Joint pain causes significant morbidity in osteoarthritis (OA). The aetiology of joint pain in OA is not well understood. The synovial membrane as an innervated joint structure represents a potential source of peripheral pain in OA. Here we analyse, using a hypothesis-free next generation RNA sequencing, the differences in protein-coding and non-coding transcriptomes in knee synovial tissues from OA patients with high knee pain ( = 5) compared with OA patients with low knee pain ( = 5), as evaluated by visual analogue scale (VAS). We conduct Gene Ontology and pathway analyses on differentially expressed mRNA genes. We identify new protein-coding, long non-coding RNA and microRNA candidates that can be associated with OA joint pain. Top enriched genes in painful OA knees encode neuronal proteins that are known to promote neuronal survival under cellular stress or participate in calcium-dependent synaptic exocytosis and modulation of GABA(γ-aminobutyric acid)ergic activity. Our study uncovers transcriptome changes associated with pain in synovial microenvironment of OA knees. This sets a firm ground for future mechanistic studies and drug discovery to alleviate joint pain in OA.