D Zal Een Astronaut Zijn Die Lelijk Zijn Enkel Verzwikt in De Krater, Die Hij Als Eerste Te Laat Ontdekte

D zal een astronaut zijn die lelijk zijn enkel verzwikt in de krater, die hij als eerste te laat ontdekte.

Mischa Andriessen

I declare that this dissertation and the data presented are the result of my own work, as developed between 2010 and 2014 in the laboratory of Dr. Lars Jansen at the Instituto Gulbenkian de Ciência in Oeiras, Portugal. Where appropriate, specific contributions by colleagues and collaborators are acknowledged in the Author Contributions section and by co- authorship.

Declaro que esta dissertação é da minha autoria e que os dados aqui incluídos são o resultado de trabalho original por mim desenvolvido entre 2010 e 2014 no laboratório do Dr. Lars Jansen no Instituto Gulbenkian de Ciência em Oeiras, Portugal. Sempre que apropriado, contribuições específicas dos colegas e colaboradores são reconhecidos na seção Author Contributions e por co-autoria.

Financial support was granted by Fundação para a Ciência e a Tecnologia, doctoral fellowhip SFRH/BD/74284/2010.

Apoio financeiro da FCT e do FSE no âmbito do Quadro Comunitário de apoio, BD nº SFRH/BD/74284/2010.

To be defended at the Instituto Gulbenkian de Ciência in Oeiras, Portugal on the 8th of June 2015, before a jury composed of:

Prof. Bill Earnshaw (Wellcome Centre for Cell Biology, Edinburgh, UK); Prof. Kerry Bloom (UNC, Chapel Hill, NC, USA); Dr. Reto Gassmann (IBMC, Porto, PT); Dr. Jorge Carneiro (IGC, Oeiras, PT); Dr. Lars Jansen (IGC, Oeiras, PT);

and presided over by a yet to be determined representative of ITQB

Printed in February, 2015

Dani Bodor

Table of C0ntents

Summary — p.ii; Resumo em Português — p.iii; Acknowledgements — p.iv; List of Publications — p.ix

1. General Introduction: Epigenetics, Centromeres, and Quantitative Biology P.1

Epigenetics — p.3; Centromeres — p.19; Quantitative Biology — p.34; References — p.46

2. Analysis of Protein Turnover by Quantitative SNAP-Based

Pulse-Chase Imaging P.71

Introduction — p.73; Pulse-Chase — p.77; Quench-Chase-Pulse — p.81; Combining SNAP Experiments with Cell Synchronization and RNAi — p.85; Live Imaging of Pulse Labeled Cells — p.91; Automated Quantification of SNAP- Tagged Protein Turnover at Centromeres — p.95; Supporting Protocols — p.102; Background Information — p.108; References — p.117; Appendix: Maps of SNAP- and SNAPf-tags — p.120

3. Assembly in G1 phase and Long-Term Stability are Unique

Intrinsic Features of CENP-A Nucleosomes P.125

Introduction — p.127; Results — p.129; Discussion— p.144; Material and Methods — p.147; References — p.151; Supplementary Figures — p.155; Appendix: The Role of CENP-C in CENP-A Dynamics— p.158

4. The Quantitative Architecture of Centromeric Chromatin P.163

Introduction — p.165; Results — p.166; Discussion— p.184; Material and Methods — p.190; References — p.199; Figure Supplements — p.207

5. General Discussion; Or, What I’ve Learned and What I Have to

Say about It P.215

Non-Centromeric CENP-A — p.217; The Ultrastability of CENP-A — p.221; Mass Action vs. Ultrastability — p.228; The Critical Amount of CENP-A — p.232; Concluding Remark — p.237; References — p.238

Summary

A PhD is like a box of chocolates, …… and in this thesis I will present what I got. My work has been focused on a cellular structure that is essential for accurate genome inheritance: the centromere. Centromeres are chromosomal domains that do not rely on the presence of any specific DNA sequence. Rather, they are determined by the presence of a histone variant called CENP-A. Stable transmission of CENP-A containing chromatin is accomplished through 1) an unusually high level of protein stability, 2) self- directed recruitment of nascent CENP-A near existing molecules, and 3) strict cell cycle regulation of assembly. Together, these features lead to a self-sustaining loop that allows for epigenetic maintenance of centromeres. My own contributions to the understanding of epigenetic centromere inheritance are of a quantitative nature. To put my work in context, I will start with an extensive INTRODUCTION of epigenetics, centromeres, and quantitative biology. Next, in CHAPTER 2, I will detail two of the main methodologies that have allowed for the quantitative analysis of centromere inheritance in subsequent chapters. These are, firstly, fluorescent SNAP- based pulse-labeling, used to distinguish between old and new protein pools; and secondly, a macro for ImageJ that I have developed, allowing for the accurate and unbiased quantification of fluorescence signals at centromeres. In CHAPTER 3, the cis requirements for assembly and extreme stability of centromeric nucleosomes are analyzed. I demonstrate that both G1 phase loading and long-term centromeric retention are unique features of the (CENP-A/H4)2 subnucleosomal core, and are self-directed through a CENP-A encoded targeting domain. CHAPTER 4 provides a quantitative analysis of centromeric chromatin. The absolute number of CENP-A molecules at centromeres has been determined in addition to its quantitative regulatory mechanism and distribution. Finally, an overarching DISCUSSION of my results is presented, providing an outlook on how my findings can guide future centromere research.

Resumo em Português Um doutoramento é como uma caixa de chocolates, ..... e nesta tese vou apresentar o que eu consegui. O meu trabalho focou-se numa estrutura celular essencial para fidelidade do processo de herança do genoma: o centrómero. Centrómeros são regiões cromossômicas que não dependem da presença de nenhuma sequencia de ADN específica. Invés, são determinados pela presença de uma histona chamada CENP-A. A transmissão estável de cromatina contendo CENP-A é possível graças 1) a uma inusual alta estabilidade da proteina, 2) o auto recrutamento da CENP-A nascente com base na presença da proteína antiga, 3) e um alto nível de regulação da sua incor- poração durante o ciclo celular. Em conjunto, estas princípios asseguram um ciclo auto sustentável de manutenção epigenética dos centrómeros. A minha contribuição para a compreensão da herança epigenética do centrómero é de natureza quantitativa. Para contextualizar o meu trabalho, começo com uma INTRODUÇÃO extensa da epigenética, dos centrómeros, e da biologia quantitativa. No CAPÍTULO 2, detalho duas das metodologias que foram usados nos capítulos seguintes para a análise da herança centromé- rico. Estas são, primeiro, marcação fluroescente baseada em SNAP-tagging, usada para distinguir as populações de proteinas antigas e novas; e segundo, uma macro de ImageJ desenvolvida por mim, que permite a quantificação dos sinais fluorescentes do centrómero de uma maneira precisa e imparcial. No CAPÍTULO 3 são analizados os requerimentos em cis da incorporação e estabilidade extrema dos nucleossomas CENP-A. Demonstro que, ambas incorporação na fase G1 e retenção centromérica a longo prazo, são pro- priedades únicas da estrutura sub nucleossomal (CENP-A/H4)2, e definidas por um domínio intrínseco de CENP-A. O CAPÍTULO 4 fornece uma análise quantitativa da cromatina centromérica. O número absoluto de moléculas de CENP-A nos centrómeros foi determinado, assim como o aspecto quantita- tivo do mecanismo da sua regulação e distribuição. Por último é apresentada uma DISCUSSÃO abrangente dos meus resultados e do impacto que as minhas descobertas trazem na orientação da futura investigação centromérica.

iii

Acknowledgements

Honestly, I don’t really know where to begin. So many people have been helpful and supportive in so many ways. I guess maybe I should start by acknowledging those that I’m sure to forget further on: you deserve my fullest gratitude as well as my most humble apology. Also, I do apologize for this utterly unsophisticated and extensive acknowledgements section, if my (ab)use of the English language bothers you (which it probably should), please skip it; I promise that the rest of the thesis is much more eloquent. OK, moving on... Lars, I am really happy with the relationship that we’ve built up over the past 6 years. I think that from the first moment we were on a very similar wavelength regarding many things and have become even more in phase over the years. I am also very happy with the type of ‘supervision’ that I received from you: lots of hands-on support initially when I needed it; lots of independence later on when I appreciated that; always supportive to my random whims —whether to take an extra day off for yet another frisbee tournament or apply to a $10.000 course with a deadline in 2 days; you were always ok with it. I also very much appreciate the personal connection that I think we had from the beginning. I have enjoyed immensely working with and for you and couldn’t have asked for a better PI. Yet, everyone in the EpiLab has been an amazing and fruitful collaborator over the years. Ana, it’s awesome to have a great buddy like you in the lab. I love our (many many) coffee breaks with random jumps from tedious boring discussions of antibody dilutions to tales of last weekend’s drinking bouts and bitching sessions about [....CONFIDENTIAL INFORMATION...]. Filipa, it has been an absolute pleasure working with you. I could not have asked for a better student and if you have learned even half as much from me as I have from you, then I would be as proud of myself as I am of you. Mariana, thanks for welcoming me to the lab and to the country from the very beginning. I very much appreciated the heated arguments and

iv the cold beers that we often shared. Luís, thanks dude, it was really fun having you around for a while. Mariluz, Maxi, Dragan, Nuno, it has been great having worked with all of you; Ruben, Sreyoshi, Wojtek, I wish you all the luck in the EpiLab and am sorry we have only barely had an opportunity to work together. Still missing one EpiLabber, right...: João, I know that you always say that you’re just doing your job —and you probably actually really feel that you do— but you do so much more. Whenever needed, whatever’s the matter, you are always ready to be as helpful as humanly possible! Whether it is to drop me off at the airport, fix my computer over the weekend, lend me your car for random errands, or discuss for a few hours a single sentence of some random translation I need for some obscure reason, I know I can always count on you. And then I won’t even mention the immense help you are in the lab, which one could potentially argue (although I personally wouldn’t) is indeed part of your job. Please know that all this, as well as your friendship over the last years, is and always has been very much appreciated. Hangout-clan, thanks a whole frickin’ bundle for sharing the joys (not many) and pains of thesis writing. The countless screens we’ve shared as well as the p*** that we didn’t was instrumental in pulling me through and I hope it’s been as useful for you too. Ewa, thanks for your patience, advice, and help about the tedious details of putting together a representable thesis. Also thanks to the theses of Babs, Ewa, Ines, Mariana, Mariluz, and Matilde for being great examples of what my boekje should look like. I would also like to thank the IGC for having been a great host institution. The open-lab philosophy and highly interactive atmosphere created here has been extremely stimulating and productive for both work and social purposes. A special thanks goes to everyone that has passed through the Zheng-Ho wing and to the cell cycle club and chromatin club communities. Nuno, the first sentence you said to me when you saw me — “what do you think you’re doing” — and the resulting collaboration has been one of the most influential events of my entire PhD, although one of the few

v things that may parallel it was the microscopy course you organized, which taught me to think like a microscope. Alekos, Mónica, Jorge, Raquel, thanks for the many fruitful discussions we’ve had about my projects. I am also very much indebted to everyone at the 2012 Physiology course for having reshaped my way of thinking about scientific problems and solutions. Thanks also to Élio for getting me out of a pickle: I was really reluctant to sit it out and your help was probably the one thing that could’ve and did rescue me. Indeed, my mind reels with appreciation of what it means to have been able to do a PhD at ITQB. Tons of thanks also go to all the people that made my time in Portugal and at IGC soooooo much fun for such a long time. An incomplete list could be (in alphabetical order): Ana, Babs, Cláudia, Ewa, Filipa, Inês, Jess, João, João Beer, Jordi, Jorge, Krzys, Laura, Lars, Luís, Mada, Marc, Mariana, Mihailo, Nicole, Nuno, Pol, Roksana, s, Stefan, Tiago. Also lots of thanks to everyone who has kept on throwing discs at me to keep me sane all this time, especially Sof, Trick, Patrão, Carla, Cons, Rui, Fred, Rui, Pifre, Inês, Seb, Morris, and of course Filipa who introduced me to this all. ZZ, KJAJBDTK! Natuurlijk gaat er ook onwijze dikke dank naar al mijn lieve vrienden, ex-collega’s en mentoren thuis, die mij na al die lange jaren hopelijk nog niet vergeten zijn. Sander, Adri, Matilde (en alle anderen waarmee ik in Sander’s lab gewerkt heb); Paulien, Stan en Veronica; jullie hebben stuk voor stuk op een onmiskenbare manier bijgedragen aan de vorming van de wetenschapper die ik vandaag ben, en ik herken in mezelf nog steeds de specifieke invloed van ieder van jullie. Piet, Petertje, Matthia, it has been a joy and honor om samen met jullie biologisch grootgebracht te worden: onze tijden van SPI___RAAL, ik spreek Oebli-Oebli en in je broek waren onmisbaar om mij de biloloog te maken die ik vandaag ben. Sanne, Ditte, Banafsheh (waarschijnlijk de enige 3 buiten mijn familie waarvan ik ervoor zorg dat ik elke keer dat ik in Nederland ben minstens een klein beetje tijd vind om bij te kletsen): hoera!

“About 99% of everything you hear is untrue.” I think that this single sentence, which I was told probably around age 9, instantaneously transformed me into a scientist. Peter, you probably don’t even remember saying this to me, but I will never forget (at least, well, I haven’t forgotten it yet). (MaPaDaNo(SaToMi)); Worte fehlen mir... ausser: Danke für alles! Kommt noch eine Person die ich noch nicht gennant habe: Papa, ich widme dir diese Dissertation. Ich glaube es gibt niemanden auf der Welt der einen grösseren Einfluss auf meine Bildung, in jeder möglichen Hinsicht, gehabt hat. Papa, es tut mir schrecklisch leid das du nicht hast sehen können was aus mir geworden ist. So it goes

vii

viii

List of Publications

In chronological order:

Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63

Bodor DL, Rodríguez MG, Moreno N & Jansen LET (2012) Analysis of Protein Turnover by Quantitative SNAP-Based Pulse-Chase Imaging. Curr. Protoc. Cell Biol. Chapter 8: Unit8.8

Bodor DL, Valente LP, Mata JF, Black BE & Jansen LET (2013) Assembly in G1 phase and long-term stability are unique intrinsic features of CENP-A nucleosomes. Mol. Biol. Cell 24: 923–932

Bodor DL & Jansen LET (2013) How two become one: HJURP dimerization drives CENP-A assembly. EMBO J. 32: 2090–2092

Bodor DL, Mata JF, Sergeev M, David AF, Salimian KJ, Panchenko T, Cleveland DW, Black BE, Shah JV & Jansen LET (2014) The quantitative architecture of centromeric chromatin. eLife Sciences 3: e02137

CHAPTER 1

General Introduction: Epigenetics, Centromeres, and Quantitative Biology

Dani L. Bodor

Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal.

Epigenetics, Centromeres, Quantitative Biology

EPIGENETICS

Inheritance systems Inheritance from a biological perspective is the transfer of information from one (cell) generation to th e next. In order for a system of inheritance to persist, a number of criteria need to be fulfilled. The bare minimal requirement is that there is a carrier (or carriers) of information that can be propagated through generations. In addition, to allow for sustained passage of information into subsequent generations, the carrier needs to be replicated in each generation. Moreover, in many cases it is important that there is careful regulation to ensure that the correct number of heritable units is passed on. Temporal regulation can play a role in quantitative control so that e.g. one new unit is formed for each pre-existing one exactly once per cell generation. In summary, the basic properties of a successful inheritance system include: 1) propagation, 2) replication, and 3) copy- number regulation. Up to the early 1940s, there was a heated debate on the molecular nature of heritability. Two opposing ideas were that either protein or nucleic acids would be the carriers of genetic information (Deichmann, 2004). Among other factors, the low apparent complexity of DNA led to the common notion that genes were more likely composed of proteins. However, In the 1940s and ‘50s a number of breakthrough discoveries were made that irrevocably showed that, in fact, DNA was responsible for genetic inheritance. Instrumental were experiments showing that DNA is the agent that is responsible for the transformation of non-virulent into virulent pneumococcus (Griffith, 1928; Avery et al, 1944), as well the famous Hershey-Chase experiment, showing that viral DNA, but not protein, enters the host upon bacteriophage infection (Hershey & Chase, 1952). Soon afterwards, Watson and Crick published their breakthrough model of the double-helical structure of DNA, including the now famous statement “it has not escaped our notice that the specific pairing we have postulated

Chapter 1 immediately suggests a possible copying mechanism for the genetic material” (Watson & Crick, 1953a). Indeed, the semi-conservative ‘copying mechanism’ that was intended, where each of the two existing strands of DNA form the template for a nascent strand (Figure 1.1A), was later confirmed by Meselson & Stahl (1958) in what is often called ‘the most beautiful experiment in biology.’ Much later, and over the course of decades, the regulation mechanisms were elucidated, which ensure that the entire genetic complement is replicated exactly once per cell division, such that there is no under- or overduplication of the genetic material (Sclafani & Holzen, 2007). In short, once per cell division cycle, a defined number of replication origins are licensed with an initiation complex that is consumed when DNA replication begins at this site, thus ensuring that the same stretch of DNA is not replicated more than once. In addition, progression of cell division is halted until a complex machinery, called a checkpoint, has ensured that DNA replication is complete. In conclusion, although some details may still need to be resolved, a fairly good understanding of the mechanism of genetic inheritance has emerged. As is clear from the section above, DNA perfectly fits all criteria given above for the carrier of heritable information. This molecule is stably propagated when cells divide, it is replicated after each cell division, and regulated such that each molecule gives rise to only one new molecule exactly once per division. Thus, genetic inheritance is a showcase model of an effective inheritance system.

Non-genetic inheritance Ever since the discovery that DNA was the carrier of genetic information, the study of inheritance from a biological perspective has been dominated by DNA and its nucleotide sequence. This system is perfectly able to account for Mendel’s laws of inheritance (Mendel, 1866) as well as some more complex variations of these principles, which together govern inheritance of the majority of traits in sexually reproducing organisms. However, certain

Epigenetics, Centromeres, Quantitative Biology heritable features do not strictly dependent on the genetic code of a cell. This is most apparent from the fact that within a single multicellular organism there can be many different cell types with the exact same genetic material. Generally, when cells that have acquired a certain developmental status divide, they give rise to the same cell type, e.g. a dividing skin cell will not suddenly give rise to a heart muscle cell, and vice versa. In addition to such non-genetic inheritance that is contained within a single organism, a number of transgenerationally inherited traits have been described that do not seem to follow the typical laws of inheritance. One famous example is ‘helmet’ size in the waterflea Daphnia cucullata: if exposed to a predator, the size of this protective structure is altered throughout multiple generations (Agrawal et al, 1999), even in the absence of a predatory cue in the offspring. Another well-known case is the toadflax Linaria vulgaris that exists in two distinct heritable morphological states, but can spontaneously switch between generations without any apparent mutations in the responsible gene (Cubas et al, 1999). Thus, there must be other structures present in cells that are able to carry information from mother to daughter cells, or even through organismal generations. Below, some typical examples of alternative inheritance systems, and their method of transferring information, are discussed.

Self-sustaining loops Perhaps the simplest possible form of (non-genetic) inheritance is a self- sustaining loop (Figure 1.1B). If the expression of a gene is driven by its own product (protein or RNA), then the cytoplasmic inheritance of this factor during cell division will ensure that the active state of the gene will also be inherited (Rosenfeld, 2011). Gene products can either drive such feed- forward loops directly (e.g. a transcription factor that activates the gene by which it is produced), or, more commonly, indirectly (e.g. a protein that initiates a genetic cascade, ultimately leading to its own expression). In either case, gene-activity will effectively be maintained throughout generations until it is actively (or spontaneously) interrupted. This type of

Chapter 1 self-sustaining loop is common in bacteria and other unicellular organisms (Santillan et al, 2007; Jablonka & Raz, 2009), and likely contributes to the maintenance of cell identity in multi-cellular organisms as well (Hobert, 2011; Holmberg & Perlmann, 2012; Ptashne, 2013).

Figure 1.1 Examples of inheritance systems. (A) DNA is replicated in a semiconservative fashion. During replication, a single DNA duplex untwines and individual nucleotides on each strand form the template for production of a new strand of DNA (image adapted from: The Nucleus and DNA Replication, 2015). (B) Once initiated by an external cue (indicated by a bomb), gene products that maintain their own expression through a self-sustaining loop can be inherited through the cytoplasm during cell division. In this way, they maintain their activity in the next cellular generation, even in the absence of the original initiating signal (image adapted from: Jablonka & Lamb, 2006). (C) Prion transmission is an example of structural inheritance. The amyloid protein conformer (red) catalyzes conversion of native protein isoforms of identical amino acid sequence (blue balls) into its own conformation (image adapted from: Shorter & Lindquist, 2005). (D) DNA methylation is the best understood form of chromatin-based epigenetics. DNMT3 is a de novo methyltransferase that is capable of adding methyl groups (red hexagons) to cytosines on unmethylated DNA. During DNA replication, the maintenance methyltransferase DNMT1 associates with the core replication machinery and specifically methylates hemimethylated DNA, thus retaining the pre-replication methyl-pattern in the next generation. Conversely, TET enzymes can oxidize methylated cytosine into hydroxymethylcytosine (orange hexagons), which can initiate a pathway that restores unmethylated DNA (image adapted from: Li & Zhang, 2014).

Epigenetics, Centromeres, Quantitative Biology

In this inheritance system, the carrier of heritable information is the gene product (let’s call it Factor X). Factor X is propagated through the cytoplasm of a dividing cell, oftentimes by random segregation of the total pool of existing molecules (Rosenfeld et al, 2005). Replication in the next generation is achieved by activating the gene that is responsible for producing Factor X. Although in this case there is no absolute requirement for copy number regulation with a high degree of accuracy, sufficient molecules are required to ensure that each daughter sustains and perpetuates gene activity. In summary, self-sustaining loops represent a very basic example of a stable inheritance system.

Structural inheritance In some cases, a given three dimensional structure propagates itself by forming the template for assembly of the same structure. Perhaps the most elegant (and best understood) structural inheritance system is in fact genetic inheritance, where nascent strands of DNA are templated onto existing molecules (Watson & Crick, 1953a, 1953b; Meselson & Stahl, 1958). However, many additional structural inheritance systems have been described. A clear example are prions (Figure 1.1C), proteins of identical amino-acid sequence that can exist in multiple conformational states, at least one of which drives conversion of the other(s) into itself (Prusiner, 1982, 1998; Halfmann et al, 2010). Although prions are generally considered detrimental or pathogenic, it has been shown that they can have a physiological role by conferring advantageous traits in certain environments (Halfmann et al, 2010, 2012). Prion inheritance is in many ways analogous to the self-sustaining loops described above (it is itself a type of feed forward loop), as it is inherited through the cytoplasm where it will replicate by mediating a nascent protein isoform into its own conformational state. An interesting case is presented by the centrosome, the primary microtubule organizing center (MTOC) in most animal cells. A single centrosome contains two centrioles, cylindrical structures composed mainly

Chapter 1 of tubulin, each of which nucleate a nascent daughter centriole exactly once per cell division cycle (Bettencourt-Dias & Glover, 2007; Nigg & Stearns, 2011). Conversely, centrioles can also form de novo under certain conditions, although this is strongly suppressed be the presence of pre- existing centrosomes (Marshall et al, 2001; Terra et al, 2005; Rodrigues- Martins et al, 2007). However, this inheritance mechanism differs from true structural inheritance, as there is no evidence for actual templating of one centrosome against another. Rather, centrosomes are more likely sites where enzymes, regulatory, and structural proteins accumulate to regulate the biogenesis of nascent structures (Rodrigues-Martins et al, 2007, 2008), allowing for a semi-conservative replication mechanism that is carefully regulated by the cell cycle (Bettencourt-Dias & Glover, 2007; Nigg & Stearns, 2011). In this system, the carrier of heritable information are the centrioles, although it is not completely clear what the information is that they carry. Nevertheless, their replication is strictly regulated in time, space, and number to ensure the propagation of the correct number of structures to the following generation. Other examples of structural (or structural-like) inheritance systems include the organization of ciliary rows on the cell cortex of certain ciliates (Sonneborn, 1964), cellular membranes (Cavalier-Smith, 2004), certain organelles (Warren & Wickner, 1996), or even the cell as a complete entity. In summary, structural inheritance is a common mechanism to pass information from one generation to the next.

Chromatin-based epigenetics The term epigenetics was originally coined by Conrad Waddington in 1942 to indicate “the mechanism by which the genes of the genotype bring about phenotypic effects” (Waddington, 1942). In this definition, epigenetics does not refer to any heritable features, but is more similar to what today is considered gene regulation or developmental biology. However, throughout the last 70-odd years, the word epigenetics has been used and redefined in

Epigenetics, Centromeres, Quantitative Biology many different ways (Jablonka & Lamb, 2002; Bird, 2007; Marris et al, 2008). One very broad definition of an epigenetic phenomenon is: “a change in phenotype that is heritable but does not involve DNA mutation” (Gottschling, 2004). However, if taken literally, this definition encompasses certain heritable features that are usually not intended, such as traits acquired through social learning (Jablonka & Lamb, 2005; Shea, 2009) or vertically transmitted infections and symbionts (Ford-Jones & Kellner, 1995; Moran et al, 2008). Nevertheless, more recently, during a conference on chromatin-based epigenetics at Cold Spring Harbor, a consensus definition was formulated as: “a stably heritable phenotype resulting from changes in a chromosome without alterations in the DNA sequence” (Berger et al, 2009). Perhaps unsurprisingly, this consensus definition only includes what the main topic of the conference was, namely chromatin-based epigenetics (see below), while excluding all other potential forms of epigenetics, including self-sustaining loops and structural inheritance. In my own opinion, the most useful definition of epigenetic inheritance goes along the lines of: information that cells can pass to their progeny without changing their DNA sequence (paraphrased from Jablonka & Lamb, 2005, p. 113). In this case, heritable features at the cellular molecular scale (e.g. self-sustaining loops and structural inheritance) are included, while features heritable at the organismal scale (e.g. symbiosis and learning) are not. Deceptively, yet more definitions exist outside of biology, e.g. epigenetic robotics, which is related to machine learning (Prince & Demiris, 2003), and the epigenetic theory of human development, a psychological theory of transitions in human development through psycho-social crises (Erikson, 1950). Therefore, although I only partially agree, Adrian Bird makes a reasonable point when he says: “epigenetics is a useful word if you don't know what's going on — if you do, you use something else” (Marris et al, 2008). Despite the ongoing controversy on the exact meaning of epigenetics, practically speaking, chromatin-based epigenetics is the most actively

Chapter 1 studied form of non-genetic inheritance. The structure and organization of chromatin allows for a plethora of modifications, many of which can either be inherited or participate in a pathway that drives inheritance. In addition, this complex nature of chromatin allows for tight control of the transmission of the epigenetic signal. I will first proceed with a brief introduction on chromatin and then delve deeper into its role in epigenetic inheritance.

Chromatin structure Generally, the existence of chromatin is attributed to the necessity of fitting a large (eukaryotic) genome into a much smaller nucleus. If we take human cells as an example, the total length of the 46 chromosomes, together comprising over six billion base pairs of DNA, would exceed two meters if placed head-to-tail (Flicek et al, 2014). However, in analogy to packing a suitcase, it does not make much sense to lay all ones clothes in a neat line next to other and then wonder how this will ever fit into a small carry-on bag (Morse, 2013). Similarly, chromosomes are not linearly extended molecules, but are folded and packaged into three-dimensional structures. In fact, the paradox of fitting 2 meters worth of DNA into an average sized nucleus of ~7 μm in diameter is easily resolved by the fact that the volume of this nucleus is almost 30 times as big as that of the total DNA (respective volumes ~180 μm3 and ~6.3 μm3). Thus, chromatinization is a means of proper folding of the DNA, and has additional roles in organizing and regulating the genome. The primary organizational unit of chromatin is the nucleosome (Kornberg, 1974; Olins & Olins, 1974). A single nucleosome consists of ~145 bp of DNA tightly wrapped around an octamer consisting of two copies of each of the histone proteins H2A, H2B, H3, and H4 (Luger et al, 1997). The octamer itself is composed of a central (H3/H4)2 tetramer, flanked by two H2A/H2B dimers. These core histones are among the most highly conserved eukaryotic proteins (Sullivan et al, 2000, 2002; Malik & Henikoff, 2003), arguing that little structural variability is tolerated for their function. This is especially true in their histone fold domain (HFD), which form the major

Epigenetics, Centromeres, Quantitative Biology interactions between the separate histones as well as with the DNA (Luger et al, 1997) and are 100% identical between human and certain plants and fungi (Sullivan et al, 2002). Histone H1 serves as a linker-histone, which binds DNA between neighboring nucleosomes, thereby helping to stabilize the chromatin structure (Thoma et al, 1979). Further organization is likely achieved by multiple forms of higher order structures, the precise in vivo nature of which has proven to be very challenging to determine (Woodcock & Ghosh, 2010). Despite the high level of conservation and strong interaction of the histone-DNA binding, chromatin is both a heterogeneous and a dynamic structure (Gasser, 2002; Flaus & Owen-Hughes, 2004; Chakravarthy et al, 2005). Indeed, both replication and transcription machineries displace, reorganize, and remodel the nucleosomes as DNA and RNA polymerases, respectively, plough through the chromatin (Mousson et al, 2007; Groth et al, 2007). In addition, certain regions of the chromosome can be highly compacted, while flanking regions remain accessible to external factors, such as transcription factors or other DNA binding proteins (Wu et al, 1979; Larsen & Weintraub, 1982; Song et al, 2011). Furthermore, major rearrangements of this chromatin organization commonly occur, e.g. throughout the cell cycle (Reeves, 1992; Aragon et al, 2013; Raynaud et al, 2014) and during cell differentiation (Meshorer & Misteli, 2006; Kobayakawa et al, 2007; Probst & Almouzni, 2008). In summary, while composed of fairly simple units, chromatin is a highly complex structure that is regulated at the level of configuration, organization, and dynamics. Consistent with its complexity, a large variety of processes exist that help effectively regulating chromatin homeostasis and dynamics in cells. The close association of chromatin and its modifications to the genome of the cells makes it an excellent candidate for driving epigenetic inheritance, e.g. of gene activities. Three of the major mechanisms are DNA methylation, incorporation of histone variants, and modification of histone proteins. Each of these processes has the potential, supported at least by some evidence, to drive epigenetic inheritance, and will be briefly discussed below.

Chapter 1

DNA methylation DNA methylation, the covalent addition of a methyl group to the DNA backbone, is found throughout the tree of life (Colot & Rossignol, 1999; Jeltsch, 2002; Ponger & Li, 2005). However, this modification was lost multiple times in evolution, and is absent from a wide variety of species including D. discoideum, S. cerevisiae, S. pombe, and C. elegans (Ponger & Li, 2005). Methylation of DNA can affect many cellular processes, including gene-regulation, transposon silencing, heterochromatin formation, and susceptibility to restriction enzymes, depending to some extent on the species (Colot & Rossignol, 1999). In eukaryotes, methylation at carbon 5 in the pyrimidine ring of cytosine, thus creating 5-methylcytosine (meC), is the only known form of methylated DNA (Jeltsch, 2002). In plants, any cytosine in the genome has the potential to be methylated, although separate enzymes are responsible for the methylation of CG-dinucleotides (CpG), CHG-sites (where H is any non-guanine nucleotide), and CHH-sites (Law & Jacobsen, 2010). In mammalian cells, however, DNA methylation is largely restricted to CpGs (Sinsheimer, 1955), although low levels of meC can be observed on other sites, especially in germ and stem cells (Ramsahoye et al, 2000; Ichiyanagi et al, 2013). Importantly, not every potential site is methylated, e.g. ~14% of cytosines are methylated in Arabidopsis thaliana leaf tissue (Capuano et al, 2014), while ~70–80% of CpGs are methylated in somatic human tissues (Ehrlich et al, 1982; Bird, 2002). Furthermore, the pattern of methylation can differ between different cell types of the same organism and change during differentiation (Reik et al, 2001). Thus, sequence determinants are not sufficient to explain the existing pattern of DNA methylation. The vast majority of meC sites in the mammalian genome are symmetrically methylated. In other words, either both strands of a minipalindromic CpG site are methylated, or neither is (Bird, 1978). However, the process of DNA replication inevitably leads to the formation of hemimethylated DNA, where a nascent strand of unmethylated DNA is

Epigenetics, Centromeres, Quantitative Biology templated against a methylated pre-existing strand. The DNA methyltransferases DNMT1 has been shown to have a high preference for hemimethylated DNA (Bestor & Ingram, 1983) and associate with the core DNA replication machinery protein PCNA (Chuang et al, 1997) as well as with NP95, which specifically recognizes hemimethylated DNA (Sharif et al, 2007). In this way, DNMT1 is accurately targeted to hemimethylated DNA during its formation and can restore the pre-existing pattern of methylation. This shows that DNA methylation is a semiconservatively inherited epigenetic feature and intrinsically coupled to cell cycle regulation (Figure 1.1D). Although DNA methylation is generally considered a stable epigenetic modification, its genomic pattern is largely reset in each generation. Demethylation can potentially occur in two fundamentally different ways. One is the passive dilution of meC during successive rounds of DNA replication in the absence of maintenance methylation. The other is by active removal of methylated cytosines, although claims of finding such mechanisms have a history of being highly controversial (Ooi & Bestor, 2008). Only recently has a bona fide active demethylation pathway been described, where meC is iteratively oxidized into hydroxymethylcytosine (Tahiliani et al, 2009), formylcytosine, and carboxylcytosine (Ito et al, 2011; He et al, 2011), the latter two of which can be converted back to unmodified cytosine through base-excision repair (He et al, 2011; Maiti & Drohat, 2011). This pathway may explain how, in the absence of replication, methylated DNA is rapidly lost from the mouse paternal pronucleus after fertilization (Mayer et al, 2000; Oswald et al, 2000). Embryonic stem cells re-initiate a nascent pattern of DNA methylation (Jähner et al, 1982; Stewart et al, 1982) using the de novo DNA methyltransferases DNMT3a and DNMT3b (Okano et al, 1998, 1999). However, a recent analysis on the genome-wide methylation patterns of three great apes, including human, argues that methylation patterns can gradually change over generations and may ultimately even contribute to variability between species (Martin et al, 2011;

Chapter 1

Boffelli & Martin, 2012). Nevertheless, generally speaking, it appears that DNA methylation in mammals is mainly involved in epigenetic inheritance through mitotic divisions, and has a relatively minor role in transgenerational inheritance.

Histone variants As mentioned above, canonical nucleosomes contain a histone octamer consisting of four different types of histone proteins: H2A, H2B, H3, and H4. Multiple different variants exist for each of these histone proteins in most species analyzed (Talbert et al, 2012), with the exception of H4, for which a sole known non-canonical variant exists in Trypanosoma (Siegel et al, 2009). In humans, up to 47 non-allelic variants, i.e. proteins with different amino acid sequence, have been described in total for the four nucleosomal histones (Wiedemann et al, 2010; Khare et al, 2011). However, it remains unclear whether each variant actually has distinct properties, especially in cases with only one or few residues divergence. Nevertheless, one example where this is indeed the case is histone H3.3, which differs from its canonical H3.1 counterpart by a mere 5 amino acids, yet its dynamics and regulation are drastically different. H3.1 is assembled throughout the genome by the CAF complex in a strictly DNA replication- coupled manner, while H3.3 assembly occurs preferentially at specific loci by the histone chaperones HIRA, DAXX, and DEK and is independent of the cell cycle (Smith & Stillman, 1989; Ray-Gallet et al, 2002; Ahmad & Henikoff, 2002; Tagami et al, 2004; Drané et al, 2010; Goldberg et al, 2010; Sawatsubashi et al, 2010). Therefore, altered histone variant compositions of the nucleosome are good candidates as carrier of epigenetic information. The process of DNA replication forms an inherent challenge to the local heritability of histones. In order for a megadalton sized replication complex to pass through the chromatin, nucleosomes are disassembled prior to the denaturation and replication of DNA (Groth et al, 2007; Alabert & Groth,

2012). However, pre-existing subnucleosomal (H3/H4)2 tetramers are

Epigenetics, Centromeres, Quantitative Biology recycled behind the replication fork, possibly through their association with the histone chaperone Asf1 (Groth et al, 2007; Mousson et al, 2007; Alabert & Groth, 2012). Conversely, it appears that histones H2A and H2B are more dynamic than H3 and H4 (Jackson, 1987; Kimura & Cook, 2001; Bodor et al, 2013) and thus less likely to carry epigenetic information. Consistently, evidence exists that at least two variants of histone H3 are carriers of epigenetic information. The role of the centromeric variant CENP-A is described in detail in part 2 of the introduction. The replacement variant H3.3 is enriched at sites of high gene activity (Ahmad & Henikoff, 2002; Mito et al, 2005; Goldberg et al, 2010), and is enriched in post-translational modifications associated with active transcription (McKittrick et al, 2004; Hake et al, 2006). Importantly, it has been shown that H3.3 is involved in the resistance to reprogram an active gene expression profile in Xenopus after transplantation of somatic cell nuclei into oocytes (Ng & Gurdon, 2008). Interestingly, a similar role for macroH2A was found by maintaining a repressed state on the X-chromosome (Pasque et al, 2011) and on pluripotency genes (Gaspar-Maia et al, 2013). Although the precise mode of action of these histone variants remains unclear, it appears that they are somehow involved in the transmission of an epigenetic state.

Histone modifications In addition to modifying the histone variant composition of nucleosomes, each of the histones can undergo a large number of post- translational modifications (PTMs). Common modifications on histones include acetylation, methylation, phosphorylation, ubiquitylation, citrullination, biotinylation, and ADP-ribosylation (Khare et al, 2011). Most PTMs exist in the protruding N-terminal histone tails, while only few are found within the HFD (Khare et al, 2011). In some cases, a single residue is known to exist in multiple different modified forms; e.g., lysine 9 of Histone H3 (H3K9) can be mono-, di-, or trimethylated, acetylated, or biotinylated. Indeed, on histone H3 alone, there are at least 44 separate known modifications, spread over 21 individual sites, resulting in over three billion

Chapter 1 potential combinatorial states of modification on each molecule (Khare et al, 2011). Interestingly, many modifications are shown to correlate with specific (functional) states, such as high or low gene-activity, splicing, DNA repair, and DNA replication (Bannister & Kouzarides, 2011). These findings have spurred the hypothesis of a ‘histone code’ that can be read by downstream effector proteins or have a function in epigenetic memory (Strahl & Allis, 2000; Jenuwein & Allis, 2001; Turner, 2002; Rothbart & Strahl, 2014). However, because most PTMs are not exclusively associated with any one particular state (Barski et al, 2007), such a histone code can at best be seen as a highly complex combinatorial code or language (Lee et al, 2010; Rothbart & Strahl, 2014), unlike e.g. the linear genetic code (1 codon => 1 amino acid). Nevertheless, similar to histone variants, PTMs on histone tails have the potential to propagate epigenetic information. PTMs are often equated to epigenetic marks, even in the scientific literature (e.g. Turner, 2002). However, in many cases there is clear evidence that the PTMs are not inherited at all, but are transient structures that mediate e.g. cell cycle progression (Van Hooser et al, 1998), DNA replication (Benson et al, 2006), or DNA repair (Rogakou et al, 1999; Hunt et al, 2013). In addition, for many modifications that are associated to gene- activity, it remains unclear whether they are the cause or consequence of the transcriptional state (Ng et al, 2003; Soshnikova & Duboule, 2009; Muramoto et al, 2010). Nevertheless, while, at least to my knowledge, there is no direct evidence that PTMs carry and transmit epigenetic information, they remain strong candidates at least for certain modifications.

Epigenetics in evolution Above, it has been thoroughly established that heritability is not exclusively mediated by the genome. Although most examples given refer mainly to inheritance of features through mitotic divisions, i.e. within the somatic cells of a single organism, more than 100 examples of transgenerational epigenetic inheritance from 40 different species have been

Epigenetics, Centromeres, Quantitative Biology documented by Jablonka & Raz (2009). Given this wealth of epigenetic heritability, at least some of the traits must be adaptive and advantageous phenotypes to certain environments have been observed for variable methylomes in the plant species Arabidopsis thaliana (Johannes et al, 2009) and Mimulus guttatus (Scoville et al, 2011), as well as prions in S. cerevisiae (Halfmann et al, 2012), heritable antiviral RNA molecules in C. elegans (Rechavi et al, 2011), and gene silencing in D. melanogaster (Stern et al, 2012). Together, these observations lead to the interesting possibility that non-genetic inheritance can contribute to evolutionary dynamics. To illustrate that evolution can be driven by epigenetic inheritance, Jablonka and Lamb (2005) used an interesting thought-experiment approximately along the following lines:

Imagine a faraway planet that is as rich and dynamic a world as our own, featuring many different environments and climates; let’s call it CB (for Complex Biosphere). This world is inhabited by a population of creatures that does not tolerate any divergence in its genome whatsoever; let’s call them SAM (for Species in the Absence of Mutation). Given the richness of the environment, there is a great potential for SAM to adapt to many different niches. Therefore, as time goes by, SAM plays it (again) in a way that does not require any genetic change. Rather, SAM differentially produces epigenetic traits, e.g. through altering the gene methylation states, generating novel prion-like protein conformations, or activating self-sustaining loops. If advantageous in a given milieu, adapted individuals will prosper, compete more successfully for the available resources, and produce a higher number of offspring. Thus, by means of natural selection, the epigenetic diversification of SAM in different environments will ultimately be the origin of species.

Given that imagination is the only weapon in the war against reality, we do not want to argue here that actual evolution is driven solely by epigenetic changes. Nevertheless, this story does clearly make the point that adaptation, and thus evolution, can in principle occur through inheritance of variable, non-genetic traits. Accepting that “variations, however slight and from whatever cause proceeding, if they be in any degree profitable to the individuals of a species [...], will tend to the preservation of such individuals” (Darwin, 1859: p.61; emphasis mine), it is difficult to imagine that natural selection would not work on epigenetically inherited traits.

Chapter 1

The influence of epigenetic mechanisms on evolution could be very different from genetic inheritance. Importantly, reproduction of epigenetic states in the next generation is generally much less accurate than genetic inheritance. For example, while DNA replication occurs at an error rate in the range of ~10-6–10-8 (Kunkel, 2004), errors in copying DNA methylation occur as frequently as ~0.3–4% (Laird et al, 2004; Goyal et al, 2006). Although a higher error rate likely makes epigenetic traits less stable, it may also lead to a more rapid acquisition in response to changing environments (Cubas et al, 1999; Pryde & Louis, 1999). These and other epigenetic specific effects (Jablonka, 2012) make that the classical models of evolution and population dynamics need to be reevaluated. However, only recently have different aspects of epigenetics started to be integrated in such models (Tal et al, 2010; Day & Bonduriansky, 2011; Geoghegan & Spencer, 2012). In addition, epigenetic mechanisms have been proposed to have a role in speciation, macroevolution, and even the major transitions in evolution (Jablonka & Lamb, 2006; Jablonka & Raz, 2009; Boffelli & Martin, 2012; Jablonka, 2012).

Epigenetics, Centromeres, Quantitative Biology

CENTROMERES

The function of centromeres Centromeres were originally defined cytologically by Walther Flemming in the late 19th century, as the site of a ‘primary constriction’ in mitotic chromosomes (Flemming, 1880). Today, we have a fairly good understanding of what brings about this particular structure. During DNA replication, nascent sister chromatids are held together by a protein complex called cohesin (Figure 1.2A), thus preventing precocious separation and chromosome missegregation (Michaelis et al, 1997; Uhlmann & Nasmyth, 1998). Upon entry into mitosis (or meiosis), the chromosomes condense (Koshland & Strunnikov, 1996) and the majority of cohesin is removed from the chromosomes (Losada et al, 1998). However, cohesin is preferentially retained at a single site on each sister chromatid pair, the centromere (Losada et al, 2000; Waizenegger et al, 2000), where it is protected by Shugoshin proteins (Kerrebrock et al, 1995; Salic et al, 2004). Only when cells are ready to exit mitosis and segregate sister chromatids to the daughter cells is the remaining centromeric cohesin cleaved by a protein called separase (Uhlmann et al, 1999, 2000). Thus, centromere specific cohesion is responsible for the X-shaped conformation of mitotic chromosomes and Flemming’s primary constriction (Haarhuis et al, 2014). Centromeres are also the chromosomal loci that form the point of contact between the DNA and the mitotic spindle (Figure 1.2B). A large group of proteins, the constitutive centromere associated network (CCAN), are present at the centromere throughout the cell cycle (Foltz et al, 2006; Izuta et al, 2006; Cheeseman & Desai, 2008). During mitosis, the CCAN recruits a secondary protein complex known as the kinetochore, which includes the conserved microtubule-binding KMN network, consisting of the protein KNL1 as well as the Mis12 and Ndc80 complexes (Cheeseman et al, 2004, 2006; DeLuca et al, 2006). Poleward directed pulling forces are exerted on centromeres by stable binding of depolymerizing microtubules at

Chapter 1 kinetochores, which drag the sister chromatids in opposite directions during anaphase (Brinkley & Cartwright, 1975; Salmon et al, 1976; Mitchison et al, 1986; Inoué & Salmon, 1995). Thus, the centromere is the primary structure responsible for recruiting the entire chromosome segregation machinery.

Figure 1.2 Centromeres control chromosome segregation. (A) Sister chromatid cohesion is maintained specifically at centromeres during mitosis to prevent precocious chromosome separation (image adapted from: Nasmyth & Haering, 2009). (B) During mitosis, centromeres form a recruitment hub for kinetochores, including the microtubule binding Ndc80 complex, which drive chromosome segregation during anaphase (image adapted from: Cheeseman & Desai, 2008). (C) An Aurora B gradient emanating from the inner centromere destabilizes proximal kinetochore-microtubule interactions to prevent asymmetric chromosome segregation (image adapted from: Lampson & Cheeseman, 2011).

Epigenetics, Centromeres, Quantitative Biology

Finally, centromeres have an integral role in monitoring proper kinetochore-microtubule interactions. The formation of amphitelic attachments, where sister centromeres are attached to microtubules of opposing spindle poles, guarantees that chromosomes are pulled in opposite directions during anaphase (Cimini et al, 2001). The spindle assembly checkpoint (SAC), aka the mitotic checkpoint, is recruited to centromeres at the onset of mitosis (Chen et al, 1996; Li & Benezra, 1996) and monitors the attachment status of centromeres (Sacristan & Kops, 2015). Attachment of microtubules to the kinetochore allows for the active removal of SAC proteins from the centromere (Waters et al, 1998). However, kinetochore- microtubule interactions are destabilized by the Aurora B kinase (Figure 1.2C)., localized in between the sister centromeres, in a distance dependent manner often called the Aurora B gradient (Pinsky et al, 2006; Liu et al, 2009). Only upon formation of amphitelic attachments are kinetochores sufficiently distant from Aurora B to allow for stable microtubule attachments. The SAC is silenced once amphitely has been accomplished on all chromosomes, leading to the activation of APC/C, an E3 ubiquitin ligase that marks target proteins for destruction (Hardwick & Shah, 2010). Important targets include Cyclin B (Amon et al, 1994; Irniger et al, 1995; King et al, 1995; Sudakin et al, 1995), which activates the mitotic master regulator Cdk1, and securin (Zur & Brandeis, 2001), which inhibits separase from cleaving cohesin. Thus cells are inhibited from exiting mitosis until proper amphitelic attachments are made on all chromosomes and accurate chromosome segregation is ensured. In summary, centromeres play a key role in the regulation of mitotic progression. Centromeres are responsible for maintenance of sister chromatid cohesion, recruitment of the microtubule binding kinetochore complex, and monitoring proper kinetochore-microtubule attachments. Together, the concerted action of these processes allows for dividing cells to accurately segregate their chromosomes to the two nascent daughters.

Chapter 1

Specification of centromere identity

Centromeric DNA Because centromeres are chromosomal loci, the simplest possible mechanism to specify them is by a particular nucleotide sequence. Indeed, in the budding yeast S. cerevisiae, centromeric sequences consist of three elements, called CDEI, CDEII, and CDEIII (for centromeric DNA element 1– 3). CDEI (8 bp) and CDEIII (25 bp) are both highly conserved between the sixteen S. cerevisiae centromeres, and CDEII (~80–85 bp), although not well conserved, systematically has an AT-richness of >90% (Hieter et al, 1985; Niedenthal et al, 1991; Hegemann & Fleig, 1993). Mutations in any of these elements can cause a dramatic increase in chromosome loss, indicative of failure to form functional centromeres (Gaudet & Fitzgerald-Hayes, 1989; McGrew et al, 1989; Niedenthal et al, 1991; Hegemann & Fleig, 1993; Meluh & Koshland, 1995), with the most severe effects in CDEIII, where specific single point mutations can completely abolish centromere function (McGrew et al, 1986). Conversely, a naïve 125 bp sequence encompassing the three centromere elements is sufficient to operate as a functional centromere (Cottarel et al, 1989). In summary, specific DNA sequences are both sufficient and required for centromere function in budding yeast. Based on the budding yeast model system, it was originally thought that centromeres in other species would also be critically dependent on specific DNA sequences or motifs (Willard, 1990). However, unlike budding yeast, centromeres in most other species contain highly repetitive tandem repeat sequences, making them muchly much much more difficult to study. In fission yeast, for example, centromeres consist of a small complex (i.e. non- repetitive) central core (~4–7 kbp) flanked by ~40–100 kbp of repeat sequences (Fishel et al, 1988; Chikashige et al, 1989), while centromeric DNA of Drosophila is characterized by 5 bp repeats, interspersed with transposable elements (Sun et al, 1997). Human centromeres are formed by megabase-sized stretches of so-called alpha-satellite DNA, which consists of

Epigenetics, Centromeres, Quantitative Biology imperfect repeats of a 171 bp AT-rich sequence (Manuelidis & Wu, 1978; Manuelidis, 1978; Willard & Waye, 1987). Surprisingly, conservation of centromeric sequences is quite poor, even between closely related species (Haaf & Willard, 1997; Csink & Henikoff, 1998; Malik & Henikoff, 2002; Lee et al, 2005, 2011). In addition, it has been observed in multiple lineages that the position of centromeres along the chromosomes can change independently of the surrounding sequences or structural rearrangements (Montefalcone et al, 1999; Rocchi et al, 2012). Interestingly, as was first described in the long bug Protenor belfragei (Schrader, 1935), centromeres are not necessarily restricted to any one locus, but can instead be diffusely spread along the length of the chromosome in what is called a holocentric arrangement. C. elegans is probably the best known example (Albertson & Thomson, 1982), but holocentricity has been observed in many species and has evolved multiple independent times in both animals and plants (Melters et al, 2012). Given all these observations, centromeres are considered among the fastest evolving chromosomal regions in eukaryotes (Henikoff et al, 2001), which conflicts with the hypothesis that centromere identity is driven by a specific sequence context. Positive evidence against DNA sequences being essential for human centromere specification came with the discovery of centromeres on atypical loci. So-called neocentromeres were first identified in 1993 on a stably segregating fragment of chromosome 10 that lacked typical α-satellite or other centromeric sequences (Voullaire et al, 1993). Although centromere repositioning appears to be a rare event, over 130 unique human neocentromeres, spanning all chromosomes except 22, have been found to date (Marshall et al, 2008; Liehr, 2014). In the majority of cases analyzed, virtually all cells (within one lineage) contained the same neocentromere, arguing in favor of stable inheritance of the neocentric locus through mitotic divisions (Marshall et al, 2008). Moreover, at least seven independent neocentromeres have been described, which are inherited through human generations (Wandall et al, 1998; Tyler-Smith et al, 1999; Knegt et al, 2003;

Chapter 1

Amor et al, 2004; Ventura et al, 2004; Capozzi et al, 2009; Hasson et al, 2011), arguing that they are stable in meiosis as well. Importantly, large arrays of α-satellite sequences that did not display any centromeric function can be retained neocentric chromosomes, including meiotically stable ones (Bukvic et al, 1996; Tyler-Smith et al, 1999; Amor et al, 2004; Ventura et al, 2004; Capozzi et al, 2009; Liehr et al, 2010; Hasson et al, 2011). In summary, observations on neocentromeres argue that centromeric sequences are neither required nor sufficient for centromere specification in human cells. Although not strictly required for centromere identity, specific sequences cannot be excluded to have a function. Indeed, one well known feature of mammalian centromeric DNA is the recruitment of CENP-B, a sequence specific DNA binding protein that recognizes a 17 bp site found within a proportion of α-satellite monomers (Masumoto et al, 1989). Although CENP-B is non-essential (Hudson et al, 1998), it may play a role in organizing centromeric chromatin (Pluta et al, 1992; Hasson et al, 2011) and it has recently been suggested to contribute to centromere function (Fachinetti et al, 2013). Moreover, in an effort to create centromeres de novo on human artificial chromosomes, it was found that both α-satellite DNA and centromeric CENP-B binding sites are essential (Ohzeki et al, 2002). Another interesting observation is that a surprisingly high number of human neocentromeres have been found at regions that correlate with centromere positions in other primates (Ventura et al, 2003, 2004; Cardone et al, 2006; Capozzi et al, 2008, 2009). Moreover, it was found that orthologous loci have been used in multiple species for evolutionary centromere repositioning events that have become fixed in the population (Ventura et al, 2004). Together, these observations suggest that while specific sequences are dispensable for centromere function and maintenance, they appear to have at least some influence on de novo centromere formation.

Epigenetics, Centromeres, Quantitative Biology

CENP-A Because DNA sequences are not responsible for centromere identity, another defining factor must exist. Using auto-immune sera from human scleroderma patients, centromere protein A (CENP-A) was among the first proteins (together with CENP-B and CENP-C) to be identified at human centromeres (Earnshaw & Rothfield, 1985). Soon after its discovery, it was found that CENP-A has many histone-like properties and copurifies with core histone proteins (Palmer et al, 1987). In addition, it shares sequence homology to histone H3, which strongly suggested that CENP-A can replace this histone in centromeric nucleosomes (Palmer et al, 1987, 1991), which was confirmed by in vitro reconstitution studies some 10 years later (Yoda et al, 2000). The first piece of evidence indicating that CENP-A may be the defining feature for centromere identity came from the discovery that it is absent from inactive centromeres in dicentric chromosomes, but readily detected on neocentromeres (Earnshaw & Migeon, 1985; Warburton et al, 1997). In addition, clear centromere specific CENP-A homologues exist in nearly all species analyzed (Malik & Henikoff, 2003; Talbert et al, 2012), with the notable exception of kinetoplastids (Akiyoshi & Gull, 2013). Surprisingly, it was recently found that multiple holocentric insects appear to have lost CENP-A (Drinnenberg et al, 2014), although the presence of centromere specific H3 variants not matching their criteria was not excluded. Furthermore, loss of CENP-A is lethal and results in severe defects of chromosome segregation in all species analyzed (Stoler et al, 1995; Buchwitz et al, 1999; Henikoff et al, 2000; Howman et al, 2000; Blower & Karpen, 2001; Talbert et al, 2002; Régnier et al, 2005; Black et al, 2007b). Conversely, CENP-A is sufficient for the recruitment of virtually all known centromere and kinetochore proteins (Foltz et al, 2006; Heun et al, 2006; Liu et al, 2006; Okada et al, 2006; Carroll et al, 2009, 2010; Barnhart et al, 2011; Guse et al, 2011; Mendiburo et al, 2011), with the exception of the sequence specific DNA binding protein CENP-B (Pluta et al, 1992; Voullaire et al, 1993). Importantly, CENP-A nucleosomes are stably transmitted

Chapter 1 through both mitotic (Jansen et al, 2007) and meiotic (Palmer et al, 1990; Raychaudhuri et al, 2012; Dunleavy et al, 2012) cell divisions. Together, these observations have for many years spurred the hypothesis that CENP-A is primarily responsible for specifying centromeric identity. Despite these indications, direct evidence that CENP-A defines centromere identity was lacking until very recently. In a seminal study, Mendiburo et al (2011) used cultured Drosophila S2 cells in which they expressed a fusion protein of CENP-ACID and LacI that can be targeted to a chromosomally integrated LacO array. Using this cell line, the authors were able to show that ectopically targeted CENP-ACID is assembled into nucleosomes, recruits virtually all known Drosophila centromere and kinetochore proteins, stably binds kinetochore microtubules, and behaves as a functional centromere (Mendiburo et al, 2011). Most importantly, it was shown that a substantial pool of naïve CENP-ACID, which has no intrinsic affinity for LacO sequences, is present on the array up to 7 days after pulse- expression of targeted CENP-ACID-LacI (Mendiburo et al, 2011). More recently, it was shown that LacO-tethering of the CENP-A loading factor HJURP is not only sufficient to induce neocentromere formation, but it is also able to rescue chromosome stability and cell viability after deletion of the endogenous centromere in chicken DT40 cells (Hori et al, 2013). Intriguingly, this same study found similar results after tethering of CCAN components CENP-C or CENP-I. Thus, almost 15 years after the original suggestion by Warburton et al (1997), these beautiful experiments were finally able to provide compelling evidence that CENP-A is sufficient for the initiation of a feedback loop allowing for the stable inheritance of a centromere structure. The question that arises next is how CENP-A is able to specify a centromere. One controversial hypothesis is that it is integrated into a particle with a radically different conformation than canonical nucleosomes. Indeed, a number of different conformational models have been proposed (reviewed in Black & Cleveland, 2011), including heterotypic CENP-A/H3

Epigenetics, Centromeres, Quantitative Biology

nucleosomes (Lochmann & Ivanov, 2012), a stable (CENP-A/H4)2 tetramer lacking H2A and H2B (Williams et al, 2009) and the replacement of H2A and H2B by a non-histone protein (Mizuguchi et al, 2007). However, these models are supported by a very limited amount and oftentimes ambiguous evidence (Black & Cleveland, 2011). Nevertheless, the hemisome model, where particles are composed of a single copy of CENP-A, H4, H2A, and H2B, continually makes its way into high impact publications. The main argument used in favor of the existence of hemisomes is that CENP-A containing particles measured by atomic force microscopy (AFM) have a reduced height of approximately 50% as compared to canonical nucleosomes (Dalal et al, 2007; Dimitriadis et al, 2010; Bui et al, 2012). However, a recent study suggested that AFM measurements of in vitro reconstituted octameric CENP-A nucleosomes are in fact only half the size of their H3 counterparts (Miell et al, 2013), perhaps due to a more flexible packaging of DNA around the histone octamer (Palmer et al, 1987; Conde e Silva et al, 2007; Tachiwana et al, 2011; Hasson et al, 2013). However, these results have almost immediately been refuted by the Dalal and Henikoff labs, practically the exclusive proponents of the hemisome model, after measuring in vitro assembled octameric CENP-A nucleosomes at canonical size ranges (Codomo et al, 2014; Walkiewicz et al, 2014), and it thus remains unclear what the true height is of CENP-A nucleosomes (Miell et al, 2014). Additional observations used in favor of the existence of hemisomes comes from: 1) a nucleosome-crosslinking assay indicating the presence of a single copy of each histone (Dalal et al, 2007), although this could easily be the result of a missing crosslinkable lysine in CENP-ACID (Black & Bassett, 2008; Zhang et al, 2012) as cysteine-crosslinking readily produced CENP-ACID dimers (Zhang et al, 2012); 2) an apparent reversed directionality of DNA supercoiling around the CENP-ACse4 particle, which would be most consistent with a hemisomal conformation (Furuyama & Henikoff, 2009), although alternative, energetically more favorable explanations for the specific observations of the assay have been proposed (Black & Cleveland,

Chapter 1

2011); 3) questionable fluorescence microscopy analyses that are far from conclusive (Bui et al, 2012; Shivaraju et al, 2012); 4) high resolution ChIP- seq indicating that other DNA binding proteins surround an ~80 bp region protected by CENP-ACse4 (Krassovsky et al, 2012), although the results are equally consistent with nucleosomes protecting a ~120 bp region as would be expected for CENP-A (see below); and 5) mapping of genome wide histone H4 induced cleavage sites showing an atypical pattern on centromeric sequences (Henikoff et al, 2014). As opposed to these equivocal observations, there are many sources of compelling and highly reproducible evidence arguing in favor of canonical octameric CENP-A nucleosomes: 1) octamers are readily produced by in vitro reconstitution experiments (Yoda et al, 2000; Camahort et al, 2009; Sekulic et al, 2010; Kingston et al, 2011; Tachiwana et al, 2011), while hemisomes can only be produced under highly artificial conditions (Furuyama et al, 2013); 2) CENP-A readily homodimerizes in vitro through a dimerization domain analogous to that of H3, mutation of which blocks in vitro dimerization and in vivo targeting of CENP-A to centromeres (Palmer et al, 1991; Yoda et al, 2000; Black et al, 2004; Camahort et al, 2009; Bassett et al, 2012; Zhang et al, 2012); 3) CENP-A particles protect ~120-150 bp of DNA from micrococcal nuclease digestion, inconsistent with subnucleosomal sized particles (Palmer et al, 1987; Conde e Silva et al, 2007; Kingston et al, 2011; Zhang et al, 2012; Hasson et al, 2013); 4) when purified from cells, particles consistently contain two copies of CENP-A and stoichiometric levels of H4, H2A, and H2B, and display similar biochemical properties as canonical nucleosomes (Palmer et al, 1987; Shelby et al, 1997; Yoda et al, 2000; Foltz et al, 2006; Camahort et al, 2009; Zhang et al, 2012; Padeganeh et al, 2013; Lacoste et al, 2014); 5) co-immunoprecipitation of differentially tagged CENP-A shows that mononucleosomal particles contain both species of this protein (Shelby et al, 1997; Camahort et al, 2009; Zhang et al, 2012); and, most compellingly, 6) X-ray crystal structures of CENP-A nucleosomes (Tachiwana et al, 2011) and subnucleosomal CENP-A/H4-containing

Epigenetics, Centromeres, Quantitative Biology particles (Sekulic et al, 2010; Cho & Harrison, 2011) show canonical nucleosome conformations (albeit with subtle differences). Thus, although there is still no absolute consensus in the field, the sum of existing evidence strongly disfavors that centromeres are specified by CENP-A through an alternative nucleosome arrangement. So it goes. Assuming that CENP-A is part of a canonical nucleosome structure, another differentiating principle from H3 nucleosomes is required. A reasonable hypothesis is that there is an intrinsic feature of the CENP-A histone itself that defines its unique properties. While the HFD of CENP-A shares over 60% sequence identity (and ~75% similarity) with histone H3, a very low level of homology exists between the N-terminal histone tails of these two histones (Palmer et al, 1991; Sullivan et al, 1994). Surprisingly, however, using chimeric proteins of H3 and CENP-A, it was shown that the HFD rather than the tail of CENP-A is responsible for its centromere targeting (Sullivan et al, 1994). Some 10 years later, Black et al (2004) showed that the centromere targeting capacity lies within a region termed CATD (for CENP-A targeting domain), consisting of loop1 and α2-helix of the HFD (residues 75-114, containing 22 differences from H3.1). Consistently, the CATD was shown to be responsible for recognition of CENP-A by its specific histone chaperone and assembly factor HJUPR (Foltz et al, 2009; Shuaib et al, 2010). In addition, the CATD was demonstrated to confer reduced conformational rigidity to (CENP-A/H4)2 tetramers (Black et al, 2004) as well as CENP-A nucleosomes (Black et al, 2007a), albeit by distinct residues from those that are responsible for HJURP binding (Bassett et al, 2012). Mutation of yet another portion of the CATD, a 2 amino acid protruding bulge within loop 1, has been shown to reduce the stability of CENP-A (Tachiwana et al, 2011). However, not the CATD, but a C- terminal LEEGLG motif of CENP-A, absent from H3, is responsible for the recruitment of the majority of downstream centromere and kinetochore proteins (Carroll et al, 2010; Guse et al, 2011; Fachinetti et al, 2013), although contradictory evidence suggests that the CENP-N binding capacity

Chapter 1 is either conferred by the CATD (Carroll et al, 2009) or by LEEGLG (Guse et al, 2011). Remarkably, it was recently shown that a clean genetic replacement of CENP-A with H3CATD is insufficient to rescue human cells, but requires the addition of either the LEEGLG motif, or, surprisingly, the CENP-A tail to the chimera (Fachinetti et al, 2013). Together, these results strongly argue that multiple motifs or regions within CENP-A are cooperatively responsible for its different centromere defining properties that discriminate it from H3.

A model system for epigenetic inheritance As discussed in the first section of the introduction, epigenetic traits are heritable features that are not solely driven by underlying nucleotide sequences. In the case of centromeres, with the sole exception of S. cerevisiae and some closely related species, specific DNA sequences are neither necessary nor sufficient for centromere identity. Nevertheless, (neo-) centromeric loci are stably inherited throughout many divisions and even over multiple human generations. Thus, it is clear that centromeres are not only epigenetically defined, by that they are an example of transgenerational epigenetic inheritance. In addition, I discussed the basic properties of inheritance systems at the very beginning of this thesis: propagation, replication, and regulation of a carrier of information. It is now evident that the defining feature of centromeres is the presence of CENP-A nucleosomes (Mendiburo et al, 2011), as has been hypothesized for many years (Warburton et al, 1997). Only few other examples exist where it is as clear what the heritable defining mark is, although perhaps gene silencing through DNA methylation is another. Centromeric CENP-A is stably and quantitatively propagated through both mitotic and meiotic divisions (Jansen et al, 2007; Dunleavy et al, 2012; Raychaudhuri et al, 2012), with the only detectable loss of existing molecules occurring through dilution during DNA replication (Jansen et al, 2007; Dunleavy et al, 2011; Bodor et al, 2013). A CENP-A specific histone

Epigenetics, Centromeres, Quantitative Biology chaperone, HJURP, is responsible for replenishing CENP-A in each cell cycle (Dunleavy et al, 2009; Foltz et al, 2009; Shuaib et al, 2010; Barnhart et al, 2011), and is recruited to centromeres through a group of mutually interacting proteins called Mis18α, Mis18β, and M18BP1 (Fujita et al, 2007; Maddox et al, 2007; Barnhart et al, 2011; Wang et al, 2014). Additional roles, potentially for stabilizing CENP-A nucleosomes after their assembly, have been proposed for proteins of the RSF chromatin remodeling complex (Perpelescu et al, 2009) and a molecular GTPase switch, regulated by MgcRacGAP, Ect2, and Cdc42 (Lagana et al, 2010). Furthermore, assembly of nascent CENP-A at centromeres is strictly coupled to the exit of mitosis in animal cells (Jansen et al, 2007; Schuh et al, 2007; Hemmerich et al, 2008; Bernad et al, 2011; Silva et al, 2012), and regulated through the core machinery driving the cell cycle (Silva et al, 2012; McKinley & Cheeseman, 2014; Müller et al, 2014; Wang et al, 2014). Thus, all the basic properties of an inheritance system discussed in the beginning of this introduction (propagation, replication, regulation) evidently apply to CENP-A, underlining its role in centromere inheritance. Intriguingly, there is some indirect evidence that centromeres can play a role in (karyotype) evolution. Unlike most well-studied epigenetic traits, (neo-) centromeric loci can be transgenerationally inherited. In agreement with this, it was shown that presence of parental CENP-ACID is essential in Drosophila to initiate centromere functionality in embryos of the next generation (Raychaudhuri et al, 2012). Moreover, it appears that neocentromeres can rapidly become fixed in a population. Evolutionary new centromeres, where centromere positions have an independent evolutionary history from flanking chromosomal regions, have been reported for many mammals (including primates), birds, and plants (Montefalcone et al, 1999; Kasai et al, 2003; Nagaki et al, 2004; Ventura et al, 2004; Rocchi et al, 2012). Remarkably, five separate centromere repositioning events took place between zebra (Equus burchelli) and donkey (Equus asinus), which diverged from each other less than one million years ago, i.e. within a very short

Chapter 1 window of evolutionary time (Carbone et al, 2006). Moreover, multiple donkey chromosomes exist where the typical centromeric satellite DNA is present at a genomic locus that is distinct from the active centromere, which is formed on complex DNA (Piras et al, 2010), arguing that centromere repositioning was the result of neocentromere formation. Similarly, non- repetitive centromeres have been found on specific chromosomes in multiple other equine species (Carbone et al, 2006; Wade et al, 2009; Piras et al, 2010), chicken (Shang et al, 2010), and orangutan (Locke et al, 2011). In light of this, it has been argued that neocentromere formation and centromere repositioning are one and the same phenomenon, observed at different timescales (Capozzi et al, 2008). Moreover, it has been argued that neocentromere formation may have the capacity to drive, or at least potentiate, karyotype evolution through a non-Mendelian mechanism called meiotic drive: biased chromosome segregation to polar bodies during female meiosis (Henikoff et al, 2001; Amor et al, 2004). Indeed, a bias in the retention rate of Robertsonian (telomere-to-telomere) fusion chromosomes has been observed in humans (Pardo-Manuel de Villena & Sapienza, 2001a) and multiple other mammalian species (Pardo-Manuel de Villena & Sapienza, 2001b). Recently, in a groundbreaking study, it was demonstrated that meiotic drive in mice can act through differential centromere ‘strength,’ as measured by the density of the microtubule binding Ndc80-complex member HEC1 (Chmátal et al, 2014). Moreover, the authors were able to show that in several wild mouse populations, a reduced karyotype (from 2n = 40 to 2n = 22–28) was correlated with stronger centromeres on metacentric (internal centromere) Robertsonian fusion chromosomes than on chromosomes with a typical telocentric (centromere next to telomere) arrangement (Chmátal et al, 2014). Thus, a similar mechanism may act on neocentromeres, where an altered strength would lead to their preferential maintenance and, ultimately, fixation in a population. Taken together, irrespective of their hypothetical role in evolution, the evidence listed above likely makes centromeres the most stable epigenetic trait known to date.

Epigenetics, Centromeres, Quantitative Biology

Furthermore, there are practical reasons that facilitate the study of centromeres. Importantly, they are essential cellular structures for which there are clear and easily measurable functional readouts of failure: mitotic defects. Furthermore, microscopy analysis and quantification are greatly facilitated by their distinct localization pattern as subnuclear, resolution limited foci (Bodor et al, 2012). Finally, as a result of over 120 years of centromere research and almost three decades of studying CENP-A, a wealth of knowledge as well as molecular tools have become readily available to the scientific community. In summary, inherent as well as practical aspects of centromeres and CENP-A make them an excellent model system for the study of epigenetic inheritance. However, as Johan Cruijff famously said: “elk voordeel hep se nadeel” (“every advantage ‘as ‘is disadvantage”). Indeed, the study of centromeres does not come without its frustrations. Notably, the highly repetitive nature of centromeres put them among the last regions in the genomes of most species for which the sequence remains elusive (Alkan et al, 2011). This forms a great obstacle for certain types of analysis of centromeres, such as chromatin immunoprecipitation (ChIP) experiments or determining the elusive role of centromeric transcription. However, promising advances in sequencing technologies and data-analysis are starting to allow for the characterization of highly repetitive genomic regions, including centromeres (Alkan et al, 2011; Hayden, 2012; Hayden & Willard, 2012; Hayden et al, 2013; Altemose et al, 2014; Miga et al, 2014). Another difficulty is that centromeres are remarkably resistant to depletion of CENP-A (Liu et al, 2006; Black et al, 2007b; Fachinetti et al, 2013), which is likely due to the extreme stability of CENP-A nucleosomes (Jansen et al, 2007; Bodor et al, 2013). Nevertheless, despite these shortcomings, over the last few decades centromere biology has become an exciting and dynamic field of study.

Chapter 1

QUANTITATIVE BIOLOGY

Biology is not an exact science. Unlike e.g. physical and chemical processes, biological mechanisms cannot be fully captured in mathematical formulas. Similarly, measurements in biological systems suffer from a fairly large degree of biological variation. Although it is arguable that, once all of the underlying physical and chemical processes are fully understood, it is in principle possible to precisely measure biological system and express them in mathematical terms, this is practically impossible. Despite this, a wealth of knowledge can be acquired from quantification in biological research, and oftentimes surprising findings are made (e.g. Meyer-Rochow & Gal, 2003).

Why quantify biology anyway? In order to gain a proper understanding of a (biological) process, it is important to consider a number of features of the system. First, it is necessary to know the key players participating in the system. For this reason, much of biological research has been focused on finding genes and proteins that are involved in a particular process, oftentimes by performing forward or reverse genetic or proteomic screens. Next, it is important to know what each component does and how they interact with and depend on each other. A large number of techniques are used in biology to determine this, e.g. biochemical assays, in vitro reconstitutions, genetic hierarchy analysis, etc., etc. Finally, it is essential to quantify the (relative) amount of each of the components. However, in biological research, this parameter is often overlooked. Accordingly, relatively few techniques exist that allow for the accurate measurement of molecular copy numbers. Nevertheless, all of the aspects raised above are essential to fully understand what is going on. If taken to an extreme, it becomes obvious that the number of molecules is an essential parameter in the regulation of a process. In a hypothetical scenario, a given function can either be performed by a single molecule or collectively by a very large number of (identical) molecules. While even a small perturbation has a dramatic effect on a process that is dependent on a

Epigenetics, Centromeres, Quantitative Biology single molecule (e.g. if this molecule is lost or damaged), only major deviations will affect a system driven by a large population of molecules. Similarly, for structures that can exist in multiple different states (e.g. active or inactive), a system consisting of a single entity will always either be fully active or fully inactive. Conversely, the law of large numbers dictates that the higher the number of units, the closer the system will be to equilibrium at any moment (Bernoulli, 1713). Although these examples take an extreme standpoint, they do have at least some biological relevance. Indeed, there is evidence in the pathogenic yeast C. albicans that the dam1 complex, which plays a role in stabilizing kinetochore-microtubule interactions and is essential in this species, becomes redundant if the single endogenous kinetochore-microtubule is experimentally increased to more than one (Burrack et al, 2011). Thus, information about the statistical and stochastic properties of a process is obtained by determining where the number of components lies on the scale of one to infinity. Even when far removed from extreme values, knowledge of the number of molecules provides information about the physical properties of a system. Naturally, four oxen can pull a heavier load than two. Similarly, biological entities will be able to, e.g., exert or resist more force, adhere more strongly, or react to a stimulus faster, depending on the number of physical modules. A clear example is that of dynein, a molecular motor that is able to utilize energy obtained from ATP hydrolysis to transport cargo along the surface of a microtubule (Goldstein & Yang, 2000). The amount of force that each dynein molecule can generate has been carefully measured to be in the pN range (Kamimura & Takahashi, 1981; Ashkin et al, 1990). Given the high cytoplasmic viscosity as well as the large volume and mass of certain pieces of cargo, this amount of force may not suffice and in many cases multiple dynein motors act simultaneously on the same piece of cargo (Ashkin et al, 1990). Therefore, the amount of force that each motor can exert, the amount of force required to drag cargo, as well as the number of motors present are all essential factors to understand the mechanics of subcellular transport.

Chapter 1

How to measure absolute copy numbers in biological systems? The total amount of a given protein per cell can be measured using a number of different techniques. A fairly straightforward strategy is the quantitative comparison of purified (recombinant) protein of a known concentration with lytic extracts of a known number of cells, e.g. by Western blot (Higgs et al, 1999) or mass spectrometry (Gerber et al, 2003; Beynon et al, 2005). Recent advances in fluorescent immunoblotting have aided accurate quantification by increasing the linear range of detection (Schutz- Geschwender et al, 2004; Wang et al, 2007). An alternative strategy is to immobilize fluorescently tagged proteins from extracts on functionalized glass surfaces after which single molecule imaging can be used to determine the number of molecules (Jiang et al, 2010). While these methods allow for the determination of the average number of molecules in a population of cells, information of the variance, and thus of the actual number of molecules in any cell, is lost. Thus, to avoid averaging over a large population, single cell techniques have been developed, e.g. by using microfluidic chambers (Huang et al, 2007). However, these whole cell quantification methods usually don’t give information about the number of molecules that actually take part in any single structure or event.

Figure 1.3 (next page) Methods that allow for the determination molecular copy numbers. (A) Fluorescence correlation spectroscopy measures the autocorrelation of fluorescence intensity over time within a minute volume. Stars represent fluorophores; arrows represent movement over (discrete) time steps; red stars and line segments represent fluorescently active molecules. (Sample data and conversion function were adapted from: Weidemann & Schwille, 2009) (B) Stepwise photobleaching is used to measure discreet steps in fluorescence decay until background intensity is reached (image adapted from: Leake et al, 2006). (C) Superresolution microscopy can be used to count individual fluorescence activation events (image adapted from: Gunzenhäuser et al, 2012). (D) Fluorescent standards are used as a reference of comparison to signal intensities of a structure of interest. In this case, Cse4-GFP intensity is compared to 4 different molecular standards. Images of purified GFP molecules are averaged over 8 frames and acquired at 2.5 fold higher exposure times. Graph shows the average fluorescence intensity per GFP molecule (in grey) for the different standards as well as Cse4-GFP count (in red) based on these particular standards (image adapted from: Lawrimore et al, 2011). *: note that the maximum possible number of LacI-GFP molecules on a 4 kb Lac array is indicated and used as fluorescent standard. (E) Internally calibrated ratiometry determines the relative fluorescence of a structure of interest compared to the fluorescence of the entire cell. In combination with measurements of the total amount of proteins present in the cell (in this case by comparative western blot against purified protein, right panel), this allows for copy number measurements that are independent of external references (image adapted from: Bodor et al, 2014).

Epigenetics, Centromeres, Quantitative Biology

Chapter 1

A number of additional difficulties arise when molecule copy numbers are interrogated at subcellular locations. Analysis usually relies on imaging- based methods, for which cells must retain a certain level of integrity – ideally live cells are used— rather than using protein extracts. In addition, every single copy of the molecule of interest must be accounted for, which ideally requires the genetic replacement of an endogenous protein with a (fluorescently) tagged version. Alternatively, in specific cases electron microscopy can be used (e.g. Ashkin et al, 1990), although complex and intrusive preparation techniques are required, which may likely affect the number of molecules detected. Below, a number of strategies to determine local molecular copy numbers using fluorescence microscopy are discussed. Fluorescence correlation spectroscopy (FCS) is a well-known method to determine local protein concentrations (Figure 1.3A). This technique measures the fluctuation of fluorescence from molecules that pass through a sub-femtoliter volume, i.e. ~5 orders of magnitude smaller than a eukaryotic cell (Schwille, 2001). In effect, this allows for the determination of fluorophore copy numbers within the excitation volume (Koppel et al, 1976), and can be repeated in different cellular regions to determine the distribution throughout the cell (Heinrich et al, 2013). One shortcoming of FCS is that is relies on Brownian motion of fluorophores and therefore is not applicable if the proteins are relatively immobile and/or stably bound to large structures. Stepwise photobleaching is a method that relies on the stochastic irreversible bleaching of individual fluorophores due to light exposure (Leake et al, 2006). By continuously exciting samples at a low intensity, fluorophores will bleach at a low frequency such that it is possible to determine the number of events that occurred before background levels are reached (Figure 1.3B). However, it becomes progressively more difficult to separate individual bleaching events with increasing number of fluorophores (Ulbrich & Isacoff, 2007). Thus, it has been estimated that the maximum number of molecules that can be accurately counted by stepwise photobleaching, even after mathematical extrapolations, lies around 30 (Coffman & Wu, 2012).

Epigenetics, Centromeres, Quantitative Biology

More recently, the principle of super-resolution microscopy has been applied to determine molecule copy numbers (Gunzenhäuser et al, 2012; Lando et al, 2012). In this case, somewhat opposite to stepwise photobleaching, activation events of photoconvertible fluorescent proteins are counted (Figure 1.3C). Individual fluorophores are successively activated, counted, and bleached prior to activation of a subsequent fluorophore at the same site. Gunzenhäuser et al (2012) were able to convincingly show that, by combining usage of optimal photo-convertible fluorescent proteins with specific buffer and imaging conditions and sophisticated analysis techniques, accurate counts of up to ~1000 molecules can be produced. However, due to the complex nature of the experimental techniques, microscope setups, and image analysis, it will likely take some time before this strategy will become common practice in the scientific community. The use of fluorescent standards is a fairly straightforward way to measure fluorophore copy numbers. Structures containing a known number of fluorophores, either determined by independent methods or synthesized to contain a calibrated number of molecules, called fluorescent standards, are imaged alongside with a fluorescent structure of interest. If imaged under identical conditions, their relative fluorescence is a direct readout of the ratio of fluorescent molecules between the two structures (Coffman & Wu, 2012). However, the fluorescence properties of most fluorophores are affected by their local environment (Suhling et al, 2002), most notably by the pH (Campbell & Choy, 2001; Griesbeck et al, 2001; Suhling et al, 2002). Similarly, maturation dynamics of fluorescent proteins (i.e. the time between protein production and emergence of their fluorescent potential) have been shown to depend on external conditions, such as temperature (Macdonald et al, 2012) or growth media supplements (Hebisch et al, 2013). Because both the environment and its effect on the fluorophore are hard to determine in vivo, a potential effect on ratiometric measurements of fluorescence intensities cannot be excluded. Nevertheless, recently, four

Chapter 1 highly diverse fluorescent standards (purified EGFP; virus like particles; bacterial flagellar motor proteins; and a calibrated LacO/LacI-system) were used to measure the number of centromeric CENP-ACse4 molecules in budding yeast (Figure 1.3D), all of which essentially producing the same result (Lawrimore et al, 2011). This indicates that environmental factors may not significantly affect these measurements, at least for the particular fluorescent protein (EGFP) used. Internally calibrated fluorescence comparisons (Figure 1.3E) provide perhaps the most elegant method to determine local protein abundance. In this case, fluorescence measurements are made both of the total cellular volume and of the specific region of interest. Combining the ratio of fluorescence between these two with a measurement of the total protein concentration (e.g. by western blot as described above), gives a direct readout of local protein copy numbers (Wu & Pollard, 2005; Wu et al, 2008). Although performing all required corrections and determining complex cell shapes is not trivial, the main advantage of this method is that it is fully internally controlled. It goes without saying that this inventory of potential methods to quantify molecular copy numbers is far from complete. Nevertheless, for most techniques, there are relatively few examples in the literature where they have been used, likely due to their complex nature. Importantly, given that each method has its own pitfalls and shortcomings, ideally a combination of strategies should be used to gain confidence in the measurements.

How many CENP-A molecules are there in a centromere? One specific case for which it is important to know the number of molecules present is centromeric CENP-A. Because CENP-A chromatin constitutes an epigenetic mark, an essential molecular unit of information that cannot be lost, the fidelity of centromere propagation is ultimately dependent on the number of nucleosomes present. For this (and other)

Epigenetics, Centromeres, Quantitative Biology reasons, many attempts have been made to measure the centromeric abundance of CENP-A in a variety of species. Consistent with the difficulty of performing copy number measurements, as described above, discrepancies between measurements performed in different studies exist in many cases. Below, I will give an overview of all the different measurements performed to my knowledge to date (see also Table 1.1) and discuss potential reasons for disagreement. Budding yeast was the first species where careful analysis of the number of CENP-ACse4 nucleosomes per centromere was performed. Given that centromeric DNA is non-repetitive in this species, ChIP experiments showed a strong enrichment of CENP-ACse4 for the ~125 base pair centromere core, although some binding of neighboring sequences was also detected (Meluh et al, 1998). A very elegant follow-up experiment by Furuyama and Biggins (2007) showed that ChIPped fragments of mononucleosomal size contain centromeric DNA, but not the surrounding sequences, strongly suggesting that budding yeast centromeres harbor a single CENP-ACse4 nucleosome. Given this apparently clean biochemical evidence, centromeric foci of this species (containing 16 clustered centromeres and 2 CENP-ACse4 molecules per nucleosome) have been extensively used as a molecular standard for 32 fluorescent molecules (Joglekar et al, 2006, 2008; Johnston et al, 2010; Schittenhelm et al, 2010). However, this may not have been the most reliable choice, as the one-nucleosome hypothesis has been challenged recently by two microscopy-based studies that used external fluorescent standards to determine that budding yeast centromeres contain on average 3.5–8 CENP-ACse4 molecules (Coffman et al, 2011; Lawrimore et al, 2011). Furthermore, it was shown that the amount of CENP-ACse4 can be reduced by ~40–60%, without affecting kinetochore-microtubule attachments (Haase et al, 2013), inconsistent with a single nucleosome per centromere. Mathematical simulations argue that due to the relatively high detection limit, CENP-ACse4 outside of centromeres would not be observed in the ChIP experiments of Furuyama & Biggins (2007) if their nucleosome positions are

Chapter 1 sufficiently variable (Lawrimore et al, 2011). Nevertheless, yet other recent analyses maintain that budding yeast centromeres contain a single nucleosome, based on high sensitivity ChIP-Seq experiments (Henikoff & Henikoff, 2012), FCS measurements (Shivaraju et al, 2012), stepwise photobleaching of the kinetochore protein Spc24 at a single centromere (Aravamudhan et al, 2013), or fluorescence comparison to TetR on an intergrated TetO array of carefully determined size (Wisniewski et al, 2014). Potential explanations for the discrepancy between the different studies can be sought in the use of different strains (as argued by Lawrimore et al, 2011); potential pre-nucleosomal or unincorporated CENP-ACse4 at centromeres (as argued by Henikoff & Henikoff, 2012), potential artifacts induced by fluorescent tags (as argued by Henikoff & Henikoff, 2012 and Wisniewski et al, 2014), and/or complex dynamics of photochemical maturation times of fluorescent proteins at the budding yeast centromere (Wisniewski et al, 2014), perhaps in combination with measurement inaccuracy of often very dim signals. Nevertheless, although the verdict is still out on the precise number of CENP-ACse4 molecules per centromere, a general agreement exists that few (≤4) nucleosomes are present (Table 1.1). The amount of CENP-A per centromere was analyzed in two other yeast species. CENP-ACse4 was used as a fluorescent standard to measure the amount of CENP-A at C. albicans and fission yeast centromeres. After correction for the number of centromeres per cluster in each species, CENP-ACaCse4 was found to be ~4 times as abundant at C. albicans centromeres as CENP-ACse4 in budding yeast (Joglekar et al, 2008). Therefore, depending on the true number in budding yeast, C. albicans has between 8 and 32 molecules of CENP-ACaCse4 (4–16 nucleosomes) per centromere. For fission yeast, the authors found that CENP-ACnp1 is ~2.5 times as abundant at the centromeres of this species as in budding yeast (Joglekar et al, 2008), arguing that there are on average 5–20 molecules per centromere (2.5–10 nucleosomes). However, it was recently found that the fission yeast strain used for these comparisons likely expressed competing

Epigenetics, Centromeres, Quantitative Biology wildtype CENP-ACnp1 in addition to the measured CENP-ACnp1-GFP (Coffman et al, 2011), which would confound the measurements. To reevaluate the measured numbers, stepwise photobleaching was performed to calibrate a different fluorescent standard, the bacterial flagellar motor protein MotB (Leake et al, 2006; Coffman et al, 2011), and was used to show that 226 molecules of CENP-ACnp1 are present per centromere in a clean genetic substitution strain of fission yeast (Coffman et al, 2011). More recently, a super-resolution-based method was employed to count the amount of CENP-ACnp1 at centromeres and found ~20 molecules per centromere (Lando et al, 2012). In addition, using high-resolution ChIP-Seq, the same study showed that, in total, the central domains of all three fission yeast centromeres only displayed 64 discrete peaks of CENP-ACnp1 (Lando et al, 2012). The authors used this result to argue that no more than 128 molecules of CENP-ACnp1 can be present at centromere foci (~43 per centromere), although it must be noted that a substantial number of peaks in the outer repeats of fission yeast centromeres were ignored. At present, given the rather large discrepancies observed between studies, it is difficult to make a final conclusion as to what the correct number of CENP-ACnp1 molecules per fission yeast centromere is (Table 1.1). Only few efforts have been reported to measure the amount of CENP-A at metazoan centromeres. One study used, yet again, CENP-ACse4 as a fluorescent standard to measure CENP-ACID levels in Drosophila wing imaginal discs (Schittenhelm et al, 2010). According to their measurements, 84–336 molecules of CENP-ACID are present per centromere, depending on how many there are in budding yeast. It must be noted however, that there are multiple experimental issues that may confound the results in this study. Importantly, rather than quantifying centromere specific signals, measurements were made on regions that are larger than an entire nucleus and thus including fluorescence derived from non-centromeric CENP-A, the levels of which can be surprisingly high and even exceed the centromeric levels (Bodor et al, 2014; Lacoste et al, 2014). In addition, it remains

Chapter 1 unexplained why a large variation in the extent of fluorescence reduction at increasing focal depth is observed for different proteins. In C. elegans, a holocentric organisms (Albertson & Thomson, 1982), ChIP experiments show that CENP-AHCP-3 can be found on ~40–60% of the genome (Gassmann et al, 2012). However, the total chromatin-bound pool is only sufficient to represent 3–4% of all nucleosomes (Gassmann et al, 2012), arguing that the precise location of CENP-A is highly variable between individuals. Two studies have reported numbers on the amount of CENP-A in chicken DT40 cells, although both caution that the presence of untagged CENP-A likely confounds their copy number measurements. Johnston et al (2010) rounded up the usual suspect, CENP-ACse4, as a fluorescent standard and report that there are at least 62 molecules of CENP-A per centromere (based on a single CENP-ACse4 nucleosome). Ribeiro et al (2010) count photoblinking events of a photoconvertible fluorescent protein to estimate that there are between 25 and 40 molecules of CENP-A-Dronpa present (in addition to the unlabeled CENP-A), although they admit that their measurements are further hampered by the fact that the photoblinking properties of this probe are quite variable (Habuchi et al, 2005; Flors et al, 2007). Prior to my own work, no careful quantification has been made for human CENP-A on a per centromere basis. In fact, to my knowledge, the only reported estimation comes from a study where the total cellular pool of CENP-A, measured at 2×106 molecules per HeLa cell, was divided over the average number of chromosomes present in this cell line and states that the maximum amount of CENP-A per centromere is ~30.000 (Black et al, 2007b). As discussed extensively in Chapter 4, I have now carefully measured the centromeric CENP-A copy number in human RPE cells to be on average 400, although minor differences exist between specific cell lines (Bodor et al, 2014; and see Table 1.1).

Epigenetics, Centromeres, Quantitative Biology

Table 1.1 Overview of published number of CENP-A molecules per centromere for different species

all centromeresall were toable

Xrn

and and

Pat

ChIP

reduction in reduction

Despite a

HeLa (263); GM06170 (336); primary fibroblasts U2OS (570); PDNC-4 (579)

- Cell lines measured (molecules/CEN):- Cell lines DLD-1 (100); HCT-116 (177);

- Described in chapter 4 chapter - Described in

- Maximum would be estimationcentromere CENP-A if all cellular localized

- Correct number depends on the amount present in budding yeast on- Correct the amountbudding depends number in present

- Presence CENP-A is not accounted forof untagged

- Number of photoblinking events of - Number molecule photoblinking per are erratic

- Presence CENP-A is not accounted forof untagged

- Unconventional corrections were performed

pool

- Measurement of centromere total rather than CENP-A chromatin bound specific

- Correct number depends on the amount present in budding yeast on- Correct the amountbudding depends number in present

procedures

- Data was not the experimentshown and the wasexperimental in not described

- A substantial number outerof number peaks in ChIP-Seq repeats were - A substantial ignored

- MotB copy by was stepwise number calibrated photobleaching

expressed in addition to protein in addition expressed GFP tagged

- Measurements may that by is potentially be confounded non-fluorescent CENP-A

- Correct number depends on the amount present in budding yeast on- Correct the amountbudding depends number in present

- TetO-arrays of sizes were different used

determined in Joglekar et Joglekar determined in al (2006)

- BiFC argues a minimum argues of- BiFC 2 molecules Spc24:CENP-A ratio CEN. per was

make connectionsstable microtubule

heterogeneous

- Nup49 is a suboptimal standard, as the signal is highly dispersed and and dispersed is highly as the signal is a suboptimal - Nup49 standard,

measurements, were but not taken into account

- Changes in centromere in - Changes morphology mitosis throughout would affect FCS

hemisomal conformation canonical and

- The authors argue nucleosome,that there - The argue is a single authors that it but cycles between a

- Data in a way is to presented that is hard evaluate their argumentation

- Fluorescent standards: mGFP, like MotB, - Fluorescent particles, standards: Virus LacI

- Different amounts onused the strain werespecific observed depending

- MotB copy by was stepwise number calibrated photobleaching

detected after CENP-A

- CEN DNA, but not surrounding sequences, sequences, of mononucleosomal not DNA, but surrounding - CEN were size

Notes

(Nup49)

Method

CENP-A)

mutants

ChIP-Seq

ChIP-PCR

(multiple)

BiFC + stepwiseBiFC

(bacterial MotB)(bacterial

PALM / ChIP-Seq

(budding yeast Cse4) (budding

Photoblinking eventsPhotoblinking

Fluorescent standard Fluorescent standard

Fluorescent standards Fluorescent standards

(budding yeast Ndc80) (budding

Reduction ofReduction CENP-A in

3 independent methods 3 independent

photobleaching of Spc25 photobleaching

Fluorescent standard (RPE Fluorescent standard

FCS + fluorescent standard + fluorescent FCS standard

Whole cell immunoblotting

Fluorescent standard (TetR) Fluorescent standard

248

579

≤

000

336

104

–

(1)

≥

227

2.25

~ 400

≤

100

~20 /

≥

molecules/CEN

(nucleosomes/CEN)

Reference

Yao et al (2013)

Bodor et al (2014)

Lando et al (2012)

Haase etHaase al (2013)

Black Black et al (2007b)

Ribeiro etRibeiro al (2010)

Joglekar et Joglekar al (2008)

Coffman et al (2011)

Johnston etJohnston al (2010)

Shivaraju et al (2012)Shivaraju

Lawrimore et al (2011)

Wisniewski et al (2014)

Schittenhelm et Schittenhelm al (2010)

Aravamudhan et al (2013)Aravamudhan

Furuyama & Biggins (2007) & Furuyama Biggins

Henikoff & Henikoff (2012)

CID

Cse4

Cnp1

CENP-A / CENP-A

ggCENP-A

Cse4 / CaCse4

Spcies specificSpcies

name of CENP-A of name

)

human human

Species

C. albicans

(S. pombe)

fission yeastfission

S. cerevisiae

human (RPE)human

budding yeast budding

human (HeLa) human

(

chicken (DT40)chicken

D. melanogaster

(multiple cell (multiple lines)

(wing imaginal disc) imaginal (wing : table corresponds The shownto in this numbers either molecules centromereper or nucleosomes centromere,per to the relevant corresponding in what reference was described a

Chapter 1

REFERENCES

Agrawal AA, Laforsch C & Tollrian R (1999) Transgenerational induction of defences in animals and plants. Nature 401: 60–63 Ahmad K & Henikoff S (2002) The histone variant H3.3 marks active chromatin by replication-independent nucleosome assembly. Mol. Cell 9: 1191–1200 Akiyoshi B & Gull K (2013) Evolutionary cell biology of chromosome segregation: insights from trypanosomes. Open Biol. 3: 130023 Alabert C & Groth A (2012) Chromatin replication and epigenome maintenance. Nat. Rev. Mol. Cell Biol. 13: 153–167 Albertson DG & Thomson JN (1982) The kinetochores of Caenorhabditis elegans. Chromosoma 86: 409–428 Alkan C, Cardone MF, Catacchio CR, Antonacci F, O’Brien SJ, Ryder OA, Purgato S, Zoli M, Valle GD, Eichler EE & Ventura M (2011) Genome-wide characterization of centromeric satellites from multiple mammalian genomes. Genome Res. 21: 137–145 Altemose N, Miga KH, Maggioni M & Willard HF (2014) Genomic characterization of large heterochromatic gaps in the human genome assembly. PLoS Comput. Biol. 10: e1003628 Amon A, Irniger S & Nasmyth K (1994) Closing the cell cycle circle in yeast: G2 cyclin proteolysis initiated at mitosis persists until the activation of G1 cyclins in the next cycle. Cell 77: 1037–1050 Amor DJ, Bentley K, Ryan J, Perry J, Wong L, Slater H & Choo KHA (2004) Human centromere repositioning ‘in progress’. Proc. Natl. Acad. Sci. U. S. A. 101: 6542– 6547 Aragon L, Martinez-Perez E & Merkenschlager M (2013) Condensin, cohesin and the control of chromatin states. Curr. Opin. Genet. Dev. 23: 204–211 Aravamudhan P, Felzer-Kim I & Joglekar AP (2013) The Budding Yeast Point Centromere Associates with Two Cse4 Molecules during Mitosis. Curr. Biol. 23: 770–774 Ashkin A, Schütze K, Dziedzic JM, Euteneuer U & Schliwa M (1990) Force generation of organelle transport measured in vivo by an infrared laser trap. Nature 348: 346–348 Avery OT, MacLeod CM & McCarty M (1944) Studies on the Chemical Nature of the Substance Inducing Transformation of Pneumococcal Types Induction of Transformation by a Desoxyribonucleic Acid Fraction Isolated from Pneumococcus Type III. J. Exp. Med. 79: 137–158 Bannister AJ & Kouzarides T (2011) Regulation of chromatin by histone modifications. Cell Res. 21: 381–395 Barnhart MC, Kuich PHJL, Stellfox ME, Ward JA, Bassett EA, Black BE & Foltz DR (2011) HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochore. J. Cell Biol. 194: 229–243

Epigenetics, Centromeres, Quantitative Biology

Barski A, Cuddapah S, Cui K, Roh T-Y, Schones DE, Wang Z, Wei G, Chepelev I & Zhao K (2007) High-resolution profiling of histone methylations in the human genome. Cell 129: 823–837 Bassett EA, DeNizio J, Barnhart-Dailey MC, Panchenko T, Sekulic N, Rogers DJ, Foltz DR & Black BE (2012) HJURP uses distinct CENP-A surfaces to recognize and to stabilize CENP-A/histone H4 for centromere assembly. Dev. Cell 22: 749– 762 Benson LJ, Gu Y, Yakovleva T, Tong K, Barrows C, Strack CL, Cook RG, Mizzen CA & Annunziato AT (2006) Modifications of H3 and H4 during chromatin replication, nucleosome assembly, and histone exchange. J. Biol. Chem. 281: 9287–9296 Berger SL, Kouzarides T, Shiekhattar R & Shilatifard A (2009) An operational definition of epigenetics. Genes Dev. 23: 781–783 Bernad R, Sánchez P, Rivera T, Rodríguez-Corsino M, Boyarchuk E, Vassias I, Ray- Gallet D, Arnaoutov A, Dasso M, Almouzni G & Losada A (2011) Xenopus HJURP and condensin II are required for CENP-A assembly. J. Cell Biol. 192: 569–582 Bernoulli J (1713) Ars conjectandi. Basileae, Switzerland: Impensis Thurnisiorum, Fratrum Bestor TH & Ingram VM (1983) Two DNA methyltransferases from murine erythroleukemia cells: purification, sequence specificity, and mode of interaction with DNA. Proc. Natl. Acad. Sci. U. S. A. 80: 5559–5563 Bettencourt-Dias M & Glover DM (2007) Centrosome biogenesis and function: centrosomics brings new understanding. Nat. Rev. Mol. Cell Biol. 8: 451–463 Beynon RJ, Doherty MK, Pratt JM & Gaskell SJ (2005) Multiplexed absolute quantification in proteomics using artificial QCAT proteins of concatenated signature peptides. Nat. Methods 2: 587–589 Bird A (2002) DNA methylation patterns and epigenetic memory. Genes Dev. 16: 6–21 Bird A (2007) Perceptions of epigenetics. Nature 447: 396–398 Bird AP (1978) Use of restriction enzymes to study eukaryotic DNA methylation: II. The symmetry of methylated sites supports semi-conservative copying of the methylation pattern. J. Mol. Biol. 118: 49–60 Black BE & Bassett EA (2008) The histone variant CENP-A and centromere specification. Curr. Opin. Cell Biol. 20: 91–100 Black BE, Brock MA, Bédard S, Woods VL & Cleveland DW (2007a) An epigenetic mark generated by the incorporation of CENP-A into centromeric nucleosomes. Proc. Natl. Acad. Sci. U. S. A. 104: 5008–5013 Black BE & Cleveland DW (2011) Epigenetic Centromere Propagation and the Nature of CENP-A Nucleosomes. Cell 144: 471–479 Black BE, Foltz DR, Chakravarthy S, Luger K, Woods VL & Cleveland DW (2004) Structural determinants for generating centromeric chromatin. Nature 430: 578– 582

Chapter 1

Black BE, Jansen LET, Maddox PS, Foltz DR, Desai AB, Shah JV & Cleveland DW (2007b) Centromere identity maintained by nucleosomes assembled with histone H3 containing the CENP-A targeting domain. Mol. Cell 25: 309–322 Blower MD & Karpen GH (2001) The role of Drosophila CID in kinetochore formation, cell-cycle progression and heterochromatin interactions. Nat. Cell Biol. 3: 730–739 Bodor DL, Mata JF, Sergeev M, David AF, Salimian KJ, Panchenko T, Cleveland DW, Black BE, Shah JV & Jansen LET (2014) The quantitative architecture of centromeric chromatin. eLife 3: e02137 Bodor DL, Rodríguez MG, Moreno N & Jansen LET (2012) Analysis of Protein Turnover by Quantitative SNAP-Based Pulse-Chase Imaging. Curr. Protoc. Cell Biol. Chapter 8: Unit8.8 Bodor DL, Valente LP, Mata JF, Black BE & Jansen LET (2013) Assembly in G1 phase and long-term stability are unique intrinsic features of CENP-A nucleosomes. Mol. Biol. Cell 24: 923–932 Boffelli D & Martin DIK (2012) Epigenetic Inheritance: A Contributor to Species Differentiation? DNA Cell Biol. 31: S–11 Brinkley BR & Cartwright J (1975) Cold-Labile and Cold-Stable Microtubules in the Mitotic Spindle of Mammalian Cells*. Ann. N. Y. Acad. Sci. 253: 428–439 Buchwitz BJ, Ahmad K, Moore LL, Roth MB & Henikoff S (1999) Cell division: A histone-H3-like protein in C. elegans. Nature 401: 547–548 Bui M, Dimitriadis EK, Hoischen C, An E, Quénet D, Giebe S, Nita-Lazar A, Diekmann S & Dalal Y (2012) Cell-cycle-dependent structural transitions in the human CENP-A nucleosome in vivo. Cell 150: 317–326 Bukvic N, Susca F, Gentile M, Tangari E, Ianniruberto A & Guanti G (1996) An unusual dicentric Y chromosome with a functional centromere with no detectable alpha-satellite. Hum. Genet. 97: 453–456 Burrack LS, Applen SE & Berman J (2011) The Requirement for the Dam1 Complex Is Dependent upon the Number of Kinetochore Proteins and Microtubules. Curr. Biol. 21: 889–896 Camahort R, Shivaraju M, Mattingly M, Li B, Nakanishi S, Zhu D, Shilatifard A, Workman JL & Gerton JL (2009) Cse4 is part of an octameric nucleosome in budding yeast. Mol. Cell 35: 794–805 Campbell TN & Choy FY (2001) The effect of pH on green fluorescent protein: a brief review. Mol. Biol. Today 2: 1–4 Capozzi O, Purgato S, D’Addabbo P, Archidiacono N, Battaglia P, Baroncini A, Capucci A, Stanyon R, Della Valle G & Rocchi M (2009) Evolutionary descent of a human chromosome 6 neocentromere: a jump back to 17 million years ago. Genome Res. 19: 778–784 Capozzi O, Purgato S, Verdun di Cantogno L, Grosso E, Ciccone R, Zuffardi O, Della Valle G & Rocchi M (2008) Evolutionary and clinical neocentromeres: two faces of the same coin? Chromosoma 117: 339–344

Epigenetics, Centromeres, Quantitative Biology

Capuano F, Mülleder M, Kok R, Blom HJ & Ralser M (2014) Cytosine DNA Methylation Is Found in Drosophila melanogaster but Absent in Saccharomyces cerevisiae, Schizosaccharomyces pombe, and Other Yeast Species. Anal. Chem. 86: 3697–3702 Carbone L, Nergadze SG, Magnani E, Misceo D, Francesca Cardone M, Roberto R, Bertoni L, Attolini C, et al (2006) Evolutionary movement of centromeres in horse, donkey, and zebra. Genomics 87: 777–782 Cardone MF, Alonso A, Pazienza M, Ventura M, Montemurro G, Carbone L, Jong PJ de, Stanyon R, et al (2006) Independent centromere formation in a capricious, gene-free domain of chromosome 13q21 in Old World monkeys and pigs. Genome Biol. 7: R91 Carroll CW, Milks KJ & Straight AF (2010) Dual recognition of CENP-A nucleosomes is required for centromere assembly. J. Cell Biol. 189: 1143–1155 Carroll CW, Silva MCC, Godek KM, Jansen LET & Straight AF (2009) Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N. Nat. Cell Biol. 11: 896–902 Cavalier-Smith T (2004) The Membranome and Membrane Heredity in Development and Evolution. In Organelles, Genomes and Eukaryote Phylogeny, Hirt RP & Horner DS (eds) pp 335–352. Boca Raton, FL: CRC Press Chakravarthy S, Park Y-J, Chodaparambil J, Edayathumangalam RS & Luger K (2005) Structure and dynamic properties of nucleosome core particles. FEBS Lett. 579: 895–898 Cheeseman IM, Chappie JS, Wilson-Kubalek EM & Desai A (2006) The conserved KMN network constitutes the core microtubule-binding site of the kinetochore. Cell 127: 983–997 Cheeseman IM & Desai A (2008) Molecular architecture of the kinetochore- microtubule interface. Nat. Rev. Mol. Cell Biol. 9: 33–46 Cheeseman IM, Niessen S, Anderson S, Hyndman F, Yates JR, Oegema K & Desai A (2004) A conserved protein network controls assembly of the outer kinetochore and its ability to sustain tension. Genes Dev. 18: 2255–2268 Chen R-H, Waters JC, Salmon ED & Murray AW (1996) Association of Spindle Assembly Checkpoint Component XMAD2 with Unattached Kinetochores. Science 274: 242–246 Chikashige Y, Kinoshita N, Nakaseko Y, Matsumoto T, Murakami S, Niwa O & Yanagida M (1989) Composite motifs and repeat symmetry in S. pombe centromeres: Direct analysis by integration of Notl restriction sites. Cell 57: 739– 751 Chmátal L, Gabriel SI, Mitsainas GP, Martínez-Vargas J, Ventura J, Searle JB, Schultz RM & Lampson MA (2014) Centromere Strength Provides the Cell Biological Basis for Meiotic Drive and Karyotype Evolution in Mice. Curr. Biol. 24: 2295–2300 Cho U-S & Harrison SC (2011) Recognition of the centromere-specific histone Cse4 by the chaperone Scm3. Proc. Natl. Acad. Sci. U. S. A. 108: 9367–9371

Chapter 1

Chuang LS, Ian HI, Koh TW, Ng HH, Xu G & Li BF (1997) Human DNA-(cytosine-5) methyltransferase-PCNA complex as a target for p21WAF1. Science 277: 1996– 2000 Cimini D, Howell B, Maddox P, Khodjakov A, Degrassi F & Salmon ED (2001) Merotelic Kinetochore Orientation Is a Major Mechanism of Aneuploidy in Mitotic Mammalian Tissue Cells. J. Cell Biol. 153: 517–528 Codomo CA, Furuyama T & Henikoff S (2014) CENP-A octamers do not confer a reduction in nucleosome height by AFM. Nat. Struct. Mol. Biol. 21: 4–5 Coffman VC & Wu J-Q (2012) Counting protein molecules using quantitative fluorescence microscopy. Trends Biochem. Sci. 37: 499–506 Coffman VC, Wu P, Parthun MR & Wu J-Q (2011) CENP-A exceeds microtubule attachment sites in centromere clusters of both budding and fission yeast. J. Cell Biol. 195: 563–572 Colot V & Rossignol J-L (1999) Eukaryotic DNA methylation as an evolutionary device. BioEssays 21: 402–411 Conde e Silva N, Black BE, Sivolob A, Filipski J, Cleveland DW & Prunell A (2007) CENP-A-containing nucleosomes: easier disassembly versus exclusive centromeric localization. J. Mol. Biol. 370: 555–573 Cottarel G, Shero JH, Hieter P & Hegemann JH (1989) A 125-base-pair CEN6 DNA fragment is sufficient for complete meiotic and mitotic centromere functions in Saccharomyces cerevisiae. Mol. Cell. Biol. 9: 3342–3349 Csink AK & Henikoff S (1998) Something from nothing: the evolution and utility of satellite repeats. Trends Genet. TIG 14: 200–204 Cubas P, Vincent C & Coen E (1999) An epigenetic mutation responsible for natural variation in floral symmetry. Nature 401: 157–161 Dalal Y, Wang H, Lindsay S & Henikoff S (2007) Tetrameric structure of centromeric nucleosomes in interphase Drosophila cells. PLoS Biol. 5: e218 Darwin C (1859) On the Origin of Species by Means of Natural Selection Or: The Preservation of Favoured Races in the Struggle for Life. 1st ed. London: J. Murray Day T & Bonduriansky R (2011) A unified approach to the evolutionary consequences of genetic and nongenetic inheritance. Am. Nat. 178: E18–36 Deichmann U (2004) Early responses to Avery et al.’s paper on DNA as hereditary material. Hist. Stud. Phys. Biol. Sci. 34: 207–232 DeLuca JG, Gall WE, Ciferri C, Cimini D, Musacchio A & Salmon ED (2006) Kinetochore microtubule dynamics and attachment stability are regulated by Hec1. Cell 127: 969–982 Dimitriadis EK, Weber C, Gill RK, Diekmann S & Dalal Y (2010) Tetrameric organization of vertebrate centromeric nucleosomes. Proc. Natl. Acad. Sci. U. S. A. 107: 20317–20322

Epigenetics, Centromeres, Quantitative Biology

Drané P, Ouararhni K, Depaux A, Shuaib M & Hamiche A (2010) The death- associated protein DAXX is a novel histone chaperone involved in the replication- independent deposition of H3.3. Genes Dev. 24: 1253–1265 Drinnenberg IA, deYoung D, Henikoff S & Malik HS (2014) Recurrent loss of CenH3 is associated with independent transitions to holocentricity in insects. eLife 3: e03676 Dunleavy EM, Almouzni G & Karpen GH (2011) H3.3 is deposited at centromeres in S phase as a placeholder for newly assembled CENP-A in G(1) phase. Nucl. Austin Tex 2: 146–157 Dunleavy EM, Beier NL, Gorgescu W, Tang J, Costes SV & Karpen GH (2012) The Cell Cycle Timing of Centromeric Chromatin Assembly in Drosophila Meiosis Is Distinct from Mitosis Yet Requires CAL1 and CENP-C. PLoS Biol 10: e1001460 Dunleavy EM, Roche D, Tagami H, Lacoste N, Ray-Gallet D, Nakamura Y, Daigo Y, Nakatani Y & Almouzni-Pettinotti G (2009) HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell 137: 485–497 Earnshaw WC & Migeon BR (1985) Three related centromere proteins are absent from the inactive centromere of a stable isodicentric chromosome. Chromosoma 92: 290–296 Earnshaw WC & Rothfield N (1985) Identification of a family of human centromere proteins using autoimmune sera from patients with scleroderma. Chromosoma 91: 313–321 Ehrlich M, Gama-Sosa MA, Huang L-H, Midgett RM, Kuo KC, McCune RA & Gehrke C (1982) Amount and distribution of 5-methylcytosine in human DNA from different types of tissues or cells. Nucleic Acids Res. 10: 2709–2721 Erikson EH (1950) Growth and crises of the ‘healthy personality.’ In Symposium on the healthy personality pp 91–146. Oxford, England: Josiah Macy, Jr. Foundation Fachinetti D, Diego Folco H, Nechemia-Arbely Y, Valente LP, Nguyen K, Wong AJ, Zhu Q, Holland AJ, Desai A, Jansen LET & Cleveland DW (2013) A two-step mechanism for epigenetic specification of centromere identity and function. Nat. Cell Biol. 15: 1056–1066 Fishel B, Amstutz H, Baum M, Carbon J & Clarke L (1988) Structural organization and functional analysis of centromeric DNA in the fission yeast Schizosaccharomyces pombe. Mol. Cell. Biol. 8: 754–763 Flaus A & Owen-Hughes T (2004) Mechanisms for ATP-dependent chromatin remodelling: farewell to the tuna-can octamer? Curr. Opin. Genet. Dev. 14: 165– 173 Flemming W (1880) Beiträge zur Kenntniss der Zelle und ihrer Lebenserscheinungen. Theil II. Arch. Für Mikrosk. Anat. 18: 151–259 Flicek P, Amode MR, Barrell D, Beal K, Billis K, Brent S, Carvalho-Silva D, Clapham P, et al (2014) Ensembl 2014. Nucleic Acids Res. 42: D749–D755 Flors C, Hotta J, Uji-i H, Dedecker P, Ando R, Mizuno H, Miyawaki A & Hofkens J (2007) A stroboscopic approach for fast photoactivation-localization microscopy with Dronpa mutants. J. Am. Chem. Soc. 129: 13970–13977

Chapter 1

Foltz DR, Jansen LET, Bailey AO, Yates JR, Bassett EA, Wood S, Black BE & Cleveland DW (2009) Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell 137: 472–484 Foltz DR, Jansen LET, Black BE, Bailey AO, Yates JR & Cleveland DW (2006) The human CENP-A centromeric nucleosome-associated complex. Nat. Cell Biol. 8: 458–469 Ford-Jones EL & Kellner JD (1995) ‘Cheap torches’: an acronym for congenital and perinatal infections. Pediatr. Infect. Dis. J. 14: 638–639 Fujita Y, Hayashi T, Kiyomitsu T, Toyoda Y, Kokubu A, Obuse C & Yanagida M (2007) Priming of Centromere for CENP-A Recruitment by Human hMis18α, hMis18β, and M18BP1. Dev. Cell 12: 17–30 Furuyama S & Biggins S (2007) Centromere identity is specified by a single centromeric nucleosome in budding yeast. Proc. Natl. Acad. Sci. U. S. A. 104: 14706–14711 Furuyama T, Codomo CA & Henikoff S (2013) Reconstitution of hemisomes on budding yeast centromeric DNA. Nucleic Acids Res. Furuyama T & Henikoff S (2009) Centromeric nucleosomes induce positive DNA supercoils. Cell 138: 104–113 Gaspar-Maia A, Qadeer ZA, Hasson D, Ratnakumar K, Adrian Leu N, Leroy G, Liu S, Costanzi C, et al (2013) MacroH2A histone variants act as a barrier upon reprogramming towards pluripotency. Nat. Commun. 4: 1565 Gasser SM (2002) Visualizing chromatin dynamics in interphase nuclei. Science 296: 1412–1416 Gassmann R, Rechtsteiner A, Yuen KW, Muroyama A, Egelhofer T, Gaydos L, Barron F, Maddox P, et al (2012) An inverse relationship to germline transcription defines centromeric chromatin in C. elegans. Nature 484: 534–537 Gaudet A & Fitzgerald-Hayes M (1989) Mutations in CEN3 cause aberrant chromosome segregation during meiosis in Saccharomyces cerevisiae. Genetics 121: 477–489 Geoghegan JL & Spencer HG (2012) Population-epigenetic models of selection. Theor. Popul. Biol. 81: 232–242 Gerber SA, Rush J, Stemman O, Kirschner MW & Gygi SP (2003) Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proc. Natl. Acad. Sci. 100: 6940–6945 Goldberg AD, Banaszynski LA, Noh K-M, Lewis PW, Elsaesser SJ, Stadler S, Dewell S, Law M, et al (2010) Distinct factors control histone variant H3.3 localization at specific genomic regions. Cell 140: 678–691 Goldstein LSB & Yang Z (2000) Microtubule-Based Transport Systems in Neurons: The Roles of Kinesins and Dyneins. Annu. Rev. Neurosci. 23: 39–71 Gottschling DE (2004) Summary: Epigenetics—from Phenomenon to Field. Cold Spring Harb. Symp. Quant. Biol. 69: 507–520

Epigenetics, Centromeres, Quantitative Biology

Goyal R, Reinhardt R & Jeltsch A (2006) Accuracy of DNA methylation pattern preservation by the Dnmt1 methyltransferase. Nucleic Acids Res. 34: 1182–1188 Griesbeck O, Baird GS, Campbell RE, Zacharias DA & Tsien RY (2001) Reducing the environmental sensitivity of yellow fluorescent protein. Mechanism and applications. J. Biol. Chem. 276: 29188–29194 Griffith F (1928) The Significance of Pneumococcal Types. J. Hyg. (Lond.) 27: 113– 159 Groth A, Rocha W, Verreault A & Almouzni G (2007) Chromatin challenges during DNA replication and repair. Cell 128: 721–733 Gunzenhäuser J, Olivier N, Pengo T & Manley S (2012) Quantitative Super- Resolution Imaging Reveals Protein Stoichiometry and Nanoscale Morphology of Assembling HIV-Gag Virions. Nano Lett. 12: 4705–4710 Guse A, Carroll CW, Moree B, Fuller CJ & Straight AF (2011) In vitro centromere and kinetochore assembly on defined chromatin templates. Nature 477: 354–358 Haaf T & Willard HF (1997) Chromosome-specific α-satellite DNA from the centromere of chimpanzee chromosome 4. Chromosoma 106: 226–232 Haarhuis JHI, Elbatsh AMO & Rowland BD (2014) Cohesin and Its Regulation: On the Logic of X-Shaped Chromosomes. Dev. Cell 31: 7–18 Haase J, Mishra PK, Stephens A, Haggerty R, Quammen C, Taylor RM 2nd, Yeh E, Basrai MA & Bloom K (2013) A 3D map of the yeast kinetochore reveals the presence of core and accessory centromere-specific histone. Curr. Biol. CB 23: 1939–1944 Habuchi S, Ando R, Dedecker P, Verheijen W, Mizuno H, Miyawaki A & Hofkens J (2005) Reversible single-molecule photoswitching in the GFP-like fluorescent protein Dronpa. Proc. Natl. Acad. Sci. U. S. A. 102: 9511–9516 Hake SB, Garcia BA, Duncan EM, Kauer M, Dellaire G, Shabanowitz J, Bazett-Jones DP, Allis CD & Hunt DF (2006) Expression Patterns and Post-translational Modifications Associated with Mammalian Histone H3 Variants. J. Biol. Chem. 281: 559–568 Halfmann R, Alberti S & Lindquist S (2010) Prions, protein homeostasis, and phenotypic diversity. Trends Cell Biol. 20: 125–133 Halfmann R, Jarosz DF, Jones SK, Chang A, Lancaster AK & Lindquist S (2012) Prions are a common mechanism for phenotypic inheritance in wild yeasts. Nature 482: 363–368 Hardwick KG & Shah JV (2010) Spindle checkpoint silencing: ensuring rapid and concerted anaphase onset. F1000 Biol. Rep. 2: 55 Hasson D, Alonso A, Cheung F, Tepperberg JH, Papenhausen PR, Engelen JJM & Warburton PE (2011) Formation of novel CENP-A domains on tandem repetitive DNA and across chromosome breakpoints on human chromosome 8q21 neocentromeres. Chromosoma 120: 621–632 Hasson D, Panchenko T, Salimian KJ, Salman MU, Sekulic N, Alonso A, Warburton PE & Black BE (2013) The octamer is the major form of CENP-A nucleosomes at human centromeres. Nat. Struct. Mol. Biol. 20: 687–695

Chapter 1

Hayden KE (2012) Human centromere genomics: now it’s personal. Chromosome Res. Int. J. Mol. Supramol. Evol. Asp. Chromosome Biol. 20: 621–633 Hayden KE, Strome ED, Merrett SL, Lee H-R, Rudd MK & Willard HF (2013) Sequences Associated with Centromere Competency in the Human Genome. Mol. Cell. Biol. 33: 763–772 Hayden KE & Willard HF (2012) Composition and organization of active centromere sequences in complex genomes. BMC Genomics 13: 324 Hebisch E, Knebel J, Landsberg J, Frey E & Leisner M (2013) High Variation of Fluorescence Protein Maturation Times in Closely Related Escherichia coli Strains. PLoS ONE 8: e75991 Hegemann JH & Fleig UN (1993) The centromere of budding yeast. BioEssays 15: 451–460 Heinrich S, Geissen E-M, Kamenz J, Trautmann S, Widmer C, Drewe P, Knop M, Radde N, Hasenauer J & Hauf S (2013) Determinants of robustness in spindle assembly checkpoint signalling. Nat. Cell Biol. 15: 1328–1339 Hemmerich P, Weidtkamp-Peters S, Hoischen C, Schmiedeberg L, Erliandri I & Diekmann S (2008) Dynamics of inner kinetochore assembly and maintenance in living cells. J. Cell Biol. 180: 1101–1114 Henikoff S, Ahmad K & Malik HS (2001) The Centromere Paradox: Stable Inheritance with Rapidly Evolving DNA. Science 293: 1098–1102 Henikoff S, Ahmad K, Platero JS & Steensel B van (2000) Heterochromatic deposition of centromeric histone H3-like proteins. Proc. Natl. Acad. Sci. 97: 716–721 Henikoff S & Henikoff JG (2012) ‘Point’ Centromeres of Saccharomyces Harbor Single Centromere-Specific Nucleosomes. Genetics 190: 1575–1577 Henikoff S, Ramachandran S, Krassovsky K, Bryson TD, Codomo CA, Brogaard K, Widom J, Wang J-P & Henikoff JG (2014) The budding yeast Centromere DNA Element II wraps a stable Cse4 hemisome in either orientation in vivo. eLife 3: e01861 Hershey AD & Chase M (1952) Independent Functions of Viral Protein and Nucleic Acid in Growth of Bacteriophage. J. Gen. Physiol. 36: 39–56 Heun P, Erhardt S, Blower MD, Weiss S, Skora AD & Karpen GH (2006) Mislocalization of the Drosophila centromere-specific histone CID promotes formation of functional ectopic kinetochores. Dev. Cell 10: 303–315 He Y-F, Li B-Z, Li Z, Liu P, Wang Y, Tang Q, Ding J, Jia Y, et al (2011) Tet-Mediated Formation of 5-Carboxylcytosine and Its Excision by TDG in Mammalian DNA. Science 333: 1303–1307 Hieter P, Pridmore D, Hegemann JH, Thomas M, Davis RW & Philippsen P (1985) Functional selection and analysis of yeast centromeric DNA. Cell 42: 913–921 Higgs HN, Blanchoin L & Pollard TD (1999) Influence of the C Terminus of Wiskott- Aldrich Syndrome Protein (WASp) and the Arp2/3 Complex on Actin Polymerization†. Biochemistry (Mosc.) 38: 15212–15222

Epigenetics, Centromeres, Quantitative Biology

Hobert O (2011) Maintaining a memory by transcriptional autoregulation. Curr. Biol. 21: R146–R147 Holmberg J & Perlmann T (2012) Maintaining differentiated cellular identity. Nat. Rev. Genet. 13: 429–439 Van Hooser A, Goodrich DW, Allis CD, Brinkley BR & Mancini MA (1998) Histone H3 phosphorylation is required for the initiation, but not maintenance, of mammalian chromosome condensation. J. Cell Sci. 111 ( Pt 23): 3497–3506 Hori T, Shang W-H, Takeuchi K & Fukagawa T (2013) The CCAN recruits CENP-A to the centromere and forms the structural core for kinetochore assembly. J. Cell Biol. 200: 45–60 Howman EV, Fowler KJ, Newson AJ, Redward S, MacDonald AC, Kalitsis P & Choo KHA (2000) Early disruption of centromeric chromatin organization in centromere protein A (Cenpa) null mice. Proc. Natl. Acad. Sci. 97: 1148–1153 Huang B, Wu H, Bhaya D, Grossman A, Granier S, Kobilka BK & Zare RN (2007) Counting Low-Copy Number Proteins in a Single Cell. Science 315: 81–84 Hudson DF, Fowler KJ, Earle E, Saffery R, Kalitsis P, Trowell H, Hill J, Wreford NG, et al (1998) Centromere Protein B Null Mice are Mitotically and Meiotically Normal but Have Lower Body and Testis Weights. J. Cell Biol. 141: 309–319 Hunt CR, Ramnarain D, Horikoshi N, Iyengar P, Pandita RK, Shay JW & Pandita TK (2013) Histone modifications and DNA double-strand break repair after exposure to ionizing radiations. Radiat. Res. 179: 383–392 Ichiyanagi T, Ichiyanagi K, Miyake M & Sasaki H (2013) Accumulation and loss of asymmetric non-CpG methylation during male germ-cell development. Nucleic Acids Res. 41: 738–745 Inoué S & Salmon ED (1995) Force Generation by Microtubule Assembly/Disassembly in Mitosis and Related Movements. Mol. Biol. Cell 6: 1619–1640 Irniger S, Piatti S, Michaelis C & Nasmyth K (1995) Genes involved in sister chromatid separation are needed for b-type cyclin proteolysis in budding yeast. Cell 81: 269–277 Ito S, Shen L, Dai Q, Wu SC, Collins LB, Swenberg JA, He C & Zhang Y (2011) Tet Proteins Can Convert 5-Methylcytosine to 5-Formylcytosine and 5- Carboxylcytosine. Science 333: 1300–1303 Izuta H, Ikeno M, Suzuki N, Tomonaga T, Nozaki N, Obuse C, Kisu Y, Goshima N, Nomura F, Nomura N & Yoda K (2006) Comprehensive analysis of the ICEN (Interphase Centromere Complex) components enriched in the CENP-A chromatin of human cells. Genes Cells 11: 673–684 Jablonka E (2012) Epigenetic variations in heredity and evolution. Clin. Pharmacol. Ther. 92: 683–688 Jablonka E & Lamb MJ (2002) The Changing Concept of Epigenetics. Ann. N. Y. Acad. Sci. 981: 82–96

Chapter 1

Jablonka E & Lamb MJ (2005) Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life. Cambridge, MA: MIT Press Jablonka E & Lamb MJ (2006) The evolution of information in the major transitions. J. Theor. Biol. 239: 236–246 Jablonka E & Raz G (2009) Transgenerational Epigenetic Inheritance: Prevalence, Mechanisms, and Implications for the Study of Heredity and Evolution. Q. Rev. Biol. 84: 131–176 Jackson V (1987) Deposition of newly synthesized histones: new histones H2A and H2B do not deposit in the same nucleosome with new histones H3 and H4. Biochemistry (Mosc.) 26: 2315–2325 Jähner D, Stuhlmann H, Stewart CL, Harbers K, Löhler J, Simon I & Jaenisch R (1982) De novo methylation and expression of retroviral genomes during mouse embryogenesis. Nature 298: 623–628 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805 Jeltsch A (2002) Beyond Watson and Crick: DNA Methylation and Molecular Enzymology of DNA Methyltransferases. ChemBioChem 3: 274–293 Jenuwein T & Allis CD (2001) Translating the histone code. Science 293: 1074– 1080 Jiang D, Liu C, Wang L & Jiang W (2010) Fluorescence single-molecule counting assays for protein quantification using epi-fluorescence microscopy with quantum dots labeling. Anal. Chim. Acta 662: 170–176 Joglekar AP, Bouck DC, Molk JN, Bloom KS & Salmon ED (2006) Molecular architecture of a kinetochore-microtubule attachment site. Nat. Cell Biol. 8: 581– 585 Joglekar AP, Bouck D, Finley K, Liu X, Wan Y, Berman J, He X, Salmon ED & Bloom KS (2008) Molecular architecture of the kinetochore-microtubule attachment site is conserved between point and regional centromeres. J. Cell Biol. 181: 587–594 Johannes F, Porcher E, Teixeira FK, Saliba-Colombani V, Simon M, Agier N, Bulski A, Albuisson J, et al (2009) Assessing the Impact of Transgenerational Epigenetic Variation on Complex Traits. PLoS Genet 5: e1000530 Johnston K, Joglekar A, Hori T, Suzuki A, Fukagawa T & Salmon ED (2010) Vertebrate kinetochore protein architecture: protein copy number. J. Cell Biol. 189: 937–943 Kamimura S & Takahashi K (1981) Direct measurement of the force of microtubule sliding in flagella. Nature 293: 566–568 Kasai F, Garcia C, Arruga MV & Ferguson-Smith MA (2003) Chromosome homology between chicken (Gallus gallus domesticus) and the red-legged partridge (Alectoris rufa); evidence of the occurrence of a neocentromere during evolution. Cytogenet. Genome Res. 102: 326–330

Epigenetics, Centromeres, Quantitative Biology

Kerrebrock AW, Moore DP, Wu JS & Orr-Weaver TL (1995) Mei-S332, a drosophila protein required for sister-chromatid cohesion, can localize to meiotic centromere regions. Cell 83: 247–256 Khare SP, Habib F, Sharma R, Gadewal N, Gupta S & Galande S (2011) HIstome—a relational knowledgebase of human histone proteins and histone modifying enzymes. Nucleic Acids Res.: gkr1125 Kimura H & Cook PR (2001) Kinetics of Core Histones in Living Human Cells Little Exchange of H3 and H4 and Some Rapid Exchange of H2b. J. Cell Biol. 153: 1341–1354 King RW, Peters J-M, Tugendreich S, Rolfe M, Hieter P & Kirschner MW (1995) A 20s complex containing CDC27 and CDC16 catalyzes the mitosis-specific conjugation of ubiquitin to cyclin B. Cell 81: 279–288 Kingston IJ, Yung JSY & Singleton MR (2011) Biophysical Characterization of the Centromere-specific Nucleosome from Budding Yeast. J. Biol. Chem. 286: 4021– 4026 Knegt AC, Li S, Engelen JJM, Bijlsma EK & Warburton PE (2003) Prenatal diagnosis of a karyotypically normal pregnancy in a mother with a supernumerary neocentric 13q21 → 13q22 chromosome and balancing reciprocal deletion. Prenat. Diagn. 23: 215–220 Kobayakawa S, Miike K, Nakao M & Abe K (2007) Dynamic changes in the epigenomic state and nuclear organization of differentiating mouse embryonic stem cells. Genes Cells Devoted Mol. Cell. Mech. 12: 447–460 Koppel DE, Axelrod D, Schlessinger J, Elson EL & Webb WW (1976) Dynamics of fluorescence marker concentration as a probe of mobility. Biophys. J. 16: 1315– 1329 Kornberg RD (1974) Chromatin Structure: A Repeating Unit of Histones and DNA. Science 184: 868–871 Koshland D & Strunnikov A (1996) Mitotic Chromosome Condensation. Annu. Rev. Cell Dev. Biol. 12: 305–333 Krassovsky K, Henikoff JG & Henikoff S (2012) Tripartite organization of centromeric chromatin in budding yeast. Proc. Natl. Acad. Sci. 109: 243–248 Kunkel TA (2004) DNA Replication Fidelity. J. Biol. Chem. 279: 16895–16898 Lacoste N, Woolfe A, Tachiwana H, Garea AV, Barth T, Cantaloube S, Kurumizaka H, Imhof A & Almouzni G (2014) Mislocalization of the Centromeric Histone Variant CenH3/CENP-A in Human Cells Depends on the Chaperone DAXX. Mol. Cell 53: 631–644 Lagana A, Dorn JF, De Rop V, Ladouceur A-M, Maddox AS & Maddox PS (2010) A small GTPase molecular switch regulates epigenetic centromere maintenance by stabilizing newly incorporated CENP-A. Nat. Cell Biol. 12: 1186–1193 Laird CD, Pleasant ND, Clark AD, Sneeden JL, Hassan KMA, Manley NC, Vary JC, Morgan T, Hansen RS & Stöger R (2004) Hairpin-bisulfite PCR: Assessing epigenetic methylation patterns on complementary strands of individual DNA molecules. Proc. Natl. Acad. Sci. 101: 204–209

Chapter 1

Lampson MA & Cheeseman IM (2011) Sensing centromere tension: Aurora B and the regulation of kinetochore function. Trends Cell Biol. 21: 133–140 Lando D, Endesfelder U, Berger H, Subramanian L, Dunne PD, McColl J, Klenerman D, Carr AM, Sauer M, Allshire RC, Heilemann M & Laue ED (2012) Quantitative single-molecule microscopy reveals that CENP-A(Cnp1) deposition occurs during G2 in fission yeast. Open Biol. 2: 120078 Larsen A & Weintraub H (1982) An altered DNA conformation detected by S1 nuclease occurs at specific regions in active chick globin chromatin. Cell 29: 609– 622 Law JA & Jacobsen SE (2010) Establishing, maintaining and modifying DNA methylation patterns in plants and animals. Nat. Rev. Genet. 11: 204–220 Lawrimore J, Bloom KS & Salmon ED (2011) Point centromeres contain more than a single centromere-specific Cse4 (CENP-A) nucleosome. J. Cell Biol. 195: 573– 582 Leake MC, Chandler JH, Wadhams GH, Bai F, Berry RM & Armitage JP (2006) Stoichiometry and turnover in single, functioning membrane protein complexes. Nature 443: 355–358 Lee H-R, Hayden KE & Willard HF (2011) Organization and molecular evolution of CENP-A--associated satellite DNA families in a basal primate genome. Genome Biol. Evol. 3: 1136–1149 Lee H-R, Zhang W, Langdon T, Jin W, Yan H, Cheng Z & Jiang J (2005) Chromatin immunoprecipitation cloning reveals rapid evolutionary patterns of centromeric DNA in Oryza species. Proc. Natl. Acad. Sci. U. S. A. 102: 11793–11798 Lee J-S, Smith E & Shilatifard A (2010) The language of histone crosstalk. Cell 142: 682–685 Liehr T (2014) Small supernumerary marker chromosomes. Available at: http://ssmc-tl.com/sSMC.html [Accessed October 4, 2014] Liehr T, Kosyakova N, Weise A, Ziegler M & Raabe-Meyer G (2010) First case of a neocentromere formation in an otherwise normal chromosome 7. Cytogenet. Genome Res. 128: 189–191 Li E & Zhang Y (2014) DNA Methylation in Mammals. Cold Spring Harb. Perspect. Biol. 6: a019133 Liu D, Vader G, Vromans MJM, Lampson MA & Lens SMA (2009) Sensing Chromosome Bi-Orientation by Spatial Separation of Aurora B Kinase from Kinetochore Substrates. Science 323: 1350–1353 Liu S-T, Rattner JB, Jablonski SA & Yen TJ (2006) Mapping the assembly pathways that specify formation of the trilaminar kinetochore plates in human cells. J. Cell Biol. 175: 41–53 Li Y & Benezra R (1996) Identification of a Human Mitotic Checkpoint Gene: hsMAD2. Science 274: 246–248 Lochmann B & Ivanov D (2012) Histone H3 Localizes to the Centromeric DNA in Budding Yeast. PLoS Genet 8: e1002739

Epigenetics, Centromeres, Quantitative Biology

Locke DP, Hillier LW, Warren WC, Worley KC, Nazareth LV, Muzny DM, Yang S-P, Wang Z, et al (2011) Comparative and demographic analysis of orang-utan genomes. Nature 469: 529–533 Losada A, Hirano M & Hirano T (1998) Identification of Xenopus SMC protein complexes required for sister chromatid cohesion. Genes Dev. 12: 1986–1997 Losada A, Yokochi T, Kobayashi R & Hirano T (2000) Identification and Characterization of Sa/Scc3p Subunits in the Xenopus and Human Cohesin Complexes. J. Cell Biol. 150: 405–416 Luger K, Mäder AW, Richmond RK, Sargent DF & Richmond TJ (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389: 251– 260 Macdonald PJ, Chen Y & Mueller JD (2012) Chromophore maturation and fluorescence fluctuation spectroscopy of fluorescent proteins in a cell-free expression system. Anal. Biochem. 421: 291–298 Maddox PS, Hyndman F, Monen J, Oegema K & Desai A (2007) Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 176: 757–763 Maiti A & Drohat AC (2011) Thymine DNA Glycosylase Can Rapidly Excise 5- Formylcytosine and 5-Carboxylcytosine: Potential Implications for Active Demethylation of CpG Sites. J. Biol. Chem. 286: 35334–35338 Malik HS & Henikoff S (2002) Conflict begets complexity: the evolution of centromeres. Curr. Opin. Genet. Dev. 12: 711–718 Malik HS & Henikoff S (2003) Phylogenomics of the nucleosome. Nat. Struct. Mol. Biol. 10: 882–891 Manuelidis L (1978) Chromosomal localization of complex and simple repeated human DNAs. Chromosoma 66: 23–32 Manuelidis L & Wu JC (1978) Homology between human and simian repeated DNA. Nature 276: 92–94 Marris E, Pearson H, Waldrop M, Hayden EC, Schiermeier Q, Baker M, Brumfiel G & Ledford H (2008) Language: Disputed definitions. Nature 455: 1023–1028 Marshall OJ, Chueh AC, Wong LH & Choo KHA (2008) Neocentromeres: New Insights into Centromere Structure, Disease Development, and Karyotype Evolution. Am. J. Hum. Genet. 82: 261–282 Marshall WF, Vucica Y & Rosenbaum JL (2001) Kinetics and regulation of de novo centriole assembly: Implications for the mechanism of centriole duplication. Curr. Biol. 11: 308–317 Martin DIK, Singer M, Dhahbi J, Mao G, Zhang L, Schroth GP, Pachter L & Boffelli D (2011) Phyloepigenomic comparison of great apes reveals a correlation between somatic and germline methylation states. Genome Res. 21: 2049–2057 Masumoto H, Masukata H, Muro Y, Nozaki N & Okazaki T (1989) A human centromere antigen (CENP-B) interacts with a short specific sequence in alphoid DNA, a human centromeric satellite. J. Cell Biol. 109: 1963–1973

Chapter 1

Mayer W, Niveleau A, Walter J, Fundele R & Haaf T (2000) Embryogenesis: Demethylation of the zygotic paternal genome. Nature 403: 501–502 McGrew J, Diehl B & Fitzgerald-Hayes M (1986) Single base-pair mutations in centromere element III cause aberrant chromosome segregation in Saccharomyces cerevisiae. Mol. Cell. Biol. 6: 530–538 McGrew JT, Xiao Z & Fitzgerald-Hayes M (1989) Saccharomyces cerevisiae mutants defective in chromosome segregation. Yeast 5: 271–284 McKinley KL & Cheeseman IM (2014) Polo-like Kinase 1 Licenses CENP-A Deposition at Centromeres. Cell 158: 397–411 McKittrick E, Gafken PR, Ahmad K & Henikoff S (2004) Histone H3.3 is enriched in covalent modifications associated with active chromatin. Proc. Natl. Acad. Sci. U. S. A. 101: 1525–1530 Melters DP, Paliulis LV, Korf IF & Chan SWL (2012) Holocentric chromosomes: convergent evolution, meiotic adaptations, and genomic analysis. Chromosome Res. 20: 579–593 Meluh PB & Koshland D (1995) Evidence that the MIF2 gene of Saccharomyces cerevisiae encodes a centromere protein with homology to the mammalian centromere protein CENP-C. Mol. Biol. Cell 6: 793–807 Meluh PB, Yang P, Glowczewski L, Koshland D & Smith MM (1998) Cse4p Is a Component of the Core Centromere of Saccharomyces cerevisiae. Cell 94: 607– 613 Mendel G (1866) Versuche über Pﬂanzenhybriden. Verhandlungen Naturforschenden Vereines Brünn 4: 3–47 Mendiburo MJ, Padeken J, Fülöp S, Schepers A & Heun P (2011) Drosophila CENH3 is sufficient for centromere formation. Science 334: 686–690 Meselson M & Stahl FW (1958) The replication of DNA in Escherichia coli. Proc. Natl. Acad. Sci. 44: 671–682 Meshorer E & Misteli T (2006) Chromatin in pluripotent embryonic stem cells and differentiation. Nat. Rev. Mol. Cell Biol. 7: 540–546 Meyer-Rochow VB & Gal J (2003) Pressures produced when penguins pooh— calculations on avian defaecation. Polar Biol. 27: 56–58 Michaelis C, Ciosk R & Nasmyth K (1997) Cohesins: Chromosomal Proteins that Prevent Premature Separation of Sister Chromatids. Cell 91: 35–45 Miell MDD, Fuller CJ, Guse A, Barysz HM, Downes A, Owen-Hughes T, Rappsilber J, Straight AF & Allshire RC (2013) CENP-A confers a reduction in height on octameric nucleosomes. Nat. Struct. Mol. Biol. 20: 763–765 Miell MDD, Straight AF & Allshire RC (2014) Reply to ‘CENP-A octamers do not confer a reduction in nucleosome height by AFM’. Nat. Struct. Mol. Biol. 21: 5–8 Miga KH, Newton Y, Jain M, Altemose N, Willard HF & Kent WJ (2014) Centromere reference models for human chromosomes X and Y satellite arrays. Genome Res. 24: 697–707

Epigenetics, Centromeres, Quantitative Biology

Mitchison T, Evans L, Schulze E & Kirschner M (1986) Sites of microtubule assembly and disassembly in the mitotic spindle. Cell 45: 515–527 Mito Y, Henikoff JG & Henikoff S (2005) Genome-scale profiling of histone H3.3 replacement patterns. Nat. Genet. 37: 1090–1097 Mizuguchi G, Xiao H, Wisniewski J, Smith MM & Wu C (2007) Nonhistone Scm3 and histones CenH3-H4 assemble the core of centromere-specific nucleosomes. Cell 129: 1153–1164 Montefalcone G, Tempesta S, Rocchi M & Archidiacono N (1999) Centromere Repositioning. Genome Res. 9: 1184–1188 Moran NA, McCutcheon JP & Nakabachi A (2008) Genomics and evolution of heritable bacterial symbionts. Annu. Rev. Genet. 42: 165–190 Morse C (2013) Seven Ways to Fit More Stuff in Your Suitcase. SmarterTravel.com Available at: http://www.smartertravel.com/photo-galleries/editorial/seven- ways-to-fit-more-stuff-in-your-suitcase.html?id=620&source=41718 [Accessed September 30, 2014] Mousson F, Ochsenbein F & Mann C (2007) The histone chaperone Asf1 at the crossroads of chromatin and DNA checkpoint pathways. Chromosoma 116: 79– 93 Müller S, Montes de Oca R, Lacoste N, Dingli F, Loew D & Almouzni G (2014) Phosphorylation and DNA Binding of HJURP Determine Its Centromeric Recruitment and Function in CenH3CENP-A Loading. Cell Rep. 8: 190–203 Muramoto T, Müller I, Thomas G, Melvin A & Chubb JR (2010) Methylation of H3K4 Is required for inheritance of active transcriptional states. Curr. Biol. CB 20: 397–406 Nagaki K, Cheng Z, Ouyang S, Talbert PB, Kim M, Jones KM, Henikoff S, Buell CR & Jiang J (2004) Sequencing of a rice centromere uncovers active genes. Nat. Genet. 36: 138–145 Nasmyth K & Haering CH (2009) Cohesin: Its Roles and Mechanisms. Annu. Rev. Genet. 43: 525–558 Ng HH, Robert F, Young RA & Struhl K (2003) Targeted recruitment of Set1 histone methylase by elongating Pol II provides a localized mark and memory of recent transcriptional activity. Mol. Cell 11: 709–719 Ng RK & Gurdon JB (2008) Epigenetic memory of an active gene state depends on histone H3.3 incorporation into chromatin in the absence of transcription. Nat. Cell Biol. 10: 102–109 Niedenthal R, Stoll R & Hegemann JH (1991) In vivo characterization of the Saccharomyces cerevisiae centromere DNA element I, a binding site for the helix- loop-helix protein CPF1. Mol. Cell. Biol. 11: 3545–3553 Nigg EA & Stearns T (2011) The centrosome cycle: Centriole biogenesis, duplication and inherent asymmetries. Nat. Cell Biol. 13: 1154–1160 Ohzeki J, Nakano M, Okada T & Masumoto H (2002) CENP-B box is required for de novo centromere chromatin assembly on human alphoid DNA. J. Cell Biol. 159: 765–775

Chapter 1

Okada M, Cheeseman IM, Hori T, Okawa K, McLeod IX, Yates JR 3rd, Desai A & Fukagawa T (2006) The CENP-H-I complex is required for the efficient incorporation of newly synthesized CENP-A into centromeres. Nat. Cell Biol. 8: 446–457 Okano M, Bell DW, Haber DA & Li E (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99: 247–257 Okano M, Xie S & Li E (1998) Cloning and characterization of a family of novel mammalian DNA (cytosine-5) methyltransferases. Nat. Genet. 19: 219–220 Olins AL & Olins DE (1974) Spheroid Chromatin Units (ν Bodies). Science 183: 330–332 Ooi SKT & Bestor TH (2008) The colorful history of active DNA demethylation. Cell 133: 1145–1148 Oswald J, Engemann S, Lane N, Mayer W, Olek A, Fundele R, Dean W, Reik W & Walter J (2000) Active demethylation of the paternal genome in the mouse zygote. Curr. Biol. 10: 475–478 Padeganeh A, Ryan J, Boisvert J, Ladouceur A-M, Dorn JF & Maddox PS (2013) Octameric CENP-A nucleosomes are present at human centromeres throughout the cell cycle. Curr. Biol. CB 23: 764–769 Palmer DK, O’Day K & Margolis RL (1990) The centromere specific histone CENP-A is selectively retained in discrete foci in mammalian sperm nuclei. Chromosoma 100: 32–36 Palmer DK, O’Day K, Trong HL, Charbonneau H & Margolis RL (1991) Purification of the centromere-specific protein CENP-A and demonstration that it is a distinctive histone. Proc. Natl. Acad. Sci. U. S. A. 88: 3734–3738 Palmer DK, O’Day K, Wener MH, Andrews BS & Margolis RL (1987) A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histones. J. Cell Biol. 104: 805–815 Pardo-Manuel de Villena F de & Sapienza C (2001a) Transmission ratio distortion in offspring of heterozygous female carriers of Robertsonian translocations. Hum. Genet. 108: 31–36 Pardo-Manuel de Villena F & Sapienza C (2001b) Female meiosis drives karyotypic evolution in mammals. Genetics 159: 1179–1189 Pasque V, Gillich A, Garrett N & Gurdon JB (2011) Histone variant macroH2A confers resistance to nuclear reprogramming. EMBO J. 30: 2373–2387 Perpelescu M, Nozaki N, Obuse C, Yang H & Yoda K (2009) Active establishment of centromeric CENP-A chromatin by RSF complex. J. Cell Biol. 185: 397–407 Pinsky BA, Kung C, Shokat KM & Biggins S (2006) The Ipl1-Aurora protein kinase activates the spindle checkpoint by creating unattached kinetochores. Nat. Cell Biol. 8: 78–83 Piras FM, Nergadze SG, Magnani E, Bertoni L, Attolini C, Khoriauli L, Raimondi E & Giulotto E (2010) Uncoupling of Satellite DNA and Centromeric Function in the Genus Equus. PLoS Genet 6: e1000845

Epigenetics, Centromeres, Quantitative Biology

Pluta AF, Saitoh N, Goldberg I & Earnshaw WC (1992) Identification of a subdomain of CENP-B that is necessary and sufficient for localization to the human centromere. J. Cell Biol. 116: 1081–1093 Ponger L & Li W-H (2005) Evolutionary Diversification of DNA Methyltransferases in Eukaryotic Genomes. Mol. Biol. Evol. 22: 1119–1128 Prince CG & Demiris Y (2003) Editorial: Introduction to the Special Issue on Epigenetic Robotics. Adapt. Behav. 11: 75–77 Probst AV & Almouzni G (2008) Pericentric heterochromatin: dynamic organization during early development in mammals. Differentiation 76: 15–23 Prusiner SB (1982) Novel proteinaceous infectious particles cause scrapie. Science 216: 136–144 Prusiner SB (1998) Prions. Proc. Natl. Acad. Sci. U. S. A. 95: 13363–13383 Pryde FE & Louis EJ (1999) Limitations of silencing at native yeast telomeres. EMBO J. 18: 2538–2550 Ptashne M (2013) Epigenetics: Core misconcept. Proc. Natl. Acad. Sci. 110: 7101– 7103 Ramsahoye BH, Biniszkiewicz D, Lyko F, Clark V, Bird AP & Jaenisch R (2000) Non-CpG methylation is prevalent in embryonic stem cells and may be mediated by DNA methyltransferase 3a. Proc. Natl. Acad. Sci. 97: 5237–5242 Raychaudhuri N, Dubruille R, Orsi GA, Bagheri HC, Loppin B & Lehner CF (2012) Transgenerational propagation and quantitative maintenance of paternal centromeres depends on Cid/Cenp-A presence in Drosophila sperm. PLoS Biol. 10: e1001434 Ray-Gallet D, Quivy J-P, Scamps C, Martini EM-D, Lipinski M & Almouzni G (2002) HIRA is critical for a nucleosome assembly pathway independent of DNA synthesis. Mol. Cell 9: 1091–1100 Raynaud C, Mallory AC, Latrasse D, Jégu T, Bruggeman Q, Delarue M, Bergounioux C & Benhamed M (2014) Chromatin meets the cell cycle. J. Exp. Bot.: ert433 Rechavi O, Minevich G & Hobert O (2011) Transgenerational Inheritance of an Acquired Small RNA-Based Antiviral Response in C. elegans. Cell 147: 1248–1256 Reeves R (1992) Chromatin changes during the cell cycle. Curr. Opin. Cell Biol. 4: 413–423 Régnier V, Vagnarelli P, Fukagawa T, Zerjal T, Burns E, Trouche D, Earnshaw W & Brown W (2005) CENP-A is required for accurate chromosome segregation and sustained kinetochore association of BubR1. Mol. Cell. Biol. 25: 3967–3981 Reik W, Dean W & Walter J (2001) Epigenetic reprogramming in mammalian development. Science 293: 1089–1093 Ribeiro SA, Vagnarelli P, Dong Y, Hori T, McEwen BF, Fukagawa T, Flors C & Earnshaw WC (2010) A super-resolution map of the vertebrate kinetochore. Proc. Natl. Acad. Sci. U. S. A. 107: 10484–10489 Rocchi M, Archidiacono N, Schempp W, Capozzi O & Stanyon R (2012) Centromere repositioning in mammals. Heredity 108: 59–67

Chapter 1

Rodrigues-Martins A, Riparbelli M, Callaini G, Glover DM & Bettencourt-Dias M (2007) Revisiting the Role of the Mother Centriole in Centriole Biogenesis. Science 316: 1046–1050 Rodrigues-Martins A, Riparbelli M, Callaini G, Glover DM & Bettencourt-Dias M (2008) From centriole biogenesis to cellular function: Centrioles are essential for cell division at critical developmental stages. Cell Cycle 7: 11–16 Rogakou EP, Boon C, Redon C & Bonner WM (1999) Megabase Chromatin Domains Involved in DNA Double-Strand Breaks in Vivo. J. Cell Biol. 146: 905–916 Rosenfeld N, Young JW, Alon U, Swain PS & Elowitz MB (2005) Gene regulation at the single-cell level. Science 307: 1962–1965 Rosenfeld S (2011) Critical Self-Organized Self-Sustained Oscillations in Large Regulatory Networks: Towards Understanding the Gene Expression Initiation. Gene Regul. Syst. Biol.: 27 Rothbart SB & Strahl BD (2014) Interpreting the language of histone and DNA modifications. Biochim. Biophys. Acta BBA - Gene Regul. Mech. 1839: 627–643 Sacristan C & Kops GJPL (2015) Joined at the hip: kinetochores, microtubules, and spindle assembly checkpoint signaling. Trends Cell Biol. 25: 21–28 Salic A, Waters JC & Mitchison TJ (2004) Vertebrate Shugoshin Links Sister Centromere Cohesion and Kinetochore Microtubule Stability in Mitosis. Cell 118: 567–578 Salmon ED, Goode D, Maugel TK & Bonar DB (1976) Pressure-induced depolymerization of spindle microtubules. III. Differential stability in HeLa cells. J. Cell Biol. 69: 443–454 Santillan M, Mackey MC & Zeron ES (2007) Origin of Bistability in the lac Operon. Biophys. J. 92: 3830–3842 Sawatsubashi S, Murata T, Lim J, Fujiki R, Ito S, Suzuki E, Tanabe M, Zhao Y, et al (2010) A histone chaperone, DEK, transcriptionally coactivates a nuclear receptor. Genes Dev. 24: 159–170 Schittenhelm RB, Althoff F, Heidmann S & Lehner CF (2010) Detrimental incorporation of excess Cenp-A/Cid and Cenp-C into Drosophila centromeres is prevented by limiting amounts of the bridging factor Cal1. J. Cell Sci. 123: 3768– 3779 Schrader F (1935) Notes on the Mitotic Behavior of Long Chromosomes. Cytologia (Tokyo) 6: 422–430 Schuh M, Lehner CF & Heidmann S (2007) Incorporation of Drosophila CID/CENP-A and CENP-C into centromeres during early embryonic anaphase. Curr. Biol. CB 17: 237–243 Schutz-Geschwender A, Zhang Y, Holt T, McDermitt D & Olive DM (2004) Quantitative, two-color Western blot detection with infrared fluorescence Available at: http://cdn.licor.com/bio/PDF/IRquant.pdf [Accessed October 30, 2013] Schwille P (2001) Fluorescence correlation spectroscopy and its potential for intracellular applications. Cell Biochem. Biophys. 34: 383–408

Epigenetics, Centromeres, Quantitative Biology

Sclafani RA & Holzen TM (2007) Cell Cycle Regulation of DNA Replication. Annu. Rev. Genet. 41: 237–280 Scoville AG, Barnett LL, Bodbyl-Roels S, Kelly JK & Hileman LC (2011) Differential regulation of a MYB transcription factor is correlated with transgenerational epigenetic inheritance of trichome density in Mimulus guttatus. New Phytol. 191: 251–263 Sekulic N, Bassett EA, Rogers DJ & Black BE (2010) The structure of (CENP-A- H4)(2) reveals physical features that mark centromeres. Nature 467: 347–351 Shang W-H, Hori T, Toyoda A, Kato J, Popendorf K, Sakakibara Y, Fujiyama A & Fukagawa T (2010) Chickens possess centromeres with both extended tandem repeats and short non-tandem-repetitive sequences. Genome Res. 20: 1219–1228 Sharif J, Muto M, Takebayashi S, Suetake I, Iwamatsu A, Endo TA, Shinga J, Mizutani-Koseki Y, et al (2007) The SRA protein Np95 mediates epigenetic inheritance by recruiting Dnmt1 to methylated DNA. Nature 450: 908–912 Shea N (2009) Imitation as an inheritance system. Philos. Trans. R. Soc. B Biol. Sci. 364: 2429–2443 Shelby RD, Vafa O & Sullivan KF (1997) Assembly of CENP-A into centromeric chromatin requires a cooperative array of nucleosomal DNA contact sites. J. Cell Biol. 136: 501–513 Shivaraju M, Unruh JR, Slaughter BD, Mattingly M, Berman J & Gerton JL (2012) Cell-Cycle-Coupled Structural Oscillation of Centromeric Nucleosomes in Yeast. Cell 150: 304–316 Shorter J & Lindquist S (2005) Prions as adaptive conduits of memory and inheritance. Nat. Rev. Genet. 6: 435–450 Shuaib M, Ouararhni K, Dimitrov S & Hamiche A (2010) HJURP binds CENP-A via a highly conserved N-terminal domain and mediates its deposition at centromeres. Proc. Natl. Acad. Sci. U. S. A. 107: 1349–1354 Siegel TN, Hekstra DR, Kemp LE, Figueiredo LM, Lowell JE, Fenyo D, Wang X, Dewell S & Cross GAM (2009) Four histone variants mark the boundaries of polycistronic transcription units in Trypanosoma brucei. Genes Dev. 23: 1063– 1076 Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63 Sinsheimer RL (1955) The Action of Pancreatic Deoxyribonuclease Ii. Isomeric Dinucleotides. J. Biol. Chem. 215: 579–583 Smith S & Stillman B (1989) Purification and characterization of CAF-I, a human cell factor required for chromatin assembly during DNA replication in vitro. Cell 58: 15–25 Song L, Zhang Z, Grasfeder LL, Boyle AP, Giresi PG, Lee B-K, Sheffield NC, Graf S, et al (2011) Open chromatin defined by DNaseI and FAIRE identifies regulatory elements that shape cell-type identity. Genome Res. 21: 1757–1767

Chapter 1

Sonneborn TM (1964) The Differentiation of Cells. Proc. Natl. Acad. Sci. 51: 915– 929 Soshnikova N & Duboule D (2009) Epigenetic temporal control of mouse Hox genes in vivo. Science 324: 1320–1323 Stern S, Fridmann-Sirkis Y, Braun E & Soen Y (2012) Epigenetically heritable alteration of fly development in response to toxic challenge. Cell Rep. 1: 528–542 Stewart CL, Stuhlmann H, Jähner D & Jaenisch R (1982) De novo methylation, expression, and infectivity of retroviral genomes introduced into embryonal carcinoma cells. Proc. Natl. Acad. Sci. 79: 4098–4102 Stoler S, Keith KC, Curnick KE & Fitzgerald-Hayes M (1995) A mutation in CSE4, an essential gene encoding a novel chromatin-associated protein in yeast, causes chromosome nondisjunction and cell cycle arrest at mitosis. Genes Dev. 9: 573– 586 Strahl BD & Allis CD (2000) The language of covalent histone modifications. Nature 403: 41–45 Sudakin V, Ganoth D, Dahan A, Heller H, Hershko J, Luca FC, Ruderman JV & Hershko A (1995) The cyclosome, a large complex containing cyclin-selective ubiquitin ligase activity, targets cyclins for destruction at the end of mitosis. Mol. Biol. Cell 6: 185–197 Suhling K, Siegel J, Phillips D, French PMW, Lévêque-Fort S, Webb SED & Davis DM (2002) Imaging the Environment of Green Fluorescent Protein. Biophys. J. 83: 3589–3595 Sullivan KF, Hechenberger M & Masri K (1994) Human CENP-A contains a histone H3 related histone fold domain that is required for targeting to the centromere. J. Cell Biol. 127: 581–592 Sullivan SA, Aravind L, Makalowska I, Baxevanis AD & Landsman D (2000) The Histone Database: a comprehensive WWW resource for histones and histone fold- containing proteins. Nucleic Acids Res. 28: 320–322 Sullivan S, Sink DW, Trout KL, Makalowska I, Taylor PM, Baxevanis AD & Landsman D (2002) The Histone Database. Nucleic Acids Res. 30: 341–342 Sun X, Wahlstrom J & Karpen G (1997) Molecular Structure of a Functional Drosophila Centromere. Cell 91: 1007–1019 Tachiwana H, Kagawa W, Shiga T, Osakabe A, Miya Y, Saito K, Hayashi-Takanaka Y, Oda T, Sato M, Park S-Y, Kimura H & Kurumizaka H (2011) Crystal structure of the human centromeric nucleosome containing CENP-A. Nature 476: 232–235 Tagami H, Ray-Gallet D, Almouzni G & Nakatani Y (2004) Histone H3.1 and H3.3 complexes mediate nucleosome assembly pathways dependent or independent of DNA synthesis. Cell 116: 51–61 Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, Agarwal S, Iyer LM, Liu DR, Aravind L & Rao A (2009) Conversion of 5-Methylcytosine to 5- Hydroxymethylcytosine in Mammalian DNA by MLL Partner TET1. Science 324: 930–935

Epigenetics, Centromeres, Quantitative Biology

Talbert PB, Ahmad K, Almouzni G, Ausió J, Berger F, Bhalla PL, Bonner WM, Cande WZ, et al (2012) A unified phylogeny-based nomenclature for histone variants. Epigenetics Chromatin 5: 7 Talbert PB, Masuelli R, Tyagi AP, Comai L & Henikoff S (2002) Centromeric Localization and Adaptive Evolution of an Arabidopsis Histone H3 Variant. Plant Cell Online 14: 1053–1066 Tal O, Kisdi E & Jablonka E (2010) Epigenetic Contribution to Covariance Between Relatives. Genetics 184: 1037–1050 Terra SL, English CN, Hergert P, McEwen BF, Sluder G & Khodjakov A (2005) The de novo centriole assembly pathway in HeLa cells cell cycle progression and centriole assembly/maturation. J. Cell Biol. 168: 713–722 The Nucleus and DNA Replication (2015) Hum. Anat. Physiol. Available at: http://philschatz.com/anatomy-book/contents/m46073.html [Accessed January 9, 2015] Thoma F, Koller T & Klug A (1979) Involvement of histone H1 in the organization of the nucleosome and of the salt-dependent superstructures of chromatin. J. Cell Biol. 83: 403–427 Turner BM (2002) Cellular memory and the histone code. Cell 111: 285–291 Tyler-Smith C, Gimelli G, Giglio S, Floridia G, Pandya A, Terzoli G, Warburton PE, Earnshaw WC & Zuffardi O (1999) Transmission of a Fully Functional Human Neocentromere through Three Generations. Am. J. Hum. Genet. 64: 1440–1444 Uhlmann F, Lottspeich F & Nasmyth K (1999) Sister-chromatid separation at anaphase onset is promoted by cleavage of the cohesin subunit Scc1. Nature 400: 37–42 Uhlmann F & Nasmyth K (1998) Cohesion between sister chromatids must be established during DNA replication. Curr. Biol. CB 8: 1095–1101 Uhlmann F, Wernic D, Poupart MA, Koonin EV & Nasmyth K (2000) Cleavage of cohesin by the CD clan protease separin triggers anaphase in yeast. Cell 103: 375– 386 Ulbrich MH & Isacoff EY (2007) Subunit counting in membrane-bound proteins. Nat. Methods 4: 319–321 Ventura M, Mudge JM, Palumbo V, Burn S, Blennow E, Pierluigi M, Giorda R, Zuffardi O, Archidiacono N, Jackson MS & Rocchi M (2003) Neocentromeres in 15q24-26 map to duplicons which flanked an ancestral centromere in 15q25. Genome Res. 13: 2059–2068 Ventura M, Weigl S, Carbone L, Cardone MF, Misceo D, Teti M, D’Addabbo P, Wandall A, et al (2004) Recurrent Sites for New Centromere Seeding. Genome Res. 14: 1696–1703 Voullaire LE, Slater HR, Petrovic V & Choo KH (1993) A functional marker centromere with no detectable alpha-satellite, satellite III, or CENP-B protein: activation of a latent centromere? Am. J. Hum. Genet. 52: 1153–1163 Waddington CH (1942) The epigenotype. Endeavour 1: 18–20

Chapter 1

Wade CM, Giulotto E, Sigurdsson S, Zoli M, Gnerre S, Imsland F, Lear TL, Adelson DL, et al (2009) Genome Sequence, Comparative Analysis, and Population Genetics of the Domestic Horse. Science 326: 865–867 Waizenegger IC, Hauf S, Meinke A & Peters J-M (2000) Two Distinct Pathways Remove Mammalian Cohesin from Chromosome Arms in Prophase and from Centromeres in Anaphase. Cell 103: 399–410 Walkiewicz MP, Dimitriadis EK & Dalal Y (2014) CENP-A octamers do not confer a reduction in nucleosome height by AFM. Nat. Struct. Mol. Biol. 21: 2–3 Wandall A, Tranebjærg L & Tommerup N (1998) A neocentromere on human chromosome 3 without detectable α-satellite DNA forms morphologically normal kinetochores. Chromosoma 107: 359–365 Wang J, Liu X, Dou Z, Chen L, Jiang H, Fu C, Fu G, Liu D, et al (2014) Mitotic Regulator Mis18β Interacts with and Specifies the Centromeric Assembly of Molecular Chaperone Holliday Junction Recognition Protein (HJURP). J. Biol. Chem. 289: 8326–8336 Wang YV, Wade M, Wong E, Li Y-C, Rodewald LW & Wahl GM (2007) Quantitative analyses reveal the importance of regulated Hdmx degradation for P53 activation. Proc. Natl. Acad. Sci. 104: 12365–12370 Warburton PE, Cooke CA, Bourassa S, Vafa O, Sullivan BA, Stetten G, Gimelli G, Warburton D, Tyler-Smith C, Sullivan KF, Poirier GG & Earnshaw WC (1997) Immunolocalization of CENP-A suggests a distinct nucleosome structure at the inner kinetochore plate of active centromeres. Curr. Biol. CB 7: 901–904 Warren G & Wickner W (1996) Organelle Inheritance. Cell 84: 395–400 Waters JC, Chen RH, Murray AW & Salmon ED (1998) Localization of Mad2 to kinetochores depends on microtubule attachment, not tension. J. Cell Biol. 141: 1181–1191 Watson JD & Crick FH (1953a) Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid. Nature 171: 737–738 Watson JD & Crick FH (1953b) Genetical implications of the structure of deoxyribonucleic acid. Nature 171: 964–967 Weidemann T & Schwille P (2009) Fluorescence Correlation Spectroscopy in Living Cells. In Handbook of Single-Molecule Biophysics, Hinterdorfer P & Oijen A (eds) pp 217–241. New York, NY: Springer US Wiedemann SM, Mildner SN, Bönisch C, Israel L, Maiser A, Matheisl S, Straub T, Merkl R, Leonhardt H, Kremmer E, Schermelleh L & Hake SB (2010) Identification and characterization of two novel primate-specific histone H3 variants, H3.X and H3.Y. J. Cell Biol. 190: 777–791 Willard HF (1990) Centromeres of mammalian chromosomes. Trends Genet. TIG 6: 410–416 Willard HF & Waye JS (1987) Hierarchical order in chromosome-specific human alpha satellite DNA. Trends Genet. 3: 192–198

Epigenetics, Centromeres, Quantitative Biology

Williams JS, Hayashi T, Yanagida M & Russell P (2009) Fission yeast Scm3 mediates stable assembly of Cnp1/CENP-A into centromeric chromatin. Mol. Cell 33: 287–298 Wisniewski J, Hajj B, Chen J, Mizuguchi G, Xiao H, Wei D, Dahan M & Wu C (2014) Imaging the fate of histone Cse4 reveals de novo replacement in S phase and subsequent stable residence at centromeres. eLife 3: e02203 Woodcock CL & Ghosh RP (2010) Chromatin Higher-order Structure and Dynamics. Cold Spring Harb. Perspect. Biol. 2: a000596 Wu C, M. Bingham P, Livak KJ, Holmgren R & Elgin SCR (1979) The chromatin structure of specific genes: I. Evidence for higher order domains of defined DNA sequence. Cell 16: 797–806 Wu J-Q, McCormick CD & Pollard TD (2008) Chapter 9: Counting proteins in living cells by quantitative fluorescence microscopy with internal standards. Methods Cell Biol. 89: 253–273 Wu J-Q & Pollard TD (2005) Counting cytokinesis proteins globally and locally in fission yeast. Science 310: 310–314 Yoda K, Ando S, Morishita S, Houmura K, Hashimoto K, Takeyasu K & Okazaki T (2000) Human centromere protein A (CENP-A) can replace histone H3 in nucleosome reconstitution in vitro. Proc. Natl. Acad. Sci. U. S. A. 97: 7266–7271 Zhang W, Colmenares SU & Karpen GH (2012) Assembly of Drosophila centromeric nucleosomes requires CID dimerization. Mol. Cell 45: 263–269 Zur A & Brandeis M (2001) Securin degradation is mediated by fzy and fzr, and is required for complete chromatid separation but not for cytokinesis. EMBO J. 20: 792–801

CHAPTER 2

Analysis of Protein Turnover by Quantitative SNAP- Based Pulse-Chase Imaging

Dani L. Bodor, Mariluz Gómez Rodríguez, Nuno Moreno, and Lars E.T. Jansen

Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal.

NB: This chapter is a near literal transcription of Current Protocols in Cell Biology. 55:8.8:8.8.1–8.8.34. ABSTRACT

Assessment of protein dynamics in living cells is crucial for understanding their biological properties and function. The SNAP-tag, a self-labeling suicide enzyme presents a tool with unique features that can be adopted for determining protein dynamics in living cells. Here we present detailed protocols for the use of SNAP in fluorescent pulse-chase and quench-chase-pulse experiments. These time slicing methods provide powerful tools to assay and quantify the fate and turnover rate of proteins of different ages. We cover advantages and pitfalls of SNAP-tagging in fixed and live cell studies and evaluate the recently developed fast acting SNAPf variant. In addition, to facilitate the analysis of protein turnover datasets, we present an automated algorithm for spot recognition and quantification.

Fluoresent pulse-chase imaging and quantification

INTRODUCTION

The ability to track specific populations of proteins over time in living cells is essential to gain insight into the dynamics of cellular processes. An array of methodologies exists that assess different aspects of protein dynamics in living cells. These include fluorescence recovery after photobleaching (FRAP), photoactivation, and recombination induced tag exchange (see Table 2.1 for a more extensive list). Here we discuss SNAP-based pulse-chase imaging, a powerful method to track protein dynamics with distinct advantages over traditional methods to assess protein dynamics. SNAP is a suicide enzyme protein fusion tag that catalyzes its own covalent binding to the cell permeable molecule benzylguanine (BG), and (fluorescent) derivatives thereof (Figure 2.1; Damoiseaux et al, 2001; Keppler et al, 2003, 2004). Fusion of SNAP to a protein of interest allows this protein to be (fluorescently) labeled at will in living cells. Importantly, subsequent removal of the substrate results in the specific labeling of the initial pulse labeled pool. Changes in location and turnover of this pool can be determined and quantified. Moreover, serial labeling of SNAP-tagged proteins with different SNAP substrates distinguishes proteins synthesized at different times, such that “old” and “new” pools can be detected separately (Figure 2.3A and Jansen et al, 2007).

Figure 2.1 Principle of SNAP pulse labeling. SNAP is cloned as an epitope tag to a protein of interest. Reaction of SNAP fusion proteins with benzylguanine (or labeled derivatives) results in a covalent irreversible bond between the (labeled) benzyl moiety and a reactive cysteine in SNAP.

Chapter 2

Table 2.1 Methods to Analyze Protein Turnover

2001

References

Lin etLin al, 2008

Deal et al, 2010

et al, 2011 (SNAPf)

Nishimura et al, 2009Nishimura

Verzijlbergen et al, 2010 Verzijlbergen

reviewed in O'Hare etreviewed in O'Hare al, 2007

reviewed in: Lukyanov et al, 2005

Keppler et Keppler al, 2003 (SNAP); Los and

reviewed in: Lippincott-Schwartz etreviewed al, in: Lippincott-Schwartz

Wood, 2007 (Halo); Gautier et al, 2008

(CLIP); et al, 2009 Gallagher (TMP); Sun

CFP2

2009

D-tag

al, 2010

Examples

FRAP, FLIPFRAP,

PAmCherry

mEos2, Kaede,

FCS, FCCS, RICS FCCS, FCS,

sulfatase, Q-tag sulfatase,

Dronpa, Padron,

CLIPf, TMPHalo,

ACP/PCP, Biotin,

see Lin et al, 2008

SNAP, SNAPf, CLIP, SNAPf, SNAP,

bsDronpa, PA-GFP, bsDronpa,

see Deal et al, 2010

see Nishimura et al, see Nishimura

see Verzijlbergen et see Verzijlbergen

Dendra-2, KiKGR,Dendra-2, PS-

TC-tag, 6His-tag, Poly-TC-tag, 6His-tag,

pools

surface

protein poolsprotein

Disadvantages

proteins ofproteins interest

specialized equipment specialized

All proteins are labeled are proteins labeled All

fluorescent fluorescent background

Unspecific labeling; toxicity labeling; Unspecific

requires cell by cell activationrequires

reaction; only onpossible cell

requires cell by cell activationrequires

simultaneously, thus requiring requiring simultaneously, thus

Not at timescalespossible long

of cells Cre-recombination in all

short timescales (sec-min); high

(hours-days), cell by (hours-days), cell analysis

Fluorescence Fluorescence prior to activation;

Blocks two fluorescent channels;

Does not allow measurements at

Requires introduction of introduction Requires multiple

downstream techniques downstreamto techniques purify

Only worksOnly for mobile atproteins

Analysis depends on completeness depends Analysis

Only allowsOnly ofanalysis new protein

low concentrations; highly requires

Requires foreign enzyme foreign Requires to catalyze

proteins. Only allowsproteins. ofanalysis new

tags

animals

Advantages

protein poolsprotein

Size (4-80aa's) Size

Size (6-12aa's) Size

simultaneously

activation channel

and new protein poolsand

a tagged poola tagged of protein

subcellular pool of subcellular protein

Allows ofAllows analysis a specific

pools prior ;to visaulization

Allows at Allows veryanalysis short

Allows Allows measurements at long

timescales (minutes-seconds)

accessibility intoaccessibility tissues of live

rates; single moleculerates; single sensitity

Allows Allows vsof analysis 'old' 'new'

Drug is sufficiently small is tosufficiently Drug allow

Allows Allows determinationaccurate of

No requirement forNo requirement or transgenes

Allows Allows measurements of both old

activation; in post- low background

protein concetrations and diffusion concetrationsprotein diffusion and

Allows rapid specific degradation of specific degradation Allows rapid

Allows analysis of subcellular protein ofAllows analysis subcellular

timescales, of ofnumbers cells. large

activation

expression

compounds

photobleaching

(artificial) groups (artificial)

Short description Short

methionine methionine analog

(time & extent) after

covalent binding to small covalent binding

inhibited by drug addition by drug inhibited

"change color""change laser upon

translationally modifiedtranslationally by

synthesized proteins with a synthesized

Enzymes that catalyze their

by a different tag upon Cre- by a tag upon different

cell-exogenous proteasome

Short peptide tags that have Short peptide

Metabolical labeling of newlyMetabolical labeling

Fluorescent that Fluorescent proteins can

affinity toaffinity chemical substances

proteins in a veryproteins small volume

"turned on" "turned by laser activation

Measures local turnover protein

Fluorescent thatbe Fluorescent proteins can

Self degrading protein tag that protein tag is Self degrading

Tags that can be specifically post- be Tags that can specifically

Measures diffusion ofMeasures fluorescent diffusion

Floxed protein tag that is replaced Floxed that protein tag is replaced

Inducible degradation tag, through through tag, degradation Inducible

(FRAP)

(PCFPs)

(PAFPs)

based tags based

(CATCH-IT)

(TimeSTAMP)

Self labeling tags labeling Self

Post translational translational Post

Photo-activatable Photo-activatable

Photo-convertible Photo-convertible

Spectroscopy (FCS) Spectroscopy

modification (PTM) modification

Tag Exchange (RITE)Tag Exchange

fluorescent proteins proteins fluorescent

system (AID system)

Chelation based tags based Chelation

After Photobleaching After Photobleaching

and identify turnover identify turnover and

Fluorescence RecoveryFluorescence

Auxin-inducible degron degron Auxin-inducible

Covalent attachment of of attachment Covalent

Recombination Induced Induced Recombination

tags to capture histones histones capture tags to

Fluorescence Correlation Correlation Fluorescence

Time Specific Tag Age for Measurement of Proteins of Measurement

or amino-acid analogs amino-acid or chemically modified chemically fluorescent proteins fluorescent fluorescent proteins fluorescent

Methods using other kinds of protein tags tags protein of kinds other using Methods Methods using tags that can be be can that tags using Methods Methods using inducible inducible using Methods Methods using auto- using Methods

Fluoresent pulse-chase imaging and quantification

Principle advantages of using SNAP-tagging include 1) pools of protein synthesized at different times can be specifically visualized, which allows for determining the fate of pre-existing versus newly synthesized pools of the same protein. 2) Because labeling occurs at a population basis, large numbers of cells can be analyzed in a single experiment. 3) Labeling and turnover occurs in the culture chamber rather than on the microscope stage. Therefore, cells are not continuously imaged, but sampled for imaging at any timepoint from hours to days post labeling. A more extensive comparison of SNAP with other pulse labeling techniques as well as its advantages and disadvantages can be found in Table 2.1 and below in the Background Information. In this chapter, we explain in detail how to perform a typical SNAP pulse labeling experiment in human cells. As an example, we will use HeLa cells that stably express a SNAP-tagged version of CENP-A, a centromere specific histone variant (Sullivan et al, 1994; Jansen et al, 2007). Using these CENP-A-SNAP cells, we have been able to show previously that the rate of centromeric CENP-A turnover corresponds to the rate of cell division, and thus that CENP-A turns over exclusively by dilution during DNA replication (Jansen et al, 2007). Using the same technology, we demonstrated that newly synthesized pools of CENP-A assemble specifically during G1 phase of the cell cycle (Jansen et al, 2007). The unique dynamics of CENP-A makes this an excellent illustration of the SNAP-labeling technique. However, this strategy is easily adaptable to other proteins (e.g. Figure 2.3D) as well, and similar strategies have been used by us and other investigators, in a range of organisms and for different applications (Jansen et al., 2007; Erhardt et al., 2008; McMurray and Thorner, 2008; Maduzia et al., 2010; Bojkowska et al., 2011; Campos et al., 2011; Dunleavy et al., 2011; Silva et al., in press; also reviewed in O’Hare et al., 2007). We will describe two typical types of SNAP-labeling strategies: pulse- chase (Basic Protocol 1) and quench-chase-pulse (Basic Protocol 2), which allow for the analysis of old and new protein pools, respectively. We also

Chapter 2

describe potential ways to combine SNAP labeling with cell synchronization and siRNA mediated protein depletion (Basic Protocol 3). Cells can be either analyzed by live imaging (Basic Protocol 4) or fixed and combined with standard techniques such as immunofluorescence (Supporting Protocol 2). In addition, we present an unbiased, automated algorithm that is used for fluorescence measurements to quantify protein turnover (Basic Protocol 5) Lastly, we present an evaluation of SNAP pros, cons, pitfalls and ways to troubleshoot them as well as the recently developed variant of SNAP, SNAPf.

Fluoresent pulse-chase imaging and quantification

BASIC PROTOCOL 1: PULSE-CHASE

This section describes a general method that employs a pulse-chase strategy for analysis of a specific pool of protein in living cells. By using fluorescence pulse labeling, the fate and turnover rate of a given protein can be determined at a particular subcellular location. Specifically, SNAP-tagged protein that is present at the beginning of an experiment is fluorescently labeled (pulse) followed by removal of excess dye. After a given amount of time (chase), cells are analyzed e.g. for localization or quantity of remaining protein by (quantitative) fluorescence microscopy (Figure 2.2A). An example of a typical pulse-chase experiment of CENP-A-SNAP is shown in Figure 2.2B. In the approach described here, cells are fixed and analyzed at set time points following the initial pulse. As a consequence, protein dynamics can be determined at any time frame (hours, days) post labeling. However, initial labeling and wash steps require approximately one hour, precluding analysis of highly dynamic processes that occur at a timescale of seconds to minutes.

Materials - Cells expressing SNAP-tagged fusion protein (see Supporting Protocol 1) - Trypsin (cell culture grade, Gibco) - Standard culture medium abbreviated to “CM” (see Reagents & Solutions). - TMR-Star (see Reagents & Solutions). - Sterile DMSO - Sterile 1X PBS (cell culture grade, Gibco) - 24-well plates - Clean, sterile, poly-lysine coated coverslips (12 mm ø; Thickness 1.5) - Vortex and tabletop centrifuge

Chapter 2

Figure 2.2 Pulse-chase imaging. (A) Schematic outlining an in vivo SNAP pulse labeling strategy (Basic Protocol 1).

Cells that produce SNAP-tagged protein are incubated with the SNAP substrate TMR-Star (Pulse) at time T0, rendering the available cellular pool of SNAP fluorescent. Following substrate washout (Chase), cells continue to synthesize SNAP protein (light blue) that is not labeled, while the pulse labeled pool turns over. The remaining pulse labeled pool of SNAP can be visualized and quantified at various time points (Tn) during the chase by microscopy. (B) Example of a pulse- chase experiment using cells expressing CENP-A-SNAP. CENP-A (top) localizes to centromeres , which are visualized as subnuclear, diffraction limited foci. Cells are pulse labeled at 0h with TMR-Star after which they are chased and the remaining pulse labeled pool is visualized by high magnification microscopy at indicated time points. After 72 hours a small but detectable pool of CENP-A-SNAP is still present at centromeres (inset at 72h shows rescaled CENP-A::TMR-Star and CENP-C signals). Cells were counterstained with CENP-C (green) and DAPI (blue).

Fluoresent pulse-chase imaging and quantification

Preparation of cells and SNAP-substrates 1) Prepare coverslips in separate wells of a 24-well plate to minimize the required incubation volumes. Trypsinize cells expressing SNAP-tagged fusion protein and seed onto the coverslips. Incubate at 37°C, 5% CO2 (henceforth referred to as standard growth conditions). The cell density depends on a number of factors, mainly cell type and the number of days between seeding cells and fixation. Ideally, by the time of fixation, the cell density should be high enough to capture a significant amount of cells on each frame, but not too high such that cells are fully confluent. Generally 60–80% confluency is ideal. For HeLa cells (duplication time ~1 day), we aim for having ~5·105 cells at the time of fixation. E.g., ~1·105 cells are seeded in the afternoon of day 1, if fixation will take place in the morning of day 4. 2) Dilute TMR-Star stock to 2 μM final concentration in CM. Vortex briefly to efficiently disperse the DMSO solvent into the aqueous medium. Dilute an equal volume of DMSO for mock labeling control. Prepare >200 μl per coverslip. Prepare TMR-Star working stock only as needed and use within the hour. Although labeling is not yet saturated at this concentration, we use 2 μM to balance signal intensities and costs per experiment (see Critical Parameters and Troubleshooting for more details). DMSO addition is an important initial control to determine background fluorescence unrelated to SNAP-labeling, as well as to determine the effect of DMSO on the cells. Once these factors have been established and an effect on cell viability, cell cycle progression, etc. are excluded for a given cell line, this control can be omitted from subsequent experiments. 3) Spin diluted TMR-Star for 5 minutes at maximum speed (~16.000 g) in a microcentrifuge to get rid of possible insoluble fluorescent debris. Recover as much of the supernatant as possible without disturbing the pellet (may not be visible). Omitting this step will result in occasional but very bright fluorescent aggregates that interfere with imaging and quantification of fluorescent signals.

Chapter 2

Pulse labeling and washes 4) Aspirate CM from cells and add 200 μl of CM+TMR-Star or CM+DMSO. Incubate for 15 minutes at standard growth conditions. TMR- Star treatment of cells will likely result in non-specific fluorescence (see Critical Parameters and Troubleshooting). It is therefore important to conduct pilot experiments in which the parent cells without expression of SNAP are labeled to discriminate SNAP dependent fluorescence from unspecific fluorescence. 5) Wash cells twice with 1 ml sterile PBS (pre-heated to 37°C) to wash away free substrate. Re-incubate cells in CM under standard growth conditions for an additional 30 minutes. In our experience, in experiments where the cells have undergone multiple consecutive treatments prior to labeling (e.g. synchronization, RNAi, drug treatments), it is preferable to perform the washes with CM rather than PBS in this and the following steps. This enhances cell survival.

6) Wash cells twice with 1 ml sterile PBS (pre-heated to 37°C). This second wash is important to remove any substrate that was retained in the cells after the initial wash. In our experience, omitting this step leads to a significant increase in background fluorescence. We calculate the chase period from the completion of this wash step, as this indicates the last time point during which SNAP-tagged proteins can be fluorescently labeled.

Chase and post processing 7) There are 3 general options to proceed. Details are presented in subsequent sections: a. Pulse-fix: Fix cells immediately after the second wash and either image directly or process for immunofluorescence (Supporting Protocol 2). This allows testing for SNAP-expression levels and/or serves as a control for subsequent pulse-chase experiments. b. Pulse-chase: Re-add 1 ml of CM and incubate cells in standard growth conditions for a given amount of time (chase period), after which cells are fixed and treated for immunofluorescence. c. Pulse-image: Mount cells for live imaging (Basic Protocol 4).

Fluoresent pulse-chase imaging and quantification

BASIC PROTOCOL 2: QUENCH-CHASE-PULSE

In this section we describe a general method that allows for the analysis of a ‘new’ pool of protein. Specifically, the pool of SNAP-tagged protein that is present at the onset of an experiment is labeled by a non-fluorescent SNAP-substrate (quench). Subsequently, after a given amount of time (chase), cells are labeled with a second, fluorescent substrate (pulse). In this way only the pool of protein synthesized during the chase period is fluorescently labeled and hence will be visible by microscopy (Figure 2.3A), while the initial quenched pool remains undetected (Figure 2.3B). This approach allows for e.g. quantitative and temporal analysis of protein translocation and/or assembly into subcellular domains. Examples of typical quench-chase-pulse experiments are shown in Figure 2.3C–D.

Materials - All materials used in Basic Protocol 1; in addition: - BTP (see Reagents & Solutions)

Preparation of cells and SNAP-substrates 1) Prepare coverslips and cells as in step 1 of Basic Protocol 1. 2) Dilute BTP to 2 μM final concentration in CM. Vortex briefly to efficiently disperse the DMSO solvent into the aqueous medium. Prepare >200 μl per coverslip. Prepare BTP working stock only as needed and use within the hour. We have successfully used BTP at concentrations as low as 0.2 μM, resulting in fully quenched SNAP-labeling. However, because full quenching is essential for accurate interpretation of the results, we prefer using BTP at an excess of 2 μM (see step 6 for determination of quench efficiency).

Quench labeling and washes

Quench labeling is performed much in the same way as the pulse labeling described in Basic Protocol 1. The main difference is the time of initial incubation with BTP: 30 minutes, as compared to 15 minutes for TMR-Star (compare step 3 of this protocol with step 4 of Basic Protocol 1).

Chapter 2

Fluoresent pulse-chase imaging and quantification

Figure 2.3 (previous page) Quench-chase-pulse imaging. (A) Schematic outlining an in vivo SNAP quench-chase- pulse labeling strategy (Basic Protocol 2). Cells that produce and turnover SNAP-tagged protein are incubated with a non-fluorescent SNAP substrate BTP (Quench) at time T0, rendering the available cellular pool unavailable for subsequent fluorescent labeling (dark blue). Following substrate washout (chase), cells continue to synthesize SNAP protein (light blue) that is not labeled. After a set chase time, nascent protein is specifically labeled with TMR-Star. This nascent (new) fluorescent pool of SNAP can be visualized and quantified at various time points (Tn) during the subsequent chase by microscopy. (B) Quench-pulse control. Cells expressing CENP-A-SNAP were either pulse labeled with TMR-Star (Pulse) or quenched with BTP immediately preceding the pulse labeling step (Quench-pulse) followed by immunofluorescence and imaging. While pulse labeling results in fluorescent centromeric CENP-A-SNAP, pre- incubation of cells with BTP (Quench) renders this pool undetectable. Cells are counterstained with anti-HA, which detects the total pool of (CENP-A-) SNAP. The merged image shows TMR-Star (green) and HA (red) signals together with DAPI stain (blue). (C) Cells expressing CENP-A-SNAP were subjected to a quench-chase-pulse experiment as outlined in (A), processed for immunofluorescence and imaged. Nascent CENP-A-SNAP (green) localizes to centromeres only in a subset of cells (arrow) while remaining non-centromeric in others (arrow heads) highlighting a cell cycle dependence in nascent CENP-A-SNAP dynamics (Jansen et al., 2007). Cells are counterstained with anti-tubulin (red) and DAPI (blue) to visualize microtubules and DNA, respectively. (D) Experiment as in (C) except that cells expressing SNAP-tagged histone H3.1 were subjected to the quench-chase-pulse protocol. H3.1 is a canonical histone that assembles into chromatin in S phase. Cells that either do not assemble (arrowhead) or are in various stages of nascent histone H3.1 (red) assembly (arrows) are shown. Cells are counterstained with DAPI to visualize DNA (blue). Panels B and C are adapted from Jansen et al., 2007.

3) Aspirate CM from cells and add 200 μl of CM+BTP or CM+DMSO. Incubate for 30 minutes at standard growth conditions. 4) Wash cells twice with 1 ml sterile PBS (pre-heated to 37°C) to wash away free substrate. Re-incubate cells in CM and standard growth conditions for an additional 30 minutes. In our experience, in experiments where the cells have undergone multiple consecutive treatments prior to labeling (e.g. synchronization, RNAi, drug treatments), it is preferable to perform the washes with CM rather than PBS in this and the following steps. This enhances cell survival.

5) Wash cells twice with 1 ml sterile PBS (pre-heated to 37°C). The second wash is important to remove all traces of free BTP. Omission of this wash will lead to continued quenching of a proportion of newly synthesized protein during the chase resulting in smaller pool size of subsequently labeled nascent protein. We calculate the chase period from the completion of this wash step, as this indicates the last time point during which SNAP-tagged proteins can be labeled by the non-fluorescent substrate.

Chapter 2

Quench-pulse control 6) Label at least one coverslip with TMR-Star directly following the quench step (no chase) as described in steps 2 through 7 of Basic Protocol 1. This is a very important control experiment, as it indicates whether or not the preexisting SNAP-tagged protein is fully quenched by the available BTP (Figure 2.3B). If this is not the case, results are very difficult, if not impossible, to interpret correctly. If BTP labeling is not complete, it may be necessary to increase the concentration of BTP and/or the incubation time. Once conditions that lead to a complete quenching of SNAP-tagged protein has been determined for a particular cell type and application, this control can be omitted in subsequent experiments.

Chase 7) Re-incubate cells in CM under standard growth conditions for the appropriate time. Chase times will dependent, amongst other things, on the expression levels of the protein of interest and cell type used. Typically in human cell culture a chase of several hours is required to create a pool size large enough for subsequent visualization by pulse labeling (e.g. for the case of CENP-A-SNAP, we found the minimum chase time required to detect nascent protein is 3 hours).

Pulse labeling and washes 8) For fluorescent pulse labeling and downstream applications, follow steps 2 through 7 from Basic Protocol 1.

Fluoresent pulse-chase imaging and quantification

BASIC PROTOCOL 3: COMBINING SNAP EXPERIMENTS WITH CELL

SYNCHRONIZATION AND RNAI

Protocol 3.1: Quench-Chase-Pulse In this section we describe how to combine the SNAP-labeling procedure with cell synchronization and/or siRNA mediated protein depletion in HeLa cells. We will give a full overview of multiple synchronization and depletion steps integrated into a single quench-chase-pulse experiment (Figure 2.4A). This allows for the determination of the fate of a newly synthesized pool of protein during the cell cycle and in response to protein depletions. It should be noted that depending on the specific experiment, in many cases not all steps will be required. An example of a typical synchronized quench-chase- pulse experiment is shown in Figure 2.4B.

Materials - All materials used in Basic Protocol 2; in addition: - Thymidine, stock of 50 mM in water - Deoxycytidine, stock of 24 mM in water - siRNAs and transfection reagents - Nocodazole stock 5 mg/ml and/or MG132 stock of 10mM

Preparation of cells and synchronization and RNAi 1) Prepare cells on coverslips as described in step 1 of Basic Protocol 1. 2) Perform siRNA transfection for analysis of RNAi mediated protein depletion at ~48–72 hours post transfection. This step is performed as described in the product description protocol for Oligofectamine (Invitrogen). Wait at least 4–5 hours before proceeding to step 3. Protein depletion can only be performed at this point in the protocol (of a synchronized experiment) if the depleted proteins are not involved in cell cycle progression. For proteins that are likely to interfere with S or M phase transition, siRNA transfection is best performed at a later stage in the protocol (see steps 5 and 9).

Chapter 2

Figure 2.4 Combination of SNAP labeling, synchronization, and RNAi. (A) Schematic outline of quench-chase-pulse protocol combined with double thymidine arrest and RNAi as described in Basic Protocol 3. (B) Combining quench- chase-pulse labeling with cell synchronization. CENP-A-SNAP cells arrested at the G1/S boundary by double thymidine block (as in A) were treated with BTP to quench available SNAP pools followed by release into S phase, during which new protein was synthesized. The nascent pool of SNAP was pulse labeled with TMR-Star after a 7 hour chase (end of S phase). Cells were fixed at different time points to analyze centromere localization of nascent CENP-A-SNAP in S, G2, mitosis (M), and G1 phase. While the nascent pool is labeled at 7 hours post release (G2), it does not localize to centromeres until G1. Cells are counterstained with anti-HA, which detects the total pool of SNAP. (C) Combining quench-chase-pulse and pulse-chase labeling with RNAi. Asynchronous CENP-A-SNAP expressing cells were transfected with siRNAs to block synthesis of CENP-A or of a control protein (GAPDH). Cells were pulse-chase (left) or quench-chase-pulse labeled (right) at indicated time points and assayed 48 hours after siRNA addition to determine the fate of old and new pools of protein, respectively. CENP-A-SNAP::TMR-Star signals representing old and new protein pools are shown following RNAi. Cells were counterstained with CENP-C (green) and DAPI (blue). TMR-Star centromere intensity levels at the centromere were determined by CRaQ (Basic Protocol 5). Average centromeric CENP-A-SNAP::TMR-Star signals were determined from 3 replicate experiments. Signals after GAPDH RNAi were set to 1. Error bars indicate standard error of the mean (SEM). While CENP-A RNAi impairs the synthesis and accumulation of nascent CENP-A (new pool) the pool synthesized prior to siRNA addition is unaffected, demonstrating the ability to differentially visualize old and new protein pools. Panel B is adapted from Jansen et al., 2007.

Fluoresent pulse-chase imaging and quantification

3) Add thymidine to the CM at a final concentration of 2 mM and incubate cells at standard growth conditions for 17 hours. Cells that are in S phase when thymidine is added will arrest immediately, while other cells progress until they enter S phase and arrest there. Thus, after 17 hours, all cells will be arrested in S phase albeit at different stages of S phase completion. Spike in thymidine rather than replacing the CM with CM+thymidine (if RNAi was performed during step 2), as this would wash out siRNAs from the medium and reduce the efficiency of protein depletion. If siRNAs are transfected with oligofectamine in serum free medium in step 2 then serum can be re-added (along with thymidine) at this point to a final concentration of 10%. 4) Release cells from thymidine arrest by performing two washes with CM, followed by addition of CM+deoxycytidine (24 μM final concentration). Incubate cells at standard growth conditions for 9 hours. 5) At 5 hours after release from the first thymidine arrest, siRNA transfection can be performed for analysis of RNAi mediated protein depletion at ~24–48 hours post transfection. This step is performed as described in the product description protocol for Oligofectamine (Invitrogen). Protein depletion can be performed at this point in the protocol for proteins that are (likely to be) required for mitotic progression, because significant levels of protein depletion are generally only observed at least 4–5 hours after siRNA transfection. At this point (~10 hours after release from the first thymidine arrest), most cells will have passed through mitosis already. For proteins that are not involved in cell cycle progression, siRNA transfection can be performed at an earlier point (see step 2), while proteins that are involved in S phase progression are best depleted at a later point (see step 9). 6) 9 hours after the release described in step 4, add thymidine to the CM to a final concentration of 2 mM. Incubate cells at standard growth conditions for 15.5 hours. At this time all cells will have finished DNA replication, while none have started the next S phase, regardless at which point in S phase they were arrested initially. Spike in thymidine rather than replacing the CM with CM+thymidine (if RNAi was performed during step 5), as this would wash out siRNAs from the medium and reduce the efficiency of protein depletion.

Chapter 2

If siRNAs are transfected with oligofectamine in serum free medium in step 5 then serum can be re-added (along with thymidine) at this point to a final concentration of 10%.

Quench labeling and washes 7) 15.5 hours after thymidine addition in step 6, perform quench- labeling (and 1st washout thereof) essentially as described in steps 3-6 of Basic Protocol 2, except that 2 mM thymidine is added to the CM+BTP and CM in order to maintain cells in the S phase arrest until after the labeling is complete. 8) 30 minutes after step 7, release cells from second thymidine arrest and perform second BTP washout by performing two washes with CM, followed by addition of CM+deoxycytidine (24 μM final concentration). This step combines the second wash of the BTP-labeling and release from second thymidine arrest. Cells will now (16 hours after initiation of second thymidine arrest) all be synchronously released from early S phase and will progress through the cell cycle largely synchronous for approximately one full cell cycle. Cells will enter mitosis at ~9–11 hours after release from the second thymidine arrest.

Chase 9) ~3 hours after release from the second thymidine arrest (step 8), siRNA transfection can be performed for analysis of RNAi mediated protein depletion at early timepoints post transfection. This step is performed as described in the product description protocol for Oligofectamine (Invitrogen). Protein depletion can be performed at this point in the protocol for proteins that are (likely to be) required for S phase progression, because significant levels of protein depletion are generally only observed at least 4–5 hours after siRNA transfection. At this point (~8 hours after release from the second thymidine arrest), most cells will have passed through S phase already. Since maximum protein depletion is generally observed 24–48 hours post-transfection, for proteins that are not involved in S phase progression, siRNA transfection are best performed at an earlier point (see steps 2 and 5).

Fluoresent pulse-chase imaging and quantification

Pulse labeling and washes 10) TMR-Star pulse labeling and downstream applications are performed as described in Basic Protocol 1, steps 4–7 at different time points following BTP-quench and thymidine release depending on the application. If siRNAs are transfected with oligofectamine in serum free medium in step 9 then serum can be re-added after 2nd washout of TMR-Star. 11) Optional: To gain higher synchrony in and around mitosis, cells can be arrested in mitosis by addition of the microtubule depolymerizing drug nocodazole to 250 ng/ml final concentration will result in a prometaphase arrest), or addition of nocodazole and washout of this drug into the proteasome inhibitor MG132 (24 μM final; metaphase arrest). Nocodazole can be added at any time to allow accumulation of cells in mitosis (optimal concentration will depend on cell type). MG132 will arrest cells in interphase unless added in late G2 phase in which case cells will continue to cycle until metaphase. Metaphase synchronization of cells by MG132 is therefore best combined with a (double thymidine arrest, release and) nocodazole arrest and release. Arrest from these drugs is reversible, allowing the analysis of cells that are synchronously released from mitosis. 12) Optional: 9 hours after release from the second thymidine arrest, thymidine (final concentration of 2 mM) can be re-added to collect cells synchronously at the next G1/S phase transition, 15 hours later.

Protocol 3.2: Pulse-Chase Here, we describe a different version of Basic Protocol 3, where a pulse- chase strategy is employed rather than quench-chase-pulse. This allows for tracking of a pre-existing pool of SNAP (as opposed to a newly synthesized pool) in relation to the cell cycle and in response to protein depletions. This protocol is highly similar to the Basic Protocol above and therefore we will only describe the key steps that are different between the two protocols.

Chapter 2

This alternate protocol can also be performed in parallel with Protocol 3.1, e.g. to distinguish a differential effect on separate pools of the same protein (an example is given in Figure 2.4C).

Materials - All reagents used in Protocol 3.1, except for BTP

Preparation of cells and synchronization and RNAi 1) Cells are prepared, and treated with siRNAs and synchronized with thymidine as described in Protocol 3.1 steps 1–6.

Pulse labeling and washes 2) 15h and 15 minutes after thymidine addition in step 6 of Protocol 3.1, perform TMR-Star pulse labeling (and 1st washout thereof), essentially as described in steps 4–6 of Basic Protocol 1, except that 2 mM thymidine is added to the CM+TMR-Star and CM in order to maintain cells in the S phase arrest until after the labeling is complete. 3) 30 minutes after step 2, release cells from second thymidine arrest and perform second TMR-Star washout by performing two washes with CM, followed by addition of CM+deoxycytidine (24 μM final concentration). This step combines the second wash of the TMR-Star-labeling and release from second thymidine arrest. 4) Proceed to downstream applications as described in step 7 of Basic Protocol 1.

Fluoresent pulse-chase imaging and quantification

BASIC PROTOCOL 4: LIVE IMAGING OF PULSE LABELED CELLS

This section will describe the basic procedure and considerations of imaging SNAP substrate signals in living cells. Live cell imaging of SNAP labeled proteins differs from conventional imaging of autofluorescent proteins (e.g. GFP) in that SNAP substrates generate considerable background staining, particularly in membrane compartments. This requires specific signals to be of sufficient strength to maintain an adequate signal-to-noise ratio. Despite this constraint, live cell imaging of temporally labeled SNAP-tagged proteins is a powerful approach to determine the fate of protein pools of different ages (Figure 2.5). We will discuss two different methods (Protocols 4.1 and 4.2) of preparing cells for live imaging.

Figure 2.5 Live cell imaging of SNAP labeled cells. Schematic outlines cell synchronization and quench-chase-pulse labeling steps as shown in Figure 2.4B. Following pulse labeling, cells are cycled into mitosis and mounted for live cell imaging (Basic Protocol 4). Time lapse series is shown of a cell in mitosis. At early time points, TMR-Star signals are non-centromeric, but are observed near the cell periphery, probably reflecting non-specific retention of the fluorescent substrate in cellular membranes. As cells exit from mitosis (after anaphase, t=0 minutes) TMR-Star signal accumulates at centromeres from t=50 minutes onwards. Cells express GFP-CENP-C that constitutively labels centromeres throughout the experiment. Insets show colocalization of nascent CENP-A-SNAP::TMR-Star (green) with centromeres (CENP-C, red). Image is adapted from Jansen et al., 2007.

Chapter 2

Materials - Materials and reagents for SNAP pulse labeling as described in Basic Protocol 2; in addition: - Live imaging medium, referred to as “LM” (see Reagents & Solutions) - Microscopy facilities suitable for live cell imaging (see below for some general considerations) - VALAP (see Reagents and Solutions; for Method 1 only) - Oxyrase (Oxyrase Inc.), stock of 30 U/ml (for Method 1 only) - 6-well plates(for Method 1 only) - 22x22 mm square coverslips (for Method 1 only) - Permanent double-sided tape (Scotch; for Method 1 only) - Standard glass slides (for Method 1 only) - 8-well Chambered Coverglass (Lab-Tek; for Method 2 only) - Mineral oil (for Method 2 only)

Protocol 4.1: double side sticky tape chamber This method is adapted from (Waterman-Storer & Salmon, 1997).

Preparation of cells and pulse labeling 1) Grow cells expressing SNAP-tag fusion proteins in 6-well plates onto 22x22 mm square glass coverslips in 2 ml of culture medium to 60–80% confluency. 2) Perform quench and pulse labeling steps as in Basic Protocols 1 or 2, except that labeling volumes of 600 μl are used in 6-well plates.

Mounting of live cell chambers 3) Glue 3 layers of double-sided tape, cut to ~3 mm wide, along the two long edges of the glass slide such that when a coverslip is placed on top, it is sealed on two sides (along the longitudinal end of the glass slide).

Fluoresent pulse-chase imaging and quantification

4) Mount coverslips, cells facing down, onto the glass slide prepared in step 3. 5) Slowly, flow in LM under the coverslip, until the chamber is filled by capillary action (<1 ml). Perform this step as quickly as possible after step 4 to avoid cells drying out. Phenol red is omitted from the LM to avoid background fluorescence. The use of CO2 independent medium (e.g. buffered by HEPES) is required to maintain pH in this chamber type as it is sealed from outside air contact. Optionally, 0.5 U/ml Oxyrase is included in the medium. Oxyrase is an oxygen-scavenging enzyme that helps reduce photobleaching and phototoxicity due to reactive oxygen species. 6) Seal the chamber on all sides with VALAP and image live cells on the microscope.

Protocol 4.2: 8-well coverglass slides

Preparation of cells and pulse labeling 1) Grow cells expressing SNAP-tag fusion proteins directly in an 8-well chambered coverglass slides to 60–80% confluency. 2) Perform quench and pulse labeling steps as in Basic Protocols 1 or 2, except that labeling volumes of 100 μl are used in 8-well chambered coverglass slides.

Mounting of live cell slides 3) Following labeling and washes, replace medium with LM to a final volume of 300 µl. Seal wells with 100 µl mineral oil. Due to small sample volumes it is critical to prevent evaporation of medium during the time lapse. Sealing of the medium-air interface with mineral oil is an effective method to achieve this. The use of mineral oil is compatible with the use of DIC optics during live cell imaging.

Chapter 2

General considerations regarding the microscope setup. A detailed description of microscope parameters is outside the scope of this unit. Typically, for live cell imaging of mammalian cells, a heated chamber is required to maintain both the cells and the microscope stage at the appropriate temperature. SNAP-dyes can be imaged in principle with any microscope setup as long as appropriate laser lines or filters are used. A variety of fluorescent SNAP-substrates is available from New England Biolabs and others can be found in the existing literature (e.g. Keppler et al, 2004, 2006). See also Critical Parameters and Troubleshooting below. Fluorescent SNAP substrates are based on organic dyes (e.g. TMR). Bleaching is therefore not as big a concern as with autofluorescent proteins such as GFP or RFP. However, due to non-specific labeling (of membranes), background signals are relatively high as compared to autofluorescent proteins. Exposure times, laser strength, neutral density filter settings, and choice of temporal resolution largely depend on signal strength and considerations of cellular phototoxicity.

Fluoresent pulse-chase imaging and quantification

BASIC PROTOCOL 5: AUTOMATED QUANTIFICATION OF SNAP-

TAGGED PROTEIN TURNOVER AT CENTROMERES

In this section we will present a method to perform unbiased fluorescence quantification of diffraction limited spots. We present here a case for centromeres, but this approach applies to any point source signals in living or fixed cells. To this end, we developed an automated algorithm which we name CRaQ (Centromere Recognition and Quantification). This ImageJ based macro detects spots in one channel and subsequently measures the fluorescence intensities in another. This allows for accurate detection and quantification of thousands of spots in a fast, unbiased, and effortless way. In brief, centromeres are recognized and the centroid position is determined. Next, fluorescent intensities are measured for each centromere by placing a small box around the centroid position of the centromere. The peak intensity value within the box is then corrected for local background by subtraction of the minimum pixel value. We have evaluated the accuracy of CRaQ by re-analyzing previously published quantifications that were performed by manually selecting spots (in a reference channel) by eye (Jansen et al, 2007). The results that are obtained by CRaQ are practically identical to the previously published results (Figure 2.6F). In addition, we evaluated the robustness of CRaQ by analyzing replicates samples (because CRaQ is a deterministic algorithm, re-analyzing identical datasets without changing parameters will lead to identical results). We show that quantification of replicate samples by CRaQ leads to a standard error of the mean (SEM) of ~5%, which is likely attributable to biological and/or experimental variation (Figure 2.6G). Thus, CRaQ allows for accurate and reproducible measurement of centromere specific signals.

Chapter 2

Figure 2.6 Centromere Recognition and Quantification (CRaQ). (A–E) Overview of automated steps taken by CRaQ (Basic Protocol 5). (A) DAPI images are thresholded and converted to binary masks. (B) REFERENCE images are filtered and (C) overlaid with the mask to produce a masked reference. (D) This image is again thresholded and spots that fit with the given parameter settings are exported as regions of interest, which are overlaid and measured in the DATA images (E). A blowup is displayed to show the accuracy and frequency of centromere recognition. Note that raw images are in capitals, while processed images are in lowercase letters throughout. (F) CRaQ was used to re-analyze manually selected and quantified centromeres in Jansen et al., 2007. The two methods lead to practically identical results, thus cross-validating each other. (G) Replicate samples were analyzed by CRaQ and standard error of mean (SEM) is plotted as a percentage of the average for four independent experiments, each consisting of four replicates.

Fluoresent pulse-chase imaging and quantification

Because this protocol is performed in an automated fashion, in this section we will first describe the steps that the researcher must take (preparation of the data, CRaQ initiation and parameter settings, etc). Next, we will give an overview of the actual steps that the algorithm goes through for each image (Figure 2.6A–E). This provides users with a good idea of how automated recognition and quantification is performed.

Materials - A standard computer - ImageJ software, including the “Grouped_ZProjector” plugin (both freely available from NIH, http://rsbweb.nih.gov/ij/index.html) - CRaQ plugin for ImageJ (freely available from http://uic.igc.gulbenkian.pt/micro-macros.htm) - Digital images of SNAP-labeled cells, as described in Basic Protocol 1 or 2 after fixation and antibody staining as described in Supporting Protocol 2

Input data preparation (before running CRaQ) 1) Input files should consist of all of the channels of a single frame in one file. CRaQ can use either stacks or projected images as an input. The order of images in a file should be such that the entire image sequence of one channel is followed by the image sequence of the second channel, etc. This as opposed to having all channels for one frame followed by the all the channels for the next frame. Additional channels that are not used during the quantification process can be stored in the same files and will be ignored by CRaQ. 2) Note the order in which the data, reference and DAPI channels are stored in the input files. In principle, only a data channel (the channel that will be quantified) is essential for CRaQ to run. See Critical Parameters and Troubleshooting for reasons and tips for using an independent reference channel. 3) Ideally, the order in which the images should be taken is 1st data, 2nd reference, 3rd DAPI, and any additional channels subsequently. In this way, potential bleaching of the data signal during reference or DAPI channel acquisition will occur only after the data have been collected.

Chapter 2

4) Create a “base folder” with separate subfolders that contain all the images for each condition (e.g. RNAi, replicates, cell types, cell cycle stages, etc). Any images that are located directly in the base folder will not be detected by CRaQ. If all images are to be quantified separately, they can be put into a single subfolder, as the output data file indicates which data points are derived from which image. Only files with extension “.dv” (produced by SoftWorx, Applied Precision) or “.tif” will be recognized by the macro. Thus additional files (log files, etc.) can remain in the base folder without interfering with the macro. When rerunning CRaQ on a previously analyzed data set (e.g. using different settings), make sure to copy the previous data output prior to rerunning, as all files will be overwritten.

Installing and Running CraQ: 5) Copy the CRaQ plugin into your “…/ImageJ/plugins/Analyze” folder and restart ImageJ. Run the algorithm by selecting it from the Plugins>Analyze menu inside ImageJ. 6) In the window that appears you can set the order in which the Data, Reference, and DAPI channels are stored in the input files, as well as the total number of channels. In addition, you can choose to change the standard parameter settings of CRaQ.

Setting the Parameters: The default parameters are those that we have found to work best for most purposes. However, depending on particular experiments, this will not always be the case. What follows is an explanation of each parameter and how and why to change them.

Square size. The size of the box placed around each centromere. Square size 7 means a box of 7x7 pixels. This will generally not change the results much, as only the maximum and minimum pixel values in each box are used. However, make sure that the box is big enough to contain some background pixels, but not too large, as this will make the background signal “less local” and will decrease the number of spots identified due to exclusion of overlapping boxes.

Fluoresent pulse-chase imaging and quantification

Minimum circularity. This measure helps to exclude clustered centromeres. Circularity is a measure of how much the recognized spots resemble a circle, where 1 is a perfect circle and 0 a straight line (the most imperfect circle). Since centromeres appear as diffraction limited spots, they should theoretically be perfectly circular and this measure can be set very close to 1 (most single centromeres actually have a circularity of 1). Because brighter centromeres tend to be less circular, decreasing circularity will allow you to pick up more bright centromeres, but will also increase the chance of picking up doublets, clusters or non-centromeric regions.

Max feret’s diameter. This measure is also made to exclude doublets/clusters and is required because occasionally clusters have a very high circularity. The feret’s diameter is the longest diameter of a spot. Together, stringent circularity and feret parameters are able to exclude most doublets. Increasing the maximum feret’s diameter has a similar effect to decreasing minimum circularity and vice versa.

Min/max centromere size. The minimum and maximum size a centromere can have (in total number of pixels). Basically having a larger maximum size can include both brighter centromeres and more doublets. Again, a lower max centromere size will exclude the last few doublets, but may also exclude some of the brightest (in the reference channel) single centromeres. Increasing the minimum will discard more false positive spots, but also more truly positive (dim) spots.

Threshold offset. This parameter sets the sensitivity of recognition of spots in the thresholded image. Increasing the offset makes the threshold more sensitive to lower signals. This will both increase the number of dim spots (true & false positives), and decrease the number of bright centromeres (false negatives), as these will now appear bigger and potentially less circular.

Chromatic aberration correction. If there is a constant chromatic aberration between reference and data channels, this can be corrected by CRaQ. If the reference channel has spots shifted towards the top/right, then input positive numbers. If the reference channel has spots more to the bottom/left, input negative numbers.

Chapter 2

Data output: All output files will be produced in an output folder inside the base folder. These are the different output files that will be produced by CRaQ:

A single file entitled: “logfile.txt”. This file contains the base directory and parameter settings used. Keep this file or copy info for further reference, replicate experiments or comparison between experiments and parameter settings.

One “*.txt” file for each condition (i.e. subfolder of the base folder). These files contain the actual measurements made by CRaQ with a reference to the corresponding image and centromere spot. These can be directly copied to analysis software such as Excel (Microsoft) or Prism (Graphpad) for further data processing and analysis.

One “*.zip” file for each image. This contains all the recognized spots for that image as ROI lists for ImageJ. To view spots, open the image and the corresponding *.zip file in ImageJ. A “ROI Manager” window will appear, and you can either see all spots by selecting “Show all” or select and display any individual spot.

If stacks where used as input images, a projection of each image is saved. All channels of an image will be saved together in a single *.tif file.

How it works:

1) Convert DAPI to mask (Figure 2.6A). This mask will exclude any spots that are recognized but do not overlap with DNA.

2) Signal enhancing on reference (Figure 2.6B). This allows for more accurate spot recognition.

3) Overlay the mask and the reference (Figure 2.6C). This excludes any non-DAPI signals. 4) Spots that are significantly above background and fall within the restrictions given by the parameter settings are detected and exported as ROI (region of interest) lists (Figure 2.6D). Note that generally <50% of all centromeres are found. However, the recognition of centromeres does not seem to depend on the brightness of centromeres in the reference channel, much the less in

100

Fluoresent pulse-chase imaging and quantification

the data channel. Exclusion of centromeres occurs mostly based on too close proximity to other centromeres. Even though many centromeres are excluded, these measurements will always be orders of magnitude faster and less biased than doing the same by hand.

5) Measure the centromere spots in the data channel (Figure 2.6E). A box of a set size is placed around the center of mass of a ROI. In these boxes, the maximum and minimum values of the Data channel will be measured. The minimum is subtracted from the maximum and that is represented in output. In addition, these boxes are also saved as output. Note that no transformations or background subtractions, etc are made to the Data file before measuring. This means that you are actually measuring raw data. Alterations are only made (but not saved) in the other channels, and are used to efficiently localize centromeres. To exclude overlapping boxes, thus measuring the same spot twice, each box is made black after being measured (value = 0). The macro is programmed to exclude any box containing pixels of value 0. These black boxes are not saved to the data file, so that raw data is preserved. If there is a chromatic aberration, this can be set in the parameters (see above) and boxes are shifted accordingly before measuring. The saved output boxes are the ones that correspond to the reference channel.

101

Chapter 2

SUPPORTING PROTOCOLS

Supporting Protocol 1: Expression of SNAP-fusion proteins We use SNAP source vectors that include a triple HA tag for efficient detection of SNAP-tagged proteins by immunoblotting or immunofluorescence. Maps of SNAP-3XHA, 3XHA-SNAP and 3XHA-SNAPf constructs are in Appendix. Fusion proteins are subsequently subcloned in transient expression vectors or in retroviral constructs (pBABE, see below) for stable expression. For piloting SNAP fusion performance in living cells, we use standard transient transfection methods for obtaining SNAP protein expression. We transfect cells using liposome based methods [e.g. Lipofectamine (Invitrogen) or Fugene (Roche) according to manufacturer’s instructions] and assay protein expression and SNAP activity 48 hours after transfection. For comprehensive experiments, we typically use monoclonal cell lines stably expressing SNAP fusions obtained by retroviral mediated transduction and selection. We use recombinant Moloney murine leukemia (Mo MuLV) retroviral particles for the delivery of SNAP-tagged transgenes into host cell lines (e.g. HeLa or hTERT-RPE). This system is derived from a set of pBABE retroviral vectors (Morgenstern and Land, 1990). Virus particles are assembled in HEK293-GP cells that express the essential Mo MuLV gag and pol genes along with transient delivery of the vesicular stomatitis virus G protein (VSV-G) that results in a pantropic virus with a broad host cell range (Burns et al, 1993; Yee et al, 1994).

Materials - HEK 293-GP cells (Burns et al., 1993) - Trypsin (Cell culture grade, Gibco) - Standard culture medium abbreviated to “CM” (see Reagents & Solutions) - Lipofectamine LTX (Invitrogen) and associated products - Sterile PBS (Cell culture grade)

102

Fluoresent pulse-chase imaging and quantification

- Polybrene (hexadimethrine bromide, Sigma), stock at 8 mg/ml - Selection drugs (e.g. Blasticidin S, puromycin, or hygromycin) - Bovine serum albumin (BSA) - 10 cm standard cell culture dishes; 10 ml syringes; 0.45 μm filters; 6- and 96-well plates - Single cell sorting equipment

Production of viral particles using pBABE based retrovirus 1) Trypsinize and seed one million HEK293-GP cells in a 10 cm dish and culture in CM using standard growth conditions. 2) After 24 hours cells are transfected with 5 μg pBABE + 2 μg pVSV-G using 17.5 μl lipofectamine LTX (Invitrogen), according to manufacturer’s instructions. 3) Incubate cells using standard growth conditions and replace medium with serum containing medium after 4 hours or overnight incubation. 4) Incubate cells for 3 days for viral particle production. 5) Harvest the medium directly from the cells and filter through a 0.45 μm filter using a 10 ml syringe to avoid cellular contaminants. 6) Aliquot (1 ml) and freeze viral stocks at -80°C in or use directly for infections.

Infection of target cells 7) Trypsinize and seed target cells into 2 wells of a 6-well plate, such that cells are at 30–40% confluence at time of infection. 8) Add 8 μg/ml polybrene immediately prior to virus addition. 9) Add 250 μl viral stock from step 6) to one well and 750 μl to the second well. Add CM to a final volume of 1 ml. 10) After 24 hours of infection, replace medium with CM. 11) Let cells proliferate until they reach confluency (at least 24 hours later).

103

Chapter 2

12) Trypsinize cells, combine the 2 wells, and plate in a 10 cm dish containing the appropriate drug selection. We use pBABE vectors with Blasticidine S (Blast), puromycin or hygromycin resistance cassettes. E.g. HeLa cell clones are drug selected with 5 μg/ml Blast, 5 μg/ml puromycin, or 250 μg/ml hygromycin. 13) Select cells until colonies are visible by the naked eye (10–20 days).

14) Trypsinize and pool the clones and amplify for single cell sorting. 15) To isolate monoclonal lines, cells are washed in sterile PBS, resuspended in sterile PBS + 5% BSA and sorted by standard flow sorting (using scatter to identify single cells) into 96-well plates containing conditioned culture medium (see Reagents & Solutions).

Supporting Protocol 2: Cell fixation and immuno-fluorescence In this section we describe a general method for fixation (of SNAP pulse labeled cells), immunofluorescence detection and DAPI staining. Immunofluorescence for detection of proteins unrelated to SNAP but localized at the same subcellular location allows for an independent measure to be used in image quantification using CraQ (see Basic Protocol 5, and Commentary). Please note that many other equally effective protocols for this purpose exist. As this is a general protocol we do not comment on specific antibody conditions and concentrations as this will need to be determined for each specific application.

Materials - 1X PBS - 4% Paraformaldehyde in 1X PBS, referred to as “PFA” - 0.1 M Tris-HCl, pH 7.5 - PBS-TX (1X PBS + 0.1% Triton X-100) - DAPI solution (see Reagents & Solutions) - MOWIOL (see Reagents & Solutions)

104

Fluoresent pulse-chase imaging and quantification

- Nail polish - Humid dark box: Can be made from an empty micropipette tip-box filled with a small layer of water, a thick sponge covered by a glass plate. Any transparent surface of the box is covered with aluminum foil - Parafilm - Clean, sterile, poly-lysine coated coverslips (12 mm ø; Thickness 1.5) - Fine forceps with sharp pointed ends - IF blocking buffer (see Reagents & Solutions) - Standard glass slides

Cell fixation 1) Grow and SNAP pulse label cells on glass coverslips in 24-well plates as described in Basic Protocols 1–3. 2) Wash cells twice in 1 ml PBS, pre-heated to 37°C. 3) Fix cells for exactly 10 minutes at room temperature in 500 μl PFA, pre-heated to 37°C. 4) Aspirate PFA and quench by adding 1 ml of 0.1 M Tris, pH 7.5 for 5 minutes. Cells can be stored at this point for up to a few days in PBS at 4°C, or up to 1 month in PBS + 0.04% NaN3 at 4°C.

Antibody detection 5) Permeabilize cells by washing twice in 1 ml of PBS-TX for 5 minutes. 6) Carefully lift coverslips with a forceps and move to a parafilm covered glass plate in humid dark box. Humid dark boxes prevent coverslips from drying and fluorescent dyes from photo-bleaching. Parafilm is a convenient receptacle for coverslips as its hydrophobic surface allows the application of small volumes to the coverslips without spilling over to neighboring coverslips. 7) Block cells for 30 minutes, 37°C in blocking buffer. Use 75 μl per coverslip. 8) Incubate cells with primary antibody diluted in blocking buffer for 60 minutes, 37°C. Use 30 μl per coverslip.

105

Chapter 2

9) Wash coverslips in 75 μl PBS-TX 3 times for 5 minutes at room temperature. 10) Incubate secondary fluorescent antibody diluted in blocking buffer for 45 minutes, 37°C. Use 30 μl per coverslip. Centrifuge diluted fluorescent antibodies for 5 minutes at maximum speed (~16.000 g) to deplete any fluorescent aggregates that may interfere with fluorescent imaging. Use supernatant for staining. 11) Wash coverslips in 75 μl PBS-TX 3 times for 5 minutes at room temperature. 12) Incubate cells in 50 μl DAPI (500 ng/ml final concentration) for 5 minutes at room temperature. 13) Replace DAPI solution with PBS. 14) Carefully pick up coverslips with a forceps, remove excess liquid by aspiration and/or filter paper, and mount on a glass slide (cells facing down) in ~5 μl Mowiol. Allow the Mowiol to solidify overnight at 4°C in the dark. 15) Seal coverslips using nail polish to avoid air contact during the imaging process.

Reagents & Solutions

BTP (bromothenylpteridine): A 2 mM stock is prepared by dissolving 100 nmol lyophilized SNAP-Cell Block (New England Biolabs, cat# S916S) in 50 μl DMSO (sterile). Shake for 10 minutes in an eppendorf shaker at maximum speed to dissolve. Store for 1 month at -20°C or aliquot and store at -80°C.

Conditioned culture medium (for HeLa): 50% fresh CM + 50% CM harvested from HeLa cultures in log growth phase, 0.45 μM filtered.

DAPI(4',6-Diamidino-2-phenylindole dihydrochloride): A 1 mg/ml stock is prepared in water. Store at -20°C. Dilute 2000 fold in PBS for working solution.

IF blocking buffer: 2% fetal bovine serum, 2% BSA, 0.1% Triton X-100, 0.04% NaN3, in 1X PBS.

106

Fluoresent pulse-chase imaging and quantification

Live imaging medium: phenol red-free, CO2-independent medium (e.g. DME or Leibovitz’s L-15) supplemented with 10% fetal bovine serum, 2 mM Glutamine (all from Gibco).

MOWIOL: Ingredients: Mowiol 4-88 (Calbiochem), Glycerol, DABCO (1,4- diazabicyclo[2.2.2]octane, Sigma).

1) Mix Mowiol 4-88 and glycerol in a 2:5 ratio (w/w).

2) Add 0.714 ml water/gram of Mowiol/glycerol mixture and stir overnight at room temperature.

3) Add 2 volumes of 0.2 M Tris (pH 8.5) for each volume of water added and heat at 50°C for 10 minutes with occasional mixing.

4) Centrifuge at 5.000 g for 15 minutes and remove debris.

5) Add DABCO to 2.4% and mix slowly.

6) Centrifuge at 5.000 g for 15 minutes and remove debris.

7) Aliquot and store at -20°C.

Standard culture medium (for HeLa and HEK293-GP): DMEM + 10% NCS (newborn calf serum), 100 U/ml penicillin, 100 µg of streptomycin, 2 mM Glutamine (all from Gibco). Other cell types may require different growth media.

TMR-Star: A 200 μM stock is prepared by dissolving 30 nmol lyophilized SNAP-Cell TMR-Star (New England Biolabs, cat # S9105S) in 150 μl DMSO (sterile). Shake for 10 minutes in an eppendorf shaker at maximum speed to dissolve. Store for 1 month at -20°C or aliquot and store at -80°C.

VALAP: Vaseline:lanolin:paraffin 1:1:1 (w/w).

1) Heat paraffin to 50°C in a large beaker in a water bath.

2) When paraffin is melted mix in vaseline and lanolin.

3) Stir to mix and aliquot, store at 4°C.

4) Heat to 50°C prior to use

107

Chapter 2

BACKGROUND INFORMATION

Historical background

The SNAP-tag is a modified version of human O6-alkylguanine-DNA alkyltransferase (hAGT). Endogenous hAGT is a DNA repair enzyme that removes a broad range of alkyl adducts from the O6 position of guanines in DNA. It acts as a suicide enzyme that catalyzes a covalent binding reaction between itself and the alkyl group that is removed from guanines, thereby restoring DNA integrity but inactivating its own catalytic activity (Pegg, 2000). SNAP, the modified form of hATG, has lost its affinity to DNA but efficiently reacts with soluble O6-benzylguanine (BG), of which the benzyl moiety is readily transferred to the SNAP protein (Figure 2.1; Juillerat et al, 2003; Keppler et al, 2003). The benzyl rings in BG can be coupled to a large variety of molecules (Keppler et al., 2003, 2004, 2006) that include fluorescent moieties as well as non-fluorescent ones (a selection of SNAP substrates is presented in Table 2.2).

General considerations for SNAP-based protein turnover assays A number of techniques exist to analyze protein turnover (Table 2.1). A common approach to in vivo protein turnover is the use of fluorescence recovery after photobleaching (FRAP). In this method, autofluorescent proteins are fused to proteins of interest that localize to a specific subcellular location. Local irreversible bleaching followed by repopulation of a bleached area by unbleached molecules from neighboring regions provides information of the local rate of protein turnover (Lippincott-Schwartz et al., 2001; and references therein). A reciprocal technique utilizes inducible fluorescent proteins, which can be activated by a focused laser, which allows tracking of a specific pool of photo-activated protein (Lukyanov et al., 2005; and references therein). While widely applied, FRAP and photo-activation experiments suffer from three specific drawbacks. 1) Measurement of fluorescence recovery or photoactivation typically requires continued imaging of cells, leading to problems such as photobleaching and

108

Fluoresent pulse-chase imaging and quantification

phototoxicity, thereby restricting the time in which turnover can be measured to a few hours at most. This precludes measurement of long-term turnover rates. 2) A focused laser is required to bleach or activate fluorescence preventing the analysis of large numbers of cells simultaneously. Lastly 3), the turnover rates using FRAP and photoactivation are a product of the “on” and “off” rates of a protein which cannot be assessed separately. SNAP-based pulse labeling differs from traditional FRAP experiments in that a fluorescent pool is created by pulse labeling with the addition of an external dye to the culture medium. Therefore, first and foremost, imaging and quantification of its fluorescence can commence at any time following labeling (hours, days after pulse labeling). This allows analysis of protein turnover at very long time scales. Secondly, because the entire cell population is treated with the dye in bulk, large numbers of cells are available for simultaneous imaging and analysis. Lastly, the combination of serial dark and fluorescent pulse labeling strategies (“pulse-chase” and “quench-chase-pulse”) allows for the separate determination of turnover of pre-existing pools (off-rates) and turnover of newly synthesized pools of protein (on-rates) (Figure 2.2 and 2.3). Several other methods capitalize on similar advantages such as other self-labeling or destructive enzymes (see Table 2.1). We would like to highlight one recently developed method named “Recombination Induced Tag Exchange” (RITE), which allows for similar applications as SNAP- tagging while using a fundamentally different strategy (Verzijlbergen et al., 2010). It uses recombination induced switching of expression of differentially tagged versions of the same gene. This allows for the simultaneous visualization, tracking, and/or analysis of the original (pre- switched) pool as well as a nascent one (Radman-Livaja et al., 2011). However, this method relies on tight control over induction of Cre-mediated recombination which is difficult to achieve in some systems (most metazoan cell lines).

109

Chapter 2

The advantage of assessing long-term dynamics also implies a major disadvantage of SNAP-based pulse labeling. Labeling and washing steps require approximately 1 hour rendering this method inappropriate to assess protein dynamics at short timescales (seconds to minutes), as pulse labeled proteins will have reached their steady state equilibrium before imaging can determine their dynamics. However, improvements are currently being made to both the SNAP-enzyme and the fluorescent substrates thereof, which would in principle allow labeling steps of 5 minutes without the need for any washes (see below and Sun et al., 2011).

Critical Parameters and Troubleshooting

SNAP labeling: Choice of substrate One very important parameter during the pulse-chase and quench- chase-pulse procedure in living cells is the choice of SNAP-substrate used. The limiting characteristic seems to be the ability of substrates to efficiently pass the cell membrane, as many substrates tend to strongly label the cell membrane while barely labeling intracellular SNAP proteins. In our experience, non-fluorescent benzylguanine (BG) or bromothenylpteridine (BTP) enter cells efficiently. However, addition of (bulky) side groups may impede the cell permeability. Thus, although there is a large variety of fluorescent substrates for intracellular labeling, the efficiency at which these enter the cells is not always the same. For this reason, using the optimal fluorophore for the particular microscopy and filter setup used has to be balanced with the cell permeability of this substrate. We generally obtain the best results with SNAP-Cell TMR-Star (New England Biolabs). It is for this reason that we prefer to use BTP for quench steps in the quench-chase-pulse procedures rather than using multiple different fluorescent substrates (see Basic Protocol 2), because complete labeling of the initial pool is essential to ensure visualization of the subsequent newly synthesized pool only.

110

Fluoresent pulse-chase imaging and quantification

Of special interest are a group of recently developed SNAP-substrates that display a dramatic increase in fluorescence after reaction with SNAP. These so called ‘dark-dyes’ are either quenched by guanine itself (Stöhr et al, 2010), or by a side-group fused the guanine moiety of benzylguanine (Komatsu et al, 2011; Sun et al, 2011). These dark-dyes provide a number of advantages over traditional fluorescent SNAP-substrates, most importantly leading to highly reduced (unspecific) background fluorescence. Other advantages include wash-free labeling, faster downstream applications (due to shorter wash steps), and potentially more efficient live cell imaging.

Table 2.2 Selection of SNAP-Substrates

Type of SNAP-substrate labels Examples Specifications References Used to block (quench) pre-existing pools of SNAP protein to Quenchers BG, BTP NEB, cat# S9106S prevent their detection in subsequent labeling steps Fluorescent substrates NEB, cat# S9105S; Standard fluorophores TMR-Star, BG-505 Used for most microscopy based pulse-labeling techniques. S9105S DRBGFL, CBG-549- Used for reduced backgrounds, which allows for wash-free Komatsu et al, 2011; Sun Dark dyes (induced quenching)1 QSY7 labeling and this a faster labeling procedure et al, 2011 Used for reduced backgrounds, which allows for wash-free Dark dyes (natural quenching)2 BG-MR121 Stöhr et al, 2010 labeling and this a faster labeling procedure Substrate that becomes fluorescent after UV-activation (similar to Caged dyes BG-CMNB-caged Campos et al, 2011 photo-activatable fluorescent proteins)

Superresolution dyes (double-dyes)3 BG-Cy3-Cy5 Used for PALM/STORM of SNAP labeled proteins Dellagiacoma et al, 2011

Protein purification substrates BG-Beads (agarose or NEB, cat# S9144S; Beads Used for biochemical purification of SNAP labeled proteins magnetic) S9145S Used for biochemical purification of SNAP labeled proteins using Biotinylation BG-Biotin NEB, cat# S9110S streptavidin beads Other types of Dyes Drugs BG-THL Used to deliver drugs to subcellular compartments Yang et al, 2011 Used to create self-assembling-monolayers (SAM) of SNAP-labeled Thiol BG-Thiol Kwok et al, 2011 proteins More …. 1: Fluorescent BG substrates are labeled to a second sidegroup that quenches the fluorescence by FRET. After protein labeling, the two sidegroups are spatially removed and leading to fluorescence activation. 2: Idem above, except that fluorophores are used that are naturally quenched by guananine, alleviating the need for adding a second (bulky) sidechain. 3: One reason to use these dyes for superresolution microscopy, is their increased brightness as compared to FPs; a limiting factor for these techniques.

SNAP labeling: enzyme variant Variants of SNAP have been derived by in vitro evolution. One example is the “CLIP-tag”, which is derived from SNAP and reacts specifically with a variant substrate, O2-benzylcytosine (Gautier et al, 2008). Tagging of two different proteins by SNAP and CLIP allows for simultaneous labeling of two different proteins in different colors (Gautier et al, 2008; Prendergast et al,

111

Chapter 2

2011). More recently, variants of SNAP and CLIP named SNAPf and CLIPf have been developed that present faster reaction kinetics (Pellett et al., 2011; Sun et al., 2011). We evaluated SNAPf and CLIPf performance in vivo by side-by-side comparison with SNAP and CLIP, using the intracellular protein CENP-A as a labeling target (data not shown and Figure 2.7A). While CLIPf showed only a modest improved over CLIP (not shown), SNAPf performed ~3-5 fold better across different concentrations of substrates and incubation times (Figure 2.7B). The use of SNAPf therefore allows for shorter labeling times and lower dye concentrations to yield the same signal intensity. A reduced background staining while retaining specific signals will potentially improve live cell capabilities significantly.

Figure 2.7 Evaluation of SNAPf-tag performance. (A) HeLa cells were transfected with either CENP-A-SNAP or CENP-A-SNAPf fusion proteins, and labeled with TMR-Star at different concentrations and incubation times, as indicated in the figure. Representative images of cells are shown with TMR-Star signals in green and DAPI (DNA) in blue. (B) TMR-Star and HA fluorescence intensity were determined using CRaQ (Basic Protocol 5) and TMR-Star/HA ratios are used as a measure of SNAP or SNAPf activity. Results are plotted as fold difference, normalized to signals obtained with SNAP after incubation with 2μM TMR-Star for 15 minutes (standard conditions). SNAPf outperforms SNAP in all conditions tested (3-5 fold).

112

Fluoresent pulse-chase imaging and quantification

SNAP labeling: Dye concentration, wash steps, and pool size: Depending on the cell type, expression levels, application, SNAP- substrate, etc., it is necessary to optimize substrate concentrations. Higher concentrations are not always preferable, as this can result in higher background levels and thus poorer signal-to-noise ratios. For CENP-A- SNAP we generally use a concentration of 2 μM TMR-Star as a compromise between signal-to-noise and cost (although we have found that using higher concentrations up to 5 μM increases the signal-to-noise ratio of labeling). For other purposes it may be necessary to use saturating concentrations, or conversely, it may be sufficient to use lower concentrations. We found that extensive washes after labeling (2 quick washes, an extended wash for 30 minutes at 37°C, and two additional quick washes) help to remove excess unbound substrates. This results in dramatically decreased background fluorescence after pulse labeling. During quench labeling these wash steps ensure that nascent protein synthesized during the chase is not immediately quenched which would lower the effective poolside of the new pool and specific signals in subsequent fluorescent labeling.

SNAP labeling: Chase time A critical aspect of a successful quench-chase-pulse experiment is the chase time that the cells are given to produce new protein. Although this is largely determined by the experimental conditions, one would typically seek conditions that maximizes the time for protein synthesis prior to labeling.

Imaging and quantification: Microscope For imaging of SNAP-derived and immunofluorescent signals any high resolution microscope can be used.

Imaging and quantification: Reference marker Special care should be taken to choose the marker used as a reference for spot detection. A number of options exist. 1) The signals that require quantification can be used simultaneously as a reference of spots to

113

Chapter 2

measure. However, this solution suffers from the drawback that spots with very low signals will not be detected and that the detection will be inherently biased, e.g. towards bright spots. A better option is to 2) use an antibody against SNAP (available from NEB) or HA (in case an HA-tag is incorporated in the fusion protein; see Appendix), which will detect the entire pool of SNAP tagged protein independent of time sliced signals (see e.g. Figure 2.3B and 2.4B). However, if the protein of interest forms aggregates or has multiple possible localization patterns, these will also be quantified by automated methods such as CRaQ. Thus, whenever possible, we prefer to use 3) antibodies (or autofluorescent fusion proteins) against an independent marker for the subcellular structure (e.g. centromeres by CENP-C or CENP-T; see Figures 2.2B, 2.4C, and Silva et al., 2012). This allows for specific and unbiased detection of spots. Naturally, clean references will lead to the most accurate quantifications and using antibodies that are highly specific and give little background staining will increase the quality of the data. In addition, when measuring proteins that reside inside the nucleus, an additional marker such as DAPI can be used to further exclude unspecific reference signals outside of the nucleus.

Imaging and quantification: CRaQ There are a number of critical aspects to take into account when using CRaQ. First and foremost, as this is an automated algorithm, the results should be validated by the user. After initiating the macro one can follow the screen shots that pop up to monitor which spots are recognized as reference points. If the macro is poorly tuned it may already be obvious at this early stage (e.g. recognition of the entire image). Next, after completion of the macro, data output files should be checked to validate whether the correct spots are detected (e.g. by doing this manually for a small, random subset of pictures and comparing this to the spots recognized automatically). If automated spot recognition is not accurate, the parameters should be optimized as described in Basic Protocol 5. Parameter optimization and testing is best done on a small subset of pictures to save time.

114

Fluoresent pulse-chase imaging and quantification

Evidently, using a high-end microscope with appropriate filter combinations and a sensitive camera is instrumental to obtain good fluorescence quantifications. In addition, potential chromatic aberrations between reference and data channels must be corrected for in the quantification (this can also be set as a parameter of CRaQ). One way to determine the chromatic aberration is to use beads that are fluorescent in the two channels used and determine whether and by how many pixels the center of mass is shifted between the colors. Finally, although inorganic dyes are generally very photostable, we have observed that imaging TMR-Star labeled cells as soon as possible after fixation (1–2 days) facilitates obtaining the most optimal signals.

Anticipated Results

SNAP-labeling Because SNAP substrates are added to the culture medium, virtually all SNAP-expressing cells are labeled in any given experiment. The ability to detect SNAP-tagged proteins depends on the expression level of the protein and the efficiency of SNAP substrate entry into the cells. In quench-chase- pulse experiments, the chase time during which cells synthesize and assemble new protein will determine which cells will become labeled during the second, fluorescent labeling step. In the case of CENP-A-SNAP, the appearance of centromeric signals will largely depend on cell cycle position (Figure 2.4B and 2.5). The expected results for other proteins will depend on the biological properties of the protein of interest. Many SNAP-substrates have difficulty passing through the cell membrane. For this reason it is normal to see relatively high background fluorescence, as compared to e.g. antibody or fluorescent protein detection. We try to minimize this background fluorescence by extensive washes of the fluorescent substrate after labeling is completed (steps e.g. 5–6 of Basic Protocol 1).

115

Chapter 2

Image quantification Using CRaQ we generally have very low false-positive rates, where off- target sites or doublets comprise ≪1% of all spots detected. In addition, this macro is generally able to detect a good proportion of the correct spots to be analyzed (>50%), although this largely depends on the quality of the reference signal. Using a generic present day desktop computer we can readily collect hundreds to thousands of data points in 15-20 minutes. The rate limiting steps are testing parameter settings (although generic parameter settings usually work very well) and analyzing the data generated.

Time considerations The time that is required for the experiments outlined above is highly variable and depends on the precise setup of the experiment. Quench and pulse labeling each take about 1–1.5h to perform. However, the chase time can be anywhere between a few hours and a few days. Furthermore, adding sequential steps, such as synchronization and/or RNAi procedures can increase the total time of the experiment to more than a week. Fixation and antibody labeling requires approximately 4–5 hours to perform and cells are preferentially imaged on the following day. Imaging requires roughly 30 minutes per coverslip used, although this again depends on many factors, including the microscopy system, signal intensity (i.e. exposure times needed), cell density (i.e. number of images required), sample thickness (i.e. number of slices required), etc. Running CRaQ generally takes no more than 20 minutes, even for large datasets, and validation of the output takes about the same time. Finally, processing of the output data into comprehensible tables/graphs takes about 30 minutes to 1 hour, depending on the size of the dataset.

116

Fluoresent pulse-chase imaging and quantification

REFERENCES

Bojkowska K, Santoni de Sio F, Barde I, Offner S, Verp S, Heinis C, Johnsson K & Trono D (2011) Measuring in vivo protein half-life. Chem. Biol. 18: 805–815 Burns JC, Friedmann T, Driever W, Burrascano M & Yee JK (1993) Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc. Natl. Acad. Sci. U. S. A. 90: 8033–8037 Campos C, Kamiya M, Banala S, Johnsson K & González-Gaitán M (2011) Labelling cell structures and tracking cell lineage in zebrafish using SNAP-tag. Dev. Dyn. Off. Publ. Am. Assoc. Anat. 240: 820–827 Damoiseaux R, Keppler A & Johnsson K (2001) Synthesis and applications of chemical probes for human O6-alkylguanine-DNA alkyltransferase. Chembiochem Eur. J. Chem. Biol. 2: 285–287 Dunleavy EM, Almouzni G & Karpen GH (2011) H3.3 is deposited at centromeres in S phase as a placeholder for newly assembled CENP-A in G(1) phase. Nucl. Austin Tex 2: 146–157 Erhardt S, Mellone BG, Betts CM, Zhang W, Karpen GH & Straight AF (2008) Genome-wide analysis reveals a cell cycle-dependent mechanism controlling centromere propagation. J. Cell Biol. 183: 805–818 Gautier A, Juillerat A, Heinis C, Corrêa IR, Kindermann M, Beaufils F & Johnsson K (2008) An engineered protein tag for multiprotein labeling in living cells. Chem. Biol. 15: 128–136 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805 Juillerat A, Gronemeyer T, Keppler A, Gendreizig S, Pick H, Vogel H & Johnsson K (2003) Directed evolution of O6-alkylguanine-DNA alkyltransferase for efficient labeling of fusion proteins with small molecules in vivo. Chem. Biol. 10: 313–317 Keppler A, Arrivoli C, Sironi L & Ellenberg J (2006) Fluorophores for live cell imaging of AGT fusion proteins across the visible spectrum. BioTechniques 41: 167–170, 172, 174–175 Keppler A, Gendreizig S, Gronemeyer T, Pick H, Vogel H & Johnsson K (2003) A general method for the covalent labeling of fusion proteins with small molecules in vivo. Nat. Biotechnol. 21: 86–89 Keppler A, Pick H, Arrivoli C, Vogel H & Johnsson K (2004) Labeling of fusion proteins with synthetic fluorophores in live cells. Proc. Natl. Acad. Sci. U. S. A. 101: 9955–9959 Komatsu T, Johnsson K, Okuno H, Bito H, Inoue T, Nagano T & Urano Y (2011) Real-time measurements of protein dynamics using fluorescence activation- coupled protein labeling method. J. Am. Chem. Soc. 133: 6745–6751 Maduzia LL, Yu E & Zhang Y (2010) Caenorhabditis elegans Galectins LEC-6 and LEC-10 Interact with Similar Glycoconjugates in the Intestine. J. Biol. Chem. 286: 4371–4381

117

Chapter 2

McMurray MA & Thorner J (2008) Septin stability and recycling during dynamic structural transitions in cell division and development. Curr. Biol. CB 18: 1203– 1208 O’Hare HM, Johnsson K & Gautier A (2007) Chemical probes shed light on protein function. Curr. Opin. Struct. Biol. 17: 488–494 Prendergast L, van Vuuren C, Kaczmarczyk A, Doering V, Hellwig D, Quinn N, Hoischen C, Diekmann S & Sullivan KF (2011) Premitotic assembly of human CENPs -T and -W switches centromeric chromatin to a mitotic state. PLoS Biol. 9: e1001082 Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63 Stöhr K, Siegberg D, Ehrhard T, Lymperopoulos K, Öz S, Schulmeister S, Pfeifer AC, Bachmann J, Klingmüller U, Sourjik V & Herten D-P (2010) Quenched substrates for live-cell labeling of SNAP-tagged fusion proteins with improved fluorescent background. Anal. Chem. 82: 8186–8193 Sullivan KF, Hechenberger M & Masri K (1994) Human CENP-A contains a histone H3 related histone fold domain that is required for targeting to the centromere. J. Cell Biol. 127: 581–592 Sun X, Zhang A, Baker B, Sun L, Howard A, Buswell J, Maurel D, Masharina A, Johnsson K, Noren CJ, Xu M-Q & Corrêa IR (2011) Development of SNAP-Tag Fluorogenic Probes for Wash-Free Fluorescence Imaging. Chembiochem 12: 2217–2226 Waterman-Storer CM & Salmon ED (1997) Actomyosin-based Retrograde Flow of Microtubules in the Lamella of Migrating Epithelial Cells Influences Microtubule Dynamic Instability and Turnover and Is Associated with Microtubule Breakage and Treadmilling. J. Cell Biol. 139: 417 –434 Yee JK, Miyanohara A, LaPorte P, Bouic K, Burns JC & Friedmann T (1994) A general method for the generation of high-titer, pantropic retroviral vectors: highly efficient infection of primary hepatocytes. Proc. Natl. Acad. Sci. 91: 9564 – 9568

Author contributions All experiments and protocols described have been executed, designed and/or optimized by me and LETJ, with the following exceptions: MGR performed experiments shown in Figure 2.7; NM created an initial macro that I further developed into CRaQ (Basic Protocol 5). The manuscript for this chapter was drafted and revised with help of LETJ and constructive suggestions by all authors.

118

Fluoresent pulse-chase imaging and quantification

Acknowledgements We thank Mariana Silva for valuable comments on the manuscript. DLB and MGR are supported by the Fundação para a Ciência e a Tecnologia (FCT) doctoral fellowships SFRH/BD/74284/2010 and SFRH/BD/33567/2008, respectively. This work is supported by the Fundação Calouste Gulbenkian, FCT grants BIA-BCM/100557/2008 and BIA-PRO/100537/2008, the European Commission FP7 programme, and an EMBO installation grant to LETJ.

119

Chapter 2 - Appendix

Appendix: Maps of SNAP- and SNAPf-tags

120

SNAP constructs

121

Chapter 2 - Appendix

122

CHAPTER 3

Assembly in G1 phase and Long-Term Stability are Unique Intrinsic Features of CENP-A Nucleosomes

Dani L. Bodor1, Luis P. Valente1, João F. Mata1, Ben E. Black2, and Lars E.T. Jansen1

1 Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal.

2 University of Pennsylvania, Philadelphia, PA 19104-6059, USA.

NB: This chapter is a near literal transcription of Mol. Biol. Cell April 1, 2013 vol. 24 no. 7 pp. 923-932. Noteworthy is the addition of unpublished results for depletion of M18BP1 in Figure 3.5 and accompanying text.

NB2: Unpublished results concerning depletion of CENP-C have been added in an appendix to this chapter.

ABSTRACT

Centromeres are the site of kinetochore formation during mitosis. CENP-A, the centromere specific histone H3 variant is essential for the epigenetic maintenance of centromere position. Previously, we have shown that newly synthesized CENP-A is targeted to centromeres exclusively during early G1 phase and is subsequently maintained across mitotic divisions. Using SNAP-based fluorescent pulse labeling, we now demonstrate that cell cycle restricted chromatin assembly at centromeres is unique to CENP-A nucleosomes and does not involve assembly of other H3 variants. Strikingly, stable retention is restricted to the CENP-A/H4 core of the nucleosome which we find to outlast general chromatin across several cell divisions. We further show that cell cycle timing of CENP-A assembly is independent of centromeric DNA sequences, but instead is mediated by the CENP-A targeting domain. Unexpectedly, this domain also induces stable transmission of centromeric nucleosomes, independent of the CENP-A deposition factor HJURP. This demonstrates that intrinsic properties of the CENP-A protein direct its cell cycle restricted assembly and induces quantitative mitotic transmission of the CENP-A/H4 nucleosome core ensuring long-term stability and epigenetic maintenance of centromere position.

Defining the heritable centromere core

INTRODUCTION

Centromeres are the chromosomal loci for kinetochore formation during mitosis and thus form the site of interaction between DNA and the mitotic spindle (Cheeseman & Desai, 2008). As a result, centromeres are essential for proper chromosome segregation and prevention of aneuploidy. Although human centromeres are usually assembled on alpha-satellite (alphoid) DNA, specific sequences are neither necessary nor sufficient to stably maintain a centromere. Evidence for this comes primarily from the existence of neocentromeres, where a specific centromere has repositioned to, and is stably maintained upon a naive locus that differs in DNA sequence context and is not normally associated with centromere activity (Amor et al, 2004; Marshall et al, 2008; Voullaire et al, 1993). This has led to the proposal that centromeres are specified in a sequence independent, epigenetic manner. While the vast majority of genomic DNA is packed by the canonical histones (H2A, H2B, H3.1, and H4), specific histone H3 variants package subsets of the genome. Among these, the H3.3 variant is mainly found at sites of active transcription (Ahmad & Henikoff, 2002), while centromere protein A (CENP-A) replaces H3.1 specifically in centromeric nucleosomes (Yoda et al, 2000; Foltz et al, 2006), and is required for the localization of nearly all other centromeric proteins (Foltz et al, 2006; Liu et al, 2006). Consistent with a role in epigenetic maintenance of centromere identity, CENP-A is a stable component of centromeric chromatin (Pearson et al, 2004; Schuh et al, 2007; Hemmerich et al, 2008) and is transmitted at centromeres during successive cell divisions (Jansen et al, 2007). In addition, it was recently shown in Drosophila S2 cells that targeting of CENP-ACID to ectopic loci for a short period of time is sufficient to initiate a sustainable epigenetic feedback loop, which recruits and maintains functional kinetochores for several subsequent cell division cycles (Mendiburo et al, 2011). Together, these findings strongly suggest that CENP-A plays a key role in epigenetic memory of centromere position and function.

127

Chapter 3

Consistent with a critical role in centromere specification, assembly of CENP-A is tightly regulated and restricted to a specific stage in the cell cycle in order to maintain proper CENP-A levels. In metazoans, assembly of CENP-A is uncoupled from S phase and dependent on passage through mitosis (Jansen et al, 2007; Schuh et al, 2007; Moree et al, 2011; Mellone et al, 2011; Bernad et al, 2011). We have previously shown that G1 phase restricted assembly of CENP-A in human and chicken cells is directly coupled to cell cycle progression as a result of inhibitory action of Cdk1 and Cdk2 in S phase, G2, and mitosis (Silva et al, 2012). While we have a basic understanding of the mechanism of cell cycle coupling of centromeric chromatin assembly, how this assembly is restricted to centromeres and how CENP-A chromatin is stably maintained is unclear. In this study we determine whether centromeric chromatin assembly during G1 represents a general phase of nucleosome turnover, or whether this is a unique feature of CENP-A nucleosomes. In addition, we determined whether the previously reported stable maintenance of CENP-A (Jansen et al, 2007) is a feature of centromeric chromatin in general, or whether this is an intrinsic property of CENP-A-containing nucleosomes or even subnucleosomal complexes thereof. Using SNAP-tag based fluorescent pulse labeling (Jansen et al, 2007; Bodor et al, 2012; Silva et al, 2012), we made the striking finding that CENP-A nucleosome assembly is the major form of nascent chromatin assembly in G1. This results in the formation of nucleosomes with a uniquely high in vivo stability of the CENP-A/H4 nucleosome core, a property induced in cis by residues encoded by the CENP-A protein.

128

Defining the heritable centromere core

RESULTS

G1 phase histone assembly is restricted to CENP-A and H4. We have previously used SNAP labeling to demonstrate that incorporation of nascent CENP-A is restricted to a brief window during early G1 phase (Jansen et al, 2007). SNAP is a self-labeling suicide enzyme that covalently and irreversibly reacts with benzylguanine or (fluorescent) derivatives thereof (Keppler et al, 2003, 2004). Sequential SNAP labeling steps allow for differential analysis of protein pools synthesized at distinct periods of time (Bodor et al, 2012). Timing of CENP-A assembly can be a consequence of an intrinsic property of this particular protein, or result from a general wave of histone exchange at centromeres during G1. To determine whether a G1 assembly pathway exists for other histones, we used cells stably expressing SNAP-tagged versions of a variety of histone proteins. These include two other histone H3 family members, the canonical H3.1 and the replacement variant H3.3, as well as H4, the direct binding partner of all H3 variants, and H2B, a member of the more dynamic H2A/H2B histone sub-complex (Kimura & Cook, 2001). Direct pulse labeling of the total (steady state) pool of SNAP-tagged histone showed signal in all cells, as expected (Figure 3.S1A–B). To determine the pattern of assembly of nascent histones, we performed SNAP-based quench-chase-pulse experiments [Figure 3.1A and (Bodor et al, 2012)]. To visualize stable chromatin assembly of nascent protein we pre-extracted cells prior to fixation and imaging (Ray-Gallet et al, 2011). As anticipated, due to cell cycle regulated assembly, nascent CENP-A-SNAP is found at centromeres in only a subset of cells [Figure 3.1B and (Jansen et al, 2007)]. Similarly, nascent H3.1-SNAP is found in a subset of the population (Figure 3.1B), owing to its strict replication-coupled assembly (Ray-Gallet et al, 2011). Interestingly, distinct sub-nuclear patterns of H3.1-SNAP staining can be observed, indicative of differential patterns of DNA-synthesis throughout S phase [Figure 3.1B and (Ray-Gallet et al, 2011)]. These results emphasize the power of SNAP-based

129

Chapter 3

pulse-chase assays as they reveal strikingly different patterns of localization of the same protein, but synthesized and deposited into chromatin at different times during the cell cycle. Our H3.1-SNAP cell line therefore provides a powerful and accessible tool for marking S phase progression without the need for an inducible expression system. In contrast, H3.3 (Ray- Gallet et al, 2002, 2011) and H2B (Kimura & Cook, 2001) are assembled throughout the cell cycle and, consequently, nascent protein can be observed in all cells analyzed (Figure 3.1B). Intriguingly, nascent H4-SNAP reveals a unique differential pattern of assembly, different from all other histone proteins analyzed. While all cells display assembly throughout chromatin, consistent with a role as partner of H3.1 in S phase or H3.3 throughout the cell cycle, preferential assembly at discrete foci is observed in a subset of cells (Figure 3.1B). This pool of nascent H4 specifically colocalizes with centromeres, as marked by CENP-C (Figure 3.1B, enlargement), suggesting that histone H4 has a distinct phase of centromeric assembly.

CENP-A and H4 are co-assembled during G1 phase. Prenucleosomal CENP-A forms a complex with H4 and HJURP, the CENP-A specific histone chaperone (Foltz et al, 2009; Dunleavy et al, 2009; Hu et al, 2011; Shuaib et al, 2010). In addition, the CENP-A/H4 interface forms a highly rigid structure in nucleosomes (Black et al, 2007a) as well as in prenucleosomal (CENP-A/H4)2 tetramers (Black et al, 2004) and CENP-A/H4/HJURP trimers (Bassett et al, 2012). Thus, we reasoned that centromere specific assembly of H4 results from co-assembly with CENP-A during G1 phase in vivo. To test this directly, we labeled nascent pools of SNAP-tagged CENP-A, H3.1, H3.3, H2B and H4 in cells synchronized in G2 phase of the cell cycle and analyzed assembly in the subsequent G1 phase (Figure 3.1C). Only CENP-A and H4-SNAP are assembled at centromere foci indicating that centromeric assembly of H4 is contemporaneous with CENP-A (Figure 3.1D).

130

Defining the heritable centromere core

Figure 3.1 H4, but not H3.1, H3.3 or H2B are co-assembled with CENP-A in G1 phase. (A) Outline of quench-chase- pulse labeling strategy, allowing visualization of a newly synthesized pool of SNAP, followed by Triton based pre- extraction. (B) Results of A for indicated histone-SNAP fusion proteins. Enlargement to the right shows rescaled images to indicate colocalization of newly synthesized H4-SNAP with centromeres (marked by CENP-C). Enlargements below show single focal plane images to indicate specific subnuclear assembly patterns. Blue, green, and red arrows show G1, early S, and mid/late S phase cells, respectively. (C) Outline of quench-chase-pulse experiment on synchronized cells. (D) Results of C for SNAP tagged histone proteins. CENP-C staining indicates centromere positions. Enlargement shows colocalization of newly synthesized H4-SNAP with centromeres.

131

Chapter 3

Importantly however, while nascent CENP-A-SNAP and H4-SNAP colocalize at centromeres, newly synthesized H3.1-, H3.3-, and H2B-SNAP remain diffusely localized (Figure 3.1D). This indicates that these histones are not preferentially assembled at centromeres at this stage. Importantly, this does not exclude the possibility that H2B is part of the centromeric nucleosome, nor that any of these histones are incorporated into centromeric chromatin at this time, albeit at a rate that is similar to the genome overall. This result, however, does demonstrate that the centromere is not a specialized chromatin domain that undergoes major nucleosome turnover events during G1 phase. Rather, CENP-A and H4 form a subnucleosomal core, which is specifically assembled at centromeres during G1 phase.

Figure 3.2 Assembly of CENP-A and H4 depends on passage through mitosis. (A) Outline of quench-chase-pulse in unperturbed cells, or combined with nocodazole treatment, or nocodazole treatment and washout. (B) Results of A for CENP-A-SNAP and H4-SNAP. Cyclin B and tubulin staining indicate G2 and G1 (midbodies) status, respectively. (C) Quantification of B. ~200–300 cells were analyzed for each condition. Note that during the 8 hour chase, cells transit through ~40% of the 22 hour cell cycle indicating the maximum expected percentage of cells entering G1 phase.

132

Defining the heritable centromere core

To validate that centromeric enrichment of H4 is not a consequence of the SNAP labeling procedure, we created a polyclonal HeLa cell line expressing H4-YFP. While endogenous pools of H4 are oscillating along the cell cycle, peaking in S phase (Marzluff & Duronio, 2002), the YFP tagged H4 transgene, like our SNAP-tagged H4, is expressed at a constitutive level. Consequently, the relative levels of tagged versus endogenous H4 are higher in G1 phase than in S phase. For this reason we expect that tagged H4 can be detected at centromeres, despite genome wide assembly in S phase, even without pulse-chase labeling. Indeed, when cells express low levels of H4- YFP, centromeric enrichment of this fusion protein can be detected over general chromatin (Figure 3.S1C), corroborating our observations with the SNAP-tag. Next, we determined whether centromeric H4 assembly depends on G1 phase entry. For this, we labeled nascent proteins either in an asynchronous population of cells or in cells which were prevented from exiting mitosis by addition of nocodazole (Figure 3.2A). After an 8 hour synthesis period, both CENP-A-SNAP and H4-SNAP readily assembled at centromeres in a subset of unperturbed cells. None of these cells stained positive for Cyclin B (Figure 3.2B–C), indicating that no centromere assembly occurred in late S, G2, or M phase. Consistently, virtually no cells assembled CENP-A or H4 when entry into G1 was prevented by addition of nocodazole in asynchronous cells (Figure 3.2B–C) or in a G2 synchronized population (Figure 3.S2E–F). However, nocodazole treatment or consequent mitotic arrest do not irreversibly prevent assembly, as release into G1 by nocodazole washout promptly resulted in centromere targeting of CENP-A-SNAP and H4-SNAP, exclusively in Cyclin B negative cells (Figure 3.2B–C). Analysis of cells synchronized at different stages along the cell cycle confirm that enrichment of nascent H4-SNAP at centromeres is only observed if cells cycle through G1 (Figure 3.S2) indicating that assembly of this histone at centromeres is uniquely restricted to this phase. We conclude that CENP-A and H4 assemble contemporaneously, in a manner dependent on mitotic exit. In

133

Chapter 3

addition, since G1 assembly of H4 is largely restricted to centromeres, our data strongly suggest that nucleosome assembly (of other H3 variants) throughout the rest of the genome represent a minority of assembly at this stage of the cell cycle. Thus, while only representing at most ~2% the total number of all nucleosomes (Black et al, 2007b), CENP-A nucleosome deposition represents the major form of chromatin assembly in G1. Together, these results strongly suggest that CENP-A and H4 represent the centromeric nucleosome core, which is assembled as a pre-formed complex during early G1 phase by the CENP-A loading machinery. The absence of foci of nascent H3.1, H3.3, and H2B indicates that these proteins are not preferentially assembled at centromeres, arguing against general chromatin reorganization during G1 phase.

Quantitative retention of the centromeric nucleosome core. Once incorporated into centromeric chromatin, CENP-A is stably transmitted as cells divide (Jansen et al, 2007) and diluted among nascent sister chromatids during S phase (Dunleavy et al, 2011). To test whether this is also true for other histones at the centromere, we performed pulse-chase experiments of SNAP-tagged proteins (Figure 3.3A). SNAP-based fluorescent pulse labeling followed by a chase period determines the turnover rate of the labeled protein pool in vivo (Jansen et al, 2007; Bodor et al, 2012). Remarkably, both CENP-A and H4 retain centromeric enrichment for the duration of the experiment (72 hours; Figure 3.3B), and can still be observed at even longer timescales (up to 120 hours for CENP-A and 96 hours for H4; Figure 3.S3A). To determine the relative stability of centromere enriched histones, we quantified centromeric and non- centromeric TMR-Star fluorescence intensity as a measure of the amount of protein remaining at different time points (Figure 3.3C and see methods). Strikingly, we find that centromeric pools of CENP-A and H4 are considerably more stable than H3.1 (Figure 3.3D). Moreover, while H3.1 turnover is indifferent to centromere localization, the centromeric pool of

134

Defining the heritable centromere core

H4 has an increased stability compared to H4 outside of the centromere (Figure 3.3D), indicating that CENP-A/H4 containing nucleosomes are preferentially stabilized compared to general chromatin. Similar to H3.1, no specific stability of H2B or H3.3 was observed at centromeres (Figure 3.S3B). This indicates that H2A/H2B dimers exchange on centromeric nucleosomes at similarly high rates as on conventional nucleosomes in bulk chromatin. Moreover, considering that intervening H3.1 and H3.3 nucleosomes are present at centromeres (Blower et al, 2002; Dunleavy et al, 2011), we find that long-term retention of chromatin is restricted to the CENP-A/H4 core of CENP-A nucleosomes with H3.1/H3.3 nucleosomes turning over at higher rates.

Timing of assembly and stable retention of the centromeric nucleosome core is controlled by the CENP-A targeting domain. While centromeres are maintained epigenetically, the unusual properties of CENP-A nucleosomes we uncovered may be dependent on local sequence features at centromeres. Alternatively, timing of centromere assembly and stable retention of CENP-A nucleosomes could be directed in cis by CENP-A itself. The CENP-A targeting domain (CATD), encompassing the L1 loop and α2 helix of the CENP-A histone fold domain, plays a pivotal role in the definition of centromeric chromatin. Replacement of the corresponding domain of canonical H3 with the CATD of CENP-A is sufficient to target the chimeric H3CATD to both canonical centromeres (Black et al, 2004, 2007b) and neocentromeres (Bassett et al, 2010). Furthermore, binding of prenucleosomal CENP-A to its histone chaperone HJURP is mediated through the CATD (Black et al, 2004; Foltz et al, 2009; Shuaib et al, 2010; Bassett et al, 2012). HJURP is itself recruited to centromeric chromatin in early G1 (Foltz et al, 2009; Dunleavy et al, 2009).

135

Chapter 3

Figure 3.3 CENP-A and H4 are preferentially maintained at centromeres. (A) Outline of pulse-chase experiment allowing for analysis of a pre-incorporated pool of SNAP for up to 72 hours. At each time point, cells were counted to allow accurate quantification of SNAP turnover per cell division. (B) Results of A for CENP-A-SNAP, H4-SNAP and H3.1-SNAP. Enlargements show rescaled images of remaining protein pool after 72 hours (see also Figure 3.S3A). (C). Schematic outline for calculation of histone half-life. (D) Half-life measurements of centromeric and non-centromeric histone pools as a function of time from experiment in B. Non-centromeric CENP-A is below detection and therefore not measured. Data is obtained from between 570 and 1464 (centromeric) foci for each time point.

136

Defining the heritable centromere core

We decided to test directly whether, in addition to regulating centromeric targeting itself, the CATD is sufficient to dictate histone assembly timing. When labeling a nascent pool of stably expressed H3CATD-SNAP we detected centromeric H3CATD only in Cyclin B negative cells (Figure 3.4A–B), suggesting that cells only load H3CATD into centromeres during G1 phase. In addition, as for CENP-A and H4 (Figure 3.2A), prevention of mitotic exit by nocodazole treatment abolished centromeric assembly of H3CATD-SNAP, while release from this arrest resulted in mitotic exit and concomitant assembly (Figure 3.4A–B). We conclude that, apart from centromere localization, the CATD also mediates cell cycle control of CENP-A assembly. Next, we determined whether long-term retention of CENP-A nucleosomes at centromeres is also an intrinsic property of CENP-A. We carried out pulse-chase experiments on H3CATD-SNAP expressing cells (Figure 3.4C) and analyzed retention of H3CATD over time. As for CENP-A and H4, pulse labeled H3CATD-SNAP remains detectable for multiple cell divisions up to 120 hours following labeling (Figure 3.4D and 3.S3A). To compare the stability of centromeric histones, we determined their rate of turnover as a function of the number of cell divisions expressed as the half- life (Figure 3.4E and see methods). In an extreme case where histones do not turn over at all, loss of histone proteins would be expected to occur only by redistribution among newly replicated sister chromatids during S phase (replicative dilution). In this situation, we would find a 50% reduction of fluorescence after each cell division (i.e. a histone half-life of exactly 1 division; Figure 3.4E, dashed line). For CENP-A-SNAP (experiment in Figure 3.3), we observed a half-life of 1.07 ± 0.17 divisions (mean ± SEM is indicated; Figure 3.4E), consistent with turnover by replicative dilution only. Importantly, we observed very similar behavior for both H4-SNAP and H3CATD-SNAP at centromeres, with half-lives of 0.94 ± 0.11 and 0.79 ± 0.12 divisions, respectively (Figure 3.4E). None of these values are significantly different from a theoretical replicative dilution rate of 1 cell division (one- tailed, one-sample t-test; n=3, α=0.05 in all cases).

137

Chapter 3

While the CATD has previously been implicated in rigidifying the CENP-A/H4 interface within the nucleosome particle (Black et al, 2004, 2007a; Bassett et al, 2012; Sekulic et al, 2010) how this contributes to CENP-A stability in vivo remained untested. Our results now show that

CENP-A confers long-term stability to the centromeric (CENP-A/H4)2 subnucleosome core and that this in vivo stability is encoded within the residues that constitute the CENP-A targeting domain. This feature of CENP-A ensures stable chromatin marking of centromeres across multiple divisions.

Figure 3.4 CATD determines G1 phase assembly and stable transmission of CENP-A nucleosomes. (A–B) As in Figure 3.2 for H3CATD-SNAP. (C) As in Figure 3.3B for H3CATD-SNAP. (D) Determination of centromeric histone half-life as a function of population doublings from experiments shown in Figure 3.3B and 3.4C. Dashed line (replicative dilution) indicates expected values for proteins that are never lost, but merely redistributed as cells divide. Average and SEM of 3 independent experiments is shown.

138

Defining the heritable centromere core

Quantitative retention does not require HJURP or M18BP1. We have shown that quantitative retention of CENP-A is, at least in part, directed by the CATD. However, the mechanism by which the CATD contributes to CENP-A stability remains unclear. To date, the most clearly defined function of the CATD is to provide the binding interface for the CENP-A chaperone HJURP (Foltz et al, 2009; Shuaib et al, 2010; Hu et al, 2011; Bassett et al, 2012). Interestingly, a proportion of endogenous HJURP is stably chromatin bound (Foltz et al, 2006). This raises the possibility that HJURP binding to CENP-A protects it from turning over, e.g. by binding to chromatin incorporated CENP-A or by transiently chaperoning this histone during the transition from parental chromosomes to daughter chromatids during DNA replication. In addition, a severe reduction of centromeric CENP-A levels was previously observed after depletion of M18BP1 (Maddox et al, 2007), suggesting that this protein may have a role beyond CENP-A assembly and contribute to its stable maintenance. To test this hypothesis directly, we combined SNAP labeling experiments with RNAi against HJURP and M18BP1 as detailed in Figure 3.5A. As expected, nascent centromeric CENP-A-SNAP was readily observed in all cells after siRNA mediated depletion of a control protein (GAPDH, Figure 3.5B). However, a large proportion of cells were unable to assemble nascent CENP-A-SNAP after depletion of HJURP (Figure 3.5B), as has been observed previously (Foltz et al, 2009). This result is consistent with the known role of HJURP in the assembly of CENP-A during G1 (Barnhart et al, 2011). Similar results were found for M18BP1 (Figure 3.5B). Quantification of centromeric signals shows that nascent CENP-A-SNAP levels are reduced by ~50% after depletion of HJURP or M18BP1 (Figure 3.5C). Similar results were obtained when RNAi was performed against CENP-A itself [Figure 3.5B–C and (Bodor et al, 2012)].

139

Chapter 3

Figure 3.5 HJURP and M18BP1 are dispensable for stable retention of CENP-A. (A) Outline of combined SNAP and RNAi experiment. To minimize variation of RNAi efficiency, quench-chase-pulse and pulse-chase experiments were done in parallel. (B) Results of A after depletion of indicated proteins. Images are displayed for nascent CENP-A-SNAP and the pre-incorporated pool at 24 hours post RNAi. (C) Quantification of centromeric TMR-star fluorescence of indicated CENP-A-SNAP pools after depletion of target proteins. Results were normalized against control RNAi (GAPDH). Average and SEM for at least 3 independent experiments is shown. Asterisks and “NS” respectively indicate statistically significant (p < 0.01) and non-significant (p > 0.05) differences from control samples in paired t-tests.

To test whether these loading factors are also involved in stabilizing previously incorporated CENP-A nucleosomes, we combined pulse-chase experiments with RNAi. Retention of CENP-A-SNAP at centromeres was analyzed after target protein depletion for 24, 48, and 72 hours to allow for assessment of both short- and long-term effects on CENP-A stability. In this assay, centromeric CENP-A-SNAP could be observed in all cells analyzed (Figure 3.5B) and no quantitative differences were observed between control

140

Defining the heritable centromere core

RNAi or depletion of HJURP, M18BP1, or CENP-A at any time point (Figure 3.5C). To ensure that we used conditions that effectively reduce protein levels, these pulse-chase experiments were performed in parallel with the quench-chase-pule experiments described above (Figure 3.5A). Our results strongly suggest that HJURP is dispensable for stabilizing centromeric CENP-A nucleosomes. We conclude that the long-term stability of the CENP-A/H4 nucleosome core is due to an HJURP and M18BP1 independent role of the CATD.

Timing of centromeric nucleosome assembly is independent of alphoid DNA. We have shown that the CATD of CENP-A is sufficient to direct G1 phase restricted assembly of CENP-A chromatin suggesting that temporal loading is dictated by the CENP-A protein itself. However, this does not exclude a role for local sequence context being involved in regulating cell cycle timing. Mammalian centromeres are assembled on arrays of alpha satellite DNA. While overall centromere function is not strictly dependent on this DNA sequence it may play a role in regulating centromere assembly and maintenance. This is clear from efforts to produce centromeres de novo on artificial chromosomes. While in some systems de novo centromeres can be formed on any DNA (Yuen et al, 2011), success in mammalian cells has only been reported with constructs containing large fragments of alphoid DNA (Ohzeki et al, 2002). In addition, the inner centromere component Aurora B was found to be mislocalized at a stably maintained human non-alphoid containing neocentromere, resulting in an impaired mitotic error correction mechanism (Bassett et al, 2010). Thus, although neocentromeres can exist on non-alphoid DNA, the role of DNA sequences in maintenance of existing centromeres remains elusive.

141

Chapter 3

Figure 3.6 Timing of CENP-A assembly is maintained at neocentromeres. (A) Cartoon of maternal (canonical centromere) and paternal (neocentric) chromosome 4 in PD-NC4 cells. Indicated is chromosomal position 4q21.3, the site of neocentromere formation and the hybridization site of the FISH probe used. (B) Outline of quench-chase-pulse experiment in CENP-A-SNAP expressing PD-NC4 cells. (C–D) Results of B for cells in G1 phase (C) or G2 phase (D), as indicated by nucleolar TMR staining, shown in rescaled inset). CENP-T indicates centromere positions. Enlargements display images of the hybridization sites of the FISH probe. Green arrows indicate the neocentromere, while red arrows show the homologous region on the maternal chromosome. (E) GFP-Mis18α expressing PD-NC4 cells were stained for GFP and for 4q21.3 by FISH to detect Mis18α and the NeoCEN4, respectively. Enlargements as above. Paternal (Neocentric) and maternal 4q21.3 positions are indicated by p and m, respectively, in C–E.

142

Defining the heritable centromere core

To determine the contribution of cis DNA elements in alphoid sequences on the timing of CENP-A assembly, we stably expressed CENP-A-SNAP in PD-NC4 (pseudodicentric-neocentric chromosome 4) cells. In these cells, the centromere on the paternally inherited chromosome 4 (but not the maternal one) has repositioned to chromosomal position 4q21.3, which does not contain alphoid DNA sequences (NeoCEN4) [Figure 3.6A and (Amor et al, 2004)]. By combining quench-chase-pulse experiments with FISH against 4q21.3 (NeoCEN4) we were able to determine that CENP-A assembly at neocentromeres occurred contemporaneously with canonical centromeres of the same cell (Figure 3.6B–C). Importantly, although a subset of cells displayed diffuse nucleolar staining, indicative of the prenucleosomal pool of CENP-A in G2 phase (Jansen et al, 2007; Silva et al, 2012), CENP-A assembly was never observed at the NeoCEN4 alone, i.e. when no assembly occurred on other centromeres (Figure 3.6D). To corroborate these results, we stably expressed a GFP-tagged version of Mis18α, an essential component of the Mis18 complex, in PD-NC4 cells. Interestingly, one member of this complex, M18BP1, contains a Myb-domain (Fujita et al, 2007; Maddox et al, 2007), a protein domain that is often involved in site-specific DNA binding (Lipsick, 1996). Nevertheless, GFP- Mis18α is consistently recruited to NeoCEN4 and alphoid DNA bearing centromeres simultaneously (Figure 3.6E). Together, these results show that CENP-A assembly at the NeoCEN4 occurs concurrently with canonical centromeres, indicating that temporal control of the CENP-A assembly machinery is maintained independently of alphoid DNA. This is consistent with a dominant role for the CENP-A encoded CATD in directly controlling temporal assembly of CENP-A chromatin.

143

Chapter 3

DISCUSSION

Maintenance of epigenetic identity requires the inheritance of structural information from one cell generation to the next. Chromatin proteins and their modifications have been implicated in such cellular memory (Talbert & Henikoff, 2010; Gardner et al, 2011). However, transmission of chromatin- based information faces many challenges throughout the cell cycle that may disturb epigenetic inheritance, including nucleosome disruption during DNA replication and chromatin (de)condensation during mitosis. Previous work identified an atypical timing of assembly of CENP-A, as well as centromere retention of the existing pool of CENP-A throughout the cell cycle (Jansen et al, 2007). We now extend these findings and determined that a distinct phase of centromeric loading in G1 as well as quantitative centromeric retention is restricted to CENP-A and H4, rather than being a general property of centromeric chromatin. Metabolic labeling experiments and photo bleaching studies of GFP-tagged histones have previously established that histone H3 and H4 are stable components whereas H2A and H2B are more dynamic (Kimura & Cook, 2001; Xu et al, 2010). However, apart from CENP-A itself (Jansen et al, 2007), locus specific assembly and turnover has not been previously determined for these or other histones. Our results now show that at the centromere, the CENP-A/H4 form a stable subnucleosomal core that is quantitatively retained throughout multiple cell divisions to maintain centromere identity (Figure 3.7). Retention of H4 specifically at the centromere but not elsewhere indicates that the centromeric CENP-A/H4 species is more stable than general chromatin outlasting most, if not all, other nucleosome types. Interestingly, many of the unique features of the CENP-A/H4 centromeric core are directed through the CATD region of CENP-A. It has previously been shown that this region is responsible for 1) targeting of CENP-A to centromeres (Black et al, 2004, 2007b) in a sequence independent manner (Bassett et al, 2010); 2) binding to the CENP-A specific histone chaperone HJURP (Foltz et al, 2009; Shuaib et al, 2010;

144

Defining the heritable centromere core

Figure 3.7 Model depicting unique features of centromeric nucleosomes. Cell cycle dynamics of different types of nucleosomes are indicated. CENP-A nucleosomes are assembled at centromeres in G1 phase, while H3.1 and H3.3 nucleosomes are assembled into general chromatin in S phase and throughout the cell cycle, respectively (Ray-Gallet et al, 2011). Neither H3.1 nor H3.3 nucleosomes are preferentially loaded into centromeric chromatin during G1 phase or any other cell cycle stage. While H2A and H2B are dynamic in all types of nucleosomes, the centromeric CENP-A/H4 core is stable at time scales far surpassing the cell division rate. However, H3.1, H3.3, and non-centromeric H4 turn over more rapidly than CENP-A, and no preferential centromeric maintenance of H3.1 or H3.3 is observed. Key to both temporal assembly and stable transmission is the CATD domain of CENP-A that forms a stable interface with H4 in both CENP-A and H3CATD nucleosomes (Sekulic et al, 2010; Bassett et al, 2012).

Bassett et al, 2012); 3) a unique, highly rigid, CENP-A/H4 dimerization interface (Black et al, 2004; Bassett et al, 2012; Sekulic et al, 2010); and 4) binding of CENP-N, which is in turn required for efficient centromeric recruitment of nascent CENP-A (Carroll et al, 2009). In addition, we now show that the CATD 5) is the element in CENP-A that mediates correct timing of CENP-A assembly, independently of underlying DNA sequence and that 6), critically, this region confers in vivo hyperstability to centromeric nucleosomes in a manner independent of HJURP or M18BP1. Importantly, parts of CENP-A outside of the CATD region have been shown

145

Chapter 3

to be required for kinetochore assembly, e.g. through binding of the centromere protein CENP-C to the 6 most carboxy-terminal residues of CENP-A (Guse et al, 2011; Carroll et al, 2010). Thus, while different domains of CENP-A are likely to be involved in full centromere function, all of the key properties of CENP-A for epigenetically maintaining centromere position are mediated through the CATD. Our results identify the CENP-A/H4 complex as the primary components of the centromere that are selectively assembled each cell division in a manner that leads to their long-term maintenance. A key future challenge is to determine whether this unusual stability is an intrinsic property of CENP-A nucleosomes or dependent on external factors that ensure stable transmission of CENP-A and centromere identity.

146

Defining the heritable centromere core

MATERIAL & METHODS

Constructs and cell lines

Human H2B-SNAP, H4-SNAP, and H3CATD-SNAP constructs were created by PCR cloning of histone ORFs into pSS26m (Covalys) to create C- terminal SNAP fusion proteins. A triple hemagglutinin (3XHA) tag was placed at the C-terminus of SNAP. Histone H4-YFP was generated by PCR cloning of the human H4 ORF into pEYFP-N1 (Clontech) carrying Q69M (citrin) and A206K (monomerization) mutations. The histone-SNAP-3XHA and H4-YFP ORFs were subcloned into pBABE-Blast to generate retroviral expression constructs. These constructs were delivered into HeLa cells via Moloney murine leukemia retroviral delivery, as described previously (Morgenstern & Land, 1990; Burns et al, 1993). Cells stably expressing the SNAP fusion proteins were selected with 5 μg/ml blasticidin S (Calbiochem) and were isolated and individually sorted by flow cytometry (except H4-YFP which was analyzed as a polyclonal cell population). The resulting monoclonal lines were selected for proper levels of the SNAP fusion proteins by fluorescence microscopy after TMR-Star labeling. The following clones were selected and used throughout this study: H2B-SNAP clone #5; H4- SNAP clone #3; and H3CATD-SNAP clone #37. We previously described HeLa monoclonal cell lines stably expressing H3.1-SNAP or H3.3-SNAP [clone #7 or #2, respectively; (Ray-Gallet et al, 2011)] or CENP‑A-SNAP [clone #23,

(Jansen et al, 2007)]. All HeLa cell lines were grown at 37°C and 5% CO2 in DMEM containing 10% newborn calf serum, 2 mM L-glutamine, 100 U/ml Penicillin and 100 μg/ml Streptomycin (henceforth referred to as complete medium). In addition, SNAP expressing cells were maintained by addition of 1 μg/ml blasticidin S. PD-NC4 stable transgenic cell lines were created by Moloney murine leukemia retroviral delivery of constructs expressing CENP-A-SNAP (Jansen et al, 2007), or GFP-Mis18α (gift from Iain Cheeseman) (Silva et al, 2012). PD-NC4 cells were grown at 37°C and 5%

CO2 in DMEM supplemented with 10% fetal calf serum, 2 mM L-glutamine,

147

Chapter 3

100 μg/ml neomycin, 100 U/ml Penicillin and 100 μg/ml Streptomycin. Stable transgenic PD-NC4 lines were selected with 2,5 µg/ml Blasticidin (CENP-A-SNAP) or 500 ng/ml Puromycin (GFP-Mis18α).

SNAP-labeling SNAP labeling was performed essentially as described (Jansen et al, 2007; Bodor et al, 2012). Briefly, cells were labeled for 30’ with 2 μM BTP (SNAP-Cell Block, New England Biolabs) or 15’ with 2 μM TMR-Star (New England Biolabs) in complete medium, for quench or pulse labeling, respectively, after which cells were washed twice with PBS and reincubated with complete medium. After an additional 30 minutes, cells were washed once more with PBS and either reincubated with complete medium, or fixed and further treated for analysis, as indicated.

Cell synchronization and RNAi Cells were synchronized in early S phase by double thymidine block as described previously (Jansen et al, 2007; Bodor et al, 2012). Nocodazole was used at a concentration of 500 ng/ml except for experiment in Figure 3.S2F for which 200 ng/ml was used. RNAi was performed in a 24-well format using 60 pm siRNAs using Oligofectamine (Invitrogen) according to the manufacturer’s instructions. All siRNAs were obtained from Dharmacon: SMARTpools were used to deplete HJURP, M18BP1, and GAPDH; for CENP-A depletion siRNA target 5’-ACAGUCGGCGGAGACAAGG-3’ was used.

Immunofluorescence Fixation, immunofluorescence, and DAPI staining of HeLa cells was performed as described (Bodor et al, 2012). Pre-extraction was performed for 5 minutes using 0.3% Triton X-100 (Sigma) in PBS prior to fixation. Antibodies against CENP-C (mouse monoclonal), Cyclin B (sc-245, Santa Cruz), and α-tubulin (YL1/2, Serotec) were used at a dilution of 1:10,000,

148

Defining the heritable centromere core

1:50, and 1:2,500, respectively. Fluorescent secondary antibodies were obtained from Jackson ImmunoResearch and used at a dilution of 1:200.

Immuno-FISH FISH was performed as previously described (Black et al, 2007a) with the following alterations: Upon cell fixation and the freeze/thawing cycles, cells were prepared for immunofluorescence as defined above. GFP-Mis18α was detected by immunofluorescence as GFP signal is lost during FISH fixation procedure. GFP-Booster (Chromotek), CENP-T (Barnhart et al, 2011) and anti-rabbit Dy680 (Rockland Immunochemicals) were used in a dilution of 1:100, 1:1000, and 1:50, respectively. Subsequently, cells where fixed with 2% formaldehyde for 10 minutes at room temperature and washed with PBS. FISH protocol was then continued as described (Black et al, 2007a). A chromosome 4q21.3 specific probe was generated by labeling a mixture of BAC clones (RP11-113G13, RP11-204I22, RP11-209G6, RP11- 458J15; BACPAC Resources Center, Oakland, CA) with either Tetramethyl- Rhodamine-5-dUTP or Fluorescein-12-dUTP (Roche, Indianapolis, IN), to detect co-localization with GFP-Mis18α or with CENP-A-SNAP, respectively. Coverslips were washed in 2X SSC (0.3 M NaCl, 30 mM Sodium Citrate, pH 7.0), containing 60% formamide prior to DAPI staining and mounting.

Microscopy Cells were imaged on a DeltaVision Core system (Applied Precision) controlling an inverted microscope (Olympus, IX-71), which is coupled to a Cascade2 EMCCD camera (Photometrics). Images were collected at 1x binning using a 100x oil objective (NA 1.40, UPlanSApo) with 0.2 mm Z- sections scanning the entire nucleus. Images were subsequently deconvolved using soft-WoRx (Applied Precision). Unless otherwise indicated, maximum intensity projections of deconvolved images are shown. Centromere quantification was performed using a custom made macro for ImageJ (NIH), called CRaQ (Bodor et al, 2012). For quantitative purposes, images

149

Chapter 3

were collected on a 512x512 pixel chip and flatfield and camera noise corrected during acquisition using soft-WoRx (Applied Precision). Fluorescence quantification was performed on non-deconvolved images. For centromere quantification, CRaQ was set to measure peak intensity values within a 7x7 pixel box around the centroid position of the centromere. For non-centromeric values (Figure 3.3D), a 2x2 pixel box was placed at a position shifted away from the centromere centroid by 5 pixels in both x and y. In Figure 3.3D, to enable the measurement of diffuse nuclear signals, fluorescence immediately outside nuclei was used for background correction. For Figure 3.4E, centromeric fluorescence was corrected for local background for each centromere. To quantify the rate of division of SNAP- tagged cells, we seeded one additional coverslip of CENP-A-SNAP cells for each time point, and treated it identically to the other cells throughout the duration of the experiment (TMR-Star and BTP were omitted and cells were mock treated with DMSO instead). At the time of fixation, the extra coverslip was trypsinized and cells were counted in a haemocytometer. To calculate histone half-life we measured fluorescence intensities as a function of number of cell divisions at 24, 48, and 72 hours. From this, we calculated the best fit one phase decay regression line (F = e-k·t; where F is fluorescence and t is time or number of divisions) using GraphPad Prism software (with a constrained plateau at 0 and F0 = 1). Half-life equals ln(2)/k (Figure 3.3C).

150

Defining the heritable centromere core

REFERENCES

Ahmad K & Henikoff S (2002) Histone H3 variants specify modes of chromatin assembly. Proc. Natl. Acad. Sci. U. S. A. 99: 16477–16484 Amor DJ, Bentley K, Ryan J, Perry J, Wong L, Slater H & Choo KHA (2004) Human centromere repositioning ‘in progress’. Proc. Natl. Acad. Sci. U. S. A. 101: 6542– 6547 Barnhart MC, Kuich PHJL, Stellfox ME, Ward JA, Bassett EA, Black BE & Foltz DR (2011) HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochore. J. Cell Biol. 194: 229–243 Bassett EA, DeNizio J, Barnhart-Dailey MC, Panchenko T, Sekulic N, Rogers DJ, Foltz DR & Black BE (2012) HJURP uses distinct CENP-A surfaces to recognize and to stabilize CENP-A/histone H4 for centromere assembly. Dev. Cell 22: 749– 762 Bassett EA, Wood S, Salimian KJ, Ajith S, Foltz DR & Black BE (2010) Epigenetic centromere specification directs aurora B accumulation but is insufficient to efficiently correct mitotic errors. J. Cell Biol. 190: 177–185 Bernad R, Sánchez P, Rivera T, Rodríguez-Corsino M, Boyarchuk E, Vassias I, Ray- Gallet D, Arnaoutov A, Dasso M, Almouzni G & Losada A (2011) Xenopus HJURP and condensin II are required for CENP-A assembly. J. Cell Biol. 192: 569–582 Black BE, Brock MA, Bédard S, Woods VL & Cleveland DW (2007a) An epigenetic mark generated by the incorporation of CENP-A into centromeric nucleosomes. Proc. Natl. Acad. Sci. U. S. A. 104: 5008–5013 Black BE, Foltz DR, Chakravarthy S, Luger K, Woods VL & Cleveland DW (2004) Structural determinants for generating centromeric chromatin. Nature 430: 578– 582 Black BE, Jansen LET, Maddox PS, Foltz DR, Desai AB, Shah JV & Cleveland DW (2007b) Centromere identity maintained by nucleosomes assembled with histone H3 containing the CENP-A targeting domain. Mol. Cell 25: 309–322 Blower MD, Sullivan BA & Karpen GH (2002) Conserved organization of centromeric chromatin in flies and humans. Dev. Cell 2: 319–330 Bodor DL, Rodríguez MG, Moreno N & Jansen LET (2012) Analysis of Protein Turnover by Quantitative SNAP-Based Pulse-Chase Imaging. Curr. Protoc. Cell Biol. Chapter 8: Unit8.8 Burns JC, Friedmann T, Driever W, Burrascano M & Yee JK (1993) Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc. Natl. Acad. Sci. U. S. A. 90: 8033–8037 Carroll CW, Milks KJ & Straight AF (2010) Dual recognition of CENP-A nucleosomes is required for centromere assembly. J. Cell Biol. 189: 1143–1155 Carroll CW, Silva MCC, Godek KM, Jansen LET & Straight AF (2009) Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N. Nat. Cell Biol. 11: 896–902

151

Chapter 3

Cheeseman IM & Desai A (2008) Molecular architecture of the kinetochore- microtubule interface. Nat. Rev. Mol. Cell Biol. 9: 33–46 Dunleavy EM, Almouzni G & Karpen GH (2011) H3.3 is deposited at centromeres in S phase as a placeholder for newly assembled CENP-A in G(1) phase. Nucl. Austin Tex 2: 146–157 Dunleavy EM, Roche D, Tagami H, Lacoste N, Ray-Gallet D, Nakamura Y, Daigo Y, Nakatani Y & Almouzni-Pettinotti G (2009) HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell 137: 485–497 Foltz DR, Jansen LET, Bailey AO, Yates JR, Bassett EA, Wood S, Black BE & Cleveland DW (2009) Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell 137: 472–484 Foltz DR, Jansen LET, Black BE, Bailey AO, Yates JR & Cleveland DW (2006) The human CENP-A centromeric nucleosome-associated complex. Nat. Cell Biol. 8: 458–469 Fujita Y, Hayashi T, Kiyomitsu T, Toyoda Y, Kokubu A, Obuse C & Yanagida M (2007) Priming of Centromere for CENP-A Recruitment by Human hMis18α, hMis18β, and M18BP1. Dev. Cell 12: 17–30 Gardner KE, Allis CD & Strahl BD (2011) OPERating ON Chromatin, a Colorful Language where Context Matters. J. Mol. Biol. 409: 36–46 Guse A, Carroll CW, Moree B, Fuller CJ & Straight AF (2011) In vitro centromere and kinetochore assembly on defined chromatin templates. Nature 477: 354–358 Hemmerich P, Weidtkamp-Peters S, Hoischen C, Schmiedeberg L, Erliandri I & Diekmann S (2008) Dynamics of inner kinetochore assembly and maintenance in living cells. J. Cell Biol. 180: 1101–1114 Hu H, Liu Y, Wang M, Fang J, Huang H, Yang N, Li Y, Wang J, Yao X, Shi Y, Li G & Xu R-M (2011) Structure of a CENP-A-histone H4 heterodimer in complex with chaperone HJURP. Genes Dev. 25: 901–906 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805 Keppler A, Gendreizig S, Gronemeyer T, Pick H, Vogel H & Johnsson K (2003) A general method for the covalent labeling of fusion proteins with small molecules in vivo. Nat. Biotechnol. 21: 86–89 Keppler A, Pick H, Arrivoli C, Vogel H & Johnsson K (2004) Labeling of fusion proteins with synthetic fluorophores in live cells. Proc. Natl. Acad. Sci. U. S. A. 101: 9955–9959 Kimura H & Cook PR (2001) Kinetics of Core Histones in Living Human Cells Little Exchange of H3 and H4 and Some Rapid Exchange of H2b. J. Cell Biol. 153: 1341–1354 Lipsick JS (1996) One billion years of Myb. Oncogene 13: 223–235 Liu S-T, Rattner JB, Jablonski SA & Yen TJ (2006) Mapping the assembly pathways that specify formation of the trilaminar kinetochore plates in human cells. J. Cell Biol. 175: 41–53

152

Defining the heritable centromere core

Maddox PS, Hyndman F, Monen J, Oegema K & Desai A (2007) Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 176: 757–763 Marshall OJ, Chueh AC, Wong LH & Choo KHA (2008) Neocentromeres: New Insights into Centromere Structure, Disease Development, and Karyotype Evolution. Am. J. Hum. Genet. 82: 261–282 Marzluff WF & Duronio RJ (2002) Histone mRNA expression: multiple levels of cell cycle regulation and important developmental consequences. Curr. Opin. Cell Biol. 14: 692–699 Mellone BG, Grive KJ, Shteyn V, Bowers SR, Oderberg I & Karpen GH (2011) Assembly of Drosophila centromeric chromatin proteins during mitosis. PLoS Genet. 7: e1002068 Mendiburo MJ, Padeken J, Fülöp S, Schepers A & Heun P (2011) Drosophila CENH3 is sufficient for centromere formation. Science 334: 686–690 Moree B, Meyer CB, Fuller CJ & Straight AF (2011) CENP-C recruits M18BP1 to centromeres to promote CENP-A chromatin assembly. J. Cell Biol. 194: 855–871 Morgenstern JP & Land H (1990) Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res. 18: 3587 –3596 Ohzeki J, Nakano M, Okada T & Masumoto H (2002) CENP-B box is required for de novo centromere chromatin assembly on human alphoid DNA. J. Cell Biol. 159: 765–775 Pearson CG, Yeh E, Gardner M, Odde D, Salmon ED & Bloom K (2004) Stable Kinetochore-Microtubule Attachment Constrains Centromere Positioning in Metaphase. Curr. Biol. 14: 1962–1967 Ray-Gallet D, Quivy J-P, Scamps C, Martini EM-D, Lipinski M & Almouzni G (2002) HIRA is critical for a nucleosome assembly pathway independent of DNA synthesis. Mol. Cell 9: 1091–1100 Ray-Gallet D, Woolfe A, Vassias I, Pellentz C, Lacoste N, Puri A, Schultz DC, Pchelintsev NA, Adams PD, Jansen LET & Almouzni G (2011) Dynamics of histone H3 deposition in vivo reveal a nucleosome gap-filling mechanism for H3.3 to maintain chromatin integrity. Mol. Cell 44: 928–941 Schuh M, Lehner CF & Heidmann S (2007) Incorporation of Drosophila CID/CENP-A and CENP-C into centromeres during early embryonic anaphase. Curr. Biol. CB 17: 237–243 Sekulic N, Bassett EA, Rogers DJ & Black BE (2010) The structure of (CENP-A- H4)(2) reveals physical features that mark centromeres. Nature 467: 347–351 Shuaib M, Ouararhni K, Dimitrov S & Hamiche A (2010) HJURP binds CENP-A via a highly conserved N-terminal domain and mediates its deposition at centromeres. Proc. Natl. Acad. Sci. U. S. A. 107: 1349–1354 Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63

153

Chapter 3

Talbert PB & Henikoff S (2010) Histone variants — ancient wrap artists of the epigenome. Nat. Rev. Mol. Cell Biol. 11: 264–275 Voullaire LE, Slater HR, Petrovic V & Choo KH (1993) A functional marker centromere with no detectable alpha-satellite, satellite III, or CENP-B protein: activation of a latent centromere? Am. J. Hum. Genet. 52: 1153–1163 Xu M, Long C, Chen X, Huang C, Chen S & Zhu B (2010) Partitioning of histone H3- H4 tetramers during DNA replication-dependent chromatin assembly. Science 328: 94–98 Yoda K, Ando S, Morishita S, Houmura K, Hashimoto K, Takeyasu K & Okazaki T (2000) Human centromere protein A (CENP-A) can replace histone H3 in nucleosome reconstitution in vitro. Proc. Natl. Acad. Sci. U. S. A. 97: 7266–7271 Yuen KWY, Nabeshima K, Oegema K & Desai A (2011) Rapid de novo centromere formation occurs independently of heterochromatin protein 1 in C. elegans embryos. Curr. Biol. CB 21: 1800–1807

Author contributions All experiments and analyses were performed by me, with the following exceptions: LPV performed and analyzed experiments shown in Figure 3.6; JFM performed experiment shown in Figure 3.S1; BEB created the H3CATD-SNAP cell line; LETJ performed experiments shown in Figure 3.1 and 3.S2. The manuscript for this chapter was drafted and revised with help of LETJ and constructive suggestions by all authors.

Acknowledgements We are indebted to Don W. Cleveland who hosted preliminary experiments in his laboratory. We thank Mariluz Gómez Rodríguez for help with the IF-FISH procedure and Nuno Moreno for help with image quantification. DLB and LPV are supported by the Fundação para a Ciência e a Tecnologia (FCT) fellowships SFRH/BD/74284/2010 and SFRH/BPD/69115/2010, respectively. This work is supported by NIH grant GM082989, a Career Award in the Biomedical Sciences from the Burroughs Wellcome Fund, and a Rita Allen Foundation Scholar Award to BEB and by the Fundação Calouste Gulbenkian, FCT grants BIA-BCM/100557/2008, BIAPRO/100537/2008, the European Commission FP7 programme and an EMBO installation grant to LETJ.

154

Defining the heritable centromere core

SUPPLEMENTARY FIGURES

Figure 3.S1 (related to Figure 3.1) Direct pulse labeling of SNAP-tagged histones. (A) Outline of SNAP-based pulse labeling experiment to visualize the total pool of SNAP protein. (B) Results of A for indicated histone-SNAP fusion proteins. (C) Centromeric enrichment can be observed in cells expressing low levels of H4-YFP.

155

Chapter 3

Figure 3.S2 (related to Figure 3.2) Quench-chase-pulse experiments reveal distinct assembly modes for H4 during the cell cycle. (A–E) Outlines and results for synchronized quench-chase-pulse experiments, analyzing H4-SNAP assembly for indicated portions of the cell cycle. (F) As in E, except that nocodazole was added to arrest cells upon mitotic entry.

156

Defining the heritable centromere core

Figure 3.S3 (related to Figures 3 and 4) Stable retention at centromeres is restricted to CENP-A, H4, and H3CATD. (A) Indicated cell lines were treated as in Figure 3.3A and imaged at indicated time points. Enlargements show rescaled images of centromeric signal. (B) Results of experiment as in Figure 3.3A for SNAP-tagged H3.1, H3.3, and H2B. Saturated enlargements of boxed cells are shown. No preferential retention at centromeres is observed for these histones.

157

Chapter 3 - Appendix

Appendix 3.1: The Role of CENP-C in CENP-A Dynamics

Ana Filipa David, Dani L. Bodor, and Lars E.T. Jansen

Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal

NB: While the original conception and design of the experiments described here are my own, the specific strategy was developed together with, and all experiments were performed by, Ana Filipa David, a former MSc student working under my supervision.

RESULTS

CENP-C is a member of the constitutive centromere associated network (CCAN). This protein has been shown to directly bind to CENP-A nucleosomes through its 6 terminal residues, LEEGLG (Carroll et al, 2010; Guse et al, 2011; Kato et al, 2013). In addition, CENP-C is required for recruitment of a large proportion of other CCAN members as well as the mitotic kinetochore complex (Carroll et al, 2010; Gascoigne et al, 2011; Guse et al, 2011; Przewloka et al, 2011; Screpanti et al, 2011) and can be sufficient to recruit CENP-A in de novo centromere formation (Hori et al, 2013). Indeed, depletion of CENP-C from human cells has been shown to lead to a reduction of centromeric CENP-A levels (Carroll et al, 2010). However, it remains unclear whether this results from a defect in assembly or retention of CENP-A. We combined RNAi mediated protein depletion of CENP-C with quench- chase-pulse and pulse-chase experiments to analyze the effect on nascent and pre-incorporated CENP-A, respectively. We found that the new pool of CENP-A-SNAP is significantly reduced (p < 0.01) after CENP-C depletion as compared to a control depletion (Figure 3.A, dark grey). In addition, we

158

The role of CENP-C in CENP-A dynamics

observed a small, yet non-statistically significant (p = 0.17) reduction of the old pool of CENP-A-SNAP (Figure 3.A, light grey). Our preliminary results indicate that CENP-C is involved in CENP-A assembly. In addition, it may have an independent role in the maintenance of centromeric CENP-A nucleosomes.

Figure 3.A Depletion of CENP-C leads to a reduction of both old and new pools of CENP-A, as measured after quench-chase-pulse and pulse- chase labeling of CENP-A-SNAP, respectively. Average ± SEM of 4 independent replicate experiments are shown; ~40–60 cells were analyzed per experiment. Dashed line indicates the normalized signal after scrambled control depletion.

MATERIAL & METHODS

Cell culture Experiments were performed on monoclonal HeLa cells, stably expressing CENP-A-SNAP; clone #72 (Jansen et al, 2007). HEK-293-T cells were used for production of lentiviral shRNA coding vectors (see below). Cell culture conditions are identical to those described in chapter 3.

RNAi mediated protein depletion A lentiviral-based system (Addgene) was used to deliver shRNA-coding vectors into HeLa cells. HEK-293-T cells were co-transfected with shRNA- coding plasmids (pLKO.1) and the packaging plasmids psPAX2 and pMD2.G, using the Lipofectamine LTX according to manufacturer’s protocol (Invitrogen). Transfection media was removed at 24h post-transfection and replaced with fresh culture medium. Viral particles in suspension were harvested after 24 hours and filtered through a 0.45μm filter.

159

Chapter 3 - Appendix

pLKO.1 plasmids with RNAi target sequences were purchased from Addgene and target sequences are part of an algorithm-based shRNA library (Moffat et al, 2006). The following targets were used for CENP-C depletion (manufacturer’s references are indicated): TRCN0000148798, TRCN0000150037, TRCN0000149366, TRCN0000148503, and TRCN0000146581. The following sequence was used as a scrambled control: CCTAAGGTTAAGTCGCCCTCG.

SNAP-labeling CENP-A-SNAP cells were seeded in 24 wells plates and infected on the next day with 70 μl of viral suspension of unknown titer in 2 ml of culture medium containing 8 μg/mL polybrene. At 26 hours post-infection, cells stably expressing shRNA were selected by addition of 1 μg/ml of puromycin. For pulse-chase experiments, TMR-Star labeling was performed 24h post- infection. For quench-chase-pulse experiments, BTP labeling was performed 40h post-infection and TMR-Star labeling 65h post-infection. In both cases, cells were trypsinized and transferred to 8-well glass bottom chambers (MatTek Corporation) at 67h post-infection and fixed at 72h post-infection. SNAP labeling, fixation, DAPI labeling, and imaging were performed as described in chapter 3. An automated algorithm was developed that measures the maximum and median (equivalent to background) nuclear TMR signal in maximum projected images and the difference between these respective signals was used as a measure of signal intensity.

REFERENCES

Carroll CW, Milks KJ & Straight AF (2010) Dual recognition of CENP-A nucleosomes is required for centromere assembly. J. Cell Biol. 189: 1143–1155 Gascoigne KE, Takeuchi K, Suzuki A, Hori T, Fukagawa T & Cheeseman IM (2011) Induced ectopic kinetochore assembly bypasses the requirement for CENP-A nucleosomes. Cell 145: 410–422 Guse A, Carroll CW, Moree B, Fuller CJ & Straight AF (2011) In vitro centromere and kinetochore assembly on defined chromatin templates. Nature 477: 354–358

160

The role of CENP-C in CENP-A dynamics

Hori T, Shang W-H, Takeuchi K & Fukagawa T (2013) The CCAN recruits CENP-A to the centromere and forms the structural core for kinetochore assembly. J. Cell Biol. 200: 45–60 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805 Kato H, Jiang J, Zhou B-R, Rozendaal M, Feng H, Ghirlando R, Xiao TS, Straight AF & Bai Y (2013) A Conserved Mechanism for Centromeric Nucleosome Recognition by Centromere Protein CENP-C. Science 340: 1110–1113 Moffat J, Grueneberg DA, Yang X, Kim SY, Kloepfer AM, Hinkle G, Piqani B, Eisenhaure TM, et al (2006) A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124: 1283–1298 Przewloka MR, Venkei Z, Bolanos-Garcia VM, Debski J, Dadlez M & Glover DM (2011) CENP-C Is a Structural Platform for Kinetochore Assembly. Curr. Biol. 21: 399–405 Screpanti E, De Antoni A, Alushin GM, Petrovic A, Melis T, Nogales E & Musacchio A (2011) Direct Binding of Cenp-C to the Mis12 Complex Joins the Inner and Outer Kinetochore. Curr. Biol. 21: 391–398

161

CHAPTER 4

The Quantitative Architecture of Centromeric Chromatin

Dani L. Bodor1, João F. Mata1, Mikhail Sergeev2, Ana Filipa David1, Kevan J. Salimian3, Tanya Panchenko3, Don W. Cleveland4, Ben E. Black3, Jagesh V. Shah2, and Lars E.T. Jansen1

1 Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal.

2 Harvard Medical School, Boston MA 02115, USA.

3 University of Pennsylvania, Philadelphia, PA 19104-6059, USA.

4 Ludwig Institute for Cancer Research, La Jolla, CA 92093, USA.

NB: This chapter is a near literal transcription of eLife 2014;3:e02137. Noteworthy is the change of the term “mass action mechanism” with the more appropriate “mass action-like mechanism” throughout the chapter (see Chapter 5). ABSTRACT

The centromere, responsible for chromosome segregation during mitosis, is epigenetically defined by CENP-A containing nucleosomes. The amount of centromeric CENP-A has direct implications for both the architecture and epigenetic inheritance of centromeres. Using complementary strategies, we determined that typical human centromeres contain ~400 molecules of CENP-A, which is controlled by a mass action- like mechanism. This number, despite representing only ~4% of all centromeric nucleosomes, forms a ~50-fold enrichment to the overall genome. In addition, although pre-assembled CENP-A is randomly segregated during cell division, this amount of CENP-A is sufficient to prevent stochastic loss of centromere function and identity. Finally, we produced a statistical map of CENP-A occupancy at a human neocentromere and identified nucleosome positions that feature CENP-A in a majority of cells. In summary, we present a quantitative view of the centromere that provides a mechanistic framework for both robust epigenetic inheritance of centromeres and the paucity of neocentromere formation.

Counting CENP-A molecules in human cells

INTRODUCTION

Centromeres are essential for proper cell division. During mitosis, a transient structure called the kinetochore is assembled onto centromeric chromatin, which mediates the interaction between DNA and the mitotic spindle (Allshire & Karpen 2008; Cheeseman & Desai 2008). Intriguingly, although centromeres are directly embedded in chromatin, specific DNA sequences are neither necessary nor sufficient for centromere function. This is best exemplified by the rare occurrence, within the human population, of neocentromeres: functional centromeres that have repositioned to atypical loci on the chromosome (Amor et al., 2004; Marshall et al., 2008; du Sart et al., 1997; Voullaire et al., 1993). Rather than centromeric sequences, the primary candidate for epigenetic specification of centromeres is the histone variant CENP-A, which replaces canonical H3 in centromeric nucleosomes (Henikoff et al., 2000; Palmer et al., 1987, 1991; Stoler et al., 1995; Yoda et al., 2000). CENP-A chromatin is sufficient for recruitment of the downstream centromere and kinetochore complexes (Barnhart et al., 2011; Carroll et al., 2009, 2010; Foltz et al., 2006; Guse et al., 2011; Mendiburo et al., 2011; Okada et al., 2006). In addition, CENP-A is stably transmitted at centromeres during mitotic (Bodor et al. 2013; Jansen et al. 2007) and meiotic (Raychaudhuri et al., 2012) divisions, and its assembly is tightly cell cycle controlled (Jansen et al., 2007; Schuh et al., 2007; Silva et al., 2012). Importantly, targeting of this protein to an ectopic site of the genome is sufficient to initiate an epigenetic feedback loop, recruiting more CENP-A to this site (Mendiburo et al. 2011). However, little is known about the quantity of CENP-A present at centromeres, despite this being an essential parameter for a functional understanding of both centromeric architecture and epigenetic inheritance. Here, we use multiple, independent approaches to determine the absolute copy number of CENP-A at centromeres. In addition, we provide novel insights in the mechanisms of centromere size control.

165

Chapter 4

RESULTS

Modification of endogenous CENP-A alleles in diploid human cells To determine absolute centromeric CENP-A levels in human cells we set out to build cell lines in which the entire CENP-A pool is fluorescent. To accomplish this, we removed a significant and essential portion of the CENP-A gene to create a knock-out allele in stably diploid retinal pigment epithelium (RPE) cells (Figure 4.1A, bottom). Subsequently, a fluorescent knock-in allele was created by placing GFP or YFP encoding sequences in frame with the sole remaining CENP-A gene (Figure 4.1A, middle). Specifically, we have built the following endogenously targeted RPE cell lines: CA+/-, CAG/-, CAY/-, and CA+/F [where + = wildtype; − = knock-out; G = GFP knock-in; Y = YFP knock-in; F = floxed (to control for potential gene- targeting artifacts); Figure 4.1-S1A]. Western blot analysis confirms that CAG/- and CAY/- cells exclusively contain tagged CENP-A (of ~43 kDa), while CA+/+ (wildtype), CA+/F, and CA+/- cells only express wildtype CENP-A (~16 kDa) protein (Figure 4.1B). Importantly, heterozygous expression or tagging of endogenous loci did not interfere with cell viability.

Figure 4.1 (next page) CENP-A levels are regulated by a mass action-like mechanism. (A) Schematic of gene- targeting strategy that allowed for the creation of CENP-A knockout and fluorescent knock-in alleles. The region encoding the essential CENP-A targeting domain [CATD (Black et al. 2007)] is indicated. (B) Quantitative immunoblots of CENP-A, HJURP, and Mis18BP1 in differentially targeted RPE cell lines. α-tubulin is used as a loading control. (C) Immunofluorescence images of same cell lines as in B. CENP-A intensity is represented in a heat map as indicated on the right. The fold difference ± SEM (n is biological replicates) compared to wildtype RPE cells is indicated below. Scale bar: 10 μm. Note that, in contrast to quantification of immunoblots, immunofluoresce detection of untagged and tagged CENP-A is directly comparable. (D) Quantification of centromeric CENP-A levels (from C) by immunofluorescence (IF) and total CENP-A levels (n = 4–9 independent experiments as in B) by western blot (WB). All cell lines expressing untagged CENP-A are normalized to CA+/+ while those expressing tagged CENP-A are normalized to the centromeric CAY/- levels measured in c, as indicated by dashed lines. (E) Correlation of centromeric and total cellular CENP-A levels as measured in D. Dashed line represents a predicted directly proportional relationship with indicated correlation coefficients. Throughout, the average ± SEM is indicated. (F) Quantification of centromeric CENP-A levels in synchronized HeLa cells (based on anti-CENP-A staining) within a single cell cycle after transient transfection of indicated proteins. Asterisk indicates statistically significant increase compared to control or indicated transfections (one-tailed t-test; p<0.05; n = 3); NS indicates no significant increase. Average ± SEM of three independent experiments is shown.

166

Counting CENP-A molecules in human cells

167

Chapter 4

Centromeric CENP-A levels are regulated by a mass action-like mechanism. While CENP-A is an essential and constitutive component of centromeres, how the size of the centromeric chromatin domain is controlled is not known. We analyzed the consequence of different CENP-A expression levels in the aforementioned RPE cell lines, as well as in a cells that ectopically overexpressed CENP-A-YFP (CAY/-+OE; Figure 4.1B). First, we measured the total protein pool of CENP-A by quantitative immunoblotting. While we found the detection output for CENP-A to be linear over at least a 32-fold range (Figure 4.2E), differences in protein transfer efficiencies do not allow for a comparison between proteins of different sizes, e.g. (GFP- or YFP-) tagged and untagged (wildtype) CENP-A (Figure 4.2—S3). Nevertheless, we could directly compare CAG/-, CAY/-, and CAY/-+OE cell lines and found that cellular CENP-A content spans a 6-fold range (Figure 4.1B, D). Given its essential role in centromere function, we predicted a tight control of centromeric CENP-A levels. However, instead of maintaining a fixed amount of CENP-A at centromeres, the levels varied extensively (Figure 4.1C). Both CA+/- and CAG/- cells, which contain a single intact allele, have decreased centromeric CENP-A levels, while the parental CA+/F cells maintain wildtype levels. Surprisingly, despite expressing CENP-A from a single allele, CAY/- cells have increased CENP-A levels, which may be due to adaptations that arose during the creation of this cell line. As expected, CENP-A levels are further elevated in CAY/-+OE cells (Figure 4.1C). Remarkably, we found a very high correlation (r2 = 84%) for a hypothetical directly proportional relationship between centromeric and total cellular CENP-A-GFP or -YFP levels (Figure 4.1D, E). Similarly, despite an only ~twofold range of expression, we still observe a high correlation with direct proportionality (r2 = 71%) for cells expressing untagged CENP-A (Figure 4.1D, E). Thus, our observations indicate that centromeric levels are determined by a mass action-like mechanism, where the amount of centromeric CENP-A varies in direct proportion with the cellular content.

168

Counting CENP-A molecules in human cells

An alternative hypothesis is that stable cell lines have undergone long- term adaptation to altered CENP-A expression, which has led to re- equilibrated centromeric levels. For example, proteins involved in CENP-A deposition at the centromere may have adapted to CENP-A expression levels. Indeed, we see a weak correlation between the levels of CENP-A and its histone chaperone HJURP (Barnhart et al. 2011; Dunleavy et al. 2009; Foltz et al. 2009) in our cells lines (Figure 4.1B, 4.1—S1B). Conversely, no correlation was detected for Mis18BP1 (Figure 4.1B, 4.1—S1C), another essential protein for CENP-A assembly (Fujita et al. 2007; Maddox et al. 2007), arguing that it is a non-stoichiometric component of the loading pathway. To test for long-term adaptation effects, we analyzed the consequence of CENP-A and/or HJURP overexpression in a single round of CENP-A assembly. Therefore, we transiently expressed CENP-A and/or HJURP and measured the level of centromeric CENP-A after one division in HeLa cells, which can be effectively synchronized in S phase using thymidine. While induction of CENP-A leads to a prompt increase in centromeric levels, no (additional) effect was observed by expression of HJURP (Figure 4.1F). Together, our results strongly suggest that centromeric CENP-A levels are directly regulated by its protein expression levels.

Centromeres contain ~400 molecules of CENP-A. To understand how CENP-A chromatin is self-propagated and nucleates the kinetochore, it is critical to establish the absolute amount of CENP-A present. In vertebrates, previous estimates range from a few tens of molecules [in chicken DT40 cells (Ribeiro et al. 2010)] to a potential maximum of tens of thousands [in HeLa cells (Black et al. 2007)]. To directly determine the copy number of CENP-A on human centromeres, we developed a 3D imaging strategy (Figure 4.2A), which was adapted from a method used to quantify cytokinesis proteins in fission yeast (Wu & Pollard 2005; Wu et al. 2008). In brief, we use a non-cell permeable dye (Figure 4.2A, I) to determine the 3D shape of cells (Figure 4.2A, II) and measure the

169

Chapter 4

total fluorescence within the entire cell volume (Figure 4.2A, III). Total cellular fluorescence of CAY/- cells (Figure 4.2A, III) was corrected for autofluorescence of wildtype RPEs (Figure 4.2A, IV), resulting in a measure of total YFP-derived signal. Next, centromere specific fluorescence was measured after correction for local background [Figure 4.2A, V; (Hoffman et al., 2001)] and axial oversampling. Importantly, fluorescence lifetime of CENP-A-YFP is similar between highly concentrated centromeric and diffuse cytoplasmic pools (Figure 4.2—S1), arguing that clustering does not lead to changes in fluorescence efficiency. In effect, our 3D integrated fluorescence strategy measures the fraction of centromeric-to-total CENP-A. We find that while CENP-A is enriched at centromeres, on average only 0.44% of cellular CENP-A is present per centromere in CAY/- cells (Figure 4.2B). Very similar fractions were observed in CAG/- and CAY/-+OE cells (0.38% in both cases; Figure 4.2C, 4.2-S2A, B), which provides an additional line of evidence in support of a mass action-like mechanism for CENP-A assembly. Furthermore, these findings show that a surprising minority, about one-fifth of the CENP-A protein content (0.44% x 46) is present on the functionally relevant subcellular location, i.e. at the centromeres. To convert centromeric fractions to absolute amounts, we determined the total number of CENP-A molecules in RPE cells. We prepared whole cell extracts of RPE cells and analyzed these alongside highly purified recombinant CENP-A/H4-complexes of known concentration by quantitative immunoblotting (Figure 4.2D). Importantly, we ensured that

Figure 4.2 (next page) Human centromeres contain 400 molecules of CENP-A. (A) Schematic outline of strategy allowing for the quantification of the centromeric fraction of CENP-A compared to the total cellular pool. Scale bars: 5 μm. (B) Quantification of the centromeric fraction of CENP-A in CAY/- cells. Each circle represents one centromere; circles on the same column are individual centromeres from the same cell. Dashed line indicates average of all centromeres. (C) Quantification of the centromeric fraction of CENP-A in indicated cell lines. Each square represents the average centromeric signal from one cell; squares on the same column are individual cells from the same experiment (Exp). Figure 4.2-S2 shows quantification of individual centromeres in CAG/- and CAY/-+OE cells. (D) Representative quantitative immunoblot of purified recombinant CENP-A and endogenous CENP-A from whole cell extracts (WCE). (E) Quantification of D. Solid line represents the best fit linear regression. Dashed line represents the amount of CENP-A from 150,000 cells. (F) Quantification of total cellular CENP-A copy number. Each diamond represents one replicate experiment; measurement from E is indicated as a grey diamond. (G) Calculation of average CENP-A copy number per centromere (CEN) in wildtype RPE cells. Throughout, the average ± SEM is indicated.

170

Counting CENP-A molecules in human cells

171

Chapter 4

recombinant and cellular CENP-A have the same transfer efficiency and can be directly compared to each other (Figure 4.2—S3). Fitting the cellular amount of CENP-A onto a linear regression curve of purified protein (Figure 4.2E) shows that CA+/+ cells contain an average of ~9.1 ± 1.1·104 (n = 10) molecules of CENP-A per cell (Figure 4.2F). Because the centromeric fraction of CENP-A is fixed, we can calculate the absolute amount of CENP-A per centromere (Figure 4.2G, 4.2—S2c) and show that wildtype RPE cells contain ~400 molecules of CENP-A on an average centromere. Both the expression and centromeric loading of CENP-A are cell cycle regulated (Figure 4.3A). In human cells, cellular protein levels of CENP-A peak in late G2 (Shelby et al., 2000), while centromere assembly occurs in early G1 phase (Jansen et al., 2007). Thus it is possible that part of the cell- to-cell variation of the centromeric CENP-A ratio observed in Figure 4.2C is due to differing cell cycle stages. We tested this by using the previously developed fluorescent ubiquitin-based cell cycle indicator (FUCCI), which can be used in live cells (Sakaue-Sawano et al., 2008). In particular, we used hCdt1(30/120)-RFP, which is expressed ubiquitously throughout the cell cycle, but specifically degraded in S, G2, and M phases (Sakaue-Sawano et al., 2008). As a result, protein levels increase as cells enter and progress through G1 phase, peak at the G1/S boundary, and then drop until cells reenter G1 (Figure 4.3A). We expressed this protein in CAY/- cells and tracked the RFP signal intensity over time (Figure 4.3B, 4.3-S1A) to identify cells that entered S phase (see methods for details). We compared their ratio of centromeric-to-total CENP-A to randomly cycling cells and found that neither the mean nor the variance differs significantly between these two populations of cells (Figure 4.3C). Importantly, expression of the FUCCI marker itself has no effect on the measurements performed (Figure 4.3— S1B). While the centromeric fraction of CENP-A is likely low in G2 phase and high just after assembly in early G1, we find that the variation observed in Figure 4.2C is not a consequence of such cell cycle induced effects and may instead reflect inherent variation between cells.

172

Counting CENP-A molecules in human cells

Figure 4.3 Centromeric CENP-A levels are equivalent between S phase and randomly cycling cells. (A) Cartoon depicting changes in cell morphology and nuclear levels of hCdt1(30/120)-RFP (in red) throughout the cell cycle (Sakaue-Sawano et al., 2008). Approximate timing of CENP-A expression (Shelby et al., 2000) and centromeric loading (Jansen et al., 2007) are indicated in orange and blue, respectively. The stage at which cells were analyzed to measure the centromeric fraction of CENP-A is indicated in green. (B) An example trace of a cell entering S phase (indicated by a sudden decrease in RFP levels) is shown. The centromeric fraction of CENP-A was measured at this point as outlined in Figure 4.2A. Peak expression is normalized to 100 and background fluorescence to 0. Micrographs of hCdt-1(30/120)- RFP at indicated timepoints are shown below. (C) As in Figure 4.2C. Orange squares represent cells that have passed the G1-S transition point, as indicated by decreasing levels of hCdt-1(30/120)-RFP. Grey squares represent randomly cycling cells. No statistically significant differences (NS) were observed between randomly cycling cells and S phase cells.

Although the method we employed to measure centromeric ratios is internally controlled, it relies on measurement of integrated fluorescence of whole cells, including highly dilute cytoplasmic CENP-A. To exclude potential errors in measurements of low protein concentration, we stably expressed H2B-RFP in CAY/- cells (Figure 4.4A, inset) and determined that 0.73% of nuclear CENP-A is present on each centromere (Figure 4.4A). In

173

Chapter 4

addition, low salt fractionation experiments indicate that ~74% of cellular CENP-A co-pellets with other chromatin components in CAY/-+H2B-RFP cells (Figure 4.4B), indicating that this represents the stable nuclear pool. Combined, we find a similar number of CENP-A molecules per centromere when analyzing the nuclear pool (492 molecules; Figure 4.4C) as when measuring total cellular CENP-A. This argues that the measurements performed above are not significantly influenced by a potential inaccuracy in determining the cytoplasmic pool. Interestingly, it has recently been shown that detectable levels of CENP-A are assembled into non-centromeric chromatin of HeLa cells (Lacoste et al., 2014). Indeed, we now find that, at least in RPE cells the proporation of chromatin bound CENP-A outside of the centromere is surprisingly high (~66% in this cell line).

Figure 4.4 Measurement of nuclear CENP-A confirms centromeric copy number. (A) As in Figure 4.2B, except that the centromeric fraction compared to total nuclear pool is indicated. Inset shows a representative image of a CAY/-+H2B- RFP cell (scale bar: 2.5 μm). (B) Quantitative immunoblot showing the soluble fraction and a dilution series from the insoluble fraction of CENP-A-YFP in CAY/-+H2B-RFP cells (left). Tubulin is used as marker for the soluble fraction and H4K20me2 [exclusively found in chromatin (Karachentsev et al. 2007)] for the insoluble fraction. Quantification of insoluble fraction of CENP-A is shown to the right. (C) Calculation of the average CENP-A copy number per centromere (CEN) in wildtype RPE cells, based on results from CAY/-+H2B-RFP cells.

174

Counting CENP-A molecules in human cells

CENP-A copy number confirmed by three independent methods. To further validate the strategy for measuring CENP-A copy numbers, we used two additional independent quantification methods. First, we applied a method that employs the statistical properties of fluorescence redistribution (Rosenfeld et al., 2005, 2006). This method relies on the fact that random segregation leads to each daughter receiving an (unequal) fraction of molecules, where the distribution of differences relates to the total number of molecules (as outlined in Figure 4.5A). During mitosis, sister centromeres form individually resolved spots by light microscopy, allowing us to measure the fluorescence intensity of individual sisters (Figure 4.5B). We find that rather than accurately segregating exactly half of pre-assembled CENP-A onto each daughter chromatid, the difference between sister centromeres follows a random distribution (Figure 4.5B, C). Previously, Rosenfeld et al. (2005, 2006) have provided mathematical evidence that measurements of this deviation allow for the determination of the fluorescence intensity of a single heritable, segregating unit (Figure 4.5A). We measured an average of 75.4 segregating units of CENP-A-GFP per centromere in CAG/- cells (Figure 4.5D). Because each segregating unit consists of one or more nucleosomes, containing 2 molecules of CENP-A each (Bassett et al. 2012; Hasson et al. 2013; Sekulic et al. 2010; Tachiwana et al. 2011; Padeganeh et al. 2013), an average CAG/- centromere has a minimum of 150.8 molecules of CENP-A. Correcting the amount of CENP-A measured in CAG/- cells for wildtype levels (Figure 4.1C) results in ≥377 molecules of CENP-A per centromere (Figure 4.5D, right y-axis). Importantly, these measurements differ significantly if random centromere pairs are chosen for which no statistical correlation exists (Figure 4.5—S1E). This confirms that fluorescence intensities at sister centromeres co-vary and renders this type of analysis suitable for centromere quantification. Stochastic fluctuation measurements in CAY/- and CAY/- +OE cells indicates that wildtype cells contain ≥188 and ≥149 CENP-A molecules per centromere, respectively (Figure 4.5—S1A–D). Importantly, the number of co-segregating CENP-A nucleosomes is unknown, which can

175

Chapter 4

be one or more. Therefore, despite the variation between the cell lines used here, all results obtained from this method provide a minimum estimate of the centromeric CENP-A copy number that is in agreement with the 400 centromeric molecules of CENP-A measured above (Figure 4.2G). Next, we used a yeast strain that harbors a chromosomally integrated 4kb LacO-array and expresses GFP-LacI as a calibrated fluorescent standard (Lawrimore et al. 2011). While there is a potential for 204 molecules of GFP-LacI to be bound to this array (Lawrimore et al. 2011), it is unlikely that the entire array is fully occupied at any moment. Because CAG/- cells express the same version of GFP as this yeast strain, direct comparison of fluorescent foci (Figure 4.5E) provides a maximum estimation of the centromeric CENP-A-GFP copy number. In this way, we determined that CAG/- centromeres contain at most 215 ± 32 CENP-A-GFP molecules, which translates to ≤538 CENP-A molecules in wildtype cells (Figure 4.5F). Importantly, the copy number that we measure directly by our 3D integrated fluorescence approach is in close agreement with minimum and maximum estimates of the stochastic fluctuation and fluorescent standard approaches, respectively (Figure 4.5G). This provides confidence that 400 molecules of CENP-A per centromere in wildtype RPE cells is an accurate measure.

Figure 4.5 (next page) Independent quantification methods confirm centromeric CENP-A copy number. (A) Stochastic fluctuation method: Cartoon depicting inheritance and random redistribution of parental CENP-A nucleosomes onto sister chromatids during DNA replication. A hypothetical distribution of the absolute difference between the two sister centromeres, as well as the formula for calculating the fluorescence intensity per segregating unit (α) are indicated on the right. (B) Representative image of mitotic CENP-A-YFP expressing cell. CENP-B staining allows for identification of sister centromeres. Blowup to the right represents a single slice of the boxed region showing that CENP-B is located in between the CENP-A spots of sister centromeres. (C) Frequency distribution of the difference between CENP-A-GFP intensity of sister centromeres in CAG/- cells. (D) Quantification of centromeric CENP-A-GFP based on the stochastic fluctuation method. Each circle represents one centromere; circles on the same column are individual centromeres from the same cell. Left y-axis indicates segregating CENP-A-GFP units in CAG/- cells; right y- axis shows the conversion to minimum number of centromeric CENP-A molecules in CA+/+ (WT) cells. (E) Fluorescent standard method: Representative fluorescence images of 4kb-LacO, LacI-GFP S. cerevisiae and human CAG/- cells. (F) Quantification of fluorescence signals of LacI-GFP and CENP-A-GFP spots (n = 2 biological replicates). The left y-axis indicates the fluorescence intensity normalized to LacI-GFP; the right y-axis shows the conversion to maximum number of centromeric CENP-A molecules in wildtype cells. (G) Comparison of independent measurements for the centromeric CENP-A copy number [corrected for CA+/+ levels; Stoch. fluctuations = stochastic fluctuation method (Figure 4.5A); Integr. fluorescence = integrated fluorescence method (Figure 4.2A)]. Levels from all strategies are corrected for CA+/+ (WT) levels. Throughout, the average ± SEM and scale bars of 2.5 μm are indicated.

176

Counting CENP-A molecules in human cells

177

Chapter 4

Assessing the critical number of CENP-A nucleosomes. While cells are able to survive with a 6-fold range of CENP-A levels (Figure 4.1D), centromere function may be compromised when levels are not accurately maintained. Indeed, based on a conserved ratio of centromere and kinetochore proteins and kinetochore microtubules between multiple yeast species as well as chicken DT40 cells, it has been hypothesized that centromeres form modular structures by repeating individual structural subunits (Joglekar et al., 2008; Johnston et al., 2010), as originally proposed by Zinkowski et al (1991). Thus, the amount of CENP-A would directly reflect the number of downstream centromere and kinetochore proteins and microtubule attachment sites. Conversely, experiments in human cells indicate that the centromere is assembled by multiple independent subcomplexes (Foltz et al., 2006; Liu et al., 2006). Here, we analyzed whether altering the levels of CENP-A has an effect on the recruitment of other, downstream centromere or kinetochore proteins. Both CENP-C and CENP-T rely on CENP-A for their centromeric recruitment (Fachinetti et al. 2013; Liu et al. 2006; Régnier et al. 2005) and have recently been shown to be responsible for mitotic recruitment of the KMN network (Gascoigne et al. 2011), including the key microtubule binding protein Hec1/NDC80 (Cheeseman et al. 2006; DeLuca et al. 2006). Interestingly, we found that none of these three proteins were significantly affected by altering the levels of CENP-A between 40% and 240% of wildtype levels (Figure 4.6A, 4.6—S1). In line with previous findings (Fachinetti et al. 2013; Liu et al. 2006), these results argue against a modular centromere architecture where CENP-A nucleosomes form individual binding sites for downstream components. Rather, a >2½ -fold excess of CENP-A appears to be present for recruitment of centromere and kinetochore complexes of fixed pool size.

178

Counting CENP-A molecules in human cells

Figure 4.6 Reduction of CENP-A leads to a CENP-C, CENP-T, and Hec1 independent increase in micronuclei. (A) Quantification of centromeric CENP-A (from Figure 4.1), CENP-C, CENP-T, and Hec1 levels for indicated cell lines; n = 4 independent experiments in each case. Note that cell lines carrying tagged CENP-A have a slight, yet non-significant impairment in recruiting CENP-C, CENP-T, and Hec1. However, this does not correlate with the CENP-A levels themselves. Below, representative images of indicated antibody staining from CA+/+ cells are shown. Representative images from all cell lines can be found in Figure 4.6—S1. (B) Quantification of the fraction of cells containing micronuclei (MN) for indicated cell lines. Asterisk indicates statistically significant increase compared to wildtype [paired t-test; p<0.05; n = 3–4 independent experiments (500–3000 cells per experiment per cell line)]; NS indicates no significant difference. Throughout, the average ± SEM is indicated and dashed lines represent wildtype levels. Scale bars: 5 μm.

Intriguingly, despite no quantitative effect on centromeric proteins, we observed that decreasing CENP-A levels leads to an increase in the fraction of cells containing micronuclei (MN; Figure 4.6B). MN often arise as a consequence of mitotic errors, such as lagging chromosomes during anaphase (Ford et al., 1988), breakage of anaphase bridges (Hoffelder et al., 2004), or multipolar mitoses (Utani et al., 2010). The presence of MN can be scored by DAPI staining (Figure 4.6B, bottom). MN are found in WT cells at a baseline fraction of ~0.5% (Figure 4.6B). Both cell lines with decreased CENP-A levels show a significantly increased fraction of cells with MN. Importantly, these two cell lines were derived independently from the CA+/F cell line (Figure 4.1—S1A), which has wildtype levels of CENP-A and no

179

Chapter 4

significant increase in MN (Figure 4.6). In addition, neither cell line with increased CENP-A levels have a larger fraction of MN than CA+/F cells. While the essential role for CENP-A in centromere function is well established (Black et al., 2007; Liu et al., 2006; Régnier et al., 2005), our results indicate that a critical level of CENP-A is passed after reducing the levels to 50%. However, the molecular mechanism responsible for MN formation remains unclear, as downstream centromere and kinetochore components of CENP-A remain unaffected.

The contribution of cell type and local centromere features to centromeric CENP-A levels. Interestingly, we find that not all centromeres of the same cell have equal amounts of CENP-A (Figure 4.5D). This could either be due to in cis features driving differential regulation of CENP-A on individual centromeres, or by unbiased stochastic effects at centromeres. To distinguish between these possibilities, we measured the centromeric levels of endogenous CENP-A on specific chromosomes. First, we analyzed a monoclonal HCT-116 cell line that has an integrated Lac-array in a unique position in the genome (Thompson & Compton 2011). While the site of integration is unknown, expressing LacI-GFP allows for the identification of the same chromosome in a population of cells (Figure 4.7A). Both the average and variance of CENP-A at this centromere does not differ statistically from the bulk (Figure 4.7B, 4.7—S1A), arguing against centromere specific features driving CENP- A levels on the Lac-marked chromosome. Conversely, we found that the Y- centromere, uniquely identified by the lack of CENP-B [Figure 4.7C; (Earnshaw et al., 1987)], of two independent male cell lines had a slight yet significant reduction of CENP-A (19% in wildtype HCT-116 and 13% in DLD-1; Figure 4.7D, 4.7—S1B, C), consistent with an earlier report (Irvine et al. 2004). Finally, we used a human patient derived fibroblast cell line (PDNC-4) where one centromere of chromosome 4 has repositioned to an atypical location (Amor et al. 2004), which we designate as NeoCEN-4

180

Counting CENP-A molecules in human cells

(Figure 4.7E). As has been observed in other cell lines derived from this patient (Amor et al. 2004), we found that the NeoCEN-4 has a ~25% decrease in centromeric CENP-A (Figure 4.7F 4.7—S1D). Taken together, these results show that while CENP-A expression drives centromeric levels, local sequence or chromatin features can also contribute to the average amount of CENP-A at specific centromeres. Nevertheless, even on these centromeres, the variance in CENP-A levels is maintained, indicating that other stochastic processes contribute to CENP-A levels. Next, to determine whether the CENP-A copy number of our model cell line is representative for functionally different cells, we performed comparative immunofluorescence against CENP-A (Figure 4.7G). We analyzed four different cancer cell lines (HeLa, U2OS, HCT-116, and DLD-1), as well as the PDNC-4 neocentromere cell line discussed above and primary human fibroblasts that were cultured for a limited number of passages (<15) since their isolation from a patient (Figure 4.7G). Using these cell lines, we found a 6-fold range of centromeric CENP-A levels (Figure 4.7H), indicating that there is substantial variance between different cell lines. However, we find that the primary cells have a similar amount of CENP-A as RPEs (Figure 4.7H), arguing that our measure of absolute CENP-A copy numbers made in RPE cells is relevant for healthy, human tissues as well. We combined these results with our measurements of individual centromeres and determined that, while an average centromere in PDNC-4 cells contains ~579 molecules of CENP-A, the NeoCEN-4 only contains ~432. Average Y-centromeres contain ~143 or ~87 molecules in HCT-116 and DLD-1 cells, respectively (Figure 4.7J). In conclusion, we find evidence that cis-elements can have an effect on CENP-A levels, at least on human Y- and neocentromeres.

181

Chapter 4

182

Counting CENP-A molecules in human cells

Figure 4.7 (previous page) Centromere specific distribution of CENP-A. (A, C, E) Representative image of mitotic spreads for LacI-GFP::LacO expressing HCT-116 cells (A); wildtype HCT-116 cells (C); and PDNC-4 cells (E). Blowups show the chromosome containing the integrated Lac-array (A); the Y-chromosome (CENP-B negative; outlined) as well as a CENP-B positive autosome (C); and the neocentric chromosome 4, containing 2 pairs of ACA spots (staining both CENP-A and CENP-B), but only 1 pair of CENP-A spots (E). (B, D, F) Quantification of CENP-A levels on the centromere of the chromosome containing the Lac-array [CEN-Lac; n = 29; (B)]; the Y-chromosome [CEN-Y; n = 18; (D)]; and neocentric chromosome 4 [NeoCEN-4; n = 39; (F)] of indicated cell lines compared to all other centromeres within the same cell (Other CENs; n = 1008, 620, and 1592, respectively). Median (line), interquartile distance (box), 3x interquartile distance or extremes (whiskers), and outliers (spots) are indicated. Figure 4.7—S1 shows results of individual centromeres. Asterisk indicates statistically significant difference (t-test; p<0.05); NS indicates no significant difference. (G) Representative images of CENP-A antibody staining in indicated cell types; independent images of RPEs are shown as reference. (H) Quantification of G. Mean ± SEM for n = 3–4 independent experiments is shown. Left y- axis represents centromeric CENP-A levels normalized to RPE cells; right y-axis shows number of CENP-A molecules per centromere (CEN). (J) Combined results from a-h allow for the determination of CENP-A copy numbers on individual chromosomes. (K) Statistical map of the distribution of 216 CENP-A nucleosomes on the NeoCEN-4 at three different scales. The top 216 peaks are indicated in blue. Y-axis indicates the probability of CENP-A occupancy for each nucleosome. (L) Histogram of the CENP-A nucleosome occupancy. Inset shows the distribution of 216 neocentric CENP-A nucleosomes on the 10% highest occupancy peaks (green) and 90% lowest occupancy peaks (red).

A statistical map of CENP-A at individual nucleosome positions. The number of CENP-A nucleosomes we find at individual centromeres is much smaller (~25-fold, see Figure 4.8A) than the total number of nucleosome positions on human centromeric DNA. This indicates that either CENP-A is randomly distributed at a low level throughout the centromere domain or that it occupies specific “hotspots”. However, it is not possible to map individual CENP-A nucleosomes on canonical centromeres, due to their repetitive nature. However, a recent high-resolution ChIP-seq analysis of the (non-repetitive) NeoCEN-4 identified 1113 unique CENP-A nucleosome positions spanning a ~300 kb locus (Hasson et al. 2013). By combining the relative height of individual peaks with the total number of CENP-A nucleosomes at this neocentromere, we were able to determine the fraction of cells containing CENP-A at each nucleosome position (Figure 4.7K). This statistical map of CENP-A occupancy shows that, while the median is ~6% (Figure 4.7L), individual positions feature CENP-A with a surprisingly high occupancy (up to 80% of all cells; Figure 4.7K, arrow). Remarkably, more than one third of all CENP-A nucleosomes are located on the top 10% potential positions (Figure 4.7L, inset). This strongly suggests that, at least on the NeoCEN-4, a number of nucleosome positioning sequences exist that strongly favor CENP-A over other H3 variants.

183

Chapter 4

DISCUSSION

It has been proposed that centromeres in budding yeast feature a single nucleosome of CENP-A (Furuyama & Biggins 2007; Meluh et al. 1998). For this reason, yeast centromeres have been extensively used to calibrate fluorescence intensities of CENP-A and other proteins from a number of species (Joglekar et al., 2006, 2008; Johnston et al., 2010; Schittenhelm et al., 2010). However, the single nucleosome hypothesis has recently been challenged (Coffman et al., 2011; Haase et al., 2013; Lawrimore et al., 2011). To avoid dependency on any single reference, we used three independent methods to measure the human centromeric CENP-A copy number. One strategy uses intrinsically controlled fluorescence ratios of cellular and centromeric CENP-A-YFP signals (Figure 4.2A). The second method does not rely directly on fluorescence intensities, but rather on the stochastic redistribution of CENP-A (Figure 4.5A). Finally, we compared CENP-A signals directly to a calibrated fluorescent standard (Figure 4.5E). Despite the independent nature of these strategies, they all come to a very similar conclusion. Thus, we demonstrate that typical centromeres in human RPE cells contain ~400 molecules of CENP-A. While there is some debate on the composition of CENP-A nucleosomes (Black & Cleveland 2011; Henikoff & Furuyama 2012), current evidence strongly favors a canonical arrangement harboring two copies of CENP-A (Bassett et al., 2012; Hasson et al., 2013; Padeganeh et al., 2013; Sekulic et al., 2010; Tachiwana et al., 2011). Hence, our numbers, correspond to 200 CENP-A nucleosomes in interphase, which are split into 100 nucleosomes on mitotic chromosomes (Figure 4.8B). Epigenetic centromere inheritance is achieved by quantitative redistribution of CENP-A across cell division cycles (Bodor et al., 2013; Jansen et al., 2007). We find that rather than ensuring that each daughter receives exactly half, redistribution of CENP-A is random (Figure 4.5B, C). Because this regulation has the potential for individual centromeres to stochastically inherit critical levels of CENP-A, the steady state must be sufficiently high to avoid chromosome loss. Although the critical amount of CENP-A is not

184

Counting CENP-A molecules in human cells

known, HeLa cell viability is lost if CENP-A levels are reduced to 33% (Black et al., 2007), i.e. 44 nucleosomes (see Figure 4.7H). Conversely, we show here that CAG/- cells are viable at 40% of RPE levels (80 nucleosomes). Consequently, we estimate that the critical number of nucleosomes that must be inherited, which is half of the steady state level and replenished during G1 phase, lies between 22 and 40.Using these values, we calculated the chance that a cell inherits critically low levels of CENP-A on any of its centromeres (Figure 4.8C). We demonstrate that at a steady state of 200 nucleosomes per centromere, less than one in 1016 cell divisions will give rise to a centromere containing 40 CENP-A nucleosomes or less (Figure 4.8C, left). Thus, the odds of inheriting a critical amount of CENP-A at wildtype steady state levels is negligible. Conversely, with 100 nucleosomes at steady state, the chance of a chromosome inheriting even the most stringent critical level of 22 nucleosomes is close to 10-6 (Figure 4.8C, right), which may pose a significant problem e.g. during development of an organism. Conversely, although critical levels would be reached even less frequently if centromeres contained a steady state of e.g. 300 CENP-A nucleosomes, this degree of accuracy may be superfluous and not outweigh the cost of maintaining a large centromere size. Therefore, we argue that the number of CENP-A molecules found on human centromeres is optimized for robust epigenetic inheritance and centromeric function. Previously, it has been shown that CENP-A is interspersed with both H3.1 and H3.3 at the centromere (Blower et al., 2002; Dunleavy et al., 2011; Ribeiro et al., 2010; Sullivan and Karpen, 2004; Sullivan et al., 2011). Indeed, based on the average size of the centromeric chromatin domain we estimate that 200 CENP-A nucleosomes represent only ~4% of all centromeric nucleosomes (see Figure 4.8A for calculation). Surprisingly, we find that the majority of chromatin bound CENP-A is located outside the centromere. Indeed, a recent study found that a proportion of CENP-A containing nucleosomes also exist in non-centromeric chromatin of HeLa cells, and is assembled by DAXX, a major chaperone of histone H3.3

185

Chapter 4

Figure 4.8 A quantitative view of human centromeric chromatin. (A) Estimated ratio and distribution of CENP-A (red) and H3 (grey) at the centromere and on non-centromeric loci (genome) in interphase cells. Estimations assume 2 CENP-A molecules per nucleosome (Hasson et al. 2013; Bassett et al. 2012; Sekulic et al. 2010; Tachiwana et al. 2011; Padeganeh et al. 2013); 200 bp nucleosome spacing; 2.5 x106 bp centromere domain (Lee et al. 1997; Sullivan et al. 1996), 40% of which contains CENP-A (Sullivan et al. 2011); 6 x109 bp diploid genome, 200 CENP-A nucleosomes per centromere; 2.5 x104 CENP-A nucleosomes outside of centromeres [9.1 x104 molecules per cell (Figure 4.2F), of which 74% is in chromatin (Figure 4.4B) and 0.44% at each centromere (Figure 4.2B)]. The centromeric, non-centromeric chromatin, and unincorporated fractions of CENP-A are indicated in green, blue, and black, respectively. (B) On average, ~100 CENP-A nucleosomes are present per mitotic centromere due to redistribution onto replicated sister chromatids (Bodor et al. 2013; Jansen et al. 2007), although the exact number depends on the available total pool. Excess CENP-A could either lead to an increased CENP-A domain or to a higher density of CENP-A within a domain of fixed size. This pool forms an excess to recruit downstream centromere and kinetochore complexes and ultimately provides microtubule binding sites for ~17 kinetochore microtubules (McEwen et al. 2001). To avoid mitotic errors, a critical amount of CENP-A is required (dashed lines). (C) Graph representing the chance of at least one centromere in a cell (with 46 chromosomes) reaching critically low levels of CENP-A by random segregation of pre-existing CENP-A nucleosomes. Calculations were performed for varying levels of critical nucleosome numbers at a fixed steady state of 200 (left), or by varying the steady state number at a fixed critical level of 22 (right). Red bars are identical calculations.

186

Counting CENP-A molecules in human cells

(Lacoste et al., 2014). In addition, detectable levels of non-centromeric CENP-A have been observed in budding yeast (Camahort et al., 2009) and chicken DT40 cells (Shang et al., 2013). Here, we quantify this pool in human RPE cells and while there is more than twice as many non- centromeric CENP-A nucleosomes than there are centromeric ones, this only represents <0.1% of all nucleosomes in the genome and thus CENP-A is ~50-fold enriched (per unit length of DNA) at centromeres (Figure 4.8A). This result may explain how, despite being outnumbered 25:1 by other H3 variants at the centromere, CENP-A can still accurately specify the centromeric locus. This hypothesis may be tested by creating artificial CENP-A binding sites (e.g. using the LacO/LacI system) of different known sizes and determining the threshold at which centromeres can be formed. Interestingly, the study by Lacoste and co-workers showed that the extra-centromeric CENP-A is not randomly distributed, but enriched at sites of high histone turnover (Lacoste et al., 2014). Our finding that CENP-T, CENP-C, and Hec1 do not quantitatively correlate with CENP-A levels (Figure 4.6A) argues that not each (non-centromeric) CENP-A nucleosome is able to recruit downstream centromere components. It would be interesting to determine to what extent other centromere and kinetochore proteins are present throughout the genome and whether they are also enriched at extra-centromeric CENP-A hotspots. This question is particularly relevant since it has been observed that downstream centromere components may affect centromeric CENP-A levels (Carroll et al., 2009, 2010; Hori et al., 2013; Okada et al., 2006). A critical combination of components at such ‘hotspots’ may trigger neocentreomere formation, the mechanisms of which are still unresolved. Previously, it has been observed that at very high levels of overexpression, CENP-A ceases to be centromere restricted (Gascoigne et al., 2011; Heun et al., 2006; Van Hooser et al., 2001). Instead, here we show that within a 6-fold range of expression levels, the CENP-A loading machinery has a constant efficiency, which maintains a strict ratio between

187

Chapter 4

the centromeric and total pools of CENP-A. Thus, within a physiological range, centromeric CENP-A levels are regulated by a mass action-like mechanism, where the loading efficiency is independent of the expression levels. This mechanism ensures that with fluctuating expression levels, CENP-A remains mainly centromere restricted, and may prevent potential neocentromere seeding. Remarkably, varying the amount of CENP-A at centromeres during perpetual growth in culture does not affect the levels of several other centromeric proteins. One possible explanation for this is that there is a fixed subset of ‘active’ CENP-A nucleosomes that is responsible for recruiting downstream components, even if the total amount of CENP-A is variable. Alternatively, an excess of CENP-A could form a chromatin domain that provides a ‘platform’ for recruitment of a centromere complex of fixed size. Surprisingly, however, we find that a critical amount of CENP-A for prevention of micronuclei is reached even before downstream centromere and kinetochore protein levels are affected (Figure 4.6, 4.8B). Our analysis indicates that the distribution of CENP-A among centromeres within one cell is generally uniform. However, in agreement with prior publications, we find that both the Y-centromere as well as a human neocentromere have lower CENP-A levels (Amor et al. 2004; Irvine et al. 2004). Interestingly, both these centromere types are atypical in that they are formed on relatively small genomic loci: ~600 kb for the Y- centromere (Abruzzo et al., 1996) and ~300 kb for the NeoCEN-4 (Hasson et al., 2013), whereas autosomes and the X-chromosome have alpha- sattellite arrays of several magabases in size (Lo et al., 1999; Mahtani and Willard, 1990; Wevrick and Willard, 1989). In addition, in contrast to canonical centromeres, neither the Y-centromere nor neocentromeres recruit the sequence specific DNA binding protein CENP-B (Amor et al., 2004; Earnshaw et al., 1987), which has been hypothesized to alter the 3D structure of centromeric chromatin (Pluta et al., 1992). The presence of CENP-B binding sites has recently been shown to have a role in phasing

188

Counting CENP-A molecules in human cells

CENP-A nucleosomes (Hasson et al., 2013), and to cooperate with CENP-A in kinetochore function (Fachinetti et al., 2013), and may therefore be involved in regulation of centromeric CENP-A levels as well. Furthermore, high resolution analysis of a human neocentromere reveals a non-random distribution of CENP-A (Hasson et al. 2013), where individual nucleosome positions are occupied in anywhere between 0.5% to 80% of cells (Figure 4.7K, L). Thus, despite specific DNA sequences being neither sufficient nor required for centromere identity (Amor et al., 2004; Earnshaw and Migeon, 1985; Marshall et al., 2008; Voullaire et al., 1993), the amount of CENP-A at centromeres likely results from a combination of a systematic cellular mechanism with a contribution of local sequence or chromatin features. In conclusion, several key mechanistic insights follow from our findings. First, while CENP-A nucleosomes are highly enriched at the centromere, most CENP-A is distributed at low levels throughout chromatin. This indicates that there is no exclusive pathway that restricts CENP-A assembly to centromeres. Nevertheless, we propose that the ample number of CENP-A nucleosomes facilitates a robust epigenetic signal that can absorb fluctuations in CENP-A inheritance and assembly in order to faithfully maintain centromere identity. Secondly, the requirement for a sizable number of CENP-A nucleosomes to perpetuate an active centromere reduces the likelihood for inadvertent detrimental neocentromere seeding without the need for a tightly restricted assembly mechanism. In addition, the fixed ratio between total and centromeric CENP-A levels may prevent excess CENP-A from accumulating at high density at non-centromeric loci, thus further reducing the probability of neocentromere formation. Finally, the number of centromeric CENP-A nucleosomes represents an ample pool of which only a subset is required to nucleate otherwise self-organized centromere and kinetochore complexes. In summary, from our analysis an integrated view of centromeric architecture, size, and regulation emerges (Figure 4.8) that provides a basis to explain the self-propagating nature of the epigenetic centromere.

189

Chapter 4

MATERIAL & METHODS

Cell culture and construction.

All human cell lines used were grown at 37°C, 5% CO2. Cells were grown in DMEM/F-12 (RPE), DMEM (HeLa, U2OS, PDNC-4), MEM (primary fibroblasts; Coriell GM06170), McCoy’s 5A (HCT-116), or RPMI-1640 (DLD-1) cell culture media. Media were supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 1 mM sodium pyruvate (SP), 100 U/ml penicillin and 100 μg/ml streptomycin, with the following exceptions: for RPE cells SP was substituted for 14.5mM sodium bicarbonate; for HeLa newborn calf serum was used instead of FBS; for fibroblasts 15% FBS was used; for DLD-1 cells SP was omitted; and both SP and glutamine were omitted for HCT-116 cells. During live cell imaging, culture medium was replaced with Leibowitz’s L-15 medium containing 10% FBS and 2 mM glutamine. LacI-GFP::LacO HCT-116 cells [gift from Duane Compton (Thompson & Compton 2011)] were selected alternatingly with 2 μg/ml blasticidin and 300 μg/ml hygromycin; PDNC-4 cells were selected with 100 μg/ml neomycin. All media and supplements were purchased from Gibco.

All targeted cell lines are derived from wildtype hTERT RPE (CA+/+). Gene targeting was achieved by adeno-associated virus (AAV) mediated delivery of targeting constructs essentially as described (Berdougo et al. 2009), except in the case if CAG/-cells (see below). The CA+/F cell line was created by inserting loxP sites surrounding CENP-A exons 2 and 4 as described previously (Fachinetti et al. 2013). The CA+/- cell line was created by targeting CA+/F cells with a construct lacking 1373 bp of the CENP-A gene (from 43 bp upstream of exon 2 to 134 bp downstream of exon 4) including the essential CENP-A targeting domain (Black et al. 2007). CAY/- cells were created by sequential targeting of a first CENP-A allele with the targeting construct inserting loxP sites flanking exon 3 and 4 as described above and the second allele by targeting EYFP (carrying citrine and monomerization mutations: Q69M, A206K) in frame with the CENP-A gene, immediately

190

Counting CENP-A molecules in human cells

prior to the stop codon in exon 4. The floxed allele was subsequently removed by retroviral delivery of HR-MMPCreGFP, a “Hit and Run” Cre vector, as described (Silver & Livingston 2001). CAG/- cells were created from an independent CA+/- clone where the remaining intact CENP-A allele was targeted with EGFP using a FACS-based strategy that we developed previously (Mata et al. 2012). Targeting resulted in insertion of the EGFP ORF directly downstream the last coding sequence in exon 4, just upstream of the endogenous stop codon, without insertion of any selectable marker gene. CAY/-+OE cells were created by stable transfection of and selection (5 μg/ml blasticidin) for a CENP-A-YFP expression vector (pBOS-Blast) bearing a CENP-A-YFP fusion protein identical to that of the endogenous knockin locus) in CAY/- cells. CAY/-+H2B-RFP and CA+/++H2B-RFP cell lines were created by stable transfection of and selection (5 μg/ml puromycin) for a H2B-RFP expression vector (Black et al. 2007) in CAY/- and CA+/+ cells, respectively. All cell lines were monoclonally sorted by FACS. For the transient transfection experiment (Figure 4.1F), wildtype HeLa cells were first synchronized in S phase by addition of 2 mM thymidine. After 17 hours, cells were released using 24 μM deoxycytidine and simultaneously transfected with untagged, wildtype CENP-A and/or HJURP expression vectors (or an empty vector) in combination with an EYFP- CENP-C expression vector (Shah et al. 2004) (2:2:1 proportion). 9 hours later, thymidine was re-added for an additional 15 hours, at which point cells were again released with deoxycytidine for 9 hours. A final thymidine arrest was performed and after 15 hours cells were fixed. Only cells expressing the positive transfection marker EYFP-CENP-C were analyzed. All stable and transient transfections were performed using Lipofectamine LTX (Invitrogen) according to the manufacturer’s instructions.

Immunoblotting and cell fractionation. Samples were prepared in Laemmli buffer, separated by SDS-PAGE, and transferred onto nitrocellulose membranes. Whole cell extracts (WCE) were

191

Chapter 4

prepared by lysing cells directly in sample buffer, to ensure that the entire cellular protein pool remained present in the sample. Recombinant CENP-A/H4-complexes were purified as described previously (Black et al., 2004) and mixed with WCE from chicken DT40 cells to bring the overall concentration of the purified protein preps to a level comparable to that of RPE WCE. Absence of cross-recognition of human CENP-A antibody to chicken protein was confirmed by omission of recombinant human CENP-A protein in DT40 extracts (Figure 4.2D, second lane). Alternatively, recombinant CENP-A/H4 was spiked into RPE cell extracts. Results obtained from the two methods are comparable [95.3 ± 14.0 ng (n=8) and 75.4 ± 5.4 ng (n=2), respectively; p>0.5]. Cellular CENP-A quantity was determined by comparison of fluorescence derived from cellular and purified CENP-A. The following antibodies and dilutions were used: CENP-A [Cell Signaling Technology, #2186 or (Ando et al., 2002)] at 1:1000 or tissue culture supernatant at 1:400, respectively; α-tubulin (DM1A, Sigma Aldrich) at 1:5000; HJURP [gift from Dan Foltz, (Foltz et al. 2009)] at 1:10000; Mis18BP1 (A302-825A, Bethyl Laboratories, Inc.) at 1:2000; H4K20me2 (ab9052, Abcam) at 1:1000. IRDye800CW-coupled anti-mouse or anti-rabbit (Licor Biosciences) and DyLight680-coupled anti-mouse or anti-rabbit (Rockland Immunochemicals) secondary antibodies were used prior to detection on an Odyssey near-infrared scanner (Licor Biosciences). Immunoblot signals were quantified using the Odyssey software (Licor Biosciences), and a linear response was confirmed over a 32-fold range (Figure 4.2E). Target protein signals were normalized to the α-tubulin loading control signal to correct for slight deviations in cell concentration between extracts of different RPE cell lines.

Cell fractionation was performed for CAY/-+H2B-RFP cells after cell lysis in ice cold buffer consisting of 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5 mM EDTA, 1% Triton X-100, 1 mM DTT, and a mix of protease inhibitors [1 mM PMSF, 1 μg/ml leupeptin, 1 μg/ml pepstatin, and aprotinin (Sigma A6279, 1:1000 dilution)]. Soluble proteins were separated from the

192

Counting CENP-A molecules in human cells

insoluble fraction by centrifugation at 21000 g at 4°C and resuspended in an equal volume of lysis buffer. Both supernatant and pellet fractions were incubated with 1.25 U/μl of benzonase nuclease (Novagen) on ice for 30 minutes prior to denaturation in Laemmli sample buffer.

Microscopy. Imaging was performed on an Andor Revolution XD system, controlling an inverted microscope (Nikon Eclipse-Ti), an iXonEM+ EMCCD camera (DU-897, Andor), a CSU-X1 spinning disk unit (Yokogawa), a laser combiner/multi-port switch system (Andor) and a motorized stage (Prior), controlled by MicroManager software (Edelstein et al. 2010). Images were collected using a Nikon 100X, 1.4NA, Plan Apo oil immersion objective (fixed cell imaging) or a Nikon 60X, 1.2 NA, Plan Apo VC water immersion objective (live cell imaging) at 1x binning. For live cell imaging, the temperature of the chamber was maintained at 37°C.

Fluorescence lifetime measurements. Cells grown on glass coverslips were fixed and mounted as described (Bodor et al. 2012) and imaged using a Zeiss LSM710 coupled to a motorized stage of an upright Zeiss Axio Examiner microscope equipped with a 63x Plan-Apo NA 1.4 oil immersion objective. A Coherent Chameleon Vision II multi-photon Ti-Sapphire laser was used to excite EYFP samples. All images were 512 x 512 pixels in size, with a pixel size of 0.09 μm. For all samples, an optimal setting of the laser power and PMT voltage was chosen to avoid pixel saturation and minimize photobleaching. The CLSM settings were kept constant so that valid comparisons could be made between measurements from different samples. Fluorescence lifetime imaging microscopy (FLIM) was performed by measuring the decay rate of EYFP using a Becker & Hickl time-correlated single photon counting hybrid detector coupled to the confocal LSM710 setup. The SPCImage (Becker & Hickl) software was utilized to perform single exponential fitting for each pixel location.

193

Chapter 4

Immunofluorescence and mitotic spreads. Cell fixation, immunofluorescence and DAPI staining was performed as described previously (Bodor et al. 2012). The following antibodies and dilutions were used: CENP-A [gift from Tatsuo Fukagawa (Ando et al. 2002)] tissue culture supernatant at 1:100, rabbit polyclonal CENP-B (sc22788, Santa Cruz Biotechnology) at 1:100, tissue culture supernatant from mouse hybridomas expressing monoclonal CENP-B (Earnshaw et al. 1987) at 1:4, CENP-C (Foltz et al. 2009) at 1:10000, CENP-T [gift from Dan Foltz (Barnhart et al. 2011)] at 1:1000, Hec1 (9G3.23; MA1-23308, Pierce) at 1:100, ACA (anti-centromere antibodies; 83JD, gift from Kevin Sullivan) at 1:100. Fluorescent secondary antibodies were obtained from Jackson ImmunoResearch or Rockland ImmunoChemicals and used at a dilution of 1:200. Immunofluorescence signals of Figure 4.1C, 4.5E, 4.6B, 4.7G were automatically quantified using the CRaQ method as described previously (Bodor et al. 2012) using CENP-T or CENP-C as a centromere reference. Hec1 levels were measured exclusively in prometaphase or metaphase (based on DAPI staining) of unperturbed cells. Micronuclei were scored based on DAPI staining. Mitotic spreads were performed after mitotic shake-off of cells arrested overnight (~16 hours) in 250 ng/ml nocodazole. 25000 cells/ml were swollen in 75mM KCl and 5000 cells were cytospun onto coverslips using a Cytopro 7620 cytocentrifuge (Wescor Inc.) for 4 minutes, at 1200 rpm, high acceleration. Cells were then fixed and processed for immunofluorescence as described above. Average centromere signals of both sisters were measured after background correction, by subtracting the minimum pixel value from the maximum of a box of 5x5 pixels around each sister centromere. Specific chromosomal markers were used as described in the text to detected centromeres of interest and signals were normalized to the average of all centromeres of the same cell spread.

194

Counting CENP-A molecules in human cells

Quantification of the centromeric CENP-A copy number.

CA+/+ cells were mixed with CAY/-, CAG/-, or CAY/-+OE cells at a ~1:4 ratio on 35mm glass-bottom petri dishes (MatTek Corporation). Non-cell permeable dextran-AlexaFluor647 (10000 MW, Molecular Probes) was added at 2–4 μg/ml to stain the medium outside of cells (Figure 4.2A, I). To minimize oversampling, individual live cells were imaged at 500 nm axial resolution (close to the resolution limit of the objective) spanning the entire cell volume. Images were flatfield corrected for unequal illumination using the signal of a uniform fluorescent slide and the “Shading Corrector” plugin for FIJI. For each axial section, the cell outline was determined based on absence of dextran-AlexaFluor647 staining and the integrated fluorescence intensities inside the cell outline as well as those of 1–3 independent background regions per section were determined. Background corrected signals from all sections were summed to determine the total cellular fluorescence. Fluorescence measurements of CAY/-, CAG/-, or CAY/-+OE cells were corrected for autofluorescence by subtraction of average per pixel fluorescence intensity of non-fluorescent CA+/+ cells from the same dish. Alternatively, CA+/++H2B-RFP and CAY/-+H2B-RFP cells were mixed and no dextran was added to the medium. In this case, the H2B-RFP signal was used to determine the nuclear volume and total nuclear fluorescence was determined as described above for the total cellular volume. Automated centromere detection was performed by an analogous algorithm to a previous study (Bodor et al. 2012; Bodor et al. 2013), where diffraction limited spots are detected based on their size, circularity, and feret’s diameter. Centromere signals were measured by integrating the intensity of a 5 pixel diameter surrounding each centromere in the appropriate axial section. Local background fluorescence was derived by measuring the difference in intensity between concentric circles of 5 and 7 pixel diameter, and subtracted from centromeric signals (Hoffman et al. 2001). In addition, centromeric signals were corrected for axial oversampling. For this, diffraction limited spots of yellow/green PS-Speck fluorescent beads

195

Chapter 4

(Molecular Probes) were measured in multiple plains. The sum intensity of individual beads from all these plains was compared to the signal in the plain with the maximum signal (i.e. the focal plane). The percentage of centromeric fluorescence was determined in relationship to the total fluorescence of each individual cell.

To allow for cell cycle staging of CAY/- cells, transduction with hCdt1(30/120)-RFP was performed using the BacMam2.0 baculovirus system (Invitrogen). Expression levels of transduced protein were allowed to stabilize for 3 days prior to analysis. Individual cells were followed by live cell microscopy using DIC and RFP signals. Nuclear RFP signals were tracked every ~2 hours to monitor their cell cycle progression. Imaging of YFP (CENP-A) and Cy5 (cellular volume) was performed as described above. Analysis of the centromeric CENP-A ratio was performed as described above, but restricted to cells in which RFP levels were decreasing at the specific timepoint of analysis (to exclude cells in G1 phase) and which did not enter mitosis or showed an increase in RFP levels for at least the following 3–4 hours (to exclude cells in G2 phase). Centromeric ratio was compared to non-transduced, randomly cycling cells (Figure 4.3C) or randomly cycling cells that were transduced, but not followed over time (Figure 4.3—S1). For these experiments, wildtype cells used to measure cellular autofluorescence were seeded on a separate dish.

Stochastic fluctuation measurements.

CAY/-, CAG/- or CAY/-+OE cells were treated with nocodazole (250–500 ng/μl) for 9 hours, after which cells were fixed and processed for immunofluorescence as described above. Sister centromere pairs were identified by CENP-B staining and GFP or YFP fluorescence intensity of each sister was measured and background corrected by subtracting the minimum pixel value of a 5 pixel diameter circle from the maximum value. The difference (δ) in fluorescence intensity as well as the sum (Σ) intensity of the two sisters was determined. The fluorescence intensity per segregating

196

Counting CENP-A molecules in human cells

unit (α) was determined from the average δ2/Σ of all centromere pairs of the same experiment and cell line. The number of segregating units on each centromere was calculated as Σ/α, as described previously (Rosenfeld et al., 2005, 2006) and in Figure 4.5A. In addition to sister centromeres, three independent rounds of random centromere pairing between all centromeres measured in a single experiment on CAG/- cells were performed and centromeric CENP-A-GFP units based on these pairings were quantified in Figure 4.5—S1E.

Yeast growth and imaging. 4kb-LacO, LacI-GFP S. cerevisiae [gift from Kerry Bloom (Lawrimore et al. 2011)] were grown in minimal synthetic media [Yeast nitrogen base (Sigma) + complete synthetic defined single drop-out medium lacking uracil and histidine (MP Biomedicals)], supplemented with 2% D (+)Glucose

(Merck). Prior to imaging, log phase cells (OD600 of ~0.7) were transferred onto a 2% low melting agarose pad and sealed under a coverslip with VALAP (1:1:1 vaseline:lanolin:paraffin). CAG/- cells were grown on 35mm glass- bottom petri dishes and yeast and human cells were imaged using identical settings during the same microscopy session. Fluorescence intensity of centromeres and Lac-arrays were quantified after background correction (maximum minus minimum of a 5x5 pixel box).

Integrating ChIP-seq and quantitative data of CENP-A at a human neocentromere. CENP-A ChIP-Seq data from the PDNC-4 neocentromere cell line (Accession #GSE44724) was processed as previously described (Hasson et al. 2013). Briefly, paired-end ChIP-Seq reads were aligned to the human genome build hg19 with Bowtie2 version 2.0.0 using paired-end mode. Reads were aligned by using a seed length of 50 bp, and only the single best alignment per read with up to two mismatches was reported in the SAM file. The aligned mate pairs were joined in MATLAB by requiring ≥95% overlap

197

Chapter 4

identity. The joined reads were aligned to the PDNC-4 neocentromere and only reads which mapped with 100% identity were used in the subsequent analysis. Nucleosome positions at the neocentromere were determined using the ‘findpeaks’ function in MATLAB. The probability of CENP-A occupancy at a given position was determined according to the following formula: (total reads overlying that position) X (216 CENP-A nucleosomes [Figure 4.7J]) / (total reads mapping to the entire neocentromere).

Calculation of the chance of reaching critical CENP-A levels after random segregation. All calculations represented in Figure 4.8C were performed in R. For these calculations we assume that CENP-A is inherited following a binominal distribution, consistent with our findings (Figure 4.5, 4.5—S1A, C). To determine the chance (X) of any chromosome reaching critical levels of CENP-A, the ‘pbinom’ function was used to calculate the fraction of a binomial distribution [where p = 0.5 and n (steady state number of nucleosomes) = 200 or was varied as indicated] that is either below a critical value (c = 22, or varied as indicated) or above acritical value (n−c). To determine the chance that any chromosome in a cell (containing 46 chromosomes) reaches critical levels, we calculated the chance that 46 independent centromeres do not reach critical levels and subtracted this chance from 1, i.e.: [1− (1−X)46].

198

Counting CENP-A molecules in human cells

REFERENCES

Abruzzo MA, Griffin DK, Millie EA, Sheean LA & Hassold TJ (1996) The effect of Y- chromosome alpha-satellite array length on the rate of sex chromosome disomy in human sperm. Hum. Genet. 97: 819–823 Allshire RC & Karpen GH (2008) Epigenetic regulation of centromeric chromatin: old dogs, new tricks? Nat. Rev. Genet. 9: 923–937 Amor DJ, Bentley K, Ryan J, Perry J, Wong L, Slater H & Choo KHA (2004) Human centromere repositioning ‘in progress’. Proc. Natl. Acad. Sci. U. S. A. 101: 6542– 6547 Ando S, Yang H, Nozaki N, Okazaki T & Yoda K (2002) CENP-A, -B, and -C chromatin complex that contains the I-type alpha-satellite array constitutes the prekinetochore in HeLa cells. Mol. Cell. Biol. 22: 2229–2241 Barnhart MC, Kuich PHJL, Stellfox ME, Ward JA, Bassett EA, Black BE & Foltz DR (2011) HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochore. J. Cell Biol. 194: 229–243 Bassett EA, DeNizio J, Barnhart-Dailey MC, Panchenko T, Sekulic N, Rogers DJ, Foltz DR & Black BE (2012) HJURP uses distinct CENP-A surfaces to recognize and to stabilize CENP-A/histone H4 for centromere assembly. Dev. Cell 22: 749– 762 Berdougo E, Terret M-E & Jallepalli PV (2009) Functional dissection of mitotic regulators through gene targeting in human somatic cells. Methods Mol. Biol. Clifton NJ 545: 21–37 Black BE & Cleveland DW (2011) Epigenetic Centromere Propagation and the Nature of CENP-A Nucleosomes. Cell 144: 471–479 Black BE, Foltz DR, Chakravarthy S, Luger K, Woods VL & Cleveland DW (2004) Structural determinants for generating centromeric chromatin. Nature 430: 578– 582 Black BE, Jansen LET, Maddox PS, Foltz DR, Desai AB, Shah JV & Cleveland DW (2007) Centromere identity maintained by nucleosomes assembled with histone H3 containing the CENP-A targeting domain. Mol. Cell 25: 309–322 Blower MD, Sullivan BA & Karpen GH (2002) Conserved organization of centromeric chromatin in flies and humans. Dev. Cell 2: 319–330 Bodor DL, Rodríguez MG, Moreno N & Jansen LET (2012) Analysis of Protein Turnover by Quantitative SNAP-Based Pulse-Chase Imaging. Curr. Protoc. Cell Biol. Chapter 8: Unit8.8 Bodor DL, Valente LP, Mata JF, Black BE & Jansen LET (2013) Assembly in G1 phase and long-term stability are unique intrinsic features of CENP-A nucleosomes. Mol. Biol. Cell 24: 923–932 Camahort R, Shivaraju M, Mattingly M, Li B, Nakanishi S, Zhu D, Shilatifard A, Workman JL & Gerton JL (2009) Cse4 is part of an octameric nucleosome in budding yeast. Mol. Cell 35: 794–805

199

Chapter 4

Carroll CW, Milks KJ & Straight AF (2010) Dual recognition of CENP-A nucleosomes is required for centromere assembly. J. Cell Biol. 189: 1143–1155 Carroll CW, Silva MCC, Godek KM, Jansen LET & Straight AF (2009) Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N. Nat. Cell Biol. 11: 896–902 Cheeseman IM, Chappie JS, Wilson-Kubalek EM & Desai A (2006) The conserved KMN network constitutes the core microtubule-binding site of the kinetochore. Cell 127: 983–997 Cheeseman IM & Desai A (2008) Molecular architecture of the kinetochore- microtubule interface. Nat. Rev. Mol. Cell Biol. 9: 33–46 Coffman VC, Wu P, Parthun MR & Wu J-Q (2011) CENP-A exceeds microtubule attachment sites in centromere clusters of both budding and fission yeast. J. Cell Biol. 195: 563–572 DeLuca JG, Gall WE, Ciferri C, Cimini D, Musacchio A & Salmon ED (2006) Kinetochore microtubule dynamics and attachment stability are regulated by Hec1. Cell 127: 969–982 Dunleavy EM, Almouzni G & Karpen GH (2011) H3.3 is deposited at centromeres in S phase as a placeholder for newly assembled CENP-A in G(1) phase. Nucl. Austin Tex 2: 146–157 Dunleavy EM, Roche D, Tagami H, Lacoste N, Ray-Gallet D, Nakamura Y, Daigo Y, Nakatani Y & Almouzni-Pettinotti G (2009) HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell 137: 485–497 Earnshaw WC & Migeon BR (1985) Three related centromere proteins are absent from the inactive centromere of a stable isodicentric chromosome. Chromosoma 92: 290–296 Earnshaw WC, Sullivan KF, Machlin PS, Cooke CA, Kaiser DA, Pollard TD, Rothfield NF & Cleveland DW (1987) Molecular cloning of cDNA for CENP-B, the major human centromere autoantigen. J. Cell Biol. 104: 817–829 Edelstein A, Amodaj N, Hoover K, Vale R & Stuurman N (2010) Computer control of microscopes using µManager. Curr. Protoc. Mol. Biol. Chapter 14: Unit14.20 Fachinetti D, Diego Folco H, Nechemia-Arbely Y, Valente LP, Nguyen K, Wong AJ, Zhu Q, Holland AJ, Desai A, Jansen LET & Cleveland DW (2013) A two-step mechanism for epigenetic specification of centromere identity and function. Nat. Cell Biol. 15: 1056–1066 Foltz DR, Jansen LET, Bailey AO, Yates JR, Bassett EA, Wood S, Black BE & Cleveland DW (2009) Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell 137: 472–484 Foltz DR, Jansen LET, Black BE, Bailey AO, Yates JR & Cleveland DW (2006) The human CENP-A centromeric nucleosome-associated complex. Nat. Cell Biol. 8: 458–469 Ford JH, Schultz CJ & Correll AT (1988) Chromosome elimination in micronuclei: a common cause of hypoploidy. Am. J. Hum. Genet. 43: 733–740

200

Counting CENP-A molecules in human cells

Fujita Y, Hayashi T, Kiyomitsu T, Toyoda Y, Kokubu A, Obuse C & Yanagida M (2007) Priming of Centromere for CENP-A Recruitment by Human hMis18α, hMis18β, and M18BP1. Dev. Cell 12: 17–30 Furuyama S & Biggins S (2007) Centromere identity is specified by a single centromeric nucleosome in budding yeast. Proc. Natl. Acad. Sci. U. S. A. 104: 14706–14711 Gascoigne KE, Takeuchi K, Suzuki A, Hori T, Fukagawa T & Cheeseman IM (2011) Induced ectopic kinetochore assembly bypasses the requirement for CENP-A nucleosomes. Cell 145: 410–422 Guse A, Carroll CW, Moree B, Fuller CJ & Straight AF (2011) In vitro centromere and kinetochore assembly on defined chromatin templates. Nature 477: 354–358 Haase J, Mishra PK, Stephens A, Haggerty R, Quammen C, Taylor RM 2nd, Yeh E, Basrai MA & Bloom K (2013) A 3D map of the yeast kinetochore reveals the presence of core and accessory centromere-specific histone. Curr. Biol. CB 23: 1939–1944 Hasson D, Panchenko T, Salimian KJ, Salman MU, Sekulic N, Alonso A, Warburton PE & Black BE (2013) The octamer is the major form of CENP-A nucleosomes at human centromeres. Nat. Struct. Mol. Biol. 20: 687–695 Henikoff S, Ahmad K, Platero JS & Steensel B van (2000) Heterochromatic deposition of centromeric histone H3-like proteins. Proc. Natl. Acad. Sci. 97: 716–721 Henikoff S & Furuyama T (2012) The unconventional structure of centromeric nucleosomes. Chromosoma 121: 341–352 Heun P, Erhardt S, Blower MD, Weiss S, Skora AD & Karpen GH (2006) Mislocalization of the Drosophila centromere-specific histone CID promotes formation of functional ectopic kinetochores. Dev. Cell 10: 303–315 Hoffelder DR, Luo L, Burke NA, Watkins SC, Gollin SM & Saunders WS (2004) Resolution of anaphase bridges in cancer cells. Chromosoma 112: 389–397 Hoffman DB, Pearson CG, Yen TJ, Howell BJ & Salmon ED (2001) Microtubule- dependent changes in assembly of microtubule motor proteins and mitotic spindle checkpoint proteins at PtK1 kinetochores. Mol. Biol. Cell 12: 1995–2009 Van Hooser AA, Ouspenski II, Gregson HC, Starr DA, Yen TJ, Goldberg ML, Yokomori K, Earnshaw WC, Sullivan KF & Brinkley BR (2001) Specification of kinetochore-forming chromatin by the histone H3 variant CENP-A. J. Cell Sci. 114: 3529–3542 Hori T, Shang W-H, Takeuchi K & Fukagawa T (2013) The CCAN recruits CENP-A to the centromere and forms the structural core for kinetochore assembly. J. Cell Biol. 200: 45–60 Irvine DV, Amor DJ, Perry J, Sirvent N, Pedeutour F, Choo KHA & Saffery R (2004) Chromosome size and origin as determinants of the level of CENP-A incorporation into human centromeres. Chromosome Res. Int. J. Mol. Supramol. Evol. Asp. Chromosome Biol. 12: 805–815 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805

201

Chapter 4

Joglekar AP, Bouck DC, Molk JN, Bloom KS & Salmon ED (2006) Molecular architecture of a kinetochore-microtubule attachment site. Nat. Cell Biol. 8: 581– 585 Joglekar AP, Bouck D, Finley K, Liu X, Wan Y, Berman J, He X, Salmon ED & Bloom KS (2008) Molecular architecture of the kinetochore-microtubule attachment site is conserved between point and regional centromeres. J. Cell Biol. 181: 587–594 Johnston K, Joglekar A, Hori T, Suzuki A, Fukagawa T & Salmon ED (2010) Vertebrate kinetochore protein architecture: protein copy number. J. Cell Biol. 189: 937–943 Karachentsev D, Druzhinina M & Steward R (2007) Free and chromatin-associated mono-, di-, and trimethylation of histone H4-lysine 20 during development and cell cycle progression. Dev. Biol. 304: 46–52 Lacoste N, Woolfe A, Tachiwana H, Garea AV, Barth T, Cantaloube S, Kurumizaka H, Imhof A & Almouzni G (2014) Mislocalization of the Centromeric Histone Variant CenH3/CENP-A in Human Cells Depends on the Chaperone DAXX. Mol. Cell 53: 631–644 Lawrimore J, Bloom KS & Salmon ED (2011) Point centromeres contain more than a single centromere-specific Cse4 (CENP-A) nucleosome. J. Cell Biol. 195: 573– 582 Lee C, Wevrick R, Fisher RB, Ferguson-Smith MA & Lin CC (1997) Human centromeric DNAs. Hum. Genet. 100: 291–304 Liu S-T, Rattner JB, Jablonski SA & Yen TJ (2006) Mapping the assembly pathways that specify formation of the trilaminar kinetochore plates in human cells. J. Cell Biol. 175: 41–53 Lo AW, Liao GC, Rocchi M & Choo KH (1999) Extreme reduction of chromosome- specific alpha-satellite array is unusually common in human chromosome 21. Genome Res. 9: 895–908 Maddox PS, Hyndman F, Monen J, Oegema K & Desai A (2007) Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 176: 757–763 Mahtani MM & Willard HF (1990) Pulsed-field gel analysis of alpha-satellite DNA at the human X chromosome centromere: high-frequency polymorphisms and array size estimate. Genomics 7: 607–613 Marshall OJ, Chueh AC, Wong LH & Choo KHA (2008) Neocentromeres: New Insights into Centromere Structure, Disease Development, and Karyotype Evolution. Am. J. Hum. Genet. 82: 261–282 Mata JF, Lopes T, Gardner R & Jansen LET (2012) A rapid FACS-based strategy to isolate human gene knockin and knockout clones. PloS One 7: e32646 McEwen BF, Chan GKT, Zubrowski B, Savoian MS, Sauer MT & Yen TJ (2001) CENP-E Is Essential for Reliable Bioriented Spindle Attachment, but Chromosome Alignment Can Be Achieved via Redundant Mechanisms in Mammalian Cells. Mol. Biol. Cell 12: 2776–2789

202

Counting CENP-A molecules in human cells

Meluh PB, Yang P, Glowczewski L, Koshland D & Smith MM (1998) Cse4p is a component of the core centromere of Saccharomyces cerevisiae. Cell 94: 607–613 Mendiburo MJ, Padeken J, Fülöp S, Schepers A & Heun P (2011) Drosophila CENH3 is sufficient for centromere formation. Science 334: 686–690 Okada M, Cheeseman IM, Hori T, Okawa K, McLeod IX, Yates JR 3rd, Desai A & Fukagawa T (2006) The CENP-H-I complex is required for the efficient incorporation of newly synthesized CENP-A into centromeres. Nat. Cell Biol. 8: 446–457 Padeganeh A, Ryan J, Boisvert J, Ladouceur A-M, Dorn JF & Maddox PS (2013) Octameric CENP-A nucleosomes are present at human centromeres throughout the cell cycle. Curr. Biol. CB 23: 764–769 Palmer DK, O’Day K, Trong HL, Charbonneau H & Margolis RL (1991) Purification of the centromere-specific protein CENP-A and demonstration that it is a distinctive histone. Proc. Natl. Acad. Sci. U. S. A. 88: 3734–3738 Palmer DK, O’Day K, Wener MH, Andrews BS & Margolis RL (1987) A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histones. J. Cell Biol. 104: 805–815 Pluta AF, Saitoh N, Goldberg I & Earnshaw WC (1992) Identification of a subdomain of CENP-B that is necessary and sufficient for localization to the human centromere. J. Cell Biol. 116: 1081–1093 Raychaudhuri N, Dubruille R, Orsi GA, Bagheri HC, Loppin B & Lehner CF (2012) Transgenerational propagation and quantitative maintenance of paternal centromeres depends on Cid/Cenp-A presence in Drosophila sperm. PLoS Biol. 10: e1001434 Régnier V, Vagnarelli P, Fukagawa T, Zerjal T, Burns E, Trouche D, Earnshaw W & Brown W (2005) CENP-A is required for accurate chromosome segregation and sustained kinetochore association of BubR1. Mol. Cell. Biol. 25: 3967–3981 Ribeiro SA, Vagnarelli P, Dong Y, Hori T, McEwen BF, Fukagawa T, Flors C & Earnshaw WC (2010) A super-resolution map of the vertebrate kinetochore. Proc. Natl. Acad. Sci. U. S. A. 107: 10484–10489 Rosenfeld N, Perkins TJ, Alon U, Elowitz MB & Swain PS (2006) A fluctuation method to quantify in vivo fluorescence data. Biophys. J. 91: 759–766 Rosenfeld N, Young JW, Alon U, Swain PS & Elowitz MB (2005) Gene regulation at the single-cell level. Science 307: 1962–1965 Sakaue-Sawano A, Kurokawa H, Morimura T, Hanyu A, Hama H, Osawa H, Kashiwagi S, Fukami K, et al (2008) Visualizing Spatiotemporal Dynamics of Multicellular Cell-Cycle Progression. Cell 132: 487–498 Du Sart D, Cancilla MR, Earle E, Mao JI, Saffery R, Tainton KM, Kalitsis P, Martyn J, Barry AE & Choo KH (1997) A functional neo-centromere formed through activation of a latent human centromere and consisting of non-alpha-satellite DNA. Nat. Genet. 16: 144–153 Schittenhelm RB, Althoff F, Heidmann S & Lehner CF (2010) Detrimental incorporation of excess Cenp-A/Cid and Cenp-C into Drosophila centromeres is

203

Chapter 4

prevented by limiting amounts of the bridging factor Cal1. J. Cell Sci. 123: 3768– 3779 Schuh M, Lehner CF & Heidmann S (2007) Incorporation of Drosophila CID/CENP-A and CENP-C into centromeres during early embryonic anaphase. Curr. Biol. CB 17: 237–243 Sekulic N, Bassett EA, Rogers DJ & Black BE (2010) The structure of (CENP-A- H4)(2) reveals physical features that mark centromeres. Nature 467: 347–351 Shah JV, Botvinick E, Bonday Z, Furnari F, Berns M & Cleveland DW (2004) Dynamics of centromere and kinetochore proteins; implications for checkpoint signaling and silencing. Curr. Biol. CB 14: 942–952 Shang W-H, Hori T, Martins NMC, Toyoda A, Misu S, Monma N, Hiratani I, Maeshima K, Ikeo K, Fujiyama A, Kimura H, Earnshaw WC & Fukagawa T (2013) Chromosome engineering allows the efficient isolation of vertebrate neocentromeres. Dev. Cell 24: 635–648 Shelby RD, Monier K & Sullivan KF (2000) Chromatin assembly at kinetochores is uncoupled from DNA replication. J. Cell Biol. 151: 1113–1118 Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63 Silver DP & Livingston DM (2001) Self-excising retroviral vectors encoding the Cre recombinase overcome Cre-mediated cellular toxicity. Mol. Cell 8: 233–243 Stoler S, Keith KC, Curnick KE & Fitzgerald-Hayes M (1995) A mutation in CSE4, an essential gene encoding a novel chromatin-associated protein in yeast, causes chromosome nondisjunction and cell cycle arrest at mitosis. Genes Dev. 9: 573– 586 Sullivan BA & Karpen GH (2004) Centromeric chromatin exhibits a histone modification pattern that is distinct from both euchromatin and heterochromatin. Nat. Struct. Mol. Biol. 11: 1076–1083 Sullivan BA, Schwartz S & Willard HF (1996) Centromeres of human chromosomes. Environ. Mol. Mutagen. 28: 182–191 Sullivan LL, Boivin CD, Mravinac B, Song IY & Sullivan BA (2011) Genomic size of CENP-A domain is proportional to total alpha satellite array size at human centromeres and expands in cancer cells. Chromosome Res. Int. J. Mol. Supramol. Evol. Asp. Chromosome Biol. 19: 457–470 Tachiwana H, Kagawa W, Shiga T, Osakabe A, Miya Y, Saito K, Hayashi-Takanaka Y, Oda T, Sato M, Park S-Y, Kimura H & Kurumizaka H (2011) Crystal structure of the human centromeric nucleosome containing CENP-A. Nature 476: 232–235 Thompson SL & Compton DA (2011) Chromosome missegregation in human cells arises through specific types of kinetochore-microtubule attachment errors. Proc. Natl. Acad. Sci. U. S. A. 108: 17974–17978 Utani K, Kohno Y, Okamoto A & Shimizu N (2010) Emergence of Micronuclei and Their Effects on the Fate of Cells under Replication Stress. PLoS ONE 5: e10089

204

Counting CENP-A molecules in human cells

Voullaire LE, Slater HR, Petrovic V & Choo KH (1993) A functional marker centromere with no detectable alpha-satellite, satellite III, or CENP-B protein: activation of a latent centromere? Am. J. Hum. Genet. 52: 1153–1163 Wevrick R & Willard HF (1989) Long-range organization of tandem arrays of alpha satellite DNA at the centromeres of human chromosomes: high-frequency array- length polymorphism and meiotic stability. Proc. Natl. Acad. Sci. U. S. A. 86: 9394–9398 Wu J-Q, McCormick CD & Pollard TD (2008) Chapter 9: Counting proteins in living cells by quantitative fluorescence microscopy with internal standards. Methods Cell Biol. 89: 253–273 Wu J-Q & Pollard TD (2005) Counting cytokinesis proteins globally and locally in fission yeast. Science 310: 310–314 Yoda K, Ando S, Morishita S, Houmura K, Hashimoto K, Takeyasu K & Okazaki T (2000) Human centromere protein A (CENP-A) can replace histone H3 in nucleosome reconstitution in vitro. Proc. Natl. Acad. Sci. U. S. A. 97: 7266–7271 Zinkowski RP, Meyne J & Brinkley BR (1991) The centromere-kinetochore complex: a repeat subunit model. J. Cell Biol. 113: 1091–1110

Author contributions All experiments and analyses were performed by me, with the following exceptions: JFM constructed and performed initial characterization of knockout and knockin cell lines; MS and JVS performed and analyzed experiments shown in Figure 4.2S1; AFD performed and helped analyze experiments in Figure 4.1F; KJS and BEB performed analysis shown in Figure 4.7K. LETJ is co-responsible for conception and design of the project. The manuscript for this chapter was drafted and revised with help of LETJ and constructive suggestions by all authors.

Acknowledgements We thank Tatsuo Fukagawa (National Institute of Genetics, Shizuoka, Japan), Dan Foltz (University of Virginia, Charlottesville, VA), Kevin Sullivan (National University of Ireland, Galway, Ireland), David Livingston (Dana-Farber Cancer Institute, Boston, MA), Bernardo Orr and Duane Compton (Dartmouth Medical School, Hanover, NH), and Kerry Bloom

205

Chapter 4

(University of North Carolina, Chapel Hill, NC) for reagents, Nitzan Rosenfeld (Cancer Research UK, Cambridge, UK) for advice, and Jorge Carneiro (Instituto Gulbenkian de Ciência, Oeiras, Portugal) for help using R. We thank the Confocal and Light Microscopy core facility at Dana Farber Cancer Institute (Harvard Medical School) for providing access to the FLIM setup. We are grateful to Alekos Athanasiadis and Monica Bettencourt-Dias (both at Instituto Gulbenkian de Ciência, Oeiras, Portugal) for helpful comments on the manuscript.

206

Counting CENP-A molecules in human cells

FIGURE SUPPLEMENTS

Supplement to Figure 1

Figure 4.1—S1 CENP-A expression is the rate limiting factor for centromeric CENP-A levels. (A) Pedigree of targeted RPE cell lines used in this study. Uninterrupted lines indicate single gene-targeting events, interrupted lines indicate multiple sequential gene-targeting events, and dashed lines indicate stable ectopic protein expression. (B, C) Correlation of centromeric CENP-A and total cellular HJURP (B) or Mis18BP1 levels (C). Insets show quantification of total protein levels from Figure 4.1B; n = 3–5 independent experiments. Dashed lines represent hypothetical directly proportional relationships with indicated correlation coefficients. In the insets, the average ± SEM (n = 3–5) is shown.

Supplements to Figure 2

Figure 4.2—S1 Representative fluorescence lifetime imaging (FLIM) micrograph of a CENP-A-YFP expressing cell (left) and quantification of indicated cellular regions (right).

207

Chapter 4

Figure 4.2—S2 Measurements of individual centromeres for CAG/- (A) and CAY/-+OE cells (B). Graphs as in Figure 4.2B. (C) Graph showing the absolute amount of centromeric CENP-A for indicated cell lines.

208

Counting CENP-A molecules in human cells

Figure 4.2—S3 Transfer efficiency of recombinant and cellular CENP-A. Immunoblots of recombinant and cellular CENP-A from CA+/+, CAG/-, and CAY/- cells, after protein transfer onto a stack of three membranes. The fraction of CENP-A retained on the first membrane (compared to the total signal from all three membranes) is quantified below. While YFP- or GFP- tagged CENP-A barely passes through the membrane at all, untagged CENP-A from cell extracts or recombinant protein preps is retained equally well on the first membrane.

Supplement to Figure 3

Figure 4.3—S1 hCdt-1(30/120)-RFP expression allows for accurate determination of cell cycle stages and measurements of centromeric CENP-A ratios. (A) An example trace of a cell that had entered G1 phase at the beginning of the experiment [as determined by cellular morphology (DIC)] is shown. Graph as in in Figure 4.3B. (B) Baculoviral transduction of hCdt-1(30/120)-RFP does not affect measurements of CENP-A-YFP. Centromeric CENP-A ratio measurements of non-transduced cells were compared to measurements of unstaged (i.e. randomly cycling) cells expressing hCdt-1(30/120)-RFP. Graph as in Figure 4.3C.

209

Chapter 4

Supplement to Figure 5

Figure 4.5—S1 Stochastic fluctuations of CENP-A segregation allows for copy number measurements. (A–D) Results as in Figure 4.5C–D for CAY/- (A–B) and CAY/-+OE cells (C–D). (E) Quantification of segregating units in CAG/- cells based on sister centromeres (dark green) or random centromere pairs (light green; random pairs were assigned independently three times). Asterisks indicate a significant difference from sister centromere result (t-test; p<0.0001 in all cases). Each circle represents one centromere pair. Throughout, the average ± SEM is indicated.

210

Counting CENP-A molecules in human cells

Supplement to Figure 6

Figure 4.6—S1 Representative images for quantifications in Figure 4.6B. Images of indicated cell lines are shown for immunofluorescence staining of (A) CENP-C, (B) CENP-T, and (C) Hec1 (mitotic cells). Scale bars: 5 μm.

211

Chapter 4

Supplement to Figure 7

Figure 4.7—S1 Measurements of individual centromeres for graphs in Figure 4.7. CENP-A levels are normalized to the average of each individual cell for CEN-Lac in HCT-116 cells (A), CEN-Y in wildtype HCT-116 cells (B), CEN-Y in DLD-1 cells (C), and NeoCEN-4 in PDNC-4 cells (D). Each circle represents one centromere; circles on the same column are individual centromeres from the same cell. Colored circle represents uniquely identified chromosome. Averages ± SEM are indicated. Graph to the right in C as in Figure 4.7D for DLD-1 cells (n = 26 and 927 for CEN-Y and Other CENs, respectively). Dashed line indicates average of all centromeres.

212

CHAPTER 5

General Discussion;

Or,

What I’ve Learned and What I Have to Say about It

Dani L. Bodor

Instituto Gulbenkian de Ciência, 2780-156, Oeiras, Portugal. INTRODUCTION

In this thesis I have presented my work on quantitative aspects of human centromere inheritance. Specifically, I have designed an algorithm to automatically recognize centromeric foci on fluorescent micrographs and quantify their signal intensity (Chapter 2). This algorithm was combined with SNAP-based pulse-chase experiments to analyze regulatory factors of CENP-A stability and assembly dynamics (Chapter 3). Furthermore, I have used a related quantification strategy to determine the number of CENP-A molecules at human centromeres and to elucidate the cell and chromatin distribution of this most critical of centromere proteins (Chapter 4). The previous chapters have detailed my specific methods and results. In this chapter, I will discuss on a more conceptual level what my findings have to offer to life, the universe, … and everything centromere biology-related.

Conclusions

NON-CENTROMERIC CENP-A

In 1985, a 17 kDa protein recognized by human auto-immune sera from scleroderma patients was originally baptized as CENP-A (CENtromere Protein A) for the single property of being centromere localized (Earnshaw & Rothfield, 1985). While the essential role of CENP-A in centromere function and specification is well established, in recent years, multiple independent studies have found detectable levels of nucleosomal CENP-A outside of centromeres in human cells (Hasson et al, 2013; Lacoste et al, 2014), as well as in other species (Camahort et al, 2009; Choi et al, 2012; Shang et al, 2013). In fact, the centromeric pool represents less than a third of all chromatin bound CENP-A (Figure 4.4) and approximately a fifth of the total protein pool (Figure 4.3) in human RPE cells. Given this minority population at the centromere, an extreme point of view would be that ‘centromere protein A’ is perhaps a misnomer for this particular protein. However, there is a completely different way of seeing this. Alphoid sequences represent only ~2.6% of the total human genome (Willard & Waye, 1987; Hayden et al, 2013), and a large proportion of the α-satellite DNA is devoid of CENP-A (Warburton et al, 1997; Spence et al, 2002; Hayden et al, 2013). Indeed, one publication found that the CENP-A enriched domain represents only 35-50% of the entire length of the α-satellite repeats (Sullivan et al, 2011). Taking these numbers into account, we determined that the centromeric minority (in absolute numbers) of CENP-A represents a nearly 50-fold enrichment compared to the overall genome when measured on a per nucleosome basis (Figure 4.8). Thus, given that there is currently some debate in the field regarding the nomenclature of this protein (Talbert et al, 2012; Earnshaw et al, 2013; Talbert & Henikoff, 2013; Earnshaw & Cleveland, 2013), I would like to take this opportunity to suggest that it promptly be renamed to what is, strictly speaking, the most accurate name: CENrichedP-A.

217

Chapter 5

In order to maintain this strong enrichment at centromeres, a specialized loading pathway exists that specifically targets CENP-A to the correct locus. HJURP is a CENP-A specific histone chaperone (Dunleavy et al, 2009; Foltz et al, 2009; Shuaib et al, 2010) and assembly factor (Barnhart et al, 2011), that binds to CENP-A through a number of residues lacking in other H3 variants (Hu et al, 2011; Bassett et al, 2012). Centromere recruitment of HJURP depends on the Mis18 complex members Mis18α and Mis18β (Barnhart et al, 2011; Wang et al, 2014), which are interdependent for centromere targeting with M18BP1 (Fujita et al, 2007), which is in turn recruited to centromeres through binding to CENP-C (Moree et al, 2011), itself a direct binding partner of CENP-A (Carroll et al, 2010). The exact nature and mechanisms by which these proteins are able to recruit each other are still unclear and currently under intense investigation. Nevertheless, it is clear that this this closed feedback loop is ultimately responsible for maintaining a high degree of CENP-A enrichment at centromeres. On the flipside, at least 98% of the genome is non-centromeric, and CENP-A is multiple orders of magnitude less abundant than other H3 variants1. Thus, while high specificity of HJURP to CENP-A is essential to avoid sequestration by typical H3 variants, strict evasion of CENP-A by other histone chaperones may not be of much consequence. Indeed, there are indications that multiple assembly factors are capable of some degree of CENP-A assembly into chromatin. First, when expressed from a typical H3.1 promoter, CENP-A is distributed throughout the nucleus and ceases to be centromere enriched (Shelby et al, 1997). Similar findings were made upon high levels of ectopic CENP-A overexpression from a constitutively active promoter (Van Hooser et al, 2001; Gascoigne et al, 2011), but not upon lower levels of overexpression (Shelby et al, 1997; Gascoigne et al, 2011;

1 This estimate is derived from my finding that there are <105 CENP-A molecules per RPE cell (Figure 4.2F), while there are ~3·107 nucleosome positions in a diploid human genome.

218

Conclusions

Figure 4.1C). More recently, direct evidence was found for crosstalk between DAXX, a major histone H3.3 chaperone (Drané et al, 2010; Goldberg et al, 2010), and CENP-A, which were shown to be able to interact in vitro, albeit with lower affinity than H3.3 (Lacoste et al, 2014). Moreover, upon CENP-A overexpression, it becomes enriched at typical H3.3 sites in a DAXX dependent manner and heterotypic CENP-A- and H3.3-containing nucleosomes are observed (Lacoste et al, 2014). However, although a small fraction of heterotypical CENP-A nucleosomes have been previously reported upon overexpression in an independent study (Foltz et al, 2006), they were not detected at wildtype expression levels and only minor co- enrichment of CENP-A and H3.3 was observed (Lacoste et al, 2014). Although CENP-A and H3.1 were not observed within a single nucleosome, no careful analysis was performed regarding the potential interaction between CENP-A and the canonical H3,1 assembly factor CAF (Lacoste et al, 2014). Thus, the restriction of CENP-A expression to G2 phase (Shelby et al, 1997, 2000), just prior to its loading in the beginning of the subsequent G1 (Jansen et al, 2007) and distinct from the major phase of canonical nucleosome assembly (Worcel et al, 1978), may limit its potential for misincoporation. Taken together, it appears that the correct (quantitative and/or temporal) regulation of CENP-A expression is a major driving force in preventing ectopic accumulation of CENP-A. Although the majority of CENP-A is not centromere localized, I consider the non-centromeric pool of this protein as noise. It would be difficult to imagine an efficient mechanism with such a high degree of stringency that it would ensure that >98% of the chromatin remains devoid of CENP-A. Although only 20% of CENP-A is centromere localized (Figure 4.8), I would not be surprised if the fraction of many other proteins that is active, or at least present at the functionally relevant location, is similar or lower. Importantly, although I would argue that it is unlikely that ectopic CENP-A has a direct endogenous function, this does not exclude that it can influence the regulation of non-centromeric chromatin. This may be especially

219

Chapter 5

relevant when CENP-A is overexpressed, as is the case in many cancer cells (Lacoste et al, 2014; Thiru et al, 2014). Indeed, ectopically assembled CENP-A has been shown to reduce CTCF occupancy at its typical binding sites in both naturally overexpressing cancer cell lines and upon experimentally induced CENP-A overproduction in HeLa cells (Lacoste et al, 2014). In addition, although direct evidence for a functional relationship remains elusive and many results are somewhat ambiguous, a number of studies have reported a link between non-centromeric CENP-A and DNA damage response (Zeitlin et al, 2005, 2009, 2011; Ambartsumyan et al, 2010; Lacoste et al, 2014). Nevertheless, at wildtype expression levels, CENP-A nucleosomes represent less than 0.1% of all non-centromeric chromatin (Figure 4.8), indicating a minor effect, if any, on chromatin (mis-) regulation.

220

Conclusions

THE ULTRASTABILITY OF CENP-A

It has become apparent over the last years that the chromatin dynamics of CENP-A are unique among nucleosomes. Canonical H3.1 is assembled throughout the genome during S phase, in a replication dependent manner (Worcel et al, 1978; Ray-Gallet et al, 2011), while H3.3 is preferentially assembled at specific loci throughout the cell cycle (Ahmad & Henikoff, 2002; Ray-Gallet et al, 2002, 2011; Goldberg et al, 2010). However, assembly of centromeric CENP-A is restricted to a brief period in the cell cycle, which in metazoans immediately follows mitotic exit (Jansen et al, 2007; Schuh et al, 2007; Bernad et al, 2011; Moree et al, 2011; Dunleavy et al, 2012; Silva et al, 2012). CENP-A assembly is regulated, at least in part, by phosphorylation of HJURP, the Mis18 complex, and itself by the key cell cycle kinases Cdk1, Cdk2, and Plk1 (Silva et al, 2012; McKinley & Cheeseman, 2014; Müller et al, 2014; Wang et al, 2014; Yu et al, 2015). In addition to its atypical assembly dynamics, CENP-A also displays an extreme level of chromatin maintenance, not observed for any other type of nucleosome. Indeed, while canonical histones turn over with a half-life of approximately 8 hours (Figure 3.3; Kimura & Cook, 2001), no turnover of CENP-A was detected, apart from replicative dilution, for up to 5 days in culture (Figures 3.3, 3.4, and 3.S3). However, the full mechanism leading to CENP-A ultrastability, which may exceed that of any other protein in nature, remains unclear.

Intrinsic determinants One possibility is that long-term retention is conferred onto CENP-A through a cis regulatory region that differs from other histone variants. Consistent with this hypothesis, hydrogen/deuterium-exchange experiments identified a region within the histone fold domain of CENP-A that induces an increased conformational rigidity of the CENP-A/H4 binding interface as compared to H3/H4 (Black et al, 2004, 2007a). This region, spanning loop1 and the α2-helix, was termed CENP-A targeting domain (CATD), because

221

Chapter 5

substitution of the 22 divergent amino acids into H3 induces a clear enrichment of chimeric H3CATD histones at centromeres (Black et al, 2004). Consistently, it was later shown that the CATD mediates recognition of CENP-A by its assembly factor HJURP (Foltz et al, 2009; Shuaib et al, 2010; Bassett et al, 2012). Moreover, H3CATD displays identical loading dynamics as CENP-A (Figure 3.4A–B), arguing that the correct cell cycle regulation of assembly acts upon its loading factors rather than upon CENP-A itself. However, centromere enrichment appears not to be exclusively dependent on binding to its chaperone, as specific residues of the CATD are required for centromere accumulation but not for assembly at sites of ectopically targeted HJURP (Bassett et al, 2012). Importantly, although not all properties of CENP-A are reproduced after a clean genetic substitution by H3CATD, which is insufficient to recruitment downstream centromere and kinetochore proteins, it is capable to maintain its own centromeric levels over many divisions (Fachinetti et al, 2013). Taken together, a model emerges where the CATD is primarily responsible for maintaining centromere identity, but not centromere function. In addition to its regulatory role in CENP-A assembly, the CATD also confers an increased nucleosome stability. SNAP-based pulse-chase experiments show that the long-term retention of H3CATD at centromeres approaches that of CENP-A (Figure 3.4E). One possibility is that the more rigid nucleosome structure of in vitro assembled complexes conferred by the CATD translates into an increase in protein stability in dividing cells. Interestingly, although no effect was observed on chromatin assembly at ectopic sites of HJURP tethering, conversion of six hydrophobic CATD residues to their H3 counterpart, thought to revert the rigidity, caused a severe defect in centromere accumulation (Bassett et al, 2012). Although this observation could theoretically indicate a decreased stability of otherwise properly assembled nucleosomes, it is unclear why the extent of the defect would suggest an even lower retention than expected for canonical H3 nucleosomes. An alternative interpretation is that this mutant

222

Conclusions

CENP-A is not properly targeted to centromeres, perhaps through HJURP- independent regulation. In a separate study, CENP-A lacking a small (2 residue) protrusion in loop1 of the CATD that is not present in H3 was cotransfected with wildtype protein, and it was found that the ratio of mutant-to-wildtype CENP-A at centromeres decreases over time (Tachiwana et al, 2011). Although the authors interpreted this result as a compromised centromeric stability of the mutant protein, it is equally consistent with other induced defects, e.g. mitotic arrest, conferring a growth disadvantage, or being otherwise toxic to cells. While conclusive evidence regarding which residues or subdomains of the CATD induce the increased nucleosome stability is currently lacking, this may be provided by pulse-chase analysis of SNAP-tagged mutant versions of CENP-A or H3CATD (Figures 2.2 and 3.4), potentially in combination with LacO tethering of HJURP (Barnhart et al, 2011; Bassett et al, 2012).

External binding factors Although ultrastability could be an intrinsic property of CENP-A, an alternative possibility is that other proteins also contribute to its centromeric retention. Interestingly, whereas the stability of H3CATD was not statistically different than expected for replicative dilution, it does appear slightly less stable than CENP-A (Figure 3.4). While it may be an artifact of imperfect targeting of H3CATD to centromeres, this result does indicate that introduction of the CATD may be insufficient to confer full ultrastability. Moreover, the CATD may not have a direct effect on CENP-A maintenance, but rather induce binding of other proteins that are responsible for its stable retention. Thus, external factors could either (physically or functionally) interact with the CATD or play an independent role to confer full stability. An interesting hypothesis is that there is an overlap between CENP-A assembly and maintenance factors. One indication for this is that a proportion of HJURP interacts with chromatin incorporated CENP-A (Foltz et al, 2006), although it is not clear what the functional relevance of this

223

Chapter 5

interaction is, if any. In addition, in one study, depletion of M18BP1 resulted in a dramatic decrease of steady state CENP-A levels, beyond what would be expected solely from a complete lack of nascent incorporation (Maddox et al, 2007). Nevertheless, I was able to exclude a role in CENP-A stability for either of these proteins by depleting them in CENP-A-SNAP expressing cells (Figures 3.5). Similarly, immunodepletion of HJURP from Xenopus extracts did not affect the centromeric levels of CENP-A in arrested cells (Bernad et al, 2011). Thus, despite a dependence on the CATD for both assembly and maintenance, it appears that these represent separate properties of CENP-A nucleosomes. To my knowledge, only two external factors have been reported to play a role in stable retention of CENP-A. The first is a group of proteins constituting a small GTPase switch that is required to retain nascent CENP-A at centromeres (Lagana et al, 2010). However, CENP-A that was assembled prior to depletion of MgcRacGAP, a key regulator in this process, remained unaffected (Lagana et al, 2010). Thus, it remains unclear whether this protein is truly responsible for stabilizing CENP-A nucleosomes, or perhaps somehow involved in the proper chromatin assembly of centromere targeted (non-nucleosomal) CENP-A. Irrespectively, the contribution of this GTPase switch to an effective CENP-A loading process appears to be higher than to its long-term retention. Second, in a study performed in Xenopus egg extracts, it was shown that depletion of condensins results in a reduction of CENP-A from centromeres of non-dividing cells (Bernad et al, 2011). Condensins are known to be important regulators of chromosome organization and their depletion would be expected to considerably influence the structure of (centromeric) chromatin (Hagstrom et al, 2002; Wignall et al, 2003; Oliveira et al, 2005). Although this hints at a functional relationship between condensin and CENP-A stability, quantification of the affected chromatin may be confounded by potential artifacts of these immunodepletion experiments, such as defocussing of centromeric signals or an altered antibody accessibility to CENP-A. To control for this potential

224

Conclusions

artifact, fluorescently tagged CENP-A could be used or counterstaining could be performed with an antibody against another centromere protein that would not be expected to be influenced by loss of CENP-A. Nevertheless, although it will be important to address the issue raised above, condensins remain among the strongest candidate CENP-A maintenance factors identified to date. Members of the constitutive centromere associated network (CCAN) form yet another group of candidate proteins involved in CENP-A retention. A decrease of centromeric CENP-A levels has been observed after depletion of a number of CCAN members, including CENP-H (Okada et al, 2006, 2009), CENP-N (Carroll et al, 2009), and CENP-C (Carroll et al, 2010). In addition, CENP-A can directly bind to both CENP-N (Carroll et al, 2009) and CENP-C (Carroll et al, 2010; Guse et al, 2011; Kato et al, 2013), through the CATD and C-terminal six residues (LEEGLG), respectively. It must be noted, however, that reconstitution experiments in Xenopus egg extracts indicate that recruitment of CENP-N is independent of the CATD, but depends exclusively on CENP-C (Guse et al, 2011), and it thus remains unclear what the functional relevance is of the direct interaction between CENP-A and CENP-N. Interestingly, while both CENP-C and CENP-N turn over at the centromere throughout most of the cell cycle, they become stably bound during mid–late S phase and their centromeric levels increase (Hemmerich et al, 2008; Hellwig et al, 2011; Gascoigne & Cheeseman, 2013). These specific cell cycle dynamics are somewhat suggestive for a role in CENP-A retention, because chromatin disruption by the replication machinery is one of the most challenging processes for nucleosome retention (Groth et al, 2007; Alabert & Groth, 2012) and centromeres have been shown to be relatively late replicating domains (O’Keefe et al, 1992; Shelby et al, 2000). Indeed, our preliminary experiments indicate that depletion of CENP-C results in a slightly accelerated loss of pre-incorporated centromeric CENP-A (Figure 3.A). Although CENP-N was previously shown to play a role in the CENP-A assembly pathway (Carroll et al, 2009), we did

225

Chapter 5

not observe any defect on either loading or maintenance after RNAi against CENP-N (data not shown). However, it must be noted that we did not carefully monitor the extent of protein depletion, which limits the interpretability of our results. Therefore, both CENP-N and, especially, CENP-C remain strong candidate CENP-A maintenance factors.

Open questions regarding CENP-A ultrastability As discussed above, there is evidence for both intrinsic and external contributions to the lack of CENP-A turnover. However, a number of interesting considerations regarding the nature of CENP-A ultrastability remain unanswered. First, it would be important to assess whether this protein is equally stable at non-centromeric loci as at the centromere, which will help identify regulatory processes of CENP-A maintenance. In addition, long-term retention assays have currently only been performed on human tissue culture cells and it remains unknown whether CENP-A ultrastability is specific to this system, or is conserved in other organisms as well. Interestingly, there is at least one known example of epigenetically defined centromeres where this protein is not stably retained between divisions, since it was shown that the entire pool of CENP-AHCP-3 turns over between the first and second mitotic division in C. elegans embryogenesis (Gassmann et al, 2012). However, C. elegans may be an exception, not only because of the holocentric nature of their chromosomes (Albertson & Thomson, 1982), but also because CENP-AHCP-3 is lost completely during gametogenesis and is therefore not absolutely required to specify centromeric identity (Monen et al, 2005; Gassmann et al, 2012). Conversely, in most species analyzed, CENP-A can be readily detected in both mature sperm (Palmer et al, 1990, 1991; Bernad et al, 2011; Dunleavy et al, 2012; Raychaudhuri et al, 2012; Chmátal et al, 2014) and oocytes (Dunleavy et al, 2012; Chmátal et al, 2014), and it has been shown that Drosophila CENP-ACID is required in sperm cells to specify centromeres on paternally inherited chromosomes of the next generation (Raychaudhuri et al, 2012).

226

Conclusions

Finally, we do not currently know what the dynamics of centromeric CENP-A are in long-lived post-mitotic cells such as neurons or human oocytes, which can remain arrested in meiotic prophase I for decades. To my knowledge, the only known analysis in this direction was performed on human pancreatic tissue, where it appears that centromeric CENP-A declines with age in non-dividing islet cells, but not in actively dividing exocrine cells (Lee et al, 2010). Although the results are intriguing, this study in a small number of human samples does not have the power to interrogate the molecular mechanisms of CENP-A turnover dynamics and it would be important to revisit these findings in a more amenable model. Moreover, while retention of CENP-A in post-mitotic pancreatic cells may not be essential, oocytes need to reenter the cell cycle upon fertilization and thus need to preserve functional centromeres. Indeed, in one study, non- canonical regulation of CENP-ACID assembly has been observed during both male and female meiosis in Drosophila (Dunleavy et al, 2012), although in an accompanying paper no meiotic loading was detected during spermatogenesis in this species (Raychaudhuri et al, 2012). In conclusion, although progress is being made towards understanding the regulation of CENP-A ultrastability, there is still a long way to go.

227

Chapter 5

MASS ACTION VERSUS ULTRASTABILITY

Mass action mechanisms were first described 150 years ago to explain how reversible chemical reactions ultimately result in a dynamic equilibrium (Waage & Gulberg, 1864). According to this theory, the output of a chemical reaction is directly proportional to its input, i.e. the amount of product formed depends directly on the amount of reactants added, and is dictated by the rate constants of the reaction (Guldberg & Waage, 1867). A similar relationship is observed for CENP-A, where the centromeric pool is maintained in direct proportion to varying total cellular levels (Figure 4.1). This observation argues that there is a fixed centromere targeting efficiency of CENP-A, which is independent of the amount of protein present. However, given the ultrastable maintenance of CENP-A at centromeres (see above) as opposed to a dynamic equilibrium, this regulation does not follow the same principles as a mass action mechanism. In fact, three modes of regulating CENP-A inheritance have been shown to exist. These are: 1) No exchange between non-centromeric and centromeric CENP-A pools (Figure 3.4D; Hemmerich et al, 2008); 2) stochastic redistribution of existing centromeric CENP-A over two centromeres during each division (Figure 4.5C); and 3) assembly of nascent CENP-A in direct proportion to the total cellular pool (Figure 4.1). However, if there is no form of communication between the centromeric and non- centromeric pool, individual centromeres would have the potential of reaching extreme values, which would lead to an increasing variance of CENP-A with each round of dilution and replenishment. Thus, the absence of a true mass action mechanism appears inconsistent with a fixed ratio of total-to-centromeric CENP-A.

228

Conclusions

Figure 5.1 Hypothetical scenario of density dependent CENP-A assembly. While, exclusive positive or negative feedback of CENP-A levels on incorporation of nascent protein would either lead to centromere spreading or centromere extinction, a combination of switch-like positive (green) and linear negative (red) regulation of CENP-A assembly would explain the observed maintenance of CENP-A levels. Similarly, several rules are required to properly regulate complex behavior of kinds as well (God, 1448BC). In the system described above, optimal loading efficiency is reached at intermediate densities (average centromeric CENP-A occupancy: ~4%; see Chapter 4), while virtually no loading occurs at very low levels of CENP-A, as found in generic chromatin (~0.1% occupancy; see Chapter 4), and efficiency is decreased at centromeres with a very high CENP-A occupancy. Red and grey nucleosomes represent CENP-A and H3 nucleosomes, respectively.

229

Chapter 5

One trivial solution to this paradox could be that cells that have exceeded certain boundaries are eliminated from the population. This would not require any additional CENP-A regulation, but rather a process that reacts to extreme levels. Theoretically, this could be a passive process: e.g. too little or too much CENP-A would lead to a dysfunctional centromere, which in turn leads to chromosomal instability and, ultimately, cell death. However, this would likely not be a very effective mechanism, as a low level of chromosomal instability and aneuploidy is generally tolerated by cells (Holland & Cleveland, 2009). Alternatively, a hypothetical monitoring factor could exist, which actively drives cells into programmed cell death upon extreme high or low CENP-A levels. Irrespective of the nature of the mechanism that eliminates cells with extreme levels, the variance of CENP-A would need to be kept to a minimum to avoid losing a large proportion of cells from the population. An alternative hypothesis is that there is an additional, yet to be discovered form of regulating centromeric CENP-A levels. Indeed, although the efficiency of CENP-A assembly is constant on the cellular scale, it could be regulated on the per-centromere level. Specifically, the pre-incorporated pool of CENP-A would be expected to negatively influence targeting of nascent protein. However, this hypothesis is apparently at odds with the fact that CENP-A is predominantly assembled at existing centromeres, which inherently have a higher density of CENP-A than non-centromeric loci. Thus, two opposing forces may be required to accurately regulate CENP-A levels. First, a positive regulator of CENP-A recruitment that has an almost all-or-nothing effect (Figure 5.1, green) is necessary, thus generating a minimal CENP-A threshold. Next, negative regulation would be required, the strength of which is expected to correlate with the amount of CENP-A (Figure 5.1, red). Combined, these processes would lead to the mass action type of regulation observed for maintenance of stably bound CENP-A levels (Figure 5.1).

230

Conclusions

Although it is not evident what these regulators would be, some candidates come to mind. Regarding positive regulation, it is likely that a high enough density of CENP-A creates a platform that is recognized as a centromere. In this case, the actual CENP-A level is ir relevant, as long as a certain threshold is exceeded. Indeed, I found that neither CENP-C nor CENP-T centromere levels correlate with CENP-A levels (Figure 4.6A) and a similar effect was seen for CENP-I (Liu et al, 2006). Of these CCAN members, CENP-C is an especially good candidate, as it has been proposed to recruit M18BP1 to centromeres (Moree et al, 2011; Dambacher et al, 2012). As opposed to the switch-like positive regulation, negative feedback is more likely to be linear with CENP-A levels. Thus, good candidates would be proteins that are stoichiometric and/or cosegregate with CENP-A, such as its proposed direct binding partner CENP-N (Carroll et al, 2009), or perhaps even (PTMs on) CENP-A itself. A similar hypothesis has been put forward previously, wherein microtubule-generated tension on centromeres is proportional to mitotic CENP-A levels and negatively influences assembly of nascent CENP-A in the subsequent G1 (Brown & Xu, 2009). Because the amount of CENP-A available for centromere assembly may still correlate with total protein levels in the absence of a negative regulator, assays to identify such a factor would likely need to focus on deregulated variances rather than mean CENP-A values. Together, switch-like recruitment of a positive regulator and stoichiometric recruitment of an antagonist assembly would lead to stable maintenance of steady state CENP-A levels. Above, I have presented two potential solutions to the paradoxical observations regarding the regulation of centromeric CENP-A levels. However, both are quite speculative and complex in nature and no evidence exists for either. Nevertheless, it is a well-documented fact that some people can believe as many as six impossible things before breakfast (Carroll, 1871). Thus, although a more realistic hypothesis would be preferable, mine also remain plausible.

231

Chapter 5

THE CRITICAL AMOUNT OF CENP-A

Chapter 4 of this thesis presents an analysis of the amount of CENP-A on human centromeres. In addition, initial characterizations were made of cells with varying amounts of CENP-A. The next important goal, which lies at the heart of understanding the epigenetic mechanism driving centromere inheritance, is to determine the minimum amount of CENP-A required to define a centromere. Unfortunately, however, I was not able to resolve this within the timeframe of my PhD. Nevertheless, I will discuss my ideas regarding the critical amount of CENP-A, including hints from the published literature as well as potential methods to address this experimentally.

CENP-A variance The amount of CENP-A on human centromeres varies at different levels. First, not all centromeres of one cell have the same amount of CENP-A. Although there may be some contribution of centromere specific differences (Figure 4.7C–F), 85% cells analyzed passed a normality test2, consistent with a largely stochastic nature of intracellular variability. Second, variation observed between cell averages (Figure 4.2C) is also likely to be stochastic, as all seven datasets (experiments) presented in this figure passed the normality test. Finally, substantial variation is observed between cultured cell lines (Figure 4.7H), which may represent differential expression levels of e.g. CENP-A itself, its loading, and/or other regulatory factors. These differences may have emerged during the production or in vitro evolution of the cell lines presented, although a contribution of cell type specific differences may exist, which has to my knowledge not been addressed for normal human tissues. This variance of CENP-A levels is an important issue to take into consideration when determining the critical amount of CENP-A.

2 The distribution of centromeric CENP-A levels of 94 of the 111 mitotic spreads presented in Figure 4.7S1 passed a D'Agostino & Pearson omnibus normality test using GraphPad Prism (α=0.05). This particular dataset was chosen to test for normality because clustering of multiple centromeres into a single spot is excluded and because at least 23 centromeres were measured in each cell.

232

Conclusions

Centromere maintenance Average mitotic centromeres in wildtype RPE cells contain ~200 molecules of CENP-A (Figure 4.8). However, at least two results show that lower levels are sufficient to maintain centromere identity. First is the finding that DLD-1 cells, a cultured colorectal adenocarcinoma cell line, have on average only ~25% as much CENP-A as RPE cells (Figure 4.7H). Evidently, it is possible, perhaps even likely, that this level of CENP-A leads to (mitotic) defects such as chromosome missegregation or centromere loss. Nevertheless, this clearly demonstrates that 50 CENP-A molecules are more than sufficient to stably sustain centromeric identity throughout generations. The second analysis was performed on an RPE cell line in which the CENP-A gene has been flanked by LoxP sites, allowing for its controlled deletion from the genome (Fachinetti et al, 2013). Following Cre- mediated gene ablation in these cells, stable retention of existing centromeric CENP-A molecules leads to a 50% decrease per division. Surprisingly, although the mitotic fidelity was compromised, cell duplication rates remained unaffected for at least 5 days after deletion of CENP-A, at which point the average centromeric levels were down to ~7% (Fachinetti et al, 2013). Similar results were obtained in HeLa cells, where the recruitment of a number of centromere proteins remained unaffected in cells where CENP-A levels had been depleted by RNAi to ~10% (Liu et al, 2006). These results argue that, although low levels of CENP-A affect centromere function, 14 molecules may be sufficient to maintain centromere identity. However, an alternative hypothesis could be that for a limited number of divisions, centromeres can survive independently of CENP-A, perhaps through semi-stable self-regulated recruitment of downstream CCAN proteins. Taken together, these results argue that in typical human cells the number of centromeric CENP-A molecules is substantially higher than the critical amount required for epigenetic centromere maintenance.

233

Chapter 5

Determining the true minimum amount of CENP-A required for centromere maintenance is not an easy feat to accomplish. For this, it would be necessary to differentially manipulate CENP-A levels, ideally in a well regulated and acute manner. One possibility would be to stably decrease CENP-A levels by genomic integration of an shRNA cassette (Black et al, 2007b) and selecting for cells that maintain viability with minimal protein levels. However, this approach is likely susceptible to a large degree of variation of knockdown efficiency, even in clonal cell lines. An alternative would be to use a similar system to the conditional knock-out cells of Fachinetti et al, but containing an ectopic CENP-A gene under the control of a regulatable promoter. In this case, CENP-A expression can either be maintained at differential levels or perhaps be reinitiated at different time intervals after deletion of the endogenous gene to determine at what point centromeric levels drop below a critical threshold. Another option would be to take advantage of a previously developed system where CENP-A is fused to the AID tag, which allows for inducible proteasome mediated degradation within a few hours after addition of a cell exogenous hormone (Holland et al, 2012). This system may allow for the determination of the minimum amount of CENP-A, either by washing out the inducing agent prior to complete degradation, or by co-expression of unresponsive CENP-A at differential levels (e.g. using promoters of different strength). However, a confounding factor of any of these approaches is that all centromeres are affected simultaneously, making it difficult to determine the minimum amount of CENP-A on any surviving centromere is. One method that could potentially 0vercome this drawback is chromophore-assisted laser inactivation (CALI), which applies photosensitive molecules that produce reactive oxygen species (ROS) upon light induction to selectively destroy specific proteins at high spatial resolution (Jay, 1988; Liao et al, 1994; Wang et al, 1996). Effective, genetically encoded photosensitizer protein tags have been developed (Bulina et al, 2006; Takemoto et al, 2013), which can be fused to target proteins such as CENP-A and allow for its selective destruction from

234

Conclusions

individual centromeres. However, one drawback of CALI is that it is not as well established as many other lab techniques and experiments may suffer from unanticipated obstacles. Taken together, complex experiments will be inevitable to determine the critical amount of CENP-A for centromere maintenance. Nevertheless, these experiments should be pursued, as a successful assay will provide fundamental insights regarding the epigenetic nature of centromere inheritance.

De novo centromere formation There is an inherent conflict between a high and low critical amount of CENP-A to specify exactly one centromeric locus per chromosome. On the one hand, a low threshold decreases the chance of losing a centromere due to stochastic redistribution, while on the other hand it increases the chance of forming an additional centromere on an ectopic locus due to random accumulation. Interestingly, however, differences exist between the processes of centromere maintenance and centromere formation. One clear example of this is the differential role of CENP-B, which is essential for de novo centromere formation on human artificial chromosomes (Ohzeki et al, 2002). Conversely, this protein is dispensable for maintenance of existing centromeres, as evidenced by knock-out mice, which are perfectly viable, reproductively normal, and do not show any mitotic or meiotic abnormalities (Hudson et al, 1998). Similarly, the minimum amount of CENP-A to maintain an existing centromere may differ from the critical amount required to initiate a centromere on a naïve chromatin domain. It is difficult to estimate from the existing literature how much CENP-A is required for de novo centromere formation. Stable, self-sustaining neocentromeres have been produced experimentally using a number of methods. One example is human artificial chromosomes, which are produced by introducing large fragments of alphoid DNA (~60–70 kb) into cells and selecting for their retention (Ikeno et al, 1998; Ohzeki et al, 2002, 2012). Alternatively, centromeres of existing chromosomes have been

235

Chapter 5

deleted, which allowed for the isolation of clones containing neocentromeres on a ectopic sites in fission yeast (Ishii et al, 2008), Candida albicans (Ketel et al, 2009), or chicken DT40 cells (Shang et al, 2013). Moreover, self- sustaining centromeres can be induced by tethering CENP-A to a LacO array in Drosophila S2 cells (Mendiburo et al, 2011) or tethering of HJURP, CENP-C, or CENP-I to acentric chromosomes of chicken DT40 cells (Hori et al, 2013). However, in none of these cases was there control or measurement of the amount of CENP-A recruited, especially in the initial phase of centromere formation. Determining the critical amount of CENP-A required for de novo centromere formation would ideally involve the recruitment of a controlled number of molecules to a naïve site. One option would be to tether CENP-A (or HJURP) to LacO arrays of different sizes and determine what the smallest number of binding sites is that can initiate a centromere. A similar, yet slightly more elegant strategy would be to take advantage of the CRISPR system, which allows for targeting of fusion proteins to unique loci by using guide RNAs that complement genomic sequences (Chen et al, 2013; Qi et al, 2013). Using multiple guide RNAs to target CENP-A to neighboring sites would in principle allow for the titration of the minimum amount required to initiate centromere formation. Moreover, this system could be used to determine both the ideal distribution of CENP-A (high density in a small region or lower density in a slightly larger domain) as well as the role of the genomic context (transcriptional activity, histone modification density, etc.) on the efficiency of neocentromere formation. Thus, an effective CRISPR- mediated de novo centromere formation assay would be instrumental in answering many open questions regarding the processes leading to neocentromere formation.

236

Conclusions

CONCLUDING REMARK

Centromeres have intrigued cell biologists for over a century. Since their first description by Walther Flemming (1880), a great deal has been discovered regarding the function and nature of centromeres. Key discoveries include their multiple regulatory roles ensuring accurate chromosome segregation, as well as their epigenetic mode of inheritance. I myself have also made a contribution to our understanding of centromere inheritance. Nevertheless, many key questions remain unanswered. Some of these have been discussed in detail earlier in this chapter, including why a substantial proportion of CENP-A is non-centromeric, how CENP-A protein can be indefinitely retained at centromeres, how centromeric levels are accurately maintained, and what the minimal amount of CENP-A to specify centromere identity is. Other intriguing questions include how it is ensured that there is exactly one centromere per chromosome and how centromeric loci are accurately maintained in the apparent absence of physical boundaries. Taken together, the centromere field still has a long way to go and may provide enough study material to keep us going for another hundred years.

237

Chapter 5

REFERENCES

Ahmad K & Henikoff S (2002) Histone H3 variants specify modes of chromatin assembly. Proc. Natl. Acad. Sci. U. S. A. 99: 16477–16484 Alabert C & Groth A (2012) Chromatin replication and epigenome maintenance. Nat. Rev. Mol. Cell Biol. 13: 153–167 Albertson DG & Thomson JN (1982) The kinetochores of Caenorhabditis elegans. Chromosoma 86: 409–428 Ambartsumyan G, Gill RK, Perez SD, Conway D, Vincent J, Dalal Y & Clark AT (2010) Centromere protein A dynamics in human pluripotent stem cell self- renewal, differentiation and DNA damage. Hum. Mol. Genet. 19: 3970–3982 Barnhart MC, Kuich PHJL, Stellfox ME, Ward JA, Bassett EA, Black BE & Foltz DR (2011) HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochore. J. Cell Biol. 194: 229–243 Bassett EA, DeNizio J, Barnhart-Dailey MC, Panchenko T, Sekulic N, Rogers DJ, Foltz DR & Black BE (2012) HJURP uses distinct CENP-A surfaces to recognize and to stabilize CENP-A/histone H4 for centromere assembly. Dev. Cell 22: 749– 762 Bernad R, Sánchez P, Rivera T, Rodríguez-Corsino M, Boyarchuk E, Vassias I, Ray- Gallet D, Arnaoutov A, Dasso M, Almouzni G & Losada A (2011) Xenopus HJURP and condensin II are required for CENP-A assembly. J. Cell Biol. 192: 569–582 Black BE, Brock MA, Bédard S, Woods VL & Cleveland DW (2007a) An epigenetic mark generated by the incorporation of CENP-A into centromeric nucleosomes. Proc. Natl. Acad. Sci. U. S. A. 104: 5008–5013 Black BE, Foltz DR, Chakravarthy S, Luger K, Woods VL & Cleveland DW (2004) Structural determinants for generating centromeric chromatin. Nature 430: 578– 582 Black BE, Jansen LET, Maddox PS, Foltz DR, Desai AB, Shah JV & Cleveland DW (2007b) Centromere identity maintained by nucleosomes assembled with histone H3 containing the CENP-A targeting domain. Mol. Cell 25: 309–322 Brown WRA & Xu Z (2009) The ‘kinetochore maintenance loop’—The mark of regulation? BioEssays 31: 228–236 Bulina ME, Chudakov DM, Britanova OV, Yanushevich YG, Staroverov DB, Chepurnykh TV, Merzlyak EM, Shkrob MA, Lukyanov S & Lukyanov KA (2006) A genetically encoded photosensitizer. Nat. Biotechnol. 24: 95–99 Camahort R, Shivaraju M, Mattingly M, Li B, Nakanishi S, Zhu D, Shilatifard A, Workman JL & Gerton JL (2009) Cse4 is part of an octameric nucleosome in budding yeast. Mol. Cell 35: 794–805 Carroll CW, Milks KJ & Straight AF (2010) Dual recognition of CENP-A nucleosomes is required for centromere assembly. J. Cell Biol. 189: 1143–1155 Carroll CW, Silva MCC, Godek KM, Jansen LET & Straight AF (2009) Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N. Nat. Cell Biol. 11: 896–902

238

Conclusions

Carroll L (1871) Through the Looking-Glass, and What Alice Found There. London, England: Macmillan Chen B, Gilbert LA, Cimini BA, Schnitzbauer J, Zhang W, Li G-W, Park J, Blackburn EH, Weissman JS, Qi LS & Huang B (2013) Dynamic Imaging of Genomic Loci in Living Human Cells by an Optimized CRISPR/Cas System. Cell 155: 1479–1491 Chmátal L, Gabriel SI, Mitsainas GP, Martínez-Vargas J, Ventura J, Searle JB, Schultz RM & Lampson MA (2014) Centromere Strength Provides the Cell Biological Basis for Meiotic Drive and Karyotype Evolution in Mice. Curr. Biol. 24: 2295–2300 Choi ES, Strålfors A, Catania S, Castillo AG, Svensson JP, Pidoux AL, Ekwall K & Allshire RC (2012) Factors That Promote H3 Chromatin Integrity during Transcription Prevent Promiscuous Deposition of CENP-ACnp1 in Fission Yeast. PLoS Genet 8: e1002985 Dambacher S, Deng W, Hahn M, Sadic D, Fröhlich J, Nuber A, Hoischen C, Diekmann S, Leonhardt H & Schotta G (2012) CENP-C facilitates the recruitment of M18BP1 to centromeric chromatin. Nucleus 3: 101–110 Drané P, Ouararhni K, Depaux A, Shuaib M & Hamiche A (2010) The death- associated protein DAXX is a novel histone chaperone involved in the replication- independent deposition of H3.3. Genes Dev. 24: 1253–1265 Dunleavy EM, Beier NL, Gorgescu W, Tang J, Costes SV & Karpen GH (2012) The Cell Cycle Timing of Centromeric Chromatin Assembly in Drosophila Meiosis Is Distinct from Mitosis Yet Requires CAL1 and CENP-C. PLoS Biol 10: e1001460 Dunleavy EM, Roche D, Tagami H, Lacoste N, Ray-Gallet D, Nakamura Y, Daigo Y, Nakatani Y & Almouzni-Pettinotti G (2009) HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell 137: 485–497 Earnshaw WC, Allshire RC, Black BE, Bloom K, Brinkley BR, Brown W, Cheeseman IM, Choo KHA, et al (2013) Esperanto for histones: CENP-A, not CenH3, is the centromeric histone H3 variant. Chromosome Res. 21: 101–106 Earnshaw WC & Cleveland DW (2013) CENP-A and the CENP nomenclature: response to Talbert and Henikoff. Trends Genet. TIG 29: 500–502 Earnshaw WC & Rothfield N (1985) Identification of a family of human centromere proteins using autoimmune sera from patients with scleroderma. Chromosoma 91: 313–321 Fachinetti D, Diego Folco H, Nechemia-Arbely Y, Valente LP, Nguyen K, Wong AJ, Zhu Q, Holland AJ, Desai A, Jansen LET & Cleveland DW (2013) A two-step mechanism for epigenetic specification of centromere identity and function. Nat. Cell Biol. 15: 1056–1066 Flemming W (1880) Beiträge zur Kenntniss der Zelle und ihrer Lebenserscheinungen. Theil II. Arch. Für Mikrosk. Anat. 18: 151–259 Foltz DR, Jansen LET, Bailey AO, Yates JR, Bassett EA, Wood S, Black BE & Cleveland DW (2009) Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell 137: 472–484

239

Chapter 5

Foltz DR, Jansen LET, Black BE, Bailey AO, Yates JR & Cleveland DW (2006) The human CENP-A centromeric nucleosome-associated complex. Nat. Cell Biol. 8: 458–469 Fujita Y, Hayashi T, Kiyomitsu T, Toyoda Y, Kokubu A, Obuse C & Yanagida M (2007) Priming of Centromere for CENP-A Recruitment by Human hMis18α, hMis18β, and M18BP1. Dev. Cell 12: 17–30 Gascoigne KE & Cheeseman IM (2013) CDK-dependent phosphorylation and nuclear exclusion coordinately control kinetochore assembly state. J. Cell Biol. 201: 23–32 Gascoigne KE, Takeuchi K, Suzuki A, Hori T, Fukagawa T & Cheeseman IM (2011) Induced ectopic kinetochore assembly bypasses the requirement for CENP-A nucleosomes. Cell 145: 410–422 Gassmann R, Rechtsteiner A, Yuen KW, Muroyama A, Egelhofer T, Gaydos L, Barron F, Maddox P, et al (2012) An inverse relationship to germline transcription defines centromeric chromatin in C. elegans. Nature 484: 534–537 1448BC) Decalogue. 2nd ed. Mount Sinai: Moses) (יהוה) God Goldberg AD, Banaszynski LA, Noh K-M, Lewis PW, Elsaesser SJ, Stadler S, Dewell S, Law M, et al (2010) Distinct factors control histone variant H3.3 localization at specific genomic regions. Cell 140: 678–691 Groth A, Rocha W, Verreault A & Almouzni G (2007) Chromatin challenges during DNA replication and repair. Cell 128: 721–733 Guldberg CM & Waage P (1867) Études sur les affinités chimiques. Brøgger & Christie Guse A, Carroll CW, Moree B, Fuller CJ & Straight AF (2011) In vitro centromere and kinetochore assembly on defined chromatin templates. Nature 477: 354–358 Hagstrom KA, Holmes VF, Cozzarelli NR & Meyer BJ (2002) C. elegans condensin promotes mitotic chromosome architecture, centromere organization, and sister chromatid segregation during mitosis and meiosis. Genes Dev. 16: 729–742 Hasson D, Panchenko T, Salimian KJ, Salman MU, Sekulic N, Alonso A, Warburton PE & Black BE (2013) The octamer is the major form of CENP-A nucleosomes at human centromeres. Nat. Struct. Mol. Biol. 20: 687–695 Hayden KE, Strome ED, Merrett SL, Lee H-R, Rudd MK & Willard HF (2013) Sequences Associated with Centromere Competency in the Human Genome. Mol. Cell. Biol. 33: 763–772 Hellwig D, Emmerth S, Ulbricht T, Döring V, Hoischen C, Martin R, Samora CP, McAinsh AD, Carroll CW, Straight AF, Meraldi P & Diekmann S (2011) Dynamics of CENP-N kinetochore binding during the cell cycle. J. Cell Sci. 124: 3871–3883 Hemmerich P, Weidtkamp-Peters S, Hoischen C, Schmiedeberg L, Erliandri I & Diekmann S (2008) Dynamics of inner kinetochore assembly and maintenance in living cells. J. Cell Biol. 180: 1101–1114 Holland AJ & Cleveland DW (2009) Boveri revisited: chromosomal instability, aneuploidy and tumorigenesis. Nat. Rev. Mol. Cell Biol. 10: 478–487

240

Conclusions

Holland AJ, Fachinetti D, Han JS & Cleveland DW (2012) Inducible, reversible system for the rapid and complete degradation of proteins in mammalian cells. Proc. Natl. Acad. Sci. 109: E3350–E3357 Van Hooser AA, Ouspenski II, Gregson HC, Starr DA, Yen TJ, Goldberg ML, Yokomori K, Earnshaw WC, Sullivan KF & Brinkley BR (2001) Specification of kinetochore-forming chromatin by the histone H3 variant CENP-A. J. Cell Sci. 114: 3529–3542 Hori T, Shang W-H, Takeuchi K & Fukagawa T (2013) The CCAN recruits CENP-A to the centromere and forms the structural core for kinetochore assembly. J. Cell Biol. 200: 45–60 Hudson DF, Fowler KJ, Earle E, Saffery R, Kalitsis P, Trowell H, Hill J, Wreford NG, et al (1998) Centromere Protein B Null Mice are Mitotically and Meiotically Normal but Have Lower Body and Testis Weights. J. Cell Biol. 141: 309–319 Hu H, Liu Y, Wang M, Fang J, Huang H, Yang N, Li Y, Wang J, Yao X, Shi Y, Li G & Xu R-M (2011) Structure of a CENP-A-histone H4 heterodimer in complex with chaperone HJURP. Genes Dev. 25: 901–906 Ikeno M, Grimes B, Okazaki T, Nakano M, Saitoh K, Hoshino H, McGill NI, Cooke H & Masumoto H (1998) Construction of YAC–based mammalian artificial chromosomes. Nat. Biotechnol. 16: 431–439 Ishii K, Ogiyama Y, Chikashige Y, Soejima S, Masuda F, Kakuma T, Hiraoka Y & Takahashi K (2008) Heterochromatin Integrity Affects Chromosome Reorganization After Centromere Dysfunction. Science 321: 1088–1091 Jansen LET, Black BE, Foltz DR & Cleveland DW (2007) Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176: 795–805 Jay DG (1988) Selective destruction of protein function by chromophore-assisted laser inactivation. Proc. Natl. Acad. Sci. 85: 5454–5458 Kato H, Jiang J, Zhou B-R, Rozendaal M, Feng H, Ghirlando R, Xiao TS, Straight AF & Bai Y (2013) A Conserved Mechanism for Centromeric Nucleosome Recognition by Centromere Protein CENP-C. Science 340: 1110–1113 Ketel C, Wang HSW, McClellan M, Bouchonville K, Selmecki A, Lahav T, Gerami- Nejad M & Berman J (2009) Neocentromeres Form Efficiently at Multiple Possible Loci in Candida albicans. PLoS Genet 5: e1000400 Kimura H & Cook PR (2001) Kinetics of Core Histones in Living Human Cells Little Exchange of H3 and H4 and Some Rapid Exchange of H2b. J. Cell Biol. 153: 1341–1354 Lacoste N, Woolfe A, Tachiwana H, Garea AV, Barth T, Cantaloube S, Kurumizaka H, Imhof A & Almouzni G (2014) Mislocalization of the Centromeric Histone Variant CenH3/CENP-A in Human Cells Depends on the Chaperone DAXX. Mol. Cell 53: 631–644 Lagana A, Dorn JF, De Rop V, Ladouceur A-M, Maddox AS & Maddox PS (2010) A small GTPase molecular switch regulates epigenetic centromere maintenance by stabilizing newly incorporated CENP-A. Nat. Cell Biol. 12: 1186–1193

241

Chapter 5

Lee S-H, Itkin-Ansari P & Levine F (2010) CENP-A, a protein required for chromosome segregation in mitosis, declines with age in islet but not exocrine cells. Aging 2: 785 Liao JC, Roider J & Jay DG (1994) Chromophore-assisted laser inactivation of proteins is mediated by the photogeneration of free radicals. Proc. Natl. Acad. Sci. 91: 2659–2663 Liu S-T, Rattner JB, Jablonski SA & Yen TJ (2006) Mapping the assembly pathways that specify formation of the trilaminar kinetochore plates in human cells. J. Cell Biol. 175: 41–53 Maddox PS, Hyndman F, Monen J, Oegema K & Desai A (2007) Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 176: 757–763 McKinley KL & Cheeseman IM (2014) Polo-like Kinase 1 Licenses CENP-A Deposition at Centromeres. Cell 158: 397–411 Mendiburo MJ, Padeken J, Fülöp S, Schepers A & Heun P (2011) Drosophila CENH3 is sufficient for centromere formation. Science 334: 686–690 Monen J, Maddox PS, Hyndman F, Oegema K & Desai A (2005) Differential role of CENP-A in the segregation of holocentric C. elegans chromosomes during meiosis and mitosis. Nat. Cell Biol. 7: 1248–1255 Moree B, Meyer CB, Fuller CJ & Straight AF (2011) CENP-C recruits M18BP1 to centromeres to promote CENP-A chromatin assembly. J. Cell Biol. 194: 855–871 Müller S, Montes de Oca R, Lacoste N, Dingli F, Loew D & Almouzni G (2014) Phosphorylation and DNA Binding of HJURP Determine Its Centromeric Recruitment and Function in CenH3CENP-A Loading. Cell Rep. 8: 190–203 Ohzeki J, Bergmann JH, Kouprina N, Noskov VN, Nakano M, Kimura H, Earnshaw WC, Larionov V & Masumoto H (2012) Breaking the HAC Barrier: histone H3K9 acetyl/methyl balance regulates CENP-A assembly. EMBO J. 31: 2391–2402 Ohzeki J, Nakano M, Okada T & Masumoto H (2002) CENP-B box is required for de novo centromere chromatin assembly on human alphoid DNA. J. Cell Biol. 159: 765–775 Okada M, Cheeseman IM, Hori T, Okawa K, McLeod IX, Yates JR 3rd, Desai A & Fukagawa T (2006) The CENP-H-I complex is required for the efficient incorporation of newly synthesized CENP-A into centromeres. Nat. Cell Biol. 8: 446–457 Okada M, Okawa K, Isobe T & Fukagawa T (2009) CENP-H–containing complex facilitates centromere deposition of CENP-A in cooperation with FACT and CHD1. Mol. Biol. Cell 20: 3986–3995 O’Keefe RT, Henderson SC & Spector DL (1992) Dynamic organization of DNA replication in mammalian cell nuclei: spatially and temporally defined replication of chromosome-specific alpha-satellite DNA sequences. J. Cell Biol. 116: 1095– 1110 Oliveira RA, Coelho PA & Sunkel CE (2005) The condensin I subunit Barren/CAP-H is essential for the structural integrity of centromeric heterochromatin during mitosis. Mol. Cell. Biol. 25: 8971–8984

242

Conclusions

Palmer DK, O’Day K & Margolis RL (1990) The centromere specific histone CENP-A is selectively retained in discrete foci in mammalian sperm nuclei. Chromosoma 100: 32–36 Palmer DK, O’Day K, Trong HL, Charbonneau H & Margolis RL (1991) Purification of the centromere-specific protein CENP-A and demonstration that it is a distinctive histone. Proc. Natl. Acad. Sci. U. S. A. 88: 3734–3738 Qi LS, Larson MH, Gilbert LA, Doudna JA, Weissman JS, Arkin AP & Lim WA (2013) Repurposing CRISPR as an RNA-Guided Platform for Sequence-Specific Control of Gene Expression. Cell 152: 1173–1183 Raychaudhuri N, Dubruille R, Orsi GA, Bagheri HC, Loppin B & Lehner CF (2012) Transgenerational propagation and quantitative maintenance of paternal centromeres depends on Cid/Cenp-A presence in Drosophila sperm. PLoS Biol. 10: e1001434 Ray-Gallet D, Quivy J-P, Scamps C, Martini EM-D, Lipinski M & Almouzni G (2002) HIRA is critical for a nucleosome assembly pathway independent of DNA synthesis. Mol. Cell 9: 1091–1100 Ray-Gallet D, Woolfe A, Vassias I, Pellentz C, Lacoste N, Puri A, Schultz DC, Pchelintsev NA, Adams PD, Jansen LET & Almouzni G (2011) Dynamics of histone H3 deposition in vivo reveal a nucleosome gap-filling mechanism for H3.3 to maintain chromatin integrity. Mol. Cell 44: 928–941 Schuh M, Lehner CF & Heidmann S (2007) Incorporation of Drosophila CID/CENP-A and CENP-C into centromeres during early embryonic anaphase. Curr. Biol. CB 17: 237–243 Shang W-H, Hori T, Martins NMC, Toyoda A, Misu S, Monma N, Hiratani I, Maeshima K, Ikeo K, Fujiyama A, Kimura H, Earnshaw WC & Fukagawa T (2013) Chromosome engineering allows the efficient isolation of vertebrate neocentromeres. Dev. Cell 24: 635–648 Shelby RD, Monier K & Sullivan KF (2000) Chromatin assembly at kinetochores is uncoupled from DNA replication. J. Cell Biol. 151: 1113–1118 Shelby RD, Vafa O & Sullivan KF (1997) Assembly of CENP-A into centromeric chromatin requires a cooperative array of nucleosomal DNA contact sites. J. Cell Biol. 136: 501–513 Shuaib M, Ouararhni K, Dimitrov S & Hamiche A (2010) HJURP binds CENP-A via a highly conserved N-terminal domain and mediates its deposition at centromeres. Proc. Natl. Acad. Sci. U. S. A. 107: 1349–1354 Silva MCC, Bodor DL, Stellfox ME, Martins NMC, Hochegger H, Foltz DR & Jansen LET (2012) Cdk activity couples epigenetic centromere inheritance to cell cycle progression. Dev. Cell 22: 52–63 Spence JM, Critcher R, Ebersole TA, Valdivia MM, Earnshaw WC, Fukagawa T & Farr CJ (2002) Co‐localization of centromere activity, proteins and topoisomerase II within a subdomain of the major human X α‐satellite array. EMBO J. 21: 5269– 5280 Sullivan LL, Boivin CD, Mravinac B, Song IY & Sullivan BA (2011) Genomic size of CENP-A domain is proportional to total alpha satellite array size at human

243

Chapter 5

centromeres and expands in cancer cells. Chromosome Res. Int. J. Mol. Supramol. Evol. Asp. Chromosome Biol. 19: 457–470 Tachiwana H, Kagawa W, Shiga T, Osakabe A, Miya Y, Saito K, Hayashi-Takanaka Y, Oda T, Sato M, Park S-Y, Kimura H & Kurumizaka H (2011) Crystal structure of the human centromeric nucleosome containing CENP-A. Nature 476: 232–235 Takemoto K, Matsuda T, Sakai N, Fu D, Noda M, Uchiyama S, Kotera I, Arai Y, et al (2013) SuperNova, a monomeric photosensitizing fluorescent protein for chromophore-assisted light inactivation. Sci. Rep. 3: 2629 Talbert PB, Ahmad K, Almouzni G, Ausió J, Berger F, Bhalla PL, Bonner WM, Cande WZ, et al (2012) A unified phylogeny-based nomenclature for histone variants. Epigenetics Chromatin 5: 7 Talbert PB & Henikoff S (2013) Phylogeny as the basis for naming histones. Trends Genet. 29: 499–500 Thiru P, Kern DM, McKinley KL, Monda JK, Rago F, Su K-C, Tsinman T, Yarar D, Bell GW & Cheeseman IM (2014) Kinetochore genes are coordinately up- regulated in human tumors as part of a FoxM1-related cell division program. Mol. Biol. Cell 25: 1983–1994 Waage P & Gulberg CM (1864) Studier over Afiniteten. Forh. Vidensk.-Selsk. Christiania: 35–45 Wang F-S, Wolenski JS, Cheney RE, Mooseker MS & Jay DG (1996) Function of Myosin-V in Filopodial Extension of Neuronal Growth Cones. Science 273: 660– 663 Wang J, Liu X, Dou Z, Chen L, Jiang H, Fu C, Fu G, Liu D, et al (2014) Mitotic Regulator Mis18β Interacts with and Specifies the Centromeric Assembly of Molecular Chaperone Holliday Junction Recognition Protein (HJURP). J. Biol. Chem. 289: 8326–8336 Warburton PE, Cooke CA, Bourassa S, Vafa O, Sullivan BA, Stetten G, Gimelli G, Warburton D, Tyler-Smith C, Sullivan KF, Poirier GG & Earnshaw WC (1997) Immunolocalization of CENP-A suggests a distinct nucleosome structure at the inner kinetochore plate of active centromeres. Curr. Biol. CB 7: 901–904 Wignall SM, Deehan R, Maresca TJ & Heald R (2003) The condensin complex is required for proper spindle assembly and chromosome segregation in Xenopus egg extracts. J. Cell Biol. 161: 1041–1051 Willard HF & Waye JS (1987) Hierarchical order in chromosome-specific human alpha satellite DNA. Trends Genet. 3: 192–198 Worcel A, Han S & Wong ML (1978) Assembly of newly replicated chromatin. Cell 15: 969–977 Yu Z, Zhou X, Wang W, Deng W, Fang J, Hu H, Wang Z, Li S, et al (2015) Dynamic Phosphorylation of CENP-A at Ser68 Orchestrates Its Cell-Cycle-Dependent Deposition at Centromeres. Dev. Cell 32: 68–81 Zeitlin SG, Baker NM, Chapados BR, Soutoglou E, Wang JYJ, Berns MW & Cleveland DW (2009) Double-strand DNA breaks recruit the centromeric histone CENP-A. Proc. Natl. Acad. Sci. U. S. A. 106: 15762–15767

244

Conclusions

Zeitlin SG, Chapados BR, Baker NM, Tai C, Slupphaug G & Wang JYJ (2011) Uracil DNA N-Glycosylase Promotes Assembly of Human Centromere Protein A. PLoS ONE 6: e17151 Zeitlin SG, Patel S, Kavli B & Slupphaug G (2005) Xenopus CENP-A assembly into chromatin requires base excision repair proteins. DNA Repair 4: 760–772

245