Morphological and Biochemical Diversity in Fruits of Unsprayed Rosa canina and Rosa dumalis Ecotypes Found in Different Agroecological Conditions

dumalis Ecotypes Found Abstract: The Rosa is one of the most diverse genera in the plant kingdom and, in particular, its fruits have been used for multiple purposes in different parts of the world for centuries. Within the genus, Rosa canina and Rosa dumalis are, economically, the most important species and dominate Rosa fruit production. In this study, some important fruit and shrub traits of ten Rosa canina and ten Rosa dumalis ecotypes collected from rural areas of Kars province, located in the east Anatolia region of Turkey were investigated. We found signiﬁcant differences among ecotypes in most of the morphological and biochemical traits. The ecotypes were found between 1446–2210 m altitude. Fruit weight and fruit ﬂesh ratio ranged from 2.95 g to 4.72 g and 62.55% to 74.42%, respectively. SSC (Soluble Solid Content), Vitamin C, total phenolic, total ﬂavonoid, total carotenoid, and total anthocyanin content of the ecotypes ranged from 16.9–22.7%, 430–690 mg per 100 g FW (fresh weight), 390–532 mg gallic acid equivalent per 100 g FW, 0.88–2.04 mg per g FW, 6.83–15.17 mg per g FW and 3.62–7.81 mg cyanidin-3-glucoside equivalent per kg, respectively. Antioxidant activity was determined to be between 19.7–34.7 mg ascorbic acid equivalent per g fresh weight. Rosa ecotypes contained chlorogenic acid and rutin the most as phenolic compound. Our results indicated great diversity within both R. canina and R. dumalis fruits.


Introduction
Wild plants, including edible or less known fruits have unique taste and are easily found in nature. They are an important employment and income sources for rural peoples [1][2][3][4].
Opposite to cultivars, wild edible fruits more resistant to adverse soil and climatic conditions. Most wild fruit species had high content of phytochemicals that vital for human health compared with cultivated one [5]. They have rich gene combinations that could be important for breeding new commercial cultivars with improved aroma and resistance to biotic and abiotic stressors [6].
Among wild edible fruits, the Rosa species have special importance. The species are treasured mainly for their fruits (rose hips). Apart from edible fruit, seeds, flowers etc. valued and used as food, medicine, fodder, fuel, agriculture, tools, fencing and ritualistic aspects, [7][8][9].
Over 30 Rosa species are found within Turkey's flora, Rosa canina and Rosa dumalis are the most common and recognized for their better fruit characteristics providing a rich source of Rosa products. Native to cold and highland areas in Turkey, Rosa canina and Rosa dumalis are widely cultivated in Turkey [10]. In the country, it is widespread from east Anatolia to the Aegean region and from the Mediterranean to the Black sea region, revealing high environmental adaptability [7,9]. Previous studies on rose hips in Turkey aimed to describe a certain degree of heterogeneity which is normally present in cross-pollinating plants in any natural population [7][8][9][10].
In rural areas rose hip is one of the main sources of income [10]. The hips of Rosa canina and Rosa dumalis have been traditionally used in Turkey for centuries and, in particular during winter months, they can be processed into several products such as marmalade, syrup, jam, etc. In natural growing conditions, both species show diversity in most of the morphological traits. The fruits of Rosa canina and Rosa dumalis have been known as medicinal plants for a long time [11,12]. The medicinal potential of Rosa species is based on their antioxidant effects, associated with phytochemical content includes high ascorbic acid, carotenoids and phenolic compounds [13][14][15][16].
It is obvious that there was an increasing interest in nutraceuticals and functional foods. Wild edible fruits are rich sources of nutraceuticals and studies concentrated on their quality and bioactivity [17][18][19][20]. Wild edible fruits exhibit diverse morphology, fruit quality, yield and phytochemicals, compared to cultivated ones, and all those traits can be influenced by environmental conditions and ecotypes [21][22][23][24][25]. Thus, ecotype selection is important task for appropriate varieties' development. Therefore, it is crucial to make detailed comparative studies on rose hip ecotypes related to important plant traits and phytochemical content. Thus, identification of the promising (elite) ecotypes for developing rose hip at a commercial scale is significant.
In the literature, there was limited information about the comparison of morphological and biochemical traits in fruits of R. canina and R. dumalis. Thus, in this study, we aimed to determine and compare some important morphological and biochemical features of R. canina and R. dumalis ecotypes, naturally growing in Kars province, located in east Anatolia, due to their potential application in functional foods.

Plant Material
In this study, ripe fruit of R. canina and R. dumalis naturally grown in Kars province, located in the east Anatolia of Turkey, were used. Fruits were harvested during September in 2017 and 2018 from 20 pre-selected ecotypes of R. canina and R. dumalis. They were pre-selected according to their higher yield, being pest and disease free and more attractive bigger fruit characteristics. Rosa ecotypes are generally found in rural areas in Turkey, and they were formed as a result of the scattering of seeds naturally. The materials used in this study were also carried out on ecotypes formed from naturally scattering seeds showing a heterogeneity situation.

Sampling and Morphological Parameters
Samples consisted of 80 randomly harvested fruits from different parts of shrubs. Samples were transferred to the laboratory for measurements of fruit fresh weight, flesh ratio and fruit biochemical analyses, while yield and thorn characteristics of the respective plants were determined by on-site observations. Fruit weight was determined with digital balance. To calculate the fruit-flesh ratio, the fruit flesh weight/fruit weight × 100 equation was used.

Sample Preparation and Extraction
For the analyses of biochemical content, the harvested fruit was immediately frozen and stored at −80 • C until further analysis. During the analysis, the frozen fruits were Sustainability 2021, 13, 8060 3 of 12 taken and thawed to 24-25 • C. A laboratory blender was used to homogenise the fruit samples (100 g lots of fruits per ecotype) and a single extraction procedure (taking 3 g aliquots transferred inside tubes and extracted for 1 h with 20 mL buffer including acetone, water (deionized), and acetic acid (70:29.5:0.5 v/v) [26]) was used.

Total Phenolic Contents
Folin-Ciocalteu method according to Singleton and Rossi [26] was used for determination of total phenolic content (TPC) of the samples. The TPC results was expressed as mg of gallic acid equivalents (GAE) per 100 g of fresh sample.

Total Carotenoid Content
Total carotenoid content was determined according to Lichtenthaler [27] and results of total carotenoid content was expressed as mg per g fresh fruit sample.

Total Flavonoid Content
Total flavonoid content was determined according to Chang et al. [28] and results were expressed as mg quercetin equivalent (QUE) per g FW.

Total Anthocyanin Content
pH differential method of Giusti and Wrolstad [29] was used to determine total anthocyanin amount and results was expressed as mg of cyanidin-3-glucoside equivalent in per kg of fresh sample.

Antioxidant Capacity
According to Nakajima et al. [30], DPPH scavenging activity of rosehip extracts was calculated and given as mg ascorbic acid equivalent (AAE) per g fresh weight.

Phenolic Compounds
Phenolic compounds were detected with a modified HPLC procedure suggested by Rodriguez-Delgado et al. [31]. Phenolic compounds were expressed as µg per g FW.

Statistical Analysis
The data of both years were pooled because there were no differences between years. SPSS software and procedures used for analysis and Least Significant Difference (LSD) method at p < 0.05 was used to analyze of variance tables. In addition, principal component analysis (PCA) was performed to determine the relationships among ecotypes. Furthermore, Pearson analysis was applied to identify the correlation among the measured main morphological and biochemical traits and investigate the relations between them.

Morphological Traits of R. canina and R. dumalis Ecotypes
Location, altitude, thorn, yield, fruit weight and fruit flesh ratio characteristics of 20 Rosa canina and R. dumalis ecotypes are shown in Table 1. As indicated in Table 1, Rosa canina and Rosa dumalis ecotypes prefer relatively higher altitudes, which were between 1446 and 2210 m ( Table 1). We found a wide variation in thorn, yield, fruit weight and fruit flesh ratio characteristics in ecotypes. In general, the ecotypes used in the present study exhibited low or medium thorn on its shoots. The high thorn causes difficulties in harvesting Rosa shrubs. The pre-selected ecotypes had high or very high yield characteristics, which is important for bringing them into cultivation conditions. Fruit weight was quite variable, ranging from 2.95 g (K5, Rosa dumalis) to 4.72 g (K12, Rosa canina), respectively (Table 1). Previous studies described variability in thorn and yield of Rosa ecotypes [32][33][34][35]. The fruit weight of rose hip ecotypes belonging to different species that were reported in Turkey ranged from 0.61 to 4.95 g [32,33,[35][36][37]. At the individual level, promising ecotypes were found in both R. canina and R. dumalis species. For example, K12, K11, K18, K17 and K4 ecotypes belonging to R. canina, and K7 and K19 belonging to R. dumalis, had bigger fruits and K8, K17, K12 and K4 within R. canina, and K10, K9 and K16 within R. dumalis, exhibited a higher percentage of fruit flesh ( Table 1). The promising ecotypes can be considered for future research on the rose hip cultivar development. Fruit-flesh ratios were between 62.55% (K-14) and 74.42% (K-8) ( Table 1). Previous studies also indicated high diversity within Rosa species at the individual level. Ercisli and Esitken [38] reported fruit-flesh ratios between 63.11-73.63% among Rosa canina and Rosa dumalis ecotypes. Ipek and Balta [39] found fruit-flesh ratios between 62.0-72.0% among 19 rose hip ecotypes naturally grown in middle Anatolia. Ersoy and Ozen [7] found fruit-flesh ratios between 64.8-82.8% among rose hip ecotypes grown in Western Anatolia.
Our results on fruit-flesh ratio was comparable with the above studies. Wild edible fruits had greater gene diversity and they also had unique gene combinations [40][41][42]. It is well understanding that wild species increase crop genetic diversity [39,43]. Frequent seed propagation of wild fruits and strictly out-crossing nature probably leads to an increase in the genetic diversity of wild fruits [37][38][39].

Total Flavonoid, Total Phenolic, Total Carotenoid, Vitamin C, Total Anthocyanin, Antioxidant Capacity and SSC Content
Total flavonoid, total phenolic, total carotenoid, Vitamin C, total anthocyanin, antioxidant capacity and SSC content of R. canina and R. dumalis ecotypes are shown in Table 2. Total flavonoid content was the highest in K6 (2.04 mg per g FW), which belongs to R. dumalis, followed by K3 (1.94 mg per g FW), while it was the lowest in K9 (0.88 mg per g FW) ( Table 2). In Iran, total flavonoid content was studied in different Rosa Sustainability 2021, 13, 8060 5 of 12 species and reported from 0.70 (R. hemisphaerica) to 2.53 mg quercetin equivalent per g FW (R. canina) [44]. Flavonoids impact the aroma and color of fruits and against biotic and abiotic stressors [45]. Medveckiene et al. [46] reported that total flavonoid content is species dependent in Rosa, and the highest total flavonoid content found as 41.59 mg/100 g DW. Rosehips accumulated a higher amount of total flavonoids ranged from 52 mg/100 g to 56 mg/100 g [20,47]. We found a wide variation in total phenolic content among ecotypes, which varied from 390 mg GAE per 100 g (K7) to 519 mg GAE per 100 g (K5). Previously, total phenolic content was found between 177-816 mg GAE per 100 g fresh samples between Rosa species grown in different parts of the world [48][49][50][51]. Koczka et al. [52] reported 150-299 mg/100 g DW total polyphenols in fruit flesh of rosehips. Medveckiene et al. [46] used several species of Rosa and reported that total polyphenol content in flesh varied from 150 to 299 mg/100 g DW. Fascella et al. [53] showed between 4057 and 6784 mg GAE/100 g DW total phenolic content among Rosa species. Alp et al. [54] found that R. dumalis fruits had total phenolic content between 297 and 403 mg GAE/100 g FW. The total phenolic content results obtained in our study were comparable with the above studies and indicate total polyphenol richness of rose hips. High polyphenol content of rose hips comparable with high polyphenol fruits such as blueberry, elderberry, black currant, blackberry and raspberry. The differences among studies may be attributed to different species, accessions and ecotypes used. Harvest period, different extraction methods, ecological conditions, etc., could also affect total phenolic content [48][49][50][51]. Table 2 shows the total carotenoid contents in the 20 ecotypes of Rosa fruits. The total carotenoid content ranged between 6.83 (K9) and 15.17 (K3) mg per g FW. Previously, Shameh et al. [44] reported variable total carotenoid content among 21 Rosa ecotypes belonging to different species. Andersson et al. [48] reported total carotenoids between 297 (R. spinosissima) and 1020 (R. dumalis) µg per g dry weight base in Rosa. Medveckiene et al. [46] reported total carotenoid content in the fruits of five Rosa species, ranging from 8.67-49.61 mg/100 g DW. Fascella et al. [53] found that hips of Rosa sempervirens and R. canina showed the highest total carotenoid values (1235 and 1204 mg/100 g DW, respectively), whereas those of R. corymbifera and R. micrantha had lower values (1072 and 1061 mg/100 g DW, respectively). Al-Yafeai et al. [55] revealed variable carotenoid contents among rosehip species. Previous study is also indicating rich carotenoid content of rosehips than the other fruits [56]. Carotenoids responsible yellow, orange and red pigments in fruits and offers human health benefits [44,57]. Genetic background, cultivars, harvest period, extraction and analytical method etc. affect fruit carotenoid content [48].
Ascorbic acid of the R. canina and R. dumalis ecotypes are given in Table 2. The ecotypes showed ascorbic acid content between 430 and 690 mg per 100 g of fresh fruit. Ascorbic acid content in rose hip is mainly species and ecotype dependent. Our results are comparable with other investigations. For example, Roman et al. [58] determined that the amount of ascorbic acid ranged between 112-360 mg per 100 g of fresh rose hips belonging to different species grown in Romania. Celik et al. [35] revealed ascorbic acid content in fruits of different Rosa species in high altitude (between 1650-1900 m) in the Eastern Anatolia region of Turkey between 604 and 1032 mg/100 g. Medveckiene et al. [46] reported that the ascorbic acid content in flesh of five rosehips species grown in Lithuania ranged between 385 mg to 736 mg/100 g FW. The vitamin C contents of rose hips in Italy were reported between 222-513 mg/100 g [53]. Alp, et al. [54] reported that vitamin C content of hips genotype dependent in Rosa dumalis varied from 402 to 511 mg/100 g fresh weight. Ercisli and Esitken [38] found vitamin C content in rose hips between 1222 to 2557 mg/100 g FW, indicating higher values than our samples. Rosu et al. [59] found vitamin C between 616 and 866 mg/100 g FW in fruits of Rosa caesia and R. rubiginosa in Romania. Rose hip species growing in high altitude regions are rich in ascorbic acid due to higher light exposure and lower oxygen amounts. Light exposure increases the amount of carotene and thus protects ascorbic acid in the fruit, while the lack of oxygen reduces oxidative stress and lessens ascorbic acid breakdown [32]. Vitamin C in fruits specie and genotype dependent and shows high heritability [60,61].
The total anthocyanin content of ecotypes ranged from 3.62 mg (K12) to 7.81 (K5) mg cyanidin-3-glucoside equivalent per kg of fresh fruit (Table 2). Previous studies reported that the major anthocyanin in the Caninae section including R. canina and R. dumalis fruits was cyanidin-3-glucoside and was mostly found in seed coat and fruit peel [12]. Guerrero et al. [62] found that the total anthocyanin content in over maturated rose hip fruit was 0.38 mg/100 g. Cyanidin-3-glucoside was previously reported to have the highest oxygen radical scavenging effect [63]. Shameh et al. [44] reported that the total anthocyanin content in the various species of Rosa was between 1.80 and 15.86 mg per kg. In Italy, anthocyanins content was found between 0.55 mg CGE/100 g DW in R. corymbifera fruits and 3.66 mg CGE/100 g DW in R. canina fruits. Murathan et al. [64] reported that total anthocyanin specie dependent in Rosa and varied from 2.45 to 3.72 mg CGE/100 g. Yildiz and Alpaslan [65] found total anthocyanins in Rosa fruits as 28.2 mg per kg. Anthocyanins give red to blue colors of many plants. Environment, specie, cultivar, altitude, cultivation techniques etc. strongly effect its concentration in fruits [62,64,66]. DPPH values of ecotypes are shown in Table 2 and the results indicate differences among ecotypes in the antioxidant capacities. Antioxidant capacity was the lowest for the K7 ecotype as 19.7 mg ascorbic acid equivalent per g FW and was the highest for K6 ecotype as 34.7 mg ascorbic acid equivalent per g FW (Table 2). Shameh et al. [44] showed significant differences in antioxidant activity among Rosa ecotypes. They reported that R. hemisphaerica fruits had the lowest (3.80 mg AAE per g FW), and R. canina fruits had the Sustainability 2021, 13, 8060 7 of 12 highest (37.60 mg AAE per g FW), antioxidant capacity. These differences may be caused by factors such as sample type used, the species differences, geographical area, the degree of ripening, climate and experimental conditions. Cunja et al. [67] reported that the highest antioxidant capacity was observed in R. canina fruit harvested in September and after that antioxidant capacity decreased. Okatan et al. [37] reported significant variability in DPPH scavenging activity of rosehip ecotypes ranging from 98.1 to 254.8 µg/mL. Demir et al. [13] found that DPPH scavenging activities of five Rosa species between 161 and 278 µg/mL. Soare et al. [68] reported that rose hip ecotypes show great diversity in DPPH values in Romania.
Soluble solid content (SSC) of ecotypes varied from 16.9% (K8) to 22.7% (K10), respectively (Table 2). Previously, SSC contents of rose hip fruits were very variable and reported between 12-36% [33,34,69]. For rose hip (Rosa spp.) genotypes from different areas, SSC were reported as 31-36.7% [38], 24.1-30.5% [7], 10-18% [68] and 17.6-22.8% [54]. Our results on SSC were in agreement with the above literature. Consumers not only emphasis on external quality criteria such as color, shape, size, but also on internal quality of fruits, including taste, sugar degree, acidity, etc. [69]. Table 3 shows phenolic compounds of 20 Rosa ecotypes. The major phenolic compounds in fruits of R. canina and R. dumalis ecotypes were chlorogenic acid, and followed by gallic acid, rutin, p-coumaric acid and caffeic acid, respectively (Table 3). Those phenolic acids are highly variable in fruits of different ecotypes of R. canina and R. dumalis. The highest chlorogenic acid content was found in ecotype K5 as 81.3 µg per g FW, while the lowest value was observed in fruits of K10 ecotypes as 27.8 µg per g FW, respectively (Table 3).

Correlations between Traits
In the study, correlation analysis was performed to determine the relationships between fruit characteristics in rose hip ecotypes (Table 4). Correlation analysis showed that FW had positive relationships with FFR (r = 0.34), Vit. C (r = 0.57 *), TPC (r = 0.07) and DPPH (r = 0.13), and negative relationships with SSC (r = −0.25), TFC (r = −0.09), TC (r = −0.03) and TA (r = −0.04), respectively (Table 4).   (Table 4). Previously, Guler et al. [9] used a number of rose hips and reported negative relationships between fruit SSC content and flesh weight, and positive correlation with FW and FFR. Cosmulescu et al. [75] reported that there was strong relationship between antioxidant activity and total phenolic content in Rosa fruits.

Principal Compenent Analysis (PCA)
In the PCA analysis we found 75.68% variability in the traits. First three components namely PC1, PC2 and PC3 explained 40.11%, 23.42% and 8.15% of variations, respectively. PC1 is identified with the fruit weight, fruit flesh ratio and vitamin C, while PC2 is related to the SSC, total carotenoids and total phenolics. The analysis clearly grouped wild grown rose hip into bigger fruited ones, sweeter ones or those with a higher content of polyphenolic compounds (Figure 1). Previous studies also indicated large diversity among ecotypes in wild and cultivated plants by using PCA analysis [76][77][78][79][80][81][82]. with Vit. C (r = 0.63 *) but had negative correlation with TPC, TFC, TC, TA and DPPH. Vit. C had a positive correlation with TPC, TFC, TC and DPPH, but was negatively related to TA. TPC showed strong relationships with TFC (r = 0.74 **) and DPPH (r = 0.84 **) (Table  4). Previously, Guler et al. [9] used a number of rose hips and reported negative relationships between fruit SSC content and flesh weight, and positive correlation with FW and FFR. Cosmulescu et al. [75] reported that there was strong relationship between antioxidant activity and total phenolic content in Rosa fruits.

Principal Compenent Analysis (PCA)
In the PCA analysis we found 75.68% variability in the traits. First three components namely PC1, PC2 and PC3 explained 40.11%, 23.42% and 8.15% of variations, respectively. PC1 is identified with the fruit weight, fruit flesh ratio and vitamin C, while PC2 is related to the SSC, total carotenoids and total phenolics. The analysis clearly grouped wild grown rose hip into bigger fruited ones, sweeter ones or those with a higher content of polyphenolic compounds (Figure 1). Previous studies also indicated large diversity among ecotypes in wild and cultivated plants by using PCA analysis [76][77][78][79][80][81][82].

Conclusions
As a result, we identified valuable rose hip ecotypes belonging to R. canina and R. dumalis among natural rose hip populations in Kars province where no previous studies were conducted on rose hip ecotypes. At present study, especially K2, K8, K11 and K18 were found promising, with high fruit weight (over 4 g), K1, K8 and K10 with a high fruitflesh ratio (over 70%), and K4, K12 and K17 with both high fruit weight and a fruit-flesh ratio. It is also thought that these promising ecotypes can be cultivar candidates and can be produced economically in Turkey. The data showed that the analysed naturally grown hips, particularly K1, K2, K3, K4, K5, K6, K11 and K15, have good nutritional quality, making them suitable as functional foods.

Conclusions
As a result, we identified valuable rose hip ecotypes belonging to R. canina and R. dumalis among natural rose hip populations in Kars province where no previous studies were conducted on rose hip ecotypes. At present study, especially K2, K8, K11 and K18 were found promising, with high fruit weight (over 4 g), K1, K8 and K10 with a high fruitflesh ratio (over 70%), and K4, K12 and K17 with both high fruit weight and a fruit-flesh ratio. It is also thought that these promising ecotypes can be cultivar candidates and can be produced economically in Turkey. The data showed that the analysed naturally grown hips, particularly K1, K2, K3, K4, K5, K6, K11 and K15, have good nutritional quality, making them suitable as functional foods.