Expression of vascular endothelial growth factor isoforms and platelet-derived endothelial cell growth factor in bladder cancer

Urologic Oncology 6 (2001) 10–15

Original article

Expression of vascular endothelial growth factor isoforms and plateletderived endothelial cell growth factor in bladder cancer Ning-Chen Li, M.D., Kazuya Kanda, M.D., Tomoharu Fukumori, M.D., Yoshio Inoue, M.D., Masaaki Nishitani, M.D., Hiro-omi Kanayama, M.D., Ph.D., Susumu Kagawa, M.D., Ph.D.* Department of Urology, School of Medicine, The University of Tokushima, 3-18-15 Kuramoto, Tokushima 770-8503, Japan Received 13 September 1999; received in revised form 28 December 1999; accepted 15 April 2000

Abstract We analyzed the expression of vascular endothelial growth factor (VEGF) messenger ribonucleic acid (mRNA) isoforms and plateletderived endothelial cell growth factor (PDECGF) mRNA in bladder cancer. We also attempted to determine if correlation exists between their expression level and conventional clinical variables in patients with bladder cancer. Tissues obtained from 60 patients with bladder carcinoma were used for analysis. Expression levels of VEGF isoforms and PDECGF were examined using reverse transcriptionpolymerase chain reaction (RT-PCR). Correlations between the expression levels of each VEGF isoform and PDECGF and histopathologic findings were evaluated. Four VEGF isoforms corresponding to VEGF121, 165, 189, and 206 were detected in bladder cancer tissue by RT-PCR. Gene expression of all VEGF isoforms as a ratio of the target to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) showed no correlation with pathologic stage of bladder cancer. However, with regard to relative expression levels of VEGF isoform, which is the ratio to the sum of total VEGF isoforms, the levels of VEGF206 and VEGF189 in tumor samples of grade pT2 or higher were significantly lower than those in tumors of grade pT1 or lower (P⬍.05). In contrast, the levels of VEGF121 in ⭓pT2 tumors tended to be higher than those in ⭐pT1 tumors (P⫽.056). The expression level of PDECGF as a ratio to GAPDH in pT2⭐ tumors was significantly higher than that in either pTa or pT1 tumors (P⬍.05). Moreover, a higher expression level of PDECGF was observed in G3 tumors than in G1 tumors (P⬍.05). The results indicated that gene expression of VEGF isoforms do not play a significant role in tumor progression or invasion; however, the distribution of VEGF isoforms may play a role in tumor progression of bladder cancer. A high expression level of PDECGF correlated significantly with the tumor progression of bladder cancer. © 2000 Elsevier Science Inc. All rights reserved. Keywords: Vascular endothelial growth factor (VEGF); Platelet-derived endothelial cell growth factor (PDECGF); Isoform; RT-PCR; Bladder cancer

Angiogenesis is a prerequisite for tumor growth and metastasis [1]. Increased vascular density has been shown to correlate with a higher incidence of tumor metastasis and a worse prognosis in various tumors, including bladder cancer [2]. Angiogenesis is regulated by a number of both stimulating and inhibiting angiogeneic factors [3,4]. One important factor is vascular endothelial growth factor (VEGF), which is a specific mitogen for endothelial cells. VEGF promotes angiogenesis by stimulating capillary proliferation, migration, and permeability [5,6], and plays an important role in neovascularization of various neoplasms. The overexpression of VEGF has been demonstrated in many types of cancers, including colon cancer, breast cancer, lung cancer, ovarian cancer, and bladder cancer [7–11]. Five VEGF isoforms of 121, 145, 165, 189, and 206 amino acids are known to be produced from a single

* Corresponding author. Tel: ⫹81-088-633-7159; fax: ⫹81-088-6337160.

human gene containing eight exons as a result of alternative splicing [12–14]. Different VEGF isoforms have different biological activities [12–15]. Recently, studies have shown that the expressions of certain VEGF isoforms are correlated with tumor progression of colon cancer, lung cancer, and breast cancer [8,9,16]. Platelet-derived endothelial cell growth factor (PDECGF) is a potent angiogenic factor that has been shown to be a thymidine phosphorylase [17,18]. This enzymatic activity is crucial to its angiogenic activity [19]. The expression of PDECGF has been reported to be elevated in several cancers [20,21] and this increased expression is correlated with tumor progression [21,22]. In bladder cancer, the level of PDECGF expression has been shown to correlate with tumor progression [23], and an increased expression level of PDECGF predicted early recurrence in stage Ta bladder cancers [24]. To date, research has indicated that the VEGF gene is frequently overexpressed in transitional cell carcinoma (TCC) of the bladder [11] and is a predictor of relapse and

1078-1439/01/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S1078-1439(00)00 0 9 5 - 8

N.-C. Li et al. / Urologic Oncology 6 (2001) 10–15

progression in noninvasive bladder cancer [25]. To our knowledge, no studies focusing on the VEGF isoform expression pattern in bladder cancer have been reported. Thus, the present study analyzed the expression of VEGF messenger ribonucleic acid (mRNA) isoforms and PDECGF mRNA by reverse transcription-polymerase chain reaction (RT-PCR) assay in bladder cancer tissues. We also attempted to determine if any correlation exists between the expression level of VEGF isoforms or PDECGF and conventional histopathologic variables in patients with bladder cancer. 1. Materials and methods 1.1. Tissue and patient characteristics A total of 60 tumor tissue samples obtained at surgery from 60 patients with bladder cancer were used for analysis. Patient characteristics and tumor findings are shown in Table 1. A preoperative evaluation of tumors was performed using cystourethroscopy and computed tomography, and all tumors were diagnosed as localized bladder cancer (Ta3N0M0). Histologic grade and both clinical and pathologic stages were evaluated according to the criteria of the World Health Organization [26] or TNM [27] classifications. The 60 tumors included 8 grade pTa, 18 pT1, 3 pT2, 22 pT3, and 9 pT4 tumors. Tumor samples for RNA extraction were taken from tissue that grossly and clearly indicated it to be tumor. These samples were fixed in 10% formalin and represented histologically by hematoxylin and eosin staining. 1.2. Quantitation of gene expression by RT-PCR analysis Tumor tissues for RNA extraction were promptly frozen in liquid nitrogen and stored at ⫺80⬚C until use. Total RNA was isolated from tumor tissues (50–100 mg) using the ISOGEN reagent technique (ISOGEN, Nippon Gene, Tokyo, Japan). For RT-PCR, single-strand complementary DNA (cDNA) was synthesized from 1 ␮g of total RNA using a cDNA synthesis kit (First-Strand cDNA Synthesis Kit, Pharmacia Biotech, Gaithersburg, MD USA). The primers for VEGF were synthesized as 5⬘-CTGCCATCCAATCGAGACCC-3⬘ (sense) and 5⬘-ATGGTGATGGTGTGGTGGCG-3⬘ (antisense) according to the VEGF gene structure [6]. The expected size of the PCR products of the VEGF121, VEGF145, VEGF165, VEGF189, and VEGF206 isoforms were 409 bp, 481 bp, 541 bp, 613 bp, and 664 bp, respectively. The primers for PDECGF were synthesized as 5⬘-CTTCGGGGCATGGATCTGGA-3⬘ (sense) and 5⬘-TAACGTCCACCACCAGAGCG-3⬘ (antisense) according to the PDECGF gene structure [28]. The expected size of the PCR products was 466 bp. RT-PCR was performed as described previously [29]. The reaction mixture contained a cDNA derived from 50 ng of total RNA, 20 pmol of the sense and antisense primers, 1 unit of Taq DNA polymerase, and 10 nmol of each deoxynucleotide triphosphate (dNTP), in a final volume of 50 ␮l. A commercial kit

11

(TaKaRa Taq, Takara Shuzo, Kyoto, Japan) was used for Taq DNA polymerase, reaction buffer, and dNTP. Amplification was performed in sequential cycles, including 1 min of denaturation at 94⬚C; 1 min of primer annealing at 59⬚C for VEGF and 58⬚C for PDECGF; and 1 min of extension at 72⬚C. For the first cycle, a 3-min denaturation on the thermal cycler was performed (GeneAmp PCR system 2400, Perkin Elmer, Oak Brook, IL USA). Amplification was carried out for 30 cycles in each case. After the last cycle, all samples were incubated for an additional 4 min at 72⬚C. RT-PCR for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was also performed using the primers 5⬘-ACGCATTTGGTCGTATTGGG-3⬘ (sense) and 5⬘-TGATTTTGGAGGGATCTCGC-3⬘ (antisense) [29]. Following PCR, 10 ␮l aliquots of each PCR product were electrophoresed on 4% (for VEGF) and 2% (for PDECGF) agarose gel, and the gel was stained with ethidium bromide. The highlighted band in each sample was scanned using an image scanner (Gel Printer, Taitec, Saitama, Japan) and was measured using MacBAS software (version 2.3, FUJIX, Tokyo, Japan) and a personal computer. Expression levels of each VEGF isoform and PDECGF were expressed as a ratio of the target to GAPDH gene product. We also examined the ratio of each VEGF isoform to the total VEGF isoforms value, which was the sum of four isoforms. 1.3. Statistical analysis Results are expressed as means⫾S.D. Differences in the expression levels of two independent groups were evaluated using the Mann-Whitney U-test. Statistical analyses were performed with StatView 4.5 software (Abacus Concepts, Berkeley, CA USA). Statistical significance was established at the P⬍.05 level.

Table 1 Characteristics of 60 patients with bladder cancer Characteristics Age 46–87 yrs (mean 66.8) Sex Male Female Cell type TCC TCC and SCC Histologic grade (G) G1 G2 G3 Pathologic stage (pT) pTa pT1 pT2 pT3 pT4 SCC ⫽ squamous cell carcinoma.

Number of cases

52 8 56 4 14 18 28 8 18 3 22 9

12

N.-C. Li et al. / Urologic Oncology 6 (2001) 10–15

Fig. 1. Gene expression of VEGF, PDECGF, and GAPDH in bladder cancer (1–8) by RT-PCR assay. The PCR products of VEGF were electrophoresed on 4% agarose gel. Four VEGF isoforms corresponding to VEGF121, VEGF165, VEGF189, and VEGF206 were detected. PDECGF products were electrophoresed on 2% agarose gel.

2. Results 2.1. Correlation between the gene expression of VEGF isoforms and both clinical and pathologic characteristics in bladder cancer Four bands of VEGF isoform gene expression were detected in bladder cancer by RT-PCR, corresponding to VEGF121, 165, 189, and 206, respectively (Fig. 1). All tumor specimens expressed VEGF121, 165, and 189, whereas VEGF206 expression was not clear in some cases. The mean levels of each VEGF isoform in pTa, pT1, and ⭓pT2 bladder cancers are listed in Tables 2 and 3. The gene expression levels of VEGF121 were highest in all of the four isoforms analyzed. Gene expression of the VEGF isoforms did not differ significantly in pTa, pT1, and ⭓pT2 tumors (Fig. 2). Even when the VEGF isoforms were separated into two groups of ⭐pT1 tumors and ⭓pT2 tumors, no significant differences were observed. Furthermore, the sum for total VEGF isoforms did not differ significantly in any of the subgroups (data not shown). We also evaluated the relative expression levels of all VEGF isoforms, which was the ratio to the total of all

Table 2 Genes’ expression levels of VEGF isoforms and PDECGF gene in pTa, pT1, and ⭓pT2 bladder cancer tissue VEGF121 VEGF165 VEGF189 VEGF206 PDECGF

pTa

pT1

⭓pT2

0.794 ⫾ 0.427 0.501 ⫾ 0.286 0.507 ⫾ 0.221 0.313 ⫾ 0.126 0.355 ⫾ 0.241

0.926 ⫾ 0.608 0.608 ⫾ 0.452 0.635 ⫾ 0.426 0.382 ⫾ 0.291 0.413 ⫾ 0.333

1.082 ⫾ 0.715 0.647 ⫾ 0.481 0.638 ⫾ 0.456 0.362 ⫾ 0.302 0.860 ⫾ 0.680

Expression levels of VEGF isoforms and PDECGF are shown as a ratio of the target to GAPDH gene product (means⫾S.D.).

VEGF isoforms. Relative expression levels of VEGF206 and VEGF189 in ⭓pT2 tumor samples were significantly lower than those in ⭐pT1 tumors (P⬍.05). In contrast, VEGF121 expression in ⭓pT2 tumors tended to be higher than that in ⭐pT1 tumors (P⫽.056) (Fig. 3). However, no significant relationship was observed between VEGF isoform expression and histologic grade (data not shown). 2.2. Correlation between PDECGF gene expression and both clinical and pathologic characteristics in bladder cancer As expected, a 466 bp band was detected as PDECGF gene expression in bladder cancer tissue (Fig. 1). High expression of the PDECGF gene was significantly correlated with pathologic stage. Invasive (⭓pT2) tumors showed significantly higher expression levels than did both pTa and pT1 tumors (Table 2, Fig. 4A). Moreover, a higher expression level of PDECGF was observed in G3 tumors than in G1 tumors (.86⫾.71 versus .39⫾.26, P⬍.05, Fig. 4B). No significant difference in PDECGF expression was observed between G1 and G2 tumors (.39⫾.26 versus .56⫾.49, P⫽.54) or G2 and G3 tumors (.56⫾.49 versus .86⫾.71, P⫽.10, Fig. 4B).

Table 3 Genes’ expression levels of VEGF and PDECGF in ⭐pT1 and ⭓pT2 bladder cancer tissue

VEGF121 VEGF165 VEGF189 VEGF206

⭐pT1

⭓pT2

P-value

0.365 ⫾ 0.071 0.231 ⫾ 0.030 0.251 ⫾ 0.037 0.153 ⫾ 0.045

0.411 ⫾ 0.093 0.233 ⫾ 0.038 0.233 ⫾ 0.037 0.123 ⫾ 0.052

P ⫽ 0.0562 P ⫽ 0.8404 P ⫽ 0.0481 P ⫽ 0.0354

Expression levels of each VEGF isoform are shown as the ratio to the sum of all VEGF isoforms expression (means⫾S.D.).

N.-C. Li et al. / Urologic Oncology 6 (2001) 10–15

13

Fig. 2. Gene expression levels of each VEGF isoform in pTa (n⫽8), pT1 (n⫽18), and ⭓pT2 (n⫽34) bladder cancer tissues. The results are determined by means of a box and whisker plot analysis. The bottom and top edges of the box represent the 25th and 75th percentiles, and the central horizontal line represents the median. The central vertical lines from the box extend to the 10th and 90th percentiles.

3. Discussion As an angiogenic growth factor, VEGF expression has been studied at both the mRNA and protein levels in many different tumors [7–11]. In humans, five different VEGF isoforms have been identified: VEGF121, VEGF145, VEGF165, VEGF189, and VEGF206. It has been shown that different VEGF isoforms have different biochemical and biological properties: whereas VEGF121 and VEGF165 are secreted in soluble form, the two larger isoforms (VEGF189 and VEGF206) remain cell-associated, given that they contain basic, heparin-binding peptides [6,12,13]. Moreover, VEGF145 has been shown to be expressed by several cell lines derived from carcinoma of the female reproductive system [14]. Recently, studies have shown that the expression of certain VEGF isoforms are related to tumor progression. Some authors reported that the expression of VEGF189 is correlated with poor prognosis in colon cancer, esophageal cancer, and non-small cell lung cancer [16,30,31]. Another study found that the VEGF121 mRNA expression level in lung cancer showing lymph node metastasis was higher than in cancer without metastasis [9]. In bladder cancer, overexpression level of the VEGF gene was observed in TCC by Northern blot hybridization [11] and VEGF has been reported to be a predictor of relapse and progression in noninvasive cancer by RNase protection assay [25]. However, based on our data using RT-PCR, no correlation was present between total VEGF expression and clinical findings of bladder cancer. Few studies have focused on the ex-

pression of VEGF isoforms in bladder cancer. In the present study, mRNA expression of four VEGF isoforms was detected in bladder cancer, corresponding to VEGF121, 165, 189, and 206, respectively. All tumor specimens expressed VEGF121, VEGF165, and VEGF189, but did not express VEGF145. The VEGF121 mRNA isoform showed the highest relative expression level among the four isoforms. With regard to gene expression of the VEGF isoforms, no significant differences were observed between VEGF isoform and either pathologic stage or grade. Notably, significantly lower ratios of VEGF206 and VEGF189 to total VEGF isoforms were observed in invasive (⭓pT2) tumors than in ⭐pT1 tumors and the ratio of VEGF121 tended to be higher in ⭓pT2 tumors than in ⭐pT1 tumors of bladder cancer (P⫽.056). This finding suggests that the distribution of VEGF isoforms may play roles in bladder cancer progression. O’Brien et al. [23] reported that the expression of VEGF in noninvasive bladder cancer tissue was fourfold higher than in invasive tumors according to RNase protection assay. However, Sato et al. [11] demonstrated that invasive TCCs showed greater VEGF gene expression than did noninvasive TCCs. In the present study, VEGF121 was expressed at a higher level in invasive bladder cancer tissue, whereas VEGF206 and VEGF189 were expressed at a higher level in noninvasive tumors. Thus, in studying the role of VEGF in tumor progression of bladder cancer, it is important to investigate the expression of different VEGF isoforms.

14

N.-C. Li et al. / Urologic Oncology 6 (2001) 10–15

Fig. 3. Relative expression levels for the ratio of each VEGF isoform gene to the total VEGF expression in noninvasive (⭐pT1, n⫽26) and invasive (⭓pT2, n⫽34) bladder cancer tissues. The results are determined by means of a box and whisker plot analysis.

Several authors have shown that PDECGF, as well as VEGF, is an important angiogenic factor in tumor progression and that its expression level correlates to clinical characteristics and prognosis of certain types of cancer, including bladder cancer [21–24,32]. O’Brien et al. [23] also reported that the expression of PDECGF in invasive bladder

cancers was 33-fold higher than in noninvasive tumors according to RNase protection assay. Moreover, Mizutani et al. [24] reported that the expression of PDECGF was fourfold higher in invasive bladder carcinoma than in noninvasive carcinoma according to high performance liquid chromatography. The present results by RT-PCR assay were in

Fig. 4. Correlation between PDECGF gene expression and (A) pathologic stage and (B) histologic grade in bladder cancer tissues. The results are determined by means of a box and whisker plot analysis.

N.-C. Li et al. / Urologic Oncology 6 (2001) 10–15

agreement with these previous studies [23,24]. Significantly higher expression of PDECGF was observed at a late pathologic stage and high histologic grade of bladder cancer. In conclusion, the present study demonstrated that four isoforms of VEGF can be detected in bladder cancer by RTPCR. No significant differences were observed between gene expression of VEGF isoforms and either pathologic stage or histologic grade in bladder cancer tissues. However, the ratio of VEGF isoforms to the sum of total VEGF isoforms may be differentially correlated with pathologic stage of bladder cancer. This suggests that different distributions of VEGF isoforms may play different roles in tumor progression of bladder cancer, although the exact mechanism remains unknown. A high expression level of PDECGF correlated significantly with tumor progression of bladder cancer. Gene expression of PDECGF appears to be a more important factor than does the VEGF gene in bladder cancer progression.

[13]

[14]

[15]

[16]

[17]

[18] [19]

Acknowledgment

[20]

This work was supported in part by a Grant-in-Aid (11470338) for Scientific Research from the Ministry of Education, Science and Culture of Japan.

[21]

References

[22]

[1] Folkman J. What is the evidence that tumours are angiogenesis dependent? N Engl J Med 1990;285:82–6. [2] Bochner BH, Cote RJ, Groshen S, et al. Angiogenesis in bladder cancer: relationship between microvessel density and tumor prognosis. J Natl Cancer Inst 1995;87:1603–12. [3] Hanahan D, Folkman J. Patterns and emerging mechanism of the angiogenic switch during tumorigenesis. Cell 1996;86:353–64. [4] Rak J, Filmus J, Finkenzeller G, Grugel S, Marme D, Kerbel RS. Oncogenes as inducers of tumor angiogenesis. Cancer Metast Rev 1995; 14:263–77. [5] Keck PJ, Hauser SD, Krivi G, et al. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science 1989;246: 1309–12. [6] Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989;246:1306–9. [7] Takahashi Y, Kitadai Y, Bucana CD, Cleary RK, Ellis LM. Expression of vascular endothelial growth factor and its receptor, KDR, correlates with vascularity, metastasis, and proliferation of human colon cancer. Cancer Res 1995;55:3964–8. [8] Scott PAE, Smith K, Poulsom R, De Benedetti A, Bichnell R, Harris AL. Differential expression of vascular endothelial growth factor mRNA vs protein isoform expression in human breast cancer and relationship to elF-4E. Br J Cancer 1998;77:2120–8. [9] Ohta Y, Watanabe Y, Murakami S, et al. Vascular endothelial growth factor and lymph node metastasis in primary lung cancer. Br J Cancer 1997;76:1041–5. [10] Boocock AC, Charnock-Jones SD, Sharkey MA, et al. Expression of vascular endothelial growth factor and its receptors flt and KDR in ovarian carcinoma. J Natl Cancer Inst 1995;87:506–16. [11] Sato K, Sasaki R, Ogura Y, et al. Expression of vascular endothelial growth factor gene and its receptor (flt-1) gene in urinary bladder cancer. Tohoku J Exp Med 1998;185:173–84. [12] Houck KA, Ferrara N, Winer J, Cashianes G, Li B, Leung DW. The

[23]

[24]

[25]

[26] [27] [28]

[29]

[30]

[31]

[32]

15

vascular endothelial growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA. Mol Endocrinol 1991;5:1806–14. Ferrara N, Houck KA, Jakeman L, Leung DW. Molecular and biological properties of the vascular endothelial growth factor family of protein. Endocrinol Rev 1992;13:18–32. Poltorak A, Cohen T, Sivan R, et al. VEGF145, a secreted vascular endothelial growth factor isoform that binds to extracellular matrix. J Biol Chem 1997;272:7151–8. Houck KA, Leung DW, Rowland AM, Winer J, Ferrara N. Dual regulation of vascular endothelial growth factor bioavailability by genetic and proteolytic mechanisms. J Biol Chem 1992;267:26031–7. Tokunaga T, Oshika Y, Abe Y, et al. Vascular endothelial growth factor (VEGF) mRNA isoform expression pattern is correlated with liver metastasis and poor prognosis in colon cancer. Br J Cancer 1998;77:998–1002. Ishikawa F, Miyazono K, Hellman U, et al. Identification of angiogenic activity and the cloning and expression of platelet-derived endothelial cell growth factor. Nature 1989;338:557–62. Furukawa T, Yoshimura A, Sumizawa T, Haraguchi M, Akiyama S. Angiogenic factor. Nature 1992;356:668. Moghaddam A, Zhang H-T, Fan T-PD, et al. Thymidine phosphorylase is angiogenic and promotes tumor growth. Proc Natl Acad Sci USA 1995;92:998–1002. Pauly JL, Schuller MG, Zelcer AA, Kirss TA, Gore SS, Germain MJ. Identification and comparative analysis of thymidine phosphorylase in the plasma of healthy subjects and cancer patients. J Natl Cancer Inst 1977;58:1587–90. Reynolds K, Farzaneh F, Collins WP, et al. Correlation of ovarian malignancy with expression of platelet-derived endothelial cell growth factor. J Natl Cancer Inst 1994;86:1234–8. Fujimoto K, Hosotani R, Wada M, et al. Expression of angiogenic factors, vascular endothelial cell growth factor and platelet-derived endothelial cell growth factor in human pancreatic cancer, and its relationship to angiogenesis. Eur J Cancer 1998;34:1439–47. O’Brien T, Cranston D, Fuggle S, Bicknell R, Harris AL. Different angiogenic pathways characterize superficial and invasive bladder cancer. Cancer Res 1995;55:510–3. Mizutani Y, Okada Y, Yoshida O. Expression of platelet-derived endothelial cell growth factor in bladder carcinoma. Cancer 1997;79: 1190–4. Crew JP, O’Brien T, Bradburn M, et al. Vascular endothelial growth factor is a predictor of relapse and stage progression in superficial bladder cancer. Cancer Res 1997;57:5281–5. Mostofi MK, Sobin LH, Torloni H. Histological typing of urinary bladder tumors. Geneva: World Health Organization, 1973. International Union Against Cancer (UICC). TNM classification of malignant tumours. 5th ed. L.H. Sobin and Ch. Wittekind, 1997. Hagiwara K, Stenman G, Honda H, et al. Organization and chromosomal localization of the human platelet-derived endothelial cell growth factor gene. Mol Cell Biol 1991;11:2125–32. Kanayama H, Yokota K, Kurokawa Y, Murakami Y, Nishitani M, Kagawa S. Prognostic values of matrix metalloproteinase-2 and tissue inhibitor of metalloproteinase-2 expression in bladder cancer. Cancer 1998;82:1359–66. Tokunaga T, Kijima H, Oshika Y, et al. Aberrant isoform of vascular endothelial growth factor 189 expression is correlated with xenotransplantability of human esophageal cancer. Oncol Rep 1998;5: 1115–8. Oshika Y, Nakamura M, Tokunaga T, et al. Expression of cell-associated isoform of vascular endothelial growth factor 189 and its prognostic relevance in non-small cell lung cancer. Int J Oncol 1998;12: 541–4. O’Brien TS, Fox SB, Dickinson AJ, et al. Expression of the angiogenic factor thymidine phosphorylase/platelet-derived endothelial cell growth factor in primary bladder cancer. Cancer Res 1996;56: 4799–804