Expression of protease-activated receptor-2 (PAR-2) in patients with nasopharyngeal carcinoma: Correlation with clinicopathological features and prognosis

ARTICLE IN PRESS

Pathology – Research and Practice 205 (2009) 542–550 www.elsevier.de/prp

ORIGINAL ARTICLE

Expression of protease-activated receptor-2 (PAR-2) in patients with nasopharyngeal carcinoma: Correlation with clinicopathological features and prognosis Zhi Li, Li-Juan Bian, Yang Li, Ying-Jie Liang, Hui-Zhen Liang Department of Pathology, 1st Affiliated Hospital, Sun Yat-sen University, 58# Zhongshan Road II, Guangzhou 510080, China Received 27 November 2008; received in revised form 17 January 2009; accepted 26 January 2009

Abstract We aimed at determining whether the expression of protease-activated receptor 2 (PAR-2) is involved in the progression of nasopharyngeal carcinoma (NPC) and correlated with latent membrane protein 1 (LMP-1), matrix metalloproteinases-9 (MMP9), and angiogenesis of tumor. PAR-2, LMP-1, and MMP9 expressions were detected in 57 biopsies of primary NPC by immunohistochemistry. The presence of Epstein-Barr virus (EBV) was determined using EBER in situ hybridization, and intratumoral microvessels were highlighted by staining endothelial cells for anti-CD34. The correlations with immunostainings and clinicopathological factors, as well as the follow-up data of patients, were analyzed statistically. Strong expression of PAR-2 in 61.4% (35/57) of the biopsies was correlated with extensive lymph node metastasis and advanced stage of NPC. The patients with PAR-2/LMP-1 or PAR-2/MMP9 dual high-expression tumors had a significant worse prognosis than those with single protein high expression and dual low or negative expression tumors (P ¼ 0.013 and 0.004, respectively). Angiogenesis in the tumor is related to overall survival of NPC patients (P ¼ 0.001), and exhibits strong PAR-2 expression or LMP-1 expression in tumors associated with increased intratumoral microvessel density (P ¼ 0.026 and 0.006, respectively). PAR-2 is a possible mediator cooperating with LMP-1 and MMP9 to influence the progression of NPC by inducing angiogenesis and promoting lymph node metastasis. Crown Copyright r 2009 Published by Elsevier GmbH. All rights reserved. Keywords: Nasopharyngeal carcinoma (NPC); Protease-activated receptor-2 (PAR-2); Latent membrane protein-1 (LMP-1); Matrix metalloproteinases-9 (MMP9); Angiogenesis

Introduction Nasopharyngeal carcinoma (NPC) is a unique epithelial neoplasm with a high incidence in southern China [35]. NPC shows more highly invasive and metastatic features than other head and neck carcinomas. However, Corresponding author. Tel.: +8620 8733 0890; fax: +8620 8733 1780. E-mail address: [email protected] (Z. Li).

the mechanism by which metastasis occurs remains unknown [1,6]. There is a growing interest in the activation of protease-activated receptor-2 (PAR-2) for its effect on metastasis of tumors. PAR-2 is one of the members of the receptor superfamily coupled to G-proteins [4,7]. PAR-2 has multiple functions, and is involved in pathophysiological processes, including inflammation and pain [17,19]. Expression of PAR-2 has been reported in human colonic and gastric cancer cell lines [2,25], and higher

0344-0338/$ - see front matter Crown Copyright r 2009 Published by Elsevier GmbH. All rights reserved. doi:10.1016/j.prp.2009.01.015

ARTICLE IN PRESS Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

expression of PAR-2 in cancer cells, implicating a poor prognosis, has been linked to breast and gastric cancer [3,18]. However, little attention has been paid to the question of whether PAR-2 is responsible for the invasion and metastasis of NPC. The role of PAR-2 in NPC cells remains unclear. Epstein-Barr virus (EBV) is a close factor in the development and progression of NPC. EBV-encoded protein, latent membrane protein 1 (LMP-1), has been considered to be an important contributor to the development of NPC [32]. However, there is no study to demonstrate whether or not LMP-1 can promote tumor progression by regulating the expression of PAR-2. Matrix metalloproteinase-9 (MMP9) can degrade extracellular matrix (ECM) and basement membrane (BM), thus playing an essential role in metastasis [24]. Previous studies have associated MMP9 with the invasion and metastasis of a large variety of cancers, such as lung, prostate, breast, colon cancers, and NPC [33,38]. Recent studies have reported a PAR-2-mediated MMP9 release in airway epithelial cells [10], and trypsinactivated MMP-9 stimulates invasiveness in tongue cancer cells [22]. These findings suggest that PAR-2 and MMP9 may play a co-operative role in the invasion and metastasis of tumors. However, no study has investigated this in NPC. Angiogenesis in tumor increases the opportunity for tumor cells to enter the circulation. VEGF, bFGF, and IL-8 have been demonstrated to influence microvessel synthesis in various tumors [13,21]. Several studies have shown that PAR-2 activation can induce the release of VEGF, IL-6, and IL-8 [20,30,37]. It still needs to be clarified whether PAR-2 activation induces microvesssel synthesis in NPC and influences its prognosis. In this study, immunohistochemical staining was done in 57 NPC biopsies to detect the expression of PAR-2, LMP-1, and MMP9, as well as intratumoral microvessel density in the tumor. We hypothesized that PAR-2 might play roles in the progression of NPC by influencing lymph node metastasis and modulating angiogenic activity of tumors with LMP-1 or MMP9 protein.

Materials and methods Patients and clinicopathological findings Fifty-seven primary NPC patients were admitted to the 1st Affiliated Hospital and Cancer Center of Sun Yat-sen University during the period of June 1999 to October 2003. There were 44 males and 13 females with an average age of 56.2 years (range 22–72 years). The patients were followed-up from 2 to 50 months, with a

543

mean period of 36 months, and 14 patients (24.6%) had died as a result of recurrence of tumor in the follow-up period. All of the primary NPC biopsies were obtained by endoscopic examination before radiotherapy. After confirmation by histopathology, all patients were treated with standard curative radiotherapy with or without chemotherapy. The biopsies were fixed with 10% neutral-buffered formalin and paraffin-embedded. Histological classification of NPC was done according to WHO criteria: keratinizing squamous cell carcinoma (KSCC) (5.3%, 3/57), differentiated non-keratinizing carcinoma (DNKC) (24.6%, 14/57), and undifferentiated non-keratinizing carcinoma (UNKC) (70.1%, 40/57). In the present study, 5 of the 57 patients had clinical stage I, 15 stage II, 15 stage III, and 22 stage IV cancers according to the staging criteria of UICC (International Union Against Cancer). T1–T2 was identified as small tumor size, N0–N1 was local lymph node metastasis, and clinical stage I–II was regarded as early stage of tumor. By contrast, T3–T4 was identified as larger tumor size, N2–N3 as extensive lymph node metastasis, and stage III–IV as advanced stage of tumor.

Detection of latent EBV infection The latent presence of EBV in NPC biopsies was detected using EBER in situ hybridization as described previously [34]. Briefly, sections were predigested with proteinase K (1:30) and incubated with probe and alkaline phosphatase-conjugated anti-fluorescein antibody, in turn, according to the instruction of EBV Probe ISH Kit (Novocastra). 5-Bromo-4-chloro-3-indolylphosphate nitroblue tetrazolium was used as a chromogen, and sections were counterstained with light 1% methyl green stain. Dark blue staining was identified as the positive hybridization signal. Scoring was defined as follows: () negative, (+) o10% positive tumor nuclei, (++) 10–50% positive tumor nuclei, or (+++) 450% positive tumor nuclei. Tumors scored as ++++ were considered EBER-positive.

Immunohistochemical staining and counting of microvessels Four-micrometer-thick sections from 57 samples were dewaxed and rehydrated routinely before antigen retrieval by microwave. The sections were incubated with diluted primary antibodies (PAR-2 goat polyclonal antibody 1:100, LMP-1 mouse monoclonal antibody 1:100, MMP9 mouse monoclonal antibody 1:50 and CD34 mouse monoclonal antibody 1:100, Santa Cruz Biotechnology) at 41 overnight. LSAB+ system (Dako Com., CA) was used to detect these antigens in NPC. The sections were developed with DAB, and lightly counterstained with hematoxylin. PAR-2-, LMP-1-, and

ARTICLE IN PRESS 544

Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

MMP9-positive signals in NPC cells were scored by the previously described criteria [3]: (0) negative, (1) 10% up to 20%, (2) 20% up to 50%, and (3) 50% or more. Immunoreactive intensity was graded as (0) negative, (1) weakly yellowish, (2) brown-yellow, (3) dark brown. The multiplier of the above two scores was defined as the protein expression in tumor. PAR-2 or MMP9 was classified as high expression when the scoring multiplier was equal or more than 4. LMP-1 protein was considered as expression when the scoring was equal to or more than 1.

The counting of microvessels in NPC was evaluated by the previously reported method [14]. Briefly, intratumoral microvessel density (IMD) was observed in areas of most intense neovascularization or hotspots in tumor by light microscopy. After determining the area of highest neovascularization, single microvessels were manually counted on a 200  field by two different observers without knowledge of patient outcome. Any brown-stained endothelial cell or cell cluster clearly separated from adjacent microvessels was considered as a single, countable microvessel.

Fig. 1. EBER, PAR-2, MMP9, and LMP-1 expression in NPC patient biopsies. EBER-positive cells showed dark blue signal in tumor cell nuclei to indicate the latent presence of EBV in NPC (A,  200, in situ hybridization). PAR-2 immunoreactivity was intense at the membrane and cytoplasm of NPC tumor cells. Stromal cells, including fibrous cells and lymphocytes, had weaker positive signals (B,  200, IHC staining). In KSCC, the keratinizing pearl (*) showed very weak immunostaining for PAR-2, but PAR-2 immunoreactivity was intensely strong in surrounding tumor cells (C,  400, IHC staining). MMP9 immunoreactivity in tumor cells and stromal cells with cytoplasmic pattern (D,  200, IHC staining), but LMP-1 positive signal was localized mainly in the cytoplasm of tumor cells (E,  200, IHC staining). All microvessels in tumor were highlighted by staining endothelial cells for anti-CD34. Any brown-staining endothelial cell or cell cluster that was clearly separated from adjacent microvessels was considered as a single, countable microvessel. (F,  400, IHC staining).

ARTICLE IN PRESS Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

Statistical analysis

545

Expression of PAR-2, LMP-1, and MMP9 in NPC biopsies, and their correlation

All statistical analyses were carried out using SPSS 13.0 software for Windows. The chi-square test was used to assess PAR-2, LMP-1, and MMP9 expression with clinicopathological characteristics. Univariate analysis by Student’s t test was used to assess protein expression in relation to angiogenesis of NPC. The survival curve of patients was determined using the Kaplan–Meier method and Cox regression, and statistical evaluation was performed using the log rank test. A P-value o0.05 was considered statistically significant.

A total of 61.4% of tumor tissues (35/57) showed high expression for PAR-2. PAR-2 expression was intense at the cell membrane and cytoplasm of tumor cells, and signals were weaker in non-tumor ciliated columnar cells, metaplastic squamous cells, fibrous cells, and immune cells (Fig. 1B and C). PAR-2 high expression was significantly correlated with extensive lymph node metastasis and advanced stage of NPC. MMP9 immunoreactivity was also detected in the tumor cell and infiltrating inflammatory cells. High expression of MMP9 (57.9%, 33/57) was found to have a positive significant correlation with extensive lymph node metastasis (Fig. 1D). LMP-1 positive expression was found in 42.1% of (24/57) tumor tissues. LMP-1 and MMP-9 could be detected in a concomitant expression in the same portion of tumor cell on serial sections. LMP-1 expression was found to be closely related to the expression of MMP9 in tumor (Fig. 1E) (Table 1). The patients with either PAR-2, LMP-1, or MMP9 high expression tumors alone had no poorer prognosis

Result EBER expression in NPC patient biopsies In this study, EBER in situ hybridization signal intensity was detected in all of the NPC biopsies, regardless of the histological subtype based on WHO classification (Fig. 1A). This indicated that all the patients’ tumors were EBV-positive.

Table 1. Correlation of PAR-2, LMP-1 and MMP9 expression and clinicopathological findings. (The values in this Table show the number of patients. Ps are results from the comparisons between the two groups. Chi-square test). Variables

Total patients Histological type KSCC DNKC UNKC Tumor size Small (T1–2) Large (T3–4) Lymph node metastasis Local (N0–1) Extensive (N2–3) Clinical stage Early (I–II) Advanced (III–IV) Sex Male Female Age (mean) o56 years 456 years MMP9 expression ()/low High PAR-2 expression ()/low High

PAR-2 expression ()/low

High

22

35

1 5 16

2 9 24

14 8

MMP9 expression P-value

()/low

High

24

33

0.824

2 8 14

1 6 26

18 17

0.281

12 12

14 8

12 23

0.017

11 11

9 26

16 6

LMP-1 expression P-value

P-value

()

(+)

33

24

0.200

2 8 23

1 6 17

0.827

20 13

0.305

17 16

15 9

0.227

15 9

11 22

0.006

16 17

10 14

0.442

0.020

10 14

10 23

0.257

12 21

8 16

0.742

28 7

0.357

20 4

24 9

0.209

25 8

19 5

0.579

10 12

11 24

0.169

9 15

12 21

0.926

13 20

8 16

0.311

11 11

13 22

0.237

19 14

5 19

0.001

12 21

10 14

0.588

ARTICLE IN PRESS Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

IMD in relation to survival of NPC patients and its relationship with PAR-2, LMP-1, and MMP9 IMD in the 57 NPC tissues varied from 12.5 to 72.5 with a mean value of 30.70714.88. There was a statistically significant correlation between IMD and overall survival of NPC patients, P-value was 0.001 as tested by the Cox regression method (Table 3). The tumors with either PAR-2 high expression or LMP-1 expression exhibited significantly higher tumor angiogenesis; however, there was no correlation between MMP9 high expression and angiogenesis (Fig. 3). We examined the relationship between other clinicopathological variables and IMD in tumor; no significant correlations were found. Therefore, angiogenesis in NPC appears to be independent of tumor size, clinical stage, lymph node metastasis, histological type, sex, and age of patients.

100

80 Cum survival (%)

by Cox regression model analysis. However, the patients with PAR-2/LMP-1 or PAR-2/MMP9 dual high expression tumors had significantly worse prognoses than those with single protein high expression or dual lower expression tumors (Fig. 2). A significant correlation existed between PAR-2/LMP-1 or PAR-2/MMP9 expression status and lymph node metastasis, as well as advanced tumor stage (Table 2).

60

40 Log Rank test P = 0.004 20

Dual PAR-2 MMP9 negative or low expression PAR-2 or MMP9 single high expression Dual PAR-2 and MMP9 high expression

0 0

10

20 30 40 Survival time (months)

50

60

100

80 Cum survival (%)

546

60

40 Log Rank test P = 0.013 Dual PAR-2 LMP-1 negative or low expression Single PAR-2 high expression or LMP-1 positive Dual PAR-2 high expression and LMP-1 positive

20

Discussion 0

The specific factors responsible for metastasis in NPC have not been identified, although certain prognostic factors (clinical stage, lymph node status) have been well-established as reliable indicators of NPC. The present study has shown that PAR-2 may play a critical role in the mechanism of tumor metastasis in NPC. A recently conducted study reveals that PAR-2 actually functions as signal molecule at the cell surface [27]. Very high expression of PAR-2 has been found in certain tumor cell lines derived from the lung, colon, and pancreas, and PAR-2-mediated tumor cell migration and metastasis have also been proved in human tumor cells in vitro [8,36]. In the current study, the result of the relationship between PAR-2 high expression and lymph node metastasis, as well as the advanced stage of tumor, indicated that PAR-2 might be a possible mediator to contribute to the extensive lymph node metastasis and to accelerate the poor progression of NPC. However, PAR-2 high expression in tumors did not induce a poorer prognosis of patients independently, and a significantly shortened survival time occurred in patients only when the tumor co-expressed PAR-2and EBV-encoded LMP-1 protein. These results indicate that PAR-2 high expression in NPC cells is not an

0

10

20 30 40 Survival time (months)

50

60

Fig. 2. Kaplan–Meier survival analysis in patients with NPC. The mean survival time of dual negative or low PAR-2/MMP9 expression was 49 months (95%CI 46–52), and 46 months (95%CI 40–52) for PAR-2/MMP9 in single high expressed patients showing strong expression, but only 36 months (95%CI 30–43) in patients with strong dual PAR-2/MMP9 expression. A significant difference could be found when compared with the two other groups of patients (A). The mean survival time was 47 months (95%CI 41–52) for patients with dual negative or low PAR-2/LMP-1 expression, and 44 months (95%CI 40–48) for patients with single high or positive PAR-2/LMP-1 expression. However, the mean survival time in patients with dual high or positive PAR-2/ LMP-1 expression was only 36 months (95%CI, 27–45). It was significantly decreased in the two other groups (B).

independent factor to influence the prognosis of NPC patients. In contrast to our results, however, Fujimoto et al. [3] recently reported a significantly poorer prognosis in 183 gastric cancer patients with PAR-2 high expression. Further investigations are necessary to

ARTICLE IN PRESS Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

Table 2.

Correlation of PAR-2/LMP-1 and PAR-2/MMP9 co-expression status with clinical findings in NPC.

Variables

PAR-2/MMP9 co-expression

PAR-2/LMP-1 co-expression

Dual lowa

Single high

11

24

22

Tumor size Small (T1–2) Large (T3–4)

7 4

12 12

13 9

Lymph node metastasis Local (N0–1) Extensive (N2–3)

8 3

13 11

Clinical stage Early (I–II) Advanced (III–IV)

6 5

9 15

Total patients

a

547

Dual lowa

Single high

12

31

14

0.305

8 4

15 16

9 5

0.075

5 17b

0.001

9 3

12 19

5 9b

0.001

5 17b

0.015

7 5

9 22

4 10b

0.010

Dual high

P-value

Dual high

P-value

Dual low expression contains negative expression. Showing significant difference to other groups (Chi-square test).

b

Table 3. Cox regression model correlating PAR-2, MMP9, LMP-1 expression, IMD, and clinical parameters with overall survival of patients. Variables

95.0% confidence interval

Pvalue

PAR-2 expression alone MMP9 expression alone LMP-1 expression alone Dual high PAR-2/LMP-1 expression Dual high PAR-2/MMP9 expression IMD Age Tumor size Lymph node metastasis Clinical stage

0.654–8.425 0.833–10.781 0.914–8.166 0.091–0.747

0.191 0.093 0.072 0.012

0.068–0.694

0.010

1.040–1.113 0.433–4.412 0.074–0.961 0.933–12.056 0.765–15.360

0.001 0.585 0.053 0.064 0.107

NOTE: Result of Cox regression model analyzing PAR-2 and MMP9 expression (high expression versus low or negative expression), LMP-1 (positive versus negative), lymph node metastasis (N0–1 versus N2–3) and stage (I/II versus III/IV), dual high PAR-2/LMP-1(MMP9)expression (dual high expression versus dual low or single high expression), age (o56 versusX56), tumor size (T1–2 versus T3–4).

reveal whether tumor species and sample size are responsible for these discrepancies. LMP-1 is an important EBV-encoded protein for the development and progression of NPC. There is no data to demonstrate whether or not LMP-1 could induce extensive lymph node metastasis and advanced stage of NPC by upregulating the expression of PAR-2 in tumor. In the present study, EBV infection could be detected in all of NPC samples, but there was no evidence given

that PAR-2 high expression was directly promoted by LMP-1 in tumor. However, co-expression of PAR-2 and LMP-1 indeed advances the tumor stage and exhibits a significantly poorer prognosis. These findings suggest that PAR-2 and LMP-1 may co-operatively influence the prognosis of NPC through the different modulating pathways. On the one hand, high expression of PAR-2 and LMP-1 in tumor cells might contribute to the neovascularization, which had a critical effect on the outcome of NPC patients. The association of neovascularization with both angiogenic and lymphatic metastasis has been examined in many malignant tumors. In this study, angiogenesis in tumors is indeed independent of tumor size and stage in relation to the prognosis of NPC patients, which is a finding similar to that of previous studies [15,23]. Moreover, increased neovascularization could be found in the tumor with high expression of PAR-2 or LMP-1 expression. These results suggest that PAR-2 and LMP-1 may contribute to angiogenesis in NPC. It is well-known that the number of angiogenic molecules, such as VEGF, bFGF, IL-6, and IL-8, contribute to microvessel synthesis in various tumors [13,16,21]. Recent studies have shown that activation of PAR-2 can release VEGF, IL-6, and IL-8 in tumor cell lines [20,30,37]. The role of PAR-2 in tumor advancement via angiogenic activity has also been found in uterine endometrial cancers [12]. As for LMP-1, several studies have shown that LMP-1 also induces VEGF through cyclooxygenase-2 activation and IL-8 through NF-kappa B activation [26,31]. Therefore, the results of the present study suggested that PAR-2 high expression in NPC, like LMP-1, might also contribute to promoting angiogenesis by increasing the release of angiogenic molecules.

ARTICLE IN PRESS 548

Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

80 Intratunormal microvessel density

Negative or low expression High expression 60 *

**

40

20

0 MMP9

PAR-2

LMP-1

Fig. 3. PAR-2, MMP9 and LMP-1 expression in relation to neovascularization of NPC. The intratumoral microvessel density (IMD) varied in tumor; IMD was significantly higher in LMP-1 positive-tumors (*, P ¼ 0.006) and in tumors with high expression of PAR-2 (**, P ¼ 0.026) when compared with their negative or low expression cases, respectively. However, MMP9 high expression in tumor cells was not associated with higher microvessel proliferation (P ¼ 0.345). (Independent samples test).

On the other hand, a tendency towards high LMP-1 expression cases in relation to higher MMP9 expression might be another important reason to elucidate the cooperative effect of LMP-1 and PAR-2 on the overall survival of NPC patients. MMP9 has been reported to be markedly associated with invasion and metastasis in human cancers [5,11]. It is mainly due to MMP9degraded ECM and BM, and induces tumor cells to migrate from the capillaries to surrounding tissues [24]. The results of this study disclosed that high MMP9 expression in tumor cells was a major contribution to tumor metastasis, and MMP9 high expression was closely related to LMP-1 expression in tumor. In fact, our data are similar to those of pervious studies that LMP-1 can activate MMP9, and induction of MMP9 by LMP-1 contributes to the metastatic potential of NPC [9,29]. However, in our study, high MMP9 expression alone had no effects on overall survival of NPC patients. There was a significantly poorer prognosis only in NPC patients with PAR-2 and MMP9 dual high expression. In addition, strong co-expression of PAR-2/MMP9 was closely related to extensive lymph node metastasis and advanced tumor stage, but there was no significant correlation between PAR-2 and MMP9 expression in tumor. These findings indicate that PAR-2- and LMP-1induced MMP9 expression in NPC also exerts a cooperative effect on tumor progression. PAR-2 expression in NPC could not directly mediate the altered expression of MMP9, even though this pathway has been established in airway epithelial cells [10] and prostate cancer cell lines [28]. LMP-1 expression in NPC might include a potential for metastasis and

poorer prognosis by activating MMP9 expression in tumor. Therefore, co-expression of PAR-2 and LMP-1 in tumor might be associated with a worse prognosis in patients with NPC. From this standpoint, diminishing the expression of PAR-2 or interfering with the activation of PAR-2 in tumor appears to be a promising strategy for developing new therapeutic approaches in the treatment of malignant tumors. In conclusion, our data indicate that PAR-2 participates in the progression of NPC by influencing the lymph node metastasis of tumor cells and by promoting angiogenesis of tumor. Therefore, PAR-2 could be a potential marker for the prognosis of NPC. It is likely that PAR-2 and LMP-1 exert their influence on the progression of NPC through independent pathways, but there are some cross effects, such as promoting microvessel proliferation, in the pathways of PAR-2 and LMP-1 for tumorigenesis of NPC. Although PAR-2 is not an independent predicting factor cooperating with LMP-1 and MMP9, it appears to have an essential role in the prognosis of NPC.

Acknowledgments This study was supported by a grant from the Natural Science Foundation of Guangdong Province, China (5001744). The main work of the study was carried out in collaboration with researchers from the Clinical Division of the Chinese Medicine School, Hong Kong Baptist University. We would like to express our deep appreciation to these individuals.

References [1] A. Fandi, M. Altun, N. Azli, J.P. Armand, E. Cvitkovic, Nasopharyngeal cancer: epidemiology, staging, and treatment, Semin. Oncol. 21 (1994) 382–397. [2] D. Darmoul, J.C. Marie, H. Devaud, V. Gratio, M. Laburthe, Initiation of human colon cancer cell proliferation by trypsin acting at protease-activated receptor-2, Br. J. Cancer 85 (2001) 772–779. [3] D. Fujimoto, Y. Hirono, T. Goi, K. Katayama, K. Hirose, A. Yamaguchi, Expression of protease activated receptor-2 (PAR-2) in gastric cancer, J. Surg. Oncol. 93 (2006) 139–144. [4] E.A. Mayer, S.M. Collins, Evolving pathophysiologic models of functional gastrointestinal disorders, Gastroenterology 122 (2002) 2032–2048. [5] E.I. Deryugina, J.P. Quigley, Matrix metalloproteinases and tumor metastasis, Cancer Metastasis Rev. 25 (2006) 9–34. [6] E.T. Chang, H.O. Adami, The enigmatic epidemiology of nasopharyngeal carcinoma, Cancer Epidemiol. Biomarkers Prev. 15 (2006) 1765–1777.

ARTICLE IN PRESS Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

[7] F. Mule, M.C. Baffi, M.C. Cerra, Dual effect mediated by protease-activated receptors on the mechanical activity of rat colon, Br. J. Pharmacol. 136 (2002) 367–374. [8] G.M. Hjortoe, L.C. Petersen, T. Albrektsen, B.B. Sorensen, P.L. Norby, S.K. Mandal, U.R. Pendurthi, L.V. Rao, Tissue factor-factor VIIa-specific up-regulation of IL-8 expression in MDA-MB-231 cells is mediated by PAR-2 and results in increased cell migration, Blood 103 (2004) 3029–3037. [9] H. Takeshita, T. Yoshizaki, W.E. Miller, H. Sato, M. Furukawa, J.S. Pagano, N. Raab-Traub, Matrix metalloproteinase 9 expression is induced by Epstein-Barr virus latent membrane protein 1 C-terminal activation regions 1 and 2, J. Virol. 73 (1999) 5548–5555. [10] H. Vliagoftis, A. Schwingshackl, C.D. Milne, M. Duszyk, M.D. Hollenberg, J.L. Wallace, A.D. Befus, R. Moqbel, Proteinase-activated receptor-2-mediated matrix metalloproteinase-9 release from airway epithelial cells, J. Allergy Clin. Immunol. 106 (2000) 537–545. [11] H. Zheng, H. Takahashi, Y. Murai, Z. Cui, K. Nomoto, H. Niwa, K. Tsuneyama, Y. Takano, Expressions of MMP-2, MMP-9 and VEGF are closely linked to growth, invasion, metastasis and angiogenesis of gastric carcinoma, Anticancer Res. 26 (2006) 3579–3583. [12] I. Jahan, J. Fujimoto, S.M. Alam, E. Sato, H. Sakaguchi, T. Tamaya, Expression of protease activated receptor-2 related to angiogenesis in tumor advancement of uterine endometrial cancers, Oncol. Rep. 17 (2007) 345–350. [13] J.A. Belperio, M.P. Keane, D.A. Arenberg, C.L. Addison, J.E. Ehlert, M.D. Burdick, R.M. Strieter, CXC chemokines in angiogenesis, J. Leukoc. Biol. 68 (2000) 1–8. [14] L. Rubio, J.S. Burgos, C. Morera, F.J. Vera-Sempere, Morphometric study of tumor angiogenesis as a new prognostic factor in nasopharyngeal carcinoma patients, Pathol. Oncol. Res. 6 (2000) 210–216. [15] L. Rubio, J.S. Burgos, J.A. Lopez-Guerrero, C. Morera, F.J. Vera-Sempere, Expression of p53 protein and tumor angiogenesis as prognostic factors in nasopharyngeal carcinoma patients, Pathol. Res. Pract. 198 (2002) 97–102. [16] M. Presta, P. Dell’Era, S. Mitola, E. Moroni, R. Ronca, M. Rusnati, Fibroblast growth factor/fibroblast growth factor receptor system in angiogenesis, Cytokine Growth Factor Rev. 16 (2005) 159–178. [17] M.D. Hollenberg, S.J. Compton, International union of pharmacology. XXVIII. Proteinase-activated receptors, Pharmacol. Rev. 54 (2002) 203–217. [18] M.R. D’Andrea, C.K. Derian, R.J. Santulli, P. AndradeGordon, Differential expression of protease-activated receptors-1 and -2 in stromal fibroblasts of normal, benign, and malignant human tissues, Am. J. Pathol. 158 (2001) 2031–2041. [19] N. Vergnolle, Modulation of visceral pain and inflammation by protease-activated receptors, Br. J. Pharmacol. 141 (2004) 1264–1274. [20] O. Ikeda, H. Egami, T. Ishiko, S. Ishikawa, H. Kamohara, H. Hidaka, M. Takahashi, M. Ogawa, Signal of proteinase-activated receptor-2 contributes to highly malignant potential of human pancreatic cancer by up-

[21] [22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

549

regulation of interleukin-8 release, Int. J. Oncol. 28 (2006) 939–946. P. Carmeliet, R.K. Jain, Angiogenesis in cancer and other diseases, Nature 407 (2000) 249–257. P. Nyberg, M. Moilanen, A. Paju, A. Sarin, U.H. Stenman, T. Sorsa, T. Salo, MMP-9 activation by tumor trypsin-2 enhances in vivo invasion of human tongue carcinoma cells, J. Dent. Res. 81 (2002) 831–835. R.L. Foote, N. Weidner, J. Harris, E. Hammond, J.E. Lewis, T. Vuong, K.K. Ang, K.K. Fu, Evaluation of tumor angiogenesis measured with microvessel density (MVD) as a prognostic indicator in nasopharyngeal carcinoma: results of RTOG 9505, Int. J. Radiat. Oncol. Biol. Phys. 61 (2005) 745–753. S. Curran, G.I. Murray, Matrix metalloproteinases: molecular aspects of their roles in tumour invasion and metastasis, Eur. J. Cancer 36 (2000) 1621–1630. S. Miyata, N. Koshikawa, H. Yasumitsu, K. Miyazaki, Trypsin stimulates integrin alpha(5)beta(1)- dependent adhesion to fibronectin and proliferation of human gastric carcinoma cells through activation of proteinase-activated receptor-2, J. Biol. Chem. 275 (2000) 4592–4598. S. Murono, H. Inoue, T. Tanabe, I. Joab, T. Yoshizaki, M. Furukawa, J.S. Pagano, Induction of cyclooxygenase2 by Epstein-Barr virus latent membrane protein 1 is involved in vascular endothelial growth factor production in nasopharyngeal carcinoma cells, Proc. Natl. Acad. Sci. USA 98 (2001) 6905–6910. S.K. Bohm, K. McConalogue, W. Kong, N.W. Bunnett, Proteinase-activated receptors: new functions for old enzymes, News Physiol. Sci. 13 (1998) 231–240. S.R. Wilson, S. Gallagher, K. Warpeha, S.J. Hawthorne, Amplification of MMP-2 and MMP-9 production by prostate cancer cell lines via activation of proteaseactivated receptors, Prostate 60 (2004) 168–174. T. Horikawa, T. Yoshizaki, T.S. Sheen, S.Y. Lee, M. Furukawa, Association of latent membrane protein 1 and matrix metalloproteinase 9 with metastasis in nasopharyngeal carcinoma, Cancer 89 (2000) 715–723. T. Li, S. He, Induction of IL-6 release from human T cells by PAR-1 and PAR-2 agonists, Immunol. Cell Biol. 84 (2006) 461–466. T. Yoshizaki, T. Horikawa, R. Qing-Chun, N. Wakisaka, H. Takeshita, T.S. Sheen, S.Y. Lee, H. Sato, M. Furukawa, Induction of interleukin-8 by Epstein-Barr virus latent membrane protein-1 and its correlation to angiogenesis in nasopharyngeal carcinoma, Clin. Cancer Res. 7 (2001) 1946–1951. T. Yoshizaki, Promotion of metastasis in nasopharyngeal carcinoma by Epstein-Barr virus latent membrane protein-1, Histol. Histopathol. 17 (2002) 845–850. T.D. Kim, K.S. Song, G. Li, H. Choi, H.D. Park, K. Lim, B.D. Hwang, W.H. Yoon, Activity and expression of urokinase-type plasminogen activator and matrix metalloproteinases in human colorectal cancer, BMC Cancer 6 (2006) 211. W. Shi, I. Pataki, C. MacMillan, M. Pintilie, D. Payne, B. O’Sullivan, B.J. Cummings, P. Warde, F.F. Liu, Molecular pathology parameters in human nasopharyngeal carcinoma, Cancer 94 (2002) 1997–2006.

ARTICLE IN PRESS 550

Z. Li et al. / Pathology – Research and Practice 205 (2009) 542–550

[35] W.C. Cho, Nasopharyngeal carcinoma: molecular biomarker discovery and progress, Mol. Cancer 6 (2007) 1. [36] X. Shi, B. Gangadharan, L.F. Brass, W. Ruf, B.M. Mueller, Protease-activated receptors (PAR1 and PAR2) contribute to tumor cell motility and metastasis, Mol. Cancer. Res. 2 (2004) 395–402. [37] Y. Liu, B.M. Mueller, Protease-activated receptor-2 regulates vascular endothelial growth factor expression

in MDA-MB-231 cells via MAPK pathways, Biochem. Biophys. Res. Commun. 344 (2006) 1263–1270. [38] Z. Li, Y. Ren, Q.C. Wu, S.X. Lin, Y.J. Liang, H.Z. Liang, Macrophage migration inhibitory factor enhances neoplastic cell invasion by inducing the expression of matrix metalloproteinase 9 and interleukin-8 in nasopharyngeal carcinoma cell lines, Chin. Med. J. (Engl.) 117 (2004) 107–114.