Inflammatory bowel diseases (IBD), are characterized by an inflammatory cascade of mediators capable of degrading and modifying bowel wall structure as well as inducing the formation of chronic inflammatory lesions of the digestive tract. The inflammatory cell infiltrate observed in chronic mucosal inflammation is associated with changes in epithelial proliferation and migration and accompanied by intensive remodeling of the subepithelial connective tissue, which in turn leads to increased turnover of extracellular matrix (ECM) components [1, 2]. A disturbance in the balance between synthesis and degradation of ECM components can result either in progressive organ destruction, as seen in ulcer formation, or fibrosis due to excessive deposition of collagen [1, 2].

Metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) show a regulated and coordinated pattern of activity which allows tissue degradation and remodeling but at the same time it prevents tissue damage [3, 4]. TIMPs are the natural inhibitors of MMPs found in most tissues and body fluids. Currently, four TIMPs (TIMP-1, -2, -3 and -4) are identified. Like MMPs, the expression of TIMPs in the tissue is also controlled to maintain a balance in the metabolism of the ECM [4]. Disruption of this balance may result in a number of pathogenic processes. The 21–34 kDa proteinic molecules of TIMPs express an inhibitory activity which is facilitated by their ability to form high-affinity non-covalent complexes with the carboxyl-terminal domains of pro-MMPs [3]. TIMP-1, TIMP-2 and TIMP-4 are present in soluble forms, while TIMP-3 is tightly bound to the matrix [4]. Numerous studies have indicated that, independently of MMP inhibition, TIMPs are multifunctional proteins involved not only in tissue remodeling and wound healing but also in many other physiological and pathological processes such as angiogenesis, steroidogenesis, hematopoiesis, cell growth and cell survival [4].

The TIMP-1 levels in the mucosa and plasma of IBD patients have been shown to be elevated, in previous studies [5–8]. Moreover, the TIMP-1 plasma levels demonstrate a significant positive correlation with the endoscopic degree of mucosa injury, disease activity and C-reactive protein concentration, in UC patients [8]. In contrast, TIMP-2-mRNA levels remained unchanged and no remarkable expression of TIMP-2 was observed in inflamed mucosa of IBD patients [6, 7]. To our knowledge, no data regarding TIMP-4 in IBD have been previously presented. On the other hand, the assessment of serum TIMPs levels has not been performed to date in IBD patients. The aim of the present study was to evaluate serum levels of TIMPs -1, -2 and -4 in IBD patients and to examine possible correlations with known inflammatory markers as well as with the clinical characteristics of patients.

Our study provides new data on the tissue remodeling process in IBD with regard to serum TIMPs levels. We confirmed an increased production of TIMP-1 in IBD patients, especially in active disease and showed that this increase correlated with the production of the well-known inflammatory markers CRP and SAA. On the other hand, a decreased production of TIMP-4 was detected in these patients, which was independent of disease activity along with a strong correlation between TIMP-4 levels and the studied inflammatory markers. What seemed to be of particular interest was also the observation that males in comparison with females, tend to show an increased production of TIMP-1, accompanied by a decreased production of serum TIMP-4.

It has been shown that the uncontrolled and sustained inflammatory cascade, observed in IBD, gives rise to the production of MMPs and TIMPs which in turn can induce tissue degradation and lesion development [1, 2]. We agree that the determination of serum MMP levels would add to the impact of this study. Unfortunately this was not included in the initial study design. The sera from these 150 patients are not available anymore, as they have been utilized for other determinations. On the other hand, it would be biased to obtain new sera because the inflammatory status (activity) of these patients has changed.

An increased production of TIMP-1 at the mucosa and plasma of IBD patients has been previously described [5–8]. It has also been demonstrated that in IBD, TIMP-1 is expressed by inflammatory cells, fibroblasts and vascular smooth muscle cells most prominently in actively inflamed areas in ulcer bases [7]. In an ex vivo study using tissue cultures of intestinal biopsies from IBD patients, TIMP-1 was detected in significant titers in inflamed mucosa of both CD and UC specimens, in contrast with uninflamed mucosa [5]. In the above-mentioned study, TIMP-1 showed strong correlation with proinflammatory cytokines IL-6, IL-1β and IL-10. In another study, an increased expression of TIMP-1 mRNA in inflamed and especially ulcerated colon mucosa of IBD patients was evident [6].

Increased levels of TIMP-1 have also been associated with the presence of fibrotic strictures in CD [11]. It is well known that, TIMPs are important markers of fibrosis. In addition, CD and especially its stenotic form are characterized by increased fibrosis. It was thus expected to find a more pronounced increase in TIMP-1 levels, in these subgroups. In our study, a trend towards higher TIMP-1 levels was observed in CD compared with UC patients which was not statistically significant. This could be attributed to the rather small number of patients which did not allow reliable subgroup analysis. In a recent study, TIMP-1 plasma levels in UC patients were found to correlate positively with scored endoscopic degree of mucosal injury, disease activity indices, clinical activity indices and CRP concentration, thus implying that TIMP-1 plasma concentration may be a possible biomarker of disease activity [8]. External stimuli such as growth factors, phorbol esters and cytokines (IL-6, IL-1 and IL-1b) are all well-known triggers of TIMP-1 expression in various cell types [4, 12]. Erythropoietin (EPO) a hormone that also serves as a promoter of TIMP-1 secretion, is increased in IBD patients, as we have demonstrated in a previous study [13]. Thus, it is possible that EPO and TIMP-1 act in a collaborative manner and play a key role not only in the erythroid cell physiology but also in several pathophysiological events in IBD.

No statistically significant difference in TIMP-2 levels was recorded between IBD patients and HC, in our study. TIMP-2 mRNA levels measured by polymerase chain reaction (PCR) in biopsies from IBD patients, remained unchanged in inflamed and uninflamed mucosa, in the study of von Lampe et al [6]. Finally, no remarkable expression of TIMP-2 was observed in inflamed mucosa of IBD patients, in an immunohistochemical study by Arihiro et al [7]. However, it is not safe to draw any conclusions based solely on these data thus, this subject requires further studies.

One major finding presented in this study is that TIMP-4 serum levels are decreased in IBD patients, irrespective of disease activity. When interpreting this result many considerations need to be taken into account. TIMP-4, the newest member of TIMPs, is a tissue-specific regulator of EMC remodeling, promoting inhibition of the MMP-2 activation [3]. On the other hand, MMP-2 activity in IBD patients has been shown to be rather excessive and it could be held accountable for TIMP-4 consumption which in turn could result in lower TIMP-4 levels. The enhanced MMP-2 activity in IBD may be of significant importance, as it is a key player in proper wound healing, angiogenesis and re-epithelization, as well as in the regulation of epithelial barrier function of the intestine [1, 2]. It has also been suggested that the increased TIMP-4 expression in normal heart tissue could explain why myocardial tumors are so rare [14]. On the other hand, there is evidence that low TIMP-4 levels result in an anti-apoptotic effect [15]. These reports, taken together with the data of the present study, may support the importance of low TIMP-4 levels in IBD-related tumorigenesis. Additionally, TIMP-4 has been shown to inhibit platelet aggregation, thus implying TIMP-4 involvement in the regulation of platelet aggregation and recruitment [16]. These platelet aggregation responses proved to be enhanced in IBD patients, even in inactive disease [17]. These findings suggest that, the decreased TIMP-4 serum levels observed in IBD patients, even those with inactive disease, have an aggravating influence on the platelet aggregation in IBD. To confirm these speculations, further studies are necessary. In any case, the imbalance of TIMP-1 and TIMP-4 serum levels observed herein, may reflect the paradoxical effects of these inhibitors, which may act as either promoters or suppressors in many important pathophysiological processes involved in inflammation and wound healing in IBD.

We have also showed that, gender has an influence on TIMP-1 and TIMP-4 serum levels. This gender-related difference may prove interesting but, how this is controlled is the subject of many differing lines of research. The involvement of TIMP-1 in steroidogenesis was first evidenced in Leyding cells and ovarian granulose cells and led to the conclusion that it could affect germ cell development, by regulating steroid concentrations in both males and females [18]. Moreover, it has been shown that TIMP-1 is not a prerequisite for overall germ cell development and that it could be substituted by one of the other TIMPs [19]. According to our results, TIMP-4 could be a candidate for such an action. In a recent study with a relatively small sample size, no significant difference in TIMP-1 plasma levels between females and males was recorded [20]. Given the fact however, that only a few TIMP-related studies with focus on gender have been published, future studies are required so that our findings could be clarified.

In summary, our data may reflect the continuous matrix turnover occurring in IBD, a process strongly associated with an imbalance of TIMP-1 and TIMP-4 serum levels. The increase of TIMP-1 serum levels in active disease and their significant correlation with the production of known inflammatory markers such as CRP and SAA, indicate that TIMP-1 levels could be used not only for diagnostic purposes but also for the assessment of activity in IBD. These findings bring into question the potential role of TIMPs in the pathogenesis, activity and therapy of IBD, thus emphasizing the need for further studies on this subject. Perhaps, in the future, the modulation of MMP and TIMP activities as well as ECM turnover may become part of a more rational, mechanism-based therapeutic manipulation in IBD.