Introduction
The matrix metalloproteinases are a family of neural endopeptidases which catabolise extracellular matrix components. Their activity is dependent upon calcium and zinc. They act by denaturing collagen in the extracellular matrix, via collagenase (gelatinase) activity, leading to the accumulation of extracellular collagen.Reference Egeblad and Werb1 Matrix metalloproteinases are released by stromal fibroblasts and inflammatory cells.Reference Yoon, Park, Yun and Chung2
Tissue inhibitor matrix metalloproteinase acts to inhibit tissue levels of matrix metalloproteinase. Concentrations of matrix metalloproteinase and tissue inhibitor matrix metalloproteinase are usually balanced to ensure a steady state. However, in some diseases (e.g. atherosclerosis and osteoarthritis) matrix metalloproteinase activity increases.Reference Doğu, Yıldıran, Güloğlu, Çipe, Yüksek and Babacan3 Pathogenetic similarities between atherosclerosis and tympanosclerosis have been described; the pathogenesis of both diseases appears to be related to matrix metalloproteinases.Reference Pirodda, Feri, Bruzzi, Marini and Faggioli4 We have previously described a relationship between tympanosclerosis and matrix metalloproteinase 2 (also known as collagenase A), matrix metalloproteinase 9 (also known as collagenase B) and tissue matrix metalloproteinase inhibitor 1.
Materials and methods
The study was approved by the İzmir Atatürk Training and Research Hospital ethics committee.
We included in the study 40 patients who had undergone surgery within the ENT department of İzmir Atatürk Training and Research Hospital between 2002 and 2007. Patients' ages ranged from 13 to 63 years. Twenty patients had uncomplicated chronic otitis media and 20 had tympanosclerosis.
We also included in the study 20 individuals with no history of otic complaints or systemic or infectious disease, as controls.
In the tympanosclerosis group, the male/female ratio was 14/6 and the mean age was 31.8 years (range 16–53). In the chronic otitis media group, the male/female ratio was 12/8 and the mean age 36.9 years (range 13–63). In the control group, the male/female ratio was 10/10 and the mean age 35 years (range 17–61).
Five millilitres of blood was taken from all subjects and stored at −20°C. Levels of matrix metalloproteinase 2, matrix metalloproteinase 9 and tissue inhibitor matrix metalloproteinase 1 were measured using the enzyme-linked immunosorbent assay method.
Matrix metalloproteinase 2
Serum matrix metalloproteinase 2 concentrations were measured using the Micro-Elisa micro enzyme-linked immunosorbent assay system with Invitrogen solid phase sandwich (Invitrogen, Camarilllo, California, USA). Results were expressed as ng/ml. Standards and samples were pipetted into the pre-prepared wells of the Micro-Elisa system. Any matrix metalloproteinase 2 was bound to immobilising antibodies present in the wells. Wells were washed and biotinylated anti human matrix metalloproteinase 2 antibodies were added. The streptavidin-peroxidase enzyme was then added; this bound to matrix metalloproteinase 2 antibody complexes. Unbound enzymes were removed, and a substrate tetramethylbenzidine solution (TMB) designed to generate a coloured product in the presence of streptavidin-peroxidase was added to the medium. The intensity of colour thus generated was directly proportional to the concentration of matrix metalloproteinase 2 in the sample.
A standard curve of matrix metalloproteinase 2 concentration versus colour intensity was calculated as follows. Standard human matrix metalloproteinase 2 (50 ng/ml), produced with standard diluent buffer, was used to prepare serial dilutions (50, 25, 12.5, 6.25, 3.13, 1.56 and 0.78 ng/ml). Following the process described above, colour intensities were measured and plotted against the known matrix metalloproteinase 2 concentration to generate a standard curve.
Serum samples were prepared by diluting 1:15 in standard diluent buffer and mixing well with a vortex mixer. Following the process described above, the colour intensity of each sample was measured. The matrix metalloproteinase 2 concentration was then estimated from the standard curve, and multiplied by 15 to obtain the actual serum concentration.
Matrix metalloproteinase 9
Serum matrix metalloproteinase 9 concentrations were measured using the RayBio® human matrix metalloproteinase 9 enzyme-linked immunosorbent assay system (RayBiotech, Norcross, Georgia, USA). Standard and samples were pipetted into the pre-prepared wells of the system. Any matrix metalloproteinase 9 was bound to the immobilising antibodies present in the wells. Wells were washed and biotinylated anti human matrix metalloproteinase 9 antibodies were added. Unbound biotinylated antibodies were removed by washing, and colour-generating substrate solution TMB (tetramethylbenzidine) was added. The colour intensity generated was proportional to the amount of bound matrix metalloproteinase 9 present. The intensity of generated colour was measured at 450 nm.
A standard curve of matrix metalloproteinase 9 concentration versus colour intensity was calculated using a standard solution of human matrix metalloproteinase 9 (6000 pg/ml) prepared as serial dilutions (6000, 2000, 666.7, 222.2, 74, 24.69, 8.23 and 0 pg/ml). The colour intensities generated by the various matrix metalloproteinase 9 concentrations were measured to generate a standard curve, as for matrix metalloproteinase 2.
The matrix metalloproteinase 9 concentrations in samples were then estimated by comparing each sample's colour intensity to the standard curve.
Tissue inhibitor matrix metalloproteinase 1
The concentration of serum tissue inhibitor matrix metalloproteinase 1 was measured using the RayBio human tissue inhibitor matrix metalloproteinase 1 enzyme-linked immunosorbent assay system. Standard and samples were pipetted into the pre-prepared wells of the system. Any tissue inhibitor matrix metalloproteinase 1 present was bound to the immobilising antibodies present in the wells. Wells were washed and biotinylated anti human tissue inhibitor matrix metalloproteinase 1 antibodies were added. Unbound biotinylated antibodies were removed by washing, and a substrate solution was added which generated colour in proportion to the amount of bound tissue inhibitor matrix metalloproteinase 1 present. The intensity of generated colour was measured at 450 nm.
A standard curve of tissue inhibitor matrix metalloproteinase 1 concentration versus colour intensity was calculated using a standard solution of human tissue inhibitor matrix metalloproteinase 1 (18000 pg/ml) prepared as serial dilutions (18000, 6000, 2000, 666.7, 222.2, 74.07, 24.69 and 0 pg/ml), as for matrix metalloproteinases 2 and 9.
The tissue inhibitor matrix metalloproteinase 1 concentrations in samples were then estimated by comparing each sample's colour intensity to the standard curve.
Statistical analysis
The Windows 16.0 version of the Statistical Package for the Social Sciences software program was used for statistical analysis. One-way analysis of variance was used to compare quantitative data. The post hoc Bonferroni statistical analysis method was used to determine the difference between the groups. A value of p < 0.05 was considered statistically significant.
Results and analysis
In the control group, the mean serum matrix metalloproteinase 2 concentration was 78 ng/ml, the mean matrix metalloproteinase 9 concentration 102 ng/ml and the mean tissue inhibitor matrix metalloproteinase 1 concentration 142 ng/ml.
In the chronic otitis media group, the mean serum matrix metalloproteinase 2 concentration was 98 ng/ml, the mean matrix metalloproteinase 9 concentration 109 ng/ml and the mean tissue inhibitor matrix metalloproteinase 1 concentration 149 ng/ml.
In the tympanosclerosis group, the mean serum matrix metalloproteinase 2 concentration was 135 ng/ml, the mean matrix metalloproteinase 9 concentration 182 ng/ml and the mean tissue inhibitor matrix metalloproteinase 1 concentration 132 ng/ml.
The serum matrix metalloproteinase 2 and 9 levels in the tympanosclerotic group were found to be statistically significantly greater than those in the chronic otitis media group and the control group. There was no statistically significant difference between the tissue inhibitor matrix metalloproteinase 1 levels of the three groups (Tables I and II). There was no statistically significant difference between the matrix metalloproteinase 2 and 9 levels of the control and chronic otitis media groups.
Table I Matrix metalloproteinase 2 and 9 and tissue inhibitor matrix metalloproteinase 1 concentrations in the three groups
Data represent ng/ml, except for p values. *Statistically significant at p < 0.05. SD = standard deviation; min = minimum; max = maximum; MMP2 = matrix metalloproteinase 2; COM = chronic otitis media; TS = tympanosclerosis; MMP9 = matrix metalloproteinase 9; TIMP1 = tissue inhibitor matrix metalloproteinase 1
Table II Matrix metalloproteinase 2 and 9 and tissue inhibitor matrix metalloproteinase 1 concentrations: comparison of group means
*Statistically significant at p<0.05. Grp = group; diff btw = difference between; TIMP1 = tissue inhibitor matrix metalloproteinase 1; COM = chronic otitis media; TS = tympanosclerosis; MMP2 = matrix metalloproteinase 2; MMP9 = matrix metalloproteinase 9
Discussion
Increased matrix metalloproteinase activation results in: the destruction of matrix components within the extracellular compartment; the migration and proliferation of vascular smooth muscle cells; and the growth and proliferation of tumour cells. Such increased matrix metalloproteinase activity is triggered by reactive oxygen radicals, via a reaction with thiol groups within the matrix metalloproteinase.Reference Van Wart and Birkadel-Hansen5 The enzymes xanthine oxidase and (Nadph) nicotinamidadenindinücleotidphosphate oxidase have also been implicated.Reference Valentin, Bueb, Kieffer, Tschirhart and Atkinson6 Recent studies have demonstrated a causative relationship between increased concentrations of free oxygen radicals within the middle ear and the development of myringosclerosis and tympanosclerosis.Reference Mattson, Magnuson and Hellström7
In the light of this evidence, the current study aimed to investigate matrix metalloproteinase levels in patients with tympanosclerosis.
Recent research has also demonstrated increased matrix metalloproteinase activity in patients with cholesteatoma and otitis media with effusion.Reference Jennings, Guo, Collins and Birchall8 Antonelli et al. have shown that matrix metalloproteinase activity is significantly inhibited by some protease inhibitors.Reference Antonelli, Schultz, Kim, Cantwell, Sundin and Pemberton9 It has been suggested that endogenous tissue inhibitors may potentiate this effect by chelating metal ions within the extracellular matrix.Reference Gomez, Alonso, Yoshiji and Thorgeirsson10 Moon et al. have suggested that the tympanic membrane changes seen in cases of otitis media with effusion may be related to matrix metalloproteinases; furthermore, these authors have suggested that such changes represent a complex process which also involves other proteases, otic pressure changes and extracellular mediators.Reference Moon, Linthicum, Yang, Lee and Park11 Moon et al. also analysed samples of effusion fluid taken from these same patients, and found high levels of matrix metalloproteinase 9 in mucous effusions and high levels of tissue inhibitor matrix metalloproteinase 2 in serous effusions.
In the current study, we found no statistically significant difference between the tissue inhibitor matrix metalloproteinase 1 levels of our patient and control groups.
Wilmoth et al. exposed the tympanic membranes of 48 Mongolian gerbils to tumour necrosis factor α and bacterial endotoxin ex vivo, and then analysed the tympanic membrane matrix metalloproteinase activity. They concluded that the accumulation of these mediators of inflammation within the tympanic membrane extracellular matrix led to increased destruction of the lamina propria, resulting in irreversible pathology (e.g. tympanic membrane retraction, atelectasia and cholesteatoma).Reference Wilmoth, Schultz and Antonelli12
High levels of matrix metalloproteinases and their inhibitors have been found in patients with chronic otitis media, nasal polyposis and Sjögren syndrome.Reference De, Fenton and Jones13 High levels of matrix metalloproteinase 2 and 9 have been found in patients with juvenile angiofibroma, a tumour rich in collagen.Reference Duerr, Wendler, Aigner, Karosi and Schick14
In patients with arteriosclerosis (one of the disease processes involving dystrophic calcification), matrix metalloproteinases 2 and 9 have been shown to have elastase activity which results in destruction of the lamellar elastin layer of the arterial wall.Reference Longo, Xiong, Grenier, Zhao, Fiotti and Baxter15 Other authors have shown that matrix metalloproteinase 9 causes arterial wall destruction and plaque formation, via increased elastase activity and increased collagen accumulation at the site of arterial wall destruction.Reference Ducharme, Frantz, Aikawa, Rabkin, Lindsey and Rohde16, Reference Luttun, Lutgens, Manderveld, Maris, Collen and Carmeliet17
We found high levels of matrix metalloproteinase 9 in our patients with tympanosclerosis, suggesting a similar involvement of this enzyme in the development of tympanosclerotic plaques.
Yasmin et al. found high levels of matrix metalloproteinases 2 and 9 in the serum of patients with isolated systolic hypertension due to atherosclerosis (mean matrix metalloproteinase 2 levels were 177 ng/ml in patients and 151 ng/ml in controls, while mean matrix metalloproteinase 9 levels were 157 ng/ml in patients and 119 ng/ml in controls).Reference Yasmin, McEniery, Wallace, Dakham, Pulsakar, Maki-Petaja, Ashby, Cockroft and Wilkinson18 In our study, the difference in matrix metalloproteinase levels between the patient groups and the control group was similarly statistically significant.
The relationship between tympanosclerosis and atherosclerosis has been investigated, since their pathogenesis is similar. Koç et al. found myringosclerosis in 66.6 per cent of 1024 atherosclerotic patients who underwent otoscopic inspection, and suggested that there may be a genetic predisposition linking both conditions.Reference Koç and Üneri19
Another disease process involving dystrophic calcification is osteoarthritis. Increased concentrations of free oxygen radicals and some cytokines have been found in patients with osteoarthritis, and steroid injections have been used as treatment.Reference Ravaud, Moulinier and Giraudeau20 Tetracyclines, which chelate calcium ions and inhibit tissue metalloproteinases, have also been used for osteoarthritis. Tetracyclines have previously been shown in vitro to be potent anticalcification agents in tissues with ectopic calcification.Reference Cohen, Solomon and Alferiev21 Doxycycline has been found to reduce the severity of osteoarthritis in dogs.Reference Brandt, Yu and Amith22 Doxycycline and tetracycline are used as antibiotics, but are also non-selective matrix metalloproteinase inhibitors. Axisa et al. showed that doxycycline accumulates in atherosclerotic plaques and prevents plaque progression and rupture by inhibiting matrix metalloproteinase 1 within the plaque.Reference Axisa, Loftus, Naylor, Goodall, Jones and Bell23 Ozcan et al. investigated acute otitis media in 25 adult guinea pigs after myringotomy; the animals' right ears were treated with topical doxycycline while their left ears were left as controls.Reference Ozcan, Selcuk, Ozcan, Akdogan, Giray and Dere24 At the end of the sixth week, histological evidence of myringosclerosis was found in significantly less of the ears treated with doxycycline. These findings support our own discovery of high levels of matrix metalloproteinases in patients with tympanosclerosis.
Topical applications of ascorbic acid, N-acetyl-cystein and vitamin E have all been used in an attempt to prevent myringosclerosis; all these agents are reported to act by reducing the concentration of reactive oxygen radicals.Reference Uneri, Bağlam and Yazıcı25 We can say this agents decrease the oxigen free radicals so matrix metalloproteinase levels and decrease myringosclerosis.
• Surgical treatment of tympanosclerosis has a poor success rate, as the primary pathological process is still active
• This study found significantly higher matrix metalloproteinase concentrations in tympanosclerotic patients, compared with normal controls and patients with chronic otitis media
• Matrix metalloproteinase inhibitors have begun to be used to restrict the spread of some cancer types, and may also be useful in the treatment of tympanosclerosis
Other methods of inhibiting matrix metalloproteinase activity have been reported, such as the use of angiotensin-converting enzyme inhibitors following acute myocardial infarction.Reference Papadopoulos, Economou, Makris, Kapetanios, Moyssakis and Votteas26 In this clinical context, these drugs are proposed to balance the concentrations of matrix metalloproteinase 1 and tissue inhibitor matrix metalloproteinase 1 (MMP1).Reference Papadopoulos, Economou, Makris, Kapetanios, Moyssakis and Votteas26
Since matrix metalloproteinases also promote tumour invasion and progression, it has been postulated that tissue matrix metalloproteinase inhibitors may be useful during the follow up of most types of cancer, to prevent metastases and relapses. High concentrations of matrix metalloproteinases have been found in many cancer types (e.g. prostate, pancreas, lung, breast and colon) and are considered to indicate a poor prognosis.Reference Murray, Duncan, O'Neil, Melvin and Fothergill27 Synthetic matrix metalloproteinase inhibitors have recently begun to be used in cancer treatment; preparations include Batimastat (British Biotec, Oxford, UK), which inhibits matrix metalloproteinases 1, 2, 3 and 9, and Marimastat, its second generation derivative.Reference Kallakury, Karikehalli, Haholu, Sheehan, Azumi and Ross28
We suggest that the use of matrix metalloproteinase inhibitors in atherosclerosis, osteoarthritis and tympanosclerosis may reduce matrix metalloproteinase concentrations and thereby retard and attenuate the pathological mechanism of disease.
Conclusion
The main treatment in tympanosclerosis is surgery. Because the process continues to be active. Therefore the medical treatment methods are still being investigated. We think that using synthetic matrix metalloproteinase inhibitors in the future will be efficient in such kinds of these patients.
Acknowledgement
This study was sponsored by the Turkish Ministry of Health. We thank Drs A E Ilknur and H E Gogus for their assistance with data collection.
