Abstract
Introduction
During development, the intervertebral disc (IVD) forms by segmentation of the embryonic axial skeleton which comprises the rod-like notochord (NC) surrounded by the somatic mesenchyme of the sclerotome. NC-derived signals are critical for induction and patterning of the sclerotome, which gives rise to the vertebral body. As the vertebrae form, the NC is segmented, forming the nucleus pulposus (NP) of the future IVDs. Embryonic notochordal cells (NCCs) have been hypothesized to be the precursors of adult NP cells (NPCs). We and others have shown from lineage tracing in mice that NP cells are of notochordal origin, suggesting NCCs are the progenitors of NP cells. We aim to define the molecular signature of the NC lineage at different developmental stages.
Material and Methods
We generated mice in which NC and NP cells are marked by GFP by crossing our NC specific Foxa2 MNE-cre mice to Z/EG mice and isolated pure populations of cells by FACS. We have generated transcriptome data from isolated mouse embryonic (at E8.5-E10) NC and postnatal (at P2) NP cells. RNAseq data were obtained from these cells and analyzed bioinformatically.
Results
In analyses of the transcriptomes of NC and NP cells, we have focused on (1) specific molecules that constitute the NC/NP cell niche (secreted factors, receptors, extracellular matrix [ECM] components), (2) unique markers for isolating the NC/NP cells at different time points, and (3) transcription regulators are essential for their differentiation. The rationale are as follows: many of the genetic risk factors identified so far for intervertebral disc degeneration (IDD) affect the ECM (e.g., COL9A1, aspirin, CHST3CILP, and ACAN) and knowledge of the composition of the extracellular niche would be important for interpreting genetic association data and will also guide us in designing appropriate ECM and growth factor support for maintaining isolated NCCs and NPCs. Our NCC data compared with published E8.0–E8.5 NC microarray expression data showed a high degree of similarity (> 0.8 Spearman rank correlation coefficient). Comparison of NCC transcriptome data with published data for mouse ES/iPS/EpiS cells showed a much lower correlation consistent with the progressive differentiation from early stages of development to NCCs. For NCCs, Wnt signaling was the major pathway, expression of genes involved in oxidative phosphorylation, and cell cycle/proliferation was enriched. The transcription factors characteristic of NCCs were Brachyury and Foxa2. There was enrichment for Hox genes. Hox genes control the body plan along the anterior and posterior axis, but their roles in NCC fate is not clear. For NPCs, the TGF-β, FGF, and VEGF pathways were dominant. Unlike NCCs, for NPCs, genes involved in ECM cell interaction (integrin signaling and focal adhesion) and Klf transcription factors dominated.
Conclusion
The spectrum of ECM genes expressed in NCCs differs significantly from NPCs, suggesting very different ECM niches for these cells. NCCs and NPCs also differed in terms of signaling pathways. The NPC characteristic genes correlated well with the known genetic risk factors identified so far for IDD and will therefore be a rich source of additional candidate genes for genetic risk factors for IDD.