Collagen X Is Dispensable for Hypertrophic Differentiation and Endochondral Ossification of Human iPSC-Derived Chondrocytes

Abstract

Collagen X is a non-fibril collagen produced by hypertrophic chondrocytes and was believed to associate with the calcification process of growth plate cartilage. The homozygous loss of Col10a1 gene in mice, however, demonstrated no remarkable effects on growth plate formation or skeletal development. To investigate the role of collagen X in human chondrocytes, we established human induced pluripotent stem cells (hiPSCs) with heterozygous (COL10A1 ^+/-) or homozygous (COL10A1 ^-/-) deletions of COL10A1 gene using the dual sgRNA CRISPR/Cas9 system. Several mutant clones were established and differentiated into hypertrophic chondrocytes by a previously reported 3D induction method. No remarkable differences were observed during the differentiation process between parental and mutant cell lines, which differentiated into cells with features of hypertrophic chondrocytes, indicating that collagen X is dispensable for the hypertrophic differentiation of human chondrocytes in vitro. To investigate the effects of collagen X deficiency in vivo, chondrocyte pellets at the proliferating or prehypertrophic stage were transplanted into immunodeficient mice. Proliferating pellet-derived tissues demonstrated the zonal distribution of chondrocytes with the transition to bone tissues mimicking growth plates, and the proportion of bone tended to be larger in COL10A1 ^-/- tissues. Prehypertrophic pellet-derived tissues produced trabecular bone structures with features of endochondral ossification, and there was no clear difference between parental- and mutant-derived tissues. A transcriptome analysis of chondrocyte pellets at the hypertrophic phase showed a lower expression of proliferating-phase genes and a higher expression of calcification-phase genes in COL10A1 ^-/- pellets compared with parental cell pellets. These in vitro and in vivo data suggested that collagen X is dispensable for the hypertrophic differentiation and endochondral ossification of human iPSC-derived chondrocytes, though it may facilitate the differentiation process. Thus, COL10A1 ^-/- iPSC lines are useful for investigating the physiological role of collagen X in chondrocyte differentiation. © 2023 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

Keywords: BONE MATRIX; CHONDROCYTE AND CARTILAGE BIOLOGY; COLLAGEN; DISEASES AND DISORDERS OF/RELATED TO BONE; GROWTH PLATE.

Fig. 1

Generation of COL10A1‐KO cell lines from human iPS cell lines. (A) Schematic diagram of the COL10A1 locus before (upper) and after (lower) genome editing using N‐ and C‐sgRNA. Gray and hatched boxes indicate noncoding and coding regions, respectively. F1, F2, R1, and R2 indicate the location of PCR primers for the evaluation of the genome editing. (B) Demonstration of an 8.1 kb deletion in the COL10A1 locus by genomic DNA PCR. Deletion is confirmed by the loss of a longer fragment (9.3 kb). Shorter fragments from N‐terminal and C‐terminal regions are amplified in cases with a heterozygous deletion but not with homozygous deletions. (C) Sequences of the longer and shorter fragments demonstrated in (B). Solid and dotted lines indicate conserved and deleted regions, respectively. Inserted sequences are highlighted in red.

Fig. 2

Induction of hypertrophic chondrocytes from parental (414C2), COL10A1 ^+/−, and COL10A1 ^−/− iPSC lines. (A) Phase contrast images and (B) Safranin‐O and Fast‐Green staining of cell pellets at different days during the induction. Scale bar = 500 μm. Similar results were obtained in three independent experiments. (C) mRNA expression of growth plate–related genes during the induction. RNAs were extracted from 3D chondrocyte pellets at each time point and assessed for the expression of each gene by qRT‐PCR. The expression level was normalized to that of parental iPSC‐derived pellets at day 0, except COL10A1, for which the level at day 28 was used for the normalization. Statistical analysis was performed using a two‐way ANOVA. Data are presented as boxplots presenting all points from the minimum to maximum values (n = 3, independent experiments).

Fig. 3

Expression of cartilage matrix at day 42 in parental (414C2), COL10A1 ^+/−, and COL10A1 ^−/− iPSC‐derived 3D chondrocyte pellets. (A) Histological evaluation of parental iPSC‐derived pellets. Cell pellets at each time point were stained with Safranin‐O and Fast‐Green (SOFG) and antibodies against COL I, COL II, or COL X. Scale bar = 500 μm. (B) Histological evaluation of parental, COL10A1 ^+/−, and COL10A1 ^−/− iPSC‐derived pellets at day 42. Cell pellets were stained with SOFG and antibodies against COL I, COL II, or COL X. Scale bar = 500 μm. (C) Capillary‐based immunoassay of proteins extracted from cell pellets. Cellular proteins were extracted from pellets at day 42 and analyzed by Western blot using antibodies against COL I, COL II, or COL X. The experiments were performed three times with similar results.

Fig. 4

Evaluation of chondrocyte pellets at the proliferating stage in vivo. (A) Cell pellets derived from parental (414C2), COL10A1 ^+/− , and COL10A1 ^−/− iPSC lines at day 14 were transplanted into immunodeficient mice, and ossification in the transplants was analyzed by X‐ray biweekly. (B) Comparison between original pellets and samples collected 98 days after the transplantation. The size of the cutting surface area was enlarged 18.8 times. Pellets were stained with Safranin‐O and Fast‐Green. Scale bar = 500 μm. (C) Histology of parental iPSC‐derived transplants. The collected samples were stained with hematoxylin–eosin (HE), Safranin‐O and Fast‐Green (SOFG), or an antibody against human nuclear antigen (HNA). Scale bar = 500 μm. (D) Histology of parental, COL10A1 ^+/− , and COL10A1 ^−/− iPSC‐derived transplants. The collected samples were stained with HE, SOFG, and antibodies against COL I, COL II, or COL X. Scale bar = 500 μm. Similar results were obtained in three independent experiments.

Fig. 5

Evaluation of chondrocyte pellets at the prehypertrophic stage in vivo. (A) Cell pellets derived from parental (414C2), COL10A1 ^+/− , and COL10A1 ^−/− iPSC lines at day 28 were transplanted into immunodeficient mice, and ossification in the transplants was analyzed by X‐ray biweekly. (B) Comparison of original pellets and samples collected 56 days after the transplantation. The size of the cutting surface area was enlarged 1.9 times. Pellets were stained with Safranin‐O and Fast‐Green. Scale bar = 500 μm. (C) Histology of parental iPSC‐derived transplants. The collected samples were stained with hematoxylin–eosin (HE), Safranin‐O and Fast‐Green (SOFG), or an antibody against human nuclear antigen (HNA). Scale bar = 500 μm. (D) Histology of parental, COL10A1 ^+/− , and COL10A1 ^−/− iPSC‐derived transplants. The collected samples were stained with HE, SOFG, and antibodies against COL I, COL II, or COL X. Scale bar = 500 μm. Similar results were obtained in three independent experiments.

Fig. 6

Comparison of mRNA expression profiles between parental and mutant iPSC‐derived 3D chondrocyte pellets at day 42. RNAs were extracted from parental, COL10A1 ^+/− , and COL10A1 ^−/− iPSC‐derived pellets at day 42 and processed for RNA sequencing analyses. (A, B) PCA of 414C2‐derived cells (A) and 1231A3‐derived cells (B) (n = 3, independent experiments). (C, D) Venn diagrams showing the overlap of DEGs for up‐ (C) and downregulated (D) genes in COL10A1 ^−/− pellets. (E, F) The list of up‐ (E) and downregulated (F) genes in COL10A1 ^−/− pellets. (G) Comparison of mRNA expression levels of growth plate–related genes between parental (414C2) and COL10A1 ^−/− ‐derived pellets. Expression levels of each gene are shown as TPM (transcript per million). Statistical analysis was performed using a one‐way ANOVA. Data are represented as boxplots presenting all points from the minimum to maximum (n = 3, independent experiments).