SEARCH

SEARCH BY CITATION

Keywords:

  • Canine;
  • Genetics;
  • Pancreatitis;
  • SPINK1

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Background

Variants in the serine protease inhibitor Kazal type 1 (SPINK1) gene have been associated with pancreatitis in Miniature Schnauzers. Replication of the association in an independent population is necessary to determine if genetic screening for SPINK1 variants should be considered in clinical practice.

Hypothesis

An association between the SPINK1 exonic variant c.74A > C and pancreatitis exists in Miniature Schnauzers. In addition, the variant is absent or rare in Standard Schnauzers, a related breed that is not reported to have an increased risk for pancreatitis.

Animals

Case-control study. Seventeen Miniature Schnauzers with pancreatitis (cases), 60 mature Miniature Schnauzers with no substantial history of gastrointestinal signs in their lifetime (controls), and 31 Standard Schnauzers of unknown pancreatitis status.

Methods

A PCR-RFLP assay was used to genotype dogs for the c.74A > C SPINK1 variant. Allele and genotype frequencies were reported for Schnauzers and compared between case and control Miniature Schnauzers.

Results

The c.74A > C variant was the major allele in both Schnauzer breeds with a frequency of 0.77 in Miniatures and 0.55 in Standards. The allele and genotype frequencies were similar between Miniature Schnauzers with and without a history of pancreatitis and did not impart an increased risk for pancreatitis.

Conclusions and Clinical Importance

Genotyping a larger population of the Miniature Schnauzer breed than a previous study, along with a Standard Schnauzer cohort, demonstrated that the SPINK1 c.74A > C variant is a common polymorphism in the Schnauzer lineage. Furthermore, we were unable to confirm a relationship between the variant and clinically detectable pancreatitis in Miniature Schnauzers.

Abbreviations
Bp

base pair

cPL

canine pancreatic-specific lipase

LD

linkage disequilibrium

PCR

polymerase chain reaction

PCR-RFLP

PCR-restriction fragment length polymorphism

Miniature Schnauzers are reported to be at increased risk for pancreatitis,[1-3] but the underlying basis for this breed predisposition is not clear. One theory to explain the high risk is that pancreatitis is a consequence of hypertriglyceridemia, a prevalent problem in the breed.[4, 5] A 2nd theory is that Miniature Schnauzers suffer from a heritable form of pancreatitis. In support of the latter, 3 variants in the SPINK1 gene recently were identified in association with pancreatitis in Miniature Schnauzers: 2 missense mutations in exon 2 and a polyT insertion in intron 3.[3] SPINK1 encodes pancreatic secretory trypsin inhibitor, which protects the pancreas from damage caused by trypsin activation. Mutations in this gene have been described in people with chronic pancreatitis.[6]

The reported data on SPINK1 variants in Miniature Schnauzers are preliminary, but could have important clinical implications. If the genetic variants truly impart a major risk for pancreatitis, genotyping results could be used in breeding decisions or to identify dogs that may benefit from preventative strategies, such as a controlled, low-fat diet.[2] Thus, further investigation into the relationship between the variants and pancreatitis is crucial for determining if SPINK1 genetic testing should be offered in a clinical setting.

A major limitation of the previous SPINK1 study was a low number of appropriate controls. Pancreatitis is diagnosed at an average age of 8–9 years.[1, 2, 7, 8] Ideal controls in genetic studies should be well beyond the average age of disease onset without ever having demonstrated clinical signs during their lifetime. Only 11 of 25 control dogs in the previous study were more than 5 years old. Therefore, some of these control dogs could still go on to develop pancreatitis later in life. Furthermore, controls were only required to be free of clinical signs for the 3 months before study recruitment, and dogs with signs of pancreatitis earlier in life could have been erroneously included. These types of phenotyping errors can lead to false negatives and decrease the ability to detect an association between the variants and pancreatitis. The use of stricter control criteria could reveal an even stronger, more clinically relevant relationship.

The aim of this study was to test a large group of carefully phenotyped Miniature Schnauzer controls and a smaller group with a history of pancreatitis for one of the exonic (protein coding) SPINK1 variants previously associated with pancreatitis risk. Because the 3 variants were reported to be in nearly complete linkage disequilibrium (LD) with each other,[3] genotyping for one should accurately predict the presence of the others. We hypothesized that the association between the variant and clinical pancreatitis in Miniature Schnauzers would be confirmed and strengthened by this study.

A secondary aim was to evaluate for the SPINK1 variant in Standard Schnauzers. The Miniature Schnauzer is said to be derived from the Standard Schnauzer,[9] and a close relationship is supported by genetic marker clustering.[10] To the authors' knowledge, the Standard Schnauzer has not been included in any published reports on pancreatitis, and we hypothesized that the SPINK1 variant would be infrequent in the breed.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Phenotyping

Miniature Schnauzers were recruited from the University of Minnesota (UMN) Veterinary Medical Center (VMC) and a veterinary referral practice in Eden Prarie, Minnesota over an approximately 3-year period (March 2009–June 2012) for participation in genetic studies on metabolic disorders in this breed. Breeders from Minnesota and surrounding states (Iowa, Wisconsin, North Dakota and South Dakota) also were contacted and asked to refer dogs to the UMN VMC for participation. A clinical history and medical records were obtained for each dog and used to assign dogs to a pancreatitis or control group. All phenotype assignments were made by a single board-certified small animal internist (E.F.) before genotyping. Pedigrees were acquired when available and used to prevent inclusion of dogs that were related within 2 generations.

Pancreatitis was defined as compatible clinical signs (eg, anorexia, vomiting, diarrhea, abdominal pain) and fulfillment of at least one of the following criteria: (1) ultrasonographic findings consistent with pancreatitis as determined by a board-certified radiologist or small animal internist, (2) histopathologic changes diagnostic for pancreatitis, (3) Spec cPL1 >400 μg/L, positive SNAP cPL2 or both with clinical exclusion of other causes of similar clinical signs as determined by a review of abdominal radiographs, serum biochemistry results, and at least 1 year of follow-up.

Pancreatitis controls were defined as dogs at least 9 years of age with no history of clinically relevant gastrointestinal signs at any point in their lifetime. Clinically relevant gastrointestinal signs were defined as vomiting, diarrhea, or anorexia lasting for > 48 hours or requiring veterinary evaluation. Complete medical records were required to verify owner recall. Rescue dogs were excluded from the control group because their full history was not available. Dogs with diabetes mellitus also were excluded from the control group due to the potential for subclinical pancreatitis as the underlying cause of diabetes.

Stored DNA samples from 31 Standard Schnauzers additionally were tested. These dogs were not assigned to phenotype categories because minimal clinical information was available.

Blood Collection and DNA Extraction

Two to 4 mL of EDTA anticoagulated blood was obtained from each dog. Genomic DNA was extracted from whole blood using a commercially available kit.3 Sample acquisition was approved by the Institutional Animal Care and Use Committee, and informed consent was obtained from owners of study participants.

Genotyping

NEBcutter4 was used to identify differences in restriction enzyme sites between the canine reference sequence of SPINK1 exon 2 and the 2 previously described variants, c.60C > A and c.74A > C. No enzyme was identified with differential recognition of the first variant compared to the reference. One enzyme, XceI (NspI), was found to recognize and cut the reference sequence c.74A, but not the variant c.74C.

The canine reference sequence was used to design primers to amplify a 374 base pair (bp) product encompassing SPINK1 exon 2: forward primer 5′-CCCTCTGCCTATGTCTCTGC-3′ and reverse primer 5′-GGGCACTGCTGTTACTTTGC-3′. Standard PCR amplification was performed with 30 cycles and a 60°C annealing temperature on a MJ Research PTC-100 thermal cycler.5 The PCR product was incubated with 5 units of the Fermentas XceI6 enzyme at 37°C, overnight.

The PCR-RFLP (restriction fragment length polymorphism) assay products were resolved using gel electrophoresis. Dogs homozygous for the reference nucleotide c.74A had both a 172 bp and a 202 bp product, dogs homozygous for the c.74C allele had a single 374 bp product, and dogs heterozygous for the variant (AC) had all 3 products (172, 202, and 374 bp). Samples from 1 dog of each genotype (AA, AC, and CC) were directly sequenced with a standard Sanger sequencing technique and used as positive controls for all assays. Each result was reviewed independently by one of the researchers (E.F.) and a laboratory technician; both were masked to the phenotype assignment of the dog at the time the genotype call was made.

Statistical Analysis

A computer program[11] was used to perform an a priori power analysis with the following assumptions: power of 0.8, alpha of 0.05, and a 3:1 ratio of controls to cases. To replicate the previously reported risk (odds ratio, OR = 21.7) of pancreatitis in dogs with at least 1 copy of the variant allele,[3] 15 cases and 45 controls are needed. To replicate the reported risk (OR = 25) of pancreatitis in dogs with 2 copies of the variant allele, 6 cases and 18 controls are needed.

The frequency of the c.74C allele was calculated as follows: [(the number of AC dogs) + 2 × (the number CC dogs)]/2 × (total dogs). Chi-square tests were used to compare the frequency of the c.74C variant, the proportion of dogs with at least 1 copy of the variant, and the proportion of dogs homozygous for the variant between Miniature Schnauzer case and control groups. ORs and 95% confidence intervals were calculated. A 2-tailed Fisher's exact test was used when the expected counts were low (<5). A P value of < .05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Blood samples from 158 Miniature Schnauzers were collected, but only 77 dogs fulfilled inclusion criteria. Seventeen dogs met the criteria for the pancreatitis group. The group comprised of 11 spayed females and 6 neutered males. The mean age at the time of pancreatitis diagnosis was 8.7 years old. Ten dogs had ultrasound findings compatible with pancreatitis, and, of these, 2 were confirmed with histopathology (1 by laparoscopic biopsy and the other by necropsy), 2 had positive SNAP cPL results, and 1 had TLI above the reference range. The other 7 cases did not have ultrasonographic imaging, but were diagnosed by high canine-specific pancreatic lipase concentrations (5 positive SNAP cPL results, a Spec cPL = 461 μg/L, and a Spec cPL > 1,000 μg/L).

Sixty dogs met the criteria for the control group. The group consisted 27 spayed females and 33 neutered males. All dogs were over 9 years old, and the mean age at the time of phenotyping was 10.6 years old.

Pedigrees were available for 9 of 17 case and 50 of 60 control dogs. According to the inclusion criteria, no dog was related to another within 2 generations. Based on evaluation of 5 generation pedigrees, 26 extended families were represented in the study population.

Genotyping data for the SPINK1 exon 2 polymorphism are summarized in Table 1. The c.74C allele was the major allele in the Miniature Schnauzer population with an overall frequency of 0.77. The variant allele frequency was not significantly different between the pancreatitis case and control groups (0.82 and 0.75, respectively, = .4), and did not increase the risk for pancreatitis (OR = 1.6, 95% CI: 0.59–4.1). The proportion of dogs with at least 1 copy of the c.74C allele was not different between groups (P = .6), but too few dogs were free of the c.74C allele to calculate an OR for pancreatitis risk. No increased risk was observed for dogs homozygous for the variant allele (OR = 1.5, 95% CI: 0.49–4.6, P = .5). Of note, two of the control dogs that were homozygous for the c.74C allele had histopathologically normal pancreata on necropsy.

Table 1. SPINK1 c.74 allele frequency and genotype data for Miniature Schnauzers with and without pancreatitis and Standard Schnauzers of unknown clinical status
 Miniature SchnauzersStandard Schnauzers
CasesControlsTotal
  1. Genotypes are reported as the number of dogs (percentage).

  2. Standard Schnauzers had a significantly lower c.74C allele frequency (P = .002). None of the comparisons between Miniature Schnauzer case and control groups were significant at P < .05.

c.74C allele0.820.750.770.55*
Genotypes
  C/C11 (65)33 (55)44 (57)9 (29)
  C/A6 (35)24 (40)30 (39)16 (52)
  A/A0 (0)3 (5)3 (4)6 (19)
  Total17607731

Thirty-one Standard Schnauzers of unknown pancreatitis status were genotyped for the SPINK1 nucleotide c.74C. The allele was significantly less common in the Standard Schnauzer group (overall frequency = 0.55) compared to all Miniature Schnauzers (P = .002), but it represented the major allele in the breed. Pedigrees were not available for the Standard Schnauzers.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

This study failed to confirm an association between a coding variant in the SPINK1 gene and clinical pancreatitis in Miniature Schnauzers. The c.74A > C variant was previously described as a putative susceptibility mutation for pancreatitis in Miniature Schnauzers.[3] Unexpectedly, we found that c.74C was the major allele in both case and control Miniature Schnauzers and was present at similar frequency in both phenotype groups. The variant also was identified as the major allele in a population of Standard Schnauzers. These findings demonstrate that the SPINK1 c.74C allele is a common, potentially benign, polymorphism in the Schnauzer lineage. However, multiple factors should be considered in the interpretation of these results.

Breed predispositions indicate familial aggregation and support a genetic component to disease. However, the relationship between the genetic risk factor and disease may not be direct. Miniature Schnauzers have a strikingly high prevalence of idiopathic hypertriglyceridemia, with nearly 80% of dogs affected by 9 years of age.[12] Hypertriglyceridemia is a well-established risk factor for pancreatitis in people. The definitive mechanism for this risk is not known, but it has been proposed to be the result of pancreatic damage induced by free fatty acids.[13] Two recent studies have provided evidence for a link between hypertriglyceridemia and pancreatitis in Miniature Schnauzers.[4, 5] There is a significant positive correlation between triglyceride concentration and cPL concentration in the breed,[4] and Miniature Schnauzers with a history of pancreatitis are 5 times more likely to have abnormally high triglyceride concentrations than are age-matched controls.[5] This observation suggests that pancreatitis in Miniature Schnauzers is secondary to hypertriglyceridemia rather than a primary genetic disorder. The absence of an association between the pancreatic gene SPINK1 and pancreatitis in this study is compatible with this hypothesis.

In acute pancreatitis in humans, a cystic fibrosis transmembrane conductance regulator gene mutation is reported to impart risk only in the context of hyperlipidemia.[14] Triglyceride concentrations were not available for the majority of dogs in this study, and they also were not reported in the study that discovered the SPINK1 variants in Miniature Schnauzers.[3] Without this information, we cannot definitively rule out an interaction between triglyceride concentrations and SPINK1 variants in Miniature Schnauzers.

The high prevalence of the c.74C allele in both Miniature and Standard Schnauzers further suggests that it is a benign ancestral polymorphism of the Schnauzer lineage. Ninety-six percent of Miniature Schnauzers in this study had at least 1 copy of the variant, and 57% were homozygous. The allele frequency is dramatically higher than the reported prevalence of clinical pancreatitis in the breed (4.4%).[3] Thus, if this variant were causative for pancreatitis, the disease penetrance would have to be very low regardless of the inheritance pattern. Standard Schnauzers also had a high frequency of the c.74C allele (55%) in this study, but are not among the breeds reported to be at increased risk for pancreatitis.[2, 7, 15, 16] We were not able to find Standard Schnauzer mentioned in any published report on canine pancreatitis. However, the relatively low popularity of the Standard Schnauzer breed may limit detection of disease risk.[17] In addition, clinical data were not available for the Standard Schnauzers in this study, and pancreatitis status was therefore not known. Finally, as discussed above, it is possible that SPINK1 variants only impart risk in the presence of hypertriglyceridemia.

This study only investigated 1 of 3 SPINK1 variants that previously were associated with pancreatitis in Miniature Schnauzers.[3] Although it is possible that the true pancreatitis risk comes from 1 of the 2 unevaluated variants, we believe this scenario to be unlikely. Nearby alleles at a locus tend to be inherited together and show a statistical relationship referred to as linkage disequilibrium (LD). The 3 SPINK1 variants were reported to be in nearly complete LD in the previous study. Therefore, 1 variant should serve as a highly accurate marker for the presence of other 2 variants in Miniature Schnauzers.

Several hypotheses could account for the disparate findings of our study compared with the previous SPINK1 study on pancreatitis in Miniature Schnauzers. The overall c.74C allele frequency found in this study actually was similar to the results of the previous study (0.77 and 0.72, respectively). The difference in disease risk association between cases and controls in the previous study was due to a lower c.74C frequency in the control group (0.58). Only 25 control dogs were included in the initial study compared with 60 in this study. The researchers attempted to control for relatedness through the exclusion of dogs related within 2 generations. However, simply by chance, the control group may have included dogs of an ancestrally distinct origin that happened to have a lower prevalence of the variants. The control phenotyping for clinical pancreatitis was stricter in the current study, with an older age requirement and careful exclusion of dogs with clinically relevant gastrointestinal signs at any point in their lifetime. Thus, false negative controls are unlikely to account for the absence of disease association in this study. Although this study had a large control population, it was limited to 17 case dogs compared to 39 dogs in the former study. An a priori power analysis was used and determined that only 15 cases would be required to detect the previously reported differences in genotype frequencies between groups. However, if the true effect of the SPINK1 variant(s) is smaller than predicted in the power calculation, our sample size may not have been sufficient. For example, if the true pancreatitis OR for dogs with at least 1 copy of the variant is only 10 (half of what was predicted), 19 cases and 57 controls would be required for a power of 0.8 with an alpha of 0.05.

The biggest limitation of this study is the absence of a highly sensitive test to rule out subclinical pancreatitis in the control group. A subset of the control dogs had histopathology, ultrasonographic imaging of the pancreas, or gross evaluation, but the majority were not subject to gross or microscopic pancreatic inspection. Subclinical pancreatitis may be common. Studies of canine pancreatic samples from random postmortem examinations found histopathologic evidence of chronic pancreatitis in 34–50% of samples.[16, 18] We therefore cannot exclude a relationship between SPINK1 variants and subclinical pancreatic inflammation. Furthermore, most of the case dogs in this study did not have a SPEC cPL test performed, and none of the control dogs had this test performed. The SPEC cPL test was the primary method of phenotyping in the previous study. It is possible that the SPINK1 variants affect cPL concentrations without necessarily causing clinical signs of pancreatic disease.

In conclusion, the c.74A > C SPINK1 variant is a common polymorphism in Schnauzers, and we did not observe an association between the variant and clinically detectable pancreatitis in Miniature Schnauzers. Given the high LD among all 3 reported SPINK1 variants, it is likely that a haplotype encompassing all 3 of these variants is widespread in the population and does not impact disease risk. We cannot rule out a subclinical biological effect of the variant or a required interaction with hypertriglyceridemia for disease susceptibility. Nevertheless, the current data suggest that variants in the SPINK1 gene do not impart substantial risk for clinically detectable pancreatitis in Miniature Schnauzers, and a genetic screening test is not warranted at this time.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The authors thank Dr James R. Mickelson for providing comments on the manuscript, Katie M. Minor and Jessica A. Rick for technical assistance, and Dr L. Noelani Reinker for her aid in recruiting dogs.

Conflict of Interest: Authors disclose no conflict of interest.

Footnotes
  1. 1

    Spec cPL, Idexx Laboratories, Westbrook, ME

  2. 2

    SNAP cPL Test, Idexx Laboratories

  3. 3

    Puregene blood core kit, Qiagen Sciences, Germantown, MD

  4. 4

    tools.neb.com/NEBcutter2

  5. 5

    MJ Research, Inc., Watertown, MA

  6. 6

    Fermentas Life Sciences, Vilnius, Lithuanaia

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References