Tumor Biol. (2011) 32:189–196 DOI 10.1007/s13277-010-0112-y

RESEARCH ARTICLE

Changes in CXCL12/CXCR4-chemokine expression during onset of colorectal malignancies Vilma Oliveira Frick & Claudia Rubie & Pirus Ghadjar & Sabrina K. Faust & Mathias Wagner & Stefan Gräber & Martin K. Schilling

Received: 2 August 2010 / Accepted: 9 September 2010 / Published online: 24 September 2010 # International Society of Oncology and BioMarkers (ISOBM) 2010

Abstract Chemokines have been proposed to contribute to tumour growth and metastatic spread of several cancer entities. Here, we examined the relative levels of CXCL12/ CXCR4 in resection specimens from patients with different malignant and non-malignant colorectal diseases as well as colorectal liver metastases (CRLM). CXCL12/CXCR4 mRNA and protein expression profiles were assessed by quantitative real-time PCR, Western blot analysis, enzymelinked immunosorbent assay (ELISA) and immunohistochemistry in resection specimens from patients with ulcerative colitis (UC; n=15), colorectal adenoma (CRA; n = 15), colorectal adenocarcinoma (CRC; n = 47) and CRLM (n=16). Corresponding non-affected tissues served as control. In contrast to UC tissues, CXCL12 showed a distinct down-regulation in CRA, CRC and CRLM specimens, whereas the corresponding receptor CXCR4 demonstrated a significant up-regulation in CRC and CRLM

V. Oliveira Frick (*) : C. Rubie : S. K. Faust : M. K. Schilling Department of General, Visceral, Vascular and Pediatric Surgery, University of the Saarland, Building 57, Homburg/Saar 66421, Germany e-mail: [email protected] P. Ghadjar Department of Radiation Oncology, Inselspital, Bern University Hospital and University of Bern, Bern 3010, Switzerland M. Wagner Institute of Pathology, University of the Saarland, Homburg/Saar 66421, Germany S. Gräber Institute of Medical Biometrics, Epidemiology, and Medical Informatics (IMBEI), University of the Saarland, Homburg/Saar 66421, Germany

related to corresponding non-affected tissues (p<0.05, respectively). Our results strongly suggest an association between CXCL12/CXCR4 expression and the induction of CRA, CRC and the development of CRLM. Therefore, CXCR4 may be a potential target for specific therapeutic interventions. Keywords Chemokines . Colorectal cancer . Colorectal liver metastases

Introduction Despite new chemotherapeutic regimes and improved surgical outcomes, colorectal cancer (CRC) remains one of the three leading causes of cancer-related death among men and women worldwide [1]. Although the 5-year survival rate for patients with local and early stage CRC approaches 90%, the spread of the disease to distant sites decreases the 5-year survival rate dramatically to 19% [1–3]. Currently, tumour growth and metastatic dissemination are accepted as a result of a complex, dysregulated molecular machinery leading to several phenomena, such as the resistance of tumour cells to apoptosis, tumour cell migration, tumour cell invasion, and tumour cell immune escape mechanisms. Regarding these aspects, recent data suggest that chemokines and chemokine receptors may direct lymphatic and hematogenous spreading and may additionally influence the sites of metastatic growth of different tumours [4]. Originally, chemokines and their G protein-coupled receptors were reported to mediate different pro- and anti-inflammatory responses [5]. As demonstrated by Muller et al. [6], two chemokine receptors, i.e. CXCR4 and CCR7, were involved in the metastasis of human breast cancer cells to distant organs that express and

190

secrete their respective ligands, CXCL12 and CCL21, thus, showing that the expression of chemokine receptors in these malignancies is not random. Various surveys demonstrated the importance of CXCR4 for the metastasis of kidney cancer [7], acute myeloid leukaemia [8] and prostate cancer, where it is likely to be involved in the ability of this cancer to migrate to the bone marrow [9]. Other studies observed a high expression of CXCR4 in CRC [10], glioblastoma [11], pancreatic cancer [12], non-Hodgkin lymphoma [13] and non-small cell lung cancer, where metastasis could be blocked using neutralizing antibodies against mouse CXCL12 [14]. Moreover, recently CXCR4 has been proposed as a prognostic factor for patients with CRC [15], indicating that CXCR4 expression increases the risk for recurrence and for poor survival [16]. Furthermore, recent studies show that nuclear expression of CXCR4 is associated with advanced CRC [17] and that expression of CXCL12 and nuclear CXCR4 predicts lymph node metastasis in CRC [18] Despite the increasing number of studies indicating a role for CC- and CXC-chemokines in different cancer types, it still remains unclear, whether chemokine expression is related to cancer induction, progression and the metastatic potential in colorectal carcinoma. Therefore, the purpose of the study was to examine the expression profile of CXCL12/CXCR4 across the inflammatory disease to the adenoma and the adenocarcinoma sequence and to evaluate their role in the regulation of CRC and the development of colorectal liver metastasis (CRLM).

Material and methods Patient selection Surgical specimens and corresponding normal tissue from the same samples were collected from patients with ulcerative colitis (UC; n=15), colorectal adenomas (CRA; n=15), colorectal carcinomas of different tumour categories (CRC; n = 47) and primary colorectal tumours with corresponding synchronous or metachronous CRLM (n =14 and 16, respectively) who underwent surgical resection at our department between 2002 and 2006. The study was approved by the local ethics commission of the Ärztekammer des Saarlandes, and written informed consent for tissue procurement was obtained from all patients. The clinical data and patient characteristics for the different malignant and non-malignant entities were obtained prospectively from the clinical and pathological records and are summarized in Tables 1 and 2. The data obtained are in accordance with the UICC/TNM classification [19].

Tumor Biol. (2011) 32:189–196

Tissue preparation Tissue specimens were collected immediately after surgical resection, snap frozen in liquid nitrogen and then stored at −80°C until they were processed under nucleic acid sterile conditions for RNA extraction. Tumour samples were taken from vital areas of histopathologically confirmed CRC and CRLM, respectively. As corresponding normal tissue we used adjacent unaffected mucosa, 2–3 cm distal to the resection margin from the same resected adenocarcinoma or liver specimens, respectively. All tissues obtained were reviewed by an experienced pathologist and examined for the presence of tumour cells. As minimum criteria for usefulness for our studies, we only used tumour tissues in which tumour cells constituted at least >80% of the tumour biopsy. Isolation of total protein Protein lysates from frozen tissue were extracted with the radioimmunoprecipitation buffer containing Complete, a protease inhibitor cocktail (Roche Diagnostics, Penzberg, Germany). Total protein quantification was performed

Table 1 Clinical characteristics of patients with colorectal carcinomas and colorectal liver metastases Factor Localization of primary tumour Colon Rectum Gender Male Female Age (years)d Largest tumour diameter (cm)d UICC stadium of primary tumour I II III IV Grading I II III Chemotherapy before operation Radiotherapy before operation a

CRCa n=47

CRLMb n=16c

26 21

6 8

31 16 63.7 (47–78) 4.6 (1.2–9.1)

8 4 60.1 (41–76) 4.2 (1.5–5.5)

7 13

1 2

19 8

11 0

1 18 28 0 0

0 4 10 2 2

Colorectal carcinoma

b

Colorectal liver metastases

c

16 CRLM originating from 14 CRC patients

d

Median with range in parentheses

Tumor Biol. (2011) 32:189–196

191

Table 2 Clinical characteristics of patients with colorectal adenomas and ulcerative colitis Factor Localization of disease Colon Rectum

Gender Male Female Age (years)c a

CRAa n=15

UCb n=15

11 4 Tubulovillous CRA 13 Villous CRA 2

12 3

10 5 65.3 (41–75)

9 6 49.8 (23–78)

Colorectal adenoma

b

Ulcerative colitis

c

Median with range in parentheses

using the BCA protein assay reagent kit (Pierce, Rockford, Il, USA). Sandwich-type ELISA The chemokine protein levels in the different tissue lysates were determined by sandwich-type ELISA according to the manufacturer´s instructions. Samples were assayed in duplicate with all values calculated as the mean of the two measurements. CXCL12 levels were assayed using a validated commercial ELISA (DuoSet; R&D Systems, Minneapolis, MN, USA). The absorbance was read at 450 nm using a 96-well microtiter plate reader. The chemokine concentration from each tissue lysate was normalized to the total protein content of the sample. Isolation of total RNA and single-strang cDNA synthesis Total RNA isolation and the cDNA synthesis were performed as described earlier [10].

was omitted to assure absence of genomic DNA contamination in each RNA sample. For signal detection, the ABI Prism 7900 sequence detector (Applied Biosystems) was programmed to an initial step of 10 min at 95°C, followed by 40 thermal cycles at 15 s at 95°C and 10 min at 60°C and the log-linear phase of amplification was monitored to obtain CT values for each RNA sample. Gene expression of all target genes was analyzed in relation to the levels of the slope matched housekeeping gene cyclophilin C [21]. Data analysis was performed according to the relative standard curve method. Data are presented in relation to the respective housekeeping genes. Immunohistochemistry Operative specimens were routinely fixed in formalin and subsequently embedded in paraffin. Before staining, 4 μm paraffin-embedded tissue sections were mounted on Superfrost Plus slides, deparaffinised, and rehydrated in graded ethanol to deionized water. The sections were submitted to microwave antigen retrieval (Target Retrieval Solution, Dakocytomation, Carpinteria, CA, USA), and after blocking of the endogenous peroxidise activity with 3% hydrogen peroxide, the section were further blocked for 30 min at room temperature with normal horse or rabbit serum. Overnight incubation at 4°C with primary mouse monoclonal anti-human CXCL12 antibody (MAB350, 8 μg/ml, R&D Systems, Minneapolis, MN, USA), or primary goat polyclonal anti-human CXCR4 antibody (sc-6279, 10 μg/ml, Santa Cruz Biotechnology, Santa Cruz, CA, USA) was followed by incubation with secondary biotinylated horse anti-mouse or rabbit antigoat IgG antibody and avidin-biotin-peroxidase reaction (Vectastain Elite ABC kit; Vector Laboratories, Burlingame, CA, USA). After colour reaction with aminoethylcarbazole solution (Merck, Darmstadt, Germany), tissues were counterstained with haematoxylin. Negative controls were performed in all cases, omitting primary antibody. Western blot analysis

Real-time PCR All Q-RT-PCR assays containing the primer and probe mix were purchased from Applied Biosystems (Applied Biosystems, Foster City, CA) and utilized according to the manufacturer´s intructions. PCR reactions were carried out using 10 μl 2×Taqman PCR Universal Master Mix No AmpErase® UNG (Applied Biosystems) and 1 μl gene assay, 8 μl RNase-free water and 1 μl cDNA template (50 ng/μl). The theoretical basis of the Q-RT assays is described in detail elsewhere [20]. All reactions were carried out in duplicate along with no template controls and an additional reaction in which reverse transcriptase

Total protein (25 μg/lane) was separated by SDS-PAGE using a 10% gel and blotted onto nitrocellulose membranes (Hybond ECL, Amersham Biosciences, Piscataway, NJ, USA). Membranes were blocked by incubation in Trisbuffered saline (TBS) containing 5% nonfat dry milk and 0.1% Tween 20 for 2 h at room temperature and then incubated overnight at 4°C with rabbit anti-human CXCR4 antibody (diluted 1:500, AHP442, Serotec, Oxford, UK). Blots were then washed and incubated at room temperature for 1 h with goat anti-rabbit HRP antibody (diluted 1:50,000, 170-6515, BioRad, Munich, Germany). Bands were visualized by ECL Western blotting analysis systems

192

(Amersham Biosciences, Piscataway, NJ, USA). The human cell lysate HL-60 (sc-2209, Santa Cruz Biotechnology, Santa Cruz, CA, USA) served as positive control. Calculations and statistical analysis The expression profiles of CXCL12/CXCR4 ligand/receptor system in the different malignant and non-malignant entities are presented as mean and standard error of the mean (SEM). All statistical calculations were done with the MedCalc (MedCalc software, Mariakerke, Belgium) software package [22]. The parametric Student’s t test was applied, if normal distribution was given, otherwise, the Wilcoxon’s rank-sum test was used. P values <0.05 at an α<0.05 were considered significant.

Tumor Biol. (2011) 32:189–196

bodies revealed weak to intermediate signals within the tumour lesions (Fig. 4c). In tumour neighbour tissues of CRLM we detected no substantial CXCR4 reactivity. Only insular punctual, cytoplasmic signals were randomly interspersed in some hepatocytes (Fig. 4d). In contrast, we observed intensive staining signals in epithelial and mesenchymal cells within the tumour invasion front of CRLM (Fig. 4e). CXCL12 immunostaining of non-diseased areas of CRA and CRC tissues displayed a specific staining pattern, where the apical intestinal epithelial cells displayed the highest CXCL12 expression and lower CXCL12 expression could be found in the less differentiated epithelial cells at the base of the crypts (Fig. 4f). In contrast, only very weak CXCL12 staining signals were found in the epithelium of CRA tissues (Fig. 4g) and in most CRC samples no substantial CXCL12 reactivity was detected (Fig. 4h). In tumour neighbour tissues of CRLM

Results Relative CXCR4 and CXCL12 expression, RNA level

n-fold gene expression related to normal tissue

Quantitative RT-PCR analysis comparatively performed on UC, CRA, CRC and CRLM specimens (n=15, 15, 47 and 16, respectively) showed significant down-regulation of CXC-chemokine ligand CXCL12 in CRA, CRC and CRLM in contrast to a significant overexpression of the respective chemokine receptor CXCR4 in CRC and CRLM samples related to the corresponding non-affected neighbour tissues (P<0.05, respectively) (Figs.1a, 2a). In consistence with the results obtained at the mRNA level, the CXCL12 protein expression, showed a significant down-regulation in the CRA, CRC and CRLM tissue specimens in comparison to the unaffected corresponding tissues, as demonstrated in Figs. 1b and 2b). As assessed by western blot analysis CXCR4 protein expression was detectable in all colorectal disease entities under investigation, namely in UC, CRA, CRC and CRLM tissue specimens as shown for representative patients in Fig. 3. The obtained protein expression data for CXCR4 confirmed the RNA transcript level analysis. To further identify the subset of cells expressing CXCL12 and CXCR4 the expression of CXCL12/CXCR4 has been evaluated by immunohistochemistry in human CRA, CRC and CRLM tissue samples and in the corresponding non-affected neighbour tissues. Immunohistochemical analyses revealed that CXCR4 was expressed in the corresponding normal areas of CRA and CRC tissue samples (Fig. 4a), where positive signals were observed in epithelial cells as well as to a lesser extent in mesenchymal cells. In contrast, CRA samples displayed only weak CXCR4 signals, which were restricted to a few areas (Fig. 4b) and in CRC specimens CXCR4-specific anti-

a 7 6 5

*

4 3 2

*

*

1 0 UC

CRA

CRC

CXCR4

b CXCL12 in pg/ml per mg total protein

CXCL12/CXCR4 are inversely expressed in CRA, CRC and CRLM

CXCL12

Absolute CXCL12 expression in UC, CRA and CRC, protein level 300

200

* *

100

0 UC

CRA

corresponding non-affected tissue

CRC diseased tissue

Fig. 1 Expression of CXCR4 and CXCL12 in ulcerative colitis (UC; n=15), colorectal adenoma (CRA, n=15) and colorectal carcinoma (CRC; n=47) tissue specimens as determined by a Q-RT-PCR and b enzyme-linked immunosorbent assay (ELISA). All data are expressed as mean±standard error of the mean. Fold increase above 1 in the QRT-PCR data indicates chemokine overexpression in affected tissues related to unaffected neighbour tissues. ELISA results are presented as absolute values of picograms per millilitre chemokine ligand per milligramme total protein in UC, CRA, and CRC tissue specimens and unaffected neighbour tissues, respectively. *P<0.05

Tumor Biol. (2011) 32:189–196

a

193

n-fold gene expression related to normal tissue

4

* *

3

2

1

*

*

primary CRC

Livermetastasis

CXCR4

b CXCL12 in pg/ml per mg total protein

growth. In order to evaluate the potential contribution of chemokines in disease pathophysiology, we analysed the expression of the CXC-chemokine member CXCL12 and its receptor CXCR4 in patients with CRC and CRLM and distinct inflammatory and non-malignant colorectal diseases, like UC and CRA, both representing known premalignant conditions often preceding the formation of colorectal malignancies. Here, we provide evidence that significantly downregulated CXCL12 expression associates with CRA, CRC and CRLM. In contrast, recent studies investigating the influence of CXCL12 expression on angiogenesis and vascularisation of colorectal tumours provide evidence that CXCL12 enhances these processes [23]. However, neutralization of CXCL12 after hepatectomy was shown to be not capable of inhibiting angiogenesis and growth of extrahepatic colorectal tumours because it is counteracted by compensatory actions through an alternative VEGF-dependent pathway. Moreover, blockade of the corresponding CXCL12 receptors, CXCR4 and CXCR7, significantly reduced tumour capillary density and tumour growth [24]. With respect to patient prognosis, Akishima-Fukasawa et al. demonstrated significantly shorter survival times for patients with high CXCL12 expression in combination with high-grade tumour budding in comparison to patients with low CXCL12 expression and low tumour budding grade. Thus, it was concluded that CXCL12 expression in CRC cells and sides of budding were significant prognostic factors [25]. In addition, a recent study observed that pretreatment of HEP-G2 cells with CXCL12 enhanced tumour cell extravasation and the activation of small GTPases [26]. Likewise, treatment of cells with an antiCXCR4 antibody impaired tumour cell extravasation [26]. In many studies, CXCR4 has been correlated with metastasis and poor prognosis in various tumour types [7–9, 16–18]. Here, we have shown that significant upregulation of CXCR4 is associated with the transition from premalignat conditions to CRC and CRLM. These findings are in line with previous studies indicating an inverse expression pattern for CXCL12 and CXCR4 in CRC [27], and CRLM specimens [28]. One study including normal mucosa, polyps, CRC and CRLM identified significant CXCR4 overexpression in cancerous lesions compared

CXCR4 and CXCL12 expression in CRC and CRLM, RNA level

CXCL12

Absolute CXCL12 expression in CRC and CRLM, protein level 200

150

100

* 50

* 0 primary CRC

Livermetastasis

corresponding non-affected tissue

diseased tissue

Fig. 2 Expression of CXCR4 and CXCL12 in primary colorectal tumours (CRC; n=14) and corresponding colorectal liver metastases (CRLM; n=16) as determined by a Q-RT-PCR and b ELISA. All data are expressed as mean±standard error of the mean. Fold increase above 1 in the Q-RT-PCR data indicates chemokine overexpression in affected tissues related to unaffected neighbour tissues. ELISA results are presented as absolute values of picograms per millilitre chemokine ligand per milligramme total protein in CRC and CRLM tissue specimens and unaffected neighbour tissues, respectively. *P<0.05

we observed medium to high CXCL12 staining intensities in the bile duct cells (Fig. 4i), while CRLM specimens were strictly negative for CXCL12 (Fig. 4j).

Discussion Chemokines are known to play an important role in the process of leukocyte trafficking and malignant tumour

Control kDa

HL60

UC N

CRA P

N

50

CRC P

N

CRLM P

N

P CXCR4

40 Fig. 3 Expression of chemokine receptor CXCR4 in UC, CRA, CRC and CRLM specimens as determined by Western blot analysis. Total cell lysates of tissues of representative patients of each disease entity

were immunoblotted with an antibody specifically recognizing chemokine receptor CXCR4. Acute leukaemia cell line HL60 served as a positive control for the detection of CXCR4

194

Tumor Biol. (2011) 32:189–196

Fig. 4 Detection of CXCR4 and CXCL12 protein expression in representative CRA, CRC and CRLM specimens as assessed by immunohistochemical staining with CXCR4- and CXCL12specific antibodies, respectively. a Weak cytoplasmic signals of CXCR4 in epithelial and mesenchymal cells within unaffected neighbour tissues of CRA and CRC, b weak CXCR4 immunostainig in CRA tissues, which was restricted to some areas, c weak to intermediate immunostaining of CXCR4 in epithelium of CRC tissues, d weak to no substantial CXCR4 reactivity in corresponding non-diseased neighbour tissues of CRLM and e intense CXCR4 reactivity in epithelial and mesenchymal cells within the invasion front of CRLM. CXCL12 staining revealed a similar distribution pattern in adjacent non-affected areas of CRA and CRC tissues, displaying the highest CXCL12 expression in the apical intestinal epithelial cells and lower CXCL12 expression in the less differentiated epithelial cells at the base of the crypts (f), weak to no substancial staining signals in the epithelium of CRA tissues (g), most CRC samples showed no substancial CXCL12 reactivity (h), medium to high staining intensities in bile duct cells in neighbour tissues of CRLM (i) and CRLM specimens were strictly negative for CXCL12 (j)

with non-cancerous samples [29]. Moreover, in dysplastic polyps CXCR4 was expressed at a higher level compared with hyperplastic polyps. We have also shown a significant change in CXCR4 expression from the non-malignant status to the malignant status. Thus, a significant change in

CXCR4 expression was observed from UC tissues to CRC and CRLM tissues as well as from CRA tissues to CRC and CRLM tissues. Moreover, immunohistochemical staining revealed weak to intermediate cytoplasmic immunostaining of CXCR4 in epithelium of CRC tissues and intense CXCR4

Tumor Biol. (2011) 32:189–196

reactivity in epithelial and mesenchymal cells within the invasion front of CRLM. With respect to the CRC tissues our findings correlate well with previous studies where predominantly a cytoplasmic staining of CRC specimens and in a few specimens also an additional weak membranous location of CXCR4 was reported [30]. In addition, recent findings also indicate nuclear CXCR4 staining of CRC specimens, which may be associated with advanced CRC [17] and the prediction of lymph node metastasis [18]. However, to date no other study described CXCR4 reactivity in cells within the tumour invasion front of CRLM. Thus, Kim et al. describe membrane and cytoplasmic patterns of immunostaining of CXCR4 protein in CRC cells within liver metastasis specimen and weak or no detectable staining in hepatocytes of normal liver [31]. In contrast, we observed intensive CXCR4 staining signals in epithelial and mesenchymal cells within the tumour invasion front of CRLM tissue specimens. In conclusion, we have demonstrated a significant upregulation in CXCR4 expression from the non-malignant status to the malignant status and a reverse expression pattern, a significant down-regulation in CXCL12 expression from the non-malignant status to the malignant status in colorectal diseases. Thus, our results strongly indicate involvement of the CXCL12/CXCR4 system in the carcinogenesis and induction of human CRC and it may be speculated that inhibition of CXCR4 signaling may be useful in preventing further progression from adenoma to colorectal malignancies. In this respect several CXCR4 antagonists are presently under trial. For example, MDX-1338, an anti-CXCR4 monoclonal antibody with a direct effect on CXCR4 signaling is currently being developed for the disruption of survival signals from the bone marrow stroma to the tumour cell in blood cancers and in solid tumours. Thus, the aim is to keep tumour cells from migrating to sites of metastasis by interfering with the homing effects of CXCR4 in acute myeloid leukaemia [32]. With respect to these studies, our results may provide a basis for further studies aiming to interrupt CXCR4-signaling in CRC patients to prevent further disease progression. Acknowledgments We thank C. Blinn, B. Kruse and C. Weber for excellent technical assistance.

References 1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer statistics 2007. CA Cancer J Clin. 2007;57:43–66. 2. Booth RA. Minimally invasive biomarkers for detection and staging of colorectal cancer. Cancer Lett. 2007;249:87–96. 3. McLeod HL, McKay JA, Collie-Duguid ESR, Cassidy J. Therapeutic opportunities from tumour biology in metastatic colon cancer. Eur J Cancer. 2000;36:1706–12. 4. Bogenrieder T, Herlyn M. Axis of evil: molecular mechanisms of cancer metastasis. Oncogene. 2003;22:6524–36.

195 5. Zlotnik A, Yoshie O. Chemokines: a new classification system and their role in immunity. Immunity. 2000;12:121–7. 6. Muller A, Homey B, Soto H, Ge N, Catron D, Buchanan ME, et al. Involvement of chemokine receptors in breast cancer metastasis. Nature. 2004;410:50–6. 7. Schrader AJ, Lechner O, Templin M, Dittmar KE, Machtens S, Mengel M, et al. CXCR4/CXCL12 expression and signalling in kidney cancer. Br J Cancer. 2002;86:1250–6. 8. Voermans C, van Heese WP, De Jong I, Gerritsen WR, van Der Schoot CE. Migratory behavior of leukemic cells from acute myeloid leukemia patients. Leukemia. 2002;16:650–7. 9. Taichman RS, Cooper C, Keller ET, Pienta KJ, Taichman NS, McCauley LK. Use of the stromal cell-derived factor-1/CXCR4 pathway in prostate cancer metastasis to bone. Cancer Res. 2002;62:1832–7. 10. Rubie C, Kollmar O, Frick VO, Wagner M, Brittner B, Gräber S, et al. Differential CXC receptor expression in colorectal carcinomas. Scand J Immunol. 2008;68:635–44. 11. Sehgal A, Keener C, Boynton AL, Warrick J, Murphy GP. CXCR4, a chemokine receptor is overexpressed in and required for proliferation of glioblastoma tumor cells. J Surg Oncol. 1998;69:99–104. 12. Koshiba T, Hosotani R, Miyamoto Y, Ida J, Tsujy S, Nakajima S, et al. Expression of stromal cell-derived factor 1 and CXCR4 ligand receptor system in pancreatic cancer: a possible role for tumor progression. Clin Cancer Res. 2000;6:3530–5. 13. Bertolini F, Dell’Agnola C, Mancuso P, Rabascio C, Burlini A, Monestiroli S, et al. CXCR4 neutralization, a novel therapeutic approach for Non-Hodgkin’s lymphoma. Cancer Res. 2002;62:3106– 12. 14. Phillips RJ, Burdick MD, Lutz M, Belperio JA, Keane MP, Strieter RM. The stromal derived factor –1/CXCL12-CXC chemokine receptor 4 biological axis in non-small cell lung metastases. Am J Respir Crit Care Med. 2003;167:1667–86. 15. Schimanski CC, Galle PR, Moehler M. Chemokine receptor CXCR4-prognostic factor for gastrointestinal tumors. World J Gastroenterol. 2008;14:4721–4. 16. Kim J, Takeuchi H, Lam ST, Turner RR, Wang HJ, Kuo C, et al. Chemokine receptor CXCR4 expression in colorectal cancer patients increases the risk for recurrence and for poor survival. J Clin Oncol. 2005;23:2744–53. 17. Wang SC, Lin JK, Wang HS, Yang SH, Fen-Yau Li A, Chang SC. Nuclear expression of CXCR4 is associated with advanced colorectal cancer. Int J Colorectal Dis. 2010;25:1185–91. 18. Yoshitake N, Fukui H, Yamagishi H, Sekikawa A, Fujii S, Ichikawa K, et al. Expression of SDF-1α and nuclear CXCR4 predicts lymph node metastasis in colorectal cancer. Br J Cancer. 2008;98:1682–9. 19. Sobin LH. TNM, sixth edition: new developments in general concepts and rules. Semin Surg Oncol. 2003;21:19–22. 20. Bustin SA. Absolute quantification of mRNA using real time reverse transcription polymerase chain reaction assays. J Mol Endocrinol. 2000;25:169–93. 21. Rubie C, Kempf K, Hans J, Su T, Tilton B, Georg T, et al. Housekeeping gene variability in normal and cancerous colorectal, pancreatic, esophageal, gastric and hepatic tissues. Mol Cell Probes. 2005;19:101–9. 22. Schoonjans F, Zalata A, Depuydt CE, Comhaire FH. MedCalc: a new computer program for medical statistics. Comput Methods Programs Biomed. 1995;48:257–62. 23. Kollmar O, Rupertus K, Scheuer C, Junker B, Tilton B, Schilling MK. Stromal cell-derived factor-1 promotes cell migration and tumor growth of colorectal metastasis. Neoplasia. 2007;9:862–70. 24. Kollmar O, Rupertus K, Scheuer C, Nickels RM, Haberl GCY, Tilton B, et al. CXCR4 and CXCR7 regulate angiogenesis and CT26.WT tumor growth independent from SDF-1. Int J Cancer. 2010;126:1302–15.

196 25. Akishima-Fukasawa Y, Nakanishi Y, Ino Y, Morija Y, Kanai Y, Hirohashi S. Prognosticsignificance of CXCL12 expression in patients with colorectal carcinoma. Am J Clin Pathol. 2009;132:202–10. 26. Gassmann P, Haier J, Schlüter K, Domikowsky B, Wendel C, Wiesner U, et al. CXCR4 regulates the early extravasation of metastatic tumor cells in vivo. Neoplasia. 2009;11:651–61. 27. Brand S, Dambacher J, Beigel F, Olszak T, Diebold J, Otte JM, et al. CXCR4 and CXCL12 are inversely expressed in colorectal cancer cells and modulate cancer cell migration, invasion and MMP-9 activation. Exp Cell Res. 2005;310:117–30. 28. Rubie C, Frick VO, Wagner M, Weber C, Kruse B, Kempf K, et al. Chemokine expression in hepatocellular carcinoma versus colorectal liver metastases. World J Gastroenterol. 2006;12:6627– 33.

Tumor Biol. (2011) 32:189–196 29. Ottaiano A, di Palma A, Napoletano M, Pisano C, Pignata S, Tatangelo F, et al. Inhibitory effects of anti-CXCR4 antibodies on human colon cancer cells. Cancer Immunol Immunother. 2005;54:781–91. 30. Schimanski CC, Schwald S, Simiantonaki N, Jayasinghe C, Gönner U, Wilsberg V, et al. Effect of chemokine receptor CXCR4 and CCR7 on the metastatic behavior of human colorectal cancer. Clin Cancer Res. 2005;11:1743–50. 31. Kim J, Mori T, Chen SL, Amersi FF, Martinez SR, Kuo C, et al. Chemokine receptor CXCR4 expression in patients with melanoma and colorectal cancer liver metastases and the association with disease outcome. Ann Surg. 2006;244:113–20. 32. Garber K. First results for agents targeting cancer-related inflammation. J Natl Cancer Inst. 2009;101:1110–2.