Periostin, a matricellular protein in the fasciclin family, is expressed in tissues subjected to constant mechanical stress. Periostin modulates cell-to-extracellular matrix interactions and can bind to collagen, fibronectin, tenascin-C and several integrins. Our objective was to evaluate whether periostin is expressed in the human intervertebral disc. Immunohistochemical localization of periostin was carried out in tissue of human lumbar discs and lumbar discs of the sand rat (Psammomys obesus). Human discs also were examined for periostin gene expression. Immunohistochemical localization demonstrated periostin in the cytoplasm of annulus and nucleus cells, and occasionally in the surrounding pericellular and interterritorial extracellular matrix. Periostin distribution in the human disc was distinctive. Outer annulus contained the highest proportion of periostin-positive cells (88.8%), whereas inner annulus contained only 61.4%. The nucleus pulposus contained the fewest periostin-positive cells (18.5%). There was a significant negative correlation between the percentage of cells positive for periostin in the inner annulus and subject age. Periostin gene expression in the human disc also was confirmed using molecular microarray analysis. Because work by others has shown that periostin plays an important role in the biomechanical properties of other connective tissues (skin, tendon, heart valves), future research is needed to elucidate the role of periostin in disc, loading, aging and degeneration.

Mouse class II major histocompatibility complex genes have been shown to be regulated at the level of transcription for both tissue-specific and inducible expression. In particular, IFN-gamma induction of the class II genes has been shown to occur at the transcriptional level, although the role that additional post-transcriptional mechanisms of regulation may play in this induction is not known. To evaluate IFN-gamma effects on transcriptional and post-transcriptional events of class II gene expression, we examined the rate of decline of class II transcription, steady-state mRNA, and cell surface protein following the removal of IFN-gamma from maximally stimulated WEHI-3 cells (an IFN-gamma inducible, myelomonocytic cell line). We determined that transcription of class II genes almost completely returned to baseline levels eight hours after removal of IFN-gamma. However, the steady-state level of class II mRNA's required 4 days, and membrane Ia expression required 5 days to return to baseline levels. This decay was linear and allowed us to determine a half-life value of 16-20 h for class II transcripts. These data demonstrate that, following removal of IFN-gamma from fully stimulated cells, transcription of the class II genes declined rapidly, but mRNA was quite long-lived. We also assessed the class II mRNA stability in unstimulated WEHI-3 cells and the B-cell lymphoma. A20/2J, by actinomycin D treatment and northern blot analysis. In agreement with the IFN-gamma washout experiments, transcripts from all four class II genes were quite long-lived in these cell types, with a half-life greater than ten hours. These data support the concept that IFN-gamma acts primarily at the level of class II transcription and argues against IFN-gamma playing a major role in post-transcriptional modulation of class II expression.