Radiation induced MMP expression from rectal cancer is short lived but contributes to in vitro invasion

EJSO (2005) 31, 869–874

www.ejso.com

Radiation induced MMP expression from rectal cancer is short lived but contributes to in vitro invasion W.J. Speakea, R.A. Deana, A. Kumara, T.M. Morrisa, J.H. Scholefieldb, S.A. Watsona,* a

Academic Unit of Cancer Studies, Queen’s Medical Centre, University Hospital, D Floor, West Block, Nottingham NG7 2UH, UK b Division of GI Surgery, University of Nottingham, Nottingham NG7 2UH, UK Accepted for publication 31 May 2005 Available online 2 August 2005

KEYWORDS Matrix metalloproteinases; Rectal cancer; Radiation

Abstract Aims: Matrix metalloproteinase (MMP) activity is increased after radiation. The aims of this study were to assess the time course of this increase and its effects on malignant cell invasion. Methods: Colorectal cancer (HCT116, LoVo, C170HM2, CaCO-2), fibroblast (46-BR. IGI, CCD-18Co) and fibrosarcoma (HT1080) cell lines were irradiated at 4 gray (4 Gy) and matrix metalloproteinase gene and protein expression examined over a 96 h period by real time polymerase chain reaction and gelatin zymography. Invasion was assessed on Matrigel. Human rectal tumour MMP expression was compared before and after long course radiotherapy. Results: Radiation increased MMP gene expression of tumour cell lines, and resulted in increased MMP protein activity in the HT1080 line. HT1080 and HCT116 in monoculture and LoVo in co-culture were more invasive after radiation at 48 h in vitro, but long course radiotherapy did not result in a consistent increase in MMP expression from human rectal tumour biopsies. Conclusions: Radiation results in increased MMP expression for a limited time period. This results in an early increase in cell line invasion. Further clinical research is required to clarify if MMP inhibition given perioperatively following radiotherapy decreases local recurrence rates. Q 2005 Elsevier Ltd. All rights reserved.

Introduction * Corresponding author. Tel.: C44 115 9709248; fax: C44 115 9709902. E-mail address: [email protected] (S.A. Watson).

Matrix metalloproteinases (MMP) are proteolytic enzymes that have key roles in metastasis. Initially MMPs were thought to simply breakdown components

0748-7983/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2005.05.016

870 of the extracellular-matrix thus allowing invasion and metastasis, however, recently extended roles have been confirmed.1 These include precise MMP localization on a cells invasive front, the exposure of key components in the extracellular-matrix turning it from a barrier into a scaffold for invasion, and cleavage of free insulin-like growth factors resulting in cell growth, division and a prevention of apoptosis.1–5 In vitro and in vivo experiments have demonstrated clear benefits of MMP inhibition in limiting tumour progression with particularly encouraging results when cytostatic MMP inhibitors are used in combination with cytotoxic agents.6 Unfortunately the anticipated success of MMP inhibitors (MMPI) in Phase III trials has not been realized. The drugs were used on patients with advanced disease and therefore the processes where they might have been beneficial may have already occurred. This has been borne out by the use of marimastat (an orally active hydroxamate MMPI) in advanced gastric cancer, in subgroups where disease volume was low and, with the addition of chemotherapy there were significant improvements in survival.7 An adverse effect of MMPIs, in the form of musculoskeletal side-effects has resulted in dose reduction in the majority of trials which may have resulted in suboptimal tissue levels.8,9 This was demonstrated in a randomised controlled trial of Marimastat versus placebo for inoperable colorectal hepatic metastases. In patients who developed musculoskeletal side-effects there was a significant survival advantage over those who did not.10 Pre-operative short course radiotherapy decreases local recurrence rates with optimal rectal cancer surgery from 8.2 to 2.4% and has also been shown to improve survival.11,12 In spite of these proven benefits we have previously shown an increase in the expression and activation of gelatinase MMPs from rectal cancer biopsies after pre-operative short course radiotherapy.13,14 The clinical importance of this finding is unclear, although potentially, additional MMP inhibition may reduce local recurrence further. This study investigates the time course of this increase in MMPs after radiation from cell lines and human rectal tumour biopsies, and studies the effects on invasion in vitro.

W.J. Speake et al. epidermal fibroblast (46-BR.IGI) and a fibrosarcoma (HT1080) cell line were used for the study. All cell lines were human in origin, and originated from within our laboratory or from the European collection of animal cultures (ECACC), Wiltshire, UK. Cells were cultured using standard techniques. Irradiation Cells were irradiated at 4 gray (Gy) using a 137 Caesium source (Gammacell 1000 Elite, Nordion, Kanata, Canada). A radiation dose of 4 Gy was chosen as this dose had been found by flow cytometry to have maximal effects on the cell cycle without excessive death (data not shown). Gelatin zymography Zymography was performed as previously described15 using precast zymograms and reagents (Novex, Frankfurt, Germany). This is a semiquantitative electrophoretic process which distinguishes active from inactive gelatinase MMPs. RNA extraction and reverse transcription Total RNA was extracted by standard techniques using RNAzol B (Biogenesis, Poole, Dorset). From each sample of RNA, positive and negative reverse transcription samples were produced using pd(N)6 sodium salt random hexamer primer (Amersham Pharmacia Biotech) and a commercially available reaction mix (GibcoBRLe, Paisley, UK). Real time polymerase chain reaction Real time PCR was performed on a GeneAmpw 5700 Sequence Detection System (Applied Biosystems, CA, USA). Primer sequences were designed with the use of Primer Expressw software (Applied Biosystems) for MMP-2 and MMP-9. Reaction mix constituents and plates were supplied by PE Biosystems (Warrington, UK). Gene expression was compared to the housekeeping gene glyceraldhyde-3-phosphate dehydrogenase (GAPDH). Matrigel invasion assay Matrigel (Becton Dickinson Labware, Two Oak Park, Bedford, MA 01730) inserts were used under standard conditions. The plates were incubated for 48 h and invaded cells were retrieved and counted.

Materials and methods Clinocopathological studies In vitro studies Cell culture and reagents Four colorectal cancer (C170HM2, LoVo, HCT116, Caco-2), one colonic fibroblast (CCD-18Co), one

Biopsy samples Hospital Ethics Committee approval was sought and obtained for the study. Patients with histological confirmed rectal adenocarcinomas who were

Radiation induced MMP expression from rectal cancer undergoing long course radiotherapy (25 fractions of 2 Gy) were recruited. Tumour and corresponding normal mucosa biopsies from within the radiotherapy field were collected under direct vision by a single investigator using a rigid sigmoidoscope. Post-radiotherapy samples were collected at a minimum of 4 weeks after the radiotherapy course. Samples were immediately snap frozen in liquid nitrogen and stored at K85 8C until the time of analysis.

Statistical analysis Data were analysed using Mann–Whitney and Wilcoxon tests as all data were non-parametric. Averages are means and experiments were repeated at least three and generally four times.

Results Cell line studies Effect of irradiation on MMP gene expression Irradiation increased MMP gene expression in all the tumour cell lines studied and, with the exception of C170HM2, this appeared to peak at earlier time periods and fall back to control levels by day 4 (Table 1). Irradiation had no significant effect on the 46-BR.IGI fibroblast cell line (data not shown). The colonic fibroblast cell line (CCD-18Co) only expressed MMP-2, and in decreased levels after radiation. Table 1 Irradiation increases MMP gene expression from colorectal cancer cell lines in a time-limited manner after exposure to 4 Gy of irradiation Cell line

MMP-2 expression time (days) 1 2 3

4

HT1080 LoVo C170HM2 CaCo-2

2.9 3.5 3.3 2.2

1.1 2.9 5.6 1.4

Cell line

MMP-9 expression time (days)

HT1080 HCT116 C170HM2 CaCo-2

1.6 4.4 2.5 1.8

1.4 7.7 1.3 2

1

2

3

4

15.3 1.9 0.7 1.6

6.3 14 4.6 0.5

1 2.7 0.3 1.11

2.3 0.2 1.4 0.6

HCT116 only expressed MMP-9, and LoVo only expressed MMP-2 in quantifiable levels (Figures in bold correspond to p!0.05).

871 Effect of irradiation on MMP-2 and MMP-9 protein expression of cancer cell lines Maximal increases in HT1080 Pro-MMP-9 and ProMMP-2 were seen at 1 and 2 days, with a 1.6-fold (p!0.05) increase at these timepoints. At later timepoints smaller increases were seen but these were not significant. In monoculture, MMP-2 activation was also significantly increased at 48 h. In the colorectal tumour cell lines, gelatinase MMP protein expression was very low and it was not possible to quantify any change. MMP expression from both fibroblast lines was unaffected by radiation. Effect of coculture of tumour cells and fibroblasts on activation status of MMP-2 following irradiation In an attempt to examine stromal interactions, cocultures of tumour cells with fibroblasts were performed. The fibrosarcoma cell line HT1080 was cocultured with skin fibroblasts 46-BR.IGI in a 50:50 coculture, MMP expression was examined 48 h after 4 Gy irradiation as this time period had the maximal effects in the above study. Similar increases in ProMMP-2, Pro-MMP-9 and in activation of MMP-2 were seen (Fig. 1). Therefore, it appears that activation in this cell line was not dependent on, or increased by addition of the fibroblasts. Colorectal cell lines were co-cultured with varying ratios of the colonic fibroblast cell line CCD-18Co. Each cell line had maximal increase in MMP expression after radiation at different ratios, although none were significant or consistent across cell lines (data not shown). Effect of irradiation on invasion of tumour cells on Matrigel The invasive potential of tumour cell lines on Matrigel after radiation at 4 Gy was examined. Cells were studied in monoculture and in coculture as described above. Radiation at 4 Gy decreased

Figure 1 Co-culture of HT1080 and 46-BR.IGI cell lines at 48 h following 4 Gy irradiation (*pZ0.04).

872

W.J. Speake et al.

viable tumour cell numbers by approximately 50% at 48 h but in three of the four cell lines studied the invasive potential of the surviving cells was increased. HT1080 in monoculture had a 1.5-fold increase in invasion (pZ0.04) after irradiation, in coculture the increase in invasion was smaller (1.2fold) but still significant (pZ0.04) suggesting that it was the tumour cells that were invading. HCT116 was invasive on Matrigel in monoculture and had increased invasion after irradiation (1.7-fold, pZ0.01), this effect was not seen in coculture with fibroblasts. LoVo was poorly invasive in monoculture but had dramatically increased invasion (2.7-fold) on Matrigel when cocultured with fibroblasts. C170HM2, both in mono and coculture, and the fibroblast cell lines all had relatively low basal invasive potential that was unchanged by radiation.

Clinical material Effect of irradiation on MMP protein expression and activation after long course radiotherapy Six complete sets of rectal tumour and normal mucosa biopsies were collected from patients before and after long course radiotherapy. The median time to the post-radiotherapy biopsy was 62 days (range 28–80). Normal mucosa only expressed Pro-MMP-2 and Pro-MMP-9 and their levels were low (as compared to tumour samples) and not affected by radiotherapy. Table 2 shows that there was a variable response in MMP protein expression at this delayed period after the radiotherapy insult. This suggests that the effects of radiation on MMP expression are short lived being limited to a matter of days as opposed to weeks (further evidence of this is seen in Table 2 where the two earliest biopsies had an increase in all three

MMP subtypes studied, this not being seen consistently in the later biopsies).

Discussion In this study in vitro work has shown a self-limiting increase in MMP gene expression by colorectal tumour cell lines following radiation. Where MMP protein expression and activation was expressed in detectable levels, by the HT1080 cell line, a time limited increase was also seen. The viable cell population surviving radiation showed an increased rate of invasion on Matrigel in three out of four tumour cell lines. It appeared that increased MMP expression is self-limiting in the clinical scenario, as human rectal tumour biopsies taken at least 4 weeks after long course radiotherapy did not express increased levels of MMPs. This compares to a significant increase seen at early time periods (2–3 days) after short course radiotherapy.13 Several recent studies have demonstrated an increase in gelatinase (MMP-2 and MMP-9) MMP expression after radiotherapy. Nirmala has shown an increase in pro MMP-2 and MMP-9 from glioblastoma cell lines after 10 Gy radiation at 72 h.16 Pro-MMP-2, Pro-MMP-9 and active MMP-2 have all been shown to increase in benign rectal mucosal biopsies in patients undergoing radiotherapy for prostate carcinoma.17 The latter study demonstrated a 2–3-fold increase during the radiotherapy course but did not examine the expression once the treatment was concluded. A similar increase has been reported from colorectal cancer and fibrosarcoma cell lines with 5 Gy of radiation at short time periods up to 24 h.18 Two studies have examined the effects of radiation on invasion. Two out of three pancreatic cell lines had increased invasion on Matrigel at

Table 2 Expression of MMP protein before and after pre-operative long course radiotherapy from human rectal tumour biopsies Patient no.

1 2 3 4 5 6 p value

Days from radiotherapy to biopsy

55 69 80 29 69 28

Pre-radiotherapy MMP expression

Post-radiotherapy MMP expression

Pro-MMP-9

Pro-MMP-2

ActiveMMP-2

Pro-MMP-9

Pro-MMP-2

Active MMP-2

100 68 164 77 157 102

61 9 22 26 39 51

75 0 ND 25 73 76

37 44 160 100 46 142 0.3

14 14 80 39 41 73 0.6

12 7 ND 46 12 46 0.2

ND, none detected, units correspond to band density. No active MMP-9 was detected.

Radiation induced MMP expression from rectal cancer doses of radiation up to 10 Gy, this was dependent on MMPs as the addition of an MMP inhibitor blocked the effect.19 Wild-Bode similarly demonstrated an increased MMP expression and invasion on Matrigel from three glioma cell lines after 3 and 6 Gy at 24 h. Effects were again blocked with the addition of an MMP inhibitor.20 Potential mechanisms exist after radiation that can upregulate MMPs. The tumour suppressor gene p53 has been said to be crucial by some, but not all, investigators.18,21 Other molecular pathways include the transcription factors NFkB and activator protein 1 (AP-1), and the downstream effects of various interleukins and growth factors.21–25 The novel finding of the limited time course of increased MMP expression in cell lines and in human rectal tumour biopsies raises questions as to its clinical relevance. A short response may allow a time window for tumour cells to invade as demonstrated by the increased tumour cell invasion in vitro. It is important to bear in mind the strict radiation dose and timepoints of the in vitro studies and because of this direct clinical extrapolation may not be appropriate. The in vitro data does however mirror the clinical scenario in that in both situations MMPs are increased briefly at similar timepoints. A temporary increase in MMPs shortly after radiotherapy with an attendant increase in invasion suggests they may contribute to (re) invasion of micrometastases or spilt tumour cells at the time of surgery. Clearly not all of these tumour cells will have long-term viability, but some may, suggesting that temporary MMP inhibition around the time of radiotherapy may have additional benefits. Recently developed MMP inhibitors, that reach adequate tissue levels with improved adverse effect profiles, and given at an appropriate stage in the disease process, could further improve the role of adjuvant radiotherapy with surgery. Further studies need to be performed to clarify if MMP inhibition following radiotherapy would decrease tumour cell invasion in a clinical scenario.

References 1. Stetler-Stevenson WG, Yu AE. Proteases in invasion: Matrix metalloproteinases. Semin Cancer Biol 2001;11: 143—52. 2. Stetler-Stevenson WG. Matrix metalloproteinases in angiogenesis: A moving target for therapeutic intervention. J Clin Invest 1999;103:1237—41. 3. Fowlkes JL, Enghild JJ, Suzuki K, Nagase H. Matrix metalloproteinases degrade insulin-like growth factor-binding protein-3 in dermal fibroblast cultures. J Biol Chem 1994; 269:25742—6. 4. Fowlkes JL, Suzuki K, Nagase H, Thrailkill KM. Proteolysis of insulin-like growth factor binding protein-3 during rat pregnancy: A role for matrix metalloproteinases. Endocrinology 1994;135:2810—3.

873 5. Manes S, Llorente M, Lacalle RA, Gomez-Mouton C, Kremer L, Mira E, et al. The matrix metalloproteinase-9 regulates the insulin-like growth factor-triggered autocrine response in DU-145 carcinoma cells. J Biol Chem 1999;274:6935—45. 6. Watson SA, Tierney GM. Matrix metalloproteinase inhibitors. BioDrugs 1998;9:325—35. 7. Fielding J, Scholefield J, Stuart R, Hawkins R, McCulloch P, Maughan T, et al. A randomized double-blind placebocontrolled study of marimastat in patients with inoperable gastric adenocarcinoma. Proc ASCO 2000;19:240. 8. Tierney GM, Griffin NR, Stuart RC, Kasem H, Lynch KP, Lury JT, et al. A pilot study of the safety and effects of the matrix metalloproteinase inhibitor marimastat in gastric cancer. Eur J Cancer 1999;35:563—8. 9. Primrose JN, Bleiberg H, Daniel F, Van Belle S, Mansi JL, Seymour M, et al. Marimastat in recurrent colorectal cancer: Exploratory evaluation of biological activity by measurement of carcinoembryonic antigen. Br J Cancer 1999;79:509—14. 10. King J, Zhao J, Clingan P, Morris D, Wick W, Wick A, et al. Randomised double blind placebo control study of adjuvant treatment with the metalloproteinase inhibitor, marimastat in patients with inoperable colorectal hepatic metastases: Significant survival advantage in patients with musculoskeletal side-effects. Anticancer Res 2003;23:639—45. 11. Improved survival with preoperative radiotherapy in resectable rectal cancer. Swedish Rectal Cancer Trial. N Engl J Med 1997; 336:980—987. 12. Kapiteijn E, Marijnen CA, Nagtegaal ID, Putter H, Steup WH, Wiggers T, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001;345:638—46. 13. Kumar A, Collins HM, Scholefield JH, Watson SA. Increased type-IV collagenase (MMP-2 and MMP-9) activity following preoperative radiotherapy in rectal cancer. Br J Cancer 2000;82:960—5. 14. Kumar A, Collins H, Van Tam J, Scholefield JH, Watson SA. Effect of preoperative radiotherapy on matrilysin gene expression in rectal cancer. Eur J Cancer 2002;38:505—10. 15. Heussen C, Dowdle EB. Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates. Anal Biochem 1980;102:196—202. 16. Nirmala C, Jasti SL, Sawaya R, Kyritsis AP, Konduri SD, AliOsman F, et al. Effects of radiation on the levels of MMP-2, MMP-9 and TIMP-1 during morphogenic glial-endothelial cell interactions. Int J Cancer 2000;88:766—71. 17. Hovdenak N, Wang J, Sung CC, Kelly T, Fajardo LF, HauerJensen M, et al. Clinical significance of increased gelatinolytic activity in the rectal mucosa during external beam radiation therapy of prostate cancer. Int J Radiat Oncol Biol Phys 2002;53:919—27. 18. Wang JL, Sun Y, Wu S. Gamma-irradiation induces matrix metalloproteinase II expression in a p53-dependent manner. Mol Carcinog 2000;27:252—8. 19. Qian LW, Mizumoto K, Urashima T, Nagai E, Maehara N, Sato N, et al. Radiation-induced increase in invasive potential of human pancreatic cancer cells and its blockade by a matrix metalloproteinase inhibitor, CGS27023. Clin Cancer Res 2002;8:1223—7. 20. Wild-Bode C, Weller M, Rimner A, Dichgans J, Wick W. Sublethal irradiation promotes migration and invasiveness of glioma cells: Implications for radiotherapy of human glioblastoma. Cancer Res 2001;61:2744—50. 21. Araya J, Maruyama M, Sassa K, Fujita T, Hayashi R, Matsui S, et al. Ionizing radiation enhances matrix metalloproteinase2 production in human lung epithelial cells. Am J Physiol Lung Cell Mol Physiol 2001;280:L30—L8.

874 22. Fisher GJ, Talwar HS, Lin J, Lin P, McPhillips F, Wang Z, et al. Retinoic acid inhibits induction of c-Jun protein by ultraviolet radiation that occurs subsequent to activation of mitogen-activated protein kinase pathways in human skin in vivo. J Clin Invest 1998;101:1432—40. 23. Bond MFR, Baker AH, Newby AC. Synergistic upregulation of metalloproteinase-9 by growth factors and inflammatory

W.J. Speake et al. cytokines: An absolute requirement for transcription factor NF-kappa B. FEBS Lett 1998;435:29—34. 24. Kheradmand F, Werner E, Tremble P, Symons M, Werb Z. Role of Rac1 and oxygen radicals in collagenase-1 expression induced by cell shape change. Science 1998;280:898—902. 25. Nagase H, Woessner Jr JF. Matrix metalloproteinases. J Biol Chem 1999;274:21491—4.