Fig. 4: Morphological characterization of the GDF5 R4 enhancer and rs6060369 variant in patients and the mouse model. | Nature Communications

Fig. 4: Morphological characterization of the GDF5 R4 enhancer and rs6060369 variant in patients and the mouse model.

From: Joint disease-specificity at the regulatory base-pair level

Fig. 4

a Occurrence of ‘T’ risk allele frequency at rs6060369 in subsampled OAI patients stratified by OA progression (see text for p-values). Points represent individual sub-samples (n = 200) of individuals taken from patient groups presenting with no/moderate-OA (KL = 0/1) (n = 1207), progressing to significant OA (n = 208), and presenting with significant OA (KL ≥ 2) (n = 1119). A two-sided Student’s T-test was used to compare risk-allele occurrence in subsamples of each group, with FDR correction for n = 3 tests. Central bar represents mean of distribution, lower/upper bars indicate first and third quartiles, respectively. b Significantly different anatomical features of the knee in OAI patients (WT n = 76, HET n = 161, HOMO n = 146). ANOVA with Dunnet post-hoc was used for pairwise comparisons to wild type. All p-values are two-sided and indicate significant differences compared to individual’s wild type for the non-risk allele. Bars indicate medians. c µCT measurements of significantly different anatomical features in rs6060369 single base-pair replacement mice at 1-year (WT n = 6, HET n = 15, HOMO n = 14). ANOVA with Dunnet post-hoc was used for pairwise comparisons to wild type. All p-values are two-sided and indicate significant differences compared to wild type. Bars indicate medians. d Histological analysis of rs6060369 single base-pair replacement mice at 1 year, showing coronal sections at mid-knee level. The sections represent knees with mild, moderate, and severe knee OA. Sections are color coded to their corresponding overall score listed in the graph in (e). The scale bars are 250 µm. e OARSI scores on wild type and homozygous rs6060369 single base-pair “T” replacement knees at 1 year (WT n = 5, HOMO n = 10). Mann–Whitney test with two-sided p values were used for comparison. f 3D comparative analysis indicating the anatomical locations of largest morphological differences between wild type R4+/+ and homozygous R4-/- hind limbs (top) as well as between wild type R4rs6060369-A/rs6060369-A and homozygous R4rs6060369-T/rs6060369-T hind limbs (bottom) at 1-year. All p-values are two-sided. See Supplementary Figs. 89, Supplementary Tables 1015 and the Supplementary Information for related analyses.

Back to article page