A 3.7 Mb deletion encompassing ZEB2 causes a novel polled and multisystemic syndrome in the progeny of a somatic mosaic bull

Abstract

Polled and Multisystemic Syndrome (PMS) is a novel developmental disorder occurring in the progeny of a single bull. Its clinical spectrum includes polledness (complete agenesis of horns), facial dysmorphism, growth delay, chronic diarrhea, premature ovarian failure, and variable neurological and cardiac anomalies. PMS is also characterized by a deviation of the sex-ratio, suggesting male lethality during pregnancy. Using Mendelian error mapping and whole-genome sequencing, we identified a 3.7 Mb deletion on the paternal bovine chromosome 2 encompassing ARHGAP15, GTDC1 and ZEB2 genes. We then produced control and affected 90-day old fetuses to characterize this syndrome by histological and expression analyses. Compared to wild type individuals, affected animals showed a decreased expression of the three deleted genes. Based on a comparison with human Mowat-Wilson syndrome, we suggest that deletion of ZEB2, is responsible for most of the effects of the mutation. Finally sperm-FISH, embryo genotyping and analysis of reproduction records confirmed somatic mosaicism in the founder bull and male-specific lethality during the first third of gestation. In conclusion, we identified a novel locus involved in bovid horn ontogenesis and suggest that epithelial-to-mesenchymal transition plays a critical role in horn bud differentiation. We also provide new insights into the pathogenicity of ZEB2 loss of heterozygosity in bovine and humans and describe the first case of male-specific lethality associated with an autosomal locus in a non-murine mammalian species. This result sets PMS as a unique model to study sex-specific gene expression/regulation.

Figure 1. Clinical features of Polled and Multisystemic Syndrome.

(A) Two-and-half-year old wild-type heifer that was mechanically dehorned when approximately one-year old. (B) Two-and-half-year old affected heifer. Note the slender build, the shaggy hair coat demonstrating the bad health condition, and the hypotonia of hind limbs. (C) Upper part of the skull of the same affected heifer. Note the absence of corneous growth, the ridge-shaped extra bone deposition along the frontal suture and the narrowness of the muzzle insertion. (D) On-farm performance testing statistics of affected (PMS) and wild-type half-sibs. Values expressed as: means ± standard deviation (number of observations). *p<0.05, **p<0.01 and ***p<0.001 versus wild-type half-sisters (Welch’s t-test). Weaning corresponds to 210 days of age. (E and F) Ovaries of the affected (+/−) heifer displayed in (B) and (C). (I) Ovary of a wild-type (+/+) matched control. (G and H, and J and K) Histological analyses of the ovaries displayed in (E) and (I) respectively. (H) and (K) are higher magnifications (X5.5) of (G) and (J). Note the numerous large lacunae surrounded by connective tissue and the absence of follicles in the ovary from the affected heifer. Follicles are surrounded with a green dotted line in the photography of the wild-type ovary. Scale bars represent 1 cm in (F), (E) and (I); 500 µm in (G) and (J); and 50 µm in (H) and (K).

Figure 2. Mapping and characterization of the causative mutation for PMS syndrome.

(A) and (B) Results of Mendelian error mapping using the Illumina 50 K and 777 K SNP beadchips, respectively. Markers displaying Mendelian errors between at least one PMS heifer and her sire are represented in purple whereas markers for which at least one of the three PMS animals is heterozygous are represented in blue. Other markers are not represented. (C) Plot of the whole-genome sequencing read depth coverage on the same region. *: artifact due to a local error in genome assembly. (D) Gene content of the region. (E) FISH-mapping with BAC clones located in the deleted region (labeled in red) and in the juxtacentromeric region of BTA2 (labeled in green) on fibroblasts of a PMS animal. (F) Magnification of (E) showing normal (above) and deleted (below) BTA2 chromosomes. (G) Genotyping of PMS using a three-primer PCR system (see methods). Neg.: negative control.

Figure 3. Real-time PCR expression analyzes of the deleted genes in different affected tissues at 90 dpc.

ZEB2, GTDC1, ARHGAP15 (exon13), and KYNU (absent from the deleted fragment) expression is measured in various tissues from wild type (+/+; black histograms) or mutant (+/−; white histograms) fetuses.

Figure 4. Histological analyses of wild-type (+/+) and PMS (+/−) horn bud and forehead skin.

(A) and (B) Head of PMS (+/−) and wt (+/+) fetuses. Horn bud of wt fetus is indicated by an arrow. (C) and (D) Histological sections of the “horn bud” of PMS and wt fetuses respectively. (E) Magnification (X10) of (C) showing one hair follicle primordium. (F) and (G) Magnifications (X10 and X3 respectively) of (D) showing respectively keratinizing epidermal cells and clusters of dermal cells displaying glandular/ductal differentiation. (H) and (I) Histological sections of the forehead skin of PMS and wt fetuses respectively. (J) and (K) Magnifications (X10) of (H) and (I) showing hair follicles primordial; note the slight difference between PMS and wt genotypes; statistical analysis also showed a significant difference in epidermis thickness: 15.8±3.0 µm in PMS vs 22.1±3.7 µm in wt; p-value = 2.3e-28 (Welch’s t-test). Scale bars in (C), (D), (H) and (I) represent 1 mm whereas scale bars in (E), (F), (G), (J) and (K) represent 100 µm.

Figure 5. Distribution of PMS at different developmental stages in V.’s progeny.

Histograms represent percentages relative to the largest category at each developmental stage. The numbers are indicated on each histogram. (1) and (2) results of the first and second sperm-FISH experiments (see methods). Blastocysts correspond to 7 dpc embryos.