Victoria P. Spreyer1; Erica K. Creighton1; Barbara Gandolfi1; Richard Malik2; Leslie A. Lyons1
Ehlers-Danlos syndrome (EDS) is a heterogeneous group of inherited connective tissue disorders characterized by altered mechanical properties of skin, joints, blood vessels, and ligaments. Ehlers-Danlos syndrome is the most common inherited connective-tissue disorder and has six main types: classic type (EDS I and II), hypermobility type (EDS III), vascular type (EDS IV), kyphoscoliosis type (EDS VI), arthrochalasia type (EDS VIIA and VIIB), and dermatosparaxis type (EDS VIIC). The majority of dermatosparaxis cases in humans are caused by variants in the gene ADAMTS2. These variants result in a defect in the processing of type I procollagen to collagen where molecules retain the amino-terminal propeptide. They then accumulate in tissues and self-assemble into ribbon-like fibrils that fail to provide normal tensile strength.
In the Hereford breed, onset is at approximately 2 months of age, with severe skin fragility leading to large open lesions exposing the dermis on the head and body, as well as joint laxity. Interestingly, while the cattle model mutations in ADAMTS2 cause the N-terminus of procollagen to be barely processed in skin, the procollagen in bone and tendon is more extensively cleaved. Since the mutation should prevent the synthesis of the active enzyme, this suggests another enzyme can remove the amino-terminus of type I collagen. Several cases of bovine dermatosparaxis are analogous to the human syndrome and were found to be caused by mutations in the procollagen I N-proteinase (pnPI) or ADAMTS2 gene, though mutations in other sites are likely responsible for other types of dermatosparaxis. In all species, the genetic cause of dermatosparaxis is monogenic. Recently, three human EDS VIIC cases were reported with normal collagen pattern and no mutation was detected in the ADAMTS2 gene. This suggests that mutations in other genes can alter the processing of procollagen molecules and cause EDS VIIC. Ehlers-Danlos syndrome in these cases may be due to a mutation in another protein that would be required for optimal ADAMTS-2 function. The complex list of proteins that play a central role in the pathway is not yet known. ADAMTS-2 interacts with other proteins; in fact, it is found in a complex with other proteins when precipitated and isolated. Further genetic studies in humans or unexplored model animals will help to better understand the collagen pathway and find novel genes involved in this process.
In cats, EDS is considered a heritable disorder affecting Himalayan and Burmese breeds and most closely resembles dermatosparaxis. Burmese cat EDS is distinct from other cat breeds and has atypical presentation, such as lesions consisting of necrotic eschars or atrophic alopecia that develop in the absence of skin lacerations (Figure 1).
Figure 1. Unaffected and affected Burmese cats
a: Unaffected sable Burmese cat
b: Affected Burmese cat displaying large, ischemic skin lesions on the base of the head. Such lesions occur often during induction of anesthesia due to scruffing the cat.
c: Affected Burmese cat displaying hyperextensible skin and small areas of alopecia
The age of presentation seems to vary in cats from a few months to two years. Affected cats are reported to have "bleeding" skin wounds and skin that is easily injured. The frequency and severity of skin lesions and their tendency to recur in response to minor trauma (especially scratching) seem to be the critical reason for euthanasia in many of these cases. To date, the underlying genetic variant(s) causing Burmese cat EDS is unknown. Possible candidate genes and regions have been described clinically and genetically in cattle, humans, and sheep, with causative mutations found in the genes ADAMTS2, PNPI, EPYC, COL5A1, COL3A1, COL1A1, and COL1A2. These mutations cause alterations in either the structure or the post-translational processing of collagen or other proteins that interact with collagen.
Two paired-end PCR-free libraries with insertion sizes of 350 bp and 550 bp were constructed per cat. Then, 40x coverage of Illumina HiSeq 2500 was generated with 100-bp paired-end reads. Each lane on the HiSeq instrument yields ∼100 Gb of sequence, corresponding to an average of 40x coverage of the whole cat genome (~ 4 Gb). Data were transferred to Maverix Biomics, and each sample was aligned to the reference genome cat (Ver6.2 assembly). Differences or variants between the reference and the affected cats were called using the program Platypus. Reads and variant calls were uploaded to a UCSC-type genome browser. Lists of the variants were generated which will include the following information: variant position in the genome, impact of the change to gene, base change, and amino acid change.
The 14,014 exonic variants found in the affected EDS Burmese cats were filtered based on expected phenotypes using the Maverix-provided variant tool which can sort variants by minor allele frequency (MAF) in each cat. 9,702 variants were homozygous in the affected Burmese cats. A mutation was considered a candidate for causation if found to be homozygous (MAF 100%) for the alternative allele (compared to the reference) in the two affected cats and absent or heterozygous in the healthy control cats. Twelve exonic variants in 11 different genes segregated concordantly with the disease phenotype. Of the 12 variants, a variant in a gene involved in collagen synthesis was predicted to produce a frameshift mutation, resulting in a severely truncated protein. The variant was concordant in 38 additional Burmese cats. Further population screenings will be performed to confirm this variant as a causal for Burmese EDS. Understanding the underlying genetics of Burmese EDS offers a novel animal model, and a genetic test can be developed to achieve better general health of the breed.