Article Genetic Diversity and Association Analysis among Germplasms of Diospyros kaki in Zhejiang Province Based on SSR Markers

Yang Xu , Wenqiang Cheng, Chunyan Xiong, Xibing Jiang, Kaiyun Wu and Bangchu Gong *

Research Institute of Subtropical Forestry, Chinese Academy of Forestry, Hangzhou 311400, China; [email protected] (Y.X.); [email protected] (W.C.); [email protected] (C.X.); [email protected] (X.J.); [email protected] (K.W.) * Correspondence: [email protected]; Tel.: +86-571-63310045

Abstract: In subtropical to temperate regions, persimmon (Diospyros kaki Thunb.) is an economically important fruit crop cultivated for its edible fruits. Persimmons are distributed abundantly and widely in Zhejiang Province, representing a valuable resource for the breeding of new cultivars and studying the origin and evolution of persimmon. In this study, we elucidated the genetic structures and diversity patterns of 179 persimmon germplasms from 16 different ecologic populations in Zhejiang Province based on the analysis of 17 SSR markers. The results show that there was a medium degree of genetic diversity for persimmon found in Zhejiang Province. With the exception of the Tiantai Mountain and Xin’an River populations, we found extensive gene exchange had occurred among the other populations. The 179 D. kaki germplasms from the 16 populations could be separated into three distinct clusters (I, II, and III) with a higher mean pairwise genetic differentiation index



(FST) (0.2714). Nearly all samples of Cluster-I were distributed inland. Cluster-II and Cluster-III
 contained samples that were widely distributed throughout Zhejiang Province including all samples

Citation: Xu, Y.; Cheng, W.; Xiong, from the coastal populations and the Northeast Plain populations. In addition, we performed C.; Jiang, X.; Wu, K.; Gong, B. Genetic association mapping with nine traits (fruit crude fiber content, fruit calcium content, fruit water Diversity and Association Analysis content, fruit longitudinal diameter, fruit aspect ratio, seed width, seed length, leaf aspect ratio, among Germplasms of Diospyros kaki and number of lateral veins) using these markers. This led to the identification of 13 significant in Zhejiang Province Based on SSR marker–trait associations (MTAs; p < 0.00044, 0.1/228) using a general linear model, of which, six Markers. Forests 2021, 12, 422. MTAs with a correlation coefficient (R2) >10% were consistently represented in the general linear https://doi.org/10.3390/f12040422 model with p < 0.00044 in the two models. The genetic structures and diversity patterns of the persimmon germplasms revealed in this study will provide a reference for the efficient conservation Academic Editor: Claudia Mattioni and further utilization of persimmon germplasms. The MTAs identified in this study will be useful for future marker-assisted breeding of persimmon. Received: 2 March 2021 Accepted: 29 March 2021 Keywords: Diospyros kaki; microsatellite markers; genetic diversity; population structure; marker Published: 1 April 2021 trait association

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- 1. Introduction iations. Persimmon (Diospyros kaki Thunb.) is a widespread temperate and subtropical species of the genus Diospyros belonging to the Ebenaceae family [1]. Persimmon was first cul- tivated in eastern China between the first to second centuries, corresponding to about 2000 years of planting history. So far, persimmon has been one of the most important fruit Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. crops in China, Japan, and South Korea due to its deciduous fruit [2,3]. Approximately This article is an open access article 1000 varieties of D. kaki have been found in China. The abundant diversity of the per- distributed under the terms and simmon germplasm represents a valuable resource for the breeding of new cultivars and conditions of the Creative Commons studying the origin and evolution of persimmon [4,5]. Attribution (CC BY) license (https:// The genetic diversity among persimmon germplasms has been reported using dif- creativecommons.org/licenses/by/ ferent marker techniques such as plant morphology [6], sequence-related amplified poly- 4.0/). morphisms (SRAPs) [7], sequence-specific amplified polymorphism (SSAPs) [8], amplified

Forests 2021, 12, 422. https://doi.org/10.3390/f12040422 https://www.mdpi.com/journal/forests Forests 2021, 12, 422 2 of 20

fragment length polymorphisms (AFLPs) [9,10], and simple sequence repeats (SSRs) [11,12]. However, the relationships among D. kaki germplasms have not yet been completely clarified due to the low sampling density of the germplasm resource collection in previous studies. Zhejiang Province is located in southeast China with a long history of persimmon cultivation. The previous evaluation of phenotypic traits demonstrated that Diospyros germplasm resources are abundant and widely distributed in Zhejiang Province, and there were rich variations of agronomic traits [13–15], which may contain an abundance of desired genes/alleles such as for stress resistance and association with excellent flavor as a result of unique climate and diverse ecology. So far, little research has focused on evaluating and mining the genetic structures, diversity patterns, and desired genes/alleles of these sources, particularly through the use of codominant molecular markers. The identification of loci influencing agronomic traits is the bases of marker-assisted breeding, which can effectively improve breeding efficiency [16]. Association mapping (AM) is a faster and more efficient approach for the high-resolution evaluation of com- plex traits and appears as a promising tool to circumvent the limitations of linkage map- ping [16–18]. AM can effectively be used to associate genotypes with phenotypes in natural populations and identify marker–trait associations for several agronomic traits in various plant species [16–24]. However, due to the lack of persimmon genomic information, only a few molecular markers have been identified as linked to natural astringency loss [25–27] or sexual traits [28,29], which implies perhaps quality inheritance. Association studies of other quantitative traits have not been reported. Of the many available DNA markers, microsatellites, also known as simple sequence repeat (SSR) markers, are an appropriate choice in association analysis compared to other types of molecular markers as they possess more characteristics that are advantageous for this type of application [18–22] such as random and wide distribution in the genome, which is genetically codominant, highly reproducible, multi-allelic, and perfectly suitable for high-throughput genotyping [18–22,30–32]. Microsatellites have been successfully used for plant crop and tree characterization and studies of genetic diversity and gene association analysis [18–22]. In the present study, we elucidated the genetic structures and diversity patterns of 179 persimmon sources in Zhejiang using SSR markers. We performed association mapping and identified markers associated with eleven traits of agronomic importance to determine marker–trait associations (MTAs). We explored the genetic diversity and marker–trait associations of the Zhejiang persimmon sources, which will be useful for future evolution studies and marker-assisted breeding of persimmon.

2. Materials and Methods 2.1. Plant Materials We collected 179 semi-wild Diospyros kaki germplasms from 16 different eco-regions in Zhejiang Province from 2014 to 2015. These samples originated from the local sampling site. Each plant was collected at least 30 m away from its nearest neighbor. The 16 eco-regions is according to the long-term habitual classification of terrain in Zhejiang Province based on geographical and environmental parameters (Table1). The climatic conditions for each collection site were estimated by the Zhejiang Climate Data 2019 [33]. We calculated the following environmental factors; detailed information of the 179 sampled individuals are listed in Table S1. All 179 D. kaki germplasms were grafted and kept during the spring of 2015 in the Plant Nursery of Lanxi City in Zhejiang Province (29◦250 N, 119◦510 E). All stocks were “YLNO.6”, a superior line selecting from Diospyros kaki var. Silvestris. There were less differences both in the terrain and environment of the Plant Nursery. Each germplasm contained six plants at a spacing of 3 m × 4 m. The trial layout was a randomized block design with six replicates of single-tree plots. Persimmon tree management followed the standard of LY/T 1887–2010 [34]. All germplasms were in a stable fruiting period during the autumn of 2019. Six to eight normal leaves and fruits of every plant were collected, in Forests 2021, 12, 422 3 of 20

total about 40 leaves and fruits of every germplasm, to measure six agronomic traits (fruit longitudinal diameter (FLD), fruit aspect ratio (FAR), seed width (SW), seed length (SL), leaf aspect ratio (LAR), and number of lateral veins (NLV)) of persimmon. The 6–8 normal fruits of every plant were pooled as one repetition to measure three agronomic traits (fruit crude fiber content (FCF), fruit calcium content (FCC), and fruit water content (FWC)). The correlation between traits was assessed by the Spearman correlation test using R software [35]. The mean values (Table S2) of these agronomic traits were used for the association study.

Table 1. Abbreviation and full name of population and climatic factor in this research.

Pop No. Full Name of Population Abbreviation Full Name of Climatic Factor Hill north of Hang Jia Hu Precipitation in spring (March to Pop 1 SprPR Plain May) Precipitation in summer (June to Pop 2 Hang Jia Hu Plain SumPR August) Precipitation in autumn Pop 3 Ningshao Plain AutPR (September to November) Precipitation in winter Pop 4 Western Hang Jia Hu Plain WinPR (December to February) Solar radiation in spring (March Pop 5 Eastern Zhejiang Hill SprSR to May) Solar radiation in summer (June Pop 6 Tiantai Mountain SumSR to August) Solar radiation in autumn Pop 7 Thousand Island Lake AutSR (September to November) Solar radiation in winter Pop 8 Xin’an River WinSR (December to February) Mean temperature in spring Pop 9 Eastern Qujing Basin SprMT (March to May) Mean temperature in summer Pop 10 JingQu Basin SumMT (June to August) The eastern hills of JingQu Mean temperature in autumn Pop 11 AutMT Basin (September to November) Xianxia Mountain and Qian Mean temperature in winter Pop 12 WinMT Li Gang (December to February) Pop 13 Wenhuang Plain Songgu Basin and Lishui Pop 14 Basin Yandang Mountain and Pop 15 Wenrui Plain Pop 16 Cave mountain range

2.2. Simple Sequence Repeat (SSR) Genotyping Genomic DNA was extracted from selected tender leaf tissue using an improved Cetyltrimethylammonium Bromide (CTAB) method [36] and the quality assessed using 1% agarose gel electrophoresis. The concentration was then determined using ultraviolet spectrophotometry. The 17 microsatellite markers were selected from various sources [37–39] (Table S3). The amplification of the SSR reaction was carried out in an ABI Applied Biosystems Poly- merase Chain Reaction (PCR) machine. The 10 µL reaction volume contained 50–100 ng total genomic DNA, 50 pmol primers, 0.25 U Taq DNA polymerase (Promega, Sunnyvale, CA, USA), 10× buffer, 200 µmolL−1 dNTP, and 2 mmolL−1 Mg2+. The PCR conditions were as follows: predenaturation at 94 ◦C for 3 min followed by 30 cycles of PCR amplifica- tion. Each cycle consisted of denaturation at 94 ◦C for 30 s, annealing for 30 s using the designated temperature depending on the primer pair, and extension at 72 ◦C for 1 min. The PCR products were finally extended at 72 ◦C for 5 min and then stored at 4 ◦C. The PCR products were resolved on 8% denaturing polyacrylamide gels, which were silver stained to detect SSR bands [35]. A 100 bp DNA ladder was included for the estimation of allele sizes. When estimating allele doses in polyploids, the situation is more complicated because there may be genotypic ambiguity even for codominant markers [40]. In addition we still do not know if the persimmon is allohexaploid or autohexaploid, or whether Forests 2021, 12, x FOR PEER REVIEW 4 of 22

Each cycle consisted of denaturation at 94 °C for 30 s, annealing for 30 s using the desig‐ nated temperature depending on the primer pair, and extension at 72 °C for 1 min. The PCR products were finally extended at 72 °C for 5 min and then stored at 4 °C. The PCR products were resolved on 8% denaturing polyacrylamide gels, which were silver stained Forests 2021, 12, 422 to detect SSR bands [35]. A 100 bp DNA ladder was included for the estimation of allele4 of 20 sizes. When estimating allele doses in polyploids, the situation is more complicated be‐ cause there may be genotypic ambiguity even for codominant markers [40]. In addition we still do not know if the persimmon is allohexaploid or autohexaploid, or whether dif‐ differentferent germplasms germplasms are are mostly mostly like like different different types types of hexaploidy of hexaploidy [25]. Therefore, [25]. Therefore, all alleles all allelesof the of SSR the markers SSR markers were regarded were regarded as the asdominant the dominant markers markers according according to previous to previ- re‐ oussearch research of polyploids of polyploids such suchas potato as potato (Solanum (Solanum tuberosum tuberosum L.) (tetraploid)L.) (tetraploid) [41], cotton [41], (Goss cotton‐ (Gossypiumypium hirsutum hirsutum L.) (allotetraploid)L.) (allotetraploid) [42], [wheat42], wheat (Triticum (Triticum aestivum aestivum L.) (hexaploid)L.) (hexaploid) [43], per [43‐ ], persimmonsimmon (hexaploid) (hexaploid) [12], [12 and], and sugarcane sugarcane (Saccharum (Saccharum officinarum officinarum L.) (octoploid)L.) (octoploid) [44,45]. [44 The,45 ]. Theresult result obtained obtained at the at the DKMP8 DKMP8 marker marker is displayed is displayed in Figure in Figure 1. SSR1. data SSR datawere werescored scored vis‐ visuallyually for for allele allele size size and and presence presence or absence or absence of each of each primer. primer. SSR bands SSR bands were scored were scored as a asqualitative a qualitative character character (e.g., (e.g., present present (1) or absent (1) or (0)) absent to create (0)) toa binary create matrix. a binary All matrix.genotypic All genotypicdata can be data found can in be Table found S4. in Next, Table the S4. phenotypes Next, the of phenotypes the 0/1 matrix of were the used 0/1 matrix for genetic were useddiversity for genetic assessments diversity and assessments association analyses and association following analyses the analytical following methods the analyticalof dom‐ methodsinant markers. of dominant markers.

FigureFigure 1. Polyacrylamide 1. Polyacrylamide gel gel obtained obtained in in simple simple sequence sequence repeat repeat (SSR) (SSR) markermarker A1. The The number number at at the the top top of of the the figure figure represents the sample order; the number beside the right of the figure represents the size of DNA marker; each locus of represents the sample order; the number beside the right of the figure represents the size of DNA marker; each locus of the the DKMP8 SSR marker is labeled in the figure with the black arrow. Sample genotypic data of DKMP8 were created DKMP8based SSR on markerthe presence is labeled (1) or in absence the figure (0) of with each the loci black (e.g., arrow. in sample Sample 48, the genotypic genotypic data data of of DKMP8 DKMP8 were marker created are DKMP8 based‐ on the presence284 (0), DKMP8 (1) or‐ absence294(1), DKMP8 (0) of each‐308(0), loci DKMP8 (e.g., in‐320(0), sample DKMP8 48, the‐328(1), genotypic DKMP8 data‐336(0), of DKMP8 DKMP8 marker‐352(0), are DKMP8 DKMP8-284‐390(1). (0), DKMP8-294(1), DKMP8-308(0), DKMP8-320(0), DKMP8-328(1), DKMP8-336(0), DKMP8-352(0), DKMP8-390(1). 2.3. Data Analysis 2.3. Data Analysis POPGENE 1.31 (University of Alberta, Edmonton, Canada) [46] was used to evaluate thePOPGENE following genetic 1.31 (University diversity parameters: of Alberta, Edmonton,the allele frequency Canada) (MAF [46] was), number used to of evaluatealleles the(Na following) and effective genetic number diversity of alleles parameters: (Ne) per thelocus, allele heterozygosity frequency ( observedMAF), number (Ho), expected of alleles (Naheterozygosity) and effective ( numberHe), Nei’s of gene alleles diversity (Ne) per index locus, (Nei heterozygosity), and Shannon’s observed diversity (Ho index), expected (I). heterozygosityThe polymorphic (He ),information Nei’s gene content diversity (PIC index) was calculated (Nei), and for Shannon’s each locus diversity using the index online ( I). Theprogram polymorphic PICcalc information [47]. The formula content of ( PICmain) was genetic calculated variation for parameters each locus usingis listed the as online fol‐ programlows. The PICcalc information [47]. The of formulaallele frequencies of main genetic is shown variation in Table parameters S5. is listed as follows. The information of allele frequencies is shown in Table S5. MAF = Common_allele/n (1) where Common_allele is the numberMAF of= highestCommon_allele genotypes/n detected at a site, and n is the(1) total number of genotypes detected at a site. where Common_allele is the number of highest genotypes detected at a site, and n is the i total number of genotypes detected at a site. 2 Ne  1/  Pi (2) ii1 2 Ne = 1/∑ Pi (2) i=1

i 2 He = 1 − ∑ pi (3) i=1

where i is the number of distinct alleles at a locus and Pi (i = 1,2, 3, . . . , I) is the frequency of allelei in the population. Nei = (2N/(2N − 1)) × He Forests 2021, 12, 422 5 of 20

where N is the number of samples.

I = −∑ (Pi × LnPi) (4)

where Pi is the frequency of the ith allele. GenAlEx v.6.5 was used to determine the average pairwise genetic differentiation index (FST) between populations [48]. The analysis of molecular variance (AMOVA) was used to partition the genetic variance and was carried out using Arlequin 3.11 [49]. Genetic relationships between the sampled germplasms were elucidated through the construction of an unrooted neighbor joining (NJ) dendrogram based on simple matching coefficients using MEGA5 [50]. A bootstrap value of 1000 replicates was used to test the reliability of the NJ dendrogram. Principal component analysis (PCA) was performed using EIGENSOFT [51]. To summarize the patterns of variation in the multi locus data set, principal coordinate analysis (PCoA) was performed by GenAlEx version 6.5 based on the pairwise FST matrix. The isolation-by-distance pattern (IBD) was detected by Mantel tests [52] with 1000 permutations based on matrices of pairwise genetic distance (FST/1-FST) and geographic distance among populations performed in GenAlEx 6.5. The population genetic structure was studied using the Bayesian clustering method implemented in STRUCTURE version 2.3.4 [53]. To determine the optimum number of subgroups (K), three independent runs were performed to estimate each K value from 1 to 10. For each run, a burn-in length of 5000 followed by 50,000 Markov chain Monte Carlo (MCMC) iterations were performed assuming an admixture model with default settings and allele frequencies. Structure Harvester [54] was used to visualize the best K value based on delta K (∆K) and maximum log likelihood L (K). Repeated sampling analysis and genetic structural plots were analyzed by CLUMPP [55]. POPHelper [56] was used to render the histogram base on the K value matrix after merging. The associations between SSR markers and traits were tested using TASSEL 3.0 soft- ware [57]. Association mapping analysis was conducted using TASSEL through the general linear model (GLM) with the Q matrix as well as the mixed linear model (MLM) with the Q matrix and kinship (K matrix) procedures. The significant threshold for the association was set at different levels p < 0.00043 (0.01/number of loci). The phenotypic variation explained by each marker–trait association was studied using the correlation coefficient (R2).

3. Results 3.1. Characterization of Morphological Traits Persimmon Germplasms The range and mean values of the nine morphological traits of the persimmon germplasms are provided in Table2. All traits showed broad ranges across the germplasm accessions. Eight morphological traits had about two to six times difference (maxi- mum/minimum). The highest FCR was approximately sixteen times the lowest. There were many different responses of correlations among the morphological traits and environ- mental factors (Table3). There were no significant relationships between SumPR, SumMT, WinMT, altitude, and morphological traits. SprPR, WinPR, SumSR, AutSR, and SprMT showed negative correlation to more than one morphological trait. AutPR, longitude, WinSR, and AutMT showed a positive correlation to the morphological traits. Latitude showed a negative correlation to FCC and FCR, and positive correlation to LAR. Perhaps because of the influences of environmental factors on phenotypes, the morphological traits of the accessions were significantly different by geographical area (Table2). FCC and FWC were both significantly high in germplasms from the Eastern Coastal Plain (e.g., Pop 15 and Pop 3). The lowest FCR, FCC, and SW were in both germplasms from Pop 8. These fruit and seed shape indices (e.g., FLD, FAR, and SL) had similar geographical distribution patterns that were higher in eastern coastal areas (e.g., Pop 3 and Pop 13), and lower in Pop 10 and the JingQu Basin overall (e.g., Pop 12). The correlation among the nine agronomic traits was also analyzed. The pairwise correlation coefficients among fruit aspect ratio and fruit longitudinal diameter, fruit Forests 2021, 12, 422 6 of 20

longitudinal diameter, fruit aspect ratio, and seed length were greater than 0.5, indicating strong associations (correlations) between each pair of these traits (Table S6).

Table 2. Nine morphological traits present in persimmon germplasms.

FCR FCC Pop No. FWC (%) FLD FAR SW (mm) SL (mm) LAR NLV (mg/100 g) (mg/100 g)

0.89 ± 0.26 7.81 ± 2.45 78.35 ± 2.14 b,c 0.93 ± 0.18 11.27 ± 1.3 20.75 ± 4.35 a e Pop 1 a,b,c a,b,c,d b,c,d,e 50.65 ± 8.26 a,b a,b b,c,d,e 1.74 ± 0.2 3.61 ± 0.26

0.88 ± 0.39 8.71 ± 1.92 77.11 ± 6.23 c b 11.19 ± 1.83 18.15 ± 3.59 a 4.22 ± 0.73 Pop 2 a,b,c a,b,c c,d,e 49.87 ± 10.26 0.86 ± 0.24 a,b e 1.74 ± 0.13 b,c,d

0.75 ± 0.15 8.97 ± 1.81 83.39 ± 1.51 a,b,c 0.91 ± 0.25 11.66 ± 2.62 22.06 ± 3.79 a a Pop 3 a,b,c,d a,b a 53.68 ± 9.51 a,b a,b a,b,c,d 1.66 ± 0.22 4.91 ± 0.28

1.18 ± 0.29 7.66 ± 2.94 79.44 ± 2.72 a,b,c 1.03 ± 0.15 a 20.1 ± 1.62 a 4.41 ± 0.4 Pop 4 a,b a,b,c,d b,c,d 55.66 ± 8.72 a,b 12.1 ± 2.03 c,d,e 1.77 ± 0.14 b,c,d

0.67 ± 0.07 6.46 ± 1.73 78.89 ± 3.17 c 1.03 ± 0.25 11.63 ± 4.65 21.08 ± 3.96 a 4.6 ± 0.93 Pop 5 c,d b,c,d b,c,d,e 49.57 ± 6.8 a,b a,b a,b,c,d,e 1.65 ± 0.07 a,b,c

0.91 ± 0.33 7.71 ± 1.28 80.66 ± 3.06 a a 10.69 ± 1.22 24.55 ± 3.58 a,b 4.19 ± 0.81 Pop 6 a,b,c a,b,c,d a,b 61.62 ± 8.24 1.16 ± 0.27 a,b a 1.6 ± 0.12 b,c,d

0.77 ± 0.19 6.19 ± 3.26 75.94 ± 3.96 a,b,c 1.08 ± 0.18 b 22.06 ± 2.76 a 4.08 ± 0.49 Pop 7 a,b,c,d c,d e 55.79 ± 8.87 a,b 9.75 ± 1.11 a,b,c,d 1.77 ± 0.23 d,e

d d 80.4 ± 1.57 a,b,c 0.96 ± 0.17 10.08 ± 0.82 19.79 ± 3.75 a e Pop 8 0.43 ± 0.09 5.34 ± 2.03 a,b 52.56 ± 8.23 a,b b c,d,e 1.64 ± 0.2 3.61 ± 0.31

0.79 ± 0.18 6.42 ± 2.62 77.91 ± 1.2 a,b,c a 10.15 ± 1.13 19.1 ± 4.86 a 4.15 ± 0.42 Pop 9 a,b,c,d b,c,d b,c,d,e 56.04 ± 13.46 1.12 ± 0.36 b d,e 1.68 ± 0.11 c,d

0.69 ± 0.14 7.18 ± 2.24 80.32 ± 2.47 c a,b 11.09 ± 1.4 21.08 ± 2.84 a 3.95 ± 0.15 Pop 10 c,d a,b,c,d a,b,c 47.49 ± 9.55 0.92 ± 0.2 a,b a,b,c,d,e 1.65 ± 0.17 d,e

0.85 ± 0.34 8.75 ± 3.02 78.49 ± 4.18 c 1.04 ± 0.26 10.48 ± 1.67 19.45 ± 3.19 a 4.45 ± 0.42 Pop 11 a,b,c,d a,b,c b,c,d,e 48.14 ± 9.67 a,b a,b c,d,e 1.7 ± 0.17 a,b,c,d

a 7.04 ± 1.59 76.23 ± 2.65 b,c a 10.14 ± 1.23 19.41 ± 1.9 1.61 ± 0.13 4.68 ± 0.48 Pop 12 1.19 ± 1.38 a,b,c,d d,e 50.59 ± 6.66 1.12 ± 0.14 b c,d,e a,b a,b

0.74 ± 0.17 6.97 ± 0.99 80.01 ± 0.48 a,b a 10.85 ± 1.29 23.86 ± 2.48 1.61 ± 0.17 4.43 ± 0.33 Pop 13 b,c,d a,b,c,d b,c 60.82 ± 11.49 1.17 ± 0.34 a,b a,b a,b b,c,d

0.84 ± 0.22 8.71 ± 3.2 76.66 ± 4.29 a,b,c a b 21.02 ± 2.82 a,b d,e Pop 14 a,b,c,d a,b,c d,e 54.49 ± 9.39 1.13 ± 0.22 9.8 ± 1 a,b,c,d,e 1.61 ± 0.2 4.09 ± 0.4

0.81 ± 0.16 a 80.94 ± 2.21 a,b,c 1.09 ± 0.28 11.21 ± 1.62 22.85 ± 3.99 a 4.42 ± 0.47 Pop 15 a,b,c,d 9.18 ± 2.49 a,b 57.25 ± 11.72 a,b a,b a,b,c 1.72 ± 0.16 b,c,d

0.9 ± 0.41 6.74 ± 2.55 78.72 ± 2.79 c a,b b 21.09 ± 3.02 b e Pop 16 a,b,c a,b,c,d b,c,d,e 49.36 ± 12.42 1 ± 0.2 9.92 ± 1.07 a,b,c,d,e 1.44 ± 0.12 3.66 ± 0.55 All 0.82 ± 0.42 7.74 ± 2.62 79.16 ± 3.66 53.39 ± 10.18 1.03 ± 0.24 10.74 ± 1.84 21.21 ± 3.57 1.67 ± 0.18 4.26 ± 0.58 Min * 0.32 2.61 64.13 24.9 0.63 10.11 6.46 2.97 1.11 Max # 5.32 16.01 85.63 81.56 1.74 30.98 21.98 6 2.16 Note: Different lowercase letters for each parameter indicate significant differences among populations according to two-way analysis of variance (ANOVA) followed by Duncan multiple- range test (α = 0.05); * represents the minimum value of morphological traits in all 179 persimmon germplasms; # represents the maximum value of morphological traits in all 179 persimmon germplasms.

Table 3. Correlation analysis between the morphological traits and environmental factors for the 179 persim- mon germplasms.

Item SprPR SumPR AutPR WinPR SprSR SumSR AutSR WinSR SprMT SumMT AutMT WinMT Longitude Latitude Altitude FCR 0.081 −0.089 −0.031 0.013 −0.056 0.014 0.04 −0.026 0.071 0.064 0.095 0.06 −0.093 −0.028 0.021 −0.159 −0.187 −0.232 −0.155 −0.190 FCC 0.122 0.190 * −0.082 0.156 * −0.142 0.037 0.079 0.125 0.250 ** −0.004 * * ** * * −0.248 0.369 −0.195 −0.270 0.208 −0.312 FWC 0.091 −0.142 0.116 −0.018 0.061 0.003 0.410 ** −0.052 −0.09 ** ** ** ** ** ** −0.225 FLD −0.085 −0.09 0.127 0.058 −0.109 −0.122 0.098 −0.082 −0.015 0.102 0.142 0.198 ** −0.023 −0.041 ** −0.177 −0.148 FAR 0.117 −0.076 −0.042 −0.146 −0.091 0.019 0.016 0.063 −0.083 −0.018 0.067 0.006 0.071 * * −0.218 0.228 −0.179 −0.175 SW −0.094 −0.04 0.161 * −0.056 0.044 0.13 0.07 0.026 0.233 ** 0.077 −0.141 ** ** * * −0.187 −0.171 SL −0.133 0.04 0.188 * 0.004 −0.121 0.111 −0.061 0.063 0.164 * 0.146 0.187 * −0.118 −0.075 * * −0.218 0.217 LAR −0.096 0.039 0.052 0.166 * 0.145 −0.055 −0.133 −0.05 0.14 0.041 −0.075 −0.011 −0.076 ** ** −0.206 −0.173 NLV −0.067 −0.002 0.320** −0.046 0.106 −0.081 −0.111 0.059 0.009 0.044 0.329 ** −0.006 −0.039 ** * Note: * represents significant difference at p < 0.05; ** represents significant difference at p < 0.01.

3.2. Characterization of Microsatellite Marker Polymorphism A total of 228 polymorphic alleles with high polymorphic rates were generated for 179 D. kaki using 17 SSR markers (Table S7). The number of alleles per locus ranged from seven (MDP21) to 22 (DKMP13), with an average of 13.41 alleles per locus. Due to the hexaploid nature of the D. kaki genome, other parameters (e.g., the expected heterozygosity) Forests 2021, 12, 422 7 of 20

were not calculated. The high polymorphic levels of the examined SSR markers are reflected by the PIC values for each marker, which ranged from 0.8566 to 0.9437; all were above 0.5 with an average of 0.8965 alleles per marker. We found 25 unique alleles for five markers at eight loci. Two unique alleles (Mdp16-150 and Mdp16-158) were only found in Pop 2 (Hanning4). One unique allele (Mdp8-284) were only found in Pop 7 (ChunAn8). As hexaploid species, high Na and Ho were observed within each accession, which ranged from 67 to 132 and 0.6471 to 1, respectively (Table S8). In addition, we detected significant relationships between Na and environmental factors (such as SprPR, AutPR, WinPR, SumSR, AutSR, WinSR, SprMT, longitude, and altitude). However, Ho was not related (Table S9). We observed that the Na of Mdp21 showed significant relationships only with altitude; another fifteen of these markers showed significant relationships with more than two factors. Therefore, these SSR markers were suitable for studying the genetic diversity of persimmon germplasms. We further analyzed the correlation between pairwise FST and geographical distance (Figure2) using the Mantel test. The results showed that there was a significant correlation between genetic distance and geographic distance Forests 2021, 12, x FOR PEER REVIEW 8 of 22 (r = 0.117, p = 0.010), indicating a clear geographic origin-based structuring or predominate isolation by distance among the investigated germplasms.

Figure 2. Mantel correlation between the genetic and geographic distances. Figure 2. Mantel correlation between the genetic and geographic distances. 3.3. Establishment of Genetic Information in Germplasms Populations The genetic3.3. diversity Establishment of 179 of genotypesGenetic Information of D. kaki in Germplasmssampled from Populations 16 areas was analyzed and the results are shownThe genetic in Table diversity4. The of number 179 genotypes of polymorphic of D. kaki sampled alleles from ( Np )16 ranged areas was from analyzed 143 to 215, withand an the average results ofare 189.4 shown alleles in Table for 4. oneThe number population. of polymorphic There was alleles considerable (Np) ranged from 143 to 215, with an average of 189.4 alleles for one population. There was considerable diversity both in the value of Nei, PIC and I among these populations. The Nei, PIC, He, diversity both in the value of Nei, PIC and I among these populations. The Nei, PIC, He, and I ranged from 0.2242 to 0.3718, 0.1725 to 0.2831, 0.2082 to 0.3571, and 0.336 to 0.5268, and I ranged from 0.2242 to 0.3718, 0.1725 to 0.2831, 0.2082 to 0.3571, and 0.336 to 0.5268, with averages ofwith 0.3279, averages 0.2473, of 0.3279, 0.3109, 0.2473, and 0.3109, 0.4604, and respectively. 0.4604, respectively. An examination An examination of the of the correlation betweencorrelation population between genetic population diversity genetic parameters diversity parameters and population and population size showed size showed that only Np wasthat significantly only Np was relatedsignificantly to the related population to the population size (Figure size3 ).(Figure Interestingly, 3). Interestingly, the the highest Ne (1.6224highest and Ne 1.6287), (1.6224 andNei (0.36941.6287), Nei and (0.3694 0.3718), andPIC 0.3718),(0.2831 PIC and (0.2831 0.2813), and 0.2813), and I and I (0.5268 and 0.5232)(0.5268 were and all 0.5232) found were in the all populationsfound in the populations of the eastern of the hills eastern of JingQu hills of Basin JingQu as Basin well as for JingQuas well Basin as overall.for JingQu Low Basin indexes overall. were Low found indexes in were Pop found 6 (the in lowest), Pop 6 (the Pop lowest), 8, Pop Pop 8, 5, Pop 9, and PopPop 13, 5, Pop which 9, and were Pop all 13, located which were in the all eastern located hillsin the area eastern of Zhejiang hills area of Province. Zhejiang Prov‐ ince.

Table 4. Comparison of the genetic diversity for the germplasms populations.

Pop No. Sample size Np Ne Nei I He PIC Pop 1 8 190 1.5826 0.3524 0.4836 0.3304 0.2606 Pop 2 7 196 1.5963 0.3655 0.4975 0.3394 0.2681 Pop 3 18 208 1.5932 0.3507 0.5042 0.341 0.2707 Pop 4 8 189 1.5624 0.3424 0.4721 0.321 0.254 Pop 5 7 174 1.4984 0.311 0.4275 0.2887 0.2299 Pop 6 7 151 1.3241 0.2242 0.3236 0.2082 0.1725 Pop 7 16 204 1.5388 0.325 0.4711 0.3149 0.2519 Pop 8 9 143 1.384 0.2379 0.336 0.2247 0.1799 Pop 9 6 177 1.5078 0.3237 0.4392 0.2967 0.2367 Pop 10 12 210 1.6287 0.3718 0.5232 0.3563 0.2813 Pop 11 15 215 1.6224 0.3694 0.5268 0.3571 0.2831 Pop 12 11 197 1.565 0.341 0.4814 0.3255 0.2586 Pop 13 7 178 1.5251 0.3252 0.4449 0.3019 0.2395

Forests 2021, 12, 422 8 of 20

Table 4. Comparison of the genetic diversity for the germplasms populations.

Pop No. Sample Size Np Ne Nei I He PIC Pop 1 8 190 1.5826 0.3524 0.4836 0.3304 0.2606 Pop 2 7 196 1.5963 0.3655 0.4975 0.3394 0.2681 Pop 3 18 208 1.5932 0.3507 0.5042 0.341 0.2707 Pop 4 8 189 1.5624 0.3424 0.4721 0.321 0.254 Pop 5 7 174 1.4984 0.311 0.4275 0.2887 0.2299 Pop 6 7 151 1.3241 0.2242 0.3236 0.2082 0.1725 Pop 7 16 204 1.5388 0.325 0.4711 0.3149 0.2519 Pop 8 9 143 1.384 0.2379 0.336 0.2247 0.1799 Pop 9 6 177 1.5078 0.3237 0.4392 0.2967 0.2367 Pop 10 12 210 1.6287 0.3718 0.5232 0.3563 0.2813 Pop 11 15 215 1.6224 0.3694 0.5268 0.3571 0.2831 Pop 12 11 197 1.565 0.341 0.4814 0.3255 0.2586 Pop 13 7 178 1.5251 0.3252 0.4449 0.3019 0.2395 Pop 14 14 191 1.5446 0.3243 0.4624 0.3127 0.2482 Pop 15 25 208 1.6113 0.3527 0.5076 0.3456 0.2725 Forests 2021, 12, x FOR PEER REVIEWPop 16 9 199 1.5281 0.3293 0.466 0.31110 of 0.249522 Mean 11.19 189.37 1.5383 0.3279 0.4604 0.3109 0.2473

FigureFigure 3. The 3. correlationThe correlation coefficient coefficient between between population population genetic genetic diversity diversity parameters parameters and the sample sample size size of of germplasm germplasm pop- populations. ulations. 3.4.The Establishment differentiation of Genetic among Information the 16 populationsin Clusters was measured by pairwise FST (Table5 ). High FSTUsingvalues the SSR (>0.25), genotypic which data, are an representative unrooted neighbor of a very joining high (NJ level) dendrogram of genetic (Fig differentia-‐ tionure as 4) suggested was constructed by Wright showing [58], the were genetic mainly relationships found between among Pop the 6populations. and other populations,In the anddendrogram, between Pop the 8179 and genotypes other populations. of D. kaki were The clustered highest into pairwise four groupsFST value(I, II, III, of and 0.497 was foundIV). betweenGroup‐I Popwas 6comprised and Pop 8of and 48 indicates germplasms, strong accounting genetic differentiation. for 26.8% of Otherthe total pairwise germplasms and comprising most germplasms (90%) from Pop 14, while about 60% of germplasms were from Pop 7, 9, and 16. Group‐II comprised 22 germplasms, comprising most germplasms (89%) from Pop 8, while about 60% of germplasms were from Pop 7, 9, and 16. Group‐III was found to contain 61 germplasms including six materials from Pop 5, which accounted for 85.7% of this population as well as the majority (≥50%) of germplasms from Pop 1, 3, 4, and 13. Group‐IV contained 48 germplasms including the majority from Pop 6 and 15 with the remainder from Pop 1, 3, 9, 10, 11, and 16, but no germplasms were found to originate from the mountains in the northeast (Pop 8 and 9), southwest (Pop 12) and south (Pop 14) of JingQu Basin.

Forests 2021, 12, 422 9 of 20

FST values between populations were nearly all lower than 0.25, indicating that the weak genetic differentiation among these populations may be due to existing extensive gene exchange.

Table 5. Pairwise genetic differentiation index (FST) estimates for groups resulting from the cluster analyses.

Pop P7 P16 P9 P14 P1 P10 P4 P2 P11 P12 P8 P5 P3 P13 P15 P6 P7 0 P16 0.0826 0 P9 0.0321 0.0597 0 P14 0.061 0.056 0.0462 0 P1 0.0679 0.0991 0.0722 0.0749 0 P10 0.0706 0.0743 0.0583 0.0755 0.0436 0 P4 0.1185 0.145 0.13 0.1284 0.0401 0.0528 0 P2 0.1066 0.116 0.1279 0.1189 0.0446 0.0608 0.0665 0 P11 0.0708 0.0762 0.0494 0.0551 0.0547 0.0458 0.0981 0.0867 0 P12 0.0665 0.0744 0.0414 0.0447 0.0851 0.0612 0.1468 0.1144 0.0289 0 P8 0.3013 0.336 0.3243 0.3002 0.3095 0.2351 0.3348 0.3023 0.1947 0.2296 0 P5 0.1791 0.1943 0.1727 0.161 0.0974 0.0918 0.1222 0.1725 0.1279 0.1542 0.3337 0 P3 0.1508 0.1592 0.169 0.1611 0.0486 0.0743 0.0567 0.0496 0.1182 0.1653 0.3095 0.1294 0 P13 0.1876 0.1981 0.2193 0.193 0.0866 0.1044 0.0986 0.0951 0.1583 0.2083 0.3912 0.1511 0.0552 0 P15 0.1639 0.1539 0.1769 0.1646 0.0787 0.0834 0.0891 0.0628 0.1224 0.1659 0.3101 0.1563 0.0427 0.0664 0 P6 0.3144 0.3277 0.3579 0.3233 0.2509 0.2453 0.2305 0.2063 0.2717 0.3362 0.497 0.3797 0.177 0.2236 0.151 0 Note: All FST estimates were highly significant (p < 0.01).

3.4. Establishment of Genetic Information in Clusters Using the SSR genotypic data, an unrooted neighbor joining (NJ) dendrogram (Figure4 ) was constructed showing the genetic relationships among the populations. In the dendrogram, the 179 genotypes of D. kaki were clustered into four groups (I, II, III, and IV). Group-I was comprised of 48 germplasms, accounting for 26.8% of the total germplasms and comprising most germplasms (90%) from Pop 14, while about 60% of germplasms were from Pop 7, 9, and 16. Group-II comprised 22 germplasms, comprising most germplasms (89%) from Pop 8, while about 60% of germplasms were from Pop 7, 9, and 16. Group-III was found to contain 61 germplasms including six materials from Pop 5, which accounted for 85.7% of this population as well as the majority (≥50%) of germplasms from Pop 1, 3, 4, and 13. Group-IV contained 48 germplasms including the majority from Pop 6 and 15 with the remainder from Pop 1, 3, 9, 10, 11, and 16, but no germplasms were found to originate from the mountains in the northeast (Pop 8 and 9), southwest (Pop 12) and south (Pop 14) of JingQu Basin. Principal component analysis (PCA) and STRUCTURE analysis were also used to evaluate the relationships among germplasms. In the PCA (Figure5), all germplasms were more clearly divided into three groups than the NJ classification. STRUCTURE analysis also indicated the optimal cluster number as K = 3 based on delta K (Figures6 and7). However, the membership of the STRUCTURE and PCA clusters corresponded closely with those delineated by the NJ dendrogram. The slight difference in group 1 and group 2 by the NJ dendrogram corresponded to one cluster of STRUCTURE and PCA. Therefore, these germplasms were grouped into three clusters based on the comprehensive results of the NJ dendrogram, PCA, and STRUCTURE analyses. All persimmon germplasms, labeled in blue, red, and black, according to the three gene pools determined by STRUCTURE analyses, were then positioned on a map based on their origins. As Figure8 shows, germplasms of the same group were mostly admixtures of different clusters (e.g., the samples of Pop 6 were divided into Cluster-I and Cluster-II, while Pop 5 consisted of samples of Cluster-I and Cluster-II). Notably, the classification pattern of the clusters was consistent with the geographical distribution. Nearly all samples of Cluster-I were those distributed inland while samples of Cluster-II and Cluster-III were mainly from coastal areas. Forests 2021, 12, 422 10 of 20 Forests 2021, 12, x FOR PEER REVIEW 11 of 22

Forests 2021, 12, x FOR PEER REVIEW 12 of 22

Significant differences of genetic parameters were observed between clusters and the primary population. The Ne, Nei, He, PIC, and I among clusters tended to be more uniform with values of 0.2985–0.3211, 1.86–1.96, 0.2954–0.3188, 0.2367–0.2546, and 0.443–0.4781, re- spectively (Table 6), which demonstrated considerable diversity among the populations. The high pairwise FST value (Table S10) between clusters indicated that each cluster was clearly differentiated from the others. The percentage of variation also increased from 9.87% among populations (Table S11) to 26.23% among clusters (Table S12). All clusters contained private alleles, with four (DK26-152, DK26-126, Mdp8-200 and Mdp8-224) found in Cluster-I, four (Mdp16-158, Mdp16-150, DKMP13-212, and DKMP13-230) in Cluster-II, two (Mdp8-284 and DKMP15-500) in Cluster-III. While, four, five, twelve al- leles were not found in Cluster-I, Cluster-II, and Cluster-III, respectively (Figure 9).

Table 6. Comparison of genetic diversity parameters for the germplasm clusters.

Sample Cluster Np Ne Nei I He PIC Size Cluster-I 70 213 1.5127 0.3041 0.4554 0.3015 0.2422 Cluster-II 61 220 1.5493 0.3211 0.4781 0.3188 0.2546 Figure 4. A neighborFigure 4. A joining neighborCluster-III dendrogram joining dendrogram48 showing showing 198 the genetic the 1.5026 genetic affiliation affiliation 0.2985 of the of 1790.4433 the persimmon179 persimmon 0.2954 germplasms. germplasms. 0.2367 AllAll germplasms germplasms were dividedMean into four59.7 groups (I,210.33 II III, and 1.5215IV). 0.3079 0.45893 0.3052 0.2445 were divided into four groups (I, II III, and IV). Principal component analysis (PCA) and STRUCTURE analysis were also used to evaluate the relationships among germplasms. In the PCA (Figure 5), all germplasms were more clearly divided into three groups than the NJ classification. STRUCTURE analysis also indicated the optimal cluster number as K = 3 based on delta K (Figures 6 and 7). However, the membership of the STRUCTURE and PCA clusters corresponded closely with those delineated by the NJ dendrogram. The slight difference in group 1 and group 2 by the NJ dendrogram corresponded to one cluster of STRUCTURE and PCA. Therefore, these germplasms were grouped into three clusters based on the comprehensive results of the NJ dendrogram, PCA, and STRUCTURE analyses. All persimmon germplasms, la‐ beled in blue, red, and black, according to the three gene pools determined by STRUC‐ TURE analyses, were then positioned on a map based on their origins. As Figure 8 shows, germplasms of the same group were mostly admixtures of different clusters (e.g., the sam‐ ples of Pop 6 were divided into Cluster‐I and Cluster‐II, while Pop 5 consisted of samples of Cluster‐I and Cluster‐II). Notably, the classification pattern of the clusters was con‐ sistent with the geographical distribution. Nearly all samples of Cluster‐I were those dis‐ tributed inland while samples of Cluster‐II and Cluster‐III were mainly from coastal areas.

FigureFigure 5. Genetic 5. Genetic structure structure of the 179 of persimmon the 179 germ persimmonplasms revealed germplasms by principal revealed component by principal component analysis (PCA). analysis (PCA).

Forests 2021, 12, x FOR PEER REVIEW 13 of 22 Forests 2021, 12, x FOR PEER REVIEW 13 of 22

Forests 2021, 12, 422 11 of 20

FigureFigure 6. 6. DiagnosticDiagnostic plots plots of delta of‐K delta-K from the from STRUCTURE the STRUCTURE analysis of analysis the 179 D. of kaki the germplasms 179 D. kaki Figure 6. Diagnostic plots of delta‐K from the STRUCTURE analysis of the 179 D. kaki germplasms populations.germplasms populations. populations.

FigureFigure 7. 7. GeneticGenetic structure structure of of the the 179 179 persimmon persimmon germplasms germplasms revealed revealed by by STRUCTURE STRUCTURE analysis. analysis. Figure 7. Genetic structure of the 179 persimmon germplasms revealed by STRUCTURE analysis. Significant differences of genetic parameters were observed between clusters and the primary population. The Ne, Nei, He, PIC, and I among clusters tended to be more uniform with values of 0.2985–0.3211, 1.86–1.96, 0.2954–0.3188, 0.2367–0.2546, and 0.443–0.4781, respectively (Table6), which demonstrated considerable diversity among the populations. The high pairwise FST value (Table S10) between clusters indicated that each cluster was clearly differentiated from the others. The percentage of variation also increased from 9.87% among populations (Table S11) to 26.23% among clusters (Table S12). All clusters contained private alleles, with four (DK26-152, DK26-126, Mdp8-200 and Mdp8-224) found in Cluster-I, four (Mdp16-158, Mdp16-150, DKMP13-212, and DKMP13-230) in Cluster-II, two (Mdp8-284 and DKMP15-500) in Cluster-III. While, four, five, twelve alleles were not found in Cluster-I, Cluster-II, and Cluster-III, respectively (Figure9).

3.5. Association Mapping We identified 18 significant marker–trait associations (p < 0.00044, 0.1/228) for the nine traits using GLM and MLM models in association mapping. A greater number of significant marker–trait associations were observed using GLM (13) compared to MLM (5). Six of the detected MTAs exhibited R2 ≥ 10% in the general linear model or p < 0.00044 in the two models and were retained for further analysis (Table7). As an exception, we included the significant MTAs for loci DKMP17-282 associated with FCR, DK11-297 and DK11-271 associated with NLV both in GLM and MLM analysis with R2 values ranging from 9% to 9.2%, 7.63% to 7.73%, 6.87% to 8.48%, 9.48% to 9.51%, and 10.18% to 10.19%, respectively. Notably, the SSR loci MDP18-310 was associated with both SL and FAR, with R2 ranging from 10.01% to 13.53% in the GLM models. A significant MTA for DKMP18-156 with FCC was also detected using GLM analysis with R2 as 10.19%.

Forests 2021, 12, 422 12 of 20 Forests 2021, 12, x FOR PEER REVIEW 14 of 22

FigureFigure 8. Distribution 8. Distribution and quantityand quantity of 179 of 179D. kakiD. kakisamples samples from from different different regions regions in in Zhejiang Zhejiang Province.Province. Each Each circle circle repre represents‐ sents different ecologic regions of sampled D. kaki germplasms. P1–P16 represents the 16 populations of germplasms. The different ecologic regions of sampled D. kaki germplasms. P1–P16 represents the 16 populations of germplasms. The pie pie chart with different color parts represents the different clusters, and the blue parts represent Cluster‐I (C‐I), the red chart withparts differentrepresent colorCluster parts‐II (C represents‐II), and the the black different parts represent clusters, Cluster and the‐III blue(C‐III). parts Every represent dot represents Cluster-I every (C-I), germplasm, the red parts representthe blue Cluster-II dot represent (C-II), andthe germplasm the black parts of Cluster represent‐I, the red Cluster-III dots represent (C-III). the Every germplasm dot represents of Cluster every‐II, and germplasm, the black dots the blue dot representrepresent the the germplasm germplasm of Cluster Cluster-I,‐III. the red dots represent the germplasm of Cluster-II, and the black dots represent the germplasm of Cluster-III.

Table 6. Comparison of genetic diversity parameters for the germplasm clusters.

Cluster Sample Size Np Ne Nei I He PIC Cluster-I 70 213 1.5127 0.3041 0.4554 0.3015 0.2422 Cluster-II 61 220 1.5493 0.3211 0.4781 0.3188 0.2546 Cluster-III 48 198 1.5026 0.2985 0.4433 0.2954 0.2367 Mean 59.7 210.33 1.5215 0.3079 0.45893 0.3052 0.2445

Table 7. Marker–trait associations identified through the general linear model (GLM) and mixed linear model (MLM).

GLM MLM Organ Traits SSR Locus * p-Value R2 (%) p-Value R2 (%) DKMP17-282 0.00024404 9.2 0.00033412 9 Fruit crude fiber content DK11-197 0.00024404 8.56 0.00216 6.5 Fruit calcium content DKMP18-156 0.000072152 10.19 0.01611 4.24 Fruit Fruit water content DKMP13-440 0.0004181 6.28 0.01158 4.02 Fruit longitudinal diameter DK26-184 0.00022922 7.42 0.01737 3.4 Fruit aspect ratio MDP18-310 7.6122 × 10−7 13.53 0.00056065 7.39 DKMP7-600 0.0001777 7.81 0.0021 6.14 Seed width MDP8-288 0.000073631 8.69 0.00046859 8.01 Seed DKMP18-190 0.00028986 7.94 0.01601 3.87 Seed length MDP18-310 0.000042848 10.01 0.00603 5.06 Leaf aspect ratio DKMP15-514 0.00020945 9.29 0.01005 4.81 Leaf DK11-297 0.00013218 9.48 0.00027251 9.51 Number of lateral veins DK11-271 0.000072327 10.18 0.00016836 10.19

Note: * SSR locus, SSR maker-the size of allele (bp). Forests 2021, 12, 422 13 of 20 Forests 2021, 12, x FOR PEER REVIEW 15 of 22

Figure 9. Distribution of of alleles alleles among among clusters clusters per per locus. locus. The The pie pie chart chart with with different different color color parts parts representsrepresents the different allele numbers of of each each locus locus among among clusters, clusters, and and the the green green pie pie represents represents zero alleleallele ofof thethe lociloci existingexisting in in the the cluster, cluster, the the red red pie pie represents represents one one allele allele of of the the loci loci existing existing in in the the cluster, and the yellow pie represents more than one allele of the loci existing in the cluster. cluster, and the yellow pie represents more than one allele of the loci existing in the cluster.

4.3.5. Discussion Association Mapping PlantWe identified germplasms 18 significant are valuable marker–trait resources thatassociations can be used (p < for0.00044, the breeding 0.1/228) offor new the cultivarsnine traits as using well asGLM evolution and MLM studies. models Persimmon in association is abundantly mapping. and A widelygreater distributed number of insignificant Zhejiang marker–trait Province [13 ,associations14], which may were represent observed an using abundant GLM variability (13) compared resource to MLM com- prising(5). desired genes/alleles due to unique climate and diverse ecology. In the study, all morphologicalSix of the detected traits of theMTAs persimmon exhibited germplasms R2 ≥ 10% in the showed general broad linear ranges, model indicating or p < 0.00044 that thein the association two models panel and includes were retained diverse germplasmsfor further analysis that can (Table be used 7). in As breeding an exception, programs. we Therefore,included the the significant elucidation MTAs of genetic for loci structures DKMP17‐ and282 associated diversity patterns with FCR, of DK11 these‐ sources297 and DK11‐271 associated with NLV both in GLM and MLM analysis with R2 values ranging

Forests 2021, 12, 422 14 of 20

and the identification markers associated with agronomic traits using 17 SSR markers was performed in this study. In accordance with the present results, previous studies [12,13,15] have demonstrated that there are many differences in the morphological traits of persimmon germplasm from different eco-regions population, while few analyses have looked into the relationship between morphological traits and environmental factors in persimmon and similar species. In this study, mean temperature of the quarter showed little significance in morphological traits, perhaps due to all sampling sites belonging to the subtropical monsoon climate and no strict temperature limit for persimmon survival. Consistent with previous studies of many fruit trees, which revealed that light [59–61] and water supply [62–66] played an important cue in both leaf and fruit development, and precipitation and solar radiation were the main climate factors affecting the morphological traits of persimmon in Zhejiang Province. There were also longitudinal differences in morphological trait diversity. These can be attributed to the comprehensive climatic and terrain factors at the establishment site. The studied SSR markers exhibited a high level of genetic diversity. The mean PIC values were 0.8965 (i.e., all were above 0.5), the threshold indicating the existence of a very high level of marker polymorphism as suggested by Botstein [67]. The mean PIC value was much higher than those of available reports in Luo within a few germplasms [68], and comparable to 0.72 and 0.827 in the evaluation reports of 228 persimmon accessions through 12 gSSR markers [12] and 71 persimmon accessions through 19 gSSR markers [69], respectively, although this may be due to the diverse accessions in the present and other studies. The abundance of genetic information as demonstrated by the highly polymorphic SSRs will be helpful in understanding the genetic structure of the persimmon population in our study. Compared to the nationwide accessions or germplasms, the population mean value of I (0.3694) from the persimmon sources in Zhejiang Province were lower than the I (1.6014) in 133 accessions originating from China, Japan, and America using 17 neutral SSR markers in Liang’s report [11]. The diversity of persimmon sources in Zhejiang Province was higher than in the evaluation of 189 D. kaki germplasms in Guangxi Province through 13 gSSR markers, with a higher mean value of Nei (0.3694) and I (0.3694) in this study compared with the Nei (0.149) and I (0.231) in Guangxi Province [70]. This indicates that persimmon sources were distributed differently in the two provinces, which may have been caused by differences in the terrain and persimmon source evolutionary history or different SSR markers used in these studies. In addition, we detected significant relationships between Na and environmental factors. These results seem to be consistent with other research that found AFLP loci in Eu- ropean beech (Fagus sylvatica L.) [71] and European white birch (Betula pendula Roth.) [72], SSR loci in Eucalyptus gomphocephala DC. [73] and American chestnut (Castanea dentata (Mar- shall) Borkh.) [74]. Among them, the most highly correlated climatic factor was WinPR, WinSR, and SprMT, while SumMT and SumPR did not show any significance. This result indicates that the climate conditions in winter were more correlated to genetic traits than those in summer. Specifically, less precipitation and more solar radiation in the winter were indicative of a greater number of alleles within a population. SprMT was also a close factor affecting genetic parameters. A possible explanation for this might be that a cold spell in later spring is a serious obstacle for persimmon leaf germination. The distribution of persimmon resources might result from the natural selection and adaptation to the local climate, which may be one of the major causes of the longitudinal gradient in genetic diversity. A significant correlation between genetic distance and geographic distance was detected by the Mantel test. Additionally, a clear geographic origin-based population structuring was performed through comparing the genetic parameters and morphological traits among the population. For example, coastal populations, with low polymorphism levels but high fruit calcium content, fruit water content, and fruit and seed shape indices, may be marginal populations according to the central-marginal hypothesis that marginal populations show less genetic variation and higher differentiation than central popula- Forests 2021, 12, 422 15 of 20

tions [75–77]. The formation reason may partly be explained as these populations being located in the seaside, and these high mountains in eastern coastal areas also dampen the genetic exchange between coastal populations and inland populations, which was also supported by the low pairwise FST values. Meanwhile, there was extensive genetic ex- change among populations that originated inland or in a neighboring province though the long history of persimmon cultivation. The similar geographic patterns of gene exchange have also been found in China fir (Cunninghamia lanceolata (Lamb.) Hook.) [78], Sakhalin fir (Abies sachalinensis (F.Schmidt) Mast.) [79], and Mongolian oak (Quercus mongolica Fisch. ex Ledeb.) [80]. The 179 D. kaki germplasms from the 16 populations were broken up into three distinct clusters (I, II, and III) based on the comprehensive results of the NJ dendrogram, PCA and STRUCTURE, with a higher mean FST (0.2714). Nearly all samples of Cluster-I were distributed inland, possibly due to a lack of adaptation genes (such as resistance to salinity or to higher groundwater levels). The samples from coastal populations were mostly classified as Cluster-II and Cluster-III. The samples from Cluster-II and Cluster-III were widely distributed throughout Zhejiang Province. There may still be less gene (or germplasm) exchange between these coastal populations and inland populations. These germplasms from Cluster-I may be eliminated by coastal climate and environmental conditions (e.g., many typhoons, highly alkaline soils), and samples from the coastal region may contain advantageous genes that allow for survival under a wide range of environments. Although the formation mechanism of the distribution and genetic structure characteristics of persimmon resources in Zhejiang Province require further research, the information in the germplasms suggests that a gene locus of differential adaptability in a corresponding cluster will be an important resource for future environmental adaptation research in persimmon. Identification of the loci influencing agronomic traits is the basis of marker-assisted breeding. However, due to a lack of genomic information, association studies of other quan- titative traits have not been reported in persimmon. In the present study, we performed association mapping and identified markers associated with nine traits of agronomic importance. Improvement in fruit taste was the main breeding target of persimmon fruit. The fruit crude fiber content and calcium content were both closely related to fruit taste [81]. The DKMP17-282 loci was strongly correlated with FCR, as supported by a high correlation coefficient in both GLM and MLM analysis (R2 ranging from 9% to 9.2%). The DKMP18-156 (R2 = 10.19%) loci were identified as strongly correlated with the FCC in the GLM analysis. Hence, these SSR locus can be used in molecular-assisted breeding of persimmon fruit to improve their taste. Among horticultural crops, fruit size and shape are also important considerations for consumers. However, little has been done regarding studies on persimmon fruit shape due to the difficulty of quantitatively characterizing fruit shape features. By using association mapping in this study, we identified loci DK26-184 and MDP18-310 as linked with the FLD and FAR in persimmon, respectively. Among these loci, MDP18-310 (R2 = 7.39–13.53%) was consistently associated with the FAR in both models. In the present study, strong positive correlations between the shape traits of fruit and seed, with higher pairwise correlation coefficients and some geographic distribution pattern, consistent with previous observations in F1 progenies [82] and germplasm resources [15,83]. These results might be explained in part by immature seeds producing a mass of a gibberellin-like substance to promote fruit development [84]. Intriguingly, among the four MTAs of seed width or length, the loci MDP18-310, associated with the FAR, was also associated with SL in GLM analysis (R2 = 10.01%). Our results further support Haruka’s description [85] that variations in fruit and seed shapes among cultivars are coordinately controlled, and that MDP18-310 may be located in the genome region for controlling both these traits. Due to the complexity of the observed quantitative traits, these MTAs explain a small portion of phenotypic variance (about 10%) consistent with previous association mapping Forests 2021, 12, 422 16 of 20

studies [86–89] while, the breeding of quantitative traits in plants usually utilized the significant association combinations by cumulative genetic effects. In this study, the loci were both strongly correlated with the NLV, which was significantly associated with two MTAs (DK11-297 and DK11-271) with additive effects, which could add up to explain 19.7% of phenotypic variation without considering the intermarker effect. Although these significant marker–trait associations have still not been documented sufficiently due to lack of information on genomic location and gene function of these SSRs, our study will provide a reference for a genome-wide association study (GWAS) in persim- mon. These MTAs will be useful for the future marker-assisted breeding of persimmon.

5. Conclusions In this work, large-scale evaluation of persimmon germplasm in Zhejiang Province based on SSR markers was carried out for the first time. The results showed that there was a medium degree of genetic diversity found for persimmon in Zhejiang Province. The 179 D. kaki germplasms from the 16 populations could be separated into three clusters with distinct geographical pattern distribution. These provide a good basis for further research on the genetic basis of adaptation in D. kaki, and therefore the selection of suitable environment-marker associations for further breeding. In our study, 13 significant marker– trait associations were identified through the general linear model with nine traits of agronomic importance, of which, six MTAs with correlation coefficients (R2) > 10% were consistently represented in the general linear model or p < 0.00044 both in the general linear model and mixed linear model. As the first association mapping in D. kaki, the study will provide a reference for large scale MTA identification though a genome-wide association study (GWAS) in persimmon. The potential marker–trait associations identified in this study will be useful in facilitating marker-assisted breeding for D. kaki improvement and identification of candidate genes for trait variability in D. kaki.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/f12040422/s1, Table S1: Environmental factors, population, group, and cluster information of the 179 sampled D. kaki germplasms; Table S2: The statistics of the nine agronomic traits of the 179 D. kaki germplasms; Table S3: Characteristics of the 17 SSR markers used in this study; Table S4: All genotypic data across 179 D. kaki germplasms by 17 SSR markers; Table S5: The information of allele frequencies; Table S6: Correlation analysis between morphological traits for the 179 persimmon germplasms; Table S7: Genetic diversity of the 17 SSR markers in 179 persimmon germplasms; Table S8: The Na and Ho within each accession; Table S9: Correlation analysis between heterozygosity observed, number of alleles, and environmental factors for the 179 persimmon germplasms; Table S10: Pairwise FST estimates for groups resulting from the cluster analyses; Table S11: The molecular analyses of variance (AMOVA) among the germplasms according to origin of the populations; Table S12: Molecular analyses of variance (AMOVA) among the persimmon germplasm clusters. Author Contributions: Y.X. and B.G. conceived and designed the study; Y.X., B.G., X.J., and K.W. organized and conducted the trial sampling and managed the samples; Y.X., W.C., and C.X. carried out the laboratory analyses; Y.X. and W.C. collated, managed, and analyzed data; Y.X. and B.G. wrote the paper. All authors have read and agreed to the published version of the manuscript. Funding: The study was financially supported by the Fundamental Research Funds of CAF (CAFYB B2017ZA004-3 and CAFYBB2017SY015) and the Key Agricultural New Varieties Breeding Projects (New Varieties of Persimmon and Jujube Breeding Projects) funded by the Zhejiang Province Science and Technology Department. Data Availability Statement: The data presented in this study are available in the supplementary material here. Acknowledgments: The survey and sample collection work was kindly supported by numerous local forestry sectors such as the Fuyang Forestry Bureau and Lanxi Forestry Bureau. We are particularly grateful to the Plant Nursery of Lanxi City for their efforts in maintaining living plant materials for this study. Conflicts of Interest: The authors declare no conflict of interest. Forests 2021, 12, 422 17 of 20

References 1. Duangjai, S.; Wallnöfer, B.; Samuel, R.; Munzinger, J.; Chase, M.W. Generic delimitation and relationships in Ebenaceae sensu lato: Evidence from six plastid DNA regions. Am. J. Bot. 2006, 93, 1808–1827. [CrossRef] 2. Luo, Z.R.; Wang, R.Z. Persimmon in China: Domestication and traditional utilizations of genetic resources. Adv. Hortic. Sci. 2008, 22, 239–243. 3. Yonemori, K.; Sugiura, A.; Yamada, M. Persimmon genetics and breeding. Plant Breed. Rev. 2000, 19, 191–225. 4. Tang, D.L.; Zhang, Q.L.; Xu, L.Q.; Guo, D.L.; Luo, Z.R. Number of species and geographical distribution of Diospyros L. (Ebenaceae) in China. Hortic. Plant J. 2018, 5, 59–69. [CrossRef] 5. Wang, R.Z.; Yang, Y.; Li, G.C. Chinese persimmon germplasm resources. Acta Hortic. 1997, 436, 43–50. 6. Yang, Y.; Yang, T.T.; Jing, Z.B. Genetic diversity and taxonomic studies of date plum (Diospyros lotus L.) using morphological traits and SCoT markers. Biochem. Syst. Ecol. 2015, 61, 253–259. [CrossRef] 7. Guo, D.L.; Luo, Z.R. Genetic relationships of some PCNA persimmons (Diospyros kaki Thunb.) from China and Japan revealed by SRAP analysis. Genet. Resour. Crop. Evol. 2006, 53, 1597–1603. [CrossRef] 8. Du, X.Y.; Zhang, Q.L.; Luo, Z.R. Development of retrotransposon primers and their utilization for germplasm identification in Diospyros spp. (Ebenaceae). Tree Genet. Genomes 2009, 5, 235–245. [CrossRef] 9. Yonemori, K.; Honsho, C.; Kitajima, A.; Aradhya, M.; Giordani, E.; Bellini, E.; Parfitt, D.E. Relationship of European persimmon (Diospyros kaki Thunb.) cultivars to Asian cultivars, characterized using AFLPs. Genet. Resour. Crop. Evol. 2008, 55, 81–89. [CrossRef] 10. Parfitt, D.E.; Yonemori, K.; Honsho, C.; Nozaka, M.; Kanzaki, S.; Sato, A.; Yamada, M. Relationships among Asian persimmon cultivars, astringent and non-astringent types. Tree Genet. Genomes 2015, 51, 1–9. [CrossRef] 11. Liang, Y.Q.; Han, W.J.; Sun, P.; Liang, J.; Wuyun, T.N.; Li, F.D.; Fu, J.M. Genetic diversity among germplasms of Diospyros kaki based on SSR markers. Sci. Hortic. 2015, 186, 180–189. [CrossRef] 12. Guan, C.F.; Zhang, P.X.; Hu, C.Q.; Chachar, S.; Riaz, A.; Wang, R.Z.; Yang, Y. Genetic diversity, germplasm identification and population structure of Diospyros kaki Thunb. from different geographic regions in china using SSR markers. Sci. Hortic. 2019, 251, 233–240. [CrossRef] 13. Zhao, X.M.; Gong, B.C.; Wu, K.Y.; Chen, H.X.; Lv, X.; Wang, N. Research on quantitative classification of native persimmon varieties in Zhejiang province. For. Res. 2012, 25, 77–87. (In Chinese) 14. Zhao, X.M.; Gong, B.C.; Wu, K.Y.; Jiang, X.B.; Deng, Q.E.; Xiong, C.Y.; Zheng, X.F.; Chen, W.H. Variation analysis of fruit nutrients of native persimmon in Zhejiang province. J. Northwest. A. F. Univ. (Nat. Sci. Ed.) 2015, 43, 125–133. (In Chinese) 15. Sun, W.M.; Xu, Y.; Gong, B.C.; Wu, K.Y. Study on fruit shape diversity of persimmon germplasm resources in Zhejiang province. For. Res. 2019, 32, 42–50. (In Chinese) 16. Yu, J.; Buckler, E.S. Genetic association mapping and genome organization of maize. Curr. Opin. Biotechnol. 2006, 17, 155–160. [CrossRef][PubMed] 17. Abdurakhmonov, I.Y.; Abdukarimov, A. Application of association mapping to understanding the genetic diversity of plant germplasm resources. Int. J. Plant. Genom. 2008, 574927. [CrossRef][PubMed] 18. Ambreen, H.; Kumar, S.; Kumar, A.; Agarwal, M.; Jagannath, A.; Goel, S. Association mapping for important agronomic traits in safflower (Carthamus tinctorius L.) core collection using microsatellite markers. Front. Plant. Sci. 2018, 9, 402. [CrossRef][PubMed] 19. Mei, Y.J.; Yu, J.W.; Xue, A.G.; Fan, S.L.; Song, M.Z.; Pang, C.Y.; Pei, W.F.; Yu, S.X.; Zhu, J. Dissecting genetic network of fruit branch traits in upland cotton by association mapping using SSR markers. PLoS ONE 2017, 12, e0162815. [CrossRef] 20. Zheng, T.C.; Qin, B.; Li, S.Z.; Cai, M.; Zhang, Q.X. Screening of applicable SSR molecular markers linked to creeping trait in crape myrtle. Forests 2019, 10, 429. [CrossRef] 21. Liao, G.L.; Li, Z.Y.; Huang, C.H.; Zhong, M.; Tao, J.J.; Qu, X.Y.; Chen, L.; Xu, X.X. Genetic diversity of inner quality and SSR association analysis of wild kiwifruit (Actinidia eriantha). Sci. Hortic. 2019, 248, 241–247. [CrossRef] 22. Breria, C.M.; Hsieh, C.H.; Yen, J.Y.; Nair, R.; Schafleitner, R. Population structure of the world vegetable center mungbean mini core collection and genome-wide association mapping of loci associated with variation of seed coat luster. Trop Plant. Biol. 2019, 13, 1–12. [CrossRef] 23. Ali, N.; Li, D.; Eltahawy, M.S.; Abdulmajid, D.; Bux, L.; Liu, E.; Dang, X.; Hong, D. Mining of favorable alleles for seed reserve utilization efficiency in Oryza sativa by means of association mapping. BMC Genet. 2020, 21, 4. [CrossRef] 24. Liu, N.; Cheng, F.Y. Association mapping for yield traits in Paeonia rockii based on SSR markers within transcription factors of comparative transcriptome. BMC Plant. Biol. 2020, 20, 245. [CrossRef][PubMed] 25. Akagi, T.; Tao, R.; Tsujimoto, T.; Kono, A.; Yonemori, K. Fine genotyping of a highly polymorphic astringency-linked locus reveals variable hexasomic inheritance in persimmon (Diospyros kaki Thunb.) cultivars. Tree Genet. Genomes 2012, 8, 195–204. [CrossRef] 26. Onoue, N.; Kobayashi, S.; Kono, A.; Sato, A. SSR-based molecular profiling of 237 persimmon (Diospyros kaki Thunb.) germplasms using an astringency-linked marker. Tree Genet. Genomes 2018, 14, 28. [CrossRef] 27. Pei, X.; Zhang, Q.L.; Guo, D.Y.; Luo, Z.R. Effectiveness of the RO2 marker for the identification of non-astringency trait in Chinese PCNA persimmon and its possible segregation ratio in hybrid F1 population. Sci. Hortic. 2013, 150, 227–231. [CrossRef] 28. Akagi, T.; Kajita, K.; Kibe, T.; Morimura, H.; Tsujimoto, T.; Nishiyama, S.; Kawai, T.; Yamane, H.; Tao, R. Development of molecular markers associated with sexuality in Diospyros lotus L. and their application in D. kaki Thunb. J. Jpn. Soc. Hortic. Sci. 2013, 83, 214–221. [CrossRef] Forests 2021, 12, 422 18 of 20

29. Zhang, P.X.; Yang, S.C.; Liu, Y.F.; Zhang, Q.L.; Xu, L.Q.; Luo, Z.R. Validation of a male-linked gene locus (OGI) for sex identification in persimmon (Diospyros kaki Thunb.) and its application in F1 progeny. Plant. Breed. 2016, 135, 721–727. [CrossRef] 30. Tong, Y.W.; Lewis, B.J.; Zhou, W.M.; Mao, C.R.; Wang, Y.; Zhou, L.; Yu, D.P.; Dai, L.M.; Qi, L. Genetic Diversity and Population Structure of Natural Pinus koraiensis Populations. Forests 2020, 11, 39. [CrossRef] 31. Zuzana, B.; Jiri, K.; Jakub, D.; Jan, B.; Dagmar, Z.; Vaclav, J.; Milan, L. Genetic structure of norway spruce ecotypes studied by SSR markers. Forests 2020, 11, 110. 32. Shi, X.M.; Wen, Q.; Cao, M.; Guo, X.; Xu, L.A. Genetic diversity and structure of natural Quercus variabilis population in china as revealed by microsatellites markers. Forests 2017, 8, 495. [CrossRef] 33. Zhejiang Climate. Available online: http://zj.weather.com.cn/hyqx/index.shtml (accessed on 16 February 2020). 34. State forestry administration of the People’s Republic of China. Technical Regulation of Cultivation for Japanese Persimmon (Diospyros kaki Thunb); LY/T 1887-2010 [S]; Standards Press of China: Beijing, China, 2010; pp. 06–10. 35. Ross, I.; Robert, G. R: A Language for Data Analysis and Graphics. J. Comput. Graph. Stat. 1996, 5, 299–314. 36. Doyle, J.J.; Doyle, J.L. Isolation of plant DNA from fresh tissue. Focus 1990, 12, 13–15. 37. Soriano, J.M.; Pecchioli, S.; Romero, C.; Vilanova, S.; Llácer, G.; Giordani, E.; Badenes, M.L. Development of microsatellite markers in polyploid persimmon (Diospyros kaki Lf) from an enriched genomic library. Mol. Ecol. Notes 2006, 6, 368–370. [CrossRef] 38. Ruan, X.F.; Gabi, K.; Yang, Y. Isolation of Microsatellite in Diospyros kaki by magnetic beads. J. Northwest. A. F. Univ. (Nat. Sci. Ed.) 2008, 36, 97–102. 39. Guo, D.L. Establishment of Several Molecular Markers and Analysis of Genetic Relationships in Diospyros Linn. Ph.D. Thesis, Huazhong Agricultural University, Wuhan, China, 2006. 40. Hubisz, M.J.; Falush, D.; Stephens, M.; Pritchard, J.K. Inferring weak population structure with the assistance of sample group information. Mol. Ecol. Resour. 2009, 9, 1322–1332. [CrossRef] 41. Wang, Y.; Rashid, M.A.R.; Li, X.; Yao, C.; Lu, L.; Bai, J.; Li, Y.; Xu, N.; Yang, Q.; Zhang, L.; et al. Collection and Evaluation of Genetic Diversity and Population Structure of Potato Landraces and Varieties in China. Front. Plant. Sci. 2019, 10, 139. [CrossRef] 42. Baytar, A.A.; Erdogan, O.; Frary, A.; Frary, A.; Doganlar, S. Molecular diversity and identification of alleles for Verticillium wilt resistance in elite cotton (Gossypium hirsutum L.) germplasm. Euphytica 2017, 213, 31. [CrossRef] 43. Al-Ashkar, I.; Alderfasi, A.; Romdhane, W.B.; Seleiman, M.F.; EI-Said, R.A.; Al-Doss, A. Morphological and Genetic Diversity within Salt Tolerance Detection in Eighteen Wheat Genotypes. Plants 2020, 9, 287. [CrossRef][PubMed] 44. Singh, R.B.; Singh, B.; Singh, R.K. Development of potential dbEST-derived microsatellite markers for genetic evaluation of sugarcane and related cereal grasses. Ind. Crop. Prod. 2019, 128, 38–47. [CrossRef] 45. Singh, R.B.; Mahenderakar, M.D.; Jugran, A.K.; Singh, R.K.; Srivastava, R.K. Assessing genetic diversity and population structure of sugarcane cultivars, progenitor species and genera using microsatellite (SSR) markers. Gene 2020, 753, 144800. [CrossRef] [PubMed] 46. Yeh, F.C.; Boyle, J. POPGENE, the user-friendly shareware for population genetic analysis. Mol. Biol. Biotechnol. 1997, 434, 724–731. 47. Nagy, S.; Poczai, P.; Cernák, I.; Gorji, A.M.; Heged˝us,G.; Taller, J. PICcalc: An online program to calculate polymorphic information content for molecular genetic studies. Biochem. Genet. 2012, 50, 670–672. [CrossRef] 48. Smouse, P.E.; Banks, S.C.; Peakall, R. Converting quadratic entropy to diversity: Both animals and alleles are diverse, but some are more diverse than others. PLoS ONE 2017, 12, e0185499. [CrossRef] 49. Excoffier, L.; Laval, G.; Schneider, S. Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evol. Bioinform. 2005, 1, 47–50. [CrossRef] 50. Tamura, K.; Peterson, D.; Peterson, N.; Stecher, G.; Nei, M.; Kumar, S. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 2011, 28, 2731–2739. [CrossRef] [PubMed] 51. Price, A.L.; Patterson, N.J.; Plenge, R.M.; Weinblatt, M.E.; Shadick, N.A.; Reich, D. Principal components analysis corrects for stratification in genome-wide association studies. Nat. Genet. 2006, 38, 904–909. [CrossRef] 52. Manel, S.; Schwartz, M.K.; Luikart, G.; Taberlet, P. Landscape genetics combining landscape ecology and population genetics. Trends Ecol. Evol. 2003, 18, 189–197. [CrossRef] 53. Pritchard, J.K.; Stephens, M.; Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 2000, 155, 945–959. 54. Earl, D.A.; Vonholdt, B.M. Structure Harvester: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 2012, 4, 359–361. [CrossRef] 55. Jakobsson, M.; Rosenberg, N.A. CLUMPP: A cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 2007, 23, 1801–1806. [CrossRef] 56. Francis, R.M. Pophelper: An R package and web app to analyse and visualize population structure. Mol. Ecol. Resour. 2017, 17, 27–32. [CrossRef] 57. Bradbury, P.J.; Zhang, Z.; Kroon, D.E.; Casstevens, T.M.; Ramdoss, Y.; Buckler, E.S. TASSEL: Software for association mapping of complex traits in diverse samples. Bioinformatics 2007, 23, 2633–2635. [CrossRef][PubMed] 58. Wright, S. Genetic structure of populations. Br. Med. J. 1950, 2, 36. [PubMed] Forests 2021, 12, 422 19 of 20

59. Quian-Ulloa, R.; Stange, C. Carotenoid Biosynthesis and Plastid Development in Plants: The Role of Light. Int. J. Mol. Sci. 2021, 22, 1184. [CrossRef] 60. Mpai, S.; Sivakumar, D. Stimulation of Light-Emitting Diode Treatment on Defence System and Changes in Mesocarp Metabolites of Avocados Cultivars (Hass and Fuerte) during Simulated Market Shelf Conditions. Agronomy 2020, 10, 1654. [CrossRef] 61. Peavey, M.; Goodwin, I.; McClymont, L. The Effects of Canopy Height and Bud Light Exposure on the Early Stages of Flower Development in Prunus persica (L.) Batsch. Plants 2020, 9, 1073. [CrossRef][PubMed] 62. Reid, M.; Kalcsits, L. Water Deficit Timing Affects Physiological Drought Response, Fruit Size, and Bitter Pit Development for ‘Honeycrisp’ Apple. Plants 2020, 9, 874. [CrossRef][PubMed] 63. Silva, R.B.; Gabira, M.M.; Prado, D.Z.; Uesugi, G.; Simões, D.; Silva, M.R. Influence of Mean Leaf Angles and Irrigation Volumes on Water Capture, Leaching, and Growth of Tropical Tree Seedlings. Forests 2020, 11, 1198. [CrossRef] 64. Anna, K.-I.; Sylwia, Ł.; Marcin, Z.; Ewa, S.-O.; Wojtan, B. Variability of Leaf Wetting and Water Storage Capacity of Branches of 12 Deciduous Tree Species. Forests 2020, 11, 1158. [CrossRef] 65. Arenas-Navarro, M.; García-Oliva, F.; Terrazas, T.; Torres-Miranda, A.; Oyama, K. Leaf Habit and Stem Hydraulic Traits Determine Functional Segregation of Multiple Oak Species along a Water Availability Gradient. Forests 2020, 11, 894. [CrossRef] 66. Kim, C.-B.; Kim, Y.S.; Choi, H.T.; Kim, J.; Kim, S.; Cha, S.; Gao, G.-L.; Bao, Y.-F.; Son, Y.; Kwon, J.; et al. Sand Dune Height Increases Water Use Efficiency at the Expense of Growth and Leaf Area in Mongolian Pine Growing in Hulunbeier Steppe, Inner Mongolia, China. Forests 2019, 10, 558. [CrossRef] 67. Botstein, D.; White, R.L.; Skolnick, M.; Davis, R.W. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 1980, 32, 314–331. 68. Luo, C.; Zhang, F.; Zhang, Q.L.; Guo, D.Y.; Luo, Z.R. Characterization and comparison of EST-SSR and TRAP markers for genetic analysis of the Japanese persimmon Diospyros kaki. Genet. Mol. Res. 2013, 12, 2841–2851. [CrossRef] 69. Naval, M.D.M.; Zuriaga, E.; Pecchioli, S.; Llácer, G.; Giordani, E.; Badenes, M.L. Analysis of genetic diversity among persimmon cultivars using microsatellite markers. Tree Genet. Genomes 2010, 6, 677–687. [CrossRef] 70. Deng, L.B.; Liang, Q.Z.; He, X.H.; Luo, C.; Chen, H.; Qin, Z.S. Investigation and Analysis of Genetic Diversity of Diospyros Germplasms Using SCoT Molecular Markers in Guangxi. PLoS ONE 2015, 10, e0136510. [CrossRef] 71. Jump, A.S.; Hunt, J.M.; Martínez-Izquierdo, J.A.; Peñuelas, J. Natural selection and climate change: Temperature-linked spatial and temporal trends in gene frequency in Fagus sylvatica. Mol. Ecol. 2006, 15, 3469–3480. [CrossRef] 72. Kelly, C.K.; Chase, M.W.; De Bruijn, A.; Fay, M.F.; Woodward, F.I. Temperature-based population segregation in birch. Ecol. Lett. 2003, 6, 87–89. [CrossRef] 73. Bradbury, D.; Smithson, A.; Krauss, S.L. Signatures of diversifying selection at EST-SSR loci and association with climate in natural Eucalyptus populations. Mol. Ecol. 2013, 22, 5112–5129. [CrossRef] 74. Müller, M.; Nelson, C.D.; Gailing, O. Analysis of Environment-Marker Associations in American Chestnut. Forests 2018, 9, 695. [CrossRef] 75. Chen, Z.; Wu, B.; Li, J.; Zhao, T.; Zhou, X.; Zhang, Y. Germination of the seeds and growth of the seedlings of Cinnamomum camphora (L.) Presl. Plant. Species Biol. 2004, 19, 55–58. [CrossRef] 76. Doh, E.J.; Kim, J.H.; Oh, S.E.; Lee, G. Identification and monitoring of Korean medicines derived from Cinnamomum spp. by using ITS and DNA marker. Genes Genom. 2017, 39, 101–109. [CrossRef] 77. Panetta, F.D. Seedling emergence and seed longevity of the tree weeds Celtis sinensis and Cinnamomum camphora. Weed Res. 2010, 41, 83–95. [CrossRef] 78. Xu, Y.; Chen, J.H.; Zhao, Y.Q.; Wang, Y.; Wang, X.M.; Liu, W.D.; Shi, J.S. Variation of EST-SSR molecular markers among provenances of Chinese fir. J. Nanjing For. Univ. (Nat. Sci. Ed.) 2014, 57, 1–8. 79. Kitamura, K.; Uchiyama, K.; Ueno, S.; Ishizuka, W.; Tsuyama, I.; Goto, S. Geographical Gradients of Genetic Diversity and Differentiation among the Southernmost Marginal Populations of Abies sachalinensis Revealed by EST-SSR Polymorphism. Forests 2020, 11, 233. [CrossRef] 80. Zeng, Y.F.; Wang, W.T.; Liao, W.J.; Wang, H.F.; Zhang, D.Y. Multiple glacial refugia for cool-temperate deciduous trees in northern East Asia: The Mongolian oak as a case study. Mol. Ecol. 2015, 24, 5676–5691. [CrossRef] 81. Aghdam, M.S.; Hassanpouraghdam, M.B.; Paliyath, G.; Farmani, B. The language of calcium in postharvest life of fruits, vegetables and flowers. Sci. Hortic. 2012, 144, 102–115. [CrossRef] 82. Katayama-Ikegami, A.; Yonemori, K.; Sato, A.; Yamada, M.; Habu, T.; Kitajima, A. Relationship between astringency type and fruit shape in progenies of Chinese PCNA persimmon. Hortic. Res. 2013, 12, 29–34. (In Japanese) [CrossRef] 83. Maeda, H.; Akagi, T.; Tao, R. Quantitative characterization of fruit shape and its differentiation pattern in diverse persimmon (Diospyros kaki) cultivars. Sci. Hortic. 2018, 228, 41–48. [CrossRef] 84. Yamamura, H.; Naito, R. GA-like substances in immature fruit of kaki (Diospyros kaki L.). J. Jpn. Soc. Hort. Sci. 1973, 42, 1–6. [CrossRef] 85. Maeda, H.; Akagi, T.; Onoue, N.; Kono, A.; Tao, R. Evolution of Lineage-Specific Gene Networks Underlying the Considerable Fruit Shape Diversity in Persimmon. Plant. Cell Physiol. 2019, 60, 2464–2477. [CrossRef] 86. Resende, R.T.; Resende, M.D.; Silva, F.F.; Azevedo, C.F.; Takahashi, E.K.; Silva-Junior, O.B.; Grattapaglia, D. Regional heritability mapping and genome-wide association identify loci for complex growth, wood and disease resistance traits in Eucalyptus. New Phytol. 2017, 213, 1287–1300. [CrossRef] Forests 2021, 12, 422 20 of 20

87. Cao, K.; Zhou, Z.; Wang, Q.; Guo, J.; Zhao, P.; Zhu, G.; Fang, W.; Chen, C.; Wang, X.; Wang, X.; et al. Genome-wide association study of 12 agronomic traits in peach. Nat. Commun. 2016, 7, 13246. [CrossRef][PubMed] 88. Farneti, B.; Di Guardo, M.; Khomenko, I.; Cappellin, L.; Biasioli, F.; Velasco, R.; Costa, F. Genome-wide association study unravels the genetic control of the apple volatilome and its interplay with fruit texture. J. Exp. Bot. 2017, 68, 1467–1478. [CrossRef] [PubMed] 89. Herrera, C.M. Post-floral perianth functionality: Contribution of persistent sepals to seed development in Helleborus foetidus (Ranunculaceae). Am. J. Bot. 2005, 92, 1486–1491. [CrossRef]