High-throughput chemometrically assisted flow-injection method for the simultaneous determination of multi-antiretrovirals in water

Highlights

• Development of an analytical method for the simultaneous determination of five antiretroviral drugs

• Spectral-pH data obtained by a flow injection system.

• Identification of a second pK_a value not reported for nevirapine.

• MCR-ALS chemometric data modeling of highly overlapped spectral data.

• Environmental aqueous sample analysis without pre-treatment steps

Abstract

This work is focused on the simultaneous determination of different nucleosides (lamivudine, abacavir, zidovudine) and non-nucleoside (nevirapine, efavirenz) reverse transcriptase inhibitors (antiretrovirals) in river, tap and well water by pH-gradient flow-injection analysis with diode-array detection (FIA-DAD) and further multivariate curve resolution-alternating least square (MCR-ALS) processing. Previous to the development of the analytical method with quantitative purpose, the acid-base characterization was conducted for all the analytes by determination of the experimental pK_a. The pK_a values, obtained by UV spectrometric titration coupled to MCR-ALS data modeling, were in agreement with those found in the literature, and, interestingly, a non-reported pK_a of 12.27 was identified for nevirapine. Eventually, the analytical method was developed and its performance on the quantitation of the antiretrovirals was evaluated through its application to validation and environmental samples. The recovery studies showed values between 85 and 115% with relative errors of prediction below 10%. The figures of merit were satisfactory for all the studied drugs, with limits of detection between 25 and 40 μg L⁻¹. Therefore, the proposed method, which is based on the generation of high-throughput data, is a simple and suitable alternative to determine the concentration of antiretroviral drugs in complex matrices, without pre-concentration or extraction steps. Additionally, MCR-ALS demonstrated to be a useful tool for solving mixtures of analytes with high spectral overlapping, in the presence of unknown interferences.

Introduction

The acquired immune deficiency syndrome (AIDS) caused by the human immunodeficiency virus (HIV) is one of the greatest lethal diseases in the world. It is estimated that causes about 1 million deaths per year worldwide [1,2]. The most widely used treatment, known as highly active antiretroviral therapy, combines at least two nucleoside analogue reverse transcriptase inhibitors, such as lamivudine (3TC), abacavir (ABC) or zidovudine (AZT), with one non-nucleoside reverse transcriptase inhibitor, such as nevirapine (NVP) or efavirenz (EFV), in a fixed dose combination [3]. The synergistic action provided by this therapy has allowed reducing both mortality and morbidity rates among HIV-infected people, improving their life quality.

Antiretroviral drugs are considered emerging contaminants since they were found in the environment and can cause adverse ecological and/or human health effects. However, they are not usually monitored due to the lack of regulations [4]. Although these drugs are generally present at rather low levels (from ng L⁻¹ to μg L⁻¹) in wastewater and surface waters [[5], [6], [7]], the continuous input can increase the concentrations bringing potential negative effects on the environment, e.g., antimicrobial resistance and toxicity to sensitive aquatic organisms. Their simultaneous determination in environmental samples, as well as in pharmaceutical dosage, biological fluids, animal tissue, synthetic mixtures, has been performed by near-infrared spectroscopy [8], high performance liquid chromatography with diode array detection (HPLC-DAD) [9], U-HPLC with tandem mass spectrometry detection (MS/MS) [10] and pH-gradient flow injection analysis (FIA) with DAD [11].

The aforementioned antiretrovirals (3TC, ABC, AZT, NVP and EFV) exhibit acid-base properties that cause UV spectral changes when variations in the pH of the medium are produced. It is known that the pK_a of the analytes plays an important role from an analytical point of view. Therefore, prior to the development of a pH-gradient with spectral detection method, an in-depth bibliographic search related to the pK_a values of the five herein analyzed drugs was conducted and summarized in Table 1. It is clear that the available information is scarce or incomplete, especially due to the lack of description of the experimental conditions in which they were determined.

Analytical methods used for pK_a determinations of several drugs include potentiometry [12,13], capillary electrophoresis (CE) with DAD [14], UV–Vis spectroscopy [15] and FIA-DAD [11], among others. The most commonly used estimator for the pK_a is the well-known Henderson-Hasselbalch equation [16], whereas chemometric resolution of spectrophotometric second-order data constitutes an alternative way of calculation. In this sense, MCR-ALS [[17], [18], [19]] has been widely applied to calculate pK_a from second-order data gathered by different instrumentations in several research fields, i.e. UV spectroscopy [20,21], FIA-DAD [22,23], CE-DAD [24] and fluorescence spectroscopy [20,25].

In this work, the apparent pK_a values of 3TC, ABC, AZT, NVP and EFV were determined by spectrophotometric titrations and chemometric data analysis. Then, based on this physicochemical information, a FIA-DAD method with MCR-ALS modeling for the simultaneous quantitation of these five drugs in aqueous environmental samples was developed.