Sialic acid-binding lectin from bullfrog eggs inhibits human malignant mesothelioma cell growth in vitro and in vivo

Malignant mesothelioma is an aggressive cancer that results from exposure to asbestos. The therapeutic options for this type of cancer are limited; therefore, the development of novel therapeutic agents is urgently required. Sialic acid-binding lectin isolated from Rana catesbeiana oocytes (cSBL) is a novel therapeutic candidate for cancer, which exhibits antitumor activity mediated through RNA degradation. In the present study, we evaluated the effect of cSBL in vitro and in vivo. Xenograft-competent H2452 and MSTO human mesothelioma cell lines were treated with cSBL, and the pathway by which cSBL induces apoptosis was analyzed. In vivo studies were performed using nude mice inoculated with one of the two cell lines, and the effects of cSBL and pemetrexed were monitored simultaneously. Furthermore, the pharmacological interactions between the three agents (pemetrexed, cisplatin and cSBL) were statistically assessed. It was demonstrated that cSBL treatments caused morphological and biochemical apoptotic changes in both cell lines. Caspase cascade analysis revealed that an intrinsic pathway mediated cSBL-induced apoptosis. The administration of cSBL significantly inhibited tumor growth in two xenograft models, without any adverse effects. Furthermore, the combination index and dose reduction index values indicated that the cSBL + pemetrexed combination showed the highest synergism, and thus potential for reducing dosage of each drug, compared with the other combinations, including the existing pemetrexed + cisplatin regimen. cSBL exerted prominent antitumor effects on malignant mesothelioma cells in vitro and in vivo, and showed favorable effects when combined with pemetrexed. These results suggest that cSBL has potential as a novel drug for the treatment of malignant mesothelioma.


Introduction
Malignant mesothelioma is an aggressive cancer of the mesothelial cells of serous membranes, involving the pleural and peritoneal spaces, which results from exposure to asbestos [1][2][3]. The mechanisms underlying the induction of DNA damage by asbestos fibers in mesothelial cells remain unclear. The production and use of asbestos is now forbidden in the majority of PLOS ONE | https://doi.org/10.1371/journal.pone.0190653 January 3, 2018 1 / 15 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 (mitochondrial) pathway [32]. The RNase activity was also determined to be critical for apoptosis induction in MDA-MB231 human breast cancer cells, as an amino acid-replaced mutant of cSBL that lacked RNase activity did not exhibit the apoptosis-inducing effect, even when internalized into the cells like native cSBL [33]. The efficacy of cSBL on malignant mesothelioma cells has previously been reported [34,35]; Even though cSBL hardly show cytotoxicity to normal mesothelial cell Met5A, it efficiently reduced the viability of H28 malignant mesothelioma cells, and exhibited synergistic effects with TRAIL and pemetrexed on these cells. In our previous study, in vivo experiments with cSBL were performed using mice transplanted with related ascites carcinoma, Ehrlich, Mep II and Sarcoma 180 cells. cSBL prolonged their survival at non-toxic dose levels [25]. However, to date, the effect of cSBL on human malignant mesothelioma cells in vivo has not been elucidated. In the present study, to assess the therapeutic potential of cSBL on malignant mesothelioma, we conducted an in vivo study of cSBL using human malignant mesothelioma cell xenografts, and analyzed its antitumor effects on these xenograft-competent cells.

Animals
Eggs-bearing bullfrogs (domestically caught) and 5-week-old male nude mice (BALB/c nu/nu Slc) were purchased from Japan SLC, Inc (Shizuoka, Japan). All animal experiments were carried out in accordance with the Guidelines for Animal Experiments of the Tohoku Medical and Pharmaceutical University (permission number: A16012-cn). Housing condition of the mice was kept under standard conditions approved by the institutional guidelines with free food-and water-consumptions.
In vivo experiment H2452 (5×10 6 cells) and MSTO (2×10 6 cells) cells were mixed with an equal volume of icecooled Corning™ Matrigel™ Basement Membrane Matrix (Corning, NY, USA), and an aliquot (100 μL) of suspended cells was subcutaneously injected into the lower backs of the mice. 2-4 weeks after inoculation (day 1), mice bearing tumors of 100-150 mm 3 in volume were randomly divided into three groups, with 10 mice in each group. Group 1 was injected with 1 mL/ kg vehicle (PBS) as the control. Group 2 was daily injected intraperitoneally with 100 mg/kg pemetrexed dissolved in sterile PBS on days 1-5 and 15-19. The method of administration the dosage of pemetrexed selected were based on the previously reported maximum tolerated dosage [36,37]. Group 3 was injected intratumorally with 2.5 mg/kg cSBL, twice weekly for four weeks. Body weights and tumor sizes were measured twice weekly. Solid tumor volumes were calculated as follows: 0.4×A×B 2 , with A and B representing the long and short tumor diameters (measured in mm), respectively. Tumor growth and body weight changes were evaluated as the ratio of each value against the baseline (day 1).

Drug combination studies
The effect of combination treatment on cell viability was determined using a WST-8 assay. Cells (5×10 4 cells/mL) were cultured in 96-well plates (100 μL/well). The concentration of pemetrexed, cisplatin, or cSBL was based on the IC 50 values obtained in the single-treatment experiments conducted in our prior study [35]. After 72 h, the cells were incubated with Cell Count Reagent SF (Nacalai Tesque Inc., Kyoto, Japan) at 37˚C in a 5% CO 2 atmosphere for 1-4 h. The absorbance of the resulting product at 450 nm was measured, and the background absorbance at 650 nm was subtracted. Combination Index (CI) and Dose Reduction Index (DRI) values were calculated using CompuSyn software (ComboSyn, Inc., Paramus, NJ), as described by Chou et al [38]. The experiments were conducted in triplicate. CI = 1 indicated an additive effect; CI<1 indicated a synergistic effect; CI>1 indicated an antagonistic effect. DRI = 1 indicated no dose reduction, whereas DRI>1 and <1 indicated favorable and unfavorable dose reductions, respectively.

Statistical analysis
The results from !3 independent experiments, each performed in triplicate, are expressed as the mean ± standard deviation. Statistical analyses were conducted using GraphPad Prism 5.0, and comparisons were made using one-way analysis of variance (ANOVA) or two-way ANOVA followed by Bonferroni's post hoc test. A P-value of <0.05 was considered statistically significant. In vitro/vivo antitumor effect of cSBL

cSBL induces apoptosis to H2452 and MSTO cells
In order to investigate the antitumor activity of cSBL on xenograft-competent malignant mesothelioma cells, H2452 and MSTO cells were treated with cSBL and the antitumor mechanisms were analyzed. The percentage of Annexin V-positive cells was significantly increased in both H2452 (16.13%, 72 h) and MSTO (40.05%, 72 h) cells (Fig 2A and 2B). In addition, chromatin condensation and nuclear collapse were observed in the two cell types treated with cSBL In vitro/vivo antitumor effect of cSBL ( Fig 2C). Alterations to nuclear morphology were detected by High-Content Analysis Systems and numerically output as fragmentation indexes. As shown in Fig 2D and 2E, cSBL provoked significant nuclear morphology changes in time-dependent manner.

cSBL-induced apoptosis is mediated by the intrinsic pathway
To obtain further insight into the mechanisms of cSBL-induced apoptosis in H2452 and MSTO cells, the activation of three key caspases was analyzed chronologically. The expression levels of activated caspase-9, -8 and -3 were detected by western blotting, and the substantial enzymatic activities of these caspases were evaluated by fluorometric analysis. As shown in Fig  3A and 3B, increased levels of activated caspase-9 were observed from 6 h and 1 h in H2452 and MSTO cells, respectively. After that, activated caspase-8 began to be observed from 24 h in H2452 and 6 h in MSTO cells. The appearance of activated caspase-3 was recorded from 48 h in H2452 and 24 h in MSTO cells. Consistently, the enzymatic activity of caspase-9 was significantly enhanced from 1 h in H2452 and MSTO cells, and the levels of caspase-8 and -3/7 increased almost simultaneously (H2452: 48 h; MSTO: 24 h; Fig 3C and 3D). Thus, caspase-9 was activated prior to caspase-8 and -3, indicating that the intrinsic apoptotic pathway was involved.

cSBL inhibits cancer cell proliferation without inducing weight loss in xenograft models
To examine the effects of cSBL on tumor growth in vivo, nude mice were inoculated with H2452 and MSTO cells. cSBL was administered intratumorally and the effects of pemetrexed were also assessed according to previously reported experimental conditions [36,37] (Fig 1). The rate of changes to animal weights and tumor volumes were monitored following the administration of each agent. As shown in Fig 4A and 4B, body weight changes were not observed in any of the groups. In the H2452 xenograft model, cSBL and pemetrexed each significantly inhibited tumor growth compared with the PBS group (P<0.05), and the cSBL group showed a growth inhibition effect earlier (after 36 days of treatment) than that of the pemetrexed group (after 47 days of treatment) (Fig 4C). Conversely, for the MSTO xenograft group, significant inhibition of cancer growth was observed only in the cSBL-treated group after 29 days of treatment (P<0.05; Fig 4D).

cSBL and pemetrexed exhibit a strong synergistic effect in H2452 and MSTO cells
Finally, we performed the in vitro combination study of cSBL with other reagents. In addition to pemetrexed, cisplatin, an existing drug for malignant mesothelioma usually used in combination with pemetrexed, was chosen for the test reagent. Pharmacological interactions between these three agents were investigated by evaluating the viability of H2452 and MSTO cells treated with pemetrexed + cisplatin, cSBL + pemetrexed, and cSBL + cisplatin. The drug concentration in each combination regimen was based on the IC 50 value for each agent previously determined via single treatments [35]. The viability curves for each drug in single or In vitro/vivo antitumor effect of cSBL combination treatments are presented in Fig 5. In H2452 cells, all combinations decreased cell viability to a greater extent than each single treatment over the whole concentration range. Similar tendencies were observed in MSTO cells over a wide concentration range, although the combination effects appeared to saturate at a higher concentration. To evaluate the synergistic effect of each drug combination, CI values were calculated. At each experimental concentration (Fig 6A) in H2452 cells, the CI values for all combinations were <1, indicating that all combinations were synergistic. cSBL + pemetrexed showed the highest synergistic effect at all concentration points. In MSTO cells, CI values at the highest two concentration points of all combinations exhibited antagonism rather than synergism; however, cSBL-containing combinations (cSBL + pemetrexed; cSBL + cisplatin) exhibited high synergism in the mid-low concentration range. The pemetrexed + cisplatin combination in MSTO cells showed dispersion and high CI values at various concentrations. Furthermore, we calculated DRI values, representing the index of the fold-number that each drug combination dose could be reduced by compared with that of each drug alone (Fig 6B). In H2452 cells, all combinations had high DRI values. In MSTO cells, high DRI values were observed in the cSBL-containing combinations, particularly in the low concentration range. However, the pemetrexed + cisplatin combination had low DRI values at all concentrations.

Discussion
We previously demonstrated that cSBL induces apoptosis in H28 (sarcomatoid histological type), MESO-1 and MESO-4 (epithelioid type) cells, but not in normal Met5A mesothelial cells, by detecting elevated proportions of Annexin V positive cells following cSBL treatment [34]. Furthermore, from the investigations in which H2452 (epithelioid type) and MSTO (biphasic type) were utilized, in addition to aforementioned cell lines, higher cancer-selectivity of cSBL was observed compared with either pemetrexed or cisplatin in their antiproliferative effects [35]. The antitumor mechanism of cSBL in malignant mesothelioma has been well-documented in H28 cells; it was revealed that cSBL treatment activates the caspase cascade, the proapoptotic Bcl-2 family proteins Bik and Bim, as well as JNK and p38 MAPKs, consequently inducing apoptosis in these cells. However, although the effectiveness of cSBL against mesothelioma in vitro has been reported, the in vivo efficacy of cSBL has not been investigated to date.
Although H28, MESO-1, and MESO-4 cells did not show tumorigenicity in the nude mice used, we succeeded in establishing malignant mesothelioma xenografts with H2452 and MSTO cells. First, the antitumor effects of cSBL on these two cell lines were investigated in vitro. cSBL induced typical apoptotic changes, such as phosphatidylserine externalization, In vitro/vivo antitumor effect of cSBL nuclear condensation and fragmentation, in both cells in a time-dependent manner (Fig 2). Moreover, caspase-9 was activated by cSBL treatment earlier and more strongly than caspase-8, indicating that apoptosis was induced through the intrinsic pathway (Fig 3). In the in vivo studies, no obvious toxicities or body weight changes were observed during the experimental period in any group (Fig 4A and 4B). In both types of xenograft, significant tumor growth suppression was observed in cSBL-treated groups compared with control groups. In H2452 xenografted groups, cSBL showed a tumor-suppressive effect earlier than that of the pemetrexedtreated group, and the antitumor effect of pemetrexed was not observed in the MSTO xenografts (Fig 4C and 4D). The reason for the lack of effect by pemetrexed is uncertain; we speculate that the high growth rate of MSTO cells in the xenograft model may contribute to this phenomenon. We were unable to compare the effects of cSBL and pemetrexed directly due to the differences in the dosing conditions; however, our observations indicate that cSBL could potentially inhibit the tumor growth of mesothelioma without any toxicity, even if previously established pemetrexed administration had little or no effect. From these results, it was suggested that cSBL had the capability to inhibit tumor growth in xenografted mice. Thus, cSBL may be safely used, and further studies are required to determine the maximal tolerated dose of cSBL in order to optimize its efficacy.
Combination therapy, a treatment modality that combines two or more therapeutic agents to reduce the risk of drug resistance or adverse effects while simultaneously providing therapeutic anti-cancer benefits, is a mainstay of current cancer therapy [39]. In fact, combination treatments comprising pemetrexed and cisplatin are used for the treatment of mesothelioma as a standard regimen. We previously demonstrated that the cSBL + pemetrexed combination exerted stronger cytotoxicity and synergism compared with the pemetrexed + cisplatin combination in H28 cell lines. The cytostatic effect of pemetrexed and the cytotoxic effect of cSBL cooperated without any repulsion, although the effects of pemetrexed and cisplatin on cyclin A expression were counteractive when used in combination [35]. In the present study, we evaluated the generality of the prominent synergistic effect of the cSBL + pemetrexed combination, utilizing H2452 and MSTO cells, by calculating CI and DRI values. The cSBL + pemetrexed combination exhibited the highest synergism of the three combinations tested in both cell lines ( Fig 5B). Surprisingly, in MSTO cells, the pemetrexed + cisplatin combination appeared to be antagonistic rather than synergistic or additive at the most of concentration points tested. High DRI values ( Fig 5C) were calculated for all combinations, except for pemetrexed + cisplatin in MSTO cells. These results suggest that cSBL + pemetrexed may be a rational treatment combination for several types of malignant mesothelioma. On the other hand, the current gold-standard regimen for malignant mesothelioma, pemetrexed + cisplatin, may be ineffective, depending on the cell type, with respect to synergism (i.e., undesired adverse effects may easily occur in some circumstances, depending on the patient).
Although the combination of pemetrexed and cisplatin has been demonstrated to prolong the survival of patients with malignant mesothelioma, the median survival is only 12 months, and the response rate is~40% [12]. Thus, almost half of all mesothelioma patients are initially resistant, and all eventually develop resistance [40]. Therefore, researches to improve the malignant mesothelioma therapy have been actively attempt. The combinations of carboplatin and pemetrexed, or gemcitabine and cisplatin showed comparable outcomes with pemetrexed and cisplatin combination in phase-II trials [41][42][43]. The French Mesothelioma Avastin Cisplatin Pemetrexed Study (MAPS) demonstrated a statistically significant improvement in the median overall survival time using a combination of cisplatin, pemetrexed and bevacizumab, a monoclonal antibody that binds VEGF and blocks its interaction with the VEGF receptor [44]. In addition, several other molecular targeting and immunotherapeutic agents, such as anti-EGFR signaling agent and anti-programmed cell death 1 (PD-1) antibodies, are currently being investigated in clinical trials [43]. In this study, cSBL was demonstrated to induce apoptosis and inhibit tumor cell growth in xenografted mice. CI analysis also evidenced a prominent combinatory effect of cSBL with pemetrexed. As cSBL is a novel candidate anti-cancer agent that exerts antitumor activity through targeting RNA (which represents a novel class of potential therapeutic targets), it may provide a new option for the chemotherapeutic treatment of malignant mesothelioma. This is particularly true among patients with pemetrexed resistance, as cSBL was effective in pemetrexed-resistant cells, as shown in the current study and in previous report [35].

Conclusions
cSBL induced apoptosis in H2452 and MSTO cells via the intrinsic apoptotic pathway. In vivo, cSBL treatment inhibited tumor growth in multiple xenograft models without any undesirable adverse effects. A higher efficacy was achieved by the use of cSBL + pemetrexed in mesothelioma cells compared with pemetrexed + cisplatin. To the best of our knowledge, this is the first report to demonstrate the antitumor efficacy of cSBL in human malignant mesothelioma xenograft models. cSBL has potential as a novel treatment for malignant mesothelioma.