Figure 1.
Client-owned US Labradors share similar morphological and histopathological features with French CNM dogs from the experimental pedigree.
French CNM (A–C) and US HMLR (D–F) affected dogs have atrophic skeletal muscles, the most affected being those of pelvic limbs (e.g. biceps femoris muscle, arrows in A,D). B,C,E,F are Hematoxylin-Eosin-stained transverse sections of the biceps femoris muscle from 6-month-old (B, FR-4; E, US-18) or 10-year-old (C,F) affected Labradors. Early signs include groups of atrophic fibers, surrounded by endomysial (e) and perimysial (p) fibrosis. In older dogs, increased internalized or centralized nuclei (asterisks) and fatty infiltration (f) are observed. Scale bar = 50 µm.
Table 1.
Number of genotyped dogs used in this study.
Table 2.
Numbers by genotype of Labradors diagnosed with HMLR or phenotypically similar myopathy.
Figure 2.
PTPLA mutation in the initial confirmation panel of CNM dogs.
(A) Wild-type (wt = 610 bp) and PTPLAcnm (cnm = wt+238 bp) alleles in a healthy carrier (FR-1) and an affected (FR-2) Labrador from the French experimental pedigree. (B) Segregation of the PTPLAcnm allele in a four-generation pedigree of a client-owned US proband female (arrow). (C) Genotypes of client-owned Labradors from several countries, diagnosed with CNM-related myopathies (asterisks; Table S1). US-6 is a champion known to have produced CNM pups; DE-5 is a control affected by a neuropathy and FR-2 was reloaded for size comparison.
Figure 3.
Percentage of wild-type homozygous (+/+), healthy carriers (+/cnm) and CNM affected (cnm/cnm) Labradors tested for medical or breeding purposes.
The total number of dogs for each period is indicated above histograms. Additional dog samples from Australia (n = 15), New-Zealand (n = 2), Puerto Rico (n = 1) and Argentina (n = 1) were tested; they were all homozygous for the wild-type allele (+/+).
Table 3.
Numbers by genotype of Labradors tested for medical or breeding purposes.
Figure 4.
A 3.8-Mb haplotype is highly associated with CNM.
The acrocentric region of the PTPLA locus within canine chromosome 2 (CFA2) is depicted. Positions of genotyped SNPs are indicated. The short and long haplotypes associated with CNM are shown in green and red, respectively. For each SNP, the allele detected in the CNM associated haplotype is indicated and represented as a grey box. The alternative allele is represented as a white box. For each SNP, the minor allele frequency (MAF) in the healthy population of Labradors is given. The PTPLAcnm allele is represented by a black dot (•) and the wild-type PTPLA+ allele by a “+”. Frequencies of long 3.8-Mb haplotypes in each population of CNM or healthy dogs are given below each haplotype. For haplotypes with frequencies >10%, width of haplotypes is proportional to its frequency. Haplotypes with frequencies below 3% have been omitted and are detailed in Figure S3.
Figure 5.
Hierarchical clustering from the 81 dogs at k = 8.
The analysis was based on genotypes obtained for ten loci (SNP 20687 to SNP 24518). Hubert Gamma values are indicated for k≥2 on the top left panel. The scale on the right axis represents the genetics distances calculated by PLINK software. In the dendogram, each vertical line represents a dog and colors reflect the eight clusters obtained by the analysis. Grey dash lines indicate common ancestors inferred from the analysis. Below the dendogram, dogs are named by their unique identifier. The “CNM_” prefix was added to the name of affected Labradors.