Fast Separation of Capsaicinoids from Peppers by Reversed Phase Ultra-Performance Liquid Chromatography: Comparation with Traditional High-Performance Liquid Chromatography Methods

Abstract

A new chromatographic method for the separation of major capsaicinoids in peppers has been developed. Nordihydrocapsaicin, capsaicin, dihydrocapsaicin, homocapsaicin, and homodihydrocapsaicin have been separated by reversed-phase ultra-performance liquid chromatography. A gradient method has been developed using two solvents: 0.1% acetic acid in water and 0.1% acetic acid in methanol. The developed method allows the full separation of capsaicinoids in less than 3 min, with high reproducibility (relative standard deviation < 4.3%) and repeatability (relative standard deviation < 3.6%). Robustness regarding the total amount of methanol in the sample was determined. Comparison with previous reversed-phase high-performance liquid chromatography methods using both monolithic and conventional columns was also studied. Finally, the method was applied in the determination of major capsaicinoids in 16 hot pepper samples produced in Spain.

Keywords:

• Capsaicinoids
• Peppers
• Fast separations
• UHPLC
• HPLC

INTRODUCTION

Capsaicinoids are the pungent compounds responsible for the hot flavor of peppers. Among these compounds, there are two major capsaicinoids, i.e., capsaicin and dihydrocapsaicin (DHC), which represent around 90% of total capsaicinoids present in most hot varieties of peppers. Besides these two major capsaicinoids, other minor capsaicinoids have been identified in peppers: nordihydrocapsaicin (n-DHC) I and II, homocapsaicin (h-C) I and II, homodihydrocapsaicin (h-DHC) I and II, and nonivamide, among others, along with more than 20 others in several pepper varieties.^[1–4]

Capsaicinoids are widely used in food in most parts of the world due to their pungent properties.^[5,6] Capsaicinoids also have several properties and biological effects regarding human health. Chemopreventive and anticarcinogenic properties have been established.^[7] Antioxidant properties,^[8] participation in the regulation of fat metabolism^[9] and anti-inflammatory properties have also been demonstrated.^[10] The huge importance and wide use of these compounds in both food and medicine makes it of interest to develop rapid methods for the determination of these compounds.

Numerous methods have been applied in the determination of capsaicinoids in both peppers and biological fluids, and these include thin-layer chromatography (TLC),^[11] gas chromatography (GC),^[12,13] and high-performance liquid chromatography (HPLC).^[14,15] By far the most commonly used technique for the identification and quantification of these compounds is reversed-phase high-performance liquid chromatography (RP–HPLC). There are several separation methods that employ different equipment, columns, solvents, and elution modes.^[16–19] Both ultraviolet (UV)-visible and fluorescence detection systems have been used in the RP–HPLC methods.

HPLC using conventional and monolithic columns have previously been used in the determination of capsaicinoids from peppers. The total time of analysis has been reduced using monolithic columns for the 5 major capsaicinoids present in peppers with a good resolution obtained in a time of 7.2 min.^[14] However, this kind of method consumes large amounts of solvents because of the high flow rates (3.5–6 mL min^–1). This article concerns the application of ultra-performance liquid chromatography (UHPLC) in the development of a method for the separation and quantification of the main capsaicinoids, including not only capsaicin but n-DHC, DHC, h-C, and h-DHC, in a shorter time and also using lower amounts of solvents.

The UHPLC method has been developed.^[20] Some UHPLC methods for the determination of capsaicinoids have already been published; however, in some of them, only capsaicin and DHC were determined,^[21] or they separate the major capsaicinoids in a large time of analysis.^[22] In the current article, we have separated the major capsaicinoids in a time less than 3 min.

MATERIALS AND METHODS

Reagents

The reference standards for capsaicinoids, capsaicin (97%) and DHC (90%), were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). The water was obtained from a Milli-Q water deionization system (Millipore, Bedford, MA, USA). Methanol and acetic acid were HPLC grade and were obtained from Merck (Darmstadt, Germany).

Plant Material

An extract from Hot Cayenne Pepper (Capsicum annuum) was used as the sample to develop the separation method for capsaicinoids. To obtain the extract, fresh peppers were peeled and the stems and seeds were separated from the rest. Only the pericarp and the placenta of the pepper were used in this study. Both the pericarp and placenta were ground using a conventional blender until a homogeneous sample was obtained for analysis. The crushed sample was kept in the freezer at –20°C until analysis. The final method was applied in the separation and quantification of the five major capsaicinoids present in 16 varieties of hot peppers cultivated in Spain, of which nine are of the species Capsicum annuum, five of the species Capsicum frutescens, and two of the species Capsicum baccatum.

Extraction Procedures

Microwave-assisted extraction (MAE) was used to obtain the samples. A previously developed extraction method was applied.^[23] The MAE process was performed in an Ethos 1600 microwave oven (Millestone, Shelton, CT, USA). All extracts were filtered through a 0.22 μm nylon syringe filter (Millex-HN, Ireland) before chromatographic analysis.

HPLC Separation

HPLC analyses were carried out on a Dionex system consisting of an autosampler (ASI 100), pump (P680), chromatographic oven (TCC-100), and a photodiode array detector (PAD100). UV absorbance was monitored from 200 to 400 nm. The software for the control of the equipment and data acquisition was Chromeleon version 6.60.

A Chromolith TH Performance RP-18e column (4.6 mm, 100 mm, Merck) was used for the HPLC method with the monolithic column method. The gradient elution used two solvents: acidified water (0.1% acetic acid, solvent (A) and acidified methanol (0.1% acetic acid, solvent (B), working at a flow rate of 6 mL min^–1. The gradient method utilized was the following: 0 min, 10% B; 2 min, 50% B; 4 min, 50% B; 4.5 min, 55% B; 5.5 min, 55% B; 6 min, 60% B; 7 min, 60% B; 9 min, 70% B; 10 min, 100% B; 15 min, 100% B. The temperature of the column was held constant at 30°C. The injection volume was 25 µL.

A Luna (150 mm × 30 mm, 5 μm) column (Phenomenex) was used for the HPLC with the conventional column method. A gradient method, using acidified water (0.1% acetic acid, solvent A) and acidified methanol (0.1% acetic acid, solvent B), working at a flow rate of 0.4 mL min^–1, was utilized for the chromatographic separation. The gradient employed was the following:  0 min, 0% B; 1 min, 0% B; 5 min, 30% B; 8 min, 50% B; 16 min, 70% B; 20 min, 70% B; 28 min, 90% B; 30 min, 90% B; 32 min, 100% B; 42 min, 100% B. The injection volume was 25 µL.

UHPLC Separation

The UHPLC analyses were carried out on an Acquity UPLC System (Waters, Milford, MA, USA) equipped with a Model 2996 PDA detector. The wavelength used was 280 nm for quantification of capsaicinoids. Empower 2 software (Waters) was used to control the equipment and for data acquisition. Capsaicinoids were analyzed on a Waters BEH C18 column (50 × 2.1 mm I.D., particle size 1.7 µm). The injection volume was 1.5 µL. The final gradient elution used acidified water (0.1% acetic acid, solvent A) and acidified methanol (0.1% acetic acid, solvent B), working at a flow rate of 0.8 mL min^–1. The gradient was as it follows (min, 0% solvent B): 0.85 min, 55%; 1.60 min, 55%; 1.95 min, 60%; 2.45 min, 63%; 2.80 min, 70%; 3.00 min, 70%; 4.00 min, 100%.

Identification of Capsaicinoids by Liquid Chromatography Coupled to Mass Spectrometry (MS)

Five major capsaicinoids found in peppers were identified using an HPLC–MS system. The following m/z signals were used for their identifications; Nordyhydrocapsaicin: M+1: 294; Capsaicin: M+1: 306; Dyhydrocapsaicin: M+1: 308; h-C: M+1: 320; h-DHC: M+1: 322. The HPLC–MS analyses of hot pepper extracts were carried out on a Finnigan LCQ-coupled LC-MS system (Thermo Electron Co., San José, CA). This equipment was fitted with a Spectra SYSTEM 2000 model gradient pump (Thermo Separation Products, Fremont, USA) and a mass detector (model LCQ), consisting of an electrospray interface and an ion trap mass analyser. Xcalibur version 1.2 software was used to control the equipment and for the acquisition and treatment of data. The sample injection volume was 25 µL. Interface conditions: positive ionization, temperature of the capillary: 220ºC, spray voltage: 20 kV, capillary voltage: –5 V, focus gas flow: 80 (arbitrary units) and auxiliary gas flow: 10 (arbitrary units). Atmospheric pressure ionization-mass spectrometry (API-MS) data were acquired in the m/z range of 50–400.

A gradient method, using acidified water (0.1% acetic acid, solvent A) and acidified methanol (0.1% acetic acid, solvent B), working at a flow rate of 0.2 mL min^–1, was employed for the chromatographic separation. The gradient employed was as follows: 0 min, 0% B; 1 min, 0% B; 5 min, 30% B; 8 min, 50% B; 16 min, 70% B; 20 min, 70% B; 28 min, 90% B; 30 min, 90% B; 32 min, 100% B; 42 min, 100% B. A C-18 column (Luna 5 μm, 150 × 3 mm, Phenomenex) was used to achieve the separation. The capsaicinoids identified in the extracts of peppers analyzed were n-DHC, capsaicin, DHC, h-C, and h-DHC.

UHPLC Calibration

The UHPLC method was used to prepare calibration curves for capsaicin and DHC (y = 866.59x – 1690.4 for capsaicin and y = 821.77x – 2350.2 for DHC), which are the two commercially available capsaicinoid standards. Regression equations and the correlation coefficient (R²; 0.9998 for capsaicin and 0.9997 for DHC), limits of detection (0.556 mg L^–1 for capsaicin and 0.896 mg L^–1 for DHC) and quantification (1.852 mg L^–1 for capsaicin and 2.988 mg L^–1 for DHC) were calculated using the ALAMIN software package.^[24] All analyses were carried out in triplicate.

Quantification of the Capsaicinoids

The five major capsaicinoids (n-DHC, capsaicin, DHC, h-C, and h-DHC) present in the hot chili peppers analyzed were quantified using this method. Capsaicin and DHC were quantified from the calibration curves obtained from the standard solutions. Since there are no commercial standards for n-DHC, h-C, and h-DHC, these compounds were quantified from the calibration curve of DHC (for n-DHC and for h-DHC) and from the calibration curve of capsaicin (for h-C), given the structural similarities between these molecules and taking into account their molecular weights. All analyses were run in triplicate.

RESULTS AND DISCUSSION

UHPLC Gradient Method Optimization

The previous HPLC method using monolithic columns^[14] was used as the starting point for the new UHPLC method. In this way a gradient method was developed that uses water as solvent with 0.1% acetic acid (solvent A) and methanol with 0.1% acetic acid (solvent B). The optimum conditions of separation were determined with a column temperature of 50°C and a solvent flow of 0.8 mL min^–1. The gradient method developed for these working conditions were as follows (min, % solvent B): 0 min, 0%; 0.85 min, 55%; 1.60 min, 55%; 1.95 min, 60%; 2.45 min, 63%; 2.80 min, 70%; 3.00 min, 70%; 4.00 min, 100%.

The previously mentioned gradient method was used to obtain full separation (Fig. 1) for the five major capsaicinoids. The resulting chromatogram showed good resolution and an analysis time that was eight times shorter than that obtained on using a conventional C-18 column by RP–HPLC and less than half the time taken when using a monolithic column. The retention times for the five capsaicinoids studied using the method developed by reversed-phase UHPLC are shown in Table 1

FIGURE 1 UHPLC Chromatogram of hot pepper extract [1-nordihydrocapsaicin (n-DHC); 2-capsaicin (C); 3-dihydrocapsaicin (DHC); 4-homocapsaicin (h-C); 5-homodihydrocapsaicin (h-DHC)]. Absorbance detection at 280 nm.

TABLE 1 Retention times in minutes (n = 3) of the chromatographic peaks corresponding to the capsaicinoids studied with each chromatographic separation method [reversed-phase HPLC using a C-18 convenctional column (rp-HPLC cc), reversed-phase HPLC using a C-18 monolithic column (rp-HPLC-mc) and reversed-phase UHPLC (rp-UHPLC)]

	n-DHC	C	DHC	h-C	h-DHC
rp-HPLC cc	14.03 ± 0.01	14.84 ± 0.01	19.42 ± 0.01	20.35 ± 0.01	24.67 ± 0.01
rp-HPLC mc	4.29 ± 0.05	4.61 ± 0,05	5.78 ± 0.04	6.12 ± 0.03	7.21 ± 0.01
rp-UHPLC	1.97 ± 0.01	2.06 ± 0.01	2.48 ± 0.01	2.58 ± 0.01	2.97 ± 0.01

Repeatability and Reproducibility of the Method

The repeatability and reproducibility of the method developed was studied in respect of the peak area and the retention time of each capsaicinoid. A total of 48 UHPLC analyses were run on the same sample, an extract of hot cayenne pepper, on three successive days (16 analyses per day). The relative standard deviations intraday and interday in respect of the retention time are shown in Table 2. Regarding retention time, the resulting relative standard deviation (RSD) values were less than 0.06% for repeatability and less than 0.46% for reproducibility. The relative standard deviations intraday and interday in respect of the peak area are shown in Table 2. The resulting RSD values were less than 3.6% for repeatability and less than 4.33% for reproducibility.

TABLE 2 Intraday (%) and interday (%) relative standard deviations in respect of the retention time and peak area

		n-DHC	C	DHC	h-C	h-DHC
Retention time	RSD Intraday (%)	0.05	0.05	0.05	0.06	0.03
	RSD Interday (%)	0.46	0.43	0.30	0.46	0.32
Peak area	RSD Intraday (%)	2.07	1.66	1.69	2.46	3.60
	RSD Interday (%)	2.22	1.97	2.23	3.20	4.33

Robustness of the Method Versus Sample Solvent

Most samples to be analyzed using the developed method are expected to include some amount of organic solvent, as they were obtained by extraction methods that employ organic solvents, mainly methanol. Additionally, the total percentage of organic solvent can have different values, because different mixtures with water are usually applied in the extraction methods.^[23] Furthermore, in several methods, a final concentration step is included to increase the analytical signal in the detection systems, thus changing the initial solvent mixtures. As a result, the robustness of the chromatographic method related to the sample composition should be checked. Different percentages of methanol in water (25, 50, and 75%) and also pure methanol were studied as sample solvents.

The robustness for chromatographic resolutions, peak area and also peak retention time for capsaicinoids was established. This required an analysis of the variance of a factor for each of the parameters to be studied. The volume of extract injected was 1.5 µL. Each analysis was performed in triplicate.

Regarding peak resolution, the developed method showed full robustness on changing the sample solvent composition between 25% methanol in water to 100% methanol. Significant differences were not found for the resulting chromatographic peak resolutions (p < 0.05).

The developed method also showed full robustness regarding retention times on changing the sample solvent composition from 25% up to 75% methanol in water, with the only exception being h-C. Significant differences were found when the solvent in the sample was 100% methanol for all assayed capsaicinoids. The data are shown in Table 3. Therefore, the developed method shows full robustness for the main chromatographic properties for samples having between 25 and 75% methanol in water as solvent. As a consequence, this method can be applied to this kind of sample regardless of the percentage of methanol in water.

TABLE 3 Analysis of the variance of a factor for the peak time and peak area, with different percentages of methanol (25–100%) in the extracts

	% MeOH	n-DHC	C	DHC	h-C	H-DHC
Retention time	25	2.059^a,b	2.145^a,b	2.558^a,b	2.659^a,c	3.021^a
	50	2.046^a	2.132^a	2.543^a	2.644^a	3.010^a
	75	2.038^a,b	2.124^a,b	2.534^a,b	2.634^b,c	3.004^a,b
	100	2.034^b	2.119^b	2.529^b	2.629^c	3.000^b
Peak area	25	14783^a	53259^a,b	43488^a,b	4816^a	3753^a
	50	14655^a	54814^a	45027^a	5081^a	3929^a
	75	14980^a	55146^a	45237^a	4894^a	3680^a
	100	14258^a	52719^b	43279^b	4817^a	3561^a

Means with the same superscripts (a, b, c) in the same column are not statistically different (p < 0.05).

Comparison with Previous HPLC Methods

The developed method was applied to the same samples as the previously employed HPLC methods, using both a conventional column^[25] and a monolithic column^[14] in order to compare these approaches. The results are shown in Table 4. The new UHPLC method provides a dramatic reduction in the analysis time for major capsaicinoids in peppers. An analysis time of only 3 min is needed on using UHPLC instead of the 25 min on using HPLC with a conventional column. The total time required on using HPLC with a monolithic column (7.2 min) was also reduced markedly to 41% of the total time. This reduction in analysis time was achieved without losing chromatographic peak resolution. Regarding costs, the new UHPLC method also represents a very interesting advance in comparison to the monolithic column-based method, because the flow rate was only 0.8 mL min^–1 for the UHPLC method while the flow rate for the previous method was 6 mL min^–1.

TABLE 4 Resolution of the chromatographic peaks corresponding to the capsaicinoids studied with each chromatographic separation method [reversed-phase HPLC using a C-18 conventional column (rp-HPLC cc), reversed-phase HPLC using a C-18 monolithic column (rp-HPLC-mc) and reversed-phase UHPLC (rp-UHPLC)]. n.a. = not available, peaks partially overlapping

	n-DHC/C	C/DHC	DHC/h-C	h-C/h-DHC
rp-HPLC cc	1.56 ± 0.02	8.47 ± 0.08	n.a.	n.a.
rp-HPLC mc	1.77 ± 0.04	6.62 ± 0.09	1.86 ± 0.03	5.90 ± 0.10
rp-UHPLC	0.71 ± 0.05	3.02 ± 0.25	0.86 ± 0.06	4.18 ± 0.16

Quantification of the Capsaicinoids Present in Different Varieties of Hot Chili Peppers Cultivated in Spain

The UHPLC method was applied in the analysis of 16 different pepper samples in order to determine five major capsaicinoids (Table 5). The capsaicinoid distribution in real samples shows that capsaicin and DHC were the major compounds in all analyzed samples. In most cases capsaicin shows higher values than DHC, with only four samples (Samples 10, 11, 12, and 14) from C. frutescens variety showing different behavior. The varieties with the highest levels of capsaicinoids also belong to the C. frutescens variety, although this information does not allow the characterization of different varieties.

TABLE 5 Quantity (µmol) of capsaicinoid per kilogram of fresh pepper in the samples analysed (n = 3)

Species	Sample	n-DHC	C	DHC	h-C	h-DHC	Total
C. annuum	1	141.19 ± 3.13^a	579.03 ± 11.41^a	528.65 ± 11.79^a	22.47 ± 0.72^a	43.43 ± 1.88^a	1314.77 ± 28.93^a
	2	150.61 ± 3.34^a	593.23 ± 11.69^a	56.17 ± 1.25^b	15.99 ± 0.51^a	37.80 ± 1.64^a	853.8 ± 18.43^b
	3	34.91 ± 0.78^b	190.30 ± 3.75^b	140.75 ± 3.14^c	7.65 ± 0.24^b	14.98 ± 0.65^b	388.59 ± 8.56^c
	4	114.10 ± 2.53^c	554.38 ± 10.92^a	440.74 ± 9.83^a	9.67 ± 0.31^b	37.39 ± 1.62^a	1156.28 ± 25.21^a
	5	53.56 ± 1.19^d	278.34 ± 5.48^c	262.87 ± 5.86	15.43 ± 0.49^a	20.28 ± 0.88^b	630.48 ± 13.91^d
	6	13.34 ± 0.30 ^e	60.08 ± 1.18^d	52.64 ± 1.17^b	n.d.	3.52 ± 0.15^c	129.58 ± 2.81^e
	7	30.57 ± 0.68^b	130.20 ± 2.56^e	88.71 ± 1.98^d	5.36 ± 0.17^b	13.22 ± 0.57^b	268.06 ± 5.97^f
	8	238.49 ± 5.29	1539.10 ± 30.32^f	1048.8 ± 23.39^e	66.67 ± 2.13	82.76 ± 3.58^d	2975.82 ± 64.72^g
	9	55.15 ± 1.22^d	282.54 ± 5.57^c	195.41 ± 4.36^f	9.87 ± 0.32^b	19.77 ± 0.86^b	562.74 ± 12.32^h
C. frutescens	10	329.98 ± 7.33^f	974.61 ± 19.20^g	1070.74 ± 23.88^e	34.45 ± 1.10^c	52.37 ± 2.27^a	2462.15 ± 53.77^i
	11	271.74 ± 6.03^f	800.93 ± 15.78^g,h	857.83 ± 19.13^g	78.20 ± 2.50^d	95.60 ± 4.14^d	2104.30 ± 47.58^i
	12	143.10 ± 3.18^a	1956.81 ± 38.55^i	1394.18 ± 31.09^h	64.48 ± 2.06^d	50.99 ± 2.21^a	3609.56 ± 77.09^j
	13	198.16 ± 4.40^g	706.09 ± 13.91^h	779.58 ± 17.38^g	47.96 ± 1.53^e	56.93 ± 2.47^a	1788.72 ± 39.69^i
	14	451.46 ± 10.02^h	1151.76 ± 22.69	1404.00 ± 31.31^h	71.28 ± 2.28^d	156.06 ± 6.76	3234.56 ± 73.06^j
C. baccatum	15	24.36 ± 0.54^e	158.88 ± 3.13^b,e	95.39 ± 2.13^d	10.80 ± 0.35^b	10.94 ± 0.47^e	300.37 ± 6.62^f
	16	54.00 ± 1.20^d	198.64 ± 3.91^b,c	196.23 ± 4.38^f	12.50 ± 0.40^b	20.61 ± 0.89^b	481.98 ± 10.78^c

Means with the same superscripts (a–j) in the same column are not statistically different (p < 0.05). Samples 1–2: Bola Pepper varieties; Samples 3–5: Jeromin Pepper varieties; Samples 6–9: Jariza Pepper varieties; Samples 10–14: Malagueta Pepper varieties; Samples 15–16: Bishop Crown varieties.

CONCLUSIONS

The UHPLC method developed for the separation and quantification of the five major capsaicinoids present in peppers leads to dramatically shorter analysis times when compared to the HPLC method with a monolithic column. Additionally, similar chromatographic peak resolution was obtained. The time required for the proper separation of the five major capsaicinoids using the RP-HPLC column is reduced by eight times compared to the conventional C-18 method. The developed method shows good repeatability (<3.6%) and reproducibility (<4.4%). The robustness of the method in terms of changing the percentage of methanol in the sample is high with respect to the resolution of the peaks for all percentages studied. This is also a robust method with respect to peak area and time of separation in the range of 25 to 75% methanol. These findings enable this method to be used on several types of extracts.

FUNDING

This work forms part of the RTA2011-00118 project funded by the National Institute for Agriculture and Food Research and Technology (INIA) and co-financed by the European Fund for Regional Development (FEDER).

ORCID

Gerardo Barbero

http://orcid.org/0000-0001-7302-6605

Marta Ferreiro-González

http://orcid.org/0000-0003-0870-403X

Miguel Palma

http://orcid.org/0000-0001-8509-4226

Carmelo Barroso

http://orcid.org/0000-0003-2284-2244

Additional information

Funding

This work forms part of the RTA2011-00118 project funded by the National Institute for Agriculture and Food Research and Technology (INIA) and co-financed by the European Fund for Regional Development (FEDER).