1. Introduction
Oxidative stress can be associated with many pathologies [
1,
2]. Together with other compounds (vitamin C, vitamin E, thioctic acid, glutathione, etc.), zinc provides protection against oxidative damage [
1,
2,
3].
Zinc is the second most abundant metal in the human body [
4]. Practically, it is an oligo-element essential for human health because it is a component of proteins involved in cell structures [
5] and on the other hand it is a cofactor for more than 300 enzymes involved in many processes, including the cellular respiration, immunity, DNA and protein synthesis, metabolism, cell division, etc. [
3,
4,
5,
6,
7]. Thus, its deficiency can lead to serious damage, being responsible for different pathologies, for example, the acute deficiency decreases the innate and adaptive immunity and chronic deficiency increases the production of pro-inflammatory cytokines and leads to severe inflammatory diseases such as rheumatoid arthritis [
5]. Obese patients showed alteration of zinc metabolism and that can lead to endocrine disorders. Zinc-α2-glycoprotein (Zn-2α-G) is an adipokine with lipid-mobilizing and anti-inflammatory activity. Thus, low levels of serum zinc and Zn-2α-G were identified in obese patients [
8]. On the other hand, blood clotting disturbances are related to zinc intake: hypozincemia leads to poor aggregation of platelets and prolongation of bleeding time and, opposite, hyperzincemia leads to increased coagulability [
9].
The zinc supplementation is efficient in the management of patients with type 2 diabetes mellitus (DM2), which exbibit hypozincemia (possibly due to reduced gastrointestinal absorption). It was observed that zinc supplementation in DM2 patients resulted in reduced glycated hemoglobin percentage [
10]. Likewise, zinc supplementation was proved to reduce severity and duration of diarrhea in children because of intestinal mucosa regeneration and facilitation of electrolyte transport [
11]. According to another study, zinc is linked to oxidative stress and may be important in prevention and treatment of sarcopenia [
12]. Additionally, zinc proved to be effective in the treatment of acne vulgaris, in several clinical trials [
13]. Active transport of zinc inside prostate cells is performed by a zinc transport protein (ZIP1). ZIP1 downregulation may be involved in the decrease in zinc inside malignant cells. Zinc treatment of human prostate tumors was demonstrated [
14].
Nevertheless, zinc may exhibit several adverse events and may interact with quinolones and tetracycline (decreasing antibiotic absorption) [
15]. On the other hand, using zinc in the treatment of neonatal hyperbilirubinemia is not encouraged [
16]. Therefore, dietary supplements (DS) can be useful in treatment of several pathologies, but special attention must be paid to nutrivigilance issues [
15].
Several zinc complexes with active pharmaceutical ingredients were synthetized and tested in order to improve biological properties of the parent molecules [
17,
18,
19,
20,
21]. Thus, for improving their efficiency and reducing the risks, the quality control of DS is essential for their biopharmaceutical profile. Many papers were published describing quantitative determination of zinc in pharmaceutical and biological matrices: neutron activation analysis [
22], AAS [
23,
24], voltammetry [
25,
26], differential-pulse polarography [
27], spectrofluorimetric analysis [
28], electrodialysis coupled with AAS [
29], ion-selective electrode [
30], capillary zone electrophoresis [
31], solid phase photometry [
32,
33], derivative spectrophotometry [
34,
35,
36,
37], and 0 order spectrophotometry [
37,
38,
39,
40,
41,
42,
43,
44,
45,
46,
47,
48,
49,
50,
51,
52,
53,
54]. Nowadays pharmacopoeias recommend quantification on zinc in dietary supplements by means of atomic absorption spectrometry (AAS) [
55]. The AAS method requires expensive equipment and relatively expensive reagents.
Zinc forms binary [
56] and ternary [
57] complexes with Xylenol Orange (XyO) in acidic buffered media. A ternary complex forms in micellar media with cetylpyridinium chloride (CtyPyChl) in the ratio 1:2:4 (Zn
2+:XyO:CtyPyChl). This method was used for quantitative determination of zinc in an ointment [
57].
Considering these, the aim of the present study is the development and validation of a cheap and fast spectrophotometric method for the determination of zinc (by means of a ternary complex with XyO) from a DS formulated as a solid pharmaceutical dosage form for oral administration, taking into consideration the latest validation guidelines and protocols [
58,
59,
60,
61].
2. Materials and Methods
2.1. Reagents for Spectrophotometric Method
Xylenol Orange (XyO) was produced by Sigma-Aldrich, cetylpyridinium chloride (CtyPyChl) was purchased from Merck, Chimopar supplied zinc sulphate heptahydrate, sodium acetate trihydrate, glacial acetic acid, sodium hydroxide, and hydrochloric acid 0.1 M. All reagents were analytical grade (purity > 99.5%), and in all experiments, ultrapure water was used (conductivity 0.05 µS/cm).
2.2. Reagents for Atomic Absorption Spectrometry Method
Multielement standard solution (zinc concentration 100 mg/L) and 37% hydrochloric acid were produced by Merck and potassium chloride was supplied by Sigma Aldrich. All reagents were analytical grade (purity > 99.5%), and in all experiments, ultrapure water was used (conductivity 0.05 µS/cm).
2.3. Raw Materials and Industrial Batches
A synthetic mixture of dietary supplement components (SMDSC) was prepared with calcium carbonate (Shanghai Nuocheng, Shanghai, China), magnesium oxide (Dr. Paul Lohmann, Emmerthal, Germany), mannitol (Roquette, Lestrem, France), pregelatinized corn starch (Colorcon, Harleysville, PA, USA), Kollidon 30 (Basf, Ludwigshafen, Germany), Aerosil 200 (Evonik Industries, Essen, Germany), zinc oxide (Dr. Paul Lohmann, Emmerthal, Germany), glyceryl dibehenate (Gattefossé, Saint-Priest, France), and cholecalciferol (DSM Nutritional Products Ltd., Sisseln, Switzerland) and four industrial batches of the DS “Calciu + Magneziu + Zinc + Vitamina D3” (Polisano Pharmaceuticals S.A., Sibiu, Romania) were used for method development.
2.4. Apparatus and Software
An atomic absorption spectrometer NovAA® 400 with integrated software from Analytik Jena, was equipped with zinc hallow cathode lamp (also from Analytik Jena, Germany) and was operated for the flame technique (acetylene–air). The used spectrophotometer was: (s1) UV-1900 spectrophotometer with UVProbe 2.7 software (Shimadzu, Kyoto, Japan). For robustness, a second spectrophotometer was used: (s2) Thermo Evolution 300 with VISION Software Suite (Thermo Fisher Scientific, Waltham, MA, USA). Weighing was performed on Mettler Toledo balances (ME104T/00 and ME2002T/00) and calcination of samples took place in a Naberthem furnace. pH readings were performed with SevenExcellence (Mettler Toledo, Columbus, OH, USA) multiparameter equipped with a glass electrode and for heating and sonication a LBS2 4.5L ultrasonic water bath from Falc Instruments was used.
2.5. Synthetic Mixture of Dietary Supplement Components Preparation
The SMDSC was prepared according to the notification no. 3083/25.07.2016 (issued by the Ministry of Health from Romania) for the dietary supplement “Calciu + Magneziu + Zinc + Vitamina D3” (notification holder Polisano Pharmaceuticals S.A., Sibiu, Romania). Appropriate amounts of raw materials were weighed on a three-decimal balance. The raw materials were mixed in a mortar with a polyethylene terephthalate card, for 20 min.
2.6. Calibration, Sample Preparation, and Spectra Recording for the Spectrophotometric Method
A series of stock solutions were prepared by weighing suitable amounts of zinc sulphate heptahydrate (A), XyO (B), and CtyPyChl (C). Concentrations of stock solutions were 10 mM for Zn2+, 1 mM for XyO, and 10 mM for CtyPyChl. Aliquots ranging from 20 to 60 μL (A) were transferred to 25 mL volumetric flasks. In each flask, 2 mL of stock B and 1 mL of stock C were transferred. Each flask was filled to the mark with acetate buffer (0.5 M, pH = 5.5). Suitable amounts of real samples (both SMDSC and industrial batches) were weighed and transferred to 250 mL volumetric flasks. In each flask, 100 mL of hydrochloric acid 0.1 M was added. Flasks were sonicated for 15 min at 30 °C and filled up to the mark with the same solvent. Solutions were filtered and aliquots of 2 mL were transferred in 50 mL volumetric flasks together with 4 mL of stock B and 2 mL of stock C.
Spectra of calibration solutions and real samples were recorded, in glass 1 cm cuvettes, considering a reagent blank, in the wavelength interval 400–600 nm.
2.7. Calibration and Sample Preparation for the AAS Method
From multielement standard solution (100 mg/L), 5 aliquots (ranging 100 to 1200 μL) were transferred to 100 mL volumetric flasks together with 100 mg of potassium chloride. Flasks were filled to the mark with hydrochloric acid 0.125 M. Blank solution was prepared from 100 mg of potassium chloride dissolved in hydrochloric acid 0.125 M, in a 100 mL volumetric flask. Suitable amounts of real samples (industrial batches) were transferred in porcelain crucibles and calcinated for at least 6 h at 550 °C. After cooling, 15 mL of 37% hydrochloric acid was added to each crucible heating slowly on the water bath for 30 min. The mix was transferred to a 100 mL volumetric flask and the flask was filled to the mark with purified water. Then, 800 μL of this solution was transferred to a 100 mL volumetric flask together with 100 mg of potassium chloride and the flask was filled to the mark with hydrochloric acid 0.125 M. Absorbances were measured at 213.9 nm using a zinc hallow cathode lamp and deuterium background.
2.8. Spectrophotometric Method Validation Protocol
2.8.1. Linearity
Linearity was tested at 5 concentration levels (k = 5) in triplicate for both standard and SMDSC. Linearity was tested on three different days. Calibration curves (3 for standard and 3 for SMDSC) were plotted by linear regression. The correlation coefficient (r) should be above 0.99. For linearity, the following statistical tests were performed: Cochrane test (for intragroup variances homogeneity testing), Fisher test (slope significance evaluation and regression curves validity), t Student test (intercept comparison with null). Furthermore, t Student tests were performed (regression curves slopes and intercepts comparison) for SMDSC. In all tests, a 5% error probability was considered for all k = 5 concentration levels.
2.8.2. Precision
In order to prove the precision of the proposed method, repeatability and reproducibility were evaluated. Determinations were performed on a single concentration level (100%), 6 replicates per day on three different days (a total of 18 samples). The following statistical tests/parameters were performed/calculated: intragroup variances homogeneity (Cochrane test), repeatability variation coefficient (CVr%), and reproducibility variation coefficient (CVR%). In all tests, a 5% error probability was considered for all 18 samples.
2.8.3. Accuracy
Accuracy was estimated in terms of recovery using a calibration curve (constructed as described in the linearity protocol). Statistical analysis of accuracy consisted of s Cochrane test (for intragroup variance homogeneity), Fisher test (medium recovery validity), and t Student test (confidence interval for mean recovery), considering a 5% error probability.
2.8.4. Limit of Detection and Quantification
Limit of detection (LOD) and limit of quantification (LOQ) are parameters related to analytical method sensitivity and were calculated based on Equation (1).
where σ is the standard deviation of the intercept of the calibration curve and s is the slope of the calibration curve.
2.8.5. Robustness
For the study of robustness, two parameters were varied: (1) apparatus and (2) zinc to other components ratio. The calibration curves were determined on two spectrophotometers (s1 and s2). Two SMDSCs were prepared by using zinc levels of 80% and 120% considering the industrial formula as the 100% level. Six samples were analyzed for each level (80% and 120%).
2.9. Statistical Comparison of Spectrophotometric and AAS Results
Comparison of results generated by the proposed method and AAS method was performed by means of a paired t Student test, considering a 5% error probability.
All statistical tests described in the validation protocol and comparison of spectrophotometric and AAS results were performed using Microsoft Excel® (Office 365 personal edition).
4. Discussion
Due to its important biological role, zinc is often used as a dietary supplement. In order to use safe and efficient products for improving the quality of patients’ lives, the manufacturer should improve the quality control process using reliable, robust, easy to apply, and inexpensive methods of analysis. Thus, several modern analytical techniques are used for zinc determination from pharmaceutical forms. These are complex and difficult to apply in situ. On the other hand, they require expensive laboratory equipment, specialized staff, and a long period of time for processing. Spectrophotometric methods have a series of advantages over other advanced techniques, such as: simplicity, speed, low cost and maintenance, portable instrumentation, etc. Additionally, these methods should be validated, because they represent real competitive techniques for quality control of dietary supplements, if the experimental conditions are rigorously controlled [
62].
Table 6 presents a comparison of current analytical method characteristics with previously published 0 order spectrophotometric methods, for Zn
2+ determination in solid pharmaceutical dosage forms.
As presented in
Table 6, the proposed method has a comparable range with other published 0 order spectrophotometric methods. The novelty of the method we described consists of a better characterization of the ternary complex (absorption minimum and isosbestic point). Quantitative determination could be performed at minimum wavelength and the presence of an isosbestic point could be valuable for the development of methods for quantitation of a mixture of ions that can form complexes with Xylenol Orange, in the future.
The current work did not try to study the whole range for which Beer’s law was obeyed, but to cover the linearity interval 80–120%, as recommended by the ICH guideline. Several statistical tests were applied in order to perform a statistical evaluation of linearity (Cochrane, Fisher, and t-Student).
The correlation coefficient is higher than 0.999 in the case of all presented spectrophotometric methods. Despite some of the methods presented in
Table 6, the current method does not involve digestion with concentrated acids, calcination, or liquid–liquid extraction with toxic solvents.
Additionally, the current method is quite sensitive, considering its LOD.
In terms of precision, the intragroup variances are homogeneous (Cochrane test was passed). Moreover, CVr% and CVR% were 1.57% and 1.77%, respectively.
The statistical evaluation of accuracy revealed that intragroup variances are homogeneous (Cochrane test was passed), medium recovery is valid (as Fisher test showed), and confidence interval of the mean recovery is narrow and very close to 100%.
For the robustness study, the two variated parameters were apparatus and zinc to other components ratio. The intercept and slope of the calibration curves determined by means of the two spectrophotometers are comparable. Additionally, the confidence intervals of the mean recovery regarding the analyzed zinc levels (80% and 120%) are comparable to the 100% one.
Finally, statistical comparison of the results obtained by means of the current method and AAS method revealed no differences between results.