A novel oxygen carrier “YQ23” suppresses the liver tumor metastasis by decreasing circulating endothelial progenitor cells and regulatory T cells

Buffalo rats (Male, 8-10weeks, 350-400 g) were obtained from lab animal unit, The University of Hong Kong. Before operation, rats were anesthesia with pentobarbitone (40 mg/kg, ip) and after operation buprenorphine (0.05-0.1 mg/kg/12 hours) were given for attenuating the analgesics. Rats were housed in a standard animal laboratory with free activity and access to water and food. They were kept under constant environment conditions with a 12-hour light–dark cycle. All operations were performed under clean conditions. The study had been licensed according to Animal Ordinance Chapter 340 by the Department of Health, Hong Kong Special Administrative Region (ref.: (12–222) in DH/HA&P/8/2/3 Pt. 39). Hepatocellular carcinoma cell line McA-RH7777 (Purchased from the American Type Culture Collection, Number CRL1601, ATCC, Manassas, VA, USA) were used to establish the orthotopic liver cancer model in Buffalo rat [20]. Two weeks after orthotopic liver tumor implantation, The branch of hepatic artery and portal vein to right and triangle lobes are clamped for 30 minutes (ischemia duration) following by reperfusion (release of the clamp). Major hepatectomy of tumor-bearing lobe (left lobe) is performed during the ischemia duration.

YQ23 products were obtained from New B Innovation Limited. YQ23 product is the stabilized non-polymeric cross-linked tetrameric hemoglobin (65 kDa) with undetectable/low level of dimeric hemoglobin (32 kDa), phospholipid, DNA impurities and protein impurities. The concentration of YQ product is 10 g/dL and its pH range is 7.2-7.8. The osmolality and viscosity (at 37°C) are >250 mOsm/kg and 0.9 centipoise respectively. The p50 value is ~ 40 mmHg. The information for YQ product is shown in patent no. US7,932,356 B1, US 8,048,856 B1 and PCT/US12/46130. YQ23 (0.2 g/kg, treatment group) or the same volume of ringer’s acetate buffer (control group) were administrated intravenously at 1 hour before ischemia. An additional 0.2 g/kg YQ23 or the same volume of ringer’s acetate buffer was injected through hepatic portal vein immediately after reperfusion. Blood samples were taken at days 0, 1, 7, 14, 21 and 28 from tail vein for detecting the populations of circulating EPCs and Tregs. And then Buffalo rats were sacrificed at 4 weeks after major hepatectomy and partial hepatic I/R injury to monitor liver tumor metastasis. Liver and lung were sampled for further investigation.

The histological changes were detected by H& E staining and the expression of CD34 (Santa Cruz Biotechnology) were detected by immunohistochemical staining. The details of H & E and IHC staining were described in our previous paper [20]. In brief, after rehydrated in water, the paraffin sections placed in citric buffer (pH 6.0) and treated in a microwave. Afterwards, the sections underwent blocking with 10% FBS and then primary antibodies were applied (incubated at 4°C overnight). Then the sections underwent blocking with 3% peroxidase for 30 min and secondary antibodies from Dako EnVision System (DakoCytomation) were applied. Signals were developed with 3,3’-diaminobenzidine substrate solution (DakoCytomation).

Microvessel density (MVD) of intrahepatic metastatic tumor sections was evaluated [21]. The mean number of tumor nodules in the lungs and liver, as well as tumor volume of intrahepatic metastatic nodules (L × W²/2) was calculated and expressed as average [22].

The number of circulating EPCs and Tregs were detected by flow cytometry. The details of flow cytometry were described in our previous paper [20]. For analysis of EPC cell surface molecules, cells were stained with the following antibodies: unconjugated rabbit anti-CD133 (Abcam, Cambridge, UK), PE-conjugated anti-CD34 (Santa Cruz Biotech, Santa Cruz, CA), VEGFR2⁺ (BD Pharmingen, San Diego, CA) and goat anti-rabbit FITC secondary antibody (Abcam, Cambridge, UK). For analysis of Treg cells, cells were stained with PE-Cy5-conjugated anti CD4 (eBioscience, San Diego, CA) and PE-conjugated anti CD25 (BD Pharmingen, San Diego, CA), and then permeabilized with fixation/permeabiliation working solution and incubated with FITC-conjugated anti-Foxp3 (eBioscience, San Diego, CA).

Liver tissue oxygenation (Liver pO₂) was directly monitored by the OxyLab® in vivo monitoring system (Oxford Optronix, UK) during the ischemia and reperfusion procedures in another group of Buffalo rats. Briefly, a large-area-surface (LAS) oxygen sensor (Oxford Optronix, UK) was placed between the right hepatic lobe and triangle lobe of the rat livers. The branch of hepatic artery and portal vein to right and triangle lobes were clamped for 30 minutes following by reperfusion. YQ23 (0.2 g/kg, treatment group) or ringer’s acetate buffer (control group) were administered intravenously at 1 hour before ischemia. An additional 0.2 g/kg YQ23 or ringer’s acetate buffer was injected through hepatic portal vein immediately after reperfusion. Liver oxygen tension was continuously measured during hepatic ischemia reperfusion injury: 1) baseline; 2) after infusion of YQ23 or ringer’s acetate buffer; 3) during ischemia; and 4) after onset of reperfusion.

In order to investigate the effect of YQ23 on expressions of cytokines/chemokines, a new group of Buffalo rats were included for major hepatecotmy and partial hepatic I/R injury. YQ23 or ringer’s acetate buffer were administered at 1 hour before ischemia and immediately after reperfusion. Liver samples were collected at 6 hours after reperfusion and gene expressions were detected by reverse transcription-polymerase chain reaction (RT-PCR) [20]. Gene expression levels were expressed as the folds relative to the normal liver. The sequences of the primers were listed as follows: CXCR3: Left AGCACAGACACCTTCCTGCT, Right CAGAGACCAGAGCCGAAAAC; TNF-α: Left GTCTGTGCCTCAGCCTCTTC, Right CCCATTTGGGAACTTCTCCT. IL6: Left GCCCTTCAGGAACAGCTATG, Right GTCTCCTCTCCGGACTTGTG; HO1: Left GAAGAAGATTGCGCAGAAGG, Right TTCATGCGAGCACGATAGAG.

In order to investigate the effect of YQ23 on the metastasis of liver tumor after liver surgery, we established a rat orthotopic liver tumor model with local and distant metastatic potentials. The results showed that, after YQ23 treatment, the incidence of intrahepatic metastasis reduced from 69.2% (9 of 13) to 36.4% (4 of 11). Similarly, YQ23 treatment also decreased the lung metastasis from 53.8% (7 of 13) to 36.4% (4 of 11) (Table 1). More importantly, the tumor volume of intrahepatic metastatic nodules was significantly decreased in YQ23 treatment group (0.201 cm³ vs 0.059 cm³, p = 0.045) (Table 1 and Figure 1A). Furthermore, the number of metastatic tumor nodules in liver and lung were also reduced after YQ23 treatment (intrahepatic metastasis: 3.5 vs 0.6, p = 0.15; lung metastasis: 2.2 vs 1.0, p = 0.4) (Table 1). Intrahepatic and lung metastasis were further confirmed by histology examination (Figure 1B).

At four weeks after major hepatectomy and partial hepatic I/R injury, IHC staining revealed that strong expression of CD34 in intrahepatic metastatic tumor nodules were detected in the control group. On the contrary, only weak CD34 expression was found in the YQ23 treatment group (Figure 2A). Furthermore, YQ23 treatment significantly reduced MVD in intrahepatic metastatic tumor nodules (36.4 vs 72.96; p < 0.01) (Figure 2B).

EPCs play important roles in tumor vasculogenesis and tumor growth at early phase by providing structural support to nascent vessels and the release of pro-angiogenic cytokines. In order to explore the underlying mechanism of YQ23 on suppressing liver tumor metastasis, we detected the levels of the circulating EPCs at different time points after major hepatectomy and I/R injury. The level of circulating EPCs was decreased at day 1 after the major hepatectomy and partial I/R injury in both control and treatment group. After that, compared to control group, the level of circulating EPCs was significantly reduced after YQ23 treatment at day 7 and day 28 (day7: 44.6 vs 190.9/10⁵ PBMC cells, p = 0.001; day28: 155.0 vs 496.6/10⁵ PBMC cells, p = 0.005) (Figure 3A). At day 14 and day 21, the treatment of YQ23 also effectively decreased the circulating EPCs levels (day14: 167.4 vs 278.8/10⁵ PBMC cells, p = 0.06; day21: 296.1 vs 506.2/10⁵ PBMC cells, p = 0.13) (Figure 3A).

Tregs play a critical role in antitumor immune responses. They can suppress the immune responses against tumors and thereby promote tumor growth. The results showed that the number of circulating Tregs was reduced at day 1 after the major hepatectomy and partial I/R injury in both groups. The level of circulating Tregs was significantly decreased by YQ23 treatment from day 7 to day 28 (day7: 134.6 vs 192.9/10⁵ PBMC cells, p = 0.003; day14: 68.8 vs 164.7/10⁵ PBMC cells, p = 0.002; day21: 153.2 vs 322.2/10⁵ PBMC cells, p = 0.004; day28: 165.9 vs 419.4/10⁵ PBMC cells, p < 0.001) (Figure 3B).

Accumulating evidence indicated that tissue ischemia and hypoxia, through elevated levels of HIF-1α responsive chemokines such as SDF-1α and VEGF, stimulate the release and recruitment of EPCs and Tregs from the bone marrow. In order to explore the underlying mechanism of YQ23 on suppression of circulating EPCs and Tregs, pO2 levels in liver tissue during hepatic IR injury were compared between control and treatment group. Representative continuous measurement of liver pO2 in rats following intravenous administration of YQ23 and control during the ischemia reperfusion procedures was presented in Figure 4A. At baseline, the average liver pO₂ did not show significant difference between the treatment and control group (15.7 vs 15.4 mmHg; p = 0.93). One hour after YQ23 or buffer injection, YQ23 treatment effectively increased liver pO₂ level (22.3 vs 13.7 mmHg, p = 0.19) although the values did not reach a statistical significant level (Figure 4B). During ischemia, although liver pO₂ was decreased in both treatment and control group, YQ23 treatment resulted in a significantly higher pO₂ (11.2 mmHg) when compared with control group (0.1 mmHg, p = 0.045). After reperfusion, the pO₂ level in liver was also found to be elevated in YQ23 treatment group (18.7 vs 6.5, p = 0.036) (Figure 4B).

In order to explore the underlying mechanism of YQ23 on suppression of circulating EPCs and Tregs, the mRNA expression levels of CXCR3, TNF-α, IL6 and HO1 in liver tissue were compared between control group and treatment group. YQ23 significantly increased HO1 expression and down-regulated expressions of CXCR3, TNF-α and IL6 in liver tissue at 6 hours after major hepatectomy and hepatic I/R injury compared to control group (Figure 5).