Potential scrapie-associated polymorphisms of the prion protein gene (PRNP) in Korean native black goats

Small ruminants, including sheep and goats are natural hosts of scrapie, and the progression of scrapie pathogenesis is strongly influenced by polymorphisms in the prion protein gene (PRNP). Although Korean native goats have been consumed as meat and health food, the evaluation of the susceptibility to scrapie in these goats has not been performed thus far. Therefore, we investigated the genotype and allele frequencies of PRNP polymorphisms in 211 Korean native goats and compared them with those in scrapie-affected animals from previous studies. We found a total of 12 single nucleotide polymorphisms (SNPs) including 10 nonsynonymous and 2 synonymous SNPs in Korean native goats. Significant differences in allele frequencies of PRNP codons 143 and 146 were found between scrapie-affected goats and Korean native goats (p < 0.01). By contrast, in PRNP codons 168, 211 and 222, there were no significant differences in the genotype and allele frequencies between scrapie-affected animals and Korean native goats. To evaluate structural changes caused by nonsynonymous SNPs, PolyPhen-2, PROVEAN and AMYCO analyses were performed. PolyPhen-2 predicted “possibly damaging” for W102G and R154H, “probably damaging” for G127S. AMYCO predicted relatively low for amyloid propensity of prion protein in Korean native black goats. This is the first study to evaluate the scrapie sensitivity and the first in silico evaluation of nonsynonymous SNPs in Korean native black goats.

Small ruminants, including sheep and goat, are natural hosts of scrapie belonging to the group of transmissible spongiform encephalopathies (TSEs), which also includes bovine spongiform encephalopathy (BSE) in cattle, chronic wasting disease (CWD) in deer and elk, and kuru, fatal familial insomnia (FFI), Gerstmann-Sträussler-Scheinker syndrome (GSS) and Creutzfeldt-Jakob disease (CJD) in humans [1][2][3][4][5][6][7] . TSEs are characterized by neurodegenerative symptoms in brain tissue and are attributed to the conformational change from the normal prion protein (PrP C ) to the deleterious isoform of prion protein (PrP Sc ), which entails distributional changes in secondary structure of PrP C 8 .
Prion protein gene (PRNP) polymorphisms have a critical effect on prion disease susceptibility among a wide range of hosts 9,10 . In our previous studies, PRNP polymorphisms in Korean people, chickens, horses and cattle have been reported, and evaluation for susceptibility to prion diseases was performed [11][12][13][14][15][16][17][18][19] . In addition, several polymorphisms in paralogs of the PRNP gene have been reported in cattle and goats [20][21][22][23] . In humans, distributions of the single nucleotide polymorphisms (SNPs) at codons 129 and 219 of the PRNP gene are correlated with the susceptibility to sporadic and variant CJD 11 . For example, M129V and E219K heterozygotes are protective against the development of sporadic CJD. In addition, all variant CJD patients were 129MM homozygous 24 . In cattle, insertion/deletion polymorphisms in the regulatory region of the PRNP gene may influence the expression level of PrP C , leading to different incubation periods and an increase in BSE susceptibility 25,26 . According to recent studies, 23-bp insertion/deletion polymorphisms in the promoter region and 12-bp insertion/deletion polymorphisms in the promoter region of intron 1 of the PRNP gene are associated with BSE susceptibility 4,12,27 .
In sheep, several nonsynonymous SNPs of the PRNP gene have been identified at codons M112T, A136V, M137T, S138N, L141F, R151C, R154H, Q171R/H, N176K, and R211Q 5, [28][29][30] . Among these alleles, heterozygosity Results investigation of genetic characteristics of the PRNP gene in 211 Korean native black goats. We performed automatic direct sequencing at the open reading frame (ORF) of the PRNP gene in 211 Korean native black goats. The sequenced ORF in the Korean native black goats was 771 bp in length and homologous with the PRNP gene of Capra hircus registered in the GenBank website (Gene ID: EU870890.1). We found a total of 12 SNPs, including 10 nonsynonymous SNPs. The genotype and allele frequencies of the caprine PRNP gene are shown in Table 1 We also investigated the LD among the 12 SNPs of the caprine PRNP gene by analyzing Lewontin's D' (|D'|) values ( Table 2). The SNP c.379G > A showed low LD with 3 other SNPs (c.126G > A (0.511), c.414T > C (0.548), and c. 461G > A (0.492). In addition, c.718C > T showed low LD with c.461G > A (0.375). The remaining SNPs showed strong LD with a score range of 0.9-1.0.
Evaluation of potential scrapie susceptibility in Korean native black goats. To evaluate potential scrapie susceptibility in Korean native black goats, we compared the genetic distribution of scrapie-associated SNPs (R143H, N146S, R154H, P168Q, R211Q and Q222K) between Korean native black goats and scrapie-affected goats in other countries. These 6 scrapie-associated SNPs of the caprine PRNP gene, which were reported previously, were selected to evaluate potential scrapie susceptibility in Korean native black goats 3,6,33,34,[37][38][39]41,42 . In PRNP codons 143 and 146, there was a significant difference in allele frequencies between scrapie-affected Greek goats and healthy Greek goats (p < 0.01). In addition, a significant difference in allele frequencies of PRNP codons 143 and 146 was found between scrapie-affected Greek goats and Korean native black goats (p < 0.01) (Fig. 1A,B). The allele frequencies at PRNP codon 154 showed significant differences between scrapie-affected French goats and healthy French goats (p = 0.0011). Interestingly, the allele frequencies of this SNP were not significantly different between scrapie-affected French goats and Korean native black goats (p = 1.0) (Fig. 1C).  Table 2. Linkage Disequilibrium (LD) of twelve PRNP polymorphisms in Korean native black goats.  Table 3. Haplotype frequencies of 10 nonsynonyomous single nucleotide polymorphisms of PRNP gene in Korean native black goats. Others a contain rare haplotype with frequency < 0.01. www.nature.com/scientificreports www.nature.com/scientificreports/ The allele frequencies at PRNP codon 168 showed significant differences between scrapie-affected Greek goats and healthy Greek goats (p < 0.05). Interestingly, the allele frequencies of this SNP were not significantly different between scrapie-affected Greek goats and Korean native black goats (p = 0.109) (Fig. 1D). In French goats, a significant difference in allele frequencies at codon 211 of the PRNP gene was found between scrapie-affected goats and healthy goats (p < 0.001). There was no significant difference in the allele frequencies between French scrapie-affected goats and Korean native black goats (Fig. 1E). In PRNP codon 222 of French and Greek goats, significant differences in allele frequencies were detected between scrapie-affected goats and healthy goats. In addition, a significant difference in allele frequencies at this codon of the PRNP gene was found between French and Greek healthy goats and Korean native black goats (p < 0.001) (Fig. 1F).

Comparison of genetic distributions of PRNP codons 136, 154 and 171 in Korean native black goats and goats in other countries.
We investigated the caprine PRNP haplotypes of codons 136, 154 and 171 in Korean native goats. In addition, the PRNP haplotypes were compared with those previously reported in 4 countries, namely, Pakistan, China, Japan and France 33,[43][44][45] . In all countries, the ARQ haplotype was distributed over 97% of goats. In Pakistani goats, Chinese goats and Japanese goats, the frequency of the detected AHQ haplotype was less than 1%. Notably, ARR was found in over 1% of Japanese goats, and AHQ was found in 2.39% of French goats. In French goats, a significant difference was found in the allele distribution of the AHQ haplotype between scrapie-affected goats and healthy goats (p < 0.001) 33 . In addition, more than 95% of the ARQ/ ARQ genotype was distributed in all countries, and significant differences in the distribution of the ARQ/AHQ genotype in French goats were found between scrapie-affected goats and healthy goats. (p < 0.001). There were few ARQ/ARR genotypes in Korea (0.47%) and Japan (1.34%), and the ARR/ARR genotype was found only in Japan (0.67%) ( Table 4).
Lastly, we investigated amyloid propensity of goat prion protein according to alleles of nonsynonymous SNPs. Previous studies have been reported that prion protein with alleles of 143R, 146S, 154H, 211Q and 222 K (RSHQK) was highly resistant to conformational change for becoming deleterious form of prion protein.  (Fig. 2).

Discussion
Polymorphisms of the PRNP gene are major genetic determinants of the susceptibility to scrapie in sheep and goats. Although the prion protein in goats shares 99% protein sequence identity with that in sheep, the amino acid residues related to scrapie susceptibility are not identical 22,35 . Previous studies reported that the ovine PRNP gene was highly polymorphic and that the distributions of genotype and haplotype frequencies at codons 136, 154, and 171 were strongly related to the susceptibility of the scrapie progression 2,34,48,49 .
We examined the genotype and haplotype distributions of caprine PRNP codons 136, 154 and 171 from other countries, including Pakistan, China, Japan and France. As a typical feature, the Val allele of PRNP codon References Pakistani goats Chinese goats Japanese goats French goats

p-value
Korean native black goats 43 44 45 33 In this study ARR/ARR 0 (0.00) 0 (0.00) 2 (0.67) 0 (0.00) 0 (0.00) 0 (0.00) AHQ/AHQ 0 (0.00) 0 (0.00) 1 (0.33) 0 (0.00) 0 (0.00) 0 (0.00) www.nature.com/scientificreports www.nature.com/scientificreports/ 136, which is known to be associated with susceptibility to scrapie in sheep, was not found in goats, and no polymorphisms of codon 136 were found (Table 4). Interestingly, PRNP codon 171 polymorphism was found only in Korea and Japan. The major haplotype of the caprine PRNP gene was ARQ, and ARR haplotypes were  www.nature.com/scientificreports www.nature.com/scientificreports/ distributed at the lowest level in Japan and Korea. The major genotype ARQ/ARQ was distributed over 97% in Pakistani goats, Chinese goats, Japanese goats, French goats and Korean native black goats. In addition, there was a statistically significant difference in allele distribution between scrapie-affected and healthy goats in France with distributions of 2.39% for the AHQ allele (p < 0.001) and 4.78% for the ARQ/AHQ genotype (p < 0.001). The haplotypes and genotype distributions of French goats were 100% of ARQ and 100% of ARQ/ARQ, respectively, in scrapie-affected animals. The haplotype distribution in Korean native black goats was similar to that in Pakistani goats, Chinese goats and Japanese goats, and the genotype distribution in Korean native black goats was similar to that in Pakistani goats and Chinese goats (Table 4).
Next, we tried to evaluate the potential scrapie susceptibility by comparing distributions of PRNP polymorphisms in Korean native black goats with those of scrapie-affected goats. From previous studies, homozygotes of codons 143H, 146N, 154R, 211R and 222Q in the PRNP gene are known to be susceptible to goat scrapie 3,33,41,[50][51][52] . On the other hand, heterozygotes of codons H143R, N146S, R154H, R211Q, and Q222K in the PRNP gene are associated with a lower risk of developing classical scrapie 3,38,41,[52][53][54] . The PRNP codon 168 polymorphism has been reported in Greece, Italy, and Cyprus goats, and 168P homozygotes were associated with scrapie susceptibility 3,37,39,50 . Korean native black goats have 100% of the Ile allele in codon 142, and we did not find genetic polymorphisms. Because the 142I allele was associated with a shorter incubation period in experimentally challenged goats with the TSE isolate 42 , and the I142M heterozygote had a lengthened incubation period after experimental inoculation with BSE and scrapie 42 . In addition, Korean native black goats had similar allele distributions of the PRNP gene in codons 154R, 168P, 211R, and 222Q with those of scrapie-affected goats (Fig. 1C-F). In contrast, comparing the allele distributions of PRNP polymorphisms between healthy goats and Korean native black goats, codons 143H and 146S of Korean native black goats showed similar distributions to healthy goats (Fig. 1A,B). However, because the caprine PRNP gene has too many polymorphisms involved in susceptibility to scrapie, evaluating the susceptibility by comparing the genetic distribution of individual SNPs is difficult. Therefore, we analyzed LD and haplotype among PRNP SNPs.
In Korean native black goats, 10 polymorphisms showed strong LD among PRNP SNPs (Table 2). We found 4 major haplotypes, QWGHNRQRIP (0.363), QWGRNRQRIP (0.303), QWGHNRQRIS (0.146) and QWGHSRQRIP (0.045), by analyzing the haplotype distributions. (Table 3). Using the haplotype found in this study, a scrapie reagent inoculation test will be helpful for evaluating scrapie susceptibility in Korean native black goats in the future. However, since it cannot be ruled out that the differences of genotype, allele and haplotypes frequencies in the current study were a consequence by the lack of selective pressure due to negligible exposure to scrapie in Korea, additional confirmation studies are needed in the future.
In addition, we predicted the effects of nonsynonymous SNPs on caprine PRNP gene by PolyPhen-2 and PROVEAN analyses. In the PolyPhen-2 assay, G127S (0.992) was 'probably damaging' and W102G (0.603) and R154H (0.934) were 'possibly damaging' . Interestingly, all 10 nonsynonymous SNPs of PRNP gene were predicted to be 'neutral' by PROVEAN. The inconsistencies in the results of PolyPhen-2 and PROVEAN are due to the differences in algorithms that translate the impact on the protein function 46,55 . Because the PolyPhen-2 analysis is based on the structural influence of nonsynonymous SNPs on the protein and the PROVEAN analysis is based on phylogenetic differences among species, we presume that two nonsynonymous SNPs may have an effect on the structure of goat prion protein. Point mutation at the codon 102 of human PRNP gene (codon 105 in goat) has been linked to GSS of human familial prion diseases. In addition, polymorphism at the codon 154 of ovine PRNP gene (codon 154 in goat) has been associated with the susceptibility of scrapie in sheep. Thus, further investigation of caprine PRNP at the codons 102 and 154 is highly desirable in the future.
Next, we evaluated potential scrapie susceptibility according to alleles of nonsynonymous SNPs of caprine PRNP gene. Interestingly, the degree of amyloid formation of prion protein according to haplotypes of the PRNP gene of the Korean native black goats was lower than or equal to that of prion protein which are known to be resistance to prion diseases. Since scrapie in Korean native black goats has not been reported thus far, this result seems likely to be consistent with the data of AMYCO analysis. Thus, to the best our knowledge Korean native black goat presumed to have resistance to scrapie thus far. However, to verify the nonsynonymous effect of the goat prion protein, it is necessary to perform in vivo or in vitro experiments using models that contain these two polymorphisms.
Numerous scrapie-related SNPs have been investigated in small ruminants for identifying correlations between the genetic diversity of the PRNP gene and scrapie susceptibility in various countries; however, Korean native black goats have not been tested for scrapie susceptibility thus far. Here, we suggest that the multiple PRNP alleles of Korean native black goats can predict potential scrapie susceptibility and may contribute to the onset of the disease. In addition, to exclude the probability of quite different frequencies of PRNP gene polymorphisms in other regions of Korea, we collected 41 blood samples of Korean native black goats from a slaughter house of another region, which is located in Jecheon-si, Chungcheongbuk-do, Republic of Korea and investigated genotype, allele and haplotype frequencies of caprine PRNP gene. The genotype, allele and haplotype frequencies of caprine PRNP gene showed similar distributions between Chungcheongbuk-do goats and Jeollanam-do goats (Supplementary Table 1). Since direct transmission of classical scrapie to primates after a 10 years incubation period was recently reported 56 , it is very important to investigate the scrapie susceptibility of Korean native black goats to prevent the possibility of scrapie transmission through the species barrier to humans.
In conclusion, we performed direct sequencing and investigated the genotype and allele distributions of caprine PRNP gene polymorphisms in 211 Korean native black goats. We also reported the distributions of 4 major haplotypes and the strong LD among PRNP SNPs. Using comparative analysis of the genetic distributions in PRNP codons 136, 154 and 171 that are major contributors to the scrapie incidence of sheep, between sheep and Korean native black goats, we confirmed that sheep and Korean native black goats had significant differences in genetic distributions in PRNP codons 136, 154 and 171. In addition, we performed a potential scrapie susceptibility test for the first time in Korean native black goats through a comparison of the allele frequencies (2019) 9:15293 | https://doi.org/10.1038/s41598-019-51621-y www.nature.com/scientificreports www.nature.com/scientificreports/ of previously reported scrapie-associated SNPs. Furthermore, we investigated the damaging impact of nonsynonymous SNPs found in Korean native black goats for the first time using in silico analysis tools, PolyPhen-2 and PROVEAN. To the best of our knowledge, we conducted the first PRNP genetic study in Korean native black goats.

Methods ethical statement. All experimental procedures were approved by the Chonbuk National University
Institutional Animal Care and Use Committee (IACUC number: CBNU 2017-0076). All experiments using Korean native black goats were performed in accordance with the Korea Experimental Animal Protection Act.
Blood sample collection and DnA extraction. Blood samples of 211 Korean native black goats were collected from a slaughter house of Hwasun-gun, Jeollanam-do, the Republic of Korea. These samples were provided from 8 farms, which is located in Jeollanam-do. We collected samples 5 times from March 2016 to June 2016. The sample size used in the present study may be enough to identify rare polymorphisms, including below 1% genotype frequency. In addition, the sample size can also represent the total population of Korean native black goats with a 95% confidence level and a confidence interval of 7. Whole blood samples were treated with ethylenediaminetetraacetic acid (EDTA) and were frozen at −80 °C prior to analysis. Genomic DNA was purified from 200 μl frozen blood using the QIAamp DNA Blood Mini Kit (Qiagen, Valencia California, USA) following the instructions from the supplier.
Polymerase chain reaction (PCR) and DNA sequencing. To amplify the caprine PRNP gene, PCR was performed with gene-specific primers as follows: caprine PRNP-Forward (5′-ATTTTGCAGAGAAGT CATCATGGTGA-3′) and caprine PRNP-Reverse (5′-AACAGGAAGGTTGCCCCTATCCTA-3′). The primers were designed based on the genomic sequence of the caprine PRNP gene, which was registered in the GenBank website (Gene ID: EU870890.1). The PCR mixture contained 10 μM of each primer, 2.5 μl of 10 × Taq polymerase reaction buffer containing 25 mM of MgCl 2 , 2.5 mM of each dNTP mixture, and 2.5 units of Solg TM Taq DNA polymerase (SolGent, Daejeon, Republic of Korea). The PCR was carried out as follows: predenaturation at 95 °C for 2 min, 30 cycles of denaturation at 95 °C for 20 sec, annealing at 58 °C for 40 sec, extension at 72 °C for 1 min, and final extension at 72 °C for 5 min. The purification of PCR products for sequencing analysis was performed with a QIAquick Gel Extraction Kit (Qiagen, Valencia California, USA). The PCR products were directly sequenced by an ABI 3730XL sequencer (Applied Biosystems, Foster City, California, USA).
Statistical analysis. Analysis of HWE, LD and haplotype distributions of the PRNP gene in Korean native black goats were performed using Haploview Version 4.2 (Broad Institute, Cambridge, MA, USA). Differences in genotype, allele and haplotype frequencies of the PRNP gene were tested by chi-squire test (χ 2 ) or Fisher's exact test using SAS 9.4 Software (SAS Institute Inc., Cary, NC, USA).

Evaluation of nonsynonymous SNPs in the caprine prion protein.
We evaluated a total of 10 nonsynonymous SNPs of the PRNP gene using PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/) and PROVEAN (http://provean.jcvi.org/index.php). PolyPhen-2 utilizes a naive Bayes classifier to provide independent count (PSIC) scores using information from sequence alignment and protein structural properties 46,47 . PolyPhen-2 predictions are subdivided into three types, "benign", "probably damaging", and "possibly damaging", according to a numerical score ranging from 0.0 to 1.0. PROVEAN calculates the impact score by predicting the functional effect on all classes of protein functions for protein sequence changes such as deletion and multiple substitution as well as single amino acid substitutions 47 . The PROVEAN program predicts the biological impacts of nonsynonymous SNPs, including effects on protein function and structure. Score predictions form PROVEAN have two types of threshold; scores below −2.5 are considered "deleterious", and scores above −2.5 are considered "neutral". Amyloid propensity of caprine prion protein according to alleles of PRNP SNPs was analyzed by AMYCO (http:// bioinf.uab.cat/amyco) 57 . AMYCO is the algorithm used to predict amyloid fibril propensity from amino acid sequences.