Functional biology of the Steel syndrome founder allele and evidence for clan genomics derivation of COL27A1 pathogenic alleles worldwide

Previously we reported the identification of a homozygous COL27A1 (c.2089G>C; p.Gly697Arg) missense variant and proposed it as a founder allele in Puerto Rico segregating with Steel syndrome (STLS, MIM #615155); a rare osteochondrodysplasia characterized by short stature, congenital bilateral hip dysplasia, carpal coalitions, and scoliosis. We now report segregation of this variant in five probands from the initial clinical report defining the syndrome and an additional family of Puerto Rican descent with multiple affected adult individuals. We modeled the orthologous variant in murine Col27a1 and found it recapitulates some of the major Steel syndrome associated skeletal features including reduced body length, scoliosis, and a more rounded skull shape. Characterization of the in vivo murine model shows abnormal collagen deposition in the extracellular matrix and disorganization of the proliferative zone of the growth plate. We report additional COL27A1 pathogenic variant alleles identified in unrelated consanguineous Turkish kindreds suggesting Clan Genomics and identity-by-descent homozygosity contributing to disease in this population. The hypothesis that carrier states for this autosomal recessive osteochondrodysplasia may contribute to common complex traits is further explored in a large clinical population cohort. Our findings augment our understanding of COL27A1 biology and its role in skeletal development; and expand the functional allelic architecture in this gene underlying both rare and common disease phenotypes.

X-rays of lateral spine and forearms at 5 months of age revealed tall vertebral bodies, long vertebral pedicles and possibly dislocated radial heads. Robinow syndrome and campomelic dysplasia were the initial diagnoses considered based on the physical examination. However, SOX9 analysis and exome sequencing showed no rare variants in the known genes associated with these disorders. WES analysis of the proband identified a homozygous variant in exon 19 of COL27A1 [hg19:g.chr9:116999951(G>A); c.2683G>A; p.Gly895Arg].

Targeting vector and allele modification to generate Col27a1 (p.Gly682Arg) knock-in mutant
mice. To model the human p.Gly697Arg variant, a guanine-to-cytidine mutation, resulting in a p.Gly682Arg substitution (orthologous to the human variant), was introduced in exon 7 of 61 (exonID: ENSMUSE00001215121 [build GRCm38]) of the mouse gene. Briefly, a DNA construct (GenScript) containing Col27a1 exon 7 and surrounding intronic sequences was synthesized to include the desired G>C mutation, as well as two small intronic deletions (away from known regulatory sites) to accommodate loss-of-allele (LOA) TaqMan assays TU and TD.  (with 2mM glutamine, 5% FBS) at 5% CO 2 and 37 °C, respectively. W20 cells were plated at 40000cells/cm2 on glass bottom, poly-l-lysine coated Ibidi chamber slides (Cat # 80427). Cells were transfected with ER-E2-Crimson (pEF.myc.ER-E2-Crimson, Addgene # 38770) along with pUC19, pCMV6-AC-mGFP-COL27A1-WT, pCMV6-AC-mGFP-COL27A1-G697R or pCMV6-AC-mGFP-COL27A1-G895R using X-tremeGENE HP DNA transfection agent (Roche #06 366 244 001) at 1:4 (DNA:transfection agent) ratio. Media was replaced a day after transfection, and the following day 50 µg/ml of L-ascorbic acid was added to the media. A day later cells were fixed with freshly prepared 4% paraformaldehyde in DPBS with calcium and magnesium for 15 minutes at room temperature, cells were either imaged directly or processed for immunostaining using anti-GFP antibody (CST #2956, 1:100 dilution) as per the manufacturers recommendation. Images were acquired on a Zeiss LSM 880 laser scanning microscope. Similarly, ATDC5 cells were plated at 40000cells/cm2 on poly-l-lysine coated, glass bottom Ibidi slides and co-transduced using adenoviral particles encoding ER-RFP [multiplicity of infection (MOI) 20] along with COL27A1-WT, COL27A1-G697R or COL27A1-G895R at MOI ranging from 10-100 per cell. The cells were then fixed with 4% paraformaldehyde in DPBS with calcium and magnesium for 10 minutes at room temperature and then imaged using a Zeiss LSM 880 microscope.
Cell transduction and histochemical analyses. ATDC5 cells were plated at 4000 cells/well in a 12well plate and transduced the following day with adenoviral particles encoding COL27A1-WT, COL27A1-G697R or COL27A1-G895R. The next day, they were changed into culture medium containing ITS supplement (Gibco #41400045) to induce chondrogenic differentiation as per Shukunami et al. On day 30 of chondrogenic induction, the cells were fixed in 95% methanol at room temperature for 20 minutes, stained with 1% Alcian blue 8GX in 0.1M HCl overnight, and imaged. The Alcian blue was then extracted by incubation with 6M guanidine HCl for 6 hours at 6 room temperature, and quantified through absorbance measurement of the resulting solutions at 630nm, using a SpectraMax M4 microplate reader.
In vitro cell culture of mouse primary chondrocytes. Primary chondrocytes were collected from mouse neonates (P1) after euthanasia following approved IACUC guidelines and according to established protocols (Gosset et al, 2008). Epiphyseal cartilage from femur, tibia, and humerus was collected using a dissecting microscope. Cartilage was further dissected into 2-3 pieces per sample and digested in 5ml of 0.25% trypsin in HBSS for 1hr at 37 °C. Supernatant was discarded from sedimented samples, remaining cartilage pellets were further digested overnight at 37 °C using 10ml of 0.75mg/ml of collagenase I (Worthington Biochemical Corporation, CLS1). Upon overnight digestion, chondrocytes were dissociated from the cartilage pieces by repeated pipetting (20 times) using a glass Pasteur pipette. Dissociated cells were passed through a cell strainer (80-100µm). Cells were subsequently washed twice with 10ml growth media (10% MEM). Cells were seeded at 150000/well in a collagen-coated 24-well plate. Upon confluency, 50 µg/ml of L-ascorbic acid was added to the growth media. After two-weeks in culture, cells were fixed in 10% neutral buffered formalin for 15 minutes at room temperature. After multiple washes with double distilled water, cells were treated with 70% ethanol for 5 minutes at roomtemperature. Subsequently, cells were washed with double distilled water and incubated with 5% acetic acid (pH: 1.0) for 5 minutes at room temperature. Immediately after, cells were incubated with 1% Alcian blue 8GX (SIGMA A3157) for 2 hours at room temperature. Cells were de-stained in water (5 washes, 10min/wash), followed by an overnight wash at 4 °C. Cells were imaged using an EVOS digital microscope.

Supplementary Figures
Supplementary Figure 1