PTEN, PTENP1, microRNAs, and ceRNA Networks: Precision Targeting in Cancer Therapeutics

Simple Summary

The PTEN gene is an important and well-characterised tumour suppressor, known to be altered in many cancer types. Interestingly, the effect of the loss or mutation of PTEN is not dichotomous, and small changes in PTEN cellular levels can promote cancer development. Less well-known mechanisms regulating PTEN, with emerging importance, include the PTEN–miRNA–PTENP1 axis, which has been shown to play a critical role in the fine tuning of PTEN cellular levels. This mechanism, working at the post-transcriptional level, involves the interplay and competition between the PTEN transcript, its pseudogene long non-coding RNA transcripts, PTENP1, and microRNAs. Our growing knowledge of this mechanism has opened avenues for the development of strategies to alter the cellular levels of PTEN, miRNAs, and PTENP1 as a new frontier in cancer therapy.

Abstract

The phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is a well characterised tumour suppressor, playing a critical role in the maintenance of fundamental cellular processes including cell proliferation, migration, metabolism, and survival. Subtle decreases in cellular levels of PTEN result in the development and progression of cancer, hence there is tight regulation of the expression, activity, and cellular half-life of PTEN at the transcriptional, post-transcriptional, and post-translational levels. PTENP1, the processed pseudogene of PTEN, is an important transcriptional and post-transcriptional regulator of PTEN. PTENP1 expression produces sense and antisense transcripts modulating PTEN expression, in conjunction with miRNAs. Due to the high sequence similarity between PTEN and the PTENP1 sense transcript, the transcripts possess common miRNA binding sites with the potential for PTENP1 to compete for the binding, or ‘sponging’, of miRNAs that would otherwise target the PTEN transcript. PTENP1 therefore acts as a competitive endogenous RNA (ceRNA), competing with PTEN for the binding of specific miRNAs to alter the abundance of PTEN. Transcription from the antisense strand produces two functionally independent isoforms (PTENP1-AS-α and PTENP1-AS-β), which can regulate PTEN transcription. In this review, we provide an overview of the post-transcriptional regulation of PTEN through interaction with its pseudogene, the cellular miRNA milieu and operation of the ceRNA network. Furthermore, its importance in maintaining cellular integrity and how disruption of this PTEN–miRNA–PTENP1 axis may lead to cancer but also provide novel therapeutic opportunities, is discussed. Precision targeting of PTENP1-miRNA mediated regulation of PTEN may present as a viable alternative therapy.

1. Introduction

The phosphatase and tensin homolog deleted on chromosome 10 (PTEN), also known as mutated in multiple advanced cancers 1 (MMAC1) and TGFß-regulated and epithelial cell-enriched phosphatase 1 (TEP-1) [1,2,3], is a well-known tumour suppressor gene located on chromosome 10q23.31 [2]. The gene and its protein product play a vital role in cell proliferation, migration, and survival [2,4,5,6,7]. As an antagonist of phosphoinositide 3-kinase (PI3K), PTEN dephosphorylates its substrate PIP₃ to PIP₂, thereby negatively regulating the pro-proliferative and anti-apoptotic PI3K/Akt pathway to maintain cellular homeostasis [8,9]. The regulation of PTEN cellular levels is critical in the negative modulation of tumorigenesis with disruption of PTEN signalling leading to significant cellular changes. Interestingly, subtle decreases in cellular levels of PTEN can result in malignancy and tight regulation of the expression, function, and cellular half-life of PTEN, at the transcriptional, post-transcriptional, and post-translational levels is necessary in the prevention of carcinogenesis [10,11]. PTEN is frequently mutated and/or deleted in the inherited PTEN hamartoma tumour syndromes (PHTS) [12,13] and multiple sporadic human malignancies, including those from the brain, breast, prostate [1], endometrium [14], skin (melanoma) [15], and colon [6].

Less well-known regulatory mechanisms of PTEN with emerging importance include the PTEN–miRNA–PTENP1 axis, which has been shown to play a critical role in the fine tuning of PTEN regulation and cellular integrity. PTENP1 is a processed pseudogene of PTEN termed the phosphatase and tensin homolog pseudogene 1 (PTENp1, PTENpg1, PTENP1, PTH2, and ψPTEN), which is located on 9p13 (Gene ID: 101243555) [16,17,18]. This pseudogene is transcribed to produce sense and antisense transcripts with the sense transcript showing high sequence similarity with the PTEN transcript; however, unlike PTEN, this transcript is not translated to produce a protein [19]. Although PTENP1 protein is undetected in cells, when transcribed in vitro as a fusion protein, the product is viable and has comparable phosphatase activity to the wild-type PTEN [19]. The sense and antisense long non-coding RNAs (lncRNA) produced from PTENP1 are important in the modulation of PTEN expression at the transcriptional and post-transcriptional levels, respectively. The PTENP1 sense transcript (PTENP1-S), acting as a competitive endogenous RNA (ceRNA) of PTEN, leads to alterations in PTEN cellular abundance. The characteristics of this PTEN pseudogene lncRNA include similarities in their microRNA (miRNA) binding sites, and as such, PTENP1 can act as a decoy or ‘sponge’, competing for miRNAs that target PTEN. Disruption of the PTEN–miRNA–PTENP1 axis and ceRNA networks in carcinogenic progression is contemporary and is an exciting area in the discovery of regulatory mechanisms that are altered in cancer. In addition to its regulation of PTEN expression, PTENP1 is able to act as a tumour suppressor independent of its PTEN regulatory function as described in a recent review of the role of PTENP1 in human disorders with a focus on its tumour suppressor functionality [20].

In this review, we outline the importance of PTEN regulation in cancer development/progression through the well-known mechanisms of mutation, deletion, and alterations of PTEN structure and function, with a major focus on the role of the PTEN–miRNA–PTENP1 axis. The mechanisms of post-transcriptional regulation of PTEN, through interaction with its processed pseudogene (PTENP1) transcript (expressed as a lncRNA) and the cellular miRNA milieu, in the context of a cellular ceRNA network is discussed. Knowledge of the working of this regulatory mechanism will allow the identification of potential future novel therapeutic options. Precision targeting of the PTEN–miRNA–PTENP1 axis is important for the regulation of PTEN and may present as a viable alternative therapy to increase endogenous wild-type PTEN in tumours shown to have reduced PTEN levels.

2. PTEN and Cancer: From Mutations to a Continuum Model of Tumorigenesis

Germline and somatic mutation of PTEN is known to contribute to many cancers, highlighting the importance of this tumour suppressor in cancer initiation, progression, and metastasis. Germline mutations of PTEN are the cause of four autosomal dominant inherited syndromes: Cowden syndrome (CS) [21], Bannayan–Riley–Ruvalcaba syndrome (BRRS) [22,23], Proteus syndrome (PS), and PS-like syndrome [24], which share common features, including the development of multiple benign hamartomas, and are all classified under the umbrella term of the PTEN hamartoma tumour syndromes (PTHSs) [12,13]. PTHS patients have an increased lifetime risk of developing specific malignancies, mainly breast cancer (approximately 80%) [12,13], thyroid cancer (approximately 30%) [12,13], renal cell carcinoma (approximately 34%) [13], endometrial cancer (approximately 28%) [13], and colorectal cancers (approximately 9%) [13]. In individual PHTS patients exhibiting clinical phenotypes, PTEN germline mutations are reported in 25-85% of CS patients [21,25,26], 60% of BRRS [21,22,25,27], up to 20% of PS [28], and between 50 and 67% of PS-like syndrome patients [24]. Interestingly, germline PTEN mutations are also associated with a subset of patients with autistic behaviour and extreme macrocephaly [29].

Somatic mutations of PTEN are frequently associated with tumorigenesis with somatic alterations of PTEN being described in over 50% of cancers of various types [30]. PTEN somatic mutations are most prevalent in prostate cancer [31], endometrial cancer [32], melanoma [33,34], non-small-cell lung cancer [35,36], kidney [37], breast cancer [38], and glioblastoma [39]. PTEN somatic alterations include the complete loss or inactivation of one allele (functional haploinsufficiency) due to point mutations and/or deletions and/or epigenetic silencing through hypermethylation of the PTEN promoter, which is characteristic of some advanced and metastatic cancers [1,4]. Deletion of both alleles of PTEN occurs at a lower incidence but is seen mostly in metastatic breast cancer, melanomas, and glioblastomas [1,4,40]. In contrast, a recent study showed that patients with high PTEN expression levels in endometrial cancer had low tumour malignancy, decreased cancer cell proliferation and had a better prognosis [41]. There are different mechanisms of PTEN loss or inactivation, with some being more prevalent in specific tumour types (

Table 1. Mechanism and frequency (%) of PTEN loss in various cancer types.

Cancer Type	Mutation		Deletion		Loss of Protein		Promoter Methylation
Glioblastoma	30%	[2,3,42,44,45,46]	78%	[44,45,46,47]	65%	[48]	6%	[49]
Breast	3%	[42,50,51]	27%	[38,52]	40%	[42]	35%	[53,54]
Prostate	13%	[55,56,57,58]	51%	[56,57,58]	54%	[55,56,57,58]	<5%	[42,59,60,61]
Colorectal	7%	[6,42,62,63,64,65,66]	8.7%	[42,62,63]	40%	[67]	17%	[68]
Lung	8%	[42]	34%	[42]	56%	[42]	38%	[69]
Endometrial	41%	[14,42,70]	48%	[14,42,70]	45%	[14,41]	19%	[42,71]
Ovarian	16%	[42,43,72,73,74,75,76]	48%	[42,43,72,73,74,75,76]	44%	[42,43,72,73,74,75,76]	10%	[42,77]

Note: Where multiple references are provided, the frequencies of mutation, deletion, and promoter methylation are an approximate average across the relevant publications.

) [30,42,43].

The effect of the loss or mutation of PTEN is not dichotomous, and subtle changes in PTEN cellular levels have been shown to lead to deleterious consequences relating to tumour incidence, penetrance, and aggressiveness in several epithelial cancers [11,78]. In the hypomorphic transgenic Pten mouse, it has been shown that in susceptible organs such as the prostate, PTEN protein expression levels need to reach dramatically low levels (reduced by 70% compared to normal levels) to initiate tumorigenesis, however, in the mammary glands, a more subtle reduction (reduced by 20% compared to normal levels) can initiate tumorigenesis [78]. Thus, PTEN does not follow the ‘two-hit’ paradigm or stepwise model of tumour suppressor gene function but rather presents a new continuum model of tumorigenesis whereby tumorigenesis occurs in an incremental dose-dependent manner [11,78]. This has been evidenced in gastric cancer, where PTEN expression was shown to gradually decrease with increasing gastric cancer progression [79].

PTEN Loss, Tumour Immune Evasion, and Therapy Resistance

There are several recent studies that have explored the relationship between PTEN loss and tumour immunity, showing PTEN loss contributes to alterations in the tumour microenvironment (TME) to produce an immunosuppressive niche. The evidence suggests that PI3K signalling may influence the composition and functionality of the TME, thereby modulating the immune response in cancer. Vidotto et al. (2023) analysed PTEN copy number in 9793 cases from 30 tumour types, derived from the Cancer Genome Atlas, and showed that reduced tumour PTEN expression occurs with hemizygous loss leading to tumour anti-cancer immune responses [80]. In another integrative analysis of TCGA samples, Lin et al. (2021) found that both PTEN loss and activation of the PI3K pathway were associated with reduced T-cell infiltration and an enhanced immunosuppressive status in multiple tumour types [81]. Overall, the effect of PTEN loss of function in the different cellular compartments swings the balance towards an immunosuppressive TME [82]. There was also a correlation between PTEN loss and poor response to immunotherapy [81]. Interestingly, PTEN loss has also been shown to promote resistance to therapy in breast cancer. Reducing PTEN levels in breast cancer cells conferred resistance to trastuzamab, and patients with PTEN-deficient breast cancers showed poorer therapeutic responses with this drug. Thus, PTEN deficiency has become a good predictor for trastuzumab resistance [83,84]. Reduced PTEN expression has been shown in vivo, in mouse models, to be due to specific miRNAs. An example being PTEN as a target of mi-R22 in breast and prostate cancers, which have been shown to have a strong influence in a cancer immune TME, playing a role in cancer initiation, progression, and metastasis [85]. Importantly, in vivo, knockdown of miR-22 appears to invoke tumour resistance in an immunocompetent environment [85]. These findings open new avenues for immuno-targeting, such as modulating miRNAs targeting PTEN, hence improving the efficacy of immunotherapy and overcoming therapy resistance.

3. Post-Transcriptional Regulation of PTEN by microRNAs and Pseudogene lncRNAs

PTEN is constitutively expressed in normal cells and, due to its critical role in several cellular processes, is closely regulated at the transcriptional, post-transcriptional. and post-translational levels to modulate expression, activity, and cellular half-life [86]. One of the most contemporary findings in PTEN regulation is the post-transcriptional regulation by its pseudogene long non-coding RNA (lncRNA) and microRNA (miRNA) [87,88,89,90]. This PTEN–miRNA–PTENP1 ceRNA network is discussed in detail in the sections below.

3.1. microRNAs Regulate PTEN Expression at the Post-Transcriptional Level

MicroRNAs are single-stranded RNAs comprised of 19-23 nucleotides. These small endogenous RNAs bind to complementary regions within the 3’ untranslated region (UTR) of their mRNA targets, whereby perfect complementarity leads to target degradation and imperfect complementarity leads to the suppression of translation [87,88,89,90] and an overall decrease (or increase in some cases) in target mRNA abundance [87,88]. Mature miRNAs known to repress PTEN include, but are not limited to, miR-17, miR-19, miR-21, miR-26, and miR-214 [17]. miRNAs can act as either tumour suppressor miRNAs or tumour promoting miRNAs (oncomiRs), depending on their modulating effect on the expression of their target gene(s). For example, miR-130 acts as a promoter of malignancy through the downregulation of PTEN expression in bladder cancer [91], invasive breast carcinoma [92], renal cell carcinoma [93], gastric cancer [94], gliomas [95], lung adenocarcinoma [96], and in colon adenocarcinoma [97]. Expression of miR-130 is lower in both non-small cell lung cancer (NSCLC) cell lines and tissues, and miR-130 overexpression results in cell growth inhibition and enhanced cell apoptosis, through increasing PTEN levels in NSCLC, thus miR-130 acts as a tumour suppressor in this context [98]. Classification of miRNAs into oncomiRs or tumour suppressor miRNAs is complicated, as several miRNAs have been shown to act as either tumour suppressors or oncomiRs in different tumour types [99,100,101].

3.2. PTENP1: A Processed Pseudogene of PTEN Produces Bidirectional Transcripts

PTENP1 is a processed pseudogene of PTEN, transcription of which produces unique, bidirectional, sense and antisense, transcripts [102]. Transcription from the sense strand produces a sense transcript (PTENP1-S), which is a pseudogene lncRNA with high sequence similarity to the PTEN transcript. There is 97.8% sequence similarity within the corresponding coding region (CDS) (with only 18 mismatches) of the two transcripts [17,19,103], and the 3’-UTR of the PTENP1-S transcript is approximately 1 kb shorter than that of the PTEN transcript. Overall, the 5’-UTR, the CDS, and the early sections of the 3’-UTR of PTEN and PTENP1–S share high sequence similarity (approximately 95%), however, the level of sequence similarity decreases drastically (approximately 50%) towards the later sections and end of the 3’-UTR [17].

Due to the high sequence similarity between the two transcripts, the PTENP1-S transcript shares common miRNA binding sites with the PTEN mRNA, particularly at the start of the 3’-UTR, resulting in the ability of PTENP1-S and PTEN to compete for the binding of common miRNAs. It is now well known that PTENP1-S acts as a miRNA sponge to protect PTEN from translational repression in a competitive manner, resulting in a positive impact on PTEN expression levels. This competition involves the participation of the PTEN and PTENP1-S transcripts, as well as the miRNAs targeting these transcripts, in a ceRNA network. Experimentally validated miRNAs that have been shown to participate in the PTEN and PTENP1-S ceRNA network in various cancer types are summarised in

Table 2. PTEN and PTENP1-targeting miRNAs identified and experimentally validated in various cancer types.

Disease	microRNAs (miR) *	References
Prostate cancer	miR-17-5p miR-19-3p miR-21-5p miR-26a-5p miR-214-3p	[17] [17] [17] [17] [17]
Hepatocellular carcinoma	miR-17-5p miR-19b-3p miR-20a-5p miR-193a-3p miR-21	[107] [107] [107] [108] [104]
Clear cell renal carcinoma	miR-21	[105]
Breast cancer	miR-19b miR-20a	[109,110] [111]
Bladder cancer	miR-17	[112]
Glioma	miR-10-5p	[113]
Endometrial cancer	miR-200c	[114]
Cervical cancer	miR-106b	[115]
Gastric cancer	miR-106b miR-93	[116] [116]
Oral squamous cell carcinomas	miR-21-5p	[106]

* All miRNAs target the 3’-UTR of PTEN and PTENP1.

. miR-21 is a common miRNA shown to target PTEN and PTENP1-S in prostate cancer [17], hepatocellular carcinoma [104], clear cell renal carcinoma [105], and oral squamous cell carcinoma [106]. Gaining an understanding of this ceRNA network presents the possibility for future manipulation of the network in the treatment of cancers to achieve positive therapeutic outcomes, and this is explored further below.

In addition to the sense transcript of PTENP1, transcription from the antisense strand produces two functionally independent isoforms, PTENP1-antisense-alpha (PTENP1-AS-α) and PTENP1-antisense-beta (PTENP1-AS-β) [102]. Both isoforms are expressed from convergent promoters and share a cis overlap with the PTENP1-S transcript and the 5’-UTR of PTEN [102,117,118]. The PTENP1-AS-α isoform shares high sequence similarity with the 5’-UTR of PTEN and is most abundant in the nucleus [102,117,118]. Thus, PTENP1-AS-α binds to the 5’-UTR of PTEN-associated transcripts, which allows for the localisation of PTENP1-AS-α to the PTEN promoter region and, in turn, recruits epigenetic modifiers, including chromatin remodelling proteins EZH2 and DNMT3A, which induce the H3K27me3 post-translational histone modification at the PTEN promoter, consequently leading to the negative transcriptional regulation of PTEN expression [117,118]. The PTENP1-AS-β transcript binds to the PTENP1-S transcript, which lacks a poly-A tail and provides stability to the PTENP1-S transcript through the formation of a PTENP1-S and PTENP1-AS-β complex that is exported into the cytoplasm, where PTENP1-S acts as a miRNA sponge to post-transcriptionally regulate PTEN through participation in the ceRNA network [102,117,118] (Figure 1).

4. PTEN, miRNA, PTENP1, and the Endogenous Competitive RNA (ceRNA) Binding Hypothesis

The endogenous competitive RNA (ceRNA) binding hypothesis, first postulated by Pandolfi and colleagues, states that endogenous RNAs, including mRNAs, transcribed pseudogenes, protein-coding genes, lncRNAs, and circular RNAs, compete to regulate each other through binding or sponging of shared miRNAs from the same cellular miRNA pool [119,120]. In this context, PTEN has been shown to be regulated by the PTENP1 sense transcript as PTENP1-S acts as a decoy to sequester miRNAs that would otherwise target and repress PTEN mRNA translation, thus maintaining or restoring PTEN protein levels [17]. This paradigm challenges previous ideas of sequence conservation working solely to influence the regulation of gene targets by ncRNAs and introduces an additional layer of complexity to the PTEN and PTENP1 ceRNA regulatory network.

Despite the myriad of papers confirming the involvement of PTEN and PTENP1 in a ceRNA network in cervical cancer [115,121], breast cancer [109,111], gastric cancer [116,122], oral squamous cell carcinoma [106], clear cell renal cell carcinoma [105], and in bladder cancer cells [112], there has been controversy from validation consortiums conducting replication studies [123,124]. Such controversy has, however, been recently cleared somewhat by evidence of PTEN and PTENP1 functioning as ceRNAs in studies using CRISPR knockdown and silencing of PTEN and PTENP1 in DU145 prostate cancer cells [125]. PTENP1 knockdown resulted in the repression of PTEN expression [125]. Additionally, the silencing of PTENP1-AS alpha and beta isoforms resulted in the downregulation of both PTENP1 and PTEN [125], confirming the regulation of PTEN and PTENP1 by the anti-sense isoforms [102]. Further to this, the knockdown of PTEN and PTENP1 resulted in the repression of the PTENP1-AS transcripts, while the opposite effect was seen when PTENP1 was upregulated [125]. This was further evidence for the involvement of PTEN and PTENP1 in a ceRNA network through the mutual regulation of each other’s expression levels [125].

Other ceRNAs Regulating PTEN Expression in Cancer

While PTEN is regulated by its first identified ceRNA, PTENP1, other ceRNAs in the form of lncRNAs and proteins that regulate PTEN, have since been identified [17]. A number of lncRNAs regulate PTEN through ceRNA mechanisms, which include, but are not limited to, FER1L4 competing with PTEN for miR-106a [126] in gastric cancer and miR-18a-5p in osteosarcoma, resulting in the suppression of PTEN [127]. The lncRNA MEG interacts with PTEN in a ceRNA manner to bind to miR-19a in glioma [128]. PTEN has been shown to be downregulated by the lncRNA HOTAIR through interaction with miR-29b in laryngeal squamous cell carcinoma [129]. Additional lncRNAs regulating PTEN through ceRNA mechanisms include Linc-USP16, which competes for miR-21 in hepatocellular carcinoma, with PTEN expression increasing upon its overexpression [130,131]. CASC2 is another proposed coregulator of PTEN through a ceRNA regulatory mechanism as PTEN and CAS2 both possess miR-21 binding sites [130,132,133]. Other lncRNAs reported to regulate PTEN through the ceRNA mechanism include LINC00702 [134], NEAT1 [135], RP11-79H23.3 [136], TP73-AS1 [137], and ORLNC1 [138]. Additionally, bioinformatics approaches have identified many other genes as PTEN ceRNA competitors, including TNRC6B, RB1, TP53, NRAS, KLF6, HIF1A, HIAT1, CTBP2, and TNKS2 [139,140,141]; however, these are yet to be experimentally validated.

5. Expression of PTEN and PTENP1 in Cancer

PTENP1 copy number loss and decreased PTENP1 expression have been reported in conjunction with PTEN loss and decreased expression in several cancer types as the result of either deletion or silencing due to promoter hypermethylation. The PTENP1 promoter has been shown to be hypermethylated in pancreatic adenocarcinoma, breast cancer, cervical cancer, ovarian cancer, and hepatocellular carcinoma cell lines [142], as well as in lymphoma [143], colorectal cancer [143], clear cell renal carcinoma cells [105,143], and NSCLC tissues [69]. In breast cancer cell lines expressing PTEN protein, PTENP1 was found to be methylated in MDA-MB-231 cells but unmethylated in MCF-7 cells [144]. PTENP1-S was methylated in endometrial cancer and hyperplasia but not in normal tissue [145]. Interestingly, a recent study showed that methylation of PTENP1 elevated PTENP1-S expression in normal endometrium tissue and endometrial hyperplasia from women aged 45 and over, and/or women approaching, or in, menopause [142,146] (

Table 3. PTENP1 methylation status in various cancer cell lines and cancer tissue types.

Cancer Tissue Type/Cancer Cells	PTENP1 Promoter Methylation Status	Reference(s)
Breast cancer	Hypermethylated	[142]
MDA-MB-231 breast cancer cells	Hypermethylated	[144]
MCF-7 breast cancer cells	Unmethylated	[144]
Cervical cancer	Hypermethylated	[142]
Ovarian cancer	Hypermethylated	[142]
Hepatocellular carcinoma cell lines	Hypermethylated	[142]
Lymphoma	Hypermethylated	[143]
Colorectal cancer	Hypermethylated	[143]
Clear cell renal carcinoma cells	Hypermethylated	[105,143]
Endometrial cancer and hyperplasia	Hypermethylated	[142,146]

).

PTENP1 is lost in several cancers and is known to be under selective pressure to undergo copy number loss in cancer. PTENP1 is lost in melanoma, breast cancer, sporadic colon cancers [17], and in endometrioid endometrial carcinoma [114,147]. Additionally, the low PTENP1 expression in endometrioid endometrial carcinoma and leukemia cells was shown to be associated with genomic copy number loss of PTENP1 [148]. In head and neck squamous cell carcinoma cell lines, complete and partial losses of PTENP1 are known to be frequent; however, the deletion of genomic PTEN is not common, further providing evidence for PTENP1 being under selective pressure to undergo copy number loss in cancer [149]. Furthermore, in studies showing lower levels of PTENP1 in cancer, it has been predicted to be a promising candidate as a future prognostic biomarker [150].

Generally, PTENP1-S expression levels are low compared to PTEN. However, the levels of PTENP1 vary depending on the cell lines and tissues being tested. Some studies have been completed, and the results of the expression of PTENP1 transcripts relative to PTEN, and between the PTENP1 transcripts in various cell lines and tissues, are presented in

Table 4. Expression levels of PTEN and PTENP1 in various cancer types.

PTEN:PTENP1 Relative Expression Ratio *	Cell Line or Tissue Type	References
↑PTEN:PTENP1-S	Osteosarcoma cell lines	[152]
	Melanoma cell lines	[151]
	Breast cancer cell lines and tissue samples	[102,109,110,111]
	Bladder cancer tissue	[112]
	Gastric cancer cells and tissues	[116]
	Oral squamous cell carcinoma cells	[106]
	Hepatocellular carcinoma cell lines and tissues	[107,108]
	Head and neck squamous cell carcinoma cells	[149]
	Glioma tissue	[112]
	Prostate cell lines	[17]
	Cervical cancer cells	[102,121]
	Endometrioid endometrial carcinoma cells	[114]
	Melanoma	[151]
↓PTEN:PTENP1-S	Some prostate cancer tissue samples, gastric cancer cell line, AGS, endometrioid endometrial carcinoma cell lines, RL-952, and JEC	[17] [116] [114]
↓PTENP1-S:PTENP1-AS	Kidney, HEK-293T, breast, MCF-7, cervix, HeLa, bone, U-2OS	[102] [102] [102] [102]

* Please note: The up and down arrows indicate either an increase or a decrease, respectivly, in the relative expression ratio of the relevant transcripts (PTEN, PTENP1-S and PTENP1-AS) as indicated.

. A quantitative study carried out in a limited number of cell lines showed that the PTENP1-AS transcript was more highly expressed compared to the PTENP1-S transcript [102]. Additionally, increased expression of the PTENP1-AS transcript resulted in lowered PTEN cellular levels due to transcriptional downregulation of PTEN by the PTENP1-AS transcript [102]. A recent study in melanoma cells showed that increased expression of PTENP1-AS resulted in the induction of BRAF inhibitor resistant cells and is likely to be due to the recruitment of epigenetic modifiers to the PTEN promoter region, resulting in reduced PTEN expression [151]. Furthermore, high levels of PTENP1-AS in stage III melanoma patient samples correlated with poor patient survival [151]. To date, all PTEN and PTENP1 expression studies have utilised relative quantitation methods (RT-qPCR) and true cellular levels are yet to be determined. In this context, the use of absolute quantitation methodologies would allow determination of the true cellular levels of PTEN, PTENP1-S, and PTENP1-AS transcripts in normal and cancer cells to help provide an understanding of the perturbations of these levels in cancer cells of various types. This information would be of great importance for our understanding of the contributions of these transcripts in cancer development and progression and would help form the basis of potential future transcript ratio altering therapies for cancer treatment.

5.1. PTENP1 also Functions Independently of the PTEN ceRNA Network

Interestingly, the PTENP1-S transcript is capable of functioning as a tumour suppressor independently of its PTEN regulatory effects and has been shown to have a growth suppressive role in numerous cancers including prostate [17], gastric carcinomas [116], clear cell renal carcinomas [105], gliomas [153], hepatocellular carcinoma [108], bladder [111,112,154], breast [109,110,155], cervical [115], melanoma [156], and colon cancer [17]. In these cancers, overexpression of PTENP1 led to a decrease in cell proliferation, suppression of cell migration and invasion, and induction of apoptosis through downregulation of the AKT and MAPK signalling pathways as well as downregulation of critical cell cycle proteins cyclin A2 and CDK2, in breast cancer [157]. This further validates the use of PTENP1 expression levels as a potential future candidate prognostic biomarker due to its tumour suppressor activity. A systematic review and meta-analysis of the data from the literature has recently been published revealing the prognostic value of PTENP1 expression in cancer. Dai et al. indicate that low expression of PTENP1 might predict poor prognosis for various carcinomas [150].

Additionally, PTENP1 has been shown to regulate PTEN outside the context of cancer. In smooth muscle cells, PTENP1 inhibits proliferation and enhances apoptosis [158]. In spinal cord injury, PTENP1 expression has also been shown to affect recovery by modulating the levels of miR-19b and miR-21 [159]. Interestingly, PTENP1 expression has implications for fertility as it was shown to regulate the human endometrial epithelial adhesive capacity in vitro by regulating miR-590-3p, and PTENP1 was shown to be one of a set of highly expressed lncRNAs in human endometrial epithelial cells subjected to blastocyst conditioned medium [160].

5.2. Evolution of PTENP1 and cross Regulation of PTENP1 by PTEN

The importance of this functional pseudogene and its regulation of PTEN is shown in its evolutionary history and conservation. Tang et al. (2016) have reported the identification of 37 PTEN pseudogenes (PTENPs) in 65 mammalian genomes, predominantly in primates and rodents [161]. While some PTENPs were shared among primates and rodents, others were shown to be species-specific. Of interest, these authors reported the presence of 17 copies of PTENPs in the naked mole rat, an anticancer model organism, with all genes sharing common miRNA binding sites with their PTEN counterpart [161]. While regulation of PTEN by PTENP1 is well established, a recent report has shown regulation of PTENP1 by PTEN in glioblastoma cells [162]. Using targeted methylation and demethylation of the PTENP1 CpG island, Kovalenko et al. showed that DNA methylation increases PTENP1-S expression in the presence of WT-PTEN protein but decreases PTENP1-S expression if there is an absence of PTEN protein. They further demonstrated that the PTEN protein binds to the promoter region of PTENP1 and inhibits PTENP1-S expression if its CpG island is demethylated. Thus, in glioblastoma cells, PTENP1 is a downstream target of PTEN.

6. Overexpression of PTENP1 or Its 3’-UTR: Prelude to Cancer Therapy?

Overexpression of the PTENP1 3’-UTR resulted in sequestration of miRNAs, showing PTENP1 to be an important gene in the regulation of PTEN. Overexpression of the 3’-UTR of PTENP1 in cell lines and in vivo studies has been shown to upregulate PTEN, thereby blocking the PI3K/Akt pathway and decreasing cell proliferation and metastasis, and increasing apoptosis in prostate (DU145) [17], renal (ACHN and SN12MP6) [105], liver (SK-Hep1 and SMMC-7721) [108], breast (MCF-7 and MDA-MB-231) [110,111,157], bladder (T24 and T5637) [112], gastric (MGC803 and BGC823) [116], oesophageal (Eca19), cervical (CasKi and HeLa) [115], and endometrial (RL-952, JEC and HEC-1B) [114] cancer cell lines. However, overexpression of PTENP1 did not successfully restore PTEN to normal levels in head and neck squamous cell carcinoma cell lines, HN13 and HN30 [149]. Additionally, overexpression of PTENP1 in oesophageal squamous cell carcinoma cells led to increased PTEN levels in Eca19 cells but not in TE-1 cells [163]. In the case of breast cancer, however, it was shown that the levels of PTEN are governed by the estrogen receptor (ER) status of the cells [144]. When PTENP1 was overexpressed in ER-positive breast cancer cells (MCF-7 and T-47D), PTEN expression decreased, and tumour growth was reported to be accelerated in MCF-7 cells [144]. Contrastingly, overexpression of PTENP1 in ER-negative breast cancer cells (MDA-MB-231 and C3HBA), led to increased PTEN expression and inhibition of tumour progression [109,110,111,144]. Similarly, another study in endometrioid carcinoma cells showed that an increase in miR-200c increased estrogen, resulting in an observed decrease in PTEN and PTENP1 expression in cells [114]. Therefore, estrogen plays an essential role in the occurrence of endometrioid carcinoma and affects the negative feedback loop of PTEN-miR-200c-PTENP1 [114]. All the above studies assess the implications of one or two miRNAs targeting PTEN and PTENP1, neglecting the multiplicity of miRNAs that are able to target and act as either tumour suppressors or oncomiRs, depending on the cell/tissue type in which they are present [101]. An examination of the miRNA expression profiles in a cell/tissue-specific manner will aid our understanding of the miRNAs that are positively and negatively regulated in cancer cells and potentially influencing the PTEN–PTENP1 ceRNA network.

7. Manipulating PTEN, PTENP1, and miRNA Levels as Potential Cancer Therapies

7.1. Increasing PTEN Levels Directly

The restoration of functional PTEN has been difficult; however, PTEN mRNA levels have been shown to be restored in PTEN null cells both in vitro in prostate cancer cells and in vivo in mouse models of prostate cancer through the use of nanoparticles delivering PTEN mRNA [164]. Additionally, the restoration of PTEN in PTEN null cells resulted in inhibition of the PI3K–Akt pathway and also increased apoptosis. This work represents a new approach to PI3K–Akt pathway inhibition through the restoration of PTEN mRNA. Recently, PTEN mRNA was delivered via nanoparticles to restore PTEN levels and enhance anti-tumour immunity in melanoma and prostate cancer mouse models [165]. Furthermore, prostate cancer progression has been shown to be inhibited in mice and in a subcutaneous tumour xenograft mouse model by the intraprostatic and intertumoral injection of recombinant adeno-associated virus 9 expressing PTEN [166].

7.2. Increasing PTEN Levels Indirectly via PTENP1 as an miRNA Competitor

Transfection of baculovirus packaged PTENP1 into hepatocellular carcinoma (HCC) cells resulted in increased PTENP1 levels in cells [107]. The injection of the PTENP1 expressing baculoviral vector into mice with HCC tumours also reduced tumour growth and cell proliferation, induced apoptosis and autophagy, and inhibited HCC cell properties [107]. Additionally, exosomal PTENP1 has been transferred from normal cells to bladder cancer cells, which resulted in a reduction in the progression of bladder cancer in vitro and in vivo [154]. PTENP1 packaged into exosomes has also been transferred to U87MG glioblastoma cells to sponge miR-10a-5p and stabilise PTEN levels in a competitive manner [113]. The success of altering PTEN levels through the delivery of PTEN mRNA or PTENP1 in both in vitro and in vivo studies is a promising start for these candidates for future gene therapies in clinical trials (Figure 2).

7.3. Altering Levels of miRNAs Targeting PTEN and PTENP1

miRNAs are powerful gene regulators and are emerging as promising therapeutics in various diseases [167]. Targeting PTEN and PTENP1 with microRNAs has tremendous potential in cancer therapeutics. Increases or decreases in specific microRNAs can lead to an increase or decrease in the levels of PTEN and PTENP1. For example, in the development of bladder cancer, miR-107 has been shown to be sponged by the LncRNA RP11-79h23.3 in a ceRNA manner to positively regulate PTEN expression [136]. In endometrial cancer development, lncRNA LA16C-313D11.11 acts as a ceRNA in the miR-205-5p–PTEN axis by inhibiting miR-205-5p and thereby increasing the expression of PTEN [168]. lncRNA GAS5 modulates miR-21 in NSCLC cells by increasing PTEN expression [169]. In NSCLC, lncRNA FER1L4 partakes in the positive regulation of PTEN by inhibiting cell proliferation and promoting apoptosis in NSCLC [170]. In breast cancer, PTEN is influenced positively by the increased expression of PTENP1 and decreasing miR-20a levels [111]. miR-200 is known to target PTEN, a key suppressor of the PI3K/AKT pathway [171,172]. miR-200a negatively targets PTEN in endometrial cancer [173] and oesophageal carcinoma [174]. Additionally, miR-200b targets PTEN in endometrial cancer [173], along with miR-200c [114], which similarly targets PTEN in head and neck carcinoma [175]. miR-429 targets PTEN in NSCLC [176]. Thus, in a cancer-dependent context, lncRNAs may be overexpressed to either inhibit or decrease the level of PTEN targeting microRNAs, thereby increasing PTEN expression and activity as a tumour suppressor. Furthermore, miRNAs can be depleted using miRNA inhibitors, or ‘sponges’ [177], in order to increase PTEN or PTENP1 levels in cells.

On the other hand, microRNAs are known to act as tumour suppressors, whereby they increase the levels of PTEN. A study in endometrioid endometrial carcinoma cells showed that an increase in miR-200c increased estrogen, resulting in a decrease in PTEN and PTENP1 expression in cells [114]. Estrogen plays an essential role in the occurrence of endometrioid carcinoma and affects the negative feedback loop of PTEN-miR-200c-PTENP1 [114]. Thus, the upregulation of certain miRNAs using miRNA mimics [177] could be a useful potential cancer therapy depending on the cancer type.

There are various delivery methods for miRNAs, including virus-based, anti-miRNA oligonucleotide delivery systems. The viral-based delivery systems include the use of retroviral, lentiviral, adenoviral, adeno-associated, and bacteriophage-based vectors [178]. The viral-based miRNA delivery systems are highly immunogenic, toxic, and have size limitations, therefore there is a need to introduce non-viral-based methods for the delivery of miRNAs and anti-miRNA oligonucleotides. The methods utilised have involved the use of lipids, polymers, inorganic and extra-cellular vesicle carriers [178]. There are still many challenges with the use of miRNAs in human trials [167] despite the advancements in miRNA delivery systems, and miRNA drugs have yet to reach phase III human trials [177]. For example, in solid tumours such as hepatocellular carcinomas, the trial of the MRX34 drug, which delivers miR-34 via liposomes intravenously to patients, was terminated due to immune-related severe adverse events [179]. There is therefore a need to understand the regulatory mechanism(s) behind the miRNA [178] actions and their effects on PTEN and PTENP1 in order to use them for a therapeutic advantage.

Alterations in the levels of PTEN, microRNAs, and PTENP1 are a new frontier in cancer therapeutics with the potential to reverse the cancer phenotype by positively manipulating the PTEN–microRNA–PTENP1 axis in favour of a precancerous cellular phenotype. In the future, successful precision therapeutic targeting in human trials delivering miRNAs and/or PTEN and PTENP1 transcripts will have the possibility of treating various cancers.

8. Conclusions

Alterations of the cellular levels of PTEN, miRNAs, and PTENP1 presents a new frontier in cancer therapeutics with the potential to reverse the cancer phenotype through the positive manipulation of the PTEN–miRNA–PTENP1 axis in favour of pre-cancer levels and induce a pre-cancerous cellular phenotype. The importance of PTEN cellular activity and function has been highlighted in the myriad of studies showing the loss of PTEN expression and/or function as the cause of PHTS and many cancers of various tissue origins. Knowledge of these new mechanisms of post-transcriptional regulation of PTEN has opened new avenues for development of novel PTEN-restoring cancer therapies through manipulation of the PTEN–miRNA–PTENP1 axis. Whether through the introduction of PTEN mRNA, to increase PTEN cellular concentration, increasing or decreasing PTENP1 expression, and/or altering the level(s) of specific PTEN-regulating miRNAs, it is tempting to consider these future therapies that may allow the fine tuning of PTEN cellular levels to achieve and maintain pre-cancerous levels. While manipulating the PTEN–miRNA–PTENP1 axis holds great promise for the future of cancer therapies, our knowledge of the mechanisms of post-transcriptional regulation of PTEN, the various competing components, and the complexities of their interactions needs further study to allow this to become a future reality. Notwithstanding this, future successful human trials delivering miRNAs and/or PTEN and PTENP1 transcripts have great potential in precision therapeutic targeting and the treatment of a broad range of PTEN-related malignancies.