Background There is significant correlation between chronic lower back pain (LBP) and degeneration of the intervertebral disc (DIVD). Current treatment approaches for LBP have been ineffective but cell-based approaches show greater restorative potential. Notochordal (NC) cells have been shown to possess appropriate therapeutic characteristics. However, limited cell yields from native tissue isolations and the lack of an appropriate methodology for cell expansion has prevented their full characterisation and practical application. This study aimed to develop a culture system that would allow for the expansion of NC cell populations while retaining NC-like phenotype using porcine NC cells as an exemplar. Additionally, this work aimed to improve current NC cell isolation methodology to enhance cell yields and give greater confidence in generation of pure NC cell populations, a particularly necessary advancement with the aim of utilising human NC cells for research and therapy. Methods Porcine NC cells were isolated from 6 week post-natal discs and cultured in vitro utilising different conditions: (1) Media: DMEM vs Î±MEM; (2) Surface coating: laminin-521, fibronectin, gelatin and uncoated tissue culture plastic; (3) Hypoxia: 2% O2 vs normoxia; (4) Media osmolarity: Î±MEM (300 mOsm/L) vs Î±MEM (400 mOsm/L); (5) Surface stiffness: 0.5 kPa, 4 kPa and standard tissue culture-treated polystyrene. Each condition was investigated sequentially. NC cells were cultured in alginate beads as a control. Adherence, proliferation, cell viability, morphology and expression of known NC cell markers (KRT8, KRT18, KRT19, CD24, T) were assessed over a 14-day culture period. IHC and flow cytometry were performed with a number of cell lines to identify and optimise the use of potential NC cell surface markers (CD13, CD24, CD69, CLDN12, CDH6) for NC cell isolation. Results NC cell cultures with Î±MEM and on laminin-521-coated surfaces achieved the greatest cell adherence, proliferation, viability and retention of NC cell morphology and phenotype over a 14-day culture period. NC cell phenotype was retained comparable to alginate culture under these conditions. Culture under hypoxia was shown to further improve porcine NC cell population expansion and culture on 0.5 kPa culture surfaces enhanced retention of NC-like gene expression profile and morphology. CLDN12 and CDH6 were identified in foetal human NP tissue. It was indicated that CLDN12, and possibly CDH6, will be useful for NC cell isolation alongside CD24. Discussion This study has demonstrated an optimised system which not only retains phenotype comparable to the current gold standard, but also it allows for proliferation of NC populations. Additionally, findings described here suggest additional NC cell surface markers for use in improved isolation methodology. Application of, and further developments to, this isolation and optimised NC cell culture system will enable more detailed study of NC cells from all sources, and represent a critical step in the research and therapeutic application of human NC cells.