Colorectal cancer (CRC) is ranked as the third most commonly diagnosed cancer type after the lung and breast cancers worldwide beyond its geographical variations and localized temporal trends (1). The global burden of CRC is expected to increase due to the huge rise in its yearly incidences around the world making it the fourth cause of cancer deaths worldwide (2). Despite the advancements in pharmacokinetics with introduction of novel therapeutics, nearly half of the CRC patient population is reported to have the disease relapse (3, 4). Chemoresistance, a feature by which the cancer cells become less sensitive to chemotherapeutic drugs forms one of the major reason for this disease relapse and pose a major challenge in cancer treatment regime (5). The factors contributing to chemoresistance are many, with cancer stemness, impaired drug transport as well as targeted gene mutations being the most highlighted ones (6). Though mutations in several genes has been extensively correlated to CRC development, TP53 (p53) forms one of the most commonly mutated genes in cancer (7). About 50% of CRC cases are known to pose this mutation (8). An impaired cross talk between mutant p53 and its apoptotic target genes with no transactivation and enhanced tumorigenic properties such as survival, proliferation, angiogenesis, and chemoresistance has been reported to contribute to CRC disease progression (9, 10).

p53 is regarded to be a tumor suppressor transcription factor, which upon stress particularly DNA damage inducing stimuli gets stabilized by posttranslational modifications primarily via phosphorylation of serine and threonine residues (11). This stabilization transactivates several target genes that regulates various cellular functions including cell cycle, apoptosis, senescence, DNA repair, and cellular metabolism which functionally leads to tumor suppression (12). Cancer related p53 mutations can lead to loss of its function as tumor suppressor while is reported to acquire a function termed as gain-of-function (GOF) that enriches the mutant p53 with oncogenic properties which promotes tumorigenesis (13, 14). With GOF, mutant p53 not only gets dominated with its pro-oncogenic properties but also, facilitates drug resistance (15, 16).

Oxaliplatin (OX), also known as Eloxatin is a third generation platinum based chemotherapeutic agent widely used in clinics for the treatment of CRC (17). Structurally, OX has a core platinum atom bound to an oxalate group-also called as the displacement group and diaminocyclohexane (DACH) group-which is a carrier ligand (18). OX is classified as an alkylating agent that exerts its cytotoxicity effects via introducing DNA damage in target cells (19). OX is known to produce DNA mono/di-adducts via forming DNA intra-/inter-strand crosslinks as well as DNA–protein crosslinks (20). DNA di-adducts formed by binding to N (7) site of the guanine are fatal, can lead to DNA lesion further inducing G2/M phase arrest or apoptosis (21). Despite its efficient mode of action, most of the cancer cells apart from its primary sensitivity to OX is found to ultimately acquire resistance (17). Most of the in vitro studies on the OX resistance in colon cancer cells are carried out in stable resistantly raised cell lines (17, 22).

The current study aimed to identify any existing link between p53 status of colon cancer cells and their sensitivity towards the chemotherapeutic drug OX. The response to genotoxic stress induced by OX in human derived HCT 116 (wild-type p53) and HT 29 (mutant p53) colon cancer cell lines were evaluated using various biochemical assays and the underlying molecular mechanisms were elucidated.

Colorectal cancer is one of the highest ranked cancer types in the world with its increasing incidence and mortality rates despite the advancements in cancer therapeutics (1). Over 50% of human CRCs are reported to have defective p53 expression ensued by a mutated TP53 gene and this defect often contributes to drug resistance, challenging the clinicians in devising chemotherapeutic strategies for management of cancer patients with this mutation (15, 37). In normal cells, p53 expression is induced in response to stressful conditions and the p53 activation indeed induce cell death via processes such as cell cycle arrest, DNA repair, senescence or apoptosis (38–40).

In the case of cancers, cells bearing wild-type TP53 gene expressing functional p53 protein is regarded to be a strong tumor suppressor while, mutated TP53 gene expressing a mutant p53 protein with GOF tactically overcomes the cell death pathways and make cells resistant to chemotherapeutic drug treatments (41, 42). The p53 mutations being one of the most frequently occurring mutations in colon cancer with its proven role in inducing resistance to therapeutic drugs, the mutant p53 marks itself as a prominent predictive molecular marker for chemoresistance (43–45). Hence, unraveling the mechanisms by which mutant p53 harboring cancer cells evade the cellular death processes induced by a genotoxic stress via chemoresistance is of clinical priority, so as to develop strategies that could enhance sensitization of these cells to various chemotherapeutics.

The current study was aimed to investigate the cytotoxic effects induced by the chemotherapeutic drug OX in wild-type TP53 harboring HCT 116 and mutant TP53 harboring HT 29 colon cancer cells as well as to elucidate the underlying molecular mechanisms in correlation to their p53 status.

In our study, the HCT 116 cells 48 h posttreatment with OX resulted in 70% growth inhibition, while the growth inhibition in HT 29 cells was significantly reduced accounting to only 40% (Figure 1A). This implied that wild-type p53-HCT 116 cells were sensitive to OX treatment while, the mutant p53-HT 29 cells exhibited an impaired sensitivity to the drug. The wild-type p53 expression is functionally renowned to be a tumor suppressor and is known to induce cell death via cell cycle arrest or apoptosis (46). While the mutant p53 accumulation and its GOF is reported to favor tumorigenesis by acquiring oncogenic properties such as survival, proliferation, migration and resistance to drugs (47).

In this study, although OX induced apoptosis in both the colon cancer cell lines after 48 h of treatment, the total apoptosis was comparatively higher in HCT 116 cells than HT 29 cells. As the wild-type p53 expression is known to be a transactivator of the Bax gene (48), we attempted to evaluate the expression of pro-apoptotic “Bax” gene product. The ratio of “Bax/Bcl-2,” a positive indicator of apoptotic process was estimated and we found the ratio to be significantly increased with the OX treatment in HCT 116 cells while, this ratio remained unaltered in HT 29 cells with OX treatment.

Studies have shown that a low Bax/Bcl-2 ratio can favor tumor survival as well as impart resistance to various kinds of cell death stimuli including genotoxic, radiation or hypoxia (49). Moreover, it has been reported that wild-type p53 protein is capable of binding to the promoter of BAX gene and transactivates its expression while, a mutant p53 protein lacks this functionality (24). Taken together, our data indicates that, the wild-type p53 in HCT 116 cells has transactivated the expression of Bax in HCT 116 cells making them sensitive to OX. The mutant p53 in HT 29 cells lacking transactivation function, along with the endogenous expression of anti-apoptotic protein Bcl-2 could be attributed to the impaired sensitivity of these cells to OX. Hence, our study implies the link between p53 status and the Bax/Bcl-2 protein expression ratio in determining the sensitivity of colon cancer cells to the genotoxic stress induced by OX.

The functional mechanisms of wild-type p53 expression is known to be mediated by the activation of its primary effector target p21 – a cyclin-dependent kinase inhibitor (CDKi) (26). Hence, we further investigated the expression profile of p21 protein in both the colon cancer cell types. Interestingly, we found a concomitant increase in the expression of p21 along with p53 after OX treatment in HCT 116 cells (Figures 3A,B) while, in HT 29 cells the OX treatment did not induce any p21 expression wherein, the p53 expression was observed to be constitutive rather than inducive (Figures 3C,D). These findings indicate that, OX induced apoptosis in wild-type p53-HCT 116 cells was mediated by p53/p21 pathway. The possible mechanism being that, the genotoxic stress by OX have induced wild-type p53 expression followed by p21 expression and the p53/p21 complex subsequently mediated the expression of Bax leading to apoptosis. The endogenous expression of mutant p53 protein in HT 29 cells, it’s accumulation in the absence of p21 along with the GOF with oncogenic features could be attributed to the impaired sensitivity to OX displayed by these cells.

Our data also indicated that in HCT 116 cells OX treatment inhibited the expression of anti-apoptotic protein-Bcl-2 where in, both p53 and p21 expression were induced by drug treatment (Figure 2D). In HT 29 cells OX treatment induced the expression of Bcl-2, with no p21 expression, while p53 expression was found to be endogenous (Figure 2F). A study by Kim et al. (50) has shown Bcl-2 as a target for both p53 and p21, as well as the p53/p21 complex together to regulate cancer cell invasion and apoptotic cell death by targeting Bcl-2 protein. Hence, our data further confirms the functional role of wild-type p53 to form p53/p21 complex inhibiting the expression of anti-apoptotic protein Bcl-2 leading to sensitization of HCT 116 cells by OX. While, in HT 29 cells, the endogenous expression of mutant p53, with no functional p53/p21 complex formation along with Bcl-2 expression could have attributed to the impaired sensitivity of these cells to OX.

The p53-dependent-p21 expression mediated growth inhibition could be ascribed to its mechanistic role either as “cell cycle regulator” inducing arrest of cells in G1 or G2 phase (51, 52) or “pro-apoptotic regulator” inducing apoptosis by upregulation of Bax (53). In accordance to this, we performed cell cycle analysis of both the colon cancer cell lines treated with OX using flowcytometry. Apparently, no significant cell cycle arrest in G1 or G2 phase was observed in any of these cell lines with the OX treatment (data not shown). All these data collectively suggest that, the significant growth inhibition of HCT 116 cells (harboring wild-type TP53) resulted from OX treatment was due to induction of apoptosis mediated by p53/p21-Bax pathway.

Posttranslational modifications including phosphorylation, acetylation, ubiquitination, methylation, and sumoylation of p53 protein are known to regulate its function inside a cell (54). Among them phosphorylation forms the key modification that stabilizes the p53 protein and thereby activates its functional role as a tumor suppressor in response to cellular stress or DNA damage (55). The stabilization and activation of p53 in response to platinum-based drug treatments (DNA-damaging) are reported to be facilitated through the phosphorylation of N-terminus serine residues especially Ser15, Ser20, and Ser37 by reducing p53 affinity to its negative regulator Hdm2 (29, 56).

In our study, we found a significant increase in the phosphorylation levels of p53 protein at serine residues: Ser15, Ser46, and Ser392 in HCT 116 cells 48 h posttreatment with OX compared to the untreated as identified using phosphoproteome profiler antibody array analysis (Figures 4A,B). While in HT 29 cells, the mutant p53 was found to be constitutively phosphorylated at serine residues: Ser15, Ser46, and Ser392 with no differential phosphorylation levels being observed between untreated vs. OX treated cells as evident from the phosphoproteome profiler arrays (Figures 4C,D).

Previous studies have evidenced that in wild-type p53, phosphorylation of N-terminus serine residues: Ser15, 20, and 37 occurs in response to DNA-damaging stresses such as gamma/UV irradiation and metal cadmium (57–59). This phosphorylation in turn is believed to loosen the affinity of p53-Hdm2 interaction leading to stabilization and activation of p53. The phosphorylation of Ser46 of wild-type p53 has been reported to be critical for the induction of apoptosis via transactivation of pro-apoptotic protein p53AIp1 (p53-regulated apoptosis-inducing protein 1) (60). An in vivo study by Feng et al. (61) has demonstrated the physiological role of Ser46 phosphorylation in regulating apoptosis in mice. The study showed that Ser46-p53 mutant mice had reduced transcription rates of certain pro-apoptotic genes resulting in compromised apoptosis compared to knock-in mice expressing wild-type human TP53 gene.

The serine residue Ser392 forms one of the most conserved phosphorylation sites in C-terminus of p53 protein and its phosphorylation is believed to facilitate the DNA binding capability of p53 in vitro (62). It has been reported that DNA-damaging signals including IR and UV radiation can induce phosphorylation on this site of wild type p53 leading to transactivation of its apoptotic function (63, 64). While, this phosphorylation event has been reported to promote the tumor suppressive function of p53 (65), lack of phosphorylation at Ser389 (serine residue in mice equivalent to human Ser392) of p53 protein has led to the development of bladder tumor in mice (66).

Our study indicates that chemosensitivity of colon cancer cells to OX is dependent on their p53 status and the phosphorylation of N-terminus: Ser15, Ser46 as well as the C-terminus: Ser392. In wild-type p53 this contributes to the transactivation of the apoptotic function of p53 making the cells sensitive to OX. This is evident from the sensitization of wild-type p53 bearing HCT 116 cells to OX being achieved via increased phosphorylation of p53 at the aforementioned 3 serine residues, leading to the activation of p53 and mediating apoptosis. The mutant p53 in HT 29 cells with its constitutive phosphorylation at Ser15, Ser46, and Ser392 and accumulation, with no transactivation function but with GOF enriched oncogenic properties could be attributed to the impaired sensitivity of these cells to OX.

We demonstrate for the first time that, endogenous phosphorylation of C-terminus serine residue Ser392 of mutant p53 protein could contribute to the impaired sensitivity of the mutant p53 bearing colon cancer HT 29 cells to OX. This dysregulated phosphorylation would have resulted in a non-activated dysfunctional p53 with reduced DNA-binding activity and hence been less responsive to the DNA-damaging drug-OX. This would fall well in agreement with certain studies showing that Ser392 phosphorylation regulated the oncogenic function of mutant p53 (67) as well as the Ser392 hyper-phosphorylation correlated to poor prognosis with tumor stage and tumor grade in p53-positive cancers (68, 69).

Autophagy an intracellular self-degradative process, forms a mechanism responsible for recycling cellular metabolic substances to maintain homeostasis, while to ascertain its role as antitumor, or tumorigenic is dependent on cellular context (70, 71). It is known that in cancerous cells, autophagy forms an alternative cellular source of energy under metabolic stress and thereby adapts to a smarter response with cancer therapies (30, 72). Certain studies suggests that autophagy can act as a tumor-suppressive factor in case of solid tumors, for instance in thyroid cancer (73), and also can increase cancer cell chemo-sensitivity even in hematological malignancy such as lymphoma (74). However, in disparity, a number of study findings support the role of autophagy in facilitating tumorigenesis by attenuating drug induced cytotoxicity augmenting chemoresistance in solid tumors (75–77) as well as in hematological malignancies (70, 78). In general, autophagy is doomed to be an adaptive response to stressful conditions such as starvation or hypoxia in tumors wherein it forms a pro-survival mechanism promoting tumor cell survival (79). Whilst under genotoxic stress induced by chemotherapeutic drugs, autophagy forms a cytoprotective mechanism in cancer cells favoring chemoresistance (80).

In our study, the conversion of LC3-I protein to LC3-II: a hallmark of autophagy induction was observed with OX treatment in HT 29 (mutant p53) colon cancer cells (Figures 5B,C). A time-dependent accumulation of LC3-II protein with OX treatment was observed in HT 29 cells while, no such accumulation of LC3-II protein was found in HCT 116 cells indicating no induction of autophagy in response to OX treatment (Figure 5A).

Many studies have described the role of autophagy in tumor chemoresistance as context-dependent (42) suggesting that wild-type p53 induces autophagy (81), whereas mutant p53 represses (82, 83) it. However, studies depicting autophagy as a cytoprotective mechanism of chemoresistance in colon cancer cells being linked their p53 status are limited or nearly nil. In this context, our findings demonstrate for the first time, “autophagy” as a possible cytoprotective mechanism for impaired sensitivity of mutant p53 harboring HT 29 colon cancer cells to OX.

There is accumulating evidence that exists on the drug-resistant property exhibited by cancerous cells harboring mutant p53 (15, 84, 85). Studies also indicate that the oncogenic properties such as tumorigenesis, anti-apoptotic as well as drug resistance featured by certain cancer cell types can be attributed to the presence of CSC as well as mutant p53 protein accumulation (34, 86). CSCs are known to be potentially resistant to chemotherapy and hence, pose the risk of disease recurrence as the conventional chemotherapies mostly fails to eliminate the CSC (33, 87). Many previous reports have also suggested the possible link between CSC and p53 status, with CSC markers such as CD133 and CD44 being repressed in wild-type p53 while, their expression contributing to the CSC characteristics in various mutant p53 cancer cell types (88, 89).

In the current study, we found that cancer specific stemness markers: CD133 and CD44 in wild-type p53-HCT 116 cells were significantly down-regulated in their expression with OX treatment (Figures 6A,B). While in mutant p53 HT 29 cells, OX treatment did not modulate the expression of stemness marker proteins (Figures 6C,D). This perhaps indicates the existence of a subpopulation of cancer cells with self-renewal capability and these CSCs along with the accumulation of mutant p53 protein (as shown before) could be attributed to the impaired sensitivity to OX displayed by HT 29 cells.

In line with the emerging evidences highlighting the association of cancer stemness and chemotherapy resistance (90–93), various studies have also underpinned the link between Wnt/β-catenin signaling pathway, CSC and chemoresistance. For instance, Urushibara et al. (35) tried to inhibit the expression of CSC proteins using a Wnt/β-catenin signaling inhibitor “IC-2” and found enhanced cytotoxicity to the drug 5-FU (fluorouracil) by colon cancer cells. Another study by Chen et al. (36) has shown that suppression of Axin2 (a negative feedback regulator of Wnt/β-catenin signaling) by micro RNA: miR-103/107 enhanced chemoresistance of colon cancer cells via inducing stemness.

A significant accumulation of β-catenin in response to OX treatment in mutant p53-HT 29 cells revealed the involvement of Wnt/β-catenin pathway in attributing HT 29 cells impaired sensitivity to OX. In contrast, this accumulation was not observed in wild-type p53-HCT 116 cells which were sensitive to OX.

Though studies have reported the role of high Wnt activity as well as the involvement of Wnt/β-catenin signaling in regulating CSC properties (94–96), to our knowledge our study forms the first ever report demonstrating cancer stemness and Wnt/β-catenin signaling as factors contributing to the impaired sensitivity of the mutant p53 harboring colon cancer HT 29 cells to OX.

In conclusion, our study demonstrates for the first time that dysregulated phosphorylation of the N-terminus and C-terminus serine residues of mutant p53 protein critically contributes to impaired sensitivity of mutant p53-harboring HT 29 colon cancer cells to OX (Figure 7). Our findings also demonstrate for the first time, autophagy as a cytoprotective mechanism for impaired sensitivity towards OX in mutant p53-harboring colon cancer cells. Our study also suggests cancer stemness and Wnt/β-catenin signaling as factors contributing to the impaired sensitivity of mutant p53-harboring colon cancer cells to OX which may have therapeutic value for treating colon cancers with OX in future.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

RK conceptualized and designed the study. SU and MS mentored the study. LT, JH, and FS performed the experiments. RK, LT, and KS analyzed the data. LT, JH, and RK made the figures. RK created the illustration. LT, JH, and RK wrote the manuscript with comments and input from all co-authors. MK and JB provided linguistic support. All authors have read and approved the final version of 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.