Abstract
BACKGROUND:
Abundant germplasm resources of Vaccinium uliginosum are present in Changbai Mountain in China, which have considerable research and development value. They include the naturally distributed alpine V. uliginosum, the white-fruited V. uliginosum variety, and various fruit shapes dominated by ellipses. However, the relevant research is limited, and the taxonomic status of various specific germplasm resources has not been addressed.
OBJECTIVE:
The genetic relationship and taxonomic status of four species of V. uliginosum germplasm resources in Changbai Mountain were investigated.
METHODS:
The chloroplast matK gene and SRAP molecular markers of 94 V. uliginosum samples were analyzed.
RESULTS:
According to the analysis of chloroplast matK gene, all four types were divided into three branches. The results showed low variation among natural samples of V. uliginosum in China and a relationship between haplotype distribution and geographical distribution. The results of the SRAP molecular marker analysis divided the V. uliginosum population into five branches. The multiple individuals of the same trait were clustered together, and different partial samples were clustered together.
CONCLUSIONS:
Diploid alpine V. uliginosum and tetraploid V. uliginosum were identified as two subspecies of V. uliginosum. The elliptical fruit shapewas the variant of tetraploid round blue fruit V. uliginosum, and the white-fruited V. uliginosum variety is a variant.
Keywords
Introduction
Vaccinium uliginosum L., a perennial deciduous shrub belonging to the family Ericaceae, is the most widely distributed species of Vaccinium [1]. It is naturally distributed among the forested areas of Da Xing’an and Xiao Xing’an Mountains, Inner Mongolia, and Changbai Mountain in northeast China [2, 3]. It is an important wild berry resource in China. The primary processed products of V. uliginosum mainly include frozen fruit, concentrated fruit juice, and jam, and the further processed products mainly include fruit wine, cake, ice cream, natural pigments, and compound products prepared with fruit juice as an ingredient [4]. The fruit of V. uliginosum is rich in flavonoids, particularly anthocyanins, proanthocyanidins, and flavonols [5, 6], which have anti-inflammatory properties [7], antioxidant properties [8], soften blood vessels [9], prevent arteriosclerosis, and reduce cholesterol accumulation [10]. In addition, the fruit juice possesses the ability to decompose the carcinogen nitrite amine [11] and has high economic, nutritional, and medical value.
Using isozyme analysis, Alsos et al. [12] showed that the genetic diversity of V. uliginosum within the population and at the species level is low. Using RAPD marker technology, Albert et al. [13] inferred the genetic structure of two V. uliginosum populations and considered that the population exhibited genetic diversity. Alsos et al. [14] analyzed the intraspecific genetic diversity of V. uliginosum using sequences of the noncoding region of chloroplast DNA and chloroplast spacer regions such as trnH-psbA to determine the origin and evolution of V. uliginosum. They also explored the community composition and genetic characteristics of V. uliginosum, detect 18cpDNA haplotypes, and identify three pedigrees of V. uliginosum: Amphi-Atlantic, Beringian and Arctic-Alpine lineages. In 2007, Eidesen et al. [15] identified two cpDNA haplotypes of V. uliginosum. “Subspecies” is the taxonomic unit under “species”. After geographical, ecological, or seasonal isolation, the intraspecific individuals in different distribution areas have certain differences in morphological characteristics, emphasizing unique geographical scope or habitat and unique natural history [16]. Varieties of V. uliginosum have varying fruit characters in different ecological environments. According to the different fruit shapes and sizes of fruit, V. uliginosum has 4–8 subspecies or varieties, but the exact number of subspecies and varieties is debatable [17]. There are natural distributions of V. uliginosum L. var. alpinum and V. uliginosum L. var. leucocarpum in China, but their subspecies nomenclature is controversial. Some researchers believe that they are varieties of V. uliginosum [17–19], but no further research report is available. V. uliginosum L. var. leucocarpum may have formed due to somatic mutation. This resource has potential values in anthocyanin regulation and synthesis [20].
DNA barcode is a relatively short DNA fragment that can represent the species in organisms and is easy to amplify. It is widely used in species identification and biogeography. Haplotype refers to the genotype composed of several closely linked genes that determine the same trait, and haplotype analysis is an important means of analyzing DNA bar code [21, 22]. The matK gene, an intron of the lysine gene trnK in the chloroplast genome, is a single-copy protein-coding gene with rapid evolution [23, 24], and widely used in genetic diversity analysis [25]. In previous studies, matK was found to be more suitable for studying the genetic diversity of V. uliginosum than other chloroplast coding genes, such as rbcL (Yang, 2020, unpublished). Molecular marker technology is widely employed to determine genetic variations in plant germplasm resources at the DNA level [26–28]. SRAP labeling is a PCR amplification-based labeling technology that selectively amplifies these regions by targeting the functional regions of the genome and regions with gene expression potential [29]. It has been successfully applied to the analysis of genetic diversity and the construction of genetic map of species [30, 31].
There are many germplasm resources of V. uliginosum in the Changbai Mountain area of China, which are naturally distributed as alpine V. uliginosum, the white-fruited V. uliginosum variety, and various fruit shape variations dominated by ellipses. These specific germplasm resources have high research and development value, but there are few relevant studies, and the taxonomic status of various germplasm resources needs to be addressed. Therefore, in this study, the genetic relationship and taxonomic status of specific germplasm resources of V. uliginosum in Changbai Mountain was determined using the chloroplast matK gene and nuclear gene SRAP molecular marker technology. The results of this study can serve as a theoretical foundation for the research, protection, rational development, and utilization of rare germplasm resources and future breeding work.
Materials and methods
Plant materials
Four types of wild V. uliginosum, including alpine V. uliginosum, white-fruited V. uliginosum variety, elliptical blue fruit, and round blue fruit collected from Lanjia forest farm in Wangqing County, Jilin Province, and Dongfanghong forest farm in Changbai Mountain, were used as materials (Fig. 1). The leaf sampling period is the fruit maturity period. In total, 94 V. uliginosum samples were collected, the main morphological characteristics are shown in Table 1. The elliptical blue fruit and round blue fruit samples were determined by naked eye observation and fruit shape index measurement. Overall, 10–30 samples of each type were collected. When sampling each individual, the distance between each plant was taken into consideration (more than 30 m) to avoid collecting the same asexual propagation plant. Young leaves on each individual plant were collected as test materials and taken back to the laboratory after quick freezing wsith dry ice and stored in a low-temperature refrigerator at –80°C until use.

Four typical photographs of Vaccinium uliginosum (A: round fruit type; B: elliptical fruit type; C: white-fruited V. uliginosum variety; D: alpine V. uliginosum).
Sample information of four types of wild Vaccinium uliginosum
DNA was extracted from fresh leaves of 94 V. uliginosum samples according to the instructions of the DNA extraction kit of Beijing Sunbiotech Co., Ltd., and the extracted DNA products were detected using 1% agarose gel electrophoresis and stored at –20°C for further experiments.
PCR amplification and sequencing of matK
The primers for matK amplification were designed using the chloroplast matK sequence of V. uliginosum in GenBank (KT307930). The upstream primer sequence was CGTACAGTACTTTTGTGTTTACGAG; the downstream primer sequence was ACCCAGTCCATGTGGAAATCTTGGTTC. The primers were synthesized by Beijing Sunbiotech Co., Ltd.
The extracted DNA served as the template for PCR amplification of the chloroplast matK gene fragment. The PCR reaction system (25μL) had the following components: template DNA (20 ng μL–1) 1μL; dNTP (2.5 mmol·L–1) 1.5μL; upstream primers (10μmol·L–1) 1μL; downstream primers (10μmol·L–1) 1μL; rTaq enzyme (5 U·μL–1) 0.2μL; 10×buffer 2.5μL; and ddH2O 17.8μL. Reagents such as rTaq enzyme and dNTPs were purchased from Takara Biomedical Technology Co. Ltd., Beijing, China.
The amplification program was as follows: predenaturation at 95°C for 5 min; 35 cycles of denaturation at 94°C for 30 s, annealing at 56°C for 60 s, extension at 72°C for 60 s, and final extension at 72°C for 5 min. Subsequently, the amplified products were stored at 4°C. The amplification products were detected using 1% agarose gel electrophoresis and observed using a gel imaging system. If the size of the fragment was consistent with the expectation, the PCR product was sent to Beijing Sunbiotech Co., Ltd., for two-way sequencing.
SRAP amplification based on genomic DNA
The SRAP reaction system and amplification were performed according to the method described by Li and Quiros [29] and were optimized based on SRAP-PCR. The reaction system (25μL) consisted of template DNA 1μL, dNTP 1.5μL, upstream primer 1μL, downstream primer 1μL, rTaq enzyme 0.2μL, 10×buffer 2.5μL and ddH2O 17.8μL. The amplification procedure was as follows: predenaturation at 94°C for 5 min; five cycles of denaturation at 94°C for 45 s, annealing at 35°C for 45 s, and extension at 72°C for 90 s; 35 cycles of denaturation at 94°C for 45 s, annealing at 50°C for 45 s, and extension at 72°C for 90 s, and final extension at 72°C for 7 min. The amplified products were detected using 2% agarose gel electrophoresis.
One hundred fifty-four pairs of primers were selected by free combination of 14 upstream primers and 11 downstream primers referring to the published SRAP universal primers [29]. All the primers were screened by the PCR reaction system to screen sequence primers with high polymorphism. All the primers required were synthesized by Beijing Sunbiotech Co., Ltd.
Data recording and analysis
cpDNA haplotype data and cluster analysis
All consistent sequences obtained by sequencing were compared with default parameters in Clustalx version 1.83 [32], and the sequence results were analyzed. The obtained sequences were divided into haplotypes (H) by DNAsp6 [33], and the nucleotide diversity (π) and haplotype diversity (Hd) were calculated. Mega 11 software [34] was used to construct the neighbor-joining cluster tree of V. uliginosum based on the p-distance between V. uliginosum samples, and V. caespitosum (GenBank accession No.: MT593009.1) was used as the outgroup.
SRAP-PCR data recording and cluster analysis
Gel-Pro analyzer 32.0 software (Media Cybernetics) combined with manual correction was used for statistical analysis of the PCR amplification results. The binary metadata was obtained according to the mobility and presence of fragments. At the same migration position, the band is assigned as “1” and “0”, and the statistical results were transformed into a data format suitable for genetic diversity and cluster analysis using NTsys-pc 2.1 software [35].
Results
Haplotype distribution and cluster analysis of cpDNA
cpDNA haplotype distribution
The sequence ends of 94 V. uliginosum matK genes were truncated to 832 bp. There were seven base site mutations and parsimony information site mutations at 55 bp, 109 bp, 370 bp, 420 bp, 495 bp, and 704 bp and a single mutation site at 460 bp. From these mutation sites, three haplotypes were identified and named H1-H3 (Table 2). A few haplotypes are formed by allelic variation of matK in V. uliginosum. H2 is the main haplotype of four V. uliginosum types, which is present in 45% of individuals, and H3 is the main haplotype of the alpine dwarf type G in Changbai Mountain Nature Reserve, which is present in 39% of individuals. All samples of the white-fruited V. uliginosum variety type B were distributed in haplotype H2. In addition, most of the elliptical blue fruit type T samples were distributed in haplotype H2. In contrast, the haplotype distribution of circular blue fruit type L was relatively scattered, which may be caused by different sampling sites.
Variable sites recorded in the matK cpDNA spacer region sequences in Vaccinium uliginosum
Variable sites recorded in the matK cpDNA spacer region sequences in Vaccinium uliginosum
Type L contained three haplotypes and had the highest nucleotide polymorphism [π(10–3)=0.322]. Types L and T contained three haplotypes, which is the highest number of haplotypes among the four types (Table 3). The Hd value of each type ranged from 0.0000 to 0.7000. The Hd value of T type was 0.3467, and the Hd value of L type was 0.6690. However, the haploid diversity of type L (Hd = 0.6690) was much higher than that of T type (Hd = 0.3467). This type showed the highest haploid diversity. On the other hand, only primitive haplotypes (H3 and H2) were detected in types G and B, and these types had the lowest levels of haploid and nucleotide diversity (Hd = 0, π = 0). Therefore, we inferred that alpine V. uliginosum type G and the white-fruited V. uliginosum variety type B may be variant types.
Distribution of matK haplotypes and index of genetic diversity in four types of Vaccinium uliginosum
Note: N: Sample number; n: Haploid number; Hd: Haplotype diversity; π: Nucleotide diversity
Based on the matK gene sequences of 94 V. uliginosum samples, the genetic distance coefficient between V. uliginosum individuals was determined by calculating the p-distance (Supplementary Table 1). The genetic distance between genes was 0.000–0.0084, and the average genetic distance was 0.0039. The genetic distance of most genes was generally higher than the average genetic distance, indicating that the frequency difference of each gene was small; the genetic difference between chloroplast genes was not apparent. An adjacency tree was constructed using MEGA 11 software. The results are shown in Fig. 2. All types could be divided into three branches in the cluster diagram. Among them, alpine V. uliginosum type G located in Dongfanghong forest farm in Changbai Mountain was clustered into one branch, and the white-fruited V. uliginosum variety type B located in Wangqing County of Jilin Province in Lanjia forest farm was clustered into one branch. The results indicated that there was a slight difference in the chloroplast internal variation and DNA level between the two types. Types L and T were mixed and scattered, indicating that there was little difference between them. The difference between type G and B samples was large, haplotype composition relationship was small, and diversity was poor. The 30 type L samples were sampled from six geographical locations, with five samples sampled from each location (Table 1). According to the cluster diagram, although type L was scattered in different groups, the samples in the same sampling area were generally clustered in one branch, indicating that the differences between samples in the same sampling area are small, and geographical isolation may be the main reason for the observed genetic diversity.

N-J clustering tree of Vaccinium uliginosum types based on the P-distance of matK gene.
The SRAP-PCR amplification system screened 154 pairs of primers, and 15 pairs of primer combinations with clear bands, moderate number, and good brightness were selected to amplify the genes of 94 V. uliginosum samples. Figure 3 shows the amplification electrophoresis of 94 V. uliginosum samples amplified using the ME9–EM10 primer combination. The polymorphism of different primer combinations among the samples was relatively rich.

Amplification electrophoresis image of the ME9–EM10 primer pair on 94 Vaccinium uliginosum samples.
The number of bands amplified by each primer was counted (Table 4). In total, 143 clear and usable bands were amplified by 15 primer pairs, including 1 common band and 142 polymorphic bands in 94 samples. The ratio of polymorphic bands was 99.3%. The number of bands amplified by each primer ranged from 8 to 11, with an average of 9.5 polymorphic bands per primer pair. These polymorphic bands could be used to analyze the genetic relationship among the 94 samples.
Bands of 94 Vaccinium uliginosum samples amplified by 15 pairs of primers
Among the 94 samples, the genetic similarity coefficient ranged from 0.3007 to 0.9860, and the average genetic similarity coefficient was 0.6238 (Supplementary Table 2). The variation range of the genetic similarity coefficient (i) between the round blue fruit type L and the alpine V. uliginosum type G was 0.3636–0.7972, (ii) between the round blue fruit type L and the white-fruited V. uliginosum variety B was 0.3846–0.7273, and (iii) between the round blue fruit type L and the elliptical blue fruit type T was 0.3706–0.7622. The average genetic similarity coefficient was 0.6062, 0.5782, and 0.5860, respectively. Thus, there are large individual differences within V. uliginosum species; the difference in similarity coefficients among the different types was minor. The average genetic similarity coefficient of each type was as follows: white-fruited V. uliginosum variety type B (0.8098)>elliptical blue fruit type T (0.6930)>alpine V. uliginosum type G (0.6881)>round blue fruit type L (0.6362). Combined with the sampling sites of each type, it can be inferred that the genetic diversity of V. uliginosum is related to its geographical distribution. The phylogenetic trees of the V. uliginosum samples were constructed using the unweighted pair group method with arithmetic mean (Fig. 4). All the evaluated types were divided into five categories. The first category comprised round blue fruit type L, the second was dominated by the alpine V. uliginosum type, with only one round blue fruit sample, the third was dominated by the elliptical fruit type, including three alpine V. uliginosum samples, and the fourth was dominated by the white-fruited V. uliginosum variety type and included three elliptical fruit samples. The fifth category mainly comprised the round blue fruit and elliptical fruit types. Thus, the clustering relationship among individuals of the germplasm resources was complex. Nonetheless, multiple individuals of the same trait were found to cluster together. The variation between samples was relatively stable, whereas the distribution of the round blue fruit type L in different sampling areas was relatively scattered, indicating that the intraspecific genetic variation of samples in different sampling areas is relatively high. However, samples collected at the same sampling site are generally clustered together, which is consistent with the results shown in the clustering diagram (Fig. 2.).

Phylogenetic tree of 94 Vaccinium uliginosum samples.
The screening of plant DNA barcodes mostly focuses on chloroplast genes, including matK, rbcL, and trnH-psbA [36–38]. Owing to the differences in evolution rates among genes and the frequent interspecific hybridization of plants, the optimal DNA barcode of plants vary greatly [39]. Because of the multicopy characteristics of the nuclear genome and large intraspecies variation, the universality of primers is poor and high-quality template DNA is required during amplification, which is not suitable for DNA degradation materials [40]. Therefore, the most suitable barcode in plants should be selected from the chloroplast genome [37]. As a PCR-based molecular marker technology, SRAP can be used to amplify the region between introns and exons of the plant genome through a unique primer design. The difference in the lengths of amplified fragments caused by differences in sequences among varieties generates polymorphism. This technology is simple to operate and efficient. Markers can be displayed in batches and are easy to distinguish [41]. However, limited research has been conducted regarding markers in V. uliginosum.
There are abundant phenotypic variations in the natural population of V. uliginosum. The common types are generally distributed at an altitude of approximately 700 m. The plant height is generally 50–80 cm, the fruit shape is typically round, and the common varieties are elliptical and cone-shaped [17]. In addition to variations in fruit shape, there are also differences in chromosome ploidy, including diploid, tetraploid and hexaploid, of which tetraploid V. uliginosum is the most widely distributed [42]. Alpine V. uliginosum is naturally distributed at an altitude of 2000–2550 m in Changbai Mountain Nature Reserve in Jilin Province, with obvious geographical distribution differences. It is a unique germplasm resource in the Arctic and alpine areas. It has a short pedicel and short tree body, with a height of only 5–10 cm. It exhibits creeping and lying down growth, and its chromosome ploidy is different from other types, generally diploid. It has fewer leaves and flowers than tetraploid plants. It also exhibits significantly differences in pollen grain size and wall thickness. Because of the harsh natural environment at high altitudes, the growth period is very short, and it is an extremely rare germplasm resource [19, 43]. The study found that the genome size and ploidy level of Vaccinium are unified, and the genome size of diploid V. uliginosum from Turkey and Andorra is significantly smaller than that of tetraploid V. uliginosum species [44]. Sultana et al. [45] identified the phylogenetic relationship of four species of Vaccinium genus such as V. uliginosum by using satellite repeat diversity analysis, but there is still great controversy about the taxonomic status of alpine V. uliginosum, most researchers have identified it as a variety of V. uliginosum (V. uliginosum L. var. alpinum) [17, 19]. The white-fruited V. uliginosum variety exists in few natural populations, concentrated in a small range, and it has no clear zonal geographical distribution. However, the fruit color is significantly different from that of natural V. uliginosum, which is silvery white and has stable genetic traits [18]; it is named V. uliginosum L. var. leucocarpum. The elliptical blue fruit and round blue fruit varieties are mixed. This variation type can be found in most natural populations, but there are slight differences in fruit morphology.
Based on its geographical distribution, morphological characteristics, cpDNA clustering, and phylogenetic tree analysis of SRAP molecular markers, alpine V. uliginosum can be defined as a subspecies rather than a variety of V. uliginosum and should be named V. uliginosum L. ssp. alpinum, which is contrary to the previous research conclusion by Hao and other researchers [17, 19]. Although it is a variety of V. uliginosum, other tetraploid types of V. uliginosum are defined as another subspecies (V. uliginosum L. ssp. uliginosum) [42]. In the cluster analysis and phylogenetic tree, a few samples of the round blue fruit and elliptical type are mixed, indicating that differences between the samples was not significant. Therefore, it can be defined as the round blue fruit variant, named V. uliginosum L. f. ellipticum. In addition, in the long-term investigation and collection, our research team found that in the natural population of V. uliginosum, other variant fruit shapes with the same geographical distribution as the elliptical fruit shape, such as conical, oblong, and oblate, can also be defined as the round blue fruit shape. The variation of V. uliginosum L. var. leucocarpum is small, which is significantly different from the other types. It is generally clustered into one branch, and it is considered a variety of the round blue fruit, which supports the research conclusion by Zhang [18].
Author contributions
C-W.Z. conceived this study. J-Z.C. wrote the paper. J-Z.C. and T.L. carried out most of the experiments and data analysis. X.T. and Y-H.Z. prepared the tables and figures. H-N.C revised the paper. All authors contributed to revising the paper and approved the manuscript.
Footnotes
Acknowledgments
We thank the staff of Lanjia forest farm in Wangqing County, Jilin Province, and the staff of Dongfanghong forest farm in Changbai Mountain for their help in collecting the samples.
Funding
This study was supported by the National Natural Science Foundation of China (Nos. 31760557 and 32160690).
Conflict of interest
The authors have no conflict of interest to report.
