**4.2 Perlecan-dependent anti-angiogenic mechanism by NK4**

Vascular EC highly express MET, while HGF stimulates mitogenic and morphogenic activities in EC (Nakamura *et al*., 1996), thus suggesting that NK4 could inhibit HGFinduced angiogenesis. Actually, NK4 potently inhibited the HGF-mediated proliferation of EC *in vitro* (Jiang *et al*., 1999b). Strikingly, NK4 also inhibited microvascular EC proliferation and migration, induced by other angiogenic factors, such as b-FGF and vascular endothelial

endocrine mechanism of the HGF delivery system. In this regard, it is known that peripheral blood monocytes produce HGF, contributing to the increase in blood HGF levels via an endocrine mechanism (Beppu *et al*., 2001). Overall, production of HGF by inflammatory cells is involved in carcinoma invasion and metastasis (*i.e.*, local system), while peripheral blood monocytes seem to prevent tumor cell anoikis during metastasis, possibly by a release of

HGF is a stromal-derived paracrine factor that has stimulated cancer invasion at least *in vitro* (Matsumoto *et al*., 1994; Matsumoto *et al*., 1996a; Nakamura *et al*., 1997). Clinical studies suggest that the degree of serum HGF and Met expressions in cancer tissues appears to correlate with a given prognosis (Yoshinaga *et al*., 1993; Osada *et al*., 2008). Thus, it is hypothesized that *in vivo* inhibition of HGF-MET signaling may be a reasonable strategy to prohibit cancer metastasis. To test this hypothesis, we prepared NK4 as an intra-molecular fragment of HGF via a chemical digestive process (Date *et al*., 1997; Matsumoto *et al*., 1998). As expected, NK4 bounded to MET and inhibited HGF-MET coupling as a competitive inhibitor. An additional "unexpected" value was that NK4 inhibited tumor angiogenesis via a MET-independent pathway. This section focuses on the biological value of NK4 as an

NK4 was initially purified as a fragment from elastase-digested samples of recombinant human HGF (Date *et al*., 1997). The N-terminal amino acid sequence of NK4 and of the remnant fragment, assumed to be composed of an HGF β-chain, revealed that NK4 is cleaved between the 478th valine and the 479th asparagine. The N-terminal amino acid sequence of NK4 revealed that the N-terminal structure of NK4 is the same as undigested HGF (*i.e*., 32nd pyroglutamate), indicating that NK4 is composed of the N-terminal 447 amino acids of the α-chain of HGF and contains the N-terminal hairpin domain and four kringle domains (thus designated NK4) (**Fig. 2A**). The binding domains that are responsible for high-affinity binding to MET are the N-terminal hairpin and the first kringle domains in NK4 (and HGF). MET tyrosine phosphorylation occurs in A549 lung carcinoma within 10 minutes after HGF addition, while NK4 inhibits the HGF-mediated MET activation (**Fig. 2B**). Actually, NK4 functions as an HGF-antagonist: HGF induces invasion and migration of the gallbladder and bile duct carcinoma cells in ECM-based gels, while NK4 inhibits HGF-induced invasion in a dose-dependent manner (**Fig. 2C**) (Date *et al*., 1998). These anti-invasive effects of NK4 are seen in distinct types of cancer cells (Hiscox *et al*., 2000; Maehara *et al*., 2001; Parr *et al*., 2001), strengthening the common

Vascular EC highly express MET, while HGF stimulates mitogenic and morphogenic activities in EC (Nakamura *et al*., 1996), thus suggesting that NK4 could inhibit HGFinduced angiogenesis. Actually, NK4 potently inhibited the HGF-mediated proliferation of EC *in vitro* (Jiang *et al*., 1999b). Strikingly, NK4 also inhibited microvascular EC proliferation and migration, induced by other angiogenic factors, such as b-FGF and vascular endothelial

HGF into blood (*i.e*., systemic system).

**4. Structure and activity of NK4 as HGF antagonist** 

HGF-antagonist and as an angiogenesis inhibitor.

**4.1 Structure and anti-invasive function of NK4** 

role of NK4 during cancer migration.

**4.2 Perlecan-dependent anti-angiogenic mechanism by NK4** 

Fig. 2. Preparation of NK4 as an HGF-antagonist and its inhibitory effects on tumor invasion *in vitro*. (A) Preparation and structure of NK4. NK4 is generated via a cleavage of HGF between 478th Val and 479th Asn. (B) Inhibition of HGF-mediated MET tyrosine phosphorylation by NK4 in lung carcinoma cells. (C) Biological activity of NK4. Cancer cell invasion (upper chamber) is induced across a Matrigel layer when fibroblasts (FB) are placed on a lower chamber. In this co-culture system, NK4 inhibits FB-induced tumor cell invasion in a dose-dependent manner.

growth factor (VEGF) (**Fig. 3A**) (Kuba *et al*., 2000). When a pellet containing b-FGF was implanted under the rabbit cornea, angiogenesis was rapidly induced. In this model, NK4 inhibited b-FGF-induced angiogenesis (**Fig. 3B**). *In vitro* models of EC proliferation, HGF and VEGF phosphorylate MET and KDR/VEGF receptor, respectively, whereas NK4 inhibits HGF-induced MET tyrosine phosphorylation, but not VEGF-induced KDR phosphorylation (Kuba *et al*., 2000). Nevertheless, NK4 inhibited the VEGF-mediated EC proliferation without modification of VEGF-mediated ERK1/2 (p44/42 mitogen-activated protein kinase) activation. These results suggest the presence of another mechanism whereby NK4 inhibits VEGF- and b-FGF-mediated angiogenesis.

The fibronectin-integrin signal is essential for the spreading and proliferation of EC. Based on this background, we demonstrated that NK4-mediated growth arrest of EC is due to a loss of the fibronectin-integrin signal. Affinity purification with NK4-immobilized beads revealed that NK4 binds to perlecan (Sakai *et al*., 2009). Consistent with this result, NK4 was co-localized with perlecan in EC. Perlecan is a multi-domain heparan sulfate proteoglycan that interacts with basement membrane components such as fibronectin. Of interest, knockdown of perlecan expression by siRNA diminished the fibronectin assembly and EC spreading, indicating an essential role of fibronectin-perlecan interaction during EC movement. A recent report described that NK4-perlecan interaction suppressed the normal assembly of fibronectin by perlecan (Sakai *et al*., 2009). As a result, FAK activation became faint in EC after NK4 treatment. Under such a loss of fibronectin-integrin signaling by NK4, EC growth and motility were suppressed, even in the presence of b-FGF or VEGF. This is the reason why NK4 arrests b-FGF- or VEGF-mediated angiogenesis (**Fig. 3C**).

Endocrine Delivery System of NK4, an HGF-Antagonist and

studies using tumor-bearing animals, as described below.

growth can be suppressed during NK4 supplemental therapy.

**5.2 Inhibition of tumor angiogenesis by NK4 treatment** 

during NK4 supplemental therapy.

Anti-Angiogenic Regulator, for Inhibitions of Tumor Growth, Invasion and Metastasis 127

tumor invasion and metastasis. This hypothesis is widely demonstrated through extensive

HGF, or co-cultured fibroblasts, are known to induce invasion of gallbladder carcinoma cells (GB-b1) across Matri-gel basement membrane components (Li *et al*., 1998). NK4 competitively inhibits the binding of HGF to MET on GB-d1 cells. As a result, NK4 diminishes HGF-induced, or fibroblast-induced, motogenic activities (Date *et al*., 1998), thus suggesting that stroma-derived HGF is a key conductor for provoking tumor invasion. Such an important role of HGF was also demonstrated *in vivo*. Subcutaneous inoculations of human gallbladder carcinoma GB-d1 cells in nude mice allow for primary tumor growth and invasion to adjacent muscular tissues. Using this conceptual model, we provided the first evidence of NK4 as an anti-tumor drug (Date *et al*., 1998). Recombinant NK4 has inhibited the growth and muscular invasion in a mouse model of gallbladder carcinoma. Consistent with tumor growth arrest, apoptotic change becomes evident during NK4 injections. Since HGF has an anti-apoptotic effect on cancer cells (Zeng *et al*., 2002), reverse of HGF-induced protection by NK4 may be one of the mechanisms whereby carcinoma

In a culture of EC, NK4 produces anti-angiogenetic effects via a MET-independent pathway (Kuba *et al*., 2000; Nakabayashi *et al*., 2003). These effects are also observed in animal models of malignant tumors: administration of recombinant NK4 suppressed primary tumor growth, metastasis of Lewis lung carcinoma, and Jyg-MC(A) mammary carcinoma implanted into mice (Kuba *et al*., 2000), although neither HGF nor NK4 affected proliferation and survival of these tumor cells *in vitro*. NK4 treatment resulted in a remarkable decrease in microvessel density and an increase in apoptotic tumor cells in primary tumors, suggesting that the inhibition of tumor growth by NK4 may be achieved by the suppression of tumor angiogenesis (Kuba *et al*., 2000). The anti-angiogenic effects of NK4 are widely demonstrated in various types of cancers [see our review articles (Matsumoto & Nakamura, 2005; Matsumoto *et al*., 2008a,b)]. Because the inhibition of angiogenesis by NK4 leads to tumor hypoxia, hypoxia-primed apoptosis may contribute to a reduction in tumor size

**5.3 Delayed NK4 therapy for attenuation of end-stage pancreas carcinoma** 

Anti-tumor effect of NK4 is also observed in a mouse model of advanced pancreas carcinoma (Tomioka *et al*., 2001). When NK4 treatment was initiated on day 10, a time when cancer cells were already invading surrounding tissues, NK4 potently inhibited the tumor growth, peritoneal dissemination, and ascites accumulation at 4 weeks after the inoculation. Such an anti-tumor effects of NK4 correlated with decreased vessel density in pancreatic tumors. In an end-stage of pancreas cancer, NK4 inhibited the malignant phenotypes, such as peritoneal dissemination, invasion of cancer cells into the peritoneal walls and ascites accumulation (Tomioka *et al*., 2001). As a result, NK4 prolonged the survival time of mice at an end-stage of cancer (**Fig. 4**). Because effective systemic therapy for pancreatic cancer is currently not available, and diagnosing pancreatic cancer in its early stages is difficult, the highly invasive and metastatic behaviors of pancreatic cancer lead to difficulty in attaining a

**5.1 First evidence of NK4 for inhibition of carcinoma progression** *in vivo* 

Fig. 3. Anti-angiogenic effects of NK4 via a perlecan-dependent mechanism. (A) NK4 suppresses HGF-, b-FGF-, and VEGF-induced proliferation of EC *in vitro* (Kuba *et al*., 2000). (B) Inhibition of b-FGF-induced corneal neovascularization by NK4 treatment in rabbits. (C) Involvement of perlecan (PC) in NK4-mediated growth arrest of EC. Left: Cell surface PC is required for the binding of fibronectin and α5β1-integrin, leading to FAK phosphorylation and crosstalk of VEGF-VEGF receptor (KDR) signaling. Right: NK4 binds to PC, and then the binding of integrin to fibronectin is impaired. As a result, VEGF fails to elicit G1/S progression of EC in the presence of NK4 (Sakai *et al*., 2009).

We have accumulated *in vitro* evidence showing that HGF-MET system may elicit cancer invasion via a paracrine loop of stroma-carcinoma interaction. This phenomenon is also demonstrated *in vivo*: anti-HGF antibody potently suppressed the tumor invasion in a mouse model of pancreas cancer (Tomiola *et al*., 2001). On the other hand, several investigators proposed, in the late-1990's, a new concept that tumor angiogenesis inhibition leads to the arrest of cancer growth and metastasis (Yancopoulos *et al*., 1998). Inhibition of tumor angiogenesis leads to local hypoxia, and then apoptotic death of cancer cells is associated with the arrests of tumor growth and metastasis (*i.e*., cytostatic therapy). In this regard, NK4 also elicits an anti-angiogenic effect via perlecan-dependent mechanism. Thus, bi-functional properties of NK4 as an HGF antagonist and angiogenesis inhibitor raise a possibility that NK4 may prove therapeutic for cancer patients, as follows.
