Simultaneous quantification by HPLC of the phenolic compounds for the crude drug of Prunus serotina subsp. capuli

Abstract

Context: Prunus serotina Ehrenb. subsp. capuli (Cav.) McVaugh (Rosaceae), commonly known as “capulin”, is a native North American tree, commercialized and used in folk medicine for the treatment of the hypertension, gastrointestinal illnesses, and cough.

Objective: This work developed a suitable HPLC method for quantifying the major active constituents of the infusion of P. serotina, the most important preparation consumed by populations around the world.

Materials and methods: The analytical method was performed using a Fortis-RP column (150 mm × 4.6 mm; film thickness 5 µm). The mobile phase consisted of an isocratic acetate buffer solution (pH 2.7; A) and methanol (B) (65:35 v/v) at a flow rate of 1.0 mL min⁻¹.

Results: The proposed method was applied to the quantification of 1–3 in several samples of the leaves of P. serotina. The results indicated that amounts of 1–3 in the samples analyzed are uniform, and greater amounts of chlorogenic acid (2; 479.9 ± 33.6 µg g⁻¹, dry matter) along with hyperoside (1; 185.7 ± 55.3 µg g⁻¹, dry matter) were present. On the other hand, benzaldehyde (3; 118.2 ± 12.1 µg g⁻¹ dry matter) was found to be in lower concentration.

Conclusions: A simple, sensitive, precise, and reproducible HPLC method for the simultaneous quantification of 1–3 in P. serotina was developed and validated. This is the first report on the quantification of 1–3 as active principles, and compound 1 was selected as a marker of P. serotina, which could be useful to guarantee the quality of the crude drug and herbal products.

• HPLC
• Prunus serotina
• quantification

Introduction

Prunus serotina Ehrenb. subsp. capuli (Cav.) McVaugh (Rosaceae), commonly known as “capulin”, “cerezo”, “cerezo negro”, “chencua”, “cusabi”, and “taunday”, among others, is a 60–90 cm foot-tall native North American tree that grows in the oak and pine woodlands. Leaves and stem bark of the species are extensively commercialized and used in folk medicine, mainly for treating cough. In combination with other herbs, such as black cherry, it has been used to alleviate gastrointestinal illnesses, kidney complaints, rheumatism, jaundice, and as a blood tonic (Argueta, 1994; Martínez, 1991). For centuries, Native Americans have used the plant for indigestion, worms, bums, labor pains, diarrhea, headache, bronchitis, and tuberculosis (Moerman, 1998). Besides their use in traditional medicine, the fruits of this plant are also a part of Mexican diet, and are consumed fresh, dried, or prepared in jam (Argueta, 1994; Martínez, 1991).

Phytochemical studies carried out on P. serotina resulted in the isolation of several flavonoids (Olszewska, 2005a,b), ursolic acid derivatives (Biessels et al., 1974), chlorogenic acid (2) (Olszewska, 2007), and prunasin (Horsley & Meinwald, 1981). In addition, P. serotina seeds contain some cyanogenic glycosides as the major secondary metabolites and the fruits mainly contain anthocyanins (Ordaz-Galindo et al., 1999; Sang et al., 2002; Santamour, 1998).

Pharmacological evaluation on the methanol extract prepared from the stem-bark displayed a significant anti-proliferative activity in cancer cells of human colon (Yamaguchi et al., 2006). Furthermore, the crude extract of the plant, as well as some isolates showed noted vasorelaxant activity. In particular, hyperoside (1), prunin, and ursolic acid displayed an important concentration-dependent relaxation of vascular smooth muscle; this pharmacological action is in agreement with the popular use of the plant for the treatment of hypertension (Ibarra-Alvarado et al., 2009).

Despite the unceasing popularity of P. serotina for primary healthcare needs, there are no procedures to guarantee the quality of the drug crude and their phytopreparations. Herein, an attempt to establish quality control and/or standardization procedures for its drug and herbal preparations for the assurance of safety, quality, and efficacy of herbal products has become an important concern. This study was undertaken in order to develop a suitable analytical high-performance liquid chromatography (HPLC) method for simultaneous determination of the major active constituents of the infusion of P. serotina, the most important preparation consumed by populations around the world. The developed method was fully validated and successfully applied for the analysis of several samples.

Materials and methods

Instrumentation and chromatographic conditions

HPLC analyses were conducted on a SHIMADZU 10AV HPLC system equipped with an UV/Vis dual detector (SPD-10 A), a quaternary pump (LC-10ATvp), vacuum degasser, auto sampler (SIL-10ADvp), column compartment (CTO-10 A), and ChemStation for Lab Solution/LC Solution software (Columbia, MD) was used for data handling. The development and quantification of the analytical method were carried out on a Fortis-RP reversed-phase column (150 mm × 4.6 mm; film thickness 5 µm; Fortis Technologies Ltd, Cheshire, UG). The mobile phase consisted of an isocratic acetate buffer solution (pH 2.7; A) and methanol (B) (65:35, v/v) at a flow rate of 1.0 mL min⁻¹, 10 µL of sample was injected and the detection wavelength was 254 nm in all cases. The column was kept at room temperature throughout all analyses.

Chemicals and plant samples

All HPLC and AR-grade solvents were acquired from Honeywell B & J (Seelze, Germany). Sodium acetate (AR grade), acetic acid (AR grade), chlorogenic acid (2), and benzaldehyde (3) were acquired from Sigma-Aldrich-Fluka (St. Louis, MO). Hyperoside (1) was isolated and characterized in our laboratory from the leaves of P. serotina (Ibarra-Alvarado et al., 2009). The purity and identity of 1 was confirmed by HPLC, spectroscopic, and spectrometric methods and was calculated as 99.1%.

Leaves of P. serotina (Ps-1) were purchased from the Sonora Market, Mexico City, in November 2010. A voucher specimen (Sandoval 15089) was deposited at the National Herbarium (MEXU), Instituto de Biologia, Universidad Nacional Autonoma de Mexico in Mexico City. In addition, five samples were purchased from various local markets (Ps-2-Ps-6) in Mexico City, in December 2010. All samples were authenticated by Dr. Estela Sandoval, Instituto de Biologia, UNAM.

Sample and stock solutions preparation

The crude drug of the species (292 g) were ground into powder (top size 2 mm) and extracted by maceration with dichloromethane–methanol (1:1, v/v) at room temperature. After filtration, the extract (EPS) was concentrated under reduced pressure to yield 30.4 g of a green residue. The infusion from the plant (IPS) were prepared by pouring 250 mL of boiling water on 10 g of the dried and grounded plant material; after 20 min, the aqueous extract was filtered. After filtration, 1 mL of the aqueous extract was passed through 0.45 µm nylon membrane filter, prior to HPLC analysis.

Stock standards solutions of hyperoside (1), chlorogenic acid (2), and benzaldehyde (3) were prepared in methanol at a final concentration of 1 mg mL⁻¹ for 1 and 2 and 0.5 mg mL⁻¹ for 3. The calibration curves for each standard solution were prepared by stepwise dilution from methanol stock solutions to obtain five concentration levels within the range of 100–1000 µg mL⁻¹ for 1 and 2 and 100–500 µg mL⁻¹ for 3.

Isolation and characterization of hyperoside (1)

EPS (200 g) was suspended in water and sequentially partitioned using dichloromethane and ethyl acetate. The ethyl acetate-soluble fraction (70 g) was subjected to open column chromatography on silica gel eluting with gradients of ethyl acetate–methanol (from 10 → 0 to 0 → 10) yielded nine secondary fractions (F₁–F₉). From fraction F_5, 230 mg of 1 (m.p. 235–236 °C) was crystallized spontaneously. The isolate was analyzed by IR, MS, and NMR and identified by comparing their spectral data with those of the authentic sample (Ibarra-Alvarado et al., 2009).

Method validation

The developed HPLC method was fully validated including the determination of precision, accuracy and linearity according to the International Conference on Harmonization guidelines (ICH, 2005). System suitability was assessed by six replicate analyses of the system suitability solution (0.5 mg mL⁻¹ for 1 and 2 and 0.3 mg mL⁻¹ for 3). The acceptance criterion was <2% for the percentage relative standard deviation (%RSD) of peak area and retention times. The resolution, capacity factor and tailing factor were also determined.

The linearity of the system, the least square line, and the correlation coefficient were calculated from calibration curves using the software Statgraphics Centurion XV (StatPoint Technologies Inc, 2009, Warrenton, VA) (ICH, 2005). Limits of detection (LOD) and quantification (LOQ) for 1–3 were determined at signal-to-noise ratios (S/N) of 3 and 10, respectively. The linearity of the method was tested by recovery, assaying independently three amounts equivalent to 50% (0.25 mg), 100% (0.5 mg) or 150% (1.0 mg) for 1 and 2 and 50% (0.15 mg), 100% (0.3 mg) or 150% (0.45 mg) for 3. At each level, 1–3 were added simultaneously to the infusion of P. serotina. Each sample was injected twice and analyzed according to the development method.

The repeatability and the inter-day intermediate precision were tested by six identical samples analyzed on two different days and by two different analysts by triplicate. The standard deviation (SD) and coefficient of variation (%RSD) were calculated for each day. Finally, selectivity was assessed analyzing all standards in several conditions including acid hydrolysis (1.0 M), base hydrolysis (1.0 M), and chemical oxidation (H₂O₂). Samples after 90% of degradation have been achieved. Briefly, to accomplish suitable resolution for 1–3 and its degradation products, some chemical variations in the composition of the mobile phase (acetate buffer pH 2.6 and methanol) were investigated; all mobile phases were filtered through a 0.45 µm membrane and degassed prior to use. Acceptable retention times for all eluting components were set within 30 min. For this endeavor, if the peak resolution (Rs) is greater than two then the run time is adequate. Rs is defined as follows: where W_a and W_b are the width of the two peaks at the baseline obtained from the chromatogram; whereas R_ta − R_tb are the retention times.

Application

The developed and validated method was applied to analyze the contents of the three studied components in six authentic plant material samples (Ps-1–Ps-6). All samples were powdered, extracted, and analyzed as describe in the “Materials and methods” section.

Results

Method validation

A typical chromatogram of 1–3 and P. serotina infusions are showed in Figure 1. The results for testing system suitability of the method are given in Table 1. System suitability was evaluated as the mean of the column efficiency parameter. RSD values (%) for retention times and peak area within 0.6% indicated a short variation of the measured values. The tailing factors (T) were, in all the cases, 1.1 (T < 2); resolution (R) was >2, and column efficiency, expressed by the number of theoretical plates (N), was more than 2100, indicating good separation of the applied HPLC system and analytical conditions.

Figure 1. Typical HPLC chromatogram of the infusion obtained from the leaves of P. serotina. (a) Hyperoside (1); (b) chlorogenic acid (2); (c) and (d) Ps-1 and Ps-4 samples, respectively.

Table 1. Validation parameters evaluated for quantifying 1–3 in P. serotina using the developed HPLC method.

	LC method
Parameter	1	2	3
r^2 (n = 6)^a	0.9986	0.9968	0.9985
Regression equation	25446x + 174439	11128x + 30447	64095x + 862067
Working range, µg mL^−1	100–1000	100–1000	100–500
LOD, µg mL^−1 (n = 3)^b	8.8	4.5	10.5
LOQ, µg mL^−1 (n = 3)^c	15.4	14.5	13.1
Average recovery, % (n = 3)	99.8–100.9	99.5–101.0	98.9–101.2
Recovery RSD, % (n = 3)^d	0.9	0.6	1.3
Reproducibility RSD, % (n = 3)	0.4	0.6	1.2
Theoretical plates	2197	2363	2693
Symmetry factor	1.3	1.3	1.8
Tailing factor	1.12	1.11	1.23
Resolution	2.9	3.9	2.6

^ar² = determination coefficient.

^bLOD = limit of determination, 3.3 × (SD of the response/slope of the calibration curve).

^cLOQ = limit of quantification, 10 × (SD of the response/slope of the calibration curve).

^dRSD = relative standard deviation.

Linearity

The linearity of the system was determined by using calibration curves. All calibration curves were plotted based on the linear regression analysis of peak area (y) versus concentration (x; µg/mL) of each standard solution at six different concentrations range between 0.1 and 1.0 mg mL⁻¹ for 1 and 2, and 0.1–0.5 mg mL⁻¹ for 3, and were founded to be linear (r² = 0.998, 0.997, and 0.998, respectively). The linear regression equations for 1–3 were: y = 25446x + 174439, y = 11128x + 30447, and y = 64095x + 862067, respectively. These results indicated that there was an excellent correlation between the peak area and concentration.

Accuracy

The accuracy of the method was assessed by the determination of recovery by means of the standard addition method assaying independently three levels in the range from 50 to 150% of the working concentration for 1–3. The linear regression equations for 1–3 were found to be y = 0.923 x + 9.25, y = 0.946 x – 12.55, and y = 0.9321x – 8.6374, respectively. The recovery ranges were expressed as the concentration detected as a percentage of the expected concentration and were 99.8–100.9, 99.5–101.0, and 98.9–101.2% for 1–3, respectively (Table 1), indicating that the proposed method is accurate for the analysis of the three standards in the infusion of P. serotina.

Reproducibility and repeatability

The reproducibility of the analytical method was evaluated in terms of the intermediate precision by performing three replicate analyses for the sample 1 (0.5 mg mL⁻¹), 2 (0.5 mg mL⁻¹), and 3 (0.3 mg mL⁻¹) by two different analysts on two consecutive days. The RSD values (%) were less than 0.4, 0.6, and 1.2% for 1–3, respectively. The repeatability was assessed by analyzing a set of six samples in triplicate for 1 (0.5 mg mL⁻¹), 2 (0.5 mg mL⁻¹), and 3 (0.3 mg mL⁻¹) on the same day. The coefficient of variation (CV) was less than 0.6% at each concentration level analyzed, confirming that the method was precise.

Limits of detection and quantification

Three different levels of the standard stock solution 0.01, 0.05, and 0.1 mg mL⁻¹ for 1 and 2 and 0.01, 0.03, and 0.05 mg mL⁻¹ for 3 were prepared in triplicate and calibration curves were constructed by plotting peak areas against concentration. The LOD and LOQ values were calculated from the equations LOD = 3.3 SD/s and LOQ = 10 SD/s, which represent the signal-to-noise ratio of 3 and 10, respectively. LOD values were 8.8, 4.5, and 10.5 µg mL⁻¹ for 1–3, respectively, whereas the LOQ values were 15.36, 14.47, and 13.09 µg mL⁻¹, respectively.

Selectivity

To assess selectivity, stock solutions for 1–3 were analyzed under conditions described for validation of the method. After degradation treatment, all samples were analyzed and the degradation products were observed at retention times of 3.58 and 16.25 for 1, 5.6 for 2 and 1.88 for 3. The results obtained showed that none of the peak generated by chemical treatment interfere with the peaks corresponding to 1–3, therefore the method showed the capability for the quantification of 1–3 in the presence of some interferences. Also, the degradation behavior of 1–3 generated important information about the best way to handle samples, which is quite useful in specific tests, such as biological investigation, in which pH variations are common.

Application of the method

Finally, the validated HPLC method was used for the simultaneous quantification of 1–3 from P. serotina infusions. The results are shown in Table 2 as the mean values from three replicate injections.

Table 2. Content of hyperoside (1), chlorogenic acid (2), and benzaldehyde (3) in P. serotina.

		Content (µg g^−1)^b
Sample^a	Source (year of harvest)	1	2	3
Ps-1	Market I (2011)	166.9 ± 5.25	500.5 ± 10.33	114.1 ± 6.94
Ps-2	Market II (2011)	162.1 ± 8.45	433.8 ± 13.64	106.4 ± 7.82
Ps-3	Market III (2011)	164.0 ± 2.58	516.5 ± 10.57	127.7 ± 5.55
Ps-4	Market IV (2010)	288.8 ± 12.6	503.0 ± 14.47	119.4 ± 9.92
Ps-5	Market V (2010)	201.6 ± 6.52	480.8 ± 15.67	105.6 ± 8.54
Ps-6	Market VI (2010)	130.8 ± 2.01	444.9 ± 19.03	136.1 ± 6.32

^aPs-1 to Ps-6 leaves of P. serotina.

^bThe value is the mean ± SD (n = 3)

Discussion

Optimization of suitable HPLC conditions

Due to the increased use of herbal medicine worldwide, the safety and quality of medicinal plants has become a major concern for health authorities and valuable sources for general analytical procedures are needed. In this context, we developed both a suitable identity and assay tests useful for quality control of the crude drug of P. serotina.

Prunus serotina possesses high amount of phenolic components, of this, hyperoside (1) and chlorogenic acid (2) are the major active principles. It is well-recognized that hyperoside (1) had protective effects to improve the heart function and it is widely used to relieve pain and also as an antioxidant agent. Therefore, 1 was selected as the active marker for quantitative analysis and LC method was chosen, as it is one of the key methods for estimating the content of marker principles in herbal preparation.

In order to develop a suitable HPLC method for the simultaneous quantification of hyperoside (1), chlorogenic acid (2), and benzaldehyde (3) in P. serotina, first, the appropriate optimal HPLC separation conditions were achieved to obtain chromatograms with an excellent resolution of adjacent peaks within a short analysis time. For this endeavor, several mobile and stationary phases were examined comprising mixtures of numerous buffer solutions (acetate, carbonate, phosphate) and organic solvents (acetonitrile, methanol), and flow rates (from 0.4–1.0 mL min⁻¹) were examined to achieve baseline separation for the major components in the infusion. The results obtained showed that an isocratic mobile phase prepared from a binary system eluting with acetate buffer solution pH 2.7 (A) and methanol (B) (65:35, v/v) on a reversed-phase Fortis-RP column provides the best separation for 1–3. Under the proposed analytical conditions, the determination of 1–3 was not subjected to interferences from other components in the complex matrix (Figure 1). The retention times (R_T) of 1–3 were found to be 4.1, 11.6, and 22.3 min, respectively. The proposed method was validated in terms of linearity, accuracy, precision, and repeatability.

Finally, the proposed method was applied to the quantification of 1–3 in six different samples of P. serotina. The results summarized in Table 2 indicated that amounts of 1–3 in the samples analyzed are uniform. This indicates that the quality of this drug could be invariable. Greater amount of chlorogenic acid (2) with a mean concentration of 479.9 ± 33.6 µg g⁻¹, dry matter, along with hyperoside (1) (185.7 ± 55.3 µg g⁻¹, dry matter) was present in the samples acquired at local markets. On the other hand, benzaldehyde (3) (118.2 ± 12.1 µg g⁻¹, dry matter) was found to be in lower concentration (Table 2).

Conclusions

Altogether, the results led to establish a simple, sensitive, precise, and reproducible HPLC method for the simultaneous quantification of hyperoside (1), chlorogenic acid (2), and benzaldehyde (3) in P. serotina. This is the first report on the quantification of 1–3 as active principles and compound 1 was selected as a marker of P. serotina that could be useful to guarantee the safety, efficacy, and stability of the crude drug and herbal products. The method was successfully used for the analysis of several samples acquired commercially in local markets.

Declaration of interest

The authors report no declaration of interest. This work was supported by grant of Dirección General de Asuntos del Personal Académico (DGAPA grant IA 200512).