Chen P et al. (2019), Nucleoplasmin is a limiting component in the sc...

XB-ART-56397

J Cell Biol 2019 Oct 02;21812:4063-4078. doi: 10.1083/jcb.201902124.

Show Gene links Show Anatomy links

Nucleoplasmin is a limiting component in the scaling of nuclear size with cytoplasmic volume.

Chen P , Tomschik M , Nelson KM , Oakey J , Levy DL .

???displayArticle.abstract???
How nuclear size is regulated relative to cell size is a fundamental cell biological question. Reductions in both cell and nuclear sizes during Xenopus laevis embryogenesis provide a robust scaling system to study mechanisms of nuclear size regulation. To test if the volume of embryonic cytoplasm is limiting for nuclear growth, we encapsulated gastrula-stage embryonic cytoplasm and nuclei in droplets of defined volume using microfluidics. Nuclei grew and reached new steady-state sizes as a function of cytoplasmic volume, supporting a limiting component mechanism of nuclear size control. Through biochemical fractionation, we identified the histone chaperone nucleoplasmin (Npm2) as a putative nuclear size effector. Cellular amounts of Npm2 decrease over development, and nuclear size was sensitive to Npm2 levels both in vitro and in vivo, affecting nuclear histone levels and chromatin organization. We propose that reductions in cell volume and the amounts of limiting components, such as Npm2, contribute to developmental nuclear size scaling.

???displayArticle.pubmedLink??? 31636119
???displayArticle.pmcLink??? PMC6891103
???displayArticle.link??? J Cell Biol
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
GO keywords: chromatin [+]

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

	Figure 1. Cytoplasmic volume contributes to nuclear size scaling in X. laevis embryo extracts. (A) Top: Schematic diagram of the experimental approach. Stage 10â€“10.5 embryos were arrested in late interphase with cycloheximide. Embryonic extract containing endogenous nuclei was encapsulated in droplets using microfluidic devices. Nuclei were visualized by uptake of GFP-NLS. Bottom left: In vivo nuclear size scaling data for X. laevis stages 8 to 10.5 (JevtiÄ‡ and Levy, 2015). Bottom right: Blue and orange represent cytoplasm and nuclei, respectively. (B) Spherical extract droplets were incubated at room temperature. (C) Spherical droplet data. At each time point, 18â€“104 nuclei were quantified (57 nuclei on average). Each curve corresponds to a different extract. (D) Flattened droplet data. The ratio of the long axis to the short axis was on average âˆ¼1.7. At each time point, 10â€“117 nuclei were quantified (35 nuclei on average). Each curve corresponds to a different extract. (E) The fold change in nuclear volume was calculated by dividing maximum nuclear volume by initial nuclear volume at t = 0. Best-fit logarithmic regression curves are displayed. For each time point of each experiment, 7â€“311 nuclei were quantified (63 nuclei on average). Data are shown for 26 different spherical droplet volumes (blue) and 12 different flattened droplet volumes (orange), using 23 different extracts. Error bars represent SD.
	Figure 2. Cytoplasmic composition contributes to nuclear size scaling in X. laevis embryo extracts. (A) Left: Based on the data presented in Fig. 1, Câ€“E, maximum nuclear volume is plotted as a function of droplet volume for stage 10 embryo extract droplets (blue). Right: Nuclear-to-cytoplasmic (N/C) volume ratios were calculated by dividing maximum nuclear volume by droplet volume. A best-fit power regression curve is displayed for the droplet data. Also plotted are previously reported in vivo nuclear size scaling data and nuclear-to-cytoplasmic volume ratios for X. laevis stages 8 to 10.5 (JevtiÄ‡ and Levy, 2015). (B) Stage 10 nuclei were isolated and resuspended in cytoplasmic extract from different embryonic stages. Extract droplets were incubated at room temperature. (C) At each time point, 10â€“168 nuclei were quantified (80 nuclei on average). The stage 10 extract data are the same shown in Fig. 1 C. (D) Stage 10 extract and nuclei were mixed with stage 5 cytoplasmic extract or egg extract at a 1:10 ratio. After a 2-h incubation, nuclei were fixed and stained with NPC antibody mAb414. Nuclear CS areas were measured and the fold change was calculated relative to the preincubation nuclear size. At least 700 nuclei were quantified for each condition. Data from two independent experiments are shown. Two-tailed Studentâ€™s t tests assuming equal variances; ***, P â‰¤ 0.001. Error bars represent SD.
	Figure 3. Fractionation approach to identifying factors limiting for nuclear growth. (A) X. laevis egg extract was subjected to high-speed centrifugation to separate the extract into cytosol (Cyto), light membrane (LM), and heavy membrane (HM). Stage 10 nuclei were incubated with the indicated fractions for 120 min. IBB, added at âˆ¼30 ÂµM, inhibits nuclear import (Weis et al., 1996). Nuclear CS areas were measured for at least 200 nuclei per condition and normalized to the preincubation nuclear size. Data from two independent experiments are shown. (B) Experiments were performed as in Fig. 1, Câ€“E, except that extracts were supplemented with IBB or wheat germ agglutinin (WGA) to block nuclear import (Cox, 1992). For each experiment, 7â€“212 nuclei were quantified at each time point (53 nuclei on average). Data from three different extracts are shown. The control data are the same shown in Fig. 1 E. (C) Schematic diagram of the fractionation approach. See Materials and methods for details. (D) Mono Q fractions were dialyzed into XB and concentrated âˆ¼10- to 20-fold. Stage 10 embryo extract and nuclei were supplemented with equivalent volumes of XB or Mono Q fractions and incubated for 90 min. Nuclei were fixed and stained with NPC antibody mAb414. Nuclear CS areas were measured and the fold change was calculated relative to the preincubation nuclear size. Means and SDs from three independent fractionation experiments are shown. At least 160 nuclei were quantified for each condition. The dotted line indicates the top of the error bar for the XB control. We selected fractions with the largest fold changes in nuclear size and with SD error bars above the dotted line, namely, fractions 2 and 9, which were shown by mass spectrometry to contain Npm2. Error bars represent SD.
	Figure 4. Npm2 levels affect in vivo nuclear size. (A) One-cell X. laevis embryos were microinjected with equivalent volumes of XB, recombinant Npm2 protein (to increase the Npm2 concentration by 3.5 ÂµM) or recombinant Npm-core protein (3.5 ÂµM final), allowed to develop to stage 10â€“10.5, and arrested in G2 with cycloheximide. Isolated nuclei were stained with an NPC antibody (mAb414), and nuclear volumes were measured for at least 74 nuclei per condition per experiment. 25 embryos on average were microinjected per condition. Data from four independent experiments and representative images are shown. Representative images to the right show nuclei stained with an antibody against Npm2. (B) Microinjections were performed as in A without cycloheximide arrest and including, where indicated, importin-Î±-E mRNA (350 pg total, Imp Î±). Importin-Î±-E is a phosphomimetic version of human importin Î±2 with reduced affinity for membranes (Levy and Heald, 2010; Wilbur and Heald, 2013). At least 330 nuclei were quantified per condition per experiment. Data from two independent experiments and representative images are shown. Orange bars represent sizes for nuclei with Npm2 staining intensity values >1 SD above the XB control (see also panel C). The dotted horizontal line at 3.5 pl corresponds to the predicted stage 10 nuclear volume resulting from a 3.5-ÂµM introduction of Npm2 (see also Fig. S2 F). (C) Nuclei from microinjections described in B were stained with an antibody against Npm2. Nuclear volumes and total nuclear Npm2 staining intensities were measured for â‰¥330 nuclei per condition per experiment. 25 embryos on average were microinjected per condition, and two independent experiments were performed. For the Npm2 + importin Î± condition, individual nuclear volume was plotted as a function of nuclear Npm2 staining intensity. For XB-microinjected embryos, nuclear Npm2 staining intensity was 2.05 Â± 1.35 (average Â± SD). In panel B here and in Fig. S4 D, the orange bars represent data for nuclei with Npm2 staining intensity values greater than one SD above the XB control; therefore, nuclei with Npm2 staining intensities >2.05 + 1.35 = 3.4 (indicated by the open bracket on the x axis in C). (D) Equivalent volumes of egg and embryo extracts were analyzed by Western blot. Npm2 band intensities were normalized to Î²-tubulin levels and used to estimate the Npm2 concentration for each stage, given an egg extract Npm2 concentration of 4.2 ÂµM. (E) Embryonic nuclei were stained for Npm2, and total nuclear Npm2 staining intensities were quantified for â‰¥148 nuclei per stage. (F) Microinjections were performed as in A with 1 ng of the indicated Npm2 mRNAs or an equivalent volume of water as a control. Mutants 1 and 2 have mutations in the histone-binding region of Npm2 (Warren et al., 2017), and mutant 3 is defective in histone chaperone activity (Salvany et al., 2004). We verified that these mutants are nuclear-import competent (data not shown). At least 900 nuclei were quantified per condition per experiment. Data from two independent experiments and representative images are shown. All statistical comparisons are to the water control. Two-tailed Studentâ€™s t tests assuming equal variances; ***, P â‰¤ 0.001; ns, not significant. Error bars represent SD.
	Figure 5. Npm2 levels correlate with nuclear histone levels and altered chromatin topology. (A) Nuclei assembled in X. laevis egg extract for 30, 60, and 90 min were stained with an H2B antibody and Hoechst. Total nuclear H2B staining intensities were measured for â‰¥60 nuclei per condition and normalized to the 30-min time point. For each nucleus, two Hoechst intensity line scans were acquired through the middle of the nucleus. The SD of all intensity values along each line was calculated and normalized to the average intensity to obtain a value we term the chromatin heterogeneity index. Larger values correspond to a more heterogeneous chromatin distribution. 54â€“68 line scans were quantified per condition (60 on average). (B) Different stage nuclei were stained with an H2B antibody and Hoechst. Total nuclear H2B staining intensities were quantified for â‰¥100 nuclei per stage and normalized to stage 12. For chromatin heterogeneity indexes, 22â€“79 line scans were quantified per condition (45 on average). (C) Microinjection experiments were performed as in Fig. 4 A. Isolated nuclei were stained with an H2B antibody and Hoechst. Total nuclear H2B staining intensities were measured for â‰¥97 nuclei per condition and normalized to the XB-microinjected controls. For chromatin heterogeneity indexes, 34â€“52 line scans were quantified per condition (41 on average). Data from four independent experiments are shown. Two-tailed Studentâ€™s t tests assuming equal variances; , P â‰¤ 0.05; , P â‰¤ 0.01; **, P â‰¤ 0.001; ns, not significant. Error bars represent SD.
	Figure 6. Npm2 increases nucleosome occupancy and decreases euchromatin. (A) On the left, one-cell X. laevis embryos were microinjected with equivalent volumes of XB or recombinant Npm2 protein (to increase the Npm2 concentration by 3.5 ÂµM). On the right, one-cell X. laevis embryos were microinjected with 1 ng wild-type Npm2 mRNA or an equivalent volume of water. Embryos were allowed to develop to stage 10â€“10.5 and arrested in G2 with cycloheximide. Isolated nuclei were digested with MNase for the indicated times, and purified DNA was separated on 2% agarose gels and stained with ethidium bromide. On average, 60 embryos were microinjected per condition. Band intensities were quantified and used to calculate the average nucleosome number per fragment (see Materials and methods). Representative experiments are shown. (B) The MNase digestion assays described in A were repeated three times for Npm2 protein microinjection and three times for Npm2 mRNA microinjection. Maximum fold changes in nucleosome occupancy between control and Npm2-microinjected embryos were quantified for each experiment and averaged. (C) One-cell X. laevis embryos were microinjected with equivalent volumes of XB or recombinant Npm2 protein (to increase the Npm2 concentration by 3.5 ÂµM), allowed to develop to stage 10â€“10.5, and arrested in G2 with cycloheximide. On average, 25 embryos were microinjected per condition. Isolated nuclei were stained with an acetyl-histone H3 antibody, and representative images are shown. Total integrated nuclear acetyl-histone H3 staining intensity was measured for â‰¥168 nuclei per condition. One representative experiment of two is shown. Two-tailed Studentâ€™s t tests assuming equal variances; , P â‰¤ 0.01; *, P â‰¤ 0.001. Error bars represent SD.

References [+] :

Amodeo, Histone titration against the genome sets the DNA-to-cytoplasm threshold for the Xenopus midblastula transition. 2015, Pubmed, Xenbase