The First Report of Polymorphisms and Genetic Features of the prion-like Protein Gene (PRND) in a Prion Disease-Resistant Animal, Dog

Prion disease has displayed large infection host ranges among several species; however, dogs have not been reported to be infected and are considered prion disease-resistant animals. Case-controlled studies in several species, including humans and cattle, indicated a potent association of prion protein gene (PRNP) polymorphisms in the progression of prion disease. Thus, because of the proximal location and similar structure of the PRNP gene among the prion gene family, the prion-like protein gene (PRND) was noted as a novel candidate gene that contributes to prion disease susceptibility. Several case-controlled studies have confirmed the relationship of the PRND gene with prion disease vulnerability, and strong genetic linkage disequilibrium blocks were identified in prion-susceptible species between the PRNP and PRND genes. However, to date, polymorphisms of the dog PRND gene have not been reported, and the genetic linkage between the PRNP and PRND genes has not been examined thus far. Here, we first investigated dog PRND polymorphisms in 207 dog DNA samples using direct DNA sequencing. A total of four novel single nucleotide polymorphisms (SNPs), including one nonsynonymous SNP (c.149G>A, R50H), were identified in this study. We also found two major haplotypes among the four novel SNPs. In addition, we compared the genotype and allele frequencies of the c.149G>A (R50H) SNP and found significantly different distributions among eight dog breeds. Furthermore, we annotated the c.149G>A (R50H) SNP of the dog PRND gene using in silico tools, PolyPhen-2, PROVEAN, and PANTHER. Finally, we examined linkage disequilibrium between the PRNP and PRND genes in dogs. Interestingly, we did not find a strong genetic linkage between these two genes. To the best of our knowledge, this was the first genetic study of the PRND gene in a prion disease-resistant animal, a dog.

According to previous studies, susceptibility to prion disease can be influenced by three major factors. One is the amount of prion protein expression. The 12-bp and 23-bp insertion/deletion (indel) polymorphisms in the promoter region of the bovine PRNP gene have been strongly associated with the expression level of bovine prion protein and the susceptibility to prion disease in cattle. Overexpression of prion protein based on the haplotype of 23-bp and 12-bp polymorphisms has been considered a vulnerability factor for BSE [8,16,17]. In addition, ablation of prion protein in knockout animals did not cause infection with prion disease [18]. However, there was no significant difference in the expression level of prion protein in the brain between dogs and other animals [19]. Another factor is the genetic characteristic of the prion protein. Polymorphisms of the prion protein gene (PRNP) in several species have been shown to be strongly correlated with prion disease susceptibility. In sheep, the haplotypes of codons 136, 154 and 171 affect susceptibility (VRQ, ARQ) or resistance (ARR) to scrapie [20,21]. The goat prion protein codon M142 extends the scrapie incubation period. In addition, codons S146N, R154H and Q222K protect against scrapie [22][23][24][25][26]. In humans, the PRNP genotypes of codons 129 and 219 are well known for being factors of CJD susceptibility [27,28]. In prion-resistant species, the dog prion protein D163 residue strongly contributes to prion disease resistance [29][30][31][32][33][34][35]. However, because a high dose of PrP Sc infection converted dog PrP C to PrP Sc , it is not sufficient to fully explain prion disease resistance in dogs [19]. The other factor is other candidate genes besides prion protein that can affect the susceptibility to prion diseases. Among them, the prion-like protein gene (PRND) is a potent candidate gene that may play a role in prion disease susceptibility. According to case-controlled studies comparing the genetic distribution of PRND gene polymorphisms in codons 26, 56, 132 and 174 and 3' untranslated region (UTR) +28, these polymorphisms were involved in the susceptibility to scrapie, BSE and sporadic CJD in ruminants and humans [36][37][38][39][40]. In addition, recent studies have reported a strong linkage disequilibrium (LD) between the PRNP gene and PRND gene in prion disease-susceptible species, sheep and goats [38,41]. However, no study has been conducted on the genetic characteristics of PRND in dogs. Thus, the investigation of the genetic characteristics of the PRND gene in dogs as a prion disease-resistant species will be a very important baseline study to obtain clues on the progression of prion disease.
In the present study, we investigated the dog PRND genotype, allele and haplotype frequencies of single nucleotide polymorphisms (SNPs). We also annotated nonsynonymous SNPs using in silico analysis tools, PolyPhen-2 [42], PROVEAN [43,44] and PANTHER [45]. In addition, we performed LD tests among PRND SNPs and analyzed major haplotypes of PRND SNPs. Furthermore, we measured the LD value between the PRNP gene and PRND gene in dogs.

Investigation of Genetic Characteristics of the PRND Gene in the 207Dogs
The dog PRND gene is composed of two exons. To investigate polymorphisms of the PRND gene in dogs, we performed direct sequencing analysis targeting exon 2, which contains the full length ORF. We found a total of four novel SNPs: c.149G>A, c.447T>C, c.465C>T in the ORF and c.556G>C in the 3 UTR of the PRND gene (Figure 1a  Among the four SNPs, c.149G>A (R50H) is a nonsynonymous SNP. Detailed values of the genotype and allele frequencies of the dog PRND gene are described in Table 1. We also investigated LD among the four dog PRND SNPs using (|D'|) and r 2 values. All four SNPs have strong LDs with D' values of 1.0 according to the (|D'|) value. In addition, the r 2 value showed a value of 1.0 between c.149G>A and c.556G>C and between c.447T>C and c.465C>T (Table  2). Next, we performed haplotype analysis of the dog PRND gene. The two major haplotypes were identified, including GGTC and ACTC, and the GGTC haplotype had the highest frequency (87.4%) in the dog PRND gene (Table 3).   Among the four SNPs, c.149G>A (R50H) is a nonsynonymous SNP. Detailed values of the genotype and allele frequencies of the dog PRND gene are described in Table 1. We also investigated LD among the four dog PRND SNPs using (|D'|) and r 2 values. All four SNPs have strong LDs with D' values of 1.0 according to the (|D'|) value. In addition, the r 2 value showed a value of 1.0 between c.149G>A and c.556G>C and between c.447T>C and c.465C>T (Table 2). Next, we performed haplotype analysis of the dog PRND gene. The two major haplotypes were identified, including GGTC and ACTC, and the GGTC haplotype had the highest frequency (87.4%) in the dog PRND gene (Table 3). Table 2. Linkage Disequilibrium (LD) among four polymorphisms of PRND gene in dogs. Next, we compared the genotype and allele frequencies of the c.149G>A (R50H) SNP in the dog PRND gene among eight dog breeds using the chi-square test. Maltese has no significant difference from Pomeranian (p = 0.454), Chihuahua (p = 1.0), Mixed (p = 0.155), and Cocker Spaniel (p = 0.488) in genotype frequency. However, Maltese has a significantly different genotype distribution with Shih Tzu (p = 0.0019), Toy Poodle (p = 0.0036) and Yorkshire Terrier (p = 0.0067). In addition, the allele frequency of Maltese was significantly different from that of Shih Tzu (p = 0.0014), Toy Poodle (p = 0.0058) and Yorkshire Terriers (p = 0.005) ( Figure 2).
Next, we compared the genotype and allele frequencies of the c.149G>A (R50H) SNP in the dog PRND gene among eight dog breeds using the chi-square test. Maltese has no significant difference from Pomeranian (p =0.454), Chihuahua (p =1.0), Mixed (p =0.155), and Cocker Spaniel (p =0.488) in genotype frequency. However, Maltese has a significantly different genotype distribution with Shih Tzu (p =0.0019), Toy Poodle (p =0.0036) and Yorkshire Terrier (p =0.0067). In addition, the allele frequency of Maltese was significantly different from that of Shih Tzu (p =0.0014), Toy Poodle (p =0.0058) and Yorkshire Terriers (p =0.005) (Figure 2).

Analysis of the Genetic Linkage between SNPs of the PRNP and PRND Genes
To examine whether dog PRND SNPs have a genetic linkage with SNPs of the dog PRNP gene, we carried out LD analysis between SNPs of these two genes. Figure 3 summarizes LD analysis using the r 2 value. Interestingly, all PRND SNPs showed weak LD with PRNP SNPs (r 2 value: below 0.3). Detailed values are described in Supplementary Table 1 and Supplementary Table 2.

Analysis of the Genetic Linkage between SNPs of the PRNP and PRND Genes
To examine whether dog PRND SNPs have a genetic linkage with SNPs of the dog PRNP gene, we carried out LD analysis between SNPs of these two genes. Figure 3 summarizes LD analysis using the r 2 value. Interestingly, all PRND SNPs showed weak LD with PRNP SNPs (r 2 value: below 0.3). Detailed values are described in Supplementary Tables S1 and S2.  (Figure 2).

Analysis of the Genetic Linkage between SNPs of the PRNP and PRND Genes
To examine whether dog PRND SNPs have a genetic linkage with SNPs of the dog PRNP gene, we carried out LD analysis between SNPs of these two genes. Figure 3 summarizes LD analysis using the r 2 value. Interestingly, all PRND SNPs showed weak LD with PRNP SNPs (r 2 value: below 0.3). Detailed values are described in Supplementary Table 1 and Supplementary Table 2.

Measurement of Protein Functional Alterations Induced by Nonsynonymous SNPs
We measured the damage of nonsynonymous SNPs (R50H) using PolyPhen-2, PROVEAN and PANTHER. PolyPhen-2 predicted R50H to be 'benign' and scored 0.051. The PROVEAN program analyzed R50H to be 'neutral', with a score of -1.065. PANTHER predicted R50H to be 'probably benign' and scored 30 (Table 4).

Measurement of Protein Functional Alterations Induced by Nonsynonymous SNPs
We measured the damage of nonsynonymous SNPs (R50H) using PolyPhen-2, PROVEAN and PANTHER. PolyPhen-2 predicted R50H to be 'benign' and scored 0.051. The PROVEAN program analyzed R50H to be 'neutral', with a score of -1.065. PANTHER predicted R50H to be 'probably benign' and scored 30 (Table 4).

The Sequence Alignments of Doppel Protein among Several Species
Finally, we performed amino acid sequence alignment of prion-like protein (Doppel) between dogs and other species (human, mouse, sheep, goat, rabbit, and horse). When compared to other species, dog prion-like protein has eight dog-specific amino acids, including leucine (L) in codon 18

Discussion
The PRND gene is in the same family of genes as the PRNP gene. Structural similarities with the PRNP gene and an association with prion disease have been reported in previous studies [44]. Recently, a strong genetic linkage between the PRNP gene and PRND gene was identified, and scrapie-associated SNPs were strictly linked to the genotype of the PRND gene [38,41]. Because those studies have been performed in prion disease-susceptible species, it is elusive whether strong

Discussion
The PRND gene is in the same family of genes as the PRNP gene. Structural similarities with the PRNP gene and an association with prion disease have been reported in previous studies [44]. Recently, a strong genetic linkage between the PRNP gene and PRND gene was identified, and scrapie-associated SNPs were strictly linked to the genotype of the PRND gene [38,41]. Because those studies have been performed in prion disease-susceptible species, it is elusive whether strong genetic linkage was a prion disease-susceptible factor or a general property among prion gene families. Here, we first investigated dog PRND gene polymorphisms and analyzed the genetic linkage between the dog PRNP gene and the dog PRND gene. We found a total of four novel SNPs, including one nonsynonymous SNP. Among the four SNPs, the R50H SNP has already been identified in cattle. In addition, four PRND SNPs have a strong genetic linkage and construct two major haplotypes (Tables 2 and 3). Interestingly, genotype and allele frequencies are significantly different among eight dog breeds ( Figure 2). Next, we investigated LD analysis between the PRNP gene and PRND gene. Notably, there was no strong LD between PRND and PRNP SNPs using r 2 analysis. Previous studies have indicated that prion disease-susceptible species, such as sheep and goats, have a strong LD between the PRNP gene and the PRND gene, which is very interesting. Furthermore, because dogs have a relatively short genetic distance between the PRNP gene and PRND gene compared to those of other species (cattle: 26 kb; goat: 21 kb; sheep: 25 kb; human: 20 kb; horse: 16 kb; dog: 17 kb), weak LD between the PRNP and PRND genes was assumed not to be induced by the genetic distance between the PRNP gene and PRND gene. In addition, it is notable that prion disease-resistant animals, including horses and dogs, have a relatively short genetic distance between the PRNP gene and PRND gene compared to prion disease-susceptible animals. Because only three studies have performed (including the present study) LD analysis between the PRNP and PRND gene, further investigation of LD in a wide range of animals in the future is highly desirable. The important thing is that dogs did not have strong LD between the PRNP gene and the PRND gene, unlike prion disease-susceptible species.
Finally, we estimated the nonsynonymous SNP R50H using PolyPhen-2, PROVEAN and PANTHER. Three in-silico estimation programs did not predict R50H to be deleterious to doppel protein function. However, previous studies indicated that synonymous SNPs located in codon 26 of the PRND gene can impact Doppel function, especially the reproductive ability of sperm [46,47]. In addition, synonymous SNPs may impact transcription efficiency and can affect the phenotype of a protein [48]. Thus, in vivo or in vitro confirmation studies based on genotype, allele and haplotype distributions of the dog PRND gene will be needed in the future. In recent studies, the prion gene family member prion-related protein gene (PRNT), which is located downstream of the PRND gene, was expressed in reproductive organs in a manner similar to the PRND gene, and a strong genetic linkage with the PRND gene has also been reported [49][50][51][52]. Because a strong genetic LD block has been detected in sheep and goats, further confirmation of the genetic linkage between these adjacent genes is also needed in the future.

Genetic Analysis
Genomic DNA was extracted from 200 µL of a whole blood sample using a Hi Yield Genomic DNA Mini Kit (Real Biotech Corporation, Taipei, Taiwan) and a Bead Genomic DNA Prep Kit (Biofact, Daejeon, Korea). Polymerase chain reaction (PCR) was performed with forward primer: 5 -AGAAAGTAACTGCCCCGAGC-3 and reverse primer: 5 TTTGGTACCTTGGGGACACG-3 . These primers were designed based on the PRND gene sequence from GenBank (Gene ID: 485782) and amplified DNA sequences containing the entire open reading frame (ORF) of the dog PRND gene. The length of the PCR products was 688 bp. A 25 µL reaction mixture containing 2.5 µL of 10X Taq DNA polymerase, 1 µL of genomic DNA, 10 pmol each primer, 0.5 µL of a 0.2 µM dNTP mixture, 0.2 µL of Taq DNA polymerase, and sterile deionized water, was used. The PCR conditions were as follows: denaturing at 95 • C for 2 min, followed by 34 cycles of 95 • C for 20 s, 63 • C for 30 s, and 72 • C for 1 min 30 s and one cycle of 72 • C for 5 min. The PCR products were separated on a 1% agarose gel stained with ethidium bromide (EtBr), and the PCR products were purified using a FavorPrep GEL/PCR Purification Mini Kit (FAVORGEN, Pingtung County, Taiwan). Purified PCR products were directly sequenced using an ABI 3730 sequencer (ABI, Foster City, CA, USA); sequencing results were visualized using Finch TV software (Geospiza Inc., Seattle, WA, USA), and genotyping was performed.

Statistical Analysis
Genotype and allele frequencies of the dog PRND gene were compared among eight dog breeds by chi-square test using SAS 9.4 software (SAS Institute Inc., Cary, NC, USA). Haplotype analysis and linkage disequilibrium calculated by Lewon-tin's D' (|D'|) and pairwise linkage disequilibrium (r 2 ) were performed using Haploview version 4.2 (Broad Institute, Cambridge, MA, USA).

Analysis of the Genetic Linkage between SNPs of the PRNP and PRND Genes
LD analysis was performed between PRNP and PRND SNPs. LD scores of the PRNP and PRND genes were calculated in 174 animals. Next, the genotype distributions of PRND were compared with those of the PRNP gene, and the difference of distribution was calculated using the chi-square test. All statistical analyses were calculated by Statistical Analysis Software (SAS), version 9.4 (SAS Institute Inc., Cary, NC, USA), and statistically significant differences were determined by p value < 0.05.

Conclusions
In summary, we investigated SNPs of the dog PRND gene. We identified four novel SNPs, including 1 nonsynonymous SNP. A total of 4 SNPs constitute two major haplotypes with strong LD. In addition, we found significantly different distributions of genotype and allele frequencies among eight dog breeds. Furthermore, we evaluated the LD value between the PRNP gene and the PRND gene in dogs and found relatively weak LD compared to that in prion disease-susceptible animals, such as sheep and goats. Lastly, we performed in silico estimation of the nonsynonymous SNP of the PRND gene R50H using PolyPhen-2, PROVEAN and PANTHER. To the best of our knowledge, this is the first genetic study of the dog PRND gene.