Background
Kallistatin, a serine proteinase inhibitor, is first identified as a tissue kallikrein-binding protein, and has emerged as a novel inhibitor of angiogenesis. Kallistatin exerts a variety of biological effects in physiologic and pathologic responses, such as blood pressure regulation, inflammation and anti-angiogenesis [
1‐
5]. It has been reported that kallistatin inhibited vascular endothelial growth factor (VEGF)-induced or basic fibroblast growth factor (bFGF)-induced the proliferation, migration and adhesion of endothelial cells and attenuated bFGF-induced capillary density in mice. Furthermore, a growing body of evidence indicates that chronic inflammation is considered to be one of the most important factors contributing to tumor development and progression. Most solid tumors contain many non-malignant cells, including immune and endothelial cells, which are important in inflammation. Inflammatory cells provide proteases that facilitate tumor invasion and matrix remodeling, accompanying with chemokines, growth factors, and angiogenic factors [
6]. Kallistatin prevented inflammatory responses by reducing the accumulation of macrophages [
7]. Based on these findings, kallistatin has the potential as a therapeutic agent for the treatment of tumor. We have recently reported adenoviral vector-mediated kallistatin expression ameliorated disease progression in the rat model of rheumatoid arthritis and osteoarthritis [
2,
8]. In this study, we exploited lentiviral vectors carrying
kallistatin gene (LV-Kallistatin) as an antitumor agent in syngeneic murine tumor models. Via dual effect of anti-angiogenic and anti-inflammatory activities, LV-Kallistatin has the therapeutic potential for treatment of lung tumors.
Discussion
Results of this study show that LV-Kallistatin inhibits the growth of orthotopic lung tumor via systemic administration. The expression of kallistatin in the tumor-bearing mice, which leads to decreased intratumoral microvessel density and chronic inflammation, may contribute to the antitumor effect of LV-Kallistatin on lung tumor. We demonstrated that intravenous administration of LV-kallistatin inhibited the tumor growth. However, complete tumor regression was not observed in the LV-Kallistatin-treated mice, and mice eventually died though the tumor growth was delayed. Different tumors may respond diversely to the angiogenic inhibitors [
17,
18]. Bergers
et al. showed that various angiogenic inhibitors have distinct efficacy profiles depending on the stage of tumor development and probably the kinetics of cell growth [
19].
Moreover, the anti-vector immunity is a potential issue for cancer gene therapy based on multiple systemic injections. The immunocompetent host previously exposed to the vector may cause their relative lack of efficacy. The lentiviral vectors, peudotyped with vesicular stomatitis virus glycoprotein (VSV-G), have been shown to be less sensitive to anti-vector neutralizing antibody, while displaying desirable characteristics, such as transduction of non-dividing cells, and long-term transgene expression [
20,
21]. However, in our previously studies, we injected adenovirus carrying
kallistatin gene into knees of rats once per week for 3 weeks, and found the expression of transgene 10 days after the last virus administration. Our studies did not find any evidence of anti-kallistatin antibody in the treated animals [
8]. A possible explanation for the limited transgene-triggered immune response is that the
kallistatin gene is conserved in chimpanzee, dog, rat, and mouse [
22,
23]. The human kallistatin is very similar to mouse endogenous kallistatin. The anti-kallistatin neutralizing antibodies may not be produced dramatically during the short course of treatment. Accordingly, multiple injections of the recombinant lentiviruses are most likely to increase the kallistatin production, and enhance the antitumor activity in the short course of treatment.
Previously, human kallistatin levels were determined by ELISA in various human organs. The kidney had highest concentration of kallistatin, followed by liver, lung, prostate gland, and colon (34.8~130.5 ng/mg protein) among human tissues [
24]. In our system, the tissue distribution of kallistatin in control mice was determined at higher concentrations in livers (27.0 ± 7.3 ng/mg), lungs (12.1 ± 3.9 ng/mg), and a lower concentrations in tumor sites (7.0 ± 5.0 ng/mg) by using human kallistatin ELISA system. Because we used polyclonal antibody against human kallistatin in ELISA system, the signals appeared in saline- and LV-GFP-treated mice were due to a cross-reaction. The expression of transgene in mouse tissues after systemic administration of LV-Kallistatin was determined by using human kallistatin ELISA system. The tissue distribution of mouse kallistatin remains to be investigated by using specific mouse kallistatin detective system.
Meanwhile, we found that kallistatin could inhibit the proliferation and migration of tumor cells. Indeed, it has been demonstrated that VEGF directly stimulates the growth of tumor cells [
25]. Kallistatin inhibits the proliferation of endothelial cells and reduces production of paracrine factors, thus suppresses the proliferation of tumor cells. Our results found that VEGF could induce the migration and invasion of tumor cells (Figures
2B and
2C). Kallistatin may act by competing with VEGF and binding to heparin-sulfate proteoglycans [
5]. Kallistatin suppresses VEGF-binding activity and angiogenic signaling cascades induced by VEGF. The anti-angiogenic effect of kallistatin may be similar to anti-VEGF antibody treatment pruning immature vessels in tumor sites. The vessel normalization and restoration of pressure gradients induced by VEGF blockade may explain the increased uptake of antitumor drugs and oxygen in tumor sites [
26]. Furthermore, it has been reported that meloxicam augmented the antitumor activity of kallistatin [
27]. Anti-angiogenic agents may contribute to improve the hypoxic condition of tumor sites by vascular normalization [
28]. Hypoxia, a hallmark of many solid tumors, was reduced by angiogenic inhibitors [
29]. Therefore, it was showed that kallistatin had the ability to reduce hypoxia inducible factor (HIF)-1 α expression [
4], and may improve the hypoxic condition in the tumor microenvironment and increase the radiation or chemotherapy effects.
Several activities of kallistatin contribute to its antitumor effects. Kallistatin has ability to inhibit the inflammation and also reduce the intracellular superoxide formation [
7]. Regulations of reactive oxygen species activity by kallistatin probably contribute to the anti-inflammatory activity. The effector cells involved in enhancing tumor growth appear to be macrophages, which are recruited to the tumor sites and produce TNF-α to stimulate tumor growth [
30,
31]. Our results found that kallistatin reduced the infiltrating macrophages, TNF-α production and NF-κB transcriptional activity in tumor sites. These findings suggest that kallistatin may affect the macrophages and provide a new insight regarding to the relation between inflammation and tumor. Therefore, it is plausible that the antitumor effect of LV-Kallistatin may be attributed not only to its effects on tumor endothelial cells, but also to its ability to down-regulate inflammation within the tumor sites. In agreement with previous reports, our results show that systemic administration of anti-angiogenic inhibitors to tumor-bearing animal results in down-regulation of VEGF expression in tumors [
32,
33]. In our studies, we did not observe any toxicity/side effects in animals, and the healthy lung did not show any toxic effect after LV-Kallistatin treatment. Therefore, combining our previous data with present ones, we suggest that lentivirus carrying
kallistatin gene may be further explored as an anticancer agent for primary and metastatic tumors.
Acknowledgements
We thank D. Trono (School of Life Sciences and Frontiers in Genetics, National Center for Competence in Research, Swiss Federal Institute of Technology, Lausanne, Switzerland) for generously providing lentiviral vectors (pWPXL, psPAX2, and pMD2.G).
This work was supported by grants from National Science Council (NSC 97-3112-B-006-001, and NSC 99-2320-B-039-001-MY2) and China Medical University (CMU-98-N2-06).
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
ALS, CLW, CRW, JLH, MYC and, CJC designed the study. MLT and CHL did the experiments and drafted the manuscript. JC and LC provided the materials. All authors approved the final version of the manuscript.