Green Analytical Method for Simultaneous Determination of Glucosamine and Calcium in Dietary Supplements by Capillary Electrophoresis with Capacitively Coupled Contactless Conductivity Detection

The need for analytical methods that are fast, affordable, and ecologically friendly is expanding. Because of its low solvent consumption, minimal waste production, and speedy analysis, capillary electrophoresis is considered a “green” choice among analytical separation methods. With these “green” features, we have utilized the capillary electrophoresis method with capacitively coupled contactless conductivity detection (CE-C4D) to simultaneously determine glucosamine and Ca2+ in dietary supplements. The CE analysis was performed in fused silica capillaries (50 μm inner diameter, 40 cm total length, 30 cm effective length), and the analytical time was around 5 min. After optimization, the CE conditions for selective determination of glucosamine and Ca2+ were obtained, including a 10 mM tris (hydroxymethyl) aminomethane/acetic acid (Tris/Ace) buffer of pH 5.0 as the background electrolyte; separation voltage of 20 kV; and hydrodynamic injection (siphoning) at 25 cm height for 30 s. The method illustrated good linearity over the concentration range of 5.00 to 200 mg/L of for glucosamine (R2 = 0.9994) and 1.00 to 100 mg/L for Ca2+ (R2 = 0.9994). Under the optimum conditions, the detection limit of glucosamine was 1.00 mg/L, while that of Ca2+ was 0.05 mg/L. The validated method successfully analyzed glucosamine and Ca2+ in seven dietary supplement samples. The measured concentrations were generally in line with the values of label claims and with cross-checking data from reference methods (HPLC and ICP-OES).


Introduction
Glucosamine (2-amino-2-deoxy-D-glucose) is an amino monosaccharide with essential roles in the biochemical synthesis of glycosylated proteins and lipids [1]. Intensive studies have proved that the exogenous use of glucosamine can relieve osteoarthritis (OA) symptoms and restore articular functions [2][3][4][5]. It also has many antiinfammatory and antioxidative efects with minimal adverse impact on human health [6][7][8][9][10]. Tus, glucosamine is suggested for relieving pain and preventing or slowing the breakdown of cartilage in OA by the Osteoarthritis Research Society International (OARSI) [11]. Te European League Against Rheumatism (EULAR) recommended glucosamine as a treatment for knee OA [12]. Te United States Food and Drug Administration (FDA) also proposed glucosamine as a dietary supplement used to manage and treat OA [13].
Because of its benefts, glucosamine products are one of the most popular over-the-counter (OTC) dietary supplements worldwide. Besides its use as a single supplement in the form of glucosamine hydrochloride or glucosamine sulfate, it can be mixed with other nutraceuticals, vitamins, and minerals. Among them, calcium is commonly combined with glucosamine because of its crucial role in preventing osteoporosis [14]. Given that there are diferent combinations of glucosamine and other supplements, more than one method is required to analyze these products. For example, the most popular method for glucosamine analysis is reversed-phase high-performance liquid chromatography (RP-HPLC), coupled with various detectors like ultraviolet, electrochemical, and mass spectrometry [15][16][17][18][19]. However, this method is not suitable for analyzing minerals like calcium, which have commonly been determined by inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma optical emission spectrometry (ICP-OES), or fame atomic absorption spectrometry (F-AAS). Due to the complexity of current methods for the quantifcation of glucosamine products, there is a need to develop a simple technique to simultaneously determine the amounts of glucosamine and other components in the food supplement.
Capillary electrophoresis (CE) can be considered a more economical alternative for developing simple and cheap analytical techniques. CE instruments are coupled with various detection methods such as ultraviolet, laser-induced fuorescence, mass spectrometry, voltammetry, and amperometry. CE coupled with capacitively coupled contactless conductivity detection (C 4 D) has recently been highlighted as an economical and efcient method for environmental monitoring, [20][21][22], food control [23][24][25], forensics [26], and clinical analysis [27][28][29]. Te working principles and applications of CE-C 4 D have been reviewed by Hauser and Kubáň [30] and others [29,[31][32][33]. Tese techniques have been applied to analyze inorganic, organic ions, and biomolecules. Because of the ability to analyze the broad range of targets, we have developed a protocol to simultaneously quantify both glucosamine and calcium in dietary supplements by CE-C 4 D. A simple procedure has been reported, with cross-checking data using standard high-performance liquid chromatography with a fuorescence detector (HPLC-FLD) and ICP-OES methods being provided to prove the reliability of the CE-C 4 D results. We also applied this protocol for the quality control of seven dietary supplements available in Vietnam.

Materials.
All chemicals used were of analytical grade and were used as received without any further purifcation. All solutions were prepared with deionized water of resistivity not less than 18.2 MΩ cm −1, which was purifed by the WaterPro RO system (Labconco Cooperation, Kansas, MO, USA).
Both glucosamine and calcium nitrate were purchased from Merck KGaA (Germany). Te stock solutions of glucosamine (1000 mg/L) and Ca 2+ (1000 mg/L) and standard solutions of calibration curves were prepared with deionized water in volumetric fasks and stored at 4°C before further use.

Sample Preparation.
Dietary supplement samples, including tablets, and hard capsules, were purchased from pharmacies in Hanoi, Vietnam. Te sample preparation procedure was described in previous studies [34,35]. Briefy, each sample was prepared with the contents of 20 whole tablets or capsules. Te tablets were ground into a fne powder using a ceramic mortar and pestle. Te hard capsules were opened to collect powder; the samples were ground if necessary. Te sample was accurately weighed (about 0.5 g) before being transferred into a 25-mL volumetric fask; 4% TCA was added to the mark. Te fask was ultrasonicated for 30 min by an ultrasonic vibrator bath (BRANSON 521) and centrifuged at 3000 rpm for 15 min. After centrifugation, the aqueous phases were collected and passed through a 0.45-μm syringe flter. Te fltrated solution was transferred to a 25 mL volumetric fask and flled to the mark with 4% TCA. Te solution could be diluted with deionized water (if necessary) before analysis by CE-C 4 D. Concentrations of glucosamine and Ca 2+ were determined by using the standard addition method.

Instrument.
Te in-house CE-C 4 D instrument was presented in the previous studies [30,36,37]. Generally, the compact CE instrument was built in-house using high voltages of a maximum of 25 kV (UM25 * 4, 12 V, Spellman, Pulborough, UK) and combined with a commercial C 4 D detector (ER815, eDAQ, Denistone East, NSW, Australia). Te system was operated with a 220-VAC power supply. Te silica capillary had a total length (L tot ) and efective length (L ef ) of 40 cm and 30 cm, respectively, and an inner diameter of 50 μm.

Method Development and Validation.
Analytical parameter optimization followed strategies in our earlier works [26,34,35,38,39]. Overall, various factors of BGEs (composition, pH, and concentration), separation voltages, and injection conditions (siphoning height and injection time) were investigated to develop an efective CE-C 4 D method for the simultaneous determination of glucosamine and Ca 2+ . A single analytical parameter was changed at a time while fxing all the others. Te background noise, signal, and migration times of analytes were calculated from the CE electropherograms to identify the optimized condition (Tables S1-S6). Because background noise may change during the recording of an electropherogram, the average noise was measured near the analyte peak regions. For a reproducible and accurate measurement, the signal's peak area was used in all this study's steps. After comparing the analytes' signals, background noise, and retention time under diferent analytical conditions, the optimized CE-C 4 D conditions are summarized in Table S7.
Based on optimized CE-C 4 D conditions, the analytical method was validated for linearity, detection and quantifcation limits, repeatability, and recovery by using the analytical results of standard solutions and matrix-spike samples. In pharmaceutical analysis, guidelines from the International Council for Harmonization (ICH) and EURACHEM are commonly used to calculate the limit of detection (LOD) [40,41]. Tey defne the LOD as the concentration of analyte that generates a signal at least 3 times the signal noise of the baseline (S/N � 3); for LOQ, the S/N ratio equals 10. Tus, in this study, to determine the LOD of each analyte (calcium or glucosamine), a standard solution was repeatedly diluted until its signal was 3 times higher than the background noise (S/N � 3). Te LOQ value of each analyte was measured by repeat dilution until the signal of the diluted solution was 10 times higher than the background noise (S/N � 10). All LOD/LOQ measurements were conducted under optimum conditions in Table S7.
Cross-check analysis was performed for both sample types (i.e., tablets and capsules). Glucosamine was analyzed with HPLC coupled with a fuorescence (FLD) detector, while ICP-OES detected Ca 2+ at the National Institute for Food Control (NIFC), Hanoi, Vietnam, according to methods reported elsewhere [42,43]. All measurements generally followed the AOAC Ofcial Method 2016.14. Detailed information about the reference methods was listed in Table S8.

Efects of Background Electrolytes.
Te background electrolytes (BGEs) are one of the most critical factors in CE-C 4 D method development [44]. In this work, we investigated four standard BGE systems, including Arg/Ace, His/Ace, Tris/Ace, and CAPS/Ace (Figure 1(a), Table S1). Te initial concentrations of the Arg, His, Tris, and CAPS solutions were 10 mM. All solutions were frst adjusted to a pH of 5.0 by acetic acid. In four BGE systems, both glucosamine and Ca 2+ were eluted before the electroosmotic fow (EOF) signal, indicating that the two analytes have positive charges under all studied conditions. Negative-going peaks of glucosamine show a conductivity reduction of the background bufer when this compound passes the detector. Although all BGE systems give a sharp peak of Ca 2+ , only the Tris/Ace solution provides a high signal of glucosamine (Figure 1(a)). Other BGE solutions cause low peak signals for glucosamine, leading to insufcient sensitivity. Tus, the BGE system containing Tris/Ace was selected because of the sufcient peak signal for Ca 2+ and glucosamine.
After selecting the BGE composition, the pH of the solution was further investigated (Figure 1(b)). Tis factor signifcantly infuences the glucosamine signal. In general, both peaks of Ca 2+ and glucosamine were far from the EOF signals at all fve pH conditions (pH 6.5-4.5). Te high pH declines the migration time of both analyzes because of the rise in the EOF mobility. At pH 6.0 and higher, the peak separation of glucosamine appears, possibly due to the conversion of glucosamine forms at diferent pHs. While β-anomer is dominant for nonprotonated neutral glucosamine, the protonated glucosamine mainly exists as α-anomer [45][46][47]. Te two peaks of glucosamine are more resolved in higher pH conditions. Two isomers of glucosamine also generate two peaks in HPLC analysis; thus, the sum of the area of these two peaks should be used to quantify the glucosamine [15,48,49]. Te pH of 4.5 to 5.5 is suitable to accurately detect total glucosamine by CE-C 4 D since glucosamine generates a single peak at these conditions, leading to a more straightforward calculation. All three pH conditions gave the stable baseline and the practical time for bulk analysis (Table S2). Te content of Ca 2+ is much lower than that of glucosamine in commercial supplements; thus, a condition that shows the best Ca 2+ signal should be chosen for the simultaneous determination of both analytes in supplement samples. Among all pH conditions, the pH of 5.0 exhibited the highest peak area of Ca 2+ signals and a single peak of glucosamine. Accordingly, a pH of 5.0 was chosen for further studies.
Tris concentration (from 8 to 20 mM) was optimized for simultaneous determination of glucosamine and Ca 2+ (Figure 1(c) and Table S3). Generally, the increase in Tris concentration shows a minor efect on the migration times of glucosamine and Ca 2+ . In the capillary tube, a high concentration of ions causes a change in the magnitude of the electric double layer; therefore, the sample moves unevenly, causing background noise, poor separation, and unbalanced peaks. Although high concentrations of Tris (≥10 mM) in BGE provide a stable baseline and balanced peaks, the intensity of the glucosamine peak declines at high Tris concentration, decreasing the sensitivity. Tus, compared to other conditions, the 10 mM Tris/Ace electrolyte solution was selected for further investigations because of both analytes' stable baselines and balanced peaks.

Efects of Separation Voltages.
Various applied voltages from 10 to 25 kV were then investigated to evaluate their impacts on electropherograms of glucosamine and Ca 2+ (Figure 2). Firstly, higher voltages resulted in shorter migration times and sharper peaks, but higher background noise was observed. Voltages of 10 or 15 kV can lead to broadening peaks of Ca 2+ and glucosamine, which could cause false results when analyzing real sample matrices.
Moreover, a high separation voltage raises the background noise (Table S4). Te noise at +25 kV showed the highest level, around 2.06 mV, while the background noise was only 0.67 mV at +10 kV. On the other hand, the area of the peaks from glucosamine and Ca 2+ decreased at a high applied voltage (Table S4). Lastly, the high voltage (higher than 20 kV) could generate audible noise, especially in high humidity conditions in a tropical country like Vietnam. Tus, it is recommended to use a low voltage in capillary electrophoresis if possible for safety. Given these points, a separation voltage of 20 kV is optimal for simultaneously determining glucosamine and Ca 2+ .

Efects of Injection
Conditions. Next, hydrodynamic injection conditions, including injection times and siphoning heights, were studied (Figure 3, Tables S5 and S6).
Tese parameters control the sample amounts injected into the capillary, afecting the separation and detection efciency. Peak areas are directly proportional to injection times and siphoning heights. Although longer injection times and higher injection heights can lower detection limits, too large injected sample amounts may degrade the peak shape and introduce elevated interferences. Considering these facts, we injected samples at a siphoning height of 25 cm in 30 s.

Selectivity.
Most commercial glucosamine supplements combine chondroitin and methylsulfonylmethane (MSM) to treat painful joints. Tus, the interference of chondroitin and MSM in the signals of glucosamine and Ca 2+ was investigated ( Figure 4). Tere is no detectable signal for chondroitin and MSM under experimental conditions. While a positive potential was applied to detect glucosamine and Ca 2+ (positive ions), chondroitin and MSM cannot form a positive charge under the studied condition. Chondroitin is a highly negative polysaccharide [50], and MSM, an organosulfur compound, is considered chemically inert with a pKa of 28. Tus, it is apparent that we cannot observe any signal of these two chemicals in the analyzing time (around 5 min). Moreover, the presence of chondroitin and MSM showed no interference with the peak signals of glucosamine and Ca 2+ . Tus, the proposed method is suitable for simultaneously determining glucosamine and Ca 2+ in commercial supplements.

Calibration Curves.
A series of mixture standard solutions ranging from 5.00 mg/L to 200 mg/L and from 1.00 mg/L to 100 mg/L of glucosamine and Ca 2+ , respectively, were analyzed at optimized conditions to construct two calibration curves of two analytes ( Figure 5). Te  regression equations of the calibration curves represent relationships between analyte concentrations (mg/L) and peak areas (mV.s). Good linearity was observed for both glucosamine (R 2 � 0.9994), and Ca 2+ (R 2 � 0.9994) over the concentration ranges ( Figure 5). Concentrations of analytes in analytical solutions (including real samples and standard addition samples) should be within the linear ranges.

Limits of Detection and Quantifcation.
Limits of detection (LODs) of glucosamine and Ca 2+ were determined using signal-to-noise ratios equal to 3 (S/N � 3). A standard solution of each analyte was repeatedly diluted until its signals were 3 times higher than the background noise measured near the peak area (S/N � 3) (Table S9). Te concentration of the analyte at S/N � 3 was recorded as the LOD value. LODs of glucosamine and Ca 2+ were 1.00 and 0.05 mg/L, respectively. For the LOQ calculation, we did a similar experiment until the analyte's signal was 10 times higher than the background noise (Table S9). LOQs were estimated to be 3.30 mg/L for glucosamine and 0.17 mg/L for Ca 2+ . Te LOD of glucosamine obtained by our method was higher than that reported by Akamatsu and Mitsuhashi (0.01 mg/L) [51]. However, their method required in-capillary derivatization with o-phthalaldehyde, which may not be suitable for determining both Ca 2+ and glucosamine in that condition. Our LOD value was comparable to the LODs of other studies using CE and HPLC (0.25 to 1.0 mg/L) [52,53]. Our method not only detects both analytes in a single run but also achieves low LOD and LOQ, which are compatible with pharmaceutical and nutraceutical analysis.

Repeatability.
Standard solutions with three concentration levels of glucosamine and Ca 2+ (20,40, and 80 mg/L) were chosen for replication analysis (n � 5) for  repeatability evaluation (Table S4). For Ca 2+ , relative standard deviations (RSD) of peak areas ranged from 0.92% to 2.16%. Otherwise, the RSD of the glucosamine peak areas ranged from 0.80% to 2.52%. Both glucosamine and Ca 2+ migration times in standards had an RSD <2.5% at all concentration levels. Tese method repeatability results meet the standard precision criteria proposed by AOAC International for concentration ranges from 1 ppm (11%) to 100 ppm (5.3%) [54].

Recovery.
Recovery is an essential parameter in evaluating the method of accuracy. In this study, we used a tablet sample without glucosamine and Ca 2+ (with no label claim of these components and confrmed with reference methods) to make matrix-spike samples. Tree spiking levels of glucosamine (40,80, and 100 mg/L), and Ca 2+ (4, 8, and 10 mg/L) were performed in triplicate. Te sample matrix (about 0.5 g) was spiked with a standard solution containing glucosamine and Ca 2+ , equilibrated for at least 30 min at room temperature, and treated according to the procedure described in Section 2.2. Recoveries of glucosamine and Ca 2+ ranged from 96.9% to 102.0% and from 98.0% to 103.0%, respectively (see Table S5). Tese recovery rates satisfy the AOAC requirement for expected recovery at concentration levels from 10 ppm (80-100%) to 100 ppm (90-107%) [54], indicating high accuracy with an insignifcant compound loss of our analytical procedure.

Method Application and Cross-Checked Results.
Te validated method was applied to seven real samples collected from pharmacies in Hanoi, Vietnam, including four tablets and three hard capsules (Table 1 and Figure 6(a)). To verify if signal peaks from samples were generated by glucosamine and Ca 2+ , known amounts of standard solutions were added to the sample before analyzing by CE-C 4 D (Figure 6(b)). Te intensities of both signals in the standard addition samples increased compared to those in the samples. Te migration times of glucosamine and Ca 2+ peaks were similar between standard and standard addition samples. It is shown that the supplement additives have a minimal impact on the migration time of both analytes. Tus, the content of glucosamine and Ca 2+ in supplements could be determined using the standard addition method. Based on the peak area obtained and the standard addition equation, the content of calcium and glucosamine in the functional food samples will be calculated.
Te contents of Ca 2+ in the samples measured by our CE-C 4 D method ranged from 7.82 to 55.43 mg per unit (e.g., tablet or capsule), with relative deviations compared to label claims ranging from −7.62% to +5.20%. Te contents of glucosamine ranged from 111.0 to 643.0 mg per unit, and measured-label deviations ranged from −5.26% to +8.10% (Table 1). Tere was an acceptable agreement between measured and label claim contents because deviations are less than ±10% in all samples. Te cross-checked analysis was performed for all the samples (Table 1). Te relative deviations between the results of CE and ICP-OES ranged from −8.08% to +5.62% for Ca 2+ , and between CE and HPLC, ranged from −5.93% to +6.47% for glucosamine. Good agreement between analyte contents measured by our method and the reference methods indicates that CE-C 4 D could be an alternative tool for conventional methods such as HPLC and ICP-OES in specifc applications.

Conclusions
Te present study developed an analytical method for the simultaneous determination of glucosamine and Ca 2+ using CE-C 4 D in commercial dietary supplements. Our method was demonstrated to be simple, fast, and costefective with adequate validation parameters such as selectivity, linearity, detection limits, repeatability, and recovery. Concentrations of glucosamine and Ca 2+ in supplement samples can be simultaneously determined in a single run analysis within fve min. Te validated method was applied to analyze samples, including tablets and hard capsules. Analytical results obtained by this CE-C 4 D method were in good agreement with labeling contents and those measured by reference methods (i.e., HPLC-FLD for glucosamine and ICP-OES for Ca 2+ ). Overall, our study has revealed another application of CE-C 4 D as an alternative green analytical tool for conventional dietary supplement quality control methods, with the advantages of high efciencies, low sample and electrolyte consumption, and minimal chemical waste. Because one simple analytical procedure can be applied for both analytes (organic compound and metal ion) with diferent properties, our method signifcantly reduced analysis time and labor cost compared to using one conventional method for each analyte. Te proposed CE-C 4 D method for glucosamine and Ca 2+ analysis also provided reliable results compared to standard techniques.

Data Availability
Te fndings of the study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest. Table S1. Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent BGE conditions. Table S2: Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent pH conditions. Table S3: Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent Tris concentrations. Table S4: Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent seperation voltages. Table S5: Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent injection time (siphoning at 25 cm). Table S6: Te background noise, peak area, and migration time of glucosamine and Ca 2+ at diferent siphoning height (injection time 30 s). Table S7. Te optimal conditions of the CE-C 4 D method for simultaneous determination of glucosamine and Ca 2+ . Table S8. Te optimal conditions for determination of glucosamine by HPLC-FLD and Ca 2+ by ICP-OES. Table S9. LOD and LOQ of glucosamine and Ca 2+ . Table S10. Te repeatability evaluation of peak area (mV.s) and migration time (min) for simultaneous determination of glucosamine and Ca 2+ by CE-C 4 D.