International review of cell and molecular biology, volume 322

Functional Significance of Nuclear Size and Morphology 33 4.1 Chromosome Positioning, Chromatin Organization, and Gene Expression 33 International Review of Cell and Molecular Biology, V

I NTERNATIONAL R EVIEW OF CELL AND MOLECULAR

BIOLOGY

International Review of Cell and Molecular Biology

Editorial Advisory Board

ROBERT A BLOODGOOD MICHAEL MELKONIAN

WALLACE F MARSHALL

I NTERNATIONAL R EVIEW OF CELL AND MOLECULAR BIOLOGY

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO

SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier

50 Hampshire Street, 5thFloor, Cambridge, MA 02139, USA

525 B Street, Suite 1800, San Diego, CA 92101-4495, USA

125 London Wall, London EC2Y 5AS, UK

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

Copyright © 2016 Elsevier Inc All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing As new research and experience broaden our understanding, changes in research methods, profes- sional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein In using such information or methods they should

be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors,

or editors, assume any liability for any injury and/or damage to persons or property

as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the mate- rial herein.

ISBN: 978-0-12-804809-2

ISSN: 1937-6448

For information on all Academic Press publications

visit our website at http://store.elsevier.com/

Fiorenza Accordi

Department of Biology and Biotechnology Charles Darwin, Sapienza University of Rome, Italy

Jaap D van Buul

Department of Molecular Cell Biology, Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, The Netherlands

Department of Molecular Biology, University of Wyoming, Laramie, WY, United States of America

ix

Cortney Chelise Winkle

Neuroscience Center and Curriculum in Neurobiology, University of North Carolina, Chapel Hill, NC, United States of America

Huajiang Xiong

Hull York Medical School, University of Hull, Hull, United Kingdom

New Insights into Mechanisms

and Functions of Nuclear Size

Regulation

Lidija D Vuković, Predrag Jevtić, Lisa J Edens, Daniel L Levy*

Department of Molecular Biology, University of Wyoming, [18_TD$DIFF]Laramie, WY, United States of America

*Corresponding author E-mail: dlevy1@uwyo.edu

4 Functional Significance of Nuclear Size and Morphology 33 4.1 Chromosome Positioning, Chromatin Organization, and Gene Expression 33

International Review of Cell and Molecular Biology, Volume 322

1

in the regulation of cell and nuclear function and speculate on the potential functional significance of nuclear size in chromatin organization, gene expression, nuclear mechanics, and disease With many fundamental cell biological questions remaining

to be answered, the field of nuclear size regulation is still wide open.

1 INTRODUCTION

Cell and nuclear sizes differ greatly among different species, as well aswithin the same organism when comparing different cell types Even in thesame tissue, cell and nuclear sizes can vary depending on the developmentalstage, state of cell differentiation, a variety of external factors, and cellulartransformation How nuclear size and shape affect cell physiology is stillunclear, but it is certainly possible that nuclear morphology impacts chro-matin organization and gene expression Elucidating the functional signifi-cance of nuclear size necessitates an understanding of the mechanisms thatcontrol nuclear size This is particularly important in the case of pathologies

in which nuclear morphology is altered, most notably cancer, where it isunclear if changes in nuclear size and shape are a cause or consequence ofdisease

In this review, we first provide a general overview of the diverse cellularstructures and activities that are relevant to the regulation of nuclear size,

including ploidy, chromatin condensation, the nucleocytoplasmic ratio,and nuclear transport (Fig[21_TD$DIFF] 1) We next turn to different model systems thathave and will continue to shed light on mechanisms of nuclear size regulation(Fig[21_TD$DIFF] 2) We review how nuclear size is regulated by soluble transport factors,structural components of the nuclear envelope[22_TD$DIFF](NE), signaling pathways, andextranuclear structures like the endoplasmic reticulum and cytoskeleton.Lastly, we discuss demonstrated or proposed roles for nuclear size in chro-mosome organization, gene expression, nuclear mechanics, and pathology.

In the last decade, the great complexity of nuclear structure and function hasbegun to emerge, and here we provide a broad overview of how the regu-lation of nuclear morphology contributes to this complexity

2 OVERVIEW OF CELLULAR STRUCTURES AND

ACTIVITIES THAT CONTRIBUTE TO NUCLEAR SIZE

DETERMINATION

Organelle size and morphology must have important implications fororganelle and cellular function (Edens et al., 2013; Heald and Cohen-Fix,2014; Jevtic et al., 2014) Understanding these functional implications firstrequires elucidation of the mechanisms that control organelle size When itcomes to nuclear size regulation, a number of potential mechanisms havebeen implicated including structural components of the nucleus, effects ofDNA amount and chromatin compaction, signaling and transport pathwaysthat impinge on the nucleus, extranuclear structures, and cell cycle state Inthis section, we provide a general overview of these diverse mechanisms thatcontribute to nuclear size control Each subsection deals with a generaltheme relevant to the regulation of nuclear size At the end of each subsec-tion, we refer to later sections in the review that delve into greater detail withrespect to that mechanism

2.1 Nuclear[23_TD$DIFF]Structure and Models of Organelle Size Control

The NE is composed of a double lipid bilayer The outer nuclear membrane(ONM) is a continuous extension of the endoplasmic reticulum (ER) Inmetazoan nuclei, the inner nuclear membrane (INM) is lined on its nucle-oplasmic face by the nuclear lamina, a meshwork of intermediate laminfilaments The INM and ONM are fused at sites of nuclear pore complex(NPC) insertion The NPC mediates nucleocytoplasmic transport of pro-teins and mRNA [24_TD$DIFF]Linker of nucleoskeleton and cytoskeleton (LINC)

[(Figure_1)TD$FIG]

[3_TD$DIFF]2, the ploidy example demonstrates that a 16-fold increase in ploidy (bottommost cell) in S pombe does not affect nuclear size [4_TD$DIFF][Image was adapted from (Neumann and Nurse, 2007), made available by creativecommons.org/licenses/by-nc-sa [5_TD$DIFF]/3.0/.] The cell cycle example shows that when REEP3/4 are knocked down, membrane fails to be cleared from metaphase chromosomes, leading to nuclear morphology defects in the subsequent interphase [4_TD$DIFF][Image was used with permission from ( Schlaitz et al., 2013 [6_TD$DIFF])] The signaling image shows that in Xenopus, nuclear cPKC levels are low in early development (a) and high later in development (b), correlating with reductions in nuclear size [4_TD$DIFF][Image was adapted from (Edens and Levy, 2014a), made available by creativecommons.org/licenses/by-nc-sa [7_TD$DIFF]/3.0/.] For section 3, the budding yeast example shows electron microscopy images of two different size wild-type G1 cells generated by varying the growth conditions Nuclei are outlined Scaling of nuclear size with cell size is evident [4_TD$DIFF][Image was used with permission from ( Jorgensen et al., 2007 [6_TD$DIFF])] In the Xenopus example, cell and nuclear sizes become smaller during progression from early to later stages of development [4_TD$DIFF][Image was used with permission from ( Jevtic and Levy,

2015 [6_TD$DIFF])] In the mammalian example, U2OS tissue culture cells overexpressing the ER-tubule shaping protein Rtn4 exhibit reduced nuclear expansion rates and smaller nuclei [8_TD$DIFF][The image was adapted from ( Anderson and Hetzer, 2008), made available by creativecommons.org/ licenses/by-nc-sa[9_TD$DIFF]/3.0/.] For section 4, the laminopathy example demonstrates that nuclear morphology is highly disrupted in HGPS patients (bottom nucleus) (Scaffidi et al., 2005), made available by creativecommons.org/licenses/by/4.0/ The cancer example shows our unpublished data in which nuclei are enlarged in metastatic melanoma cells compared to normal melanocytes In the cell migration example, HT1080 fibrosarcoma cells are shown migrating through microfluidic channels of different dimensions The shape of the nucleus must change to pass through narrow channels [8_TD$DIFF][The image was used with permission from (Denais and Lammerding, 2014[10_TD$DIFF])]

Figure 2 Model organisms and factors that control nuclear size [2_TD$DIFF]The left column shows model organisms that have been used to elucidate mechanisms of nuclear size regulation The other columns list how manipulating specific proteins and structures affects nuclear size and morphology The relevant references are included throughout the text.

complexes span the NE establishing connections between the cytoskeletonand nuclear interior (Rothballer and Kutay, 2012a,[25_TD$DIFF]b; Simon and Wilson,2011; Wilson and Berk, 2010) The usual shape of the nucleus is spherical orellipsoid (Walters et al., 2012), and in some cases extensions of the NE reachinto the interior of the nucleus in the form of a nucleoplasmic reticulum(Malhas et al., 2011) (Fig[21_TD$DIFF] 1).

A number of different models for organelle size control have been posed, and many of these may be relevant to the nucleus (Chan and Marshall,2012; Marshall, [26_TD$DIFF]2002, 2012) One model is that the sizes of individualcomponents of the structure act as rulers to dictate the overall size of thestructure (Tskhovrebova and Trinick, 2012) Such a ruler mechanism hasbeen demonstrated for measuring the length of cilia and flagella

(McCloskey et al., 2012) Another model for organelle size control is one

in which fixed amounts of organellar building blocks determine its ultimatesize (Goehring and Hyman, 2012) These limiting component models arerelevant to size control of the mitotic spindle (Good et al., 2013; Hazel et al.,

2013) and Golgi (Ferraro et al., 2014; Romero et al., 2013) Regulatedsynthesis of organelle structural components can also control size, as in thecase of lipid droplets (Wilfling et al., 2013)

More dynamic mechanisms for organelle size control have also beenproposed, for instance[27_TD$DIFF], balancing rates of organelle assembly and disassemblythat determine steady-state size Examples of size control that fit this modelinclude flagella (Ludington et al., 2012) and peroxisomes (Mukherji andO’Shea, 2014), and likely most membrane-bound organelles In [28_TD$DIFF]case ofthe nucleus, invoking several of these models may be necessary to fullyaccount for nuclear size regulation, and we will touch on these modelsthroughout the review Also see Sections[29_TD$DIFF]3.2–3.5, 3.7, 3.8, and4.1–4.4

2.2 Genome Size and Ploidy

The correlation between genome and nuclear size has been known for over acentury (Gregory, 2001, 2011; Umen, 2005) It is therefore tempting tospeculate that nuclear size is determined by the amount of nuclear DNA.However, abundant phenomenological and experimental evidence demon-strates that other factors must contribute to nuclear size Different cell typeswithin the same species exhibit nuclear size differences, despite having thesame genome content, and nuclear size varies during early development

while the DNA amount per cell remains constant (Altman and Katz, 1976;Butler et al., 2009; Faro-Trindade and Cook, 2006; Oh et al., 2005;

has a minimal impact on nuclear size In fission yeast, a 16-fold increase innuclear DNA amount did not affect nuclear size (Neumann and Nurse,

2007) (Fig[21_TD$DIFF] 1) Furthermore, an abrupt increase in nuclear size was notobserved at the time of DNA replication as might be expected if DNAamount significantly impacted nuclear size (Jorgensen et al., 2007;

mam-malian tissue culture cells expanded normally during interphase even whenDNA replication was blocked (Maeshima et al., 2010)

Nonetheless, ploidy changes can have important implications for cellularfunction Programmed polyploidization in mammalian cells is an adaptiveresponse to stress and injury (Pandit et al., 2013) In Drosophila melanogaster,polyploidization of glia is necessary to maintain integrity of the blood-brainbarrier (Unhavaithaya and Orr-Weaver, 2012) and plays a role in wound healing

in the adult epithelium (Losick et al., 2013) In different diatom species, genomesize correlates with cell division rates (Sharpe et al., 2012), and altered ploidy insalamanders impacts cell and animal size (Fankhauser, 1939, 1945a, b) It isunknown whether these functional effects might be mediated through changes

in nuclear size Also see Sections3.1, 3.2, 3.5, 3.7[207_TD$DIFF], and4.3

2.3 Chromatin[208_TD$DIFF]State

In addition to DNA amount, chromatin compaction is another feature thatpotentially impacts nuclear size and morphology The large number ofproteins known to interact with and modify chromatin complicates thisquestion (Kustatscher et al., 2014), although roles for condensins and his-tones have emerged For example, increasing condensin II-mediated chro-matin compaction in Drosophila caused distortion of NE morphology(Buster et al., 2013) An analysis of 160 eukaryotic genomes showed that

as genome size increased during evolution, the amino terminus of histoneH2A has acquired arginine residues that confer increased chromatin com-paction Addition of arginine residues to the yeast H2A resulted in increasedchromatin compaction and reduced nuclear volume, while mutating argi-nine residues in human H2A led to chromatin decompaction and increasednuclear size (Macadangdang et al., 2014)

It is worth [31_TD$DIFF]noting that chromatin compaction might also indirectlyimpact nuclear size Yeast cells increase compaction of long chromosome

arms during mitosis to ensure complete chromosome segregation(Neurohr et al., 2011), whereas Drosophila cells transiently elongate duringanaphase (Kotadia et al., 2012) In both instances, this might affect nuclearsize in the subsequent interphase Histone H3 methylation status has beenshown to dictate chromatin regions that associate with the nuclear lamina, socalled lamina-associated domains (LADs) (Harr et al., 2015), and certain longnoncoding RNAs regulate histone methylation (Wang et al., 2011b).Chromatin organization might, in turn, affect nuclear size Also seeSections3.1, 3.2, 3.4, 3.7, 3.8, 4.1[32_TD$DIFF], and4.3.

2.4 Cell Size and Nucleocytoplasmic Ratio

It has long been recognized that the nuclear-to-cytoplasmic (N/C) volumeratio is maintained at a roughly constant value in normal cells (Wilson, 1925),and this ratio is often perturbed in cancer cells (Chen et al., 2010; Hokamp

that nuclear size is dynamically sensitive to cytoplasmic volume When henerythrocytes were fused with HeLa cells, the erythrocyte nuclei grew largerand became transcriptional active (Harris, 1967) Somatic nuclei injectedinto Xenopus eggs or oocytes also expanded, with nuclei exposed to largercytoplasmic volumes enlarging more (Gurdon, 1976; Merriam, 1969).Manipulating cytoplasmic partitioning in sea snail embryos demonstratedthat nuclei within larger cytoplasmic volumes were larger than nuclei withinsmall cytoplasmic volumes (Conklin, 1912) Yeast studies have also shownthat nuclear size is sensitive to cytoplasmic volume (Jorgensen et al., 2007;

sensing and regulating the N/C ratio are not yet fully understood

An equally important question in the context of the N/C ratio is how cellsize is controlled In principle, the two relevant parameters are cell growth rateand cell division rate What are the mechanisms responsible for sensing cell size(Umen, 2005)? In fission yeast, two models have emerged By one model, agradient of Pom1, a cell polarity kinase located at the cell ends, acts as a sensor

of cell size As cells elongate, Pom1 levels decrease at the center of the cell andupon reaching a critical low level, mitosis is induced (Martin and Berthelot-Grosjean, 2009; Moseley et al., 2009) In another model, total cell surface area

is sensed by Cdr2, a peripheral membrane kinase (Pan et al., 2014) In buddingyeast, accumulation of G1 cyclins appears to act as a proxy for cell size (Cross,1988; Nash et al., 1988; Zapata et al., 2014) As in budding yeast, cyclinexpression controls the number of times erythroid precursors divide during

differentiation, influencing erythrocyte size and number (Sankaran et al., 2012).The situation in mammalian cells is likely more complex than in yeasts (Echave

et al., 2007; Kafri et al., 2013; Tzur et al., 2009)

Cell size certainly has important implications for cell function that mightinfluence nuclear size For instance, RNA and protein synthesis tend to scalewith cell size (Kempe et al., 2015; Marguerat and Bahler, 2012; Sato et al.,

transcription factors levels and the action of cytoskeletal motor proteins(Albus et al., 2013) Wnt signaling concomitantly increases cellular proteincontent and cell size (Acebron et al., 2014) A genetic screen in Jurkat cellsidentified Largen as a protein that increases cell size through increasedexpression of mitochondrial proteins (Yamamoto et al., 2014) Conversely,experimentally increasing cell size in mouse hepatocytes led to a reduction inmitochondrial gene expression (Miettinen et al., 2014)

Cell size and rates of cell division are also coupled to developmentalmorphogen gradients, environmental signals, and growth conditions

cytokinesis rates appear to be independent of cell size, due to the fact thatlarger contractile rings constrict faster than smaller ones (Calvert et al., 2011;Carvalho et al., 2009; Turlier et al., 2014) Active spindle positioning andplasma membrane expansion mechanisms ensure that cell division occurs inthe middle of the cell to generate daughter cells of equal size (Kiyomitsu andCheeseman, 2013) Cell shape control is also critical for cell function Failurefor cells to round at mitosis causes defects in spindle assembly and mitoticprogression (Cadart et al., 2014; Lancaster et al., 2013; Ramanathan et al.,

2015), and the ability of cells to switch between discrete cell shapes appears to

be under genetic control (Yin et al., 2013) Open questions remain abouthow cell size and shape impinge on the control of nuclear morphology Alsosee Sections[33_TD$DIFF]3.2–3.5, 3.7, 3.8, and4.3

2.5 Nucleocytoplasmic[34_TD$DIFF]Transport

One mechanism that contributes to the regulation of nuclear size in a variety

of systems is nucleocytoplasmic transport Classical nuclear import of proteinscontaining a nuclear localization signal (NLS) is mediated by karyopherins

of the importin α/β families Expression of different importin isoformscontrols nuclear import of specific cargo molecules Nuclear import beginswhen NLS cargos interact either directly with importin β or indirectlythrough association with an importinα adapter Importin β interacts with

nucleoporins (Nups) of the NPC that contain phenylalanine/glycine (FG)repeats Once within the nucleus, the importin/cargo complex dissociates

nucleus Ran is a small monomeric GTPase critical to[39_TD$DIFF] directional cytoplasmic transport Intranuclear Ran is predominantly GTP boundbecause the Ran guanine nucleotide exchange factor (RanGEF), RCC1,

nucleo-is bound to the chromatin In nuclear export, proteins containing a nuclearexport signal (NES) complex with an exportin, such as CRM1, and Ran-GTP for directed transport to the cytoplasm through the NPC

local-ized to the cytoplasmic surface of the NPC, so upon export Ran’s boundnucleotide is hydrolyzed to GDP and the exported cargo is released in thecytoplasm (Alberts et al., 1994)

Dedicated transport factors play important roles in regulating the cytoplasmic distribution of the importins and Ran After a round of nuclearimport, CAS is responsible for recycling importinα back to the cytoplasm foradditional cycles of import NTF2 is a protein that largely associates with theNPC and is responsible for recycling of cytoplasmic Ran-GDP back into thenucleus, where is it converted to Ran-GTP by RCC1 (Bayliss et al., 1999;Clarkson et al., 1997; Morrison et al., 2003; Smith et al., 1998) Kinetics ofnuclear import based on cargo size and competition may have importantimplications for the regulation of nuclear size (Feldherr et al., 1998; Hodel

nucleo-et al., 2006; Hu and Jans, 1999; Lane nucleo-et al., 2000; Lyman nucleo-et al., 2002; Mincer andSimon, 2011; Timney et al., 2006) It is worth [31_TD$DIFF]noting that many nuclearproteins do not possess a canonical NLS, and importin-independent pathwayshave recently been elucidated (Lu et al., 2014) Similarly, some proteins of theINM reach the interior of the nucleus through importin-independent pathways(Boni et al., 2015; Katta et al., 2013; Ungricht et al., 2015), and large ribonu-cleoprotein complexes are able to leave the nucleus by budding of the NE,bypassing the NPC entirely (Speese et al., 2012) Redistribution of nucleartransport factors that occurs during differentiation and stress could contribute toconcomitant changes in nuclear size (Andrade et al., 2003; Huang and Hopper,2014; Kose et al., 2012; Lieu et al., 2014; Rother et al., 2011; Whiley et al.,2012; Yasuda et al., 2012) Also see Sections[40_TD$DIFF]3.1–3.4, 3.7, and4.3

2.6 Intranuclear [41_TD$DIFF]Structures

Related to nuclear size is the question of how the sizes of intranuclearstructures are determined and whether nuclear size impacts intranuclear

structure A variety of different membraneless RNA/protein bodies arefound within the nucleus, such as nucleoli, speckles, and Cajal, histonelocus, and PML bodies Surprisingly, many of these structures have beenshown to exhibit fluid-like properties, with important implications for thenumber and size distribution of these bodies within the nucleus

within the large germinal vesicle (GV) of Xenopus oocytes can fuse

individual nucleoli separated, and disruption of actin causes nucleoli tosettle to the bottom of the nucleus under the force of gravity where theyfuse into one large nucleolus (Feric and Brangwynne, 2013) Clearly this is

an extreme example as the GV is on the order of 0.5 mm in diameter, but itdemonstrates that nuclear size has important implications for intranuclearstructure and organization Mechanisms that regulate the assembly of suchribonucleoprotein bodies are beginning to be elucidated (Nott et al.,

2015) Also see[42_TD$DIFF]Sections 3.2and 3.4

2.7 Extranuclear[41_TD$DIFF]Structures

The endoplasmic reticulum (ER) has also been shown to influence nuclearsize The ER is an interconnected network of lipid bilayer membrane sheetsand tubules that is continuous with the NE It has been proposed that there is

a tug-of-war relationship between the NE and ER membrane systems.Altering the relative proportions of ER tubules and sheets can concomitantlyaffect nuclear size (Anderson and Hetzer, 2007, 2008; Webster et al., 2009).Proteins in the reticulon (Rtn) and REEP families shape ER membranes intotubules and also stabilize membrane curvature at the edges of ER sheets(Friedman and Voeltz, 2011; Shibata et al., 2010; Voeltz et al., 2006; West

nuclear size Future experiments will address if other morphological proteins

of the ER also impact nuclear size, such as CLIMP63 that dictates ER sheetspacing (Shibata et al., 2010), atlastins that mediate fusion of ER tubules intothree-way junctions (Hu et al., 2009; Orso et al., 2009), and Lunapark thatregulates three-way junction dynamics (Chen et al.,[43_TD$DIFF]2012b, 2015a).The microtubule and actin cytoskeletons also impact nuclear size andmorphology One intriguing example involves cytoplasmic streaming, aform of intracellular transport found in plants that is generated by themovement of organelles, including the nucleus, along actin filaments underthe action of myosin XI The exact functional role of this streaming is largelyunknown High-speed and low-speed versions of[44_TD$DIFF]Arabidopsisthaliana myosin

XI were generated by varying the motor domain Expression of high-speedchimeric myosin accelerated cytoplasmic steaming and led to increased celland plant size Conversely, low-speed myosin slowed cytoplasmic streamingand cells and plants were smaller (Tominaga et al., 2013) Because there is arelationship between cell and nuclear size, we speculate that the kinetics ofcytoplasmic streaming might influence nuclear size, although this was notexplicitly examined in this study Also see Sections[45_TD$DIFF]3.2–3.8and4.2

2.8 Cell-Cycle Effects

The kinetics of some cell cycle events can have important implications fornuclear size and morphology During the metazoan cell cycle, nuclear enve-lope breakdown (NEBD) occurs prior to mitosis This is in contrast to theclosed mitoses of many yeasts in which the nucleus remains intact duringmitosis (Sazer et al., 2014) NEBD is initiated by a variety of different kinasesincluding cyclin-dependent kinases (Cdks), protein kinase C (PKC), andNIMA-related kinases Key nuclear substrates in NEBD are lamins andnucleoporins (Laurell et al., 2011; Mall et al., 2012) Interestingly, thisprocess seems to have been adapted during lens epithelial cell differentiation

to remove nuclei entirely from these cells (Chaffee et al., 2014) NEBD instarfish oocytes is driven by an F-actin shell that fragments the NE membrane(Mori et al., 2014), while an in vitro Xenopus assay for NEBD implicated themicrotubule cytoskeleton and Ran in NE rupture (Muhlhausser and Kutay,

2007) LINC complex proteins that connect the nucleus to the cytoskeletonare also important in NEBD, and impairing NEBD can lead to mitoticdefects (Turgay et al., 2014)

After NEBD, mitotic spindle assembly and chromosome segregationensue Components of the NE are important for this process, for example

lamins contribute to formation of a spindle matrix (Johansen et al., 2011;Tsai et al., 2006) Interestingly, nuclear and cytoplasmic proteins do notappear to mix homogeneously after NEBD, and this may have importantimplications for spindle assembly and the subsequent cell cycle (Pawar

et al., 2014) In radial glial progenitor cells, dynein-dependent nuclearmigration during the G1 phase of the cell cycle is necessary for entry intomitosis (Hu et al., 2013) Aside from the nucleus, other organelles must also

be faithfully segregated between the two daughter cells during mitosis(Jongsma et al., 2015)

After anaphase, NE assembly occurs around the segregated daughterchromosomes through targeting of ER tubules to the chromatin and spread-ing of membrane across the chromosomes (Anderson and Hetzer, 2007;Clever et al., 2013; Schooley et al., 2012) It was recently demonstrated thatcomponents of the ESCRT machinery are responsible for the membranefusion that gives rise to an intact NE (Olmos et al., 2015; Vietri et al., 2015).Distinct mechanisms are responsible for NPC assembly during NE forma-tion and for the insertion of NPCs into the intact NE during interphase

Nuclear expansion during interphase occurs through a nuclear dependent process In mammalian tissue culture cells, the number ofNPCs and nuclear volume were observed to double during interphase,and Cdk activity was involved in interphase NPC formation Interestingly,Cdk inhibition disturbed new NPC assembly but did not block nuclearexpansion, suggesting that NPC doubling during interphase is not requiredfor normal nuclear growth (Maeshima et al.,[47_TD$DIFF]2010, 2011)

import-Proper NE assembly depends on the removal of microtubules and brane from chromosomes REEP3/4 are ER proteins required to clearmembranes from metaphase chromosomes through a microtubule-depen-dent process Failure of this process leads to defective NE architecture, theformation of intranuclear membrane structures, and defects in chromosomesegregation and the separation of daughter nuclei (Schlaitz et al., 2013) (Fig[21_TD$DIFF]

mem-1) BAF is a chromatin-bound protein key to NE assembly that is usuallyremoved from the chromatin prior to mitosis Mutations causing constitutivechromatin association of BAF led to highly aberrant nuclear morphology,likely resulting from a failure to clear membrane from the mitotic chromo-somes (Molitor and Traktman, 2014) While microtubules are essential tochromosome segregation during mitosis, microtubule removal later in mito-sis is required for normal nuclear morphology in the subsequent interphase

In Xenopus egg extracts, chromatin-bound Dppa2 was shown to be required

to inhibit microtubule polymerization around the chromosomes during NEassembly, and persistent microtubules led to the formation of small, mis-shapen nuclei (Xue and Funabiki, 2014; Xue et al., 2013) Also see Sections

3.2, 3.4, 3.7[48_TD$DIFF], and4.1

2.9 Signaling Pathways

While manipulating the levels or activities of NE components can alter thesize and shape of the nucleus, relatively few studies address mechanisms ofnuclear size regulation in a physiological context The Xenopus embryoprovides an excellent model system for studying nuclear size regulation.Upon fertilization, the single-cell Xenopus embryo undergoes a series ofrapid divisions to generate a few thousand cells, while the overall size ofthe embryo remains unchanged Dramatic reductions in cell size during earlydevelopment are associated with changes in nuclear size and dynamics,without changes in nuclear DNA content (Jevtic and Levy, 2015; Levyand Heald, 2010) (Fig[21_TD$DIFF] 1) The Xenopus egg and embryo extract systemsconstitute undiluted cytoplasms that have been extensively used to studyvarious cellular activities including nuclear assembly and import, mitoticspindle regulation, and chromosome structure (Chan and Forbes, 2006;Edens and Levy, 2014b; Good et al., 2013; Hara and Merten, 2015; Hazel

et al., 2013; Kieserman and Heald, 2011; Levy and Heald, 2010; Loughlin

et al., 2011; Wilbur and Heald, 2013)

We sought to develop nuclear re-sizing assays, using Xenopus embryoextracts, in order to identify novel regulators of nuclear size We found thatlarge nuclei, assembled in Xenopus egg extract, became smaller when incu-bated in cytoplasm isolated from late stage embryos Using this system, wedetermined that this nuclear shrinking activity was regulated by the activa-tion and nuclear translocation of conventional protein kinase C (cPKC),leading to removal of lamins from the NE During development, nuclearcPKC activation and localization increase, correlating with decreasednuclear size (Fig[21_TD$DIFF] 1) Furthermore, we showed that this signaling pathwaywas also important for proper nuclear size regulation in vivo in the embryoduring interphase (Edens and Levy, 2014a) While PKC activity has previ-ously been implicated in NEBD during mitosis and nuclear export of largeribonucleoprotein complexes and certain viruses (Hatch and Hetzer, 2014;Leach and Roller, 2010; Milbradt et al., 2010; Park and Baines, 2006; Speese

et al., 2012), our data suggest that interphase nuclear cPKC activity plays arole in steady-state nuclear size regulation (Fig[21_TD$DIFF] 3)

Phospholipid double bilayer

Lamin protein dimers, tetramers etc.

Importin α

Figure 3 [11_TD$DIFF]Models of nuclear size regulation during Xenopus development [12_TD$DIFF](A) In the MBT embryo, nuclear growth is mediated by greater importin α activity and import of lamin B3 (B) In the post-MBT embryo, nuclear shrinking is mediated by increased nuclear cPKC localization and activity, and subsequent dissociation of lamins from the

pre-NE (C) A balance of import and cPKC-mediated shrinking determines steady-state nuclear size This model is based on our studies of nuclear size regulation in Xenopus ( Edens and Levy, 2014a; Levy and Heald, 2010 ).

While nuclear lamins are known targets for PKC phosphorylation

shrinking is mediated by direct PKC phosphorylation of lamins or diate signaling proteins Consistent with our results, the phosphorylationstate of lamin A in HeLa cells mediates its assembly dynamics with the laminameshwork during interphase (Kochin et al., 2014) To account for themechanical forces associated with lamin removal during nuclear shrinking,

interme-we envision a model wherein lamin density must remain roughly constant forthe structure of the nucleus to maintain its integrity and withstand cytoskel-etal forces (Buxboim et al., 2010) By this model, increased lamina dynamicsand loss of lamins from the NE are compensated by a retraction of NEmembrane back into the ER, thus maintaining a constant nuclear lamindensity as nuclear size becomes smaller

While diverse models of organelle size regulation have been discussedelsewhere (Chan and Marshall, 2012; Goehring and Hyman, 2012; Webster

et al., 2009), we envision two plausible mechanisms for how nuclear size isdetermined during early Xenopus development The first model posits thatthere is a limiting component for nuclear growth and as this component isdistributed among greater numbers of cells during development, nucleireach a steady-state size that scales smaller with cell number Possible limitingcomponents include molecules that directly inhibit cPKC or an inhibitor of

an upstream cPKC activator One might also consider the affinity of PKC forits various substrates, which can vary drastically and is important in differ-ential spatial and temporal regulation of PKC activity (Fujise et al., 1994).Over the course of development, changes in the relative abundances ofdifferent PKC substrates with varying affinities might shift PKC activitytoward lower affinity substrates relevant to nuclear shrinking

A second model is based on the idea that there is an equilibrium balancebetween nuclear growth and contraction Such equilibrium balance modelshave been applied to the mitotic spindle (Loughlin et al., 2011), flagella(Marshall et al., 2005), mitochondria (Rafelski et al., 2012), vacuoles(Chan and Marshall, 2014), nucleus (Edens and Levy, 2014a), and others

organ-elle like the nucleus, the simplest equilibrium balance model is one where aconstant rate of nuclear growth, mediated by nuclear import, is balanced by aproportional contraction rate mediated by cPKC translocation to the NE Bybalancing nuclear import-mediated growth, nuclear shrinking causes nuclei

to reach a steady-state size (Fig[21_TD$DIFF] 3) It is important to note that the rium balance model is not mutually exclusive with limiting component

equilib-models of size regulation (Goehring and Hyman, 2012) Important questionsremain about what upstream regulatory signals might control nuclear size inthis system Although several factors that influence nuclear size are known,

an integrated model of the regulatory mechanisms controlling nuclear sizehas yet to be described

3 MODEL SYSTEMS TO ELUCIDATE MECHANISMS

OF NUCLEAR SIZE REGULATION

Cells tend to maintain their nuclear size within a defined range.Changes in cell size that occur during development, cell division, anddifferentiation are accompanied by dynamic nuclear size adjustments inorder to establish appropriate N/C volume relationships Mechanisms thatregulate proper nuclear size and the functional significance of this regulationare largely unknown Aberrations in nuclear size are associated with certaindisease states, most notably cancer It seems likely that nuclear size and the N/

C volume ratio affect cell physiology, for instance through altered chromatinorganization and gene expression In this section we focus our attention onstudies from different eukaryotic model experimental systems includingTetrahymena, yeasts and fungi, plants, worms, flies, fish, frogs, mice, andmammalian tissue culture Research based on these model systems has elu-cidated some important molecular mechanisms of nuclear size regulation(Fig[21_TD$DIFF] 2)

3.1 Tetrahymena thermophila

The ciliateTetrahymena thermophila has two morphologically and functionallydistinct nuclei both located within the same cell The bigger somatic mac-ronucleus (MAC) is polyploid and transcriptionally active, while the smallergerminal micronucleus (MIC) is diploid and transcriptionally inactive duringthe vegetative growth cycle (Frankel, 2000) The two nuclei also differ inprotein composition Macronuclear linker histone H1 is localized to theMAC and its deletion results in enlargement of only the MAC.Conversely, micronuclear linker histone (MLH) is unique to the MIC andits deletion leads to enlargement of the MIC but not MAC (Allis et al.,[49_TD$DIFF]1979,1980; Shen et al., 1995; White et al., 1989)

Among all identified nucleoporins, four homologues of Nup98 werefound to have strict nuclear selectivity, with two localized to the MAC

(MacNup98A and MacNup98B) and the other two localized to the MIC(MicNup98A and MicNup98B) MacNup98A and MacNup98B have typ-ical GLFG N-terminal repeat domains, while MicNup98A and MicNup98Bhave unusual repeats of NIFN or SIFN To elucidate the function of MAC-specific and MIC-specific repeats, swapping experiments of N-terminalrepeat domains between MacNup98A and MicNup98A were performed.Chimeric proteins composed of the N-terminal NIFN repeat domain ofMicNup98A and the C-terminal domain of MacNup98A (BigMac) exclu-sively located to the MAC NPC, and chimeric proteins composed of the N-terminal GLFG repeat domain of MacNup98A and the C-terminal domain

of MicNup98A (BigMic) predominantly located to the MIC NPC.Overexpression of BigMic caused a 2-fold increase in the size of MIC and

a dramatic decrease in MIC localization of MIC-specific MLH proteins.BigMac overexpression led to increased MAC size and decreased MAClocalization of macronuclear histone H1 (Iwamoto et al., 2009) These datademonstrate that, to enable nucleus-selective import of different proteins,the MAC and MIC utilize unique Nups

Karyopherins also contribute to the regulation of nuclear size In total,Tetrahymena encodes 13 putative importinα-like proteins and 11 importinβ-like proteins Nine importin α proteins are MIC specific, while all impor-tinβ proteins localize to both the MIC and MAC Knockdown of IMA10, aMIC-specific importin α, caused MIC division defects including laggingMIC chromosomes, loss of MIC DNA content, and abnormal nuclearnumber and morphology This suggests that IMA10 plays a MIC-specificrole in regulating MIC division and nuclear morphology Transport to theMIC and MAC are mediated through different subsets of importinα trans-port receptors that are uniquely targeted to each nucleus, and this likely hasimportant implications for how nuclear size is regulated in the two types ofnuclei, for instance through regulated import of different histone H1 iso-forms (Malone et al., 2008)

Treatment ofTetrahymena with low concentrations of the DNA polymerase

α inhibitor aphidicolin led to cell division arrest and, surprisingly, rounds ofMAC endoreduplication and cell size increase Upon resumption of celldivision, large extrusion bodies formed from dividing MACs and the size ofextrusion bodies correlated with the duration of aphidicolin pre-treatmentand the amount of MAC DNA (Kaczanowski and Kiersnowska, 2011) Thesedata suggest that compensatory mechanisms exist to regulate the level of ploidyinTetrahymena, thereby influencing nuclear morphology and cell size

3.2 Yeasts and[54_TD$DIFF]Fungi

In budding yeast, cell growth during the cell cycle correlated withincreased nuclear size, maintaining a constant N/C volume ratio(Jorgensen et al., 2007) (Fig[21_TD$DIFF] 1) In fission yeast, different cell size mutantsexhibited differently sized nuclei, again maintaining similar N/C volumeratios A 16-fold increase in DNA content did not affect nuclear size, andnuclear sizes and N/C volume ratios were similar between polyploidmutants and wild-type cells, strongly suggesting that ploidy does not affectnuclear size in this system (Fig[21_TD$DIFF] 1) In a multinucleate cytokinesis mutant,nuclei exposed to larger volumes of cytoplasm grew larger than nuclei

2007) These data demonstrate that the N/C volume ratio is a tightlyregulated cellular feature and that cytoplasmic volume contributes to theregulation of nuclear size

The Ran GTPase cycle plays an important role in NPC assembly and themaintenance of normal NE morphology in budding yeast Disrupting Ran,RanGAP, RanGEF, or NTF2 led to NPC assembly defects, cytoplasmicaccumulation of Nup containing vesicles, and alterations in nuclear sizeand shape (Ryan et al., 2003) Furthermore, cells expressing a mutant version

of importinβ (kap95-E126K) or deleted entirely for importin β showed severenuclear architecture distortion, accumulation of extended sheets of cyto-plasmic membranes, and mislocalization of Nups Thus both Ran andimportinβ contribute to proper NPC assembly and the formation of nor-mally sized and shaped nuclei (Ryan et al., 2007)

Genetic screens in budding yeast designed to identify regulators of NPCstructure and assembly uncovered several components of the RSC chromatinremodeling complex as being important for normal Nup localization andnuclear morphology Following up on these results, thin section transmissionelectron microscopy was performed on RSC mutants, revealing NPC mis-localization, severe nuclear morphology alterations, accumulation of exten-sive sheets of nuclear and cytoplasmic membrane and vesicles, and alteredchromatin structure and morphology Because RSC complex loss-of-func-tion resulted in NPC and nuclear morphology perturbations, this suggeststhat normal nuclear morphology is functionally linked to proper chromatinstructure (Titus et al., 2010)

In yeasts that have a closed mitosis, spindle pole bodies (SPBs) are insertedinto the NE, connecting cytoplasmic and intranuclear microtubules to drivespindle assembly and elongation In budding yeast, overexpression of a

dominant negative version of the SUN-domain protein Mps3 led to defects

in SPB duplication and spindle formation, causing massive over-proliferation

of the nuclear membrane This alteration in nuclear morphology and SPBduplication could be rescued by treatment of cells with chemicals that altermembrane composition or fluidity, suggesting that SUN proteins regulateinsertion of protein complexes such as the SPB by modulating NE lipidcomposition (Friederichs et al., 2011)

In budding yeasts, the mother cell divides asymmetrically to produce asmaller daughter cell with a smaller nucleus The transcription factor Ace2concentrates in the daughter nucleus, contributing to the asymmetry ofbudding yeast division Ace2 activates transcription of several daughter-specific genes and regulates physical separation of the daughter cell fromthe mother cell (Dohrmann et al., 1992) During anaphase, the nucleusadopts a dumbbell-like shape, including a characteristic internuclear bridge.The length and width of the internuclear bridge appears to limit theexchange of nucleoplasm between the two halves of the dumbbell-shapednucleus, potentially contributing to differences in nuclear size between themother and daughter cells (Boettcher et al., 2012)

Even though all nuclei in the multinucleate fungus Ashbyagossypii share acommon cytoplasm, they are non-randomly positioned and divide asyn-chronously Functionally distinct zones are established through a microtu-bule-dependent mechanism whereby the nuclei actively repel each other.The spacing of these autonomous nuclear territories increases approachingmitosis Interestingly, after nuclear division, the two daughter nuclei assumetheir own territories but maintain similar cell cycle timings This suggeststhat nuclei inherit a characteristic cell cycle timing using a microtubule-based repulsion mechanism to form different subpopulations within a com-mon cytoplasm (Anderson et al., 2013) We speculate that nuclei in mono-nucleated cells might utilize similar cytoskeleton-based mechanisms to sensethe available cytoplasmic volume in order to adjust their size according to cellsize

3.3 Plants

Plant nuclei roughly resemble animal nuclei, but many proteins and tures that regulate their nuclear morphology appear to be different Plantproteins that constitute the nuclear lamina share a similar domain structurewith animal lamins and belong to the NMCP/LINC/CRWN protein family(nuclear matrix constituent proteins/little nuclei proteins/crowded nucleus

struc-proteins) (Ciska and Moreno Diaz de la Espina, 2014) Originally namedLINC and later renamed CRWN, these proteins contain a large centralcoiled-coil region and NLS, and localize to the nuclear periphery.Decreasing the expression of A[58_TD$DIFF].thaliana LINC1 or LINC2 led to a reduction

in nuclear size, and the double mutant showed an additive effect with muchsmaller nuclei, altered nuclear shape, and pronounced whole plant dwarfism(Dittmer et al., 2007) When all four members of the CRWN protein familywere examined, single mutants deficient in either CRWN1 or CRWN4 haddecreased nuclear size, while loss of CRWN2 or CRWN3 had no effect.Reducing levels of CRWN1 with either CRWN2 or CRWN3 causedfurther nuclear size reductions, while depleting levels of CRWN4 in com-bination with either CRWN2 or CRWN3 did not have an additive effect onnuclear size Reducing levels of both CRWN1 and CRWN4 decreasednuclear size the most These data identify CRWN1 and CRWN4 as themost important regulators of nuclear size in this system (Wang et al., 2013).The structure and composition of the NPC in A.thaliana is very similar tothe animal NPC, except for a plant-specific nucleoporin, Nup136 Based onits localization within the NPC and known functions, Nup136 is likely afunctional homolog of vertebrate Nup153, known to regulate classicalnuclear import pathways (Shah and Forbes, 1998) Nup136 may also mediate

an interaction between the NPC and plant-like lamina structures Alteringthe levels of Nup136 in vivo altered nuclear size and shape Overexpressionincreased nuclear size and elongation in many tissues, whereas reducingNup136 expression resulted in smaller, more spherical nuclei (Tamura

et al., 2010; Tamura and Hara-Nishimura, 2011) In A thaliana there are 3plant specific KASH proteins: AtWIP1, AtWIP2, and AtWIP3 These pro-teins bind to AtSUN1 and AtSUN2 at the NE and are important foranchoring RanGAP to the NE Disrupting these interactions led to alterednuclear morphology and reduced nuclear elongation (Zhou et al., 2012)

3.4 Caenorhabditis elegans

During rapid early embryonic cell divisions, there is a dramatic decrease inboth cell and nuclear sizes What is the interplay between chromosomecondensation and nuclear size? In early[59_TD$DIFF]Caenorhabditis elegans embryos, con-densed mitotic chromosome length decreases during early embryonic pro-gression This difference is not due to different chromosome condensationdynamics because the speed and duration of chromosome condensation werefound to be similar between different stages of early development Because

nuclear size decreases during development, intra-nuclear DNA densityincreases, potentially influencing chromosome condensation before mitosis.

To test this idea, ploidy was varied in embryos Condensed chromosomelength was larger in haploid embryos than in diploid embryos of the samestage, and chromosomes in polyploid embryos were shorter Furthermore,decreasing or increasing nuclear size in embryos decreased or increasedcondensed chromosome length, respectively Similarly, decreasing nuclearsize in Xenopus egg extract led to shorter chromosomes compared to control

DNA density, controlled by DNA amount and/or nuclear size, influenceschromosome compaction and length Consistent with studies in Xenopus,depleting an importin α protein in C elegans resulted in smaller nuclei.Interestingly, mitotic chromosome length also scaled smaller in thesemutants Reductions in nuclear and chromosome sizes were also observedfor embryos depleted of RCC1 or NTF2, although the scaling relationshipsbetween cell, nuclear, and chromosome sizes differed depending on themutant examined (Ladouceur et al., 2015)

C elegans has one lipin homologue, LPIN-1, a phosphatidic acid phatase with a role in lipid homeostasis and glycerolipid biosynthesis LPIN-

phos-1 downregulation led to disruption of ER structure, accumulation of ERsheets, defects in NE disassembly and chromosome segregation, aberrant NEreassembly, and irregularly shaped nuclei of variable size (Golden et al.,

2009) CNEP-1 is a NE-localized activator of lipin Deletion of theCNEP-1 gene in one-cell embryos caused formation of binucleate cells(twinned nuclei) at the 2-cell stage, and the nuclei were misshapen as aconsequence of defective NE disassembly Transmission electron microscopyrevealed that in CNEP-1Δ embryos, the nucleus was encased by extra ERmembrane, leading to defective NE disassembly and the twinned nucleiphenotype Expression of wild-type, but not phosphatase-defective,CNEP-1 could rescue nuclear morphology defects in two-cell stageembryos Furthermore, reducing the elevated levels of phosphatidylinositolfound in CNEP-1Δ embryos prevented ER sheet accumulation and NEBDdefects (Bahmanyar et al., 2014)

Gp210 is an integral membrane protein that regulates NPC assembly andspacing, as its downregulation in both C.elegans and human cells led to NPCclustering, weaker staining for FG repeat Nups, Nup aggregation, andreduced viability Gp210 depletion also caused aberrant nuclear morphology,enlargement of the lumen between the ONM and INM, and defects inchromatin morphology and distribution In human cells, gp210 depletion

impaired nuclear import and led to reduced NE staining for lamins(Cohen et al., 2003) These findings implicate gp210 in membrane fusion,

NE formation, and the regulation of normal nuclear morphology Loss of the

Morales-Martinez et al., 2015)

By quantifying cell volumes and cell cycle durations in three C elegansembryonic lineages, a power law relationship was derived with larger cellsundergoing more rapid cell cycles, consistent with what was previouslyobserved in Xenopus embryos (Wang et al., 2000) Elongation of the cellcycle in smaller cells was due to elongation of interphase and not mitosis.Furthermore, nuclear size was found to correlate with cell size, with cellcycle timing strongly correlating with the N/C volume ratio (Arata et al.,

2014) These correlations may have important implications for the properregulation of developmental transitions, such as the midblastula transition(MBT)

The nucleolus is a membraneless intranuclear organelle whose size scaleswith nuclear and cell sizes (Boisvert et al., 2007; Jorgensen et al., 2007;

64-cell stage, nucleolar size directly scales smaller as cell size is reduced Totest the relationship between nucleolar size and cell size, RNAi knockdownswere performed to generate embryos that were bigger and smaller than wild-type Surprisingly, nucleolar size scaled inversely with cell size across RNAiconditions at the same developmental stage Embryos with decreased sizeshowed increased nucleolar intensity, and embryos with increased size hadsmaller nucleoli compared to normally-sized control embryos of the samestage Further study showed that embryos from different RNAi conditionsare loaded with a fixed amount of nucleolar components originating fromthe oocyte, which then results in concentration differences between differ-ently sized embryos Therefore, cell and nuclear sizes can impact the nuclearconcentration of nucleolar components, thereby controlling nucleolar size

scaling of other cellular organelles

3.5 Drosophila melanogaster

In D melanogaster, early embryogenesis is characterized by 13 syncytialnuclear divisions that precede the process of cellularization When cytoplasmwas extracted from[61_TD$DIFF]preblastoderm embryos and encapsulated in droplets ofdefined size, nuclei spread throughout the entire available space after a few

divisions, suggesting that early stage cytoplasm has the intrinsic ability todistribute nuclei homogeneously This spacing was shown to depend on bothmicrotubules and actin filaments (Telley et al., 2012) In other systems, it hasbeen shown that the amount of accessible cytoplasm contributes to nuclearsize (Gurdon, 1976; Hara and Merten, 2015; Neumann and Nurse, 2007), so

it will be interesting to test if this spacing mechanism is important forregulating nuclear size in the early Drosophila embryo

During cellularization, nuclear morphology changes from spherical toelongated, lobulated, and grooved, concomitant with an 84% increase in NEsurface area During this process, microtubules (MTs) form a ring structurearound the nuclei and they appear enriched in regions with NE grooves.Treatment of embryos during early cellularization with MT depolymerizingdrugs inhibited nuclear elongation, decreased the rate of NE surface expan-sion, impaired NE dynamics, and reduced groove formation In untreatedembryos, MT bundles lining the NE were observed to be very dynamic.Stabilizing these MTs by taxol injection reduced MT dynamics, decreasedfluctuations of the NE, and impaired formation of NE grooves These dataindicate that specific localization and organization of dynamic MTs at the

NE are required for NE remodeling during cellularization The minus-enddirected MT motor dynein has also been implicated in this process, as dyneininhibition decreased the tubulin signal at the NE, reduced the MT lifetimewithin bundles, and impaired nuclear elongation and groove formation Alsoinvolved in this process is the farnesylated INM protein Kugelkern (Kuk)which is upregulated during cellularization, increasing stiffness of the NEand helping to maintain NE grooves once induced by the polymerization of

MT bundles Ultimately, nuclear morphology changes that occur at thisstage of development contribute to altered chromatin dynamics and organi-zation, and may be important for regulating the onset of zygotic geneexpression (Brandt et al., 2006; Hampoelz et al., 2011; Polychronidou andGrobhans, 2011; Polychronidou et al., 2010)

As a normal part of Drosophila development, cells in the salivary glandsundergo endoreduplication, giving rise to large cells and nuclei containingpolytene chromosomes (Hochstrasser and Sedat, 1987) In adults, puncturewounding of the abdominal wall caused quiescent epithelial cells close to thewound to reenter the cell cycle Surprisingly, these cells did not divide.Instead, epithelial cells migrated toward the wound site and fused to formlarge syncytia containing clustered nuclei A different group of enlarged cellsarose near the wound by undergoing endoreduplication These polyploidcells were more than two times larger than normal epithelial cells and

contained one large nucleus Reinjuring the same site led to further increases

in the sizes of surviving epithelial cells Both cell fusion and polyploidizationwere shown to be controlled by the Hippo signaling effector Yorkie, andblocking the formation of these large cells led to defects in wound healing.These data demonstrate that cell mass lost by wounding can be replaced bycell enlargement ([62_TD$DIFF]ie, hypertrophy and cell fusion) rather than by mitoticproliferation (Losick et al., 2013)

Nuclear roles for classically cytoplasmic proteins are beginning toemerge In particular, actin and proteins that regulate actin polymerizationare found within the nucleus (Belin et al., 2013; Belin and Mullins, 2013;

was shown to interact with B-type lamins Wash knockdown altered nuclearmorphology, leading to abnormal, wrinkled nuclei This, in turn, alteredchromatin organization and gene expression (Verboon et al., 2015)

3.6 Zebrafish

During early embryonic cleavage stages of Zebrafish (Danio rerio) ment, individual or groups of chromosomes are enclosed within NE at theend of mitosis, forming intermediate nuclear structures called karyomeres.Eventually these micronuclei fuse to form a mononucleus Brambleberry(bmb) was identified as a protein required for karyomere fusion and propernuclear morphology during early development Bmb mutant embryosarrested after the MBT with clustered micronuclei consisting of separate,NE-encased chromatin structures that failed to fuse In wild-type embryosduring metaphase, bmb localized within the mitotic spindle near the chro-mosomes At anaphase, NE and bmb assembled around elongated condensedchromosomes forming early karyomeres During the process of karyomererounding, bmb localized to foci at the periphery of micronuclei at karyo-mere-karyomere contact sites These foci correlated with sites of karyomeremembrane fusion Taken together, these data illustrate a role for bmb in theformation of mononuclei of the correct size and morphology during rapidearly embryonic cell cycles (Abrams et al., 2012)

develop-3.7 Xenopus

Xenopus egg extracts provide a robust biochemical system to study isms of nuclear size regulation Xenopus egg extract is a cell-free system whereproteins of interest can be immunodepleted or inhibited with specific anti-bodies or compounds, and recombinant proteins can be added Xenopus egg

extracts lack egg chromosomes but contain membranes and all cytoplasmicproteins necessary to assemble nuclei in vitro Addition of demembranatedXenopus sperm to interphasic egg extract stimulates nuclear formation.Amphibian eggs are arrested in metaphase-II of the meiotic cell cycle, socell-free extracts are easily manipulated and can be arrested at different stages

of the cell cycle (Edens and Levy, 2014b; Jevtic and Levy, 2014; Levy andHeald, 2012)

X laevis is a frog species widely used as a model organism, and its closerelative, X tropicalis, is a much smaller animal Interspecies nuclear scalingwas studied using X laevis and X tropicalis egg extracts Nuclei assembledwith X.laevis sperm were larger in X.laevis extract than in X.tropicalis extract,and mixing the two extracts produced a graded effect on nuclear size X.tro-picalis sperm has approximately half the DNA content of X laevis sperm,and adding X tropicalis sperm to X laevis extract produced only slightlysmaller nuclei This demonstrated that titratable cytoplasmic factors regulatenuclear size in this system, and DNA amount has a much smaller effect.Nuclear import rates and the levels of importinα and NTF2 were found todiffer between the two extracts X.laevis extract had higher levels of importin

α and exhibited faster rates of nuclear import The NTF2 concentration wasgreater in X tropicalis extract, correlating with slower import of large cargomolecules like lamin B3 (LB3) Manipulating the levels of importinα andNTF2 in X tropicalis extract was sufficient to generate X laevis sized nuclei

In addition, reductions in nuclear size during early Xenopus developmentcorrelated with reductions in cytoplasmic importinα levels and bulk import(Levy and Heald, 2010)

Xenopus early development represents a powerful in vivo system to studythe functional significance and mechanisms of nuclear size regulation.Xenopus embryos can be microinjected with mRNA to overexpress proteins

of interest, morpholino oligonucleotides to inhibit translation of targetedmRNAs, or inhibitory antibodies and proteins Furthermore, embryoextracts can be isolated at desired stages of development, providing a bio-chemically tractable approach to studying developmental processes In X.laevis, the fertilized egg undergoes 12 rapid cleavages to produce ∼ 4000cells at the midblastula transition (MBT) The MBT is the first major devel-opmental transition and is characterized by the onset of zygotic transcriptionand cell cycle lengthening One mechanism that has been proposed toregulate the MBT is the DNA-to-cytoplasm ratio (Edgar et al., 1986;Kobayakawa and Kubota, 1981; Newport and Kirschner, 1982a,[64_TD$DIFF]b) By thismodel, as the total DNA amount in the embryo increases though rapid

rounds of DNA replication and cell division, a maternal factor loaded in theegg is titrated by DNA, leading to onset of the MBT Recently, a number ofpotential limiting factors have been identified that may act in this way,including DNA replication initiation factors (Collart et al., 2013), proteinphosphatase PP2A (Murphy and Michael, 2013), and histones (Amodeo

et al., 2015)

It is also possible that the N/C volume ratio contributes to MBT timing,

in addition to the DNA-to-cytoplasm ratio During the cleavage stages ofdevelopment and up to the MBT, average nuclear volume decreases[65_TD$DIFF]three-fold while average cytoplasmic volume shows a much more dramatic∼70-fold reduction Consistent with this observation, the N/C volume ratioincreases rapidly prior to the MBT We showed that nuclear size and theN/C volume ratio could be manipulated in embryos by microinjectingmRNAs to ectopically alter the expression of nuclear transport factors,nuclear lamins, or tubule-shaping components of the ER Increasing nuclearsize in early stage embryos was sufficient to increase the N/C volume ratio tothat found at the MBT and to cause premature onset of zygotic transcriptionand lengthening of cell cycles Conversely, decreasing nuclear size in MBTembryos lowered the N/C volume ratios to values found in cleavage stageembryos resulting in delayed MBT onset (Jevtic and Levy, 2015) (Fig[21_TD$DIFF] 4)

A number of interesting questions remain with respect to MBT timing inXenopus What are the relative contributions of nuclear size and DNAamount to MBT timing? Do changes in nuclear size mediate the MBT byaltering chromatin compaction, organization, and structure? How mightchanges in NE surface area affect import capacity and the import of limitingDNA binding components that in turn regulate MBT timing? Perhapschanges in nuclear volume alter the intranuclear concentrations of limiting,maternally-derived DNA binding factors In this way, the MBT might beregulated not only by the total amount of these maternally deposited limitingcomponents but by their nuclear concentrations, as determined by changes

in total nuclear volume during early embryogenesis

Reductions in nuclear size at the MBT correlate with reductions in bulkimport and cytoplasmic importinα levels, and ectopic expression of impor-tinα was sufficient to increase nuclear size in MBTembryos (Levy and Heald,

2010) To investigate mechanisms that might account for the 3-fold tion in nuclear surface area after the MBT, we developed an in vitro assay inwhich large nuclei assembled in egg extract were observed to shrink whenincubated in late embryo extract This shrinking activity depended onconventional protein kinase C (cPKC) In post-MBT embryos, nuclear

cPKC activity and localization increased concomitantly with a decrease inthe nuclear localization of lamins (Fig[21_TD$DIFF] 1) Manipulating cPKC activity invivo altered interphase nuclear size, suggesting a mechanism that contributes

to post-MBT nuclear size reductions (Edens and Levy, 2014a) (Fig[21_TD$DIFF] 3).Lamin intermediate filament proteins constitute the nuclear lamina, themesh-like structure on the internal face of the NE that connects the INMand chromatin Studies in Xenopus implicate the lamins in nuclear sizecontrol [67_TD$DIFF]Postmitotic NE reformation, nuclear lamina polymerization, and

Figure 4 [14_TD$DIFF]The N/C volume ratio controls MBT timing in Xenopus [2_TD$DIFF]Key results from our study of MBT timing in Xenopus are depicted ( Jevtic and Levy, 2015 ) The MBT normally occurs after 12 rapid cell cleavages and is marked by the onset of zygotic transcription and longer cell cycles Embryos microinjected with factors that increase nuclear size and the N/C volume ratio exhibit hallmarks of the MBT prior to the 12 [15_TD$DIFF]th cell division Conversely, microinjections that decrease nuclear size and the N/C volume ratio delay MBT onset.

nuclear expansion were all shown to be dependent on the lamin C-terminalIg-fold domain (Shumaker et al., 2005) The B-type lamins contain a C-terminal CaaX motif that is modified by farnesylation and is responsible fortargeting lamins to the INM where interactions between the nuclear lamina

Furthermore, overexpression of B-type lamins resulted in invaginations oflamina membrane structures that were attached to, but not continuous with,the INM (Ralle et al., 2004) In vitro nuclear assembly in Xenopus egg extractdepleted of embryonic LB3 resulted in small nuclei that failed to expandnormally (Newport et al., 1990) Conversely, ectopic addition of LB3increased the rate of nuclear expansion (Levy and Heald, 2010)

The nucleoporin building blocks of the NPC can dramatically affectnuclear size In Xenopus egg extract the depletion of Nup188 led to thederegulation of INM protein import and a concomitant increase in nuclearsize (Theerthagiri et al., 2010) A dominant-negative form of POM121, anintegral membrane Nup, inhibited the classical nuclear import pathway and

NE membrane expansion in Xenopus egg extract (Shaulov et al., 2011).Xenopus studies have also revealed how mitotic proteins and events caninfluence interphase nuclear size During mitosis, Tpx2 plays an importantrole in spindle assembly and also regulates spindle length (Helmke and Heald,

2014) During interphase, Tpx2 is localized to the nucleus and binds tolamina-associated polypeptide 2 (LAP2), a protein involved in nuclearassembly Depletion of Tpx2 from Xenopus egg extract led to the formation

of small nuclei, suggesting that Tpx2 and its interaction with LAP2 arerequired for normal nuclear size regulation (O’Brien and Wiese, 2006).Consistent with this model, addition of the chromatin-binding or lamin-binding domain of LAP2 to Xenopus egg extract inhibited nuclear growth

(Yang et al., 1997) These data suggest that lamina-chromatin interactionscontribute to proper nuclear size determination

Post-mitotic MT dynamics also influence nuclear morphology.Developmental pluripotency-associated 2 (Dppa2) protein binds to chro-matin and inhibits MT polymerization in vitro Depletion of Dppa2 fromXenopus egg extract led to the formation of abundant MTs surroundingthe chromatin after mitosis This resulted in the formation of small mis-shapen nuclei, and this nuclear morphology defect was rescued by treatmentwith MT depolymerizing drugs These data demonstrate that the regulatedreduction in MTs around post-mitotic chromosomes is important for propernuclear reassembly (Xue et al., 2013) Similarly, clearance of membrane from

metaphase chromatin is required for proper nuclear morphology in thesubsequent interphase Using Xenopus egg extracts, REEP4 was identified

as a novel ER protein that links MTs and membrane HeLa cells depleted ofREEP3/4 exhibited inappropriate membrane accumulation around meta-phase chromosomes leading to misshapen nuclei with intranuclear mem-branes in the subsequent interphase (Schlaitz et al., 2013) (Fig[21_TD$DIFF] 1)

Decondensation of [69_TD$DIFF]postmitotic chromosomes has also been studied inXenopus Highly condensed chromosome clusters isolated from HeLa cellsdecondense when added to interphase Xenopus egg extract, in a process thatrequires ATP and GTP hydrolysis Through biochemical fractionation, theATPase activity of RuvB-like ATPases was identified as being required fordecondensation RuvBL1 and RuvBL2 are AAA+ ATPases that associatewith different chromatin remodeling complexes and function individually inthe process of chromatin decondensation Inhibition of these proteins led todefects in chromatin decondensation and the formation of misshapen nuclei(Magalska et al., 2014) These data demonstrate that the formation of inter-phase nuclei of normal size and shape depends on proper chromatin decon-densation during late mitosis

3.8 Mammalian[70_TD$DIFF]Model Systems

Lamin genes in mammals include LMNB1 and LMNB2, as well as LMNAwith its 7 differently spliced proteins (Goldman et al., 2002) In laminknockout experiments, cells appear to have small fragile nuclei, and nuclearshape can also be affected (Misteli and Spector, 2011; Young et al., 2014).Mouse embryonic fibroblasts depleted of lamin A/C exhibited abnormalnuclear shapes and decreased nuclear stiffness (Lammerding et al., [71_TD$DIFF]2004,

2006), and lamin A overexpression caused invaginations of the NE (Friedl

et al., 2011; Prufert et al., 2004) Silencing LMNB1 in HeLa cells alteredlamina structure, leading to enlarged holes within the lamina meshwork andbleb formation (Shimi et al., 2008) In keratinocytes lacking all lamins, NEand ER membranes spread into the chromatin, suggesting that the laminameshwork is required to prevent the nuclear invasion of cytoplasmic orga-nelles (Jung et al., 2014) Knockout mice lacking B-type lamins surviveduntil birth but died immediately due to failure to breath, with defects in thedevelopment of multiple organs (Kim et al., 2011) Further studies willdetermine if any of these observed phenotypes result from altered nuclearsize The contribution of lamins to nuclear morphology and cell function arediscussed further in the context of laminopathies (see Section4.4)

During neutrophil differentiation, nuclei acquire a lobulated shape comitant with increased expression of lamin B receptor (LBR) and decreasedexpression of lamin A Knockdown of LBR led to the formation of hypo-lobulated nuclei in human promyelocytic leukemia (HL-60) cells, but thesecells maintained their ability to pass through micron-scale constrictions Onthe other hand, overexpressing lamin A in HL-60 cells prevented nuclearlobulation and caused impaired neutrophil migration, suggesting that lamin

con-A levels are important for both nuclear morphology and deformability(Rowat et al., 2013)

Interactions between the nucleus and cytoskeleton, mediated by LINCcomplex proteins, have also been implicated in nuclear size determination.LINC complexes are composed of INM SUN-domain proteins linked toONM KASH-domain proteins The contribution of LINC complex pro-teins to nuclear size regulation was studied in human keratinocytes (HaCaTcells) F-actin depolymerization resulted in small, highly dysmorphic nuclei,while microtubule depolymerization increased nuclear size Notably, depo-lymerization of both cytoskeletal components decreased nuclear size(Lu et al., 2012), suggesting that nuclear connections with the actin cyto-skeleton may preferentially determine nuclear size Overexpression of eitherthe Nesprin-2 actin binding domain (ABD) or Nesprin-2 C-terminalKASH domain led to increased nuclear size in HaCaT cells Conversely,nuclear size was reduced by overexpression of a Nesprin-2 mini construct,consisting of a fusion between the ABD and KASH domains and lackingmost of the centrally located rod domain spectrin repeats Furthermore, itwas demonstrated that Nesprin-3 interacts with the ABD of Nesprin-1/-2,and Nesprin-2 mini nuclear size reductions were further enhanced whenNesprin-3 was [72_TD$DIFF]co-overexpressed or actin filaments were depolymerized.Interestingly, these observed nuclear size changes correlated with analogouscell size changes (Lu et al., 2012) These results suggest that the spectrin roddomain within Nesprin-2 and interactions between Nesprins and the cyto-skeleton are important for nuclei to attain a certain size Nesprins may form abelt-like filamentous structure on the outside of the nucleus that regulatesnuclear size, similar to the intranuclear role of the lamina in controllingnuclear size Consistent with these data, mice lacking the ABD ofNesprin-2 exhibited increased epithelial nuclear size that correlated withthickening of the epidermis (Luke et al., 2008)

During neuronal differentiation, nuclear size rapidly increases, ing with increased expression of the methyl-CpG-binding protein MeCP2,

correlat-an abundcorrelat-ant chromatin-associated protein When mouse embryonic stem

cells lacking MeCP2 were induced to differentiate into neuronal cells, theresulting neuronal nuclei failed to increase in size, and this nuclear sizephenotype could be rescued by overexpression of MeCP2 Consistent withthe in vitro data, neuronal nuclei in mice lacking MeCP2 were smaller andless transcriptionally active compared to wild-type neurons (Yazdani et al.,

2012) Future work will address the relative contributions of MeCP2 andnuclear size to the regulation of gene expression during neuronal celldifferentiation

4 FUNCTIONAL SIGNIFICANCE OF NUCLEAR SIZE

AND MORPHOLOGY

Studies on the functional significance of nuclear size are rather limited.This is due in part to our incomplete understanding of the mechanisms thatregulate nuclear size While some progress has been made in identifyingphysiological nuclear size regulators, these factors can affect other cellularprocesses as well as nuclear size Because of these potentially pleiotropiceffects, interpreting the functional significance of nuclear size by manipu-lating these proteins can be problematic Nonetheless, some correlativestudies implicate nuclear size in cell and nuclear function Here we reviewthese studies and also speculate on the potential functional significance ofnuclear size in chromatin organization, gene expression, nuclear mechanics,and disease

4.1 Chromosome[73_TD$DIFF]Positioning, Chromatin Organization,

and Gene Expression

Chromosome positioning is not random within the three-dimensional ume of the nucleus Rather, chromosomes occupy specific positions, calledchromosome territories (Cremer and Cremer, 2010) Chromosomal move-ments are generally limited to short distances (Marshall et al., 1997), althoughsome genes can change their position within the nucleus from the periphery

vol-to the center without large-scale changes in chromosome positioning(Albiez et al., 2006; Misteli, 2007) Long distance chromosome movementsare rare and are usually observed during cell differentiation or transformation(Zink et al., 1998) High mobility chromosome regions can move at[209_TD$DIFF]0.1–0.5μm/s (Bystricky et al., 2005; Chubb and Bickmore, 2003).[75_TD$DIFF]Noncodingchromosomal regions, such as centromeres (Aquiles Sanchez et al., 1997;Hou et al., 2012) and telomeres (Ebrahimi and Donaldson, 2008), tend to be

located at the nuclear periphery in association with the NE Certain mosomes exhibit tissue-specific positioning (Kim et al., 2004; Parada et al.,

chro-2004) For example, in lymphoblasts and fibroblasts, chromosomes with largenumbers of genes are located in the center of the nucleus, while gene-poorchromosomes are more peripheral, and these positions are independent ofchromosome size (Boyle et al., 2001) An open question is how nuclearvolume influences these chromosome territories and dynamics

Whether nuclear volume impacts chromatin condensation is anotherpoint of interest Different cell types exhibit different patterns of hetero-chromatin distribution For instance, fibroblasts have an even distribution ofheterochromatin, while neurons have very little heterochromatin and hepa-tocytes show a patchy distribution Experiments with endothelial cellsgrown on micropatterned substrates demonstrated that chromatin conden-sation changed with nuclear volume During cell spreading in the G1 phase

of the cell cycle, nuclear volume increased and chromatin became more

Roca-Cusachs et al., 2008) Similar results were obtained in retina cells of aSCA7 mutant mouse model Mutant nuclei were twice as large in volume

as wild-type nuclei and euchromatin occupied four times more of thenuclear territory A redistribution of heterochromatin was also observed,from the center of the nucleus to the periphery, correlating with a reduction

in histone H1c levels (Kizilyaprak et al., 2011)

Ultimately, it will be important to determine if nuclear size-mediatedchanges in chromatin organization lead to changes in gene expression Cell

Remodeling fibroblast geometry with micropatterned substrates affectednuclear size and shape, actin polymerization and contractility, the nucleo-cytoplasmic distribution of histone deacetylase 3, and the level of histonedeacetylation This, in turn, led to altered chromatin condensation and geneexpression, with larger cells exhibiting larger nuclei and greater histoneacetylation (Jain et al., 2013)

Certain cancers exhibit characteristic alterations in chromatin tion (Parada and Misteli, 2002) One protein that plays a role in chromatinorganization and nuclear architecture is CDCA2 (cell division cycle associ-ated 2) or Repo-Man This protein is involved in NE assembly and chro-matin remodeling In HeLa cells, activated Repo-Man binds to PP1γ (pro-tein phosphatase 1γ) leading to dephosphorylation of histone H3 at the end

organiza-of mitosis Repo-Man also binds importinβ and recruits it to chromosomesduring anaphase, promoting NE formation In cancer cells, Repo-Man is