The Nuclear Envelope Protein, LAP1B, Is a Novel Protein Phosphatase 1 Substrate

Protein phosphatase 1 (PP1) binding proteins are quintessential regulators, determining substrate specificity and defining subcellular localization and activity of the latter. Here, we describe a novel PP1 binding protein, the nuclear membrane protein lamina associated polypeptide 1B (LAP1B), which interacts with the DYT1 dystonia protein torsinA. The PP1 binding domain in LAP1B was here identified as the REVRF motif at amino acids 55-59. The LAP1B:PP1 complex can be immunoprecipitated from cells in culture and rat cortex and the complex was further validated by yeast co-transformations and blot overlay assays. PP1, which is enriched in the nucleus, binds to the N-terminal nuclear domain of LAP1B, as shown by immunocolocalization and domain specific binding studies. PP1 dephosphorylates LAP1B, confirming the physiological relevance of this interaction. These findings place PP1 at a key position to participate in the pathogenesis of DYT1 dystonia and related nuclear envelope-based diseases.


Introduction
Reversible protein phosphorylation is a major mechanism controlling key intracellular events that are essential for cell health and viability [1][2][3]. Protein phosphatase 1 (PP1) is a ubiquitous serine/threonine phosphatase that is estimated to dephosphorylate about one third of all proteins in eukaryotic cells [4,5]. PP1 regulates a variety of cellular functions, such as glycogen metabolism, transcription, protein synthesis, cellular division and meiosis [4,6,7]. In mammalian cells, three genes encode the three PP1 isoforms: PP1alpha (PP1α), PP1Beta/ Delta (PP1β/δ) and PP1gamma (PP1γ). Furthermore, the PP1γ gene undergoes alternative splicing to originate a ubiquitous PP1γ1 variant and a PP1γ2 variant that is enriched in testis [8,9]. PP1 isoforms are expressed in virtually all tissues but exhibit different expression levels depending on the tissue, and different subcellular distribution [9][10][11]. The versatility of PP1 is largely determined by the binding of its catalytic subunit to different specific regulatory subunits, that are responsible for the directed targeting of PP1 to a particular subcellular compartment and also determine its substrate specificity and activity [4,6,7,12]. More than 200 binding/regulatory subunits have been already described, making PP1 an essential protein in many distinct cellular processes [2,13]. Most PP1 binding proteins interact with the PP1 catalytic subunit through a conserved PP1 binding motif termed the RVxF motif, which has the consensus sequence [R/K] X A(0-1) [V/I] X B [F/W], where X A is any amino acid and X B is any amino acid except proline [14]. More [16]. The binding of PP1 to regulatory proteins through the RVxF motif does not cause major effects on the conformation and activity of PP1, but mediates the initial anchoring of regulatory subunits and thereby promotes the interaction at secondary binding sites [7,16,17]. Subsequently other PP1 binding motifs have also been reported, such as the apoptotic signature F-X-X-[KR]-X-[KR] [18], the SILK and the MyPhoNE motifs [16]. The existence of these conserved binding sites within regulatory subunits explains the ability of PP1 catalytic subunit to interact with numerous regulatory proteins and consequently the binding of most regulatory subunits is mutually exclusive [7]. Since the regulatory subunits control the specificity and the diversity of PP1 activity, the key to understanding PP1 function lies in studying these regulatory subunits and their cellular functions. Several yeast two-hybrid (YTH) screens of a human brain cDNA library were performed using the three PP1 isoforms (PP1α, PP1γ1 and PP1γ2) as baits. From each screen many positive clones were identified encoding several different proteins. Among them, were proteins already known as PP1 regulators and also novel putative PP1 regulators [19,20]. The majority of the novel putative PP1 regulators comprise at least one of the conserved PP1 binding motifs. One of the novel PP1 regulators isolated in the three independent YTH screens was torsinA interacting protein 1 (TOR1AIP1) or lamina associated polypeptide 1B (LAP1B). LAP1B belongs to a family of integral proteins of the inner nuclear membrane, named lamina associated polypeptide 1 (LAP1). Members of the LAP1 family (LAP1A, B and C) were initially identified using monoclonal antibodies generated against lamina-enriched fractions of rat liver nuclei [21]. The three LAP1 isoforms result from alternative splicing of the TOR1AIP1 gene [22] and have been poorly studied. Moreover, the cDNA for LAP1B isoform was, to date, the only human isoform completely sequenced [23]. The function of LAP1B is poorly understood but it is known that it binds to lamins and chromosomes and that it is phosphorylated during interphase and mitosis [24]. As indicated by the nomenclature, LAP1 was found to interact with torsinA [25,26], the central protein of a neurologic disorder known as DYT1 dystonia [27]. However, the physiological relevance of this novel complex has not yet been determined. In the present study we validated the novel complex LAP1B:PP1 using several in vitro and in vivo techniques, namely a blot overlay assay, coimmunoprecipitation (co-IP) and yeast co-transformation. Furthermore, the PP1 binding motif responsible for the interaction was mapped. The functional relevance of this complex was pursued and we determined that LAP1B is dephosphorylated by PP1 in vitro.
Additionally a LAP1B (ΔA185) mutation was introduced into pET-LAP1B by site-directed mutagenesis using the primers described in Table S1. The constructs were all verified by DNA sequencing using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Porto, Portugal).

Expression of recombinant proteins in Escherichia coli
Escherichia coli Rosetta cells (DE3) were transformed with one of the pET-LAP1B constructs and grown overnight in 5 mL of Luria-Bertani/Kanamycin medium at 37°C. Aliquots (500 µL) were transferred to 50 ml of Luria-Bertanini/Kanamycin until the OD 600 was around 0.5-0.6. Expression was induced by adding 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) to the culture at 37°C for different periods of time (1, 3 and 5 hours) with shaking. Cells were recovered by centrifugation (4000 rpm for 10 min), resuspended in 500 µL of 1x PBS and lysed by sonication. Cells were then centrifuged (13200 rpm for 30 min), the supernatant was transferred to a new microtube (soluble fraction) and the pellet resuspended in 500 µL of boiling 1% SDS (insoluble fraction).

Blot overlay assays
LAP1B deletion mutants. Bacterial extracts, prepared as described above, were separated on a 10% SDS-PAGE and LAP1B Is a Novel PP1 Regulatory Protein PLOS ONE | www.plosone.org the proteins were subsequently transferred to a nitrocellulose membrane. The membrane was overlaid with 1 µg/mL of purified recombinant PP1γ1 protein [28]. After washing to remove excess protein, bound PP1γ1 was detected by incubating the membrane with PP1γ antibody and developed by enhanced chemiluminescence (ECL, GE Healthcare).
LAP1B isoforms. LAP1B and LAP1B (ΔA185) proteins were generated by in vitro transcription/ translation (IVT) from pET-LAP1B and pET-LAP1B(ΔA185) expression vectors, respectively, using the TnT-coupled transcription/translation kit (Promega), according to the manufactures' instructions. For the overlay assays, two samples of 250 ng of purified recombinant PP1γ1 protein [28] were separated on a 12% SDS-PAGE. Both proteins were transferred to a nitrocellulose membrane but one was overlaid with LAP1B-IVT, while the other was overlaid with LAP1B (ΔA185)-IVT. The bound proteins were detected by incubating the membrane with LAP1 antibody and developed by ECL.

Brain dissection
Winstar rats (9-12 weeks) were obtained from Harlan Interfaune Ibérica, SL. All experimental procedures observed the European legislation for animal experimentation (2010/63/ EU). No specific ethics approval under EU guidelines was required for this project, since the rats were only euthanized, by cervical stretching followed by decapitation, for brain removal. This is within the European law (Council Directive 86/609/EEC) and during this procedure we took all steps to ameliorate animal suffering and used the minimum number of animals possible. The procedures were approved and supervised by our Institutional Animal Care and Use Committee (IACUC): Comissão Responsável pela Experimentação e Bem-Estar Animal (CREBEA).

Co-immunoprecipitation
COS-7 cells transfected with Myc-LAP1B were collected in lysis buffer (50 mM Tris-HCl pH 8, 120 mM NaCl, 4% CHAPS) containing protease inhibitors (1 mM PMSF, 10 mM Benzamidine, 2 µM Leupeptin, 1.5 µM Aprotinin, 5 µM Pepstatin A). Dynabeads Protein G (Dynal, Invitrogen) were washed in 3% BSA/1x PBS. Primary antibodies were crosslinked to Dynabeads according to the manufacter's protocol. Cell lysates were precleared with 20 µL (0.6 mg) Dynabeads for 1 hour and then incubated with antibody-dynabeads for 2h at 4°C. The immunoprecipitates were washed in 1x PBS and proteins eluted by boiling in loading buffer before SDS-PAGE and immunoblotting analysis. For analysis of endogenous proteins, COS-7 cells and SH-SY5Y cells were collected in lysis buffer and immunoprecipitated as described above, but cell lysates were incubated with antibody-dynabeads overnight at 4°C. HEK293 immunoprecipitation was carried out as previously described [30] Immunocytochemistry Once fixed as described above, HeLa cells were permeabilized with methanol for 2 min. Cells were first incubated with one of the primary antibodies (anti-PP1γ or PP1α) for 2 hours, followed by Alexa 594-conjugated secondary antibody. After washing with 1x PBS, cells were subsequently incubated with a second primary antibody (antimyc) for 2 hours, followed by anti-mouse fluorescein isothiocyanate (FITC) conjugated secondary antibody. Preparations were washed with PBS, mounted using Vectashield mounting media with DAPI (Vector) and visualized using an LSM510-Meta confocal microscope (Zeiss) and a 63x/1.4 oil immersion objective. The argon laser lines of 405 nm, 488 nm, and a 561nm DPSS laser were used. Microphotographs were acquired in a sole section in the Z-axis (xy mode) and represent a mean of 16 scans. Profiles were acquired using the Zeiss LSM 510 4.0 software as previously described [29,31].

SDS-PAGE and Immunoblotting
Samples were separated on SDS-PAGE and electrophoretically transferred onto nitrocellulose, followed by immunological detection with specific antibodies as indicated. Membranes were saturated in 5% non-fat dry milk in TBS-T for 3 hours and further incubated with primary antibodies. The incubations with the CBC2C, CBC3C and LAP1 antibodies were performed overnight. The incubations with the His-tag and Myc-tag antibodies were carried out for 2 hours. Detection was achieved using horseradish peroxidase-conjugated antirabbit or anti-mouse IgGs as secondary antibodies and proteins visualized by ECL (GE Healthcare).

Identification of LAP1B as a novel putative PP1 regulatory protein
In order to identify novel potential PP1 regulatory proteins, the three PP1 isoforms (PP1α, PP1γ1 and PP1γ2) were used as baits to screen a human brain cDNA library by the YTH system. From these screens 298, 241 and 228 positive clones were identified using PP1α, PP1γ1 and PP1γ2 as baits, respectively [19,20]. Detailed analysis revealed 14, 4 and 2 positive clones, encoding a potentially novel PP1 regulatory protein -LAP1B, for the PP1α, PP1γ1 and PP1γ2 screens, respectively (Table S2) [32]. Sequencing of the identified clones determined that all correspond to the LAP1B variant 1, recently reported in the GenBank database (NM_001267578.1). The variant 1 differs from the variant 2 (NM_015602) only by a CAG insertion, which results in an additional alanine in the coding sequence, otherwise the sequence is identical to the first human LAP1B sequence reported in 2002 [23]. Moreover, others reports showed that TOR1AIP1 gene possesses a 3' tandem splice site, TAGCAG, at the exon 3 boundary, which results in an amino acid insertion or deletion in the encoded protein [33,34]. From all LAP1B clones obtained we selected a full-length clone (clone 135; Table S2) for further studies. The latter comprises a short 5' untranslated region followed by the ATG start codon, an open reading frame of 2000 nucleotides, that encodes 584 amino acids, followed by a stop codon and a short 3' untranslated region ( Figure 1A). LAP1B was previously described as a type 2 integral membrane protein, with an Nterminal nucleoplasmic domain, one predicted transmembrane domain (TM) and a C-terminal lumenal domain [23]. Our in silico analysis also revealed that LAP1B has three conserved PP1 binding motifs (BM): REVRF (amino acids 55-59), KVNF (amino acids 212-215) and KVKF (amino acids 538-541), that were called BM1, BM2 and BM3, respectively. BM1 and BM2 are localized in the nucleoplasm, while BM3 is localized in the lumen of the perinuclear space. Additionally, a second generic PP1 binding motif termed SILK (amino acids 306-309) was also identified ( Figure 1A). Further in silico characterization of these four potential PP1 binding motifs (BM1, BM2, BM3 and SILK) was achieved using the ClustalW algorithm, which allowed for the determination of the homology between species ( Figure  1B). Interestingly, the BM1 and BM3 are totally conserved motifs among the species analyzed (Human, Chimpanzee, Orangutan, Mouse, Rat). The other binding motifs (BM2 and SILK) are also completely conserved between Human, Chimpanzee and Orangutan but do deviate markedly in mouse and rat; with two residues common to all species in the SILK domain and only one residue is common to all species for the BM2 domain ( Figure 1B). Subsequent to the in silico LAP1B characterization, the novel LAP1B:PP1 complex was validated in vitro and in vivo.

LAP1B and PP1 interact in vitro
The in vitro validation of this novel complex as well as the identification of the binding motif responsible for the interaction was accomplished using a blot overlay assay. Several deletion mutants were prepared as described above, each comprising at least one of the PP1 binding motifs as represented in Figure  2A. The importance of the TM was also addressed by evaluating the binding in the presence or absence of the latter (Figure 2A). Thus, both full-length LAP1B and the different deletion mutants were expressed in bacteria as 6×His-tag fusion proteins by adding 1 mM IPTG to the growing cultures at 37°C. All recombinant proteins (full-length and mutants) were efficiently expressed in bacteria, including the deletion mutants with or without the TM domain (data not shown). The proteins thus expressed were used to carry out the blot overlay assay. Briefly, the different recombinant proteins were separated by SDS-PAGE and electrotransferred onto nitrocellulose membranes. Recombinant protein detection was achieved with the His-tag antibody ( Figure 2B) and for the overlay assay the purified PP1γ1 protein was applied ( Figure 2C). The negative controls used included the pET-28c vector without an insert or with a non-induced insert. As a positive control the pET28c-Nek2A was used, given that Nek2A is a well known protein that strongly interacts with PP1 [35]. The immunoblotting using Histag antibody ( Figure 2B) of both full-length and deletion mutants revealed that all the expressed proteins have the expected molecular weight as indicated in Figure 2A. Additional bands were sometimes detected for LAP1B, LAP1B-BM1 and LAP1B-BM2, these could possibly correspond to proteolytic fragments ( Figure 2B). The overlay assays revealed increased PP1γ1 binding concomitant with increasing amounts of recombinant full-length LAP1B ( Figure 2C). Additionally, we also demonstrated that PP1γ1 binds to the deletion mutant that comprises the residues 1-209 encompassing the BM1, and also to the construct that comprises both BM1 and BM2 with or without the TM ( Figure 2C). However, PP1γ1 does not bind to the recombinant protein comprising only the BM2 or the BM3 ( Figure 2C). Thus, using in vitro techniques we established that LAP1B binds to PP1γ1 through the BM1 (REVRF).

The novel complex LAP1B:PP1 is also formed in vivo
Having shown that the LAP1B:PP1 complex is formed in vitro the ability of the same to be formed in vivo was addressed. Yeast co-transformations were carried out in order to validate the novel complex formation and to simultaneously confirm relevant domains in LAP1B for the interaction, as well as to test different PP1 isoforms. Firstly, co-transformations were carried out with full-length LAP1B and different PP1 isoforms (α, γ1 or γ2), the assay took advantage of α-Galactosidase activity in the LAP1B Is a Novel PP1 Regulatory Protein PLOS ONE | www.plosone.org   presence of the blue chromogenic substrate X-α-gal, producing a blue colour in the event of an interaction. The results ( Figure  3) clearly show that LAP1B interacts with all PP1 isoforms tested (PP1α, PP1γ1 and PP1γ2; Figure 3A) but does not interact with the C-terminal portion of PP1γ2 isoform (PP1γ2end; Figure 3A). Subsequently, the relevant domain in LAP1B for the interaction was again tested using the in vivo conditions. These LAP1B deletion mutants, previously produced, were co-transformed in yeast with the PP1γ1 isoform and the results are presented in Figure 3B. Unequivocally, the results were positive for the deletion mutant that comprises the residues 1-209 including the BM1 ( Figure  3B) and also with the construct that comprises both BM1 and BM2 (BM1/2). In sharp contrast the results were negative with the LAP1B-BM2 and LAP1B-BM3 ( Figure 3B). The positive and negative controls are presented in Figure 3C. These results show that it is the BM1 (REVRF) that mediates the interaction between LAP1B and PP1.
Additionally, the in vivo occurrence of the complex was further investigated and distinct models were used, namely a non-neuronal cell line (COS-7), a neuronal-like cell line (SH-SY5Y) and rat brain. Thus, co-IPs using the specific antibodies against PP1α, PP1γ and LAP1B were carried out. Firstly, COS-7 cells were transfected with Myc-LAP1B and Myc-LAP1B deletion mutants (LAP1B-BM2 and LAP1B-BM3) and further immunoprecipitated using the PP1γ antibody ( Figure  4A). The Myc-LAP1B deletion mutants are similar to those tested in yeast co-transformation assays. Our results showed that PP1γ binds only to the full-length LAP1B but not to LAP1B deletion mutants (LAP1B-BM2 and LAP1B-BM3). Once more our results clearly showed that PP1γ only binds LAP1B when the BM1 is present. Non-transfected COS-7 ( Figure 4B) and SH-SY5Y ( Figure 4C) cellular extracts were also immunoprecipitated with PP1α and PP1γ antibodies. Rat cortical extracts were also immunoprecipitated with the same antibodies as well as with the LAP1 antibody and the results are presented in Figure 4D. Endogenous LAP1B was also  Figure  4B) and SH-SY5Y cells ( Figure 4C). Finally, using rat cortical brain extracts we confirmed that the LAP1B:PP1 complex is also formed in brain, since we detected LAP1B after IP with PP1α or γ antibodies. Further, when we immunoprecipitated with LAP1 antibody we could also detect both PP1 isoforms ( Figure 4D). The complex was also detected in other brain regions, namely rat striatum and cerebellum (data not shown) were PP1 is highly abundant.

Both LAP1B isoforms bind to PP1
Thus far we have unequivocally shown that LAP1B is a novel PP1 regulatory protein and that the complex formed by the two proteins can occur both in vitro and in vivo. Given the high degree of similarity between both LAP1B variants and the fact that both comprise the BM1 (REVRF) responsible for the interaction, it is reasonable to deduce that both variants interact with PP1. In order to confirm this hypothesis we generated, by site directed mutagenesis, the LAP1B (ΔA185) construct that corresponds to LAP1B variant 2 reported in GenBank. The interaction of both LAP1B isoforms with PP1 was indeed confirmed by in vitro overlay assay. Briefly, 250 ng of recombinant purified PP1γ1 protein were separated by SDS-PAGE and electrotransferred to a nitrocelulose membrane that was subsequently overlaid with each of the LAP1B variants. As expected both LAP1B variants are able to interact with PP1γ1 since we can detect a band of approximately 37 kDa ( Figure 5) that corresponds to the molecular weight of PP1γ1.

Localization of the LAP1B:PP1 complex
Given the confirmation that this novel complex is formed in vivo, it is evident that these two proteins are functionally related and therefore it becomes important to describe their localization. Co-localization of both proteins indicated that the complex exists in vivo and has physiological relevance. HeLa cells were transiently transfected with Myc-LAP1B and subjected to immunocytochemistry using a Myc-tag antibody. The Myc-LAP1B was mainly found in the nuclear envelope and also in the nucleus where it co-localizes with lamin B1 ( Figure  6A). Endogenous PP1γ and PP1α were detected using specific antibodies. All proteins have the expected subcellular distribution; each has been previously described individually [10,21,23]. The PP1γ was found predominantly in the nucleus, including the nucleolus, and throughout the cytoplasm ( Figure  6B). The PP1α was also found predominantly in nucleus, excluding the nucleolus, and throughout the cytoplasm ( Figure  6B). Despite the quite different subcellular distribution of LAP1B and PP1 isoforms it is evident that they co-localize at specific points within the nucleus and very near to the nuclear envelope, as observed by the yellow color indicated in the ROIs ( Figure 6B). The co-localizing points were confirmed by confocal profiling (denoted with *), where we demonstrated that the fluorescence intensity of the two channels correlated, particularly in those points, through the white line represented in the image ( Figure 6C and D). The co-localization quantitative analysis was performed ( Figure 6E) using a specific colocalization software (Zeiss LSM 510 4.0 software) as previously described [29,31]. Essentially, LAP1B co-localizes similarly with both PP1 isoforms (27.6 ± 0.65% and 27.1 ± 0.8% with PP1γ and PP1α, respectively). The percentage of PP1γ and PP1α that co-localize with LAP1B is lower (13.5 ± 0.48 and 13.8 ± 0.8, respectively).

PP1 specifically binds to LAP1B
Many nuclear membrane proteins are primarily or ultimately linked to each other via direct or indirect interactions involving the nuclear lamina. To test the specificity of the LAP1B:PP1 complex we tested whether other inner nuclear membrane proteins (NPC62, emerin and LAP2β) also bound to PP1. Co-IPs were performed using PP1 antibody and the potential binding partners were further screened using specific antibodies (Figure 7). The results presented, clearly show that PP1 only binds to LAP1 and not to the other proteins tested (NPC62, emerin and LAP2β). These results strengthen our hypothesis that LAP1 and PP1 are functionally associated, since they bind specifically in regions near the nuclear envelope ( Figure 6).

LAP1B is a novel substrate for PP1
It is well established that PP1 versatility is largely determined by its regulatory proteins. The latter define subcellular targeting, substrate specificity and even the activity PP1. Having determined that LAP1B is a novel PP1 regulatory protein, since the complex is formed both in vitro and in vivo and the two proteins co-localize in human cell lines, it is reasonable to deduce that the two proteins are functionally associated. We went on to test if LAP1B is a substrate for PP1, given that LAP1B can be phosphorylated at several residues [36][37][38]. SH-SY5Y cells were incubated with two different concentrations of OA (a protein phosphatase inhibitor), followed by IP with LAP1 antibody and further incubation with recombinant purified PP1γ1 protein (Figure 8). Analysis of Figure 8 showed that when we inhibit PP2A and PP1 (500 nM OA) a slight decrease in the migration of LAP1B is detected, consistent with an increase in its protein phosphorylation level, this is not evident when only PP2A is inhibited (0.25 nM OA) (Figure 8, lysates). In contrast, when we immunoprecipitate LAP1B from cells treated with 500 nM OA and incubated the resulting immunoprecipitates with 100 ng of purified PP1γ1 protein (Figure 8, IP) a clear shift in the opposite direction is observed, indicating an in vitro dephosphorylation by PP1γ1.

Discussion
PP1 regulates numerous cellular functions by binding to different regulatory subunits, which are, in turn, responsible for the targeting of PP1 to a particular subcellular compartment, but also determine substrate specificity and activity [4,6,7,12]. In this study we show that LAP1B is a novel PP1 regulatory protein. This LAP1B interaction was originally identified in YTH screens of a human brain cDNA library using three PP1 isoforms (α, γ1 and γ2) as baits [19,20]. Moreover, LAP1B LAP1B Is a Novel PP1 Regulatory Protein PLOS ONE | www.plosone.org variant 1, from all these screens, was isolated and fully sequenced [32]. In each case the variant obtained was the one recently reported in the GenBank database (NM_001267578) and differs from the LAP1B variant 2 (NM_015602) by only a 3  nucleotide (CAG) insertion, which results in the introduction of a single alanine residue at position 185 of the coding sequence. Typically such variants are generated by alternative splicing [28,33] and this has been shown to be so for the TOR1AIP1 gene. Despite this subtle difference, both LAP1B variants have the same conserved PP1 binding domains and bind to PP1 ( Figure 5). Consistent with the YTH results the data here presented, for the yeast co-transformation assays, confirmed that LAP1B is able to bind to all the PP1 isoforms tested (α, γ1 and γ2), but not to the C-terminus of PP1γ2 isoform ( Figure 3). Thus, this novel LAP1B:PP1 complex was validated not only in vivo but also in vitro by a blot overlay assay ( Figure 2). LAP1 is expressed in both non-neuronal and neuronal tissues [30], as are PP1α and PP1γ1; even though the latter has higher expression levels in brain [9]. Thus, we performed co-IP assays in COS-7 cells (non-neuronal cell line), SH-SY5Y cells (neuronal-like cell line) and rat cortex ( Figure  4). Moreover, the complex LAP1B:PP1 was also found in rat hippocampus and striatum, regions where PP1α and PP1γ are particularly enriched (data not shown) LAP1B is not a well characterized protein and its main cellular function remains to be defined. However, it is an integral membrane protein of the inner nuclear membrane that interacts with lamins and chromosomes and this is functionally relevant for the maintenance of the nuclear architecture during interphase and mitosis [24]. Recently, it was demonstrated that LAP1 recruits torsinA to the nuclear envelope [25], although the latter is normally located at the endoplasmic reticulum. In contrast, a torsinA mutant form (ΔE302/303-torsinA), which is present in DYT1 dystonia patients, is primarily relocated to the nuclear envelope [39,40]. Protein: protein interactions can be regulated by protein phosphorylation and consequently by protein kinases and phosphatases. It is therefore highly relevant that LAP1B should bind to PP1. Thus, several LAP1B deletion mutants were generated to determine the binding region responsible for the binding to PP1. In silico analysis revealed that LAP1B has three potential PP1 binding RVxF motifs: REVRF (amino acids 55-59) and KVNF (amino acids 212-215) located in the nucleoplasm and KVKF (amino acids 538-541) located in the lumen of the perinuclear space. An additional generic PP1 binding motif termed SILK (Figure 1) is also present in the nucleoplasmic domain. We defined that LAP1B:PP1 interaction occurs primarily through a region that comprises the REVRF domains (BM1). These results ( Figures  2C, 3B, 4A) support an interaction, which in topological terms provides physiological relevance, given that the BM1 is located in the nucleoplasmic portion of LAP1B and PP1 is particularly abundant in the nucleus. The BM2 and the SILK motif are also located in the nucleoplasm but our results clearly showed that these motifs do not mediate the interaction between LAP1B and PP1. BM3 is located in the perinuclear space making this an unlikely domain for the association of both proteins. Moreover, BM1 is conserved among different species ( Figure  1B) while BM2 and SILK are not. Another remarkable aspect is that RVxF motif is often N-terminal flanked by basic residues and C-terminal flanked by acidic residues and this affects the binding affinity for the RVxF motif [14]. Concordantly, BM1 is preceded by basic residues (arginine) and followed by acidic residues (aspartic and glutamic acid) ( Figure 1B). Further, the SILK motif is always N-terminal to the RVxF motif [16], but this is not the case with LAP1B where in fact it is C-terminal to the BM1, reinforcing that the SILK motif is not important for the interaction.
PP1 regulatory proteins can be substrates that directly associate with the PP1 catalytic subunit, but can also be substrate specifiers and/or targeting proteins. Having established that LAP1B is a novel PP1 regulatory protein and that the BM1 is responsible for the interaction, the physiological significance of this complex was addressed. Immunolocatization of LAP1B and PP1 (for both PP1α and PP1γ isoforms) showed that they co-localize near the nuclear envelope ( Figure 6). These results are in agreement with PP1 interacting with LAP1B through its nucleoplasmic domain, as discussed above. Furthermore, LAP1B is well described to interact with lamin A/C and B1 [21,24]. Lamins are not only located at the nuclear membrane periphery but are also found within the nucleus [41,42]. Indeed, we showed that LAP1B colocalizes with lamin B1 at the nuclear envelope and also in specific intranuclear areas ( Figure 6). Originally, nuclear lamins were proposed to have a role in supporting the nuclear envelope and binding to chromatin. However, recent reports suggest many other roles for nuclear lamins, namely in DNA replication, transcription, mitosis, apoptosis and cell differentiation [43]. Potentially, and in a similar fashion, LAP1B may be involved in other nuclear functions. LAP1B is known to be phosphorylated during both interphase and mitosis and several phosphorylated residues have been identified [36][37][38]. However, the involvement of specific kinases and/or phosphatases had not hitherto been documented. Since we validated the interaction between LAP1B and PP1 it is reasonable to deduce that LAP1B may indeed be a substrate for PP1. When PP1 is inhibited with OA a slight shift in the LAP1B migration is observed, consistent with an increase in its protein phosphorylation state ( Figure 8). Additionally, upon adding purified PP1γ1 protein to LAP1B immunoprecipitates, from cells previously incubated with 500 nM OA, an increase in the migration of LAP1B was detected, indicating that in vitro PP1 dephosphorylates LAP1B.
Many PP1 regulators identified thus far are located in the nucleus and are involved in diverse cellular functions, such as, cell cycle regulation, splicing and transcription [4,5,20].
However, the only known PP1 regulator located specifically at the nuclear membrane is AKAP-149. AKAP149 is a component of the endoplasmic reticulum/nuclear system but the discovery that it interacts with lamins A/C and B [44], suggests that AKAP149 is associated with both the outer and inner nuclear membranes. AKAP-149 recruits PP1 to the nuclear envelope upon nuclear envelope assembly in vitro and promotes lamin B dephosphorylation and polymerization [45,46]. Indeed, PP1 may mediate nuclear lamina reassembly, in part by dephosphorylation of lamin B at the end of mitosis [47]. Following nuclear envelope disassembly at mitosis, LAP1B and lamin B have similar localization and both reassemble around the nuclear envelope during telophase [48].
In conclusion, we have identified a novel PP1 regulatory protein, LAP1B. Both proteins co-localize in the close proximity of the nuclear envelope. LAP1B is mainly localized to the nuclear envelope whereas PP1 is rich in the nucleus. Colocalization of the LAP1B:PP1 to the nuclear envelope is indicative of specific protein recruitment for signaling events, and this will be addressed in the future. This interaction occurs through the REVRF motif, located in the nucleoplasmic domain of LAP1B. Further we also determined that LAP1B is a substrate for PP1 and PP1 in turn will dephosphorylate LAP1B.