High-Performance Liquid Chromatographic Ultraviolet Determination of Memantine Hydrochloride after In Vitro Transdermal Diffusion Studies

1 Departamento de Farmacia, Instituto de Ciencias Biomédicas, Facultad de Ciencias de la Salud, Universidad CEU Cardenal Herrera, Avenida Seminario s/n, Valencia, 46113 Moncada, Spain 2 Instituto de Reconocimiento Molecular y Desarrollo Tecnológico, Centro Mixto Universidad Politécnica de Valencia, Universidad de Valencia andDepartamento de Farmacia y Tecnología Farmacéutica, Universidad de Valencia, Avenida Vicente Andrés s/n, Valencia, 46100 Burjassot, Spain


Introduction
In recent years, the population has aged considerably.Advanced age is associated with damage of brain function, and the prevalence of degenerative diseases such as Alzheimer's disease has, therefore, risen.Alzheimer's disease is a progressive and fatal brain disease that manifests itself through many symptoms.e earliest observable symptom is memory loss, which becomes more intense as the disease advances, eventually constituting a severe state of amnesia [1,2].
Memantine (a noncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist) was developed in order to achieve better results in the treatment of this pathology.An alteration of the glutamatergic system provokes neurotoxicity and cognitive failure [3].Memantine acts on the glutamatergic system by blocking NMDA receptors and inhibiting their overstimulation by glutamate [4].Due to its mechanism of action, memantine can delay an Alzheimer's patient's cognitive and functional failure [5].
Randomized, placebo-controlled trial clinical studies have demonstrated the efficacy and safety of memantine in patients with mild to moderate vascular dementia [6].As a result, this antagonist, which has a different mechanism of action to that of acetylcholinesterase inhibitors, has been approved for the treatment of the symptoms that characterize the moderate-serious state of Alzheimer's disease [7].
e tolerability and safety data for memantine obtained in clinical trials show that adverse events due to this antagonist are infrequent and mild and include confusion, dizziness, headache, and fatigue [8].
e purpose of our research was to study the transdermal administration of memantine.e samples obtained in these experiments included endogenous substances released from the skin that could have interfered with subsequent analyses.
A review of the literature reveals that different analytical methods have been described for the determination of memantine hydrochloride.ey include HPLC with UV/Visible detection, alone [9,10] or in tandem with mass spectrometry (MS) [11,12], and �uorescence detection [10,[13][14][15].But they rely on an infrastructure that is not available to small laboratories, present too many interferences from endogenous compounds present in the skin, or provide an elevated range of concentration, thus not suitable for the purpose of our research.
In the present work, we have developed and validated a speci�c, accurate, precise, and robust HPLC method for the quantitative determination of memantine hydrochloride aer in vitro transdermal diffusion studies by means of ultraviolet detection aer dansyl derivatization procedure.
Six standard solutions (50, 25, 10, 5, 1, and 0.5 g/mL) were prepared by further dilution of the memantine stock solution with an isotonic buffer.ese standard fresh solutions were used for the calibration curves.(180-220 L) or an alkaline buffer solution [KOH (1.5 M)-KHCO 3 (1 M) (pH 10.6)] (200 L).It was then vortexextracted for 1 minute and centrifuged at 13000 rpm for 1.5 minutes.Following centrifugation, two phases were clearly separated; the lower layer contained the aqueous phase and the upper layer contained the organic phase, which consisted of ACN containing the extracted memantine.Different protocols were also assayed in order to optimize our derivatization procedure.An aliquot of 70 L of the acetonitrile containing the extracted memantine was removed from the upper layer and placed in a PCR tube (0.2 mL) or an Eppendorf tube (1.5 mL) and 40-60 L of an alkaline-buffered solution [KOH (1.5 M)-KHCO 3 (1 M) (pH 10.6)] or 50 L of a KOH solution (3 M) were added.e solution was then mixed with the solution of the derivatizing agent [40-60 L of a dansyl chloride solution (1.7 mg/mL in ACN), were created immediately prior to being used].e mixture was shaken vigorously for 5 seconds and maintained in the dark at 50 ∘ C during the derivatization reaction.Aer derivatization (35-55 minutes), the mixture was cooled in ice water for 10 minutes.Once again, two phases were obtained.An aliquot of 70 L of the acetonitrile was removed from the upper layer, and 10 L of HCl (1 N) was added to halt the reaction.

Extraction and
e derivatization reaction is shown in Figure 1, which shows how memantine was derivatized into a dansylmemantine complex that provided the chromophoric groups that enabled the UV/visible detection [16].
Once the chemical procedure had terminated, the derivatized memantine solutions were stored at −20 ∘ C until HPLC analysis.

Instrumentation and Chromatographic Conditions. A
Waters system equipped with a Waters 600 quaternary pump (including a Waters 996 UV/VIS diode-array detector), and a Waters 717 Plus automatic injector was employed for HPLC analysis.Acquisition and treatment of computerized data were performed with the Millenium Chromatography Soware.e instrumentation components were purchased from Waters Cromatografía, S. A.
Chromatographic separation of the analytes was achieved at room temperature (25 ∘ C ± 2 ∘ C) using a Kromasil C18 (150 × 4.6 mm) reverse-phase column packed with 5 m C18 silica particles.A mixture of hydrochloric acid water solution (0.01 M� pH 2.4)-methanol (15 : 85, v/v) pumped at a �ow rate of 1.2 mL⋅min −1 was used as a mobile phase.e injection volume was 50 L.Absorbances were measured at 218 nm.
2.5.Validation.e analytical method was validated with six different concentrations of memantine hydrochloride (0.5, 1, 5, 10, 25, and 50 g/mL).Each concentration was assayed �ve times to determine within-day reproducibility.In order to detect inter-day variation, the procedure described previously was repeated on ten different days.
Calibration curves were obtained by least square linear regression analysis of the peak area obtained as a function of the concentration of the drug.An average curve was constructed using all the data from all the calibration curves obtained.
Absence of interference was investigated by analyzing 10 blank isotonic solutions in order to determine the speci�city of the methods.e isotonic buffer used to prepare the standard solutions for calibration [NaCl (150 mM)-HEPES (20 mM) (pH 7.4)] was considered a blank isotonic solution.
e linearity of the calibration curves was tested by statistical comparison (Student's -test) of the slopes and intercepts of the calibration curves with zero and the correlation coefficient with 1, respectively.
Accuracy expresses the closeness of agreement between the calculated value and the accepted reference value, namely the conventionally true value.It was de�ned as the relative error of known concentration solutions.Measures had to be within ±10% for all concentrations to be considered acceptable [17,18].e precision of an analytical method expresses the closeness of agreement between a series of measurements obtained from multiple sampling of the same homogeneous drug solution under prescribed conditions.It provides information regarding random error.e variance of repeatability and intermediate precision, as well as the corresponding relative standard deviation (R.S.D.), were calculated from the estimated concentrations.To be considered acceptable, the R.S.D. had to be lower than 10% at all the concentrations analyzed [17,18].e detection limit (LOD) is the lowest amount of an analyte that can be detected but not necessarily quanti�ed as an exact value.e quanti�cation limit (LOQ) is the lowest amount of the analyte that can be quantitatively determined with de�ned precision under the experimental conditions.Assuming a normal distribution of measured concentration values, LOD and LOQ were calculated from the residual standard deviation of the regression data according to the criteria LOD = 3.3×(S.D./ and LOQ = 10×(S.D./ [19], where S.D. was the residual standard deviation and  was the slope of the linear regression equation.
e robustness of a method is its ability to remain unaffected by small changes in operating conditions.To determine robustness, the experimental conditions were deliberately altered at three different levels and evaluated for chromatographic response and retention time [20][21][22].
Factors were changed one at a time in order to study the effect.ese alterations were carried out by varying the detector wavelength (±2 nm (216 nm-220 nm)), the buffer and methanol ratio of the composition of the mobile phase (82 : 18-88 : 12, v/v), the pH of the mobile phase (2.2-2.6), and the �ow rate of the mobile phase (1.0-1.4 mL⋅min −1 ).
2.6.Stability Test.Six groups of 100 g/mL memantine hydrochloride saline-buffered solution were prepared to determine the stability of the dansylated memantine complex.
e �rst group was prepared by dilution of a 5 mg/mL memantine stock solution in order to obtain solutions at the appropriate concentrations.Once extraction and derivatization had �nalized, solutions were quanti�ed by HPLC in the dark.e rest of the groups were prepared in an identical manner and following extraction and derivatization, the solutions were stored in darkness under different conditions: freezer (−80 ∘ C and −20 ∘ C), refrigerator (4 ∘ C), laboratory stove (37 ∘ C), and at ambient temperature (25 ∘ C ± 2 ∘ C).All these groups were assayed to determine the residual concentration of memantine at 8, 24, 48, and 120 hours aer preparation.

Application of the Method to Determine In Vitro Transdermal Diffusion of Memantine
Hydrochloride. e validated HPLC method was used to quantify memantine hydrochloride in the samples obtained from in vitro transdermal diffusion studies.ese experiments were performed employing vertical standard diffusion cells (Franz type) purchased from DISA (Milan, Italy) with a diffusion area of 0.567 cm 2 and skin from pig ear as the membrane.
Pig ears were generously provided by the Faculty of Medicine, University of Valencia (Valencia, Spain) immediately following the death of the animal.
One mL of a solution of memantine hydrochloride (2.7 mg/mL) prepared in an isotonic buffer [NaCl (150 mM)-HEPES (20 mM), pH 7.4] was placed in the donor compartment.e receptor compartment (4.2 mL volume) was �lled with the same isotonic buffer (pH 7.4), which was thermostated at (36.93 ± 0.13) ∘ C ( = 36) and stirred by a rotating Te�on-coated supermagnet placed inside the cell to prevent boundary layer effects.Two hundred L samples were removed manually from the receptor chamber at predetermined time intervals.e volume of sample removed was replaced with the same volume of buffer pH 7.4.e memantine hydrochloride contained in each sample was recorded in order to calculate the accumulative amount in the receptor compartment.At the end of the in vitro transdermal diffusion experiments the amount of drug retained in the skin was determined by the HPLC method aer extraction by shaking the skin for 12 h with 1 mL of the isotonic buffer (pH 7.4).e transdermal �ux () was estimated from the slope of the linear region (steady-state portion) of the plot of the accumulated amount of memantine hydrochloride (g/cm 2 ) against time (hours).

Results and Discussion
In this work, we describe a new, speci�c, accurate, precise, and robust HPLC method for the qualitative and quantitative analysis of samples of a noncompetitive NMDA receptor antagonist, namely, memantine hydrochloride.
As previously described, because of memantine hydrochloride's on-existent UV absorbance, a conjugation reaction (dansyl derivatization) is necessary in order to introduce a chromophoric group that allows UV/Visible detection (see Figure 1).For this purpose, an extraction procedure was performed before the derivatization reaction to avoid interference peaks of skin endogenous compounds and to concentrate the memantine presented in isotonic solutions.e results of the optimization of the memantine extraction procedure are shown in Table 1.
ere are several factors that can have a bearing on the extraction of memantine.e volume of the aliquot determines the total amount of memantine to be extracted but also can modify the separation of the organic and aqueous layers and their appearance.Two other relevant factors are the volume of the organic dissolvent (acetonitrile (ACN)) and the volume and nature of the alkaline solution (KOH or KOH-KHCO 3 ) employed, as the procedure can alter the extent of the separation of the organic and aqueous layers.
e best protocol for optimizing the extraction procedure consisted of mixing an aliquot of 200 L of the memantine solution with 180 L of ACN.In these conditions, the concentration of memantine was sufficient for the analysis of the most diluted isotonic solutions, and the volume of ACN allowed the upper layer to be recovered adequately.
e addition of 200 L of an alkaline buffer solution [KOH (1.5 M)-KHCO 3 (1 M) (pH 10.6)] did not produce an appropriate separation of the organic and aqueous layers.For this reason the solution was rendered alkaline by adding 200 L of a KOH solution (3 M). is volume was selected because higher or lower volumes did not modify the separation of the layers or, indeed, the concentration recovery.
Once the memantine was extracted from the isotonic solutions, the derivatization procedure was performed.As in the previously described protocol, several factors can modify the derivatization of the memantine extracted from isotonic solutions.Many conditions can modify the output of dansyl derivatization, such as the capacity of the tube in which the reaction takes place, the volume and nature of the alkaline solution used (pH of medium), the volume of the dansyl chloride solution, and the reaction time [23].e results of the optimization of the memantine conjugation reaction are shown in Table 2.
PCR tubes (0.2 mL) were selected because when the reaction was carried out in Eppendorf tubes (1.5 mL), the concentration recovery obtained was 19.8 ± 4.4%.is low value was probably due to the temperature at which the reaction took place (50 ∘ C), which would have led to part of the solvent (especially ACN) evaporating inside the tube.When an Eppendorf tube was used, an empty space appeared high up inside the tube, which interfered with the reaction; this did not occur when a PCR tube was employed, as the recipient was practically full with the reagents.Cleaner chromatograms can be obtained if a minimum amount of dansyl is used [24], but it is important that there is enough reagent to ensure the complete dansylation of all the memantine present in isotonic solutions.Other authors have selected different concentrations for derivatization in the range of 2-30 mg/mL of dansyl in ACN [23,[25][26][27], but the concentration used in our study (1.7 mg/mL) provided enough dansyl for an optimal derivatization of memantine and did not produce any interference with the separation and detection achieved with HPLC.e volume of the dansyl ACN solution that we employed (50 L) produced a concentration recover of 100% (see Table 2).
As stated previously, reaction time and temperature are also important factors that determine the efficacy of a derivatization procedure.Other authors have employed different temperatures and reaction times to those used herein; Kang [25,26], and the derivatization procedure of Escribano and Legaz lasted 16 hours and was carried out at room temperature [27].As these results show, the higher the temperature applied, the less time was required for the reaction to be completed.We have determined a temperature of 50 ∘ C and a period of 45 minutes as optimal conditions (see Table 2).Although higher temperatures can reduce the reaction time, they also increase the rate at which the solvent evaporates.
e alkaline buffer solution employed and the volume employed were found to be relevant factors.e addition of 50 L of a KOH solution (3 M) produced a concentration recovery of 10.0 ± 1.0%.When an alkaline buffer solution [KOH (1.5 M)-KHCO 3 (1 M) (pH 10.6)] was used, the optimal volume for the reaction was 50 L (see Table 2).
We employed a 150 mm reverse phase column (Kromasil C18) to produce the highest resolution of the dansylated memantine peaks.Dansylated memantine spectra resulted in three maximum peaks of absorbance at 218.0, 249.7, and 332.6 nm (Figure 2).A wavelength of 218 nm was selected since it permitted the quanti�cation of lower concentrations of memantine without interferences of skin endogenous compounds.A representative dansylated memantine chromatogram is shown in Figure 3. e retention time of dansylated memantine obtained was 6.95 ± 0.01 minutes (mean ± S.D.;  𝑛 10).
Ten blank isotonic solutions were analyzed to evaluate speci�city.No signi�cant interfering peaks were observed at the retention time for the dansylated memantine obtained.
Ten calibration curves were assayed; these calibration curves were constructed by plotting the concentration of the analyte against the peak area response.e linearity between the response () and the corresponding concentration () over the 0.5-50 g/mL concentration range was evaluated.
e results of the least square linear regression analysis revealed correlation coefficients ≥0.999.e slope of the calibration curve was statistically different from zero (  0.05), and the intercept was not statistically different from zero (  0.05).
Intraday and interday precision and accuracy were evaluated by analyzing each concentration of memantine standard ) F 4: e accumulated amounts in receptor compartments (g/cm 2 ) plotted against time (hours).
and its relative error values.Sets of �ve replicates were studied on the same day and on ten different days.Accuracy and precision were within acceptable limits: the values obtained for all concentrations were below 10% (see Table 3).e limit of detection of dansylated memantine was found to be 0.144 g/mL, and the limit of quanti�cation was established as 0.437 g/mL.e results of the robustness of the method are shown in Table 4. Variation of the detector wavelength, composition, pH, and �ow rate of the mobile phase had no signi�cant effect on the retention time and chromatographic response of the method, thus indicating that the method was robust.Six groups, each stored under different conditions, were employed to evaluate the stability of the dansyl-memantine complex.As can be seen in Table 5, all the assayed conditions maintained a concentration of memantine-derivate above 90% for at least �ve days.
e potential use of this method has been demonstrated by the study of in vitro transdermal permeation of memantine hydrochloride.Memantine contained in samples obtained from in vitro diffusion experiments has been �uanti�ed by means of the HPLC method validated and reported in this paper.e accumulated amount of memantine in receptor compartment (g/cm 2 ) plotted against time (hours) is shown in Figure 4. e memantine transdermal �ux value across porcine skin was 7. ± . g/cm 2 ⋅h [mean ± S.D.;   8], and the amount of drug retained in skin aer 10 hours of diffusion was 86.3 ± 18.2 g/cm 2 [mean ± S.D.;   8].

Conclusions
We have developed and validated a new isocratic HPLC-UV method for the acute and precise determination of memantine hydrochloride.e results of the validation prove the method to be satisfactory in terms of speci�city, limit of detection and �uanti�cation, accuracy, precision, and robustness.erefore, the method is suitable for the routine determination of memantine hydrochloride presented in samples obtained from transdermal diffusion studies.

F 1 :
Derivatization.Different protocols were carried out to optimize the extraction procedure, all of which consisted of mixing an aliquot of the memantine solutions (which varied between 180 and 220 L) with acetonitrile (in a volume of 160 to 200 L) in an Eppendorf tube (1.5 mL).In order to separate the organic and aqueous layers, the mixture was rendered alkaline by adding a KOH solution (3 M) Diagram of the dansylation reaction with memantine.
et al. chose a reaction time of 30 minutes at 80 ∘ C [23], Minocha et al. and Smith and Davies established a reaction time of 60 minutes at 60 ∘ C

T 1 :
Optimization of the memantine extraction process.
T 3: Intraday and interday precision and accuracy of the HPLC method.Stability of the dansyl-memantine complex in solution.Concentration of the molecule complex is expressed as the percentage respective to the initial concentration (mean ± S.D.).