Silencing Id-1 inhibits lymphangiogenesis through down-regulation of VEGF-C in oral squamous cell carcinoma

Oral Oncology 47 (2011) 27–32

Contents lists available at ScienceDirect

Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology

Silencing Id-1 inhibits lymphangiogenesis through down-regulation of VEGF-C in oral squamous cell carcinoma Zuoqing Dong a, Fengcai Wei a, Chengjun Zhou b, Tomoki Sumida c, Hiroyuki Hamakawa c, Yingwei Hu a, Shaohua Liu a,⇑ a b c

Department of Oral and Maxillofacial Surgery, Qilu Hospital, and Stamatology Research Center, Shandong University, 107#, Wenhua Xi Road, Jinan 250012, PR China Department of Pathology, The Second Hospital of Shandong University, Jinan 250033, PR China Department of Oral and Maxillofacial Surgery, Ehime University Graduate School of Medicine, Ehime 791-0295, Japan

a r t i c l e

i n f o

Article history: Received 9 September 2010 Received in revised form 24 October 2010 Accepted 25 October 2010 Available online 15 December 2010 Keywords: Id-1 Lymphangiogenesis RNA interference VEGF-C Oral squamous cell carcinoma Oral cancer

s u m m a r y Our previous study demonstrated that overexpression of Id-1 (inhibitor of differentiation/DNA binding) was associated with lymphatic metastasis in human oral squamous cell carcinoma (OSCC). In this study, we further unveiled the association of Id-1 with vascular endothelial growth factor-C (VEGF-C) and peritumoral lymphatic vessel density (PLVD), and the effect of silencing Id-1 on inhibiting lymphangiogenesis in OSCC. We found that Id-1 was associated with VEGF-C (r = 0.569, p < 0.001) and PLVD (r = 0.240, p < 0.001) in OSCC. Lentivirus-mediated RNA interference targeting Id-1 in an OSCC cell line Tca8113 resulted in down-regulation of VEGF-C (p = 0.003, 0.007). Moreover, when Id-1 was suppressed by injecting Id-1-siRNA-lentivirus into the transplanted tumors in nude mice, VEGF-C was down-regulated (p = 0.018) and the PLVD decreased (p = 0.001). Our results suggest that Id-1 was correlated with lymphangiogenesis in OSCC. Silencing Id-1 could inhibit lymphangiogenesis through down-regulation of VEGF-C and it might be a promising treatment modality for the lymphatic metastasis of OSCC. Ó 2010 Elsevier Ltd. All rights reserved.

Introduction Oral squamous cell carcinoma (OSCC) is a common malignant tumor with poor clinical outcome and poor prognosis.1 OSCC usually metastasizes via lymph vessels, which is the common cause of death. Lymphatic metastasis depends on lymphangiogenesis,2,3 which is regulated by various lymphangiogenic factors. Vascular endothelial growth factor-C (VEGF-C) plays an important role in lymphangiogenesis. It has been confirmed that cancer cells could induce lymphangiogenesis and promote lymphatic metastasis by expressing VEGF-C.4,5 Id-1 (inhibitor of differentiation/DNA binding) is a transcription factor of the helix-loop-helix (HLH) family. It plays an important role in the development of cancer,6 and is associated with poor tumor differentiation,7 lymph node status8 and poor clinical prognosis.9 Our previous study found that Id-1 was overexpressed in OSCC and was associated with lymphatic metastasis.10 Though the association between Id-1 and lymphatic metastasis has been well documented, the molecular mechanism is not clear, and whether there is some correlation between Id-1 and lymphangiogenesis needs to be investigated further.

⇑ Corresponding author. Tel.: +86 13791122835; fax: +86 53182169274. E-mail address: [email protected] (S. Liu). 1368-8375/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.oraloncology.2010.10.012

To our knowledge, until now there was never any evidence that has shown a relationship between Id-1 and lymphangiogenesis in OSCC. In this study, using immunohistochemical methods, we investigated the relationship between Id-1 and VEGF-C and peritumoral lymphatic vessel density (PLVD), then small interfering RNA (siRNA) technique using lentivirus vector was applied to specifically inhibit the expression of Id-1 in an OSCC cell line Tca8113, and the expression of VEGF-C was detected. Finally in vivo model was used to evaluate the effect of silencing Id-1 on inhibiting lymphangiogenesis. Materials and methods Tumor samples and immunohistochemistry Tumor samples were obtained from 128 patients with OSCC who were treated at Qilu Hospital of Shandong University from 2004 through 2008. All of the patients received local tumor resection and synchronal neck dissection. Informed consent was obtained from all patients. Paraffin-embedded sample sections (4 lm thickness) were dewaxed and rehydrated. Id-1, VEGF-C and PLVD were detected using primary antibodies of Id-1 (rabbit polyclonal antibody, dilution 1:400, Santa Cruz, USA), VEGF-C (goat polyclonal antibody, dilution 1:50, R&D, USA) and LYVE-1 (rabbit polyclonal antibody,

28

Z. Dong et al. / Oral Oncology 47 (2011) 27–32

dilution 1:100, R&D, USA), respectively. The avidin–biotin technique was applied using DAB for visualization and hematoxylin for counterstaining. The immunoreactivity of Id-1 and VEGF-C was graded based on the intensity and the percentage of positive cells.11,12 PLVD was determined by counting the number of LYVE1 positive vessels per HPF in the tumor.13

1:2000, Zhongshan Goldenbridge Biotechnology, China) at room temperature for 1 h. Signals were detected by enhanced chemiluminescence (ECL).

Cell culture

The cell culture supernate were removed and assayed to detect secretory VEGF-C proteins. VEGF-C proteins were analyzed by ELISA using Quantikine Immunoassay Kits (R&D, USA). ELISA was performed according to the manufacturer’s instruction. After the colorimetric reaction, the optical density (OD) at 450 nm was quantified by an eight-channel spectrophotometer, and the OD readings were converted to picograms per milliliter (pg/ml) on the basis of the standard curves.

Tca8113 cell line (derived from human tongue carcinoma) was grown in DMEM supplemented with 10% fetal calf serum. The cells were cultured at 37 °C in a humidified atmosphere of 5% CO2. RNA interference Tca8113 cells were cultured at a concentration of 2  105 cells per well in six-well plates. Growing to 30–40% confluence, the cells were transfected with Id-1-siRNA-lentivirus in the presence of 5 lg/ml polybrene at a multiplicity of transfection (MOI) of 50. As mock transfection, cells were treated with NC-siRNA-lentivirus. And the control group was used maternal cells without any treatment. Cells were used in assays after 96–120 h. The siRNA sequences (Shanhai GeneChem, China) were as followed: Id-1-siRNA sequence: 50 -CATGAACGGCTGTTACTCA-30 .

NC-siRNA sequence: 50 -TTCTCCGAACGTGTCACGT-30 . Real-time PCR analysis Total RNA was prepared using TRIZOL reagent (Invitrogen, USA) according to the manufacturer’s protocol, and cDNA was synthesized using standard procedures. Real-time RT-PCR (Biored) was carried in the presence of SYBR green to measure the levels of Id1 and VEGF-C mRNAs. The primers sequences used in this study were shown in Table 1. Real-time PCR was performed using the cycle profile: 1 cycle at 94 °C for 1 min, followed by 40 cycles of 94 °C for 20 s; 52 °C (for Id-1), 54 °C (for VEGF-C) or 50 °C (for b-actin) for 30 s, and 72 °C for 30 s; and then a final extension at 60 °C for 5 min. Melting curve analysis was performed to verify the specificity of the amplification reactions. The amplification curves were analyzed and Ct values were determined. Western blotting analysis Proteins were extracted from the cells with Protein Extraction Kit, and quantified using BCA Protein Assay Reagent Kit (Nanjing, China). Proteins were separated by 10% SDS–PAGE and transferred onto PVDF membranes (Nippon Genetics, Japan) at 90 V for 35 min. The membranes were blocked, washed, and incubated with the primary antibodies of Id-1 (dilution 1:1000), VEGF-C (dilution 1:1000) and b-actin (dilution 1:1000, Santa Cruz, USA) respectively, overnight at 4 °C. The next day, the membranes were washed and incubated with HRP-conjugated IgG (dilution

Secretory VEGF-C protein quantitation by ELISA

In vivo treatment Tca8113 cell suspensions of 2  105 cells (0.1 ml) were submucosally injected into lingual central part in 30 BALB/c nu/nu male mice (Shanghai, China). The mice were 5-weeks old, with a body weight of 18–22 g. When the tumors were about 3.0–5.0 mm in diameter, 21 mice were divided into three groups. Tumors of each group were injected with DMEM, NC-siRNA-lentivirus (108 TU) and Id-1-siRNA-lentivirus (108 TU), respectively. Injection was performed intratumorally at several points, twice 1 week for 2 weeks. Three weeks later the mice were sacrificed and the tumor tissues were collected. Each tumor was divided into two halves, one was used to determine mRNA expression of Id-1 and VEGF-C, and the other was used to observe the expression of Id-1, VEGF-C proteins and PLVD by immunohistochemistry using primary antibodies of Id-1 (dilution 1:400), VEGF-C (dilution 1:50) and LYVE-1 (mouse polyclonal antibody, dilution 1:100, Santa Cruz, USA), respectively.

Statistical analysis The results were expressed as mean ± SD. All statistical analyses were carried out using SPSS software (version 16.0, USA). Spearman’s coefficient of correlation was used for bivariate correlation comparison. The Mann–Whitney test and Student’s t-test were used to examine the association between Id-1 expression, VEGFC expression and PLVD. P-value <0.05 was considered to be significant.

Table 2 Immunohistochemistry analysis of Id-1, VEGF-C and PLVD in orthotopic tumors of nude mice (n = 7, value indicated with mean ± SD).

Id-1 VEGF-C PLVD

Control

NC-siRNA

Id-1-siRNA

P-value

5.43 ± 0.98 3.57 ± 0.53 23.71 ± 4.61

5.52 ± 0.53 3.43 ± 0.53 22.29 ± 5.41

3.71 ± 0.49 2.57 ± 0.53 15.43 ± 1.62

0.003 0.018 0.001

Table 1 Primers sequences and size of PCR product. Gene

Accession No.

Primer sequence

PCR protocol

Size (bp)

Id-1

NM002165

50 -TCTACGACATGAACGGCTG-30 50 -GGTCCCTGATGTAGTCGAT-30

94 °C for 1 m, followed by 40 cycles of 94 °C for 20 s

117

VEGF-C

NM005429

50 -CACTTGCTGGGCTTCTTCT-30 50 -CACAGACCGTAACTGCTCCT-30

52 °C (Id-1), 54 °C (VEGF-C), or 50 °C (b-actin) for 30 s

186

b-Actin

NM001101

50 -GCCAACACAGTGCTGTCT-30 50 -AGGAGCAATGATCTTGATCTT-30

72 °C for 30 s, then 60 °C for 5 m

114

Z. Dong et al. / Oral Oncology 47 (2011) 27–32

Results Id-1 expression was correlated with VEGF-C expression and PLVD in OSCC tumor samples In this study, we evaluated 128 tumor samples of OSCC by immunohistochemistry. The expression of Id-1 and VEGF-C was found in most tumors10 (Fig. 1A and B). There was a significant correlation between the expression of Id-1 and VEGF-C (r = 0.569, p < 0.001) (Fig. 1A and B). Tumors with higher Id-1 expression also showed higher VEGF-C expression (3.08 ± 0.76 vs. 1.87 ± 1.30, p < 0.01). Lymphatic vessels were detected in most tumor samples. The intratumoral lymphatic vessels were small and collapsed, while the peritumoral ones were large and dilated (Fig. 1C and D). According to previous report,14 we used PLVD to assess lymphangiogenesis in this study. We found that tumors with higher Id-1 expression also showed significantly higher PLVD (16.06 ± 6.16 vs. 13.21 ± 5.53, p = 0.012), there was a significant correlation between Id-1 expression and PLVD (r = 0.240, p < 0.001) (Fig. 1A and D). Lentivirus-mediated siRNA suppressed Id-1 expression in Tca8113 cells Real-time PCR and Western blotting were performed to determine the transfection efficacy of lentivirus. We found that there was no difference of Id-1 expression between the control group and the NC-siRNA group, but there was significant difference between the Id-1-siRNA group and the control group (p < 0.01). The results showed that Id-1-siRNA-lentivirus could significantly inhibit Id-1 expression (Fig. 2A and C). Silencing Id-1 down-regulated the expression of VEGF-C in Tca8113 cells By real-time PCR, we found that the expression of VEGF-C mRNA decreased in the Id-1-siRNA group compared with the control group (p = 0.003) (Fig. 2B). The expression of VEGF-C protein

29

was assessed by Western blotting and ELISA. By Western blotting, VEGF-C protein decreased obviously in the Id-1-siRNA group (Fig. 2C). By ELISA, the amount of secretory VEGF-C protein was relatively fewer in the Id-1-siRNA group than in the control group (p = 0.007). The above data showed that Id-1 suppression led to down-regulation of VEGF-C in Tca8113 cells, both in mRNA and protein levels. Silencing Id-1 inhibited lymphangiogenesis in vivo The tongue orthotopic tumor models were successfully established (Fig. 3A and B). RT-PCR results showed that the expression of Id-1 mRNA decreased in the Id-1-siRNA group (p < 0.01) (Fig. 4A), and the expression of VEGF-C mRNA decreased accordingly (p < 0.01) (Fig. 4B). Moreover, the expression of Id-1 and VEGF-C proteins decreased after injecting Id-1-siRNA-lentivirus into the tumors (p = 0.003, 0.018) (Fig. 5A–D). The lymphatic vessels localized mainly in the periphery of the tumor. Lymphatic vessels in the control group were large and dilated, and tumor embolus could even be found in the open vessel (Fig. 5E). But in the Id-1-siRNA group, the lymphatic vessels were small and collapsed (Fig. 5F). Tumors injected Id-1-siRNA-lentivirus showed significantly fewer PLVD than the control group (15.43 ± 1.62 vs. 23.71 ± 4.61, p = 0.001) (Table 2). Discussion It has been reported that high level of Id-1 expression is associated with lymph node status.8 Our previous study also found that overexpression of Id-1 was associated with lymphatic metastasis in OSCC.10 But the molecular mechanism is not clear. Lymphatic metastasis depends on lymphangiogenesis.2,3 VEGF-C, an important lymphangiogenic factor, plays a critical role in lymphangiogenesis. By expressing VEGF-C, cancer cells could induce lymphangiogenesis and promote lymphatic metastasis.4,5 In OSCC, VEGF-C could stimulate lymphangiogenesis, and tumors with higher VEGF-C expression were prone to metastasize via lymph ves-

Figure 1 Id-1 (A) and VEGF-C (B) were overexpressed in OSCC. Lymphatic vessels in the tumor were small and collapsed (C) while in the periphery ones were large and dilated (D). DAB (brown) served as chromogen. A and B, 40 and C and D, 64. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

30

Z. Dong et al. / Oral Oncology 47 (2011) 27–32

Figure 2 Id-1, VEGF-C mRNAs (A and B) and proteins (C) were down-regulated in the Id-1-siRNA group (n = 6, value indicated with mean ± SD).

Figure 3 The orthotopic tumor model was established (A) and tumors were showed 3 weeks after treatment (B) (n = 7).

Figure 4 Id-1 mRNA (A) and VEGF-C mRNA (B) were down-regulated in the Id-1siRNA group of the nude mice (n = 7, value indicated with mean ± SD).

sels.15,16 In the light of these studies, we hypothesized that Id-1 might be associated with lymphangiogenesis. To confirm this, we investigated the correlation between the expression of Id-1,

VEGF-C and PLVD in OSCC, which was further confirmed by in vitro and in vivo studies. We collected 128 OSCC tumor samples and observed the association between Id-1, VEGF-C and PLVD by immunohistochemistry. We found that there was a significant correlation between Id-1 expression and VEGF-C expression. Moreover, PLVD was higher in the tumors with higher Id-1 expression, which showed that Id-1 was associated with lymphangiogenesis. From these results we thought that Id-1 might promote lymphangiogenesis through activating the expression of VEGF-C, and silencing Id-1 might inactivate VEGF-C expression thereby inhibit lymphangiogenesis. To confirm this, we first performed in vitro test. We found that the expression of Id-1 was inhibited by lentivirus-mediated Id-1-siRNA in Tca8113 cells. With the suppression of Id-1, the expression of VEGF-C was down-regulated. Down-regulation of VEGF-C was found not only in mRNA level but also in protein level. Furthermore, the secretory VEGF-C decreased in culture medium, which meant that the ability of cancer cells to induce lymphangiogenesis decreased. These results suggested that silencing Id-1 might inhibit lymphangiogenesis via inactivating the expression of VEGF-C. To study the metastasis and the effect of agent that inhibits metastasis, the orthotopic tumor models are needed.17 In this study, the tongue orthotopic tumor models were successfully established, which reproduced the organ-specific clinical features of OSCC. RNA interference was performed by injecting Id-1-siRNA-lentivirus into the tumors. The results showed that Id-1 was

Z. Dong et al. / Oral Oncology 47 (2011) 27–32

31

Figure 5 Id-1 and VEGF-C were strongly expressed in the control group of the nude mice (A and C), weakly expressed in the Id-1-siRNA group (B and D). The peritumoral lymphatic vessels were large and dilated, and tumor embolus could be found in the control group (E), while in the Id-1-siRNA group the vessels were small and collapsed (F). DAB (brown) served as chromogen, 40. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

suppressed both in mRNA and protein levels, which meant that RNA interference in solid tumors was effective. Moreover, the expression of VEGF-C also decreased, which confirmed the results obtained from in vitro study. LVD was commonly used to assess lymphangiogenesis in cancers. Higher LVD always means more lymphangiogenesis and higher rate of lymphatic metastasis.3 In this study, PLVD was used to assess lymphangiogenesis in nude mice models, we found that lymphatic vessels localized mainly in the periphery of the transplanted tumors and tumor embolus could be found in the lymphatic vessels. This suggested that the transplanted tumor could invade the lymphatic vessels and metastasize via them. But in the tumors injected Id-1-siRNA-lentivirus, we did not find any tumor embolus in the lymphatic vessels, and the PLVD was less compared to the tumors without injection. This indicated that silencing Id-1 could inhibit lymphatic metastasis by reducing lymphangiogenesis in transplanted tumors. Since the expression of VEGF-C down-regulated simultaneously, we thought that the mechanism that silencing Id-1 could inhibit tumor lymphatic metastasis might be that the expression of VEGF-C was down-regulated and its ability to promote lymphangiogenesis decreased. In conclusion, we demonstrated for the first time that Id-1 was associated with tumor lymphangiogenesis in OSCC. Silencing Id-1 could inhibit lymphangiogenesis through down-regulation of

VEGF-C. RNA interference targeting Id-1 might be a novel strategy in the treatment of lymphatic metastasis in OSCC. Conflict of interest statement None declared. Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 30672339) and the Project of Shandong Medical Science and Technological Development (No. 2009HZ044). References 1. Vokes EE, Weichselbaum RR, Lippman SM, Hong WK. Head and neck cancer. N Engl J Med 1993;328(3):184–94. 2. Yuan P, Temam S, El-Naggar A, Zhou X, Liu DD, Lee JJ, et al. Overexpression of podoplanin in oral cancer and its association with poor clinical outcome. Cancer 2006;107(3):563–9. 3. Zhao D, Pan J, Li XQ, Wang XY, Tang C, Xuan M. Intratumoral lymphangiogenesis in oral squamous cell carcinoma and its clinicopathological significance. J Oral Pathol Med 2008;37(10):616–25. 4. Miyahara M, Tanuma J, Sugihara K, Semba I. Tumor lymphangiogenesis correlates with lymph node metastasis and clinicopathologic parameters in oral squamous cell carcinoma. Cancer 2007;110(6):1287–94.

32

Z. Dong et al. / Oral Oncology 47 (2011) 27–32

5. Liang X, Zhou H, Liu X, He Y, Tang Y, Zhu G, et al. Effect of local hyperthermia on lymphangiogenic factors VEGF-C and -D in a nude mouse xenograft model of tongue squamous cell carcinoma. Oral Oncol 2010;46(2):111–5. 6. Ouyang XS, Wang X, Lee DT, Tsao SW, Wong YC. Over expression of ID-1 in prostate cancer. J Urol 2002;167(6):2598–602. 7. Langlands K, Down GA, Kealey T. Id proteins are dynamically expressed in normal epidermis and dysregulated in squamous cell carcinoma. Cancer Res 2000;60(21):5929–33. 8. Nishimine M, Nakamura M, Mishima K, Kishi M, Kirita T, Sugimura M, et al. Id proteins are overexpressed in human oral squamous cell carcinomas. J Oral Pathol Med 2003;32(6):350–7. 9. Schoppmann SF, Schindl M, Bayer G, Aumayr K, Dienes J, Horvat R, et al. Overexpression of Id-1 is associated with poor clinical outcome in node negative breast cancer. Int J Cancer 2003;104(6):677–82. 10. Dong Z, Liu S, Zhou C, Sumida T, Hamakawa H, Chen Z, et al. Overexpression of Id-1 is associated with tumor angiogenesis and poor clinical outcome in oral squamous cell carcinoma. Oral Oncol 2010;46(3):154–7. 11. Schoppmann SF, Schindl M, Bayer G, Aumayr K, Dienes J, Horvat R, et al. Overexpression of Id-1 is associated with poor clinical outcome in node negative breast cancer. Int J Cancer 2003;104(6):677–82.

12. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis-correlation in invasive breast carcinoma. N Engl J Med 1991;324(1):1–8. 13. De la Torre NG, Buley I, Wass JA, Turner HE. Angiogenesis and lymphangiogenesis in thyroid proliferative lesions: relationship to type and tumour behaviour. Endocr Relat Cancer 2006;13(3):931–44. 14. Gombos Z, Xu X, Chu CS, Zhang PJ, Acs G. Peritumoral lymphatic vessel density and vascular endothelial growth factor C expression in early-stage squamous cell carcinoma of the uterine cervix. Clin Cancer Res 2005;11(23): 8364–71. 15. Siriwardena BS, Kudo Y, Ogawa I, Udagama MN, Tilakaratne WM, Takata T. VEGF-C is associated with lymphatic status and invasion in oral cancer. J Clin Pathol 2008;61(1):103–8. 16. Shintani S, Li C, Ishikawa T, Mihara M, Nakashiro K, Hamakawa H. Expression of vascular endothelial growth factor A, B, C, and D in oral squamous cell carcinoma. Oral Oncol 2004;40(1):13–20. 17. Kim S. Animal models of cancer in the head and neck region. Clin Exp Otorhinolaryngol 2009;2(2):55–60.