Soft-tissue sarcomas have an incidence of about two to four cases/10⁵ people and account for approximately 2.6% of sarcomas [1,2]. Treatment of soft-tissue sarcomas with doxorubicin and ifosfamide is still barely satisfactory [3]. New treatment strategies are needed, including new chemotherapeutic agents like tumor necrosis factor (TNF)-related apoptosis in esophageal, colon, pancreas, and liver carcinomas in vitro and in vivo, which was also analyzed and suggested to be an inducer of apoptosis in human fibrosarcomas [4]. To the present time, therapy of fibrosarcomas by surgical resection has been one of the choices. However, most sarcomas are located in the limbs, vessels, or nerves, which makes treatment very difficult [5,6]. Although perfusion with chemotherapeutic agents such as melphalan and TNF-α appears efficacious in some cases, melphalan and TNF-α induce side effects, and the possible loss of the treated limb is still a problem [7,8]. Therefore, establishing an available option for using antimicrobial peptides (AMPs) as a drug to cure fibrosarcomas would be a potentially valuable clinical application.

Pardaxin, a member of a family of antimicrobial peptides (AMPs) from a marine fish (Pardachirus marmoratus), was studied as an amphipathic polypeptide neurotoxin composed of 33 amino acids [9]. The structural features of pardaxin that are essential for its cytolytic and cytotoxic activities were defined. For example, in the lipopolysaccharide (LPS)-pardaxin 4 complex, pardaxin 4 assumes a unique helix-turn-helix conformation that looks like a “horseshoe”. The LPS-bound structure of pardaxin 4 shows noteworthy differences from structures analyzed and determined in lipid micelles and organic solvents. Solid-state nuclear magnetic resonance (NMR) experiments indicated that the membrane orientation and interactions of pardaxins were dependent on compositions of lipids in the bilayers demonstrated C-terminal helix and a flexible turn/helix structure at the N-terminal [10]. Therefore, pardaxin forms stable or transient pores in zwitterionic lipid vesicles which causes them to release their contents without loss of vesicle integrity, suggesting that pardaxin permeabilizes vesicles more efficiently by pore formation than by disruption [11]. We have previously shown that synthetic pardaxin inhibits the proliferation of HT1080 cells in a dose-dependent manner and induces programmed cell death in HeLa cells. DNA fragmentation and increases in the subG1 phase and caspase-8 activities suggested that pardaxin caused HeLa cell death by inducing apoptosis, but it had a different mechanism in HT1080 cells [12]. Pardaxin-treated HT1080 cells showed elevation of caspase-3/7 activities, disruption of the mitochondrial membrane potential, and accumulation of reactive oxygen species (ROS) production, which suggested that pardaxin may be a potential anticancer agent for selectively inducing apoptosis in cancer cells [13]. Although many AMPs such as pardaxin were identified as candidate drugs for anti-cancer treatment, there are no reports of pardaxin being used to treat tumors in a murine system.

Mutatect MN-11 tumor cells were derived from a poorly differentiated fibrosarcoma from a C57BL mouse and were grown subcutaneously in syngeneic C57BL/6 mice [14]. A histochemical examination of tumors from injected MN-11 tumor cells in mice revealed the presence of infiltrating host cells, neutrophils, macrophages, lymphocytes, and mast cells in the tumor microenvironment [15]. Neutrophils can produce superoxide, nitric oxide, and myeloperoxidase that cause an influx of neutrophils, which is associated with necrosis. These can be a source of chemotactic factors that promote the infiltration of neutrophils, which can cause chronic inflammation [15]. Therefore, appropriate activation of macrophages or immune-related gene expressions prior to surgery [16] is one method to reduce some of the detrimental effects caused by surgical operations to remove fibrosarcomas from patients. Recently, experimental results confirmed the genomic instability of cancer cells and showed a feature of tumor progression [14,15]. The contribution of AMPs to inhibiting tumor growth and reducing products of inflammatory cells has not been assessed in experimental studies. Most AMPs possess antibacterial and immunomodulatory functions against pathogen infections in hosts [17,18,19]. Treating fibrosarcoma cells with pardaxin can be employed to trigger caspase-dependent and ROS-mediated apoptosis [13]; hence, using an AMP to treat cancer patients either by inhibiting tumor growth or preventing bacterial infections may play an important role in clinical applications.

The present study was designed to further explore the mechanism of action of pardaxin therapy against fibrosarcomas. For this purpose, we established in vitro and in vivo mouse models of fibrosarcomas and examined the effect of treatment on the expressions of immune-related genes. We thus performed in vitro and in vivo analyses to demonstrate that pardaxin suppressed tumor growth and was involved in immune-related gene expression in fibrosarcoma HT-1080 cells. Our results showed that synthetic pardaxin-induced apoptosis in HT-1080 cells was caspase-dependent and ROS-mediated.

Pardaxin was first isolated from P. marmoratus in the early 1980s, and its biological properties were initially noted: Inducing transcription of a vesicular stomatitis virus in the absence of nonionic detergents and making the virion membrane permeable to nucleoside triphosphates [20,21]. Subsequently, pardaxin became of significant interest as a potential biophysical model to study ionic channel selectivity [22] and as a potential drug for new target sites in bacteria [23]. In addition, it also showed signs of antitumor activity in in vitro studies with several cancer cell lines [13].

Several studies have suggested that pardaxin possesses potent antitumor activity in cancer cell lines [12,13], although the effects in vivo and the mechanisms behind this activity are still not very clear. MN-11 cells were derived from an MC1A fibrosarcoma in a male C57BL/10 mouse. They are very sensitive to mutations by clastogenic agents in vitro and in vivo [14]. Recent studies have suggested that pardaxin induces apoptosis by triggering caspase-dependent and ROS-mediated apoptosis in human fibrosarcoma HT-1080 cells [13]. In the present study, pardaxin caused striking concentration- and time-dependent growth inhibition of MN-11 cells. A 24-h exposure to 13 µg/mL resulted in a great inhibition ratio. Pardaxin also presented apoptosis-inducing activity. Pardaxin treatment of MN-11 cells resulted in morphological changes typical of apoptosis; these changes included condensed chromatin, nuclear envelope lysis, and shrunken cytoplasm. A colony formation assay demonstrated that pardaxin inhibited MN-11 invasion. In previous studies, pardaxin disrupted the lipid bilayer dependent on membrane components. Those results demonstrated that pardaxin was located on the surface of lipid bilayers constructed of POPC and was inserted into lipid bilayers devised of DMPC [24]. Moreover, plasma membranes of cancer cells are composed of anionic PS and O-glycosylated proteins [25,26]. Comparison of these features suggested that AMPs could not electrostatically interact with non-transformed cells such as tumor cells, the plasma membranes of which are composed of zwitterionic membrane components, such as sphingomyelin, phosphatidylethanolamine [27], and phosphatidylcholine. It was established that possessing the lipid composition allows AMPs to selectively kill malignant cells. This may be one of the mechanisms for the inhibition of tumor cell proliferation. Recently biophysical studies on pardaxin showed that pardaxins physically interact with cell membranes by forming pores or voltage-gated ion channels that disrupt cellular functions. It was shown that the membrane composition plays a vital role in its antimicrobial activity [28]. Cholesterol plays a very important role in the antitumor activities of a peptide. The addition of cholesterol and pardaxin to dimyristoyl phosphatidylcholine resulted in a very marked lowering of the transition temperature, which suggested that the peptide promoted redistribution of cholesterol in the membrane [29]. Moreover, the presence of cholesterol significantly reduced the backbone motion and the tilt angle of the C-terminal amphipathic helix of pardaxin, suggesting the involvement of cholesterol in the selectivity of the broad-spectrum antimicrobial activities of pardaxin [30].

At the moment, all methods of cancer treatment, including chemotherapy, radiotherapy, suicide gene therapy, immunotherapy, and surgery, strongly activate certain signal pathways such as linking to apoptosis to kill tumor cells. On the other hand, there is an urgent need for more effective and less toxic therapies for solid malignancies. Chemotherapy alone shows high toxicity and a low survival rate that may be due to cancer cells developing drug resistance [31,32]. In this study, to overcome such shortcomings, pardaxin was used as a drug for injection into the intratumoral space to eradicate surviving clonogenic cells, and showed significant results in preventing tumor growth. Pardaxin injected into tumor sites of MN-11 cells in mice also did not induce any side-effects according to our results (Supplementary Figure S1). Recently, cisplatin rapidly changed from the interstitial space to the tumor site, which prevented potentially fatal toxicities in a model of feline solid tumors with electrochemotherapy (ECT) treatment [33,34]. The ECT concept was established to increase susceptibility of tumor cells to permeabilizing pulses compared to normal tissues, and it produced semi-selective drug delivery [35,36,37], thereby diminishing side effects. Pardaxin is a membrane-lysing peptide, which indicates that a majority of host-defense peptides exert their action by permeabilizing microbial or tumor membranes [38]. Conformational analysis of pardaxin peptides and a study of their model membrane-permeabilizing activities indicated that their selective activities could be explained by their biophysical properties [39]. As described above, our research results suggested that pardaxin might be able to replace the ECT method when pardaxin is directly added to the surgical protocol. There was unanimous support of results obtained in a previous study by our group with another AMP as a drug directly injected into tumor regions which reduced tumor growth in a mouse system [40]. Indeed, one of the aims of our study was to evaluate whether AMPs (such as pardaxin) can be a future anticancer protocol through injection into tumor sites. The simple strategy of using pardaxin provides another choice for cancer therapy, while avoiding harmful or fatal side effects. It is very important to develop compounds such as AMPs that can be used to selectively kill tumor cells [41]. There are other AMPs that have been shown to have antimicrobial activities such as LL-7–27, a 21-residue peptide segment (LL-7–27; RKSKEKIGKEFKRIVQRIKDF), has been shown to exhibit potent activity against microbes but not against erythrocytes. Another 15-residue peptide named G15 was shown to exert potent activity against microbes but not against human Jurkat cells. These results show the susceptibility of bacteria and the resistance of erythrocytes and human cells and may have implications for designing membrane-selective therapeutic agents [42,43].

In view of emerging data that pardaxin treatment caused increased ROS production, followed by the triggering of mitochondrion-dependent apoptosis by activating caspase-3/7 activities [13], our results showed that treatment with pardaxin for 7 days downregulated IFN-β1 and Bcl-2 (Figure 5 and Figure 7a). Bcl-2 family proteins are essential regulators of mitochondrion-related apoptosis via modulation of mitochondrial membrane permeabilization to control the release of apoptosis inducers such as apoptosis-inducing factor (AIF) and cytochrome [44,45]. Downregulation of Bcl-2 gene expression may be the reason why pardaxin inhibited the invasive and proliferative properties of MN-11 cells and inhibited colony-forming and angiogenic activities [46]. Intratumoral administration of pardaxin inhibited tumor growth, possibly by downregulating Th1-type cytokine production (TNF-α and IFN-γ) and inducing apoptosis of fibrosarcomas by decreasing the Bax/Bcl-2 ratio and activating caspase-3. In this study, after treatment with pardaxin for 14 days, we found that downregulation of SOCS-3 gene expression may have led to increased death of MN-11 cells in mice (Figure 5 and Figure 7b). We found a considerable increase in activation of the proapoptotic caspase-7. We showed that pardaxin could inhibit IFN-γ and downregulate SOCS3 expression.

Pardaxin is a peptide that effectively inhibited MN-11 cells in C57BL/6 mice. Nowadays, it is imperative to find novel anticancer drugs with increased selectivity. In the present study, we show that pardaxin induced apoptosis through an intrinsic apoptotic pathway in cancer cells, thus showing its potential to serve as an anticancer drug in in vivo experiments. To sum up, our study provides evidence from molecular and animal studies, which shows that pardaxin induces death of murine fibrosarcoma MN-11 cells. The data suggest that pardaxin retards tumor development through suppressing both cell proliferation and tumor vascularization. These findings suggest that pardaxin is a peptide that can be used as an effective drug to treat solid neoplasms. Further studies with a greater number of feline or dog patients are needed, also with a view of possible translation to therapies for human fibrosarcomas.