An investigation of genetic diversity in three Dezhou donkey original breeding farms

Dezhou donkey is one of the excellent large donkey breeds in China. In our study, eight microsatellite markers were used to genotype from each of three populations of Chinese Dezhou donkeys: 67 individuals in Liaocheng (pop1), 103 individuals in Binzhou 1 (pop2), and 102 individuals in Binzhou 2 (pop3), in order to study the genetic diversity of these populations. A total of 213 alleles were detected, and the PIC results showed that eight loci were all highly polymorphic. The means of Ho and He of pop2 were the highest, which were 0.646 and 0.717, respectively. The PCoA analysis demonstrated that the samples from three conservation farms were mixed together. The phylogenetic tree showed that pop 2 and pop 3 were closely related. The phylogenetic tree results clustered that 272 donkeys were divided into six groups. AMOVA analysis showed that the genetic variation was mainly concentrated within the population and the genetic differentiation among the populations was low. Fst values between populations also indicated that genetic differentiation between populations was too small to be considered. This indicated a low probability of inbreeding within the population. And this also showed that the conservation and breeding of Dezhou donkeys in recent years have achieved excellent results. The investigation of genetic diversity in three Dezhou donkey original breeding farms can provide reference data for the selection and breeding of excellent breeds of Dezhou donkey.

www.nature.com/scientificreports/ horses using microsatellite markers. Kefena et al. 8 , Zhang et al. 9 , and Aranguren-méndez et al. 10 have analyzed the genetic diversity and variability of different donkey breeds. However, the analysis of genetic diversity among different Dezhou donkey populations by microsatellite markers has not yet been done and reported. In this study, the polymorphism of eight microsatellite loci in Dezhou donkey was investigated by capillary electrophoresis sequencing, and genetic diversity as well as genetic distance in three Dezhou donkey populations were analyzed. This provides the basis for the conservation of biodiversity and genetic resources of the Dezhou donkey.

Materials and methods
Moral statement. The experimental animals and methods used in this study were approved by the Animal Policy and Welfare Committee of Liaocheng University (no. LC2019-1). The care and use of laboratory animals were in full compliance with local animal welfare laws, guidelines and policies.
Sample collection. Samples were collected from three Dezhou donkey original breeding farms. One of the three breeding farms is located in Liaocheng City, Shandong Province, and two are located in Binzhou City, Shandong Province. A total of 272 individuals were randomly collected from three breeding farms, including 67 individuals in Liaocheng (pop1), 103 individuals in Binzhou 1 (pop2), and 102 individuals in Binzhou 2 (pop3), which represent the major genetic resources of Dezhou donkey. The farm location is shown in Fig. 2. Blood samples were collected from the jugular vein using an EDTA blood collection tube and stored immediately in a -20 °C refrigerator until use 11 . DNA extraction, PCR amplification and Microsatellite genotyping. DNA was extracted from blood using the M5 FlexGen Blood DNA kit (TIANGEN, Beijing, China). The DNA purity (OD260/OD280) was measured by spectrophotometer (B500, Metash, Shanghai, China), and the quantity of genomic DNA extracted was measured using 1% agarose gel 12 .
According to the results of previous studies 13 , eight microsatellite loci were analyzed. The primer sequences are shown in Table S1. The total volume of PCR reaction was 25 μL, including 12.5 µL PCR Mix (Mei5bio, Beijing,   PCR products were detected on 1% agarose gels, and those with specificity were separated by capillary electrophoresis using an ABI 3730xlDNAAnalyzer (Applied Biosystems, Foster City, CA, USA) 12 , the reagents were as follows: Applied Biosystems POP-7 Polymer and Hi-Di Formamide, running parameters: Run Voltage: 15,000, Injection Voltage: 1600, Injection Duration: 15, Temperature: 63, Laser Power: 12. Fluorescently labeled fragments were detected and sized using GeneMarker version 2.6.4 (Applied Biosystems, USA).
Population analyses. The Hardy-Weinberg equilibrium (HWE), Observed Heterozygosity (Ho), Expected Heterozygosity (He), No. Effective Alleles (Ne), No. Alleles (Na), number of private alleles (NPA) and principal coordinates analysis (PCoA) were calculated using GenAlEx version 6.51 15 . Polymorphic information content (PIC), the number of alleles (K), and the number of individuals at the locus (N) were calculated using Cervus version 3.0.7 16 . Percentage of estimated genetic distance from the Fst Values for Dezhou donkeys from the three populations were calculated using GenAlEx version 6.51 15 . Arlequin version 3.1 software was used for AMOVA analysis 17 . Popgene version 1.32 software was used to generate the phylogenetic tree of three Dezhou donkey populations 12 . Mega version 11.0 software and PowerMarker version 3.25 software were used to generate the phylogenetic tree of 272 Dezhou donkeys 18,19 . Institutional review board statement. Experiments and animal care in this study were approved by the Animal Welfare and Ethics Committee of Institute of Animal Sciences, Liaocheng University (no. LC2019-1). This study in accordance with the arrival guidelines.

Results and discussion
Genotyping results and polymorphism of microsatellite loci. Eight microsatellite markers were amplified in three Dezhou donkey populations. Figure 3A shows the genotyped heterozygous peak, and Fig. 3B shows the genotyped homozygous peak. Table 1 shows a summary of the genotyping results. The HMS2 locus had the highest number of alleles, while the HMS6 locus had the lowest number of alleles in pop 1 and pop 3. The TKY343 locus had the highest number of alleles, while the TKY337 locus had the lowest number of alleles in pop 2. The number of alleles at each of the eight loci in pop 2 and pop 3 was higher than that in pop 1, which may be due to the fact that pop 1 had fewer individuals than pop 2 and pop 3. Shang et al. 20 constructed a set of multiplex PCR systems for horses, which can amplify multiple microsatellite marker loci by a single PCR reaction, with the advantage of saving time and effort, but the study has not been reported by relevant studies on donkeys, and the method needs our further study on donkeys.
PIC is an indicator of gene abundance and its level reflects the diversity of the genetic base of a breed, the PIC reflects the genetic variation of microsatellite loci 21 . All eight loci in population 2 were highly polymorphic, all loci except TKY337 and HMS6 were highly polymorphic in pop 1, and all loci except HMS7 were highly polymorphic in pop 3 (PIC > 0.5), indicating that they were all highly polymorphic. In this study, abundant polymorphism has been observed for all three populations. Zhang et al. 9 found that the mean PIC for the seven polymorphic loci of the Dezhou donkey population was 0.4561, in the present study the PIC results for all three of our populations were higher than that result, this may be due to the large sample size. Consistent with the results of Kim et al. 7 , AHT4, ASB23, HMS2 and HMS3 were highly polymorphic in the donkey population.
Except for HMS7 and HMS6, the remaining loci showed different levels of Hardy-Weinberg equilibrium in the three populations. Four loci (TKY343, HMS2, ASB23, and HMS3) in population 1 were not in Hardy-Weinberg equilibrium, which may be due to the smaller sample size of pop 1 compared with pop 2 and pop 3. All eight loci in pop 3 were in Hardy-Weinberg equilibrium. Pop 2 had the highest number of private alleles, and the TKY343 locus had eight private allele numbers (Table S2). Six private alleles had frequencies greater than 5% (ASB23, 159 bp with a frequency of 28.8%; ASB23, 161 bp with a frequency of 14.4%; ASB23, 165 bp with a frequency of 17.4%; ASB23, 167 bp with a frequency of 25.0%; ASB23, 169 bp with a frequency of 6.8%; HMS3, 153 bp with a frequency of 8.2%). The private alleles with genotype frequencies > 5% were all in pop 1, which may be due to the relatively distant geographical location of pop 1 from population 2 and pop 3. Aranguren-Méndez et al. 10 . found that only one private allele (HTG6) had a frequency greater than 5% in Encartaciones breed. This may be due to the large number of populations selected for this study and the relatively long geographical distance of the populations.
Genetic diversity of three populations of Dezhou donkey. Genetic variability parameters were calculated for the three populations ( Table 2 and Table S3). The mean Na for three Dezhou donkey populations was 8.875, ranging from 6.750 (pop1) to 10.625 (pop 2). The Ne was the highest in the pop 2 (3.787) and lowest in the pop 1(3.366), with a mean of 3.527. The Ho and He of pop 2 were the highest, which were 0.646 and 0.717, respectively. The Ho and He of pop 1 were the lowest, which were 0.498 and 0.646, respectively. Interestingly, these findings were also found in other studies, for example, Aranguren-Méndez et al. 10 reported an average He in 13 microsatellite markers ranging between 0.637 and 0.684 for five endangered Spanish donkey breeds, which is basically consistent with our results. Bordonaro et al. 22 found that the He of the endangered Pantesco and two other Sicilian donkey breeds was lower than that of our study population, ranging from 0.471 for Pantesco to 0.589 for Grigio.    Figure 5 shows a phylogenetic tree of 272 Dezhou donkeys from 3 populations that was constructed by using the NJ method. The phylogenetic tree results clustered that the 272 donkeys were divided into 6 groups. Consistent with the PCoA results, pop 2 and pop 3 were closely related. One branch of the phylogenetic tree has only five Dezhou donkeys from pop 3, suggesting that the donkeys may be descended from the same breeding male. Two Dezhou donkey populations, pop 2 and pop 3, clustered together apart from pop 1 (Fig. 6). This may be due to the geographic proximity of pop 2 and pop 3 and the relatively distant geographic location of pop 1 from pop 2 and pop 3. There are only pop 2 and pop 3 on one of the branches of the phylogenetic tree, which indicates that there is gene exchange between pop 2 and pop 3. In the same way, Kim et al. 7 found that thoroughbreds overlap with Jeju Halla horses in the phylogenetic tree.
The genetic distances according to Nei (1972) are summarized in Table 3. In total, the three Dezhou donkey populations showed high genetic similarity (genetic distances between 0.053 and 0.229). The genetic distance between pop 3 and pop 1 and the genetic distance between pop 2 and pop 1 were higher than those between pop 1 and pop 2. This is consistent with the phylogenetic tree results. Wang et al. 10 found that the genetic distance of the Spanish donkey breeds ranged from 0.057-0.093, this is lower than our results.
The AMOVA analysis is a satisfactory grouping criterion for evaluating the variation within and among populations 23 . AMOVA analysis showed that 0.14382 of the genetic variation occurred among populations and 2.70564 of the genetic variation occurred within the population, indicating that the genetic variation was mainly concentrated within the population and the genetic differentiation among the populations was low ( Table 4 (Table 5). When the Fst value is 0-0.05, it indicates that the genetic differentiation among populations is very small and cannot be considered. Genetic structure coefficients, as described by F-statistics, are presented in Table 6

Conclusions
In conclusion, we evaluated genetic diversity in three Dezhou donkey breeding farms using microsatellite markers. The results showed that abundant polymorphism and low inbreeding levels were observed in all three populations, this provides a direction for the conservation of genetic diversity of Dezhou donkeys. However, this study was limited by the low sample size, and further investigation is needed to expand the sample size in the future.