Requirement for Lamin B Receptor and Its Regulation by Importin β and Phosphorylation in Nuclear Envelope Assembly during Mitotic Exit*

Lamin B receptor (LBR), a chromatin and lamin B-binding protein in the inner nuclear membrane, has been proposed to target the membrane precursor vesicles to chromatin mediated by importin β during the nuclear envelope (NE) assembly. However, the mechanisms for the binding of LBR with importin β and the membrane targeting by LBR in NE assembly remain largely unknown. In this report, we show that the amino acids (aa) 69–90 of LBR sequences are required to bind with importin β at aa 45–462, and the binding is essential for the NE membrane precursor vesicle targeting to the chromatin during the NE assembly at the end of mitosis. We also show that this binding is cell cycle-regulated and dependent on the phosphorylation of LBR Ser-71 by p34cdc2 kinase. RNAi knockdown of LBR causes the NE assembly failure and abnormal chromatin decondensation of the daughter cell nuclei, leading to the daughter cell death at early G1 phase by apoptosis. Perturbation of the interaction of LBR with importin β by deleting the LBR N-terminal spanning region or aa 69–73 also induces the NE assembly failure, the abnormal chromatin decondensation, and the daughter cell death. The first transmembrane domain of LBR promotes the NE production and expansion, because overexpressing this domain is sufficient to induce membrane overproduction of the NE. Thus, these results demonstrate that LBR targets the membrane precursor vesicles to chromatin by interacting with importin β in a LBR phosphorylation-dependent manner during the NE assembly at the end of mitosis and that the first transmembrane domain of LBR promotes the LBR-bearing membrane production and the NE expansion in interphase.

Nucleus, the largest organelle that contains the genome of the eukaryotic organisms, is surrounded by a continuous nuclear envelope (NE) 2 composed of a pair of inner and outer nuclear membranes, studded by numerous nuclear pore com-plexes. Through the nuclear pore complexes, the NE controls the flow of molecules between the nucleus and the cytoplasm in a tightly regulated manner (1)(2)(3)(4)(5). The NE is highly dynamic during the cell cycle. It disassembles into membranous vesicles or tubules and disperses into the cytoplasm at the onset of prometaphase and reassembles around the newly replicated chromosomes at the end of mitosis using their previously disassembled components (2,4,6). Recently, it was shown that Ran, a small Ras-like nuclear GTPase, and its binding proteins regulate the NE assembly process (7)(8)(9)(10)(11)(12)(13). Ran exists in GTP-and GDP-bound states that interact differently with effectors. Conversion between these two states and the assembly or disassembly of the effector complexes requires the interaction of regulator proteins, the nucleus-based guanine nucleotide exchange factor RCC1 and the cytoplasm-localized GTPase-activating protein RanGAP1 (6,14). The compartmentalization of the regulators produces a high concentration gradient of Ran-GTP in the nucleus and Ran-GDP in the cytoplasm. In mitosis, after the NE breakdown, the nucleoplasm and the cytoplasm are mixed; Ran diffuses into the cytoplasm with many other soluble nuclear proteins; RCC1 dynamically binds with the mitotic chromosomes and promotes the establishment of a concentration gradient of Ran-GTP/GDP where it is found at highest near the chromosomes and lowest at the centrosomes. During the NE assembly at the end of mitosis, Ran rebinds the chromatin and regulates the NE assembly, which depends on its GTPase cycle and its regulators (6 -11, 14).
Although Ran and its regulators have been shown crucial for the NE assembly, the mechanism for the NE precursor vesicle recruitment to chromatin and the downstream effectors remain poorly understood (8 -11, 15). Importin ␤ interacts with the FXFG domain of nucleoporins through its FXFG domainbinding sites (16,17). Importin ␤ likely works with other Ranbinding proteins to regulate the nuclear pore complexes and NE formation through interacting with the FXFG domain nucleoporins and targeting them to the chromatin (8,13). For example, importin ␤ depletion in Xenopus egg extracts blocks the NE assembly, whereas adding back exogenous importin ␤ rescues the NE formation in an appropriate concentration-dependent manner (8). Moreover, importin ␤-coated beads alone could induce the NE assembly in Xenopus egg extracts under the regulation of Ran (8). In Caenorhabditis elegans embryos, importin ␤ knockdown by RNAi also blocks the NE assembly (18). We have recently reported that importin ␤ plays a crucial role in the recruitment of the lamin B receptor (LBR)-bearing NE precursor membrane vesicles to the chromatin surface and the NE assembly through a direct interaction with the N-terminal spanning domain of LBR at amino acids (aa) 45-90 (15). Furthermore, these activities were carried out in a Ran-sensitive and importin ␣-independent manner (15).
LBR appears to be a central player in targeting nuclear membranes to the chromatin (4,15). LBR is a polytopic inner nuclear envelope protein with eight predicted hydrophobic transmembrane domains (76). The hydrophilic N-terminal spanning region of LBR is composed of ϳ200 aa and exposed to the nucleoplasm, where it interacts with chromosome/chromatin proteins such as B-type lamins, HA95, and heterochromatin protein 1 (HP1) (19 -30). The interaction of LBR with HP1 may not be direct, but is mediated through the core histones H3/H4 (31). The N-terminal spanning region of LBR contains a conserved arginine/serine (RS)-rich region, which may be phosphorylated on some serine and threonine residues by multiple RS kinases in a cell-cycle-dependent manner (32,33). LBR phosphorylation appears to regulate its binding to chromatin proteins (32)(33)(34)(35)(36)(37). The C terminus of LBR is a short hydrophilic spanning region facing to the nucleoplasm with an unknown function (38). The large middle part from aa 211-615 of the LBR houses the 8 transmembrane domains. The first transmembrane domain, but not the others, is critical for confining this protein to the inner nuclear membrane (39). During the NE assembly at the end of mitosis, it is found that LBR plays a central role in targeting the NE precursor membrane vesicles to the chromatin mediated by importins. Overexpression of LBR in yeast cells induces over-generation of membrane and formation of membrane stacks in the cytoplasm (40). In sea urchins, LBR is required for targeting the LBR-bearing membrane vesicles to the chromatin and the NE assembly during male pronuclear formation (41). In culture cells, LBR fused with the green fluorescent protein (GFP) has been used to follow the dynamics of the NE assembly and the mechanism investigation of the NE membrane precursor targeting to the chromatin (42,43). In a more recent study we found that, during the NE reassembly, the targeting of the NE membrane precursor vesicles to the chromatin is mediated by the interaction of the N-terminal spanning region with importin ␤ (15). Once meeting with Ran-GTP on the surface of the chromatin generated by RCC1, importin ␤ preferentially binds with Ran-GTP and releases the LBR-bearing vesicles to assemble the NE. With the hydrolysis of the GTP molecule of Ran, the membrane vesicles fuse to form the bi-layered NE and importin ␤ leaves Ran for next cycle (6,15). However, how the interaction between importin ␤ and LBR is regulated and why overexpression of LBR causes overgeneration of the membrane remain unclear.
In this work, we report that the targeting of the LBR-bearing NE precursor membrane vesicles to the chromatin during the NE assembly is mediated by importin ␤ through interaction of its aa 45-462 with the aa 69 -90 of LBR. LBR knockdown causes the NE assembly failure and the cell death in early G 1 phase. We also show that the phosphorylation status of LBR serine 71 regulates the interaction between LBR and importin ␤. Deletion of the N-terminal spanning domain or aa 69 -73 of LBR reduces the binding strength of LBR to the chromatin in interphase, increases the lateral movement potential of LBR, prevents successful targeting of the LBR-bearing NE precursor membrane vesicles to the chromatin at the end of mitosis, and causes the cell death of the daughter cells in next early G 1 phase. We also find that the first transmembrane domain of LBR is responsible for the membrane production. Together, our results demonstrate that LBR performs a crucial role in recruiting the LBRbearing NE precursor membrane vesicles to the chromatin during the NE assembly mediated by importin ␤ in a phosphorylation-dependent manner in mitosis and that the first transmembrane domain of LBR is responsible for the membrane production and the targeting of this protein to the NE during the nuclear growth and the NE expansion in interphase.
Protein Expression and Purification-Escherichia coli strain BL21 (pLys) was transformed with His-xLBR 1-210 , His-xLBR 1-210-S71A , His-xLBR 1-210S71D , GST-importin ␤ 1-876 ,  , and GST-importin ␤ 71-380 . To produce the recombinant His-tagged LBR proteins, the bacterial cells were grown to an A 600 of ϳ1.0. Isopropyl-␤-D-thiogalactopyranoside was then added to a final concentration of 0.1 mM, and the cells were incubated at 30°C for more than 5 h to induce protein expression. To produce recombinant GST-tagged importin ␤ proteins, the bacterial cell cultures were grown to an A 600 of ϳ0.5. Isopropyl-beta-D-thiogalactopyranoside was then added at a final concentration of 0.1 mM, and the cultures were incubated at 17°C for Ͼ6 h to induce the protein expression. The bacterial cells were pelleted by centrifugation at 6,000 ϫ g for 10 min at 4°C. Proteins were purified with either Talon-Resin (BD Bioscience) or Glutathione-Sepharose 4B (Pharmacia Biotech Inc.) following the manufacturer's instructions.
Cell Culture, Transfection, Synchronization, and RNAi-HeLa cells were grown in DMEM medium (Invitrogen), supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin (Sigma), on 35-mm diameter Petri dishes at 37°C in a 5% CO 2 atmosphere and transfected using the calcium precipitation method as described previously (44). Briefly, the precipitated plasmid DNA was left with the cells in the Petri dishes for 6 h. Cells were then washed with PBS and grown for another 24 -48 h before direct observation and fixation for immunofluorescence microscopy.
For cell synchronization, HeLa cells were grown in DMEM, treated in thymidine (2.5 mM) for 20 h, released in fresh medium for 4 h, and then treated with nocodazole (50 nM) for 12 h. Almost all the cells were blocked at metaphase.
For co-immunoprecipitation assays, the clarified cell lysates were incubated with indicated antibodies for 60 min on ice. Then 15 l of protein A-or G-Sepharose (75% slurry) was added, and the mixtures were rotated for 2 h at 4°C. The beads were washed three times with lysis buffer and harvested by brief centrifugation and finally suspended in gel sample buffer.
For GST pulldown assay, 5 g of soluble GST or GST-fused proteins bound to 15 l of Glutathione-Sepharose beads (75% slurry) were incubated with lysates from HeLa cells for 3 h at 4°C. The beads were washed five times with lysis buffer and harvested by brief centrifugation and finally suspended in gel sample buffer. The beads-bound proteins were separated on 10% SDS-PAGE gels and analyzed by immunoblotting (IB) with the indicated antibodies.
Gel Electrophoresis and IB-After being separated on 10% SDS-PAGE gels, the protein samples were transferred onto nitrocellulose filters in the transfer buffer (25 mM Tris, 192 mM glycine, and 20% methanol) for 1 h at 100 V. The filters were blocked in TTBS (20 mM Tris-HCl (pH 7.4), 500 mM NaCl, and 0.3% Tween 20) containing 5% nonfat milk for 1 h at room temperature and probed with anti-GFP monoclonal antibody (Santa Cruz Biotechnology, diluted 1:1000 in TTBS with 5% nonfat milk) or probed with anti-importin ␤ monoclonal antibody (BD Transduction, diluted 1:1000 in TTBS with 5% nonfat milk) overnight at 4°C. The filters were then washed three times and blocked again for 30 min in TTBS containing 5% nonfat milk and incubated with horseradish peroxidase (HRP)conjugated goat anti-mouse secondary antibody (Jackson, diluted 1:1000 in TTBS with 5% nonfat milk) for 1 h at room temperature. After a thorough wash in TTBS, the filters were developed for visualization by enhanced chemiluminescence (Sigma) and x-ray films.
Immunofluorescence Microscopy-Transfected cells were grown to 60% confluency on 35-mm diameter Petri dishes, washed three times with PBS, and fixed at room temperature in 4% paraformaldehyde for 15 min. The cells were then incubated with primary antibodies diluted in PBS containing 3% BSA for 1 h at room temperature or overnight at 4°C. The antibody against cleaved caspase 3 was from Cell Signaling Technology and diluted at 1:1000. The antibody against hLBR was generated by injecting rabbits and diluted 1:500. The cells were then washed five times in PBS and incubated with secondary antibodies diluted in PBS containing 3% BSA (TRITC goat anti-rabbit Ig (DAKO), diluted 1:200) at room temperature for 45 min to 1 h at room temperature. The cells were then washed five times in PBS and mounted in Mowiol (Sigma) containing DAPI. Samples were viewed under a Zeiss immunofluorescence microscope 200M equipped with a 63ϫ objective. Images were captured using a cooled charged-coupled device AxioCamMRm camera.
Live Imaging-HeLa cells grown in a glass-bottomed Petri dish in a temperature-controlled chamber at 37°C with 5% CO 2 were transfected with GFP-xLBR 211-621 , GFP-xLBR 1-621⌬69 -73 , or full-length GFP-xLBR 1-621 (as control). The images of the cells expressing the GFP fusion protein were acquired every 5 min by a cooled charged-coupled device AxioCamMRm camera on an Axiovert 200M microscope. The Axiovert software was used to collect and process the data.

Preparation of Xenopus Egg Extract and Demembraned Xenopus Sperm Chromatin and the NE Assembly in Vitro-Low
speed Xenopus egg extracts were prepared as described previously (45)(46)(47). The extracts were then frozen and stored in aliquots in liquid nitrogen until use.
Sperms were released from testes by gently squeezing in nuclear isolation buffer (NIB, 15 mM NaCl, 60 mM KCl, 15 mM Tris, pH 7.5, 1 mM dithiothreitol, 0.5 mM spermine, and 0.25 M sucrose) and centrifuged at 100 ϫ g for 1 min at 4°C to remove somatic tissues. Sperm-containing supernatant was centrifuged at 1,500 ϫ g for 10 min at 4°C. The pellet was re-suspended in NIB and incubated in the presence of 0.05% lysolecithin for 8 -10 min at 23°C. The reaction was stopped by adding 3 volumes of cold NIB plus 3% BSA. After washing for three times with NIB, the demembraned sperm chromatin were stored in aliquots in liquid nitrogen at 10 8 /ml. Low speed egg extract was mixed with an ATP-regenerating system and the demembraned sperm chromatin. The reaction system contained 20 l of low speed egg extract, 1 l of ATPregenerating system, and ϳ2000/l demembraned sperm chromatin. The proteins His-xLBR 1-210 , His-xLBR 1-210-S71A , His-xLBR 1-210-S71D , or EB buffer (extract buffer) alone was added into the reaction system to a final concentration of 10 M. The samples were removed after incubation at 23°C for the time indicated and stained on a slide with 3,3-dihexyloxacarbocyanine without fixation. The samples were immediately viewed under a Zeiss 200M immunofluorescence microscope equipped with a 63ϫ objective. Images were captured using a cooled charged-coupled device AxioCamMRm camera.
FRAP Analysis-Fluorescence recovery after photobleaching (FRAP) was performed with a Leica confocal microscope using a 488-nm line of a 400-milliwatt Kr/Ar laser in conjunction with a 100ϫ objective for optimum resolution. For quantitative essay, the framed areas were photobleached at full laser power (100% power and 100% transmission) for 2 s, and recovery of the fluorescence was monitored by scanning the whole cell at low laser power (19% power and 19% transmission) in a 20-s intervals. The scanning laser intensity did not significantly photobleach the specimen over the time course of the experiment.
MS Analysis of the Phosphorylated Proteins-2 g of His-hLBR 1-210 was incubated with cyclin B/Cdc2 (New England Biolabs) in 50 mM Tris, pH 7.5, 10 mM MgCl 2 , 2 mM EGTA, 5 mM DTT, 100 mM ATP, for 30 min at 30°C. The reactions were subjected to SDS-PAGE, and the gels were processed for MS analysis. The targeted gel band was cut and destained by 25 mM NH 4 HCO 3 in 50% acetonitrile. The gel band was dried in 100% acetonitrile and digested in 25 mM NH 4 HCO 3 (pH 8.0) with modified porcine trypsin (Promega, Madison, WI) using a trypsin-to-protein ratio of 1:50 (w/w) overnight at 37°C. The peptide mixtures were extracted from the trypsin-digested gel band using 50% acetonitrile/1% trifluoroacetic acid, redissolved with 10 l of 0.2% formic acid, and analyzed by online nanoLC (Microtech Scientific) connected to an LTQ Orbitrap mass spectrometer (ThermoFisher Scientific). Samples were eluted with a linear gradient from 2% to 35% buffer B (100% acetonitrile and 0.1% acetic acid) at a flow rate of 500 nl/min over 40 min using a 150-m inner diameter ϫ 150-mm reverse-phase C18 columns (Microtech Scientific). The MS was operated in data-dependent MS/MS mode in which each full MS scan was followed by five MS/MS scans where the five most abundant peptide ions were selected for collision-induced dissociation using a normalized collision energy of 35% in the ion trap. Survey MS scans were acquired in the Orbitrap analyzer with resolution of 60,000 (m/z 400) in the positive ion mode. Raw MS files were processed with Proteome Discoverer 1.0 (ThermoFisher Scientific), and searched against the IPI-human (version 3.55 European Bioinformatics Institute) containing both forward and reversed protein sequences. Initial maximum precursor and fragment mass deviations were set to 7 ppm and 0.5 Da, respectively. The search parameters included variable modifications for oxidation of methionine, and phosphorylation of serine, threonine, and tyrosine and variable modifications for cysteine carbamidomethylation. The maximum peptide false discovery rate was set to 0.01.

LBR Interacts with Importin ␤ at Mitosis and They Become
Disassociated at the End of Mitosis-The NE assembly is one of the key steps in generating daughter nuclei during the cell division in eukaryotic cells. LBR has been shown to play a vital role in targeting the membrane vesicles to the chromatin during the NE assembly at the end of mitosis (15). Although it is known that the targeting of the membrane to the chromatin by the LBR is carried out through interaction of the aa 45-90 of its N-terminal spanning region with importin ␤, the regulation of the interaction remains unclear. To investigate how the interaction of LBR with importin ␤ is regulated, we cloned the N-terminal spanning region aa 1-210 of Xenopus and human LBR genes and transiently expressed both GFP-xLBR 1-210 and GFP-hLBR 1-210 in human HeLa cells (Fig. 1). By immunostaining with an anti-importin ␤ antibody, we observed that GFP-LBR 1-210 co-localized with importin ␤ in the cytoplasm and spindle in mitosis and in the nucleus in interphase (Fig. 1A). We then selected the low-to-middle level GFP-LBR-expressing cells and synchronized them to interphase or mitosis and performed coimmunoprecipitation experiments using an anti-GFP antibody, followed by Western blots using the anti-importin ␤ antibody to reveal the interaction of GFP-LBR with importin ␤. The results showed that the endogenous importin ␤ co-precipitated with both GFP-xLBR 1-210 and GFP-hLBR 1-210 from the mitotic but not the interphase cell lysates (Fig. 1B). This result indicates that LBR interacts with importin ␤ mostly during mitosis.
We then set out to investigate if the interaction between importin ␤ and LBR is dynamic during mitosis. We synchronized GFP-xLBR 1-210 -expressing HeLa cells to various phases of the cell cycle by double thymidine block/release and performed pulldown assays with GST-importin ␤ 45-462 as the bait FIGURE 1. LBR 1-210 interacts with importin ␤ at mitosis and dissociates at the end of mitosis. A, HeLa cells, transiently transfected with GFP-xLBR 1-210 , were co-stained with anti-importin ␤ and DAPI for DNA. Note that this truncate LBR localizes to the nucleus in interphase and moves to mitotic spindle and other cytoplasmic place where it co-localizes with importin ␤ in mitosis. B, co-immunoprecipitation of importin ␤ with GFP-xLBR. The clarified xLBR 1-210expressing cells in both interphase and mitosis were lysed and incubated with a rabbit anti-GFP or rabbit-IgG (as control) antibody for 60 min on ice. Then 15 l of protein A-or G-Sepharose (75% slurry) were added, and the mixtures were rotated for 2 h at 4°C. The beads were washed three times with lysis buffer and harvested by brief centrifugation and finally suspended in gel sample buffer. The protein samples were separated by SDS-PAGE gel and processed for Western blot with anti-importin ␤ or anti-GFP antibody. Note that the binding of the truncate GFP-xLBR 1-210 with importin ␤ occurred only in mitosis. C, HeLa cells were arrested by double thymidine block/release to various cell cycle phases and lysed to generate time-course samples, and the time at the metaphase onset is referred to as 0 min. The cell lysates were processed for GST pulldown assay between GFP-xLBR 1-210 and GST-importin ␤ 45-462 using an anti-GFP antibody, followed by Western blot. D, HeLa cells were arrested by thymidine/nocodazole block at metaphase and released. Cell lysates were prepared at the time indicated after release from metaphase and GST pulldown assay between GFP-xLBR 1-210 and GST-importin ␤  was performed using anti-GFP antibody. The lysate GFP-LBR 1-210 in GST pulldown assay was used as control. Note that the interaction of GFP-LBR 1-210 became weaker along with the release from metaphase.
in corresponding cell lysates, followed by Western blot using an anti-GFP antibody. The result showed that a weak interaction between importin ␤ and the LBR started during the S and G 2 phases, then became stronger as the cells entered mitosis, reached peak at 90 min after entry into mitosis, and declined until it disappeared during the G 1 phase (Fig. 1C). We also performed similar pulldown assays using metaphase-arrested and -released cells expressing GFP-xLBR 1-210 and GST-importin ␤  . Consistent with the previous observations, we found that a strong interaction between LBR and importin ␤ occurred within 30 min after metaphase onset. Following that, the interaction attenuated till it became undetectable after 50 min when most of the cells entered the G 1 phase (Fig. 1D). Together, these results demonstrate that the strong binding of LBR with importin ␤ is mitosis-specific and that they dissociate at the end of mitosis.  Fig. 2A). Then we performed pulldown assays using these GST-importin ␤ truncate proteins and the lysate of mitotic HeLa cells transfected with GFP-xLBR 1-210 to determine the interaction between importin ␤ and LBR. Our results showed that the binding affinity of GST-importin ␤ 1-462 , GSTimportin ␤ 45-876 , or GST-importin ␤ 45-462 with LBR was as the same as that of full-length importin ␤, whereas the binding affinity of other truncated importin ␤ proteins with LBR was not detectable (Fig. 2B). These results indicate that the aa 45-462 region of importin ␤ is the minimal length required for binding with LBR (Fig. 2, A and B).
We then asked whether disrupting the interaction between LBR and importin ␤ interferes the NE assembly. We transiently expressed the full-length mutants GFP-LBR 1-621-S71A , GFP-LBR 1-621-S71D , and the wild-type GFP-LBR 1-621 in HeLa cells followed by fluorescence microscopy. The result showed that, like the wild-type, both GFP-LBR 1-621-S71D and GFP-LBR 1-621-S71A caused no significant defects to the cells at interphase (data not shown). However, once the cell entered mitosis, the unphosphorylation-mimicking mutant GFP-LBR 1-621-S71A showed a perturbed distribution and often aggregated into membrane stacks in the cytoplasm. At the end of mitosis (telophase/G 1 ), this mutant protein did not become associated with the NE but remained in the membrane stacks (Fig. 2F). In contrast, the phosphorylation-mimicking mutant GFP-LBR 1-621-S71D showed a similar NE distribution to the wild type at the end of mitosis (telophase/G 1 ), except that the GFP-LBR 1-621-S71D -expressing metaphase cells showed a likely continuous membrane around the aligned metaphase chromosomes (Fig. 2F). To explore and confirm whether serine 71 of LBR is phosphorylated during mitosis in HeLa cells, we performed phosphorylation reactions in vitro using his-hLBR 1-210 and Cdc2-cyclin B. The sample was separated by SDS-PAGE, and the targeted gel band was cut, destained, and trypsin-digested. Then we analyzed the tryptic peptide mixtures using high resolution LTQ Orbitrap mass spectrometry and found that serine 71 was indeed phosphorylated (Fig. 2G). Together, these results indicate that the interaction of LBR with importin ␤ regulated by phosphorylation at LBR Ser-71 is required for the NE membrane precursor vesicle targeting to the chromatin and the NE assembly. 1 Due to the Failure of the NE Assembly-Knowing that the N-terminal spanning region of LBR is critical for its interaction with importin ␤ and the NE assembly, we set out to examine the effects to the cells when deleting this N-terminal spanning region. We transfected the cells with the truncate GFP-xLBR 211-621 and the full-length GFP-xLBR 1-621 as control. We observed that, similar to the full-length GFP-xLBR 1-621 , the truncate GFP-xLBR 211-621 caused no visible defects to the interphase cells at current cell cycle, no matter at low or high level expression (Fig. 3A). We immunostained the cells with a human LBR-specific antibody (Fig. 3B) and found that the endogenous LBR (in red) was interweaved with the truncate exogenous xLBR 211-621 or the fulllength foreign xLBR 1-621 in the nuclear membrane (Fig. 3A,  arrows). It is noteworthy that, similar to the full-length GFP-xLBR 1-621 , the truncate GFP-xLBR 211-621 also induced the membrane stack formation in the cytoplasm and interweaved with the endogenous human LBR (red in Fig. 3A, arrowheads).

Deletion of the N-terminal Spanning Region Causes LBR Lateral Mobile and the Cell Death in Early G
It has been reported that one role of the N-terminal region of LBR is to anchor this protein and the NE to the chromatin (39,48,49). In this work, we set out to determine whether the protein is mobile when the function of N-terminal region of LBR is disturbed. We carried out a FRAP experiment using the cells with low-to-middle level expression of the full-length GFP-xLBR 1-621 as a control and the truncate GFP-xLBR 211-621 (Fig.   FIGURE 2. Phosphorylation of LBR Ser-71 is required for the interaction of LBR with importin ␤, the NE precursor membrane vesicle targeting to the chromatin, and the NE assembly. A, schematic representation of the importin ␤ truncates and their respective binding ability to GFP-xLBR 1-210 . The gray zone indicates the minimal length (aa 45-462) of the truncated importin ␤ proteins necessary for binding with LBR. B, GST pulldown assay between the GSTimportin ␤ proteins and GFP-xLBR 1-210 . 5 g of soluble GST or GST-importin ␤ proteins bound to 15 l of glutathione-Sepharose beads (75% slurry) was incubated with lysate of HeLa cells expressing GFP-xLBR 1-210 for 3 h at 4°C. The bound proteins of the beads were processed for Western blot using an anti-GFP antibody. Note that importin ␤ 45-462 was the minimum size to bind with GFP-xLBR 1-210 , and any truncate containing this fragment such as importin ␤ 1-462 and importin ␤ 45-876 as well as the full-length importin ␤ 1-876 could bind with GFP-xLBR 1-210 equally. C, part of xLBR amino acid sequence, showing the serine-rich conserved region (SR-rich region). D, pulldown assay using GST-importin ␤ 45-462 -coated beads and mitotic HeLa cell lysates expressing different GFP-xLBR truncated proteins or GFP as control, followed by a Western blot analysis with an anti-GFP antibody. Note that the aa 69 -90 region of xLBR is required for its binding with importin ␤ 45-462 and its serine 71 is critical for this binding. E, pulldown assay using GST-importin ␤ 45-462 -coated beads and the lysate of mitotic HeLa cells expressing the wild-type truncate GFP-xLBR 1-210 , the unphosphorylation-mimicking mutant GFP-xLBR 1-210-S71A or the phosphorylationmimicking mutant GFP-xLBR 1-210-S71D , or GFP alone as control, followed by a Western blot analysis with an anti-GFP antibody. Note that importin ␤ 45-462 only bound with wild-type GFP-xLBR   3, C and D). As expected, the fluorescence in the bleached NE regions of the full-length GFP-xLBR 1-621 -expressing cells could not recover within Ͼ300 s after the photo-bleaching, indicating that the full-length GFP-xLBR in the NE was totally immobilized as previously reported (48). In contrast, the majority of the truncate GFP-xLBR 211-621 in the NE was mobile, as indicated by the fast recovery of the fluorescence of the photobleached region (Fig. 3D). This result suggests that the truncate GFP-xLBR 211-621 is laterally mobile and that it moves rapidly from ER to the NE, although this truncate protein has no "root" to anchor itself to the chromatin and/or the nuclear lamina.
Because the truncate GFP-xLBR 211-621 was mobile within the NE and showed no visible harm to the cell in current cell cycle, we wished to investigate if it is harmful to the cell in the next cell cycle. We transiently expressed the truncate GFP-xLBR 211-621 , full-length xLBR 1-621 in HeLa cells and followed the behavior of the dividing cells. We found that, although GFP-xLBR 211-621 is not significantly toxic to the cells in current cell cycle as aforementioned (Fig. 4A), the daughter cells overexpressing the truncate GFP-xLBR 211-621 could not spread well after division (Fig.  4B), compared with the daughter cells overexpressing the full-length xLBR (data not shown). Furthermore, the truncate GFP-xLBR 211-621 protein seemed to form vesicle-like or aggregate structures (Fig. 4B). By DAPI staining to visualize DNA, we found that the chromatin in these daughter cells could not decondense well, the NE around the chromatin could not be well assembled and the cells were likely dying (Fig.  4C). We repeated these experiments three times and found that the likely dying cells stood for ϳ70% of the paired G 1 cells overexpressing the truncate GFP-xLBR 211-621 , whereas only 3% were dying among the paired cells expressing full-length xLBR   (Fig. 4D). By time-lapse microscopy, we examined the behavior of the paired daughter cells during division (Fig.  4, E-G). We observed that the truncate GFP-xLBR 211-621 and GFP-xLBR 1-621⌬69 -73 did not take part in the NE assembly around the daughter nuclei, and the nuclei could not be well assembled, compared with those in the cells overexpressing full-length GFP-xLBR. Furthermore, the daughter cells overexpressing GFP-xLBR 211-621 and GFP-xLBR 1-621⌬69 -73 could not spread well and finally showed apoptotic characteristics such as bubbling, chromatin condensing, and apoptotic body-like structure formation (Fig. 4,  E and G).
To confirm if the dying paired G 1 cells were apoptotic, we immunostained these cells with an antibody against the cleaved caspase 3, a marker for apoptotic cells (Fig. 5, A and B). We observed that the cleaved caspase 3 staining was totally negative in the mother cells (in the current cell cycle) and clearly positive in the paired cells (G 1 phase of the next cell cycle). These results indicate that N terminus-deleted LBR and GFP-xLBR 1-621⌬69 -73 cannot target the NE precursor membrane vesicles to the chromatin to form the NE, and the daughter cells without a functional NE would eventually die in G 1 phase through apoptosis. Then we performed LBR RNAi and rescue experiment in HeLa cells to see if the function of LBR is critical for the NE assembly. At 72 h after siRNA transfection, the LBR protein level was decreased substantially detected by immuno- A, HeLa cells overexpressing full-length GFP-xLBR  or the N-terminal spanning region-deleted truncate GFP-xLBR 211-621 were fixed and immunostained with hLBR-specific antibody, and counterstained with DAPI for DNA. Note that both the full-length and the truncated LBR localized on the NE (arrows) and induced membrane overproduction and stack formation containing endogenous LBR in the cytoplasm (arrowheads). B, characterization of anti-human hLBR antibody used in A. This antibody specifically recognizes endogenous hLBR but not the exogenous GFP-xLBR. C, FRAP assay for the full-length GFP-xLBR 1-621 on the NE. Note that the fluorescence in the photo-bleached regions could not recover within as long as 300 s, indicating that the full-length xLBR  was static on the NE. D, FRAP assay for the GFP-xLBR 211-621 on the NE. Note that the fluorescence in the photo-bleached regions recovered rapidly, indicating that GFP-xLBR 211-621 was mobile on the NE. Bars, 10 m. blotting (Fig. 5C). We also co-transfected the cells with siRNA and RFP-H2B as a transfection reporter for microscopy. After 72 h, the cells were stained with an antibody against cleaved caspase 3. We found that LBR RNAi caused apoptosis of the daughter cells (Fig. 5D) and that the cells could be rescued by expressing RNAi-resistant xLBR (Fig. 5E). Taken together, these results demonstrate that the function of LBR is crucial for the NE assembly.
The First Transmembrane Domain of LBR Induces the NE Membrane Production and Targeting to the Chromatin-It has been previously shown that overexpression of exogenous full-length or N-terminal spanning region-deleted LBR in HeLa cells results in membrane overproduction and stack formation in the cytoplasm (15). To reveal how the membrane overproduction and stack formation were induced, we constructed GFP-fused fulllength xLBR 1-621 and a number of truncated proteins: GFP-xLBR  (the N-terminal spanning region), GFP-xLBR 1-260 (the N-terminal spanning region plus the first transmembrane domain), GFP-xLBR 211-621 (xLBR without the N-terminal spanning region), GFP-xLBR 261-621 (xLBR without the N-terminal spanning region and the first transmembrane domain), GFP-xLBR ⌬211-260 (xLBR without the first transmembrane domain), and GFP-xLBR ⌬211-289 (xLBR without the first and the second transmembrane domains) (Fig. 6A). We transiently expressed these proteins in HeLa cells and observed their distributions in the cells. The results showed that overexpressing fulllength GFP-xLBR 1-621 and GFP-xLBR 211-621 efficiently induced membrane over-generation and stack formation and that the N-terminal spanning region was situated in the nucleus when expressed (Fig.  6B). Furthermore, we observed that the truncate xLBR  , which only contains the N-terminal spanning region and the first transmembrane domain, not only localized to the NE, but also efficiently induced the membrane stack formation, and that the truncate GFP-xLBR 261-621 , which lacks the first transmembrane domain (aa211-260), distributed throughout the whole cell and caused no any membrane stack formation (Fig. 6B). Consistently, we also observed that both the truncate proteins GFP-xLBR ⌬211-260 (the first transmembrane domain deletion truncate mutant) and GFP-xLBR ⌬211-289 (both the first and the second transmembrane domain deletion truncate mutant) localized throughout the cell but not the NE and did not induce membrane stack formation (Fig. 6B). These results not only confirm that the first transmembrane domain of LBR is essential for the NE localization of this protein as reported previously (39,50), but also demonstrate that the transmembrane domain induces the membrane overproduction and the stack formation.

Phosphorylation of LBR Ser-71 Regulates the NE Assembly around the Sperm Chromatin in Xenopus
Egg Extracts-To further study the role of the N-terminal spanning region of LBR in the NE assembly, we decided to take advantage of the Xenopus egg extracts system to examine the NE assembly (14). We prepared the His-tagged truncate LBR proteins His-xLBR 1-210 , His-xLBR 1-210-S71A , and His-xLBR 1-210-S71D and the interphase Xenopus egg extract and the sperm chromatin for the nuclear assembly (46,47). When the sperm chromatin were added into the extract, a complete NE assembled around the sperm chromatin within 150 min, visualized by lipid staining with a lipophilic fluorescent dye 3,3-dihexyloxacarbocyanine (Fig. 7A). When the wild-type N-terminal  A, the upper panel shows a mother cell, and the lower panels show three paired daughter cells. Note that overexpression of GFP-xLBR 211-621 does not significantly harm the mother cell but causes the daughter cell death during the next G 1 phase through apoptosis, judging by positive staining of the cleaved caspase 3, an apoptosis marker. Bars, 10 m. B, the upper panel shows a mother cell, and the lower panels show three paired daughter cells. Note that Overexpression of GFP-xLBR 1-621⌬69 -73 does not significantly harm the mother cell but causes the daughter cell death during next G 1 phase through apoptosis. Bars, 10 m. spanning region His-xLBR 1-210 and its mutants were added, we found that the wild-type and the mutant S71D, but not S71A, blocked the NE assembly around the sperm chromatin (Fig. 7,  B-D). To understand these results, we supposed that, because S71A mutant did not bind with importin ␤ (see Fig. 2, D and E), it may not block the targeting of LBR-containing membrane vesicles to the chromatin for the NE assembly by importin ␤. Because S71D could bind with importin ␤ (see Fig. 2E), it may block the targeting of LBR-containing membrane vesicles to the chromatin by importin ␤ for the NE assembly. Therefore, these results demonstrate that, even in the in vitro NE assembly system based on Xenopus egg extracts, phosphorylation of Ser-71 of LBR also regulates the binding of LBR and importin ␤ for the LBR-bearing membrane targeting to the chromatin and the NE assembly.

DISCUSSION
Importin ␤ is an evolutionarily conserved protein that serves as a main nucleocytoplasmic transport receptor to regulate the nuclear import together with importin ␣ regulated by Ran (51)(52)(53)(54)(55)(56). Importin ␤ can also mediate the nuclear import of some proteins independently of importin ␣ (57)(58)(59)(60)(61)(62)(63). It is also found that, in early mitosis, importin ␤ plays an inhibitory role for mitotic spindle assembly by interacting with and suppressing the spindle assembly factors (64), and this inhibitory role can be eliminated by Ran-GTP, which interacts with importin ␤ and releases the spindle assembly factors (6). At the end of mitosis, importin ␤ takes part in the NE assembly (9,12,15,65). Importin ␤ was first reported to take a role in the NE assembly in Xenopus egg extracts in an appropriate dosage-dependent manner, and if its concentration was increased in the extract by adding excess importin ␤ protein, it inhibits the NE assembly (9,12,13). In our previous report, we found that one role of importin ␤ in the NE assembly is to target the NE membrane precursor vesicles to the chromatin through interaction with the membrane-born LBR (15). We found that depletion of LBR causes the cell death via apoptosis at early G 1 phase due to the NE assembly failure. We have pinpointed the binding regions, which are aa 69 -90 in LBR and aa 45-462 in importin ␤. We also demonstrate that the interaction of both proteins is regulated by phosphorylation of LBR at Ser-71 and that the interaction plays a crucial role in mediating the LBR-containing membrane vesicle targeting to the chromatin for the NE assembly.
LBR has been implicated in a number of important functions. It plays roles in sustaining the NE architecture, in NE assembly, in the lobulated (segmented) nuclear shape formation in the development of the neutrophiles, and in the regulation of proliferative and functional responses for innate immunity (66 -69). It is also involved in some diseases, such as Greenberg skeletal dysplasia and Pelger-Huet anomaly (38, 70 -72). However, the detailed analyses of LBR have been performed using its soluble and functional N-terminal spanning region, which provides essential chromatin docking sites at the NE (22,24,30,73). During NE assembly, the N-terminal spanning region of LBR targets its associated membrane vesicles to the chromatin by interacting with importin ␤ (15). In this work, we have identified the interacting sites in both importin ␤ and LBR Nterminal spanning region. We observed that, without the N-terminal spanning region, LBR is mobile compared with the stock-still full-length LBR at the NE. Even so, the N-terminal spanning region-deleted LBR is not immediately lethal to the cell in the current cell cycle, and the cell could go to mitosis. However, when the cell divides, its daughter cells could not FIGURE 6. The first transmembrane domain is essential for LBR to localize to the NE and for the membrane stack formation. A, schematic description of the GFP-fused xLBR constructs used to transfect the cells. M1 to M8 indicate the eight putative transmembrane domain positions. B, the cells expressing full-length or truncate GFP-xLBR proteins were counterstained with DAPI for DNA. Note that only the wild-type and the first transmembrane domain-containing truncate proteins GFP-xLBR 1-260 and GFP-xLBR 211-621 localize to the NE and induce membrane stack formation in the cytoplasm, whereas the other truncate proteins did not localize to the NE and induce the membrane stack formation. When the first transmembrane domain was deleted or codeleted with the second transmembrane domain, the deletion mutants showed no specific membrane distribution in the cells. Together, these results demonstrate that the first transmembrane domain is sufficient to induce the membrane formation and targets the membrane to the NE. Bars, 20 m. organize their complete and functional NE due to the LBRbearing NE membrane precursor vesicles not being targeted onto the chromatin, and their chromatin could not be fully decondensed. Finally, these daughter cells will die in early G 1 phase by apoptosis. This result demonstrates that LBR does play a role in the NE assembly. Besides, LBR may play a role in chromosome decondensation and nuclear structure organization through anchoring the chromatin to the NE and organizing local domains through binding some factors like lamin B, HP1, and chromatin/DNA for some special nuclear functions (25,28,30). To perform these functions, LBR may be regulated through some modifications. During interphase, the RS region of LBR is mainly phosphorylated by NE-bound kinase-RS kinases; whereas in mitosis, the RS region is phosphorylated by p34 cdc2 at the site serine 71. RS kinases also phosphorylate other sites of LBR in interphase, and this process may be antagonized by phosphatase PP1 (34,36,74). The modification of the RS region is believed to modulate interaction between LBR and other nuclear proteins. Further phosphorylation of LBR by RS kinase completely abolishes the binding of p34/p32, indicating a distinct regulation mechanism between interphase and mitosis (32)(33)(34)(35)75). The function of these distinct kinases in regulating LBR activity so far remains unclear. In our present study, we found that Ser-71 phosphorylation of LBR by p34 cdc2 enhances the binding of LBR and importin ␤ during mitosis. As it has been reported that the PP1 stimulate the binding of LBR to the chromatin (74), here we propose that, during the NE assembly at the end of mitosis, once importin ␤/LBR-bearing membrane vesicle complex approaches the surface of the chromatin, importin ␤ preferentially binds with Ran-GTP on the surface of the chromatin generated by RCC1 and releases the LBR-bearing membrane vesicle. Simultaneously, PP1 dephosphorylates LBR at Ser-71 and prevents it from rebinding with importin ␤ but to the chromatin to form the NE.
Once the NE is assembled around the daughter nuclei, it grows to meet the requirement of the chromatin decondensation to a regular size and the function of the interphase nucleus. It has been shown that overexpression of LBR in yeast cells induced membrane overproduction and stack formation (40). We have previously reported that, in mammalian cells, overexpression of LBR resulted in overproduction of the NE, which either invaginated into the nucleus or budded off from the NE into the cytoplasm in an expression quantity-dependent manner (15). In this work, we found that the first transmembrane domain is an essential part for the NE growth and the membrane stack formation. With the N-terminal spanning region, the first transmembrane domain could induce the membrane stack formation when it is overexpressed. If the first transmembrane domain is deleted, the deletion mutant will not be able to localize to the NE, and its overexpression will not cause any membrane overproduction and membrane stack formation. Therefore, we propose that the first transmembrane domain of LBR is responsible for the NE generation and localization to the NE.
In summary, we found that the interaction between importin ␤ and the N-terminal spanning region of LBR occurs between aa 45-462 of importin ␤ and aa 69 -90 of LBR and that the interaction is regulated by the phosphorylation at the amino acid serine 71 of LBR. Disruption of this interaction by deletion of the N-terminal spanning region or 69 -73 aa of LBR inhibits the NE assembly. We also found that the first transmembrane domain of LBR is responsible for the NE generation. We propose that, during the NE assembly at the end of mitosis, LBRcontaining NE precursor membrane vesicles are targeted to the chromatin by the N-terminal spanning region of LBR through binding with importin ␤ in a LBR Ser-71 phosphorylation-dependent manner. Once it approaches the surface of the chromatin at the end of mitosis, the complex of importin ␤/LBR disassociates through the preferential binding of importin ␤ with Ran-GTP and the dephosphorylation of the released LBR FIGURE 7. Phosphorylation of LBR Ser-71 regulates the NE assembly around the sperm chromatin in Xenopus egg extract. When Xenopus egg extract was added with the sperm chromatin and the ATP-regenerating system and incubated at 23°C for a certain period, a complete NE could be assembled around the sperm chromatin. A, a time course of the NE assembly in the egg extract. The EB buffer was added into the extract as control. Note that a complete NE assembled around the well decondensed sperm chromatin, shown by fluorescent lipophilic dye 3,3-dihexyloxacarbocyanine staining for the NE and DAPI for DNA. B, when 10 M truncated wild-type protein His-xLBR 1-210 was added into the extract, the NE assembly was significantly blocked and the chromatin decondensation was not completed. C, when 10 M truncated unphosphorylation-mimicking mutant protein His-xLBR 1-210-S71A was added to the extract, the NE assembly and the chromatin decondensation were not seriously affected. D, when 10 M truncated phosphorylation-mimicking mutant protein His-xLBR 1-2101-210-S71D was added to the extract, the NE assembly and the chromatin decondensation were almost totally blocked. Together, these data indicate that phosphorylation of LBR Ser-71 regulates the NE assembly around the sperm chromatin in Xenopus egg extract. Bar, 20 m. by PP1 to prevent LBR from rebinding to importin ␤. The unphosphorylated LBR with the membrane vesicles binds to the chromatin through its N-terminal spanning region, and the vesicles fuse to form the NE. Without a functional N-terminal spanning region, the NE assembly by LBR-containing NE precursor membrane vesicles and importin ␤ would be inhibited, and the daughter cells would die in the next G 1 phase through apoptosis. LBR may also perform a role for the nuclear growth and the NE expansion through its first transmembrane domain in interphase.