Development of the methods for atomoxetine identification suitable for the chemical and toxicological analysis

The cases of acute and lethal poisonings by atomoxetine, an antidepressant, have been registered. Therefore, development of the available and reliable methods for its chemical and toxicological analysis is the topical issue. Aim. To develop the conditions for atomoxetine detection in the presence of a number of its pharmacological and structural analogs when performing the general TLC-screening and identification of the antidepressant by UV spectrophotometry. Materials and methods. Chromatographic mobility of antidepressants in thin sorbent layers was studied in 11 mobile phases, including those recommended by TIAFT, using five types of chromatographic plates. А number of chromogenic reagents was used for visualization. The UV-spectrum of atomoxetine was studied in 0.1 mol ∙ L-1 solution of hydrochloric acid. Results and discussion. Chromatographic systems with the low correlation of Rf values for atomoxetine that make them suitable for the general TLC-screening have been found. The UV spectrum of atomoxetine in the acid solution had light absorption maxima at wavelengths of 270 (ε = 1300; А1 = 45) and 277 nm. Сonclusions. A combined use of three mobile phases: methanol – 25 % ammonia solution (100 : 1.5), cyclohexane – toluene – diethylamine (75 : 15 : 10), and toluene – acetone – 25 % ammonia solution (45 : 45 : 7.5 : 2.5) provides a reliable atomoxetine detection in the presence of a number of its pharmacological and structural analogs when performing the general TLC-screening. The presence of specific light absorption in the UV region of the spectrum for atomoxetine makes the UV spectrophotometry method suitable for identification of the antidepressant in toxicological studies.


Ключевые слова: атомоксетин; ТСХ-скрининг; УФ-спектр
Atomoxetine is a thymoleptic from the group of selective norepinephrine reuptake inhibitors (central acting indirect sympathomimetic) [1] approved by the USA FDA in late 2002 [2]. It is used in the pharmacotherapy of attention deficit hyperactivity disorder [1], as well as in treatment-resistant depression, psychosis, bipolar disorders, epilepsy as an individual drug or as part of a complex treatment [3]. Unlike traditional psychostimulants, atomoxetine does not refer to narcotic substances and shows no abuse potential associated with them [4].
Atomoxetine has some side effects [1,5], among them the appearance of suicide thoughts is the most serious complication [6]. This drug is incompatible with monoamine oxidase inhibitors [7].
Most bioanalytical methods described in the literature for atomoxetine determination are based on using HPLC with different types of detection: UV [10,11,12], mass spectrometry (MS) [13,14] fluorescence [15] or capillary electrophoresis [16]. These methods of analysis are not always available for toxicological laboratories; they require expensive equipment and the appropriate skill level of the staff.
The combination of TLC during the toxicological screening and absorption spectrophotometry in the UV region for the confirmatory study is an available approach for use in the chemical and toxicological analysis [17,18]. Information on using TLC and UV spectrophotometry for detection and identification of atomoxetine has not been found in the available literature.
The aim of the study was to develop the conditions for atomoxetine detection in the presence of a number of its pharmacological and structural analogs when performing the general TLC-screening and the antidepressant identification by UV spectrophotometry.

Materials and methods
The pure substance of atomoxetine isolated from the medicine "Strattera" (7 capsules, 60 mg) produced by "Lilly" (Czech Republic) was used for the study.
Extraction of the atomaxetine substance from capsules. Triturate the content of 7 capsules in a porcelain mortar, add 50 mL of absolute ethanol, and then filter the mixture through a folded filter paper in a porcelain evaporating dish. Wash the residue on the filter once with 20 mL of absolute ethanol, and combine the ethanol filtrates. Evaporate the dish content on a water bath at 40-60 °C to complete removal of the organic solvent. Dry the residue in the loss-on-drying oven and weigh.
Using this method 400 mg of atomoxetine hydrochloride was obtained. The purity of the substance was tested by TLC, UV spectrophotometry and HPLC, and its compliance with the quality requirements of the SPhU was determined.
The TLC ultraviolet lamp was 254/365 nm.
In the study the water bath LW-4 (Bytom, Poland) and a spectrophotometer (SF-46, LOMO, USSR) with the spectral measurement range of 190-1100 nm were also applied.
Developers recommended by the Committee of Systematic Toxicological Analysis of the International Association of Forensic Toxicologists (TIAFT) for the ge- neral TLC-screening [17,18], and reagents proposed by the WHO and the UNO [19] to identify the main groups of potent substances (Tab. 1) were studied as chromogenic reagents. According to the recommendations of UNODC [20] determination 10 samples containing the drug from 1.25 • LOD to 2.0 • LOD were studied to determine the limit of detection (LOD). The number of false negative results was not more than 2 out of 10 experiments (RSD ≤ 20 %). Chromogenic reagents were prepared as follows [17,18]. Results and discussion Chromatographic mobility of atomoxetine, venlafaxine, fluoxetine, fluvoxamine, amitriptyline, melipramin and doxepin in thin sorbent layers were studied in the m. ph. recommended by TIAFT for the general TLCscreening of drugs (m.ph. No. 1-9) [17,18] and in the m.ph. widely used in the national practice of forensic toxicological studies (m. ph. No. 10, 11) [21,22]. The Rf values of atomoxetine, venlafaxine, fluoxetine, fluvoxamine, amitriptyline, melipramine and doxepin are shown in Tab. 2.
Atomoxetine exhibited R f values acceptable for identification in m.ph. No. 4, 7, 8, 10, 11. In m. ph. No. 7 separation of all antidepressants under study was observed. The use of m.ph. No. 8 gave the possibility of appropriate separation of atomoxetine from other drugs under study, excluding fluoxetine and fluvoxamine. In the mobile phase mentioned atomoxetine, fluoxetine and fluvoxamine showed the low correlation of Rf values compared to m.ph. No. 7. Separation of atomoxetine from other antidepressants studied, with the exception of fluoxetine, was also observed in m.ph. No. 10. Moreover, in the m.ph. No. 10 atomoxetine and fluoxetine had the inverse correlation of chromatographic mobility compared to m.ph. No. 7. It greatly increases the reliability of identification of these antidepressants. Therefore, the use of three m.ph. No. 7, 8 and 10 are recommended for atomoxetine identification in the presence of a number of its structural and pharmacological analogs when performing TLC-screening. According to the TIAFT recommendations the use of several chromatographic systems preferably with the low correlation of Rf values significantly increases the reliability for identification of substances by TLC [17,18].
The UV-light and Dragendorff reagent with Munier modification were the most sensitive for detecting atomoxetine and other antidepressants studied. For atomoxetine LOD was 0.3 µg in the sample in UV light at 254 nm, and 0.5 µg in the sample at 365 nm using Dragendorff reagent with Munier modification. High sensitivity compared to atomoxetine also was found when using Froehde reagent (1.0 µg in the sample), which formed products of a blue color with the antidepressant. Thus, Froehde reagent can be recommended as a relatively selective reagent to biological admixtures for detection of atomoxetine in biological extracts. Melipramine and doxepin also formed blue coloration with Froehde reagent. But they did not give color with Marquis reagent, a pink color was observed under its action on atomoxetine.
Antidepressants containing in its structure the secondary (atomoxetine, venlafaxine, fluoxetine) or primary (fluvoxamine) amino group with ninhydrin solution gave intense coloration with different colors. Specific color transitions were observed while processing of atomoxetine spots with Mandelin reagent and formaldehyde vapor consequentially (modified Mandelin reagent) (Tab. 1).
Thus, when performing the TLC-screening for detection and identification of atomoxetine in the presence of its structural and pharmacological analogs (venlafaxine, fluoxetine, fluvoxamine, amitriptyline, melipramine and doxepin) it is appropriate to use chromogenic reagents in the following sequence: UV light, ninhydrin solution, Dragendorff reagent (it can be used after Table 1 Sensitivity of the reactions (LOD, µg per sample) and color of the reaction products of atomoxetine, venflaxine, fluoxetine, fluvoxamine, amitriptyline, melipramine and doxepin with the chromogenic reagents     Notes: * -the type of the sorbent layer is given in the section "Materials and methods".
pre-treatment of spots with ninhydrin), Froedhe reagent, Marquis reagent, modified Mandelin reagent. According to the TIAFT recommendations on the reliable identification of toxic substances in TLC-screening the acceptable condition is to use at least four reagents on the same chromatographic plate consequentially [17,18]. While conducting the TLC-screening simultaneously the separation of the substances studied from the components of the biological matrix (TLC purification) occurs, and it is a prerequisite for the use of UV spectrophotometry to identify the toxic substance at the confirmation stage of toxicological studies.
The UV-spectrum of atomoxetine in 0.1 Mol • L -1 hydrochloric acid solution was studied, and the presence of specific absorption was determined (Fig.). The molar absorptivity (ε м ) and the specific absorption coefficient (А 1 1 ) for the higher intensity absorption peak at 270 nm were calculated using the absorbance values of three WSS given above. Absorption maxima for atomoxetine solutions in 0.1 mol•L -1 hydrochloric acid (λ max ± 2 (ε м ; А 1 1 )) were observed at the wavelengths of 270 (1,300; 45) and 277 nm. CONCLUSIONS 1. The parameters of chromatographic mobility of atomoxetine and a number of its structural and pharmacological analogs (venlafaxine, fluoxetine, fluvoxamine, amitriptyline, melipramine and doxepin) have been determined in thin sorbent layers using mobile phases recom-mended by TIAFT for the general TLC-screening and mobile phases, which are widely used in toxicological screening of the basic drugs. Chromatographic mobility has been studied using five types of chromatographic plates.
3. Sensitive and specific chromogenic reagents for atomoxetine identification in the presence of a number of its structural and pharmacological analogs have been determined. It has been recommended to use the chromogenic reagents in the following sequence: UV light, ninhydrin solution, Dragendorff reagent, Froedhe reagent, Marquis reagent, modified Mandelin reagent.
4. The presence of specific light absorption in the UV region of the spectrum for atomoxetine solution in 0.1 Mol • L -1 hydrochloric acid at 270 (ε m = 1,300; А 1 1 = 45) and 277 nm has been found, it makes the UV spectrophotometry method suitable for identifying the antidepressant in toxicological studies.
Conflict of Interests: authors have no conflict of interests to declare.