The composition of potentially bioactive triterpenoid glycosides in red raspberry is influenced by tissue, extraction procedure and genotype

Gordon J. McDougall *a, J. William Allwood a, Gema Pereira-Caro b, Emma M. Brown c, Cheryl Latimer c, Gary Dobson a, Derek Stewart ad, Nigel G. Ternan c, Roger Lawther e, Gloria O'Connor e, Ian Rowland f, Alan Crozier g and Chris I. R. Gill c
aEnvironmental and Biochemical Sciences Group, The James Hutton Institute, Invergowrie, Dundee DD2 5DA, Scotland, UK. E-mail:; Fax: +44 (0)844928 5429; Tel: +44 (0)1382 568782
bPostharvest, Technology and Agrifood Industry Area, IFAPA, Córdoba, Spain
cNutrition Innovation Centre for Food and Health (NICHE), Centre for Molecular Biosciences, University of Ulster, Cromore Road, Coleraine, N. Ireland BT52 1SA, UK
dNIBIO, Norsk Institutt for Bioøkonomi, Pb 115, NO-1431 Ås, Norway
eAltnagelvin Area Hospital, Western Health and Social Care Trust, Glenshane Road, Londonderry, BT47 6SA, UK
fHugh Sinclair Unit of Human Nutrition, Department of Food and Nutritional Sciences, University of Reading, P.O. Box 226, Whiteknights, Reading, UK
gDepartment of Nutrition, University of California, Davis, California 95616, USA

Received 9th June 2017 , Accepted 24th August 2017

First published on 29th August 2017

The beneficial effects of consumption of berry fruits on a range of chronic diseases has been attributed (at least in part) to the presence of unique phytochemicals. Recently, we identified novel ursolic acid-based triterpenoid glycosides (TTPNs) in raspberry fruit and demonstrated their survival in human ileal fluids after feeding which confirmed their colon-availability in vivo. In this paper, in vitro digestion studies demonstrated that certain TTPNs were stable under gastrointestinal conditions and confirmed that these components may have been responsible for bioactivity noted in previous studies. Sequential extractions of raspberry puree, isolated seeds and unseeded puree showed that certain TTPN components (e.g. peak T1 m/z 679, and T2 m/z 1358) had different extractabilities in water/solvent mixes and were differentially associated with the seeds. Purified seed TTPNs (mainly T1 and T2) were shown to be anti-genotoxic in HT29 and CCD841 cell based in vitro colonocyte models. Further work confirmed that the seeds contained a wider range of TTPN-like components which were also differentially extractable in water/solvent mixes. This differential extractability could influence the TTPN composition and potential bioactivity of the extracts. There was considerable variation in total content of TTPNs (∼3-fold) and TTPN composition across 13 Rubus genotypes. Thus, TTPNs are likely to be present in raspberry juices and common extracts used for bioactivity studies and substantial variation exists in both content and composition due to genetics, tissue source or extraction conditions, which may all affect observed bioactivity.


Dietary guidance is consistent in recommending greater consumption of fruit and vegetables to promote health. A recent US report1 noted that higher fruit and vegetable intake was the only characteristic of dietary patterns to be consistently identified in all beneficial health outcomes. While the report does not recommend individual fruits, accumulating evidence suggests that the phytochemical composition of berries differentiates them from other fruits with their consumption having beneficial effects on a range of chronic diseases2–4 and their underlying pathophysiologies.5–7 It is, therefore, unsurprising phytochemicals in berries are of interest with regard to their bioactivity. Notably, berries accumulate high levels of (poly)phenols; for example, total phenol contents of 100–300 mg per 100 g fresh weight are commonly reported for strawberries, blackcurrants and raspberries.8 Berry (poly)phenol composition is strongly influenced by both environmental and genetic factors9 including cultivation methods, weather, ripeness at and time of harvesting,10 duration and conditions of storage11,12 as well as species and variety,13 all of which in consequence may influence bioactivity.

In addition to (poly)phenols, the presence of triterpenoids (TTPNs) in plants is also well established and they too have been ascribed various biological activities.14–16 TTPNs are abundant particularly in the skins of commonly consumed fruits such as apples, grapes, tomatoes and olives17 but they also occur in Prunus avium (sweet cherry) and Pyrus species (pear). TTPNs have been detected in berries such as cranberry and other Vaccinium species18 as well as sea buckthorn.19 These components are generally associated with the cuticular wax layers and tend to be hydrophobic components.20 The presence of cuticular TTPNs has been confirmed in leaves and flowers of common red raspberry cultivars.21 In raspberry and other Rubus species, a range of TTPNS have been identified22–25 but often from leaves, roots or other non-edible tissues. Evidence for potentially bioactive TTPNs in Rubus fruits is less common although they have been identified in blackberry26 and Rubus rosifolius.27 Recently, we identified novel TTPN glycosides from red raspberries after gastrointestinal digestion in vivo and confirmed that they would be colon-available following berry consumption.28 Thus, TTPN glycosides will come into direct contact with the colonic epithelium and so be potentially able to modulate risk factors for gut health including DNA damage, permeability and inflammation,29–31 as well as interacting with the gut microbiota.

This discovery has highlighted a knowledge gap regarding the presence of TTPN glycosides within key UK raspberry varieties. Such information could inform breeding strategies to increase TTPN levels in new enriched Rubus varieties as one route to improving consumption-related health benefits either as whole fruit or juice. Through re-examination of previous data-sets in the light of our discovery combined with new experiments, we report on the levels and diversity of novel putative TTPNs across a diverse range of important raspberry varieties and highlight variation due to tissue choice and extraction procedures on the potential bioactivities of raspberry extracts.

Materials and methods


The synthetic TTPN, 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid (CDDO) was purchased from Cambridge Bioscience Ltd, UK. HPLC grade ethanol, acetone and acetonitrile were purchased from VWR Ltd (Poole, UK). Formic acid (Optima™ LC/MS Grade) was purchased from Fisher Scientific Ltd (Loughborough, UK).

Raspberry puree and seed extractions

The bulk of studies were carried out on raspberries (Rubus idaeus var. Glen Ample; 12 kg) that were purchased from local farmers at harvest in 2013 and frozen in 500 g aliquots. Frozen raspberries (∼500 g) were thawed on ice then pureed in a Waring blender. A portion of the puree was sieved to remove seeds (1 mm size) and the unseeded puree collected. The seeds were washed with ice-cold distilled water then dried on paper towels. Seeded and unseeded purees (5 mL) and seeds (1 g fresh weight) were extracted with 10 mL of 0.1% aqueous formic acid for 30 min at 5 °C with end-over-end mixing in a blood rotator then centrifuged (2500g, 5 min at 5 °C) and the extracts removed to new tubes. The extraction was repeated with 0.1% aqueous formic acid then sequential extractions were carried with 80% acetonitrile containing 0.1% formic acid and finally with 10 mL 50% aqueous acetone. Extract aliquots (1 mL) were dried in a Speed-Vac, re-suspended in 5% aqueous acetonitrile containing 0.1% formic acid before LC-MSn analysis.

Preparation and sequential extraction of raspberry seeds

Eight kg of frozen raspberries (var. Glen Ample) were pureed in 250 g batches and the seeds separated (seed yield ∼4% w/w). Seeds (250 g) were extracted with 1 L of 0.1% aqueous formic acid in ultrapure water (UPW) for 60 min at 5 °C with orbital rotation at 90 rpm and the extract obtained by filtering through a glass sinter (porosity 3). The seeds were then extracted twice with 500 mL of 80% ethanol and finally with 500 mL of 50% aqueous acetone. These fractions were assayed for total phenol content and 1 mL aliquots of the water, ethanol and acetone fractions were dried and re-suspended as above for LC-MSn analysis.

Following this work, a TTPN-rich fraction (TRF; >95% purity by LC-MSn) was purified from raspberry seeds using the procedure outlined previously.28

Preparation of raspberry extract and in vitro digestion procedure

Phenolic-rich extracts of the berries were obtained using the in vitro digestion method of McDougall et al.32 as modified as described previously.33 Briefly, an extract (20 ml) was obtained by homogenising raspberries with an equal volume of solvent (2% (v/v) acetic acid in acetonitrile) in a Waring blender (full power, 3 × 15 s), the homogenate filtered through muslin, and the filtrate subjected to rotary evaporation to remove the solvent. This extract then underwent a two-stage in vitro procedure that simulates the digestive process. For gastric digestion, the extract (4 × 20 mL adjusted to total phenol content similar to the original extract) was adjusted to adjusted to pH 1.8 using 5 N HCl, 100 μL of pepsin (18.2 mg ml−1) was added and incubated at 37 °C for 2 h with orbital shaking at 80 rpm. For the pancreatic digestion, 4 mL of pancreatin/bile salts (4 mL of 4 mg ml−1 and 25 mg ml−1 respectively) was added then sufficient 0.5 M NaHCO3 was added to raise the pH to small intestine levels (assumed to be pH 7.4). The NaHCO3 was added in dialysis tubing to ensure that the pH was raised gradually. After 2 h incubation at 37 °C, the IVD samples were removed, aliquoted and stored at −20 °C.

Extraction of triterpenoid rich fractions (TRF) from fruit of a number of raspberry genotypes

Various Rubus idaeus (red raspberry) genotypes, a mixture of named Hutton cultivars (e.g. the Glen lines), other cultivars and some breeding lines, were grown at the James Hutton Institute in 2013 as part of the raspberry breeding consortia effort. A black raspberry (Rubus occidentalis) cultivar, Cumberland, which is present in the pedigrees of many red raspberry varieties,34 was also sampled. Fruit was harvested at full ripeness and frozen. Following freeze drying, the berries were milled to pass 0.5 mm sieves and the powders frozen prior to analysis.

Triplicate 100 mg aliquots of powders and extracted with 5 mL 50% aqueous acetonitrile containing 0.2% (v/v) formic acid. The sample was vortexed then extracted for 30 min on a blood rotator (100 rpm) at 5 °C. After centrifugation (2500g, 10 min, 5 °C), the supernatant was removed and the extraction repeated on the residue. The combined supernatants were combined and 1 mL aliquots frozen then dried in a SpeedVac. The dried extracts were re-suspended in 5% acetonitrile in 0.1% formic acid containing 25 μg mL−1 morin as an internal standard. The samples were subjected to LC-MSn analysis and the ratio of peak areas for selected components to the internal standard calculated.

LC-MSn analysis

LC-MSn analysis of most samples was performed on an HPLC system consisting of an Accela 600 quaternary pump, and an Acella PDA detector coupled to an LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Hemel Hempstead, UK) with accurate mass capabilities, operated under the Xcalibur software package as described previously.35 Some samples were also analysed using an LCQ-DECA ion trap mass spectrometer.35

The samples were prepared in filter vials and replicate 20 μL samples were injected onto a 2 × 150 mm (4 μm) Synergy Hydro-RP 80 fitted with a C18 4 × 2 mm Security Guard cartridge (Phenomenex Ltd). Autosample and column temperatures were maintained at 6 and 30 °C, respectively. The samples were analysed at a flow rate of 0.3 mL min−1 using a binary mobile phase of (A) 0.1% aqueous formic acid and (B) 0.1% formic acid in acetonitrile/water (1[thin space (1/6-em)]:[thin space (1/6-em)]1, v/v) and the following gradient: 0–4 min, 5% B; 4–22 min, 5–50% B; 22–32 min, 50–100% B. Mass detection was carried out in both negative and positive ESI modes. Two scan events were employed; full-scan analysis was followed by data-dependent MS/MS of the three most intense ions using collision energies of 45 eV source voltage (set at 3.4 kV) in wide-band activation mode. Prior to analysis, the mass accuracy of the Orbitrap XL instrument was assured by calibration following the manufacturer's protocols.

Tissue culture

Human colon cell lines HT29 (from a colon adenocarcinoma in a Caucasian female) and CCD 841 CoN (isolated from normal colon tissue in a Caucasian female)28 were acquired from European Collection of Cell Cultures and American Type Culture Collection respectively. HT29 cells were cultured in Dulbecco's Minimum Essential Medium (DMEM) supplemented with 10% FBS and 100 U l−1 penicillin/streptomycin. CCD 841 CoN cells were maintained in DMEM supplemented with 10% FBS, 100 U L−1 penicillin/streptomycin, 1% sodium pyruvate, 1% Non-Essential Amino Acids (NEAA) and were used between passage 15–25. Both cell lines were incubated at 37 °C with 5% CO2 and grown as monolayers in large roux flasks. Cells were sub-cultured every 3–4 days by the addition of trypsin (0.25% trypsin-EDTA) at 37 °C for 5 min. Cells were centrifuged at 1200 rpm for 3 min, the supernatant decanted and cells re-suspended in the appropriate medium. Both cell lines were treated with either 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid (CDDO), or the TTPN-rich fraction (TRF) purified from raspberry seeds. Both CDDO and TRF were prepared fresh in the respective cell culture media at a final concentration of 100 nM for use in the in vitro experiments. 100 nM was selected as a physiologically relevant concentration to test bioactivity (cytotoxicity, genotoxicity and antigenotoxicity) for the TRF based upon levels of the compound detected in ileal fluid following consumption of 300 g of raspberries by ileostomates as described by McDougall and co-workers.35 CDDO was tested at the equivalent concentration to serve as a comparator.

Cytotoxicity and genotoxicity of triterpenoids

Cytotoxicity assay. The effects of CDDO and raspberry TTPN extracts on the viability of HT29 and CCD841 cells were determined using the widely utilised 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.36 The assay is based on the ability of living cells to metabolize the tetrazolium salt, MTT by mitochondrial activity to formazan, a blue dye36 which can be measured spectrophotometrically. Cells were seeded in 96 multi-well plates (Costar, Cambridge, MA, USA) at a concentration of 1.5 × 104 HT29 and 3.0 × 104 CCD 841 cells per well respectively. After 48 h incubation at 37 °C, media was replaced with 100 nM of either TRF or CDDO in media then incubated for 24 h. The wells were washed and cells incubated for a further 48 h in fresh media. Thereafter 15 μL of MTT was added to each well. After 4 h, lysis was carried out with 100 μL solubilizing solution to free the product formazan. Formazan was measured using a microtiter plate reader (Alpha, SLT Rainbow Thermo, Antrim, UK) at a wavelength of 560 nm. The survival of the cells in cell media only was set as 100% viability. Each treatment was performed in octuplicate and the experiment was carried out on 3 separate occasions.
Genotoxicity assay. The COMET assay was performed using the well-established HT29 adenocarcinoma cell model for colonic DNA damage37 and the normal colonocyte cell line CCD841 CoN. In brief, both cell lines were incubated for 24 h with 100 nM of either TRF or CDDO. To assess the anti-genotoxic potential of the treatments, the cells were treated with hydrogen peroxide (75 μM, H2O2 for HT29 and 25 μM H2O2 for CCD 841) for 5 min at 4 °C, then centrifuged for 5 min at 258g. The cell pellet was re-suspended in 85 μL of 0.85% low melting point agarose (LMPA) in PBS and maintained in a water bath at 40 °C. The suspension was added to previously prepared gels (1% normal agarose) on frosted slides and coverslips were added. Slides were subsequently immersed in lysis buffer (2.5 M NaCl, 100 mM Na2EDTA, 10 mM TRIS, pH 10) for 1 h at 4 °C then placed in electrophoresis buffer and allowed to unwind for 20 min before running at 26 V (300 mA) for 20 min. After electrophoresis, gels were washed thrice in neutralisation buffer (0.4 M Tris, pH 7.5) at 4 °C. The percentage tail DNA was recorded using Komet 5.0 image analysis software (Kinetic Imaging Ltd, Liverpool, UK). For each slide, 50 cells were scored. Data represents the mean percent tail DNA of triplicate gels per treatment from three independent experiments. To assess the genotoxic potential of the compounds, cells were treated as above omitting the H2O2 challenge. Positive (H2O2) and negative controls (PBS) were included in all experiments (cells without CDDO/TRF pre-treatment).

Statistical analysis

The mean of each data set was used for statistical analysis and experiments were carried out as independent triplicates. The Shapiro–Wilk test was used to test for normality. Analysis of variance was applied to test for significant differences between means compared to control using Dunnett T post-hoc test. Significance was accepted at p < 0.05. Analysis was carried out using SPSS (version 20 for Windows).

Results and discussion

Previously we reported a novel compound with a characteristic [M − H] signal at m/z 679 (peak T1; see ESI Fig. S1) in ileal fluids from 11 volunteers after raspberry supplementation.35 The m/z 679 signal had 0.5 amu variants indicating a doubly charged entity with a true mass of ∼1360. Compounds that were good matches for the predicted formula (MW C36H56O12) obtained for [M − H]m/z 679 were isomers of ursolic acid-based TTPN glucosides that differed in the position of attachment of glucose, hydroxyl or methyl groups. TTPNs have been identified in both raspberry leaves and flowers, including in commercial varieties of Rubus ideaus.21 Ursolic acid-based TTPN glycosides have been noted in leaves of R. coreanus, for example, suavissimoside R1.38 Initial 1H NMR spectra analysis was consistent with the T1 peak being an ursolic acid-based TTPN.28

In the light of the discovery of this putative TTPN, we re-examined the composition of raspberry extracts (cultivar Glen Ample) that had previously exhibited protective effects on colonocyte cell models.33,39 The raspberry extract contained TTPN peak T1 and another five putative TTPN components (Fig. 1; Table 1). This diversity of putative TTPNs was not noted in our earliest studies of ileal fluids,35 although one of these putative TTPNs (T2, peak 16) was identified in pilot studies of TTPN composition in raspberry seeds28 and may, in fact, be a dimer of peak T1. These TTPN-like components share important common attributes; they have little discernible UV absorbance above that of the acetonitrile solvent but have strong MS signals, particularly in negative mode. Many present as formate adducts and on fragmentation show loss of 46 amu (formate) followed by loss of 162 amu (hexose) to form the putative TTPN aglycone (Table 1). The initial analysis was carried out using an ion-trap MS which lacked the ability to obtain exact mass data, and thus molecular formula data was not obtained. TTPN glycosides that share many of these mass spectral properties have been noted previously.40 The putative components with apparent aglycones at m/z 503 (peak 12), m/z 501 (peaks 13 and 14) and 487 (peak 15) are 14, 16 and 30 amu respectively smaller than the aglycone for peak 11 (T1) at m/z 517 and these differences could be caused by a “loss” of a methyl group (e.g. CH3 to H, 14 amu) or one oxygen (OH to H, 16 amu) or a methoxy group (CH3O to H, 30 amu). “Loss” of these groups would alter the polarity of the components and thus change their chromatographic behaviour.

image file: c7fo00846e-f1.tif
Fig. 1 UV and MS traces of raspberry extract. Parentheses denote main anthocyanins peaks. Peaks with blue numbers are putative triterpenoid glycosides. These are present at similar MS intensity to many of the phenolic components. The putative identities are shown in Table 1.
Table 1 Major peaks noted in raspberry extract
Peaka RT PDA [M − H] (m/z) MS2[thin space (1/6-em)]b Putative identity
a Peaks with blue numbers are putative triterpenoid glycosides. b The figures in bold are the major MS2 fragments. Analysis carried out on the LCQ-FLEET ion-trap MS system that lacks accurate mass capabilities.
1 17.26 260–280 1567 1265, 1103, 933, 897, 631 Sanguiin H10
2 17.78 350 771 609, 591, 301 Quercetin glucosyl rutinoside
3 18.54 240–260 1401 1869, 1567, 1250, 1099, 934, 897 Lambertinian C
4 18.92 240–260 1869 1567, 1235, 1103, 934, 933 Sanguiin H6
5 19.81 360 433 301 Ellagic acid pentose
6 20.13 360 433 301 Ellagic acid pentose
7 20.92 355 463 301 Quecetin hexoside
477 301 Ellagic acid glucuronate
8 22.13 360 447 315 Methyl ellagic acid pentose
9 23.18 315 475 301 UK
10 24.13 280–310 677 645, 627, 520 UK
11 24.50 679 517, 455, 437 TTPN peak T1
12 25.45 711 665, 503 TTPN
13 26.01 709 663, 501 TTPN (709a)
14 26.97 709 663, 501 TTPN (709b)
15 29.14 695 649, 487 TTPN
16 30.17 1358 679, 517, 499, 455 TTPN peak T2

In earlier work, we subjected raspberry extracts to an in vitro digestion procedure that simulated gastrointestinal conditions and demonstrated that the IVD samples retained protective effects on cellular models of colon cancer.33 We examined these IVD extracts for the stability of putative TTPN components. It was notable that peak T1 (m/z 679) was quite stable under in vitro digestion conditions (Fig. 2). This experiment had focused on polyphenolic components33 and the extracts were dried to equivalent phenol content prior to LC-MSn analysis or application to cell models, which amounted to a ∼2-fold concentration. Nevertheless, the presence of TTPN components after in vitro digestion was confirmed and suggested that the protective effects noted on colonocyte cell models may have been partly due to the presence of the novel putative TTPN components, possibly acting in concert with polyphenol constituents. The low levels of recovery of the other TTPNs may be due to degradation under the mildly alkaline pancreatic conditions, or as result of binding to proteins. The higher recovery of the m/z 679 TTPN may be as a result of degradation of the m/z 1358 TTPN as it is hypothesised to be an ester-linked dimer akin to Coreanoside F141 which could be degraded under gut conditions to two monomers at m/z 679. Indeed, our original work on ileostomy subjects35 showed that peak T1, but not peak T2, was apparent in ileal fluids and thus would be available to enter the colon. The two putative TTPN components with m/z 709 (peaks 13 and 14) exhibit differing stabilities and it would be intriguing to examine if these possibly isomeric components can be interconverted. Further work on purified components under simulated gastrointestinal conditions would be required to dissect out these possibilities.

image file: c7fo00846e-f2.tif
Fig. 2 Effect of in vitro digestion (IVD) on triterpenoid composition in raspberry extracts. Data are average MS peak area for triplicate samples ± SE.

Previous work suggested that the seeds may be a particularly rich source of the TTPNs (especially T1 and T2).28 Sequential water, acetonitrile and acetone extractions of raspberry puree, isolated seeds and “unseeded” puree showed that peaks T1 and T2 were present in both whole and unseeded purees but with higher levels in the whole puree and seeds. For peak T1, the whole puree appears to be a better source than the unseeded puree. Indeed, the yield of both peaks T1 and T2 was ∼1.7 fold higher in the whole rather than in the unseeded puree (see insert, Fig. 3). In both the whole puree and the seeds, extraction with acetonitrile improved the yield of these TTPNs but the proportion of peak T2 was markedly increased by use of acetonitrile, and also acetone. This strongly suggests that TTPN T2 is more tightly associated with, or bound to, the seed surface and perhaps more hydrophobic than T1.

image file: c7fo00846e-f3.tif
Fig. 3 Distribution of TTPN peaks T1 and T2 in sequential extracts of whole raspberry puree, unseeded puree and isolated raspberry seeds. (Insert shows whole and unseeded puree at a different scale)

We further examined the composition of TTPNs and their extractability from seeds using similar sequential water to solvent procedures (see ESI, Fig. S2). In this case, we used ethanol rather than acetonitrile extraction as ethanol can be used for food grade work. A total of seven peaks with MS properties ascribable to putative TTPN glycosides (peaks 18–24; Fig. 4 and Table 2) were noted upon LC-MSn analysis of the extracts. Some of these had been noted in earlier studies28 and some occurred in the raspberry extracts (Table 1). The main differences were that raspberry seeds contained a putative TTPN with m/z 741 but lacked the TTPN with m/z 711 found in the whole raspberry extract (Table 1). In addition, there was a minor peak (peak 24; Table 2) with m/z 1387 that was only present in the ethanol and acetone extracts of the seeds.

image file: c7fo00846e-f4.tif
Fig. 4 Effect of sequential extraction conditions on polyphenol and triterpenoid composition from raspberry seeds. Peaks are putatively identified in Table 1. Peaks A–C were more abundant in the water extracts.
Table 2 Mass spectroscopic properties of selected peaks from raspberry seeds
Peak RT [M − H] (m/z) MS2 Predicted formula Putative identityb
All predicted formula derived at <2 Δ ppm mass accuracy data.a [M − H]2− ion, has 0.50 amu variants. Ef = epiafzelechin; EC = epicatechin. Bold MS2 fragments = most abundant signals. Putative triterpenoid peaks are shaded.b Godevac et al., (2009);48 McDougall et al., (2014).35c Perret et al., (1999).40
1 2.19 191.0128 173, 111 C6H7O7 Citric acid
2 8.03 181.0442 151, 133 C9H9O4 UK
3 8.95 783.0402 481, 301 C34H23O22 Peduncalagin-like ellagitannin
4 10.81 783.0405 481, 301 C34H23O22 Peduncalagin-like ellagitannin
5 11.60 577.1143 451, 425, 407, 289 C30H25O12 EC dimer
6 12.55 577.1145 451, 425, 407, 289 C30H25O12 EC dimer
7 13.10 121.0254 93 C7H5O2 Benzoic acid
8 13.38 289.0612 245, 205 C15H13O6 Epicatechin (EC)
9 13.69 561.1202 543, 407, 289 C30H25O11 EfEC dimer
10 14.33 849.1727 723, 561, 407, 289 C45H37O17 EfECEC trimer
11 14.51 934.0376 1567, 1250, 897, 633, 301 None Sanguiin H-6
12 15.16 833.1780 815, 707, 561, 289 C45H37O16 EfEfEC trimer
13 15.88 1121.2309 1103, 1023, 951, 833, 561 C60H49O22 EfECEfEC tetramer
14 16.23 833.1784 815, 707, 561, 289 C45H37O16 EfEfEC trimer
15 16.69 1105.2369 951, 833, 815, 707, 561 C60H49O21 EfEfEfEC tetramer
16 16.96 147.0400 119, 85 C9H7O2 Cinnamic acid
17 17.36 313.0969 295, 283, 253 C14H17O8 Unknown
18 18.87 741.3436 695, 579, 451 C38H59O14 Formate adduct of triterpenoid glycosidec
19 (T1) 19.38 679.3451 559, 541, 517, 499 C36H55O12 Triterpenoid glycoside
20 20.85 709.3550 663, 501 C37H57O13 Formate adduct of triterpenoid glycosidec
21 22.91 695.3763 649, 487 C35H53O11 Formate adduct of triterpenoid glycoside
22 23.18 709.3553 663, 501 C37H57O13 Formate adduct of triterpenoid glycosidec
23 (T2) 24.00 1358.6842 1313, 1019, 679, 517, 455 None Triterpenoid glycoside dimer
24 24.43 1387.6913 1369 None Triterpenoid glycoside dimer

TTPNs from commonly eaten fruits other than berries (e.g. apple, grape and tomatoes) are associated with the skins and tend to be mainly hydrophobic in nature.17 They are largely non-glycosylated and their hydrophobicity and low solubility affects their extractability and may limit their bioavailability and possible effectiveness in vivo.42 Indeed, improving the drug-like properties of these TTPN aglycones such as ursolic, oleanolic and betulinic acids through nanotechnological approaches is an active area of research.43,44 The relative ease of extraction of the TTPN glycosides from raspberry suggests that these largely hydrophilic components would be readily bioaccessible.30

The choice of solvent affected both yield and composition of putative TTPNs, in addition to composition and yield of the phenolic components. For example, peak T1 (peak 19; m/z 679) was readily extracted using UPW but extraction of T2 (peak 23) was much improved by ethanol (Fig. 3). The extraction of most of the other putative TTPNs was improved by use of ethanol, and yet further with acetone. Some TTPNs (peaks 18 & 24) were much more apparent in the acetone extracts, suggesting that they may be particularly hydrophobic and tightly bound to the seed surface. The extraction was carried out on intact seeds and little or no damage to seeds was noted, and thus we conclude that the extracted components must be available on the surface of the seeds. The viability of seeds was not tested but considering that raspberry seeds are routinely scarified with 95% sulphuric acid prior to germination,45 it is likely that the alcohol-extracted seeds, at least, would be unaffected. Indeed, the presence of phenolic components, such as proanthocyanidins, has been implicated in seed coat permeability and the control of germination in Rubus species.46 The role of TTPNs in seed coat biology is less well defined but they may provide a protective antimicrobial activity.47

The phenolic components from seeds were mainly composed of proanthocyanidins (of both procyanidin and propelargonidin types) and ellagitannins,35 which have been noted previously.48 Most phenolic components were more effectively extracted using ethanol rather than UPW, apart from the propelargonindin dimer (peak 9; m/z 561) which was readily extracted using UPW. The levels of some proanthocyanidins and ellagitannins (e.g. peaks 6, 10, 11 & 12) were further improved by extraction with acetone.

Therefore, a balance can be struck between the extractability of TTPNs and extraction of phenolic components from seeds. In our previous work,28 use of low ethanol concentrations (i.e. 10% v/v) gave effective extraction of the major TTPNs but minimised extraction of phenolics and thereby simplified purification. Neither the TTPN-rich fraction purified from seeds (TRF; >95% of peaks T1 and T2) nor 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid were significantly cytotoxic in either of the cell lines tested compared to the control (Table 3) (e.g. HT29 cells ABS 1.5 ± 0.08, CCD 841 ABS 0.92 ± 0.04). TRF reduced H2O2-induced DNA damage to HT29 adenocarcinoma cells and also to the normal colonocyte cell line CCD 841 CoN (Table 4). At 100 nM, both the synthetic TTPN 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid and TRF exerted a significant anti-genotoxic effect against H2O2 challenge in the two cell lines (Table 4). With HT 29 cells, both treatments reduced tail DNA by ∼40% compared to the untreated control, whilst in CCD 841 CoN cells, DNA damage was reduced by ∼50%. Neither TRF nor 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid were genotoxic in either of the cell lines tested and efficacy between cell lines was not significantly different. This data confirms that these TTPNs had protective effects in both normal and adenocarcinoma cell lines, and furthermore at levels likely to be physiologically achievable given TTPN survival under both simulated in vitro digestion and in ileal fluids in vivo.28 Similarly albeit at 50-fold higher concentration of 5 μM, the triterpenoid ursolic acid was reported to decrease H2O2-induced DNA damage in the colonic cell line Caco-2 by >40%.49 Triterpenoids from Rubus rosifolius fruits have also previously been shown to have beneficial effects against human colon cancer HCT-116 cells;27 triterpenoids from Scoparia dulcis roots exert anti-mutagenic activity50 and triterpenoid-rich fractions from Korean raspberries have potent anti-inflammatory effects.51

Table 3 The cytotoxic effects of triterpenoid rich fraction (TRF) and the synthetic triterpenoid CDDO after 24 h pre-incubation on HT29 cells and CCD 841 cells
  HT29 cells CCD841 cells
Treatment % Cell viability S.D. % Cell viability S.D.
Data is presented as mean of 3 independent experiments + SD. Untreated control normalised to 100%. Data is presented as mean of 3 independent experiments + SD. One-way ANOVA and Dunnett T test, * p < 0.05, significance is compared to untreated control.
Untreated control 100.0 100.0
100 nM CDDO 94.1 3.7 89.6 7.0
100 nM TRF 88.8 3.2 92.6 5.4

Table 4 The anti-genotoxic/genotoxic effects of triterpenoid rich fraction (TRF) and the synthetic triterpenoid CDDO after 24 h pre-incubation on HT29 cells and CCD 841 cells challenged with H2O2 (75 μM HT29, 25 μM CCD841)
  HT29 cells CCD841 cells
Treatment % Tail DNA S.D. % Tail DNA S.D.
Data is presented as mean of 3 independent experiments + SD. One-way ANOVA and Dunnett T test, * p < 0.05, significance is compared to untreated control.
Untreated control (+H2O2) 53.0 3.4 49.9 0.4
100 nM CDDO (+H2O2) 28.1* 8.6 22.2* 3.2
100 nM TRF (+H2O2) 30.2* 3.7 26.3* 2.8
Untreated control 4.4 0.9 3.3 1.2
100 nM CDDO 5.8 3.6 3.7 0.8
100 nM TRF 5.1 3.4 3.4 1.3

Assessment of diversity of TTPNs in raspberry varieties

Six putative TTPN peaks (at m/z 679 (T1), 1358 (T2), 695, 709a, 709b & 711) were consistently noted in extracts from the various Rubus genotypes (see ESI; Fig. S3; Table S1). These match those noted in the raspberry extract from Glen Ample (Table 1). Most of these peaks were also observed in the analysis of seed extracts (Table 2) but the differences reflect the different source material (whole fruit vs. seeds) and quite possibly the influence of different extraction procedures.

The different Rubus genotypes (grown in the same location and in the same season, extracted and analysed under identical conditions) showed considerable variation in both the amount of, and composition of TTPN-like components (Fig. 5). Compared to Glen Ample, the total amount of TTPNs ranged between 34% (Cumberland) to 144% (Glen Cally; see ESI Fig. S4a). The compositional variation was also substantial (Fig. 3). For example, in breeding line 0485K-1, the predominant TTPN was peak 709a. The levels of peak T1 were lowest in Glen Ample and even Cumberland, which had low levels of total TTPNs, still had higher levels of peak T1 than Glen Ample. The ratio of peaks T1 to T2 varied considerably and was around 2–3 in many genotypes but reached >11-fold in breeding line 0534RB-1 (ESI; Fig. S4b). Such differences in the levels and composition of putative TTPNs could influence the bioactivities noted for different varieties in previous studies. The relative amounts of dimeric TTPNs (e.g. peak T2) could influence potential bioactivity as Nguelefack et al.23 reported that the dimeric TTPN glucoside, Coreanoside F1, was a more potent antioxidant than equivalent monomeric forms isolated from R. rigidus var. camerunensis. Very minor structural differences may also influence potential bioactivities. For example, the position of methyl groups in the triterpenic dialcohols, uvaol and erythrodiol, from olive oil altered bioactivity from genoprotective to genotoxic in both normal and breast cancer cell lines (MCF10A & MDA-MB-231).52 Furthermore, TTPNs isolated from Rubus rosifolius fruits exhibited differing abilities to inhibit the growth of colon cancer cells in vitro27 of 8 TTPN tested, only 2 – hyptatic acid and a glycoside of this compound, 4-epi-nigaichigoside F1 – were effective but, notably, positional isomers of both compounds were inactive. Ono and co-workers26 showed that blackberry leaf TTPNs showed great variation (4–90% inhibition) in their ability to inhibit foam cell formation in human monocyte-derived macrophages. For example, pomolic acid was 3 times more effective than tormentic acid which differ by one hydroxyl group.

image file: c7fo00846e-f5.tif
Fig. 5 Diversity of TTPN composition in raspberry varieties and breeding lines. Data are derived as average peak areas for each TTPN from triplicate samples ± SE expressed as a % of the internal peak area for each sample.

The variation in putative TTPNs between Rubus genotypes suggests that there may be some genetic control of the accumulation of these components which might be exploited to breed new varieties with elevated levels of TTPNs. The pedigree of these varieties is complicated and initial examination does not indicate that a simple model for TTPN inheritance could readily be derived. However, it is notable that Glen Cally and Glen Ericht, which are sister varieties derived from the same parents, had very similar TTPN profiles, (Table 5). Although more qualitative in nature, the overall % TTPN composition for the varieties analysed (ESI, Fig. S4c) provides a graphic illustration of the relative compositions. Other genetic relationships are more distant; e.g. Tulameen is a parent of both O46OF-5 and O534RB-1 and, although not identical by any means, these genotypes do share some similarities in the pattern of TTPN composition. The seed content of the fruits may also be relevant, as many of the genotypes differ in fruit size and have varying seed content. The outlying TTPN composition of Cumberland may arise because it is a different species (an old black raspberry cultivar) but also because it has a high proportion of seeds and, consistent with data presented in Fig. 3, higher relative amounts of seeds may affect the TTPN profile. However, genotypes with relatively low seed content (Glen Ample, Glen Fyne, Octavia and 9911C-1) did not exhibit similar TTPN profiles, and therefore it is possible that genotype may be the more influential factor. Further studies into the genetic basis of TTPN accumulation are warranted, perhaps through examination of TTPN diversity in raspberry progeny collections and integration with the genetic maps available at the Hutton.53,54

Table 5 Pedigree and other information on genotypes
Genotype Pedigree Fruit size Seed number
a Old blackberry cultivar (R. occidentalis) used in developing many UK red raspberry varieties (Dale et al., 1993).34 b Sister of Glen Cally, with Glen Fyne in background. c Sister of Glen Ericht, with Glen Fyne in background.
Cumberlanda Not known Very small High
Glen Doll Glen Rosa × 8605C-2 Small Moderate
0460F-5 0019B11 × Tulameen Medium High
0534RB-1 9764F-3 × Tulameen Very large Moderate
Tulameen Nootka × Glen Prosen Medium High
Glen Ample 7326E1 × 7412H16 Large Low
Octavia Glen Ample × Malling Hestia Very large Low
9455F-2 9072RE-1 × 9029RB-4 Large Moderate
0485K-1 0030E-12 × 0039F-2 Medium High
9911C-1 8914RB3 × 9238D5 Very large Low
Glen Erichtb 9422C-4 × 9434B-1 Large Moderate
Glen Callyc 9422C-4 × 9434B-1 Medium Moderate
Glen Fyne 8631D-1 × 8605C-2 Medium Low


This study has confirmed that potentially bioactive TTPN glycosides are present in fruits from a range of common Rubus genotypes including some widely-available commercial raspberry cultivars. The TTPNs are compositionally diverse and differ between genotypes and with extraction procedures. However most components are readily soluble in water and would be present and bio-accessible in products such as juices and purees as well as the intact fruits. Certain of these components were confirmed as colon-available in vivo28 and in vitro, studies suggest that others could also contribute to potential effects on colon cancer in vivo. These bioavailable, bioactive phytochemicals require greater attention to their possible biological effects.


TRFTriterpenoid-rich fraction
CDDO2-Cyano-3,12-dioxoolean-1,9-dien-28-oic acid
FAVFruit and vegetables

Conflicts of interest

The authors declare no conflicts of interest.


We would like to thank the volunteers for participating in the study. The authors thank Dr Nikki Jennings of James Hutton Limited for the expert advice and pedigree information on the Rubus varieties. CG, RL and AC acknowledge funding from the National Processed Raspberry Council. GMcD, GD, DS and JWA acknowledge funding from the Scottish Government's Rural and Environment Science and Analytical Services (RESAS) Division. DS and GMcD also acknowledge funding from BachBerry (Project No. FP7-613793).


  1. US Dept of Health & Human Services; UDSA, Scientific Report of the 2015 Dietary Guidelines Advisory Committee, 2015.
  2. A. Hjartåker, M. D. Knudsen, S. Tretli and E. Weiderpass, Consumption of berries, fruits and vegetables and mortality among 10,000 Norwegian men followed for four decades, Eur. J. Nutr., 2015, 54, 599–608 CrossRef PubMed.
  3. A. M. Rodriguez-Mateos, C. Heiss, G. Borges and A. Crozier, Cardioprotective effects of berry (poly)phenols, J. Agric. Food Chem., 2014, 62, 3842–3851 CrossRef CAS PubMed.
  4. S. Zanini, M. Marzotto, F. Giovinazzo, C. Bassi and P. Bellavite, Effects of dietary components on cancer of the digestive system, Crit. Rev. Food Sci. Nutr., 2015, 55, 1870–1885 CrossRef CAS PubMed.
  5. D. Graf, S. Seifert, A. Jaudszus, A. Bub and B. Watzl, Anthocyanin-Rich juice lowers serum cholesterol, leptin, and resistin and improves plasma fatty acid composition in Fischer rats, PLoS One, 2013, 8, e66690,  DOI:10.1371/journal.pone.0066690.
  6. T. E. Tjelle, L. Holtung, S. K. Bøhn, K. Aaby, M. Thoresen, S. Å. Wiik, I. Paur, A. S. Karlsen, K. Retterstøl, P. O. Iversen and R. Blomhoff, Polyphenol-rich juices reduce blood pressure measures in a randomised controlled trial in high normal and hypertensive volunteers, Br. J. Nutr., 2015, 114, 1054–1063 CrossRef CAS PubMed.
  7. S. V. Joseph, I. Edirisinghe and B. M. Burton-Freeman, Berries: Anti-inflammatory effects in humans, J. Agric. Food Chem., 2014, 62, 3886–3903 CrossRef CAS PubMed.
  8. N. Deighton, R. Brennan, C. Finn and H. V. Davies, Antioxidant properties of domesticated and wild Rubus species, J. Sci. Food Agric., 2000, 80, 1307–1313 CrossRef CAS.
  9. M. J. Anttonen and R. O. Karjalainen, Environmental and genetic variation of phenolic compounds in red raspberry, J. Food Compos. Anal., 2005, 18, 759–769 CrossRef CAS.
  10. D. R. Castrejón, I. Eichholz, S. Rohn, L. W. Kroh and S. Huyskens-Keil, Phenolic profile and antioxidant activity of highbush blueberry (Vaccinium corymbosum L.) during fruit maturation and ripening, Food Chem., 2008, 109, 564–572 CrossRef.
  11. S. H. Häkkinen, S. O. Kärenlampi, H. M. Mykkänen and A. R. Törrönen, Influence of domestic processing and storage on flavonol contents in berries, J. Agric. Food Chem., 2000, 48, 2960–2965 CrossRef.
  12. A. M. Connor, J. J. Luby, J. F. Hancock, S. Berkheimer and E. J. Hanson, Changes in fruit antioxidant activity among blueberry cultivars during cold-temperature storage, J. Agric. Food Chem., 2002, 50, 893–898 CrossRef CAS PubMed.
  13. M. Mikulic-Petkovsek, V. Schmitzer, A. Slatnar, F. Stampar and R. Veberic, Composition of sugars, organic acids, and total phenolics in 25 wild or cultivated berry species, J. Food Sci., 2012, 77, C1064–C1070 CrossRef CAS PubMed.
  14. K. T. Liby, M. M. Yore and M. B. Sporn, Triterpenoids and rexinoids as multifunctional agents for the prevention and treatment of cancer, Nat. Rev. Cancer, 2007, 7, 357–369 CrossRef CAS PubMed.
  15. J. M. Patlolla and C. V. Rao, Triterpenoids for cancer prevention and treatment: current status and future prospects, Curr. Pharm. Biotechnol., 2012, 13, 147–155 CAS.
  16. C. C. Neto, Ursolic acid and other pentacyclic triterpenoids: Anticancer activities and occurrence in berries, in Berries and Cancer Prevention, ed. G. D. Stoner and N. P. Seeram, Springer, 2007,  DOI:10.1007/978-1-4419-7554-62.
  17. N. Han and M. Bakovic, Biologically active triterpenoids and their cardioprotective and anti-inflammatory effects, J. Bioanal. Biomed., 2015, S12, 005,  DOI:10.4172/1948-593X.
  18. A. Szakiel, C. Paczkowski and S. Huttunen, Triterpenoid content of berries and leaves of bilberry, Vaccinium myrtillus, from Finland and Poland, J. Agric. Food Chem., 2012, 60, 11839–11849 CrossRef CAS PubMed.
  19. Z. G. Yang, H. R. Li, L. Y. Wang, Y. H. Li, S. G. Lu, X. F. Wen, J. Wang, A. Daikonya and S. Kitanaka, Triterpenoids from Hippophae rhamnoides L. and their nitric oxide production-inhibitory and DPPH radical-scavenging activities, Chem. Pharm. Bull., 2007, 55, 15–18 CrossRef CAS PubMed.
  20. A. Szakiel, C. Paczkowski, F. Pensec and C. Bertsch, Fruit cuticular waxes as a source of biologically active triterpenoids, Phytochem. Rev., 2012, 11, 263–284 CrossRef CAS PubMed.
  21. D. W. Griffiths, G. W. Robertson, T. Shepherd, A. N. Birch, S. C. Gordon and J. A. Woodford, Comparison of the composition of epicuticular wax from red raspberry (Rubus idaeus L.) and hawthorn (Crataegus monogyna Jacq.) flowers, Phytochemistry, 2000, 55, 111–116 CrossRef CAS PubMed.
  22. P. Chen, H. J. Yan, Q. X. Mei, Y. F. Chen, X. L. Wang, L. Ming and Z. J. Fang, Triterpenoids from the roots and stems of Rubus alceaefolius, Chem. Nat. Compd., 2016, 52, 248–251 CrossRef CAS.
  23. T. B. Nguelefack, F. H. K. Mbakam, L. A. Tapondjou, P. Watcho, E. P. Nguelefack-Mbuyo, B. K. Ponou, A. Kamanyi and H. J. Park, A dimeric triterpenoid glycoside and flavonoid glycosides with free radical-scavenging activity isolated from Rubus rigidus var. camerunensis, Arch. Pharm. Res., 2011, 34, 543–550 CrossRef CAS PubMed.
  24. W. Li, H. W. Fu, H. Bai, T. Sasaki, H. Kato and K. Koike, Triterpenoid saponins from Rubus ellipticus var. obcordatus, J. Nat. Prod., 2009, 72, 1755–1760 CrossRef CAS PubMed.
  25. J. R. Vera, C. G. Dacke, A. V. Patel and G. Blunden, Triterpenoids, including one with smooth muscle relaxant activity, from Rubus idaeus (raspberry) leaves, Nat. Prod. Commun., 2006, 1, 613–617 CAS.
  26. M. Ono, S. Yasuda, H. Komatsu, Y. Fujiwara, M. Takeya and T. Nohara, Triterpenoids from the fruits and leaves of the blackberry (Rubus allegheniensis) and their inhibitory activities on foam cell formation in human monocyte-derived macrophage, Nat. Prod. Res., 2014, 28, 2347–2350 CrossRef CAS PubMed.
  27. C. S. Bowen-Forbes, V. Mulabagal, Y. B. Liu and M. G. Nair, Ursolic acid analogues: Non-phenolic functional food components in Jamaican raspberry fruits, Food Chem., 2009, 116, 633–637 CrossRef CAS.
  28. G. J. McDougall, J. W. Allwood, G. Pereira-Caro, E. M. Brown, S. Verrall, D. Stewart, C. Latimer, G. McMullan, R. Lawther, G. O'Connor, I. Rowland, A. Crozier and C. I. R. Gill, Novel colon-available triterpenoids identified in raspberry fruits exhibit antigenotoxic activities in vitro, Mol. Nutr. Food Res., 2017, 61 DOI:10.1002/mnfr.201600327.
  29. D. Del Rio, A. M. Rodriguez-Mateos, J. P. Spencer, M. Tognolini, G. Borges and A. Crozier, Dietary (poly)phenolics in human health: structures, bioavailability, and evidence of protective effects against chronic diseases, Antioxid. Redox Signaling, 2013, 18, 1818–1892 CrossRef CAS PubMed.
  30. T. Bohn, G. J. McDougall, A. Alegría, M. Alminger, E. Arrigoni, A. M. Aura, C. Brito, A. Cilla, S. N. El, S. Karakaya, M. C. Martínez-Cuesta and C. N. Santos, Mind the gap - deficits in our knowledge of aspects impacting the bioavailability of phytochemicals and their metabolites: A position paper focusing on carotenoids and polyphenols, Mol. Nutr. Food Res., 2015, 59, 1307–1323 CAS.
  31. G. Yang, S. Bibi, M. Du, T. Suzuki and M. J. Zhu, Regulation of the intestinal tight junction by natural polyphenols: a mechanistic perspective, Crit. Rev. Food Sci. Nutr., 2016, 23 DOI:10.1080/10408398.2016.1152230.
  32. G. J. McDougall, P. Dobson, P. Smith, A. Blake and D. Stewart, Assessing potential bioavailability of raspberry anthocyanins using an in vitro digestion system, J. Agric. Food Chem., 2005, 53, 5896–5904 CrossRef CAS PubMed.
  33. E. M. Brown, G. J. McDougall, D. Stewart, G. Pereira-Caro, R. González-Barrio, P. Allsopp, P. Magee, A. Crozier, I. Rowland and C. I. Gill, Persistence of anti-cancer activity in berry extracts after simulated gastrointestinal digestion and colonic fermentation, PLoS One, 2012, 7, e49740,  DOI:10.1371/journal.pone.0049740.
  34. A. Dale, R. P. Moore, R. J. McNicol, T. M. Sjulin and L. A. Burmistrov, Genetic diversity of red raspberry varieties throughout the world, J. Am. Soc. Hortic. Sci., 1993, 118, 119–129 Search PubMed.
  35. G. J. McDougall, S. Conner, G. Pereira-Caro, R. Gonzalez-Barrio, E. M. Brown, S. Verrall, D. Stewart, T. Moffet, M. Ibars, R. Lawther, G. O'Connor, I. Rowland, A. Crozier and C. I. Gill, Tracking (poly)phenol components from raspberries in ileal fluid, J. Agric. Food Chem., 2014, 62, 7631–7641 CrossRef CAS PubMed.
  36. M. B. Hansen, S. E. Nielsen and K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell-growth/cell kill, J. Immunol. Methods, 1989, 119, 203–210 CrossRef CAS PubMed.
  37. E. M. Brown, C. Latimer, P. Allsopp, N. G. Ternan, G. McMullan, G. J. McDougall, D. Stewart, A. Crozier, I. Rowland and C. I. Gill, In vitro and in vivo models of colorectal cancer: Anti-genotoxic activity of berries, J. Agric. Food Chem., 2014, 62, 3852–3866 CrossRef CAS PubMed.
  38. X. H. Zhou, R. Kasai, K. Ohtani, O. Tanaka, R. Nie, C. Yang, J. Zhou and K. Yamasaki, Oleanane and ursane glucosides from Rubus species, Phytochemistry, 1992, 31, 3642–3644 CrossRef CAS.
  39. E. M. Coates, G. Popa, C. I. Gill, M. J. McCann, G. J. McDougall, D. Stewart and I. Rowland, Colon-available raspberry polyphenols exhibit anti-cancer effects on in vitro models of colon cancer, J. Carcinog., 2007, 18, 1–6 Search PubMed.
  40. C. Perret, J. L. Wolfender and K. Hostettmann, LC/ES-MS analysis of triterpene glycosides: Rapid estimation of the saponin content of dried berries of Phytolacca dodecandra, Phytochem. Anal., 1999, 10, 272–278 CrossRef CAS.
  41. J. Chen, X. Wang, Y. Cai, M. Tang, Q. Dai, X. Hu, M. Huang, F. Sun, Y. Li and P. Xia, Bioactivity-guided fractionation of physical fatigue-attenuating components from Rubus parvifolius L, Molecules, 2013, 18, 11624–11638 CrossRef CAS PubMed.
  42. K. Valdes, J. Morales, L. Rodriguez and G. Gunther, Potential use of nanocarriers with pentacyclic triterpenes in cancer treatments, Nanomedicine, 2016, 11 DOI:10.2217/nnm-2016-0251.
  43. M. Zhou, R. H. Zhang, M. Wang, G. B. Xu and S. G. Liao, Prodrugs of triterpenoids and their derivatives, Eur. J. Med. Chem., 2017, 131, 222–236 CrossRef CAS PubMed.
  44. T. Garanti, A. Stasik, A. J. Burrow, M. A. Alhnan and K. W. Wan, Anti-glioma activity and the mechanism of cellular uptake of asiatic acid-loaded solid lipid nanoparticles, Int. J. Pharm., 2016, 500, 305–315 CrossRef CAS PubMed.
  45. M. Contreras, P. D. Loeza, J. Villegas, R. Farias and G. Santoyo, A glimpse of the endophytic bacterial diversity in roots of blackberry plants (Rubus fruticosus), Genet. Mol. Res., 2016, 15, 3,  DOI:10.4238/gmr.15038542.
  46. S. Wada, J. A. Kennedy and B. M. Reed, Seed-coat anatomy and proanthocyanidins contribute to the dormancy of Rubus seed, Sci. Hortic., 2011, 130, 762–768 CrossRef CAS.
  47. M. Ono, S. Yasuda, K. Nishi, K. Yamamoto, S. Fuchizaki, S. Higuchi, H. Komatsu, M. Okawa, J. Kinjo, H. Yoshimitsu and T. Nohara, Two new triterpenoids from the seeds of blackberry (Rubus fructicosus), Nat. Prod. Res., 2016, 30, 904–911 CrossRef CAS PubMed.
  48. D. Gođevac, V. Tešević, V. Vajs, S. Milosavljević and M. Stanković, Antioxidant properties of raspberry seed extracts on micronucleus distribution in peripheral blood lymphocytes, Food Chem. Toxicol., 2009, 47, 2853–2859 CrossRef PubMed.
  49. A. A. Ramos, C. Pereira-Wilson and A. R. Collins, Protective effects of ursolic acid and luteolin against oxidative DNA damage include enhancement of DNA repair in Caco-2 cells, Mutat. Res., 2010, 692, 6–11 CrossRef CAS PubMed.
  50. P. L. de Freitas, A. C. Dias, V. R. Moreira, S. G. Monteiro and S. R. Pereira, Anti-mutagenic action of the triterpene betulinic acid isolated from Scoparia dulcis (Scrophulariaceae), Genet. Mol. Res., 2015, 14, 9745–9752 CrossRef CAS PubMed.
  51. J.-S. Shin, E.-J. Cho, H.-E. Choi, J.-H. Seo, H.-J. An, H.-J. Park, Y.-W. Cho and K.-T. Lee, Anti-inflammatory effect of a standardized triterpenoid-rich fraction isolated from Rubus coreanus on dextran sodium sulfate-induced acute colitis in mice and LPS-induced macrophages, J. Ethnopharmacol., 2014, 158, 291–300 CrossRef CAS PubMed.
  52. C. Sánchez-Quesada, A. López-Biedma and J. J. Gaforio, The differential localization of a methyl group confers a different anti-breast cancer activity to two triterpenes present in olives, Food Funct., 2015, 6, 249–256 Search PubMed.
  53. P. Dobson, J. Graham, D. Stewart, R. Brennan, C. A. Hackett and G. J. McDougall, Over-seasons analysis of quantitative trait loci affecting phenolic content and antioxidant capacity in raspberry, J. Agric. Food Chem., 2012, 60, 5360–5366 CrossRef CAS PubMed.
  54. S. McCallum, M. Woodhead, C. A. Hackett, A. Kassim, A. Paterson and J. Graham, Genetic and environmental effects influencing fruit colour and QTL analysis in raspberry, Theor. Appl. Genet., 2010, 121, 611–627 CrossRef CAS PubMed.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c7fo00846e

This journal is © The Royal Society of Chemistry 2017