Cancer is a disease characterized by abnormal and uncontrollable cell division. It has become the second leading cause of death globally, with 9.6 million deaths reported by the WHO in 2018 (Bray et al., 2018). Conventional cancer treatments, such as radiotherapy and chemotherapy, are unavoidably accompanied with adverse effects as they are designed for mass killings of cancer cells while lacking specificity. To further complicate matters, cancer cells may acquire resistance to chemotherapeutic treatments. Thus, the pursuit for a cancer-specific, non-resistant, and potent compound has been the cornerstone in anticancer drug discovery (Liberio et al., 2013). In this context, snake venom proteins are promising candidates for novel anticancer agents (Li et al., 2018). Snake venoms consist of evolutionary fine-tuned, pharmacologically active proteins and peptides to most mammalian cells (Mackessy, 2009; Casewell et al., 2013). The mammalian system-targeting property of snake venom proteins is typically specific and selective, thus making them a repertoire of bioactive molecules for the discovery of new therapeutic agents (Lewis and Garcia, 2003).

In Asia and Africa, envenomation by cobras (Naja spp.) causes fatal paralysis and debilitating complications resulted from extensive tissue necrosis (Ranawaka et al., 2013; World Health Organization [WHO], 2016). The tissue-necrotizing effect observed in cobra envenomation suggests cobra venoms contain cytotoxic components that can be further explored for anticancer potential. Studies have shown that virtually all cobra species produce a significant amount of cytotoxins in their venom proteomes, with abundances ranging from 20% to as high as ∼80% of the total venom proteins (Petras et al., 2011; Huang et al., 2015; Dutta et al., 2017; Tan et al., 2019; Tan et al., 2020). Cobra venom cytotoxins are polypeptides of 60–70 amino acid residues held by four disulfide crosslinks, forming a three-protruding fingers-like structure from the core of the molecule and is grouped under the three-finger toxin superfamily (Antil-Delbeke et al., 2000; Servent et al., 2000). Its unique structure gives rise to its diverse pharmacological actions, including cytolytic activity through pore formation on cell membranes, and disruption of membrane lipid bilayer integrity, thus initiating cell death (Konshina et al., 2011; Chan et al., 2016). Its cytotoxic properties, nevertheless, imply the possibility to harness cobra cytotoxins in the development of new anticancer drugs.

Earlier studies had shown that cytotoxins from a few cobra species, i.e., Naja atra (Chen et al., 2009; Tsai et al., 2016), Naja haje (Feofanov et al., 2005), Naja naja (Das et al., 2013), and Naja oxiana (Ebrahim et al., 2016) were cytotoxic to some cancer cell lines. The anticancer properties of various cytotoxin subtypes from most other cobra species, however, remain largely uncharacterized. In this regard, the two most common cobra species in Southeast Asia, i.e., Naja sumatrana (equatorial spitting cobra) and Naja kaouthia (monocled cobra), which cause significant tissue necrosis in envenomed patients (MOH Malaysia, 2017), are valuable sources of venoms to elucidate the prospect of cytotoxin-derived anticancer therapeutics (Ebrahim et al., 2016). This study set to investigate the anticancer properties of the venoms of N. sumatrana and N. kaouthia as well as their purified cytotoxins in the breast, prostate, and lung cancer cell lines. The study further delved into the specificity and selectivity of the cytotoxins and verified the cell death mechanisms underlying their cytotoxic activities.

The present study demonstrated the cytotoxic activities of N. sumatrana and N. kaouthia venoms in lung (A549), prostate (PC-3), and breast cancer (MCF-7) cell lines. The venom cytotoxic effects were more prominent in A549 and PC-3 cell lines than MCF-7 cell line, suggesting higher tissue specificity toward the lung and prostate cancer cell lines. The major cytotoxins from both venoms were purified to homogeneity, validated by LCMS/MS, and investigated for their cytotoxic properties in the cancer cell lines. The respective cytotoxins also exhibited a similar trend of tissue specificity as with their venoms toward the A549 and PC-3 cancer cell lines, unveiling the therapeutic potentials of cobra-derived cytotoxins against these cancers. The purified CTXs were apparently more potent (with lower IC₅₀) compared to the respective crude venoms, supporting that these proteins are the main cytotoxic component in the cobra venoms implicated in tissue necrosis (Tan and Tan, 2016). Sequences of the two cytotoxins indicated that the NS-CTX and NK-CTX belong to P-type and S-type subgroups of CTX, respectively. The P-type and S-type cytotoxins vary in their amino acid sequence at the tip of loop II: the former is characterized by the presence of Pro³¹ within the phospholipid binding site near the tip of loop II, whereas the latter has Ser²⁹ within the same but more hydrophilic region. The binding of virtually all CTX to phospholipid membranes involves a phospholipid binding site at loop I; however, the presence of Pro³¹ in P-type CTX gave rise to an additional phospholipid binding site at loop II, thus enabling the P-type CTX to exhibit higher fusion and binding activity than the S-type CTX to cell membrane (Chien et al., 1994; Suzuki-Matsubara et al., 2016). In this study, the higher potency of NS-CTX compared with NK-CTX is therefore consistent with the differential cytotoxicity reported previously for the P-type and S-type of CTX. P-type cytotoxins such as NS-CTX presumably exert a deeper penetration into the lipid bilayer of anionic phospholipid-containing membrane and destabilize the membrane more efficiently (Chien et al., 1994; Almeida et al., 2002; Dubovskii et al., 2005).

Recent proteomic studies validated that CTXs are indeed abundant in N. sumatrana and N. kaouthia venoms of various locales, constituting at least 20% of the total venom proteins (Yap et al., 2014; Tan et al., 2016a; Liu et al., 2017; Xu et al., 2017). The two major subgroups of CTX, i.e., P-type and S-type CTX, distribute variably among different Naja species. In this study, the purification of the P-type CTX and S-type CTX from N. sumatrana and N. kaouthia venoms, respectively, is consistent with the venom proteomes (Yap et al., 2014; Tan et al., 2015) and venom-gland transcriptomes of the respective species (Tan K.Y. et al., 2017; Chong et al., 2019). The genome of the Indian cobra (N. naja), on the other hand, indicated that the genes of both P-type and S-type CTX could well be conserved in other cobra species (Suryamohan et al., 2020). Nevertheless, environmental factors, particularly diet, can impact protein expression and modulate the dominant phenotypic expression of CTX in the cobra venom, contributing to the differential cytotoxicity of the venom as observed in this study.

The discovery of toxin compounds with low IC₅₀ comparable to or superior than commercially available anticancer drugs is not uncommon, and this has piqued great interest into the medicinal attribute of toxins as anticancer therapeutics (Ebrahim et al., 2016; Attarde and Pandit, 2017). The potent cytotoxicity of toxins, however, may cause collateral damage to normal or non-cancerous cells, defying the therapeutic use of these new drug candidates. Selectivity study is hence warranted, and in this study, both NS-CTX and NK-CTX exhibited significantly higher IC₅₀ values in the normal lung NL20 cell line compared with the cancerous A549 cell line, indicating good selective cytotoxicity for lung cancer cell line. The selectivity of cytotoxins, however, was absent in the prostate (PC-3) and breast (MCF-7) cancer cell lines, implying limited application of the toxins against these cancer cells. The higher selective cytotoxicity of the cytotoxins in A549 cell lines could be attributed to the unique characteristic of non-small lung cancer cell membrane, in which phosphatidylinositols are present abundantly (Marien et al., 2015). Earlier, Wang et al. (2006) reported that cobra cytotoxin interacted and bound preferentially with negatively charged lipids like phosphatidylinositol. It is possible that the preferred binding facilitated the membrane insertion and internalization of CTX into the A549 cancer cell lines.

In flow cytometry, both NS-CTX and NK-CTX induced late apoptosis in PC-3 and MCF-7. The finding is consistent with a cytotoxin isolated from N. naja (Das et al., 2013), N. atra (Yang et al., 2006), and N. oxiana (Ebrahim et al., 2014), which demonstrated late apoptosis as the primary cell death mechanism in CTX-treated leukemia and breast cancer cell lines. However, the late apoptosis was not observed along with the upregulation of caspase 3/7 activity in the present study. This finding indicated that the late apoptosis in PC-3 and MCF-7 by NS-CTX and NK-CTX is likely mediated through caspase-independent pathways, possibly through AIF, endonuclease G, or HtrA2 (Lorenzo and Susin, 2004; Chen and Wong, 2009; Kim et al., 2009). The elucidation of the exact pathways requires further investigations. Although NS-CTX and NK-CTX showed commendable cell-type selectively toward the lung A549 cell line, the flow cytometric finding revealed that the cells predominantly underwent necrosis, consistent with the absence of phosphatidylserine exposure and caspase activation.

To further understand the anticancer properties and cell death mechanisms of cobra venoms and their cytotoxins, we reviewed and tabulated findings from related studies involving a variety of Naja spp. reported thus far, which included N. atra, N. haje, N. naja, Naja nigricollis, N. oxiana, and N. sumatrana as well as N. kaouthia (current study), tested on assorted cancer cell lines (Table 3). Presently, most studies on the cytotoxicity and the implicated cell-death mechanisms were reported for N. atra cytotoxins (Tsai et al., 2006; Chen et al., 2009; Chien et al., 2009; Chiu et al., 2009; Lin et al., 2010; Wu et al., 2013; Liu et al., 2019) and N. naja (Das et al., 2013), while studies on other Naja spp. cytotoxins were largely limited to cytotoxicity IC₅₀ reporting in cancer cells (Feofanov et al., 2005; Ebrahim et al., 2016; Attarde and Pandit, 2017; Conlon et al., 2020). The present study is the first to elucidate the cytotoxicity, selectivity, and cell death mechanism of the purified cytotoxins of N. sumatrana and N. kaouthia, tested in lung (A549), prostate (PC-3), and breast (MCF-7) cancer cell lines. Overview of the IC₅₀ of cytotoxins from various Naja spp. showed wide-ranging differential cytotoxicity (IC₅₀ values between 0.88 and 12.23 μg/ml) in the different mammalian cell lines, treated with cytotoxin between 12 and 72 h. However, a much broader range of IC₅₀ was seen in the purified cytotoxins of N. oxiana, N. haje, and N. nigricollis, with values wide ranging from 0.33 to 132 μM tested in lung, peripheral blood, breast, and colon cancer cell lines (Feofanov et al., 2005; Conlon et al., 2020). The cytotoxic potentials and cell death pathways were, apparently, highly variable depending on the types of cell, concentrations of cytotoxin, durations of treatment, and most likely, species of the cobra too. Although all cobra cytotoxins share homologous sequences, subtle differences in their conformational structures could have contributed to variable structure–activity relationships as the toxins act on different cell lines. On the other hand, the selectivity of cobra cytotoxin was infrequently reported, limiting the interpretation of the prospect of the toxin in anticancer drug development. The present study clearly demonstrated selectivity of NS-CTX and NK-CTX for the lung cancer cells but not for the prostate and the breast cancer cell lines. The finding, however, varied from that reported by Ebrahim et al. (2016) and Attarde and Pandit (2017) who observed selectivity of cobra cytotoxins in the breast cancer cell line MCF-7 (SI = 4.33–10). The discrepancy could be due to the fact that the cytotoxins were from different species (N. naja and N. oxiana), and that the normal breast cell line tested in the present study (184B5) was indeed the corresponding normal mammary gland epithelial cells, in contrast to the MCF10A cell line which was the immortalized, normal-like mammary gland epithelial cells (with fibrocystic disease background) used in the previous studies (Ebrahim et al., 2016; Attarde and Pandit, 2017).

In drug discovery, the cell death mechanism underlying the anticancer activity of cytotoxin is as important as its selectivity for cancer cells. Apoptosis is, in general, preferred over necrosis as necrosis-induced inflammation can be hazardous to normal tissues. The mechanistic study of cancer cell death induced by cobra cytotoxins has revolved mainly around N. atra and, to a small extent, N naja but virtually never been investigated for cytotoxins from other cobra species. The majority of cancer cell lines (breast, colon, oral cavity, and pleural effusion) treated with N. atra and N. naja cytotoxins appeared in the late apoptotic stage (Table 3), consistent with the present finding of late apoptosis in MCF-7 cell line treated with NS-CTX and NK-CTX. Furthermore, in the A549 lung cancer cell line, we observed necrosis in the cells upon treatment with NS-CTX and NK-CTX, a finding similar to that reported by Liu et al. (2019) in KG1a leukemic cell line treated with N. atra cytotoxin 1. The cytotoxin necrotizing pathway probably lacks attraction for further elucidation, although the authors suggested that N. atra cytotoxin 1 initiated necroptosis in KG1a cells that may have involved the stimulation of interferon, tumor necrosis factor, and T cell receptors (Liu et al., 2019).

The development of peptide-based anticancer drug has been gaining momentum in recent years due to the possibility of improved potency, efficacy, and minimal systemic toxicity. The cobra cytotoxins have such potentials but their highly evolved, diet-adapted cytotoxic nature might have limited the intended application as a safe and effective drug. The present study demonstrated such challenges and dilemma in developing these cytotoxic proteins as anticancer therapies. The cytotoxins from N. sumatrana (NS-CTX) and N. kaouthia (NK-CTX) were highly potent in inhibiting the growth of the lung (A549), prostate (PC-3), and breast (MCF-7) cancer cell lines but lack selectivity for the latter two cell lines. Although NS-CTX and NK-CTX exhibited good selectivity in the lung A549 cancer cell line (sparing the normal NL20 cell line), the cell death pathway was predominantly by necrosis, questioning its use as a safe anticancer candidate. The study underscores the need for a comprehensive, fundamental investigation that addresses the selectivity and cell death mechanism of venom proteins, in the quest of novel anticancer agents. The anticancer potential of the toxins, however, may be further improved through structural modification to deliver a more specific, molecularly targeted cancer therapy in the future.

All datasets presented in this study are included in the article/Supplementary Material.

CHT and KYT conceptualized the study, designed the study and methodology, and provided resources. HPC was involved in the investigation, analysis, data validation, and writing the original draft of manuscript. All authors reviewed, edited, and approved the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.