Chemicals and Bioactivity Discrimination of Syconia of Seven Varieties of Ficus deltoidea Jack via ATR-IR Spectroscopic-Based Metabolomics

Yunusa, Rashid, Mat, Abu Bakar, and Ali: Chemicals and Bioactivity Discrimination of Syconia of Seven Varieties of Ficus deltoidea Jack via ATR-IR Spectroscopic-Based Metabolomics

Authors

https://s3-us-west-2.amazonaws.com/jourdata/pj/pharmacognj-10-6s-147_g000.jpg

INTRODUCTION

FTIR spectroscopy is a fast, non-destructive and sensitive technique, which has been widely used for chemicals fingerprinting. It is suitable for natural products bioactive compounds analysis,1 since no two compounds will have the same spectra.2 It has been used for comparison and discrimination of plant/food from different origins and cultivars, such as almond cultivars subjected to oxidative treatments.3

Ficus deltoidea belongs to the family of Moraceae. It is distributed throughout the Southeast Asia as well as Africa.4 In Malaysia, it is called mas cotek or sempit-sempit, while in Indonesia is known as tabat barito. Its other vernacular names are Delta fig, Fig shrub and Mistletoe.4 It is a herb with various therapeutic benefits such as for reducing diabetic symptoms and other health risk related conditions such as obesity.5 Various studies have applied spectroscopy coupled with MVDA to examine the differences/similarity in cultivars or plant varieties.6 Previously, the six varieties that showed leaf morphological variations by quantitative measurement on different parts of the plant, are: var. deltoidea, var. angustifolia, var. trengganuensis, var. bilobata, var. intermedia and var. kunstleri.7

However, to the best of our knowledge, no literature is available for the discrimination of syconia of different varieties of F. deltoidea. Therefore, this study was aimed to chemically discriminate the syconia of seven varieties of F. deltoidea and to determine the correlation with their bioactivities.

MATERIALS AND METHODS

Sample collection and preparation

Syconia of seven varieties of Ficus deltoidea (3 accessions from each variety) were collected from the germplasm of Faculty of Bioresources and Food Industry, Universiti Sultan Zainal Abidin. All syconia were carefully selected to be identical in terms of colour and ripening stage. The syconia of the seven varieties were deposited as voucher specimens at the Faculty’s herbarium. The syconia samples were dried and then ground into powder form and stored at –20ºC before extraction.

Chemical and reagents

Ferric chloride, 2, 4, 6-tris(2-pyridyl)-s-triazine, 2,2-Diphenyl-1-picrylhydrazyl (DPPH), sodium acetate trihydrate, were purchased from Sigma-Aldrich Co. (Switzerland). Acetic acid was purchased from R & M Chemicals. Hydrochloric acid was purchased from Merck (Germany). α-glucosidase enzyme and p-nitrophenyl-α-D-glucopyranose (PNPG) were supplied by Sigma-Aldrich (St. Louis, MO, USA). All solvents used for extraction and other chemicals were of analytical grade.

Extraction

Dried and powdered syconia were extracted individually with ethanol in a soxhlet apparatus. The extraction was repeated twice after decanting off prior extracts. Each combined ethanol were filtered, pooled and concentrated under reduced pressure at 40°C to yield crude ethanolic extracts (EE). The dried extracts were stored at -20 °C before used.

Biological activities assays

DPPH radical scavenging ability of samples were measured following,8 and modified by.9 FRAP was measured according to,10 and modified by.11 The alpha glucosidase assay, was performed according to,12 using alpha glucosidase enzymes from Saccharomyces cerevisiae at different concentrations (1.56 to 100 μg/mL) as substrate and quercetin as standard.

ATR-FTIR analysis

ATR-FTIR spectra of extracts were collected by using a Bruker Analitik IFS 66 FTIR spectrometer (Ettlingen, Germany) equipped with a DTGS KBr detector and a Golden Gate Single Reflection Diamond ATR accessory (incident angle of 45°). Spectra were recorded in the absorbance mode at mid infrared region (4000–400 cm-1) by using 16 scans and 4 cm-1 resolution. Three spectra were obtained for each sample.

Statistical analysis

Results are presented as the mean values of three different experiments and are presented as the mean ± standard deviation (SD). Analysis of variance (ANOVA) was used to determine the differences on antioxidant activities and alpha glucosidase inhibition activity. The alpha level of all analysis was at 0.05.

Data pre-processing and multivariate analysis

Each FTIR spectrum was baseline corrected and smoothed to seven data points,13 by using a software to minimize the differences between spectra due to the baseline shifts. The spectra were then exported to ASCII file and excel files were prepared for chemometric analysis. The multivariate analyses ((PCA, OPLS-DA, OPLS-DA-HCA and PLS models) of FTIR data were produced using SIMCA-P software version 13 (Umetrics, Umea Sweden).

RESULTS AND DISCUSSION

Most of the accessions showed DPPH free radical scavenging activity comparable to quercetin (standard). The IC50 values of extracts (100 μg/mL) ranged from 13.58 ± 3.56 μg/mL to 79.3 ± 3.2 μg/mL, though they were less than that of quercetin (Table 1). In regards to the syconia, only few studies on some varieties of F. deltoidea have been found compared to the leaves part such as DPPH radical scavenging activity of syconia of var. deltoidea and var. intermedia (both with IC50; 7.8 μg/mL).14

Table 1

DPPH free radical scavenging activity, ferric reducing power (FRAP) of ethanolic extracts of syconia of seven varieties of Ficus deltoidea.

Accession CodeDPPH scavenging activity (IC50; μg/mL)FRAP (mmol Fe2+/g)
T00231.83 ± 5.48cde6.13 ± 0.33ef
T01960.42 ± 2.67b2.82 ± 0.36iJ
T020ND2.27 ± 0.21jkl
K00313.58 ± 3.57fg9.72 ± 0.70a
K21718.17 ± 3.21efg7.91 ± 0.37cd
K31317.83 ± 3.62efg7.64 ±0.43cd
A17123.83 ± 5.58def6.47 ± 0.29e
A32136.50 ± 5.41cd5.02 ± 0.01g
A29565.33 ± 13.01ab2.67 ± 0.23ijk
D00615.83 ± 3.82efg8.64 ± 0.64b
D17214.50 ± 2.18fg7.39 ± 0.54d
D15625.17 ± 6.66cdef8.14 ± 1.05bc
B01338.42 ± 9.74cd3.27 ± 0.30i
B01416.75 ± 2.46efg7.35 ± 0.35d
B37822.42 ± 3.02defg5.50 ± 0.25fg
I32340.75 ± 1.64c4.19 ± 0.16h
I38668.83 ± 6.17ab2.82 ± 0.24ij
I38779.33 ± 3.21a2.73 ± 0.18ij
M234A77.20 ± 21.77a1.84 ± 0.28l
M234BND1.11 ± 0.23m
M234C63.33 ± 24.77ab1.96 ± 0.09kl
Quercetin5.50 ± 0.87gNT

Values are the means ± standard deviation based on three different experiments. Superscript letters refer to significant different (p<0.05) by comparing among syconia accessions. Means with different superscript letters were significantly different (p<0.05). IC50: inhibition concentration. ND: not detected, NT: not tested, T: var. trengganuensis, K: var. kunstleri, A: var. angustifolia, D: var. deltoidea, B: var. bilobata, I: var. intermedia, M: var. motleyana.

In FRAP assay, the ability of the plant/food extracts to reduce Fe3+ to Fe2+ were evaluated. The results showed that almost all extracts have strong reducing power. FRAP values of the extracts ranged from 1.1 (M234B) to 9.72 (K003) mmol/g which in comparison indicated significant differences (p<0.05) between all varieties (Table 1). FRAP of var. angustifolia and var. kunstleri were found to be significantly lower (1.8 and 1.3 mmol/g)15 than the similar varieties used in our study.

The current study indicated strong α-glucosidase inhibition activity in most of the syconia extracts of the seven F. deltoidea varieties. Potential activity of all accessions was detected (IC50) except from var. motleyana. The representatives from the seven varieties with highest inhibition are shown in Figure 1. Among all accessions tested, the extracts of var. trengganuensis (T002) and var. kunstleri (K003) at different concentrations (1.56 to 100 μg/mL) exhibited the strongest activity with IC50 of 36.7 and 37.8 μg/mL, respectively (Table 2). Our findings are consistently agreed with16 who reported ethanol extract of Ipomoea aquatic had stronger α-glucosidase inhibition, in concentration dependent manner.

Figure 1

Bioactivities of ethanolic extracts of the seven varieties of Ficus deltoidea syconia. A) DPPH free radical scavenging activity B) α-glucosidase inhibition. Data represented by the most active accession of each variety.Values are the means ± standard error based on three different experiments.

https://s3-us-west-2.amazonaws.com/jourdata/pj/pharmacognj-10-6s-147_g001.jpg
Table 2

α-glucosidase inhibition of ethanolic extracts of syconia of seven varieties of Ficus deltoidea.

Accession Codeα-Glucosidase Inhibition (IC50; μg/mL)
T00236.75 ± 2.75f
T019ND
T020ND
K00337.67 ± 4.22f
K21744.33 ± 12.34ef
K31358.00 ± 17.84cdef
A17156.17 ± 10.68def
A32163.75 ± 18.10bcde
A295ND
D00681.50 ± 12.26abc
D17277.83 ± 5.01abcd
D15671.83 ± 12.92abcd
B013ND
B01470.67 ± 12.50abcd
B37883.83 ± 19.66ab
I32376.67 ± 1.53abcd
I386ND
I38789.50 ± 4.27a
M234AND
M234BND
M234CND
Quercetin65.17 ± 18.87bcde

Values are the means ± standard deviation based on three different experiments. Superscript letters refer to significant different (p<0.05) by comparing among syconia accessions. Means with different superscript letters were significantly different (p<0.05). IC50: inhibition concentration. ND: not detected, T: var. trengganuensis, K: var. kunstleri, A: var. angustifolia, D: var. deltoidea, B: var. bilobata, I: var. intermedia, M: var. motleyana.

The mean spectra of three different measurements were given in Figure 2. The spectra showed broad peaks at wavenumbers in range 3600-3000 cm-1 and sharp peaks at 1604.77 and 1038 cm-1. They suggested the presence of O-H bonds of either alcohols, phenols or water fraction at 3292.2 cm-1, the sp3 and sp2 stretching of C-H bonds at bands ranging from 3000 cm-1 to 2850 cm-1 and the saturated fatty acids fraction at 2923 cm−1.17 The C=O stretching of aldehyde group presence at 1734.3 cm-1,1 C=C stretching of alkenes and N-H bending of amines and amides at 1612.4 cm-1,1 and the vibration of N=O and stretching of C=C in aromatics at 1514.2 cm-1,1 were also assigned. Besides, other peaks at 1416 cm-1 were assigned as the alkane C-H bending, at 1368.6 cm-1 represent S=O, at 1162 cm-1 for carboxylic acids and 1043.5 cm-1 C-N for vibration of amines. In addition, the fingerprints of 1400 to 900 cm−1 may also be the carbohydrates signals.17

Figure 2

ATR-FTIR spectra of ethanolic extracts of syconia of the seven varieties of Ficus deltoidea.

https://s3-us-west-2.amazonaws.com/jourdata/pj/pharmacognj-10-6s-147_g002.jpg

FTIR data sets were subjected to principal component analysis (PCA) for unsupervised classification of syconia of the seven varieties of F. deltoidea. The first two principal components explained 80.2% of the total variance. The model revealed goodness of fit R2X (cum) and prediction of the model Q2 (cum) with values of 0.986 and 0.973, respectively. The loading line plots showed that 2918, 2849, 1730 and 1708 cm-1 were the fingerprints contributed to the variation along PC1 Meanwhile, the loading plot along PC2 revealed that 1602 and 1440 cm-1 contributed the most to the variation (data not shown). In the present study, clustering of groups of varieties in PCA model of FTIR spectra were observed, though some of the varieties accessions were overlapped (Figure 3A). The overlapping of spectra was suggested to be due to similarity in their phytochemicals content. Meanwhile, the clusters differences were possibly due to other distinctive metabolites variation such exhibited by Sri Lanka green tea that clustered away from teas of other origins.18

Figure 3

Unsupervised and supervised multivariate data analysis (MVDA) of ATR-IR spectra to discrimination of chemical variability and the relationship between the alpha glucosidse inhibition, antioxidant activity and Ficus deltoidea varieties. A) Principal component analysis (PCA), B) Orthogonal partial least square discriminant analysis (OPLS-DA), C) Orthogonal partial least square discriminant analysis hierarchical cluster analysis (OPLS-DA-HCA) and D) Partial least square (PLS) model indicating the relationship between chemicals and extracts bioactivities.

https://s3-us-west-2.amazonaws.com/jourdata/pj/pharmacognj-10-6s-147_g003.jpg

Orthogonal partial least square–discriminant analysis (OPLS-DA) model was established as the supervised discrimination of syconia of seven varieties of F. deltoidea. The model was confirmed by the Q2Y and R2Y cumulatives, in which R2Y (cum) was 0.964 and Q2 (cum) was 0.914. The first two PCs explained 39.2% of the variance. Six clusters were identified on PC1. The score plot showed that var. intermedia and var. bilobata were clustered together, revealing that the two varieties have phytochemicals similarity (Figure 3B). Var. trengganuensis, var. angustifolia and var. motleyana, discriminated in the positive quadrant of PC1, while, var. intermedia, var. bilobata, var. kunstleri and var. deltoidea were located into the negative quadrant of PC1, due to fingerprints at 1600, 1515, 1448, 1095 and 762 cm-1. In addition, 1705, 443, 453, 484 cm-1 were the fingerprints that contributed to the discrimination of var. intermedia and var. angustifolia into the positive quadrant of PC2 (data not shown). Interestingly, var. bilobata and var. intermedia formed one cluster and discriminated themselves from var. kunstleri and var. deltoidea along PC2. The loading plot reveals that 1729, 1448, 1095, 1705, 443 and 453 cm-1 were the fingerprints with the highest contribution to the discrimination ((loading plot data not shown)). OPLS-DA-HCA was performed and similarities among the varieties were calculated using Euclidean distance and the clusters were established using ward hierarchical agglomerative method. The resulting dendrogram of the OPLS-DA shows the clustering of the seven varieties of F. deltoidea extracts. Three clusters (C1-C3) were observed. Accessions of var. trengganuensis and var. motleyana clustered in C1; var. kunstleri and var. deltoidea in C2; and var. angustifolia, var. intermedia and var. var. bilobata in C3 (Figure 3C). HCA model of OPLS-DA gave distinctive clusters involving all accessions of each variety. This indicated that supervised HCA of OPLS-DA model was a better representation than HCA of PCA model in discussing hierarchy of chemicals clusters of the syconia of seven varieties of F. deltoidea.

Partial least square (PLS) model was established to investigate the relationship between the bioactivity and FTIR fingerprints of different varieties of F. deltoidea. PLS model consisted of two variables; the biological activities (DPPH, FRAP and α-glucosidase inhibitory) are represented by the y-axis while the set of wavenumbers are the x-axis (4000-400 cm-1). The PLS model was cross-validated using 100 permutation tests with all Q2 values lower than those of original ones. The cumulative R2X, R2Y and Q2 values indicated good predictive abilities (R2X = 0.904; R2Y = 0.863; Q2 = 0.806). In the bi-plot, accessions exhibited the most potent biological activities (DPPH, FRAP and α-glucosidase) include var. kunstleri (K003, K217, K313), var. deltoidea (D006, D172, D156), var. bilobata (B014, B378), var. angustifolia (A171, A321) and var. intermedia (I323, I386, I387); were located on the right side of PC1 which separated from other accessions having weaker biological activities on PC1 left side (Figure 3D). Interestingly, in this study, PLS model of FTIR spectra of syconia of the seven varieties of F. deltoidea showed discrimination which corresponded to bioactivity results (Figure 3D).

CONCLUSION

In general, the results show potential α-glucosidase inhibitory and antioxidant activities of syconia extracts of almost all varieties. Among all Ficus deltoidea varieties, most promising antioxidant and anti hyperglycemic activities were found in var. kunstleri (K003, K217, K313), var. trengganuensis (T002) and var. angustifolia (A171, A321). OPLS-DA model of FTIR data successfully discriminated the varieties, while PLS model had identified the correlation between var. kunstleri, var. deltoidea and var. intermedia bioactivity and their respective chemicals fingerprints, those were also identified. Thus, their syconia part were determined to be a good source of natural antioxidants as well as α-glucosidase inhibitors. Instead of being underutilized, the syconia of these targeted varieties (elite accessions) could be developed as ingredient in food products and health supplement in local agro-industry besides the widely studied Ficus deltoidea leaves.

ACKNOWLEDGEMENT

This project was funded by the Ministry of Agriculture and Agro-Based Industry, Malaysia under NKEA EPP#1 (NRGS), Universiti Islam Malaysia and the Ministry of Higher Education of Malaysia, under a research grant Project Number-RR107. We also thank Universiti Sultan Zainal Abidin (UniSZA) and the Institute of Bioscience (IBS), Universiti Putra Malaysia for the laboratory facilities. The first author would also like to thank the Kano State Scholarships board for the PhD scholarship.

CONFLICT OF INTEREST

The authors declare no conflicts of interest

ABBREVIATIONS

ATR-IR

Attenuated Total Reflectance-Infrared

FTIR

Fourier Transform Infra-Red

MVDA

Multi Variate Data Analysis

DPPH

2,2′-diphenylpicryl hydrazyl

FRAP

Ferric Reducing Antioxidant power

PCA

Principal Component Analysis

HCA

Hierarchical Cluster Analysis

OPLS-DA

Orthogonal Partial Least Square–Discriminant Analysis

PLS

Partial Least Square

OPLS-DA-HCA

Orthogonal Partial Least Square– Discriminant Analysis-Hierarchical Cluster Analysis

REFERENCES

1 

Easmin S, Sarker MZI, Ghafoor K, Ferdosh S, Jaffri J, Ali ME, Mirhosseini H , authors. et al. Rapid investigation of α-glucosidase inhibitory activity of Phaleria macrocarpa extracts using FTIR-ATR based fingerprinting. J Food Drug Anal. 2016; 25 (2): 306 –15

2 

Fadzlillah NA, Rohman A, Ismail A, Mustafa S, Khatib A , authors. Application of FTIR-ATR spectroscopy coupled with multivariate analysis for rapid estimation of butter adulteration. J Oleo Sci. 2013; 62 (8): 555 –62

3 

Sanahuja AB, Moya MSP, Pérez SEM, Teruel NG, Carratalá MLM , authors. Classification of four almond cultivars using oil degradation parameters based on FTIR and GC data. JAOCS, J Am Oil Chem Soc. 2009; 86 (1): 51 –8

4 

Hanani SH, Ware I, Yaakob H, Mukrish H, Roji SM , authors. Antioxidant and anti cancer actvity of standardized extracts of three varieties of Ficus deltoidea’s leaves. J Teknol. 2015; 77 (3): 19 –25

5 

Woon SM, Seng YW, Ling APK, Chye SM, Koh RY , authors. Anti-adipogenic effects of extracts of Ficus deltoidea var. deltoidea and var. angustifolia on 3T3-L1 adipocytes. J Zhejiang Univ Sci B. 2014; 15 (3): 295 –302

6 

Lawal U, Maulidiani M, Shaari K, Ismail IS, Khatib A, Abas F , authors. Discrimination of Ipomoea aquatica cultivars and bioactivity correlations using NMR-based metabolomics approach. Plant Biosyst. 2016; 3504: 1 –11

7 

Mat N, Rosni NA, AB Rashid NZ, Nor ZM, Nudin NFH, Yunus AG , authors. et al. Leaf morphological variations and heterophylly in Ficus deltoidea jack (Moraceae). Sains Malaysiana. 2012; 41 (5): 527 –38

8 

Shimada K, Fujikawa K, Yahara K, Nakamura T , authors. Antioxidative properties of xanthan on the autoxidation of soybean oil in cyclodextrin emulsion. J. Agric. Food Chem. 1992; 40 (6): 945 –8

9 

Farsi E, Ahmad M, Hor SY, Ahamed MBK, Yam MF, Asmawi MZ , authors. Standardized extract of Ficus deltoidea stimulates insulin secretion and blocks hepatic glucose production by regulating the expression of glucose-metabolic genes in streptozitocin-induced diabetic rats. BMC Complement Altern Med. 2014; 14 (1): 220

10 

Benzie IFF, Strain JJ , authors. The ferric reducing ability of plasma (FRAP) as a measure of antioxidant power: The FRAP assay. Anal Biochem. 1996; 239 (1): 70 –6

11 

Thaipong K, Boonprakob U, Crosby K, Cisneros-Zevallos L, Hawkins BD , authors. Comparison of ABTS, DPPH, FRAP and ORAC assays for estimating antioxidant activity from guava fruit extracts. J Food Compos Anal. 2006; 19 (6-7): 669 –75

12 

Mohd KS, Azemin A, Rosli AS, Zakaria AJ, Ismail Z , authors. Comparison of chemical profile and biological activities of different plant parts of Ficus deltoidea Jack var. trengganuensis. Int J Pharmacogn Phytochem Res. 2015; 7 (2): 325 –32

13 

Kaya-Celiker H, Mallikarjunan PK, Kaaya A , authors. Characterization of Ininvation of genus Aspergillus on peanut seeds using FTIR-PAS. Food Anal Methods. 2016; 9 (1): 105 –13

14 

Dzolin S, Ahmad R, Mat Zain M, Ismail MI , authors. Flavonoid distribution in four varieties of Ficus deltoidea (Jack). J Med Plant Herb Ther Res. 2015; 3: 1 –9

15 

Misbah H, Aziz AA Aminudin N , authors. Antidiabetic and antioxidant properties of Ficus deltoidea fruit extracts and fractions. BMC Com and Alt Med. 2013; 13 (1): 118

16 

Abu BSA, Abas F, Ismail A, Khatib A , authors. Effect of different drying treatments and solvent ratios on phytochemical constituents of Ipomoea aquatica and correlation with α-glucosidase inhibitory activity. Int J Food Prop. 2016; 19 (12): 2817 –31

17 

García AV, Sanahuja AB, Garrigós SMDC , authors. Characterization and classification of almond cultivars by using Spectroscopic and thermal techniques. J Food Sci. 2013; 78 (2): 138 –44

18 

Fraser K, Lane GA, Otter DE, Hemar Y, Quek SY, Harrison SJ , authors. et al. Analysis of metabolic markers of tea origin by UHPLC and high resolution mass spectrometry. Food Res Int. 2013; 53 (2): 827 –35

GRAPHICAL ABSTRACT

https://s3-us-west-2.amazonaws.com/jourdata/pj/pharmacognj-10-6s-147_g004.jpg

SUMMARY

  • FTIR Metabolomics approach was used to understand the variability of syconia of seven varieties of Ficus deltoidea.

  • FTIR combine with chemometric successfully discriminated the seven varieties of Ficus deltoidea.

  • PLS models were developed and identified the correlation between samples and bioactivity and the fingerprints contributed.