Skip to main content
NCBI home page
Journal of Food Science and Technology logoLink to Journal of Food Science and Technology
. 2017 Jun 6;54(9):2814–2824. doi: 10.1007/s13197-017-2719-0

Comparative analysis of carotenoid content in Momordica cochinchinensis (Cucurbitaceae) collected from Australia, Thailand and Vietnam

Dilani Wimalasiri 1, Robert Brkljača 3, Terrence J Piva 2, Sylvia Urban 3, Tien Huynh 1,✉
PMCID: PMC5583111  PMID: 28928521

Abstract

Momordica cochinchinensis (Cucurbitaceae) is the richest source of lycopene and β-carotene of all known fruits but the influences of collection sites, variety and environment on carotenoid accumulation is unknown. This study analysed the carotenoid content of 44 M. cochinchinensis aril samples collected from Australia, Thailand and Vietnam using HPLC, UV–visible spectrophotometry and compared with the colorimetry method. The highest lycopene content was observed in samples collected from Ha Noi (7.76 mg/g) of Northern Vietnam and Lam Ha (6.45 mg/g) and Lam Dong (6.64 mg/g) provinces of Central Vietnam. The highest β-carotene content was observed in a sample from Nam Dinh (9.60 mg/g) in Northern Vietnam while a variety from Hoa Binh province in Northern Vietnam had high contents of both lycopene (5.17 mg/g) and β-carotene (5.66 mg/g). Lycopene content was higher in samples collected from low temperatures (<14 °C) and higher elevations whilst β-carotene content was greatest at temperatures between 27 and 33 °C. Crop improvement for increased lycopene and β-carotene requires rapid and accurate methods of quantification. All three analytical methods utilised were in agreement for lycopene quantification. The (a*/b*)2 transformed colour value resulted in more linear relationship for lycopene indicating that colorimetry method could potentially be developed to select lycopene rich fruits in the field.

Electronic supplementary material

The online version of this article (doi:10.1007/s13197-017-2719-0) contains supplementary material, which is available to authorized users.

Keywords: Momordica cochinchinensis, Collection sites, β-carotene, High performance liquid chromatography (HPLC), Lycopene

Introduction

Carotenoids are fat soluble pigments synthesized in plants, photosynthetic bacteria and fungi through the mevalonic acid pathway using an isopentenyl pyrophosphate (IPP) precursor (Rodriguez-Amaya and Kimura 2004). The main carotenoids are lycopene and β-carotene, which are the compounds responsible for the red and yellow-orange colour, respectively. Both lycopene and β-carotene have numerous bio-medicinal and nutraceutical benefits (Rao and Rao 2007). Both pigments are highly effective antioxidants due to their conjugated double bond system and free radical scavenging activity. Amongst carotenoids, lycopene has the highest singlet oxygen-quenching rate (Perkins-Veazie et al. 2001) and is responsible for the reduction of cancers, cardiovascular diseases and macular dysfunction (Rao and Rao 2007). Lycopene is the precursor of most cyclic carotenoids including β-carotene, which converts to produce vitamin A, important for the development of cell membranes, immunity and skin health (Hammond and Renzi 2013; Rodriguez-Amaya and Kimura 2004).

Momordica cochinchinensis based products have potential in the nutraceutical and functional food industries due the high carotenoid concentration in this fruit (Supplementary Fig. 1). Currently, the commonly recognised food source of lycopene is tomatoes and for β-carotene it is carrots. M. cochinchinensis have the highest content of dietary β-carotene and lycopene, with contents 10–70 fold more than all known fruits and vegetables (Aoki et al. 2002) including tomatoes (8.8–42.0 µg/g FW lycopene) (Rao and Rao 2007) and carrots (47–11,210 µg/g FW β-carotene) (Gul et al. 2015). Despite its superior carotenoid content and phytochemical potential (Chuyen et al. 2015), M. cochinchinensis is genetically diverse and geographically restricted (Bootprom et al. 2012; Wimalasiri et al. 2016); it is not widely distributed outside of South East Asia (Aoki et al. 2002). There is limited information on varietal differences and the downstream effects on carotenoid accumulation, which hinder the selection of breeding stock.

Carotenogenesis in other economically important fruits, including apricots (Dragovic-Uzelac et al. 2007), carrots (Leja et al. 2013) and tomatoes, has been shown to be influenced by geographical and environmental factors, including elevation (Rodriguez-Amaya and Kimura 2004), rainfall (Brandt et al. 2003) and temperature (Leoni 1992; Dumas et al. 2003), with carotenoid synthesis impacted more by geographical location than varietal differences (Aherne et al. 2009) or vice versa (Leja et al. 2013). M. cochinchinensis has a diverse distribution in South East Asia from tropical to temperate regions with varying eco-geographical profiles (Wimalasiri et al. 2016), which could account for the nutritional variations previously reported (Chuyen et al. 2015). Cultivar selection and optimal growth conditions are essential for producing consistently higher quantities of carotenoids but in the case of M. cochinchinensis these parameters are unknown.

Development of cultivars with higher carotenoid content using traditional breeding approaches requires rapid screening methods. The carotenoids in M. cochinchinensis are commonly quantified using High Performance Liquid Chromatography (HPLC) (Ishida et al. 2004; Kubola and Siriamornpun 2011; Vuong et al. 2006) or spectrophotometric methods using fruit extracts. Although these methods allow accurate quantification of individual carotenoids and separation of isomers; it is very laborious and this standardised methodology is not available to primary producers in developing nations where M. cochinchinensis is endemic. In contrast, colorimetry methods (Barba et al. 2006) could be used for rapid screening of carotenoid rich fruits since they are simple, cost effective and routinely used in the quantification of lycopene and β-carotene in other common fruits and vegetables (Biehler et al. 2010; Fish 2012) but not yet known for M. cochinchinensis.

The aims of this study were firstly, to elucidate the variation of lycopene and β-carotene concentration of M. cochinchinensis collected from Thailand, Vietnam and Australia. Secondly, to determine whether a colorimetry method could yield estimates of lycopene and β-carotene with comparable accuracy as methods accruing chemical extractions such as HPLC and spectrophotometry. The findings from this study will assist the breeding selection of cultivars with higher carotenoid levels and optimum growth environments to ensure consistent higher carotenoid contents. Furthermore, results from this study will also assist in the selection of carotenoid-rich fruits in resource-poor or technologically challenged situations.

Materials and methods

Sample collection

Mature fruits of 36 M. cochinchinensis samples belonging to 17 provinces of Thailand and Vietnam were collected (Table 1). All the fruits collected were at the same maturity stages of ripeness with orange or red coloured skin with soft texture and were collected by Dr Tien Huynh from their natural habitats, as reported by locals, during harvesting season (December–February). Eight fruits from a cultivated variety from Australia (A5-12) was provided by the Department of Primary Industries; NSW and used for comparison (Table 1). After collection, all the fruits were cleaned; the aril was separated from the seeds, transported to RMIT University in an insulated bag, separated into aliquots, covered with aluminium foil and stored at −20 °C in darkness until required. The geographical (altitude, latitude and longitude) and ecological (rainfall of wettest and driest month, observed minimum and maximum temperature, annual temperature range) data were obtained for each province using the DIVA-GIS spacial analysis software (Hijmans et al. 2004) (Table 1).

Table 1.

Geographical and environmental climate of M. cochinchinensis from Vietnam (southern, northern and central), Thailand and Australia (Wimalasiri et al. 2016). Bioclimatic data was obtained from DIVA-GIS spacial analysis software (Hijmans et al. 2004)

Country Province Sample Latitude Longitude Elevation (m) Annual temperature °C Precipitation (mm) Climatea
Max Min Difference Annual Wettest month Driest month
S Vietnam Can Tho VS1 10.034320 105.775130 1 32.6 22.2 10.4 1524 254 2 Tropical
Can Tho VS2 9.996850 105.665870 3 32.4 22.0 10.4 1623 269 3 Tropical
Vinh Long VS3a, 3b 10.248758 105.969566 2 33.2 21.9 11.3 1413 244 2 Tropical
Tra On VS4 9.965880 105.925140 3 32.5 22.1 10.4 1520 259 2 Tropical
HCM City VS7 10.757410 106.673439 6 34.7 21.1 13.6 1851 310 2 Tropical
N Vietnam Hung Yen VN8 20.814159 106.125992 4 33.1 14.4 19.0 1679 316 15 Temperate
Hung Yen VN9a, 9b 21.648399 105.028999 71 32.2 11.6 20.6 1774 325 18 Temperate
Hai Duong VN10 20.897711 106.423561 3 32.8 14.2 18.6 1648 311 17 Temperate
Hai Duong VN11 20.919382 106.419062 3 33.0 13.5 19.5 1692 312 14 Temperate
PhuTho VN12 21.331640 104.953453 296 30.8 10.5 20.3 1574 300 10 Temperate
PhuTho VN13-15 21.445990 105.094419 296 32.3 11.3 20.9 1734 315 19 Temperate
Ha Noi VN16-20 20.866699 105.750000 9 33.0 13.5 19.5 1700 314 14 Temperate
Ha Noi VN21 20.985260 105.858501 9 33.1 13.5 19.6 1673 324 15 Temperate
Hoa Binh VN22 20.829601 105.332893 195 31.6 11.7 19.9 1801 352 8 Temperate
Nam Dinh VN24-27 20.422930 106.173630 4 33.4 14.7 18.7 1663 336 27 Temperate
C Vietnam Lam Dong VC28, 29 11.835772 108.336275 1112 27.9 13.3 14.6 1651 248 7 Temperate
Lam Ha VC32 11.793576 108.268610 925 28.1 13.8 14.3 1658 252 8 Temperate
Thailand Nakhon Pathom T1, T3, T4 13.828790 100.039620 8 35.9 18.7 17.2 1237 257 6 Tropical
Samut Prakan T5 13.576110 100.665420 2 34.5 20.4 14.1 1439 325 10 Tropical
Chiang Mai T6, T8 18.767000 99.026161 316 36.2 12.9 23.3 1163 244 7 Temperate
Chanthaburi T7 12.878721 102.129135 79 32.0 17.4 14.6 1751 316 10 Tropical
Australiab Newcastle A5–A12 33.212281 151.225069 NA 25.0 18.0 NA NA NA NA Tropical
Open in a new tab

aRegions with a minimum temperature <15 °C in coldest month was selected as temperate climate

bPlants grown in greenhouse condition with an average temperature of 25 °C

Extraction of carotenoids

Carotenoids of the M. cochinchinensis aril were extracted as previously described (Barba et al. 2006). The aril (2 g) was placed in a vessel protected from sunlight and mixed with 100 mL of extraction solvent (hexane/acetone/ethanol: 50:25:25 v/v/v) (Barba et al. 2006). The mixture was ultra-sonicated (Unisonics, Australia) for 30 min and then 15 mL of distilled water was added to enhance phase separation, where by the upper hexane layer contained the carotenoids and the bottom water layer contained hydrophilic compounds and cell debris. 1 mL of the upper organic layer was used for UV–visible spectroscopic analysis; 10 mL of the remaining hexane extract was evaporated to dryness using a rotary evaporator (Büchi Labortechnik AG, Australia). The residue was dissolved in THF/acetonitrile/methanol: 50:25:25 v/v/v to a final volume of 4 mL for HPLC analysis. The final extract was filtered through a 0.45 µm PTFE membrane filter (Labquip, Australia) and a 20 µL injection was employed for HPLC analysis.

Analyses of carotenoids

High performance liquid chromatography (HPLC)

Carotenoid extracts were analysed via reverse phase analytical HPLC using solvents and conditions described previously (Barba et al. 2006). Samples were analysed using an isocratic method (90% CH3CN/H2O) on an Alltech Alltima HP C18 (250 × 4.6) 5 μm column at a flow rate of 1.0 mL/min. Analytical HPLC analyses were performed on a Dionex P680 solvent delivery system equipped with a PDA100 UV detector (operated using “Chromeleon” software). The column temperature was 30 °C and UV detection was monitored at 465 nm (provided highest peak areas for both lycopene and β-carotene). The lycopene and β-carotene contents was quantified with reference to samples of commercial standards of lycopene and β-carotene (Sigma Chemical, St. Louis, USA) of known concentrations ranging from 10 to 400 µg/mL which were used to obtain a calibration plot (Supplementary Fig. 2). The regression equation was used to determine the lycopene and β-carotene content of the samples where, y = 1.16x − 4.369, R2 = 0.971 for lycopene and y = 0.540x + 0.458, R2 = 0.984 for β-carotene. The analyses were performed in triplicate for all the samples and the results were presented as fresh weight basis (FW).

UV–visible spectrophotometry

Carotenoid extracts of M. cochinchinensis were placed in a 1 mL quartz cuvette and analysed using UV–visible spectrophotometry in a Cary® 50 UV–Vis spectrophotometer (Varian Inc, USA). UV visible spectra were recorded between 200 and 800 nm in hexane using a 1 nm interval.

The carotenoid content of the samples was calculated at the absorption maxima of lycopene (502 nm) and β-carotene (450 nm) using the equation: C = (Aλ × Fd)/ε (Biehler et al. 2010), where C was the unknown content of carotenoids in mol/L; Aλ was the absorbance maximum of lycopene and β-carotene; F was a dilution factor adjusting for extraction, drying and reconstitution processes; ε was the molar extinction coefficient in hexane (ε (L/mol) = 169,000 for lycopene and 139,000 for β-carotene) (Zechmeister et al. 1943); and d was the width of the cuvette (1 cm). Using the molar mass of lycopene (537 g/mol) and β-carotene (537 g/mol), the results were converted to mg/g fresh weight. The spectroscopic analyses were performed in triplicate for all M. cochinchinensis samples.

Colorimetry

The colour of the frozen aril samples was measured in triplicate using a tristimulus colorimeter (Konico Minolta, Chroma meter—CR 400, Japan) with a 8 mm measuring area and diffuse illumination/0° viewing angle. The data were recorded as L*, a*, b*, corresponding to indices of lightness, redness and yellowness, respectively, according to the CIE L*a*b* scale (Hunter 1987). The L* coordinate indicates darkness or lightness of colour and ranged from dark (0) to light (1). The a* and b* coordinates indicate the colour directions where a* > 1 is the red direction and a* < 0 is the green direction; b* > 1 is the yellow direction and b* < 0 is the blue direction (Itle and Kabelka 2009). Chroma indicates the saturation of colour, whereas as the value of chroma increases, the colour becomes more intense and as it decreases, the colour becomes duller. The hue was used to interpret the actual colour, for example, 0° = pure red, 45° = orange, 90° = yellow, 180° = pure green (Arias et al. 2000; Itle and Kabelka 2009). Both chroma and hue were derived from a* and b* values (Arias et al. 2000) where, chroma = (a* 2 + b* 2)1/2 and hue = tan−1(b*/a*). The contents of lycopene and β-carotene measured by HPLC and UV–visible spectrophotometry were compared graphically and statistically against colour values (L*, a*, b*, chroma and hue) for the linear relationship using Minitab 17 software. For relationships that appeared to be non-linear, colour values were transformed (e.g. a*/b*, (a*/b*) 2, a*4) to improve the linear relationships with the lycopene and β-carotene contents that was quantified by HPLC and UV–visible spectrophotometry.

Statistical analysis

All analytical experiments were performed in triplicate and the values for the contents of lycopene and β-carotene were analysed using the Minitab statistical software (version 17). For normally distributed data, the means were compared using the one-way analysis of variance (ANOVA). Non-normally distributed data were transformed (e.g. log10) or analysed using the non-parametric Kruskal–Wallis test and the significance difference between the means were analysed using Fisher’s post hoc test. Statistical values of p ≤ 0.05 were considered as significantly different. 36 M. cochinchinensis samples grown in its natural habitat in Vietnam (southern, central and northern) and Thailand was used to correlate carotenoid contents with eco-geographical factors. The Pearson’s correlation was used to determine whether the carotenoids correlated with any of the eco-geographical factors (latitude, longitude, maximum temperature, minimum temperature, annual temperature range, annual precipitation, rainfall of wettest and driest month and elevation) or with the different analysis methods used (HPLC, UV–Vis spectrophotometry and colorimeter). The r value represented the correlation coefficient between two variables (1 = total positive correlation, 0 = no correlation, −1 = total negative correction) and the statistically significant level was set at p ≤ 0.05. Principal component analysis (PCA) for lycopene and β-carotene content of 44 M. cochinchinensis samples was carried out using Minitab statistical software (version 17).

Results

Variation of carotenoid content in M. cochinchinensis based on collection sites

The carotenoid content of all tested samples was highly variable and ranged between 0.27 to 7.76 mg/g for lycopene and 0.20 to 8.27 mg/g for β-carotene (Fig. 1). The highest lycopene content was observed in samples collected from the Ha Noi (7.76 mg/g) of Northern Vietnam and Lam Ha (6.45 mg/g) and Lam Dong (6.64 mg/g) provinces of Central Vietnam. These samples contained lycopene at contents more than 28 times higher than those from the Can Tho province (0.27 mg/g) in Southern Vietnam (Fig. 1). The highest β-carotene content was observed in a sample from Nam Dinh (9.60 mg/g) in Northern Vietnam which was 37 times higher than those from Chiang Mai province (0.20 mg/g) of Thailand (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Variation in lycopene and β-carotene content of 44 M. cochinchinensis samples collected from Australia (AU), Thailand (TH) and Vietnam (northern = VN, central = VC and southern = VS) detected by HPLC. Data are shown as mean + standard deviation (SD) of three analytical replicates for an individual sample. * and # indicated statistical significance at p ≤ 0.05 for lycopene and β-carotene, respectively compared with HPLC

The proportions of lycopene and β-carotene were highly variable within the analysed fruits. Some fruits possessed high lycopene and low β-carotene (Lam Ha 6.46 mg/g lycopene and 1.77 mg/g β-carotene) contents whilst others had equally high (Hoa Binh 5.17 mg/g lycopene and 5.66 mg/g β-carotene) or equally low (Chanthaburi 0.88 mg/g lycopene and 0.82 mg/g β-carotene) contents. The fruits collected from most regions contained more lycopene than β-carotene (38 of 44 samples), with the highest difference being a 38-fold from the Chiang Mai region (Thailand).

The PCA analysis grouped the three Central Vietnam samples together (VC28, VC29 and VC32) and collectively contained the highest lycopene contents and was in agreement with the UPGMA cluster analysis previously reported (Wimalasiri et al. 2016). Samples from Southern Vietnam, Thailand and Australia was grouped together (Fig. 2) but the T1 and T3 samples were separated from the other Thailand samples due to the high β-carotene contents. The highest variation in lycopene and β-carotene content was observed in Northern Vietnam samples and was in agreement with genetic and morphological clusters (Wimalasiri et al. 2016). The Lam Dong province of Central Vietnam and the Hoa Binh province of Northern Vietnam possessed genotypes that contained high contents of lycopene and β-carotene, respectively. Therefore, these varieties should be considered for future commercial production. Up to date, no apparent genetic marker is identified for the superior lycopene and β-carotene contents in M. cochinchinensis from Central and Northern Vietnam. The contents of lycopene and β-carotene have been linked with the genes involved in their biosynthesis pathway in tomato (Smita et al. 2013), which has facilitated the identification of genotypes producing high carotenoid contents for commercial breeding programs (Smita et al. 2013). Future studies focused on the expression of key genes involved in lycopene (CRTISO PSY) and β-carotene (LCYB) accumulation will facilitate the selection of breeding stock of M. cochinchinensis for the development of consistently high carotenoid synthesising varieties.

Fig. 2.

Fig. 2

Open in a new tab

Score plot of lycopene and β-carotene of 44 M. cochinchinensis samples collected from Vietnam, Thailand and Australia constructed using the Principal Component Analysis method (PCA) with Minitab statistical software (Version 17). Country of collection was indicated as different colours; orange Southern Vietnam, blue Northern Vietnam, red Central Vietnam, green Thailand and olive green Australia

The effect of environment factors on the variation of carotenoid content of M. cochinchinensis

The samples from Australia was excluded from analysis to determine the correlation between carotenoid content and eco-geographical factors since they were grown in controlled environments compared to the samples from Vietnam and Thailand which were grown in their natural habitats. Of the seven environmental factors investigated in this study, temperature negatively influenced lycopene and β-carotene accumulation whilst elevation and precipitation positively influenced lycopene and β-carotene accumulation, respectively (Table 2; Supplementary Fig. 3).

Table 2.

Pearson correlation coefficients (r) and coefficient of determination (r 2) of carotenoids between eco-geographical factors (maximum and minimum temperature, annual temperature range, elevation and precipitation) and the carotenoid content of 36 M. cochinchinensis samples as measured by HPLC

Factor Details Correlation coefficient (r) Coefficient of determination (r 2)
Lycopene β-carotene Lycopene β-carotene
Temperature Minimum −0.53a −0.27NS 0.28 0.07
Maximum 0.32NS −0.42a 0.10 0.17
Range 0.40a 0.28NS 0.16 0.08
Elevation Elevation 0.38a −0.06NS 0.14 0.00
Precipitation Annual 0.25NS 0.40a 0.06 0.16
Driest month 0.29NS 0.40a 0.09 0.16
Wettest month 0.21NS 0.37a 0.04 0.14
Open in a new tab

a Significant at p < 0.05

NS Not significant p > 0.05

Temperate cooler climates with low temperatures, high temperature fluctuations and high elevations were correlated with a high lycopene content (Table 2; Fig. 1). The highest lycopene content was observed in samples from Northern and Central Vietnam, which are cooler regions with low minimum temperatures (<14 °C) and high temperature ranges (>18 °C) (Table 1; Fig. 1). The correlation between minimum temperature and lycopene contents was moderate, with an r 2 value of 0.28, which can interpret that 28% of the variability of lycopene contents in M. cochinchinensis was due to the minimum temperature (Table 2). Samples from high elevations correlated with a high lycopene content, such as those from Lam Dong and Lam Ha, which are situated at 925–1112 m above sea level (Table 1; Fig. 1). In comparison, a low content of lycopene was seen in samples from Southern Vietnam, which has high minimum temperatures (>20 °C) reminiscent of tropical climates and consequently low annual temperature ranges (<11 °C) and low elevations (1–6 m).

The β-carotene content was significantly reduced in warmer tropical climates where the maximum temperatures exceeded 34 °C (p < 0.05, r = −0.416) (Table 2) (e.g. Samut Prakan and HCM city). In contrast, a high β-carotene content was seen in samples from provinces that have cooler temperate climates (e.g. Lam Dong and Lam Ha). However, the β-carotene content was not correlated with elevation (Tables 1, 2). A high β-carotene content was also inversely correlated with moist conditions as measured by the annual precipitation (p < 0.05, r = −0.400) and the precipitation during the wettest (p < 0.05, r = −0.402) and driest month (p < 0.05, r = −0.371).

Comparisons between quantification assays

The lycopene content measured by UV–visible spectrophotometry and colorimetry methods was not significantly different (p < 0.05) from colorimetry. L*, a* and b* values showed significant relationships with lycopene measured by HPLC but the correlation was low. However, (a*/b*)2 transformation of colour value resulted a more linear relationship (Fig. 3). The estimated content of lycopene by UV–visible spectrophotometry (r = 0.802, p ≤ 0.001) and (a*/b*)2 value of colorimetry, (r = 0.734, p ≤ 0.001) was significantly correlated with HPLC (Table 3; Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Scatter plots showing correlations between carotenoids (lycopene and β-carotene) of 44 M. cochinchinensis aril extracts and analytical methods. a Scatter plot showing correlations between lycopene and UV–Vis spectrophotometry. b Scatter plot showing correlations between lycopene and (a*/b*)2. c Scatter plot showing correlations between β-carotene and UV–Vis spectrophotometry. d Scatter plot showing correlations between β-carotene and (a*)4 × 106

Table 3.

Pearson correlation coefficients (r) between colorimetry, spectrophotometry and chromatography (HPLC) of 44 M. cochinchinensis samples

Analytical test Details Correlation (r) with carotenoids (HPLC) Correlation of determination (r 2) with carotenoids (HPLC)
Lycopene β-carotene Lycopene β-carotene
Colorimetry L* −0.512a −0.031NS 0.26 –
a* 0.311a 0.408a 0.10 0.17
b* −0.412a 0.127NS 0.17 0.02
a*/b* 0.686a 0.069NS 0.46 –
(a*/b*)2 0.734a 0.063NS 0.54 –
Hue −0.161NS −0.017NS – –
Chroma −0.216NS 0.188NS – –
(a*)2 0.302a 0.418a 0.09 0.17
(a*)4 0.288NS 0.427a – 0.18
(a*)0.5 0.315a 0.402a 0.10 0.16
a*b* −0.216NS 0.283NS – –
Spectrophotometry 502 nm 0.802 NA 0.64 –
450 nm NA 0.490 – 0.24
Open in a new tab

a Significant at p < 0.05

NS Not significant at p > 0.05

NA Not available

With respect to β-carotene quantification, there was a low correlation between the spectrophotometry and chromatography (r = 0.49, p ≤ 0.001) (Table 3) results. The untransformed a* value and (a*)4 value showed significant relationship with β-carotene measured by HPLC but the correlation was low (Table 3) indicating that colorimetry measurements are not suitable for the estimation of β-carotene content in M. cochinchinensis.

Discussion

Variation of carotenoid content in M. cochinchinensis based on collection sites

The geographical distribution of the collected M. cochinchinensis fruit was related to the carotenoid content, with the best levels observed in samples from specific provinces; the diversity of carotenoid level was vast, ranging from 0.88 to 6.30 mg/g for lycopene and 0.25 to 5.70 mg/g for β-carotene. The carotenoid content was significantly different between provinces with the highest lycopene contents found in the samples from Central Vietnam in Lam Ha province and the highest β-carotene contents measured in the samples from Northern Vietnam in Hoa Binh province. This distribution of carotenoid content was related to that of the morphology from the same provinces, with a notable character being that Central Vietnam samples had the heaviest and largest seeds (Wimalasiri et al. 2016). The same samples were genetically and morphologically clustered (Wimalasiri et al. 2016), and was reflected in PCA analysis with high lycopene and β-carotene contents. The highest diversity in lycopene and β-carotene contents was seen in the samples from Northern Vietnam and was in agreement with genetic and morphology clustering (Wimalasiri et al. 2016). However, this diversity might be attributed by the larger number of Northern Vietnamese samples analysed in this study. Interestingly, the samples collected from Australia had similar contents of lycopene and β-carotene as did the Southern Vietnam samples and was grouped together indicating that the variety might have an influence on the nutrition content. However, the Thailand samples were not separated from the samples from Australia and Southern Vietnam as did for the genetic and morphology clustering indicating that there might be other factors, such as soil nutrition, micro-climatic differences and limited sampling per each location that are responsible for variations in carotenoids.

All samples analysed were genetically diverse based on molecular markers (Wimalasiri et al. 2016). However, there is no apparent genetic marker identified for lycopene content in M. cochinchinensis, which could make the plant produce fruits with consistent higher yields of lycopene as observed for samples from Lam Ha province compared with Chanthaburi province. Studies on other carotenoid rich foods, such as sweet potato, pumpkin squash and tomato, have also found that the genotype and variety determined the lycopene, phenolic compounds and antioxidant superiority of these fruits (Chandra et al. 2012; Lenucci et al. 2006; Martínez-Valverde et al. 2002; Rodriguez-Amaya and Kimura 2004). However, this is contradictory to other studies, which have suggested that the same varieties grown in separate countries with different climates had a diverse carotenoid content, indicating that geographical location had a greater impact than the variety (Nishiyama et al. 2005; Raffo et al. 2006; Rodriguez-Amaya and Kimura 2004; Sass-Kiss et al. 2005). This information on varietal differences and downstream influences on carotenoid accumulation was previously unknown for M. cochinchinensis. Further experiments are pivotal to elucidate if genetically identical varieties could be encouraged to produce higher lycopene contents for this potentially valuable crop.

The carotenoid contents of M. cochinchinensis samples analysed in this study were higher than expected and were 3 times higher for lycopene and 8 times higher for β-carotene compared to that previously reported for M. cochinchinensis (Aoki et al. 2002; Ishida et al. 2004; Kubola and Siriamornpun 2011). The global carotenoid market value was estimated to be $1.5 billion in 2014 and it is expected to reach nearly $1.8 billion in 2019, with an annual growth rate of 3.9% (BCC 2015). M. cochinchinensis can contribute to this expanding carotenoid market as an alternate source of lycopene and β-carotene. This information will be imperative for primary producers to facilitate the conservation, sharing and breeding of better varieties of M. cochinchinensis with consistently higher carotenoids.

The effect of environmental conditions on variation of carotenoid content in M. cochinchinensis

Of the seven environmental factors analysed, the temperature, elevation and precipitation were related to the carotenoid content of the M. cochinchinensis aril and therefore, they may have influenced the specific metabolic pathways for the production of lycopene, β-carotene or both. The optimum temperatures for lycopene synthesis are known to be between 16 and 21 °C for tomatoes (Martínez-Valverde et al. 2002) and, at temperatures above 30 °C, lycopene is known to be converted to β-carotene (Brandt et al. 2006; Dumas et al. 2003; Nishiyama et al. 2005). Therefore, high temperatures may have caused the low lycopene content observed in the samples from tropical Southern Vietnam and Thailand provinces, where the maximum temperatures exceed 30 °C compared to the samples from the temperate regions of Central Vietnam where the temperatures does not exceed 28 °C. Low temperatures at high altitudes with high UV radiation, particularly in Central Vietnam where the varieties are grown in mountainous provinces, may explain the high contents of carotenoids. This was consistent with other studies, which indicated that low temperatures, high altitudes and high UV radiation increased secondary metabolite synthesis in other fruits such as tomato, grape berry and passionfruit (Jaakola and Hohtola 2010; Rodriguez-Amaya and Kimura 2004). The influence of geography and climate has been found to override genetic control in tomatoes (Aherne et al. 2009) and this finding has been implemented using seasonal and environmental manipulations to improve lycopene synthesis and bio-accumulation (Raffo et al. 2006).

Interestingly, a M. cochinchinensis variety originated from Southern Vietnam (Wimalasiri et al. 2016) but grown in temperature-controlled glasshouses at 25 °C in Australia did not produce higher contents of lycopene than the samples collected from Southern Vietnam. This suggests that variables other than temperature may be involved in carotenoid synthesis and bioaccumulation in M. cochinchinensis. Since there is a lack of genetic markers to predict higher carotenoid content, eco-geographical contributions could assist primary producers to maximise the carotenoid production. More experiments are currently underway with the same genotypes growing in different provinces to determine the dominance of eco-geographic versus genetic influences on lycopene accumulation.

Comparisons between quantification assays

Selection of fruits and cultivars with higher carotenoid content require rapid analysis methods which are suitable to use in field conditions. The estimation of lycopene in the aril of M. cochinchinensis was consistent across the chromatographic, colorimetric and spectrophotometric methods. High performance liquid chromatography (HPLC) is the standard method to quantify carotenoids in fruits and vegetables but it is too costly and inefficient for resource poor farmers. The use of colorimetry and spectrophotometry to estimate lycopene is more feasible and this has been implemented in field applications for tomatoes (Arias et al. 2000) and cucurbits such as pumpkins and squash (Itle and Kabelka 2009). These methods have been used to detect differences among genotypes (Hyman et al. 2004). Therefore, colorimetry method can be used to select lycopene rich M. cochinchinensis fruits in farms and genotypes with higher lycopene as breeding stock.

In comparison, the estimation of β-carotene was consistent with spectrophotometry but not colorimetry, compared with HPLC. The inconsistency of the colorimetry results may be due to interference from a high lycopene content. This has been confirmed in other carotenoid studies when lycopene, xanthophyll and chlorophyll were co-localised in homogenates (Arias et al. 2000; Fish 2012; Hyman et al. 2004). Therfore, the colorimetry method may require additional separation, fractionation and purification for the accurate quantification of β-carotene in M. cochinchinensis.

Conclusion

This study confirms the nutritional variation of M. cochinchinensis from wild and cultivated samples obtained from Vietnam and Thailand. The fruit contained up to 200 times more lycopene than hydroponic tomatoes and 54 times more β-carotene than carrots than previously reported. The carotenoid contents in this study were more than expected; they were 3 times higher in lycopene and 8 times higher in β-carotene than the highest content previously reported for M. cochinchinensis. Varieties with high carotenoids (lycopene and β-carotene) and grown under optimal eco-geographical parameters (optimum temperature, elevation and precipitation) were identified. These findings will be pivotal for the agricultural development of this fruit for the food and nutriceutical industries.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOCX 78 kb) (78.2KB, docx)
Supplementary material 2 (DOCX 77 kb) (77.6KB, docx)
Supplementary material 3 (DOCX 18 kb) (18.7KB, docx)

Acknowledgement

We are grateful for the support from Thu Nga Nguyen and Huong Nguyen in (Hanoi University of Agriculture), Nonnipa Labbunruang and Wanisika Phadungsii (Tammasat University) and Thuy-em Pham (Saigon) for collection assistance; Duc Nguyen (Hanoi University of Agriculture) for access and the use of laboratory facilities, Cuong Nguyen (Hanoi University of Agriculture) for agronomy advice; Rattanapong Charntawong (Thailand gac farm), Dr Sophie Parks (NSW Department of Primary Industries, Australia) and Professor Minh Nguyen for providing M. cochinchinensis fruit samples. We thank RMIT University for the travel grant awarded to Tien Huynh to permit the samples to be collected.

Footnotes

Electronic supplementary material

The online version of this article (doi:10.1007/s13197-017-2719-0) contains supplementary material, which is available to authorized users.

References

  1. Aherne SA, Jiwan MA, Daly T, O’Brien NM. Geographical location has greater impact on carotenoid content and bioaccessibility from tomatoes than variety. Plant Foods Hum Nutr. 2009;64:250–256. doi: 10.1007/s11130-009-0136-x. [DOI] [PubMed] [Google Scholar]
  2. Aoki H, Kieu NTM, Kuze N, Tomisaka K, Chuyen NV. Carotenoid pigments in GAC fruit (Momordica cochinchinensis Spreng) Biosci Biotechnol Biochem. 2002;66:2479–2482. doi: 10.1271/bbb.66.2479. [DOI] [PubMed] [Google Scholar]
  3. Arias R, Lee T-C, Logendra L, Janes H. Correlation of lycopene measured by HPLC with the L*, a*, b* color readings of a hydroponic tomato and the relationship of maturity with color and lycopene content. J Agric Food Chem. 2000;48:1697–1702. doi: 10.1021/jf990974e. [DOI] [PubMed] [Google Scholar]
  4. Barba A, Hurtado MC, Mata M, Ruiz VF, Tejada M. Application of a UV–Vis detection-HPLC method for a rapid determination of lycopene and β-carotene in vegetables. Food Chem. 2006;95:328–336. doi: 10.1016/j.foodchem.2005.02.028. [DOI] [Google Scholar]
  5. BCC R (2015) Report FOD025E The global market for carotenoids. http://www.bccresearch.com/market-research/food-and-beverage/carotenoids-global-market-report-fod025e.html
  6. Biehler E, Mayer F, Hoffmann L, Krause E, Bohn T. Comparison of 3 spectrophotometric methods for carotenoid determination in frequently consumed fruits and vegetables. J Food Sci. 2010;75:C55–C61. doi: 10.1111/j.1750-3841.2009.01417.x. [DOI] [PubMed] [Google Scholar]
  7. Bootprom N, Songsri P, Suriharn B, Chareonsap P, Sanitchon J, Lertrat K. Molecular diversity among selected Momordica cochinchinensis (Lour.) spreng accessions using rapd markers. SABRAO J Breed Genet. 2012;44:406–417. [Google Scholar]
  8. Brandt S, Lugasi A, Barna É, Hóvári J, Pék Z, Helyes L. Effects of the growing methods and conditions on the lycopene content of tomato fruits. Acta Alimentaria. 2003;32:269–278. doi: 10.1556/AAlim.32.2003.3.6. [DOI] [Google Scholar]
  9. Brandt S, Pék Z, Barna É, Lugasi A, Helyes L. Lycopene content and colour of ripening tomatoes as affected by environmental conditions. J Sci Food Agric. 2006;86:568–572. doi: 10.1002/jsfa.2390. [DOI] [Google Scholar]
  10. Chandra HM, Shanmugaraj BM, Srinivasan B, Ramalingam S. Influence of genotypic variations on antioxidant properties in different fractions of tomato. J Food Sci. 2012;77:C1174–C1178. doi: 10.1111/j.1750-3841.2012.02962.x. [DOI] [PubMed] [Google Scholar]
  11. Chuyen HV, Nguyen MH, Roach PD, Golding JB, Parks SE. Gac fruit (Momordica cochinchinensis Spreng.): a rich source of bioactive compounds and its potential health benefits. Int J Food Sci Technol. 2015;50:567–577. doi: 10.1111/ijfs.12721. [DOI] [Google Scholar]
  12. Dragovic-Uzelac V, Levaj B, Mrkic V, Bursac D, Boras M. The content of polyphenols and carotenoids in three apricot cultivars depending on stage of maturity and geographical region. Food Chem. 2007;102:966–975. doi: 10.1016/j.foodchem.2006.04.001. [DOI] [Google Scholar]
  13. Dumas Y, Dadomo M, Di Lucca G, Grolier P. Effects of environmental factors and agricultural techniques on antioxidant content of tomatoes. J Sci Food Agric. 2003;83:369–382. doi: 10.1002/jsfa.1370. [DOI] [Google Scholar]
  14. Fish WW. Refinements of the attending equations for several spectral methods that provide improved quantification of β-carotene and/or lycopene in selected foods. Postharvest Biol Technol. 2012;66:16–22. doi: 10.1016/j.postharvbio.2011.08.007. [DOI] [Google Scholar]
  15. Gul K, Tak A, Singh A, Singh P, Yousuf B, Wani AA, Yildiz F. Chemistry, encapsulation, and health benefits of β-carotene—a review. Cogent Food Agric. 2015;1:1–12. [Google Scholar]
  16. Hammond BR, Renzi LM. Carotenoids. Adv Nutr. 2013;4:474–476. doi: 10.3945/an.113.004028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hijmans RJ, Guarino L, Bussink C, Mathur P, Cruz M, Rojas E (2004) DIVA-GIS 5.2 mannual
  18. Hunter RS. The measurement of appearance. New York, USA: Wiley; 1987. [Google Scholar]
  19. Hyman JR, Gaus J, Foolad MR. A rapid and accurate method for estimating tomato lycopene content by measuring chromaticity values of fruit puree. J Am Soc Hortic Sci. 2004;129:717–723. [Google Scholar]
  20. Ishida BK, Turner C, Chapman MH, McKeon TA. Fatty acid and carotenoid composition of gac (Momordica cochinchinensis Spreng) fruit. J Agric Food Chem. 2004;52:274–279. doi: 10.1021/jf030616i. [DOI] [PubMed] [Google Scholar]
  21. Itle RA, Kabelka EA. Correlation between L* a* b* color space values and carotenoid content in pumpkins and squash (Cucurbita spp.) HortScience. 2009;44:633–637. [Google Scholar]
  22. Jaakola L, Hohtola A. Effect of latitude on flavonoid biosynthesis in plants. Plant Cell Environ. 2010;33:1239–1247. doi: 10.1111/j.1365-3040.2010.02154.x. [DOI] [PubMed] [Google Scholar]
  23. Kubola J, Siriamornpun S. Phytochemicals and antioxidant activity of different fruit fractions (peel, pulp, aril and seed) of Thai gac (Momordica cochinchinensis Spreng) Food Chem. 2011;127:1138–1145. doi: 10.1016/j.foodchem.2011.01.115. [DOI] [PubMed] [Google Scholar]
  24. Leja M, Kamińska I, Kramer M, Maksylewicz-Kaul A, Kammerer D, Carle R, Baranski R. The content of phenolic compounds and radical scavenging activity varies with carrot origin and root color. Plant Foods Hum Nutr. 2013;68:163–170. doi: 10.1007/s11130-013-0351-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leoni C. Industrial quality as influenced by crop management. Acta Hortic. 1992;301:177–184. doi: 10.17660/ActaHortic.1992.301.20. [DOI] [Google Scholar]
  26. Lenucci MS, Cadinu D, Taurino M, Piro G, Dalessandro G. Antioxidant composition in cherry and high-pigment tomato cultivars. J Agric Food Chem. 2006;54:2606–2613. doi: 10.1021/jf052920c. [DOI] [PubMed] [Google Scholar]
  27. Martínez-Valverde I, Periago MJ, Provan G, Chesson A. Phenolic compounds, lycopene and antioxidant activity in commercial varieties of tomato (Lycopersicum esculentum) J Sci Food Agric. 2002;82:323–330. doi: 10.1002/jsfa.1035. [DOI] [Google Scholar]
  28. Nishiyama I, Fukuda T, Oota T. Genotypic differences in chlorophyll, lutein, and β-carotene contents in the fruits of Actinidia species. J Agric Food Chem. 2005;53:6403–6407. doi: 10.1021/jf050785y. [DOI] [PubMed] [Google Scholar]
  29. Perkins-Veazie P, Collins JK, Pair SD, Roberts W. Lycopene content differs among red-fleshed watermelon cultivars. J Sci Food Agric. 2001;81:983–987. doi: 10.1002/jsfa.880. [DOI] [Google Scholar]
  30. Raffo A, Malfa GL, Fogliano V, Maiani G, Quaglia G. Seasonal variations in antioxidant components of cherry tomatoes (Lycopersicon esculentum cv. Naomi F1) J Food Compos Anal. 2006;19:11–19. doi: 10.1016/j.jfca.2005.02.003. [DOI] [Google Scholar]
  31. Rao A, Rao LG. Carotenoids and human health. Pharmacol Res. 2007;55:207–216. doi: 10.1016/j.phrs.2007.01.012. [DOI] [PubMed] [Google Scholar]
  32. Rodriguez-Amaya DB, Kimura M. HarvestPlus handbook for carotenoid analysis. Washington: International Food Policy Research Institute (IFPRI); 2004. [Google Scholar]
  33. Sass-Kiss A, Kiss J, Milotay P, Kerek M, Toth-Markus M. Differences in anthocyanin and carotenoid content of fruits and vegetables. Food Res Int. 2005;38:1023–1029. doi: 10.1016/j.foodres.2005.03.014. [DOI] [Google Scholar]
  34. Smita S, Rajwanshi R, Lenka SK, Katiyar A, Chinnusamy V, Bansal KC. Comparative analysis of fruit transcriptome in tomato (Solanum lycopersicum) genotypes with contrasting lycopene contents. Plant Mol Biol Rep. 2013;31:1384–1396. doi: 10.1007/s11105-013-0613-0. [DOI] [Google Scholar]
  35. Vuong LT, Franke AA, Custer LJ, Murphy SP. Momordica cochinchinensis Spreng. (gac) fruit carotenoids reevaluated. J Food Compos Anal. 2006;19:664–668. doi: 10.1016/j.jfca.2005.02.001. [DOI] [Google Scholar]
  36. Wimalasiri D, Piva T, Urban S, Huynh T. Morphological and genetic diversity of Momordica cochinchinenesis (Cucurbitaceae) in Vietnam and Thailand. Genet Resour Crop Evol. 2015;63(1):19–33. doi: 10.1007/s10722-015-0232-8. [DOI] [Google Scholar]
  37. Zechmeister L, LeRosen AL, Schroeder WA, Polgár A, Pauling L. Spectral characteristics and configuration of some stereoisomeric carotenoids including prolycopene and pro-γ-carotene. J Am Chem Soc. 1943;65:1940–1951. doi: 10.1021/ja01250a039. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material 1 (DOCX 78 kb) (78.2KB, docx)
Supplementary material 2 (DOCX 77 kb) (77.6KB, docx)
Supplementary material 3 (DOCX 18 kb) (18.7KB, docx)

Articles from Journal of Food Science and Technology are provided here courtesy of Springer

RESOURCES