Lamina-associated polypeptide-1 interacts with the muscular dystrophy protein emerin and is essential for skeletal muscle maintenance

Abstract

X-linked Emery-Dreifuss muscular dystrophy is caused by loss of function of emerin, an integral protein of the inner nuclear membrane. Yet emerin null mice are essentially normal, suggesting the existence of a critical compensating factor. We show that the lamina-associated polypeptide1 (LAP1) interacts with emerin. Conditional deletion of LAP1 from striated muscle causes muscular dystrophy; this pathology is worsened in the absence of emerin. LAP1 levels are significantly higher in mouse than human skeletal muscle, and reducing LAP1 by approximately half in mice also induces muscle abnormalities in emerin null mice. Conditional deletion of LAP1 from hepatocytes yields mice that exhibit normal liver function and are indistinguishable from littermate controls. These results establish that LAP1 interacts physically and functionally with emerin and plays an essential and selective role in skeletal muscle maintenance. They also highlight how dissecting differences between mouse and human phenotypes can provide fundamental insights into disease mechanisms.

Figure 1. LAP1 and Emerin Co-immunoprecipitate from Cell Extracts

(A) Endogenous emerin was co-immunopreciptated with overexpressed V5-tagged LAP1 in HEK 293 cells. Protein extracts from un-induced (− DOX) or induced (+ DOX) cells were collected and immunoprecipitated (IP) with anti-V5 antibodies. Protein inputs (1% of total lysate) and precipitated products were separated by SDS-PAGE, transferred to nitrocellulose and probed with antibodies against V5, emerin, Man1, LAP2β and lamin A, the migrations of which are indicated. Migrations of molecular mass standards in KDa are indicated at the left of the blots. (B) Protein extracts from HEK 293 cells were incubated with anti-emerin (4G5), isotype matched anti-immunoglobulin G (IgG) and agarose beads with no antibody for co-immunoprecipitation. LAP1 co-immunoprecipitated with emerin but LAP2β did not as indicated in the immunoblot. Double band of emerin may be due to differential phosphorylation. Bands corresponding to IgG heavy and light chains are indicated and migrations of molecular mass standards in KDa are indicated at the left of the blots. (C) In vitro pull-down assay showed that LAP1 binds to the nucleoplasmic domain of emerin. Immunoblot shows the input and GST-LAP1 and V5-tagged LAP1 purified from HEK 293 cells fused to agarose beads pulled down purified MBP-emerin fusion protein but not purified MBP. Analysis of bacterially expressed MBP-emerin showed intact protein (*) and smaller fragments (bracket). Migrations of molecular mass standards in KDa are indicated at the left of the blots. (D) Schematic diagram of full-length human LAP1 (LAP1-FL), the nucleoplasmic domain plus transmembrane segment (LAP1-Nuc) and luminal domain plus transmembrane segment (LAP1-Lum) expressed in HEK 293 cells for the co-immunoprecipitation experiments shown in panel E. Red indicated V5 epitope and blue the transmembrane segment; amino acid residues 1, 330, 360 and 584 are indicated. The blue mesh indicates the relative position of the nuclear lamins and green rectangles the inner (INM) and outer (ONM) nuclear membranes [not to scale]. (E) Full-length LAP1 and the nucleoplasmic domain plus transmembrane segment bind emerin but the luminal domain plus the transmembrane segment does not. HEK 293 cells were transfected with vector only or plasmids expressing V5 tagged fusions of LAP1 as indicated. Protein extracts were collected and immunoprecipitated (IP) with anti-V5 antibodies. Protein inputs (1% of total lysate) and precipitated products were separated by SDS-PAGE, transferred to nitrocellulose and probed with antibodies against V5 or emerin as indicated. Migrations of molecular mass standards in KDa are indicated at the left of the blots. See also Table S1.

Figure 2. FRET Analysis of LAP1 and Emerin Association in Cells

(A) Representative images of a cell expressing RFP-emerin (Acceptor) and GFP-LAP1 (Donor). Images were taken pre (upper panel) and post (lower panel) photobleach of RFP-emerin in the boxed region of interest (ROI). In the two right panels (LAP1 and LAP1 Zoomed), intensities of GFP-LAP1 signals were mapped as colors in a rainbow spectrum from low and high fluorescence intensity as indicated; red signals demonstrate increased GFP-LAP1 fluorescence and show increased fluorescent intensity after a photobleaching. Bar: 5 μm. (B) Calculated FRET energy transfers between the protein pairs indicated. Values are means ± SEMs; n = 15, ***P<0.001. (C) Acceptor (RFP) fluorescence intensities prior to photobleaching (x-axis) plotted against energy transfers (y-axis) for the protein pairs indicated. Linear regression analysis of the plotted data sets of the indicated protein pairs showed no correlation between fluorescence intensities and FRET % energy transfers, confirming that the binding of protein pairs was specific.

Figure 3. Mislocalization of Emerin and Lamin A in LAP1 Null Fibroblasts

(A) Confocal immunofluorescence micrographs showing emerin localization in fibroblasts from wild type (WT) and LAP1 null mice. Cells were labeled with antibodies against emerin (green) and LAP1 (red) with co-localization showing as yellow (Overlay); in overlays counterstaining of nuclei with DAPI (blue) is also shown. Bar: 10 μm. (B) Confocal immunofluorescence micrographs showing the localization of emerin and other indicated proteins in LAP1 null fibroblasts. Cells were labeled with antibodies against emerin (green) and the other indicated protein (red) with co-localization showing as yellow (overlay). Bar: 10 μm. (C) Proteins in extracts from fibroblasts of 3 different LAP1 null and control mice were collected and subjected to immunoblot analysis using antibodies against LAP1, emerin, lamin A/C, lamin B1 and GADPH. Migrations of molecular mass standards in KDa are indicated at the left of the blots. (D) FRAP analysis of GFP-emerin in transfected wild type and LAP1 null mouse fibroblasts. Curve shows normalized fluorescence versus time after photobleaching an area of the nuclear envelope. Data at each measured time point are means ± SD (n = 20 for control cells; n = 22 for LAP1 null cells). See also Figure S1.

Figure 4. Striated Muscle-selective Deletion of LAP1 in Mice Produces Shortened Lifespan with Progressive Muscle and Body Mass Loss

(A) Immunoblots of protein extracts of lateral quadriceps muscle from normal adult humans and 6-week old wild type C57/B6 mice. Blots were probed with antibodies against emerin, LAP1 and GAPDH. Each lane is a sample from a different subject. (B) Quantification of emerin and LAP1 expression normalized to GAPDH in protein extracts of mouse (n = 6) and human (n = 4) muscle samples. Values are means ± SEM; ***P < 0.0001, **P < 0.005. (C) Photos of a littermate control (Tor1aip1^f/f) and an M-CKO (Mckcre^+/−;Tor1aip1^f/f) mouse at 12 weeks of age. (D) Body mass of male M-CKO and control (Tor1aip1^f/f) mice versus age. Values are means ± SEM; *P < 0.05, **P < 0.005, ***P < 0.0005. (E) Kaplan-Meier survival curves for male Tor1aip1^f/f control and male M-CKO mice. (F) Photos of skinned hindlimbs from a littermate control (Tor1aip1^f/f) and an M-CKO mouse at 16 weeks of age showing decreased muscle in the M-CKO mouse. (G) Muscle mass to tibia length ratio of quadriceps (Q), gastrocnemius medialis (Gs-m), tibialis anterior (Ti) and soleus (Sol) muscles of Tor1aip1^f/f control (n = 5) and M-CKO (n = 7) mice at 16 weeks of age. Values are means ± SEM; ***P < 0.0005. (H) Serum creatine phosphokinase (CPK) activities in Tor1aip1^f/f control (n=11) and M-CKO (n=8) mice at 8–9 weeks of age. The upper limit cut-off of the assay is 1,200 U/L. Values for each individual control (circles) and M-CKO (triangles) mouse are given and the horizontal bars are the mean values; *P < 0.05. (I) Grip strength (kg force per kg) in Tor1aip1^f/f control (n = 8) and M-CKO (n = 6) mice at 8–9 and 11–12 weeks of age. Values are means ± SEM; **P < 0.005, ***P < 0.0005. See also Figure S2.

Figure 5. Histopathological Analysis of Skeletal Muscle from M-CKO Mice

(A) Representative hematoxylin and eosin-stained cross and longitudinal sections of quadriceps muscle from Tor1aip1^f/f control and M-CKO mice at 9 (9W) and 12(12W) weeks of age (3 mice were analyzed per each group). Black arrows indicate degenerative/necrotic fibers, black arrowheads indicate central nuclei and white arrowheads indicate myonuclei arrays. Bars: 50 μm. (B) Mean ± SEM of cross sectional fiber areas in quadriceps from Tor1aip1^f/f control and M-CKO mice at 6 (6W), 9 (9W) and 12 (12W) weeks of age. Three different regions of each section from three different animals per group (n = 9) were analyzed. Values are means ± SEM; *P< 0.01; **P< 0.001. (C) Quantification of central nuclei in sections of quadriceps from Tor1aip1^f/f control and M-CKO mice at 9 (9W) and 12 (12W) weeks of age. Three different regions from each section were counted (3 sections per animal; 3 animals per each age group). Note that about 40% of myofibers have central nuclei in quadriceps of 12-week old M-CKO mice. Values are means ± SEM; ***P< 0.0001. (D) Representative micrographs of sections of quadriceps muscle stained with Masson’s trichrome from Tor1aip1^f/f control mice (Control-16W) and M-CKO mice (M-CKO-16W) at 16 weeks of age. Areas of fibrosis stain blue; Bar: 50 μm. (E) Hue histograms of micrographs in Panel D in which colors are represented from 0–255 units (blue 140–200 units) versus area of tissue section. Y-axis gives area (pixels) and X-axis the color spectrum with red color for muscle tissue and blue color (140–200 units) for connective tissue. (F) Percentage of fibrosis per surface area of muscle. Sections of quadriceps from 16 week-old Tor1aip1^f/f mice (Control-16W) and M-CKO mice were stained with Masson’s trichrome measured as described in previous panel. Three different regions from each section from three different animals per group (n = 9) were analyzed. Values are means ± SEM; **P < 0.005. (G) Representative electron micrographs of sections of quadriceps muscle from Tor1aip1^f/f control mice (Control-12W) and M-CKO mice (M-CKO-12W) at 12 weeks of age. Bars: 2 μm. White arrowhead indicates an enlarged nucleus, black bracket indicates dispersed glycogen particles and black arrows indicate enlarged mitochondria in muscle sections from a M-CKO mouse. See also Figure S3.

Figure 6. Analysis of Mice with Liver-selective Deletion of LAP1

(A) Photos of a littermate control (Tor1aip1^f/f) and L-CKO (Albcre^+/−;Tor1aip1^f/f) mouse at 16 weeks of age. (B) Body mass of male L-CKO and control (Tor1aip1^f/f) mice versus age. Values are means ± SEM; differences between genotypes were not significant. (C) Photos of livers from a littermate control (Tor1aip1^f/f) and an L-CKO mouse at 16 weeks of age. (D) Liver to body mass ratio of Tor1aip1^f/f control (n = 6) and L-CKO (n = 10) mice at 16 weeks of age. Values are means ± SEM; NS = not statistically significant. (E) Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities and albumin and bilirubin concentrations in Tor1aip1^f/f control littermates (n = 9) and L-CKO (n = 9) mice at 16 weeks of age and normal reference ranges for these parameters in the Comparative Pathology Laboratory at Columbia University Medical Center. Values are means ± SEM. (F) Representative hematoxylin and eosin-stained sections of liver from Tor1aip1^f/f control mice and L-CKO mice at 16 weeks of age; Bar: 50 μm. (G) Bar graph shows mean ± SEM of nuclear area per hepatocyte for Tor1aip1^f/f control mice and L-CKO mice at 16 weeks of age; n = 100, **P < 0.005. (H) Kaplan-Meier survival curves for male Tor1aip1^f/f control and male L-CKO mice. Mice were monitored up to 690 days of age; one mouse per each group died at ages greater than 400 days. (I) Immunoblots of protein extracts of skeletal muscle (Sk. Muscle) and liver from 6-week old wild type C57/B6 mice. Blots were probed with antibodies against emerin and LAP1. Ponceau S-stained nitrocellulose sheet shows similar amounts of protein loading (20 μg of protein/lane) for each sample. Migrations of molecular mass standards in KDa are indicated at the left. Signal intensity was measured and averaged for emerin and LAP1 bands (bottom panel). Values are means ± SEM. ** P < 0.005. (J) Immunoblots of protein extracts of human skeletal muscle (Sk. Muscle) and liver. Blots were probed with antibodies against emerin and LAP1. Ponceau S-stained nitrocellulose sheet shows similar amounts of protein loading (20 μg of protein/ lane) for each sample. Migrations of molecular mass standards in KDa are indicated at the left. Signal intensity was measured and averaged for emerin and LAP1 bands (bottom panel). Values are means ± SEM. *** P < 0.0001. See also Figure S4.

Figure 7. Shortened Lifespan and More Severe Myopathy in Mice with Combined Striated Muscle-selective LAP1 and Emerin Depletion

(A) Kaplan-Meier survival curves for male M-CKO, M-CKO;Emd^−/y and Tor1aip1^f/f;Emd^−/y (Control;Emd^−/y) mice. Log-rank test was used to compare survival curves. The difference between M-CKO and M-CKO;Emd^−/y mice was statistically significant (P<0.0001). (B) Representative hematoxylin and eosin-stained transverse and longitudinal sections of quadriceps from male M-CKO and M-CKO;Emd^−/y mice. Bar: 50 μm (C) Myopathy severity scores of histopathological examinations for skeletal muscle from male M-CKO and M-CKO;Emd^−/y mice. Individual scores for samples from 7 male M-CKO and 7 M-CKO;Emd^−/y mice and mean ± SEM score given by two independent, blinded neuromuscular pathologists are shown. P-values for severity score differences between M-CKO and M-CKO;Emd^−/y mice of each pathologist are shown. (D) Hematoxylin and eosin-stained cross and longitudinal sections of quadriceps muscle from Control;Emd null (Tor1aip1^f/f;Emd null) and M-CKO het;Emd null (Mck cre^+/−;Tor1aip1^f/f;Emd null) mice at their 10 weeks of ages. Muscle sections from n=8 mice per each group were blindly examined by a pathologist. The sections from Control;Emd null group showed no pathology, while the sections from two out of eight M-CKO het;Emd null mice displayed occasional degenerative/necrotic fibers as indicated by arrowheads. (E) Serum creatine phosphokinase (CPK) activities in Control;Emd null (n=12) and M-CKO het;Emd null mice (n=15) at their 8–11 weeks of ages. The upper limit cut-off of the assay is 1,200 U/L. Values for each individual Control;Emd null (circles) and M-CKO het;Emd null (triangles) mouse are given and the horizontal bars are the mean values; *P <0.05.