Role of melatonin in regulating matrix metalloproteinase-9via tissue inhibitors of metalloproteinase-1 during protectionagainst endometriosis

Introduction

Endometriosis is an invasive but benign gynecologicaldisorder of reproductive women and is characterized by the

growth of endometrial glands and stroma outside theuterine cavity [1]. It is estimated to affect 10–20% of womenof reproductive age and affects quality of life negatively as it

is associated with chronic pelvic pain, severe dysmenorrhea,dyspareunia and infertility [2–4]. The progression of endo-metriosis shows a strong correlation to estrogen levels and

cytokines [5–7]. The etiology and pathogenesis of endo-metriosis are still unclear. However, Sampson�s theory ofretrograde menstruation is the most accepted theory to

explain the disease, which arises from the ectopic implan-tation of endometrial tissue that reaches the peritoneum byretrograde menstruation followed by acquisition of a newblood supply through angiogenesis [8–10]. Thus, endo-

metriosis has the unique characteristics of invasion andrequires remodeling of the extracellular matrix (ECM).

Matrix metalloproteinases (MMPs) are a series of zinc-

requiring proteolytic enzymes that are secreted in latentpro-enzyme form and are involved in remodeling anddegradation of ECM [11–13]. Because endometrium is a

tissue that undergoes ECM remodeling periodically, preciseregulation of MMPs is essential in orchestrating proper

physiological functioning of the endometrium and hence,derangement of MMP regulation is considered to be acritical factor in the development of pathological conditions

like endometriosis [14, 15]. The active forms of MMPs areregulated by complexing stoichiometrically with tissueinhibitors of matrix-metalloproteinases (TIMPs) [16, 17].

Several diseases including rheumatoid arthritis, gastriculceration and endometriosis have been shown to beassociated with MMP and TIMP imbalance [18–20]. Pres-ence of active MMPs especially MMP-9 (92-kDa gelatinase

B) and MMP-3 (stromelysin-1) has been demonstrated inendometrial carcinoma and endometrial stromal tumors [21,22]. Very recently, Collette et al. [23] reported that eutopic

endometrium of women with endometriosis shows increasedactivity of MMP-9. A few studies also demonstratedelevated expression of MMP-3 in ectopic endometrial

tissues of rats surgically induced with endometriosis [24,25]. Additionally, the activity and expression of severalMMPs are regulated by different hormones [26, 27].

Matrix metalloproteinase activity can also be modulatedby oxidative stress [28, 29]. Reactive oxygen species (ROS)are capable of converting the latent pro-enzymes to activeforms by cleavage of thiol linkage [30–32]. Both MMP-2

and MMP-9 are activated by ROS, and their expressionsseem to be regulated by oxidant stress [30]. The role of ROS

Abstract: Endometriosis is a gynecological disease of women and plausibly

regulated by matrix metalloproteinases (MMPs). However, mechanisms of

alterations in MMPs during endometriosis remain unclear. Human

endometriotic tissues possessing varying degrees of severity were examined

for expression of MMPs and tissue inhibitors of metalloproteinase (TIMP)-1.

In addition, endometriosis was generated in mice and endometriotic tissues

were tested for MMP-9 activity. Results show significant upregulation of

secreted and synthesized proMMP-9 activity with duration and severity

of endometriosis. Along with upregulation of activity, the expression of

proMMP-9 was found increased while TIMP-1 expression followed an

inverse trend. The effect of melatonin, a major secretory product of the

pineal gland, on endometriosis was examined in preventive and

therapeutic models in mice. The results show that melatonin arrested

lipid peroxidation and protein oxidation and downregulated proMMP-9

activity and expression in a time and dose-dependent manner while

protecting and regressing peritoneal endometriosis. Moreover, the

attenuated activity and expression of proMMP-9 were associated with

subsequent elevation in the expression of TIMP-1. Our study reveals for the

first time the role of melatonin in arresting peritoneal endometriosis in mice

and a novel marker, expression ratio of proMMP-9 versus TIMP-1, was

identified for assessing severity and progression of endometriosis.

Sumit Paul1, Anamika VivekSharma1, Pramathes DasMahapatra2, ParthaBhattacharya2, Russel J. Reiter3

and Snehasikta Swarnakar1

1Department of Physiology, Indian Institute of

Chemical Biology; 2Spectrum Research Clinic,

Kolkata, India; 3Department of Cellular and

Structural Biology, University of Texas Health

Science Center, San Antonio, TX, USA

Key words: endometriosis, matrix

metalloproteinase, melatonin, oxidative

stress, severity index, tissue inhibitor of

metalloproteinase

Address reprint requests to Snehasikta

Swarnakar, PhD, Scientist and Head, Depart-

ment of Physiology, Indian Institute of Chemi-

cal Biology, 4, Raja S.C. Mullick Road,

Jadavpur, Kolkata 700032, India.

E-mail: snehasiktas@hotmail.com

Received September 28, 2007;

accepted November 27, 2007.

J. Pineal Res. 2008; 44:439–449Doi:10.1111/j.1600-079X.2007.00547.x

� 2008 The AuthorsJournal compilation � 2008 Blackwell Munksgaard

Journal of Pineal Research

439

in increasing growth and adhesion of endometrial cells inthe peritoneal cavity during endometriosis has been docu-mented [33, 34]. Administration of antioxidant enzymes like

superoxide dismutase and catalase prevented intraperito-neal (i.p.) adhesions of endometriotic tissues in the perito-neal cavity of rabbit [35].Melatonin (N-acetyl-5-methoxytryptamine) is the major

secretory product of the pineal gland and the function of itand its metabolites as antioxidants in protecting cellularcomponents such as nuclear DNA, membrane lipids,

cytosolic proteins from free-radical damage are well doc-umented [36–39]. Melatonin binding sites and melatoninreceptors in uterus and endometrial carcinoma cell lines

have been documented [40, 41]. It also stimulates secretionof progesterone and has oncostatic, antiproliferative andantiestrogenic effects [42–45]. Although, endometriosis likeendometrial cancer and breast cancer is known to be an

estrogen-dependent pathological condition and is associ-ated with oxidative stress, the antioxidant role of melatoninin healing of endometriosis, has not been reported.

We document here a novel role of melatonin in preven-tion and regression of peritoneal endometriosis in mice.Melatonin acts as antioxidant and protects endometrium

from oxidative stress as reflected by the inhibition of lipidperoxidation and protein oxidation during protectionagainst endometriosis. Moreover, it downregulates proM-

MP-9 and upregulates TIMP-1expression dose-dependentlyduring protection against endometriosis. The study unveilsthe oxidative stress response associated with endometriosisand the role of melatonin in the protection and regression

of peritoneal endometriosis in mice. We report here for thefirst time the upregulation of proMMP-9 activity andexpression that parallels with severity of endometriosis. The

other finding includes the robust increase in proMMP-9versus TIMP-1 expression ratio with the progression andseverity of the disease.

Materials and methods

Chemicals

Gelatin from porcine skin, Triton X-100, protease inhibi-tors mixture, and 5-bromo-4-chloro-3-indolyl phosphate/

nitro blue tetrazolium were obtained from Sigma, St. Louis,MO, USA. Human and mice reactive polyclonal anti-MMP-9, polyclonal anti-TIMP-1 and monoclonal anti-b

actin antibody were purchased from Santa Cruz Biotech-nology, Santa Cruz, CA, USA. All other chemicals werepurchased from a local company.

Collection of human samples

Patients with symptoms suggestive of endometriosis attend-

ing the gynecology unit of Spectrum Research Clinic,Kolkata, India were included in the study. Endometriosiswas identified during laparoscopy in women consulting

for infertility and/or pelvic pain. Normal women exhibitedno visible evidence of endometriosis upon laparoscopy.Endometriotic tissues were collected after approval of the

Human Ethics Committee and consent from patients. Adetailed demographic profile of the study population isgiven in Table 1. Briefly, we collected three categories ofendometriotic samples, e.g., severe endometriosis (stage

III–IV), moderate endometriosis (stage II) and mild endo-metriosis (stage I). Eutopic endometrium from normalwomen was used as control. The stages of endometriosis

were indexed according to the score of revised AmericanFertility Society (rAFS) [46]. After collection, all biopsieswere stored at )70�C for future experiments.

Animals used

Female BALB/c mice of 6–8 wk old, bred in house with freeaccess to food and water were used in all experiments.Experiments were designed to minimize animal sufferingand to use the minimum number associated with valid

statistical evaluation. Animals were anesthetized by keta-mine (12 mg/kg b.w.) and killed by cervical dislocation.Animal (n = 79) experiments were carried out following

the guidelines of the animal ethics committee of theinstitute. Animals of both control and experimental groupswere kept separately under standard controlled conditions.

Induction of peritoneal endometriosis in BALB/cmice and its protection by melatonin

Induction of peritoneal endometriosis was performed usinga modified method of the Somigliana et al. [47]. Briefly, onday 0 the donor mice were anesthetized by ketamine

(12 mg/kg b.w.) and killed by cervical dislocation. Using asterile technique, the uterine horns were removed andplaced in a petri-dish containing phosphate buffered saline

Table 1. Demographic profile of study population

Characteristics Normal (n = 14) Fibroid (n = 10) Severe Endo (n = 28) Moderate Endo (n = 24) Mild Endo (n = 10)

Age (mean) Range 32.9 ± 4.8 (19–45) 43 ± 5.2 (27–45) 30.12 ± 4.3 (17–45) 28.4 ± 3.9 (26–41) 33.42 ± 4.44 (22–42)Fertile 6 3 4 20 4DysmenorrheaPresent 4 2 19 12 3Absent 10 8 9 12 7

CPPPresent 1 1 21 5 2Absent 13 9 7 19 8

Endometriotic tissues on the basis of severity (as standardized by rAFS) were collected (n = 86) and grouped as severe, moderate and mild.Age, fertility status, dysmenorrhoea and chronic pelvic pain (CPP) were recorded for the study population. All individuals of the studypopulation were in proliferative phase of the menstrual cycle.

Paul et al.

440

(PBS). After removing the muscle layers, the endometriumwas finely chopped using a sharp razor blade. Endometrialfragments thus obtained were suspended in 0.6 mL of

sterile PBS and inoculated into the peritoneal cavity ofrecipient mice with a ratio of one donor to two recipients.Mice inoculated with sterile PBS containing no endometrialfragments were kept in animal house and were used as

control. Mice were killed on 7th (Endo 7), 15th (Endo 15)and 21st day (Endo 21) of postinduction of endometriosis.Uterine tissue of control mice killed on the day of

endometriosis induction (Endo 0) was used as control.Different doses of melatonin in 15% ethanol, were admin-istered i.p. 30 min prior to inoculation of endometrial

extract and once daily for the next 3 days to test theirprotective effects in endometriosis (n = 39).

Therapeutic model of peritoneal endometriosis inmice

Peritoneal endometriosis was induced in mice as described

above. Mice were induced with peritoneal endometriosisand were administered with melatonin (48 mg/kg b.w.) aswell as vehicle intraperitoneally, once daily for the

following 10 and 20 days from the 15th day of posten-dometriosis induction. Regression of endometriotic lesionsin melatonin-treated and vehicle-treated mice was moni-

tored after killing them on the 10th and 20th day(n = 40).

Tissue extraction

Human biopsy samples and mouse endometriotic tissueswere suspended in PBS containing protease inhibitors,

minced and incubated for 10 min at 4�C [48]. The suspen-sion was centrifuged at 12,000 g for 15 min, and superna-tant was collected as PBS extracts. The pellet was further,

extracted in lysis buffer (10 mm Tris–HCl, pH 8.0, 150 mm

NaCl, 1% Triton X-100, and protease inhibitors) andcentrifuged at 12,000 g for 15 min to obtain Triton X-100(TX) extracts. Both PBS and TX extracts were preserved

at )70�C for future studies.

Measurement of lipid peroxidation

The cytosolic and microsomal fractions were collectedfrom tissue homogenate obtained from both human and

peritoneal endometriosis in mice. These were used formeasurement of lipid peroxide content as thiobarbituricacid-reactive species (TBARS) [48]. Briefly, 1 mL of the

cytosolic fraction and 1 mL of microsomal fraction wereallowed to react with 2 mL of TCA–TBA–HCl (15% TCA,0.375% TBA, 0.25 N HCl) reagent, heated in a boilingwater-bath for 15 min, cooled and centrifuged. The absor-

bance of the supernatant was measured for nanomoles ofTBARS at 535 nm (e =1.5 · 105 m/cm).

Measurement of protein carbonyl content

Protein oxidation was measured as carbonyl content in the

low speed supernatant of the endometriotic tissues of humanand mice [48]. The tissues were homogenized in 50 mm

sodium phosphate buffer, pH 7.4, in a Potter–Elvehjem glasshomogenizer for 2 min to get 20% homogenate. Aftercentrifugation at 600 g for 10 min, the proteins from

1.0 mL of the supernatant were precipitated with 10%trichloroacetic acid and allowed to react with 0.5 mL of10 mm 2, 4-dinitrophenylhydrazine for 1 hr. After precipi-tation with 20% trichloroacetic acid, the protein was washed

thrice with amixture of ethanol–ethyl acetate (1:1), dissolvedin 1.0 mL of a solution containing 6 m guanidine HCl in20 mm potassium phosphate adjusted to pH 2.3 with

trifluroacetic acid, and centrifuged, and the supernatantwas read for carbonyl content at 362 nm (e = 22,000 m/cm).

Gelatin zymography

For assay of MMP-9 activity, tissue extracts were electro-phoresed in a 8% SDS-polyacrylamide gel containing

1 mg/mL gelatin, under nonreducing conditions [48]. Forhuman tissues 30 lg and for mice tissues 70 lg protein wereloaded equally in all lanes. The gels were washed twice in

2.5% Triton X-100 and then incubated in calcium assaybuffer (40 mm Tris–HCl, pH 7.4, 0.2 m NaCl, 10 mm

CaCl2) for 18 hr at 37�C. Gels were stained with 0.1%

Coomassie blue followed by destaining. The zones ofgelatinolytic activities appeared as negative staining. Quan-tification of zymographic bands was performed using

densitometry linked to proper software (Lab Image,Kapelan Gmbh, Germany).

Western blotting

Tissue extracts (50 lg/lane) were resolved by 8% reducingSDS-polyacrylamide gel electrophoresis and transferred to

nitrocellulosemembranes [48]. Themembranes were blockedfor 2 hr at room temperature in 3% BSA solution in 20 mm

Tris–HCl, pH 7.4 containing 150 mm NaCl and 0.02%

Tween 20 (TBST) followed by overnight incubation at 4�C in1:200 dilution of the respective primary antibodies in TBSTcontaining 0.2% BSA. The membranes were washed fivetimes with TBST and then incubated with alkaline phos-

phatase-conjugated secondary antibody (1:2000) for 1.5 hr.The bands were visualized using 5-bromo-4-chloro-3-indolylphosphate/nitro blue tetrazolium substrate solution. The

blots shown in this study are representative replicatesselected from at least three experiments.

Statistical analysis

Data for the activity and expression of MMP-9 and

expression of TIMP-1 were fitted using Sigma plot. Datawere represented as the mean values ± S.E.M. P < 0.05was accepted as level of significance. The statistical analysisof the data was carried out using GraphPad Instat 3

software (San Diego, CA, USA). Comparison betweengroups was carried out using one-way analysis of variance(ANOVA) followed by Student–Newman–Keuls test.

Results

To test whether peritoneal endometriosis in mice is asso-ciated with MMP-9 activity at the level of secretion and

Arresting endometriosis by melatonin

441

synthesis, we performed gelatin zymography using PBS andTX extracts of endometriotic tissues, respectively. We firstexamined the time dependent regulation of MMP-9 in both

PBS and TX extracts (Fig. 1A,B). It was seen that fromseventh day onwards activity of secretory proMMP-9gradually increased from approximately fivefold andattained approximately ninefold increment on the 21st

day of postendometriosis induction as compared to control(Fig. 1C). Similarly Fig. 1D shows that synthesized proM-MP-9 activity was elevated by �10-fold on the 15th day

which further increased to �14-fold on 21st day ascompared to control. Our results show a strong correlationbetween the duration of disease and the activity of

proMMP-9 at the level of both synthesis and secretion.The activities of MMPs are regulated by TIMPs in varioussystems. To address the question whether there is a directrelationship between secreted MMP-9 activity and the

expression of TIMP-1, PBS extracts were subjected toWestern blot analysis (Fig. 2A). It was interesting toobserve that TIMP-1 expression decreased with time

(Fig. 2A,B) in mouse endometriotic tissues. Thus, it isinteresting to note that when we plotted the expression ratioof proMMP-9 and TIMP-1 in a histogram (Fig. 2C) it

increased parallely with duration of disease states andelevated maximally to �20-fold on the 21st day.To compare the activities of MMP-9 in eutopic endo-

metrium of women with and without endometriosis, thePBS tissue extracts were subjected to gelatin zymography(Fig. 3A). To find out the molecular markers of invasive-ness associated with endometriosis we have also used a

nonendometriotic fibroid tissue. Our results show a mod-erate (�2.5-fold) upregulation in the activity of secretedproMMP-9 in eutopic endometrium of diseased women as

compared to fibroid as well as normal (Fig. 3B). Thus,eutopic endometrium from patients with endometriosis

showed higher gelatinolytic activity as compared to controlgroups at the level of secretion. We next compared theactivity of proMMP-9 in endometriotic tissues possessing

varying degrees of severity indices as standardized by AFS(Table 1). Fig. 4A,B show a gradual increase in the activityof proMMP-9 with the increasing severity of endometriosis.From Fig. 4C,D it is revealed that the activity of secreted

and synthesized proMMP-9 were gradually elevated frommild to severe endometriosis with maximum upregulationof �8.5-fold in each case as compared to normal eutopic

endometrium. Thus, our results demonstrate a severtitydependent upregulation in the activity of proMMP-9. Afollow-up experiment was also performed for monitoring

MMP-9 and TIMP-1 expression in human endometrioticbiopsies. We found that proMMP-9 expression wasincreased with increased severity of endometriosis alongwith decreased TIMP-1 expression (Fig. 5A). Histographic

analysis revealed that the expression of TIMP-1 followed adistinct inverse ratio with proMMP-9 (Fig. 5B). Fig. 5Cshows that there was a sharp (�40-fold) elevation of

proMMP-9 versus TIMP-1 ratio in severe endometriosis ascompared to that of normal eutopic endometrium. Inaddition, a �10-fold elevation in moderate and an approx-

imately fivefold elevation in mild endometriosis were alsoobserved in the expression ratio of proMMP-9 versusTIMP-1. We can directly correlate the ratios of proMMP-9

to TIMP-1 with rAFS score as well as the severity of thedisease from Table 2 and it was seen that the ratio ofproMMP-9 to TIMP-1 was increased �43-fold in severe(rAFS; 45 ± 6.98), �12-fold in moderate (rAFS;

12 ± 5.87) and approximately fivefold in mild (rAFS;8 ± 5.62) endometriosis.To examine the role of oxidative stress in the patho-

physiology of endometriosis, experiments were performedto measure protein oxidation and lipid peroxidation in

Fig. 1. Time-dependent increase in secretory and synthesized proMMP-9 activity in endometriotic tissues of mice. Gelatin zymography asdescribed in �Materials and methods� was performed to monitor the secretory (A) and synthesized (B) activity of proMMP-9 using equalamounts (70 lg) of PBS and TX extracts. Histographic representation of secretory (C) and (D) synthesized proMMP-9 activity against timeas measured by Lab Image software from the above zymograms and three other representative zymograms from independent experiments ineach case. Values are ±S.E.M. of the above (A) and (B) gelatin zymogram and three other representative gelatin zymogram fromindependent experiments. Sample number n = 39. *P < 0.001 versus the appropriate control using ANOVA followed by Student–New-man–Keuls test.

Paul et al.

442

control, endometriotic and melatonin-pretreated endo-metriotic mouse tissues. Protein carbonylation and lipidperoxidation were significantly increased in edometriotic

tissues of mouse with endometriosis as compared withcontrol one. Increased protein oxidation and lipid peroxi-dation were also observed in human endometriotic tissueswith severe and moderate syndrome (Table 3). Table 4

shows the antioxidant activity of melatonin in protectingperitoneal endometriosis in mice through inhibition ofprotein oxidation by �80% and lipid peroxidation by

�90%. To investigate the role of melatonin on proMMP-9activity and expression during protection of endometriosiswe first designed an experiment to test the efficacy of

varying doses of melatonin in a preventive model ofperitoneal endometriosis in mice. To test the effectof melatonin on secretory and synthesized MMP-9, we

performed gelatin zymography using PBS and TX extractsof endometriotic tissues from melatonin-pretreated andvehicle-treated mice that were induced with peritonealendometriosis. Fig. 6A,B show that there was significant

reduction in the activity of both secretory and synthesizedproMMP-9 with increased doses of melatonin during theprevention of peritoneal endometriosis in mice. Melatonin

administration with the highest i.p. dose of 48 mg/kg b.w.successfully prevented peritoneal endometriosis throughdownregulation (�90%) of both secreted and synthesized

proMMP-9 activity (Fig. 6C,D). To test whether melato-nin has any effect on the expression of MMP-9, we usedPBS extracts for Western blot analysis (Fig. 6E). Ourresults clearly provide evidence that melatonin pretreat-

ment blocked the expression of proMMP-9 by �90%(Fig. 6F) compared to 15th day postendometriosis mice.As MMP-9 activity is coupled to TIMP-1 expression, we

investigated the role of melatonin in the expression ofTIMP-1. We noted that melatonin stimulated the expres-sion of TIMP-1 by approximately twofold as compared to

nontreated ones (Fig. 6E,F). Western blot of b-actin was

Fig. 2. Role of TIMP-1 on MMP-9 upregulation during induction of peritoneal endometriosis in mice. PBS extracts of different tissues weresubjected to Western blot as described in �Materials and methods� and probed with polyclonal anti-MMP-9, polyclonal anti-TIMP-1 andmonoclonal anti-b actin antibody. (A) Representative western blot are shown in all cases. Histographic representation of (B) fold changes atprotein level for proMMP-9 and TIMP-1 and (C) protein expression ratio of proMMP-9 to TIMP-1 versus time as measured by Lab Imagesoftware from the above blots and three other representatives blots from independent experiments in each. Sample number, n = 39. Errorbars = ±S.E.M. *P < 0.001, **P < 0.01 versus the appropriate control using ANOVA followed by Student–Newman–Keuls test.

Fig. 3. Comparison of proMMP-9 activity in eutopic endometriumof women with and without endometriosis. Eutopic endometriumobtained from women with (n = 28) and without endometriosis(n = 14) as well as women with fibroid (n = 10) were extracted inPBS. ProMMP-9 activity was detected in (A) gelatin zymogram asdescribed in text under �Materials and methods�. (B) Densitometricanalysis of proMMP-9 lysis bands following zymography. Valuesare ±S.E.M. of the above gelatin zymogram and three otherrepresentative zymograms from independent experiments.*P < 0.001, and NS, nonsignificant versus the appropriate controlusing ANOVA followed by Student–Newman–Keuls test.

Arresting endometriosis by melatonin

443

performed to confirm equal protein loading in the aboveblots. Indeed, melatonin reverted back to control level the

dysregulated expression ratio of proMMP-9 versus TIMP-1 and inhibited disease progression(Fig. 6G).

To investigate whether melatonin is effective in regressionof peritoneal endometriosis, we designed a therapeutic

model as described in �Materials and methods� and assayedthe time dependent effect of melatonin on proMMP-9

Fig. 4. Severity dependent upregulation in secretory and synthesized proMMP-9 activity in endometriotic tissues of women with endo-metriosis as compared to normal endometrium. To detect the activity of secreted and synthesized proMMP-9, biopsies were extracted inPBS and TX buffer and gelatin zymography was performed as described earlier. Gelatin zymogram shows the activities of (A) secretory and(B) synthesized proMMP-9 in different severities of endometriosis. Densitometric analysis of (C) secreted and (D) synthesized proMMP-9lysis bands following zymography. Activities were measured by using Lab Image densitometry program. Values are ±S.E.M. of the above(A) and (B) gelatin zymogram and three other representative gelatin zymogram from independent experiments. Sample number, n = 86.*P < 0.001 versus the appropriate control using ANOVA followed by Student–Newman–Keuls test.

Fig. 5. Severity dependent elevation of proMMP-9 and TIMP-1 expression ratio in women with endometriosis as compared to normalendometrium. PBS extracts of varying severity of endometriotic tissues were subjected to Western blot as described in �Materials andmethods� and probed with polyclonal anti-MMP-9, polyclonal anti-TIMP-1 and monoclonal anti-b actin antibody. (A) Representativewestern blot are shown in all cases. (B) Histographic representation of fold changes at protein level for proMMP-9 and TIMP-1 and (C)Protein expression ratio of proMMP-9 to TIMP-1 as measured by Lab Image densitometry values from the above blots and two otherrepresentative blots from independent experiments in each case. Sample number, n = 86. Error bars = ±S.E.M. *P < 0.001 versus theappropriate control using ANOVA followed by Student–Newman–Keuls test.

Paul et al.

444

activity. Fifteen days postendometriosis, the mice wereintraperitoneally administered with melatonin once daily

for the following 20 days and were killed on day 10 and day20. It is evident that melatonin arrested the increasedactivity of both secreted and synthesized proMMP-9 with

time (Fig. 7A,B). Fig. 7C,D show that melatonin progres-sively inhibited the activity of secreted as well as synthesizedproMMP-9 by �45% and �90% on the 25th and 35th day

of postendometriosis as compared to vehicle-treated oneson the 25th and 35th day, respectively.

Discussion

Because accumulating evidence supports the role ofenhanced MMP activity in endometriosis, we evaluated

the gelatinolytic activity in both peritoneal endometriosisinduced in mice as well as in human endometriotic tissues.In the present study, proMMP-9 induction was observed in

both mice and human endometriotic tissues but not innormal endometrium. Moreover, the eutopic endometriumof women suffering from endometriosis exhibited 2.5-fold

upregulation in the activity of proMMP-9 as compared toendometrium of control group. Our choice of fibroid tissueas a reference for endometriotic samples is to compare aninvasive disease (endometriosis) of the uterus to noninva-

sive and nonendometriotiotic ones, e.g., fibroid. Since wehave found almost no activity and expression of proMMP-9 in fibroid that implies the involvement of this molecule in

invasiveness procedure associated with endometriosis.MMP-9 is an important enzyme for the degradation ofthe basement membrane, primarily collagen type IV, and

plays crucial role in rheumatoid arthritis, embryogenesisand in cancer metastasis [18, 49, 50]. Therefore, theobserved increase in proMMP-9 activity and expressionmay have important consequences for the development of

vascular complications associated with endometriosis. Forinstance, Collette et al. [51] demonstrated that an elevatedMMP-9 activity is responsible for increase in the proteolytic

activity and plays an important role in vascular remodelingduring tissue invasion at ectopic sites in endometriosis.Recent studies have also established that MMP activity is

required for angiogenesis [52–55]. Therefore, enhancedMMP-9 activity may be critical for neovascularizationand proliferation of endothelial cells as well as continuous

growth and stabilization of endometriotic lesions at ectopicsites [52–55]. It is interesting to note that eutopic endo-metrium of diseased group in human biopsies showssignificantly high proMMP-9 activity as compared to

Table 2. Correlation of MMP-9/TIMP-1expression ratio to rAFS score Samples Eutopic Endo Severe Endo Moderate Endo Mild Endo

rAFS (Mean) 0 45 ± 6.98 12 ± 5.87 8 ± 5.62Histopathology Yes Yes Yes YesproMMP-9/TIMP-1 1 ± 0.23 42.91 ± 4.89* 11.92 ± 3.46* 5.34 ± 2.65*

Endometriotic samples having varying severity according to the revised score of AmericanFertility Society (rAFS) were characterized on the basis of their protein expression ratio ofMMP-9 to TIMP-1. Histopathology was performed to indicate the severity of endometri-osis. Results are reported as the mean values ± S.E.M. *P < 0.001 versus the appropriatecontrol using one-way analysis of variance (ANOVA) followed by Student–Newman–Keulstest.

Table 3. Estimation of protein oxidation and lipid peroxidation invarious tissues

SamplesProtein carbonylnmol/mg protein

Lipid peroxidation (nmolTBARS/mg protein)

Cytosolic Microsomal

Eutopic Endo(normal)

1.2 ± 0.06 1.22 ± 0.28 1.32 ± 0.13

Fibroid 1.27 ± 0.06 NS 1.29 ± 0.56 NS 1.38 ± 0.87 NSEutopic Endo(diseased)

1.23 ± 0.04 NS 1.28 ± 0.42 NS 1.42 ± 0.29 NS

Ectopic Endo(severe)

1.96 ± 0.08* 2.29 ± 0.26* 2.55 ± 0.39*

Ectopic Endo(moderate)

1.28 ± 0.06 NS 2.23 ± 0.31* 2.42 ± 0.44*

Ectopic Endo(mild)

1.22 ± 0.05 NS 1.32 ± 0.39 NS 1.44 ± 0.23 NS

Lipid peroxidation, and protein oxidation were measured in severe(n = 15), moderate (n = 15) and mild Endo (n = 10) as describedin the text. TBARS, thiobarbituric acid reactive substance. Resultsare reported as the mean values ± S.E.M. *P < 0.001 and NS,nonsignificant versus the appropriate control using ANOVA fol-lowed by Student–Newman–Keuls test.

Table 4. Effect of melatonin on protein oxidation and lipid per-oxidation in endometriotic tissues

SamplesProtein carbonyl(nmol/mg protein)

Lipid peroxidation (nmolTBARS/mg protein)

Cytosolic Microsomal

Endo 0 0.19 ± 0.02 1.25 ± 0.28 1.46 ± 0.29Endo 15 0.29 ± 0.08* 2.25 ± 0.34* 2.88 ± 0.47*Endo 15 +melatonin(48 mg/kg b.w.)

0.21 ± 0.03NS (80%)

1.34 ± 0.32NS (�90%)

1.61 ± 0.35NS (�90%)

Peritoneal endometriosis was induced in mice and protectionexperiment was carried out using melatonin (48 mg/kg b.w.) asdescribed in �Materials and methods�. Mice were killed on the 15thday postendometriosis induction. Lipid peroxidation (n = 14),protein carbonyl content (n = 14) from tissue homogenates asdescribed in the text. In the last row of each column, the valuegiven in parentheses represents the percentage reduction of lipidperoxidation and protein oxidation by melatonin. The percentagereduction was measured using the following formula: [(value ofEndo 15 ) value of melatonin)/value of Endo 15 ) value of con-trol)·100]. Results are reported as the mean ± S.E.M. and*P < 0.001 and NS, nonsignificant versus the appropriate controlusing ANOVA followed by Student–Newman–Keuls test.

Arresting endometriosis by melatonin

445

fibroid patient or normal women, suggesting endometrium

of women with endometriosis is prone to peritonealimplantation. We next sought to determine the involvementof TIMPs for the enhanced proMMP-9 activity and

expression in endometriotic tissues.The proteolytic activities of MMPs are tightly controlled

during activation from their proenzymes to active forms bythe combination of endogenous activators (e.g., membrane-

type MMPs and urokinase plasminogen activators) andinhibitors (e.g., a2-macroglobulins and tissue inhibitors ofmetalloproteinases) [16, 17]. In the present study, the

upregulation of proMMP-9 was found to be directly relatedto severity of endometriosis, which follows a paralleldownregulation of TIMP-1 expression. It is documented

that TIMP-1 inhibits the majority of the MMPs includingMMP-9 and MMP-3 [16, 17]. In this context, the presentstudy supports the regulatory role of TIMPs on MMPs asthe data show that the MMP-9/TIMP-1 expression ratio

was far above the value of unity that is required for properphysiological functioning. Thus, the enhanced expression

ratio of MMP-9/TIMP-1 is indicative of a novel marker to

monitor the progression of endometriosis. Although, thedownregulation of TIMP-1 expression might partiallyaccount for upregulation in the activity of MMP-9, there

are several other potential sources, which can upregulateMMP-9 expression or activity. Moreover, MMP regulationoccurs at the level of gene transcription and chemical orenzymatic activation of proMMPs [13]. Various stimuli

including oxidative stress, hormones induce MMP geneexpression [26, 27, 30]. The MMP-9 promoter regioncontains nuclear factor-jB, activator protein-1, stimulatory

protein-1, and phorbol ester-responsive elements [56, 57].Previous findings indicate that nuclear factor-jB andactivator protein-1 are redox sensitive and offer a potential

mechanism by which oxidative stress may regulate MMP-9transcription and activity [58].Although, only a few studies have found free radical

generation during endometriosis [59–62], the present study

provides further evidence that oxidative stress has a positivecorrelation with the severity of endometriosis. Endometrial

Fig. 6. Inhibition of MMP-9 activity andexpression and elevation of TIMP-1expression by melatonin pretreatmentduring blocking of peritoneal endometri-osis induced in mice. Gelatin zymographywas conducted to monitor the activity ofsecretory (A) and synthesized (B) proM-MP-9 activity, as mentioned in �Materialsand methods� using PBS and TX extractsof endometriotic tissues from melatonin-pretreated endometriosis induced group ofmice. Histographic representation ofactivity values of secreted (C) and (D)synthesized proMMP-9 against melatonindoses. PBS extracts of melatonin-pre-treated endometriotic tissues from micewere subjected to Western blot andprobed with polyclonal anti-MMP-9,polyclonal anti-TIMP-1 and monoclonalanti-b actin antibody. (E) Representativewestern blot are shown in all cases.Histographic representation of (F) foldchanges at protein level for proMMP-9and TIMP-1 and (G) expression ratio ofproMMP-9 and TIMP-1 after melatoninpretreatment as measured by Lab Imagesoftware from the above blots and threeother representatives blots from indepen-dent experiments in each case. Errorbars = ±S.E.M. Sample number,n = 39. *P < 0.001 versus Endo 0 andEndo 15 and NS, nonsignificant versusEndo 0 using ANOVA followed by Stu-dent–Newman–Keuls test.

Paul et al.

446

lesions that mainly consist of apoptotic endometrial tissuesand damaged erythrocytes cause ROS production inendometriosis patients either due to free iron overload orpartial reduction of molecular oxygen by the auto oxidation

of flavoproteins [32–34]. We found significant increase inlipid peroxidation and protein oxidation in mouse endo-metriotic tissues as well as in human endometriotic biopsies

especially in severe endometriosis. In the present study,melatonin protected against oxidative stress through thereduction of protein carbonyl formation and lipid perox-

idation. We postulate that besides the antioxidant role ofmelatonin it seems to serve as an antiestrogenic agent inregression of endometriotic lesions in mice. It has already

been demonstrated that melatonin can inhibit estrogenreceptor a via calmodulin and has oncostatic and antipro-liferative role on endocrine responsive neoplasms [44].Additionally, melatonin can regulate MMP expression via

regulation of progesterone secretion [26, 42].In the present study, we explored the role of melatonin in

the protection and regression of peritoneal endometriosis in

mice. Our results show that melatonin downregulates theactivity and expression of proMMP-9 in a time and dosedependent manner. The downregulation in the activity and

expression of proMMP-9 is associated with an increase ofTIMP-1 expression. In conclusion, our study demonstratesthe gradual upregulation of proMMP-9 activity andexpression in peritoneal endometriosis of mice in a time

dependent manner and identifies a novel diagnostic marker,MMP-9/TIMP-1 expression ratio in judging disease pro-gression and severity. This study is the first report to show

the role of melatonin in prevention and regression of

endometriotic lesions and opens up future avenues to unveilthe antiestrogenic property of melatonin via regulation ofMMPs.

Acknowledgements

This study was supported by grant MLP13 and IAP 001 of

Council of Scientific and Industrial Research and GAP209of Indian Council of Medical Research, New Delhi, India.SP and AVS are recipient of senior research fellowship and

research associateship respectively from Council of Scien-tific and Industrial Research, India.

References

1. Giudice LC, Tazuke SI, Swiersz L. Status of current

research on endometriosis. J Reprod Med 1998; 43:252–262.

2. Wheeler JM. Epidemiology of endometriosis-associated

infertility. J Reprod Med 1989; 34:41–46.

3. Adamson GD, Nelson HP. Surgical treatment of endome-

triosis. Obstet Gynecol Clin North Am 1997; 24:375–409.

4. Kettel LM, Hummel WP. Modern medical management of

endometriosis. Obstet Gynecol Clin North Am 1997; 24:361–

373.

5. Bulun SE, Zeitoun K, Takayama K et al. Estrogen pro-

duction in endometriosis and use of aromatase inhibitors to

treat endometriosis. Endocr Relat Cancer 1999; 6:293–301.

6. Akoum A, Lemay A, Paradis I et al. Secretion of interleukin-

6 by human endometriotic cells and regulation by proinflam-

matory cytokines and sex steroids. Hum Reprod 1996;

11:2269–2275.

Fig. 7. Time dependent downregulation of secreted and synthesized proMMP-9 activity by melatonin during healing of peritoneal endo-metriosis in mice. For therapeutic effects of melatonin the animals were administered intraperitoneally with melatonin (48 mg/kg b.w.) 15thday of postendometriosis for another extra 10 and 20 days. The animals were killed on day 10 and 20 postmelatonin or vehicle treatmentand gelatin zymography was performed as described in �Materials and methods� to detect the activity of secretory (A) and synthesized (B)proMMP-9. Histographic representation of secreted (C) and synthesized (D) proMMP-9 activity against time for therapeutic potential ofmelatonin, as measured by Lab Image software from the above zymograms and three other representative zymograms from independentexperiments in each case. Sample number n = 40. Error bars = ±S.E.M. *P < 0.001 and NS, nonsignificant versus Endo 0 usingANOVA followed by Student–Newman–Keuls test.

Arresting endometriosis by melatonin

447

7. Tseng JF, Ryan IP, Milam TD et al. Interleukin-6 secretion

in vitro is up-regulated in ectopic and eutopic endometrial

stromal cells from women with endometriosis. J Clin Endo-

crinol Metab 1996; 81:1118–1122.

8. Sampson JA. Peritoneal endometriosis due to menstrual dis-

semination of endometrial tissue into the peritoneal cavity. Am

J Obstet Gynecol 1927; 14:422–469.

9. Lebovic DI, Bentzien F, Chao VA et al. Induction of an

angiogenic phenotype in endometriotic stromal cell cultures by

interleukin-1beta. Mol Hum Reprod 2000; 6:269–275.

10. Rogers PA, Abberton KM, Susil B. Endothelial cell

migratory signal produced by human endometrium during the

menstrual cycle. Hum Reprod 1992; 7:1061–1066.

11. Matrisian LM. Metalloproteinases and their inhibitors in

matrix remodeling. Trends Genet 1990; 6:121–125.

12. Birkedal-Hansen H, Moore WG, Bodden MK et al. Matrix

metalloproteinases: a review. Crit Rev Oral Biol Med 1993;

4:197–250.

13. Nagase H. Activation mechanisms of matrix metalloprotein-

ases. Biol Chem 1997; 378:151–160.

14. Hulboy DL, Rudolph LA, Matrisian LM. Matrix metallo-

proteinases as mediators of reproductive function. Mol Hum

Reprod 1997; 3:27–45.

15. Sillem M, Prifti S, Neher M et al. Extracellular matrix

remodelling in the endometrium and its possible relevance to

the pathogenesis of endometriosis. Hum Reprod Update 1998;

4:730–735.

16. Woessner JF Jr. Matrix metalloproteinases and their inhibi-

tors in connective tissue remodeling. FASEB J 1991; 5:2145–

2154.

17. Murphy G. Matrix metalloproteinases and their inhibitors.

Acta Orthop Scand Suppl 1995; 266:55–60.

18. Martel-Pelletier J, Mccollum R, Fujimoto N et al.

Excess of metalloproteases over tissue inhibitor of metallo-

protease may contribute to cartilage degradation in osteoar-

thritis and rheumatoid arthritis. Lab Invest 1994; 70:807–815.

19. Kundu P, Mukhopadhyay AK, Patra R et al. Cag patho-

genicity island-independent up-regulation of matrix metallo-

proteinases-9 and -2 secretion and expression in mice by

Helicobacter pylori infection. J Biol Chem 2006; 281:34651–

34662.

20. Chung HW, Lee JY, Moon HS et al. Matrix metallopro-

teinase-2, membranous type 1 matrix metalloproteinase, and

tissue inhibitor of metalloproteinase-2 expression in ectopic

and eutopic endometrium. Fertil Steril 2002; 78:787–795.

21. Di Nezza LA, Misajon A, Zhang J et al. Presence of active

gelatinases in endometrial carcinoma and correlation of matrix

metalloproteinase expression with increasing tumor grade and

invasion. Cancer 2002; 94:1466–1475.

22. Liokumovich P, Goldberg I, Davidson B et al. Expression

of metalloproteinases endometrial stromal sarcoma: immuno-

histochemical study using image analysis. J Clin Pathol 1999;

52:198–202.

23. Collette T, Maheux R, Mailloux J et al. Increased

expression of matrix metalloproteinase-9 in the eutopic endo-

metrial tissue of women with endometriosis. Hum Reprod

2006; 21:3059–3067.

24. Cox KE, Piva M, Sharpe-Timms KL. Differential regulation

of matrix metalloproteinase-3 gene expression in endometriotic

lesions compared with endometrium. Biol Reprod 2001;

65:1297–1303.

25. Uzan C, Cortez A, Dufournet C et al. Eutopic endo-

metrium and peritoneal, ovarian and bowel endometriotic

tissues express a different profile of matrix metalloproteinases-

2, -3 and -11, and of tissue inhibitor metalloproteinases-1 and -

2. Virchows Arch 2004; 445:603–609.

26. Marbaix E, Donnez J, Courtoy PJ et al. Progesterone reg-

ulates the activity of collagenase and related gelatinases A and

B in human endometrial explants. Proc Natl Acad Sci U S A

1992; 89:11789–11793.

27. Razandi M, Pedram A, Park ST et al. Proximal events in

signaling by plasma membrane estrogen receptors. J Biol

Chem 2003; 278:2701–2712.

28. Pustovrh MC, Jawerbaum A, Capobianco E et al. Oxidative

stress promotes the increase of matrix metalloproteinases-2

and -9 activities in the feto-placental unit of diabetic rats. Free

Radic Res 2005; 39:1285–1293.

29. Mori K, Shibanuma M, Nose K. Invasive potential induced

under long-term oxidative stress in mammary epithelial cells.

Cancer Res 2004; 64:7464–7472.

30. Rajagopalan S, Meng XP, Ramasamy S et al. Reactive

oxygen species produced by macrophage-derived foam cells

regulate the activity of vascular matrix metalloproteinases in

vitro. Implications for atherosclerotic plaque stability. J Clin

Invest 1996; 98:2572–2579.

31. Belkhiri A, Richards C, Whaley M et al. Increased

expression of activated matrix metalloproteinase-2 by human

endothelial cells after sublethal H2O2 exposure. Lab Invest

1997; 77:533–539.

32. Somerville RP, Oblander SA, Apte SS. Matrix metallo-

proteinases: old dogs with new tricks. Genome Biol 2003;

4:216.

33. Arumugam K, Yip YC. De novo formation of adhesions in

endometriosis: the role of iron and free radical reactions. Fertil

Steril 1995; 64:62–64.

34. Alpay Z, Saed GM, Diamond MP. Female infertility and free

radicals: potential role in adhesions and endometriosis. J Soc

Gynecol Invest 2006; 13:390–398.

35. Portz DM, Elkins TE, White R et al. Oxygen free radicals

and pelvic adhesion formation: I. Blocking oxygen free radical

toxicity to prevent adhesion formation in an endometriosis

model. Int J Fertil 1991; 36:39–42.

36. Tan DX, Chen LD, Poeggeler B et al. Melatonin: a potent

endogenous hydroxyl radical scavenger. Endocr J 1993; 1:57–

60.

37. Onuki J, Almeida EA, Medeiros MH et al. Inhibition of

5-aminolevulinic acid-induced DNA damage by melatonin,

N1-acetyl-N2-formyl-5-methoxykynuramine, quercetin or res-

veratrol. J Pineal Res 2005; 38:107–115.

38. Tan DX, Manchester LC, Terron MP et al. One molecule,

many derivatives: a never-ending interaction of melatonin with

reactive oxygen and nitrogen species? J Pineal Res 2007; 42:28–

42.

39. Manda K, Ueno M, Anzai K. AFMK, a melatonin metab-

olite, attenuates X-ray-induced oxidative damage to DNA,

proteins and lipids in mice. J Pineal Res 2007; 42:386–393.

40. Zhao H, Pang SF, Poon AM. Variations of mt1 melatonin

receptor density in the rat uterus during decidualization, the

estrous cycle and in response to exogenous steroid treatment.

J Pineal Res 2002; 33:140–145.

41. Kobayashi Y, Itoh MT, Kondo H et al. Melatonin binding

sites in estrogen receptor-positive cells derived from human

endometrial cancer. J Pineal Res 2003; 35:71–74.

42. Macphee AA, Cole FE, Rice BF. The effect of melatonin on

steroidogenesis by the human ovary in vitro. J Clin Endocrinol

Metab 1975; 40:688–696.

Paul et al.

448

43. Hill SM, Spriggs LL, Simon MA et al. The growth inhibi-

tory action of melatonin on human breast cancer cells is linked

to the estrogen response system. Cancer Lett 1992; 64:249–256.

44. Del Rı́o B, Garcı́a Pedrero JM, Martı́nez-Campa C et al.

Melatonin, an endogenous-specific inhibitor of estrogen recep-

tor alpha via calmodulin. J Biol Chem 2004; 279:38294–38302.

45. Rato AG, Pedrero JG, Martinez MA et al. Melatonin

blocks the activation of estrogen receptor for DNA binding.

FASEB J 1999; 13:857–868.

46. Rock JA. The revised American Fertility Society classification

of endometriosis: reproducibility of scoring. ZOLADEX

Endometriosis Study Group. Fertil Steril 1995; 63:1108–1110.

47. Somigliana E, Vigano P, Rossi G et al. Endometrial ability

to implant in ectopic sites can be prevented by interleukin-12 in

a murine model of endometriosis. Hum Reprod 1999; 14:2944–

2950.

48. Swarnakar S, Ganguly K, Kundu P et al. Curcumin reg-

ulates expression and activity of matrix metalloproteinases-9

and-2 during prevention and healing of indomethacin-induced

gastric ulcer. J Biol Chem 2005; 280:9409–9415.

49. Behrendtsen O, Alexander CM, Werb Z. Metalloprotein-

ases mediate extracellular matrix degradation by cells from

mouse blastocyst outgrowths. Development 1992; 114:447–456.

50. Fishman DA, Bafetti LM, Banionis S et al. Production of

extracellular matrix-degrading proteinases by primary cultures

of human epithelial ovarian carcinoma cells. Cancer 1997;

80:1457–1463.

51. Collette T, Bellehumeur C, Kats R et al. Evidence for an

increased release of proteolytic activity by the eutopic endo-

metrial tissue in women with endometriosis and for involve-

ment of matrix metalloproteinase-9. Hum Reprod 2004;

19:1257–1264.

52. Vu TH, Shipley JM, Bergers G et al. MMP-9/gelatinase B is

a key regulator of growth plate angiogenesis and apoptosis of

hypertrophic chondrocytes. Cell 1998; 93:411–422.

53. Qian X, Wang TN, Rothman VL et al. Thrombospondin-1

modulates angiogenesis in vitro by up-regulation of matrix

metalloproteinase-9 in endothelial cells. Exp Cell Res 1997;

235:403–412.

54. Ria R, Loverro G, Vacca A et al. Angiogenesis extent and

expression of matrix metalloproteinase-2 and -9 agree with

progression of ovarian endometriomas. Eur J Clin Invest 2002;

32:199–206.

55. Spuijbroek MD, Dunselman GA, Menheere PP et al. Early

endometriosis invades the extracellular matrix. Fertil Steril

1992; 58:929–933.

56. Van Den Steen PE, Dubois B, Nelissen I et al. Biochemistry

and molecular biology of gelatinase B or matrix metallopro-

teinase-9 (MMP-9). Crit Rev Biochem Mol Biol 2002; 37:375–

536.

57. Sato H, Kita M, Seiki M. v-Src activates the expression of 92-

kDa type IV collagenase gene through the AP-1 site and the

GT box homologous to retinoblastoma control elements. A

mechanism regulating gene expression independent of that

by inflammatory cytokines. J Biol Chem 1993; 268:23460–

23468.

58. Li N, Karin M. Is NF-kappab the sensor of oxidative stress?

FASEB J 1999; 13:1137–1143.

59. Murphy AA, Palinski W, Rankin S et al. Evidence for

oxidatively modified lipid-protein complexes in endometrium

and endometriosis. Fertil Steril 1998; 69:1092–1094.

60. Shanti A, Santanam N, Morales AJ et al. Autoantibodies

to markers of oxidative stress are elevated in women with

endometriosis. Fertil Steril 1999; 71:1115–1118.

61. Van Langendonckt A, Casanas-Roux F, Donnez J. Oxi-

dative stress and peritoneal endometriosis. Fertil Steril 2002;

77:861–870.

62. Szczepanska M, Kozlik J, Skrzypczak J et al. Oxidative

stress may be a piece in the endometriosis puzzle. Fertil Steril

2003; 79:1288–1293.

Arresting endometriosis by melatonin

449