Materials and Methods

Sample population—Whole blood and serum samples were obtained from 39 Miniature Schnauzers with pancreatitis, 25 healthy Miniature Schnauzers, and 23 healthy dogs of other breeds. Miniature Schnauzers with pancreatitis were client-owned dogs, and healthy dogs of other breeds were owned by veterinary students and staff at the Texas A&M University College of Veterinary Medicine and Biomedical Sciences; healthy Miniature Schnauzers were owned by breeders. Informed written consent of owners was obtained for dogs enrolled in the study. The study protocol was reviewed and approved by the Texas A&M University Institutional Animal Care and Use Committee.

Pancreatitis was diagnosed in Miniature Schnauzers on the basis of the following criteria: clinical signs compatible with a diagnosis of pancreatitis (eg, vomiting, anorexia, abdominal pain, or a combination of these), an increase in serum cPLI concentration (cPLI concentration ≥ 200 μg/L), or an increase in serum cPL concentration (cPL concentration ≥ 400 μg/L).^c The database of the Gastrointestinal Laboratory at Texas A&M University was searched, and dogs with an increase in serum concentrations of cPLI or cPL were identified and recruited for inclusion in the study. The veterinarian of each identified Miniature Schnauzer was contacted and asked to complete a questionnaire regarding clinical signs and overall health of the affected dog. That veterinarian then was asked about the date of birth, sex, sexual status (sexually intact vs neutered), body weight, current diet, current medications, and medical history of the dog. When a dog met the criteria for inclusion, the veterinarian was asked to contact the owner for permission to enroll the dog in the study. When permission was granted by the owner and informed consent was obtained, each veterinarian scheduled collection of whole blood samples into tubes containing EDTA and into serum tubes. Owners were instructed to withhold food from their dog for at least 12 hours prior to the scheduled blood collection.

Samples from healthy Miniature Schnauzers were obtained as part of a separate study.⁹ These dogs were voluntarily enrolled by breeders located throughout the United States, and samples were collected by use of the same collection protocol as described for the Miniature Schnauzers with pancreatitis. Additionally, the owners were questioned and the pedigree of each dog was examined to ensure that the dogs were not related for at least 2 generations. Healthy dogs were those that did not have any prior history of pancreatitis, did not have clinical signs of any disease for 3 months prior to blood collection, and had serum concentrations of cPLI and cPL within the respective reference ranges.

Samples from healthy dogs of other breeds were obtained from dogs owned by veterinary students and staff at the veterinary college. Samples were obtained by use of the same protocol as that described for the Miniature Schnauzers with pancreatitis. This group of dogs included 5 mixed-breed dogs, 3 Chinese Shar Peis, 3 German Shepherd Dogs, 3 Lundehunds, 2 Beagles, 2 Border Collies, 1 Alaskan Malamute, 1 Boston Terrier, 1 Bull Mastiff, 1 Coonhound, and 1 Yorkshire Terrier.

Sample preparation and processing—All serum samples were analyzed for cPLI concentrations by use of an in-house ELISA (as described elsewhere)¹⁰ or for cPL concentrations by use of a commercially available assay.^c In other studies,^11–13,d investigators have reported that pancreatic lipase is specific for the pancreatic acinar cells and that the measurement of pancreatic lipase in serum is sensitive for the diagnosis of pancreatitis. Extraction of DNA from whole blood samples was accomplished by use of a commercially available kit^e; extraction was performed in accordance with the manufacturer's instructions.

Primer design, PCR assay, and sequencing—The SPINK1 gene in dogs encodes for a protein that consists of 80 amino acids. The gene is located on chromosome 2 at location 44,960,544 to 44,969,111, consists of 4 exons and 3 introns, and has a total exon length of 363 bp. Exons 1,2,3, and 4 consist of 98, 35, 107, and 123 bp, respectively. The nucleotide sequence of the SPINK1 gene in dogs is publicly available (GenBank accession No. XM_845464). This sequence was used to design primers to amplify all 4 exons and their respective intron boundaries (Appendix). Primers were designed by use of a commercially available software program.^f

A PCR assay was performed and optimized for each of the 4 exons. Briefly, for the first 3 exons, the 25-μL reaction mixture contained approximately 50 ng of genomic DNA, 10× buffer (15mM Tris HCl and 50mM KCl [pH, 8.0]), 2.5mM MgCl₂, 10μM of each deoxy-nucleoside triphosphate, 0.4μM of each sense and antisense primer, and 2 U of polymerase.^g The reaction mixture for the fourth exon was similar, except for a higher concentration of MgCl₂ (3.5mM). The PCR assay was performed by use of a commercially available thermocycler.^h Cycle conditions for exons 1 and 2 were 94°C for 4 minutes; 35 cycles at 94°C for 15 seconds, 67°C for 30 seconds, and 72°C for 45 seconds; and 72°C for 1 minute. Cycle conditions for exon 3 were 94°C for 5 minutes; 35 cycles at 94°C for 30 seconds, 67°C for 30 seconds, and 72°C for 1 minute; and 72°C for 2 minutes. Cycle conditions for exon 4 were 35 cycles at 94°C for 30 seconds, 56°C for 30 seconds, and 72°C for 30 seconds.

Aliquots of the PCR products were electrophoresed on 1% agarose gels at 100 V. Gels were stainedⁱ and exposed to UV light and then visually examined to determine purity and to ensure correct size of the amplicons. For the initial experiment, PCR amplicons were ligated into a vector^j and Escherichia coli organisms^k were transformed with the ligants. Plasmids were extracted by use of a commercially available plasmid purification kit^l in accordance with the manufacturer's instructions. The purified amplicon was then sequenced in both directions by use of a commercially available sequencing mix.^m The products were analyzed and separated on a DNA sequencer.ⁿ For the association experiment, PCR products were sequenced directly by use of the aforementioned materials and protocols.

Initially, all 4 exons and their intron boundaries were amplified from genomic DNA of 22 dogs, which consisted of 8 healthy Miniature Schnauzers, 6 Miniature Schnauzers with pancreatitis, and 8 healthy dogs of other breeds. Sequences were compared with each other and with published sequences by use of a commercially available software package.^o

Association screening—The DNA from 65 additional dogs was sequenced in 2 regions of the SPINK1 gene that were found to have 3 variants identified in the initial experiment. These additional dogs consisted of 17 healthy Miniature Schnauzers (total of 25 healthy Miniature Schnauzers in the study), 33 Miniature Schnauzers with pancreatitis (total of 39 Miniature Schnauzers with pancreatitis in the study), and 15 healthy dogs of other breeds (23 healthy dogs of other breeds in the study). Thus, serum samples from 87 dogs were analyzed for the variants of interest in the SPINK1 gene. Sequences were compared among dogs and with the published sequence. To determine if the variants found were in similar locations as those in humans and to establish whether these variants were in a conserved region of the protein, amino acid sequences were obtained from a publically available genome bank^p and aligned with a commercially available software program^q (Figure 1).

Ortholog comparison of the SPINK1 gene among species. Notice the location of 2 exon variants (normal amino acid in bold type). Numbers on the right side represent the amino acids contained in the sequence to the left. * All residues are identical. †Conserved substitution. ††Semiconserved substitution. K = Lysine. N = Asparagine. T = Threonine.

Citation: American Journal of Veterinary Research 71, 5; 10.2460/ajvr.71.5.527

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Ortholog comparison of the SPINK1 gene among species. Notice the location of 2 exon variants (normal amino acid in bold type). Numbers on the right side represent the amino acids contained in the sequence to the left. * All residues are identical. †Conserved substitution. ††Semiconserved substitution. K = Lysine. N = Asparagine. T = Threonine.

Citation: American Journal of Veterinary Research 71, 5; 10.2460/ajvr.71.5.527

• Download Figure
• Download figure as PowerPoint slide

Ortholog comparison of the SPINK1 gene among species. Notice the location of 2 exon variants (normal amino acid in bold type). Numbers on the right side represent the amino acids contained in the sequence to the left. * All residues are identical. †Conserved substitution. ††Semiconserved substitution. K = Lysine. N = Asparagine. T = Threonine.

Citation: American Journal of Veterinary Research 71, 5; 10.2460/ajvr.71.5.527

• Download Figure
• Download figure as PowerPoint slide

Statistical analysis—The proportions of Miniature Schnauzers with pancreatitis and variants of the SPINK1 gene were compared with the proportion of healthy Miniature Schnauzers with variants of the SPINK1 gene by use of the Fisher exact test, and ORs and their 95% CIs were calculated. Age of the dogs in the 2 groups of Miniature Schnauzers was compared by use of a Mann-Whitney U test and an unpaired t test. All data were tested for a normal distribution by use of the Kolmogorov-Smirnov test. Significance was set at values of P < 0.05.

Results

Analysis of sequencing results of the initial 22 dogs revealed 3 variants. In exon 2, 2 missense mutations were identified. The first was a C to A substitution at nucleotide 5, which caused an amino acid substitution for residue 20 (Asn → Lys, N20K). The second missense mutation was an A to C substitution at nucleotide 19, which caused an amino acid substitution at residue 25 (Asn → Thr, N25T; Figure 1). The third variant was located on intron 3. This variant was a poly T insertion and duplication mutation located 26 nucleotides from the end of exon 3 (IVS3+26–27ins(T)30–39,15_61dup11). All 3 variants were always detected together in healthy Miniature Schnauzers and in Miniature Schnauzers with pancreatitis. No variants were detected in the initial 6 healthy dogs of other breeds.

In the larger association screening experiment, the 2 exon variants always were detected together and in all 3 groups of dogs. The intron variant was only detected in the 2 groups of Miniature Schnauzers, where the 3 variants were detected in almost complete linkage disequilibrium with each other. Interestingly, 1 healthy Miniature Schnauzer was heterozygous for the 2 exon variants but did not have the intron variant.

Overall, the 3 variants were significantly (P = 0.030) associated with pancreatitis in Miniature Schnauzers. Miniature Schnauzers with pancreatitis were 9.5 times (95% CI, 1.0 to 87.0) as likely to have at least 1 copy of each of the 3 variant alleles as were healthy Miniature Schnauzers. Additionally, pancreatitis in Miniature Schnauzers was significantly associated with homozygosity for all 3 variants, compared with results for healthy Miniature Schnauzers. Comparing the proportion of dogs homozygous for these variants with the proportion of dogs that had wild-type alleles yielded an OR of 13.9 (95% CI, 1.4 to 135.6 [P = 0.014]). Comparing the proportion of dogs homozygous for the 3 variants with the proportion of dogs heterozygous for the 3 variants or that had wild-type alleles yielded an OR of 3.4 (95% CI, 1.1 to 9.0 [P = 0.040]). However, heterozygosity for all 3 variants was not significantly (P = 0.605) associated with pancreatitis in Miniature Schnauzers, compared with results for healthy Miniature Schnauzers. Finally, healthy Miniature Schnauzers were significantly (P = 0.001) more likely to harbor the exon variants than were the healthy dogs of other breeds (OR, 9.1; 95% CI, 2.4 to 34.3).

Median age of Miniature Schnauzers with pancreatitis differed significantly (P < 0.001) from the median age of healthy Miniature Schnauzers. Therefore, a second control group of healthy Miniature Schnauzers was generated that included only dogs > 5 years old, which have been reported¹⁴ to be at an increased risk for developing pancreatitis. Exclusion of the younger dogs resulted in a control group of 11 healthy Miniature Schnauzers. Results of an unpaired t test revealed that there was no significant difference (P = 0.661) in the mean ages of the 2 groups after excluding healthy dogs ≤ 5 years old (Table 1).

Analysis of the data with the control group of healthy Miniature Schnauzers > 5 years old revealed that Miniature Schnauzers with pancreatitis were 21.7 times (95% CI, 2.1 to 224.4 [P = 0.006]) as likely to have at least 1 copy of each of the 3 variant alleles as were the 11 healthy Miniature Schnauzers. Additionally, pancreatitis in Miniature Schnauzers was significantly associated with homozygosity for all 3 variants, compared with results for the 11 healthy Miniature Schnauzers. Comparing the proportions of dogs homozygous for these variants with the proportion of dogs that had wild-type alleles yielded an OR of 25 (95% CI, 2.2 to 284.8 [P = 0.007]). Comparing the proportion of dogs homozygous for the 3 variants with the proportion of dogs heterozygous for the 3 variants or that had wild-type alleles yielded a result that was no longer significant (P = 0.166; Fisher exact test). Similar to the results for the analysis that involved data for all 25 healthy Miniature Schnauzers, heterozygosity for all 3 variants was not significantly (P = 0.134) associated with pancreatitis in Miniature Schnauzers, compared with results for the healthy control group of 11 Miniature Schnauzers > 5 years old (Table 2).

Discussion

Pancreatitis is the disease that most commonly affects the exocrine portion of the pancreas in dogs.¹⁵ Although it has been traditionally believed that acute pancreatitis is the more common manifestation in dogs, recent results¹⁵ would suggest that chronic pancreatitis may be more common in dogs than is acute pancreatitis. Little is known about whether chronic pancreatitis is caused by recurrent bouts of acute pancreatitis or whether this condition represents a separate syndrome. Some potential risk factors for the development of pancreatitis in dogs include obesity, trauma, high-fat diets, pharmaceuticals, endocrinopathies, and hypertriglyceridemia.¹⁵ However, most cases of pancreatitis in dogs are considered idiopathic because a cause is rarely definitively identified.¹⁶ Because of the breed prevalence of pancreatitis in dogs, it has been proposed^14,17,18 that genetics play an important role in the development of pancreatitis in some breeds, such as the Miniature Schnauzer, Yorkshire Terrier, and possibly other breeds.

As mentioned previously, the SPINK1 gene in humans is believed to be one of the defensive mechanisms of the pancreas against premature activation of trypsinogen.³ Mutations of the SPINK1 gene have been reported in several studies^4,6,19,20 and are thought to be the cause of hereditary pancreatitis in humans. However, some authors^5,21 maintain that mutations found in the SPINK1 gene do not cause hereditary pancreatitis, but instead, they merely act as disease modifiers by lowering the threshold for pancreatitis or increasing the severity of pancreatitis caused by other genetic or environmental factors. For example, investigators in 1 study²² reported that when the most common mutation, N34S, was recombinantly expressed, the subsequent protein did not have altered ability to inhibit trypsin. However, this mutation is also in linkage with 4 other intron mutations, one of which may be causative.⁴ Other mutations of the SPINK1 gene have been identified and have been found to be detrimental to protein function or expression. An extensive review of these findings has been reported elsewhere.²³ Finally, mutations in other genes, such as the gene for cationic trypsinogen, have been associated with pancreatitis in humans and may be related to, or may act in conjunction with, mutations in the SPINK1 gene.^24–26

To the authors' knowledge, this is the first study in which variants of the SPINK1 gene have been associated with pancreatitis in dogs. In this study, the 2 exon variants (ie, N20K and N25T) were always detected together in all dogs. However, the intron variant could only be identified in Miniature Schnauzers and also was cosegregated with the exon variants in this breed, except for 1 healthy Miniature Schnauzer. This exception could have been attributable to a random crossover event or may have been a result of offspring from an accidental mating with a dog of another breed during a previous generation in that dog's pedigree.

The first exon variant is a highly conserved asparagine residue reported in many species (Figure 1). The second asparagine at position 25 appears to be unique to dogs. The intron variant is on intron 3 (located 26 nucleotides from the end of exon 3), and the wild-type intron sequence in the region of the variant is identical to its human ortholog, which suggests conservation among species. In the study reported here, Miniature Schnauzers homozygous for the 3 variants were significantly more likely to have pancreatitis. In addition, none of the healthy dogs of other breeds had the intron variant.

Variants found in our study differ from those described in humans with hereditary pancreatitis and mutation of the SPINK1 gene.²³ However, an intron-exon boundary mutation has been reported in humans.²⁷ This boundary mutation is near the location of the intron variant detected in the dogs of the study described here. Furthermore, cDNA analysis of this mutation in humans revealed that exon 3 was skipped in the final transcript.²⁷

Several reasons could exist for the fact that healthy Miniature Schnauzers had a high prevalence of the 3 variants. First, although the dogs were healthy at the time of sample collection, the variant may predispose them to development of pancreatitis later in life. This speculation is supported by the fact that the healthy Miniature Schnauzers were significantly younger than were the Miniature Schnauzers with pancreatitis. In this study, age-matched control dogs would have been ideal; however, it was exceedingly difficult to find healthy older Miniature Schnauzers that fulfilled our criteria for inclusion, and we believed it was important to evaluate a large number of dogs. These difficulties were attributable to the fact that most of the older Miniature Schnauzers had clinical signs of chronic diseases, such as diabetes mellitus and urolithiasis, for which Miniature Schnauzers are also predisposed.^28,29 However, after exclusion of Miniature Schnauzers ≤ 5 years old, it was found that despite having a smaller population of dogs to analyze, being homozygous for the intron variant was still significantly associated with pancreatitis.

Second, mild or subclinical pancreatitis may have easily been missed by the owner because mild pancreatitis may be manifested as occasional bouts of vomiting or other nonspecific clinical signs. Also, a prior episode of pancreatitis may have been misdiagnosed as another condition. It is also possible that the gene is highly prevalent but has a low penetrance; thus, it may be necessary for other genes or other environmental factors to be present for pancreatitis to develop. Additionally, Miniature Schnauzers have a high prevalence of a syndrome of primary hypertriglyceridemia, which could be the cause of pancreatitis or a necessary cofactor.⁹ Hypertriglyceridemia was not evaluated in the study reported here because it would have been difficult to classify the dogs as having hypertriglyceridemia prior to the onset of pancreatitis or to determine whether the hyperlipidemia was secondary to pancreatic inflammation.¹⁴ Further studies are in progress to investigate the role of hyperlipidemia in the development of pancreatitis in Miniature Schnauzers.

Finally, although the cPLI concentration is the serum marker with the highest sensitivity (64% to 82%) for the detection of pancreatitis and is believed to be extremely specific, it is not possible to exclude the fact that some of the healthy dogs may have had mild subclinical pancreatitis. ^11,30,d,r,s Histologic evaluation of the pancreas is considered to be the criterion-referenced standard for the diagnosis of pancreatitis in dogs, but it was not considered feasible in our study. Biopsy of the pancreas is an invasive, expensive procedure that is not routinely performed in dogs suspected of having pancreatitis. Although some dogs may have been misdiagnosed on the basis of the measurement of the cPLI or cPL concentrations, we believe that an elevated serum cPLI or cPL concentration in combination with clinical signs compatible with pancreatitis is the most feasible and economical method for phenotype assignment in such a large group of dogs in the United States.

In this study, other dog breeds were found to carry only the exon variants and at a lower frequency than for the Miniature Schnauzers; however, many breeds were not included in the study. It could be that the exon variants are single-nucleotide polymorphisms that have been passed on through generations and that, perhaps, the intron variant represents the causative mutation in Miniature Schnauzers with chronic pancreatitis.

One problem encountered in this study was that Miniature Schnauzers with pancreatitis were significantly older than were healthy Miniature Schnauzers. Pancreatitis develops more commonly in older dogs, and this might have accounted for the discrepancy of the prevalence of pancreatitis in the 2 populations of Miniature Schnauzers (healthy vs with pancreatitis). However, to more accurately compare the true frequency of the variants, healthy Miniature Schnauzers ≤ 5 years old were excluded and the data were analyzed again, which confirmed the original findings.

In addition, population stratification could be another cause of the higher frequency of variant alleles detected in the Miniature Schnauzers with pancreatitis because a relationship among those Miniature Schnauzers could not be conclusively excluded. However, Miniature Schnauzers with pancreatitis that were enrolled in the study came from numerous locations throughout the United States, and we believe that this helped to minimize the possibility of selection of a population that was more inbred than is the general Miniature Schnauzer population. Finally, only a few other dog breeds were included in our study; thus, the variants identified might also be found in breeds that were not included in the study reported here.

Additional studies are needed to further evaluate the pathogenetic impact of the 3 variants identified here. Such studies would include evaluation of the prevalence of the 3 variants in the general population of healthy dogs, dogs with pancreatitis, and distinct groups of dogs, such as other breeds with a suspected hereditary predisposition for developing pancreatitis, dogs with hypertriglyceridemia, and dogs receiving treatment with drugs associated with pancreatitis.³¹ Additionally, studies focusing on possible effects of these mutations on the structure and function of the SPINK1 gene, mode of inheritance, and penetrance are warranted. Finally, a genome-wide search for other genes that may be contributing to pancreatitis in Miniature Schnauzers as well as follow-up monitoring of the healthy Miniature Schnauzers with variants to determine whether they will develop pancreatitis in the future are needed.

In the study reported here, we detected 3 closely associated variants (2 exon and 1 intron) of the SPINK1 gene in healthy Miniature Schnauzers and in Miniature Schnauzers with pancreatitis. These variants were significantly associated with pancreatitis. The exon variants were also detected in healthy dogs of other breeds but at a lower frequency than in the healthy Miniature Schnauzers. We conclude that defects of the SPINK1 gene likely play a role in development of pancreatitis in Miniature Schnauzers. However, we also hypothesize that other environmental or genetic factors may contribute to this disease. Further studies of these genetic variants are warranted.