a yeast suspension adjusted to 1-5 x 162 10 6 UFC/mL and diluted 1:1000 with RPMI-1640 medium (Sigma-Aldrich). A stock solution 163 of ergosterol was prepared by dissolving 11 µg/mL in Tween 80/ethanol, and 20 µL of this 164 solution was added to each well. The controls were yeasts cultured without any inhibitor 165 (growth control) and those with an inhibitor but without sterol (growth rescue control). 166 Data are expressed as the mean of three replicates ± standard deviation (SD). Differences 168 between groups were examined with two-way analysis of variance (ANOVA), with the 169 Bonferroni correction, and a 95% confidence interval. Statistical analyses were performed 170 and graphs constructed with GraphPad Prism 5.0. Statistical significance was considered at 171 P<0.001. 172 Total sterols were extracted with a slightly modified version of the methodology reported by 174 Arthington-Skaggs et al. [15] . Briefly, C. glabrata yeasts were grown in YPD medium by LigProt+ software [22] . 206 The fibrate-based derivates 1a-c, 2a-c and 3a-c, and 1,2-dihydroquinolines 4a-d The synthesis of 5a, 5d and 6c-d has been previously reported [11, 12] . The preparation of 223 bromopyrroles 5b and 5c was achieved by treatment of compound 5a [12] with N-224 bromosuccinimide (NBS) as the brominating agent under mild reaction conditions (Scheme 225 1). Even though l.0 mol equivalent of NBS was employed, a mixture of bromopyrroles 5b 226 and 5c was obtained. Due to the fact that they were easily separated by column 227 chromatography, an excess of NBS (2.5 mol equiv.) was added to the reaction mixture to 228 give 5b and 5c in 32% and 58% yields, respectively. In the case of sensitivity tests and docking analysis, α-asarone was the control for the fibrate-309 based derivatives 1a-c, 2a-c, 3a-c (series 1) and 1,2-dihydroquinolines 4a-d (series 2) and 310 atorvastatin for the substituted pyrroles 5a-d and 6b-d (series 3). In the experiment to 311 determine the effect of the compounds on the biosynthesis of ergosterol, simvastatin and α-312 asarone were employed. It has been reported that these two compounds are capable of Resistance of Candida to azoles 330 and echinocandins worldwide Fungal co-infection in COVID-19 patients: Should we be 334 concerned? Outbreak of Candida auris infection in 337 a COVID-19 hospital in Mexico Increased antimicrobial resistance during 340 COVID-19 pandemic Statin resistance in Candida 343 glabrata Candida infections, causes, targets, and resistance 345 mechanisms: traditional and alternative antifungal agents The 3-hydroxy-3-methylglutaryl coenzyme-A 349 reductases from fungi (HMGR): a proposal as a therapeutic target as a study 350 model Recombinant 3-hydroxy 3-methyl glutaryl-CoA 353 reductase from Candida glabrata (rec-CgHMGR) obtained by heterologous expression, 354 as a novel therapeutic target model for testing synthetic drugs Simvastatin reduces ergosterol levels, inhibits 362 growth and causes loss of mtDNA in Candida glabrata Synthesis and biological activity of fibrate-based acyl-and alkyl-366 phenoxyacetic methyl esters and 1, 2-dihydroquinolines Synthesis and highly potent anti-370 inflammatory activity of licofelone and ketorolac-based 1-arylpyrrolizin-3-ones Synthesis and quantitative 373 structure-activity relationship analysis of N-triiodoallyl-and N-iodopropargylazoles. 374 New antifungal agents Quantitation of ergosterol 377 content: novel method for determination of fluconazole susceptibility of Candida Simvastatin and other inhibitors of the enzyme 3-hydroxy-3-382 methylglutaryl coenzyme A reductase of Ustilago maydis (Hmgr-Um) affect la viability 383 of the fungus, its synthesis of sterols and mating Yeast analysis, spectrophotometric semi-386 microdetermination of ergosterol in yeast Protein Structure Modeling 389 with MODELLER PROCHECK: a 392 program to check the stereochemical quality of protein structures Scalable 395 molecular dynamics with NAMD Avogadro: An advanced semantic chemical editor, visualization, and analysis 399 platform LigPlot+: multiple ligand-protein interaction 404 diagrams for drug discovery Design, synthesis and 407 antifungal activities of novel pyrrole alkaloid analogs Pyrrole antibacterial agents. 2. 4, 5-Dihalopyrrole-2-410 carboxylic acid derivatives Evolutionary emergence of drug resistance 413 in Candida opportunistic pathogens Growth inhibition 416 of Candida species and Aspergillus fumigatus by statins Rapid effect 419 of 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibition on coronary endothelial 420 function evaluating alpha-asarone-based HMG-CoA reductase inhibitors Molecular docking of the highly hypolipidemic agent alpha-429 asarone with the catalytic portion of HMG-CoA reductase Facultative 432 sterol uptake in an ergosterol-deficient clinical isolate of Candida glabrata harboring a 433 missense mutation in ERG11 and exhibiting cross-resistance to azoles and amphotericin Sterol uptake and sterol biosynthesis act coordinately to mediate antifungal resistance i n Candida glabrata under azole and hypoxic stress Exploration of virtual 441 candidates for human HMG-CoA reductase inhibitors using pharmacophore modeling 442 and molecular dynamics simulations This work was supported by CONACyT (grants CB283225, 300520 and A1-S-17131) and 316 the SIP-IPN (grants SIP20200775, SIP20210508, SIP20200227 and SIP20210700).