Objectives-To analyse the collagen composition of normal adult human supraspinatus tendon and to compare with: (1) a flexor tendon (the common biceps tendon) which is rarely involved in any degenerative pathology; (2)

TMŽ . Atlas cDNA cell interaction arrays CLONTECH were used to examine degenerate tissue from four patients with Achilles tendon disorders, in order to identify changes in expression of genes important in cell᎐cell and cell᎐matrix Ž . interactions. The greatest difference between normal post-mortem and degenerate tissue samples was in the level of MMP-3 Ž . stromelysin mRNA, which was down-regulated in all the degenerate samples. Quantitative RT-PCR assay of RNA extracted from paired 'normal' and degenerate Achilles tendon tissue samples taken from tendons during surgery mirrored the results of the arrays. Levels of MMP-3 mRNA were lower, whereas levels of type-I and type-III collagen mRNAs were significantly higher, in the degenerate compared to the 'normal' samples. Immunoblotting of proteins extracted from the same tendon samples showed that three of four normal tissue samples taken from individuals without apparent tendon disorder had much higher levels of MMP-3 protein than 'normal' or degenerate samples from patients with tendinosis. We suggest that MMP-3 may play an important role in the regulation of tendon extracellular matrix degradation and tissue remodelling. ᮊ

Objectives-To analyse the glycosaminoglycans of the adult human rotator cuff tendon matrix, to characterise changes in the glycosaminoglycan composition with age and in chronic rotator cuff tendinitis. Methods-Rotator cuff (supraspinatus) tendons (n = 84) and common biceps tendons (n = 26) were obtained from cadavers with no history of tendon pathology (age range 11-95 years). Biopsies of rotator cuff tendons (supraspinatus and subscapularis tendons, n = 53) were Conclusions-The normal supraspinatus tendon has the proteoglycan/glycosaminoglycan of tendon fibrocartilage, which it is suggested is an adaptation to mechanical forces (tension, compression and shear) which act on the rotator cuff tendons in the shoulder, although other factors such as reduced vascularity, low oxygen tension and the influence of local growth factors may also be important. This functional adaptation may have important consequences for the structural strength of the supraspinatus tendon and to influence the ability of the tendon to repair after injury. The glycosaminoglycan composition of tendon specimens from patients with chronic tendinitis is consistent with acute inflammation and new matrix proteoglycan synthesis, even in relatively old tendon specimens and after at least one injection of corticosteroid. (Ann Rheum Dis 1994; 53: 367-376) Chronic shoulder pain is a common and disabling condition that is often seen in rheumatological practice and prevalent in the elderly community.1 1 Relatively little is known about the underlying pathology, as surgical biopsies of human tendons are rarely obtained and are not required for diagnosis. The main causes of shoulder pain are lesions of the rotator cuff tendons, most commonly the supraspinatus tendon, although the subscapularis, infraspinatus and teres minor Riley, Harrall, Constant, Chard, Cawston, Hazleman tendons may also be involved.3 The clinical description of 'tendinitis' is often used to describe painful tendon lesions, although there is little evidence for an acute inflammatory process in degenerate and spontaneously ruptured tendons. Rotator cuff tendinitis does not necessarily resolve with time and is often refractory to conservative therapies, including rest, physiotherapy and local corticosteroid injections.4 5Degeneration of the tendon matrix is generally considered to predispose to 'tendinitis' and eventual rupture, as normal tendons are immensely strong under tension.6 A variety of factors including ageing, vascular insufficiency, the anatomical shape of the acromium, impingement against osteophytes, and repetitive activities have all been implicated in the pathology of rotator cuff tendinitis.7 8 Whatever the cause, the most frequent site of all tendon lesions is the 'critical zone' in the supraspinatus tendon, approximately one centimetre from the bone insertion and a region of blood vessel anastomoses.9Although they form a minor proportion of the extracellular matrix, proteoglycans and their constituent glycosaminoglycans can influence many important physiolo...

Objective. To determine whether the fluoroquinolone antibiotic ciprofloxacin, which can cause tendon pain and rupture in a proportion of treated patients, affects the expression of matrix metalloproteinases (MMPs) in human tendon-derived cells in culture.Methods. Cell cultures were derived from 6 separate tendon explants, and were incubated in 6-well culture plates for 2 periods of 48 hours each, with ciprofloxacin (or DMSO in controls) and interleukin-1␤ (IL-1␤), alone and in combination. Samples of supernatant medium from the second 48-hour incubation were assayed for MMPs 1, 2, and 3 by Western blotting. RNA was extracted from the cells and assayed for MMP messenger RNA (mRNA) by semiquantitative reverse transcription-polymerase chain reaction, with normalization for GAPDH mRNA.Results. Unstimulated tendon cells expressed low or undetectable levels of MMP-1 and MMP-3, and substantial levels of MMP-2. IL-1␤ induced a substantial output of both MMP-1 and MMP-3 into cell supernatants, reflecting increases (typically 100-fold) in MMP mRNA, but had only minor effects on MMP-2 expression. Ciprofloxacin had no detectable effect on MMP output in unstimulated cells. Preincubation with ciprofloxacin potentiated IL-1␤-stimulated MMP-3 output, reflecting a similar effect on MMP-3 mRNA expression. Ciprofloxacin also potentiated IL-1␤-stimulated MMP-1 mRNA expression, but did not potentiate the output of MMP-1, and had no significant effects on MMP-2 mRNA expression or output.Conclusion. Ciprofloxacin can selectively enhance MMP expression in tendon-derived cells. Such effects might compromise tendon microstructure and integrity.

The aim of the study was to immunolocalise interstitial collagenase (MMP-1) and the tissue inhibitor of metalloproteinases (TIMP-1) in ulcerating corneas from patients with rheumatoid arthritis, to determine whether changes in expression are associated with destructive corneal disease. Collagenase was expressed by stromal cells in 8 of 8 ulcerating corneas but was not seen in normal tissue (n = 3). TIMP-1 was abundant throughout the normal stroma, but was much reduced or absent from diseased corneas. Collagenase staining was frequently more intense near the epithelial surface and associated with a cellular infiltrate consisting of activated antigen-presenting cells (HLA-DR+), many of which were macrophages (CD68+) and derived from the epithelium or limbus (S100+). Interstitial collagenase produced by infiltrating macrophages and/or stimulated corneal fibrocytes is probably a major mediator of collagen degradation in rheumatoid corneal ulceration. In addition, reduced levels of TIMP-1 expression are consistent with collagenase activity and tissue destruction. Epithelial-stromal cell interactions and the production of local inflammatory mediators are of major importance in the pathogenesis of corneal destruction, although the precise nature of the antigenic stimulation and/or cellular interactions remains to be elucidated.