HOW miRs AND mRNA DEADENYLASES COULD POST-TRANSCRIPTIONALLY REGULATE EXPRESSION OF TUMOR-PROMOTING PROTEIN PLD

Abstract

Phospholipase D (PLD) plays a key role in both cell membrane lipid reorganization and architecture, as well as a cell signaling protein via the product of its enzymatic reaction, phosphatidic acid (PA). PLD is involved in promoting breast cancer cell growth, proliferation, and metastasis and both gene and protein expression are upregulated in breast carcinoma human samples. In spite of all this, the ultimate reason as to why PLD expression is high in cancer cells vs. their normal counterparts remains largely unknown. Until we understand this and the associated signaling pathways, it will be difficult to establish PLD as a bona fide target to explore new potential cancer therapeutic approaches. Recently, our lab has identified several molecular mechanisms by which PLD expression is high in breast cancer cells. First, PA, a mitogen, functions as a protein and mRNA stabilizer that counteracts natural decay and degradation. Second, there is a repertoire of microRNAs (miRs) that keep PLD mRNA translation at low levels in normal cells, but their effects change with starvation and during endothelial-to-mesenchymal transition (EMT) in cancer cells. Third, there is a novel way of post-transcriptional regulation of PLD involving 3'-exonucleases, specifically the deadenylase, Poly(A)-specific Ribonuclease (PARN) which tags the mRNA for degradation. This enables PLD accumulation and ultimately breast cancer cell growth, proliferation, and metastasis. We review in depth the emerging field of post-transcriptional regulation of PLD, which is only recently beginning to be understood. Since, surprisingly, so little is known about post-transcriptional regulation of PLD and related phospholipases (PLC or PLA), this new knowledge could help our understanding of how post-transcriptional deregulation of a lipid enzyme expression impacts tumor growth.

Introduction: Post-transcriptional regulation of messenger RNA

Transcriptional control is the predominant form of regulation for most genes [1]. However, after RNA polymerase has bound to the gene's promoter and RNA synthesis has started, post-transcriptional regulation can still control the amount of gene that is ultimately expressed and for many genes, post-transcriptional control is essential. Post-transcriptional control of gene expression is important for cellular functions across biological contexts. It comprises a complex regulatory network that contributes to cell-type and organism specific gene expression patterns.

Pre-mRNA synthesized in the nucleus undergoes a series of modifications that include capping, splicing, addition of poly(A) tail, RNA editing, nuclear degradation (exosome), sequence-specific nuclear export mRNA, all of which occur in the cell nucleus. More post-transcriptional control mechanisms occur in the cytoplasm, such as stability and lifetime in the cytosol and small regulatory RNAs, specifically microRNA (miRs),that can be considered translation regulators (Figure 1).

Figure 1. Post-transcriptional control of gene expression.

It comprises a complex regulatory network that contributes to cell-type and organism specific gene expression patterns. The pre-mRNA has to go through some modifications to become a mature mRNA molecule that can leave the nucleus and be translated. Those modifications include capping, splicing, addition of poly(A) tail, RNA editing, nuclear degradation (exosome), sequence-specific nuclear export mRNA, stability and lifetime in the cytosol and small regulatory RNAs, specifically microRNA (miRs). This review will focus on the three events (labeled “1”, “2” and “3”) occurring in the cytoplasm.

Microarray analysis has indicated that close to 50% of the changes in inducible gene expression occur at the level of mRNA stability [2] highlighting the exquisite level of post-transcriptional regulation. All of these post-transcriptional regulations that occur in both the nucleus and the cytoplasm determine the level of gene expression and how much of the transcripts are ultimately translated into proteins [3]. For the purpose of this review we will concentrate on the mRNA stability and miR post-transcriptional events.

PLD as an example to apply new post-transcriptional control mechanisms

Extensive studies exist on the enzymatic regulation of phospholipid phospholipases A, C and D. However, little is known about their transcriptional and especially post-transcriptional regulation. This is surprising considering the central role these phospholipases play in lipid metabolism and cell signaling. The discrete number of articles found in the scientific literature, some of which are cited below, offer a glimpse of how valuable post-transcriptional control could be for lipid enzyme signaling. For example, regarding PLA2, insulin-like growth factor-I (IGF-I) destabilizes mRNA of the type II sPLA2. Conversely, IL-1beta stimulates the transcription rate and gives rise to a very stable mRNA [4]. PLA, regulation of lipolytic activities by PLA2 depends on the transcriptional regulators LetA/S and RpoS, inducing the expression of virulence traits, and on post-transcriptional activators like the zinc metalloprotease ProA [5]. Inhibition of endogenous miR-338 with anti-miR-338 increased the mRNA and protein expression of PLA2G4B in decidual cells with a proposed role in human pregnancy and parturition [6].

Downstream phospholipid catabolism by PLA2 produces arachidonic acid that can be used in prostaglandin synthesis by PGEs. A group of selected miRs regulate mRNA expression M-type phospholipase A2 receptor (PLA2R1) in normal human mammary epithelial cells and cancer cell lines [7]. There are several RNA sequence elements within the 3'UTRs of the genes involved in the PGE(2) pathway, that are predicted to be binding sites for miRNAs and RNA-binding proteins, both of which appear to be central regulators of PGE(2) synthesis and function [8]. Regarding another phospholipase, PLC, a particular group of miRs (miR-200b, miR-200c, and miR-429) target PLC (PLCG1) during regulation of PG function [8]. Inhibition of miR-214 in C2C12 cells enhances protein expression of PLCβ1 and promotes C2C12 BMP-2-induced osteogenesis through PLCβ1 [9]. Thus, numerous microRNAs have recently emerged as post-transcriptional gene repressors for phospholipases (like PLA2 and PLCβ) or phosphatases. However, very little is known about deadenylases post-transcriptionally regulating phospholipases.

PLD background

The conversion of PC to PA is catalyzed by the enzyme PLD [10] and is, in general, dependent on the presence of the co-factor phosphatidylinositol 4,5-bisphosphate (PIP2) (an anionic lipid localized primarily to the plasma membrane) [11, 12]. PLD has been found in a variety of cells and tissues. Its activity has been reported predominantly in the plasma membrane, as well as in cytoplasmic locations, the mitochondrial membrane, the Golgi endoplasmic reticulum (ER), the nucleus, the nuclear membrane and subcellular compartments [13, 14]. There is also an interplay between PLD and Diacylglycerol Kinase (DGK). PLD generates phosphatidic acid by catalyzing the hydrolysis of phosphatidylcholine (PC), which could be de-phosphorylated to generate DAG. Inversely, DGK catalyzes the phosphorylation of DAG to synthesize PA. Thus, both enzymes regulate the levels of DAG and PA, increasing the later and decreasing the former. This is important for intracellular vesicle trafficking cycling as both PA and DAG are both needed for exocytosis [15]. Many of these functions are negated or diminished during PLD loss of function [16, 17]. In elegant studies, Ryu’s group [17] and Frohman’s group [18] have provided direct evidence of the function of PLD2 in several pathological conditions such as cancer, vascular disease, immunological disease, and neurological disease.

Little is known about PLD post-transcriptional regulation. However, recent publications from our lab are indicating a previously unsuspected level of post-transcriptional regulation of PLD mRNA that can affect how much and when PLD is expressed. As such we will discuss post-transcriptional regulation of PLD as a new paradigm in this review as a novel and interesting alternative perspective on PLD. Understanding control of PLD expression is important because PLD has been documented as having a direct involvement in promoting breast cancer cell growth, proliferation, and metastasis [19–21].

The three post-transcriptional control mechanisms considered in this review

We posit that normally operating mechanisms in non-cancerous cells keeping PLD at a desired level are deregulated in cancer cells. Thus, we will concentrate on the understudied field of PLD post-transcriptional regulation and will review the following three mechanisms: (a) miRs, (b) deadenylases, and (c) a combination of the two (Figure 1).

In (a), microRNAs are a well-studied means of post-transcriptional regulation of protein levels within the cell. The association of miR-loaded RISC on targeted mRNA functions to inhibit translation of the mRNA by either inhibiting ribosomal function or by inducing the degradation of the mRNA [22].

For the second mechanism of post-transcriptional control considered here (b), the involvement of deadenylases, specifically, 3' exonucleatic cleavage of mRNA poly-A tails by deadenylases, we will concentrate on PARN as it was downregulated in patient invasive breast carcinoma samples compared to adjacent normal control tissue and at the same time Phospholipase D (PLD) was upregulated in these same breast carcinoma samples.

For the third mechanism (c), mRNA deadenylation is under tight control of cis-acting regulatory elements, both of which are located in the 3’ UTR of eukaryotic mRNAs. In a unique and still somewhat controversial mechanism, miRs promote deadenylase-induced mRNA decay.

PLD in cancer

The mammalian phospholipase D (PLD) family members are important signaling molecules that hydrolyze membrane lipids [18, 24, 25]. The classical PLD isoforms (PLD1 and PLD2) are ubiquitously expressed and are the best characterized. PLD1 localizes predominately to cytoplasmic membranes while PLD2 localizes to the plasma membrane [18, 24, 25]. Their enzymatic function is to hydrolyze phosphatidylcholine (PC) to free choline and phosphatidic acid (PA) [18, 24, 25]. PA is a critical secondary messenger signal within the cell, regulating pathways leading to cell growth and proliferation, vesicle trafficking, and cell migration [18, 24, 25].

Several authors have reported an increase in PLD gene expression, protein expression, and enzymatic activity in multiple cancer types including breast [19, 26–30], gastric [31, 32], colorectal [33], renal [34], thyroid [35], and brain [36]. Regarding the enzymatic activity, the product PA is highly mitogenic and has been shown to be involved in regulation of tumorigenesis, cell proliferation, cell invasion, and cell movement including metastasis [37–39]. PA mediates induction of HIF-1α [40]. Regarding expression of PLD protein, it has been shown that PLD1 is elevated in breast cancer xenotransplants and Frohman’s group studied PLD1 in the tumor microenvironment (TME) and angiogenesis [28]. Our laboratory has recently shown [26, 27, 41] that increased PLD1 and PLD2 protein expression in breast cancer cells correlates with increases in PLD lipase activity [27]. Fite et al. demonstrated that several miRNAs regulate PLD2 in non-cancerous human breast cells, but upon epithelial-mesenchymal transition (EMT), these miRNA were downregulated, allowing for the increase in PLD2 protein mass observed in an invasive breast cancer cell line [27].

It is to be expected that keeping mRNA transcripts for PLD at manageable levels is desirable in normal cells, but in cancer, PLD1 and PLD2 mRNA transcripts are maintained at more elevated levels and subsequently translated into protein more efficiently.

Regulating protein expression post-transcriptionally by non-coding RNAs

Between 70–90% of the human genome is transcribed into RNA but only approximately 2% of the genome encodes protein [42], meaning the majority of RNA are non-coding RNA molecules. Non-coding RNAs are classified as either long (lncRNAs) (greater than 200 nucleotides) or short (less than 200 nucleotides) [43]. Types of long non-coding RNAs include: sense lncRNAs, antisense lncRNAs, bidirectional lncRNAs, intronic lncRNAs, and long intergenic ncRNAs (lincRNAs). Sense and antisense lncRNAs overlap exons of a protein-coding gene while bidirectional lncRNAs are transcribed opposite to another transcript and intronic lncRNAs are transcribed from within introns. LincRNA transcripts derive from regions between two protein-coding genes [44]. Long non-coding RNAs function as competitor RNA, or “sponges”, for miRNA targeting or function as mediators of epigenetic silencing [44]. Their expression is cell and tissue specific and increasing evidence supports lncRNA involvement in cancer progression [43, 45, 46]. The other category, small non-coding RNAs, include tRNAs and rRNAs, in addition to miRNAs, siRNAs, snoRNAs, snRNAs, and piRNAs [44]. MicroRNAs (miRNAs; miRs) and endogenous siRNA are means of posttranscriptional regulation within the cell and function to inhibit protein translation of target mRNA. Small nucleolar RNAs (snoRNA) are involved in post-transcriptional regulation of rRNA [47]. Small nuclear RNA (snRNA) associate with proteins of the spliceosome and therefore function in intron removal in mRNA processing [48]. PIWI-interacting RNA (piRNA) serve to silence transposable elements thereby maintaining genome integrity in germ cells [49]. We focus herein on discussing miRNA and cancer (Figure 2).

Figure 2. MicroRNA (miRNA, miR) and cancer.

miRs are short (~22 nucleotide) single-stranded RNA molecules. After transcription (1), the miR precursors undergo several processing steps in the nucleus and cytoplasm. In the nucleus a key step is their cleavage by endonucleases, such as Drosha (2) and the resulting pre-miR is exported from the nucleus (3). In the cytoplasm, Dicer (4) produces a mature miR duplex that, after unwinding (4), becomes associated with Ago2 in the RISC (RNA-induced silencing complex) (5). Once the miR is loaded onto the RISC, the entire complex targets mRNA based on sequence complementarity with the seed sequence (first 8 amino acids) of the miR. The association of miR-loaded RISC on mRNA functions to inhibit translation of the mRNA by either inhibiting ribosomal function or by inducing the degradation of the mRNA. This process is deregulated in several pathologies including cancer.

miRs in cancer

MicroRNA (miRNA, miR) are short (~22 nucleotide) single-stranded RNA molecules [50–52]. After transcription, the miR precursors undergo several processing steps in the nucleus and cytoplasm, in which they are cleaved by endonucleases, Drosha and Dicer. Cleavage by Dicer produces a mature miR that becomes associated with Ago2 in the RISC (RNA-induced silencing complex) [53–56]. Once the miR is loaded onto the RISC, the entire complex targets mRNA based on sequence complementarity with the seed sequence (first 8 amino acids) of the miR. The association of miR-loaded RISC on mRNA functions to inhibit translation of the mRNA by either inhibiting ribosomal function or by inducing the degradation of the mRNA [22]. MicroRNA are one of the main factors in post-transcriptional regulation to regulate protein levels within the cell. It is predicted that miRs regulate the majority of the human transcriptome and are therefore involved in virtually every signaling pathway. Furthermore, deregulation of miRs contributes to numerous pathologies, including cancer [22, 57].

Many investigators are pursuing the potential of microRNAs as biomarkers in various pathologies including cancer [58–61]. MicroRNAs make attractive candidates for their stability in body fluid samples such as blood and urine, which are less invasive and less painful diagnostic samples than tissue biopsies. Additionally, microRNAs are being investigated as biomarkers for response to therapy [62, 63]. Several miRNA-based interventions are currently in clinical trials. The first was Miravirsen, which is currently in Phase II clinical trial. This drug targets the liver-specific miR-122 in the context of Hepatitis C. Hepatitis C virus uses the liver miR-122 in stabilizing its genome and downregulation of miR-122 would reduce viral production [64, 65]. MiR-122 was an ideal target because it is only expressed at meaningful levels in the liver and targeting it would therefore not have abundant off-target effects in many other cell types [65, 66]. MRX34 was a miR-34 mimic intended to treat melanoma patients. However, MRX34 was pulled from Phase I clinical trial in 2016 due to severe adverse effects [67]. The delivery system of MRX34 was intravenous injection of miR-34-containing liposomes. This study highlights the challenge in designing miRNA-based interventions. Since one miRNA can have numerous targets and the targets can vary between cell types, administering a systemic miRNA-based intervention could impact many tissues.

Another recent trial took a targeted approach with TargomiRs and a Phase I clinical trial was completed in early 2017. These specific TargomiRs are miR-16 mimics designed for lung cancers including mesothelioma and non-small cell lung cancer (NSCLC) [68]. The miR mimics are packaged into nanoparticles with anti-EGFR expressed on the nanoparticle surface with the intention to target the nanoparticles to EGFR-expressing cells. Both mesothelioma and NSCLC are known for their high expression of EGFR and as such, if successful, the anti-EGFR coated nanoparticles should specifically target these cancer cells and release the TargomiRs selectively.

miRs, PLD: an interlocked positive and negative feedback mechanism

Two recent studies in our lab serve to clear the way to understand how miRs regulate PLD expression. In the first, our laboratory identified a set of microRNAs predicted to have “high score” binding abilities to PLD 3’UTRs [27]. Matrigel-based cell invasion assays and breast cancer cell lines showed cell invasion was reduced in the presence of these miRs due to a decrease in PLD protein. Additionally, expression of these miRs likely decrease as a result of the epithelial-mesenchymal transition (EMT). EMT induces genetic and epigenetic changes in the cell allowing cancer cells to adopt a more mesenchymal cell phenotype with increased ability to migrate. E-cadherin triggers expression of miRs in pre-EMT breast cancer cells, thereby keeping PLD levels low [27]. Exogenous addition of these miRs negatively affects PLD protein levels in the post-EMT MDA-MB-231 cells leading to a reduction in cell invasion. The invasive properties of MDA-MB-231 cells through Matrigel matrix were increased with PLD overexpression and then negated following miR overexpression.

In a second study from our lab [26], we asked the question of what happens to PLD and miR expression in tumor-like conditions including nutrient starvation, hypoxia, or culture cell density? PLD expression is upregulated in nutrient-starved cells. Expression of certain microRNA increases after prolonged (> 12 hours) starvation. In fact, the earlier identified repertoire of microRNAs directly target and regulate PLD itself during cancer cell starvation. In conditions of cellular stress, such as hypoxia and starvation, cancer cells increase PLD expression and activity as well as cell invasion. Prolonged starvation of cultured cells reverses that phenotype. The mechanism for this biphasic process is: Initially, PLD mediates cell invasion. PLD associates with motility proteins and PA promotes the formation of “positive” membrane curvature needed for the formation of lamellipodia and motility cellular structures in the cell membrane. With prolonged starvation, autophagy and survival become paramount to cell migration. As PLD rises, increased PA levels promote transcription of several miR genes. Although no specific transcription factor has been demonstrated, Mahankali et al [69] have demonstrated that PPAR binds to PA and regulates the expression of both PLD and EGFR. Several miRs target the 3’UTR of PLD transcripts downregulating PLD translation, so the PLD-led process comes to an end in prolonged starvation. In other words, PLD protein, activity, mRNA, and miR operate in a feedback loop in cancer cells (Figure 3).

Figure 3. PLD protein, activity, mRNA, and miR operate in a feedback loop in cancer cells.

(A) There is a causal relation between certain miRs and PLD expression. (B) During starvation, PLD (a “stress protein”) rises and mediates cell invasion. This could help cells at the core of a primary solid tumor escape hypoxia and low nutrient conditions. However, with prolonged starvation, PA levels (possibly to association to trasncription factors of the PPAR family) promote expression of several miR genes, which un turn target the 3’UTR of PLD transcripts, consequently downregulating PLD translation. In this way, the PLD-led process comes to an end in prolonged starvation.

Another way of regulating protein expression post-transcriptionally: mRNA decay regulated by deadenylases

For mRNA decay, the rate-limiting step is the shortening of mRNA poly(A) tails. This is accomplished through the action of a group of 3’ to 5’ exonucleases and is one of the most potent methods for preventing mRNA translation and induce mRNA transcript turnover [70–73]. The 3’ to 5’ exonucleases involved in this poly(A) tail shortening are known as deadenylases, of which there are many. The two major eukaryotic deadenylase complexes are known as Ccr4-Not and Pan2/Pan3 [73, 74]. A third deadenylase, poly(A)-specific ribonuclease (PARN), has also been identified [75–77].

Deadenylases (Ccr4-Not, Pan2/Pan3, and PARN)

Ccr4-Not

The Ccr4-Not complex consists of 9 components and the human orthologs are termed CNOT1-CNOT10 [78, 79]. While many studies have been in the yeast system, interest is growing in understanding the role of the human CNOT complex in physiology and pathology. The Ccr4-Not yeast complex regulates transcription by associating with transcription factors, however, this mechanism of transcriptional control is not well understood [74]. The better characterized mechanism of the complex is its deadenylation of mRNA [74], which is conserved in humans [80]. In yeast, Ccr4 contains the deadenylase activity and is a global regulator of mRNA decay in yeast. Notably, it regulates the deadenylation of mRNA encoding for ribosomal proteins and is therefore important in the control of ribosomal protein levels within the cell [81, 82].

Pan2/Pan3

Pan2/Pan3 and Ccr4-Not complexes are responsible for the majority of mRNA deadenylation activity in eukaryotes [73, 83]. Pan stands for Poly(A) Nuclease and, similar to Ccr4-Not, is most highly studied in yeast [73, 84]. The Pan complex has 3’ to 5’ exonuclease activity and can only degrade the 3’ poly(A) tail with no exonuclease activity on the rest of the mRNA [73]. Unlike the Ccr4-Not complex, the Pan complex consists of only two proteins, Pan2 and Pan3 [73], with Pan2 having exonuclease activity [85, 86]. Pan2/Pan3 predominately localizes to P-bodies in the cytoplasm and it remains unclear if the complex functions in the nucleus [73, 87].

PARN

PARN is a Mg²⁺ ion-dependent, poly(A)-specific 3’ to 5’ exonuclease [88]. PARN protein contains a large nuclease domain as well as two distinct RNA-binding domains. PARN has 5’ GTP cap-binding capabilities due to the RNA recognition motif (RRM) RNA-binding domain. This enhances PARN processivity. A second RNA binding domain, the R3H, is essential for PARN dimer stability [89]. Importantly, PARN shows a preference for the targeting of mRNAs containing AU-rich elements in their 3’ untranslated regions (UTRs) [90–93] (Figure 4).

Figure 4. The 3’-mRNA exonuclease PARN.

(A) Modular scheme of the PARN protein. The PARN protein consists of three distinct exonuclease (deadenylase) sites (Exo I, Exo II, and Exo III) contained in two nuclease regions also with two RNA binding regions (R3H and RRM). (B) PARN targets mRNA 3' UTR containing long stretches of AU combinations called, AU- rich elements (AREs). Such elements are present in the 3' UTR of pLD1 mRNA. (C) PARN contributes to mRNA degradation through the cleavage of mRNA poly(A) tails (deadenylase function) followed by decapping and by exonuclease activity; cleaved mRNAs accumulate in “P bodies” in the cell that can be observed under the microscope.

PARN has been shown to participate in non-sense mediated decay and also in the process of human telomerase RNA component (TERC) maturation. In the nucleus, nuclear cap binding protein complex (CBC) inhibits PARN deadenylation activity, which potentially interferes with PARN's function in nonsense-mediated decay and TERC maturation [94]. PARN counteracts this oligoadenylation by cleaving off the oligo-A tails on the TERC, allowing it to serve its crucial role in telomerase [95–97]. In the cytoplasm, the integration of mRNA decay and translation initiation has been proposed where PARN plays an important role [98].

PARN deficiencies have been shown to be a cause in the development of a severe form of dyskeratosis congenita, a telomere disease, due to the progressive shortening of telomeres [96, 99–102]. Previous studies have also begun to characterize PARN’s mRNA targets. PARN can target transcripts involved in cell migration, adhesion, p53 signaling, BRCA1 DNA damage response, and oncogenes such as c-myc, c-fos, and c-jun to keep these transcripts levels controlled under normal conditions [103–110].

Deadenylases in PLD-associated breast cancer

After PARN has removed the poly-A tail the remaining mRNA is “marked” for further or total degradation. PARN preferentially targets mRNA 3' UTR containing long stretches of AU combinations called, AU- rich elements (AREs). As described herein earlier, PLD is a cell-signaling molecule well known for promoting breast cancer cell growth, proliferation, and metastasis, and was found to be upregulated in human breast carcinoma samples compared to the normal adjacent breast tissues.

It is known that PARN regulates the transcript levels of several proteins involved in cancer development and progression, can be activated by the tumor-suppressor BARD1, and when overexpressed in SCC patients, they survive 7.0 months longer than in patients that underexpress it [103]. Interestingly, PARN expression levels in human breast carcinoma samples were decreased compared to the normal control tissues, the opposite results of PLD (Fig. 5), and would be an interesting area of further study.

Figure 5. An inverse relation exists between PARN and PLD expression in breast cancer.

Microarray data from the Finak Breast dataset were downloaded from the Oncomine database [115]. The data compares PARN levels of deadenylase (PARN) (A) and PLD2 (B).

miR and deanylation regulation of protein translation

While still somewhat controversial in the field, there is evidence that all three major cellular deadenylases, Ccr4-Not, Pan2/Pan3, and PARN can exert their function through miR-dependent mechanisms. Recently, PARN has been identified as capable of using a similar miR-dependent mechanism. PARN protein was shown physically bound to Ago2 in the RNA-Induced Silencing Complex (RISC). Once the RISC was loaded with miR-504 or miR-125b, the RISC, along with PARN, were brought to the TP53 mRNA 3’ UTR for which these miRs targeted. This resulted in subsequent TP53 transcript degradation under non-stress conditions. Upon UV-induced DNA damage, this degradation of TP53 mRNA was abolished, allowing TP53 translation into functional p53 protein to then exert DNA damage response [23, 106].

It is also possible that a combination of PARN deadenylase activity and a putative presence of a miR that would synergize in destabilizing PLD transcripts impeding protein translation (Fig. 6).

Figure 6. Hypothetical post-transcriptional regulation of PLD.

This is a hypothetical model showing a combination of PARN deadenylase activity and a putative presence of a miR that would synergize in destabilizing PLD transcripts hampering translation.

Future directions

In the study of PLD post-transcriptional regulation [26, 27, 41] new questions that are arising are: if PARN can regulate PLD, what controls this process and what role does PARN have in the patho-physiology and development of breast cancer? It has been shown that phosphorylation is important in PARN regulation. Under UV to induce DNA damage, PARN is phosphorylated at S557 by MK2 and no longer binds and regulates Gadd45α mRNA [111]. Under serum starvation conditions, PARN can be phosphorylated which increases its 5’ GTP cap binding affinity to out compete eIF4E for access to the cap [112]. Poly(A) binding proteins (PABPs) can inhibit PARN action by binding RNA and excluding PARN from its target [113]. Cap-binding proteins (CBPs) can also repress PARN [94, 113]. CUG-binding protein anchors PARN to mRNA targets thus increasing PARN processing [110, 114]. The role PARN has in the patho-physiology and development of breast cancer also needs further investigation.