The Effect of Bovine Cartilage on the Growth of Mouse B16F10 Melanoma Cells in vitro and in vivo

Corresponding Author: Alexander M. Abdelnoor Department of Experimental Pathology, Immunology and Microbiology, Faculty of Medicine, American University of Beirut, Beirut Lebanon Tel: (+)961 1 350000 Fax: (+)961 1 744487 Email: aanoor@aub.edu.lb Abstract: Earlier reports indicated that Bovine Cartilage (BC) had antitumor effects but only a few in vitro and in vivo investigations were conducted to assess its mechanism of action. The aim of this study was to investigate some of the proposed mechanisms of action of BC on mouse melanoma cells both in vitro and in vivo and to determine if its effect on tumor cells is selective. One hundred and ten mice were divided into 5 groups and received Intraperitoneal (IP) injections of melanoma cells followed by treatment with BC using different routes of administration. Following a daily treatment period of 16 days, serum levels of VEGF and IL-12 were determined by ELISA at 2, 4 and 6 h after the last treatment dose. Additionally, 10 mice from each group were monitored for survival for 20 days post-last treatment. Moreover, melanoma and mouse mononuclear cells were incubated separately with increasing BC concentrations for 48 h and percent viability was determined. A significant decrease in the serum levels of VEGF and significant increase in the serum levels of IL-12 were observed in the groups treated with BC. Moreover 20% survival rate was noted in the group treated with BC both orally and IP, whereas 10% survival was noted in the groups given BC either IP or orally. In vitro, total eradication of melanoma cells was observed with 1000 μg mL −1 of BC. BC was not toxic to mouse mononuclear cells. The in vivo anti-tumor effect of BC was best observed when combined IP and oral doses were given and it appears that two of its actions are by activating macrophages as indicated by increases in IL-12 levels and blocking angiogenesis as indicated by decreases in VEGF levels. Finally BC showed selective toxicity against melanoma cells.


Introduction
The use of Bovine Cartilage (BC) as a treatment for several medical conditions has been around for more than 40 years. Dr. Prudden was the first to assess the ability of BC to accelerate healing in experimental wounds (Prudden et al., 1975;Takayuki, 1960;Prudden et al., 1962). The obtained positive results drove Prudden to test the accelerating effect of BC on several different laboratory animals and on humans, where in both cases results were significant Prudden, 1964;Prudden and Allen, 1965).
A study conducted by (Houck et al., 1961) showed that BC had potent anti-inflammatory activity which was further emphasized, when a patient with psoriasis and a non-healing ulcer markedly improved upon BC treatment. Hereafter several cases of acute and chronic inflammations and skin allergies were treated with BC (Prudden and Balassa, 1974).
The initial decision to test BC in the clinical treatment of cancer was based on its efficiency in the treatment of psoriasis, on its record of being non-toxic and on its ability to cure a patient's ulcerated breast cancer. The effectiveness of BC in treating cancer was reported for 31 terminally ill patients. There were some significant decreases in a wide variety of intractable malignancies and all the patients except one survived longer than the time predicted by their clinicians (Prudden, 1985).
The mechanism of action underlying the anti-tumor activity of BC is not well defined. At least four possible mechanisms have been proposed; BC may directly induce cancer cell death (Prudden, 1985;Durie et al., 1985), it could stimulate the immune system (Rosen et al., 1998;Morell and Daniel, 2014), inhibit collagenase activity (Kuettner et al., 1977;Murray et al., 1986) and/or inhibit angiogenesis (Folkman et al., 1971;Langer et al., 1976).
The current study was carried out in an effort to investigate some of the proposed anti-tumor mechanisms of action of BC on mouse melanoma; and for this purpose an in vitro study was performed to determine if BC had a direct toxic effect on melanoma cells and if so, to determine if its toxicity is selective. Moreover, an in vivo study was done to determine survival rates of B16F10 melanomabearing mice and their serum levels of IL-12 and VEGF treated with BC compared to untreated controls.
It was anticipated that the results obtained would provide some insights on the mechanism(s) by which BC exerts its anti-tumor effect.

Materials and Methods
In vivo Studies BC used was "NOW Bovine Cartilage" (395 S. Glen Elyn Rd, Bloomingdale, IL), a commercially available extract. The powdered BC was weighed, dissolved in sterile distilled water and stored at -20°C until used.

B16F10 Melanoma Cells
B16F10 metastatic melanoma cells are syngeneic with the C57BL/6 mice. They are adherent cells and were maintained as monolayers in vitro in RPMI 1640 culture medium supplemented with 1% L-Glutamine, 1% Penicillin-Streptomycin and 10% Fetal Bovine Serum (Lonza, B-4800 Verviers, Belgium) and incubated at 37°C in a 5% CO2 incubator.
On the day of tumor induction into mice, the melanoma cells were detached with trypsin (2.5% trypsin in 10× in HBSS without calcium or magnesium, Lonza, B-4800 Verviers, Belgium). The viable cell count was determined by trypan blue exclusion method using a Neubauer Chamber and re-suspended in RPMI-1640 medium. The amount of B16F10 melanoma cells injected intraperitoneally (IP) was 3.4 million cells/Kg suspended in 0.5 mL RPMI-1640.

Challenge of Mice with Tumor Cells and Treatment
The use of mice in this study was approved by the Institutional Animal Care and Use Committee (IACUC) at the Faculty of Medicine, American University of Beirut. One hundred and ten C57BL/6 female mice, 6 to 8 weeks old, were injected IP with melanoma cells at day 0 and divided into 5 groups (twenty-two mice in each). After 2 days post melanoma injection, BC treatment was given to the five groups as shown in Table 1. Two groups served as controls; one group remained untreated and served as a control for oral BC group. Another group received daily IP injection of sterile water at a dose of 0.4 mL/mouse and served as a control for IP BC and IP and oral BC groups. The treatment was given for 16 consecutive days.

Procurement of Specimens
On day 16 and at 2, 4 and 6 h after the last BC treatment, four mice from each group were anesthetized then dissected and blood was collected by cardiac puncture. Blood from each group was pooled and serum was separated and used for Vascular Endothelial Growth Factor (VEGF) and Interleukin 12 (IL-12) quantification. The remaining 10 mice from each group were monitored for 20 days post-last treatment to determine the survival rate. Upon death, the mice were dissected to confirm that death was tumor related.

VEGF and IL-12 Quantification
VEGF mouse ELISA kit (Abcam, ab100751, USA) and IL-12p70 mouse ELISA kit (Abcam, ab119531, USA) were used to determine the serum levels of VEGF and IL-12, respectively. The procedures were performed according to the manufacturers' protocol.

In Vitro Evaluation of BC Effect
Culture of B16F10 Melanoma Cells in the Presence of BC B16F10 Melanoma cells were counted and seeded in 24-well plates, (seeding density of 2*10 4 cells/500µl/well), then incubated for 24 h at 37°C and 5% CO2. After incubation, 500 µL of five different BC concentrations (5, 10, 100, 1000 and 5000 µg mL −1 ) were added to the different wells. No BC was added to the control wells.
The samples were run in duplicates. Wells were then incubated at 37°C in a 5% CO2 incubator for 48 h. After incubation, melanoma cells were detached with 1× trypsin, then suspended in culture medium and a viable cell count was determined.

Culture of Mouse Mononuclear Cells (MNC) in the Presence of BC
Seven healthy female C57BL/6 mice were sacrificed and spleens were collected aseptically. Spleens were homogenized, treated with RBC lysis solution and a single mononuclear cell suspension was prepared in RPMI 1640, 1% L-Glutamine and 10% FBS.
The cells were counted and seeded in 24-well plates, with a seeding density of 3*10 6 cells/500µl/well. Then, 500 µL of five increasing BC concentrations (5, 10, 100, 1000 and 5000 µg mL −1 ) were added to the different wells. No BC was added to the control wells. All wells were run in duplicates. Wells were then incubated for 48 h at 37°C in a 5% CO2 incubator. After incubation, viable and dead cell count was determined.

Statistical Analysis
Whenever applicable, data were expressed as Mean ± SD. Mice survival was evaluated by generating Kaplan-Meier survival curves. The unpaired student T-test was implemented to assess the sample variations between groups using the Graph pad online software. Results were considered to be statistically significant at P value <0.05.

VEGF Serum Levels
When compared to the untreated group, significant decrease in serum VEGF levels at 2, 4 and 6 h post-last BC treatment (p-value = 0.0008, 0.0014 and 0.0002, respectively) was seen in oral BC group. Again, when comparing IP BC group, to the sterile water group, significant decrease in serum VEGF levels was seen at 4 and 6 h after the last BC treatment (p-value = 0.0148 and 0.0014, respectively). Finally when comparing the IP and oral BC group to the sterile water group, significant decrease in serum VEGF levels was seen only at 6 h post BC treatment (p-value = 0.0006) (Fig. 1). In the sterile water group and the IP and oral group, VEGF levels at 2 h after the last BC dose were out of range (higher than the highest standard's absorbance). It is worth noting that this experiment was repeated twice using different dilution factors and consistent results were obtained.

IL-12 Serum Levels
The serum levels of IL-12, at 2 h post BC treatment, were lower in IP BC and oral BC groups, compared to their respective control, sterile water and untreated groups; however IL-12 level of the IP and oral BC group, was equal to that of its control, sterile water group. At 4 h post-treatment, the levels of IL-12 significantly increased in IP BC group (P value = 0.003) and oral BC group (P value = 0.0330) when compared to their respective controls. However in the IP and oral BC group, a notable yet insignificant increase was seen. Finally, at 6 h post treatment, the IL-12 levels of IP BC and oral BC groups were lower than that of their respective controls, yet a significant increase in the serum level of IL-12 was seen in IP and oral BC group (P value = 0.0477) (Fig. 2).

Mice Survival
By day 20, 9 out of 10 mice were dead in IP BC group and oral BC group (10% survival rate in both groups); however 8 out of 10 mice were dead in IP and oral BC group (20% survival rate), yet none of the mice in the untreated and sterile water groups survived beyond day 19 (0% survival) (Fig. 3).
The survival results were further evaluated by generating the Kaplan-Meier survival curves showing the probability of survival in a given period of time. The statistical significance of the results obtained was assessed by determining the p-value; p-value≤0.05 was considered statistically significant. The IP BC group and the oral BC group did not show significant survival results when compared to the sterile water and untreated groups, respectively. However, IP and oral BC group showed statistically significant survival results (P value = 0.047) when compared to its control, sterile water group (Fig. 4). Nonetheless, the remaining one mouse from IP BC group showed complete response to BC treatment (probable cure). This mouse did not show any solid tumor growth externally unlike the remaining mice challenged with B16F10 melanoma cells. Moreover, upon dissection no necrotic organs were visible. A blood smear from the dissected mouse was prepared with trypan blue, and no large cells similar to those seen in the previous cases were observed under the microscope. These observations further led to the conclusion that this mouse might have undergone probable cure.
The linear relations between mice survival and VEGF as well as IL-12 levels were examined. The relation between the survival ratio and VEGF levels approached linearity at 6 h post BC treatment (Pearson's Correlation coefficient= -0.839, p-value = 0.0750), which may indicate that VEGF levels tend to decrease in groups with better survival (Table 2).

Viability of B16F10 Melanoma Cells Incubated with BC
After 48 h of incubation, significant decrease in viable melanoma cell count was obtained in the wells treated with 500 µL of 5 µg mL −1 (P value = 0.0003), 10 µg mL −1 (P value = 0.003) and 100 µg mL −1 (P value = 0.0001) of BC, when compared to the control wells. Total eradication of melanoma cells was obtained in the wells treated with 500 µL of 1000 and 5000 µg mL −1 of BC (Fig. 5). The in vitro experiment conducted on the viability of B16F10 melanoma cells was repeated three times and consistent results were obtained.

Viability of Mouse Mononuclear Cells Incubated with BC
When compared to the control well, the viability of mouse mononuclear cells remained relatively stable throughout the first four BC concentrations (5, 10, 100 and 1000 µg mL −1 ) after 48 h; however a slight insignificant decrease was seen with the highest BC concentration which is 5000 µg mL −1 (88% survival rate) (Fig. 5).

Discussion
The anti-inflammatory effect of BC and its ability to cure a patient's ulcerated breast cancer drove Prudden to assess the cartilage's effect in the clinical treatment of cancer (Prudden and Balassa, 1974;Morell and Daniel, 2014).
Although promising results were obtained, most scientists chose not to further investigate the anti-tumor activity of BC. For this purpose this study was carried out, in an attempt to determine the effect of BC on tumor growth in vivo, on serum levels of the pro-angiogenic cytokine VEGF and the anti-tumor cytokine IL-12 in tumor-bearing mice and on the survival of mouse melanoma and mononuclear cells in vitro.
In an attempt to test whether BC acts selectively on tumor cells, B16F10 melanoma cells and mouse mononuclear cells were separately incubated with increasing concentrations of BC for 48 h. It was observed that the viability of melanoma cells decreased when treated with BC and this decrease was further enhanced with time of exposure and amount of BC added, reaching a total eradication of tumor cells at 1000 µg mL −1 of BC after 48 h. In contrast, when mouse MNCs were incubated with increasing concentrations of BC, only a minute decrease in the viability was observed at 5000 µg mL −1 after 48 h; indicating that even at higher doses of BC MNCs remained viable. These results were consistent with (Durie et al., 1985) who demonstrated the anti-proliferative activity of BC against several tumor cell lines using the same BC concentrations. Additionally, the obtained result might further emphasize the function of BC as a strong biological modifier because it appeared that this toxicity was selective to tumor cells. Moreover, to further test this hypothesis of selective toxicity of BC, its activity in vivo was tested. Results showed that after receiving IP BC injections for 10 days consecutively, mice didn't show any weight, skin or hair loss for 3 months of observation; all of which are markers of toxicity. Hence, the observed results in-vitro and in-vivo, concurred with the hypothesis showing that BC has a selective toxicity against tumor cells and were also consistent with Prudden's result, who reported that BC presents no record of toxicity in animals and humans (Prudden, 1985).
When the anti-angiogenic activity of BC was evaluated in vivo, significant decrease in VEGF levels was seen in IP BC and oral BC groups and in the IP and oral BC group, however the significant decrease was delayed in the latter. The outcome of this experiment indicates that independent of the route of administration, BC causes a decrease in VEGF levels which is consistent with 2 previous studies; (Folkman et al., 1971) who demonstrated that an inhibitor secreted from BC was able to directly inhibit tumor capillary proliferation and (Langer et al., 1976) who purified a fraction from BC with a molecular weight of 16,000, demonstrated that it possesses the ability to inhibit cancer induced neovascularization. It is worth to note that, the reason behind the delayed but significant decrease of VEGF levels in IP and oral BC group might be due to the decreased BC administration dose divided between oral and IP.
The stimulatory effect of BC on Antigen Processing/Presenting cells was determined by quantifying the serum levels of Interleukin-12 (IL-12).
IL-12 possesses strong anti-tumor and anti-metastatic properties and it is mainly secreted by the Antigen Presenting Cells (APC) including macrophages (Zeh III et al., 1993;Valerie et al., 1993;Gately et al., 1994). The ability of BC to activate macrophages was reported (Prudden, 1985;Durie et al., 1985;Rosen et al., 1998;Morell and Daniel, 2014). In this context, our results indicated activation of macrophages indirectly through the significantly increased levels of IL-12 seen at 4 h post-BC treatment in IP BC and oral BC groups and at 6 h post-BC treatment in IP and oral BC group. The delayed increase of IL-12 levels seen in the sera of the treated groups might be explained by the fact that BC needed more time to exert its stimulatory effect on the immune system. Moreover the decreased IL-12 levels seen in the sera of IP BC and oral BC groups at 6 h post treatment might be due to what is known as the oscillating effect of a biological response after administration of the drug. The oscillating effect is when a cycle of decrease and increase in physiological responses are seen to a given drug or substance after administration. The response doesn't remain low or high it represents as a transient cycle (Rando et al., 2010).
As for the survival results, IP BC group and oral BC group did not show significant survival results when compared to their respective controls (10% survival) as opposed to the IP and oral BC group which showed significant survival rate (20%) compared to its control group. These results were consistent with the report of (Romano et al., 1985) who stated that the probable cause of the decreased cure rate obtained with his patients might be due to the utilization of the subcutaneous route only, while Prudden advised using both subcutaneous and oral administrative routes to maximize bovine cartilage's effect (Prudden, 1985). According to (Prudden, 1985), the maintenance dosage is given orally; however, every treatment should start with injections which are regarded as a "loading phase". The loading phase is considered complete when a total of 2000 mL of cartilage is injected into the patient. This might also explain why mice in IP BC group and oral BC group which showed significant decrease in serum levels of VEGF did not show significant survival rates.
The mouse from IP BC group which might have shown probable cure even after discontinuing the treatment is similar to one of Prudden's patients who was completely cured (Prudden, 1985). Additionally, the case of this mouse resembles the situation where some cancer patients respond to cancer therapy whereas others don't. The underlying cause of this incident might be multifactorial including the composition of the normal microbial flora and the important role played by some specific microorganisms present in it. These microorganisms might inhibit tumor development through immune system activation and help promote complete response to cancer therapies (Poutahidis et al., 2014;Sivan et al., 2015).

Conclusion
It appears that BC possesses selective anti-tumor activity. This activity seems to give better results when both routes of administration of BC are utilized as suggested by Prudden. Moreover, improved results could be achieved by increasing the population number, broadening the number of cytokines quantified and using more accurate oral administration methods.