Aggrecan is required for growth plate cytoarchitecture and differentiation

Abstract

The proteoglycan aggrecan is a prominent component of the extracellular matrix in growth plate cartilage. A naturally occurring, recessive, perinatally lethal mutation in the aggrecan core protein gene, cmd(bc) (Acan(cmd-Bc)), that deletes the entire protein-coding sequence provided a model in which to characterize the phenotypic and morphologic effects of aggrecan deletion on skeletal development. We also generated a novel transgenic mouse, Tg(COL2A1-ACAN), that has the chick ACAN coding sequence driven by the mouse COL2A1 promoter to enable the production of cmd(bc)/cmd(bc); Tg(COL2A1-ACAN) rescue embryos. These were used to assess the impact of aggrecan on growth plate organization, chondrocyte survival and proliferation, and the expression of mRNAs encoding chondrocyte differentiation markers and growth factors. Homozygous mutant (cmd(bc)/cmd(bc)) embryos exhibited severe defects in all skeletal elements with deformed and shortened (50%) limb elements. Expression of aggrecan in rescue embryos reversed the skeletal defects to varying degrees with a 20% increase in limb element length and near-full reversal (80%) of size and diameter of the ribcage and vertebrae. Aggrecan-null growth plates were devoid of matrix and lacked chondrocyte organization and differentiation, while those of the rescue embryos exhibited matrix production concomitant with partial zonation of chondrocytes having proliferative and hypertrophic morphologies. Deformation of the trachea, likely the cause of the mutation's lethality, was reduced in the rescue embryos. Aggrecan-null embryos also had abnormal patterns of COL10A1, SOX9, IHH, PTCH1, and FGFR3 mRNA expression in the growth plate. Expression of chick aggrecan in the rescue embryos notably increased COLX expression, accompanied by the reappearance of a hypertrophic zone and IHH expression. Significantly, in transgenic rescue embryos, the cell death and decreased proliferation phenotypes exhibited by the mutants were reversed; both were restored to wild-type levels. These findings suggest that aggrecan has a major role in regulating the expression of key growth factors and signaling molecules during development of cartilaginous tissue and is essential for proper chondrocyte organization, morphology, and survival during embryonic limb development.

Keywords: Aggrecan; Cartilage matrix deficiency; Chondrocyte morphology; Extracellular matrix; Growth plate; Proteoglycan.

Figure 1. Generation of chick aggrecan transgenic mice

(A) Schematic representation of the Col2a1-chick aggrecan DNA construct utilized for generation of Agc¹⁷/+, Agc¹⁸/+, and Agc¹²⁸³/+ transgenic mice. The construct consists of a 3 kb Col2a1 promoter (hatched box), 237 bp exon 1 (gray box), a 3.02 kb segment of Col2a1 intron 1, and the 6.3 kb chick aggrecan gene. Restriction sites used in cloning are depicted and described in the Materials and Methods section. (B) Representative Southern blot using a ³²P-dCTP labeled probe (panel A, red box) of genomic DNA from each of the Agc/+ transgenic lines following XbaI and KpnI digestion (panel A, red). Transgenic line Agc¹⁷/+ contained the highest copy number with greater than 50 copies. Representative immunohistochemistry experiments using monoclonal antibodies against chick aggrecan (S103L, C) and chondroitin-6-sulfate (D) are shown for distal femurs of 18.5 dpc embryos of each genotype from the same litter. SA=splice acceptor site

Figure 2. Skeletal element analysis in aggrecan null cmd^Bc/cmd^Bc embryos and transgenic cmd^Bc/cmd^Bc;Ag¹⁷/+ rescue mice

(A) Comparison between whole skeletons of wildtype (+/+), mutant (cmd^Bc/cmd^Bc), and rescue (cmd^Bc/cmd^Bc;Agc¹⁷/+) embryos harvested 18.5 dpc and stained with Alizarin red (presumed ossified areas) and Alcian blue (cartilage). Individual skeletal element dissections were analyzed for staining patterns and overall size for the cervical spine (B–C), lumbar spine (D), ribcage (E), forelimb digits (G), hindlimb digits (H) and sternum (I). Arrows demonstrate Alizarin red staining differences in the spine and digits, and sternum segments are labeled to demonstrate abnormal staining patterns of the third through fifth sternebrae in mutants (I). Quantification of long bone deficiencies found a 50% decrease in length of mutant limbs, with a 20% improvement in rescue mice (F). n=6–18 embryos/genotype, *p<0.001 vs. +/+, #p<0.05 vs. cmd^Bc/cmd^Bc, one-way ANOVA with Tukey’s post-hoc testing.

Figure 3. Histological examination of mutant and rescue growth plate cytoarchitecture

Representative sections of 18.5 dpc femoral growth plates stained with hematoxylin and eosin demonstrating chondrocyte morphology within each growth plate zone (A–C) for wild-type, mutant, and rescue embryos. (A’–C’) Higher magnification of the proliferative and hypertrophic regions outlined in black boxes in top row. Red outlines highlight the columnar formation of proliferative chondrocytes, and the black box denotes chondrocytes with hypertrophic morphology, both of which are absent in aggrecan-null growth plates. Note the hypercellularity and lack of extracellular matrix in mutant embryos (arrow heads) compared to wild type and rescue (arrows). (A’’–C’’) Sections of cartilaginous rings from the upper third of the trachea, showing differences in airway patency with and without aggrecan. R= resting, P= proliferative, H= hypertrophic, L= lumen, D= dorsal aspect.

Figure 4. Expression pattern of growth plate transcripts by mRNA in situ hybridization

Representative expression analyses of Col2a1 (A–C), Sox9 (D–F), Col10a1 (G–I), Ihh (J–L), Ptch1 (M–O), and Fgfr3 (P–R) transcripts in 18.5 dpc tibia sections from wild-type, mutant, and rescue embryos. The panels demonstrating Ihh data contain insets that enlarge the prehypertrophic/hypertrophic zone of expression. Results were observed consistently in five independent experiments with at least four embryos for each mRNA probed. In situ hybridizations with DIG-labeled RNA probes were performed on 40 µm sections; cartilage elements are outlined for clarity.

Figure 5. Patterns of cell death in the growth plate of mutant and rescue embryos

Paraffin sections of 18.5 dpc embryonic femurs were analyzed for apoptotic cells using TUNEL staining. Cartilaginous element limits are outlined in white. (A–C) Increased numbers of apoptotic cells are present in mutants, an occurrence not found in rescue embryos. (A’–C’) Higher magnification of the white boxes outlined in A–C, demonstrating the aberrant cell death occurring in the resting zone of mutants but not in wild-type or rescue embryos. (A’’–C’’) Negative control images of resting zone areas outlined. (D) Quantification of total TUNEL-positive cells in the growth plate. n=3–6 embryos/genotype, *p<0.001 vs. +/+, #p<0.001 vs. cmd^Bc/cmd^Bc, one-way ANOVA with Tukey’s post-hoc testing.

Figure 6. Proliferation in mutant and rescue embryonic growth plates

Embryonic femurs 18.5 dpc stained with DAPI and anti-phospho-histone H3 (PHH3) for analysis of mitotic chondrocytes within the proliferative zone. (A–C) DAPI-stained proliferative chondrocytes demonstrate partial columnar formation in rescue embryos (red arrow heads) compared to mutants. (A’’–C’’) Merged images show that rescue proliferative chondrocytes include double the percentage of pHH3-staining cells in the growth plate compared to mutants, which is comparable to wild-type levels. (D) Quantification of pHH3-positive proliferative chondrocytes in the growth plate. n=3–6 embryos/genotype, *p<0.001 vs. +/+, #p<0.01 vs. cmd^Bc/cmd^Bc, one-way ANOVA with Tukey’s post-hoc testing.