Anti-cancer properties of gastropodan hemocyanins in murine model of colon carcinoma

Figureé1 presents the synopsis of our results from the structural studies of the Hc isolated from marine snail R. thomasiana. The electron micrograph shows typical for gastropodan Hcs didecamers (Figureé1A). The quaternary structure of molluscan Hcs is a hollow cylindrical molecule, about 35énm in diameter, with an internal collar complex. 3-D reconstruction of the RtH molecule is shown on Figureé1B [21]. Molecules of molluscan Hcs are structured as decamers (in cephalopods) or didecamers (in gastropods) of subunits with molecular mass of 350é450ékDa. The polypeptide chains of subunits contain seven or eight globular functional units (FUs), depending on the species. FUs are generally termed (a - h), starting from the N-terminus of the respective subunit (Figureé1C). The FUs have a binuclear copper active site, capable of binding reversibly one oxygen molecule. The active site consists of two closely spaced copper ions (Cu_A and Cu_B) each coordinated by three histidine ligands (type 3 copper site). Figureé1D shows the crystal structure of a FU RtH2-e [22]. The different type of FUs share sequence identities of around 45%. Being similar in primary, tertiary and quaternary structure, molluscan Hcs differ in their carbohydrate content (between 2% and 9% (w/w)) and monosaccharide composition [23]. Most probably the specific carbohydrate content and composition of molluscan Hcs are the structural basis for the difference in their antigenicity.

To study the apoptotic effect of gastropodan Hcs on tumor cells we tested their ability to influence directly the C-26 cells by apoptosis assay. C-26 carcinoma cells were incubated with different amounts of HpH and RtH (ranging from 0.1 μg/ml to 100 μg/ml). The cells were further washed and the surface expression of phosphatidylserine was studied at the 24 h, 48 h and 72 h of incubation. The levels of induced early and late apoptosis as well as necrosis were analyzed by Annexin-V/Propidium iodide staining and flow cytometry.

A significant increase in percentage of early apoptosis and a slight increase of late apoptosis were observed after 48 hours of incubation with all tested amounts of Hcs. In contrast, no obvious effect of the treatment was found on the 24 h of incubation (data not shown).

High level of early apoptosis and considerable enhancement in late apoptosis was observed on the 72 h of incubation (Figure 2). Dose-dependant quantitative corelation between induced apoptosis and Hcs added to the cells was not found at all tested time points.

For selection of proper C-26 murine colon carcinoma model 4 groups of mice challenged with different cell numbers were monitored (for details see Methods section). Formation of solid tumors in all mice in the group as well as 0% survival within 3 months after cells challenge was the leading parameters for successful model establishment. In the group inoculated with 5 10⁵ cells palpable solid tumors were observed 2 weeks after C-26 challenge followed by 0% survival 3 months after cell challenge, which is acceptable for the model requested. In the other groups mice developed tumors and died too early or survived for longer period, so they weren t appropriate for further studies.

We have performed an immune system modulation analysis during tumor development when the tumor model was established to demonstrate the immune reaction to the tumor growth. Spleens from mice challenged with 5 10⁵ C-26 carcinoma cells or PBS treated only were isolated and analyzed by flow cytometry. We observed a decrease of CD4+, CD8+ (T cells), CD19+ (B cells) as well as CD94+ (NK cells and a subset of CD8+ T-cells) and slight increase of CD14+ cells (monocytes, macrophages) compared to PBS treated Balb/c mice (Figure 3A).

A cell suspension prepared from solid tumors of the studied mice showed presence of CD4+, CD8+, CD19+, CD94+ as well as CD14+ cell populations in the tumor microenvironment (Figure 3B). These results suggest a movement of immune cells populations to the developed tumors and a relevant decrease of the same populations in the spleen of the animals.

TNFα, IFNγ, IL10 and IL4 levels were measured in mice sera using Enzyme linked immunosorbent assay (ELISA) kits. Challenge with 5 10⁵ C-26 carcinoma cells and treatment with RtH or HpH did not result in a significant difference in the cytokine production between the groups (data not shown).

Groups of Balb/c mice were treated with Hcs from gastropods R. thomasiana and H. pomatia. Two groups (sensRtH, sensHpH) were injected with RtH or HpH 2 weeks before tumor cells inoculation and after establishment of palpable solid tumors the mice were injected intratumoraly once weekly with RtH or HpH.

The other two groups (RtH, HpH) were inoculated with 5 10⁵ cells/mouse of C-26 cells and after palpable solid tumor development the animals were injected once a week with RtH or HpH. Mice injected with C-26 cells only or PBS injected healthy mice were used as controls.

Collected sera from all studied groups were tested for presence of anti-C-26 antibodies.

All groups developed anti-C-26 IgG antibodies compared to the PBS treated control group (Figure 4A). The titer of these antibodies decreased in C-26 group without hemocyanin treatment during tumor development.

The level of anti-C-26 IgG antibodies in the groups immunized with RtH and HpH without priming did not differ compared to the level of C-26 control group for the first 4 weeks after cell administration. Further treatment with RtH or HpH prolonged persistence of high levels of anti-C-26 antibodies in the sera of these animals. A number of studies demonstrate the prolonged survival in case of enhanced anti-tumor antibody production [24].

The administration of RtH or HpH to sensRtH and sensHpH groups resulted in a presence of anti-C-26 antibodies even before cell inoculation which higher values were measured at the 4^th week after C-26 injection (Figure 4A). To determine whether the recognition of C-26 by antibodies generated after priming with Hcs is due to polyspecificity we tested the sera specifisity to foreign antigene (tetanus toxoid). We observed similar pattern of anibody titers to the pattern of anti-C-26 in all of the groups injected with Hcs (data not shown).

To examine the presence of cross-reactive epitopes in Hc molecules and C-26 cells we set ELISA with the HpH and RtH coated on the 96-well plates and incubated with sera from animals challenged with C-26 cells only. The results showed high levels of crossreactive IgG antibodies to HpH and RtH in the sera of animals after tumor cells inoculation (Figure 4B). These results suggest the presence of crossreactive epitopes within the hemocyanins and C-26 colon carcinoma cells. The biochemical nature of these epitopes needs to be studied and determined [25].

The survival of all tested groups was compared with the survival of animals inoculated with C-26 cells only. The results showed 15% survival of sensRtH group on week 15, 30% survival of the sensHpH group and over 30% survival of the group treated with HpH compared to the C-26 group while the group treated with RtH did not show improved survival (Figure 5A).

The dynamics of tumor growth was monitored in all Hcs treated groups and compared to the control group (C-26 only). Animals injected with RtH and HpH with or without priming developed small size tumors within the whole period of observation (Figure 5B). HpH treated group with no priming even showed a decrease in the tumor growth at week 7^th after C-26 challenge.

Splenomegaly is a characteristic feature of C-26 murine tumor model [26]. Representative numbers of mice from each group were sacrificed for organ comparison and examination. We observed a more than 3-fold significant increase of the spleens in non treated C-26-bearing mice compared to Hcs treated and mostly it was expressed in sensHpH and sensRtH groups (Figure 6A).

Lungs from all studied groups were excised and the levels of metastatic manifestation were compared to the C-26 group. The Figure 6B shows large metastatic nodules formed in the lungs of the control mice and also in the RtH-treated group compared to the markedly inhibited number of surface metastasis in the sensRtH, sensHpH and in HpH treated groups.

We have examined morphological and histological features of C-26 tumor model. The paraffin sections of solid tumors from all studied groups were stained with heamatoxilin-eosin histological staining technique and were analyzed by light microscope. The tumor necrosis indicates a poor prognosis for different cancers and is associated with the survival. We observed differences into the necrosis sections within the tumor mass of the studied animals at 6^th week of the treatment schedule. In the sensRtH and sensHpH groups the levels of necrosis were less than in the RtH, HpH groups and also in the group of animals bearing C-26 carcinoma cells only (data not shown). These results could be associated with the enhanced antibody titers that we have found in the sera of the hemocyanin-pretreated animals and with prolonged survival of these groups.

Female 10 weeks old Balb/c mice were obtained from Harlan Farm, Blackthorn, UK. The animals were kept under specific pathogen free (SPF) conditions and the manipulations were approved by the Committee on Animal Research and Ethics at the Bulgarian Food Safety Agency - Ministry of Agriculture and Food in accordance with the international regulations.

FITC (Fluorescein isothiocyanate) -conjugated anti-mouse CD8, PE (Phycoerythrin) -conjugated anti-mouse CD94 and anti-mouse CD19 and APC (Allophycocyanin) conjugated anti-mouse CD4 mAbs (eBioscience, Frankfurt, Germany) were used for FACS (Fluorescence-activated cell sorting) experiments. AP (Alkaline phosphatase) conjugated anti-mouse IgG (Sigma-Aldrich, Taufkirchen, Germany) were used for ELISA.

Murine colon carcinoma cell line C-26 (CT26.WT (ATCC CRL-2638 )) was cultured in complete RPMI (Roswell Park Memorial Institute medium) 1640 medium (Gibco, Gaithersburg, MD) containing 10% FCS (fetal calf serum), 4 mM L-glutamine, 50 μM 2-mercaptoethanol and antibiotics at 37 °C/5% CO₂. The confluent monolayer cells (90%) were trypsinized and used for animal administration and solid tumor establishment.

C-26 cell line was cultured in complete RPMI 1640 medium (see above) and 90% confluent cell monolayer was trypsinized and centrifuged (110 g, 10 min, 4°C). The pellet was collected and was put under 8-fold cycles of freeze-thaw (from −80°C to 37°C). After centrifugation (110 g, 2 min) the supernatant was kept and the total protein was determined spectrophotometrically at λ = 280 nm.

RtH was isolated from the hemolymph of marine gastropod Rapana thomasiana as described in [29]. HpH was isolated from the hemolymph of terrestrial snails Helix pomatia according to [30]. The HpH was solubilized in 100 mM sodium acetate buffer, pH 5.7, and the β-Hc isoform precipitated by dialysis of HpH against 10 mM sodium acetate buffer, pH 5.3, as described elsewhere [31]. Then β-HpH was solubilized in 100 mM sodium phosphate buffer, pH 6.5, at a concentration of 27 mg/ml (stock solution).

RtH and β-HpH were further purified by gel filtration chromatography on a Sepharose 4B column (90 2.4 cm), equilibrated and eluted with 50 mM PBS buffer, pH 7.2. The purity of isolated RtH and β-HpH was controlled by SDS- and native PAGE as described in [29],[31], and by transmission electron microscopy [32]. The protein concentrations were determined spectrophotometrically using the specific absorption coefficient A₂₇₈ ^.1% = 1.36 mg⁻¹ ml cm⁻¹ (20°C) for RtH (27) and A₂₇₈ ^.1% = 1.416 mg⁻¹ ml cm⁻¹ (20°C) for β-HpH [33], respectively.

Hc solutions were passed once through a Detoxi-Gel column in order to remove endotoxin contaminations (Detoxi-Gel column, Thermo Fisher Scientific, Rockford). The level of the remaining endotoxin was determinated by Limilus Amebocyte Lysate coatest gel (LAL) (Chromogenix AB, Molndal, Sweden).

The C-26 cells were cultured in complete RPMI 1640 medium and 90% confluent cell monolayer was incubated in the presence of different concentrations of HpH and RtH (100 μg/ml, 12.6 μg/ml, 1.6 μg/ml and 0.1 μg/ml) for 24, 48 and 72 hours at 37°C/5% CO₂. Then cells were collected, washed and stained with Annexin V-FITC apoptosis detection Kit I (BD Biosciences Pharmingen, Erenbodegem, Belguim), containing Propidium iodide as DNA-binding dye according manufacturer s instruction. The apoptosis of gated cells was analyzed within 15 minutes using flow cytometry (BD LSR II flow cytometer). The early (AnnV-FITC+, PI-) and late (AnnV-FITC+, PI+) apoptosis, as well as necrosis (AnnV-FITC-, PI+) were measured at different intervals (24, 48 and 72 h of incubation).

Four groups of Balb/c mice (10 animals each) were challenged s.c. in the right flanks with different number of C-26 murine carcinoma cells ( 1 10⁴, 1 10⁵, 5 10⁵, 5 10⁶ cells/mice). The animals were observed for 14 weeks.

Two groups of female Balb/c mice (20 animals per group) were injected with RtH (sensRtH) or HpH (sensHpH) (200 μg/mouse, i.p.) 2 weeks before tumor cells inoculation (5 10⁵ cells/mouse). After palpable solid tumor establishment the animals were injected intratumoraly once weekly with 100 μg/mouse of RtH or HpH, respectively. Another two groups were inoculated with C-26 cells (5 10⁵ cells/mouse) and after palpable solid tumor development the mice were injected every week with RtH (RtH group) or with HpH (HpH group) (100 μg/mouse, intratumoraly). Control groups of mice were challenged with C-26 cells or PBS only. Another two control groups were injected once weekly with RtH or HpH only (100 μg/mouse, i.p.). Tumor growth was monitored by measuring palpable solid tumors once a week with a microcaliper and the tumor volume was determined. Every two weeks the animals from all groups were bled and the sera were kept frozen at −70°C. All experiments were duplicated.

Spleens and solid tumors from representative number of animals (n = 4 to 6) were taken and monocellular suspension was isolated by cell strainers (BD Biosciences, Erenbodegem, Belguim). The cells were washed with PBS (containing 2.5% FCS and 0.05% sodium azide) and incubated with one of the following anti-mouse antibodies - anti-CD4-APC, anti-CD8-FITC, anti-CD19-PE, or anti-CD94-PE. Each incubation step was performed for 30 min at 4°C. Finally, the cells were washed twice and kept at 4°C. Ten thousands cells were analyzed from each sample with a BD LSR II flow cytometer using the Diva 6.1.1. software (BD Biosciences, San Jose, CA).

TNFα, IFNγ, IL10 and IL4 levels were measured in mouse sera using ELISA sets (BD OptEIA , BD Biosciences Pharmingen, Erenbodegem, Belguim) according to the manufacturer s instructions.

Solid tumors were resected from terminal animals. After fixation in phosphate-buffered formalin (10%) the tissues were embedded in paraffin and 7 μm sections were analyzed using a standard haematoxylin/eosin staining technique.

Spleens, kidneys, livers and lungs were taken from the same animals and were compared with organs from healthy mice for size differences 8 weeks after the tumor cell administration.

The values in figures correspond to mean ± SD. The Two-way ANOVA test was used to determine differences between each two groups. A value of P < 0.05 was considered as statistically significant. Survival significance was determined via analysis of survival curves with Prism software from GraphPad (San Diego, CA).