A comparative study of the chemical composition, biological and
multivariate analysis of Crotalaria retusa L. stem barks, fruits, and
flowers
obtained via different extraction protocols
Kouadio Ibrahime Sinan
a,1, Lara Saftic
b,1,
Zeljka Persuric
b, Sandra Kraljevic Pavelic
b,
Ouattara Katinan Etienne
c, Marie Carene Nancy Picot-Allain
d,
Mohamad Fawzi Mahomoodally
d,*
, Gokhan Zengin
a,*
a
Department of Biology, Science Faculty, Selcuk University, Campus, Konya, Turkey
bDepartment of Biotechnology, Centre for High-Throughput Technologies, University of Rijeka, Radmile Matejcic2, 51000 Rijeka, Croatia cLaboratoire de Botanique, UFR Biosciences, Universite Felix Houphou€et-Boigny, Abidjan, C^ote d’Ivoire
dFaculty of Science, Department of Health Sciences, University of Mauritius, Reduit, Mauritius
A R T I C L E I N F O
Article History: Received 14 August 2019 Revised 25 September 2019 Accepted 27 October 2019 Available online 12 November 2019
A B S T R A C T
Crotalaria retusa L. (Fabaceae) also known as‘rattlebox’ has been used in traditional medicine for the
manage-ment of various human ailmanage-ments. The present study comparatively evaluated thea-amylase,a-glucosidase,
acetylcholinesterase, butyrylcholinesterase, and tyrosinase inhibitory activity, antioxidant properties, as well as phytochemical profiles of extracts of C. retusa (bark, fruits, and flowers) obtained by homogenization, macer-ation, ultrasonicmacer-ation, and Soxhlet extractions. Little variation was noted between the phytochemical profiles obtained by liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) of C. retusa same plant parts extracted by different methods while the different plant parts showed specific phytochemical finger-prints. For instance, myricetin was identified in C. retusa fruits only. Fruit and bark extracts possessed the high-est concentrations of quercetin and rutin, respectively. p-hydroxybenzoic acid, a phenolic derivative of benzoic acid, was identified in all C. retusa plant parts. Spectrophotometric determinations revealed that C. retusa bark
extracts have highest concentrations of phenolic andflavonoids. Besides, C. retusa bark extracts showed highest
antioxidant capacity. The extracts showed high inhibitory activity againsta-glucosidase (21.22 4.81 mmol
acarbose equivalent/g), acetylcholinesterase (8.71 8.26 mg galantamine equivalent/g), butyrylcholinesterase (4.16 2.36 mg galantamine equivalent/g), and tyrosinase (133.11 125.26 mg kojic acid equivalent/g). Multi-variate component analysis showed that the plant part was the main factor responsible for the observed vari-ability between the extracts. Data collected proved that C. retusa has the potential for the development of novel biopharmaceutical, nutraceutical, and cosmetical products.
© 2019 SAAB. Published by Elsevier B.V. All rights reserved.
Keywords: Rattlebox Enzyme Extraction techniques Antioxidant Mass spectrometry 1. Introduction
The type of extraction technique is one of the most important steps in the preparation of plant extracts for biopharmaceutical and chemical analysis. In this respect, several methods have been designed to extract bioactive phytochemicals. Extraction methods such as soxhlet and maceration are the most commonly used meth-ods and are considered as conventional methmeth-ods of extraction. How-ever, such conventional methods tend to bear some inherent disadvantages including more time and solvent consumption. In this
sense, recent extraction techniques have focused on the use of ultra-sonicator, homogenizer and microwave assisted techniques and are currently being advocated as green extraction techniques. Taken together, comparative studies on different extraction techniques are gaining interest in the scientific platform that would establish base-line date for further biomedical and biopharmaceutical analysis (Azmir et al., 2013;Belwal et al., 2018).
Crotalaria retusa L. (Fabaceae) is a botanical remedy widely distrib-uted in tropical and subtropical regions (Arag~ao et al., 2017). Also known as“rattlebox” (Aremu et al., 2012), different parts of C. retusa have been used in traditional medicine for the treatment of numerous human ailments (Ablasse et al., 2018). For instance, an infusion of the whole plant are used to treat skin infections, the roots were used to treat hemoptysis, the leaves andflowers are used to manage fever, cold, and lung disease, powdered from C. retusa seeds boiled in milk
* Corresponding authors.
E-mail addresses:f.mahomoodally@uom.ac.mu(M.F. Mahomoodally),
gokhanzengin@selcuk.edu.tr(G. Zengin).
1
These authors contributed equally. https://doi.org/10.1016/j.sajb.2019.10.019
0254-6299/© 2019 SAAB. Published by Elsevier B.V. All rights reserved.
Contents lists available atScienceDirect
South African Journal of Botany
was used in the management of leprosy, fever,flatulence as well as an analgesic in cases of scorpion stings and snake bites (Ablasse et al.,
2018;Arag~ao et al., 2017). Over the past decades, there have been sev-eral scientific studies reporting the biological activities of species belonging to the Crotalaria genus, including C. retusa. Seeds of C. retusa showed anti-inflammatory and antinociceptive properties which were related to the inhibitory effect on neutrophil migration of lectin pres-ent in albumins (Arag~ao et al., 2017). Ethanol extract of propagated C. retusa inhibited the growth of Pseudomonas aeruginosa (zone of inhibi-tion 38 mm) and scavenged 2,2-diphenyl-1-picrylhydrazyl (DPPH) (Fifty percent of free radical inhibition (IC50): 57.6
m
g/ml) (Devendra et al., 2012). Saponins, tannins, alkaloids, and sterols were identified in various parts of C. retusa plant (Anim et al., 2016). Extracts of C. retusa, especially the C. retusa stem extracts, exhibited cytotoxicity against Jurkat, MCF 7, and PC 3 cell lines, but were also toxic against normal human liver cells (Anim et al., 2016).Despite the numerous traditional uses of C. retusa in the manage-ment of a plethora of human ailmanage-ments, scientific studies reporting the pharmacological properties of this plant are limited. In addition, as far as our literature search could ascertain, no study has been carried out to determine the effect of different extraction methods on the bioactivity of C. retusa extracts. Besides, no attempt has been made in the analysis of the bioactivity of C. retusa bark, fruits, andflowers. In light of the above, this study entails establishing a valuable baseline data on the possible enzymes
a
-amylase,a
-glucosidase, acetylcho-linesterase, butyrylchoacetylcho-linesterase, and tyrosinase inhibitory proper-ties and antioxidant activity of C. retusa bark, fruits, and flowers obtained by homogenization, maceration, ultrasonication, and Soxh-let extraction. The phytochemical profiles of the different extracts were established by LC-MS/MS. In particular, the liquid chromatogra-phy-triple quadrupole mass spectrometry/mass spectrometry (LC-QQQ MS/MS) method was used for targeted quantification of phenols, while liquid chromatography quadrupole time-of-flight mass spec-trometry/mass spectrometry (LC-QTOF MS/MS) was used for profiling of the samples. Correlations between phytochemicals and the observed biological activities were analyzed. It is expected that data gathered from the presented study provide valuable scientific data on the potential use of C. retusa in the development of novel thera-peutics, nutraceuticals, and cosmetic raw materials and products. 2. Materials and methods2.1. Collection of plant material
Sampling of the plant species was done in Gb^ek^e Region (vallee du Bandama District) of Ivory Coast in the year 2018. Botanical authenti-cation of the plant was done by the botanist Dr. Ouattara Katinan Eti-enne (Laboratoire de Botanique, UFR Biosciences, Universite Felix Houphou€et-Boigny, Abidjan, Ivory Coast). The flowers, fruits, and stem barks were dried at room temperature (in shade, about 10 days). These materials were then powdered by using a laboratory mill.
2.2. Extraction techniques
Different extraction techniques were performed and the experi-mental details were provided below. Homogenizer assisted extraction: 2 g dried plant materials were extracted with 40 ml of methanol by using an Ultra-turrax at 6000 g for 5 min; maceration: 5 g dried plant materials were macerated in a shaker (SI-300 Benchtop Shaker) with 100 ml of methanol (HPLC grade, 99.9%) for 24 h; ultrasonication assisted extraction: 2 g dried plant materials were sonicated in one ultrasonic bath (Daihan, WUC D06H) with 40 ml of methanol for 1 h at 30 °C; Soxhlet extraction: 5 g dried plant materials were extracted with 100 ml of methanol for 5 h in a Soxhlet apparatus (Isolab). All
extracts werefiltered and concentrated by using a rotary-evaporator. The obtained plant extracts were kept at +4 °C until further analysis. 2.3. Profile of bioactive compounds
The content of four major groups of bioactive compounds (phe-nols,flavonoids, phenolic acids and flavonols) in obtained extracts was determined by using spectrophotometric (Thermo Scientific Multiskan GO) methods (Zengin and Aktumsek, 2014;Zengin et al., 2018). Expression of obtained results was achieved by equivalents of standards gallic acid (in case of total phenolic), rutin (in case of totalflavonoid), caffeic acid (in case of total phenolic acids) and cate-chin (in case of totalflavonols).
2.4. Determination of antioxidant and enzyme inhibitory effects For the comprehensive insights in bio-potential of obtained extracts and influence of extraction techniques on their bioactivity antioxidant, anti-
a
-amylase, anti-a
-glucosidase, cholinesterases, and anti-tyrosinase activities assays were performed. Estimation of anti-enzy-matic activity of the extracts was done by in vitro assays previously described byUysal et al. (2017). Measurement of extracts potential to be antioxidants and scavengers of free radicals was performed by ferric reducing antioxidant power (FRAP), 2,20 -azino-bis(3-ethylbenzothiazo-line-6-sulphonic acid (ABTS), cupric reducing antioxidant capacity (CUPRAC), 2,2-diphenyl-1-picrylhydrazyl (DPPH), phosphomolydenum and metal-chelating tests. A detailed description of applied assays was given previously (Uysal et al., 2017).2.5. Quantitative and qualitative MS analysis 2.5.1. Reagents and materials
Luteolin, naringenin, apigenin, p-coumaric acid, oleuropein, 2,5-dihydroxybenzoic acid (2,5-DHBA), chrysin, 3-hydroxytyrosol, 3,4-dihydroxybenzoic acid (3,4-DHBA), quercetin, caffeic acid, diosmetin, ellagic acid, rutin, 4-hydroxybenzoic acid (pHBA), sodium tri fluoroace-tate, 2,5-dihydroxybenzoic acid used as MALDI matrix (DHB), tetrahy-drofuran (THF) and MALDI standards bradykinin protein, angiotensin II, P14 R and triolein were provided from Sigma-Aldrich (St. Louis, MO, USA). Rhamnetin and myricetin were purchased from Extrasynthese (Genay, France). Verbascoside and apigenin-7-O-glucoside were obtained from HWI Analytik GMBH (R€ulzheim, Germany).
Acetonitrile (LC-MS grade) and methanol were supplied by Hon-eywell research chemicals (Bucharest, Romania), formic acid (LC-MS grade) was obtained by Sigma Aldrich (St. Louis, MO, USA), and water (LC-MS grade) was provided by Merck (Billerica, MA, USA)
For sample filtration, microfilters (chromafil cellulose acetate (0.45
m
m, 25 mm)) (Macherey-Nagel, Germany) were used.2.5.2. Sample preparation
Crotalaria retusa extracts (15 mg) were diluted in 2 ml of methanol and sonicated in the ultrasonic bath (Sonorex digitec, Bandelin, Ger-many) for 10 min at 25°C. Afterwards, samples werefiltrated and sub-jected to LC-QQQ analyses. For ESI-QTOF MS and MS/MS analysis, samples were additionally diluted 10 times in methanol and analyzed. 2.5.3. LC-QQQ MS/MS
All quantitative LC-MS/MS experiments were carried out on an Agi-lent 1260 series HPLC chromatograph equipped with a degasser, binary pump, auto-sampler and column oven coupled to an Agilent 6460 triple quadrupole mass spectrometer equipped with Jet Stream electrospray (AJS ESI) source (Agilent Technologies, Palo Alto, CA, USA). For chromatographic separation, Zorbax SB-C18, Rapid resolu-tion HT, 600 bar column (2.1 mm I.D£ 50 mm, 1.8
m
m, Agilent Tech-nologies, Palo Alto, CA, USA) was used, and all obtained results were processed in the Mass Hunter workstation software (version B.07.00).Method was published in our previous work (Saftic et al., 2019). All additional parameters of standard optimization are summarized in Table S1. Calibration curves were designed and analytical parameters including the limit of detection (LOD) and limit of quantification (LOQ) were calculated according to the international rules (ICH, 2005). 2.5.4. LC-QTOF MS/MS
For alkaloid profiling, an Agilent 1290 series UPLC equipped with a degasser, binary pump, autosampler and column oven coupled to an Agilent 6550 iFunnel quadrupole time-of-flight mass spectrome-ter equipped with dual AJS ESI source (Agilent Technologies) was used. QTOF MS scan and auto MS/MS scan methods were applied on all C. retusa extracts. Fragmentor was set at 300 V, and collision ener-gies were 0, 10, 20, 30, 40 and 50 eV. The tested extracts were ana-lyzed in positive ion mode in m/z range of 100 1200 for MS and MS/ MS scans. The MS parameters were set as in our previous work (Saftic et al., 2019).
Mass Hunter workstation software Qualitative analysis (version B.07.00) was used for data mining, and identification of some of the obtained alkaloids was done using Mass Hunter PCDL Manager (Ver-sion B.04.00). Mass Profiler Professional (version 13.0) was used for chemometric analysis of direct infusion ESI-QTOF MS data.
2.5.5. MALDI-TOF/MSfingerprint
For MALDI-TOF/MS fingerprints, an Bruker Ultraflextreme TOF mass spectrometer (Bruker Daltonik GmbH, Bremen, Germany) equipped with a 355 nm smartbeam II laser (Nd:YAG laser) technol-ogy was used.
DHB matrix, sodium trifluoroactetate, and C. retusa plant extracts were dissolved in THF (different concentration 1, 5 and 40 mg/ml). Afterwards, plant extracts were mixed with the DHB matrix and sodium trifluoroacetate THF solutions in the ratios of 2:1:1 (v/v/v) and the solutions were applied on the MALDI ground steel plate.
Bradykinin protein, angiotensin II and P14 R were used for device cali-bration. The analysis was carried out in the positive-ion reflection mode using the FlexControl software package (version 3.4 Bruker Dal-tonics). MS analysis parameters were; ion source 1 voltage, 20 kV; lens voltage, 7.3 kV; ion source 2 voltage, 17.9 kV; mass range, 200 1500 Da. Six independent subspectra (500 shots per subtrum) were manually collected and combined to form a single spec-trum for each C. retusa plant extract.
2.6. Statistical analysis
Statistical calculations were performed under R v 3.5.1 and Xlstat 2018 softwares environment. Firstly the normality of all data was evaluated by the Shapiro-Wilk test. According to the results of the normality test, One-way ANOVA with Tukey post-hoc test or Krus-kal Wallis with Dunn multiple pairwise comparison tests were con-ducted for comparisons among samples. Both the Spearman’s and Pearson correlation coefficients were calculated among total bioac-tive compounds and biological activities. Then biological activities dataset was analysed by multivariate methods i.e PCA and PLS-DA. 3. Results and discussion
Secondary metabolites present in the extracts of the bark,flowers, and fruits of C. retusa obtained by homogenization, maceration, ultra-sonication, and Soxhlet extractions were tentatively characterized and quantified by LC-QQQ MS/MS and ESI-QTOF MS/MS methods. Addi-tionally, chemicalfingerprints were obtained by MALDI-TOF/MS and ESI-QTOF MS. Significant differences between chemical profiles of dif-ferent plant parts were observed as shown in Figs. S1 3. These differ-ences between chemical profiles were further evaluated by high-resolution ESI-QTOF/MS analysis, and the results were processed by chemometric PCA method. PCA of raw ESI-QTOF data (Fig. 1)
Fig. 1. Distribution of samples in the space of principal component 1 and principal component 2 when ESI-QTOF spectra were used. Samples were divided into 3 categories. The tags used in the annotation are: red for Crotalaria retusa bark extracts, blue forflower, and brown for fruit extracts; triangle for maceration, square for homogenization, circle for sonica-tion, and rhomb for soxhlet extraction process. (For interpretation of the references to color in thisfigure legend, the reader is referred to the web version of this article.)
confirmed a high degree of variability among plant parts, whereas the first two components describe 60.49% of total variability. As expected, PCA projection of the total direct QTOF MS profiles showed grouping of extracts obtained by the different extraction procedures, but the same plant part, and was able to differentiate different plant parts regardless of the extraction process used for sample preparation.
Quantitative estimation of phenolic compounds in the bark, flow-ers, and fruits of C. retusa were analyzed by LC-QQQ MS/MS and are summarized inTable 1. Myricetin was identified in C. retusa fruits only, the highest concentration (0.0423 mg/g) was recorded in the Soxhlet extract. Likewise, C. retusa fruit extracts contained the high-est quercetin content, particularly the Soxhlet extract (0.1365 mg/g). C. retusa bark (0.0086 0.0185 mg/g) was rich in rutin compared to theflowers (0.0002 0.0003 mg/g) and fruits (0.0038 0.0091 mg/g). The bark extracts showed the highest luteolin content (Table 1), while the lowest concentration of luteolin was found in theflower extracts. C. retusa bark extract (0.0264 mg/g) obtained by ultrasonica-tion was rich in apigenin. Significant amounts of pHBA, a phenolic derivative of benzoic acid, were detected in all the studied C. retusa plant parts. On the other hand, 2,5-DHBA and 3,4-DHBA (also known as protocatechuic acid), derivatives of benzoic acid, were present in significant amounts in C. retusa bark extracts. Additionally, C. retusa flowers and fruits were rich in p-coumaric acid compared to the bark.
Table 2describes the alkaloid compositions of C. retusa bark, flow-ers, and fruits analysed by ESI-QTOF MS/MS. Previously published papers reporting the chemical characterization of Crotalaria species revealed high concentrations of different alkaloids, specially pyrroli-zidine (Flores et al., 2009;Nakka et al., 2013). In the present study, full ESI-QTOF MS/MS spectra was used and data obtained were com-pared with the literature and databases (Avula et al., 2015;Shim et al., 2013). Different alkaloids were identified from C. retusa metha-nolic extracts, and confirmed previous results on significant differen-ces between different plant parts. Monocrotaline and its N-oxide derivative, trichodesmine, morphine derivative, and isoquinoline derivatives were present in all the C. retusa plant parts. Floridanine (m/z 442.2060), producing fragment ions at m/z 122.0954 and 118.0851, was tentatively identified in the C. retusa bark extracts only. Another alkaloid compound (m/z 316.1377), was also identified in the bark only. The alkaloids having m/z values 352.1737, were ten-tatively characterized as serecionine N-oxide, usaramine, and retro-sine and were identified in C. retusa fruits only (Table 2). An unknown alkaloid (m/z 349.1588), producing fragment ions at m/z 292.1507, 195.9648, 154.0500, 138.0911, and 119.0715, was identi-fied in C. retusa flowers only.
Quantitative determination of phenolic,flavonoids, flavonols, and phenolic acids using spectrophotometric methods was also carried out. As presented inTable 3andFig. 2, ultrasonication was the pre-ferred method for the extraction of phenolics; homogenization for the extraction of flavonoids; and Soxhlet extraction for phenolic acids. Among the different C. retusa plant part investigated, the bark was rich in phenolics, phenolic acids, andflavonoids. C. retusa bark extract obtained by ultrasonication showed the highest phenolic con-tent (40.49 mg gallic acid equivalent (GAE)/g) while extract of the same plant part obtained by homogenization was rich inflavonoid (16.47 mg rutin equivalent (RE)/g). As shown inTable 5, the extrac-tion offlavonoids was enhanced by homogenization.
The antioxidant properties of the methanolic extracts of the dif-ferent plant parts of C. retusa were evaluated using standard in vitro spectrophotometric methods. In concordance with several studies, extracts of C. retusa possessing high phenolic andflavonoid contents were showed potent antioxidant properties (Dong et al., 2019;
Ozdal et al., 2019). The total antioxidant capacity of C. retusa extracts was assessed using the phosphomolybdenum assay. As shown inTable 4, C. retusa bark extracts (1.46 1.18 mmol trolox equivalent (TE)/g) showed the highest antioxidant capacity when tested using the phosphomolybdenum method. In addition, the Table
1 Quantitative analysis of phenolic compounds in Crotalaria retusa bark, fl ower and fruit methanolic extracts. Bark (mg/g) Flower (mg/g) Fruit (mg/g) Phenolic compounds Maceration Homogenisation Sonication Soxhlet Maceration Homogenisation Sonication Soxhlet Maceration Homogenisation Soni cation Soxhlet Apigenin 0.0202 § 0.0046 0.0169 § 0.0036 0.0264 § 0.0008 0.0196 § 0.0007 0.0041 § 0.0004 0.0039 § 0.0003 0.0027 § 0.0014 0.0031 § 0.0000 0.0130 § 0.0052 0.0198 § 0.0091 0.0205 § 0.0004 0.0203 § 0.0002 Apigenin-O-7-glucoside 0.0042 § 0.0003 0.0048 § 0.0007 0.0052 § 0.0001 0.0039 § 0.0001 0.0009 § 0.0000 0.0010 § 0.0001 0.0009 § 0.0002 0.0012 § 0.0002 0.0040 § 0.0009 0.0044 § 0.0008 0.0042 § 0.0002 0.0053 § 0.0006 Caffeic acid 0.0013 § 0.0002 0.0065 § 0.0001 0.0061 § 0.0006 0.0064 § 0.0003 0.0081 § 0.0006 0.0087 § 0.0011 0.0085 § 0.0005 0.0102 § 0.0003 0.0028 § 0.0025 0.0076 § 0.0001 0.0051 § 0.0001 0.0025 § 0.0009 Chrysin 0.0002 § 0.0003 0.0003 § 0.0004 0.0000 § 0.0000 0.0000 § 0.0000 0.0003 § 0.0004 0.0017 § 0.0004 0.0008 § 0.0001 0.0005 § 0.0007 0.0007 § 0.0001 0.0005 § 0.0007 < LOQ < LOQ Luteolin 0.0106 § 0.0062 0.0132 § 0.0024 0.0153 § 0.0002 0.0140 § 0.0002 0.0012 § 0.0001 0.0009 § 0.0002 0.0011 § 0.0000 0.0005 § 0.0001 0.0067 § 0.0004 0.0070 § 0.0005 0.0060 § 0.0006 0.0110 § 0.0007 Myricetin < LOQ a < LOQ < LOQ < LOQ < LOQ < LOQ < LOQ < LOQ 0.0230 § 0.0014 0.0174 § 0.0021 0.0169 § 0.0013 0.0423 § 0.0022 Naringenin 0.0118 § 0.0004 0.0092 § 0.0052 0.0157 § 0.0001 0.0083 § 0.0000 0.0002 § 0.0000 0.0002 § 0.0000 0.0002 § 0.0000 0.0000 § 0.0001 0.0966 § 0.0018 0.0795 § 0.0052 0.0854 § 0.0047 0.1677 § 0.0102 Oleuropein 0.0049 § 0.0041 0.0065 § 0.0010 0.0090 § 0.0006 0.0028 § 0.0020 0 0 0 0 0.0020 § 0.0002 0.0049 § 0.0004 0.0021 § 0.0001 0.0022 § 0.0005 Quercetin 0.0012 § 0.0002 0.0011 § 0.0002 0.0020 § 0.0002 0.0010 § 0.0000 0.0010 § 0.0002 0.0015 § 0.0001 0.0012 § 0.0001 0.0008 § 0.0000 0.0620 § 0.0043 0.0313 § 0.0004 0.0467 § 0.0023 0.1365 § 0.0117 Rhamnetin 0.0014 § 0.0000 0.0021 § 0.0000 0.0021 § 0.0000 0.0012 § 0.0000 0 0 0 0 0.0014 § 0.0001 0.0020 § 0.0000 0.0021 § 0.0000 0.0018 § 0.0001 Verbascoside 0.0174 § 0.0018 0.0124 § 0.0042 0.0225 § 0.0015 0.0127 § 0.0003 0 0 0 0 0.0050 § 0.0003 0.0070 § 0.0041 0.0049 § 0.0002 0.0046 § 0.0000 2,5-DHBA 0.0437 § 0.0018 0.0537 § 0.0006 0.0517 § 0.0006 0.0411 § 0.0042 0.0012 § 0.0000 0.0014 § 0.0003 0.0013 § 0.0004 0.0016 § 0.0001 0.0025 § 0.0001 0.0016 § 0.0004 0.0022 § 0.0012 0.0035 § 0.0011 3,4-DHBA 0.0471 § 0.0028 0.0546 § 0.0045 0.0502 § 0.0013 0.0263 § 0.0024 0.0049 § 0.0003 0.0051 § 0.0001 0.0048 § 0.0005 0.0049 § 0.0002 0.0094 § 0.0003 0.0058 § 0.0002 0.0053 § 0.0005 0.0089 § 0.0017 3-hydroxytyrosol 0.0246 § 0.0054 0.0195 § 0.0041 0.0060 § 0.0016 0.0055 § 0.0011 0 0.0010 § 0.0014 0 0 0.0057 § 0.0004 < LOQ 0.0021 § 0.0008 0.0045 § 0.0008 Ellagic acid 0.0476 § 0.0009 0.0360 § 0.0055 0.0652 § 0.0076 0.0179 § 0.0011 0.0012 § 0.0016 0.0041 § 0.0002 0.0031 § 0.0006 0.0041 § 0.0004 0.0035 § 0.0004 0.0084 § 0.0004 0.0074 § 0.0001 0.0075 § 0.0014 p -coumaric acid 0.0137 § 0.0020 0.0147 § 0.0026 0.0119 § 0.0013 0.0097 § 0.0011 0.0326 § 0.0048 0.0293 § 0.0123 0.0349 § 0.0024 0.0350 § 0.0019 0.0472 § 0.0046 0.0484 § 0.0035 0.0415 § 0.0046 0.0672 § 0.0043 p HBA 0.0663 § 0.0001 0.0772 § 0.0017 0.0651 § 0.0096 0.0574 § 0.0028 0.0386 § 0.0022 0.0416 § 0.0050 0.0405 § 0.0086 0.0240 § 0.0099 0.0469 § 0.0019 0.0243 § 0.0020 0.0349 § 0.0004 0.0424 § 0.0053 Diosmetin 0.0017 § 0.0003 0.0024 § 0.0002 0.0029 § 0.0006 0.0019 § 0.0006 0.0004 § 0.0000 0.0004 § 0.0000 0.0003 § 0.0001 0 0.0002 § 0.0000 0.0003 § 0.0000 < LOQ 0.0002 § 0.0000 Rutin 0.0143 § 0.0040 0.0157 § 0.0009 0.0185 § 0.0012 0.0086 § 0.0004 0.0003 § 0.0000 0.0002 § 0.0001 0.0003 § 0.0001 0 0.0056 § 0.0001 0.0091 § 0.0002 0.0038 § 0.0009 0.0042 § 0.0002 *Values are expressed as a mean value § SD (mg/g). a LOQ limit of quanti fi cation
Table 2
Alkaloids characteristic for Crotalaria retusa bark,flower and fruit methanolic extracts. The table lists the monoisotopic mass of alkaloid (M), molecular formula, precursor ion m/z, mass error (ppm) and fragment ions.
Tentative identification M Molecular formula m/z precursor Error/ ppm Fragment ions Bark Flower Fruit
Monocrotaline 325.1525 C16H23NO6 326.1595 0.97 194.1157, 121.0878, 120.0803 + + + Monocrotaline N-oxide 341.1475 C16H23NO7 342.1561 4.02 236.1269, 121.0854, 120.0802 + + + Trichodesmine 353.1838 C18H27NO6 354.1902 2.59 336.1770, 198.1105, 138.0989, 137.0829, 136.0744 + + + Floridanine 441.1999 C21H31NO9 442.2060 2.62 122.0954, 118.0851 + Mecambrine 295.1208 C18H17NO3 296.1264 2.08 278.1181, 260.0936, 212.0732 + + Morphine type 380.0770 C20H14NO7 381.0852 2.36 324.1422(M-CH2CHNHCH3), 236.1260, 153.0710, 137.0826, 120.0804 + + +
Isoquinoline type 341.1416 C23H19NO2 342.1500 3.36 296.1430(M-CH4-CH2O), 236.1255, 157.1075, 137.0828,
120.0801, 119.0720
+ + +
Dimer alkaloid (342.1558) 341.1416 C23H19NO2 683.2990 342.1508, 137.0838 + + +
Isoquinoline type 294.113 C18H17NO3 295.1200 1 259.0907(M-H2O), 211.0699, 194.0439, 152.0333,
137.0884, 108.0440 + + Unknown 321.0399 C21H8NO3 322.0474 7.69 290.0214, 260.0106, 227.9849, 120.0797 + + + Unknown 383.1580 C18H25NO8 384.1638 3.9 137.0828, 120.0806 + + Unknown 315.1318 C14H21NO7 316.1377 4.37 298.1269, 201.8788, 178.0839, 158.1159, 137.0800, 119.0488 + Unknown 348.1506 C11H26NO11 349.1588 2.69 292.1507, 195.9648, 154.0500, 138.0911, 119.0715 + Senecionine N-oxide 351.1682 C18H25NO6 352.1737 5.02 334.1700, 336.2000, 138.0905, 120.0812 + Usaramine 351.1682 C18H25NO6 352.1737 5.02 334.1700, 336.2000, 138.0905, 120.0812 + Retrosine 351.1682 C18H25NO6 352.1737 5.02 334.1700, 336.2000, 138.0905, 120.0812 + Table 3
Total bioactive components of Crotalaria retusa methanolic extracts.
Samples Total phenolic content (mg GAE/g) Totalflavonoid content (mg RE/g) Totalflavonol (mg CE/g) Total phenolic acid (mg CAE/g)
Bark-Homogenization 37.72§0.31abc 16.47§1.05a 0.52§0.01ab 4.46§0.41ab Bark-Maceration 35.09§0.32abc 9.13§0.19ab 0.47§0.08ab 3.92§0.20ab Bark-Ultra sonication 40.49§0.25a 12.21§0.31ab 0.53§0.01ab 5.94§0.15a Bark-Soxhlet 39.80§0.47ab 10.88§0.25ab 0.49§0.01ab 7.16§0.44a Flowers-Homogenization 25.77§0.18abc 7.06§0.26ab 0.99§0.02ab 1.02§0.08ab Flowers-Maceration 24.52§0.26abc 6.05§0.16b 0.45§0.02b 1.02§0.02ab
Flowers-Ultra sonication 25.87§0.22abc 6.97§0.24ab 0.50§0.01ab 0.68§0.06b
Flowers-Soxhlet 21.76§0.83c 6.93§0.11ab 0.50§0.01ab 1.10§0.02ab Fruits-Homogenization 27.02§0.21abc 15.17§0.21a 0.52§0.02ab 3.24§0.24ab Fruits-Maceration 22.47§0.20bc 8.69§0.19ab 0.47§0.01ab 1.53§0.20ab
Fruits-Ultra sonication 25.63§0.09abc 10.42§0.22ab 1.69§0.01a 2.74§.012ab
Fruits-Soxhlet 27.37§0.20abc 9.63§0.13ab 0.54§0.01ab 2.85§0.45ab
Values are mean§SD of three parallel measurements. GAE: Gallic acid equivalent; RE: Rutin equivalent; CE: Catechin equivalent; CAE: Caffeic acid equivalents. Different letters indicated significant differences in the tested extracts (p < 0.05, by ANOVA test).
Fig. 2. Total bioactive compounds in extracts of Crotalaria retusa and their relationship with the studied biological activities. (A) Total phenolic,flavonoids, phenolic acids and flavo-nols content. (B) Pearson’s correlation. (C) Spearman’s correlation.
bark extracts of C. retusa proved to be potent radical scavengers and reducing agents, among which bark extract obtained by Soxh-let extraction revealed to be the most active (46.12, 79.58, 149.77, and 86.34 mg TE/g, for DPPH, ABTS, CUPRAC, and FRAP, respec-tively). The flower extracts of C. retusa showed low antioxidant activity. The metal chelating properties of the different extracts are reported inTable 4. It was noted that the fruit extracts were the most active metal chelator, where the highest result was obtained from the fruit extract obtained from homogenization (48.95 mg ethylenediaminetetraacetate equivalent (EDTAE)/g). C. retusa fruit extract obtained by homogenization was rich in flavo-noids (15.17 mg RE/g) and phenolic acids (3.24 mg caffeic acid equivalent (CAE)/g). Detailed quantitative secondary metabolite analysis of C. retusa fruit extract obtained by homogenization revealed that the extract was rich in phenolic acids, namely, caffeic acid and ellagic acid, and flavonoids, namely, rutin and apigenin. Indeed, the metal chelating potential of these phytochemicals have been reported in several previous studies (Adjimani and Asare, 2015;Cherrak et al., 2016;Srivastava et al., 2007; Symono-wicz and Kolanek, 2012).
Enzyme inhibitors have been claimed to be potential therapeutic tools for the management of several human ailments. For instance, the inhibition of carbohydrate digesting enzymes such as
a
-amylase anda
-glucosidase, is one of the pharmacological approaches used to control diabetes type II (Mollica et al., 2019,2017;Picot et al., 2017;Thilagam et al., 2013). Interestingly, multiple phytochemicals ubiqui-tously present in plants have been reported to possess enzyme
inhibitory properties. It has been proposed that phytochemicals induce enzyme inhibition by binding to the orthosteric site thereby inducing conformation changes preventing substrate binding or binding to an allosteric site and preventing substrate interaction and product formation. In the present study, the ability of the C. retusa bark,flowers, and fruits extracts to inhibit enzymes targeted in the management of Alzheimer’s disease, epidermal hyperpigmentation conditions, and diabetes type II was evaluated and summarized in
Table 5. In general, the different extracts of C. retusa showed inhibi-tion against the studied enzymes with little variainhibi-tion. For instance, no significant (p < 0.05) difference was noted for the inhibition of acetylcholinesterase by the differentflower extracts. Likewise, the inhibitory action of C. retusa fruit extracts against amylase was not significantly (p < 0.05) different. In addition, it is noteworthy men-tioning that the observed enzyme inhibitory activity might be the result of the concerted action of several phytochemicals. In general, C. retusa extracts were more active inhibitors of acetylcholinesterase compared to butyrylcholinesterase. Scientific evidences highlight the role of acetylcholinesterase and butyrylcholinesterase in the patho-genesis of neurodegenerative conditions, such as Alzheimer’s disease. While the role of acetylcholinesterase in neurotransmission regula-tion is long-established and well-defined, the exact physiological role of butyrylcholinesterase in Alzheimer’s disease remains unclear. However, it has been reported that in late-stage Alzheimer’s disease, butyrylcholinesterase increased up to two-fold (Omar et al., 2017). Similarly, C. retusa extracts showed low inhibition against
a
-amylase (0.84 0.56 mmol ACAE/g) and higher activity againsta
-glucosidaseTable 4
Antioxidant properties of the tested methanolic extracts of Crotalaria retusa.
Samples DPPH (mg TE/g)a ABTS (mg TE/g)a CUPRAC (mg TE/g)a FRAP (mg TE/g)a Metal chelating (mg EDTAE/g)a Phosphomolybdenum (mmol TE/g)b Bark-Homogenization 30.89§0.54ab 61.19§0.63ab 135.82§1.41abc 78.71§1.51abc 28.86§0.16ab 1.46§0.12a
Bark-Maceration 33.25§1.90ab 62.70§3.08ab 130.16§0.51abc 83.91§1.51abc 24.40§0.77ab 1.18§0.12ab
Bark-Ultra sonication 43.51§2.75a 72.65§3.72a 139.71§1.74ab 87.63§0.90a 26.60§1.58ab 1.39§0.10a Bark-Soxhlet 46.12§0.86a 79.58§1.56a 149.77§3.26a 86.34§0.78ab 22.08§2.19ab 1.40§0.21a Flowers-Homogenization 7.67§1.97ab 15.71§0.74ab 66.86§0.64abc 35.64§0.23abc 24.30§1.41ab 0.82§0.17c Flowers-Maceration 6.04§0.92ab 17.70§2.20ab 65.47§0.79bc 34.79§0.70bc 17.22§1.52b 0.87§0.09bc
Flowers-Ultra sonication 7.79§1.53ab 18.69§1.06ab 70.05§0.89abc 35.96§0.74abc 26.53§0.14ab 1.19§0.07ab
Flowers-Soxhlet 5.67§0.36b 10.04§1.21b 60.90§0.39c 31.25§0.62c 21.56§0.61ab 0.96§0.10bc
Fruits-Homogenization 19.19§2.24ab 26.17§0.96ab 90.83§2.19abc 48.56§0.12abc 48.95§0.08a 1.11§0.11abc
Fruits-Maceration 6.18§0.41ab 16.27§3.14ab 80.35§0.32abc 40.03§0.30abc 41.68§1.59ab 0.78§0.10c
Fruits-Ultra sonication 14.46§0.99ab 24.13§0.89ab 79.14§0.89abc 44.33§1.36abc 43.72§1.39a 0.80§0.03c Fruits-Soxhlet 16.03§2.01ab 25.83§1.93ab 87.77§4.32abc 48.78§0.44abc 39.07§0.71ab 0.81§0.10c
Values are mean§SD of three paralel measurements. TE: Trolox equivalent; EDTA: EDTA equivalent; Different letters indicated significant differen-ces in the tested extracts (p< 0.05).
a
Statistical evaluation was done by Kruskal Wallis test.
b Statistical evaluation was done by ANOVA test.
Table 5
Enzyme inhibitory effects of the tested methanolic extracts of Crotalaria retusa.
Samples AChE inhibition
(mgGALAE/g)a BChE inhibition (mgGALAE/g)b Tyrosinase inhibition (mgKAE/g)b Amylase inhibition (mmolACAE/g)a Glucosidase inhibition (mgACAE/g)a
Bark-Homogenization 8.61§0.14ab 3.02§0.04def 129.25§1.00abcd 0.84§0.01a 20.71§0.05ab
Bark-Maceration 8.26§0.18b 2.36§0.36f 131.81§2.53ab 0.62§0.01ab 21.22§0.05a Bark-Ultra sonication 8.55§0.10ab 3.30§0.10bcde 130.52§0.36abc 0.68§0.03ab 15.23§2.25abc
Bark-Soxhlet 8.37§0.14b 3.24§0.54cde 127.07§1.18cd 0.58§0.06ab 15.98§3.47abc
Flowers-Homogenization
8.71§0.03ab 3.56§0.04abcd 125.60§0.79d 0.61§0.03ab 10.44§1.36abc
Flowers-Maceration 8.61§0.07ab 3.53§0.29abcde 125.26§0.66d 0.56§0.02b 8.37§1.11bc
Flowers-Ultra sonication 8.64§0.02ab 3.32§0.40abcde 128.69§2.11bcd 0.61§0.01ab 13.62§2.20abc
Flowers-Soxhlet 8.68§0.04ab 3.90§0.28abc 126.55§0.23cd 0.64§0.01ab 18.98§0.74abc
Fruits-Homogenization 8.65§0.04ab
2.70§0.09ef
132.20§1.36ab
0.64§0.01ab
4.81§0.65c
Fruits-Maceration 8.54§0.05ab 3.92§0.30abc 130.56§1.42abc 0.60§0.02ab 19.10§0.39abc
Fruits-Ultra sonication 8.76§0.06a 4.12§0.14ab 133.11§2.19a 0.63§0.02ab 11.04§1.90abc
Fruits-Soxhlet 8.64§0.05ab 4.16§0.34a 131.49§0.23ab 0.61§0.01ab 16.46§0.68abc
Values are mean§SD of three paralel measurements. AChE: acetylcholinesterase; BChE: butyrylcholinesterase; GALAE: Galatamine equivalent; KAE: Kojic acid equivalent; ACAE: Acarbose equivalent; Different letters indicated significant differences in the tested extracts (p < 0.05).
a
Statistical evaluation was done by Kruskal Wallis test.
b
(21.22 4.81 mmol ACAE/g). It is noteworthy mentioning that the bark extracts of C. retusa were rich in dihydroxybenzoic acid (DHBA) derivatives, namely 2,5 DHBA and 3,4 DHBA, which were previously reported to exhibit inhibitory activity on
a
-glucosidase (Abdullah et al., 2016;Das et al., 2017). A group of researchers investigated the interaction of 3,4 DHBA or protocatechuic acid witha
-glucosidase using in silico molecular docking and reported interaction at the active site involving hydrogen bonds, electrostatic covalent and van der Waals interactions (Erukainure et al., 2017). As shown inTable 5, potent inhibition of tyrosinase was noted. C. retusa fruit extracts (133.11 130.56 mg KAE/g) showed slightly higher tyrosinase inhibi-tory potential. Tyrosinase inhibitors have attracted considerable interest in the formulation and development of skin whitening and skin depigmenting products. Tyrosinase is a key regulatory enzyme responsible for the synthesis of melanin, a pigment protecting the human body from harmful UV lights (Mukherjee et al., 2018). How-ever, overproduction of melanin has been related to epidermal hyperpigmentation conditions, such as age spots, melasma, and freckles. Interestingly, myricetin, present in high amounts in C. retusa fruit extracts, has been reported to exhibit inhibitory activity against tyrosinase (Souza et al., 2012;Takekoshi et al., 2014;Zolghadri et al., 2019). Besides, the observed inhibitory activity upon tyrosinase might be due to the synergistic action of different phytochemicals possessing tyrosinase inhibitory activity. In this sense, a study pub-lished by Fan et al. (2017)demonstrated the tyrosinase inhibitory action quercetin, also identified in C. retusa fruit extracts, exerted on tyrosinase (Fan et al., 2017).3.1. Multivariate analysis of the biological activities data of C. retusa The analysis of a large number of extracts of C. retusa encompassing various biological activities justified the application of multivariate analysis methods. Thus, biological activities dataset was subjected to principal component analysis (PCA) in order to discriminate the sam-ples provided by four methods of extraction and three plant parts. Four components were deemed necessary to capture approximately 88.36% of the variance (Fig. 3). Thefirst two main principal compo-nents summarized 52.20% and 16.41% of the variance respectively. As observed in the factorial map (PC1 vs PC2) three majors groups were obtained and the segregation was more according to the plant part
(Fig. 3(A)). Thisfinding suggested that the plant part was the main fac-tor responsible for the variability observed between the extracts. Thereby, regardless of the plant part, the second studied factor (meth-ods of extraction) had a relatively small impact on the observed vari-ability. This was apparent in PCA sample plot, as the four extraction methods were not clearly differentiated at each plant part level.
The multivariate unsupervised principal component analysis showed that plant part influenced biological activity. In the next step, supervised approach partial least squares discriminant analysis (PLS-DA) was applied to provide the best discriminant biological activities responsible for the observed difference between samples. The factor (a type of parts) was used as class membership criteria.Fig. 3(C) displaying the samples plot of the scores for thefirst two-component, showed a clear separation of the three plant part, with an excellent prediction accuracy (Balance error rate = 0.001;Fig. 3(D)). In addition, the ROC plot showed that the model was high prediction accuracy for all samples (Fig. 3(E)). The group I comprised the bark extracts and was separated from the other two groups along thefirst component (Fig. 3(A) and (B)). The key biological activities that contributed to this segregation were exclusively those from the antioxidant assays (DPPH, ABTS, FRAP, CUPRAC) (VIP score>1.20) (Fig. 3(F)). The groups II and III representatives forflower and fruit extracts respectively, contributed to separation, and were sepa-rated along the second component; the metal chelating ability and tyros-inase inhibitory assays allowed that discrimination (Variable Importance in Projection (VIP) score>1.20) (Fig. 3(F)). Kruskal-wallis test also con-firmed that these biological activities were significantly different (p < 0.001) (Fig. 3(F)). PLS-DA model proved the PCA results and highlighted the impact of plant part on the evaluated biological activities of C. retusa. 4. Conclusion
Presented data provide comprehensive evidence on specific dif-ferences in phytochemical profiles of C. retusa bark, flowers, and fruits which are not correlated with the extraction methods tested in the presented paper. For instance, myricetin was identified in C. retusa fruits only. The method of extraction however, affected the phytochemicals profiles of the plant material and consequently the observed bioactivity. As such, C. retusa fruit extracts contained the highest quercetin content, but C. retusa fruit extract obtained by Soxhlet extraction had the highest amount of quercetin. Pyrrolizidine
Fig. 3. Multivariate statistical analysis carried out from the biological activities dataset of Crotalaria retusa. (A) Principal component analysis (PCA) sample plot of PC1 vs PC2, The data tend to group together by organ type. (B) PCA scree plot displaying the variation captured by each principal component. (C) PLS-DA sample plot. (D) The classification error rate for two number of PLS-DA components showing Balanced and Overall Error Rates. (E) The ROC (Receiver Operating Characteristic) curves assessing the prediction accuracy of a classification mode. (F) Variable Importance in Projection (VIP) score plot showing the most discriminating biological activities and Kruskal Wallis test on biological activities with VIP>1.2.
alkaloids, including monocrotaline andfloridanine, with previously documented cytotoxic effects, were also identified in the different plant parts of C. retusa. Therefore, exact quantification studies on these alkaloids in the separate parts of C. retusa would be essential in future cytotoxicity studies of C. retusa. On the other hand, C. retusa bark, fruits, andflowers can be further evaluated as potential candi-dates for the development of novel agents in the pharmaceutical and cosmetic sectors, especially focused on the management of Alz-heimer’s disease, diabetes type II, or hyperpigmentation problems. Declaration of Competing Interest
The authors have no conflict of interest to declare. CRediT authorship contribution statement
Kouadio Ibrahime Sinan: Conceptualization and Methodology. Lara Saftic: Data curation, Methodology, Writing - original draft and Writing - review & editing. Zeljka Persuric: Data curation, Methodol-ogy and Writing - original draft. Sandra Kraljevic Pavelic: Data cura-tion, Methodology and Writing - original draft. Ouattara Katinan Etienne: Conceptualization and Methodology. Marie Carene Nancy Picot-Allain: Writing - original draft and Writing - review & editing. Mohamad Fawzi Mahomoodally: Writing - original draft and Writ-ing - review & editWrit-ing. Gokhan Zengin: Supervision, Methodology, Writing - original draft and Writing - review & editing.
Acknowledgments
We greatly appreciate the granted access to equipment owned by the University of Rijeka within the project“Research Infrastructure for Campus-based Laboratories at University of Rijeka,” financed by the European Regional Development Fund (ERDF). We also acknowl-edge the University of Rijeka research support uniri-biomed-18-133. We would like to thank Croatian Government and the European Union (European Regional Development Fund—the Competitiveness and Cohesion Operational Programme KK.01.1.1.01) for funding this research through project Bioprospecting of the Adriatic Sea (KK.01.1.1.01.0002) granted to The Scientific Centre of Excellence for Marine Bioprospecting-BioProCro.
References
Abdullah, N.H., Salim, F., Ahmad, R., 2016. Chemical constituents of Malaysian U. cor-data var. ferruginea and their in vitroa-glucosidase inhibitory activities. Molecules 21, 525.
Ablasse, R., Ouedraogo, V., Karama, I., Compaore, M., Kiendrebeogo, M., 2018. Ethno-medicinal use of Crotalaria retusa L. (Fabaceae), a pyrrolizidine alkaloid toxic plant. International Journal of Biochemistry Research & Review 23, 1–6.
Adjimani, J.P., Asare, P., 2015. Antioxidant and free radical scavenging activity of iron chelators. Toxicology Reports 2, 721–728.
Anim, M.T., Larbie, C., Appiah-Opong, R., Tuffour, I., Owusu, K., Aning, A., 2016. Phyto-chemical, antioxidant and cytotoxicity of hydroethanolic extracts of Crotalaria retusa L. World Journal of Pharmaceutical Research 5990, 162–179.
Arag~ao, D.P., da Silva Souza, B., de Brito, T.V., de Araujo Bastos Santana, L., de Paiva Silva, R.M., de Oliveira, A.P., da Cunha Pereira, A.C.T., Ferreira, G.P., dos Reis Barbosa, A.L., de Oliveira, J.S., 2017. The anti-inflammatory and antinoci-ceptive activity of albumins from Crotalaria retusa seeds. Biomed. Pharmacother-apy 93, 536–542.
Aremu, M., Bamidele, T., Amokaha, J., 2012. Compositional studies of rattle box (Crota-laria retusa L.) seeds found in Nasarawa State, Nigeria, Pakistan. Journal of Nutri-tion 11, 880–885.
Avula, B., Sagi, S., Wang, Y.-H., Zweigenbaum, J., Wang, M., Khan, I.A., 2015. Characteri-zation and screening of pyrrolizidine alkaloids and N-oxides from botanicals and dietary supplements using UHPLC-high resolution mass spectrometry. Food Chem-istry 178, 136–148.
Azmir, J., Zaidul, I., Rahman, M., Sharif, K., Mohamed, A., Sahena, F., Jahurul, M., Ghafoor, K., Norulaini, N., Omar, A., 2013. Techniques for extraction of bioactive com-pounds from plant materials: a review. Journal of Food Engineering 117, 426–436. Belwal, T., Ezzat, S.M., Rastrelli, L., Bhatt, I.D., Daglia, M., Baldi, A., Devkota, H.P.,
Orhan, I.E., Patra, J.K., Das, G., 2018. A critical analysis of extraction techniques used for botanicals: trends, priorities, industrial uses and optimization strategies. Trends in Analytical Chemistry 100, 82–102.
Cherrak, S.A., Mokhtari-Soulimane, N., Berroukeche, F., Bensenane, B., Cherbonnel, A., Merzouk, H., Elhabiri, M., 2016. In vitro antioxidant versus metal ion chelating properties of flavonoids: a structure-activity investigation. PLoS One 11, e0165575.
Das, S., Acharya, J., De, B., 2017. Metabolite profiling, antioxidant activity, and glycosidase inhibition property of the mesocarp tissue extracts of sugar date palm [Phoenix syl-vestris (L.) Roxb.] fruits. International Journal of Food Properties 20, 2982–2993. Devendra, B.N., Srinivas, N., Solmon, K.S., 2012. A comparative pharmacological and
phytochemical analysis of in vivo & in vitro propagated Crotalaria species. Asian Pacific Journal of Tropical Medicine 5, 37–41.
Dong, X., Hu, Y., Li, Y., Zhou, Z., 2019. The maturity degree, phenolic compounds and antioxidant activity of Eureka lemon [Citrus limon (L.) Burm. f.]: A negative correla-tion between total phenolic content, antioxidant capacity and soluble solid con-tent. Scientia Horticulturae 243, 281–289.
Erukainure, O.L., Hafizur, R.M., Choudhary, M.I., Adhikari, A., Mesaik, A.M., Atolani, O., Banerjee, P., Preissner, R., Muhammad, A., Islam, M.S., 2017. Anti-diabetic effect of the ethyl acetate fraction of Clerodendrum volubile: protocatechuic acid suppresses phagocytic oxidative burst and modulates inflammatory cytokines. Biomedicine and Pharmacotherapy 86, 307–315.
Fan, M., Zhang, G., Hu, X., Xu, X., Gong, D., 2017. Quercetin as a tyrosinase inhibitor: Inhibitory activity, conformational change and mechanism. Food Research Interna-tional 100, 226–233.
Flores, A.S., de Azevedo Tozzi, A.M.G., Trigo, J.R., 2009. Pyrrolizidine alkaloid profiles in Crotalaria species from Brazil: Chemotaxonomic significance. Biochemical System-atics and Ecology 37, 459–469.
ICH, I. (2005). Q2 (R1): Validation of analytical procedures: text and methodology. In International Conference on Harmonization, Geneva.
Mollica, A., Zengin, G., Durdagi, S., Ekhteiari Salmas, R., Macedonio, G., Stefanucci, A., Dimmito, M.P., Novellino, E., 2019. Combinatorial peptide library screening for discov-ery of diversea-glucosidase inhibitors using molecular dynamics simulations and binary QSAR models. Journal of Biomolecular Structure and Dynamics 37, 726–740. Mollica, A., Zengin, G., Locatelli, M., Stefanucci, A., Macedonio, G., Bellagamba, G.,
Onaolapo, O., Onaolapo, A., Azeez, F., Ayileka, A., 2017. An assessment of the nutra-ceutical potential of Juglans regia L. leaf powder in diabetic rats. Food and Chemical Toxicology 107, 554–564.
Mukherjee, P.K., Biswas, R., Sharma, A., Banerjee, S., Biswas, S., Katiyar, C.K., 2018. Vali-dation of medicinal herbs for anti-tyrosinase potential. Journal of Herbal Medicine. 14, 1–16.
Nakka, S., Kumar, B.V., Devendra, B.N., 2013. Analysis of pyrrolizidine alkaloid from Crotalaria retusa L. Der Pharma Chemica 5, 6–11.
Omar, S.H., Scott, C.J., Hamlin, A.S., Obied, H.K., 2017. The protective role of plant bio-phenols in mechanisms of Alzheimer's disease. Journal of Nutritional Biochemistry 47, 1–20.
Ozdal, T., Ceylan, F.D., Eroglu, N., Kaplan, M., Olgun, E.O., Capanoglu, E., 2019. Investiga-tion of antioxidant capacity, bioaccessibility and LC-MS/MS phenolic profile of Turkish propolis. Food Research International 122, 528–536.
Picot, M.C., Zengin, G., Mollica, A., Stefanucci, A., Carradori, S., Mahomoodally, M., 2017. In vitro and in silico studies of mangiferin from Aphloia theiformis on key enzymes linked to diabetes type 2 and associated complications. Medicinal Chemistry 13, 633–640.
Saftic, L., Persuric, Z., Fornal, E., Pavlesic, T., Kraljevic Pavelic, S., 2019. Targeted and untargeted LC-MS polyphenolic profiling and chemometric analysis of propolis from different regions of Croatia. Journal of Pharmaceutical and Biomedical Analy-sis 165, 162–172.
Shim, H.J., Lee, J.Y., Kim, B., Hong, J., 2013. General fragmentations of alkaloids in electro-spray ionization tandem mass spectrometry. Mass Spectrometry Letters 4, 79–82. Souza, P.M., Elias, S.T., Simeoni, L.A., de Paula, J.E., Gomes, S.M., Guerra, E.N.S.,
Fonseca, Y.M., Silva, E.C., Silveira, D., Magalh~aes, P.O., 2012. Plants from Brazilian cerrado with potent tyrosinase inhibitory activity. Plos One 7, e48589.
Srivastava, A., Jagan Mohan Rao, L., Shivanandappa, T., 2007. Isolation of ellagic acid from the aqueous extract of the roots of Decalepis hamiltonii: Antioxidant activity and cytoprotective effect. Food Chemistry 103, 224–233.
Symonowicz, M., Kolanek, M., 2012. Flavonoids and their properties to form chelate complexes. Food Science and Biotechnology 76 (1), 35–41.
Takekoshi, S., Nagata, H., Kitatani, K., 2014. Flavonoids enhance melanogenesis in human melanoma cells. Tokai Journal of Experimental and Clinical Medicine 39, 116–121.
Thilagam, E., Parimaladevi, B., Kumarappan, C., Chandra Mandal, S., 2013.a -Glucosi-dase anda-amylase inhibitory activity of Senna surattensis. Journal of Acupuncture and Meridian Studies 6, 24–30.
Uysal, S., Zengin, G., Locatelli, M., Bahadori, M.B., Mocan, A., Bellagamba, G., De Luca, E., Mollica, A., Aktumsek, A., 2017. Cytotoxic and enzyme inhibitory potential of two Potentilla species (P. speciosa L. and P. reptans Willd.) and their chemical composi-tion. Front Pharmaceutical 8, 290.
Zengin, G., Aktumsek, A., 2014. Investigation of antioxidant potentials of solvent extracts from different anatomical parts of Asphodeline anatolica E. Tuzlaci: an endemic plant to Turkey. African Journal of Traditional, Complementary and Alter-native Medicines 11, 481–488.
Zengin, G., Llorent-Martínez, E.J., Fernandez-de Cordova, M.L., Bahadori, M.B., Mocan, A., Locatelli, M., Aktumsek, A., 2018. Chemical composition and biological activities of extracts from three Salvia species: S. blepharochlaena, S. euphratica var. leiocalycina, and S. verticillata subsp. amasiaca. Industrial Crops and Products 111, 11–21. Zolghadri, S., Bahrami, A., Hassan Khan, M.T., Munoz-Munoz, J., Garcia-Molina, F.,
Garcia-Canovas, F., Saboury, A.A., 2019. A comprehensive review on tyrosi-nase inhibitors. Journal of Enzyme Inhibition and Medicinal Chemistry 34, 279–309.
