Marr, N.; Meeson, R.; Piercy, R. J.; Thorpe, C. T.; Hildyard, J. C. W.

Tendons are one of the major load-bearing tissues in the body; subjected to enormous peak stresses, and thus vulnerable to injury. Cellular responses to tendon injury are complex, involving inflammatory and repair components, with the latter employing both resident and recruited exogenous cell populations. Gene expression analyses are valuable tools for investigating tendon injury, allowing assessment of repair processes and pathological responses such as fibrosis, and permitting evaluation of therapeutic pharmacological interventions. Quantitative polymerase chain reaction (qPCR) is a commonly used approach for such studies, but data obtained by this method must be normalised to reference genes: genes known to be stably expressed between the experimental conditions investigated. Establishing suitable tendon injury reference genes is thus essential. Accordingly we investigated mRNA expression stability in a rat model of tendon injury, comparing both injured and uninjured tendons, and the effects of rapamycin treatment, at 1 and 3 weeks post injury. We used 11 candidate genes (18S, ACTB, AP3D1, B2M, CSNK2A2, GAPDH, HPRT1, PAK1IP1, RPL13a, SDHA, UBC) and assessed stability via four complementary algorithms (Bestkeeper, deltaCt, geNorm, Normfinder). Our results suggests that ACTB, CSNK2A2, HPRT1 and PAK1IP1 are all stably expressed in tendon, regardless of injury or drug treatment: any three of these would serve as universally suitable reference gene panel for normalizing qPCR expression data in the rat tendon injury model. We also reveal 18S, UBC, GAPDH, and SDHA as consistently poor scoring candidates (with the latter two exhibiting rapamycin- and injury-associated changes, respectively): these genes should be avoided.