Bufalin is one of the most intensively studied compounds among all TVAs (Wang et al., 2018b; Lan et al., 2019; Soumoy et al., 2022). A study by Zhang et al. (2018) showed that bufalin worked as an anticancer agent via a p53-mediated mechanism in PCa cells both in-vitro and in-vivo. In p53-mutant DU145 cells and p53-wild type LNCaP cells, bufalin (5–100 nM, 48 h) treatment could upregulate the expression of cleaved poly (ADP-ribose) polymerase (PARP), and downregulate steroid receptor co-activator 1/3 (SRC1/3), AR and prostate specific antigen (PSA). This study showed that bufalin increased p53 expression in LNCaP cells, but decreased p53 in DU145 cells, however, cleaved PARP or p53 was not observed in p53-null PC-3 cells although inhibited proliferation was identified, suggesting a p53-mediated efficacy (Zhang et al., 2018). The microarray detection of certain mRNA levels indicated that in LNCaP cells, bufalin treatment increased p53 and its transcriptional target P21CIP1, as well as mRNAs related to cellular stress and DNA damage response, and certain senescence-associated genes, such as CYR61/CCNI, CTGF/CCN2 and CDKN1A, which were then been validated by the subsequent assays of cell cycle distribution (sub G_0/1) and the presence of senescence-like phenotype (Zhang et al., 2018). The knockdown of p53 could attenuate bufalin-induced apoptosis as indicated by the decreased level of cleaved PARP. Finally, in the in-vivo model of LNCaP xenograft, bufalin (1.5 mg/kg body weight, IP, daily) for 9 weeks inhibited tumor growth, resulting in a 67% decrease as compared to untreated group, without affecting body weight significantly which might suggest a safe profile. More importantly, in bufalin-treated tumors, phospho-p53 was increased, confirming the on-target effect and a network of bufalin with p53 (Zhang et al., 2018).

A recent study by Zhang et al. (2019) found that bufalin can alter the expression of both microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) that are critical for PCa (Zhang et al., 2019b). In CRPC DU145 and PC-3 cells, bufalin suppressed the cell viability in a dose-dependent manner, with an IC₅₀ value of 0.89 and 1.28 μM, respectively. At lower than the corresponding IC₅₀ (to be more specific, at half of the corresponding IC₅₀), bufalin could significantly reduce the migration and invasion of DU145 and PC-3 cells as confirmed by the wound healing assay and transwell assay. The authors screened lncRNA alteration after bufalin treatment (0.1–5 µM) using a lncRNA microarray, and they identified that HOX transcript antisense RNA (HOTAIR) was one of the mostly reduced (Zhang et al., 2019b). HOTAIR targets and inhibits miR-520b as confirmed by RNA immunoprecipitation assay; meanwhile, miR-520b can negatively regulate the expression of fibroblast growth factor receptor 1 protein (FGFR1) which plays a pivotal role in PCa progression and metastasis (Yang et al., 2013; Wang et al., 2019). The authors also investigated and confirmed the positive correlation of HOTAIR and FGFR1 with PCa bone metastasis, and that the overexpression of HOTAIR could reverse bufalin-induced cancer-suppressing effects. Thus, this study indicated that bufalin can inhibit PCa proliferation, migration and invasion via regulating the HOTAIR-miR520b-FGFR1 loop (Zhang et al., 2019b).

MiRNA-181, composed with subunits miRNA-181a and b, targets apoptosis-associated proteins such as Bcl-2 family members, functioning as a tumor suppressor (Liu et al., 2017; Pei et al., 2020). Zhai et al. (2013) found that in bufalin (10 μM, 24 h)-treated PC-3 cells, miRNA-181a, but not the others such as miRNA-10b, −17, 18a, 20a, 21, −106, −155, −221 and −372, was markedly upregulated (5-fold), which was later confirmed to be a dose-dependent manner (1, 10 and 15 µM) (Zhai et al., 2013). In PC-3 cells, bufalin (15 µM) significantly decreased the expression of Bcl-2, an anti-apoptotic protein, accompanied with caspase-3 protein activation (via testing the level of cleaved caspase-3), which is essential in promoting apoptosis (Kesavardhana et al., 2020). At the same time, the rescue experiments showed that bufalin-induced apoptosis and caspase-3 proteins activation can be partially reversed by miR-181a inhibitor co-treatment (100 nM), validating the targeted effects of bufalin toward miR-181a (Zhai et al., 2013).

Structurally, bufalin is a hydrophobic compound that may encounter poor absorption and bioavailability (Shao et al., 2021). Thus, Liu and Huang (2016) constructed an amphiphilic targeting brush-type copolymers that can deliver bufalin to CRPC cells, which exhibited controlled drug release and higher anticancer capability than free bufalin both in-vitro and in-vivo (Liu and Huang, 2016). This constructed BUF-loaded micellar nanoparticle BUF-NP-(G3-C12) was found to have an IC₅₀ value of 8.0 ng/mL, which was lower than that of free bufalin (which was 13.3 ng/mL) in CRPC DU145 cells; and consistent results were also observed in inducing apoptosis. In DU145 xenograft model, when used by intravenous injection iv) at an equivalent 1.0 mg bufalin/kg, BUF-NP-(G3-C12) showed significantly higher tumor-inhibiting effects than that of free bufalin. Importantly, it didn’t change body weight as compared to vehicle control, suggesting its safety (Liu and Huang, 2016). Further evaluation is clearly needed to develop it as a drug candidate for PCa.

Chen et al. (2017) reported that arenobufagin, among five bufadienolides including cinobufotalin, bufarenogin, 19-oxocinobufotalin and 19-hydroxybufalin, showed the highest potency in suppressing the progress of epithelial-mesenchymal transition (EMT) in PC-3 cells, leading to decreased ability of migration and invasion (Chen et al., 2017). Arenobufagin (8 nM) time-dependently (24, 36 and 48 h) downregulated EMT markers in PC-3 cells, including slug, zinc finger E-box binding homeobox 1 (ZEB1), snail, N-cadherin, vimentin and Twist1 as confirmed by the Western bolt experiment. In addition, β-catenin was reduced at both mRNA and expression levels by arenobufagin, which then lead to the downregulation of its downstream genes including Met, LEF, TCF, c-Myc and cyclin D1. These effects can be reversed by β-catenin overexpression, suggesting the network of arenobufagin with β-catenin. Arenobufagin (1 mg/kg) reduced tumor growth without altering the body weight or causing harms to major organs including heart, liver, spleen, lung and kidney. In the in-vivo PC-3 cells pulmonary metastases model, arenobufagin markedly reduced the number and size of tumor metastatic foci in lung tissues, suggesting its dual role in preventing tumor growth and metastasis, warranting further study (Chen et al., 2017).

Niu et al. (2018) reported the anticancer effects and the mode of action of another TVA, cinobufagin, in CRPC PC-3 cells. Cinobufagin could significantly suppress PC-3 cells proliferation, with an approximately IC₅₀ of 100 nM (24 h) or 50 nM (48 h), suggesting a dose- and time-dependent manner. When tested in colony formation, cinobufagin possessed a much lower IC₅₀ (slightly lower than 5 nM). Mechanistically, cinobufagin induced apoptosis of PC-3 cells via down-regulating anti-apoptotic MCL-1 protein (Niu and Qin, 2018). Cinobufagin appears to be much more potent than bufalin, which has an IC₅₀ of 1.28 μM in PC-3 cells.

In addition to its role in working alone to suppress PCa, TVA bufalin has also been found to work as a chemo-sensitizer when combined with other conventional therapeutics.

Bufalin was identified as a possible DNA topoisomerase II (Top II) inhibitor (Hashimoto et al., 1997; Pastor and Cortes, 2003). Previous in-vitro studies showed that sequential administration of different Top isomer inhibitors exhibited improved outcomes as compared to simultaneous administration, suggesting a feasible combinational strategy (Cho and Cho-Chung, 2003; Griffith and Kemp, 2003). Recently, Gu and Zhang (2021) investigated the combination of low-dose (0.4–0.8 mg/kg) bufalin with hydroxycamptothecin, a Top I inhibitor (Gu and Zhang, 2021). In this study, CRPC DU145 cells xenograft model in nude mice were constructed and treated by hydroxycamptothecin (2 mg/kg) combined with 0.4 mg/kg, 0.6 mg/kg or 0.8 mg/kg bufalin, respectively. The results showed that among all treatments, the combination of hydroxycamptothecin with 0.6 mg/kg bufalin showed the strongest tumor-reducing effect (93% inhibition) than the other two combinations or monotherapy, bufalin at 1 mg/kg (∼30% inhibition) or hydroxycamptothecin at 2 mg/kg (58% inhibition), without altering body weight significantly (Gu and Zhang, 2021). This combination, named as H6B, induced significantly higher apoptosis but reduced proliferating cell nuclear antigen (PCNA) proteins in the tumors than the other treatments as confirmed by the TUNEL assay and immunohistochemistry, respectively. Western blot assay showed that H6B increased pro-apoptotic proteins such as Bax, p53 and programmed cell death protein 4 (PDCD4); whereas it decreased anti-apoptotic proteins such as Bcl-XL and p-AKT (Gu and Zhang, 2021). While this study presented a possible combinational treatment that was safe and can be further validated in other models and even in humans, it remains unclear if H6B inhibit Top I/II in the treated tumor tissue.

Several TVAs have been confirmed to induce cardiac toxicity. Resibufogenin (0.2 mg/kg, iv) could significantly increase heart burden as indicated by contractile force in rabbit, cat and dog, leading to delayed afterdepolarization and triggered arrhythmias (Xie et al., 2001). These effects were partially mediated by its disturbance of Na⁺/K⁺-ATPase which caused calcium (Ca⁺) overload (Xie et al., 2001). Similarly, in human cardiomyocytes model, bufalin (30–300 nM) showed a biphasic effect on the contractility, which was strengthening contractility, accelerating conduction, and increasing beating rate at the earlier stage, while in the opposite when at the later stage (Li et al., 2020).

TVAs are known to cause neuron toxicity as reported previously (Brubacher et al., 1999; Dasgupta, 2003; Ma et al., 2007). In addition to the inhibition of Na⁺/K⁺-ATPase, in rat hippocampal neurons (Wang et al., 2014a), both resibufogenin and cinobufagin could also inhibit outward delayed rectifier potassium current (Hao et al., 2011), which may work together to induce toxicity in neuron system. However, we believe that more studies are needed to reveal the doses or concentrations on inducing human neuron cells related toxicity.

A study by Li et al. (2009) found that bufalin had an inhibitory toward recombinant human CYP3A4 in vitro, with an IC₅₀ of 14.52 μM, leading to increased elimination half-time, peak plasma level of midazolam (a substrate of CYP3A4) in the rat model. Thus, when being used with combination, adverse effects due to CYP3A4 inhibition of TVAs should be monitored and prevented.

It’s known that in mouse model the median lethal dose (LD₅₀) of bufalin in nude mice is 2.2 mg/kg (Tu et al., 2000), which is pretty close to the doses of achieving therapeutic effects of tumor inhibition (normally not more than 1 mg/kg), suggesting a very narrow therapeutic window, and that the accumulative TVAs may further worsen certain toxic effects. Thus, when being tested in humans, a close monitor of serum concentration is necessary.

The above literature review has summarized the application of TVAs in PCa (Table 1; Figure 3). Generally, TVAs could suppress PCa cells proliferation via inducing apoptosis and regulating certain miRNAs and lncRNAs; meanwhile, they also show activity in reducing PCa cells migration and invasion in-vitro and in-vivo through negatively regulating critical players involved in metastasis. It’s known that bufalin targets steroid receptor coactivators SRC-3 and SRC-1 (Wang et al., 2014b), while growing evidence suggests that bufalin is a multi-targeting or multi-functional agent, especially in the treatment of cancers.

By far, except for bufalin that has been extensively studied, the therapeutic applications of other TVAs in PCa are yet to be fully revealed. While we suspect that since they all share a very similar scaffold, there may be limited differences of the underlying mechanisms, requiring further validations.

It also comes to our notice that TVAs have been approved and/or under active clinical evaluations only in China. Currently, combinational therapies of using some of TVAs are also actively tested in clinical trials, such as the combination of thalidomide with cinobufagin to treat lung cancer cachexia (Xie et al., 2018). Other TVAs-related clinical trial was either conducted more than 10 years ago (Meng et al., 2009), or using formulations made of toad venom extraction rather than isolated single component (Meng et al., 2012; Wu et al., 2020b; Tan et al., 2021). Since both the active components (major and minor) and the associated mechanisms remain largely elusive, thus, these formulations will likely meet many obstacles to be approved in other countries outside China due to different new drugs regulations. More studies using corresponding isolated pure compounds are in urgent need to support their further evaluation in humans.

While the anticancer of TVAs in PCa can be confirmed in lab (in-vitro and in-vivo), much more works are needed before they can be eventually applied in patients worldwide. The authors propose that six future directions are worth trying.

(1) Rational design of TVAs derivatives or analogs via the assistance of computer-aided drug design (CADD). These structures of bufadienolides can serve as leading compounds that may undergo structural modification for improved target-binding and anticancer effects. By far, this research area is extremely undeveloped. Very few studies have been published, and most of them are focusing on the modification of hydroxyl groups at different positions (Sampath et al., 2022). Further and varied structural modification at other positions and functional groups are necessary.

Finally, more clinical trials in PCa are necessary to test the efficacies of TVAs including their pharmaceutical formulations.