INTRODUCTION
Arboviruses are responsible for diseases such as Zika, Chikungunya, dengue and yellow fever which have spread at high rates throughout the world in the last years [1] and increasing mortality and morbidity rates associated with these diseases [2,3]. The mosquito Aedes aegypti is the primary vector responsible for the transmission of most of these viruses. The control of this species face struggles due to the great adaptability of the mosquito to urban environments in addition to other biological features [1]. Nowadays, dengue vaccine is in final stage of trials, although there are not vaccines against the other arboviruses. Thus, the most accessible alternative to decrease the transmission of these diseases remains vector control. Therefore, is urgent the search for novel larvicides to exhibit high specificity and low toxicity to minimize the emergence of resistant mosquitoes [4]. An important metabolic pathway in mosquitoes is the kynurenine pathway, which is the major tryptophan catabolism pathway in living organisms. The metabolite 3-hydroxykynurenine (3-HK) has been associated with oxidative stress caused by spontaneous oxidation of reactive oxygen species [5]. In mosquitoes, this pathway comprises the tryptophan degradation, taking to the chemically stable xanthurenic acid (XA). This acid is synthesized from the 3-HK in an irreversible reaction catalyzed by the enzyme 3-hydroxykynurenine transaminase (HKT) [6,7]. Further, previous reports have shown that xanthurenic acid is essential for the sexual reproduction of Plasmodium falciparium in the malaria vector Anopheles spp. [8]. Therefore, inhibition of the enzymatic conversion of 3-HK into xanthurenic acid in the kynurenine pathway is an attractive strategy to control Ae. aegypti development and Anopheles spp. via the accumulation of 3-HK. Before our work, only the crystallographic structure of HKT from Anopheles gambiae (AgHKT) was available, co-crystallized with the competitive inhibitor 4-(2-aminophenyl)-4-oxobutyric acid (4-OB) at 2.7 Å resolution [9]. These HKTs from An. gambiae and Ae. aegypti are orthologs with 79% of sequence identity, and identical residues within a radius of 7 Å from the active site [10]. Previously, we have recognized the striking structural similarity between 4-(2-aminophenyl)-4-oxobutyric acid (4-OB) inhibitor and 1,2,4-oxadiazole derivatives. We have shown that these heterocyclic compounds could have similar binding modes and equivalent affinity for HKT as inhibitor 4-OB. Furthermore, toxicity bioassays performed for fourth-instar larvae of Ae. aegypti have shown that several 1,2,4-oxadiazole derivatives exhibit potent activity against Ae. aegypti larvae (ca. of 15 ppm) and low toxicity in mammals [10]. In this work, we report a complete study of HKT enzyme from Aedes aegypti and the 1,2,4-oxadiazole derivatives inhibitory activity as a new weapon to fight against mosquito spread using organic synthesis, molecular biology and biochemistry approaches.
AIMS
The main objective of this work is to investigate the role of the water soluble 1,2,4-oxadiazole derivatives as inhibitors of Aedes aegypti HKT enzyme. For this purpose, the specific goals were to obtain the recombinant HKT through cloning, heterologous expression and purification technologies. Moreover, we aimed the development of a biochemical protocol for the detection of the xanthurenic acid yielded by HKT catalysis and finally, the quantification of enzyme inhibitory assays. All this work was planned to serve as a model for the design of more potent AeHKT inhibitors.
METHODS
RNA extraction, RT-PCR and cloning methods: Total RNA was extracted from RecLab strain larvae (from Recife-PE) using Ambion™TRIzol™ Reagent and chloroform solution in water, as firstly described by Romão, TP and coworkers [11]. Reverse transcription and PCR were performed as described by Maciel, LG et al [12]. PCR products were cloned into pGEM-T Easy plasmid vector and amplified. The plasmidial DNAs were purified with QIAprep Spin Miniprep kit and submitted to automatic sequencing. After confirmation, the constructs were subcloned in the plasmid pETTrxA/LIC and used for the expression of the RL_AeHKT recombinant protein. Amplification of RL_AeHKT gene to add LIC overhangs were conducted as described by Maciel, LG and coworkers [12].
Expression of recombinant HKT The expression of the recombinant protein was carried in Rosetta 2 DE3 cells transformed with HKT_pETTrxA/LIC construct. The cultures were grown in ZYM-5052 culture medium at 37 °C and expression was driven by incubation at 22 °C for 24 h. The cells were harvested by centrifugation, resuspended in lysis buffer (containing 1% Tween 20) and then lysed by sonication. Protein purification steps were executed in AKTÄ pure system. At first, ionic affinity chromatography with HisTrap HP 5 mL column was performed using buffer A (containing 20 mM Imidazole pH 8) for column equilibration, injection and washing, followed by gradient elution with buffer B (containing 500 mM Imidazole pH 8). Then, a desalting chromatography with HiTrap Desalting column was made to remove imidazole excess prior to TEV cleavage. RL_AeHKT was purified from TEV and Trx mixture by another ionic affinity chromatography, and the highly purified protein was submitted to desalting and concentration. All the steps of expression and purification were monitored by SDS-PAGE.
Synthesis and characterization of oxadiazole salts – general methods A series of eleven oxadiazole salts were synthesized by our research group. The reactions were monitored by TLC analysis with TLC plates containing GF254 from E. Merck©. Compounds were characterized by Electro-thermal Mel-temp© apparatus to determine melting points. A Varian© UNMRS 400 MHz spectrometer, referenced as follows: 1H (400 MHz), internal standard SiMe4 at ? = 0.00 ppm, 13C (100 MHz), internal standard SiMe4 at ? = 77.23 ppm was used to obtain 1H e 13C spectra, respectively. Elemental analyses were performed on Perkin Elmer© 2400 CHNS/O Series II Analyzer. Arylamidoximes were synthesized as described in literature [13].
Synthesis of 4-(3-phenyl-1,2,4-oxadiazol-5-yl)butanoic acids: Glutaric anhydride (30 mmol) and the corresponding arylamidoxime (20 mmol) were put in a round-bottom flask in an oil-bath at 473 K. After one hour, the end of the reaction was verified by TLC analysis in 7:3 hexane/ethyl acetate mobile phase. Then, 10 mL of saturated NaHCO3 solution and 10 mL of dichloromethane were added to the cold flask and the reaction mixture was allowed to stir overnight. After the phase separation, the aqueous phase was acidified until precipitation of the title compound occurred and then dissolved in dichloromethane, dried over Na2SO4 and recrystallized from a mixture of chloroform and hexane. The products were named by their respective aromatic substituent as follows: 3a – Ph; 3b – 4-OCH3Ph; 3c – 3,4-diClPh; 3d – 4-BrPh; 3e – 4-CH3Ph; 3f – 3-CH3Ph; 3g – alpha-Naphthyl; 3h – 4-CF3Ph; 3i – 4-IPh; 3j – 4-ClPh; 3k – Benzodioxolyl. Substances 3a [14], 3b [10], 3d [15], 3e [16], 3f [10], 3i [10], 3j [16] and 3k [10] were synthesized previously and their data were compared to previous literature. All spectroscopic data are available in reference 12.
Synthesis of sodium 4-(3-phenyl-1,2,4-oxadiazol-5-yl)butanoate salts: In a round-bottom flask, the respective 4-(3-phenyl-1,2,4-oxadiazol-5-yl)butanoic acid (0.8 mmol) and 3.2 mL of 1% NaOH methanol solution (freshly prepared) were mixed and the reaction was allowed to stir for one hour. After the time, methanol was evaporated and the product was recrystallized from chloroform. All spectroscopic data are available in reference 12.
Biochemical assays The developed method for the enzymatic assay was adapted from Han and coworkers [6]. Stock solutions were freshly prepared, except for 3-hydroxy-D,L-kynurenine (3-D,L-HK) which was kept in ?20 °C. 3-D,L-HK and sodium pyruvate were directly diluted in 200 mM HEPES buffer pH 7.5/100 mM NaCl, while 20 mM pyridoxal 5?-phosphate (PLP) was first dissolved in 0.1 M NaOH and then diluted in HEPES/NaCl buffer until 0.5 mM. A solution of 20 mM xanthurenic acid was prepared in 0.1 M NaOH to perform calibration curves. The reactions were carried out in a 96-well round bottom microplate containing 8 µL of 25 mM 3- D,L-HK, 8 µL of 0.5 mM PLP and 4 µL of 50 mM sodium pyruvate, 2 µg of recombinant RL_AeHKT and HEPES/NaCl buffer to a 100 µL final volume. The plates were incubated at 50 °C for 5 min and then 100 µL of 10 mM FeCl3·6H2O in 0.1 M HCl was added to interrupt the reaction and generate the XA-Fe3+ complex. Samples were performed in duplicate and absorbances were read at 570 nm. All data were analyzed using GraphPad Prism7.0 software package using the Michaelis-Menten equation. Kinetics and inhibition parameters were determined by nonlinear regression.
Kinetics analysis Kinetics assays were performed as described above. 3-D,L-HK concentrations tested are 0.5, 1.75, 3, 4, 8.5 and 10 mM in the presence of 5 mM sodium pyruvate. pH assays were performed by adding 60% of reaction volume of 200 mM Tris-HCl buffer pH 6, 7 and 8, 200 mM Borate buffer pH 9 and 200 mM Carbonate buffer pH 10, all containing 100 mM NaCl. The effect of the temperature was evaluated by incubation of all reaction components in a heater at 30, 40, 50, 60 and 70 °C. All experiments were conducted in duplicate and blank samples were analyzed simultaneously to each read.
Inhibition assays The inhibition assays were performed with pre-incubation of each compound in five concentrations ranging from 0.5 mM to 0.05 mM with PLP and RL_AeHKT. After 30 minutes of incubation at 50 °C, the samples were mixed to the other reaction components and the assays were performed as described in the Biochemical assays subsection. All compounds tested were diluted in 200 mM HEPES buffer pH 7.5/100 mM NaCl and the samples were evaluated in duplicate. In each test, the negative control was added to calculate the activity percentage.
RESULTS AND DISCUSSION
cDNA amplification via RT-PCR Samples were submitted to the reaction and a fragment of around 1100-bp was amplified. The fragments were purified, cloned in pGEM-T Easy© vector and sequenced to confirm their identity and to evaluate the integrity of these cDNA sequences. Four RL_AeHKT constructs were obtained and two of them were selected for further steps.
Expression and purification of recombinant HKT Rosetta 2 DE3 Escherichia coli cells transformed with the constructs were inoculated into the auto-inducer ZYM-5052 culture medium [17]. The recombinant RL AeHKT is N-terminally fused to Thioredoxin A (Trx) and a 6x-His tag to increase the solubility of the protein, and to allow its purification by immobilized metal affinity chromatography, respectively. RL_AeHKT-Trx exhibited a migration profile of approximately 50 kDa, which is consistent with the predicted size of 54 kDa (43 kDa [6] of HKT e 11 kDa of Trx [18]). After all the purification steps, RL_AeHKT showed integrity at the electrophoresis level and a yield of 30 mg per liter of induced bacterial culture.
Synthesis of 1,2,4-oxadiazoles-like sodium salts 1,2,4-Oxadiazoles are five-membered heterocyclic compounds, which contain an oxygen atom in position 1, and two nitrogen atoms in positions 2 and 4. The water-soluble larvicides were obtained via a three step synthesis, starting with the respective amidoxime and glutaric anhydride to form 4-[3-(aryl)-1,2,4-oxadiazole-5-yl]butanoic acids (3a-k). After recrystallization, these compounds reacted with sodium hydroxide in methanolic medium yielding the corresponding sodium salts (4a-k), which were tested for inhibitory activity against HKT.
Biochemical assays Enzymatic assays were conducted by a new standardized protocol modified from Han and coworkers [6] and developed to evaluate the conversion of 3-HK to xanthurenic acid (XA) by HKT. The assays were performed on a 96-well round bottom microplate, allowing for simultaneous reading of multiple samples, and thus for a faster and cheaper characterization of RL AeHKT activity and/or inhibition. The method is based on XA-Fe3+ complex detection, first described by Lepkovski and coworkers to detect XA in pyridoxine-deficient mice [19]. The assays were performed in duplicate, and a calibration curve was built to quantify the amount of XA generated by RL_AeHKT (R2 = 0.98). Protein samples derived from two clones obtained were tested, and no differences in XA concentrations were observed, indicating that both clones are equally functional.
Enzyme Kinetics The steady-state kinetic constants of RL_AeHKT were estimated from absorbance changes during the catalysis of the substrate 3-D,L-HK at pH 7.5 at 50 °C and 570 nm, varying 3-D,LHK concentrations from 0.5 to 10 mM. Values for Michaelis-Menten constant (KM=3.3 ± 0.8 mM) and maximal enzyme turnover (Kcat=492.3 ± 46 min-1) with their standard errors were calculated by fitting the experimental measurements to the Michaelis-Menten model. Catalytic efficiency (Kcat/KM=149.2 ± 50 min-1.mM-1) value with their standard error was obtained by fitting the experimental data to the normalized Michaelis-Menten equation. Progress curves were carried from 0 to 25 min (with 5 min interval) and the interval of the first-order reaction was observed from 0 to 5 min of incubation, and a plateau was reached after 20 min. Previously, Han and coworkers [6] measured the kinetics constants for HKT from Ae. aegypti expressed in Sf9 cells (AeHKT). Although a direct comparison of our measurements to the latter work is not straightforward due to different expression conditions and biochemical assays, we measured similar KM values for HKT expressed in E. coli (KM = 3.3 mM) compared to the enzyme expressed in Sf9 cells (KM = 3.0 mM) [6]. In contrast, AgHKT has higher affinity for 3-D,L-HK than both recombinant HKT from Aedes aegypti expressed in different systems. One of the probable reasons for this difference in substrate recognition is the presence of D- isomer in 3-D,L-HK commercial substrate, as discussed by Canavesi and coworkers [20]. Despite this difficulty, our interest here is the HKT detoxification activity in the conversion of 3-HK to XA, besides XA is the key compound in XA-Fe3+ detection. The effect of pH and temperature on RL_AeHKT activity was evaluated in the range between 6 and 10 and 30 to 70 °C, respectively. Optimal RL_AeHKT activity was observed at pH 8, slightly lower than pH 9 measured for the HKT recombinant expressed in Sf9 cells [6]. In temperature arrays, RL_AeHKT exhibited higher activity at 50 °C with a decrease in the enzymatic activity at temperatures above that value.
Inhibition assays The inhibitory activity of eleven new 1,2,4-oxadiazole derivatives was assessed in duplicate by adding 0.5–0.05 mM of each diluted salt in 200 mM HEPES buffer pH 7/100 mM NaCl. The obtained IC50 values range from 42 to 294 ?M, which are promising results for the first series of RL_AeHKT inhibitors. Eight out of the eleven inhibitors showed IC50 values below 100 ?M. The series of compounds 4a-k showed the following IC50 values: 4a – 72 ?M; 4b – 101 ?M; 4c – 64 ?M; 4d – 110 ?M; 4e – 88 ?M; 4f – 54 ?M; 4g – 63 ?M; 4h – 70 ?M; 4i – 42 ?M; 4j – 294 ?M; 4k – 87 ?M. The most potent compounds exhibited inhibition activity in the two-digits micromolar range (4i, IC50 = 42 ?M and 4f, IC50 = 54 ?M). Despite the chemical diversity of aromatic ring substituents in the synthesized compounds, there is not a clear relationship between the substitution pattern at the aromatic ring and the experimental IC50 values obtained herein. For instance, the most promising RL_AeHKT inhibitor 4i has an iodine atom in the para position, indicating that an electronegative, soft, and bulky group at this position is favorable for inhibitory activity. However, inhibitors 4d and 4j containing bromine and chlorine atoms at the same aromatic ring position exhibited only moderate inhibition. Our previous findings indicate that chemical substitutions on the aromatic ring do not significantly affect the binding of 1,2,4-oxadiazole derivatives to HKT [10]. This is so because upon binding to the enzyme, the aromatic ring in the oxadiazole scaffold is positioned partially outside the active site pocket [10]. The present findings are consistent with our previous molecular docking calculations predicting that these compounds could be potential HKT inhibitors [10]. Kinetics assays were performed for four concentrations of our canonical compound 4a at different concentrations of the substrate 3-D,L-HK. The addition of compound 4a triggered changes which are representative of a noncompetitive mixed inhibition. The nonlinear regression fit of the raw data showed a KI = 360 ?M and ?KI = 1224 ?M, indicating that the inhibitor exhibits higher affinity for the free form of the enzyme (? > 1). The KI value for compound 4a is similar to the co-crystallized inhibitor from AgHKT, 4-OB (KI = 300 ?M) [9]. This means that the series of compounds related here have similar behavior as the reference compound, showing potential to guide improvements aiming the development of more potent HKT inhibitors.
CONCLUSIONS
We are on the way to get vaccines against arboviral infections (e.g. 17D yellow fever virus, dengue virus vaccine) [21,22]. However, population control remains the most effective route to dismiss the spread of most mosquito-borne diseases. We have previously identified the enzyme 3-hydroxykynurenine transaminase (HKT) as a potential target for 1,2,4-oxadiazole derivatives with larvicidal activity against Ae. aegypti [10,19]. In the present work, we established RL_AeHKT as a molecular target for 1,2,4-oxadiazoles larvicides acting by a noncompetitive inhibition mechanism. This has been accomplished through the construction and high-yield expression of a recombinant RL_AeHKT in E. coli, and the standardization of a fast and low-cost absorbance spectroscopy biochemical protocol to determine HKT activity. The presented findings attest to the inhibitory potential of 1,2,4-oxadiazole derivatives against the enzyme HKT, which is expected to lead to the accumulation of the highly toxic substrate 3-HK in Ae. aegypti larvae.
ACKNOWLEDGMENTS
Financial support for this research project was provided by the Brazilian funding agencies FACEPE (APQ-0732-1.06/14), BioMol/CAPES (BioComp 23038.004630/2014-35), INCT-Fx/CNPq (465259/2014-6) and FAPESP (2013/07600-3).
REFERENCES.
[1] 10.1016/j.antiviral.2009.10.008.
[2] 10.4269/ajtmh.17-0106.
[3] 10.1016/j.tmaid.2017.01.004.
[4] 10.1146/annurev- ento-010814-020828.
[5] 10.1080/10715760000300021.
[6] 10.1016/S0014- 5793(02)03229-5.
[7] 10.1016/j.jinsphys.2006.09.004.
[8] 10.1074/jbc.273.20.12003.
[9] 10.1073/pnas.0510233103.
[10] 10.1016/j.bmc.2013.09.020.
[11] 10.1111/j. 1742-4658.2006.05177.x.
[12] 10.1016/j.bmc.2019.115252
[13] 10.1021/cr60216a003.
[14] 10.1021/jm800639h.
[15] Tale RHH, Rodge AHH, Keche APP, et al. J Chem Pharm Res. 2011;3(2):496–505.
[16] 10.1016/j.bmc.2013.09.040.
[17] 10.1016/j.pep.2005.01.016.
[18] Novagen I. pET system manual. http://research.fhcrc.org/content/dam/stripe/ hahn/methods/biochem/pet.pdf. Published 1991. Accessed June 18, 2019.
[19] Lepkovsky S, Roboz E, Haagen-Smit AJ. J Biol Chem. 1943;149:195–201.
[20] 10.1016/j.jpba.2019.05.025.
[21] 10.1016/j.jinf.2016. 02.017.
[22] 10.1007/s40265-016-0626-8.
[23] 10.1016/j.bmc.2012.12.006.
Comissão Organizadora
Francisco Mendonça Junior
Pascal Marchand
Teresinha Gonçalves da Silva
Isabelle Orliac-Garnier
Gerd Bruno da Rocha
Comissão Científica
Ricardo Olimpio de Moura
Complete tool for submission, evaluation and publication of proceedings for scientific events.
