Hanoverian F/W-line contributes to segregation of Warmblood fragile foal syndrome type 1 variant PLOD1:c .2032G>A in Warmblood horses

Background: Warmblood fragile foal syndrome (WFFS) is a lethal condition detected in Warmblood horses. Its origin and association with performance traits and fertility among horse populations is unknown. Objectives: To validate the previously identified WFFS type 1 (WFFST1)-associated missense variant PLOD1: c.2032G>A and to investigate its distribution among various horses with particular focus on Hanoverian breed, as well as its pathomorphological picture. The study aimed at identifying the origin of the mutant allele and its correlation with performance and fertility traits in Warmblood horses. Study design: affected by the WFFST1-missense mutation is specifically modify-ing the architecture of collagen fibrils leading to their degradation. 38 PLOD1 was shown to be an important enzyme hydroxylating lysyl resi-dues in collagen sequences, which are essential for attachments of car-bohydrate units. 38 This is in turn important for cross-linking of collagen fibrils, supporting their mechanical stability. Investigations of different connective tissues from human Ehlers-Danlos Syndrome Type VI pa-tients revealed different expression levels of PLOD1 in collagen types I and III. 39,40 These findings suggest that the signs of the disease can vary among different tissues, localisations at the body or even among different individuals dependent of the expression pattern.


| INTRODUC TI ON
Collagen fibrils in connective tissues provide structure and strength to the skin and thus are essential for the organisation of the entire dermis. 1 Disorders disturbing this type of tissue are known as Ehlers-Danlos syndromes (EDS) in human subjects 2 and are designated as EDS-like diseases in many other species like cattle, 3 sheep, 4 rabbits, 5 minks, 6 cats, 7 dogs 6,8 as well as horses. 9 In horses, several case reports on inherited connective tissue disorders exist. They report on dermatosparaxia on an EDS-like disease in Oldenburg foals, 10,11 in a Coldblood and Warmblood horse, 12,13 dermal asthenia in Warmblood horses, [14][15][16] Hereditary Equine Regional Dermal Asthenia (HERDA) in Quarter horses [17][18][19][20] and Warmblood fragile foal syndrome (WFFS) in a Westphalian. 10 In two of these EDS-like forms, HERDA and WFFS, genetic investigations supported a monogenic autosomal mode of inheritance. 21,22 Analyses of HERDA-affected Quarter horses revealed a missense mutation in peptidylprolyl isomerase B gene to be potentially causative for this disorder mostly found in older horses. 21 In contrast, WFFS was found to be a characteristically early onset form of inherited connective tissue disorders. 22 For WFFS type 1 (WFFST1) a potential causative variant was identified in the gene procollagen-lysine-2-oxoglutarate-5-dioxygenase 1 (PLOD1, PLOD1:c.2032G>A, NC_009145.3:g.39927817C>T). 22 Clinical signs of WFFST1 were shown to include fragile skin, skin and mucosa lacerations, hyperextension of the articulations, subcutaneous emphysema, oedema and haematomas. 10 The predominant manifestation of WFFST1 was reported to be lethal during late gestation or live birth of nonviable foals. 23 Initial screenings in pedigree data suggested PLOD1:c.2032G>A segregating among Hanoverian, Selle Francais, KWPN, Oldenburg and Westphalian. 22 Subsequent genetic testing proposed an allele frequency of 5.5% in Warmblood horses in Brazil and 1.2% in Thoroughbred. 24,25 Due to its clinical importance, it was postulated that further investigations on the prevalence of the mutation and on the correlation between genotype and phenotype are required. 10 Thus, the objective of this study was to validate WFFST1 variant, to investigate the prevalence among the Hanoverian and other horse breeds and its association with performance and fertility traits. We aimed to trace the WFFST1-associated allele to the most recent common ancestor (MRCA) and thus the potential origin of WFFST1.
Furthermore, the histopathological and electron microscopical presentation of this inherited connective disorder was examined in an affected foal.

| Validation of WFFST1 variant
For evaluation of variants within the region of PLOD1, whole genome sequencing (WGS) data of 77 horses of 35 breeds/populations and one donkey were available. All data are accessible at Sequence Read Archive (NCBI). Sequence IDs can be found in Table S1. Fastqfiles were mapped and underwent variant calling as previously de- In addition, all variant effects were estimated using SNPEff predictions. 27 Missense variants were further investigated using SIFT scores. 28

| Genotyping of WFFST1
In total, 1166 EDTA-blood or hair samples from 1059 horses, 76 ponies and 4 donkeys were obtained. DNA was isolated using standard saline extraction 29 and genotyped for the WFFST1 variant (NC_009145.3:g.39927817C>T) using a Competitive Allele Specific PCR (KASP) assay (LGC Genomics) run on an ABI7300 real-time system (Applied Biosystems) (Table S2).

| WFFS case
A stillborn filly, registered as Oldenburg but a descendant of two

Conclusions:
This study provides a comprehensive evaluation of WFFST1 variant and traces it back to its potential origin.

K E Y W O R D S
horse, skin, fragile foal syndrome, PLOD1, Hanoverian  32 In addition, EBVs were corrected for systematic environmental effects including age of the stallion, breeding season, period within breeding season and insemination centre. 31 A successful artificial insemination in an oestrus cycle was encoded 1, otherwise the trait value was 0. All EBVs were standardised onto a mean of 100 and a standard deviation of 20. EBVs > 100 mean higher conception rates than the population average. 31 All 195 individuals were also genotyped for the WFFST1 variant. The analysis was done to evaluate genotype effects on EBVs of Hanoverian stallions for performance and fertility traits using SAS. GLM for LS (least square) means was computed using the genotype as an independent and performance and fertility traits as dependent variates.

| Evaluation of WGS and genotyping data
In order to validate WFFST1 variant and exclude further potential candidate variants for WFFS in the region of PLOD1, a screen-

| Macroscopic, histopathological and ultrastructural findings
The stillborn filly, aborted 2 weeks before the expected date of birth,

| Association of WFFS and performance
The genotype effects on EBVs were significant for a range of traits recorded at SBI including outer appearance complying with breed characteristics (breed type) and characteristics of stallion or mare (sex type), quality of body conformation, correctness of gait, elasticity (trot), head, neck, frame, saddle area, front and hindlimbs. horse points and dressage were also significant (Table S5). No significant differentiation was observed with the inbreeding coefficient and EBVs for the embryonic and stallion effect.

| D ISCUSS I ON
In the current study, comprehensive analysis for WFFS validated the WFFST1 variant and gave a deeper view into its clinical and pathomorphological picture, genotypic distribution and origin. Furthermore, we also identified a homozygous mutant genotype in an affected foal with EDS-like signs. Histopathological investigations of the skin confirmed previous findings in an affected foal of a markedly thinned dermis and irregular collagen bundles. 10 Ultrastructural findings resembled those that have been reported elsewhere in horses, dogs, rabbits, human subjects and mice. 10,11,[33][34][35][36][37] However, besides a more severe degradation of collagen fibrils as observed in the previously investigated WFFST1-case, 10 we encountered new findings using ultrastructural examinations, which highlighted fibrillar plaques and collagen fibres embedded in an amorphous matrix. Similar observations were made in an Oldenburg foal with EDS-like signs but no genetically confirmed disease. 11 The irregular shaped electron dense material was found to represent an increased amount of collagen type I. 11 Thus, the ultrastructural findings indicate that PLOD1 protein In contrast to the severely affected horses harbouring a homozygous mutant genotype, genetic carriers of the recessively inherited disease can pass the mutant allele unnoticed to their progeny. In respect to WFFST1 mutant allele, the distribution was found to be wide in diverse Warmblood horse populations but also Thoroughbred from various countries like Germany, Brazil and USA. [23][24][25] We suppose that this wide distribution can be explained To conclude, our study confirmed WFFST1 variant in WGS data as well as in an affected foal with EDS-like signs due to severe degradation of collagen fibrils. A Hanoverian stallion, Stallion A, was shown to be the major contributor to the wide distribution of the mutant allele in the Warmblood population.

ACK N OWLED G EM ENTS
We gratefully acknowledge support from the North-German Supercomputing Alliance (HLRN) for HPC-resources that contributed to the research results. The authors would like to thank H. Rohn for their technical assistance.

CO N FLI C T O F I NTE R E S T S
No competing interests have been declared. All authors read and approved the final manuscript.

E TH I C A L A N I M A L R E S E A RCH
EDTA-blood or hair samples were obtained according to the approval of the state veterinary office Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit in Oldenburg (no. 02A/138).

OWN E R I N FO R M E D CO N S E NT
Written informed approval was acquired from the horse owners and breeders to collect samples for current research.

DATA ACCE SS I B I LIT Y S TATE M E NT
All data are accessible at Sequence Read Archive (NCBI, see summary of all IDs in Supplementary Item 1).