Mitotic Defects Lead to Pervasive Aneuploidy and Accompany Loss of RB1 Activity in Mouse LmnaDhe Dermal Fibroblasts

Background Lamin A (LMNA) is a component of the nuclear lamina and is mutated in several human diseases, including Emery-Dreifuss muscular dystrophy (EDMD; OMIM ID# 181350) and the premature aging syndrome Hutchinson-Gilford progeria syndrome (HGPS; OMIM ID# 176670). Cells from progeria patients exhibit cell cycle defects in both interphase and mitosis. Mouse models with loss of LMNA function have reduced Retinoblastoma protein (RB1) activity, leading to aberrant cell cycle control in interphase, but how mitosis is affected by LMNA is not well understood. Results We examined the cell cycle and structural phenotypes of cells from mice with the Lmna allele, Disheveled hair and ears (LmnaDhe). We found that dermal fibroblasts from heterozygous LmnaDhe (LmnaDhe/+) mice exhibit many phenotypes of human laminopathy cells. These include severe perturbations to the nuclear shape and lamina, increased DNA damage, and slow growth rates due to mitotic delay. Interestingly, LmnaDhe/+ fibroblasts also had reduced levels of hypophosphorylated RB1 and the non-SMC condensin II-subunit D3 (NCAP-D3), a mitosis specific centromere condensin subunit that depends on RB1 activity. Mitotic check point control by mitotic arrest deficient-like 1 (MAD2L1) also was perturbed in LmnaDhe /+ cells. LmnaDhe /+ fibroblasts were consistently aneuploid and had higher levels of micronuclei and anaphase bridges than normal fibroblasts, consistent with chromosome segregation defects. Conclusions These data indicate that RB1 may be a key regulator of cellular phenotype in laminopathy-related cells, and suggest that the effects of LMNA on RB1 include both interphase and mitotic cell cycle control.


Introduction
The nuclear lamina is a network of Type V intermediate filament proteins that lie primarily beneath the inner nuclear membrane. In mice and humans, it is composed of both A-type (A, AD10, C and C2) and B-type (B1 and B2) lamins. A-type lamins are generated by alternative splicing of one gene (Lmna in mice, LMNA in humans) and B-type lamins are encoded by two genes, Lmnb1/LMNB1 and Lmnb2/LMNB2 [1][2][3]. The Lamin subtypes form independent, but interconnected, lattice-like meshworks that provide structural stability to the nucleus [4]. The nuclear lamina also participates in cell cycle control, DNA replication, gene transcription, genome three-dimensional (3-D) architecture and stabilization of the nucleus in the cytoplasm [1][2][3][4][5][6][7]. Perturbations to LMNA have dire effects on the cellular processes, including the cell cycle. Although recent work has begun to describe the consequences of LMNA defects on the cell cycle, the mechanisms underlying these effects are poorly understood.
Recent work has begun to make inroads to understanding cell cycle defects in LMNA/Lmna mutant cells [6,[11][12][13][14][15][16]. Cells from Lmna knock-out mice, as well as mice deficient for LMNA interacting proteins LAP2a and ZMPSTE-24, have defects in the G 1 /S-phase transition, due to reduced levels of hypophosphorylated retinoblastoma protein (RB1). Normal interactions among RB1 and a soluble, intranuclear pool of LMNA and LAP2a are disrupted in these mutant cells [11,13,14,16]. Cells from human progeria and muscular dystrophy patients have gene expression signatures that implicate central defects in RB1 activity as well [16,17]. However, for progeria cells, a direct link to RB1 signaling has yet to be demonstrated. Cell populations from progeria patients and mice with HGPS-related lamin A alleles do grow more slowly than normal cells. This is in part due to persistent DNA damage and telomere defects, which lead to increased cellular senescence [18][19][20][21][22][23]. In addition, cells expressing LMNA HGPS have aberrant mitotic progression and aneuploidy [6,7,12,24].
At the start of mitosis the nuclear lamina must break down to allow for proper attachment of the chromatids to the mitotic spindle [25]. The checkpoints regulating chromosome spindle attachment, congression at the metaphase plate and separation into daughter cells at anaphase are highly regulated. Condensin II is a multi-subunit protein complex that condenses mitotic chromosomes prior anaphase. One subunit of this complex, the non-SMC condensin II-subunit D3 (NCAP-D3) functions at the centromeric regions of chromosomes and is required to maintain centromeric cohesion. Recent studies in primary cells and tumor cell lines have shown that decreased expression of RB1 causes decreased NCAP-D3 levels, which resulted in a more disorganized metaphase plate and chromosome missegregation [23,[26][27][28]. Thus, although RB1 is often thought to exert its influence at the G 1 /S-phase transition, perturbations to RB1 have consequences further downstream in the cell cycle, specifically during mitosis. These recent findings suggest that the perturbations to RB1 in both human and mouse lamin A mutant cells could manifest at both the G 1 /S-phase transition and in mitosis.
In this study, we examined features of the cell cycle using a newly described Lmna mouse model, Disheveled hair and ears (Lmna Dhe ), which exhibits a subset of phenotypes of HGPS and other non-muscular dystrophy laminopathies, including epidermal dysplasia, craniofacial abnormalities, and markedly thinned hypodermal fat layers [5][6][7]19,29]. The Lmna Dhe allele is a spontaneous point mutation in the first coiled-coil domain of lamin A and C (L52R), suggesting it significantly perturbs lamina structure and function [29]. Indeed, we found that dermal fibroblasts from heterozygous Lmna Dhe mice (hereafter called Lmna Dhe/+ ) had nuclear morphology defects and increased DNA damage, as well as mitotic defects, including aneuploidy, lagging chromosomes and anaphase bridges. Our data indicate that these mitotic defects are associated with low levels of activated RB1 and CAPD3, as well as other defects in the mitotic spindle checkpoint, suggesting a new role for RB1 in the mitotic defects of laminopathies.

Mice
The B6(D2)-Lmna Dhe /TyJ mice arose on the C57BL/6J segment of the BXD8/TyJ strain, and the mutation has now been made congenic on the C57BL/6J inbred strain through repeated backcrosses (N50F1). The mice were maintained under 14:10 hour light:dark cycles; autoclaved diet NIH-31 (6% fat, 18% protein, Ca:P 1:1, vitamin and mineral fortified; PMI, Richmond, IN) and HCl acidified water (pH 2.8-3.2) were provided ad libitum [29]. Mice were housed in groups of 4 or 5 within polycarbonate boxes of 51 square inch area on sterilized shavings of Northern White Pine as bedding. All procedures were approved by The Jackson Laboratory's Institutional Animal Care and Use Committee and performed in accordance with National Institutes of Health guidelines for the care and use of animals in research (ACUC Policy # 99066).

Genotyping
The following PCR primers flanking the Lmna Dhe mutation in exon 1 were used to amplify genomic DNA extracted from tail tips: dhefwd 59-ACCTGCAGGAGCTCAATGAC-39, dherev 59 -TGAACTCCTCACGCACTTTG -39. PCR products then were digested with SmaI, which produces a 240 bp DNA fragment for the Lmna +/+ allele and 186 bp and 54 bp DNA fragments for the Lmna Dhe allele [29].

Cell Culture
Primary dermal fibroblast skin explant cultures were obtained using neonatal (8 day old) Lmna +/+ and Lmna Dhe/+ mice as previously described [30]. Briefly, we excised dorsal skin from just posterior to the occipital bone to just anterior to the tail base and from 1 mm dorsal to the limbs on either side. Any remaining subcutaneous fat and muscle was then trimmed and skin was washed twice in sterile, ice-cold 16 Phosphate Buffered Saline (PBS). Skin was cut into 2 mm62 mm squares and washed in sterile, ice-cold 16 PBS. Skin explants were placed into 100 mm 2 cell culture dishes dermal side down, covered with sterile, ethanolwashed glass coverslips and Dulbecco's Minimal Essential Media (DMEM; Gibco, Carlsbad, CA) supplemented with 10% Fetal Bovine Serum (FBS; Lonza, Walkersville, MD) and 2% antibiotic/ antimycotic (Gibco, Carlsbad, CA). Plates were placed in a humidified cell culture incubator at 37uC with 5% CO 2 and 2% O 2 . Media was replaced every 2 days with fresh media for 7 days until fibroblasts emerged from edges of skin explants. Fibroblasts were then passaged to additional coverslips for immunofluorescence or harvested for protein analysis.

Immunofluorescence
Dermal fibroblasts were fixed for 10 minutes in 4% formaldehyde, 16PBS and permeabilized for 10 minutes with 0.5% Triton X-100 in 16 PBS. Next, cells were immunostained with goat anti-  Microsystems, Deerfield, IL) with a 10061.4 NA oil objective lens. Three-dimensional (3D) image stacks were taken at 200 nm steps.

Nuclear morphometry
3D wide-field epi-fluorescence images were deconvolved using AutoDeblur software (Media Cybernetics, Bethesda, MD). Anti-LMNB immunolabeled cell nuclei were scored for the presence or absence of one or more blebs, anaphase bridges and/or micronuclei in the nuclear lamina using cells from three mice from different litters for each genotype. Single optical sections through the middle of each cell nucleus were examined for blebbing, while all sections were examined for the presence of micronuclei and/or anaphase bridges. Blebs, micronuclei and anaphase bridges were scored independently for each cell counted. Therefore cells may exhibit either blebs, micronuclei, anaphase bridges or all three.

Mitotic index
Cell fluorescently labeled with both anti-a-Tubulin and LMNA antibodies were directly scored using the Zeiss Axiovert 200M inverted microscope. Approximately 1000 cells from at least 3 different pups for both Lmna +/+ and Lmna Dhe/+ cultures were counted to determine the proportion of mitotic cells present. At least 100 mitotic cells from both Lmna +/+ and mutant cultures also were scored as in prophase, prometaphase, metaphase, anaphase or telophase, based upon DAPI, LMNA and a-Tubulin labeling.
Static Adhesion Assay 50,000 dermal fibroblasts were seeded into each well of a 6-well cell culture plate. After either 3 hours (Day 0 time point) or 96 hours (Day 4 time point), media was removed and wells were rinsed three times with pre-warmed 16 PBS. Next , 500 mL of substrate solution (3.75 mM p-nitropheno-N-acetyl-b-D-glucosaminide (Sigma-Aldrich, St. Louis, MO), 0.1% Triton X-100, 50 mM citrate buffer, pH 5.0) was added to each well and the plates were incubated for 30 minutes at 37uC. After incubation, 750 mL of STOP solution (5 mM EDTA, 50 mM Glycine buffer, pH 10.4) was added to each well. Lastly, 100 mL of reaction mixture was loaded in triplicates into a 96-well plate and absorbance was measured on a microplate reader at 405 nm [32,33].

Apoptosis assay
Annexin V labeling was performed as previously described [35,36]. Briefly, fibroblasts were trypsinized, washed in 16 PBS and resuspended in 16 Annexin V Binding Buffer (BD Pharmingen, San Jose, CA) at a concentration of 1610 6 cells/mL. 1610 5 cells were removed and suspended in 5 mL PE-Annexin V (BD Pharmingen, San Jose, CA) and 0.2 mg/mL DAPI for 15 minutes at room temperature. Lastly, 400 mL of 16 Binding Buffer was added to each sample. Annexin V labeling was measured by flow cytometry as described above.

DNA content measurements
For DNA content and cell cycle analysis, cells were trypsinized washed twice in 16 PBS, and fixed in ice-cold 70% ethanol for 2 hours at 4uC. Cells were washed twice with 16 PBS, resuspended in 200 mL of propidium iodide working solution (0.1% Triton X-100, 10 mg/mL RNase A, 1 mg/mL Propidium Iodide (Sigma-Aldrich, St. Louis, MO) for 15 minutes at 37uC, and analyzed by flow cytometry as described above [37].

Spectral Karyotyping (SKY)
Metaphase spreads were prepared as previously described [38]. Metaphase slides were denatured in 70% Formamide (Sigma-Aldrich, St. Louis, MO) in 26 Saline-Sodium Citrate buffer (SSC) (Invitrogen, Carlsbad, CA) at 68uC for 30 seconds and dehydrated in a series of ethanol washes (70%, 80% and 100%). The mouse specific SKY paint (Applied Spectral Imaging, Migdal Ha'emek, Israel) was denatured in an 80uC water bath for 10 minutes and allowed to pre-anneal at 37uC for 15 minutes before being applied to chromosome preparations. Slides were hybridized overnight at 37uC in a humidified hybridization chamber. Slides were then washed in 46SSC at 72uC and 46SST (46SSC/0.1% Tween) at 45uC for 5 minutes. The slides were incubated with CAD Antibody Detection Kits (Applied Spectral Imaging, Migdal Ha'emek, Israel) at 37uC for 45 minutes and washed in 46 SST at 45uC for 5 minutes before mounting in Vectashield (Vector Labs, Burlingame, CA) with DAPI. Slides were imaged and analyzed by an ASI SKY Workstation equipped with HiSKY software (Applied Spectral Imaging, Migdal Ha'emek, Israel). Automated karyotyping was followed by manual analysis to validate the karyotype.

Aberrant nuclear lamina morphology in Lmna Dhe/+ skin fibroblasts
Previous work on Lmna Dhe/+ mice had shown skin defects, cranial abnormalities, and extensive blebbing of the nuclear membrane in cranial osteoblasts [29]. To determine whether Lmna Dhe/+ skin was similarly affected we cultured dermal fibroblasts from neonatal mice, then fixed and labeled them with anti-LMNA and -LMNB antibodies. Examination by wide-field fluorescence microscopy revealed large blebs and lobulations within Lmna Dhe/+ fibroblasts nuclear membranes. These abnormal cells were significantly more frequent (41.78%+/24.9%) in Lmna Dhe/+ cultures than in Lmna +/+ cultures (6.54%+/21.4%) (p#0.01; x 2 -test) (Figure 1 A-F). In addition, while the LMNA signal was bright around the periphery of the entire nucleus of mutant cells, the LMNB signal was depleted in at least half of the blebbed regions ( Figure 1B,D and F; white arrowheads). Many mutant nuclei appeared larger than Lmna +/+ nuclei. 3-D confocal imaging and measurement of the nuclear volume confirmed that Lmna Dhe/+ cells had a larger average nuclear volume than Lmna +/+ counterparts (p#0.001, Student's t-test). The range of nuclear volumes also was greater in Lmna Dhe/+ cells, indicating large variations in nuclear size ( Figure 1G).
We next tested whether the Lmna Dhe allele affected nuclear lamina structure. Cells labeled with anti-LMNA and -LMNB antibodies were imaged by 3D confocal microscopy. Confocal optical sections of the top and bottom of mutant nuclei revealed patches of irregularity in the normal, criss-cross pattern of the LMNA and LMNB meshworks (59.4+/24% of cells; N = 100), unlike the regularly patterned network of lamins seen in Lmna +/+ cells (2+/22.8%; N = 84; p,0.001; x 2 -test) (Figure 2 A-F). These ''holes'' in the meshworks were particularly noticeable in blebbed regions of mutant nuclei (Figure 2 B, D and F; white arrowheads). Western blotting indicated similar amounts of LMNB in normal and mutant fibroblasts ( Figure 3A). Given the greater size of Lmna Dhe/+ nuclei, these measurements suggest a lower density of LMNB in the mutant cell nuclei, consistent with the loss of protein from blebs ( Figure 1D). Notably, both Prelamin A and LMNA were less abundant in both soluble and insoluble fractions in mutant cells, suggesting that LMNA is even less dense than LMNB in the lamina meshwork ( Figure 3B, C and D).
Similar to human progeria cells, Lmna Dhe/+ fibroblasts exhibited perturbations to LMNA disassembly during mitosis ( Figure 4B, D and F) [6]. Specifically, residual LMNA aggregates clustered close to the spindle poles as well as throughout the mitotic spindle in metaphase in all observed mitotic cells. In contrast, LMNA in Lmna +/+ cells was disassembled much earlier in mitosis and was nearly undetectable at metaphase ( Figure 4A, C and E). Cytoplasmic aggregates were not detected in Lmna Dhe/+ mitotic cells, in contrast to a previous report of fibroblasts transfected with Lmna HGPS [6]. LMNB disassembled normally during mitosis in Lmna Dhe/+ cells (data not shown). Together, these data suggest that LMNA expression and assembly at the nuclear lamina is perturbed in Lmna Dhe/+ cells during interphase and mitosis. In addition, Lmna Dhe influences the structure of the LMNB meshwork, but this influence is restricted to interphase.

Lmna Dhe/+ cells exhibit extensive aneuploidy
The large nuclear size of Lmna Dhe/+ fibroblasts suggested that the cells might be aneuploid, which is often a consequence of cell cycle defects [24,[39][40][41]. To test this we used flow cytometry to determine DNA content in each cell. We found a significantly higher fraction of cells that were greater than 4C in mutant as compared to Lmna +/+ cultures. Likewise a much lower proportion of Lmna Dhe/+ cells were 2C (p,0.01; x 2 -test) ( Figure 5A and B). In addition, the peaks surrounding 2C and 4C cells were broader in Lmna Dhe/+ cells, suggesting a range of genome sizes.
Increased DNA content, as observed above, suggested that Lmna Dhe/+ cells might harbor chromosome abnormalities. Indeed, spectral karyotyping (SKY) indicated widespread aneuploidy in Lmna Dhe/+ fibroblasts (92.5%) with chromosome numbers ranging from 38 to 104 chromosomes per nucleus ( Figure 6A). In general, Lmna Dhe/+ cells tended to be tetraploid, though chromosome copy numbers up to octoploidy were detected. In contrast, aneuploidy was detected in only 20% of Lmna +/+ cells and typically involved gains or losses of a single chromosome. No specific chromosome was found to be consistently duplicated in mutant cells ( Figure 6B and C). The wide variation in chromosome number in mutant cells is consistent with the large range of nuclear volumes observed ( Figure 1G; Figure 5). The increased nuclear volume and subsequent extensive chromosome duplications suggested that the Lmna Dhe allele alters cell division in some way.

Increased DNA damage in Lmna Dhe/+ fibroblasts
The extensive nuclear blebbing in Lmna Dhe/+ fibroblasts, as well as the pervasive aneuploidy, led us to examine whether the growth of these cells is affected in similar ways to other Lmna mutations [6]. We measured the growth rate of Lmna +/+ and Lmna Dhe/+ fibroblasts in vitro and found that Lmna +/+ cells grew significantly faster than Lmna Dhe/+ cells (p#0.05, Student's t-test) (Table 1) [32,33]. We then measured the proportion of senescent and actively cycling cells in Lmna Dhe/+ cultures using senescenceassociated b-galactosidase (SA-b-gal) and Ki-67 as markers, respectively. In contrast to Lmna HGPS and some other Lmna alleles, Lmna Dhe/+ cells did not exhibit a significant proportion of cells positive for SA-b-gal as compared to Lmna +/+ fibroblast cultures (p = 0.3; Student's t-test) ( Table 1). Lmna Dhe/+ cells also did not have a significantly different proportion of cells entering apoptosis, as determined by Annexin V labeling, an early marker of apoptosis (p = 0.9, Student's t-test; Table 1). However, fewer Ki-67 positive cells were found in Lmna Dhe/+ cultures as compared to Lmna +/+ cultures (p = 0.06, Student's t-test) ( Table 1). These results indicated that a smaller fraction of Lmna Dhe/+ fibroblasts was actively cycling, with no additional senescent cells or changes in apoptosis, as compared to Lmna +/+ cultures.
The slow growth rate indicative of other Lmna/LMNA mutations is often accompanied by increased DNA damage [19][20][21]. Indeed, Lmna Dhe/+ fibroblasts had numerous foci of the DNA damage marker, phosphorylated H2AFX (cH2AFX), present in both the main body of the nucleus as well as in blebbed regions, in marked contrast to Lmna +/+ cells ( Figure 7A-F). Western blotting indicated that Lmna Dhe/+ cells also expressed higher levels of phosphorylated TRP53 then Lmna +/+ cells ( Figure 7G). These data suggest that although the apoptotic pathway is intact to the point of TRP53 activation, it is not promoting increased apoptosis in Lmna Dhe/+ cells.

Mitotic chromosome cohesion defects in Lmna Dhe/+ fibroblasts
The extensive aneuploidy and DNA damage observed in Lmna Dhe/+ cells led us to examine whether these cells also had mitotic defects, which could further decrease mutant cell growth rates. First, we determined the mitotic indices of asynchronous cultures of Lmna +/+ and mutant fibroblasts. Interestingly, we noticed two-fold more Lmna Dhe/+ mitotic cells as compared to  Culture growth over four days measured by static adhesion assay. 3 Proportion of all cells positive for SA-b-gal fluorescence versus total cells counted by flow cytometry. 4 Cycling cells identified by Ki-67 immunostaining. 5 Apoptosis measured by Annexin V labeling and flow cytometry. 6 Mitotic indices determined from cells stained with anti-Lamin A and anti-a-Tubulin antibodies and DAPI.  Lmna +/+ cells (p#0.07, Student's t-test) ( Table 1). Next, we tested whether Lmna Dhe/+ fibroblasts arrested at a particular stage in mitosis. We found no significant difference in the proportions of prophase and prometaphase cells in Lmna +/+ and Lmna Dhe/+ cells. Lmna Dhe/+ cultures had an increased proportion of cells in anaphase compare to normal cells (20% vs. 13%, p = 0.07, x 2test; Table 2). Other stages of mitosis were not significantly affected ( Table 2). An anaphase delay in Lmna Dhe/+ cells suggests chromosome segregation defects. Recent work has shown that decreased levels of RB1 caused centromere condensation and cohesion defects in anaphase due to decreased expression of NCAP-D3, the centromere-specific condensin II subunit [23,[26][27][28]. To test whether this pathway was perturbed in Lmna Dhe/+ cells, we examined the levels of RB1 and NCAP-D3 by Western blotting. The levels of the active hypophosphorylated RB1 were lower in Lmna Dhe/+ cells than those found in Lmna +/+ cells ( Figure 8A). Lmna Dhe/+ fibroblasts also expressed lower levels of NCAP-D3 than Lmna +/+ cells ( Figure 8B), consistent with the decrease in hypophosphorylated RB1. We further determined whether a spindle checkpoint protein was present in Lmna Dhe/+ fibroblasts. Immunoblotting indicated little to no detectable levels of mitotic arrest deficient-like 1 (MAD2L1), a spindle checkpoint protein that is degraded upon activation of APC/C Cdc20 after proper spindle attachment ( Figure 8C). Reduced levels of NCAP-D3 and MAD2L1 may lead to chromosome segregation defects and aneuploidy in Lmna Dhe/+ cells. Consistent with this, we observed a significant increase in the number of anaphase bridges formed as compared to Lmna +/+ cells (p#0.01, x 2 -test; Figure 9A, C, E; Table 3). In addition, we noted an increase in the number of micronuclei in Lmna Dhe/+ (11%+/21.6) cells versus Lmna +/+ (1.3%+/21.5, p,0.001, x 2 -test, Table 3; Figure 9B, D, F). These features suggest that reduced levels of NCAP-D3 and MAD2L1 are associated with aneuploidy in Lmna Dhe/+ fibroblasts.

Discussion
Using the Lmna Dhe mouse model, which recapitulates some physiological phenotypes of HGPS and other non-muscle involved laminopathies, we show here that the slowed cell growth caused by Lmna mutation is in part due to mitotic defects that include loss of the centromere-specific condensin subunit, NCAP-D3, and a defective spindle checkpoint. The mitotic defects are consistent with mis-regulation of RB1 in Lmna Dhe/+ dermal fibroblasts and subsequent affects on sister chromatid condensation and cohesion through loss of NCAP-D3. Accordingly, the mitotically defective Lmna Dhe/+ fibroblasts exhibited pervasive aneuploidy with high chromosome copy numbers. In addition, mitotic defects and increased DNA damage in interphase also were detected in Lmna Dhe/+ cells. Thus, this study begins to fill in the missing link between the loss of RB1 activity in interphase and aberrant events in mitosis observed in other laminopathies.

Lmna Dhe as a model for HGPS and related laminopathies
The Lmna Dhe mouse model studied here exhibits physiological phenotypes similar to human progeria and related laminopathies characterized by lipodystrophy, cranial abnormalities,  and skin defects [29]. Here we show that the Lmna Dhe allele confers several of the cellular phenotypes observed in these diseases. These phenotypes include nuclear morphology defects, DNA damage, mitotic defects, and aneuploidy. Unlike HGPS patients, Lmna Dhe/+ mice do not age prematurely (L.R. Donahue, unpublished). Consistent with their normal lifespan, cells from Lmna Dhe/+ mice also did not senesce prematurely, although they did exhibit other cell cycle defects. Thus, Lmna Dhe may be a useful model for teasing apart mechanisms underlying the many cellular defects of HGPS and other non-muscle affected laminopathies. We specifically used this model to extend our understanding of mechanisms underlying cell cycle delay and aneuploidy in laminopathy cells.

Mitotic defects in Lmna Dhe/+ cells
Our data demonstrate that Lmna Dhe/+ fibroblasts grow slower than Lmna +/+ counterparts in part due to mitotic defects. Defects in mitosis have been reported previously for other LMNA mutant cells [6]. Here, we further show that Lmna Dhe/+ mitotic defects were accompanied by low levels of NCAP-D3 and hypophosphorylated RB1. Recent reports indicate that these two proteins interact in primary cells and tumor cell lines, and that loss of RB1 activity subsequently perturbs chromosome condensation, sister chromatid cohesion and the spindle checkpoint. Failed interactions between RB1 and NCAP-D3 in primary and tumor cells therefore result in loss of chromosome segregation fidelity, aneuploidy, micronuclei and anaphase bridges, all of which were observed in Lmna Dhe/+ cells as well [7,[13][14][15][19][20][21][22]24]. Whereas the low levels of hypophosphorylated RB1 reported for other Lmna/LMNA deficient cells have been implicated in altering the G1/S transition, our findings with Lmna Dhe/+ fibroblasts suggest that the influence of RB1 extends to mitosis as well [11,19,26,28,[39][40][41][42].
Consistent with our findings in Lmna Dhe/+ fibroblasts, gene expression data implicate a loss of the RB1 pathway in HGPS patient cells [16,17]. Furthermore, aneuploidy has been detected in a variety of HGPS and related laminopathies as well [22,24]. For example, a mouse model of HGPS that is deficient for ZMPSTE24 exhibits aneuploidy in up to 40% of cultured cells [22]. In human premature aging syndromes, up to 30% of patient cells exhibit aneuploidy [21,22,24]. While aneuploidy in itself is not a novel phenotype for LMNA HGPS and related alleles, the extent of aneuploidy present in Lmna Dhe/+ cells is unique. Aneuploidy may be further exacerbated in Lmna Dhe/+ cells by loss of proper spindle checkpoint proteins, such as MAD2L1.
Independent, but interconnected networks at the nuclear lamina LMNA and LMNB have been described as independent, but interconnected networks of proteins comprising the nuclear lamina [4]. We show here that the Lmna Dhe allele had profound effects on both the LMNA and LMNB networks. LMNA expression was curtailed in Lmna Dhe/+ fibroblasts, resulting in an overall lower density of LMNA throughout the nucleus. Reduction in LMNB density also was detected, but to a lesser extent. In contrast to the dominant mutation described here, siRNA knockdown of LMNA does not affect the LMNB network in HeLa cells [4]. However, knockdown of LMNB1 induces large holes in the LMNA and LMNB2 network, supporting interdependency of the networks [4]. Furthermore, this interdependency may be cell cycle stage specific. Our data suggest that LMNA can influence LMNB in interphase, but that the normal dissolution of LMNB in mitosis is unaffected by Lmna Dhe , which persists in mitosis.
In addition to decreases in the overall density of LMNA and LMNB in Lmna Dhe/+ fibroblasts, perturbations to nuclear lamina structure also were detected. Specifically, the LMNB distribution was skewed toward depletion from the majority of nuclear blebs. This redistribution was not found for LMNA. Furthermore, both LMNA and the residual LMNB networks exhibited large, irregular holes in blebs, indicating aberrant assembly of lamin filaments in these regions. We envision two ways in which these networks were perturbed: (1) LMNA Dhe might have influenced the nuclear membrane to form blebs, which then indirectly pulled the lamina into a less dense network in these regions, or (2) LMNA Dhe directly caused the perturbation of both LMNA and LMNB networks. Lmna Dhe is caused by a dominant mutation in a critical leucine of the first coiled coil domain, which is involved in filament formation. Thus, Lmna Dhe might form aberrant, irregular filaments in the nucleus. These possibilities are not necessarily mutually exclusive and await testing in future studies.