Study on Supergenus Rubus L.: Edible, Medicinal, and Phylogenetic Characterization

Rubus L. is one of the most diverse genera belonging to Rosaceae; it consists of more than 700 species with a worldwide distribution. It thus provides an ideal natural “supergenus” for studying the importance of its edible, medicinal, and phylogenetic characteristics for application in our daily lives and fundamental scientific studies. The Rubus genus includes many economically important species, such as blackberry (R. fruticosus L.), red raspberry (R. ideaus L.), black raspberry (R. occidentalis L.), and raspberry (R. chingii Hu), which are widely utilized in the fresh fruit market and the medicinal industry. Although Rubus species have existed in human civilization for hundreds of years, their utilization as fruit and in medicine is still largely inadequate, and many questions on their complex phylogenetic relationships need to be answered. In this review, we briefly summarize the history and progress of studies on Rubus, including its domestication as a source of fresh fruit, its medicinal uses in pharmacology, and its systematic position in the phylogenetic tree. Recent available evidence indicates that (1) thousands of Rubus cultivars were bred via time- and labor-consuming methods from only a few wild species, and new breeding strategies and germplasms were thus limited; (2) many kinds of species in Rubus have been used as medicinal herbs, though only a few species (R. ideaus L., R. chingii Hu, and R. occidentalis L.) have been well studied; (3) the phylogeny of Rubus is very complex, with the main reason for this possibly being the existence of multiple reproductive strategies (apomixis, hybridization, and polyploidization). Our review addresses the utilization of Rubus, summarizing major relevant achievements and proposing core prospects for future application, and thus could serve as a useful roadmap for future elite cultivar breeding and scientific studies.


Introduction
Rubus L. is one of the most diverse and largest genera in the Rosaceae family. The genus consists of more than 700 shrubby or herbaceous species mainly distributed throughout the temperate zone of the northern hemisphere, with a few having expanded to the tropics and the southern hemisphere [1][2][3][4][5][6]. Species of the Rubus genus worldwide are classified into 12 subgenera [1][2][3]. However, Lu et al. [6] reclassified them into 8 subgenera, whereby only habitats in China were considered. There are two hypothetical centers of origin for Rubus: one is North America [7,8] and the other is southwestern China [9][10][11]. In addition, the pleasant flavor of the fresh Rubus fruit, its medicinal functions due to the health benefits of its very high secondary metabolite content, and its high genetic diversity and complex phylogeny rendering it suitable for scientific studies, make Rubus an important and ideal genus for breeders as well as scientists [8,12,13] (Figure 1). Furthermore, the rich secondary metabolites and the bark of Rubus are also important raw materials for cosmetics and fiber [14]. The attributes of Rubus are primarily utilized for applications involving fruit (fresh fruit, jam, and juice), medicinal compounds (fruit, leaf, and stem), and scientific studies (adaptation, reproduction, polyploidy, and evolution). In addition among the wild Rubus species, some can also be used to produce cosmetics or fiber products. This figure was created using BioRender software. Figure 1. Different aspects of Rubus species utilization. The attributes of Rubus are primarily utilized for applications involving fruit (fresh fruit, jam, and juice), medicinal compounds (fruit, leaf, and stem), and scientific studies (adaptation, reproduction, polyploidy, and evolution). In addition, among the wild Rubus species, some can also be used to produce cosmetics or fiber products. This figure was created using BioRender software.
Rubus bears aggregate drupetum fruits that have economically important edible and medicinal characteristics [12,13]. They have a pleasant flavor and have been dubbed "superfoods" due to their very high levels of secondary metabolites, such as hydrolyzable tannins, anthocyanins, polyphenols, flavanols, organic acids, and many other organic compounds [12,[15][16][17][18][19][20]. In an early investigation, Moyer et al. [15] extracted multiple anthocyanins and phenols from the ripe fruits of Rubus. Based on genomic resequencing, quadrupole time-of-flight liquid chromatography, and mass spectroscopy, 29 hydrolyzable tannins and their candidate chromosomal regions were identified by Wang et al. [20]. Because of these diverse secondary metabolites, the superfood Rubus fruits can provide anti-oxidants as well as anti-cancer, anti-microbial, and anti-complement activities, in addition to having other benefits for humans [21][22][23][24][25].
Since the publication of Darwin's On the Origin of Species, understanding the genetic basis of adaptation for the arisal of new species has been a central topic in evolutionary biology [26][27][28][29]. In the species-rich genus of Rubus, the diversity of reproductive strategy, such as through the process of hybridization, polyploidization, and apomixis, which enhances the adaptation capacity of Rubus [8,30,31]. Additionally, the various reproductive strategies also bring a huge challenge regarding the taxonomy of the Rubus genus in terms of morphological and molecular systematics [8,9,[31][32][33]. However, in order to make better use of the wild germplasm in Rubus, it is necessary to develop a better knowledge of the clear affinity of its phylogeny. Therefore, species of the Rubus genus provide a natural experimental system for studying the fundamental mechanisms of adaptation via diverse reproductive strategies and reticulate evolutionary phylogeny.
Over the past hundred years since Focke [1] published Species Ruborum, our understanding of Rubus L. has improved, including regarding its edibility [4,12,30] and in the medicinal [4,13,15,18,20,34] and phylogenetic fields [7,8,31,35,36] (Figure 1). However, most previous studies and reviews have devoted their attention to metabolic compounds of pharmacologic interest in only a very few specific species [13,19]. This review summarizes the general outcomes for fresh fruit breeding, medicinal components, and studies into phylogenetic relationships of the Rubus genus; based on the latest advances in the fields of omics (genomics, transcriptomics, proteomics, and metabolomics), CRISPR/Cas, and other genome editing technologies, experimental efficiency has improved remarkably. Finally, we point out the important value of Rubus in fruit germplasms, medicinal research, and understanding complex phylogenetic relationships resulting from diverse adaptative reproduction strategies. In short, Rubus L., in the Rosaceae family, is an ideal natural "supergenus" for breeders, pharmacologists, and evolutionary biologists. Our review addresses major questions regarding how to better exploit the wild germplasm in Rubus species and can thus serve as a useful roadmap for future breeding and fundamental scientific studies.

Studies of Edible Rubus Species
Rubus L. belongs to the Rosaceae family, from which many palatable fruit cultivars have been bred, such as the woodland strawberry (Fragaria vesca L. var. americana Porter), and the domesticated apple (Malus × domestica Borkh.), pear (Pyrus bretschneideri Rehd.), and peach (Prunus persica (L.) Batsch) [1,9,37]. Rubus species are popular for their pleasant fresh fruits, including the blackberry (R. fruticosus L.), red raspberry (R. ideaus L.), and black raspberry (R. occidentalis L.). Rubus fruits are aggregate drupetum fruits with varied colors, such as red, yellow, purple, and black [12]. Rubus fruits have been called "superfoods" because of the very high levels of beneficial secondary metabolites that they contain, including, e.g., anthocyanins, phenolic acids, flavonoids, tannins, and other essential compounds [19,24,[38][39][40] (Figure 2). The chemical composition of Rubus is influenced not only by environmental factors but also internal genetic differences. For example, studies of blackberries found that a temperate climate and higher cumulative rainfall increased the production of phenolic compounds [41]. The concentration of chemical components in Rubus is also related to the storage conditions, growing season or location, and maturity [42][43][44][45]. The genotype differences between species and cultivars in Rubus is the major internal factor that influences the divergence of chemical compositions; for instance, Skrovankova et al. [46] found that different cultivars show significant variations in the production of secondary metabolites, even when they were grown under the same environmental conditions. Furthermore, a series of studies on different Rubus genotypes, cultivars, and species produced results consistent with those of Skrovankova [41,44,[47][48][49].
Species of the Rubus genus have been cultivated and appeared in gardens for more than 15 centuries in Europe, for example in Turkey and Rome [18,38]. To date, thousands of cultivars have been bred, and these can mainly be divided into two types: the primocane-fruiting (also called annual-fruiting) type, including Heritage, Amity, Autumn Bliss, Autumn Britten, Dinkum, and Polana cultivars; and the floricane-fruiting (also called biennial-fruiting) type, including Claudia, Emily, Esta, Lauren, and Qualicum cultivars [8,12,38]. Among these, Logan, Boysen, and Marion are three elite cultivars. The major areas for growing these cultivars are Russia (125,000 t), North America (59,123 t), and Europe (43,000 t) [8,38]. Spurred by the human pursuit of fruit quality and increasing consumption, the fruit production of Rubus cultivars has rapidly expanded for the production of fresh fruit for use in jams and fruit juice [12,38,50,51] (Figure 1). In the early stages of Rubus domestication, breeders commonly selected superior individuals from wild habitats [4,5]. During the nineteenth century, approximately 30 breeding projects were conducted in North America and Europe. The elite cultivars of Preussen, Cuthbert, and Newburgh were bred by crossing between different subspecies of red raspberries [30]. However, the process of domestication has vastly reduced the morphological and genetic diversities of crops [52][53][54][55]. Current cultivars are bred from crossing or the improvement of only a few wild species, i.e., red raspberry (R. ideaus L.), black raspberry (R. occidentalis L.), and blackberry (R. fruticosus L.). In order to meet the demands of visual appeal, higher yield, greater quality, excellent health benefits, and diverse adaptation for breeders, more wild germplasms of Rubus resources and advanced biotechnology should be utilized for Rubus breeding. Based on simple sequence repeat (SSR) and amplified fragment length polymorphism (AFLP) markers, the first linkage map for Rubus was constructed based on the crossing of two red raspberry cultivars, Glen Moy and Latham, from Europe and North America, respectively [56]. Bushakra et al. [57] constructed a genetic linkage map using 1218 markers by the crossing of S1 (R. occidentalis L.) and Latham (R. idaeus L.) and compared it with genomes of other genera in the rosa family, such as Fragaria L., Malus Mill., and Prunus L. That study reported a high consistency of collinearity of genomes between different genera of Rosaceae, and hundreds of new polymorphic genetic markers were found for future quantitative trait loci mapping studies. In recent years, different high-resolution markers and advanced sequencing methods have been applied for trait mapping or new wild germplasm identification in Rubus [58][59][60][61][62].

Medicinal Studies of Rubus
Rubus L. is one of the most species-rich genera in the Rosaceae family, but only a few species have been used as medicinal herbs [13,18,63]. According to the records of the ancient pharmacopoeias in Europe and China, Rubus species have been used as medicinal herbs for several centuries. The stems and leaves of blackberry (R. fruticosus L.) were soaked with white wine for use as an astringent poultice for wound healing and for difficulties during childbirth, as suggested by Hippocrates [18]. The dried unripe fruits of "Fu-Pen-Zi" (R. chingii Hu) were used to improve and enhance liver and kidney health [20,64]. In the early stages of Rubus domestication, breeders commonly selected superior individuals from wild habitats [4,5]. During the nineteenth century, approximately 30 breeding projects were conducted in North America and Europe. The elite cultivars of Preussen, Cuthbert, and Newburgh were bred by crossing between different subspecies of red raspberries [30]. However, the process of domestication has vastly reduced the morphological and genetic diversities of crops [52][53][54][55]. Current cultivars are bred from crossing or the improvement of only a few wild species, i.e., red raspberry (R. ideaus L.), black raspberry (R. occidentalis L.), and blackberry (R. fruticosus L.). In order to meet the demands of visual appeal, higher yield, greater quality, excellent health benefits, and diverse adaptation for breeders, more wild germplasms of Rubus resources and advanced biotechnology should be utilized for Rubus breeding. Based on simple sequence repeat (SSR) and amplified fragment length polymorphism (AFLP) markers, the first linkage map for Rubus was constructed based on the crossing of two red raspberry cultivars, Glen Moy and Latham, from Europe and North America, respectively [56]. Bushakra et al. [57] constructed a genetic linkage map using 1218 markers by the crossing of S1 (R. occidentalis L.) and Latham (R. idaeus L.) and compared it with genomes of other genera in the rosa family, such as Fragaria L., Malus Mill. and Prunus L. That study reported a high consistency of collinearity of genomes between different genera of Rosaceae, and hundreds of new polymorphic genetic markers were found for future quantitative trait loci mapping studies. In recent years, different high-resolution markers and advanced sequencing methods have been applied for trait mapping or new wild germplasm identification in Rubus [58][59][60][61][62].

Medicinal Studies of Rubus
Rubus L. is one of the most species-rich genera in the Rosaceae family, but only a few species have been used as medicinal herbs [13,18,63]. According to the records of the ancient pharmacopoeias in Europe and China, Rubus species have been used as medicinal herbs for several centuries. The stems and leaves of blackberry (R. fruticosus L.) were soaked with white wine for use as an astringent poultice for wound healing and for difficulties during childbirth, as suggested by Hippocrates [18]. The dried unripe fruits of "Fu-Pen-Zi" (R. chingii Hu) were used to improve and enhance liver and kidney health [20,64]. More recently, as shown in Figure 2, many kinds of secondary metabolites with remarkable beneficial effects on humans have been extracted. Table 1 exhibits detailed information on the species, concentration, and part from which they were extracted. For instance, anthocyanins with an anti-oxidant function are extracted from the fruits of the blackberry [15,19]; flavonoids with anti-oxidant, anti-cancer, and anti-inflammatory effects are extracted from the fruits of the blackberry or Fu-Pen-Zi [16,[65][66][67]. Other organic compounds have been identified, mainly in blackberry or Fu-Pen-Zi, such as hydrolyzable tannin, glycoprotein, organic acid, and phenolic compounds [19,21,61,68,69]. In addition, reports have indicated many other kinds of beneficial effect of these secondary metabolites on humans, for example, improving mitosis and eyesight, treating or preventing cancer, back pain, and frequent urination [23,25,70,71]. To date, most of the secondary metabolites mentioned above are considered safe according to data from limited studies [13,[72][73][74][75][76]. For example, the extracted components from R. niveus Thunb. showed no statistically significant toxicity for mice [72]. Based on a cytotoxic experiment on Caco-2 cells, Ke et al. [77] found that metabolites extracted from the fruit of R. chingii Hu were safe and had a favourable effect on anti-cancer cells. Overall, the limited available data on the toxicity and allergenicity of Rubus species indicate they are safe for humans. More in-depth investigations regarding medicinal applications in pharmacology are needed.
In the genomic age, genomic sequencing of Rubus is already lagging behind compared to other major crops and fruits, such as rice, maize, cotton, apple, and pear [53,[78][79][80][81][82][83][84][85]. However, benefiting from the quick development of the cost-effective next-generation sequencing (NGS) [86,87] and transcriptome (RNA-seq) technology [88,89], nuclear or plastid genomes have been sequenced for some important medicinal species in the Rubus genus [20,[90][91][92][93][94][95]. Utilizing the RNA-seq data of the red raspberry (R. idaeus L.) fruit, Hyun et al. [90] determined the regulated candidate genes for biosynthesis of γ-aminobutyric acid and anthocyanins, which have anti-oxidant activity. Based on unripe Fu-Pen-Zi (R. chingii Hu) fruits, the chromosome-scale reference and genomic regions related to the biosynthetic pathway for hydrolyzable tannin (HT) have been reported [20]. Therefore, using the results of these studies, breeders could modify candidate genes or genomic regions of Rubus cultivars to improve the content of targeted secondary metabolites (HT, anthocyanins, etc.) using sitedirected genome editing technologies such as CRISPR/Cas. More recently, in the field of crop breeding, CRISPR/Cas genome-editing technology has been used with encouraging results [96]. However, use of this speedy, proven, and precise genome editing technology has not been reported in programs for breeding Rubus. To date, various studies taking advantage of transcriptomic analysis at different developmental stages (green; green and yellow; yellow, orange, and red) of Fu-Pen-Zi fruits have revealed that flavonoids and anthocyanins are synthesized at an early stage and their levels then decrease during subsequent development [61,97,98]. These studies also indicated that anthocyanins might not be responsible for the reddish color of ripe fruits. Thus, better characterization will require extraction of the pharmacological metabolites at the early stage of Rubus fruit development. Meanwhile, several studies on plastid genomes have focused on the pharmacological components of Rubus, i.e., R. eucalyptus Focke [93], R. rufus Focke [99], R. longisepalus Nakai and R. hirsutus Thunb. [100,101], and R. phoenicolasius Maxim. [94]. However, the detailed genetic basis and biosynthetic pathways of the pharmacological metabolites in Rubus species are still largely unclear, and further future investment and research on different aspects are needed [102][103][104].

Phylogenetic Studies of Rubus
The Rubus genus is one of the most successful models of an adaptive and evolutionary group, with distribution worldwide except for Antarctica [1,9]. As shown in Table 2, Rubus species were classified by Focke into 12 or by Lu into 8 subgenera according to worldwide distribution or distribution in China, respectively. The classification and phylogenetic construction of Rubus is a challenging task due to phenomena such as hybridization, apomixis, polyploidization, and introgression, which happen frequently in this genus. Ploidy levels among different subgenera and species are highly differentiated [7,32,33,35,58] (Table 2). More importantly, the diverse reproductive strategies may have conferred to Rubus species the ability to occupy various habitats worldwide, and thus demonstrate reticulate evolutionary phylogeny. According to previous phylogenetic analysis based on the ndhF gene, Howarth et al. [110] suggested that the Hawaiian Islands species (R. hawaiensis A. Gray and R. macraei A. Gray) originated from different ancestors, in contradiction with the morphological results. The reticulate evolution of Rubus has been indicated in recent studies. Wang et al. [31] used multiple chloroplast and nuclear genes to investigate the phylogenetic relationships of 142 Rubus taxa, which indicated reticulate evolutionary events between different subgenera and species. A study based on approximately 1000 target genes constructed the phylogenetic tree for 87 wild Rubus taxa and three cultivars, concluding that hybridization and incomplete lineage sorting (ILS) were responsible for the low resolution and topological conflicts between different subgenera, which were not caused by insufficient molecular signals [8]. Furthermore, it has been suggested that North America might be the primary center of origin of Rubus, which then expanded into Asia and Europe and finally dispersed to Oceania via birds [8]. Additionally, a series of studies on the phylogeny of Rubus detected conflicts in the phylogenetic affinities between plastid genes and nuclear genes in most cases [36,114,115]. The molecular and morphological topotaxies also appear inconsistent, which may result from the multiple reproductive strategies [36,111,[114][115][116][117]. In general, the difficulties of morphological or molecular taxonomy in subgenera and between species are not caused by lack of characteristics or signals; the real reason may be the diverse reproductive patterns that have made this group an ideal genus for investigation of the genetic basis of different reproductive and adaptive patterns.

Concluding Remarks and Future Perspectives
The Rubus genus consists of more than 700 species, but only a few of them, such as blackberry (R. fruticosus L.), red raspberry (R. ideaus L.), and black raspberry (R. occidentalis L.), have been domesticated or crossed by breeders to generate elite cultivars with excellent characteristics of strong adaptability, good storage properties, and pest or disease resistance. Unfortunately, there have been relatively few molecular breeding studies on Rubus and fewer genomic resources exist compared to other types of crops and fruits. Rubus breeders can reference those studies in order to improve the breeding methods for elite cultivars of different crops and fruits.
Furtehrmore, the situation for medicinal cultivars of Rubus is worse, and most of the investments in pharmacology have been concerned with R. ideaus L., R. fruticosus L., and R. chingii Hu, rarely involving residual species such as R. eucalyptus Focke, R. occidentalis L., and R. phoenicolasius Maxim., which have also been used as medicinal ingredients for hundreds of years. Notably, there remains a large deficiency in the study of the basic mechanisms and genetics of the active ingredients in medicinal Rubus species. Fortunately, advances in NGS and RNA-seq technologies offer an opportunity for researchers to spend less money and labor investigating the above-mentioned problems and to quickly identify and choose high-quality germplasms.
Finally, reconstructing the phylogenetic relationships for Rubus is a task made challenging by hybridization, polyploidization, apomixis, and introgression. However, researchers can also combine consideration of morphological characteristics with omics technologies (i.e., genomics, transcriptomics, proteomics, and metabolomics) to decipher the phylogenetic and evolutionary puzzles of Rubus. Consumers in the present era are increasingly demanding tastier and healthier fresh fruits. The wild species and elite cultivars of Rubus provide ideal candidates to address this demand due to their pleasant flavor and high concentrations of secondary metabolites. However, only a few wild species have been domesticated and are used in our daily food markets and medical treatment. Therefore, in order to better exploit the abundance of Rubus wild germplasms, information on their phylogenetic relationships and genetic diversity should be clarified. This study reviewed three major topics (edible, medicinal, and phylogenetic properties), but many challenges still exist in the utilization and research of Rubus. Working as a team and applying the latest omics strategies may open the door for developing a series of satisfactory elite germplasms for fruit and medicine, and reveal the central evolutionary phenomenon resulting in reticulate evolution.

Conflicts of Interest:
The authors declare no conflict of interest.