XBP1-Independent UPR Pathways Suppress C/EBP-β Mediated Chondrocyte Differentiation in ER-Stress Related Skeletal Disease

Abstract

Schmid metaphyseal chondrodysplasia (MCDS) involves dwarfism and growth plate cartilage hypertrophic zone expansion resulting from dominant mutations in the hypertrophic zone collagen, Col10a1. Mouse models phenocopying MCDS through the expression of an exogenous misfolding protein in the endoplasmic reticulum (ER) in hypertrophic chondrocytes have demonstrated the central importance of ER stress in the pathology of MCDS. The resultant unfolded protein response (UPR) in affected chondrocytes involved activation of canonical ER stress sensors, IRE1, ATF6, and PERK with the downstream effect of disrupted chondrocyte differentiation. Here, we investigated the role of the highly conserved IRE1/XBP1 pathway in the pathology of MCDS. Mice with a MCDS collagen X p.N617K knock-in mutation (ColXN617K) were crossed with mice in which Xbp1 was inactivated specifically in cartilage (Xbp1CartΔEx2), generating the compound mutant, C/X. The severity of dwarfism and hypertrophic zone expansion in C/X did not differ significantly from ColXN617K, revealing surprising redundancy for the IRE1/XBP1 UPR pathway in the pathology of MCDS. Transcriptomic analyses of hypertrophic zone cartilage identified differentially expressed gene cohorts in MCDS that are pathologically relevant (XBP1-independent) or pathologically redundant (XBP1-dependent). XBP1-independent gene expression changes included large-scale transcriptional attenuation of genes encoding secreted proteins and disrupted differentiation from proliferative to hypertrophic chondrocytes. Moreover, these changes were consistent with disruption of C/EBP-β, a master regulator of chondrocyte differentiation, by CHOP, a transcription factor downstream of PERK that inhibits C/EBP proteins, and down-regulation of C/EBP-β transcriptional co-factors, GADD45-β and RUNX2. Thus we propose that the pathology of MCDS is underpinned by XBP1 independent UPR-induced dysregulation of C/EBP-β-mediated chondrocyte differentiation. Our data suggest that modulation of C/EBP-β activity in MCDS chondrocytes may offer therapeutic opportunities.

Fig 1. Genetic and morphometric characterization of C/X mice.

(A) RT-PCR on cDNA derived from femoral epiphyseal cartilage from wildtype (Wt) and C/X to detect the full-length form of Xbp1 (Xbp1 _FL) or the inactive form of Xbp1, lacking exon 2 (Xbp1 _ΔEx2), and sequencing of cDNA from C/X femoral head cartilage to assay for the deletion of Xbp1 exon 2. (B) PCR for residual loxP site downstream of the p.Asn617Lys Col10a1 coding sequence using genomic DNA derived from Wt and C/X. (C) Alizarin red S/Alcian blue staining of skeletal preparations from newborn, 1 week, and 2 week wildtype (Wt), Xbp1 ^CartΔEx2, ColX ^N617K and C/X mice. (D-F) Quantification of (D) femoral and (E) tibial length, and (F) intercanthal distance (ICD) from legs from 2 week Wt and mutant mice–Wt, N = 8; Xbp1 ^CartΔEx2, N = 8, ColX ^N617K, N = 6; C/X, N = 8; statistical analysis performed using Student’s t test.

Fig 2. Ablation of XBP1 does not significantly affect the MCDS phenotype in C/X mice.

(A-C) Tibial epiphyseal cryosections from 2 week Wt, Xbp1 ^CartΔEx2, ColX ^N617K and C/X mice stained with (A) haematoxylin and eosin (H&E), or by immunofluorescence using (B) anti-collagen II or (C) anti-collagen X antibodies; B—Bone; HZ—Hypertrophic Zone; PZ—Proliferative Zone; SCO—Secondary Center of Ossification. (D-F) Quantification of growth plate (D) resting zone, (E) proliferative zone, and (F) hypertrophic zone lengths in mutant and Wt mice; N = 3 for each genotype; statistical analysis performed using Student’s t test.

Fig 3. Apoptosis is elevated in 2 week ColX ^N617K and C/X growth plate cartilage.

(A) Representative 2 week wildtype (Wt), Xbp1 ^CartΔEx2, ColX ^N617K and C/X tibial growth plate sagittal cryosections analysed by TUNEL with DAPI counterstaining; HZ—hypertrophic zone. Boxes inset indicate magnified areas of the hypertrophic zones containing TUNEL-positive chondrocytes. (B) TUNEL analysis of at least 6 tibial growth plate sections from each of 3 Wt, Xbp1 ^CartΔEx2, ColX ^N617K, and C/X mice, expressed as the number of TUNEL-positive chondrocytes in the hypertrophic zone as a percentage of the total number of chondrocytes per zone (as defined by DAPI-stained nuclei), and showing standard deviation around the mean. (C,D) Representative 2 week (C) ColX ^N617K and (D) C/X tibial growth plate cryosections, showing the distribution of TUNEL-positive cells along the antero-posterior axis of ColX ^N617K and C/X hypertrophic zones, as demarcated by 10 consecutive columns (1–10) of arbitrary width. Plots depict the number of TUNEL-positive chondrocytes in each column as a percentage of the total number of chondrocytes per column (as defined by DAPI-stained nuclei), from the same ColX ^N617K and C/X mice as analysed in (B), and showing standard deviation around the mean. Statistical analysis performed using Student’s t-test, ** p < 0.01, *** p < 0.001.

Fig 4. Quantitative PCR of mutant and wildtype hypertrophic zones.

qPCR with primers specific for (A) Agc1, (B) Ctgf, (C) Matn1, (D) Creld2, (E) Derl3, (F) Ero1l, (G) Fgf21, (H) Steap1, and (I) p58IPK on cDNA derived from Wt, Xbp1 ^CartΔEx2, ColX ^N617K and C/X hypertrophic zone aRNA. Plots depict mean fold differences with standard deviation from the mean, N = 3, statistical significance was determined using Student’s t test, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Fig 5. Microarray analysis of mutant and wildtype hypertrophic zones.

(A) Venn diagram depicting the relationship between probes indicating differential gene expression (fold difference ≥ 2.0, adjusted p value ≤ 0.01) following comparisons of C/X versus wildtype (Wt) (blue), Xbp1 ^CartΔEx2 versus Wt (yellow), and ColX ^N617K versus Wt (red), by whole genome microarray analysis of hypertrophic zone aRNA. (B-D) Ontological analysis of (B) all probes in cohort i in (A), or those showing (C) up-regulation or (D) down-regulation, by Functional Annotation Clustering, using DAVID v6.7 software, and depicting representative gene ontology terms from each annotation cluster achieving an enrichment score (ES) ≥ 1.3.

Fig 6. Expression of wildtype growth plate zone gene signatures in ColX ^N617K, Xbp1 ^CartΔEx2, and C/X hypertrophic zones.

Heatmaps depicting the relative fold difference (log fold change) of microarray probes representing (A) 773 wildtype (Wt) proliferative zone signature genes and (B) 510 Wt hypertrophic zone signature genes following the comparison of C/X versus Xbp1 ^CartΔEx2, ColX ^N617K versus Wt, and Xbp1 ^CartΔEx2 versus Wt hypertrophic zones; N = 3. For both heatmaps, each Wt growth plate zone signature gene is represented by a single bar, colour-coded according to relative expression as indicated, with up-regulated probes coloured yellow, and down-regulated probes coloured red.

Fig 7. Dysregulated expression of genes involved in ER stress and chondrocyte differentiation.

(A) Immunofluorescent analysis for ATF4 in tibial epiphyseal cryosections from 2 week wildtype (Wt), Xbp1 ^CartΔEx2, ColX ^N617K and C/X mice; B—Bone; HZ—Hypertrophic Zone; PZ—Proliferative Zone. (B-H) qPCR with primers specific for (B) Chop, (C) Cebpb, (D) p57 ^Kip2, (E) Gadd45b, (F) Runx2, (G) Col10a1, and (H) Mmp13 on cDNA derived from Wt, Xbp1 ^CartΔEx2, ColX ^N617K and C/X hypertrophic zone aRNA. Plots depict mean fold differences with standard deviation from the mean; N = 3; statistical analysis performed using Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. (I) Schematic diagram of proposed model to explain the molecular pathology of MCDS. Blue boxes depict genes. Red boxes depict biological processes. Green arrows depict activation or up-regulation. Red arrows depict inactivation or down-regulation. Green lines depict increased interaction between proteins. Red lines depict decreased interaction between proteins.