Plasma membrane and nuclear phosphatidylinositol 4,5-bisphosphate signalling in cancer

The role of actin in the cytoplasm is well known, but its presence in the cell nucleus has been debated for many years. Recently, the development of advanced imaging techniques has made it possible to visualise and study nuclear actin in detail [ 61 , 62 ]. Actin lacks a nuclear localisation signal (NLS); thus, its nuclear localisation depends on active transport. This shuttle transport is enabled by cofilin and importin 9, which are responsible for importing actin into the nucleus, and profilin and exportin 6, which facilitate actin export [ 63 ]. The main limitations of this process are the availability of Gactin in the cytoplasm and the level of unphosphorylated cofilin. For a long time, the inability to visualise filamentous nuclear actin led to the assumption that actin did not form filaments within the nucleus. However, there is now evidence that nuclear actin filaments can be observed during specific physiological processes or pathological conditions, although they are mainly present as monomers [ 61 , 64 ]. One of these processes is the regulation of chromatin regulatory complexes. Actin interacts in a PI(4,5)P2dependent manner with the mammalian ATPdependent chromatinremodelling complex (SWI/SNFlike BAF complex), which is involved in a variety of cellular processes ranging from gene activation and repression to cell development and differentiation. [ 65 , 66 ]. In homologydirected DNA damage repair (DDR), the N-WASP/Arp2/3 complex mediates nuclear actin polymerisation, which is required for the clustering of DNA breaks [ 63 ]. In addition, actin has been reported to be involved in the initiation and regulation of transcription mediated by all three RNA polymerases [ 67 – 69 ]. Additionally, nuclear Gactin plays a role in Pol II pause release through its interaction with PTEFb [ 70 ]. Interestingly, upon serum stimulation, Arp2/3 and N-WASP promote actin polymerisation, which plays a crucial role in enhancing Pol II clustering and significantly changing the dynamics of transcription condensates. The larger and more stable clusters transcribe genes associated with survival, proliferation, angiogenesis, migration, EMT, and drug resistance [ 71 ]. In comparison to actin, the presence of phosphoinositides in the cell nucleus is less controversial and more widely accepted [ 72 , 73 ]. Phosphoinositides are located in nucleoplasmic membraneless nuclear structures such as nuclear speckles or nucleoli [ 74 – 76 ]. Recently, researchers have started to explore the potential role of phosphoinositides in regulating nuclear processes. The most abundant phosphoinositide detected in these structures was PI(4,5)P2, which is thought to play a role in the spatial and functional organisation of the nucleus (Fig.  2 ) [ 75 ]. The cell nucleus possesses its own pool of phosphoinositidemetabolizing enzymes, which is distinct from the cytoplasmic pool and thus enables them to control their involvement in nuclear processes specifically [ 77 – 79 ]. Fig. 2 Schematic illustration of the nuclear PI(4,5)P2 role in normal cells ( A ) and cancer cells ( B ). Figure ( A ) highlights the mechanism of the transcriptional regulation of Pol II in normal cells. The myosin phosphatase Rho-interacting protein (MPRIP) mediates the interaction between Pol II initiation condensate (Tyr1P-CTD) and PI(4,5)P2-containing structures which enables the release of RNAPII from the promoter-proximal pausing. Figure ( B ) highlights the implications of the altered nuclear PI(4,5)P2 levels in carcinogenesis Schematic illustration of the nuclear PI(4,5)P2 role in normal cells ( A ) and cancer cells ( B ). Figure ( A ) highlights the mechanism of the transcriptional regulation of Pol II in normal cells. The myosin phosphatase Rho-interacting protein (MPRIP) mediates the interaction between Pol II initiation condensate (Tyr1P-CTD) and PI(4,5)P2-containing structures which enables the release of RNAPII from the promoter-proximal pausing. Figure ( B ) highlights the implications of the altered nuclear PI(4,5)P2 levels in carcinogenesis Previously, the dysregulation of the nuclear PI(4,5)P2 pool and its metabolism were overlooked, as the observed changes were more aligned with the total cellular PI(4,5)P2 levels. However, recent exploration of the nuclear PI(4,5)P2 has uncovered its role in the crucial nuclear processes, whose dysregulation is associated with various pathological conditions, including cancer [ 56 , 80 , 81 ]. These include the chromatin remodelling process mentioned above, as well as the acylation and deacylation of histones, RNA processing, DDR, regulation of Pol II transcript activity, premRNA splicing, and polyadenylation [ 19 , 74 , 82 , 83 ]. For these processes to function properly, it is essential that appropriate proteins and nucleic acids are concentrated in a specific location. This can be achieved through the formation of biomolecular condensates that concentrate molecules and separate them from their surroundings [ 84 ]. The liquid-liquid phase separation (LLPS) phenomenon plays an essential role in this organisational process [ 85 , 86 ]. The resulting condensates are usually spherical, dynamically coalesce, and undergo rapid fluorescence recovery after photobleaching, demonstrating a high degree of internal dynamics [ 87 ]. The hallmark of the LLPS process is multivalent weak inter and intramolecular interactions, where proteins undergoing LLPS are characterised by distinctive features, such as multiple structured modular domains as well as intrinsically disordered regions (IDRs) or oligomerisation domains [ 84 ]. Recently, PIPs have been suggested to be novel LLPS regulatory molecules [ 22 , 74 , 88 ]. The PI(4,5)P2 effector protein and Factin regulator, myosin phosphatase Rhointeracting protein (MPRIP), was identified as a component of the Pol II initiation condensate and as a transcription modulator with condensation capacity [ 89 ]. MPRIP recruits myosin 1 C (NM1)/Pol II complex via Pol II Tyr1PCTD to PI(4,5)P2containing structures [ 21 ]. Notably, a recent study by Dumilie et al. showed that PI(4,5)P2 is a key component of natural condensates and is enriched in mediator (MED1) condensates [ 22 ]. Together with BDR4, MED1 is a transcriptional cofactor found in superenhancers that can compartmentalise the Pol II transcriptional apparatus [ 90 ]. The BRD4MED1 SuperEnhancer condensates play crucial roles in boosting the transcription of protooncogenes, which cancer cells use to regulate cell proliferation and survival [ 91 ]. Thus, the abnormal LLPS process of PI(4,5)P2containing condensates may lead to the formation of pathological structures with abnormal physicochemical properties and trigger a malignant transcriptional program leading to disease, including cancer progression and metastasis. In addition, the crucial DDR pathway, which is critical for genome stability, may also be affected by abnormal LLPS and contribute to its dysregulation and, ultimately, to carcinogenesis. During the DDR response, the interaction of PI5P, a precursor of PI(4,5)P2, with an inhibitor of growth protein 2 (ING2) leads to the acylation of the tumour suppressor protein p53, thereby transactivating the p53dependent apoptotic pathway [ 16 , 18 ]. A previously published study showed an increase in PI(4,5)P2 levels in HPV8induced skin tumour samples as well as in HPV16positive samples from lowgrade cervical intraepithelial neoplasia (CIN I) to invasive cervical cancer [ 92 ]. HPV E6 protein is known to induce ubiquitination and degradation of the p53 protein [ 93 , 94 ], while PI(4,5)P2 is known to stabilise p53 [ 16 ]. An unknown epigenetic modification may be present in these cancers that protects p53 from degradation and simultaneously ensures cancer cell survival. The elevated nuclear PI(4,5)P2 levels observed in betaHPVpositive skin and HPV16positive cervical tumours highlight the clinical significance of these findings in HPVmediated tumorigenesis. These results established a novel link between HPV infection and nuclear PI(4,5)P2 metabolism. Given that elevated nuclear PI(4,5)P2 is detected as early as in CIN I lesions, HPVinfected cells represent a promising model for studying the role of PI(4,5)P2 in tumour progression and metastasis. The absence of elevated PI(4,5)P2 signals in other skin cancers, such as malignant melanoma and Merkel cell carcinoma, strongly indicates a specific effect on HPVassociated skin tumours [ 95 ]. The specificity of these signals warrants further investigation to determine whether similar molecular patterns exist in other cancers or are unique to HPVassociated cancers. The methodology developed by Hoboth et al. can be used for this purpose on paraffinembedded section patient material obtained from clinicians [ 96 ]. Several vital questions must be addressed in future research. First, it is important to investigate whether elevated nuclear PI(4,5)P2 levels are also associated with HPV16positive oropharyngeal squamous cell carcinoma (OPSCC) compared to nonHPVrelated OPSCC to determine whether the cancer origin and tumour grade affect the elevated nuclear PI(4,5)P2 levels. It also needs to be clarified whether an increase in nuclear PI(4,5)P2 levels is specific to oncogenic skin and mucosainfecting HPV types or if it is a characteristic of HPVinduced tumours or tumours in general. As mentioned above, PI(4,5)P2 is known to localise to nuclear speckles, nuclear lipid islets, and nucleoli [ 7 ]. The compartmentalisation of the nucleus and, thus, the existence of different pools of PI(4,5)P2 might allow local finetuning of specific nuclear processes [ 12 ]. Using superresolution threedimensional structured illumination microscopy, PI(4,5)P2 was found to localise to nuclear speckles in HPVpositive associated cancers, and the precise link between nuclear speckle function and nuclear PI(4,5)P2 remain elusive and needs to be determined. Furthermore, it should be investigated whether only nuclear PI(4,5)P2 levels are dysregulated in HPVinfected tissues or if other phosphoinositide species are also differentially regulated in HPVrelated tumours. An increase in PI(4,5)P2 may alter PI(3,4,5)P3 levels. However, it remains unclear which nuclear kinases or phosphatases are dysregulated in the presence of the virus. Further research is needed to clarify the mechanisms behind these potential changes in phosphoinositide levels and to understand their implications for the pathogenesis and progression of HPVrelated cancers. Additionally, exploring the interplay between these phosphoinositides may provide insight into early diagnosis, as nuclear PI(4,5)P2 levels changes are observed as early as CIN I or novel therapeutic targets for managing HPVassociated malignancies.

Phosphoinositides are fundamental lipids involved in various cellular processes, including the regulation of membrane dynamics and signalling cascades, as well as in nuclear processes such as chromatin organisation. Dysregulation of phosphoinositides, whether in the cytosolic or nuclear pools, can contribute to the progression of several diseases, particularly cancers. Dysregulation of the PI3K signalling pathway has been observed in various types of cancer, underscoring the potential of PI3K as a target for therapeutic intervention. While research involving therapeutic targeting of the PI3K pathway has progressed, the therapeutic efficacy of PI3K inhibitors continues to be challenged by their toxicity, acquired drug resistance, and offtarget activity. Currently, five small molecule PI3K inhibitors have been clinically approved by the U.S. Food and Drug Administration and the European Medicines Agency [ 97 ]. Nonetheless, advancement of techniques, such as highthroughput screening, have facilitated the design of novel smallmolecule inhibitors with higher specificity, selectivity, and potency. As a result, more than 60 novel smallmolecule PI3K inhibitors are currently undergoing clinical trials [ 97 ]. Recent data suggest that nuclear PI(4,5)P2 plays a crucial role in organising nuclear processes, including RNA processing, DDR, and regulation of Pol II transcript activity [ 21 , 22 , 83 , 88 , 98 , 99 ]. Moreover, elevated nuclear PI(4,5)P2 levels appear to be linked to cancer progression [ 92 ]. Studies have also suggested that PI(4,5)P2 is a key component of LLPS biocondensates [ 22 , 88 ]. However, given that the nucleoplasm is a membraneless environment, a significant question arises regarding the structural form of PI(4,5)P2 condensates. One possibility is that it forms a lipid structure, such as micelles, to hide its hydrophobic acyl chains from the unfavourable environment. Alternatively, the acyl groups of the PIP may integrate into protein condensate structures, forming hydrophobic cores. Another possible scenario is that PI(4,5)P2 may be covalently linked to certain proteins, as shown by Marx et al. [ 92 ]. Such conjugation is thought to induce a conformational change in the protein, resulting in the acyl chains being shielded from an aqueous environment. This type of conjugation would represent the novel posttranslational modification of unknown metabolic pathway. Finally, similar to SF-1 protein, certain nuclear proteins might possess hydrophobic binding pockets where they accommodate PI(4,5)P2 acyl chains, leaving the head exposed to the external environment [ 100 ]. Therefore, a deeper understanding of the metabolic pathways, structure and functions of nuclear PI(4,5)P2 may provide valuable insights into the mechanisms underlying cancer invasiveness and metastasis. Table 1 Impact of four potential PI(4,5)P2 deregulation patterns in cancer cells on Cancer Progression and Metastasis High cytosol levels– High nuclear levels Low cytosol levels– High nuclear levels: ♦ Enhanced cytoplasmic signalling (promoted PI3K/Akt signalling pathway) > increased cell survival, proliferation, angiogenesis, invasiveness. ♦ Dysregulation of actin polymerisation and cytoskeletal remodelling > enhanced cytoskeletal dynamics, cell migration and adhesion. ♦ Promotion of Factin assembly > enhanced migration. ♦ Enhanced invadopodia dynamics (formation and stability) -> ECM degradation. ♦ Enhanced/alternate transcription of genes involved in DNA damage repair, cell proliferation -> promotion of genome stability and cancer cell survival. ♦ Activation of genes involved in cytoskeletal remodelling metastasis ♦ Attenuated PI3K/Akt signalling > impaired cell migration and invasion. ♦ Altered cytoskeletal remodelling > limited dynamics, reduced actin polymerisation, decreased assembly of Factin. ♦ Accumulation of free Gactin molecules > possibility of translocation to the nucleus. ♦ Impaired invadopodia formation. ♦ Activation of signalling pathways involved in the regulation of transcription and DNA damage repair. ♦ Activation of genes involved in cytoskeletal remodelling > metastasis. High cytosol levels– Low nuclear levels Low cytosol levels– Low nuclear levels ♦ Enhanced cytoplasmic signalling (promoted PI3K/Akt signalling pathway) > increased cell survival, proliferation, angiogenesis, invasiveness. ♦ Dysregulation of actin polymerisation and cytoskeletal remodelling > enhanced cytoskeletal dynamics, cell migration and adhesion. ♦ Promotion of Factin assembly > enhanced migration. ♦ Enhanced invadopodia dynamics (formation and stability) > ECM degradation ♦ Limited/ Attenuated regulation of transcription > altered gene expression profiles. ♦ Decrease nuclear actin levels > impaired chromatin organisation. ♦ Defective DNA repair mechanism > genome instability. ♦ Attenuated PI3K/Akt signalling -> decreased proliferation, cell survival and invasiveness. ♦ Limited cytoskeletal dynamics > reduced motility ♦ Attenuated regulation of transcription -> altered gene expression profiles, decreased invasive potential. ♦ Impaired gene expression. ♦ Decrease nuclear actin levels -> impaired chromatin organisation. Impact of four potential PI(4,5)P2 deregulation patterns in cancer cells on Cancer Progression and Metastasis ♦ Enhanced cytoplasmic signalling (promoted PI3K/Akt signalling pathway) > increased cell survival, proliferation, angiogenesis, invasiveness. ♦ Dysregulation of actin polymerisation and cytoskeletal remodelling > enhanced cytoskeletal dynamics, cell migration and adhesion. ♦ Promotion of Factin assembly > enhanced migration. ♦ Enhanced invadopodia dynamics (formation and stability) -> ECM degradation. ♦ Enhanced/alternate transcription of genes involved in DNA damage repair, cell proliferation -> promotion of genome stability and cancer cell survival. ♦ Activation of genes involved in cytoskeletal remodelling metastasis ♦ Attenuated PI3K/Akt signalling > impaired cell migration and invasion. ♦ Altered cytoskeletal remodelling > limited dynamics, reduced actin polymerisation, decreased assembly of Factin. ♦ Accumulation of free Gactin molecules > possibility of translocation to the nucleus. ♦ Impaired invadopodia formation. ♦ Activation of signalling pathways involved in the regulation of transcription and DNA damage repair. ♦ Activation of genes involved in cytoskeletal remodelling > metastasis. ♦ Enhanced cytoplasmic signalling (promoted PI3K/Akt signalling pathway) > increased cell survival, proliferation, angiogenesis, invasiveness. ♦ Dysregulation of actin polymerisation and cytoskeletal remodelling > enhanced cytoskeletal dynamics, cell migration and adhesion. ♦ Promotion of Factin assembly > enhanced migration. ♦ Enhanced invadopodia dynamics (formation and stability) > ECM degradation ♦ Limited/ Attenuated regulation of transcription > altered gene expression profiles. ♦ Decrease nuclear actin levels > impaired chromatin organisation. ♦ Defective DNA repair mechanism > genome instability. ♦ Attenuated PI3K/Akt signalling -> decreased proliferation, cell survival and invasiveness. ♦ Limited cytoskeletal dynamics > reduced motility ♦ Attenuated regulation of transcription -> altered gene expression profiles, decreased invasive potential. ♦ Impaired gene expression. ♦ Decrease nuclear actin levels -> impaired chromatin organisation. Signals originating from cytosolic PIPs may also influence nuclear events and vice versa. Some PIPs enzyme precursors are known to shuttle between the nucleus and cytoplasm, allowing for a degree of interconnected regulation [ 72 ]. In summary, there are four different patterns in which deregulation of the PIP pathway and its levels can influence cancer progression. The table lists the three with the most likely metastatic ability (high cytoplasmlow nucleus, low cytoplasmhigh nucleus, and high cytoplasmhigh nucleus) and the one with the least invasive ability (low cytoplasmlow nucleus), as it is most likely to contribute to cancer cell death (Table  1 ). Deciphering the detailed mechanisms of this complex and precise system will not only enhance our understanding of PIPs metabolism in cancers but may also uncover novel therapeutic targets for malignancies characterised by aberrant nuclear lipid signalling.

Cancer is a leading cause of death worldwide [ 1 ]. In most cases, the increased invasiveness of cancer cells is responsible for their lethal potential. Metastasis is defined as the spread of cancer cells from their primary site to neighbouring tissues and other organs via blood vessels and the lymphatic system. This process can be divided into five sequential stages. Invasion, a crucial step in the progression of metastasis, followed by intravasation, survival in circulation, extravasation, and colonisation [ 2 ]. A deeper understanding of the molecular mechanisms underlying this metastatic cascade will lead to the development of effective therapies and improved patient prognosis. Metastatic progression is associated with altered composition and dynamics of the plasma membrane (PM) and cell nucleus. Given the complex nature of the PM, consisting of a lipid bilayer and various proteins, these alterations can occur as changes in membrane rigidity, increased lipid raft formation, or differences in PM signalling [ 3 ], as well as alterations in actin polymerisation and depolymerisation, which contribute to enhanced cellular invasion [ 4 ]. Lipid rafts are highly ordered microdomains enriched in sphingomyelin and cholesterol [ 5 ], serving as key platforms for physiological processes, including signalling, migration, trafficking, cellcell communication, and cellmatrix interaction [ 6 ]. Although the enrichment of sphingomyelin and cholesterol mainly characterises lipid rafts, significant amounts of phosphoinositides (PIPs) are also present [ 7 , 8 ]. Among these, phosphatidylinositol 4,5bisphosphate (PI(4,5)P2), is a crucial regulator of Factin dynamics, significantly influencing cell dynamics and invasive potential [ 9 ]. Disruption of lipid rafts within the PM, responsible for organising specific proteins and regulating lipid microdomains, can modulate membraneassociated signalling pathways, such as the phosphoinositide 3kinase/protein kinase B (PI3K/Akt) pathway [ 3 , 10 ]. This, in turn, affects the migratory and invasive behaviour of cancer cells by altering actin remodelling, promoting invadopodia formation, and enhancing their capacity to degrade the extracellular matrix (ECM). Ultimately, these changes increase the metastatic potential of cancer cells [ 4 ]. In addition, nuclear dynamics affect cancer metastasis by altering the shape and size of the nucleus and the structure of the nuclear envelope [ 11 , 12 ]. Studies have demonstrated that nuclear morphological changes are prevalent in cancer cells and may influence their invasiveness. For instance, cancer cells frequently exhibit enlarged and irregularly shaped nuclei, which have been linked to an elevated capacity for metastasis in a multitude of distinct cancer types [ 11 , 12 ]. Moreover, dysregulation of nuclear signalling pathways, including those mediated by transcription factors, chromatin modifiers, and DNA damage response proteins, can significantly affect gene expression patterns involved in cancer metastasis. Transcription factors play a critical role in the regulation of gene expression associated with cell motility, invasion, and metastasis [ 2 ]. Dysregulated expression or activity of these transcription factors can promote the expression of prometastatic genes while inhibiting the expression of genes involved in tumour suppression and cell adhesion [ 2 ]. Furthermore, alterations in chromatin structure and epigenetic modifications can also contribute to cancer metastasis by affecting gene expression patterns [ 13 ]. Epigenetic modifications, such as DNA methylation or histone acetylation, have been demonstrated to regulate the expression of genes involved in metastatic progression. For instance, aberrant DNA methylation patterns at the promoter regions of tumour suppressor genes can result in their silencing, thereby promoting metastasis [ 14 ]. Nuclear phosphoinositides have emerged as essential regulators of processes involved in metastasis and invasiveness of cancer cells [ 15 – 17 ]. These lipid molecules not only participate in signalling cascades that control gene expression and cellular responses to extracellular signals but also contribute to the regulation of nuclear envelope dynamics and chromatin organisation [ 18 – 20 ]. Recent studies have underscored the importance of nuclear phosphoinositides in modulating the activities of transcription factors and chromatin modifiers involved in metastatic progression [ 21 , 22 ]. In this review, we focus on PI(4,5)P2, the most abundant phosphoinositide and a key regulator of cell invasion. We highlight the potential implications of the interplay between cytoplasmic and nuclear PI(4,5)P2 pathways in driving cancer invasiveness.