Analysis of the contents of Ugni molinae Turcz fruits across the ripening stages

The murta fruits used in this study were harvested from wild plants that grow on small farms as a living fence in the Pelluhue town, Maule region, Chile (S 35°50′00″; W 72°38′00″). Pelluhue town has a Mediterranean climate with warm and dry summers and cold winters. The average annual temperature in Pelluhue is 20°C. For this reason, the fruits were collected at 9:00 at an atmospheric temperature ranging from 14°C to 16°C, and one or two fruits were collected per plant. The fruits were obtained from 20-year-old half sib plants of murta grown. Initially, the plants were obtained from a small group of 3–5 plants growing wild near the property that is now a commercial farm. To categorise the fruits into different developmental stages, criteria such as weight, size and fruit skin colour were considered (refer to Figure 1). The four stages are defined as follows: small green (SG) fruit, large green (LG) fruit, half ripe fruit (50%R) and fully ripe fruit (R). A total of 150 fruits were collected from each developmental stage, obtained from 20 different plants, across two consecutive growing seasons (fall of 2020 and 2021). Each fruit was rapidly frozen in liquid nitrogen and stored at –80°C until further analysis. After harvesting, the fruits were transported to the Multidisciplinary Agroindustry Research Laboratory at the Universidad Autónoma de Chile in Talca, which is approximately 150 km away from the harvest site, under cold conditions in containers having dry ice, maintaining the fruits between 4°C and 8°C.

Figure 1:

In the present study, 25 fruits from each developmental stage (free from external damage) were selected for the assessment of fruit skin colour using a Nix Pro 2 Color Sensor (Nix Sensor Ltd., Hamilton, Ontario, Canada) (https://www.nixsensor.com/free-color-converter/). The surface colour of the different fruits was characterised and represented using the Hunter scale, which includes parameters such as L*, a*, b*, C*, and h°. Additionally, the weight (in grams) and size (in millimetres) of the berries were measured. Values indicate the mean of 25 replicates from each season, and the standard deviations are also shown. To determine the pH, the soluble solids concentration (SSC) and titratable acidity (TA), the juice extracted from 2 g of frozen tissue was ground with liquid nitrogen, homogenised in 5 mL of distilled water and filtered using miracloth. The results for TA and SSC were expressed as the SSC/TA ratio, as described by Ramos et al. (2018) and Castro et al. (2023).

For the analysis of volatile compounds in murta berries during ripening, three different samples were subjected to SPME–GC–MS analysis. The stages SG and LG were combined to form a single batch representing the unripe stage, referred to as green (G) samples. Additionally, the 50% R stage was used to represent the stage before full ripening, while the R stage represented the fully ripe stage.

To perform the analysis, approximately 3 g of murta berries (consisting of 9–14 berries) were mixed with 8 mL of ultrapure water and 10 μL of a 4-methyl-2-pentanol solution (used as an internal standard) in methanol (0.1% v/v). The mixture was homogenised in a 50 mL centrifuge tube using an Ultra-Turrax^TM homogeniser (Janke & Kunkel GmbH & Co. KG, Staufen, Germany). The homogenised sample was transferred to a 20 mL glass vial and incubated for 20 min at 45°C with agitation at 500 rpm using an MPS Autosampler (Gerstel, Palo Alto, CA, USA). Subsequently, a 2 cm fibre (50/30 μm carboxen/divinylbenzene/polydimethylsiloxane [Supelco, Bellefonte, PA, USA]) was exposed to the sample for 40 min.

The analysis was performed using a 7890B Agilent GC system coupled to a quadrupole mass spectrometer Agilent 5977 inert (Agilent Technologies, Palo Alto, CA, USA). The chromatographic conditions and identification criteria were consistent with those previously published in Parra-Palma et al. (2019).

Each of the three fruit stages was analysed in triplicate, and the results are expressed as relative areas (RA). The RA values were divided by the number of berries used in each replicate to obtain results on a per berry basis. Furthermore, the values were divided by the mass of berries (in grams) to obtain results on a per gram basis. The terpene compounds were categorised based on their carbon backbone, following the classification presented in Figure 2.

Figure 2.

To extract the compounds from the berries, the entire berries were homogenised using a mortar and pestle in a 1% HCl solution prepared in methanol (5 L · kg^–1 of fruit). The homogenisation process aimed to break down the berry tissue and facilitate the extraction of the compounds. The resulting mixture was stirred for 1.5 hr at room temperature to allow for proper extraction. Subsequently, the mixture was centrifuged at 4,200 g, and the supernatant was collected for further analysis.

For each stage, three independent extractions were performed using 10 g of fruit tissue. This ensured the extraction of a sufficient amount of compounds for accurate analysis. Each extraction was carried out following the same procedure of homogenisation, stirring, and centrifugation.

The obtained extracts were utilised to determine the total phenolic content using the Folin–Ciocalteu reagent, a widely used method for quantifying phenolic compounds (Singleton et al., 1999). The total anthocyanin content was determined using the pH differential method described in Lee et al. (2008), which relies on the colour changes exhibited by the anthocyanins at different pH levels. The total flavonol content was determined using a colorimetric method after the addition of aluminium chloride, as reported by Chang et al. (2020). These methods allowed for the quantification of the total phenolic, anthocyanin, and flavonol contents present in the murta berry extracts, providing valuable information about the composition of these compounds at different developmental stages. The phenolic compounds were quantified based on a standard curve of gallic acid; the results are expressed as gram of gallic acid equivalents per kilogram of fruit (g · kg^–1 GAE) and correspond to the means ± SEs of three biological replicates with two more technical replicates. Meanwhile, the flavonol results were expressed as gram of quercetin equivalents per kilogram of fruit (g · kg^–1 quercetin). Finally, the total anthocyanin content was expressed as gram of cyanidin 3-glucoside equivalents per kilogram of fruit (g · kg^–1 Cy3G) using the absorbance values of A = (A524–A700 nm) pH 1.0 - (A524–A700 nm) pH 4.5, with a molar extinction coefficient of 26,900 according to Parra-Palma et al. (2020).

To assess the radical-scavenging ability of the methanol extracts, the 1,1-diphenyl-2-picrylhydrazyl (DPPH) discolouration assay was conducted following the method described by Cheel et al. (2007). The procedure involved mixing an aliquot of 20 μL of the methanol extract with a DPPH solution (0.5 mM, 0.25 mL) and acetate buffer (100 mM, pH 5.5, 0.5 mL), as outlined in Castro et al. (2019). The reaction mixtures were prepared in triplicate and then measured at 517 nm using an Infinite^® 200 PRO NanoQuant spectrophotometer (Infinite M200; Tecan, Zurich, Switzerland). The extent of DPPH radical scavenging was determined by comparing the results with a negative control group consisting of DPPH with methanol.

Furthermore, the ferric reducing antioxidant power (FRAP) analysis was performed using the same methodology as described in Castro et al. (2018). This assay is commonly used to measure the antioxidant capacity of samples. The FRAP analysis involves mixing the methanol extract with a ferric-tripyridyltriazine (Fe³⁺-TPTZ) complex, which is reduced by the antioxidants present in the extract. The reduction reaction results in the formation of a blue-coloured ferrous tripyridyltriazine (Fe²⁺-TPTZ) complex, which can be measured spectrophotometrically. The FRAP values obtained provide an indication of the reducing power and antioxidant capacity of the extracts.

By conducting both the DPPH discolouration assay and the FRAP analysis, the radical-scavenging and antioxidant abilities of the methanol murta berry extracts can be evaluated, providing insights into their potential health benefits and functional properties.

The size of the berries at each of the four selected developmental stages was analysed, taking into consideration their weight and width. The results are presented in Figure 3. Regarding berry mass, there is an increase from SG to LG, as well as from 50%R to R stages. However, there were no significant differences in berry mass between the LG and 50%R stages (Figure 3A). These findings suggest a growth stop similar to what has been described for grape berries (Vitis vinifera), coinciding with the onset of colour biosynthesis known as veraison (Kennedy, 2002). On the other hand, when considering the fruit width (Figure 3B), a gradual increase in berry size was observed from SG to R, although there were no statistical differences between the 50%R and R developmental stages. This contradicts the growth stop hypothesis proposed based on berry weight since an increase in size is observed between the LG (green berries, Figure 1) and 50%R (slightly coloured, Figure 1) development stages. This aspect should be considered for yield calculations. The soluble solids concentration (SSC) of the berries increases from LG to R (Figure 3E), reaching 27°Brix for fully ripe murta berries. This increase in SSC throughout the ripening aligns with the common trend observed in other fruits (Zhao, 2007). A similar increasing trend was observed for TA (Figure 3F), which rises as ripening progresses, explaining the decrease in pH (Figure 3C) as maturity advances. Finally, the SSC/ TA ratio also increases with ripening (Figure 3D), which is a common trend in fruit maturity and has been linked to the sweet–sour balance and consumer acceptance (Endrizzi et al., 2009).

Figure 3.

Currently, it is widely accepted that fruit colour serves as an attractant for predators, particularly vertebrates, to disperse seeds away from the parent plant. When the seeds are not ready to germinate, fruits tend to be green to blend in with the leaves and often contain toxic or unappealing chemical compounds to discourage dispersers. In contrast, when the seeds are ready to germinate, fruits become colourful and nutritious, enticing dispersers (Simms, 2013). Thus, the evolution of colour during berry-ripening is an important aspect that has been associated with other sensory characteristics of fruits (Li et al., 2019a). In the conditions of this study, the colour trend of murta berries follows a common pattern (Table 1), where the colour of SG and LG stages is similar, displaying green tones, while the red component increases significantly from LG to R, as clearly demonstrated by the rise in the a* component. The colour of fruits at harvest can influence consumer preference (López Camelo and Gómez, 2004; Lewers et al., 2020), which is significant considering the consumption of fresh murta fruits. When considering only fully ripe stage fruits (R) and disregarding postharvest changes, the berries exhibited vibrant colours, potentially offering a pleasant balance between sweetness and sourness, which are desirable traits for high consumer acceptance in terms of organoleptic attributes (Endrizzi et al., 2009).

The total phenolic compounds, as shown in Figure 4A, were analysed using the Folin–Ciocalteu reagent, which includes compounds such as ascorbic acid, phenolic and cinnamic acids, tannins, flavanols, flavonols and anthocyanins (Singleton et al., 1999). As seen in Figure 4A, the total phenolic content of murta berries gradually decreases from the SG to the R developmental stages. This decrease in phenolic content during ripening can be related to the increase in the fruit’s sensory appeal, as many phenolics have traditionally been associated with the perception of bitterness and astringency in food (Ferrer-Gallego et al., 2014; Soares et al., 2013). The decrease in total phenolic concentration can be attributed to several factors, such as dilution due to an increase in berry volume (often associated with an increase in fruit water content), degradation reactions (Tomás-Barberán and Espín, 2001) or lignification of seed tissues (Barros et al., 2015). A similar decreasing trend in total polyphenolic content during ripening has been observed in other berries such as blackberries, raspberries, blackberries and strawberries (Wang and Lin, 2000; Zhao, 2007; Li et al., 2019b).

The colour of the berry peel at each described development stage is closely related to its chemical composition. Specifically, the increase in the red component and the decrease in the green component have been associated with the accumulation of anthocyanins, which are a family of flavonoid natural pigments commonly found in fruits and flowers, as well as during the degradation of chlorophyll (Parra-Palma et al., 2019). The content of anthocyanins at each developmental stage is shown in Figure 4C. Anthocyanins were not detected in SG berries, indicating that the seeds were likely still immature (Wang and Lin, 2000; Zhao, 2007). Although LG berries appear green like those in the SG stage, a low content of anthocyanins was found, suggesting that this stage corresponds to the beginning of anthocyanin biosynthesis, which continues until the R stage, explaining the increasing red colour observed (Table 1). The increasing trend of anthocyanins throughout ripening has been described for other berries (Wang and Lin, 2000; Zhao, 2007) and ensures that there are sufficient colour differences between the ripe fruits and plant leaves to attract predators.

Figure 4.

Finally, considerable amounts of flavonols have been found in ripe murta berries, which were quantified using a colorimetric method after the addition of aluminium chloride (Chang et al., 2020). According to the results shown in Figure 4B, the highest content of flavonols is observed in SG berries. Flavonol content significantly decreases between the SG and LG developmental stages, and no further variations were observed from LG to R. In our conditions, flavonols and anthocyanins exhibit inverse trends: while anthocyanin contents increase throughout maturity, the flavonol contents of ripe fruits are lower than those of unripe (SG) fruits, as also reported for raspberries (Wang et al., 2009).

The antioxidant capacity of murta berries gradually increases from the SG to R developmental stages, as analysed using both the FRAP and DPPH methods, as shown in Figure 5. It is logical that both methods exhibit the same trend for antioxidant capacity throughout ripening, as a strong correlation has been established between the FRAP and DPPH methodologies (Martínez et al., 2022). However, it is noteworthy that the antioxidant capacity tends to positively correlate with the total phenolic compounds, which contradicts the results of the present study. Although the total phenols decrease during the ripening of murta berries, the anthocyanin content clearly increases concurrently, which can explain the rise in antioxidant capacity as the berries mature. The progressive increase in the antioxidant capacity observed during the development of murta berries aligns with the reported data for blackberry hybrids (Siriwoharn et al., 2004), but it does not correlate with studies on other berries, which demonstrate a decrease in antioxidant capacity from red to pink stages, followed by an increase from pink to red stages (Zhao, 2007; Wang and Lin, 2000). Nevertheless, fully ripe murta berries exhibit the highest levels of antioxidant capacity. Thus, the consumption of fully ripe murta fresh fruits, which possess desirable sensory traits as mentioned earlier, could be healthier than consuming unripe berries, at least in terms of their antioxidant activity (Shahidi and Ambigaipalan, 2015). Considering the preservative properties of phenolic compounds and the higher antioxidant activity of the ripest berries, it can be expected that preserved or fermented foods made from mature murta berries would possess superior qualities.

Figure 5.

The profile of VOCs of murta berries was performed for three developmental stages: unripe (G, corresponding to a mixture of SG and LG berries), half-ripe (50%R) and full-ripe (R) in triplicate. The obtained VOCs profiles consists of 70 compounds found in all samples, corresponding to alcohols, aldehydes, esters, and mainly terpenoids (including monoterpenoids and sesquiterpenoids), which represent more than 90% of the identified compounds. The huge amount of identified terpenoids could be surprising when compared to previously published data (Scheuermann et al., 2008), in which few terpenoids were reported. These differences could be related not only to methodology issues, but also to the origin of the berries. It has been reported that, when eucalyptus trees are in the edges of the vineyards, the vines closer to the trees contain higher contents of 1,8-cineole (Capone et al., 2012). Thus, the environment of developing fruits could affect their VOC profile, and the wild origin of the employed murta samples could explain the high account of identified terpenoids. Nevertheless, murta plants must have active biosynthetic routes for terpenoids since several works highlight its triterpenoid (Schreckinger et al., 2010) and sesquiterpenoid (Siani et al., 2016) contents.

The VOC profile of fruits is important, considering that it is related with the aromatic perception of consumers, either for fresh fruit or derivate products consumption (Xu et al., 2019). In that sense, the relative abundance of C6 compounds could be of interest, given that such compounds have been related with herbal scents (Nakamura and Hatanaka, 2002). As shown in Figure 6, the C6 compounds (including hexanal, 2-hexenal, 1-hexanol and cis-3-hexen-1-ol) dropped between 50%R and R berries when expressed as R.A. · g^–1 FW; those are related with its potential impact on the aromatic perception of murta berries. However, no statistical differences were found among the ripening stages when the relative abundance of C6 compounds was expressed as R.A. · berry^–1; these are related to the biosynthesis/degradation ratio. In contrast, the results for ethyl octanoate, the ester with the highest relative abundance, remain constant among development stages when expressed in R.A. · g^–1 FW, but gradually increase through development when expressed in R.A. · berry^–1. Esters has been related with fruity scents; specifically, the sensory descriptors for ethyl octanoate are fruity, sweet, ripe banana and pear. Thus, the results seem to indicate that the berries produce ethyl octanoate continuously throughout their development, but the increase of berry volume hides such production, maintaining constant its relative abundance. In contrast, the production of C6 compounds seems to be inactive throughout the development of fruit, decreasing its relative abundance between the 50%R and R fruits due to the dilution provoked by the increase of berry size.

The identified terpenoids correspond to monoterpenoids (C10 compounds, made up of two units of isoprene) and sesquiterpenoids (C15 compounds, made up of three units of isoprene). When expressed as RA · g^–1 FW, even acyclic (aliphatic), monocyclic and bicyclic monoterpenoids showed the same trend, slightly decreasing from G to 50%R, to increase then significantly from 50%R to R. In contrast, when expressed as R.A. · berry^–1, the acyclic monoterpenoids just increase at the late phase of ripening, between the 50%R and R stages, while the monocyclic terpenoids gradually increase from the G to R stages. The total relative abundance of monoterpenoids (Table 2) clearly increases from the 50%R to R stages, regardless of the expression units, indicating an active biosynthesis, enough to surpass the dilution due to berry growth. In contrast, sesquiterpenoids did not show any statistical differences among the developmental stages of murta berries, probably due to the fact that they are consumed and produced simultaneously.

The identified terpenoids present in the murta berries profile were grouped according to their carbon backbone as shown in Figure 6. The four groups belonging to the monoterpenoids group showed the same trend as the total monoterpenoids, increasing its relative abundances at late ripening (from 50%R to R), either expressed as RA · g^–1 FW or RA · berry^–1 (Figure 7). Otherwise, none of the groups belonging to the sesquiterpenoids showed any statistically significant differences among the development stages. Finally, the total relative abundance of all the terpenoids did not show statistical differences when expressed as RA · g^–1 FW, but it showed a rise between the 50%R and R berries when expressed as RA · berry^–1, which again indicates an active biosynthesis during the last phase of ripening.

Figure 6.

Figure 7.

Volatile terpenoids have been related with fruity and floral scents, and they have been described as the most-important group of metabolites impacting the sensory perception and the consumers’ acceptability of blueberries (Ferrão et al., 2022). Moreover, health-beneficial biological activities for these compounds have been reported, including antioxidant, anti-allergenic, anti-inflammatory, antimicrobial, and anticancer activities (Feng et al., 2020; Masyita et al., 2022). Thus, considering the contents of terpenoids during the development and ripening of murta berries, a good ripening degree of harvested fruits ensures more pleasant sensory attributes and more beneficial effects on consumers’ health, traits that favour their potential use as superfood.