Validation of an HPLC Method for the Simultaneous Quantification of Metabolic Reaction Products Catalysed by CYP2E1 Enzyme Activity: Inhibitory Effect of Cytochrome P450 Enzyme CYP2E1 by Salicylic Acid in Rat Liver Microsomes

Inhibition of cytochrome P450 (CYP) alters the pharmacokinetic parameters of the drug and causes drug–drug interactions. Salicylic acid been used for the treatment of colorectal cancer (CRC) and chemoprevention in recent decades. Thus, the aim of this study was to examine the in vitro inhibitory effect of salicylic acid on CYP2E1 activity in rat liver microsomes (RLMs) using high-performance liquid chromatography (HPLC). High-performance liquid chromatography analysis of a CYP2E1 assay was developed on a reversed phase C18 column (SUPELCO 25 cm × 4.6 mm × 5 µm) at 282 nm using 60% H2O, 25% acetonitrile, and 15% methanol as mobile phase. The CYP2E1 assay showed a good linearity (R2 > 0.999), good reproducibility, intra- and inter-day precision (<15%), acceptable recovery and accuracy (80–120%), and low detection (4.972 µM and 1.997 µM) and quantitation limit values (15.068 µM and 6.052 µM), for chlorzoxazone and 6-hydroxychlorzoxazone, respectively. Salicylic acid acts as a mixed inhibitor (competitive and non-competitive inhibition), with Ki (inhibition constant) = 83.56 ± 2.730 µM and concentration of inhibitor causing 50% inhibition of original enzyme activity (IC50) exceeding 100 µM (IC50 = 167.12 ± 5.460 µM) for CYP2E1 enzyme activity. Salicylic acid in rats would have both low and high potential to cause toxicity and drug interactions with other drugs that are substrates for CYP2E1.


Introduction
Members of the cytochrome P450 (CP450) superfamily are known as phase 1 enzymes and play a key role in the biotransformation of a large number of endogenous (steroids, hormones, bile acids, fatty acids) and exogenous (toxic chemicals, drugs, carcinogens, environmental pollutants) compounds to a more hydrophilic form [1]. Statistical results showed that more than 90% of marketed drugs are metabolized by P450s [2]. However, with more than 10 forms identified, only the CYP1, CYP2, and CYP3 families are involved in the biotransformation of the majority of clinical drugs [2].
Because P450s play a vital role in drug metabolism, co-administration of a drug with another CYP substrate may alter their metabolism, thus causing a drug-drug interaction [2]. As a definition, drug interactions take place when a certain drug interacts with another drug [3]. These interactions can result in the changing activity of one or both drugs and lead to adverse side effects [4]. Statistical evidence has shown that nearly 20-40% of elderly people in developing countries have drug-drug interactions due to poly-therapy [5]. evidence has shown that nearly 20-40% of elderly people in developing countries have drug-drug interactions due to poly-therapy [5].
Understanding the properties of adverse drug reactions (ADRs) gives us a clear indication for quantification of the side effects of a prospective drug and a good knowledge of the pathogenic pathways involved during the interaction [5]. Nowadays, adverse drug reactions seem to be the main obstacle in clinical trials, slowing down the recovery of patients in hospitals [5].
It has been reported by Badyal and Dadhich (2001) that drug-drug interactions could be the fourth or the sixth leading cause of deaths in United States. Drug-drug interactions can be the result of inhibition and induction of P450 enzyme activity [3]. Potential drug-drug interactions appear to be the most challenging part of clinical practice, resulting in pharmacokinetic and pharmacodynamic variations of the drugs and changing their overall therapeutic response [6]. On the contrary, a clear understanding in both drug metabolism and drug interaction has resulted from allotting P450 enzymes based on their amino acid chain [7]. Many drugs have been withdrawn from U.S. market, e.g., terfenadine, astemizole, and cisapride, because the inhibition of these drugs by other drugs results in arrhythmias [8].
Relevant studies have suggested that potential drug-drug interactions (pDDIs) are more likely to occur in persons suffering from cardiovascular disease compared to other major diseases [6]. The explanation for this clinical observation can refer primarily to many factors such as age, combination therapy, and the effective therapeutic management of drugs used in cardiology [6]. In fact, mibefradil, a calcium channel blocker, was withdrawn from U.S. market since it acts as a potent inhibitor that increases the toxicity level of certain cardiovascular drugs [8]. Specific enzyme activity and metabolite formation can be predicted.
Aspirin (acetyl salicylic acid) is a nonsteroidal, anti-inflammatory drug used for the treatment of fever, pain, and inflammation [9]. It is hydrolysed to salicylic acid in the liver, intestine, and plasma by means of the esterase enzyme [9]. Salicylic acid ( Figure 1) is a natural compound obtained from wintergreen leaves and white willow trees, and possesses many beneficial pharmacological activities, such as anti-inflammatory, antioxidant, and vasodilator effects [9]. It has been reported by Danchineni (2017) that salicylic acid potentially targets tumour cells by downregulating the function of cyclin A2, B1, D3, and cyclin-dependent kinases (CDKs 1,2,4, and 6) to exert a chemopreventive effect. Recently, inhibition studies demonstrated that CYP2E1 enzyme is responsible for the hydroxylation of salicylic acid in humans [9]. However, there is a lack of information concerning the effect of salicylic acid on other CYPs. There is thus an urgent need for a detailed study on the inhibitory effect of salicylic acid on other P450s enzyme activities. Our previous in vitro inhibition study showed that salicylic acid acts as a non-competitive inhibitor for CYP2C11 enzyme activity [10]. This means that salicylic acid has a low potency to cause drug interactions with other drugs that are substrates for the CYP2C11 enzyme.
The aim of this study is to investigate the inhibitory effect of salicylic acid on the metabolism of the CYP2E1 isoform in male rat liver microsomes. In this systematic study, the in vitro inhibitory effect of salicylic acid on CYP2E1 enzyme activity was evaluated to determine the potency of salicylic acid in affecting CYP mediated phase 1 metabolism in male rat microsomes, employing However, there is a lack of information concerning the effect of salicylic acid on other CYPs. There is thus an urgent need for a detailed study on the inhibitory effect of salicylic acid on other P450s enzyme activities. Our previous in vitro inhibition study showed that salicylic acid acts as a non-competitive inhibitor for CYP2C11 enzyme activity [10]. This means that salicylic acid has a low potency to cause drug interactions with other drugs that are substrates for the CYP2C11 enzyme.
The aim of this study is to investigate the inhibitory effect of salicylic acid on the metabolism of the CYP2E1 isoform in male rat liver microsomes. In this systematic study, the in vitro inhibitory effect of salicylic acid on CYP2E1 enzyme activity was evaluated to determine the potency of salicylic acid in affecting CYP mediated phase 1 metabolism in male rat microsomes, employing chlorzoxazone as a probe substrate for the CYP2E1 enzyme in the presence of different concentrations of salicylic acid.
The following figure demonstrates the measurements of maximum wavelength of each compound in the CYP2E1 assay.
According to the overlain spectra ( Figure 2), it is perceived that the maximum of the absorption band for the four components in CYP2E1 assay is 282 nm.
Molecules 2020, 25, 932 3 of 18 chlorzoxazone as a probe substrate for the CYP2E1 enzyme in the presence of different concentrations of salicylic acid.
The following figure demonstrates the measurements of maximum wavelength of each compound in the CYP2E1 assay.
According to the overlain spectra ( Figure 2), it is perceived that the maximum of the absorption band for the four components in CYP2E1 assay is 282 nm.

Specificity and Selectivity
Specificity was achieved by choosing the right mobile phase composition (60% H2O, 25% acetonitrile, and 15% methanol). It results in a good separation of the salicylic acid peak from the CYP2E1 metabolite (6-hydroxychlorzoxazone) peak at T = 25 °C using C18 (SUPELCO 25 cm × 4.6 mm, 5 µm), at a 0.7 mL/min flow rate and wavelength of λ = 282 nm. In order to determine the retention time of each compound in the CYP2E1 assay, each compound was run separately using phenacetin as an internal standard on the HPLC instrument using isocratic elution programming. Figure 3 shows the chromatogram of the response peaks of four compounds (salicylic acid, chlorzoxazone, 6-hydroxychlorzoxazone, and phenacetin) in the conditions mentioned above.

Specificity and Selectivity
Specificity was achieved by choosing the right mobile phase composition (60% H 2 O, 25% acetonitrile, and 15% methanol). It results in a good separation of the salicylic acid peak from the CYP2E1 metabolite (6-hydroxychlorzoxazone) peak at T = 25 • C using C18 (SUPELCO 25 cm × 4.6 mm, 5 µm), at a 0.7 mL/min flow rate and wavelength of λ = 282 nm. In order to determine the retention time of each compound in the CYP2E1 assay, each compound was run separately using phenacetin as an internal standard on the HPLC instrument using isocratic elution programming. Figure 3 shows the chromatogram of the response peaks of four compounds (salicylic acid, chlorzoxazone, 6-hydroxychlorzoxazone, and phenacetin) in the conditions mentioned above.

Linearity and Range
Different solution concentrations of chlorzoxazone (0, 25, 50, 100, 150, 200, 300, and 400 µM) and its metabolite 6-hydroxychlorzoxazone (0, 10, 20, 40, 60, 80, and 100 µM) were injected into the HPLC instrument for the determination of the standard calibration curves. Calibration curves were assessed by plotting the mean area peak of standards and phenacetin (internal standard) (50 µM) versus the concentration of standard (chlorzoxazone and 6-hydroxychlorzoxazone). The outcomes are listed in Table 1 and show good linearities for both chlorzoxazone and 6-hydroxychlorzoxazone, with r 2 values of 0.9997 and 0.9994, respectively. The linear regression coefficient was within the acceptable fit (r 2 > 0.99) according to the International Conference On Harmonization (ICH) guidelines. The % RSD (relative standard deviation at each different concentration (% RSD < 5%)) met the ICH guidelines. The results, presented in Table 2, demonstrate that both chlorzoxazone and 6hydroxychlorzoxazone have low detection and quantitation limit values, consistent with the literature [12].   (5) chlorzoxazone, respectively.

Linearity and Range
Different solution concentrations of chlorzoxazone (0, 25, 50, 100, 150, 200, 300, and 400 µM) and its metabolite 6-hydroxychlorzoxazone (0, 10, 20, 40, 60, 80, and 100 µM) were injected into the HPLC instrument for the determination of the standard calibration curves. Calibration curves were assessed by plotting the mean area peak of standards and phenacetin (internal standard) (50 µM) versus the concentration of standard (chlorzoxazone and 6-hydroxychlorzoxazone). The outcomes are listed in Table 1 and show good linearities for both chlorzoxazone and 6-hydroxychlorzoxazone, with r 2 values of 0.9997 and 0.9994, respectively. The linear regression coefficient was within the acceptable fit (r 2 > 0.99) according to the International Conference on Harmonization (ICH) guidelines. The % RSD (relative standard deviation at each different concentration (% RSD < 5%)) met the ICH guidelines.

Limit of Detection and Limit of Quantitation (LOD and LOQ)
The results, presented in Table 2, demonstrate that both chlorzoxazone and 6-hydroxychlorzoxazone have low detection and quantitation limit values, consistent with the literature [12].

Intra-assay Variation of Chlorzoxazone
Intra-assay variation of chlorzoxazone was determined by measuring three concentration levels (200, 100, 25 µM, or high, medium, low, respectively) three times in a single batch (n = 3). Mean activity was calculated from the following calibration curve linear equation: y = 0.0544 x + 0.0626 (r 2 = 0.9997). The outcomes, summarized in Table 3, demonstrate that the relative standard deviation or % RSD (percentage of relative standard deviation) was <5% for chlorzoxazone. The experiment revealed that there was no large variation in the intra-assay experiment. Table 3. Intra-assay variation for chlorzoxazone (n = 3).

Intra-Assay Variation of 6-hydroxychlorzoxazone Metabolite
Determination of intra-assay variation of the metabolite was assessed by injecting three levels of 6-hydroxy chlorzoxazone concentrations (low, moderate, high) into a HPLC instrument three times in a single batch (n = 3). Mean activity was calculated from the following calibration curve linear equation: y = 0.0164 x − 0.0021 (r 2 = 0.9994). The outcomes, summarized in Table 4, illustrate that the relative standard deviation or % RSD (percentage of relative standard deviation) was <5% for 6-hydroxychlorzoxazone. The experiment revealed that there is no large variation in the intra-assay experiment. Table 4. Intra-assay variation of CYP2E1 enzyme metabolite (6-hydroxychlorzoxazone) (n = 3).

Inter-Assay Variation of Chlorzoxazone
Inter-assay variation was determined by measuring chlorzoxazone standards of three concentrations levels (200, 100, 25 µM, or high, medium, low, respectively) for three consecutive days, in separate batches. Mean activity was calculated from the following calibration curve linear equations on days 1, 2, and 3: y = 0.0544 x + 0.0626 (r 2 = 0.9997) for day 1, y = 0.0522 x + 0.074 (r 2 = 0.9994) for day 2, and y = 0.024 x − 0.0268 (r 2 = 0.9994) for day 3. The results, summarized in Table 5, illustrate that the relative standard deviation or % RSD (percentage of relative standard deviation) was <10% for chlorzoxazone. The experiment revealed that there is no large variation between aliquots of the same batch sample in the inter-assay experiment. Inter-Assay Variation of 6-Hydroxychlorzoxazone Assessment of inter-assay variation of 6-hydroxychlorzoxazone was carried out by injecting three concentration levels of 6-hydroxychlorzoxazone (low, moderate, and high) for three consecutive days in separate batches (n = 3). Mean activity was calculated from the following calibration curve linear equations on days 1, 2, and 3: y = 0.0164 x − 0.0021 (r 2 = 0.9994) for day 1, y = 0.0248 x − 0.0048 (r 2 = 0.9999) for day 2, and y = 0.0277 x − 0.0141 (r 2 = 0.9996) for day 3. The outcomes are shown in Table 6, and illustrate that the relative standard deviation or % RSD was <10% for 6-hydroxychlorzoxazone. The experiment revealed that there is no variation between aliquots of the same batch sample in the inter-assay experiment. Table 6. Inter-assay performance of 6-hydroxychlorzoxazone.

Stability Test
Stability Test for Chlorzoxazone The stability of chlorzoxazone was investigated for three different concentrations (25, 100, and 200 µM) stored for 72 h at room temperature in natural light conditions. Phenacetin (used as an internal standard of 50 µM) was added to each batch. Each sample was analysed in triplicate (n = 3) for each Molecules 2020, 25, 932 7 of 18 batch. The calibration curve of chlorzoxazone was run at t = 0, t = 24, t = 48, and t = 72 h. The stability test results are summarized in the Table 7, below: The outcomes in Table 7 [12].
Stability Test for 6-hydroxychlorzoxazone Stability of 6-hydroxychlorzoxazone was investigated for three different concentrations (10, 40, and 80 µM) stored for 72 h at room temperature in natural light conditions. Phenacetin (used as an internal standard of 50 µM) was added to each batch. Each sample was analysed in triplicate (n = 3) for each batch. The calibration curve of 6-hydroxychlorzoxazone was run at t = 0, t = 24, t = 48, and t = 72 h. The stability test results are summarized in the Table 8, below. The outcomes in Table 8         Studying CYP enzyme inhibition is considered as an important route in evaluating drug-drug interactions in the pharmaceutical field and in the drug development process. Several marketed drugs have been withdrawn from the market in the past few decades because CYP enzyme inhibition causes harmful drug-drug interactions [1]. To our knowledge, this systematic study seems to be the first time that the in vitro inhibitory effect of salicylic acid on the CYP2E1 probe substrate metabolism has been investigated.
The CYP2E1 isoform appears to be the most vulnerable when salicylic acid's modulatory effect on each individual isoform is considered. Based on its characteristics, the CYP2E1 isoform has a broad substrate specificity, an ability to metabolise major marketed drugs, and is responsible for the catalysis of many carcinogens and poisons [2].
Only one enzyme exists in the CYP2E isoform and this is known as the CYP2E1 isozyme [14]. CYP2E1 has the ability to metabolize nitrosamines and short-chains such as ethanol, toluene, and paracetamol, and many other anaesthetics actively [15]. Therefore, the CYP2E1 enzyme can be included only in the metabolism of lower molecular weight drugs such as styrene, vinyl chloride, and benzene that are involved in the synthesis of many households cleaning products [16]. The substrates recognized by the CYP2E1 enzyme are chlorzoxazone, p-nitrophenol, coumarin, quinolone, and caffeine [17]. CYP2E1 can activate some of the above carcinogenic substances [14]. Bibi (2008) claimed that the expression of CYP2E1 might differ between genders. Likewise, CYP2E1 activity can be also affected by health conditions such as obesity and fasting, which may provide an apparent clarification for obesity associated with cancer [14].
In vitro evidence has also revealed that garlic and watercress serve as potent inhibitors of the CYP2E1 enzyme in the human body [18]. In addition, nervous system agents, such as isoflurane, ethanol, aflatoxin B1, balothane, and chlorzoxazone, can be preferably metabolised by the CYP2E1 enzyme [1].
Recent studies have suggested that the sensitivity of the CYP2E1 enzyme activity can be decreased by certain drugs and food stuffs for example green and black powder tea, ellagic acid, chrysin, N-acetyl cysteine, and dandelion [18]. Our previous study showed that salicylic acid acts as a non-competitive inhibitor for CYP2C11 enzyme activity [10]. Thus, salicylic acid has a low potency to cause toxicity and drug interaction with other drugs that are substrates for the CYP2C11 enzyme [10]. In this study (Figure 4; Figure 5), it is evident, based on the in vitro inhibition kinetic parameters in rat liver microsomes and Lineweaver plot shapes, that salicylic acid inhibited CYP2E1 activities with a mixed-type inhibition mode (competitive and non-competitive) with inhibition constant, K i = 83.56 ± 2.730 µM and a 50% inhibitory concentration (IC 50 ) value exceeding 100 µM (IC 5 0 = 167.12 ± 5.460 µM). In addition, salicylic acid at its concentration of 25 µM can considered as a saturated concentration, since its Michaelis constant K m value represents the highest value compared to other salicylic acid concentrations. According to Table 9, K m and the maximum rate of the reaction (V max ) hanged between each concentration of salicylic acid, and thus salicylic acid acts as a mixed inhibition for CYP2E1 enzyme activity. Thus, in this study it is considered that the in vitro inhibitory effect of salicylic acid on CYP2E1 enzyme activity will be beneficial for a future in vivo study for healthcare screening of the effective use of salicylic acid in clinics and the safe administration of salicylic acid with other drugs, consistent with in vivo drug-drug interactions pharmacokinetic parameters. Some further investigational in vitro and in vivo studies in human CYP450s are needed due to the varied CYP2E isoform expression in species.

Rat Liver Microsomes
Microsomes from liver, pooled from male rat (Sprague-Dawley) in this study were purchased from Merck (Old Brickyard, Gillingham, UK) and stored at -80 • C. The manufacturer had previously characterized the microsomes for CYP2E1 activity, CYP450 content, and protein concentrations.

Instruments
A Shimadzu LC-2010A HT high performance liquid chromatography system equipped with a low-pressure pump quaternary gradient (series 200 LC pump), a series 200 Peltier LC column oven for chromatographing the analysed solutions, a series 200 autosampler (Shimadzu, Tokyo, Japan), a degasser, and a model series 200 UV detector were used. The data were processed using the Shimadzu HPLC 2 Data Lab Solutions (Kingston University, UK) software processing system. A 5-µm C 18 SUPLCO column (25 cm × 4.6 mm) was employed from Merck (Old Brickyard, Gillingham, UK) to separate target analytes using a low-pressure isocratic elution system. A UV-VIS Spectrometry instrument with a 1-cm-length quartz cuvette was purchased from VWR International Ltd. (Magna Park, Lutterworth, Leicestershire, LE17 4XN, UK) and UV spectra were processed using a Bio 100 Cary software from Agilent Technologies LDA (Cheadle Royal Business Park, Stockport, Cheshire, SK8 3GR, UK). A 570 pH meter, purchased from JENWAY Limited (Beacon Road, Stone, Staffordshire, ST15 0SA, UK) was used.

CYP2E1 Substrate and its Metabolite
Validations were conducted using HPLC system LC-2010A HT Module (Shimadzu, Tokyo, Japan). Chromatographic experiments were evaluated in a low-pressure isocratic system. The separations of the four target components (salicylic acid as a tested inhibitor, phenacetin as an internal standard, chlorzoxazone as CYP2E1 substrate, and 6-hydroxychlorzoxazone as CYP2E1 metabolite) were injected into a SUPELCO C18 column (25 cm × 4.6 mm, 5 µm particle size). The mobile phase for the chromatographic separation of the four compounds was made up as follows: (A): 60% H 2 O, (B): 15% of methanol, and (C): 25% of acetonitrile. The flow rate was set to 0.7 mL/min, and the oven temperature was set at 25 • C. Wavelength detection was set to 282 nm. An injection volume of 10 µL was used. The mobile phase consisted of methanol: water: acetonitrile (15:60:25 % v/v). This provided a good separation and resolution of CYP2E1 components.

Inhibition of CYP2E1 Enzymatic Activity Assay
CYP2E1 activity was assessed using chlorzoxazone (substrate) via the formation of 6-hydroxychlorzoxazone (the metabolite for the CYP2E1 enzyme). Oxidative metabolism of chlorzoxazone was evaluated using a NADPH-regenerating system consisting of: 1.0 mM nicotinamide adenine dinucleotide phosphate hydrogen (NADPH), 5 mM glucose-6-phosphate (G6P), 1.7 units/mL glucose-6-phosphate dehydrogenase (G6PDH), 1.0 mM ethylenediamine tetraacetic acid (EDTA), and 3.0 mM magnesium chloride. A 500-µL incubation volume containing NADPH-regenerating system and a final concentration of 0.067 M potassium phosphate buffer at pH = 7.4 was incubated in triplicate. The incubation mixture consisted of 0.5 mg/mL of pooled liver microsomes with a serial range of chlorzoxazone concentrations (25, 50, 100, 150, and 200 µM, dissolved in mobile phase) and a serial range of salicylic acid concentrations (0, 10, 25, 40, and 60 µM, dissolved in mobile phase). Pre-incubation of all components was done for 10 min in a water bath at T = 37 • C before the addition of nicotinamide adenine dinucleotide phosphate (NADP + ) into the mixture [19]. The final concentration of organic solvent did not exceed the 1% v/v. Tubes were incubated for 40 min in an Eppendorf thermomixer (Eppendorf UK Limited, Stevenage) at 150 rpm (37 • C). Termination of the reaction was carried out by adding ice-cold acetonitrile containing 50 µM of phenacetin as an internal standard at each time interval (10,15,20,25,30,35, and 40 min). Tubes were centrifuged in a microcentrifuge (13,000× g) for 12 min to precipitate protein. Then the supernatant was collected and dissolved in a mobile phase (60% H 2 O, 15% methanol, and 25% acetonitrile) made up to 1000 µL volume. A 10-µL injection volume of the dissolved supernatant was used for HPLC analysis.

Data Analysis:
Standard/calibration curves consisted of different concentration ranges of chlorzoxazone, 6-hydroxychlorzoxane, and a fixed concentration of phenacetin (50 µM) as an internal standard. Validation parameters (Range, LOD, LOQ, % error, recovery, and accuracy) were calculated using Microsoft Excel 2010 software. All results are presented as mean ± S.D.
Formation of the 6-hydroxychlorzoxazone metabolite of the tested CYP2E1 substrate (chlorzoxazone) was measured for CYP inhibition analysis. Microsoft Excel 2010 software was used for determination of area peak ratios of both metabolite and internal standard. Pharmacokinetic parameters (V m , K m , Cl int (hepatic intrinsic clearance), K i ) were determined by nonlinear regression analysis from secondary Lineweaver-Burk and Michaelis-Menten plots of the enzyme activity-metabolite concentration data. The type of CYP2E1 inhibition was assumed to be mixed inhibition based on the shape of Lineweaver-Burk plots, the standard error, the Akaike information criterion (AIC), and Schwarz criterion (SC) values, which were fitted by a simple non-linear regression model. A non-linear regression model was chosen as a better fit because of its low AIC value. The Akaike weights (∆AIC) are the relative difference between the best model and each other model in the set. The ∆AIC provided substantial evidence for the model (∆AIC < 2). The 50% inhibitory concentration (IC 50 ) value was calculated by using the following mixed inhibition equation: where V is the observed velocity, V 0 is uninhibited velocity, S is slope factor, and I is the inhibitor concentration.

Conclusions
In conclusion, all the analytical parameters (accuracy, precision, % error, % recovery, LOD, LOQ, linear regression) for chlorzoxazone and 6-hydroxychlorzoxazone were in line with ICH guidelines. The in vitro data provided allow us to understand the drug interactions with salicylic acid. Nevertheless, our findings indicated that salicylic acid potentially acts as a mixed inhibitor (competitive and non-competitive inhibitor) for CYP2E1 enzyme activity in rat liver microsomes. This finding provides useful data for the efficacy and the safe use of salicylic acid in clinical practice. However, the extent to which our obtained data can be related to human CYP isoforms is unknown. The majority of clinically important drugs such as S-warfarin, steroids, and retinoic acid are metabolised by both rat CYP2C11 and human CYP2C9 that display a high identity of amino acid sequences. Thus, it is primarily important to investigate whether human CYP isoforms are induced by salicylic acid, and further verification of the obtained in vitro data should be done with in vivo experiments.
Author Contributions: All authors designed, conceived, and performed the experiments. All authors contributed to formal data analysis, interpretations, and writing of the original draft. All authors have read and agreed to the published version of the manuscript.
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Robustness studies were performed by changing the flow rate, and evaluate the variation in retention time and area peak of each component for CYP2E1 assay . Tables A3 and A4 show the effect of flow rate on both retention time and area peak of each compound. According to Table A3, better resolution obtained at flow rate = 0.7 mL/min compared to flow rate = 0.9 mL/min. In addition to that, area peak remains unaffected by small for all components of CYP2E1 assay.
When the flow rate decreased to 0.5 mL/min, Salicylic acid and 6-hydroxychlorzoxazone peaks were poorly resolved (Difference = 1.2323 min) compared to a flow rate = 0.7 mL/min (Difference = 1.8183 min). Additionally, area peak of each CYP2E1 compounds remains unaffected by little as shown in Table A4. Table A4. Variation in retention time, and area peak between compounds when changing the flow rate. Robustness studies were performed by changing the column Temperature, and evaluate the variation in retention time and area peak of each component for CYP2E1 assay . Tables A5 and A6 shows the effect of temperature on both retention time and area peak of each compound.  Table A6 shows the variation of area peak, retention time of CYP2E1 compounds when the column Temperature increased from 25 • C to 35 • C.

Mobile Phase composition
Overall, area peak, retention time remains unaffected for all CYP2E1 compounds when changing one of the HPLC parameters. Therefore, this method showed a robust methodology.