Edens LJ et al. (2017), PKC-mediated phosphorylation of nuclear lamins ...

XB-ART-54125

Mol Biol Cell 2017 May 15;2810:1389-1399. doi: 10.1091/mbc.E16-11-0786.

Show Gene links Show Anatomy links

PKC-mediated phosphorylation of nuclear lamins at a single serine residue regulates interphase nuclear size in Xenopus and mammalian cells.

Edens LJ , Dilsaver MR , Levy DL .

???displayArticle.abstract???
How nuclear size is regulated is a fundamental cell-biological question with relevance to cancers, which often exhibit enlarged nuclei. We previously reported that conventional protein kinase C (cPKC) contributes to nuclear size reductions that occur during early Xenopus development. Here we report that PKC-mediated phosphorylation of lamin B3 (LB3) contributes to this mechanism of nuclear size regulation. By mapping PKC phosphorylation sites on LB3 and testing the effects of phosphomutants in Xenopus laevis embryos, we identify the novel site S267 as being an important determinant of nuclear size. Furthermore, FRAP studies demonstrate that phosphorylation at this site increases lamina dynamics, providing a mechanistic explanation for how PKC activity influences nuclear size. We subsequently map this X. laevis LB3 phosphorylation site to a conserved site in mammalian lamin A (LA), S268. Manipulating PKC activity in cultured mammalian cells alters nuclear size, as does expression of LA-S268 phosphomutants. Taken together, these data demonstrate that PKC-mediated lamin phosphorylation is a conserved mechanism of nuclear size regulation.

???displayArticle.pubmedLink??? 28356420
???displayArticle.pmcLink??? PMC5426852
???displayArticle.link??? Mol Biol Cell
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: h2bc21 lmnb3 mtor ptk2

???attribute.lit??? ???displayArticles.show???

	FIGURE 1:. PKC-mediated phosphorylation of lamin B3 at S267 affects nuclear size in X. laevis embryos. (A) Schematic of the nuclear shrinking assay. Nuclei assembled in X. laevis egg extract are isolated by centrifugation and resuspended in cytoplasm derived from stage 11.5–12 X. laevis embryos (i.e., late-embryo extract). Nuclei incubated in late-embryo extract become smaller, whereas heat treatment of the extract or addition of the PKC inhibitor chelerythrine prevents nuclear shrinking. Nuclei were visualized by NPC staining (mAb414). Complete details of the nuclear shrinking assay are described elsewhere (Edens and Levy, 2014a, 2016). (B) Recombinant GFP-lamin B3 (GFP-LB3) was incubated with recombinant cPKC (Promega PepTag Non-Radioactive PKC Assay Kit) or two different late-embryo extracts for 30 min at 30°C. GFP-LB3 was isolated from the reactions using Ni-NTA beads (Qiagen), resolved on a SuperSep Phos-tag gel (Wako), and probed by Western blot using an anti-GFP antibody. (C) The shifted band from the cPKC reaction in B was excised from a Coomassie-stained gel (Supplemental Figure S2A) and subjected to phosphorylation-site mapping by mass spectrometry. The two sites with the highest Ascore values are shown in red, and complete results are in Supplemental Figure S2B. (D) The two putative phosphorylation sites identified in C are shown mapped onto the structural domains of LB3. (E–H) One-cell-stage X. laevis embryos were comicroinjected with morpholinos to knock down endogenous LB3 levels and equivalent amounts of mRNA (500 pg per embryo) expressing GFP-LB3 phosphorylation-site mutants. Stage 11.5–12 nuclei were isolated and quantified. For each sample, the cross-sectional nuclear areas of >160 nuclei were quantified, averaged, and normalized to LB3-wt (set at 1.0). (F) Cumulative LB3 phospho-null mutant data for three different fertilizations each with a minimum of 20 embryos. (G) Embryos microinjected as in F were treated with 6 nM PMA or DMSO for 90 min at room temperature before quantification. Cumulative data are shown for two different fertilizations each with a minimum of 25 embryos. (H) Cumulative LB3 phosphomimetic mutant data for two different fertilizations each with a minimum of 20 embryos. Bars, 10 µm. *p < 0.005; p < 0.01; N/S, not significant. Errors bars represent SD.
	FIGURE 2:. FRAP analysis of lamin B3 dynamics. (A) Nuclei were assembled in X. laevis egg extract at room temperature. At 30–40 min after initiating nuclear assembly, reactions were supplemented with 28 nM GFP-LB3. After an additional 60-min incubation, nuclei were isolated and resuspended in late-embryo extract as in Figure 1A. Where indicated, heat-inactivated late-embryo extract or late-embryo extract supplemented with 0.4 mM chelerythrine to inhibit PKC activity was used. Reactions were supplemented with an oxygen-scavenging mix. GFP-LB3 was imaged by confocal microscopy, and a 3.6-µm-radius circular spot on the surface of the nucleus was photobleached. After photobleaching, GFP-LB3 fluorescence recovery was detected by imaging every 5 s for a total of 120 s. Images from representative time lapses are shown (also see Supplemental Videos S1–S3). Bar, 10 µm. (B) For each FRAP time-lapse experiment, the mean GFP-LB3 fluorescence intensity of the photobleached region, entire nucleus, and a background region was measured at each time point. These data were analyzed using the easyFRAP application (Rapsomaniki et al., 2012) and subjected to double normalization. The first postbleach image was acquired at 10 s, and prebleach intensities were normalized to 100%. Five independent experiments were performed, and time lapses were acquired for 46 late-embryo extract nuclei, 48 heat-inactivated late-embryo extract nuclei, and 44 late-embryo extract plus chelerythrine nuclei. The mean normalized GFP-LB3 fluorescence intensity within the bleached regions and SEM are shown for each time point. (C, D) For each acquired time lapse, normalized GFP-LB3 fluorescence intensity measurements within the bleached region were fitted to a single exponential that was used to estimate the mobile fraction with easyFRAP (Rapsomaniki et al., 2012). The LB3-wt data with and without chelerythrine shown in D are the same data presented in C. In D, nuclei were assembled in X. laevis egg extract supplemented with the indicated GFP-LB3 phosphorylation-site mutants and then subjected to the nuclear shrinking assay and FRAP. We quantified 53 LB3-S267A, 55 LB3-S267E, and 33 LB3-S267E plus chelerythrine nuclei in four independent experiments. Note that estimated t1/2 recovery times generally supported our conclusions about wt and mutant LB3 dynamics; however, low mobile fractions prevent accurate determination of t1/2 values and so those data are not included. ***p < 0.001; N/S, not significant. Error bars represent SD.
	FIGURE 3:. Altering PKC activity in cultured mammalian cells influences nuclear size. Representative images for the cell lines, with visualization methods indicated to the left. HeLa cells stably express H2B-GFP, whereas the Ptk2 and MRC-5 nuclei were visualized either with Hoechst or H2B-mCherry introduced by cotransfection. Cross-sectional nuclear area was quantified for >150 nuclei per sample, averaged, and normalized to the relevant control (set at 1.0). (A) Nearly confluent cells grown on glass coverslips were treated with a PKC activator (6 nM PMA) or cPKC inhibitor (2 µm Gö 6976; inhibits PKC α and β) for 90 min. Cumulative data from four independent experiments. Gö 6976 was used for these experiments rather than chelerythrine, which is known to disrupt cell adhesion (Zimmerman et al., 2004; Tan et al., 2011). (B) Cells were transiently cotransfected with plasmids expressing constitutively active cPKC α or βII and H2B-mCherry. Cells were fixed and imaged, using H2B-mCherry fluorescence to identify transfected cells. Cumulative data from three independent experiments. (C) Cells were transiently cotransfected with siRNA against the indicated PKC isoforms and a plasmid expressing H2B-mCherry. Cells were fixed and imaged, using H2B-mCherry fluorescence to identify transfected cells. The PKC ζ isoform was used as a negative control, as its expression is exclusively neuronal. Cumulative data from three independent experiments. These siRNA oligos designed against human PKC sequences were ineffective in Ptk2 rat-kangaroo cells. Bars, 10 µm. *p < 0.005; p < 0.01; *p < 0.05; N/S, not significant. Errors bars represent SD.
	FIGURE 4:. Nuclear distribution of human lamin A is altered by a conserved PKC phosphorylation site. (A) Protein sequence alignment of Xenopus and human nuclear lamins. The Xenopus LB3-S267 site is highlighted in yellow and shows conservation with S268 in human LA. (B) Experiments were performed as in Figure 1B, using recombinant human LA tagged with GFP (GFP-LA) as the substrate. (C) HeLa cells were transiently transfected with the indicated GFP-LA expression plasmids, and nuclei were visualized with GFP 2 d after transfection. Representative nuclei with similar total GFP-LA intensities. Bar, 10 µm. (D) Line scans measuring GFP-LA intensity through nuclear cross sections were acquired using ImageJ. Line scans for seven to nine representative nuclei are shown in each graph, with each trace representing one nucleus. For this analysis, cells with similar GFP-LA expression levels were selected based on nuclear GFP-LA intensity quantification (unpublished data). (E) For LA-wt and LA-S268A, GFP-LA intensity line scans, like those shown in D, exhibited clear peaks at the NE. The widths of these peaks were measured at the GFP-LA intensity corresponding to the nucleoplasmic signal for a given nucleus. The widths of both peaks were measured for 40 nuclei and averaged for a given sample. Cumulative data from four independent experiments. **p < 0.01. Errors bars represent SD.
	FIGURE 5:. Phosphorylation of S268 on human lamin A influences nuclear size. HeLa cells were cotransfected with siRNA against endogenous LA and the indicated GFP-LA expression plasmids, and nuclei were visualized with GFP 2 d after transfection. The LA siRNA used here targets the 5′ UTR (Lund et al., 2013), reduced endogenous LA expression levels by 81% (Supplemental Figure S4C), and does not target plasmid-expressed GFP-LA. (A) Increasing levels of LA-wt, as determined by GFP-LA intensity, correlate with increased nuclear size. Bar, 10 µm. (B) Cross-sectional nuclear area as a function of total nuclear GFP-LA intensity. Each data point represents one nucleus. Data from one representative experiment, including 1011 GFP-LA-wt nuclei, 585 GFP-LA-S268A nuclei, and 526 GFP-LA-S268E nuclei. Best-fit linear regression lines are plotted. (C) Nuclei were divided into bins based on GFP-LA intensity levels. Cumulative data from four independent experiments. Each bin includes a minimum of 100 nuclei per condition. Within each bin, Student’s t tests were performed relative to LA-wt, and absence of asterisks indicates that the difference was not statistically significant. (D) Cumulative data from four independent experiments. More than 150 nuclei were quantified per condition and experiment, and nuclear size changes were normalized to LA-wt (set at 1.0). *p < 0.005; p < 0.01; *p < 0.05. Error bars represent SD.

References [+] :

Abdalla, Correlation of nuclear morphometry of breast cancer in histological sections with clinicopathological features and prognosis. 2009, Pubmed

	FIGURE 1:. PKC-mediated phosphorylation of lamin B3 at S267 affects nuclear size in X. laevis embryos. (A) Schematic of the nuclear shrinking assay. Nuclei assembled in X. laevis egg extract are isolated by centrifugation and resuspended in cytoplasm derived from stage 11.5–12 X. laevis embryos (i.e., late-embryo extract). Nuclei incubated in late-embryo extract become smaller, whereas heat treatment of the extract or addition of the PKC inhibitor chelerythrine prevents nuclear shrinking. Nuclei were visualized by NPC staining (mAb414). Complete details of the nuclear shrinking assay are described elsewhere (Edens and Levy, 2014a, 2016). (B) Recombinant GFP-lamin B3 (GFP-LB3) was incubated with recombinant cPKC (Promega PepTag Non-Radioactive PKC Assay Kit) or two different late-embryo extracts for 30 min at 30°C. GFP-LB3 was isolated from the reactions using Ni-NTA beads (Qiagen), resolved on a SuperSep Phos-tag gel (Wako), and probed by Western blot using an anti-GFP antibody. (C) The shifted band from the cPKC reaction in B was excised from a Coomassie-stained gel (Supplemental Figure S2A) and subjected to phosphorylation-site mapping by mass spectrometry. The two sites with the highest Ascore values are shown in red, and complete results are in Supplemental Figure S2B. (D) The two putative phosphorylation sites identified in C are shown mapped onto the structural domains of LB3. (E–H) One-cell-stage X. laevis embryos were comicroinjected with morpholinos to knock down endogenous LB3 levels and equivalent amounts of mRNA (500 pg per embryo) expressing GFP-LB3 phosphorylation-site mutants. Stage 11.5–12 nuclei were isolated and quantified. For each sample, the cross-sectional nuclear areas of >160 nuclei were quantified, averaged, and normalized to LB3-wt (set at 1.0). (F) Cumulative LB3 phospho-null mutant data for three different fertilizations each with a minimum of 20 embryos. (G) Embryos microinjected as in F were treated with 6 nM PMA or DMSO for 90 min at room temperature before quantification. Cumulative data are shown for two different fertilizations each with a minimum of 25 embryos. (H) Cumulative LB3 phosphomimetic mutant data for two different fertilizations each with a minimum of 20 embryos. Bars, 10 µm. *p < 0.005; p < 0.01; N/S, not significant. Errors bars represent SD.
	FIGURE 2:. FRAP analysis of lamin B3 dynamics. (A) Nuclei were assembled in X. laevis egg extract at room temperature. At 30–40 min after initiating nuclear assembly, reactions were supplemented with 28 nM GFP-LB3. After an additional 60-min incubation, nuclei were isolated and resuspended in late-embryo extract as in Figure 1A. Where indicated, heat-inactivated late-embryo extract or late-embryo extract supplemented with 0.4 mM chelerythrine to inhibit PKC activity was used. Reactions were supplemented with an oxygen-scavenging mix. GFP-LB3 was imaged by confocal microscopy, and a 3.6-µm-radius circular spot on the surface of the nucleus was photobleached. After photobleaching, GFP-LB3 fluorescence recovery was detected by imaging every 5 s for a total of 120 s. Images from representative time lapses are shown (also see Supplemental Videos S1–S3). Bar, 10 µm. (B) For each FRAP time-lapse experiment, the mean GFP-LB3 fluorescence intensity of the photobleached region, entire nucleus, and a background region was measured at each time point. These data were analyzed using the easyFRAP application (Rapsomaniki et al., 2012) and subjected to double normalization. The first postbleach image was acquired at 10 s, and prebleach intensities were normalized to 100%. Five independent experiments were performed, and time lapses were acquired for 46 late-embryo extract nuclei, 48 heat-inactivated late-embryo extract nuclei, and 44 late-embryo extract plus chelerythrine nuclei. The mean normalized GFP-LB3 fluorescence intensity within the bleached regions and SEM are shown for each time point. (C, D) For each acquired time lapse, normalized GFP-LB3 fluorescence intensity measurements within the bleached region were fitted to a single exponential that was used to estimate the mobile fraction with easyFRAP (Rapsomaniki et al., 2012). The LB3-wt data with and without chelerythrine shown in D are the same data presented in C. In D, nuclei were assembled in X. laevis egg extract supplemented with the indicated GFP-LB3 phosphorylation-site mutants and then subjected to the nuclear shrinking assay and FRAP. We quantified 53 LB3-S267A, 55 LB3-S267E, and 33 LB3-S267E plus chelerythrine nuclei in four independent experiments. Note that estimated t1/2 recovery times generally supported our conclusions about wt and mutant LB3 dynamics; however, low mobile fractions prevent accurate determination of t1/2 values and so those data are not included. ***p < 0.001; N/S, not significant. Error bars represent SD.
	FIGURE 3:. Altering PKC activity in cultured mammalian cells influences nuclear size. Representative images for the cell lines, with visualization methods indicated to the left. HeLa cells stably express H2B-GFP, whereas the Ptk2 and MRC-5 nuclei were visualized either with Hoechst or H2B-mCherry introduced by cotransfection. Cross-sectional nuclear area was quantified for >150 nuclei per sample, averaged, and normalized to the relevant control (set at 1.0). (A) Nearly confluent cells grown on glass coverslips were treated with a PKC activator (6 nM PMA) or cPKC inhibitor (2 µm Gö 6976; inhibits PKC α and β) for 90 min. Cumulative data from four independent experiments. Gö 6976 was used for these experiments rather than chelerythrine, which is known to disrupt cell adhesion (Zimmerman et al., 2004; Tan et al., 2011). (B) Cells were transiently cotransfected with plasmids expressing constitutively active cPKC α or βII and H2B-mCherry. Cells were fixed and imaged, using H2B-mCherry fluorescence to identify transfected cells. Cumulative data from three independent experiments. (C) Cells were transiently cotransfected with siRNA against the indicated PKC isoforms and a plasmid expressing H2B-mCherry. Cells were fixed and imaged, using H2B-mCherry fluorescence to identify transfected cells. The PKC ζ isoform was used as a negative control, as its expression is exclusively neuronal. Cumulative data from three independent experiments. These siRNA oligos designed against human PKC sequences were ineffective in Ptk2 rat-kangaroo cells. Bars, 10 µm. *p < 0.005; p < 0.01; *p < 0.05; N/S, not significant. Errors bars represent SD.
	FIGURE 4:. Nuclear distribution of human lamin A is altered by a conserved PKC phosphorylation site. (A) Protein sequence alignment of Xenopus and human nuclear lamins. The Xenopus LB3-S267 site is highlighted in yellow and shows conservation with S268 in human LA. (B) Experiments were performed as in Figure 1B, using recombinant human LA tagged with GFP (GFP-LA) as the substrate. (C) HeLa cells were transiently transfected with the indicated GFP-LA expression plasmids, and nuclei were visualized with GFP 2 d after transfection. Representative nuclei with similar total GFP-LA intensities. Bar, 10 µm. (D) Line scans measuring GFP-LA intensity through nuclear cross sections were acquired using ImageJ. Line scans for seven to nine representative nuclei are shown in each graph, with each trace representing one nucleus. For this analysis, cells with similar GFP-LA expression levels were selected based on nuclear GFP-LA intensity quantification (unpublished data). (E) For LA-wt and LA-S268A, GFP-LA intensity line scans, like those shown in D, exhibited clear peaks at the NE. The widths of these peaks were measured at the GFP-LA intensity corresponding to the nucleoplasmic signal for a given nucleus. The widths of both peaks were measured for 40 nuclei and averaged for a given sample. Cumulative data from four independent experiments. **p < 0.01. Errors bars represent SD.
	FIGURE 5:. Phosphorylation of S268 on human lamin A influences nuclear size. HeLa cells were cotransfected with siRNA against endogenous LA and the indicated GFP-LA expression plasmids, and nuclei were visualized with GFP 2 d after transfection. The LA siRNA used here targets the 5′ UTR (Lund et al., 2013), reduced endogenous LA expression levels by 81% (Supplemental Figure S4C), and does not target plasmid-expressed GFP-LA. (A) Increasing levels of LA-wt, as determined by GFP-LA intensity, correlate with increased nuclear size. Bar, 10 µm. (B) Cross-sectional nuclear area as a function of total nuclear GFP-LA intensity. Each data point represents one nucleus. Data from one representative experiment, including 1011 GFP-LA-wt nuclei, 585 GFP-LA-S268A nuclei, and 526 GFP-LA-S268E nuclei. Best-fit linear regression lines are plotted. (C) Nuclei were divided into bins based on GFP-LA intensity levels. Cumulative data from four independent experiments. Each bin includes a minimum of 100 nuclei per condition. Within each bin, Student’s t tests were performed relative to LA-wt, and absence of asterisks indicates that the difference was not statistically significant. (D) Cumulative data from four independent experiments. More than 150 nuclei were quantified per condition and experiment, and nuclear size changes were normalized to LA-wt (set at 1.0). *p < 0.005; p < 0.01; *p < 0.05. Error bars represent SD.