|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online March 5, 2003 as doi:10.1096/fj.02-0725fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Rega Institute for Medical Research, Laboratory of Molecular Immunology, University of Leuven, Minderbroedersstraat 10, 3000 Leuven, Belgium
2 Department of Morphology and Molecular Pathology, University of Leuven, Minderbroedersstraat 12, 3000 Leuven, Belgium
Correspondence should be sent to: Ghislain Opdenakker Rega Institute for Medical Research Minderbroedersstraat 10 B-3000 Leuven Belgium Phone: 32 16 33 73 41 Fax: 32 16 33 73 40 E-mail: ghislain.opdenakker{at}rega.kuleuven.ac.be
SPECIFIC AIMS
Gelatinase B (MMP-9) is a key regulator in the pathophysiology of autoimmune diseases, but its role in diabetes has not yet been investigated. We studied whether and in which cells gelatinase B is expressed in normal, acutely and chronically inflamed pancreas tissue. Since insulin-producing ß cells were found in direct contact with gelatinase B-secreting ductular cells in the islets, another aim was to define how insulin is degraded by gelatinase B.
PRINCIPAL FINDINGS
1. Gelatinase B is absent in normal and expressed in inflamed pancreas tissues
We studied by immunohistochemistry the expression of gelatinase B in chronic (n=9) and acute pancreatitis (n=6) tissues and in control pancreas tissues (n=8). Fig. 1
shows the immunohistochemical analysis on formalin-fixed and paraffin-embedded sections. Panel a illustrates the absence of gelatinase B expression in the exocrine pancreas as well as in the islets of Langerhans of a control patient. Panel b represents a biopsy from a patient in an early stage of acute pancreatitis. The microscopic slide shows marginating and extravascular leukocytes, which stained strongly with the monoclonal antibody against gelatinase B. As indicated in panels c and d, the cases of acute pancreatitis contained many polymorphonuclear and occasional mononuclear cells expressing gelatinase B. Panels e and f show a completely different picture in chronic pancreatitis. The exocrine pancreas is replaced by fibrotic tissue in which islets and ductules are spared. The ductular epithelial cells express considerable amounts of gelatinase B and within the islets also some cells stain positively for gelatinase B.
|
2. Insulin and gelatinase B are expressed by different cell types in close proximity
Double immunostainings for gelatinase B and for insulin were performed on histological sections of patients with chronic pancreatitis in order to detect temporal or spatial colocalisation. Insulin-producing cells did not show a strong positive staining for gelatinase B whereas inflammatory or ductular cells did. However, the two different cell types expressing insulin or gelatinase B were often found in close or direct proximity. In other words, after secretion of insulin by ß cells and gelatinase B by neighboring (ductular) cells, it is obvious that both molecules may interact, in particular after extensive tissue remodeling as observed in chronic pancreatitis.
3. Human insulin is degraded by gelatinase B
To define whether human insulin is targeted by active gelatinase B, both molecules were mixed at a substrate:enzyme ratio in excess of 6.5 and were incubated. Degradation of insulin was visualized by SDS-PAGE and silver staining analysis. Furthermore, the fragmentation was analyzed in detail by mass spectrometry (MS). Enzyme kinetic studies were performed after standardization of the MS analysis by co-injection of known amounts of GCP-2, a molecule with a molecular mass in the same range as intact insulin. Complete digestion of insulin was obtained after a time interval of 10 h. Incubation of insulin with an equivalent amount of stromelysin-1, which is present in the reaction mixture for the activation of progelatinase B, showed complete absence of degradation. Moreover, addition of the metalloproteinase inhibitors EDTA or o-phenantroline to the reaction mixture inhibited the cleavage completely.
4. Site-specificities of insulin degradation by gelatinase B
The reaction mixture of a complete digestion of 50 µg human insulin by gelatinase B was separated by RP-HPLC on a C18 column (Fig. 2
a) with on-line mass spectrometry analysis. First, exact molecular masses were derived from the spectra and were fitted to virtual insulin fragments. Predicted amino-acid sequences were confirmed by MS/MS analysis of the corresponding peptides. Next, a different set of fractions was analyzed after reduction with ß-mercapto-ethanol. Reduction resulted in the separation of fragments, linked by one of the three disulfide bridges present in the insulin molecule. As expected, molecular masses of the non-reduced fragments were exactly in accordance with one or the summation of two fragments in the reduced state. In Fig. 2b
, the generated insulin fragments are shown and indicated with their RP-HPLC fraction number. By indicating the amino acids following the cleavage sites (P1' residues) in green (panels b, c and d), the accessibility for gelatinase B can be demonstrated on the three-dimensional crystal structure of human insulin (Fig. 2c)
. With these data, we were able to elucidate previously undefined peaks from the mass spectra obtained after short incubation times or at low enzyme concentrations in the kinetic studies. Association of the appearance of specific peaks in function of incubation time is indicative of the relative efficiency of the respective cleavage sites. In Fig. 2d
, the 10 cleavage sites were aligned from the most to the least efficient cleavage. The P1' residue of the cleavage was six times a leucine. This clear preference for hydrophobic residues at P1' is in agreement with digestion of synthetic peptides and the previously published cleavage sites in denatured bovine collagen II. From our data, it is obvious that insulin is progressively degraded into fragments by gelatinase B, rather than processed by a unique clipping, as observed for IL-1ß, IL-8, and endothelin-1.
|
CONCLUSIONS AND SIGNIFICANCE
Gelatinase B expression has been functionally associated with various autoimmune diseases including multiple sclerosis and rheumatoid arthritis. For diabetes as an autoimmune disease, only circumstantial evidence and phenomenological data are available, but so far no functional link has been demonstrated between gelatinase B and insulin or any other islet antigen.
In this study, insulin was demonstrated to be a substrate for gelatinase B. Furthermore, insulin and gelatinase B producing cells are in close proximity in acute and chronic pancreatitis, as visualized by double immunostainings. The effects of the expression of gelatinase B, in particular the diabetogenic role, are exemplified in Fig. 3
. Whereas it is understandable that, by its virtue to cleave ECM components, gelatinase B may contribute to the aforementioned extensive tissue remodeling and fibrosis in pancreatitis, it may also exert other functions. For instance, gelatinase B has been shown to potentiate IL-1ß, the neutrophil chemokine IL-8 and endothelin-1 and it degrades other CXC chemokines. Here we demonstrate the processing of the classical polypeptide hormone insulin by gelatinase B.
|
Previously, a model how gelatinase B may play a role in autoimmunity was developed. Cytokine-regulated extracellular proteases (including gelatinase B) play a central role in this model by degrading secreted proteins into fragments that activate T lymphocytes. For instance, degradation of myelin basic protein and denatured type II collagen by gelatinase B effectively generates fragments corresponding to immunodominant peptides for multiple sclerosis and rheumatoid arthritis development, respectively. For type I diabetes, the neuroendocrine enzyme glutamic acid decarboxylase, tyrosine phosphatase and insulin are important autoantigens. For instance, the fragment of the insulin ß-chain, containing residues 9 to 23, constitutes an immunodominant epitope. From our data, it is clear that the splitting by gelatinase B in front of residue 24 is the most efficient of all 10 cleavages. This implies that such a remnant fragment, containing residues 1 to 23 of the ß chain, may be presented in MHC-II either after extracellular processing into the immunodominant epitope [residues 9 to 23] or after uptake by antigen-presenting cells. However, if the fragment 1 to 23 persists in the extracellular milieu for prolonged time intervals, then it will be further degraded by gelatinase B. In fact, human insulin is completely destroyed by gelatinase B.
Our study implies that MMP inhibitors may have a stabilizing effect on insulin levels and that in some cases of diabetes, insulin therapy may be assisted by MMP inhibition. Anyhow, insulin is an important and novel substrate of gelatinase B.
This article has been cited by other articles:
![]() |
H. Jiang, H. Zhu, X. Chen, Y. Peng, J. Wang, F. Liu, S. Shi, B. Fu, Y. Lu, Q. Hong, et al. TIMP-1 Transgenic Mice Recover From Diabetes Induced by Multiple Low-Dose Streptozotocin Diabetes, January 1, 2007; 56(1): 49 - 56. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |