Cancer Cell Biology
Identification of bradykinin receptors in clinical cancer specimens and murine tumor tissues
Article first published online: 29 NOV 2001
Copyright © 2001 Wiley-Liss, Inc.
International Journal of Cancer
Volume 98, Issue 1, pages 29–35, 1 March 2002
How to Cite
Wu, J., Akaike, T., Hayashida, K., Miyamoto, Y., Nakagawa, T., Miyakawa, K., Müller-Esterl, W. and Maeda, H. (2002), Identification of bradykinin receptors in clinical cancer specimens and murine tumor tissues. Int. J. Cancer, 98: 29–35. doi: 10.1002/ijc.10142
- Issue published online: 29 JAN 2002
- Article first published online: 29 NOV 2001
- Manuscript Accepted: 26 SEP 2001
- Manuscript Revised: 15 AUG 2001
- Manuscript Received: 1 FEB 2001
- Ministry of Education, Culture, Sports, Science and Technology of Japan. Grant Number: 06282247
- Sagawa Foundation for Frontier Science and Technology
- bradykinin receptor;
- human carcinoma;
- vascular permeability
Bradykinin (BK) has multiple pathophysiologic functions such as induction of vascular permeability and mitogenesis, and it triggers the release of other mediators such as nitric oxide in inflammatory and cancer tissues. To explore the pathophysiologic roles of BK in tumor, we examined the distribution of BK B2 receptors in human adenocarcinoma (lung, stomach), lymphoma (lymph node), hepatoma, squamous cell carcinoma (lung) and carcinoid (duodenum), and in mouse colon adenocarcinoma 38 (C-38) and sarcoma 180 (S-180) tumor tissues. Immunohistochemical staining of tumor tissues with an anti-BK B2 receptor antibody, or autoradiography with the B2 receptor antagonist [125I]HOE 140 (D-Arg-[Hyp Thi D-Tic Oic8]-BK) and the B2 receptor agonist [3H]BK indicated the presence of B2 receptors in all human tumor cells and murine S-180 and C-38 cells. Specific binding of [3H]HOE 140 was observed in S-180 cells with a Kd of 2.1 nM. Binding of [125I]HOE 140 to S-180 cells was competed by an excess amount (20–100 times) of nonradiolabeled HOE 140 or BK, but not by BK B1 receptor agonist des-Arg9-BK. These results provide direct evidence that the BK B2 receptor is expressed in human cancer and experimental murine tumors, which suggests a potential role for BK in inducing pathologic signal transduction in cancer growth and progression, nitric oxide production and vascular permeability enhancement in tumors. BK antagonists may thus have applications in the modulation of cancer growth and in paraneoplastic syndromes. © 2001 Wiley-Liss, Inc.
Kinins constitute a family of potent endogenous peptidyl vascular mediators released by the proteolytic action of kallikrein on kininogens. Among the kinins, bradykinin (BK) is generated in mammalian species from high-molecular-weight kininogen by plasma kallikrein. BK is also generated from low-molecular-weight kininogen by tissue kallikrein via kallidin or Lys10-BK precursor.1 BK causes contraction of vascular smooth muscle of the intestine, uterus and bronchus, to regulate vascular tone. It modulates glucose metabolism and blood pressure and, more importantly, induces vascular permeability in inflammation and cancer.2–6 BK also causes endothelial nitric oxide synthase (eNOS) to promote nitric oxide (NO) production, which contributes to vasorelaxation.7 In addition, BK induces cytokines (e.g., interleukin-1, -6 and -8) and prostaglandins (PGs) and affects the inflammatory response of a host8–11
We previously demonstrated very high concentrations of BK and its homologue hydroxyl prolyl-BK (Hyp3-BK) in ascitic and pleural fluids obtained from patients with ovarian, gastric, pulmonary and hepatic cancers as well as in experimental murine ascitic tumors.12–14 HOE 140 (D-Arg-[Hyp,3 Thi,5 D-Tic,7 Oic8]-BK) is a BK B2 receptor-specific antagonist.15, 16 We reported that administration of the kallikrein inhibitor soybean trypsin inhibitor (SBTI, Kunitz type) and HOE 140 significantly suppressed solid tumor vascular permeability, tumor ascitic fluid accumulation and tumor growth in mice bearing sarcoma 180 (S-180).12, 17
It is therefore essential to investigate the presence and the roles of BK receptors in cancer tissues. Trifilieff et al.18 reported the binding kinetics of BK receptors in normal and cancerous lung tissues in humans. Clements et al.19 determined by reverse transcriptase PCR and Southern blot analysis that tissue kallikrein and the B2 receptor were expressed in endometrial and prostate cancers. However, the histologic presence of BK receptors in human tumors remains to be clarified. Here we report the identification of BK receptors in various clinical specimens of adenocarcinoma, squamous carcinoma, lymphoma, hepatoma and carcinoid as well as in experimental mouse S-180 and colon adenocarcinoma 38 (C-38). For this identification, we used immunohistochemical staining with anti-BK receptor antiserum, and autoradiography and binding assays with 125I-labeled HOE 140 ([125I]HOE 140) and tritium-labeled BK ([3H]BK).
MATERIAL AND METHODS
Animals and tumors
Male ddY mice (SLC, Shizuoka, Japan), 5–6 weeks old, were inoculated with S-180 tumor cells, which were maintained by intraperitoneal passage in mice. For solid S-180 tumors, the inoculum size was 2 × 106 cells, transplanted subcutaneously. Free S-180 cells were isolated from S-180 ascitic fluid after removal of adherent cells (mostly macrophages) by incubating the ascitic fluid in serum-free RPMI-1640 medium (Gibco BRL, Rockville, MD) in a polystyrene flask 3 hr incubations repeated twice. The S-180 cells thus obtained were washed by centrifugation with RPMI-1640 medium and subjected to the binding assay and autoradiography as described below. C57BL/6n mice (Seac Yoshitomi, Fukuoka, Japan), 5–6 weeks old, were used for studies with C-38 tumor, which was maintained in our laboratory. About 3 × 3 × 3 mm3 of fresh C-38 tumor tissue was injected subcutaneously into C57BL/6n mice with a trocar. All experiments were carried out according to the guidelines of the Laboratory Protocol of Animal Handling, Kumamoto University School of Medicine.
Clinical samples of cancer tissues were obtained from the Kumamoto University Hospital in accordance with the guidelines of the Ethical Committee of Kumamoto University for handling human materials. All cancer specimens used in our study except for Ki-1 lymphoma were obtained by autopsy of the cancer patients. The clinical history and pathologic findings of each case are as follows. The specimen of the moderately differentiated adenocarcinoma was derived from the primary lesion of a gastric cancer patient (a 77-year-old woman), with the complication of metastatic peritonitis; no anticancer chemotherapy or radiotherapy was performed throughout the clinical course. The specimen of well-differentiated squamous cell carcinoma from a primary lung cancer was from a 41-year-old man with metastatic pleurisy who had undergone combination chemotherapy with cisplatin, vindesine and mitomycin C (CVM protocol) once for 1 day, but no radiotherapy. The specimen of duodenal carcinoid was taken from a hairy cell leukemia patient (a 68-year-old woman) who died of fungal meningitis; no chemotherapy had been administered for at least 4 months before the onset of the meningitis; the duodenal carcinoid was completely asymptomatic and was incidentally found during autopsy.
The lung papillary carcinoma specimen was from a primary lung cancer patient (a 70-year-old woman, pathologic diagnosis at autopsy: bronchioalveolar cell carcinoma) who died of respiratory failure due to extensive infiltration of the cancer cells throughout the bronchoalveolar spaces; UFT (a mixture of tegafur and uracil) had been given orally for 1 week prior to the death of the patient. The hepatoma specimen was obtained from a 70-year-old woman with hepatocellular carcinoma who died of intraperitoneal bleeding due to rapture of the primary tumor in the liver; this patient had undergone intraarterial administration of cisplatin/Lipiodol suspension twice and another Lipiodol formulation 1 and a half years before the tumor rupture. The Ki-1 lymphoma specimen was from the biopsy of the retroperitoneal periaortic lymph glands of an 11-year-old girl; neither chemotherapy nor radiotherapy had been administered before biopsy.
Chemicals and antibodies
HOE 140, [3H]HOE 140 with a specific activity of 22.4 GBq/mmol and [125I]HOE 140 with a specific activity of 72.5 TBq/mmol were generous gifts from Dr. B.A. Schölkens (Hoechst Marion Roussel, Frankfurt am Main, Germany). These were dissolved and diluted in physiologic saline before the experiments were done. [2,3-prolyl-3,4-3H(N)]-BK, with a specific activity of 4218 GBq/mmol, was obtained from DuPont NEN (Boston, MA). Rabbit anti-rat BK B2 receptor antiserum AS276-83 was prepared as described previously.20–22 Other compounds were of reagent grade and were from commercial sources.
Immunohistochemical staining for B2 receptors in human and murine tumor tissues
Human cancer samples were embedded in paraffin. Murine tumor tissues were excised and then embedded immediately in tissue-embedding medium (Miles, Elkhart, IN), after which they were frozen in liquid nitrogen. Sections 6 μm thick were prepared by use of a cryostat at −14°C and mounted on gelatin-coated slides. The sections were stained by the indirect immunoperoxidase method described previously.23 Briefly, after inhibition of endogenous peroxidase activity with 0.3% H2O2/methanol, sections were incubated at room temperature for 60 min with rabbit polyclonal anti-B2 receptor antibody (1:200). The sections were washed 3 times with PBS and then incubated for 60 min with peroxidase-conjugated donkey anti-rabbit immunoglobulin F(ab′)2 (Amersham, Little Chalfont, UK) diluted 1:100. For visualization of the immunostaining reaction in the tissue sections, 3,3′-diaminobenzidine (Sigma, St. Louis, MO) was used as the substrate, with 50 mM Tris-HCl buffer (pH 7.6) containing 0.01% H2O2. Hematoxylin was used for nuclear staining as usual.
In vitro HOE 140 binding assay
The binding of the B2 receptor antagonist HOE 140 to S-180 cells was assessed as reported previously with slight modifications.24–26 Briefly, S-180 cells (5 × 106/0.5 ml) were incubated on ice for 60 min with various concentrations (0–215 nM) of [3H]HOE 140 in KRP (0.12 M NaCl, 4.8 mM KCl, 0.54 mM CaCl2, 1.2 mM MgCl2, 11 mM glucose, 78 mM sodium phosphate buffer [pH 7.4]) containing 0.2% BSA in the presence or absence of 200 μM unlabeled HOE 140. After incubation, cell suspensions were centrifuged, and 50 μl of the supernatant was removed for counting the unbound [3H]HOE 140. Precipitates of S-180 cells were further washed 3 times by centrifugation with cooled KRP containing 0.2% BSA and solubilized with 2% Triton X-100. The amount of [3H]HOE in the supernatant and that bound to S-180 cells were quantified by measuring their radioactivity using a liquid scintillation counter.
Similarly, for the competition assay, S-180 cells were incubated with 0.2 nM [125I]HOE 140 in the presence of various concentrations of unlabeled HOE 140, B2 receptor agonist BK and B1 receptor agonist des-Arg9-BK. The protein content of the tumor tissue was estimated by use of a protein assay kit (DC Protein Assay, Bio-Rad, Hercules, CA), after the tissues were solubilized with 0.2 M NaOH. Specific binding of [125I]HOE 140 to BK receptors was defined as [125I]HOE 140 binding that was displaced by a 1,000-fold excess of unlabeled HOE 140.
Macroscopic and microscopic autoradiographic studies were carried out by using [125I]HOE 140 and [3H]BK according to methods described previously.24, 25 Consecutive sections of human and murine tumor tissues, 6 μm thick, as well as free S-180 cells fixed on gelatin-coated slides, were treated with [125I]HOE 140 (0.5 nM) or [3H]BK (1 nM) in the incubation buffer containing various protease and kininase inhibitors as described above. For macroscopic autoradiography, x-ray film (Kodak X-OMAT AR) was placed in contact with [125I]HOE 140-treated tissue sections, exposed for 36 hr at −70°C and developed with an automatic developer (Cerpros S, Fuji Film, Tokyo, Japan). For microscopic autoradiography with [3H]BK, slides that had been incubated were dipped into Sakura NR-M2 emulsion (Konica, Tokyo, Japan) at 50°C, dried in the dark and exposed in the dark at 4°C for 10 days. The slides were then developed and stained with hematoxylin for nuclear visualization.
Immunohistochemical staining for B2 receptors in human and murine tumor tissues
As shown in Figure 1, immunostaining clearly demonstrated B2 receptors in clinical cancer specimens of gastric adenocarcinoma (a), squamous cell lung carcinoma (b,c), duodenal carcinoid (d), lung papillary adenocarcinoma (e), hepatoma (f) and Ki-1 lymphoma (g). B2 receptors were also found in murine S-180 (Fig. 1h) and C-38 (Fig. 1i) tumors. The receptors were found mainly on tumor cells. A control gastric adenocarcinoma specimen treated with normal rabbit serum exhibited no appreciable immunostaining (Fig. 1j).
Microautoradiography with [3H]BK in human cancer samples, murine tumor tissues and S-180 cells
Figure 2 shows radiolabeled BK (1 nM) as minute dark grains in free S-180 cells (a) and S-180 tissue sections (b), C-38 tumor sections (c), human lung papillary adenocarcinoma (d) and human hepatoma (e). These results are evidence of the presence of B2 receptors in these specimens, which was further verified by competitive inhibition of radioactive BK binding with a 1,000-fold excess of nonradiolabeled BK (1 μM), as shown for each serial section (a′–e′). These data are consistent with the above immunostaining observations for B2 receptors.
Macroautoradiography with [125I]HOE 140 in S-180 and C-38 tumor tissue sections
The presence of BK receptors was also investigated by using 0.5 nM [125I]HOE 140 in S-180 (Fig. 3a) and C-38 (Fig. 3b) tumors. The competitive binding assay with a 2,000-fold excess of nonradiolabeled BK in serial sections of the same tumors showed strong inhibition of [125I]HOE 140 binding in tumor specimens (Fig. 3a′,b′). These results are consistent with the microautoradiographic observations for [3H]BK, just described.
Saturation profile and Scatchard plot analysis of [3H]HOE 140 binding to S-180 cells
The binding of radiolabeled HOE 140 to free S-180 cells in vitro increased progressively in a dose-dependent and saturable manner (Fig. 4a). In the presence of an excess amount of nonradiolabeled HOE 140, binding was greatly suppressed. Scatchard plot analysis (Fig. 4a, inset) indicated that S-180 cells have receptors with 2 different binding affinities for HOE 140: a high binding affinity with a Kd value of 2.1 nM and 7,500 binding sites per cell, and a low affinity with a Kd value of 83 nM and 48,000 binding sites per cell. A similar saturation profile with a Kd value of 1.0 nM was observed for the binding of [125I]HOE 140 to S-180 tumor tissue sections ex vivo (data not shown).
Competitive binding of [125I]HOE 140 and nonradiolabeled HOE 140, BK and des-Arg9-BK to free S-180 tumor cells
Figure 4b shows competitive inhibition of the binding of [125I]HOE 140 (0.2 nM) to B2 receptors by nonradiolabeled BK, des-Arg9-BK and HOE 140 in S-180 cells in vitro. The B1 receptor agonist des-Arg9-BK did not affect the binding of [125I]HOE 140 to S-180 cells, whereas both the B2 receptor agonist BK and the antagonist HOE 140 suppressed the binding of radiolabeled HOE 140 at 1 nM or higher concentrations in a similar, dose-dependent manner. These results indicate specific binding of HOE 140 to B2 receptors on S-180 cells.
BK has been studied extensively in inflammation and infection. However, the role of BK in tumor biology has been overlooked, and much remains to be clarified. BK-induced vascular permeability enhancement will allow a supply of nutrients and oxygen for rapid growth of solid tumors. The enhanced vascular permeability will also result in fluid accumulation in carcinomatous pleurisy and peritonitis,12–14, 17 which will produce a loss of albumin in serum, i.e., hypoalbuminemia, a cause of cachexia.27 We previously reported that a BK-generating cascade was triggered in tumor tissues, with production of excessive amounts of BK and [Hyp3]-BK in various ascitic and pleural fluids in rodents and in cancer patients.12–14 The administration of SBTI and HOE 140 reduced S-180 ascites formation in mice.12, 14, 17
Two BK receptors, B1 and B2, are classified according to their affinity for 2 different agonists: des-Arg9-BK and BK, respectively.3, 28–32 Most of the typical actions of BK are mediated through the B2 receptor, which is homologous to the G protein-coupled family receptors.29 When BK is generated and binds to the receptor, G protein-mediated signal transduction will be amplified. Thus BK triggers the production of a number of cytokines and causes eNOS and cyclooxygenase to generate NO and PGs, respectively.3, 7–11, 29 NO and PGs are involved in angiogenesis stimulated by vascular endothelial growth factor33 and in tumor vascular permeability.14, 34–37 NO reacts with superoxide anion radical very rapidly and forms a highly oxidizing and nitrating compound, peroxynitrite (ONOO−).38 We recently reported that ONOO− can activate neutrophil- and fibroblast-derived pro-matrix metalloproteinases (pro-MMPs), and therefore they may upregulate tumor metastasis and angiogensis.39 Both ONOO− and MMPs appear to have potent vascular permeability-inducing activity.40, 41 BK even appears to show angiogenic activity via binding to B142, 43 or B2 receptors43, 44 on endothelial cells. The presence of BK and its receptors in tissues and organs will therefore have profound biologic effects in pathophysiologic processes in tumor tissues.
In the present study, we provide histochemical evidence, by means of immunostaining and autoradiography, of the presence of B2 receptors in several types of clinical cancer specimens as well as experimental murine tumor cells and tissues. A few other reports have described expression of BK receptors in carcinomatous PC12 pheochromocytoma cell lines and human astrocytic tumor cell lines in vitro.45, 46 All these results strongly indicate that BK receptors are ubiquitous in cancer tissues. These results thus suggest that tumor cells may be equipped with an autocrine mechanism of BK generation to promote the action of BK, leading to signal amplification for tumor growth involving enhanced tumor vascular permeability,12–14, 18, 41 angiogenesis42–44 and MMP activation39–41, 47 (Fig. 5), even though the half-life of BK is relatively short under physiologic conditions. We previously reported that urinary-type plasminogen activator is widely expressed in tumor cells, and hence plasmin may also be involved in the activation of plasma prekallikrein, resulting in efficient BK production in tumor tissues.14, 48 BK receptor antagonists and kallikrein inhibitors may therefore be of great interest for modulation of growth of solid tumors as well as in paraneoplastic syndromes such as carcinomatous fluid accumulation in peritoneal and pleural cavities and hypoalbuminemia.
The authors thank Dr. B.A. Schölkens, Hoechst Marion Roussel, for generous experimental support and Ms. J.B. Gandy for English editing. Thanks are also due to Dr. M. Takeya at the Department of Pathology II, Kumamoto University School of Medicine, for stimulating discussion and technical assistance.
- 1Zur Kenntnis des Kallikreins. Angew Chem 1948;60: 53., .
- 2Enzymology of glandular kallikreins. In: Erdös, ed. Bradykinin, kallidin and kallikrein. New York: Springer-Verlag, 1979. 103–61..
- 3Bioregulation of kinins: kallikreins, kininogens, and kininases. Pharmacol Rev 1992;4: 1–80., , .
- 4Mechanism of bradykinin-inducd nociceptive response. Adv Exp med Biol 1989;247: 255–9., , , et al.
- 47Activation of matrix metalloproteinases by peroxynitrite-induced protein S-glutathiolation via disulfide S-oxide formation. 2001;276: 29596–602., , , et al.