VEGFR-3, VEGF-C and VEGF-D mRNA Quantification by RT-PCR in ...

ANTICANCER RESEARCH 26: 1885-1892 (2006)

VEGFR-3, VEGF-C and VEGF-D mRNA Quantification by RT-PCR in Different Human Cell Types DANY LECLERS, KARINE DURAND, JEANNE COOK-MOREAU, HÉLÈNE RABINOVITCH-CHABLE, FRANCK G. STURTZ and MICHEL RIGAUD

Department of Biochemistry and Molecular Genetics, Faculty of Medicine, Limoges, France

Abstract. The molecular events favoring lymphangiogenic pathways for tumor growth and dissemination are not perfectly understood, nor are the expression patterns of lymphangiogenic biomarkers such as the VEGFR-3 receptor and its ligands, VEGF-C and VEGF-D. In particular, VEGFR-3 expression is not restricted to the lymphatic endothelium, but is found on some cancer cells and other cell types. A quantitative RT-PCR method was set up to measure the mRNA levels of VEGFR-3, VEGF-C and VEGF-D. With this method, a very low detection threshold was obtained when tested on 17 different human cell types. It was found that, in contrast to VEGF-D mRNA, the VEGFR-3 and VEGF-C mRNAs were not expressed in all the cell types studied, and that blood cells expressed high VEGFR-3 mRNA levels compared to solid tumor cells. As a result, quantitative RT-PCR is considered to be a highly reliable and reproducible technique that could help elucidate lymphangiogenic marker patterns of expression and function in cancer. The physiological functions of the lymphatic network mainly consist of the regulation of fluid homeostasis, participation in immunological surveillance and response, and lipid absorption. During the tumoral processes, the lymphatic system is diverted by the tumor which requires endothelial microvasculature development for a nutrient supply and metastatic dissemination to distal tissues and organs (1). An important marker for the lymphatic endothelium is vascular endothelial growth factor receptor-3 (VEGFR-3 or FLT4), a tyrosine kinase receptor (2), initially thought to be expressed, in normal adults, exclusively by lymph vessel endothelium (3). However, its expression has also been reported in blood capillaries (4, 5), in some endothelial cells (6), and in a wide

Correspondence to: Karine Durand, Laboratory of Biochemistry, School of Medicine, University of Limoges, 2, Rue du Dr. Raymond Marcland, F-87025 Limoges, France. Tel: +(33)5 55 05 63 41, Fax: + (33)5 55 05 64 02, e-mail: [email protected] Key Words: Lymphangiogenesis, VEGFR-3, VEGF-C, VEGF-D, mRNA quantification.

0250-7005/2006 $2.00+.40

variety of cancer cells originating from colorectal and lung adenocarcinomas (7, 8), prostatic and breast tumors (9, 10), or myeloid leukemia (11). In addition, many tumors have been shown to express VEGF-C and/or VEGF-D, two members of the VEGF family, that bind to VEGFR-3 and then activate lymphatic endothelial cell migration to the tumor site and microvasculature establishment at the tumor periphery (12-14). The tumor lymphangiogenic processes can be inhibited through neutralization of VEGF-C and VEGF-D by a soluble form of VEGFR-3, as reported in a xenograft model of human lung cancer (15). However, if VEGF-C and/or VEGF-D expression is often associated with poor prognosis in many cancers (16, 17), the VEGFR-3, VEGF-C and VEGF-D expression panel and function in tumor processes remain unclear. Using highly accurate real-time quantitative RT-PCR (qRT-PCR), VEGF-C, VEGF-D and VEGFR-3 mRNA expressions were measured in 17 cell lines, cancer or not, of human origin. Noticeable differences were found in the VEGFR-3, VEGF-C and VEGF-D mRNA levels among the cell types studied, whether they originated from solid tumors, or from healthy or leukemic circulating blood cells.

Materials and Methods Cell description and culture. The cells used included four melanoma cell types (M1Dor, M4Beu, M3Dau and T1P26), blood cells such as normal T lymphocytes or acute leukemia cell lines (Jurkat, HEL and Molt4). Four cell lines originated from colorectal adenocarcinoma, either from primary tumors (HT29-Cl19A, WiDr, SW480) or adenocarcinoma-derived lymph nodes (SW620, established from the same patient as SW480, one year later). Other cell lines were neuroblastoma (SHSY5Y), lung epithelial carcinoma (A549), osteosarcoma (MNNG/HOS), breast adenocarcinoma (MDAMB231) and bladder carcinoma (T24). The cells were cultured in RPMI 1640 (A549, SHSY5Y, T24, Molt4, Jurkat, HEL), DMEM (MDAMB231, HT29-Cl19A), Leibovitz-15 (SW480, SW620), modified McCoy 5A (M1Dor, M4Beu, M3Dau, T1P26) and MEM medium (MNNG/HOS, WiDr). All media (GibcoInvitrogen, Carlsbad, CA, USA) contained 10% (v/v) fetal calf serum and 2 mM glutamine (Gibco-Invitrogen). For WiDr cultures, 1% (v/v) non-essential amino acids were added to the medium.

1885

ANTICANCER RESEARCH 26: 1885-1892 (2006) Table I. Program for amplification and melting curves of VEGF-C, VEGF-D, VEGFR-3 and G6PDH cDNA. Parameter Initial denaturation Cycles Target temperature, ÆC Incubation time, sec Amplification Cycles Target temperature, ÆC Incubation time, sec

Melting curve Cycles Target temperature, ÆC Incubation time, sec

Value*

1 95 (VEGF-C,VEGF-D); 96 (VEGFR-3); 98 (G6PDH) 480

50 Segment 1 95 (VEGF-C,VEGF-D); 96 (VEGFR-3); 98 (G6PDH) 10

1 Segment 1 95(VEGF-C,VEGF-D); 96 (VEGFR-3); 98 (G6PDH) 10

Segment 2 60 (VEGFR-3, G6PDH); 62 (VEGF-C,VEGF-D) 5 (VEGF-C, VEGFR-3); 6 (G6PDH); 4 (VEGF-D)

Segment 3 72

Segment 2 65 (VEGFR-3, G6PDH); 67 (VEGF-C,VEGF-D) 30

Segment 3 95(VEGF-C,VEGF-D); 96 (VEGFR-3); 98 (G6PDH) 0

8 (VEGF-C, VEGFR-3 ); 7 (VEGF-D); 11 (G6PDH)

*When values differ for VEGF-C, VEGF-D, VEGFR-3 and G6PDH amplification, respective genes are indicated in brackets.

Total RNA extraction and reverse transcription. Total RNA was extracted as previously described (18). Reverse transcription (RT) was performed according to the protocol of the Omniscript Reverse Transcriptase kit (Qiagen, Venlo, The Netherlands) with oligo(dT)12-18 primers. The reaction was performed for 60 min at 37ÆC and 5 min at 93ÆC (enzyme inactivation). Two types of controls were included: i) the RT-control for each extracted RNA contained all reagents except the reverse transcriptase and ii) the RNA-control was reverse transcribed without any RNA matrix. Real-time quantitative PCR of VEGF-C, VEGF-D, VEGFR-3 and G6PDH. PCR amplification was performed on a LightCycler (Version 3.5, Roche, Basel, Switzerland) using the Fast Start SYBR Green I kit (Roche). Each reverse-transcribed total cDNA was used for PCRs with VEGF-C (NM_005429), VEGF-D (NM_004469), VEGFR-3 (NM_182925) and housekeeping G6PDH (NM_000402) gene-specific forward (f) and reverse (r) primers, designed to bind to two different exons using specific primer analysis software (Amplify 1.2): VEGF-C: (f) 5’CACGAGCTACCTCAGCAAGA-3’, (r) 5’GCTGCCTGACACTGTGGTA-3’; VEGF-D: (f) 5’-CCTGAAG AAGATCGCTGTTC-3’, (r) 5’-GAGAGCTGGTTCCTGGAGAT3’; VEGFR-3: (f) 5’-CCTGAAGAAGATCGCTGTTC-3’, (r) 5’GAGAGCTGGTTCCTGGAGAT-3’; G6PDH: (f) 5’-TGGAGA ATGAGAGGTGGGATG-3’; (r) 5’-GAGCTTCACGTTCTT GTATCTGT-3’. The mRNA levels were expressed as normalized ratios (VEGF-C or VEGF-D or VEGFR-3 mRNA/G6PDH mRNA) in Relative Arbitrary Units (R.A.U.). For the reaction, 2 ÌL of pure or serially-diluted (1/10, 1/100, 1/1000 and 1/10000) reverse-transcribed products were incubated with 3 mM (G6PDH) or 4 mM (VEGF-C, VEGF-D, VEGFR-3) MgCl2, 2 ÌL ready-touse SYBR Green I Master mix and 0.3 ÌM (G6PDH) or 0.5 ÌM

1886

(VEGF-C, VEGF-D, VEGFR-3) forward and reverse primers in 20 ÌL final volume. The PCR program and melting curve, performed after PCR to control the specificity of the amplified fragment, are described in ∆able I. The second derivative maximum method was used to determine the crossing point (Cp) for individual samples. Quantification using standard curves was performed by the LightCycler software. Statistical analysis. The reproducibility of VEGF-C, VEGF-D, VEGFR-3 and G6PDH cDNA amplification was controlled by calculating standard deviations among duplicate amplifications of each dilution. Four different PCRs were performed for each sample and each gene of interest and standard deviations were calculated. Correlations between VEGF-C, VEGF-D and VEGFR-3 mRNA levels were analyzed by non-parametric Spearman test.

Results Standard curves for VEGFR-3, VEGF-C and VEGF-D mRNA. Among the 17 cell types studied, only the A549 cells expressed the four genes in sufficient amounts to establish the standard curve. Reverse-transcribed cDNA from A549 was used pure and serially-diluted for standard curves (Figure 1). Each cDNA sample was amplified in duplicate and showed high reproductibility for VEGFR-3, VEGF-C, VEGF-D or G6PDH (Figures 1A and 1C). The melting curves confirmed the absence of primer dimers or nonspecific amplified products in the different cDNA amplifications, as shown for VEGFR-3 (Figure 1B). For the same pool of pure cDNA, the four gene amplifications did

Leclers et al: mRNA Level of VEGFR-3 and its Ligands

Figure 1. Standard curves for VEGF-C, VEGF-D, VEGFR-3 and G6PDH mRNA. The VEGFR-3 mRNA standard curve amplification (A) and melting curve (B) are shown as an example. Pure (black line) and serially diluted (1/10: white circles, 1/100: black circles, 1/1000: white squares) cDNA solutions were used for PCR. Cp values and PCR efficacy were reported for VEGF-C, VEGF-D, VEGFR-3 and G6PDH standard curves (C). Control samples were always included in the experiments (dashed curves, B).

not start at the same Cp (VEGFR-3: 25.08±0.13; VEGF-C: 21.28±0.01; VEGF-D: 26.77±0.04; G6PDH: 15.75±0.01), reflecting different expression levels of the respective mRNAs in the A549 cells (Figure 1C). Analysis of VEGFR-3 mRNA levels. VEGFR-3 mRNA expression was analyzed in different cell types, using that of A549 as the reference value of 1 R.A.U. (Figure 2). VEGFR-3 mRNA expression could be divided into two main categories. The first category of cells expressed relatively high VEGFR-3 mRNA levels, ranging from 0.47 R.A.U. (T lymphocytes) to 53.7 R.A.U. (HEL). They belonged mainly to normal (T lymphocytes) or leukemic (Molt4, Jurkat or leukemic HEL) blood cell types, with the exception of lung adenocarcinoma A549 cells. The second group expressed little or no VEGFR-3 mRNA. Although VEGFR-3 mRNA was never detected in SHSY5Y (neuroblastoma) and colorectal adenocarcinoma cells in four different

experiments, very weak transcript levels, ranging from 0.001 R.A.U. to 0.03 R.A.U., were detected in bladder carcinoma, melanoma, breast carcinoma and osteosarcoma cell lines. Analysis of VEGF-D and VEGF-C mRNA levels. As tumor cells can express VEGF-C and/or VEGF-D, sometimes, in a concomitant manner with VEGFR-3, the VEGF-C and VEGF-D mRNA levels were measured in the 17 cell types studied (Figure 3). VEGF-D mRNA was expressed in all the studied cell types (Figure 3A). The expression was highly variable, ranging from 0.028 R.A.U. for Jurkat and Molt4 leukemic cells to 9.613 R.A.U. for the colorectal adenocarcinoma cell line SW480. Interestingly, the highest VEGF-D mRNA levels were detected in the colorectal adenocarcinoma and neuroblastoma SHSY5Y cell lines, where VEGFR-3 mRNA was not expressed. In contrast, normal and leukemic blood cells had only weak VEGF-D mRNA levels, as opposed to those of VEGFR-3. Analysis by

1887

ANTICANCER RESEARCH 26: 1885-1892 (2006)

Figure 2. VEGFR-3 mRNA expression in different cell types. VEGFR-3 mRNA levels (R.A.U) were expressed as a ratio of the A549 cell value. The cells were divided into two groups: high VEGFR-3 mRNA levels and no or weak expression. A logarithmic scale is used in the histogram.

a non-parametric Spearman correlation rank test showed that the VEGF-D mRNA levels were inversely correlated with those of VEGFR-3 mRNA (r2=–0.724; p=0.004) in the studied cell types (analysis excluded A549 as they constituted the reference value for all mRNA measurements). VEGF-C mRNA expression was then quantified in the studied cells (Figure 3B). The levels ranged from 3.10–4 R.A.U. for colorectal adenocarcinoma-derived lymph node cells SW620 to 17.87 R.A.U. for breast carcinoma MDAMB231. Leukemic Jurkat and Molt4 cells, as well as colorectal adenocarcinoma WiDr and HT29 cells, did not express VEGF-C mRNA. In contrast, the sensitivity of the method allowed detection of very weak levels of VEGF-C mRNA, 25- to 3000-fold less than that of A549, in colorectal adenocarcinoma SW480, neuroblastoma, leukemia (HEL) and melanoma cells. Cells such as bladder (T24), breast (MDAMB231) or lung (A549) carcinoma, osteosarcoma (MNNG/HOS) and, to a lesser extent, T lymphocytes and the melanoma cell lines (T1P26 and M1Dor), expressed in turn relatively high VEGF-C mRNA levels. No correlations were found between the VEGF-C and VEGFR-3 or VEGF-D mRNA levels.

Discussion As analysis of reliable angiogenesis and lymphangiogenesis markers remains an indispensable tool for cancer research and clinical oncology, a quantitative RT-PCR (qRT-PCR)

1888

method was set up to accurately measure the VEGFR-3, VEGF-C and VEGF-D mRNA levels in biological samples. Then, this method was checked by analysis of the mRNA levels in normal T lymphocytes and 16 human tumor cell lines. This technique was very sensitive and allowed detection of mRNA level variations in a broad range (up to 3.6x104-fold). Then, we found that, contrary to VEGF-D mRNA, VEGFR-3 and VEGF-C mRNA were not expressed in all cell types studied, with the highest VEGFR-3 mRNAs expression measured for blood leukemic, peripheral blood T lymphocytes and lung adenocarcinoma cells. The qRT-PCR technique combines rapid in vitro DNA amplification and real-time accurate quantification of DNA. The fluorescent DNA intercalant SYBR Green used with the Light-Cycler gives the benefit of real-time detection of an mRNA level as low as one copy. The qRT-PCR method is more sensitive and less cumbersome than other mRNA quantification methods such as Northern blot or in situ hydridization (19). As it requires limited starting material, it is particulary well adapted for small, precious amounts of mRNA obtained from human tissue samples. Furthermore, our method has the advantage of avoiding variations in qRT-PCR efficacy for different mRNAs, as the same cDNA pool was used for all four cDNA PCRs. One of the critical aspects of successful qRT-PCR is the choice of a reference gene in order to correctly normalize the mRNA level. In a preliminary experiment, we compared the normalization of


Figure 3. VEGF-C and VEGF-D mRNA expression in different cell types. VEGF-C and VEGF-D mRNA levels (R.A.U) were expressed as a ratio of the A549 cell values. In contrast to VEGF-D mRNA (A), VEGF-C mRNA (B) was not expressed in all cell types studied. The results are presented in a logarithmic scale.

VEGFR-3 mRNA levels by three currently used housekeeping genes (G6PDH, porphobilinogen deaminase (PBGD) and hypoxanthine phosphoribosyl transferase

(HPRT)) (20). As we found significant correlations between the VEGFR-3 mRNA/G6PDH mRNA and VEGFR-3 mRNA/PBGD mRNA ratios (r2=0.745, p=0.02) and

1889

ANTICANCER RESEARCH 26: 1885-1892 (2006) between the VEGFR-3 mRNA/G6PDH mRNA and VEGFR-3 mRNA/HPRT mRNA ratios (r2=0.945, p=0.003), (data not shown), G6PDH was chosen as the reference gene. Tumor markers can be quantified at mRNA and/or protein levels but the techniques used for protein level assessment (Western blotting, measure of catalytic activity, immunohistochemistry) have major drawbacks such as the quantity of biological material required or the lack of precision/sensitivity. Indeed, qRT-PCR is more accurate and easier to perform but a remaining question is whether or not the mRNAs levels reflect protein expression? Correlations between VEGF-C, VEGF-D and/or VEGFR-3 mRNA and protein levels have already been reported in different cells or tissues (21, 22). However, the correlation greatly depends on the technique, as shown for colorectal cancer (13). Moreover, mRNA levels can be a predictive or a prognostic factor by themselves, as it has been reported that VEGF-C, VEGF-D and/or VEGFR-3 mRNAs are associated with progression and metastatic processes in gastric and colorectal cancer (23-25). We quantified VEGFR-3, VEGF-C and VEGF-D mRNA levels in normal T cells and in different human tumor cell lines. The accuracy of the technique allowed the distinction of three kinds of cells: some did not (neuroblastoma SHSY5Y, colorectal adenocarcinoma) or only weakly expressed VEGFR-3 mRNA (melanoma, osteosarcoma MNNG/HOS, breast carcinoma MDAMB231), probably corresponding to a function of VEGFR-3 for tumor cells which remains unclear at the present time. The third category, including non-small cell lung cancer A549, T lymphocytes, HEL, Molt4 and Jurkat, expressed high VEGFR-3 mRNA levels. To our knowledge, this is the first report of VEGFR-3 mRNA expression in Jurkat cells, T lymphocytes or A549, whereas it has already been detected in the hematopoietic cell line HEL (11, 26), but not in Molt4 (2). If the function of VEGFR-3 mRNA expression in solid tumor cells such as A549 is unclear, the receptor mRNA expression in blood cells could be due to their hematopoietic characteristics. It is now wellestablished that VEGF receptors, notably VEGFR-2 and VEGFR-1, are expressed by endothelial cells as well as hematopoietic stem cells and their leukemic malignant counterparts, that both rise from a common precursor (hemangioblast), and so share several signaling pathways (27). Our results also showed that high VEGF-C mRNA levels were found in MDAMB231, in agreement with previous results (28, 29) and, to a lesser extent, in bladder carcinoma (T24), in lung adenocarcinoma A549 and osteosarcoma 1547 cells. In contrast, no or weak VEGF-C mRNA expressions were found for all other cell types studied. On the other hand, VEGF-D mRNA was expressed in all the cell types studied and was inversely-correlated with

1890

VEGFR-3 expression. This, to our knowledge, has never been shown, and could result from a VEGFR-3-dependent paracrine regulation mechanism. The qRT-PCR technique is a very reliable method for the detection of extremely slight variations in mRNA expression in all cell or tissue types. In fact, it is currently probably a more rapid, precise and reproducible method than the gene or protein expression detection techniques used at present, especially when determination of protein expression is not absolutely required, notably in the case of prognostic factor assessment in cancer. In this study, no significant levels of the tumor growth factors VEGF-C and VEGF-D were detected in circulating blood tumor cells, possibly as circulating cells are not dependent, as solid tumors, are on neovasculature formation. Furthermore, high VEGFR-3 mRNA levels were detected in normal or leukemic blood cells and the sensitivity of the qRT-PCR technique allowed us to distinguish very slight from non-existent mRNA levels in the different solid tumor cells tested. In addition, a large variability in VEGFR-3 mRNA expression did not depend on tumor type, which may require additional studies to analyze the hypothetical role of the receptor in cancer cell invasion or aggressiveness.

Acknowledgements This work was supported by grants from the "Ligue contre le Cancer. Comité de la Creuse", France. We thank Jeanne CookMoreau for reading this manuscript thoroughly.

References 1 Karkkainen MJ and Alitalo K: Lymphatic endothelial regulation, lymphoedema, and lymph node metastasis. Semin Cell Dev Biol 13: 9-18, 2002. 2 Aprelikova O, Pajusola K, Partanen J et al: FLT4, a novel class III receptor tyrosine kinase in chromosome 5q33-qter. Cancer Res 52: 746-748, 1992. 3 Karkkainen MJ and Petrova TV: Vascular endothelial growth factor receptors in the regulation of angiogenesis and lymphangiogenesis. Oncogene 19: 5598-5605, 2000. 4 Paavonen K, Puolakkainen P, Jussila L et al: Vascular endothelial growth factor receptor-3 in lymphangiogenesis in wound healing. Am J Pathol 156: 1499-1504, 2000. 5 Valtola R, Salven P, Heikkila P et al: VEGFR-3 and its ligand VEGF-C are associated with angiogenesis in breast cancer. Am J Pathol 154: 1381-1390, 1999. 6 Partanen TA, Arola J, Saaristo A et al: VEGF-C and VEGF-D expression in neuroendocrine cells and their receptor, VEGFR-3, in fenestrated blood vessels in human tissues. FASEB J 14: 20872096, 2000. 7 Tanno S, Ohsaki Y, Nakanishi K et al: Human small cell lung cancer cells express functional VEGF receptors, VEGFR-2 and VEGFR-3. Lung Cancer 46: 11-19, 2004. 8 Witte D, Thomas A, Ali N et al: Expression of the vascular endothelial growth factor receptor-3 (VEGFR-3) and its ligand VEGF-C in human colorectal adenocarcinoma. Anticancer Res 22: 1463-1466, 2002.


9 Li R, Younes M, Wheeler TM et al: Expression of vascular endothelial growth factor receptor-3 (VEGFR-3) in human prostate. Prostate 58: 193-199, 2004. 10 Longatto Filho A, Martins A, Costa SM et al: VEGFR-3 expression in breast cancer tissue is not restricted to lymphatic vessels. Pathol Res Pract 201: 93-99, 2005. 11 Fielder W, Graeven U, Ergun S et al: Expression of FLT4 and its ligand VEGF-C in acute myeloid leukemia. Leukemia 11: 1234-1237, 1997. 12 Achen MG, Jeltsch M, Kukk E et al: Vascular endothelial growth factor D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4). Proc Natl Acad Sci USA 95: 548-553, 1998. 13 George ML, Tutton MG, Janssen F et al: VEGF-A, VEGF-C, and VEGF-D in colorectal cancer progression. Neoplasia 3: 420-427, 2001. 14 Makinen T, Veikkola T, Mustjoki S et al: Isolated lymphatic endothelial cells transduce growth, survival and migratory signals via the VEGF-C/D receptor VEGFR-3. EMBO J 20: 4762-4773, 2001. 15 He Y, Kozaki K, Karpanen T et al: Suppression of tumor lymphangiogenesis and lymph node metastasis by blocking vascular endothelial growth factor receptor 3 signaling. J Natl Cancer Inst 94: 819-825, 2002. 16 Clarijs R, Schalkwijk L, Hofmann UB et al: Induction of vascular endothelial growth factor receptor-3 expression on tumor microvasculature as a new progression marker in human cutaneous melanoma. Cancer Res 62: 7059-7065, 2002. 17 Dias S, Choy M, Alitalo K et al: Vascular endothelial growth factor (VEGF)-C signaling through FLT-4 (VEGFR-3) mediates leukemic cell proliferation, survival, and resistance to chemotherapy. Blood 99: 2179-2184, 2002. 18 Durand-Faucher K, Rabinovitch-Chable H, Dzugan H et al: A quantitative RT-PCR method to determine topoisomerase I mRNA levels in human tissue samples. Clin Chem Lab Med 43: 707-714, 2005. 19 Bustin SA and Nolan T: Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. J Biomol Tech 15: 155-166, 2004. 20 de Kok JB, Roelofs RW, Giesendorf BA et al: Normalization of gene expression measurements in tumor tissues: comparison of 13 endogenous control genes. Lab Invest 85: 154-159, 2005.

21 Hanrahan V, Currie MJ, Gunningham SP et al: The angiogenic switch for vascular endothelial growth factor (VEGF)-A, VEGF-B, VEGF-C, and VEGF-D in the adenoma-carcinoma sequence during colorectal cancer progression. J Pathol 200: 183-194, 2003. 22 Schietroma C, Cianfarani F, Lacal PM et al: Vascular endothelial growth factor-C expression correlates with lymph node localization of human melanoma metastases. Cancer 98: 789-797, 2003. 23 Kawakami M, Furuhata T, Kimura Y et al: Quantification of vascular endothelial growth factor-C and its receptor-3 messenger RNA with real-time quantitative polymerase chain reaction as a predictor of lymph node metastasis in human colorectal cancer. Surgery 133: 300-308, 2003. 24 Kitadai Y, Kodama M, Cho S et al: Quantitative analysis of lymphangiogenic markers for predicting metastasis of human gastric carcinoma to lymph nodes. Int J Cancer 115: 388-392, 2005. 25 Onogawa S, Kitadai Y, Tanaka S et al: Expression of VEGF-C and VEGF-D at the invasive edge correlates with lymph node metastasis and prognosis of patients with colorectal carcinoma. Cancer Sci 95: 32-39, 2004. 26 Wang JF, Ganju RK, Liu ZY et al: Signal transduction in human hematopoietic cells by vascular endothelial growth factor related protein, a novel ligand for the FLT4 receptor. Blood 90: 3507-3515, 1997. 27 Dias S, Hattori K, Zhu Z et al: Autocrine stimulation of VEGFR-2 activates human leukemic cell growth and migration. J Clin Invest 106: 511-521, 2000. 28 Al-Rawi MA, Watkins G, Mansel RE et al: The effects of interleukin-7 on the lymphangiogenic properties of human endothelial cells. Int J Oncol 27: 721-730, 2005. 29 Fitzpatrick TE, Lash GE, Yanaihara A et al: Inhibition of breast carcinoma and trophoblast cell invasiveness by vascular endothelial growth factor. Exp Cell Res 283: 247-255, 2003.

Received February 27, 2006 Accepted March 20, 2006

1891