Introduction

The genetic alterations involved in prostate cancer (PC) are extremely complex, in fact, several studies have revealed the existence of different genetic subtypes of PC, particularly in localized forms considering aberrations on different oncogenic pathways [1, 2]. On the other hand, the evolution of the disease is marked by great heterogeneity, in fact, PC even diagnosed at the localized stage can lead to different clinical outcomes, considering in particular that relapses occur in 30% of men even with treatments like prostatectomy and/or radiotherapy [3]. The variability in disease progression suggests not only molecular changes impacting cancer cells but also the intricate interactions between cancer cells and their microenvironment [4]. Some studies have suggested a positive correlation between periprostatic adipose tissue (PPAT) thickness and the aggressiveness of PC [5, 6]. Another study investigated factors affecting the response to docetaxel in metastatic PC, it was found that adipocytes surrounding the tumor promote docetaxel resistance through a mechanism dependent on β-tubulin isoform 2B (TUBB2B) [7].

Considering these facts, conducting a detailed study of various clinical cases to identify genetic markers that characterize different disease stages becomes essential. These biomarkers will play a crucial role alongside established parameters [prostate-specific antigen (PSA), Gleason score] in further refining the stratification of PC. Such refinement is critical for selecting the most appropriate therapeutic strategies.

At the molecular level, numerous studies investigating signaling pathways altered in PC have revealed that genes involved in the DNA damage repair pathway are often affected by both somatic and germline mutations [8]. Among these are genes breast cancer gene 1 (BRCA1) and BRCA2, which act as tumor suppressors involved in the homologous recombination repair (HRR) process of double-strand breaks [9]. Indeed, BRCA germline mutations in PC patients have been associated with more aggressive disease and poorer clinical outcomes. Cancer-specific survival (CSS) and metastasis-free survival (MFS) at 5 years are also negatively affected, with a CSS of 8.6 years in BRCA mutation carriers compared to 15.7 years in non-carriers, and an MFS of 77% versus 93% [10]. Based on the finding that alterations in DNA repair pathways correlate with metastasis occurrence and poor prognosis, therapeutics such as PARP inhibitor [poly(adenosine diphosphate-ribose) polymerase inhibitors] targeting BRCA1 and BRCA2 genes have been developed to treat patients with metastatic castration-resistant PC (mCRPC) [10, 11]. However, since PARP inhibitor therapies do not work in all patients with DNA repair gene alterations, further investigation of other genes in this pathway is warranted to better understand the dysfunctions that may be associated with the occurrence of PC. One of the most important gene in DNA repair process is tumor protein 53 (TP53), often referred to as the guardian of the genome due to its role in defending against various external and internal stressors. It is a key gene regulating the transcription of several genes involved in important biological processes [12], which explains that gene dysfunction induced by various genetic alterations is found in more than 50% of human primary tumors [13]. The germline TP53 mutation is associated with Li-Fraumeni syndrome (LFS), an inherited autosomal dominant disorder, associated with a predisposition to several cancers like breast carcinomas, sarcomas, brain tumors, and adrenal cortical carcinomas [14].

Although mutations in the TP53 gene have traditionally been considered a late event in PC [15, 16], recent studies have reported significant frequencies of TP53 mutations in primary, and, particularly, in castration-naive metastatic PC [17, 18]. A study conducted on 175 primary PC patients, who later developed mCRPC, found that TP53 gene aberrations were the most frequent genetic events, detected in 25% of the studied population (primary PC) [19]. Another study on mutant TP53 clones from primary prostate tumors, suggested that these mutations could be the cause of metastatic spread with a lag period of 17 years [20]. Given these data, there is an emerging hypothesis that mutations in the TP53 gene may be an early event in certain forms of PC and may predispose to a lethal disease outcome.

Hence, the aim of this study was to evaluate the mutational profile of exon 5 of the TP53 gene, considering that mutation of this gene is mainly located in exons 5 and 8 [21]. The study group comprises 48 patients at various stages of PC to assess the mutational status of the TP53 gene in both advanced and localized forms.

Materials and methods

Results

The clinical characteristics of study subjects are summarized in Table 1. Individuals aged over 60 represent more than 81% of the population studied. Seventeen individuals (35%) have a Gleason score > 7, fifteen subjects (31.25%) have a score of 7, and thirteen patients (27%) have a Gleason score < 7. About the PSA value, 37.5% of the population have a PSA level ranging between 10 and 20 ng/mL. The analysis based on the clinical stage showed that fifty percent of patients are at stage T2, 29% of the population is at stage T1, and only 8% and 6.25% are at stage T3 and T4, respectively. The analysis of other collected parameters related to certain risk factors such as smoking and alcohol has shown that 46% of patients smoke, and 48% are non-smoker. Regarding alcohol consumption, 29% of patients drink alcohol, while 65% of the population doesn’t drink.

The mutations were arranged according to genotype, genome location, the effects on the amino acids, and frequency rate. Our research showed that some patients harbor known TP53 gene variants, but also identified new mutations of this gene. Among the population studied, 38 (79%) patients carry TP53 mutations. A total of 137 germline mutations of the TP53 gene were identified, including 115 new mutations. All the mutation carriers harbored at less than three mutations and some subjects exhibit a significant frequency of mutation such as case 43 with seventeen mutations. The effects of the identified alterations are of different types, but with a predominance of the frameshift type and missense mutations. Mutations were found in patients at different stages of the disease, but predominantly in stage T2 (Figure 1).

Display full size

Figure 1. 

TP53 mutations profile by tumor stage in Moroccan prostate cancer patients. The presentation illustrates the distribution of the 137 mutations found in 38 patients according to the disease stage. The graph takes into account recurring mutations carried by more than one patient. The variation in circle colors represents the frequency of mutations at each cancer stage; the more common a type of mutation is, the darker the circle

Among the new germline mutations detected in the TP53 gene, we noted a significant frequency of frameshift mutation. Sixty-one new frameshift mutations were identified in 36 patients (95% of mutation carriers), the mutation c.392delA was recorded in fifteen cases (31%); the mutations c.383delC and c.432delG observed at a frequency of 12.5% and 10% respectively.

Eighteen missense new mutations of the TP53 gene were detected in 68% of the mutation carriers group, the most frequent was the variant c.502C>A (p.His168Asn) identified in eleven patients (23%). The mutation c.382C>T resulting in the substitution of proline at position 128 by serine was observed in six cases (12.5%). The two other mutations: c.398T>G (p.Met133Arg) and c.413C>G (p.Ala138Gly) were observed in four cases (8%), whilst the 12 remaining variants were observed only once. Regarding the synonymous mutations, seven variants were recorded but the mutant c.441T>G (p.Val147=) was the only recurrent variant, identified in three cases (6%). One nonsense mutation was identified in one patient and resulted in a stop codon at position 126 (tyrosine). All codons affected by these alterations are part of the DNA binding domain of the protein. The most common alterations revealed, genome locations, and phenotypes are given in Table 2.

Among the 137 detected mutations, twenty-two (16%) were identified in 34 cases (71%) and correspond to known variants (Table 3). The most frequent mutation type was the missense variants, observed in eighteen patients. Among ten missense changes, the variant rs1057519977 (p.Cys141Trp) was the most frequent and has been recorded in nine patients. The two missense variants rs1131691037 (p.Asn131Ile) and rs138729528 (p.Arg175Gly) were recorded at a frequency of 10% and the variants rs1555526335 (p.Tyr126Cys) and rs758781593 (p.Gln136His) at a frequency of 8%. Five other missense variants were also identified but only once: rs1555526226 (p.Val147Ile), rs28934875 (p.Ala138Pro), rs587782596 (p.Arg181Ser), rs730881999 (p.Ser127Cys), rs786203071 (p.Gln144Pro). One mutation resulting in the insertion of Proline (rs786202525) and another one resulting in a truncated protein in position 144 (rs757274881) were also identified once.

The clinicopathological characteristics were compared between the mutation carrier and non-mutation carrier groups (Table 4). Fifteen patients of the mutation carriers have a PSA ranging between 10 and 20 ng/mL, while only three patients of non-carriers have this value (P = 0.614). Concerning the Gleason score, 13 patients among mutation carriers have a score greater than 7, while four among non-carriers have this score (P = 0.815). Regarding the clinical stage, 20 carriers of mutations are at the T2 stage, while four out of ten non-carriers are in this stage (P = 0.364). The smoking individuals among the mutation carriers were fifteen, while seven individuals belonged to the non-carriers group (P = 0.053). Regarding alcohol consumption, twelve of the mutation carriers are an alcohol consumer and six are from the non-carriers group (P = 0.010).

Protein sequence and mutational positions were submitted to SIFT Blink server, the server predicts if the amino acid substitution alters the protein function. The intolerant range ≤ 0.05 has been used as a limit of variants classification. More than this limit indicates as damaging to the protein function. 13 (72%) out of 18 missense variants were predicted to affect the protein function. Eight showed the highest score of 0.00. Five of the missense variants tested (28%) were predicted as tolerated. These results were combined with scores given by the Polyphen tool. The scores of this last were comprised from zero to one. Score (1.00) means that substitution is probably damaging, low scores were considered a benign effect. Out of 18 missense variants, 13 (72%) were predicted as probably damaging with values range: (0.965–1.00), two missense variants possibly damaged the protein with values of 0.466 and 0.768, the rest were benign. The results of missense variants effect prediction are given in Table 5.

Discussion

Despite the great diversity of genes involved in oncogenesis, the transcription factor TP53 remains a key tumor suppressor and a master regulator of various signaling pathways involved in this process [27]. It is considered a powerful tumor suppressor because of its various roles including the ability to induce cell cycle arrest, DNA repair, senescence, and apoptosis, to name only a few. Furthermore, TP53 mutations were reported to occur in almost every type of cancer at rates varying between 10% (e.g., in hematopoietic malignancies) [28] and close to 100% (e.g., in high-grade serous carcinoma of the ovary) [29].

Germline mutation of TP53 is also associated with cancer predisposition emphasized notably by LFS characterized by a wide spectrum of tumor types occurring over a wide age range, starting at a young age [30].

Here, we identified 22 variants in the DNA-binding domain of the TP53 gene, in Moroccan PC patients, which were previously reported. The mutation c.423C>G (p.Cys141Trp) is one of the highly frequent variants recorded in 9 patients (19%). This mutant was reported in breast cancer and LFS [31]. The mutant c.392A>T detected in 5 patients, resulted in deleterious substitution of asparginine in position 131 by isoleucine, this variant was identified first in a family with a history of multiple malignancies [32]. Another missense variant identified at the same frequency is c.523C>G mutant and affects the arginine in position 175, considered one of the hot spot residues. This alteration was identified in a French family meeting LFS criteria [33], besides, another alteration at this same amino acid position (p.R175H) is a well-characterized TP53 hotspot mutation [34]. The nonsense variant c.430C>T identified in one case, generates a premature translational stop signal (p.Gln144*) in the TP53 gene. It is expected to result in an absent or disrupted protein product. This mutant has been observed in individuals with clinical features of LFS [35]. Within the known mutations identified the intronic variant c.559+31G>A was the most frequently recorded in 26 patients (54%).

Considering previous reports on TP53 gene, the majority of mutations reported are missense mutations [36]. Indeed, among the mutants identified in our study, missense variants were very frequent and have been recorded in 27 cases. Furthermore, 79% of the population studied was found to carry germline mutation all localized at the DNA-Bainding site, according to previous studies this site hosts 90% of the mutations between residues 110 and 290, and this could therefore lead to the inhibition of its transcriptional activity [37]. Although few data are available on the association of PC with germline TP53 mutations [38, 39], but a large study by Maxwell et al. [40] identified germline TP53 mutations in 38 PC patients (0.55% prevalence) with a relative risk of having germline TP53 significantly elevated at 9.1 (95% CI 6.2–14, P < 0.0001) compared with the non-cancer population database. In the same study, the incidence of PC in LFS was assessed, and PC was identified in 31 cases of 163 (19%).

Our study identified a high frequency of new germlines mutations affecting the TP53 gene which do not appear to be associated with LFS, a similar finding was reported by the Maxwell et al. study [40], which revealed that over half of the germline TP53 variants identified in PC patients are considered attenuated or hypomorphic variants and not typically associated with classic LFS. The actual effects of these variants on the TP53 protein are unknown. However, the two prediction tools SIFT and PolyPhen-2, employed to assess the impact of missense variants, revealed that more than half of these mutants are predicted to alter the function of the TP53 protein. Thus, these mutations lead to defects in the protein pathway, affecting its role as a tumor suppressor gene. The variants predicted as the most deleterious by both SIFT and Polyphen.2 tools namely: p.Ala138Gly, p.Tyr126Arg, p.Val143Ala, p.Val143Glu, p.Gln144Pro, p.Gln144His, p.Pro152Thr, p.Gly154Asp, p.Ala159Asp, p.Lys164Gln, p.His168Asn, p.Asp184His. The codon at position 143 could be considered a critical residue. A study by Dridi et al. [41] aimed to determine the dominant-negative effect of different TP53 mutations in the near-diploid LoVo colon carcinoma cell line in response to mitotic spindle inhibitors, demonstrated that the TP53-175H and p53-143A mutant clones re-enter S phase with no apparent arrest unlike the wild type showing a tetraploid G1 cell arrest. Regarding the functional significance of TP53 mutations, missense mutation can have dominant negative effects on the transactivation of other genes containing p53-specific responsive elements. According to Forrester et al. [42] generally minimal dominant negative effects can be attributed only to codons 143ala-, 175his-, 248trp-, 249ser-, 273his-mutations in PC cell line PC-3, lacking one base pair in codon 138.

Despite the absence of a correlation between the clinicopathological features and mutational status, we noticed a significant frequency of mutations in patients at localized stages of PC T1 (29%) and T2 (50%) (Figure 1). Although alterations of this gene have been mainly associated with advanced stages of PC and constitute a late event in carcinogenesis [43, 44], substantial aberrations in TP53 have previously been reported not only in advanced PC but also in primary PC and were associated with poor patient outcomes [19, 45–47]. Taken together, our results suggested that some patients may harbor germline genetic alterations in the TP53 gene, which are likely involved in the initiation or progression of the disease. Indeed, unlike somatic mutations, which do not have a hereditary character, germline mutations involve gametes by definition and are therefore passed down to the offspring [48]. Hence, the presence of these inherited alterations could increase the risk of developing hereditary cancer.

Germline mutations affecting other genes including the BRCA2 and BRCA1 genes were also identified by the Cancer Genome Atlas Research Network in 333 patients with primary PC, with respective rates of 13% and 0.3% [45]. Recently, another study involving a population of 620 PC patients, reported prevalence rates of 24.3% and 6.4% for BRCA1 and BRCA2 respectively [49]. Indeed, alterations in BRCA1/2 genes are included in gene panels used for patient stratification in the context of PARP inhibitors, which currently do not include TP53 [50].

In addition to genetic alterations affecting crucial genes like tumor suppressors TP53, BRCA1, and BRCA2 in the DNA damage repair pathway, as well as various other genes such as PTEN and PIK3ca in the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/mTOR pathway [51], various molecular and biological aspects must be explored together, to better elucidate the onset and progression of the disease. Indeed, variations in cancer outcomes can be observed within the same phenotype, and the disease frequently progresses to severe forms, such as CRPC and mCRPC [48]. Recent research highlights the significant roles of circular RNAs (CircRNAs) in chemotherapy resistance in various cancers [52]. CircRNAs are known to be dysregulated in various malignancies, including PC [53, 54]. A recent study investigating the role of CircRNAs in CRPC identified four CircRNAs that are upregulated during the progression to castration resistance in patient-derived xenografts (PDXs) [55]. These CircRNAs can also be detected in the plasma of PC patients, suggesting their potential utility as biomarkers for CRPC [56]. Another study by Shen et al. [57] has revealed that CircRNAs are involved in determining the sensitivity of PC to docetaxel.

Another aspect that has recently been explored is the role of the tumor microenvironment (TME) in the disease. The TME, encompasses a diverse array of cell types, including cancer-associated fibroblasts (CAFs), immune cells, endothelial cells, and stromal cells. Additionally, non-cellular components, such as extracellular matrix proteins (ECM) related proteins, cytokines, chemokines, and extracellular vesicles (EVs) are integral [58]. These elements interact via a complex network and mounting evidence indicates that this crosstalk plays a crucial role in the onset and progression of CRPC [59].

Cross-referencing data relating to all characterized factors in PC, such as genetic alterations, interactions between different signaling pathways, other biomarkers like CircRNAs, and elucidating the role played by the TME, will help better characterize various pathological profiles of PC. Consequently, this will enable personalized management through targeted therapies.