Changes in Pomological and Physical Parameters in Rosehips during Ripening

Rosehips of various Rosa spp. are well known for having human health-promoting compounds like mineral nutrients, vitamins, fatty acids, and phenolic compounds. However, little is known about rosehip characteristics which describe the fruit quality and may indicate appropriate harvest times. Our study evaluated the pomological (width, length, and weight of fruits, flesh weight, and seed weight), texture, and CIE colour parameters (L*, a*, and b*), chroma (C), and hue angle (h°) of rosehip fruits of Rosa canina, Rosa rugosa, and genotypes of Rosa rugosa ‘Rubra’ and ‘Alba’, harvested at five ripening stages (I–V). The main results revealed that genotype and ripening stage significantly affected parameters. The significantly longest (R. canina) and widest fruits (R. Rugosa) were measured at ripening stage V. Genotypes of R. rugosa ‘Rubra’ and ‘Alba’ had significantly higher fruit and flesh weights at ripening stage V. Rosehips of all investigated genotypes expressed darkness (lower L*) during ripening, and had the highest hue angle h° values at ripening stage I while the lowest was at stage V. The significantly lowest skin elasticity of rosehips was found at stage V. However, R. canina was distinguished by the highest fruit skin elasticity and strength. As our results show, the desired pomological, colour, and texture features of various species and cultivars rosehips can be optimised according to the harvest time.


Introduction
With a growing global population and diminishing natural resources, food supply and demand disparity have become more pronounced [1]. Diverse species employed in agricultural systems are crucial for human nutrition and sustainable food systems. Food diversity gained from or near agricultural fields and natural environments such as forests is an additional source of food system resilience [2]. The hunt for nutrient-dense food sources is a matter of great concern in today's food-scarce, multibillion-person world. An unexploited functional food source is the Rosa species cultivated for the ornamental flowers where the fleshy fruits (hypanthium/haw), situated under the perianth, known as rosehips, are removed [3].
As interest in rosehips' nutritional worth has increased, many studies on their morphological traits, phytochemical components, antioxidant capacity, and volatile chemicals have been conducted [1]. In addition, in natural growing conditions, different species and varieties show diversity in most morphological traits [4]. These plants are widely utilised in food; however, the interest in using them in cosmetics and pharmaceutical industries as effective materials for enhancing the quality of final goods with antioxidant activity features is prevailing [5][6][7]. Rosehips have antioxidant and anti-inflammatory, antibacterial, anti-mutagenic probiotic, anti-ulcerogenic, antinociceptive, and anti-carcinogenic properties [8,9]. In a recent study, over 500 s metabolites rosehips from five species were determined, most of which were flavonoids and phenolic acids [1]. In addition, rosehips are

IV
Our results showed that the significantly longest fruits (24.93 mm) were those of Rosa canina, while the widest fruits (31.88 mm) corresponded to Rosa rugosa at ripening stage V. Rosa rugosa 'Rubra' and Rosa rugosa 'Alba' had significantly higher fruit weights (12.53 and 13.79 g, respectively) and flesh weights (11.33 and 12.65 mm, respectively) at ripening stage V than the other genotypes (Table 1).
For each species and cultivar of Rosa spp., the weight of seeds varied from 0.39 to 1.42 g, depending on the ripening stage. Rosa canina had significantly greater seed weights at ripening stage V, while R. rugosa had significantly lower seed weights at ripening stage I. Our results agree with those of previously published studies that referred to pomological characteristics, such as fruit length, width, weight, flesh weight, seed weight, and the fruit-flesh ratio [26,35,36].
The flesh ratio is one of the most important criteria regarding rosehip fruit quality. In developing rosehip cultivars suitable for the industry, cultivars with large-sized fruits and a high flesh ratio are desired [19]. Our results indicated that the flesh ratio was quite variable, ranging from 70.84 % (ripening stage I, R. canina) to 91.88 % (ripening stage V, R. rugosa 'Alba') ( Figure 1). This ratio was highest in R. rugosa, R. rugosa 'Rubra', and R. rugosa 'Alba' in ripening stage V. Previous research has also found a significant level of flesh ratio variability within Rosa L. species. Our results were comparable with the other studies [18,19,36,37], showing flesh ratios of 46.8-100% among rosehip genotypes. flesh ratio variability within Rosa L. species. Our results were comparable with the other studies [18,19,36,37], showing flesh ratios of 46.8-100% among rosehip genotypes. Figure 1. Flesh ratio of rosehip of different species/cultivars at five ripening stages (n = 60). Data are given as the arithmetic mean ± standard deviation. Averages followed by different letters are significantly different at the 5% level of probability (p < 0.05). The effects of species/cultivars, ripening stage, and their interaction are significant at p < 0.0001.

Colour Parameters of Rosehips
CIE colour parameters were determined and are shown in Table 2: L* for the value for lightness, a* value for red/green colour, b* value for yellow/blue colour, hue angle (h°), and C* for the brightness of different genotypes of rosehip fruits at each ripening stage. A two-way ANOVA showed that rosehip species/cultivar, ripening stage, and their interaction significantly influenced the CIE colour coordinates L*, a*, b*, C, h° values. According to Kazankaya et al. [37], rosehip fruits had a base colour of red, dark red, light red, and orange. Our results showed that the value of L* varied differently in all species and/or cultivars during the ripening period ( Table 2). Erogul and Oguz [38] researchers established the L* value of the investigated rosehip fruits from 32.35 to 37.62 among all genotypes. According to these authors, the differences can be caused by environmental conditions and depend on genotype. Our data showed that the fruits of R. canina, R. rugosa, R. rugosa 'Rubra', and R. rugosa 'Alba' expressed darkness during ripening, as demonstrated by the decreasing values of coordinate L*. Significantly higher rosehip lightness values L* were found for R. rugosa at ripening stages I, II, and IV (38.44, 39.24, and 38.07, respectively) and for R. rugosa cv. 'Alba' at ripening stages I, II, and III (39.31, 38.56, and 38.17, respectively). A significantly lowest coordinate L* (26.60) was found in the fruits of R. canina at ripening stage V. A study by Ercisli [20] showed higher colour L* (48.06), a* (41.70), and b* (39.39) values of R. canina rosehip fruits. Shades of red colours were found to be predominant among the investigated genotypes. Other studies showed that the L* value varied from 30.5 to 45.2 in the rosehips harvested from four Sicilian plants of Rosa canina [7].  . Data are given as the arithmetic mean ± standard deviation. Averages followed by different letters are significantly different at the 5% level of probability (p < 0.05). The effects of species/cultivars, ripening stage, and their interaction are significant at p < 0.0001.

Colour Parameters of Rosehips
CIE colour parameters were determined and are shown in Table 2: L* for the value for lightness, a* value for red/green colour, b* value for yellow/blue colour, hue angle (h • ), and C* for the brightness of different genotypes of rosehip fruits at each ripening stage. A twoway ANOVA showed that rosehip species/cultivar, ripening stage, and their interaction significantly influenced the CIE colour coordinates L*, a*, b*, C, h • values. According to Kazankaya et al. [37], rosehip fruits had a base colour of red, dark red, light red, and orange. Our results showed that the value of L* varied differently in all species and/or cultivars during the ripening period (Table 2). Erogul and Oguz [38] researchers established the L* value of the investigated rosehip fruits from 32.35 to 37.62 among all genotypes. According to these authors, the differences can be caused by environmental conditions and depend on genotype. Our data showed that the fruits of R. canina, R. rugosa, R. rugosa 'Rubra', and R. rugosa 'Alba' expressed darkness during ripening, as demonstrated by the decreasing values of coordinate L*. Significantly higher rosehip lightness values L* were found for R. rugosa at ripening stages I, II, and IV (38.44, 39.24, and 38.07, respectively) and for R. rugosa cv. 'Alba' at ripening stages I, II, and III (39.31, 38.56, and 38.17, respectively). A significantly lowest coordinate L* (26.60) was found in the fruits of R. canina at ripening stage V. A study by Ercisli [20] showed higher colour L* (48.06), a* (41.70), and b* (39.39) values of R. canina rosehip fruits. Shades of red colours were found to be predominant among the investigated genotypes. Other studies showed that the L* value varied from 30.5 to 45.2 in the rosehips harvested from four Sicilian plants of Rosa canina [7].
For all investigated Rosa spp., colour changes during ripening were characterised by a significant increase of a* (redness) ( Table 2). All investigated samples exhibited negative values of a* ranging from −1.89 to −4.52 at ripening stages I and II.
All tested rosehip samples were significantly greener at ripening stage I than those harvested at other ripening stages. However, from ripening stages III to V, there was a significant increase in the a* value, changing from negative to positive values (6.96-37.83) in all rosehips. According to the literature, colour variation in fruits is due to changes in the accumulation of pigments like carotenoids and chlorophyll degradation [39]. At ripening stage V, comparing the species and/or cultivars, the redness of rosehip fruits differed. Rosa rugosa showed a significantly higher a* value (37.83), followed by R. rugosa 'Rubra' (35.59), R. rugosa (35.00), and R. rugosa 'Alba' (32.25). Nine Rosa spp. genotypes were investigated by Bilgin et al. [40] and they found a* values from 19.31 to 34.09, which express the red colour, while the L* value was from 21.71 to 36.01 and the b* from 9.75 to 22.57. Data are given as the arithmetic mean ± standard deviation with ANOVA p-value. Averages in the column followed by the different letters are significantly different at the 5% probability level (p < 0.05). The effects of species/cultivars, ripening stage, and their interaction are significant at p < 0.0001.
The value of coordinate b* showed a high level of variation ( Table 2). The colour b* value of fruits was always positive (yellow h • ) for all species/cultivars at all ripening stages. Significantly, the highest value of coordinate b* (41.51) was found for R. rugosa fruits at ripening stage I and the lowest (22.53) for R. canina at stage V, indicating that they became less yellow. In Palermo, Fascella et al. [7] investigated colour parameters for different Sicilian rose species with intense red colour fruits. The colour b* value varied from 14.1 to 26.0. According to these researchers, the rosehip colour may be a valuable predictor of the optimal harvesting time for different Rosa spp. genotypes [7].
Ripening stage and genotype significantly affected the C* values of the rosehips ( Table 2). Our data showed that, as the fruit ripened, the C* values of all species and (or) cultivars increased significantly. For example, in the fruits of R. rugosa, the value of C* was significantly higher at ripening stages IV and V (49.87 and 48.28, respectively). According to Erogul and Oguz [39], the range of the C* value varied from 34.53 to 42.25, which is similar to ours.
In all investigated rosehip fruits, the hue angle (h • ) of all ripening stages was significantly different ( Table 2). All tested rosehip samples had the highest h • values at ripening stage I and the lowest at stage V. R. canina fruits had the significantly highest h • value of 98.39 (green) and the lowest h • value of 32.77 (red). Chae et al. [31] showed that the h • of R. rugosa rosehips also depended on the ripeness stage. h • decreased during ripening due to carotenoid accumulation, which is associated with changes in yellow and red colours [33,41].

Texture Properties of Rosehip Fruits
Ripening is the process by which fruits acquire desirable flavour, quality, colour, palatability, and other textural attributes. However, the textural aspects of the skin of rosehips during ripening have received little attention in the literature. The changes in elasticity during rosehip ripening for different cultivars and/or species are presented in Figure 2.

Texture Properties of Rosehip Fruits
Ripening is the process by which fruits acquire desirable flavour, quality, colour, pa atability, and other textural attributes. However, the textural aspects of the skin of ros hips during ripening have received little attention in the literature. The changes in elasti ity during rosehip ripening for different cultivars and/or species are presented in Figu 2. Figure 2. Elasticity of different species/cultivars at five ripening stages (n = 60). Data are given the arithmetic mean ± standard deviation. Averages followed by the different letters are signi cantly different at the 5% level of probability (p < 0.05). The effects of species/cultivars, ripenin stage, and their interaction are significant at p < 0.0001.
The elasticity of the rosehip skin had a tendency to decrease during the ripening p riod and the significantly lowest elasticity was found at stage V. According to the result Rosa canina presented the highest skin elasticity during ripening. We found significa differences in strength among the cultivars and/or species studied and ripening stage ( Figure 3). Fruits harvested at stages I and II have shown higher skin strength than at othe ripening stages. During ripening, the fruit of all investigated cultivars/species had a d crease in skin strength. The strength parameters of all species and/or cultivars als changed throughout ripening. At ripening stage V, Rosa canina fruit had the strongest sk compared with other genotypes at the same stage. A decrease in the strength of the fru skin during ripening can be associated with the metabolic pathways that are responsib for textural changes in fruits, which are believed to involve loss in turgor pressure, deg radation, and other physiological changes in the composition of membranes, degradatio of starch, and modifications in the cell wall structure and dynamics [42]. Elasticity of different species/cultivars at five ripening stages (n = 60). Data are given as the arithmetic mean ± standard deviation. Averages followed by the different letters are significantly different at the 5% level of probability (p < 0.05). The effects of species/cultivars, ripening stage, and their interaction are significant at p < 0.0001.
The elasticity of the rosehip skin had a tendency to decrease during the ripening period and the significantly lowest elasticity was found at stage V. According to the results, Rosa canina presented the highest skin elasticity during ripening. We found significant differences in strength among the cultivars and/or species studied and ripening stages ( Figure 3). Fruits harvested at stages I and II have shown higher skin strength than at other ripening stages. During ripening, the fruit of all investigated cultivars/species had a decrease in skin strength. The strength parameters of all species and/or cultivars also changed throughout ripening. At ripening stage V, Rosa canina fruit had the strongest skin compared with other genotypes at the same stage. A decrease in the strength of the fruit skin during ripening can be associated with the metabolic pathways that are responsible for textural changes in fruits, which are believed to involve loss in turgor pressure, degradation, and other physiological changes in the composition of membranes, degradation of starch, and modifications in the cell wall structure and dynamics [42].

Texture Properties of Rosehip Fruits
Ripening is the process by which fruits acquire desirable flavour, quality, colour, palatability, and other textural attributes. However, the textural aspects of the skin of rosehips during ripening have received little attention in the literature. The changes in elasticity during rosehip ripening for different cultivars and/or species are presented in Figure  2. Figure 2. Elasticity of different species/cultivars at five ripening stages (n = 60). Data are given as the arithmetic mean ± standard deviation. Averages followed by the different letters are significantly different at the 5% level of probability (p < 0.05). The effects of species/cultivars, ripening stage, and their interaction are significant at p < 0.0001.
The elasticity of the rosehip skin had a tendency to decrease during the ripening period and the significantly lowest elasticity was found at stage V. According to the results, Rosa canina presented the highest skin elasticity during ripening. We found significant differences in strength among the cultivars and/or species studied and ripening stages ( Figure 3). Fruits harvested at stages I and II have shown higher skin strength than at other ripening stages. During ripening, the fruit of all investigated cultivars/species had a decrease in skin strength. The strength parameters of all species and/or cultivars also changed throughout ripening. At ripening stage V, Rosa canina fruit had the strongest skin compared with other genotypes at the same stage. A decrease in the strength of the fruit skin during ripening can be associated with the metabolic pathways that are responsible for textural changes in fruits, which are believed to involve loss in turgor pressure, degradation, and other physiological changes in the composition of membranes, degradation of starch, and modifications in the cell wall structure and dynamics [42]. During fruit ripening, various biochemical changes occur, including seed maturation, colour change, texture softening, volatile taste generation, wax formation on the skin, tissue permeability, and changes in the content of carbohydrates, organic acids, and proteins. These ripening-related changes account for fruit skin elasticity and strength variations during fruit maturity 28]. Unfortunately, as far as we know, there is no research on rosehip fruit skins.

Correlation among Investigated Traits
The correlation matrix between the pomological, colour, and textural properties of rose hip species/cultivars is shown in Table 3. Strong positive correlations were found between fruit weight and flesh weight, seed weight, and fruit length (r = 0.999, r = 0.778, and r = 0.856, respectively). According to other researchers, the highest correlation was that between fruit weight and fruit width (r = 0.898) [40].  Our study showed that fruit length had a negative correlation with L * , b*, and h • values (r = −0.751, r = −0.641, and r = −0.933, respectively) and a high positive correlation (r = 0.910) with the a* value. In comparison, values of h • were strongly negative for a* and C (r = −0.933 and r = −0.823, respectively). Bilgin et al. [40] discovered highly negative associations between fruit weight, stone weight, and width and colour L*, a*, b*, and C* values. In our study, skin elasticity negatively correlated with fruit weight, flesh weight, and fruit width (r = −0.737, r = −0.741, and r = −0.876, respectively). As well, fruit skin strength correlated negatively with fruit weight, flesh weight, fruit length, and a * (r = −0.878, r = −0.880, r = −0.800, and r = −0.855) while it positively correlated with the h • value (r = 0.839). No correlation was found between skin elasticity, C, and rosehip L* and b* values.

Principal Component Analysis
The relationships between rosehip samples of different species/cultivars harvested at five ripening stages and the pomological (Table 1), colour (Table 2), and textural (Figures 2 and 3) properties were examined using a principal component analysis (PCA). The PCA findings showed that the first two axes (PC1: 72.45% and PC2: 16.19%) explained 88.64% of the total variance ( Figure 4). In addition, the first principal component's (PC1) and second principal component's (PC2) eigenvalues were both greater than one (8.69 and 1.94, respectively). of the total variance ( Figure 4). In addition, the first principal component's (PC1) and second principal component's (PC2) eigenvalues were both greater than one (8.69 and 1.94, respectively). All pomological characteristics and colour characteristics a* and C were strongly and positively associated with PC1. In contrast, the skin's colour parameters L*, h•, and textural properties (strength and elasticity) were negatively associated with PC1. PC2 was positively associated with colour coordinate b*. As illustrated in Figure 4, PC1 separated the rosehip samples harvested at ripening stages I, II, and III from the rosehip samples harvested at ripening stages IV and V. The fruits harvested at ripening stages IV and V are defined by higher width, length and weight, and also, greater values of colour characteristics a* and C. However, these samples contained lower values of colour parameters L*, h°, skin strength, and elasticity compared to fruits harvested at ripening stages I, II, and III. All pomological characteristics and colour characteristics a* and C were strongly and positively associated with PC1. In contrast, the skin's colour parameters L*, h•, and textural properties (strength and elasticity) were negatively associated with PC1. PC2 was positively associated with colour coordinate b*. As illustrated in Figure 4, PC1 separated the rosehip samples harvested at ripening stages I, II, and III from the rosehip samples harvested at ripening stages IV and V. The fruits harvested at ripening stages IV and V are defined by higher width, length and weight, and also, greater values of colour characteristics a* and C. However, these samples contained lower values of colour parameters L*, h • , skin strength, and elasticity compared to fruits harvested at ripening stages I, II, and III.

Field Experiments
This research was conducted in 2018-2020 at the organic farm in Pakruojis District, Lithuania (field coordinates 56 • 10 29.0" N 23 • 49 02.6" E). Two species, Rosa rugosa and Rosa canina and two cultivars, Rosa rugosa cv. 'Rubra' and Rosa rugosa cv. 'Alba' were planted in 2011. The distance between the rows was 4 m, and the distance between rosehip shrubs was 2 m. The interrow was loosened, and weeds were manually removed. The soil pH at the experimental site was from 6.8 to 7.0 mg kg −1 plant-available potassium from 97.7 to 181.0 mg kg −1 , available phosphorus from 120.6 to 153.3 mg kg −1 , and total nitrogen from 2.5%. The experimental plots were arranged in the randomised block with four replicates per treatment. Each replication consisted of four plants of each species/cultivar. Fruits were randomly collected for analysis from July to September, depending on the ripening stage. For the analyses, 1 kg of fruits was randomly harvested from each block of each treatment.
The rose fruits were harvested at five ripening stages (I-V; Figure 5): (I) was reached when fruit colour changed slightly from green to yellow, pink, or red in no less than 10% of its surface; (II) fruit colour changed from green to tarnish-yellow, pink, or red in no less than 30% of the surface; (III) fruit colour changed from green to light orange or red or a combination thereof in no less than 60% of the surface; (IV) fruit became pinkish or orange depending on the species; (V) fruit surface was red.
of each treatment.
The rose fruits were harvested at five ripening stages (I-V; Figure 5): (I) was reached when fruit colour changed slightly from green to yellow, pink, or red in no less than 10% of its surface; (II) fruit colour changed from green to tarnish-yellow, pink, or red in no less than 30% of the surface; (III) fruit colour changed from green to light orange or red or a combination thereof in no less than 60% of the surface; (IV) fruit became pinkish or orange depending on the species; (V) fruit surface was red. Compared to standard climate normal (SCN), in 2018, 2019, and 2020, the temperature was higher by 2.5, 1.4, and 0.7 ℃. During the rosehip vegetation period of 2018, 2019, and 2020, the average precipitation was 86, 106.6, and 72.5 mm less than in the SCN. In addition, in 2018, 2019, and 2020, the sunshine was 195, 118, and 8 h longer than SCN (Table 4).  Compared to standard climate normal (SCN), in 2018, 2019, and 2020, the temperature was higher by 2.5, 1.4, and 0.7°C. During the rosehip vegetation period of 2018, 2019, and 2020, the average precipitation was 86, 106.6, and 72.5 mm less than in the SCN. In addition, in 2018, 2019, and 2020, the sunshine was 195, 118, and 8 h longer than SCN (Table 4).

Soil Agrochemical Analyses
The soil agrochemical characteristic was evaluated in the Vytautas Magnus University Agriculture Academy Laboratory of Food Raw materials, Agronomical and Zoo-technical Investigations. From the soil samples, the small stones, roots, and other organic plant parts were removed, then air-dried in open plastic boxes and crushed. Homogenised soil was sieved through a 1 mm mesh size sieve. Soil samples were analysed for pH in potassium chloride (KCl) solution (pH KCl ), plant-available phosphorus, potassium, and total nitrogen. Soil pH KCl was established by the potentiometric method in 1N KCl extract [43]. Plant-available phosphorus and potassium were extracted with ammonium lactate according to the Egner-Riehm-Domingo method [44]. Total nitrogen concentration (mg kg −1 ) was determined by the Kjeldahl method using Kjeldatherm (Gerhardt, Königswinter, Germany) [45].

Pomological Parameters
Fruit length and width in mm were measured with a digital calliper. The fresh fruit biomass (g), flesh and seed weights (g) were measured using a technical balance (Radwag, WPS 510/C/1, Radom, Poland). The results are displayed in grams with a precision of 0.01 g. The fruit-to-flesh ratio was calculated by dividing flesh weight by fruit weight and multiplying by 100. The pomological fruit parameters were calculated as the mean of 60 (3 replicates × 20 fruits) fruits.
Chroma presents the quality and intensity of fruit colour. Calculations were performed on colour attributes chroma (C), using the following formula: C = (a * · 2) + (b * · 2) ·1 2 ; (1) Hue angle (h • ) was calculated using the following formula: The colour parameters were calculated as the mean of 60 (3 replicates × 20 fruits) fruits.

Fruit Skin Texture Analysis
The skin elasticity and strength of rosehips were measured using a texture analyser, TA.XTPlus, with a P/2 probe (Stable Micro Systems, Godalming, UK).
The probe descended into the sample at a speed of 1 mm s −1 and reached a depth of 5 mm. Skin strength is the force (N) required to puncture the fruit skin, whereas skin elasticity is the distance (mm) to which the skin deflects before probe penetration.
The texture parameters were calculated as the mean of 60 (3 replicates × 20 fruits) fruits.

Statistical Analysis
Data were analysed using Microsoft ® Excel ® 2016 MSO and confirmed using the STA-TISTICA 10 (StatSoft, Inc., Tulsa, OK, USA, 2010). All analyses were conducted in triplicate, with the mean and standard deviation presented. Since the analysis of variance did not show year interaction, the data are presented as two-year averages. The samples were defined by two qualitative factors (species/cultivars and ripening stage), and therefore, the reliability of the results was evaluated by a two-way analysis of variance (ANOVA) method. Tukey's test was applied to assess significant differences between the means at p < 0.05. The nature and strength of the correlations between the variables were determined using correlation analysis. Finally, the relationships between rosehip samples of different species/cultivars harvested at five ripening stages and the pomological, colour, and texture properties were evaluated with the XLSTAT 2018 (New York, NY, USA) using principal components analysis (PCA).

Conclusions
The results of this experimental investigation revealed that the genotype and ripening stage significantly affected the fruit's pomological, colour, and textural parameters. According to our results, appropriate for transportation was Rosa canina having the strongest fruit skin elasticity (3.03 mm) and strength (17.36 N). R. rugosa, R. rugosa 'Rubra', and R. rugosa 'Alba' are suitable for the industry because these species/cultivars had the largest-sized