The role of inner nuclear membrane proteins in tumourigenesis and as potential targets for cancer therapy

In addition to their roles in establishing and maintaining normal cellular structure, the nuclear envelope and associated proteins have been shown to have roles in tumour cell migration and invasion. Cellular migration and invasion are two essential factors in the metastatic cascade, where tumours have the capacity to spread from their primary site [38]. Understanding the proteins which have a key role in these pathways will not only improve our knowledge of tumour metastatic pathways, but also allow for the exploration of anti-cancer therapies which minimise the capacity for tumours to metastasise.

Recent research has demonstrated that alterations in nuclear conformation and subcellular localisation are vital for effective migration to occur. More specifically, nuclear movement secondary to protrusion of the cell’s edge by actin filaments is essential for cellular migration [38]. Several nuclear envelope proteins have been shown to have a vital role in maintaining the appropriate cytoskeletal actin structure and nuclear movement to promote cellular migration [39, 40]. It has been demonstrated that Emerin-mediated nuclear/myosin IIB coupling is essential to achieve directional flow of actin filaments during cellular migration [39]. Similarly, A-type Lamins have been shown to anchor nesprin-2G-SUN2 TAN lines to the nucleus to ensure appropriate nuclear positioning for migratory pathways [40]. Furthermore, depletion of these proteins by siRNA has been shown to impair this subcellular movement to impair cellular migration [39, 40]. Thereby, inhibition of these proteins may provide a favourable anti-cancer therapy by diminishing the capacity of tumour cells to enter the appropriate morphology for proficient migration, inevitably reducing the metastatic capacity of these tumours.

Movement of the nucleus is often considered the rate-limiting step during cellular migration. This is predominately due to the observation that cytoplasmic components of the cell can squeeze through minute spaces; however, the nucleus has significantly more limited flexibility [41]. Therefore, it is unsurprising that there are a substantial number of interphase nuclear rupture events during cellular migration due to the extensive mechanical force placed on the nucleus [42, 43]. Furthermore, Banf1 and the A-type Lamins have been shown to have an essential role in maintaining cellular integrity during migration with pressure required to induce nuclear ruptures being significantly less following depletion of Banf1 or the A-type Lamins [44].

Recently, substantial research has been conducted to further understand the mechanism by which cells repair these mechanical-stress induced ruptures. The key nuclear envelope protein, Banf1, has been shown to be essential for the rapid repair of these ruptures [42]. Immediately following rupture, cytosolic Banf1 rapidly relocalises to the rupture site due to Banf1’s affinity for double-stranded DNA [45]. Subsequently, this promotes the recruitment of other nuclear envelope repair proteins, including ESCRT III and the Lem domain protein, Lemd2, to the site of rupture [44]. Hence, inhibition or loss of Banf1 impairs nuclear envelope rupture repair; therefore, inducing catastrophic damage and subsequent cell death in migrating cells. Secondary to this, Banf1’s ability to bind dsDNA outcompetes that of cyclic guanosine monophosphate–adenosine monophosphate synthase (cGAS). cGAS is a double-stranded DNA (dsDNA) sensor, usually localised to the cytoplasm, essential for activation of stimulator of interferon genes (STING). Therefore, if cells were unable to protect their dsDNA from cGAS in the event of a nuclear rupture, this would likely result in a catastrophic event and subsequent cell death [46, 47].

Lamin A/C has an essential role in upholding nuclear stability whilst maintaining some level of flexibility to minimise catastrophe following the mechanical stress of cellular migration [48]. Supporting this, one study demonstrated that depletion of Lamin A/C decreased prostate cancer cell colony formation, migration and invasion due to upregulation of the PI3K/AKT cell survival pathway [49]. Inversely, over-expression of Lamin A was shown to promote invasion and migration in a colorectal cancer model due to the upregulation of T-plastin and subsequent downregulation of E-cadherin [50]. In contrast, another study found that expression of Lamin A/C has been shown to be negatively correlated with the probability of developing distant metastatic lesions in breast cancer. Specifically, depletion of Lamin A was shown to enhance confined cell migration capacity in breast cancer models by creating highly deformable nuclei [24]. These opposing findings suggest that Lamin A/C expression may have a multifaceted role in tumour progression, which is likely mediated by external factors.

Collectively, these findings suggest that targeting various nuclear envelope proteins may impair tumour cell migration, thereby decreasing the risk of metastatic sites by a magnitude of mechanisms. The targeting of nuclear envelope proteins may have several multifaceted effects on tumour cell migration including inhibiting subcellular re-localisation of the nucleus to prevent the initiation of migration, preventing the repair of migration induced nuclear envelope ruptures and disrupting the precise regulation of nuclear stability to promote nuclear envelope ruptures which subsequently cannot be repaired leading to tumour cell death.

Aberrant cell cycle activity is characteristic amongst tumours; therefore, cell cycle inhibitors have shown promise in the development of anti-cancer therapeutics. However, the current clinically available cell cycle kinase inhibitors have substantial drawbacks. These include unsustainable neurological and bone marrow off target toxicity, a high prevalence of acquired resistance and poor improvements in patient outcomes (reviewed in [51]). The nuclear envelope is well recognised to have an essential role in allowing for changes in nuclear shape and size throughout the cell cycle. Therefore, it is possible that down regulating the expression or activity of key nuclear envelope proteins, may act to inhibit cell cycle progression and subsequently to suppress tumour cell growth. Furthermore, targeting these nuclear envelope proteins may have less off-target effects than conventional cell cycle therapies given their altered expression in tumours.

Recently, a novel role was revealed for the Lem domain protein, Ankle2, in breast cancer tumourigenesis. They demonstrated that Ankle2 is overexpressed in breast cancer, and this overexpression facilitated breast cancer proliferation and migration. This effect on proliferation was suggested to be through promotion of the transition through the G1 to S phase of the cell cycle via the stabilisation of Erα, induction of Erα transactivation activity and promotion of Aurora A-mediated phosphorylation of Erα [17]. Based on these observations, it can be hypothesised that compounds which supress Ankle2 levels or activity may provide favourable anti-tumour activity like that of a traditional cell cycle checkpoint inhibitor. Although, given Ankle2 is substantially overexpressed in tumour cells, it is possible that specifically targeting Ankle2 may have substantially less off-target toxicity than conventional therapies.

In addition, in mouse embryotic fibroblast models, Sun1 and Sun2 null mice demonstrated significant anti-proliferative effects due to the accumulation of cells within the S phase of the cell cycle [52]. Therefore, this suggests that targeting the expression or activity of these proteins may have positive implications for tumour patients. However, the implications of this for a tumorigenic model remain to be investigated.

Similarly, several nuclear envelope proteins have been shown to have a role in the mitotic phase of the cell cycle [36, 53‐55]. This is largely because nuclear envelope breakdown and reformation is essential for mitosis to occur [56]. The exploitation of compounds which disrupt mitosis as a cancer therapy has been of scientific interest for many years, with the intent of directly inhibiting the production of new tumour cells. However, most current anti-mitotic therapeutics, such as paclitaxel and docetaxel, target microtubule polymerisation dynamics and induce substantial off-target toxicity due to their inability to distinguish between tumour cells and other highly proliferative non-tumorigenic cell types [57]. Although there is yet to be a cancer therapeutic developed, which targets the involvement of nuclear envelope proteins in mitosis, recent research has demonstrated that this may be a viable target. For instance, it has also been shown that the phosphorylation state of Banf1 is key in the regulation of cell division [36]. During mitotic entry, VRK1 phosphorylates Banf1 resulting in Banf1 being unable to maintain binding with Lem domain proteins and DNA. Thereby, this promotes the dissociation of chromosomes from the nuclear envelope during prophase. Likewise, at the termination of cell division, the nuclear envelope localised Lem domain protein, Lem4, binds to VRK1 to promote the dephosphorylation of Banf1 by PP2A allowing for the reformation of the nuclear envelope [9, 16]. Thereby, either directly or indirectly suppressing Banf1’s role in nuclear envelope breakdown and reformation may provide a novel anti-mitotic drug without inducing the off-target effects associated with the currently available therapeutics.

Similarly, in HeLa cervical cancer cells, the Lem domain protein, Emerin, was shown to relocalise to the centrosomes and microtubules during mitosis [58]. In addition, deletion of various regions of the gene encoding Emerin induced several mitosis deficient phenotypes, including tubulin network and centromere mislocalisation and prolonged time in mitosis, suggesting that Emerin is required for appropriate mitotic functioning [59]. Thereby, targeting Emerin’s role in mitosis as an anti-cancer therapy may have similar activity to that of targeting Banf1.

As previously discussed, Lamin A/C expression is frequently downregulated in tumour cells [50, 60]. In addition to having a role in tumour cell migration and invasion, the downregulation of Lamin A/C has also been shown to induce aberrations in mitosis [48, 60]. More specifically, siRNA-mediated depletion of Lamin A/C was shown to induce nuclear budding, tripolar cell division, aneuploid cells and micronuclei in ovarian carcinoma cells [60]. Taken together, these studies also implicate targeting Lamin A/C as a potential anti-cancer strategy.

Dysregulation of cell signalling cascades is well recognised to be a shared characteristic amongst cancers. It is well recognised that several nuclear envelope proteins are involved in cell signalling pathways. For example, the MAPK/ERK pathway is a complex cellular pathway with extensive involvement in cellular proliferation, apoptosis, and differentiation [61]. Dysregulated and hyper-activation of MAPK/ERK signalling has shown to promote tumourigenesis via several mechanisms (reviewed in [62]). The role of several nuclear envelope proteins in maintaining appropriate MAPK/ERK signalling has been established. For instance, it has been shown that siRNA-mediated depletion of Lamin A/C or Emerin in cervical cancer cells results in increased ERK phosphorylation and localisation to the nucleus, subsequently enabling activation of downstream transcription factors [63, 64].

Another instance of nuclear envelope proteins being involved in cell signalling is the role of the Lem domain protein, MAN1, in the transforming growth factor-β (TGF-β) signalling pathway [65]. The TGF-β pathway has significant involvement in many cellular pathways including proliferation, differentiation, adhesion and cell cycle regulation. Like the MAPK/ERK pathway, TGF-β is frequently dysregulated in cancer cells [66]. MAN1 has been shown to bind two intracellular mediators of TGF-β signalling, Smad2 and Smad3, to suppress their ability to stimulate TGF-β activity. Subsequent studies showed that MAN1 overexpression inhibits the nuclear translation of Smad2 and Smad3, consequently, inhibiting cellular proliferation [65, 67]. Despite these findings, it remains to be determined if MAN1 can be targeted as an anti-cancer therapy. The transmembrane inner nuclear membrane protein, TMEM201, has also been shown to have a role in regulation of TGF-β and Smad2/3, suggesting that multiple inner nuclear membrane proteins may regulate this pathway [68].

Upregulation of the Wnt signalling pathway has been well established to be a shared characteristic amongst many tumour types, resulting in favourable conditions for the unregulated proliferation of tumour cells [69, 70]. Emerin, an inner nuclear membrane protein, has been shown to downregulate the activity β-catenin by inhibiting its nuclear accumulation via the upregulation of β-catenin export to the cytoplasm. The accumulation of β-catenin in the nucleus inevitably enables the activation of transcription factors, thereby promoting Wnt signalling [71]. Emerin has been shown to be dynamically expressed in a wide variety of tumours, with its expression being significantly decreased in highly metastatic tumours [72]. Therefore, it is possible that the downregulation of Emerin increases Wnt signalling by minimising nuclear export of β-catenin, thereby promoting cellular migration and invasion [72].

In addition, LAP2β has been shown to be overexpressed in a variety of tumours including gastric, stomach, breast, and lung cancers [73, 74]. Like other nuclear envelope proteins, LAP2β has been shown to participate in several cell signalling pathways, including suppression of the transcriptional activity of various transcription factors. The role of p53 as a transcription factor in tumourigenesis is well documented, with p53 having an essential role in the cellular stress response to promote DNA repair, cell cycle arrest, cellular senescence and apoptotic mechanisms (reviewed in [75]). Of interest, LAP2β expression has been shown to inversely correlate with p53 transcriptional activity [76]. Thereby, the characteristic overexpression of LAP2β in tumour cells is suggested to have a role in supressing the anti-tumour activity of p53. However, it is arguable that that this may not be of therapeutic significance as greater than 50% of solid tumours have p53 mutations.

Finally, DNA repair defects are characteristic amongst a large subset of tumours. Several nuclear envelope proteins have been shown to participate in the cellular response to DNA damage, including Lamin A/C, Lemd2 and Banf1 [77‐79].

As previously discussed, Lamin A/C expression is frequently altered in tumour cells. Furthermore, Lamin A/C has been shown to be involved in numerous DNA damage response pathways, with its central role in DSB repair being the most studied. Briefly, HR is an error-free and template driven DNA double strand break repair method, that is limited to the S and G2 phases of the cell cycle. Whereas NHEJ can occur throughout all phases of the cell cycle as it does not rely on a sister chromatid as a template (reviewed in [80]). However, the lack of need for a DNA template in NHEJ does increase the likelihood of nucleotide insertions, deletions and translocation, driving genomic instability [81]. For instance, Lamin A/C depleted U-2OS cells show impaired ability to undergo APE1-mediated DNA incision and POLβ-mediated DNA polymerisation during base excision repair of oxidised DNA [82]. Furthermore, Lamin A/C has been shown to have a role in DSB repair in mouse models. Specifically, knockout of Lamin A/C significantly decreased 53BP1 expression at 1-h post irradiation, with this being predominately due to the upregulation of nuclear and lysosomal Cathepsin L activity [83]. This increased Cathepsin L activity results in the degradation of 53BP1, in addition to pRb and p107. Given 53BP1 has a key role in the facilitation of non-homologous end joining (NHEJ), Lamin A/C-deficient cells are unable to effectively complete NHEJ. In addition, the collective decrease in pRb and p107 expression allows for the binding of the p130/E2FA protein complex, which directly inhibits BRCA1 and RAD51 function. Consequently, the simultaneous BRCA1 and RAD51 downregulation inhibits homologous recombination repair [83]. Based on these findings, it is possible that targeting Lamin A/C overexpression in certain tumour subtypes may impair the cellular DNA damage response; therefore, conferring sensitivity to anti-cancer therapies which induce DNA damage.

In addition, findings from our team have identified a key role for Banf1 in the repair of DNA damage induced by oxidative stress. Initially, it was observed that Banf1 relocalised from the nuclear envelope to the nucleus following treatment with the oxidative stress inducing agents, H₂O₂ and Potassium Bromate. Subsequently, it was shown that Banf1 directly bound to PARP1 following oxidative stress to inhibit the auto-ADP-ribosylation of PARP1 and PARP1’s histone ADP-ribosylation, thereby impairing the repair of oxidative lesions [77]. In addition, it has also been observed that siRNA-mediated depletion of Banf1 significantly decreases HR activity and upregulates NHEJ due to Banf1’s role in the regulation of DNA-PK activity [78].

More recently, a novel role for the Lem domain protein, Lemd2, in nucleotide excision repair has been revealed. Nucleotide excision repair is the most frequently utilised pathway to repair bulky, helix distorting DNA lesions induced by ultraviolet light and other environmental mutagens. Although the underlying mechanism remains to be understood, it has been shown that siRNA-mediated depletion of Lemd2 enhances the anti-proliferative effect of ultraviolet-C irradiation. Lemd2-depleted cells also displayed increased γ-H2AX foci at 48 h after UV-C exposure suggesting that Lemd2 may have a role in promoting repair of these lesions [79].

The nuclear pore complexes (NPC), spanning the nuclear periphery, have also been implicated to have a role in the repair of a subset of DNA double-strand breaks [84]. Initially, these protein complexes were shown to induce sensitivity to DNA damaging agents in yeast deletion mutants, suggesting a potential role in DNA repair [64, 65]. Subsequently, it was shown that a subset of HO-induced double-strand breaks moved to the edge of the nucleus and were localised with NPCs on the nuclear periphery [22, 85]. This relocalisation of DNA double-strand breaks was later found to be dependent upon the nucleoporins NUP107, NUP153 and NUP160 in Drosophila cells [86]. While the relocalisation of double-strand breaks has been observed in several cell systems, it should be noted that only a small number of these breaks are relocalised, so it remains to be elucidated whether targeting these proteins would have a substantial effect on DNA repair in tumour cells.

In light of the evidence discussed above, when considering the potential for cancer therapies that target the nuclear envelope it is worth considering that due to the roles of some nuclear envelope proteins in DNA repair, these therapies may prove effective in combination with current DNA repair focused therapeutics, such as PARP inhibitors [87].

In summary, this demonstrates that the nuclear envelope is involved in extensive list of cellular processes, the majority of which are frequently dysregulated in cancer cells. Therefore, nuclear envelope proteins may provide a novel class of targeted anti-cancer therapies by selectively disrupting the nuclear envelope in tumour cells.