Therapeutic potential of an anti-angiogenic multimodal biomimetic peptide in hepatocellular carcinoma

Abstract

Hepatocellular carcinoma (HCC) is a major cause of cancer-related death worldwide. Due to inadequate screening methods and the common coexistence of limited functional liver reserves, curative treatment options are limited. Liver transplantation is the only curative treatment modality for early HCC. There are multidisciplinary treatment options like ablative treatments, radiation and systemic therapy available for more advanced patients or those that are inoperable. Treatment resistance and progression is inevitable for these HCC patients. Newer therapeutics need to be explored for better management of HCC. HCC is a hypervascular tumor and many pro-angiogenic proteins are found significantly overexpressed in HCC. Here we explored the therapeutic potential of the anti-angiogenic, anti-lymphangiogenic, and directly anti-tumorigenic biomimetic collagen IV-derived peptide developed by our group. Human HCC cell lines HuH7, Hep3b and HepG2 showed significant disruption of cell adhesion and migration upon treatment with the peptide. Consistent with previously described multimodal inhibitory properties, the peptide was found to inhibit both c-Met and IGF1R signaling in HepG2 cells and blocked HepG2 conditioned media stimulation of microvascular endothelial cell (MEC) tube formation. Furthermore, the peptide treatment of mouse HepG2 tumor xenografts significantly inhibited growth relative to untreated controls. The peptide was also found to improve the survival of autochthonous Myc-induced HCC in a transgenic mouse model. Mechanistically, we found that the peptide treatment reduced microvascular density in the autochthonous liver tumors with increased apoptosis. This study shows the promising therapeutic potential of our biomimetic peptide in the treatment of HCC.

Keywords: HGF; IGF1R; angiogenesis; c-Met; hepatocellular carcinoma.

Figure 1. In vitro effects of AXT050 on liver cancer cells

(A) AXT050 significantly reduced the adhesion of HuH7, Hep3b and HepG2 HCC cell lines. Data presented as mean ± SEM; HuH7 (p = <0.0001), Hep3b (p = <0.0001), and HepG2 (p = 0.0001) by 1-way ANOVA. ^*, ^**, or ^*** designate p values ≤ 0.05, 0.01, or 0.001 respectively by Tukey test for differences compared to the 0 μM AXT050 samples. (B) AXT050 significantly inhibited the migration of HuH7, Hep3b and HepG2 HCC cell lines. Data presented as mean ± SEM; HuH7 (p = <0.0001), Hep3b (p = <0.0001), and HepG2 (p = <0.0001) by 1-way ANOVA. ^*, ^**, or ^*** designate p values ≤ 0.05, 0.01, or 0.001 respectively by Tukey test for differences compared to the 0 μM AXT050 samples. (C) AXT050 treatment (10 μM, 32 μM, and 100 μM) completely inhibited tube formation by HMEC incubated in HepG2 tumor conditioned media.

Figure 2. AXT050 inhibits IGF1R and c-Met signaling in HCC cells

(A) Western blotting demonstrating dysregulation of HGF mediated pMet, pAkt and pErk signaling pathways following treatment of HepG2 cells with AXT050. The presented figures were cropped from images of the original membranes that were cut at approximately 70-75 kDa. The top portion was blotted for pMet and the bottom portion blotted sequentially for GAPDH, pAkt, and pErk 1/2 in that order. Bands for GAPDH are still visible in the pErk 1/2 image (lowest bands) because of this sequential blotting technique. (B) Western blotting showing dysregulation of IGF1 mediated IGF1R signaling following treatment of HepG2 cells with AXT050. The presented figures were cropped from images of the original membranes that were cut at approximately 70-75 kDa. The top portion was blotted for pIGF1R and the bottom portion blotted for GAPDH. (C, D) Semi-quantitative assessment of band intensity of signals of activated pMet, pAkt, pErk and pIGF1R normalized to GAPDH showing significant downregulation of Met and IGF1R signaling with increasing concentrations of AXT050 (10 μM, 32 μM and 100 μM). Data presented as mean ± SEM (N=3); pMet (p = 0.0009), pAkt (p = 0.0029), pERK 1/2 (p = 0.0002), and pIGF1R (p = 0.0002) by 1-way ANOVA. ^* and ^** designate significant (< 0.05) and highly significant (<0.01) differences respectively by Tukey test compared to the growth factor, 0 μM AXT050 treated samples. (E, F) Representative western blots showing total protein levels of Met, AKT, ERK 1/2 (E) and IGF1R (F) in HepG2 lysates following treatment with AXT050. GAPDH is provided as a loading control. Blots were otherwise prepared as described for phosphorylated equivalents (A and B).

Figure 3. AXT050 treatment in vivo delayed tumor growth of subcutaneous HepG2 xenografts

(A) Treatment schema: Tumors were allowed to grow to a volume of 75 mm³, 5 mice were randomly assigned to each of the two groups: (1) no treatment, vehicle control; (2) AXT050 treatment. Mice in the treated arm were given a single dose of 10 mg/kg AXT050 by body weight, intraperitoneally daily. (B) Significant tumor growth delay following treatment with AXT050 (at day 48 Mean ± SEM of Control= 797.9 ± 262.9, (n=5) ; Mean ± SEM of treatment= 329.9 ± 72.00, (n=5) p= 0.002). (C) Gross tumor volume of control and treatment two weeks following treatment showing reduced tumor size in treatment group. (D) Tumor weight is significantly reduced on treatment with AXT050 (Mean ± SEM of Control= 1980 ± 349.9 (n=5); Mean ± SEM of treatment = 1120 ± 174.4, (n=5) p= 0.004). (E) AXT050 inhibits tumor growth in vivo over four weeks of treatment as graphed by 4x tumor doubling. Kaplan-Meier survival analysis where the event was considered time to tumor quadrupling. AXT050 resulted in statistically significant mean tumor growth delay: AXT050 = 12 days > Vehicle Control = 8 days (p= 0.02).

Figure 4. AXT050 improved survival of mice with Myc-induced autochthonous liver tumors, reduced HCC tumor microvascular density, and increased apoptosis

(A) Treatment schema. Myc oncogene was induced in the mice by removing doxycycline from the drinking water at eight weeks of age. After four weeks physical examinations of the mouse abdomen were conducted for the presence of liver tumors. Upon confirmation of increased abdomen size, presence of ascetic fluid as depicted, the mice were randomized to treatment and control arms. Mice in the treated arm were given a single dose of 10mg/kg AXT050 by body weight intraperitoneally daily. Deaths were recorded for survival analysis. (B) Gross pathology showing reduced tumor burden in AXT050-treated mice. (C) AXT050 significantly increases the survival of mice with Myc-induced autochthonous liver tumors. Kaplan-Meier survival analysis showing a significant difference in survival of AXT050 treated animals (N=15) vs control (N=15) group of mice. Median survival AXT050 = 22 days > Vehicle Control = 8 days (p= 0.001). (D) Extensive replacement of normal hepatocytes by neoplastic cells in vehicle control (i). Treated groups show less extensive tumor infiltration and numerous apoptotic cells (arrowheads, ii). (E) Increased apoptosis in treated groups indicated by cleaved caspase 3 immunohistochemistry. (F) Histogram showing mean percentage of cleaved caspase 3 positive cells were significantly higher than the control tumors (Mean ± SEM of control= 31.10 ± 0.2982, (n=20) vs Mean ± SEM of AXT050= 44.60 ± 0.5956, (n=20), p< 0.0001). (G) Decreased vascular density in treated groups indicated by CD31 immunohistochemistry. (H) Histogram showing significantly reduced mean percentage of CD31 stains in the AXT050 treated tumors compared to control tumors (Mean ± SEM of control = 49.20 ± 0.4115, (n=20) vs Mean ± SEM of AXT050 = 37.50 ± 0.8946, (n=20), p< 0.0001).

Figure 5. Multimodal anticancer properties and possible mechanism of action of AXT050 in hepatocellular cancer cells in vitro (inhibition of cellular migration) and in autochthonous mouse model in vivo (inducing apoptosis)