Quantitation of Phenolic Benzotriazole Class Compounds in Plasma by Liquid Chromatography–Tandem Mass Spectrometry (LC-MS/MS)

Abstract

Phenolic benzotriazoles are used as UV stabilizers in consumer products and have been detected in the environment suggesting potential human exposure. Phenolic benzotriazoles were nominated to the Division of National Toxicology Program for testing based on their potential widespread human exposure and lack of adequate toxicity data. Nine chemicals were selected for toxicological evaluation, representing unsubstituted (2-(2H-benzotriazole-2-yl)phenol, (P-BZT)), monosubstituted (drometrizole; 2-(2H-benzotriazol-2-yl)-4-tert-butylphenol (tBu-BZT); octrizole), disubstituted (2-(2H-benzotriazol-2-yl)-4,6-bis(1-methyl-1-phenylethyl)phenol (diMeEtPh-BZT), 2-(2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylpropyl)phenol (ditPe-BZT); 3-(2H-benzotriazol-2-yl)-5-(1,1-dimethylethyl)-4-hydroxybenzenepropanoic acid, octylester (tBuPrOcEst-BZT) and halogenated trisubstituted (bumetrizole; 2-(5-chloro-2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylethyl)phenol (ditBuCl-BZT)) compounds. Different extraction methods were utilized and methods were developed to analyze phenolic benzotriazoles by quantitating free (unconjugated parent) and total (free and conjugated parent) analyte levels in plasma of rats to aid in interpretation of toxicity data, understanding of absorption, distribution, metabolism, and excretion differences. The calibration standard range was 1–500 ng/mL for free analytes and 1–1000 ng/mL for total analytes. The methods were linear (r² ≥ 0.99). The accuracy was determined as relative error (RE) and ranged from −18.2 to +17.8, and precision was determined as relative standard deviation (RSD) and ranged from 0.0 to 20.1% for both free and total plasma calibration standards, respectively. The limit of quantitation was ≤ 5.0 and 10.0 ng/mL and limit of detection was ≤ 1.2 and 2.0 ng/mL, for free and total analytes, respectively. These data demonstrate that the methods are suitable for quantitation of free and total analytes in rat plasma.

Introduction

Phenolic benzotriazoles are a class of chemicals that are also known as ultraviolet stabilizers/absorbers. They are used in a variety of consumer products and industrial applications including polymers, plastics, paints, orthodontic adhesives, and personal care products such as cosmetics and sunscreens (Cantwell et al. 2015). Phenolic benzotriazoles have high lipophilicity (K_ow> 4) and hence high bioaccumulation potential (Lopez-Avila and Hites 1980; Kim, Isobe, et al. 2011; Brandt et al. 2016; Chan and Webster 2019). Compounds belonging to this class are considered as persistent (Lopez-Avila and Hites 1980; Pruell et al. 1984; Health Canada 2016) and have been detected in water (Nakata Haruhiko et al. 2009; Shi et al. 2019), sediment (Reddy et al. 2000; Nakata H. et al. 2010), soil (Shi et al. 2019), various marine organisms (Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Nakata H. et al. 2012), birds (Nakata Haruhiko et al. 2009), and mammals (Nakata H. et al. 2010) suggesting potential human exposure.

There are limited toxicology studies available for phenolic benzotriazole class chemicals. Some are reported to be toxic to aquatic organisms (Lopez-Avila and Hites 1980; Kim, Chang, et al. 2011; Cantwell et al. 2015). The oral LD₅₀ values were ≥1000 mg/kg in mice and rats depending on the compound (Thomas et al. 1995). Sub-chronic and chronic exposure studies in rats and mice following oral exposure to phenolic benzotriazoles resulted in increased liver weights, body weight changes, histopathological changes, altered liver enzyme content, hematological effects, and relative organ weights (Thomas et al. 1995; Hirata-Koizumi et al. 2007; Burnett 2008; Hirata-Koizumi, Matsuyama, et al. 2008a, 2008b; Hirata-Koizumi, Ogata, et al. 2008; Hirata-Koizumi et al. 2009). Reproductive and teratological studies were also reported following oral exposure to several phenolic benzotriazoles ((2-(2H-benzotriazol-2-yl)-4,6-bis(1-methyl-1-phenylethyl)phenol (diMeEtPh-BZT), drometrizole, 2-(5-chloro-2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylethyl)phenol (ditBuCl-BZT)) in female rats and female mice (Thomas et al. 1995; Ema et al. 2006; Ema et al. 2008). While some phenolic benzotriazoles (i.e., diMeEtPh-BZT, drometrizole) did not have any reproductive effects (U. S. EPA 2010b, 2010a), exposure to other phenolic benzotriazoles (i.e., 2-(2′-hydroxy-3′,5′ -di-tert-butylphenyl)benzotriazole (HDBB), ditBuCl-BZT) resulted in decreased pup body weight, increased liver weight, decreased fetal body weight, and increased incidence in skeletal maturation delay (Ema et al. 2008; Hirata-Koizumi, Ogata, et al. 2008).

Analytical methods to quantify phenolic benzotriazoles in marine biota (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012) and environmental samples (Zhang et al. 2011; Montesdeoca-Esponda et al. 2012) have been reported in the literature. In general, methods utilized liquid (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012) (Zhang et al. 2011) and/or solid phase extraction (Zhang et al. 2011; Montesdeoca-Esponda et al. 2012) followed by further purification using silica gel chromatography (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012) and/or gel permeation chromatography (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012) to separate lipids prior to analysis by either gas chromatography- (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012) or liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) (Kim, Chang, et al. 2011; Zhang et al. 2011; Montesdeoca-Esponda et al. 2012). While there are several published methods for the analysis of some phenolic benzotriazoles from environmental matrices and in various aquatic species, to the best of our knowledge, there are few of methods (Thomas et al. 1995; Hirata-Koizumi et al. 2009) available to quantitate phenolic benzotriazoles in rodent plasma or blood. One method (Thomas et al. 1995) utilized a solvent extraction followed by LC-MS/MS to analyze HDBB in rat plasma with a limit of quantitation (LOQ) of 20 ng/mL. A second method (Hirata-Koizumi et al. 2009) utilized liquid extraction followed by thin layer chromatography and radiometric scanning to quantify 3-[3-(2H-benzotriazole-2-yl)-5-tertbutyl-4-hydroxyphenyl] propionic acid and its methyl ester in blood, however, the details of the analysis method and LOQ were not provided.

Phenolic benzotriazoles were nominated to the Division of National Toxicology Program (DNTP) for testing based on their potential human exposure and lack of adequate toxicity data (NTP 2020). Nine chemicals were selected for testing in male Hsd:Sprague Dawley^® SD^® (HSD) rats, representing unsubstituted (2-(2H-benzotriazole-2-yl)phenol, (P-BZT)), monosubstituted (drometrizole; 2-(2H-benzotriazol-2-yl)-4-tert-butylphenol (tBu-BZT); octrizole), disubstituted diMeEtPh-BZT, 2-(2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylpropyl)phenol (ditPe-BZT); 3-(2H-benzotriazol-2-yl)-5-(1,1-dimethylethyl)-4-hydroxybenzenepropanoic acid, octylester (tBuPrOcEst-BZT) and halogenated trisubstituted (bumetrizole; ditBuCl-BZT) compounds (Figure 1). Phenolic benzotriazoles that were not commercially available were custom synthesized. In this study, we report LC-MS/MS methods to quantify free (unconjugated parent) and total (free and conjugated parent) analytes of these nine chemicals in rodent plasma to aid the interpretation of toxicology studies. Furthermore, the methods developed in this study were later modified for the analysis of free analytes for a class comparison of phenolic benzotriazoles in a toxicokinetic study (Waidyanatha et al. 2021).

Figure 1.

Structures of nine phenolic benzotriazoles compounds selected for testing; unsubstituted (2-(2H-benzotriazole-2-yl)phenol, (P-BZT)), monosubstituted (drometrizole; 2-(2H-benzotriazol-2-yl)-4-tert-butylphenol (tBu-BZT); octrizole), disubstituted (2-(2H-benzotriazol-2-yl)-4,6-bis(1-methyl-1-phenylethyl)phenol (diMeEtPh-BZT), 2-(2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylpropyl)phenol (ditPe-BZT); 3-(2H-benzotriazol-2-yl)-5-(1,1-dimethylethyl)-4-hydroxybenzenepropanoic acid, octylester (tBuPrOcEst-BZT) and halogenated trisubstituted (bumetrizole; 2-(5-chloro-2H-benzotriazol-2-yl)-4,6-bis(1,1-dimethylethyl)phenol (ditBuCl-BZT))

Materials and Methods

Chemicals and Reagents

Phenolic benzotriazoles were procured from various sources and the details of the compounds including abbreviations used in the manuscript are provided in Table S1. 3-(3-(2HBenzo[d][1,2,3]triazol-2-yl)-5-(tert-butyl)-4-hydroxyphenyl)propanoic acid (tBuPrOH-BZT, Lot No. GC-08152011) was procured from Adesis (New Castle, DE). β- glucuronidase from Helix pomatia was procured from Sigma Aldrich (Saint Louis, MO). The chemicals were used to prepare the calibration standards, quality control (QC) samples, QC stability samples, and internal standards. Male Sprague Dawley (SD) rat plasma containing potassium ethylenediaminetetraacetic acid (K₃EDTA) as an anticoagulant was received from BioIVT (Westbury, NY) and was used to prepare plasma calibration standards, QC standards, and plasma blanks. HSD adult male rat plasma purchased from Bioreclamation IVT (Westbury, NY) and was used to prepare plasma QC stability standards.

Analytical Method Development

Initially, we planned to develop a single method for all nine compounds. Based on physicochemical properties of the compounds, liquid extraction followed by LC-MS/MS was used for quantitation of free and total analytes. Deconjugation with β-glucuronidase from Helix pomatia containing both glucuronidase and a sulfatase activity was used to release conjugated forms of parent for total analysis. Protein precipitation was evaluated first as the simplest extraction technique. Matrix effect was evaluated by comparing extracted matrix standards to extracted solvent standards. Based on the recovery and matrix effect for some of the compounds when utilizing protein precipitation, it was determined a more thorough clean-up procedure was needed. As a result, both solid phase extraction (SPE) and liquid to liquid extraction (LLE) methods were evaluated. Based on these results, it was determined that a single method could not be found for all compounds with the recovery and sensitivity required. Therefore, different extractions techniques that were optimal for each analyte were used instead, including protein precipitation, SPE, or LLE. The selected method for each analyte is shown Table S2. All samples were analyzed using one general LC-MS method. During the total analysis of tBuOcEst-BZT, it was discovered that the ester of the OcEst-BZT was not stable during handling and hydrolyzed to the acid (Figure 2) during extraction. Therefore, a method was developed for the analysis of tBuPrOH-BZT, the acid of tBuOcEst-BZT.

Figure 2.

Hydrolysis of 3-(2H-benzotriazol-2-yl)-5-(1,1-dimethylethyl)-4-hydroxybenzenepropanoic acid, octylester (tBuPrOcEst-BZT) to 3-(2HBenzo[d][1,2,3]triazol-2-yl)-5-(tert-butyl)-4-hydroxyphenyl)propanoic acid

Linearity and intraday accuracy (as measured by percent relative error (RE)) and precision (as measured by relative standard deviation (RSD) of the methods were evaluated by analyzing duplicate calibration standards prepared in SD male rat plasma. The concentration range for calibration standards is provided in Tables 1 and 2. Four replicates of low, and high plasma quality control (QC) samples were also evaluated for intraday accuracy (RE) and precision (RSD) of the method (Tables 1 and 2). Sensitivity of the methods was estimated by analyzing duplicate aliquots prepared at the lower limit of quantitation (LOQ, low end of the curve) for free and total analytes. The limit of detection (LOD) was calculated as three times the standard deviation of the LOQ QC samples.

Table 1.

Method Qualification Data for Free Analytes in Male SD Rat Plasma

Compound	Concentration Range (ng/mL)	LODa (ng/mL)	LOQb (ng/mL)	Standardc		QCd
				Intra-day Accuracy (RE, %)e	Intra-day Precision (RSD, %)f	Intra-day Accuracy (RE, %)e	Intra-day Precision (RSD, %)f
P-BZT	2–200	0.40	2.0	−11.2 to 10.9	2.8 to 20.7	−6.5 to 5.6	5.8 to 6.8
Octrizole	5–100	0.17	5.0	−17.0 to 14.8	1.4 to 9.1	−11.5 to 26.8	1.6 to 18.9
Drometrizole	1–100	0.60	1.0	−4.8 to 3.2	1.9 to 20.1	−12.9 to 13.9	3.2 to 10.2
tBuPrOcEst-BZT	1–500	NA	1.0	−9.0 to 15.8	0.9 to 9.0	16.3 to 32.3	8.9 to 14.3
Bumetrizole	1–100	1.21	2.0	−7.8 to 4.0	12.7 to 19.3	−37.3 to 33.5	7.0 to 34.7
DiMeEtPh-BZT	1–100	0.03	1.0	−14.1 to 10.6	1.2 to 15.2	−28.3 to 16.3	12.3 to 23.8
ditBuCl-BZT	1–100	0.13	1.0	−4.9 to 6.0	0.5 to 6.4	3.0 to 5.5	2.9 to 4.4
DitPe-BZT	5–500	1.13	5.0	−2.0 to 2.2	0.1 to 7.5	−6.8 to 0.8	3.3 to 7.2
tBu-BZT	1–100	0.48	1.0	−6.5 to 6.9	0.7 to 15.2	−2.4 to 8.5	6.5 to 11.9

^aLOD= Limit of Detection; 3×Standard Deviation for 6 lower limit of quantitation (LOQ) replicates

^bLOQ=lower limit if quantitation; is the lowest concentration of the calibration curve

^cEstimated based on plasma calibration standards, n = 2

^dEstimated based on low and high-level Quality Control (QC) samples, n = 4; 5 and 200 ng/mL for P-BZT; 9 and 75 ng/mL for Octrizole, 2 and 50 ng/mL drometrizole, bumetrizole, DiMeEtPh-BZT, and ditBu-BZT; 2 and 250 ng/mL for tBuPrOcEst-BZT; 10 and 250 ng/mL for DitPe-BZT

^eRE %= percent relative error

^f%RSD = percent relative deviation

Table 2.

Method Qualification Data for Total Analytes in Male SD Rat Plasma

Compound	Concentration Range (ng/mL)	LODa (ng/mL)	LOQb (ng/mL)	Standard c		QCd
				Intra-day Accuracy (RE, %)e	Intra-day Precision (RSD, %)f	Intra-day Accuracy (RE, %)e	Intra-day Precision (RSD, %)f
P-BZT	10–1000	1.95	10.1	−14.3 to 5.9	1.4 to 18.0	−5.1 to −1.5	8.8 to 15.1
Octrizole	5–100	1.53	5.0	−7.5 to 11.4	0.8 to 9.1	−45.5 to 2.4	8.6 to 27.8
Drometrizole	1–100	0.14	1.0	−8.8 to 4.6	0.4 to 5.0	−25.2 to −3.6	15.8 to 17.3
tBuPrOcEst-BZT	1–100	0.12	1.0g	−7.1 to 4.5 g	0.0 to 7.0	−12.2 to 0.7g, −27.3 to −20.5h	6.7 to 10.2g 2.9 to 6.0h
Bumetrizole	1–100	NA	1.0	−8.1 to 17.8	5.5 to 7.3	−17.0 to 0.5	23.1 to 28.1
DiMeEtPh-BZT	1–100	0.22	1.0	−11.5 to 14.0	2.7 to 17.1	3.0 to 4.4	9.3 to 25.7
ditBuCl-BZT	1–100	0.04	1.0	−3.4 to 3.0	1.0 to 8.2	8.5 to 3.2	5.6 to 10.0
DitPe-BZT	5–500	NA	5.0	−18.2 to 17.2	5.0 to 12.9	8.0 to 12.7	4.2 to 8.4
tBu-BZT	1–100	NA	1.0	−13.2 to 11.9	1.6 to 12.4	2.4 to 4.0	8.4 to 23.9

^aLOD= Limit of Detection; 3×Standard Deviation for 6 lower limit of quantitation (LOQ) replicates

^bLOQ=lower limit if quantitation; is the lowest concentration of the calibration curve

^cEstimated based on plasma calibration standards, n = 2

^dEstimated based on low and high-level Quality Control (QC) samples, n = 4; 20 and 800 ng/mL for P-BZT; 9 and 75 ng/mL for Octrizole, 2 and 50 ng/mL for drometrizole, bumetrizole, DiMeEtPh-BZT, and ditBu-BZT; 10 and 250 ng/mL for DitPe-BZT; and 1.5 and 80 ng/mL for total tBuPrOcEst-BZT (as tBuPrOH-BZT))

^eRE %= percent relative error

^f%RSD = percent relative deviation

^gWith respect to tBuPrOH-BZT

^hWith respect to tBuPrOcEst-BZT

Selectivity was checked by analyzing blanks with internal standard (IS) and rat matrix blank samples without IS from one lot. To determine carryover a single blank rat plasma sample was analyzed immediately following a high matrix standard.

Preparation of Solvent Calibration Standards, Plasma Calibration Standards, and Quality Control (QC) Samples

For each analyte, a combined solution of one or more of the other analytes was prepared and used as the internal standard (IS). Each internal standard was evaluated and the one that provided the best precision and accuracy was used as the IS for analysis. The IS selected for each analyte are presented in Table S2. IS stock solutions of each internal standard were prepared at 500 μg/mL in acetonitrile except for Octrizole which was prepared in methanol. Working IS (WIS) solutions of each internal standard were prepared at 40 ng/mL by diluting with acetonitrile except for free P-BZT and total tBuPrOcEst-BZT (analyzed as the acid, tBuPrOH-BZT) analysis, which were diluted with 90:10 acetonitrile: methanol and 5% formic acid in acetonitrile, respectively.

Three stock solutions of each analyte were prepared in acetonitrile at 500 μg/mL except Octrizole and tBuPrOH-BZT, which were prepared in methanol. Plasma calibration standards were prepared by spiking 450 μL blank pooled male SD rat plasma with 50 μL of the appropriate diluted standards prepared from two alternate stock solutions. Solvent standards were prepared similarly to plasma calibration standards except no plasma was used. Plasma QC samples were prepared at two concentrations (low and high QC) using an independent stock solution (Tables 1 and 2).

Preparation of Samples for Analysis of Free Analytes

The samples were extracted either by SPE or LLE for free analytes (Figure 3). The samples were prepared by SPE for free analysis of P-BZT and octrizole in the following manner: Fifty (50) μL of each calibration standard, QC sample, QC stability standard, and blank sample were transferred into vials. Ten (10) μL of buffer solution (1.0M ammonium sulfate with 3 mM sodium azide in deionized water) and 50 μL of WIS were added to each vial (except the blanks without IS where 50 μL of acetonitrile was added). The vials were mixed for ~ 30 sec and centrifuged at 22000 × g for ~ 5 min. The samples were loaded onto an Oasis Prime HLB μElution 96-Well plate (Waters, Milford, MA) and ~ 5 psi of pressure was applied between each step. Each well was washed with 200 μL of 5% formic acid in 5% methanol and analytes were eluted twice with 75 μL of 1:1 methanol:acetonitrile into a collection plate. The eluate was evaporated to dryness ~ 40°C, and reconstituted in 100 μL of 90:10 acetonitrile:methanol. A pierceable cover was placed over the collection plate and vortexed for ~ 2 min prior to analysis.

Figure 3.

Schematic for Sample Preparation and Analysis of Free and Total (Free and Conjugated) Phenolic Benzotriazoles Analytes in Rat Plasma

Samples were prepared by LLE for free analysis of drometrizole, bumetrizole, DiMeEtPh-BZT, tBu-BZT, DitPe-BZT, ditBuCl-BZT, and tBuPrOcEst-BZT in the following manner: Fifty (50) μL of each calibration standard, QC sample, QC stability standard, and blank sample were transferred into individual micro centrifuge tubes. Ten (10) μL of buffer solution (1.0M ammonium sulfate with 3 mM sodium azide in deionized water) was added to each tube for free analysis, and vortexed to mix. Fifty (50) μL of WIS was added to each micro centrifuge tube, except the blanks without IS. For the blanks without IS, 50 μL of acetonitrile was added instead of WIS. The tubes were mixed by vortex for ~ 30 sec and centrifuged at maximum setting for ~ 5 min for all analytes except tBuPrOcEst-BZT, bumetrizole, and DiMeEtPh-BZT, for which the centrifuge step was omitted. Two-hundred (200) μL of ethyl acetate was added to each tube, vortexed for ~ 2 min at maximum speed to mix, and centrifuged for ~5 min at ~ 22000 × g. The top layer (ethyl acetate, ~ 200 μL) was transferred to a 96-well collection plate and the extracts were evaporated to dryness at a temperature setting of 40°C. Each well was reconstituted with 100 μL of 90:10 acetonitrile:methanol. A pierceable cover was placed over the collection plate and vortexed for ~2 min at maximum speed prior to analysis.

Preparation of Samples for Analysis of Total Analytes

The samples were extracted either by LLE or protein precipitation for determination of free and conjugated parent (total) analytes (Figure 3). For total analysis of total P-BZT, octrizole, drometrizole, bumetrizole, DiMeEtPh-BZT, tBu-BZt, DitPe-BZT, and ditBuCl-BZT, samples were prepared by LLE in a similar manner above with minor modifications. Instead of buffer solution, 10 μL of β-glucuronidase in aqueous ~1.0 M ammonium sulfate with 3 mM sodium azide was added to each tube, vortexed to mix, and placed in a water bath set to 37°C for approximately 24 hrs. The rest of the steps were followed as described above for free analysis using LLE.

The samples were prepared by protein precipitation for total analysis of tBuPrOcEst-BZT (analyzed as acid, tBuPrOH-BZT) in the following manner: Fifty (50) μL of each calibration standard, QC standard, QC stability standard, blank, and study sample were transferred into individual micro centrifuge tubes. One-hundred (100) μL of 6N HCl was added to each tube, vortexed ~ 2 min to mix, and placed in an oven set to approximately 105°C for ~ 2 hrs. One hundred (100) μL of 6N NaOH was added to each tube and each tube was vortexed for ~ 2 min. Two hundred (200) μL of WIS was added to each micro centrifuge tube, except the blanks without IS. For the blanks without IS, 200 μL of 5% formic acid in acetonitrile was added instead of WIS. The tubes were mixed by vortex at maximum speed for ~ 2 min and centrifuged at ~22000 × g for ~ 5 minutes. The supernatant was transferred to autosampler vials. The samples were analyzed by LC-MS/MS.

LC-MS/MS Analysis and Analyte Quantitation

Chromatographic parameters were developed and optimized for the analysis of the nine phenolic benzotriazoles and tBuPrOH-BZT. Samples were analyzed using a Shimadzu Prominence (Kyoto, Japan) LC coupled to AB Sciex (Framingham, MA) API 5000 or API 5500 MS in a positive chemical ionization mode. Phenomenex Kinetex, Phenyl-Hexyl column (50 m × 2.1 mm (ID) × 2.6 μm film thickness) (Torrance, CA) and guard column Phenomenex Security Guard (C18, 4×2 mm) were used with a flow rate of 300 μL/min. The method details are provided in Table S2.

A linear regression with 1/x weighting was used for free P-BZT, free and total octrizole, free tBuPrOcEst-BZT, total tBuPrOH-BZT, free and total bumetrizole, free DiMeEtPh-BZT, free and total ditBuCl-BZT, free and total DitPe-BZT, and total tBu-BZT to relate analyte to IS peak area response ratio to analyte concentration in plasma. A linear regression with 1/x² was used for total P-BZT, free and total drometrizole, total DiMeEtPh-BZT, and free tBu-BZT to relate peak area response ratio to analyte concentration in plasma. Plasma concentration of each free and total phenolic benzotriazoles were reported as ng analyte/mL plasma.

Stability

Long term analyte stability in plasma (QC stability samples) was determined by preparing three aliquots of male HSD rat plasma standards at 10 ng/mL, which were stored up to 387 days for free and 705 days for total analytes at ~ −70 °C. Individual stability days for each free and total analyte are provided in Table 3. The stability days varied by phenolic benzotriazole based on the potential toxicology study sample analysis dates to allow longer duration storage of the samples. Samples were analyzed along with freshly prepared and processed matrix calibration standards and QC samples for both free and total analytes to determine stability.

Table 3.

Stability Data for Free and Total Phenolic Benzotriazoles in Male SD Rat Plasma

Analyte	Stability (RE, %)a	Number of Days Stored
Free Analysis
P-BZT	≤ ± 6.8	231
Octrizole	≤ ± 13.6	251
Drometrizole	≤ ± 31.6	300
tBuPrOcEst-BZT	≤ ± 22.0	330
Bumetrizole	≤ ± 66.5	387
DiMeEtPh-BZT	≤ ± 33.6	332
ditBuCl-BZT	≤ ± 29.8	362
DitPe-BZT	≤ ± 20.9	365
tBu-BZT	≤ ± 61.0	350
Total Analysis
P-BZT	≤ ±12.3	306
Octrizole	≤ ± 27.1	328
Drometrizole	≤ ± 32.9	316
tBuPrOcEst-BZT b	≤ ± 73.7	705
Bumetrizole	≤ ± 9.4	399
DiMeEtPh-BZT	≤ ± 44.2	359
ditBuCl-BZT	≤ ± 40.2	364
DitPe-BZT	≤ ± 18.4	370
tBu-BZT	≤ ± 58.7	372

^aRE %= percent loss relative to Day 0, three replicates plasma QCs at 10 ng/mL stored at ∼ −70°C

^bWith respect to tBuPrOcEst-BZT

Results and Discussion

Analytical Method Development

The analytical methods were developed to analyze free and total analytes in SD male rat plasma following a literature search on methods for quantitation of phenolic benzotriazoles in different matrices. There are several methods identified in the literature for analysis of phenolic benzotriazoles in environmental samples (Zhang et al. 2011; Montesdeoca-Esponda et al. 2012) and marine biota (Pruell et al. 1984; Nakata Haruhiko et al. 2009; Nakata H. et al. 2010; Kim, Isobe, et al. 2011; Nakata H. et al. 2012). In general, the methods reported several steps of clean up and separation techniques depending on matrices and analyte(s) of interest and analyzed by GC-MS or LC-MS/MS. For example, one method quantitated ditPe-BZT, tBu-BZT, ditBuCl-BZT and bumetrizole in sediment and sludge samples (Zhang et al. 2011) following solvent extraction and further clean up with silicagel chromatography and analysis by GC-MS. Reported LODs ranged from 0.1 to 0.5 ng/mL and LOQs from 0.3 to 1.65 ng/mL. Another method (Montesdeoca-Esponda et al. 2012) reported analysis of 8 different phenolic benzotriazoles (including drometrizole, bumetrizole, diPe-BZT, ditBu-BZT, diMeEtPh-BZT, and octriazole) in coastal marine and wastewater samples utilizing online and offline solid phase extraction (with pre-clean up and pre-concentration step) coupled with LC-MS/MS analysis. LODs reported ranged from 0.6 to 18 ng/mL and LOQs ranged from 2.1 to 61.0 ng/mL depending on analytes. These values may not be comparable to our methods due to the differences between matrices and potential matrix effects (plasma vs water samples). While neither REs nor stability data were reported, RSD values ranged from 6.2 to 13%. While literature methods reported some of the same phenolic benzotriazoles as our nine selected analytes, the methods were found not to be suitable for this study due the matrix differences, extensive clean up requirements and labor-intensive separation techniques.

At the initiation of this study, there were only two methods (Thomas et al. 1995; Hirata-Koizumi et al. 2009) in the literature for the analysis of some of the free (unconjugated parent) analytes in plasma and no methods reported the analysis of total (free and conjugated parent) analytes. One method analyzed one phenolic benzotriazole, HDBB, in male and female rat plasma (n=4/sex/group) following oral administration of HDBB (0.5, 2.5, and 12.5 mg/kg). Samples were prepared by liquid extraction and analyzed by LC-MS/MS with LOQ of 20 ng/mL. Other details such as stability, accuracy and precision of the method were not provided. Another method was published in 1995 and utilized thin layer chromatography and radiometric scanning to quantify phenolic benzotriazoles and their methyl esters in blood (Thomas et al. 1995), and the methods were not optimal for this study.

In this study, we needed to develop an efficient analytical method to evaluate internal dose to aid in the interpretation of toxicological studies following exposure to phenolic benzotriazoles. Since, these analytes have phenolic groups, they are expected to undergo phase II metabolism to generate glucuronide and sulfate conjugates. To understand the totality of exposure needed to compare the toxicology data between the class chemicals, methods were developed and evaluated for both free and total analytes in SD rat plasma. Although in our study, we developed and used the LC-MS/MS to determine concentrations of individual analytes, the method could be utilized for simultaneous analysis of all nine analytes due to the different transitions monitored. In addition, the methods developed in this study were later adapted and modified for the analysis of free phenolic benzotriazoles for a class comparison in a toxicokinetic study (Waidyanatha et al. 2021).

Preparation of Samples

For free analytes, protein precipitation was initially evaluated as the simplest extraction technique. Based on the variability in recovery (1.2–84.7% absolute recovery and 11–131.1% relative recovery) and matrix effect (e.g. matrix enhancement; compound dependent) for each of the compounds when utilizing protein precipitation, it was determined a more thorough clean-up procedure was needed. As a result, both SPE and LLE methods were evaluated for free analysis. The extraction methods for free analytes are listed in Table S2.

For each analyte, a combined solution of one or more of the other analytes was prepared and used as the IS. The internal standards were evaluated and the one that that provided the best precision and accuracy was used as the IS for analysis and are presented in Table S2.

Similar to free analysis, different extraction techniques (SPE, LLE, and protein precipitation) were evaluated for total analyte analysis, resulting in LLE being utilized for all analytes except tBuPrOcEst-BZT (as tBuPrOH-BZT), which utilized protein precipitation (Table S2). During the total analysis of tBuPrOcEst-BZT, it was discovered that the compound hydrolyzed to the acid (Figure 2); therefore, a method was developed for the total analysis of the acid (tBuPrOH-BZT).

LC-MS/MS Analysis and Analyte Quantitation

The analysis method utilized LC-MS/MS operated in positive ion mode. Instrument parameters were optimized with solvent standards based on the analyte of interest (Table S3). Transitions used for each analyte are presented in Table S3. In one publication (Montesdeoca-Esponda et al. 2012) while precursor ions for drometrizole, bumetrizole, diPe-BZT, ditBuCl-BZT, and octrizole were reported (m/z 226.2, 316.3, 352.3, 358.3, and 324.2, respectively), no product ions were reported even though LC-MS/MS was utilized for the analysis. Transitions for drometrizole, bumetrizole, diPe-BZT, ditBuCl-BZT, diMeEtPh-BZT and octrizole reported in another LC-MS/MS method (Kim, Isobe, et al. 2011) were m/z 226→98.9, 316.0→260.0, 352.1→282.1, 358.0→302.0, 448.2→370.1, and 324.0→212.0, respectively which are very similar to the current method.

The plasma standard curves had coefficient of determination (r²) values that were ≥ 0.99 for both free and total analytes. LOQs from 1 ng/mL to 5 ng/mL (for free analytes) and 1 to 10 ng/mL (for total analytes) were achieved and are presented in Tables 1 and 2. In a previously published method (Hirata-Koizumi et al. 2009), the LOQ of ditBuCl-BZT in rat plasma was reported to be 20 ng/mL, while in this study we achieved an LOQ of 1 ng/mL for free and total ditBuCl-BZT. Representative chromatograms for some analytes from each substitution group are presented in Figures S1–S6.

The intra-day accuracy determined as RE were ≤ ±17.0% for free and ≤±18.2% for total analytes based on calibration standards (Tables 1 and 2). The intra-day precision, determined as RSD were ≤ 20.7% for free and 18.0% for total analytes (Tables 1 and 2).

The intra-day accuracy and precision were also determined using plasma QC samples. The intra-day accuracy determined as RE was ≤ ±37.3% for free and ≤ ±45.5% for total analytes, respectively (Tables 2 and 3). The intra-day precision, determined as RSD was ≤ 34.7% for free and 27.8% for total analytes (Tables 1 and 2). To the best of our knowledge intra-day accuracy have not been reported in the literature methods for either of the phenolic benzotriazoles in plasma.

The sensitivity of the method was determined by the lowest reportable calibration standard (~ 20% RE) in each run and is presented in Tables 1 and 3. The estimated LOD for free and total analytes are presented in Table 1 and Table 2, respectively. The LOD value for free tBuPrOcEst-BZT could not be calculated due to an insufficient number of samples. The estimated LOD for total P-BZT, octrizole, drometrizole, tBuPrOcEst-BZT, DiMeEtPh-BZT, and ditBuCl-BZT was 1.95, 1.53, 0.14, 0.12, 0.22, and 0.04 ng/mL, respectively. The LOD values for total Bumetrizole, DitPe-BZT, and tBu-BZT could not be calculated due to insufficient number of samples.

Stability

Stability data are presented in Table 3. Storage stability QCs were evaluated, and it was found that the rat plasma extraction samples could be reinjected and analyzed with freshly prepared calibration standards when stored at ~ −70°C for up to 387 days for free and 705 days for total analytes. The storage stability accuracy (RE%) for the analytes ranged from ≤± 6.8% to 66.5% (mean RE was −59.0% to +20.5%, Table S4) for free and from ≤± 9.4 % to 73.7% (mean RE −73.5% to 25.0%, Table S4) for total analytes. For total tBuPrOcEst-BZT a higher relative error (≤±73.7%) was observed in comparison to other total values. This may be due to incomplete hydrolysis since the samples were prepared using tBuPrOcEst-BZT and analyzed for the acid following hydrolysis. No additional stability information for this class of phenolic benzotriazoles were found in the literature.

Conclusions

An LC-MS/MS method was developed and qualified to quantitate free (unconjugated parent) and total (free and conjugated parent) phenolic benzotriazoles in rat plasma to aid in interpretation of toxicity data and understanding of absorption, distribution, metabolism, and excretion (ADME) differences. In addition, free and conjugated analyte levels provide a measure of totality of exposures and allow for evaluation of potential ADME differences among phenolic benzotriazoles. The quantitation ranges for the qualified methods were 1–500 ng/mL for free analytes and 1–1000 ng/mL for total analytes, using 50 μL plasma. Rat plasma samples were evaluated found stable when stored ~ −70 °C up to 387 days for free analytes and up to 700 days for total analytes. In general, male SD rat plasma was selective with no significant (>30% of the area of the low standard) interfering peaks at the retention time of the nine analytes. The methods are suitable for determination of free and total phenolic benzotriazoles in rat plasma in support of toxicology and toxicokinetic studies.