A new approach based on isoindole formation reaction for sensitive fluorimetric assay of milnacipran in tablets and biological fluids (plasma/urine)

The current study describes a new, sensitive, and economic protocol for milnacipran analysis. Milnacipran was introduced as a therapy for fibromyalgia and depression. It acts by unique and equal inhibition of noradrenaline and serotonin neurotransmitters reuptake. In the presence of 2-mercaptoethanol, the amino moiety of milnacipran condenses with o-phthalaldehyde to generate isoindole fluorescent derivative. The isoindole product was measured at (λex 338.5 nm, λem 433.5 nm) and condensation variables were strictly optimized. The fluorescence intensity of measurements was plotted versus milnacipran concentration to give a linearity range over 200–4000 ng mL−1. The proposed approach was fully validated by the directives of ICH guidelines and applied without any influence of combined excipient for milnacipran tablet analysis. Furthermore, the procedure was applied in spiked urine and plasma analysis with excellent percentage recovery.


Introduction
Among medical problems that affect many people are bromyalgia and depression. Fibromyalgia is a complicated disease characterized by many symptoms as generalized chronic muscular pain, sleep disturbance, anxiety, fatigue, cognitive dysfunction, and stiffness. 1 About 2-4% of people are affected by bromyalgia. 2 Depression is a mood condition that is substantially different from normal mood changes in daily life. Many symptoms are accompanied by depression as weight loss, psychomotor retardation, insomnia, concentration problems, fatigue, and anxiety. In severe states, patients may have a feeling of worthlessness, hallucinations, and thoughts of suicide. 3 Milnacipran (MLC; 2-(aminomethyl)-N,N-diethyl-1phenylcyclopropane-1-carboxamide) was accepted by the FDA as a therapy for depression and bromyalgia. 4 It acts by unique and equal inhibition of noradrenaline and serotonin neurotransmitters reuptake. 5 In the previous analytical literature, MLC was assayed by various techniques which include GC-MS, 6 liquid chromatography with uorescence or UV detection, 7,8 spectrouorimetry by derivatization with ninhydrin 9 or NBD-Cl, 10 UVspectrophotometric detection at 223 nm 11,12 and colorimetry using ninhydrin 13 or ion pair complex. 14 The approach target was simple development for accurate assay of MLC in the spiked biological uid, and tablets. The method provides advantages over the published method as the heating step is absent which make the method is more facile than the reported methods, 9,10 non-extractive, 13,14 low LOD and LOQ, 9,11,13,14 and nally implemented in non-sophisticated and easily available equipment.
The assay strategy simply relies on isoindole formation reaction. At suitable basic pH, isoindole reaction is the fundamental concept of many derivative reactions used in spectro-uorimetric research. [15][16][17][18] The approach is considered an alternative way for MLC analysis with many advantages. The directives that followed for application and validation were ICH guidelines.

Apparatus
Throughout the work, uorescence intensities were monitored by a spectrouorimeter FS-2 (Scinco, Korea). The instrument is equipped with Xe-arc lamp (150 W). The voltage of PMT and both monochromators (excitation and emission) slit width and were set at 400 V and 5 nm respectively.

Chemicals and materials
We have been supplied with Milnacipran hydrochloride from Mash Premiere (Badr City, Egypt) pharmaceutical company as a gi. Milnavella (Mash Premiere) and Averomilan® (Averroespharma, Egypt) with 50 mg active milnacipran were utilized in dosage form analysis. o-Phthalaldehyde (Alpha Chemika, India) was prepared freshly by dissolving 10 mg in 5 mL methanol and volume adjusted by water to 100 mL to give concentration 7.46 Â 10 À4 M. 2-Mercaptoethanol (99%) was purchased from Sigma (MO, USA), stored at 2-8 C and diluted by water to give (0.01% v/v) solution. Organic solvent, boric acid, and sodium hydroxide were obtained from El-Nasr chemical-Co (Cairo, Egypt). Borate buffer (0.2 M) solution was prepared by combining suitable volumes of 0.2 M NaOH and 0.2 M boric acid and adjusting the intended pH by pH meter. Human plasma and urine samples were obtained from healthy male volunteers with ages ranged from 30-39 years old at Assiut University Clinics according to institutional guidelines. In all cases, informed written consent was obtained from all participants and plasma samples were kept frozen at À20 C until assessment aer gentle thawing. Urine samples were stored in the refrigerator at 4 C until analysis.

MLC standard solution
Stock MLC solution was prepared by dissolving 20 mg of MLC powder in a 100 mL ask. The powder was initially dissolved in 25 mL distilled water then completed to the mark with the same solvent. Different MLC working solutions were prepared by further dilution of the previously prepared stock solution.

General analytical procedure
The procedure was implemented in 10 mL ask. To each ask 1.4 mL borate buffer (pH ¼ 10.5) was mixed with an aliquot volume of MLC equal to (200-4000 ng) in the nal working solution. The content was thoroughly mixed with 1.2 mL of 2mercaptoethanol (0.01 v/v) followed by 1 mL o-phthalaldehyde (7.46 Â 10 À4 M), then allowed to stand 25 minutes. Finally, the nal volume was adjusted with methanol and the uorescence intensity of the isoindole product was recorded at (l ex 338.5 nm, l em 433.5 nm). All uorescence reading was corrected against an analytical blank prepared simultaneously.

Tablet formulation analysis
Ten tablets of Milnavella or Averomilan® were weighed and crushed entirely to a ne powder. A quantity of dosage form powder equal to 50 mg MLC was placed and sonicated in the ask with 30 mL distilled water for 15 min. The content of the ask was accurately ltered into another volumetric ask and completed to the mark by water. Aer suitable dilution, the percentage recovery of tablets content was estimated from the regression equation aer analysis by the recommended approach procedure.

Steps for spiked human plasma
At Minia University Hospital, the process of plasma isolation was implemented. The blood was taken from healthy volunteers (who not taken MLC) via the forearm vein, then placed in heparinized tubes. Plasma isolation was accomplished by centrifugation of the tube content at 5000 rpm for 20 min. Until the time of the assay, the plasma was held at À20 C. 19 Into a series of Eppendorf tubes (5 mL) 800 mL of the stored plasma and 200 mL of MLC (containing the suitable concentration) was mixed by the vortex. Separation of plasma protein was performed by adding 3 mL of acetonitrile followed by centrifugation at the time and value mentioned above. Ultimately, 1.0 mL from the tube content was transferred into a 10 mL analytical ask and was analyzed by the recommended procedure and the response was corrected against plasma blank.

MLC assay in spiked urine
Into 50 mL separating funnel, 800 mL human urine, 200 mL MLC solution (containing the different concentrations), and 3 mL NaOH/carbonate solution (pH ¼ 11.8) was added. Then the solution was extracted three times using 5 mL dichloromethane in each one. The collected organic layer was carefully evaporated under nitrogen. The MLC residue was reconstituted in methanol. Finally, the assay procedure was followed using a portion of the nal methanolic solution and reading corrected against urine blank reading.

Result and discussion
Several analytical approaches rely on derivatizing a primary amino group by the uorogenic reagent. Ninhydrin, NBD-Cl, NQS (1,2-naphthoquinone-4-sulfonate), acetylacetone, uorescamine, and o-phthalaldehyde are the prevalent reagents probe. o-Phthalaldehyde is more preferred over other uorogenic reagents due to the absence of the extraction step as in NQS and no heating step like that in ninhydrin, NBD-Cl, NQS, and acetylacetone. Moreover, the reagent has a low price compared to uorescamine. Although the price of both NBD-Cl and o-phthalaldehyde are relatively similar, a very low concentration is needed in the case of o-phthalaldehyde (0.01%) compared to NBD-Cl (0.1%) which would efficiently reduce the cost per run. It is known that heating and extraction steps render analytical assay tedious, and lengthy.
In the presence of a thiol-containing compound as 2-mercaptoethanol, the amino moiety of MLC condenses with ophthalaldehyde to generate isoindole uorescent derivative. The reaction proceeded in basic pH and 2-mercaptoethanol is essential for isoindole product stability. 20 The MLC derivative was monitored at (l ex 338.5 nm, l em 433.5 nm) (Fig. 1) and (Scheme 1) elucidate isoindole formation mechanism. Successful application of the approach on MLC containing dosage forms and biological uids (plasma/urine) was established aer complete validation of the method.

Reaction variables optimization
The isoindole product was inspected at a variety of wavelengths to choose the accurate wavelength (excitation, emission). Furthermore, variables that affect stability and product formation were precisely studied and optimized.

pH and borate buffer volume
Inspection of the uorescence strength of isoindole product at a variety of pH solution (8)(9)(10)(11)(12) was carried out by using borate buffer. Isoindole formation proceeded at alkaline pH and maximum strength attained at pH ¼ 10.5 (Fig. 2). Furthermore, the borate buffer volume was studied by performing several experiments by different volumes (0.2-2.2 mL). The data of analysis represented in (Fig. 2) reveal that 1.4 mL is optimum.

Reagent volume (o-phthalaldehyde/2-mercaptoethanol)
The impact of o-phthalaldehyde (7.46 Â 10 À4 M) on the derivative's uorescence was evaluated by using different volumes (0.2-2.4 mL). It was noted that FI gradually increased with increasing volume to give constant uorescence (0.8-2 mL) followed by a slight gradual decrease (Fig. 3). To stabilize isoindole product, 2-mercaptoethanol addition is an important    step, so the uorescence strength values were recorded aer adding different volumes of 2-mercaptoethanol (0.01 v/v). The high product uorescence was recorded by adding 1.2 AE 0.2 (Fig. 4).

Time and diluting solvents effect
Isoindole derivative formed by the reaction of MLC with ophthalaldehyde was diluted by a variety of solvent to select the appropriate one. The inspected solvents were isopropyl alcohol, water, DMSO, methanol, acetone, dimethylformamide, and ethanol (Fig. 5) exhibit high uorescence value with methanol. The isoindole derivative was allowed to be formed at various time intervals (5-35) minutes. The values of (Fig. 5) show that 25 minutes is adequate time for product formation (isoindole derivative) with maximum FI. Finally, the effect of temperature was tested at different values and the ambient temperature was suitable for isoindole formation.

Work validation
The procedure of the assay was strictly validated by following directives of ICH guidelines. 21 The inspected parameters were linearity range, sensitivity (LOQ, LOD), precision, accuracy, and robustness.

Linearity range and sensitivity
Construction of the calibration graph was established at seven standard MLC concentrations. Each MLC concentration was analyzed 3 times by the approach procedure. The uorescence intensities values were plotted against MLC concentration to derive regression equation. The value of correlation coefficient (0.9998) indicates excellent linearity between response and MLC concentration over the range of (200-4000 ng mL À1 ). The statistical data and regression equation was found in (Table 1). The sensitivity of the proposed approach was assessed by LOQ and LOD which is estimated by using equations of ICH guidelines. The equations state that LOQ ¼ 10 s/b and LOD ¼ 3.3 s/b. It is known that (s) is the intercept (SD) standard deviation while (b) is calibration slope. The calculated values were 150.4 and 49.6 ng mL À1 for LOQ and LOD respectively.

Accuracy
Analysis of ve MLC concentrations within the calibration curve was performed by the proposed approach. Each MLC concentration was analyzed 3 times and the results were estimated as percentage recovery and the estimated standard deviation. The data in (Table 2) exhibit excellent harmony between the estimated and actual values which conrm the accuracy of the work procedure.

Precision
Intra-day precision which represents repeatability and inter-day precision were checked at ve MLC concentrations within linearity. While intra-day precision was performed on the same day (triplicate measurement for each concentration), inter-day precision was veried on three consecutive days. Aer data processing, the calculated RSD values were lower than 2 ( Table 3) which in agreement with ICH guidance and prove the procedure precision.

Robustness
Approach robustness was studied by an intended small variation of the assay parameters. The variables varied were pH, buffer volume, o-phthalaldehyde, and 2-mercaptoethanol. The results presented in (Table 4) conrm that a small variation of the reaction conditions has no signicant change in the results which indicates the robustness of the method.

Pharmaceutical tablets application
In the Egyptian market, MLC was formulated in many tablets formulation, among them Milnavella or Averomilan®. The proposed method was applied for the determination of MLC in the cited products by following the general assay procedure. A comparison of the obtained results with the reported uorimetric method 9 was performed by calculating Student's t-test and F-test to ensure accuracy and precision. At the condence limit 95%, the calculated values were lower than the tabulated values (Table 5). This an evidence of the suitability of the approach in dosage form analysis.

Approach application in spiked urine and plasma
Aer 2 h of MLC (50 mg) tablet administration, the maximum plasma concentration (135 ng mL À1 ) is reached. Also, about 60% of MLC excreted in the urine without changing. 9 Approach application for MLC analysis in spiked (plasma/urine) was established and validated without any notable interference. The data in (Table 6) represent mean recovery of four spiked concentration was estimated from the corresponding regression equation shows the suitability of the approach as a preliminary test for MLC assay in spiked urine and plasma.

Conclusion
The emphasis in this work was directed in the development of a new, feasible, and sensitive uorimetric method. The method offers many benets like simplicity as it implemented in a onepot without heating or extraction, low reagent price, and sensitivity. Furthermore, the measurements were monitored in a spectrouorometer that is more preferable than other analytical instrumentation. The priority of spectrouorometer is related to its low price, availability in most laboratories, measurement selectivity, and not require a high expert in usage or tedious sample preparation. The concept of the analytical approach relies on the condensation of MLC amino moiety with o-phthalaldehyde to generate isoindole uorescent derivative in the presence of 2-mercaptoethanol. The isoindole product was monitored at (l ex 338.5 nm, l em 433.5 nm) and condensation variables were strictly optimized. Finally, the approach applied for MLC analysis in bulk powder and dosage form and involved in urine and spiked plasma analysis with excellent percentage recovery.