A promising blueberry from Colombia: antioxidant activity, nutritional and phytochemical composition of Cavendishia nitida (Kunth) A.C.Sm.

Many Neotropical representatives of Ericaceae have fruits with antioxidant activity and high nutritional value. However, in Colombia, these fruits are little consumed and are considered underutilized. One such example is the berries of Cavendishia nitida (Kunth) A.C.Sm. In this study, the nutritional value, the total phenolic, flavonoid and anthocyanin contents, and antioxidant activity of C. nitida fruit were performed. From the leaves, an ethanolic extract was made, which was then fractioned to measure its antioxidant activity and analyze its chemical composition. The results indicate that the fruit of C. nitida can be classified as potentially edible, due to its minerals and vitamins contents. Five anthocyanins were detected in the berries; while in the leaves extract six terpenes and one flavonol were identified. The ethyl acetate fraction of the leaves extract exhibited strong antioxidant activity with the DPPH• and ABTS•+ radicals tested. We also found a strong correlation between the total phenolic and flavonoid contents and the values of percentage of inhibition of DPPH• and ABTS•+ in all the samples tested. The results of this study suggest that the berries of C. nitida are promising as edible fruits, and beneficial for human and animal health. However, even though the communities of the region use this berry as food, the toxicity of fruits must be evaluated to confirm that their consumption is safe for humans.

Introduction

Wild edible fruits have contributed significantly to the human diet (Bacchetta et al., 2016). Colombia is a territory with a wide diversity of wild fruits, which are an integral part of the diet of rural inhabitants (López Diago and García Castro, 2021). Wild fruits are often richer in nutrients and bioactive compounds, such as anthocyanins and flavonoids, than cultivated species (Bacchetta et al., 2016; Li et al., 2016). Studies on the composition of wild fruits have increased in recent years, especially in relation to their antioxidant, antimicrobial, anti-inflammatory, and anticancer activities, showing that the consumption of these fruits is beneficial for human health and in the prevention of chronic diseases (Li et al., 2016).

Ericaceae is a large family of flowering plants with almost 128 genera and about 4000 species (Luteyn, 2002). The fruits of the Ericaceae family are known for their high antioxidant content compared to other species from different families. As a result, these berries are considered "superfruits" (Dastmalchi et al., 2011; Rashidinejad, 2020). The edible berries of the genus Vaccinium are the most widely studied and cultivated. Research on Vaccinium berries has focused on the phenolic, flavonoid, and anthocyanin contents and antioxidant capacity, given the health benefits of these bioactive compounds (Dastmalchi et al., 2011; Ma et al., 2013; Rashidinejad, 2020). However, there are more than six hundred species of berry-producing Ericaceae native to the Neotropics (Luteyn, 2002), almost 270 species are found in Colombia with very few studies on them, and it is clear that these plants are underutilized (Dastmalchi et al., 2011).

Among these Ericaceae are the species of Cavendishia, most of which are endemic to the Neotropics, predominantly in cold montane forests between 1000 m and 3000 m and several species are found above the tree line up to 4650 m in the páramo ecosystem (Luteyn, 2002). The Neotropical Ericaceae have developed a tolerance to the conditions of the mountainous regions (Luteyn, 2002), in Colombia these species have adapted to the environmental conditions of the páramo and sub-páramo ecosystems, including low temperatures, high ultraviolet radiation, fog, high precipitation levels and strong winds (Luteyn, 1999). As a consequence, these plants have developed chemical adaptations that include the accumulation of anthocyanins and phenolic compounds as a protective response to reactive oxygen species (ROS) to cope with environmental conditions and the environmental stress that these conditions can cause (Alonso-Amelot, 2008).

Although the fruits of Cavendishia are consumed by rural communities in Colombia (López Diago and García Castro, 2021), there have been few studies focusing on their chemical and nutritional properties. It is known that the fruits of Cavendishia bracteata contain flavonoids, quinones, tannins, saponins, coumarins, and triterpenoids (Sanabria-Galindo et al., 1997; Plazas González, 2015). Likewise, the leaves of species of Cavendishia have been found to contain stilbenes, flavone glycosides, flavonoids, flavonols, proanthocyanidins and terpenes (Dastmalchi et al., 2011; Garcia, 1992; Luteyn et al., 1980), while quercetin, myricetin, and some stilbenes have been isolated from the leaves of Cavendishia nitida (Luteyn et al., 1980). However, to our knowledge, there is no nutritional information, and chemical studies have been conducted on the fruit.

C. nitida, from the Ericaceae family, is a shrub that is endemic to Colombia (Figure 1), commonly known as “Uva camarona”, “uvo”, or “uva”, its fruits are in the form of red berries, which are consumed by inhabitants of the “Altiplano Cundiboyacense”, a region in the center of the country (Luteyn, 1983).

Figure 1

Flowers and green berries (A - left), and wild berries at ripening stage (B - right) of Cavendishia nitida.

In the present study, the nutritional value of C. nitida fruits is evaluated for the first time, through an analysis of its minerals, vitamins and proximate composition. The total phenolic, total flavonoid and anthocyanin content was quantified, and the antioxidant capacity was assessed. The same analyses were also performed for the leaf extract, and the chemical composition of this extract was studied.

Material and methods

Plant materials

Fruits (550 g) and leaves (2 kg) of C. nitida were collected in February 2019 at páramo El Tablazo, near the municipality of Subachoque, Cundinamarca, Colombia [5° 00′ 37″ N 74° 11′ 34″ W], which lies 2900 m above mean sea level (mamsl). Some of the fresh fruits (50 g) were freeze dried (−80 °C) for the total phenolic and total flavonoid analyses. The anthocyanin and proximal analyses were performed using fresh samples (500 g). For the leaf extract, the leaves were dried and ground before solvent extraction.

A specimen was taxonomically registered at the Herbarium of the Pontificia Universidad Javeriana, under code N. García 764 (HPUJ: 030161). The material was collected in accordance with the framework contract for Access to Genetic Resources and its derivative products No. 212 of July 19, 2018, signed by the Ministry of Environment and Sustainable Development and the Pontificia Universidad Javeriana.

pH, total soluble solids and total acidity

The pH and total soluble solids of the C. nitida fruits were determined using a pH meter (Hanna Instruments HI 2210, Woonsocket, RI) and a digital Hanna refractometer (Hanna Instruments, Woonsocket, RI), 0–85% range at 20 °C, respectively. The total titratable acidity was determined according to Association of Analytical Chemists International (A.O.A.C) method (AOAC, 2003).

Proximate analysis

The moisture content was determined by the gravimetric method at 105 °C until constant weight. The crude protein content was calculated by multiplying the nitrogen content obtained by the Kjeldahl method (International Organization for Standardization, 2009) by factor 6.25. The total fat, dietary fiber and ash contents were determined according to the respective A.O.A.C methods (AOAC, 2003).

Preparation of extract of C. nitida leaves and fruits

730 g of dried, ground leaves was extracted by maceration with ethanol at room temperature. The ethanolic extract obtained (ECn1) was flocculated with water:acetone (2:1 v/v) and subsequently filtered and fractionated.

2.5 g of freeze-dried fruits was powdered in liquid nitrogen and ultrasound-assisted extraction was carried out using methanol (ECn2).

Fractionation of the ethanolic extract of C. nitida leaves (ECn1)

84 g of ECn1 was partitioned by continuous liquid-liquid extraction with solvents of increasing polarity, obtaining five fractions: n-hexane (Fr1), dichloromethane (Fr2), ethyl acetate (Fr3), n-butanol (Fr4) and an aqueous residue (Fr5).

2.22 g of Fr1 was fractionated by Column Chromatography (CC) using mobile phase n-hexane:ethyl acetate (gradient 0–100%), obtaining a yellow oil (Cn1, 0.1142 g) which was analyzed by Gas Chromatography coupled to a Mass Spectrometry (GC-MS); a white solid (Cn2, 0.1380 g), which was analyzed by GC-MS and purified by crystallization with chloroform – hexane to obtain a solid 3; a white solid (Cn3, 0,0843 g), which was recrystallized from chloroform – n-hexane to obtain solid 5; and finally a white solid (6, 0.0923 g).

Fractions Fr3 and Fr4 were monitored by RP18 TLC (Merck), using methanol:water (8:2) as mobile phase and spraying with revealing natural product - polyethylene glycol reagent (NP/PEG). 2 g of Fr3 was fractionated by CC using sephadex LH20 and MeOH:H2O (gradient 0–100%), obtaining a yellow solid 7 (0.0638 g).

For the separations by CC silica gel 60 (63–120 μm) was used, and the TLC were performed on silica gel 60 or RP–18 chromatoplates, with detection under UV light (254, 366 nm), iodine vapors, NP/PEG reagent and vanillin/H2SO4 followed by heating. The metabolites obtained were identified by IR, UV/vis, and NMR spectroscopy, chromatography, and mass spectrometry.

The yellow oil (Cn1) and white solid (Cn2) were analyzed by GC-MS on an Agilent Technologies 7890AGC gas chromatograph coupled to a mass spectrometer Hewlett Packard 5973 with quadrupole analyzer, electron ionization (IE) at 70 eV in full scan mode at 4.75 scan/s, in the mass range m/z 20–300 uma, a split/splitless injector, and GC column Rtx-5MS of 30 m × 0.25 mm (id) x 0.25 μm (df). The oven was set to a temperature of 80 °C (1 min), then increased to 310 °C (10min) @ 10 °C min−1. The temperatures of the ionization chamber and transfer line were 230 °C and 285 °C, respectively. The gas carrier used was helium (99.995%, Aga Fano, S.A), with a constant flow of 1 mL min−1. The components present in the fractions were identified by comparison of the acquired mass spectrum with those reported in the Willey8 and NIST14 databases.

Analysis of total phenolic compounds

The total phenolic content was determined by the Folin-Ciocalteu method (Ainsworth and Gillespie, 2007; Llivisaca et al., 2018; Wootton-Beard et al., 2011). 400 μL of Folin-Ciocalteu reagent (10% v/v) was added to 200 μL of extract or fraction (2.5 mg mL−1). After 5 min, 1600 μL of Na2CO3 (7.4% w/v) was added and the solution was left to rest for 2 h at room temperature. Finally, the absorbance at 765 nm was measured. The total phenolics content was determined by the equation obtained from the calibration curve, using gallic acid as standard, and the results are expressed as mg of gallic acid equivalents (GAE) per g of fresh weight (fw).

Analysis of total flavonoids

The total flavonoid content was determined by the AlCl3 method (Woisky and Salatino, 1998). 150 μL of AlCl3 2% w/v, 150 μL of 1M HCl and 150 μL of 1M CH3COOK was added to 50 μL of extract or fraction (2.5 mg mL−1). Subsequently, 850 μL of methanol was added. The solution was left to rest, in the dark, at room temperature, for 40 min and the absorbance at 425 nm was measured. The total flavonoid content was determined by the equation obtained from the calibration curve using quercetin as standard, and the results are expressed as mg of quercetin (QE) per g of fresh weight (fw).

Analysis of anthocyanins

Analysis of total monomeric anthocyanin

The total monomeric anthocyanin content was determined by the pH differential method (Giusti and Wrolstad, 2001) and the absorbencies were measured at 510 and 700 nm. The anthocyanin content was calculated as cyanidin 3-glucoside (C3G), with the extinction coefficient (ε) of 26900 L cm−1 mol−1 and molecular mass (449.2 g mol−1). The results are expressed as mg of C3G equivalents per g of fw.

Analysis of anthocyanin contents by UPLC-DAD

The anthocyanin content was determined by UPLC-DAD (Brown and Shipley, 2011) using a Shimadzu Prominence Nexera-I (Kyoto, Japan), with an SPD-M20A DAD system. The data collection was performed using Shimadzu LC Solution software. For the sample preparation and UPLC analyses, deionized water, and solvents such as methanol, acetonitrile (ACN), and 88% formic acid (v/v), were used. A Phenomenex Kinetex® C18 column (100 mm × 2.1 mm id x 2.6 μm particle size) was used, at a temperature of 40 °C. The mobile phases were (A) formic acid (2%) and (B) acetonitrile-formic acid (1%), and the solvents used were HPLC grade (Merck, Darmstadt, Germany). For the separation, ACN/HCOOH gradient was used, starting with 5–11% B for 4.17 min, then 11–18% B for 9.58 min, 18–35% for 0.42 min, and finally 35-5% B for 2 min. The injection volume and flow rate were 3 μL and 0.5 mL min−1, respectively and detection was achieved at 520 nm. For each extract or fraction, 250 mg was weighed in 20 mL of methanol, from which 1 mL had been previously filtered.

For the compound identification, the retention times and UV/vis spectrum of each peak in the chromatogram were compared to the standards, and the anthocyanin content was calculated by the equation obtained from the calibration curve using six standards: cyanidin 3-O-galactoside (C3Ga), cyanidin 3-O-glucoside (C3G), cyanidin 3-O-arabinoside (C3A), peonidin 3-O-glucoside (P3G), malvidin 3-O-glucoside (M3G), and petunidin 3-O-glucoside (Pe3G). Table 1 shows the standard concentrations. Anthocyanin standards were obtained from Sigma-Aldrich (St. Louis, MO, USA) and PhytoLab (Vestenbergsgreuth, Germany). Due to the unavailability of some standards, such as delphinidin anthocyanins, this type of anthocyanin was tentatively identified by UPLC-ESI/MS based on accurate mass and fragmentation ion patterns of ion peaks compared with previous theoretical and experimental monoisotopic masses reported for the Ericaceae family, using a mass accuracy <5 ppm (Ma et al., 2013).

Table 1

Concentrations of the standards using in the calibration curve for the determination of anthocyanin of C. nitida.

Concentration range	Concentration of each anthocyanin standard (μg mL−1)
	C3Ga	C3G	C3A	P3G	M3G	Pe3G
C1	44.00	11.00	44.00	7.70	44.00	44.00
C2	22.00	5.50	22.00	3.85	22.00	22.00
C3	11.00	2.75	11.00	1.92	11.00	11.00
C4	5.50	1.37	5.50	0.96	5.50	5.50
C5	2.75	0.69	2.75	0.48	2.75	2.75
C6	1.37	0.34	1.37	0.24	1.37	1.37
C7	0.69	0.17	0.69	0.12	0.69	0.69

C3Ga: cyanidin 3-O-galactoside, C3G: cyanidin 3-O-glucoside, C3A: cyanidin 3-O-arabinoside, P3G: peonidin 3-O-glucoside, M3G: malvidin 3-O-glucoside, and Pe3G: petunidin 3-O-glucoside.

UPLC-ESI/MS analyses were performed on a Shimadzu® LCMS-9030 QqQ–time-of-flight (Q-TOF) LC–mass spectrometer equipped with a UV/vis detector (SPD-20AV) (Shimadzu Scientific Instruments, Columbia, MD, USA). The ESI interface was controlled by workstation LabSolutions™ LCMS for LCMS-9030. Mass spectrums were acquired in both positive and negative modes over the range m/z 100–1000. The capillary voltages were set at +3 kV and -2.8 kV, and the cone voltage was 20 V. The nebulization and desolvation gas used was nitrogen. The desolvation and cone gas flow rates were 300 and 20 L h−1, respectively. The desolvation temperature was 400 °C, and the source temperature was 120 °C. Positive and negative ESI data were collected using a scan time of 500 ms (Ma et al., 2013).

Antioxidant activities by radical scavenging capacity of DPPH• and ABTS•+ assays

The antioxidant activity of the extracts (ECn1 and ECn2) and fractions were determined by radical scavenging capacity of 1,1-diphenyl-2-picryl-hydrazyl (DPPH•) at 520 nm, and radical scavenging capacity ABTS•+ at 740 nm, both read on a FLUOStar Omega plate reader (BMG Labtech) Methanol was used as negative control, and Trolox as positive control, at concentrations of 0–10 mg L−1 (Garzón et al., 2010; Giraldo Vásquez and Ramírez Aristizabal, 2013; Guija-Poma et al., 2015; Llivisaca et al., 2018; Nuengchamnong and Ingkaninan, 2017; Wootton-Beard et al., 2011). The DPPH• and ABTS•+ free radical scavenging activity were calculated as mean inhibitory concentration (IC50).

Statistical analysis

The results are representative of three independent replicates and are expressed as mean ± SD. The correlation between antioxidant activity and phenolic and flavonoid contents of C. nitida was determined by the Pearson correlation test, previous tests of homoscedasticity, and normality of the data. The analyzes were conducted using the RStudio program, Version 1.2.1335, of the corTest package.

Results and discussion

Physicochemical characteristics

Cavendishia nitida can be considered a wild edible fruit (López Diago and García Castro, 2021) and its nutrition value has not yet been investigated. We determined some quality parameters of the fruit, which are described in Table 2. The total acidity (as citric acid %) and total soluble solids (°Brix) are related to the flavor of the fruit, and the pH gives an indication of the ability of the fruit to withstand microbial growth [26]. The color of the fruit is influenced by the pH, among other factors. Anthocyanins are red at low pH and turn blue as the pH increases (Khoo et al., 2017).

Table 2

Physicochemical and nutritional values of C. nitida fruits.

Parameters	Values∗
pH	3.31 ± 0.04
Total acidity, %	0.76 ± 0.02
Total soluble solids, °Brix	10.6 ± 3.9
Moisture, %	83.31 ± 0.03
Crude protein, g per 100g dw	0.29 ± 0.01
Crude fat, g per 100g dw	0.13 ± 0.01
Total dietary fiber, g per 100g dw	3.26 ± 0.20
Crude ash, g per 100g dw	0.30 ± 0.02
Total carbohydrates, g per 100g dw	16.0 ± 0.7
Calorific Value, Kcal per 100g dw	66.2 ± 0.6
Potassium, mg K per 100g dw	236.8 ± 17.5
Calcium, mg Ca per 100g dw	11.5 ± 2.5
Sodium, mg Na per 100g dw	114.9 ± 13.4
Iron, mg Fe per 100g dw	1.3 ± 0.2
Phosphorus, mg P per 100g dw	84.9 ± 6.7
Ascorbic acid, mg per 100g fw	62.0 ± 2.1
Thiamine, mg per 100g fw	0.05 ± 0.01
Riboflavin, mg per 100g fw	0.79 ± 0.06
Total anthocyanins, mg C3G per 100g fw	503.1 ± 34.2

Values are the average of triplicates; C3G = cyaniding 3-O-glucoside.

The results of the proximal analysis of C. nitida berries are shown in Table 2. The high moisture content (83.31%) in the fruits is similar to the values reported for other berries (Padmanabhan et al., 2015), and makes them more susceptible to spoilage due to microbial contamination, reducing the shelf life of the fresh fruit. The ash measurement indicates the mineral content in food (Rashidinejad, 2020). The ash values for C. nitida (0.30%) are similar to those of other berries (Padmanabhan et al., 2015). C. nitida fruits are a source of dietary fiber, with 3.26% of fruit weight consisting of fiber, which is similar to the values found for blueberries of Vaccinium ssp. (3–3.5%) (Michalska and Łysiak, 2015).

Enzymes require minerals (Table 2) to catalyze metabolic reactions. Potassium (236.9 mg per 100 g, dry weight) was the most abundant element in the fruits of C. nitida, followed by sodium (114.9 mg per 100 g, dry weight), phosphorus (84.9 mg per 100 g, dry weight), calcium (11.5 mg per 100 g, dry weight) and iron (1.3 mg per 100 g, dry weight).

Ascorbic acid is used as an antioxidant and preservative, helping to extend the shelf life of the fruits (Skrovankova et al., 2015; Valente et al., 2011). The content of ascorbic acid in the fruits is cultivar-dependent. In C. nitida fruit, ascorbic acid (62.0 mg per 100 g, fresh weight) (Table 2) is present in higher concentrations than the average value for blueberries (10 mg per 100 g, fresh weight) (Michalska and Łysiak, 2015; Shivembe and Ojinnaka, 2017), and is similar to the content found in strawberries and raspberries (Skrovankova et al., 2015). The vitamin content, as thiamine (0.05 mg per 100 g, fresh weight) and riboflavin (0.79 mg per 100 g, fresh weight), is similar to that of other berries (Michalska and Łysiak, 2015; Golubtsova, 2017).

Cavendishia nitida extracts and fractions

Table 3 shows the yield percentages of the crude extracts of the leaves (ECn1) and fruits (ECn2) of C. nitida, and the fractions of ECn1 (Fr1  – Fr5).

Table 3

Yield percentages of extracts and fractions of C. nitida.

Extract/Fraction	Percentage yield (%w/w)
ECn1	15.4
ECn2	48.0
Fr1	2.0
Fr2	4.2
Fr3	15.9
Fr4	13.4
Fr5	62.9

In the GC-MS analysis of Cn1 (Figure 2), two compounds were determined: Rimuene (1) and Kaur-16-ene (2) (Figure 3). Rimuene (1) is a diterpene, and is not taxonomically restricted, although it has been reported in plants of the genus Cavendishia and others of the family Ericaceae (Salasoo, 1988; Zamudio et al., 2019). It has also been isolated from plants of different families, such as Podocarpaceae, Cupressaceae, and Illiciaceae (Vijayakumar et al., 2012). Studies with extracts containing this compound have reported anticancer and antimicrobial activity, however, it is unknown whether they are caused specifically by this compound (Sylvestre et al., 2005; Vijayakumar et al., 2012). Kaur-16-ene (2), another diterpene, has been reported for other species of the genus Cavendishia (Zamudio et al., 2019), the Asteraceae family, specifically the genus Pentacalia, and the Cupressaceae family (Matsunaga et al., 2012). Kaurano-type compounds have antibacterial activity (Sylvestre et al., 2005).

Figure 2

GC-MS total ion chromatogram (TIC) of Cn1 and 1 - mass spectrum of Rimuene (1); 2 – mass spectrum of Kaur-16-ene (2).

Figure 3

Metabolites isolated from leaves of C. nitida. Compounds 1 to 7 are first reported for the species.

From the analysis by GC-MS of white solid Cn2 (Figure 4), two triterpenes were determined: α – amyrin (3) and β – amyrin (4) (Figure 3). These compounds have been reported in plants of the Ericaceae, Balanophoraceae, and Combretaceae families and there have been studies of their anti-inflammatory and antibacterial activity (Malca Garcia et al., 2015; Matulevich et al., 2016; Salih et al., 2018).

Figure 4

GC-MS total ion chromatogram (TIC) of Cn2 and 3 - mass spectrum of α-amyrin (3); 4 – mass spectrum of β-amyrin (4).

α – amyrin (3) was purified from Cn2 by crystallization with chloroform – hexane to obtain a needle-shaped white solid 3. soluble in chloroform, m.p. 182–184 °C (lit. 182–183 °C, (Sirat et al., 2010)). IR (CHCl3), ν (cm−1): 3400, 2900, 1600, 1460, 1380, 1100. 1H-NMR (300 MHz, CDCl3) (Figure 5), δH (ppm): 5.05 (t, J = 3.2, H-5), 3.20 (dd, J = 5.0; 11.0, H-3), 1.96 (td, J = 4.5; 13.5, H-15β), 1.84 (dt, J = 3.0; 8.0, H-22), 1.76 (td, J = 5.0; 13.0, H-16β), 1.00 (s, H-27), 0.94 (s, H-28), 0.93 (s, H-23), 0.90 (s, H-26), 0.85 (d, J = 6.0, H-29), 0.74 (s, H-24), 0.73 (d, J = 7.0, H-30), 0.72 (s, H-25), 0.67 (d, J = 11.6, H-5). 13C-NMR (75 MHz, CDCl3) (Figure 6), δC (ppm): 139.5 (C-13), 124.3 (C-12), 79.5 (C-6), 59.1 (C-18), 55.2 (C-5), 47.7 (C-9), 42.1 (C-14), 41.5 (C-22), 40.3 (C-20), 39.9 (C-19), 39.9 (C-8), 38.6 (C-1), 38.2 (C-4), 36.5 (C-10), 33.6 (C-17), 32.5 (C-7), 31.3 (C-21), 28.7 (C-2), 28.3 (C-28), 28.1 (C-23), 27.3 (C-15), 26.6 (C-16), 23.6 (C-11), 23.4 (C-27), 21.4 (C-30), 18.4 (C-6), 17.4 (C-29), 16.8 (C-26), 15.8 (C-25), 15.7 (C-24). The 1H and 13C NMR data agree with literature (Sirat et al., 2010). EIMS: 70 eV, m/z (% intensity): 426 M+, 408 (37), 365 (33), 218 (100), 203 (34), 189 (36), 136 (24), (Figure 7)

Figure 5

1H-NMR spectrum (CDCl3) of a-amyrin (3).

Figure 6

13C-NMR spectrum (CDCl3) of a-amyrin (3).

Figure 7

Mass spectrum of α-amyrin (3).

Friedelin (5) (Figure 3). needle-shaped white solid, recrystallized in hexane – AcOEt, m.p. 259–260 °C (lit. 261–262 °C, (Subhadhirasakul and Pechpongs, 2005)); IR (CHCl3) ν (cm−1): 2930, 1700, 1460, 1380. 1H-NMR (300 MHz, CDCl3) (Figure 8), δH (ppm): 2.42–2.40 (m, H-2), 2.38–2.36 (m, H-4), 1.95–1.93 (m, H-1), 1.75–1.34 (m), 1.25 (s, H-28), 1.16 (s, H-27), 1.03 (s, H-26), 1.02 (s, H-29), 0.94 (s, H-30), 0.90 (d, J = 6.4 Hz; H-23), 0.88 (s, H-25), and 0.71 (s, H-24). 13C-NMR (75 MHz, CDCl3) (Figure 9), δC ppm: 213.1 (C-3), 59.4 (C-10), 58.5 (C-4), 52.9 (C-8), 43.0 (C-18), 42.1 (C-5), 42.0 (C-2), 41.7 (C-6), 40.1 (C-14), 39.2 (C-22), 38.6 (C-13), 37.4 (C-9), 36.3 (C-16), 35.9 (C-11), 35.8 (C-19), 35.5 (C-29), 33.1 (C-21), 32.8 (C-15), 32.4 (C-28), 32.0 (C-30), 30.6 (C-12), 30.3 (C-17), 28.6 (C-20), 22.7 (C-1), 19.8 (C-26), 18.8 (C-27), 18.3 (C-7), 18.0 (C-25), 14.6 (C-24), 6.8 (C-23). The 1H and 13C NMR data agree with literature (Subhadhirasakul and Pechpongs, 2005). EIMS (Figure 10) m/z (% intensity): 426 ([M+], 93), 411 (28), 341 (20), 302 (56), 274 (40), 273 (70), 246 (51), 231 (39), 218 (58), 205 (51), 179 (50), 125 (100), 109 (68).

Figure 8

1H-NMR spectrum (CDCl3) of Friedelin (5).

Figure 9

13C-NMR spectrum (CDCl3) of Friedelin (5).

Figure 10

Mass spectrum of Friedelin (5).

Glutinol (6) (Figure 3). Amorphous white solid, m.p. 202–203 °C (lit. 201–203 °C, (Abdel-Sattar et al., 2015)); IR (CHCl3) ν (cm−1): 3450, 2950, 1600, 1380, 1090. 1H-NMR (300 MHz, CDCl3) (Figure 11), δH (ppm): 5.65 (t, H-6), 3.5 (s, H-3α), 1.19 (s, H-28), 1.15 (s, H-24), 1.08 (s, H-26), 1.03 (s, H-23), 1.00 (s, H-27), 0.99 (s, H-30), 0.93 (s, H-29), 0.89 (s, H-25). 13C-NMR (75 MHz, CDCl3) (Figure 12), δC ppm: 141.6 (C-5), 122.3 (C-6), 76.3 (C-3), 49.6 (C-10), 47.4 (C-8), 43.2 (C-18), 41.0 (C-4), 39.2 (C-13), 39.1 (C-22), 38.0 (C-14), 36.1 (C-16), 35,3 (C-19), 35.1 (C-9), 34.8 (C-11), 34.6 (C-29), 33.1 (C-21), 32.5 (C-30), 32,3 (C-15), 32.2 (C-28), 30.4 (C-12), 30.2 (C-17), 29.0 (C-23), 28.2 (C-20), 28.0 (C-2), 25.4 (C-24), 23.6 (C-7), 19.6 (C-26), 18.4 (C-27), 18.2 (C-1), 16,2 (C-25). The 1H and 13C NMR data agree with literature (Abdel-Sattar et al., 2015).

Figure 11

1H-NMR spectrum (CDCl3) of Glutinol (6).

Figure 12

13C-NMR spectrum (CDCl3) of Glutinol (6).

3-O-methyl quercetin (7) (Figure 3). Yellow crystals, recrystallized in methanol – H2O, m.p. 275–276 °C. UV-VIS (MeOH), λ (nm): 256, 358; (AlCl3) 270 y 433; (AlCl3/HCl) 274 y 408; (MeONa) 268 y 405; (AcONa) 273 y 400; (AcONa/H3BO3) 260 y 380. IR ν (cm−1): 3426, 1465, 1630, 1260. 1H NMR NMR (300 MHz, acetone-d6) (Figure 13), δH (ppm): 3.70 (s, H, OCH3-3″), 12.69 (s, H, OH-5), 6.30 (d, J = 1.8 Hz, H-6), 10.70 (s, H, OH-7), 6.39 (d, J = 2.01 Hz, H-8), 7.56 (d, J = 1.98 Hz, H-2′), 10.72 (s, H, OH-3′), 10.79 (s, H, OH-4′), 6.87 (d, J = 8.0 ′Hz, H-5′), 7.40 (dd, J = 1.98, 8.1 Hz, H-6′). 13C NMR (75 MHz, acetone-d6) (Figure 14), δC (ppm): 175.7 (C-4), 162.1 (C-7), 160.3 (C-5), 154.9 (C-9), 151.9 (C-2), 148.6 (C-4′), 143.1 (C-3′), 137.5 (C-3), 122.8 (C-1′), 120.2 (C-6′), 115.7 (C-5′), 115.5 (C-2′), 105.0 (C-10), 96.9 (C-8), 91.7 (C-6), 59.1 (OCH3-3″), 1H and 13C NMR data were consistent with the literature (Kumar et al., 2016). ESI-MS (m/z) 315 [M-H] + (Figure 15).

Figure 13

1H-NMR spectrum (acetone-d6) of 3-O-methyl quercetin (7).

Figure 14

13C-NMR spectrum (acetone-d6) of 3-O-methyl quercetin (7).

Figure 15

LC-MS Total ion chromatogram (TIC) and the full scan mass spectra of 3-O-methylquercetin (7).

Triterpenoids have been reported in plants of the genus Cavendishia and others of the family Ericaceae (Salasoo, 1988; Zamudio et al., 2019) and are components that have aroused great interest as anti-inflammatory agents, the presence of triterpenes in C. nitida could suggest their potential pharmacological use. In fact, some communities in Colombia mention the use of C. bracteata leaves, in decoction, to treat arthritis.

Total phenolics, flavonoids, anthocyanins, and antioxidant activity

Berries of the Ericaceae family (blueberry, cranberry) are great food sources and rich in bioactive compounds (total phenolics, flavonoids, anthocyanins, and ascorbic acid) (Skrovankova et al., 2015) Phenols and flavonoids are compounds with high pharmacological activity: they have antiviral, anti-inflammatory, antifungal activity, and antioxidant activities that regulate the formation of reactive oxygen species and decrease the degradation processes of macromolecules, such as lipid peroxidation. Due to this activity, they are used as food additives (Gutiérrez et al., 2010; Havsteen, 2002; Heim et al., 2002; Ilić et al., 2021; Pourcel et al., 2007; Shakil et al., 2008; Sharma et al., 2019). The total content of phenolic compounds in berries is cultivar-dependent, growing conditions and maturity (Skrovankova et al., 2015).

Table 4 shows the values of phenols (mg of gallic acid equivalents (GAE) per g fw) and total flavonoids (mg of quercetin equivalents (QE) per g of fw) present in the extracts and fractions analyzed. The leaves extract (ECn1) has the highest content of the two metabolites analyzed, while the fruit extract (ECn2) has the lowest content. This study showed the same behavior as Cavendishia bracteata (Plazas González, 2015). This trend could be explained by a greater presence of tannins in the leaves compared to the other organs of the plant, a tendency reported for other plants belonging to Ericaceae (Abreu Guirado et al., 2008). The total phenolic content (3.9 mg GAE per g fw) and flavonoids (2.5 mg QE per g fw) in C nitida fruits is comparable with the values reported in the literature for other berries (Dastmalchi et al., 2011; Rashidinejad, 2020; Skrovankova et al., 2015).

Table 4

Total phenols and flavonoids content in extracts and fractions of C. nitida.

Extract/Analized Fraction	Total Phenol Content (mg GAE per g fw)	Total Flavonoid Content (mg QE per g fw)
ECn1	1012.8 ± 8.8	5206 ± 10.9
ECn2	3.9 ± 0.2	2.5 ± 0.1
Fr3	572.4 ± 5.6	1206 ± 7.1
Fr4	31.6 ± 0.8	4113.6 ± 12.4

Values are the average of triplicates.

Leaves and fruits of C. nitida were collected at 2900 mamsl. At this altitude, the plants are subjected to stress for low temperature and intense sunlight, with the consequent oxidative stress (Taiz et al., 2014). Under these conditions, the synthesis of flavonoids, anthocyanins and other phenolic compounds increases, as a mechanism to mitigate damage to the plant (Ramakrishna and Ravishankar, 2011).

Total anthocyanins content

Plants of the Ericaceae family, such as blueberry or cranberry, are among the main sources of anthocyanins in edible plants (Hyun et al., 2015). Cavendishia nitida produces berries of up to 14.33 mm in diameter, with colors ranging from red to purple, indicating high anthocyanin content. The total monomeric anthocyanin content in cranberries is 25–100 mg per 100 g of fresh weight (fw) and in blueberries, it is 25–495 mg per 100 g fw. This composition is dependent on cultivar, growing conditions and maturity (Padmanabhan et al., 2015; Rashidinejad, 2020). Cavendishia nitida fruits contain 503.1 mg of C3G equivalents per g of fw (Table 2), a value that is comparable with blueberries. Due to the high antioxidant potential of anthocyanins, these compounds present in blueberries are expected to inhibit oxidative processes related to vascular diseases and cancer (Ilić et al., 2021; Martín-Gómez et al., 2021; Neto, 2007).

Analysis of anthocyanin content by UPLC-DAD.

Figure 16 shows the chromatographic profile of Cavendishia nitida fruits. Comparing the retention times and UV/vis data of the fruit extract with the data for six standards: cyanidin 3-O-galactoside (C3Ga), cyanidin 3-O-glucoside (C3G), cyanidin 3-O-arabinoside (C3A), peonidin 3-O-glucoside (P3G), malvidin 3-O-glucoside (M3G), and petunidin 3-O-glucoside (Pe3G), three anthocyanins were identified (Table 5): cyanidin 3-O-glucoside (C3G) the most abundant anthocyanin (59% of the total peak area), petunidin 3-O-glucoside and peonidin 3-O-glucoside. According to UV/vis and MS-MS data the two remaining peaks from Figure 16 were tentatively identified as delphinidin 3-O-hexoside and peonidin 3-O-galactoside. In the UPLC-ESI/MS-MS analysis of these two peaks (Table 5), delphinidin 3-O-hexoside (peak 1) revealed, in positive mode, a molecular ion at m/z 465.1013 and fragment ion at m/z 303.0504 corresponding to aglycone delphinidin, while peonidin 3-O-galactoside (peak 4) revealed, in positive mode, a molecular ion at m/z 463.1250 and fragment ion at m/z 301.0708 corresponding to aglycone peonidin. Based on Ma et al. (2013), the monoglycosides of anthocyanidin are eluted on the C-18 column in the order of increasing retention time: galactoside, glucoside, and arabinoside. Therefore, the sugar moiety ascribed to peak 4 was determined by this elution series rule as galactoside, considering that peak 5 corresponds to peonidin 3-O-glucoside. Mass spectra of delphinidin 3-O-hexoside, peonidin 3-O-galactoside, petunidin 3-O-glucoside, cianidin 3-O-glucoside, and peonidin 3-O-glucoside are presented in Figures 16, 17, 18, and 19.

Figure 16

UPLC-PDA chromatograms of anthocyanins of (A) C. nitida fruits and (B) standards detected at 512 nm.

Table 5

Anthocyanins detected in C. nitida.

# Peak	Peak area (%)	Content, μg g−1 dw	λmax, nm	tR, min	Molecular ion, m/z (MF, ppm)	Fragment ion exact mass, m/z (MF, ppm)	Tentative identification
1	22.3	-	521	3.924	465.1013 [M+] (C21H21O12, -2.58)463.0869 [M − H]− (C21H19O12, −1.62)	303.0504 [M − hexosyl]+ (C15H11O7, −0.3)499.0644 [M − H + Cl]− (C21H20O12Cl, −0.06);509.0924 [M − 2H + HCOOH]− (C22H21O14, 1.5)	Delphinidin 3-O-hexoside
2	58.9	2519	516	4.748	449.1075 [M]+ (C21H21O11, −1.97)	287.0558 [M − glucosyl]+ (C15H11O6, 0.83)	Cyanidin 3-O-glucoside
3	3.3	192	523	5.312	479.1187 [M]+ (C22H23O12, −0.52)	317.0663 [M − glucosyl]+ (C16H13O7, 0.54)	Petunidin 3-O-glucoside
4	13.1	-	520	5.527	463.1250 [M]+ (C22H23O11, 2.08)	307.0708 [M − galactosyl]+ (C16H13O6, 1.5)497.0855 [M − H + Cl]− (C22H22O11Cl, −1.0);507.1143 [M − 2H + HCOOH]− (C23H23O13, −0.85).	Peonidin 3-O-galactoside
5	2.4	115	520	6.144	463.1249 [M]+ (C22H23O11, 1.86)	301.0715 [M − hexosyl]+ (C16H13O6, 0.95)	Peonidin 3-O-glucoside

Peak area (%) = Peak area (%) UPLC-DAD; tR = retention time; MF = molecular formula.

Figure 17

Mass spectra of delphinidin 3-O-hexoside obtained in positive ion mode (A), negative ion mode (C), and (B) MS-MS spectrum of m/z 465.1013.

Figure 18

Mass spectra of peonidin 3-O-galactoside obtained in positive ion mode (A), negative ion mode (C), and (B) MS-MS spectrum of m/z 463.1250.

Figure 19

Mass spectra obtained in positive ion mode of petunidin 3-O-glucoside (A), cianidin 3-O-glucoside (B), and peonidin 3-O-glucoside (C).

Among berries, those belonging to the Ericaceae family contain the highest amount of antioxidants, mainly composed of anthocyanins, which give the ripe fruits their color. For example, the glycosides of cyanidin, delphinidin, and pelargonidin provide the red, blue, and purple colors, respectively, of fresh blueberries (Liu et al., 2020; Rashidinejad, 2020). Berries of the genus Cavendishia contain monoglycosides of delphinidin, cyanidin, petunidin, peonidin, and malvidin (Ma et al., 2013), while in C. nitida berries, no malvidin glycosides were detected. The foregoing is important, considering the antioxidant and beneficial properties for human health of these compounds, which is largely due to the consumption of its fruits (Guija-Poma et al., 2015; Jungmin, 2006; Merzlyak et al., 2008).

Brown et al. (2012) reported high values of C3G (80 μg g−1 dw), P3G (1500 μg g−1 dw), C3A (500 μg g−1 of dw) and P3A (600 μg g−1 dw) for fruits of various cultivars of Vaccinium macrocarpon compared to those calculated for C. nitida fruit.

An ecological aspect that can affect the anthocyanin content is the size of the fruit. In fruits of different species of the genus Vaccinium it was observed that there is a relationship between the accumulation of anthocyanins and the size of the fruit: the smaller the fruit, the greater the accumulation of these compounds (Gao and Mazza, 1994). This could explain the low anthocyanin content in the fruits of C. nitida, due to the berries of the Cavendishia genus are one of the largest in this plant family. According to Villagra et al. (2014) studies on the fruits of the genus Gaultheria that do not exceed 8 mm have reported values of phenolic compounds higher than those found for C. nitida.

Antioxidant activity of leaves and fruits of C. nitida

Oxidative stress occurs due to the production of radicals, and is a cause of multiple cardiovascular and neurodegenerative diseases. Ericaceae berries act as free radical scavengers, due to their well-known high antioxidant content (Dastmalchi et al., 2011).

For C. nitida, the fractions of the leaf extract enriched in flavonoids (Fr3 and Fr4) presented high antioxidant activity. Furthermore, the high anthocyanin content in the fruits of C. nitida suggests that these berries can be used as sources of bioactive compounds, with possible uses as antioxidants and colorants, and as ingredients in functional foods.

ABTS •+ and DPPH• scavenging activities of the extracts of fruits and leaves of C. nitida, and of the fractions of leaves are shown in Table 6, the EtOH crude extract leaves (ECn1) had the highest inhibitory activity, followed by EtOAc fraction (Fr3), BuOH fraction (Fr4), and finally, MeOH crude extract of the fruits (ECn2).

Table 6

DPPH• and ABTS•+ Scavenging (mg L−1) from extracts and fractions of C. nitida needed to inhibit 50% of oxidative damage (free radical).

Extract/Fraction	IC50 DPPH• (mg L−1)	IC50 ABTS•+ (mg L−1)
ECn1	8.12 ± 0.02	21.90 ± 0.22
ECn2	65.86 ± 0.13	145.43 ± 0.88
Fr3	12.14 ± 0.05	29.01 ± 0.11
Fr4	17.51 ± 0.09	67.38 ± 0.32

Table 6 shows higher IC50 values for ABTS•+, compared to those obtained for DPPH•. This trend corroborates the results obtained in a study conducted on various foods (Floegel et al., 2011). Although the two trials (ABTS•+ and DPPH• scavenging activities) are based on the same principle, this difference may have been due to three factors. The first is the different wavelengths at which the tests are carried out: ABTS•+ is measured at 740 nm, while DPPH• is measured at 520 nm. The wavelength of the ABTS•+ assay is high enough to avoid interference in the absorbance of colored compounds such as anthocyanins and carotenoids, while the DPPH• is not (Dastmalchi et al., 2011; Surveswaran et al., 2007). Secondly, the difference in the values could also be caused by the mechanism of the reaction in each case. The reaction in the DPPH• assay depends on the steric accessibility of the compounds to the radical, so small antioxidants with high accessibility will report greater antioxidant activity, but other large ones that react quickly with peroxide radicals can react more slowly, or even be inert to DPPH• due to steric inaccessibility; while for ABTS•+ this problem does not occur (Dastmalchi et al., 2011; Prior et al., 2005). Finally, the difference in values may have been due to the nature of the compounds in the sample. The ABTS•+ radical is soluble in an organic or aqueous medium, which allows it to react with lipophilic and hydrophilic compounds, while the DPPH• radical is only soluble in an organic compound, therefore it only reacts with lipophilic compounds. The antioxidant activity of hydrophilic substances will not be considered (Dastmalchi et al., 2011; Surveswaran et al., 2007).

The assumptions of normality and homoscedasticity were evaluated for the data on the antioxidant activity and phenolic and flavonoid contents, observing normal distribution and homogeneous variances (P = 0.8721). Therefore, Pearson's correlation coefficient was calculated for analysis. There was a negative correlation between the IC50 values for both ABTS•+ and DPPH• in relation to the total phenol and flavonoid contents (r < - 0.5). This is because at a higher levels of flavonoid and phenolic compounds, a lower sample concentration will be necessary to inhibit 50% of the free radicals (IC50), giving them higher antioxidant capacity (Giraldo Vásquez and Ramírez Aristizabal, 2013). The leaf extract (ECn1) and ethyl acetate fraction (Fr3) exhibited strong antioxidant activity (Table 6).

On the other hand, there was a strong correlation (Pearson's correlation coefficient) between total phenolic and flavonoid contents and the values of the percentage of inhibition of ABTS•+ and DPPH• in all samples tested (r > 0.9) (Table 7). This means that these compounds present in the extracts and fractions analyzed are the main responsible for the radical scavenging activity DPPH• and ABTS•+ (Dastmalchi et al., 2011) This has been observed in other studies carried out on plants of the genus Vaccinium (Castrejón et al., 2008; Çelik et al., 2008; Giovanelli and Buratti, 2009). The antioxidant activity of flavonoids and phenolic compounds is caused by their ability to transfer electrons to free radicals in order to stabilize them thanks to their structure (Havsteen, 2002).

Table 7

Pearson's correlation coefficient between the content of total phenols and flavonoids with the percentage of inhibition to DPPH• and ABTS•+ in extracts and fractions of C. nitida.

Extract/Fraction	Pearson correlation coefficient (r)
	DPPH•	ABTS•+
ECn1	0.9732	0.9669
ECn2	0.9735	0.9830
Fr3	0.9291	0.9774
Fr4	0.9833	0.9819

Conclusion

This study analyses, for the first time, the nutritional compositions, the total phenolics, flavonoids, and anthocyanins contents and the antioxidant activity of C. nitida fruit. The results of the nutritional analysis, bioactive compounds and antioxidant activity suggest that C. nitida fruits have potential as very promising edible fruits, useful for human and animal health. However, even though the communities of the region use this berry as food, it is necessary to carry out toxicity tests to evaluate the safety of their food use. These berries are a good source of anthocyanin and phenolics compounds, which are related to excellent antioxidant properties, making them good sources of natural antioxidants, natural colorants, and ingredients of functional foods, and providing added value to the production chain associated with the consumption of native Colombian fruits. In addition, the presence of notable antioxidant activities in the EtOH extract and EtOAc fraction indicates that C. nitida could be used as a crude drug and dietary supplement with potential health benefits. Terpenes such as rimuene, α- and β-amyrin, kaur-16-ene, friedelin and glutinol; the flavonoid 3-O-methyl quercetin and anthocyanins such as delphinidin 3-O-hexoside, cyanidin 3-O-glucoside, petunidin 3-O-glucoside, peonidin 3-O-galactoside, and peonidin 3-O-glucoside are reported here for the first time for the species.

Declarations

Author contribution statement

Elizabeth Gil Archila, Jorge Andres Carvajal Vasquez: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Felipe Rojas-Bautista: Conceived and designed the experiments; Wrote the paper.

Nestor Garcia: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by the [PPTA8374].

Data availability statement

Data included in article/supplementary material/referenced in article.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.