Characterizing the interactions between mouse nucleoplasmin and chromosomal proteins

(1)

Characterizing the Interactions Between Mouse Nucleoplasmin and Chromosomal Proteins

by Katherine Ellard

H.B.Sc., Lakehead University, 2009

A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of

Master of Science

in the Department of Biochemistry and Microbiology

(2)

Supervisory Committee

Characterizing the Interactions Between Mammalian Nucleoplasmin and Chromosomal Proteins

by Katherine Ellard

H.B.Sc., Lakehead University, 2009

Supervisory Committee

Dr. Juan Ausio (Department of Biochemistry and Microbiology) Supervisor

Dr. Caroline Cameron (Department of Biochemistry and Microbiology) Departmental Member

Dr. John Taylor (Department of Biology) Outside Member

(3)

Abstract

Supervisory Committee

Dr. Juan Ausio (Department of Biochemistry and Microbiology) Supervisor

Dr. Caroline Cameron (Department of Biochemistry and Microbiology) Departmental Member

Dr. John Taylor (Department of Biology) Outside Member

Abstract

The family of Nucleoplasmin (NPM) proteins play an important role in a number of

chromatin remodelling processes. The first NPM protein discovered in the eggs and oocytes of

Xenopus laevis was NPM2, a tissue specific histone chaperone. In Xenopus, NPM2 has been

linked to paternal chromatin decondensation following fertilization through the removal of

sperm proteins, nucleosome assembly through the storage and addition of H2A-H2B dimers

and apoptosis. In mammals, NPM2 correlates strongly with nucleolus-like bodies, and has been

suggested by various groups to differ in its roles when compared to the X. laevis homologue.

However, the exact roles of NPM2 in mammals remain to be fully elucidated. In this

dissertation, attempts are made to determine the physical interaction sites between mouse

NPM2 and core histone proteins, H2A, H2B, H3 and H4, as well as physical interactions between

mouse NPM2 and protamines (sperm proteins) P1 and P2.

Interaction sites between mouse NPM2 and various chromosomal proteins were

investigated using a number of different techniques. First, NPM2: chromosomal protein binding assays were attempted to determine the ratio of NPM2 to both core histones and protamines.

(4)

When visualized through 12% Native gels, NPM2 was determined to interact with histone

octamers at a molar ratio of 1-1.5 mol NPM2/mol histone octamer. Mouse sperm protamines were determined to form complexes with mouse NPM2 at a molar ratio of 2.5 mol

protamine/mol NPM2 (or mol protamine/0.4 mol NPM2).

Analytical Ultracentrifuge (AUC) analysis was conducted on NPM2 and chromosomal

proteins separately and in complex formation. Although determining that isolated, full length

mouse NPM2 exists in a pentamer form, attempts with AUC were unsuccessful in determining

specific NPM2:chromosomal protein binding affinity and complex formation.

Specific physical interaction sites between NPM2 and chromosomal proteins were

investigated using Cross Linking Mass Spectrometry. Here, a number of new interaction sites as

well as sites previously identified by other groups were determined. In combination, our results

present likely interaction sites between NPM2 and chromosomal proteins and represent an

(5)

List of Tables

Table 1. Cross-Linked Peptide Sequences of NPM2-Histones/Protamines Obtained from CXMS Data

(10)

List of Figures

Figure 1. The nucleosome core particle and chromatin Figure 2. NPM2 and NPM3 sequence alignment Figure 3. Sedimentation velocity data examples

Figure 4. Oligomerization and ESI-MS of full length NPM2

Figure 5. Isolation and purification of mouse protamines P1 and P2 Figure 6. Isolation of histone octamers from chicken erythrocyte cells

Figure 7. Complex formation of NPM2:chromosomal proteins at increasing NPM2 ratios Figure 8. Sedimentation coefficient (Sw,20) values for NPM2 at 20°C under various ionic strengths

Figure 9. Sedimentation coefficient (Sw,20) values of mutated X. laevis NPM2 pre (●) and post (■) anion exchange column purification obtained through AUC

Figure 10. Possible sites of interactions between NPM2 and chromosomal proteins determined through CXMS

Figure 11. Schematic representation and 3D structure of mouse NPM2 with highlighted chromosomal protein interaction sites

(11)

Abbreviations

ACN Acetonitrile

AML Acute Myeloid Leukemia Asf1 Anti-Silencing Function 1 AUC Analytical Ultracentrifuge A1 – A3 Acidic tract 1 – 3

CAF1 Chromatin Silencing Factor

CBDPS Cyanurbiotindipropionylsuccinimide

CM Chicken Histone Marker

CXMS Cross Linking Mass-Spectrometry

Da Daltons

DNA Deoxyribonucleic Acid

ESI-MS Electrospray Ionization Mass-Spectroscopy FACT Facilitates Chromatin Transcription

fNPM3 Full Length Mouse NPM3

FPLC Fast Protein Liquid Chromatography

HP1 Heterochromatin Protein 1

ITC Isothermal Calorimetry

MALDI Matrix Assisted Laser-Desorption/Ionization mass spectrometry MEP50 Methylosome Protein 50

mNPM2 Mutant Mouse NPM3

MNase Micrococcal Nuclease

Mr Molecular Weight

mxNPM2 Mutant X. laevis NPM2

NAP1 Nucleosome Assembly Protein NASP Nuclear Autoantigenic Sperm Protein NES Nuclear Export Signal

NLS Nuclear Localization Signal NLBs Nucleolus Like Bodies NLPs Nucleoplasmin Like Proteins NoLS Nucleolar Localization Signal

NPM Nucleplasmin

PM Protein Marker

PRMT5 Protein Arginine Methyltransferase PTM Posttranslational Modification

P1 Protamine 1

P2 Protamine 2

RD Replication Dependent

RI Replication Independent

RP- HPLC Reverse Phase - High Protein Liquid Chromatography SAXS Small Angle X-ray Scattering

SNBP Sperm Nuclear Basic Protein Sw,20 Sedimentation Coefficient

(12)

Acknowledgements

I would like to thank my supervisor, Dr. Juan Ausio, for his support and guidance during my time at the University of Victoria. I would also like to thank previous and current members of the Ausio and Nelson lab for their help and input into this project.

I am grateful for the time and support of my supervisory committee, Dr. Caroline Cameron and Dr. John Taylor. Also, a special thank you to Dr. Evans, Melinda, Deb and Sandra for always having time to talk or answer questions.

(13)

Chapter 1 Introduction to Chromatin and Nucleoplasmin (NPM)

Proteins

(14)

Note: In Chapter 1, sections of the following headings: Chromatin, Chromatin Remodelling

during Fertilization, Histone Chaperone Proteins, and Nucleophosmin/Nucleoplasmin (NPM) Proteins have been taken from the recently published article:

Finn, R. M., Ellard, K., Eirin-Lopez, J. M., & Ausio, J (2012). Vertebrate Nucleoplasmin

and NASP: Egg Histone Storage Proteins with Multiple Chaperone Activities. FASEB Epub

(15)

Chromatin

Within each cell, fundamental building blocks of deoxyribonucleic acid (DNA) and

proteins come together to form a macromolecular complex termed chromatin. This vast, highly

dynamic and yet remarkably organized assemblage of DNA and proteins found in all eukaryotic

cells (McGhee, Felsenfeld et al. 1980) allows for the compaction of a large quantity of DNA

within a cell’s nucleus. While the protein components of chromatin mainly consist of histones,

it can also include non-histone chromosomal proteins that contribute to both the structure and

function of a chromosome. The principal repeating unit of chromatin is termed the nucleosome

and is comprised of three major elements: the nucleosome core particle, linker DNA and linker

histones (Bednar, Horowitz et al. 1998). Repeating nucleosomes occur approximately every 200

base pairs, and with the exception of particular regions of the genome undergoing specific

genomic processes (such as actively transcribing genes or during DNA repair) all DNA in a

eukaryotic cell is assembled into nucleosomes or similar structures.

The famous “beads” on the “beads on a string” array that nucleosomes form represent nucleosome core particles (Figure 1A). These nucleosome core particles are comprised of a

histone octamer, which contains two of each canonical (or core) histones H2A, H2B, H3 and H4

around which approximately 146 base pairs of DNA are wrapped in 1.67 left-handed

superhelical turns (Luger, Mader et al. 1997). Here, two H3-H4 pairs interact through a 4-helix

bundle from H3 and H3’ ‘histone folds’, and result in the H3-H4 tetramer formation while

histones H2A-H2B, existing in a dimerized state, finish off the nucleosome core by flanking each

end of the H3-H4 tetramer through a homologous 4-helix bundle located between the H2B and

(16)

signifies the “string” between each “bead“(Figure 1A). This linker DNA can vary in length

depending on species and tissue type (Vignali and Workman 1998) and represents the binding site for linker histones (most often histone H1 and H1 variants). In doing so, linker DNA plays a

critical role in forming higher-order chromatin structures. Finally, linker histones (histone H1

and H5 family) and other non-histone chromatin-binding proteins associate with approximately

10-60 base pairs of linker DNA to construct higher order chromatin structures (Kornberg 1974;

Kornberg and Thomas 1974; Olins and Olins 1974). By acting as a shield between the negative

charge of linker DNA and adjacent nucleosomes, linker histones promote further chromatin

packaging (Thomas 1999).

The level of DNA compaction chromatin can achieve is extremely impressive; taking for

example the average human cell: each cell contains approximately 2 meters of DNA if stretched

out from end to end (Annunziato 2008). While it is estimated that the human body is comprised

of 50 trillion cells, this equates to each person containing approximately 100 trillion meters of

DNA! With this perspective, one can begin to imagine the accomplishment chromatin can

obtain. Not only are these resulting chromatin structures compacted in such a way as to meet

the physical constraints of the cell, they also confer necessary protection to the underlying

DNA. Furthermore, nucleosome structure and the arrangement of chromatin are critical in

determining the accessibility of DNA, as they can either block or allow the access of various

enzymes and factors that facilitate DNA-mediated reactions (Annunziato 2008). While essential

in eukaryotic gene regulation, higher order chromatin structures are continuously arranged and

rearranged throughout a cells life. Three major factors that can induce local and global changes

(17)

Kadonaga 2003), histone modifying enzymes (Marmorstein and Trievel 2009), and variants of

core and linker histones (Hardy and Robert 2010). In combination, these factors allow for rapid structural transitions and permit a number of genomic processes to take place successfully,

which include DNA replication, transcription, recombination and repair. A strong example of

the first group includes the Nucleoplasmin (NPM) family of histone chaperone proteins, and will

(18)

A

B

Figure 1. The Nucleosome Core Particle and Chromatin

A) Famous “beads on a string” conformation of histones and DNA viewed under Electron Microscopy by (Olins and Olins 2003). Linker histones were removed to visualize decondensed (or ‘open’) chromatin. Arrows point to the nucleosome core particle. B) Crystallographic structure of the NCP as determined by (Luger, 1997) displaying 146 bp

DNA (grey) wrapped around histones H2A (yellow), H2B (red), H3 (blue) and H4 (green). Disorganized histone tails can be seen protruding out the NCP (various colours).

H2A H2B H3 H4

Lodish, 2000

(19)

Chromatin Epigenetics

Epigenetics was first defined in the 1940s as the ‘interaction of genes with their

environment which bring the phenotype into being’. More recently, this description has

evolved to define the field of epigenetics as ‘the sum of alterations to the chromatin template

that collectively establish and propagate different patterns of gene expression and silencing

from the same genome’ (Goldberg, Allis & Bernstein, 2007). With the literal translation of ‘epi’

meaning ‘above’ in Latin, epigenetic marks along a genome greatly alter the output of genetic

information, allowing for significantly different readouts of an identical DNA sequence.

Epigenetic changes manifest themselves within the chromatin template through a number of ways, including posttranslational histone modifications (Strahl 2000), energy-dependent

chromatin-remodeling factors, the incorporation and shuffling of histone variants in and out of

the nucleosome (Hardy and Robert 2010), and the targeting role of small noncoding RNAs (Kawasaki 2005). In addition, DNA itself can be modified covalently through the methylation at

cytosine residue of (most often) CpG dinucleotides (Reik 2001), which serves a number of

biological purposes, including the silencing and activation of specific genes (Reik and Walter

2001) and genome stability (Chen, Pettersson et al. 1998). All modifications made to the

chromatin template are viewed as having a cis-effect, which alters the structure or charge due

to a change in chromatin organization, or trans-effect, which can increase affinity for

chromatin-associated proteins which exert downstream effects. Working together, these

modifications constitute the “histone code”, and are recognized and interact with specific

(20)

Histones, the protein component of chromatin, are often subjected to a variety of

covalent modifications, which include acetylation, methylation, phosphorylation, ubiquitination, ADP-ribosylation, SUMOylation, biotinylation and proline isomerization

(Goldberg, Allis & Bernstein, 2007). These modifications are major influences on the state of

chromatin, and can affect biological processes that involve gene replication, repair,

transcription and chromatin stability. However, due to the nature of histones used in this

dissertation, only relevant epigenetic marks in the context of other proteins will be discussed.

Here we isolated histones from chicken erythrocytes, and while histone octamers can be

isolated and purified from a number of different sources, chicken erythrocytes are ideal as

histones can be obtained in large quantities, are easy to process and are highly stable during

storage. Additionally, these histones contain low levels of post-translational modifications and

have an identical amino acid sequence to that of human histones, making them an attractive

source of histone octamers (Peterson and Hansen 2008).

Phosphorylation

Serine and threonine represent two of the most common residues that are subjected to

the covalent modification of phosphorylation. This addition or removal of a negatively charged

phosphate group is carried out by two types of enzymes; kinases and phosphatases. In the

context of chromatin, negative charges conveyed to DNA binding proteins (for example,

histones) can alter the nucleosome and lead to major structural and functional alterations

(21)

The phosphorylation of proteins has been linked to the regulation of signalling

processes as well as protein activation; a strong example of the latter includes phosphorylation of the histone chaperone protein Nucleoplasmin (NPM2). An oocyte specific protein, NPM2 has

previously been determined to be involved in histone storage in oogenesis, decondensation of

sperm chromatin after fertilization, and nucleosome assembly in early embryonic cells

(Dingwall C 1987; Philpott 1992). Upon phosphorylation by casein kinase II (Vancurova 1995)

and mitosis promoting factor (Cotten 1986) during egg maturation, cytoplasmic Xenopus NPM2

is able to transport into the nucleus and exert its effects on sperm chromatin. Phosphorylation

is the most prominent and functionally significant modification to affect this group of proteins,

and is strongly correlated with NPMs ability to decondense paternal chromatin in fertilized X.

laevis eggs (Banuelos, Omaetxebarria et al. 2007). Furthermore, dephosphorylation of X. laevis

NPM2 led to an inability to decondense sperm chromatin through sperm protein removal (Leno

1996).

Methylation

Methylation, the addition of a methyl group to arginine and lysine residues in histones

or to cytosine nucleotides in DNA, represents another prominent mechanism for genome

regulation. In the context of histones, methylation was one of the first covalent modifications

to be discovered by (Alfrey 1964) in histone H3 and H4. Depending on the target residue and

the degree to which the residue is methylated (mono-,di- or tri-methylation), this modification

(22)

lysine 4 is generally associated with active chromatin, methylation of histone H3 lysine 9 can

signify inactive chromatin (Volkel and Angrand 2007). Often times methylation of histones can recruit and act as docking sites for effector proteins, (for example, methylation of H3K4 and

recruitment of heterochromatin protein HP1) (Bannister, Zegerman et al. 2001) further

illustrating the complexity and significance of this modification.

While histones and DNA can undergo extensive methylation, histone chaperone

proteins are also affected by this covalent modification. Recent studies have revealed that in

the oocytes of X. laevis, NPM2 is subjected to mono- and di-methylation at arginine 187 within

the conserved motif (GRGXK) by the protein arginine methyltransferase 5 (PRMT5) and

methylosome protein 50 (MEP50) complex (Wilczek 2011). The same study also determined

that methylation of H2A and H4 in X. laevis eggs by the same methyltransferase complex was

modulated by NPM2, suggesting that NPM2 may play an important role in global histone

(23)

Histones

Histones make up the protein component of nucleosomes in almost all nucleated

eukaryotic cells, and can be generally described as small, basic, and highly conserved proteins

composed of a globular domain and flexible histone ‘tails’ (Goldberg, Allis & Bernstein, 2007).

These positively charged proteins come into contact with negatively charged DNA through

hydrogen bonds, as a total of 14 contact points have been determined through X-ray

crystallography (Luger, Mader et al. 1997). Histones can be subdivided into two major groups:

core histones constituting the nucleosome core particle, and linker histones which bind to

linker DNA. Posttranslational modifications (PTMs) of histones constitute a key mechanism by which chromatin is regulated both structurally and functionally, including the acetylation of

core histone lysines to regulate transcription or the recruitment of chromatin binding proteins

(Strahl 2000). Furthermore, a number of histone variants and histone-like proteins exist throughout the eukaryotic kingdom, which further adds to the complexity of gene regulation.

Core Histones

The core histones mentioned previously (H2A, H2B, H3 and H4) constitute the histone

octamer. Although sharing little sequence identity, specific motifs have remained preserved

among all core histones. One well characterized motif includes the ‘histone fold’, isolated from

a variety of organisms, including insects (Xie, Kokubo et al. 1996), chicken (Arents, Burlingame

(24)

al. 1996). Interestingly, archaeal bacteria contain archaeal nucleosomes homologous to the

classical H3-H4 tetramer of eukaryotic nucleosomes, a genome packaging trait not seen in the Bacteria biological domain (Pereira and Reeve 1998). These motifs that constitute the ‘histone

fold’ act as a fundamental core histone dimerization motif (often referred to as the handshake

motif) (Arents, Burlingame et al. 1991) and help dock DNA onto the histone octamer (Arents

and Moudrianakis 1993). The tertiary structure of the ‘histone fold domain’ contains three

α-helices joined together by two loops as determined through X-ray crystallography (Luger,

Mader et al. 1997).

Protruding away from the histone fold are N- and C-terminal tail sequences, which

comprise a substantial part of each core histone (approximately 28% of a histones total mass).

While all core histones contain a N-terminal tail domain, H2A is unique in that it contains a long

(approximately 15 amino acids in length) C-terminal extension that helps stabilize the

nucleosome core particle and appears to act as a recognition module for linker histone H1

(Vogler, Huber et al. 2010). Histone tails have been determined to affect nucleosome stability,

as mutations to H2A and H3 histone tails resulted in a number of nucleosome abnormalities,

including the unwrapping of DNA near the edge of the nucleosome, the rate of nucleosome

sliding on DNA and the rate of H2A-H2B dimer exchange (Ferreira, Somers et al. 2007). Histone

tails also interact with adjacent nucleosomes, as seen with the tail of histone H4 and seven

negatively charged residues within the H2A-H2B dimer (Luger, Mader et al. 1997), and can

additionally act as a ‘docking’ site for a variety of chromatin modifying complexes. To add to

their significance, histone tails are frequently targeted by the vast majority of post translational

(25)

chromatin structure. Possibly one of the best studied histone tail modifications is the

acetylation of lysine 16 on the tail of histone H4, strongly linked to transcriptional activation (along with other H3 and H4 tail lysine acetylations) (Akhtar and Becker 2000). Overall, core

histones are stable proteins, but can be degraded or evicted from the nucleosome at a higher

rate depending on certain modifications and the cellular context in which they reside. While

some groups have determined that the half-life of an average histone lies between 93 and 105

days (Piha, Cuenod et al. 1966), others have recorded upwards of 223 days(Commerford,

Carsten et al. 1982).

Linker Histones

Linker histones, in contrast to their core histone counterparts, are most notably

characterized by their winged helix domain and completely lack a histone-fold domain (Happel

and Doenecke 2009). Approximately 75 amino acids in length, linker histones are comprised of

a globular domain made of three helix bundles and three anti-parallel beta-sheets which are

flanked by a long, lysine-rich C-terminal tail and a shorter (usually basic) N-terminal extension

(Ramakrishnan 1993). Through the interactions with 10-60 base pairs of linker DNA, linker

histones allow for two full turns of DNA around the nucleosome and proper formation of the 30

nm chromatin fiber (van Holde 1988). Not only does this offer protection to the underlying

DNA, in vitro studies have shown that linker histones add further stabilization to the

nucleosome while limiting the nucleosomes mobility along the DNA template (Grigoryev 2001)

(26)

chromatin binding proteins and suggest that linker histones can play the role of transcriptional

repressor. In vitro, it has been shown that phosphorylation of linker histones contribute significantly to both the general and specific inhibition of chromatin remodeling complexes

(Horn, Carruthers et al. 2002).

Linker histones are less conserved when compared to core histones, with the highest

amount of amino acid divergence found in their unstructured N- and C-terminal regions

(Kasinsky, Lewis et al. 2001). In humans alone, 11 subtypes of linker histone H1 exist (Happel

and Doenecke 2009). Some H1 variants are redundant in nature, while others expressed only in

specific tissues, such as histone H5 from avian nucleated erythrocytes (Neelin, Callahan et al.

1964) and testis-specific H1 variants H1t, H1T2 and HILS1 (Ausio 2006).

Histone Variants

Histone variants represent yet another way in which genetic information is altered and

arranged within a eukaryotic cell. Taking the place of core histones, histone variants often act

as a signal within chromatin to carry out specialized functions, as in the case of histone variant

H3.3. Differing from histone H3 by four amino acids, H3.3 contains a large amount of ‘active’

histone modifications and thus located in a more ‘active’ and less compacted chromatin region

(Ahmad and Henikoff 2002). Within a cell, covalent ‘markers’ present on histones can act as a

signal for the replacement by histone variants, allowing for rapid turnover of histones during

(27)

Histone variants also make an appearance during one of the most dramatic examples of

chromatin remodelling: mammalian spermiogenesis and egg fertilization. As chromatin

undergoes rapid reorganization in male germ cells during sperm cell maturation, testis-specific

histone variants take the place of traditional histones, which are then replaced by small nuclear

basic proteins (SNPBs) to allow for a higher DNA compaction ratio within the sperm cell head.

Two histone H1 testis-specific variants, H1t and HILS1, occur sequentially during spermatid

nucleus elongation and condensing (Ausio 2006). An oocyte-specific linker histone also exists,

H1oo, yet its exact function remains poorly understood (Tanaka 2003). Histone H3 contains five

histone variants, with two variants making an appearance in the maturing sperm cell.

Testis-specific H3 (H3t) and non-testis Testis-specific H3.3 appear in spermatids and are linked with fertility

levels and active transcription during meiosis, respectively (Couldrey 1999; Lewis 2003). Three

testis-specific variants of histone H2B exist, which include testis-specific H2B (TH2B), human

testis-specific H2B (hTSH2B) and family member W testis-specific H2B (H2BFWT) (Ausio 2006).

To date, one H2A testis-specific variant (TH2A) has been identified in pachytene spermatocytes,

while non-testis specific H2A variants H2A.X and macroH2A also make an appearance in the

spermatozoon (Govin 2004).While some histone variants are almost identical to their

traditional counterparts (for example, H3.3 only differing by 4-5 amino acids), others are almost

(28)

Protamines

First discovered in 1874 by Swiss Biochemist Friederich Miescher (Vaughn 1966),

protamines first symbolized an ‘organic base’ found in the nuclei of salmon sperm. Belonging

to a well-studied group of Sperm Nuclear Basic proteins (SNBPs), protamines are present only in

sperm cells, and can be isolated from a wide variety of vertebrate and invertebrate systems

(Ausio 2007). These small, highly basic proteins constitute the vast majority of DNA-packaging

proteins within the mammalian sperm cell. By taking the place of histones, protamines allow for

a significantly higher compaction ratio within mature sperm (Gaucher 2009). To date, the

structure of protamine proteins remain poorly understood. However, they are believed to bind and wrap around the major groove of DNA and form an extensive network of hydrogen and

electrostatic bonds with DNA phosphate groups (Balhorn 2007).

Protamines have been identified in polyplacophors (ie. chitin), gastropods,

cephalopods a number of insects, algae and in vertebrates (Wouters-Tyrou 1998; Balhorn

2007). While rare in all other chromatin interacting proteins, cysteine’s have become

extensively incorporated into protamines over time, a unique characteristic allowing for the

formation of inter-chromatin fibre associations which in turn promote nucleoprotein

stabilization and DNA compaction (Gimenez-Bonafe 2002; Ausio 2007).The most common

amino acid present in these proteins is arginine, making up >30 mol% of all amino acids and is

most often seen in clusters (Ausio 1999). Histidine and lysine are present in relatively high

percentages as well, with serine, threonine, and glycine occurring in lower yet significant

(29)

may help in protamine deposition onto the DNA template (Lewis 2003). All protamine genes

contain a poly-adenylated (poly-A) signal (Oliva 1990), a trait seen in replication-independent histone variant proteins. Without a poly-A tail, protamine transcription and translation in sperm

cells would occur concurrently and lead to improper sperm development. During translational

repression, a poly-A binding protein binds to the poly-A tail and subsequently arrests

translation until migrating upstream to a 3’UTR region (Marzluff, Wagner et al. 2008). This

characteristic is in contrast to canonical histone proteins, which are encoded by

replication-dependent genes and contain a unique 3’stem-loop structure critical for proper regulation

(Marzluff, Wagner et al. 2008). Although certain aspects among these proteins remain

conserved, protamines are one of the fastest evolving proteins identified to date (Eirin-Lopez,

2009), and can vary extensively in size, with the smallest protamine molecules present in fish

and the largest occurring in mollusks (Daban 1995).

P1 and P2

Protamine P1 is present in all protamine-containing species. Representing the major form of protamines, P1 is believed to have given rise to the now amino acid distinct P2. While

all eutharian mammals contain the P2 gene, its expression is not uniform. Absence of P2

expression seen in some is a result of in-gene mutations (Bower 1987), while in others it is due to transcriptional and translational suppression of the P2 gene (Balhorn 2007). In mammals,

(30)

haploinsufficiency of either protamine variant can lead to chromatin instability and infertility

(31)

Chromatin Remodelling during Fertilization

Chromatin continuously undergoes arrangement and rearrangement during a number

of cellular processes, including DNA transcription, translation, replication and repair. However,

one of the most extreme examples of chromatin rearrangement is thought to take place during

spermiogenesis (the maturation of a male sperm cell) and fertilization of the female egg. As

round spermatids elongate into motile spermatozoon, paternal chromatin undergoes extensive

remodelling and condensation to result in a compact and transcriptionally silent form of DNA

able to fit within the constraints of a sperm cell head (McLay 2003), (Govin 2004), (Gaucher

2009). With sperm head size affecting both sperm motility and function (Ausio 2007; Gillies 2009), proper genome compaction within this small space is essential for successful fertilization

(Miller, Brinkworth et al. 2010). A key component in this chromatin remodelling process

includes the replacement of histones by histone variants and highly specialized SNBPs, a group consisting of protamines (Rice, Garduno et al. 1995), protamine-like proteins (Rice, Garduno et

al. 1995) and sperm-specific histone variants (Rousseaux and Khochbin 2010). While there is no

clear mechanism as to how histones are removed during spermiogenesis, it is believed to occur

in three general stages: first, the incorporation of histone variants, followed by

hyperacetylation of both histones and histone variants, and finally the replacement of both

histones and histone variants by transition proteins or protamines (Gaucher, Reynoird et al.

2010). Replacement of histones by SNBPs is thought to occur during final post-meiotic phases

of spermatogenesis and results in a ten-fold compaction of paternal chromatin (Braun 2001).

(32)

compaction ratio, forming a crystalline-like structure unique from the common nucleosome

(McLay 2003).

Shortly following fertilization, sperm chromatin must be decondensed and re-organized

by egg-supplied proteins. This remodelling of paternal chromatin allows for the rapid fusion of the male and female pronuclei to permit proper zygote development (Philpott, 2000). During

this period, a group of histone chaperones come into play. While overseeing the chromatin

remodelling in both somatic and germ cells, histone chaperones are known to displace histone

variants and protamines while incorporating core and linker histones shortly following

fertilization. This group of proteins have become extensively studied, and two chaperone

families in particular, Nucleoplasmin (NPM) and Nuclear Autoantigenic Sperm Protein (NASP),

are known to play a key role during paternal chromatin remodelling following fertilization

(Richardson, Alekseev et al. 2006; Frehlick, Eirin-Lopez et al. 2007). While NPM will be the topic

of interest throughout this dissertation, key mechanisms and the importance of chaperone

(33)

Histone Chaperone Proteins

Before the discovery of molecular chaperone proteins, chromatin formation was

believed to occur through ‘self-assembly’; a process where nucleic acids and histones

spontaneously came together to create higher order chromatin structures without the

assistance from additional proteins (Morange 2005). However, in 1978, Laskey changed the way

the scientific community looked at protein complex formation when coining the term

‘molecular chaperone’ to explain the newly discovered Nucleoplasmin (NPM) protein, a protein

determined to assist in the formation of nucleosome structures by shielding protein charges

and preventing protein aggregation (Laskey, Honda et al. 1978). R.J. Ellis further expanded on the molecular chaperone concept to include proteins that assist in the post-translational

assembly of other protein complexes (Ellis, van der Vies et al. 1989). Today, chaperones are

known to play an integral part in the formation of various cellular complexes, and are used to describe an immense group of proteins that assist in the post-translational processing of

proteins to ensure proper folding and protein complex formation. A diverse range of proteins

involved in chromatin reorganization represents a strong example of a chaperone, and are

specifically referred to as histone chaperone proteins.

In addition to histone chaperone proteins, ATP-dependent chromatin remodelling

complexes, non-histone chromatin-associated proteins and post-translational modification

enzymes interact with the nucleosome throughout various stages of a cell’s lifecycle in an effort

to deposit, evict, and store histone and histone variant proteins (Choudhary and Varga-Weisz

(34)

with the nucleosome during key cellular processes (Winkler 2011), histone chaperones are

essential for the prevention of DNA damage, genomic instability, cell cycle arrest and cell death (Alekseev, Bencic et al. 2003; Gunjan and Verreault 2003; Morillo-Huesca, Maya et al. 2010;

Singh, Liang et al. 2010). It is essential that histone chaperone proteins mediate nucleosome

assembly and disassembly in a timely fashion as well as be able to accommodate a wide array

of nucleosome structures that may be encountered across the genome at any given time

(Luger, Mader et al. 1997; Davey, Sargent et al. 2002). Additionally, structural differences

amongst various types of cells add to the complexity of the role of these proteins and are all

critical factors that must be taken into consideration during nucleosome remodelling.

Research surrounding histone chaperones have primarily targeted the characterization

of processes involved in the step-wise addition of core histones during nucleosome formation

in somatic cells. While less is known concerning the histone chaperone activity during egg

fertilization, histone chaperone proteins responsible for histone H3/H4 tetramer storage (and

possible linker histone H1) as well as histone H2A/H2B dimer storage have been identified as

the NASP (SHNi-TPR family) and the NPM family of proteins, respectively (Finn 2012).

Before newly synthesized histones can be incorporated into chromatin, they must first

be produced in the cytoplasm and protected from protein degradation and non-specific

interactions. In addition, appropriate histone levels must be strictly regulated on a translational

and post-translational level to prevent genomic instability or abnormal chromatin structures

(Gunjan, Paik et al. 2005). As histones are highly positive in nature, non-specific binding to DNA,

(35)

genome instability, DNA damage, miscreant protein-protein interactions and delays in cell cycle

progression, cell senescence or cell death (Alekseev, Bencic et al. 2003; Gunjan and Verreault 2003; Morillo-Huesca, Maya et al. 2010; Singh, Liang et al. 2010). Through the binding of newly

synthesized histones, chaperone proteins can prevent unwanted histone-protein interactions

and prepare histones for transfer to other chaperones in replication-dependent (RD) and

replication-independent (RI) pathways (Laskey, Honda et al. 1978; Laskey, Honda et al. 1978;

Laskey, Kearsey et al. 1985; Campos, Fillingham et al. 2010; Cook, Gurard-Levin et al. 2011) as

required.

Steps involving H3/H4 incorporation into the nucleosome has served as a model for how

histone chaperone networks work (Tagami, Ray-Gallet et al. 2004; Campos, Fillingham et al.

2010; Campos and Reinberg 2010; Das and Tyler 2011; Formosa 2011; Hamiche and Shuaib

2011; Keck and Pemberton 2011). Specifically, focus has been placed on the events that follow

the binding of H3.1/H4 or H3.3/H4 to anti-silencing function 1 (Asf1) (Mello, Sillje et al. 2002;

Tagami, Ray-Gallet et al. 2004). By controlling replication fork advancement and interacting

with old histones evicted from nucleosomes as well as newly synthesized histones that are on

route to nucleosome incorporation, (Groth, Rocha et al. 2007), Asf1 transfers

replication-dependent H3/H4 dimers to the chromatin silencing factor (CAF-1) complex. Following histone

transfer, CAF-1 incorporates H3/H4 tetramers onto the DNA template in a DNA-synthesis

coupled process (Mello, Sillje et al. 2002; Tagami, Ray-Gallet et al. 2004; Groth, Rocha et al.

2007). FACT (facilitates chromatin transcription) and NAP-1 (nucleosome assembly protein)

complexes are responsible for facilitating H2A/H2B dimer deposition and subsequent

(36)

et al. 2009) (Del Rosario and Pemberton 2008). While the steps following Asf1, NAP1 and to

some extent FACT’s role in nucleosome assembly have been the primary focus of research, it has come to light that other families of chaperones (ie. NPM and NASP) play important roles in

the proceeding steps leading to nucleosome formation, specifically within the egg of

vertebrates. It has also been suggested that, when under the right cellular context or cell type,

the NPM and NASP family of chaperones can take the role of FACT, Asf1, and NAP1, directly or

indirectly assembling or disassembling nucleosomes not only in vitro (Earnshaw, Honda et al.

(37)

Nucleophosmin/Nucleoplasmin (NPM) Proteins

Ubiquitously expressed throughout the animal kingdom, the NPM family of nuclear

chaperone proteins have been shown to play diverse functions during key cellular processes.

The founding member, Nucleoplasmin (sometimes referred to in the literature as NP, NPM, or

NPM2), was initially labelled as the archetypal molecular chaperone, and became a point of

reference for molecular chaperones discovered thereafter (Dingwall and Laskey 1990). Further

research went on to discover other NPM members, and today this family can be categorized

into four main groups: nucleophosmin (NPM1), nucleoplasmin (NPM2),

nucleophosmin/nucleoplasmin 3 (NPM3) and nucleoplasmin-like proteins (NLPs). This family is expressed throughout eukaryotes and share similarities both in structure and amino acid

sequence (Figure 2). Members of this family have been shown to play a role in a large variety of

cellular processes and are essential for both chromatin remodelling following fertilization in

vivo (Burns, Viveiros et al. 2003; McLay and Clarke 2003; Lindstrom 2011) and nucleosome

assembly and disassembly in germinal and somatic cells (Okuwaki, Sumi et al. 2012). While

some members are expressed throughout all tissues, others are tissue specific and may vary in

their expression levels during different stages of development. Although abbreviations for

these proteins often differ in the literature, they will be referred throughout this dissertation as

(38)

NPM Structure

Crystallographic structures of the N-terminal domains have been determined for NPM1

in Xenopus (Namboodiri 2004) and humans (Lee 2007), NPM2 in Xenopus (Dutta 2001) and

humans (Platonova 2011) and NLPs in Drosophila melanogaster (Namboodiri 2003). Each of

these proteins is comprised of a protease resistant N-terminal core and a highly disordered

C-terminal tail. The N-C-terminal core is composed of an eight-stranded, beta barrel topology that is

believed to be responsible for oligomerization of its monomer subunits (Dutta 2001)

(Namboodiri 2003) as well as providing extreme thermal stability to the protein (Hierro 2002).

Interestingly, and despite the predicted tertiary structure similarity of the N-terminal region of all the NPM members (Frehlick, Eirin-Lopez et al. 2007), mammalian NPM3 appears to lack the

ability to form homo-pentameric structures of its own (Okuwaki, Sumi et al. 2012). Protruding

from the core domain is a highly disordered C-terminal tail present in all NPMs, which has been suggested to play a key role in ligand binding (Dutta 2001; Taneva 2009; Ramos 2010; Platonova

2011). Many conserved regions exist throughout certain NPM groups, including up to three

regions rich in glutamic and aspartic acid residues (often referred to as acidic tracts), a nuclear

localization signal, a nuclear export signal and a nucleolar localization signal which will be

(39)

Acidic Tracts

Each NPM family member contains two to three conserved acidic tracts consisting

mainly of glutamic acid (highlighted boxes in Figure 2). These acidic tracts are highly disordered

in nature and have been shown to play a role in binding to both core (Dutta 2001) and linker

histones(Ramos 2005). The first short acidic tract (A1) present within the N-terminal core is

believed to interact with chromosomal proteins (Salvany 2004), giving this family its

chaperoning qualities. Furthermore, this short tract present in the N-terminal core has been

linked to NPM2’s ability to oligomerize and form a pentamer structure(Arnan, Saperas et al.

2003; Namboodiri, Akey et al. 2004). Interestingly enough, this tract is not present in mammalian NPM2. Instead, the region where the A1 is normally located only contains one

acidic residue (glutamic acid at residue 37 in Mus musculus noted by an arrowhead in Figure 2)

within the disordered A1 loop region located between two β-strands (Platonova 2011).

The large, polyglutamic tract located within the C-terminal tail of X. laevis NPM2

(denoted A2/black box in Figure 2) has also been suggested to interact with Xenopus sperm

proteins and help carry out its nucleosome disassembly abilities (Philpott 1991; Prieto 2002). In

mammals, this long A2 tract binds to core histones (H2A-H2B dimers) and form large

histone-NPM2 complexes in vitro (Platonova 2011) . Although being shown to bind histones with the

presence of only one major acidic tract, combinations of acidic patches may work synergistically

(40)

Figure 2. NPM sequence alignments

Amino acid sequence alignment of X.laevis NPM2, X. laevis NPM2(mutant), M. musculus NPM2,

M. musculus NPM1 and M.musculus NPM3 proteins. Basic (red) acidic (blue) neutral (green)

and aromatic (purple) residues are colour coded, with acidic tracts highlighted with a red (A1) black (A2) and green (A3) boxes. Conserved NoLs (blue circle) and NES (black dashes) domains are highlighted. Arrowhead denotes the single glutamic acid residue at site 37 in M. musculus NPM2. X. laevis Npm2 (NCBI Reference Sequence: NPM_Mouse, Q61937.1); M. musculus NPM2 (NCBI Reference Sequence: NP_851990.2); M. musculus NPM1(NCBI Reference Sequence: NPM_MOUSE, Q61937.1); M. musculus NPM3 (NCBI Reference Sequence: NP_032749.1).

(41)

Nuclear Localization Signal

Three of the four group members, NPM1, NPM2 and NPM3 all contain a bipartite

nuclear localization signal (NLS) (Dingwall C 1987). Located in the C-terminal tail domain, this

NLS is responsible for NPM transport into the nucleus from the cytoplasm. Carrier proteins

present in the nuclear pore complex of cells termed importins recognize the NLS in proteins and

allow for their import into the nucleus. Two importins, α-importin and β-importin are

responsible for NLS recognition and translocation of the NPM complex through the nuclear

pore, respectively (Pemberton 2005);(Dingwall 1990; Radu 1995). Both thermodynamic and

structural characterization of importin α/β with X. laevis NPM2 has been conducted using isothermal calorimetry (ITC) and small angle X-ray scattering (SAXS), and while phosphorylation

of residues flanking the NLS were taken into consideration, it does not appear that

phosphorylation modulates NPM2-importin interaction (Falces 2010).

Nucleolar Localization Signal

In addition to the NLS, NPM1 exclusively contains a a nucleolar localization signal (NoLS),

a nuclear export signal (NES) and a RNA-binding domain (Frehlick et al., 2006). Localized in the

nucleolus of cells, the NES allows for the transport of pre-ribosome subunits to the cytoplasm

by NPM1 when recognized by the nuclear export receptor CRM1 (Borer 1989), while the NoLS

and RNA-binding domain are essential for NPM1’s ribosome biogenesis function (Okuwaki

(42)

NPM1

The most extensively studied member of this family, Nucleophosmin 1 (NPM1; also

known as B23, numatrin, or NO38), is a well-characterized protein ubiquitously expressed

throughout different types of cells. A nucleolar phosphoprotein, NPM1 has many diverse roles

within the cell including ribosomal biogenesis (Yung 1985; Huang 2005); partly based on its

localization with pre-ribosomal ribonucleoprotein particles and its ability to facilitate ribosome

assembly (Szebeni and Olson 1999); genomic stability, as inactivation of the NPM1 gene in mice

led to noticeable defects in centrosome duplication (Grisendi 2005; Grisendi, Bernardi et al.

2005); DNA replication (Takemura 1999), transcriptional regulation (Swaminathan 2005); histone chaperoning and nucleosome assembly (Okuwaki 2001; Frehlick, Eirin-Lopez et al.

2007), with preferential binding seen with H3-H4 tetramers over H2A-H2B dimers and nucleic

acid binding (Wang 1994). The human NPM1 gene is a 23.8 kb gene composed of 10 exons, when spliced, at least 19 variants result with 13 of these having the potential to encode

proteins (Thierry-Mieg and Thierry-Mieg 2006). Two well-known subtypes of NPM1, B23.1 and

B23.2, differ at their C-termini and are believed to exert different effects within the cell, seen

through B23.1’s ability to bind tightly to DNA polymerase α-(dA)-(dT) complex when compared

to its B23.2 isoform (Umekawa, Sato et al. 2001).

NPM1 and Cancer

The abundance of research focussing around NPM1 is largely due to the fact that it

(43)

and correlates strongly with tumorigenesis (Nozawa, Van Belzen et al. 1996) and

haematopoietic malignancies (Falini, Mecucci et al. 2005). NPM1 has strong links to the MDM2/p53 tumor suppressor pathway and has been determined to play an important role in

p53 stability and transcriptional activation (Colombo, Marine et al. 2002). Strikingly, 55

different mutations in exon 12 (Rau and Brown 2009) of NPM1 are present in 35% of all AML

cases (Falini, Mecucci et al. 2005), suggesting that the altered NPM protein may have a direct

connection to AML progression. While mutations in the Npm1 gene correlate strongly with a

large percentage of AML cases, other events are required in addition to drive the onset of AML

in humans. Characteristic of Npm1 gene mutations include the addition of a NES motif and an

alteration in the NLS motif in the C-terminal region of the protein (Falini, Bolli et al. 2006).

Mutations leading to the disrupted NLS ultimately results in the irregular build-up of NPM1 in

the cytoplasm (Falini, Bolli et al. 2006). It still remains unclear as to whether this mutation in

the Npm1 gene can be inherited.

While mutations in NPM1 occur in a large percentage of AML cases, there appears to be

an absence of NPM1 gene mutations found in other forms of common solid cancers, including

cancer of the lungs, liver, breasts, colon and gastric system (Jeong, Lee et al. 2007). Instead,

overexpression of NPM1 has been noted in solid tumors from various histological origins

(Nozawa, Van Belzen et al. 1996; Pianta, Puppin et al. 2010). In both instances, the deregulation

of NPM1 contributes to tumorigenesis, and as NPM1 is linked to cell proliferation and growth

(Okuwaki 2001) (aspects critical for the spread of cancer cells), it is not surprising that this

(44)

NPM2

The second member of this chaperone family, NPM2, was first isolated from egg

extracts of X. laevis. Here, NPM2 was determined to be the most abundant protein in oocyte

nuclei, making up 8-10% of all protein extract (Mills 1980). While the majority of literature has

focused on the study of NPM2 in Xenopus, recent studies have turned their attention to the

role of NPM2 in human and mouse models. A tissue specific protein, NPM2 has been linked to

chromatin decondensation (Philpott 1991), nucleosome assembly (Laskey, Honda et al. 1978)

and apoptosis (Lu 2005) in X. laevis. In mammals, NPM2 has been linked to the formation of

nucleolus like bodies (Inoue 2010; Inoue 2011) and may differ in its histone chaperoning abilities compared to its Xenopus homologue (Burns, Viveiros et al. 2003). The human NPM2

gene is a 1.1 kb gene composed 10 exons that is alternatively spliced and predicted to have at

least 11 splice variants; 9 of these variants potentially encoding proteins (Thierry-Mieg and Thierry-Mieg 2006) with the predominant transcript producing a 214 amino acid protein.

NPM2 as a Histone Chaperone

One of the first described functions of NPM2 in X. laevis was its histone chaperoning

activities. Here, NPM2 was seen binding to large pools of oocyte supplied histones H2A and H2B in preparation for egg fertilization and rapid nucleosome assembly (Laskey, Honda et al. 1978;

Earnshaw, Honda et al. 1980). Upon fertilization, NPM2 is known to transfer histones H2A and

(45)

Laskey, Honda et al. 1978; Kleinschmidt, Fortkamp et al. 1985; Laskey, Kearsey et al. 1985;

Kleinschmidt, Dingwall et al. 1986; Kleinschmidt and Seiter 1988). Early on, NPM2 was shown to only bind to H2A and H2B, while H3 and H4 in Xenopus are bound to the chaperone protein N1

(NASP being its mammalian homologue) (Dilworth 1987). Together, these two proteins are able

to supply enough histones to assemble 12,000 diploid nuclei (Laskey 1993).

At the molecular level, information available on the binding of histones to NPM2 has

been mainly obtained with the Xenopus model. Despite the acidic nature of this molecule and

the basic charge of its interacting histone partners, it has been shown that hydrophobic

interactions play a critical role more so than that of the electrostatic component (Arnan,

Saperas et al. 2003; Taneva, Banuelos et al. 2009). Interestingly, functional phosphorylation

activation of NPM2 enhances the binding affinity for histones. However, such increase appears

to be related to changes in the quaternary pentameric rearrangement of the molecule rather

than in the negative charge increase associated with phosphorylation (Taneva, Munoz et al.

2008). In its pentameric form, Xenopus NPM2 can bind up to 5 molecules of H2A/H2B and linker

histones with decreasing affinity as the number of histones bound increase (Taneva, Banuelos

et al. 2009). The differential affinity of nucleoplasmin for histone H5 (linker histones) and

H2A/H2B may facilitate its dual functional role as a storage histone chaperone and as a

chromatin remodeler (Taneva, Banuelos et al. 2009).

Possible chaperoning qualities of NPM2 remain poorly understood in mammals. One

study determined that H2A/H2B deposition onto paternal chromatin was unaltered by

(46)

NPM2 in humans contain histone chaperoning capabilities through the binding of H2A/H2B

dimers to its long polyglutamic tract (Platonova 2011). Functional disparity between

mammalian and Xenopus NPM2 invariably highlights a possible difference between NPM2s role

amongst various species (Frehlick, Eirin-Lopez et al. 2006), with a need for future research to

pinpoint NPM2s exact role in the context of histone chaperoning.

NPM2 in Nucleosome Disassembly

NPM2 in X. laevis also assists in the removal of sperm proteins (Prieto 2002) before

H2A/H2B deposition onto paternal chromatin. Removal of these sperm proteins, one of the

earliest events in fertilization (Poccia 1996), is necessary for histone deposition and proper nucleosome formation. NPM2 has been shown to remove at least two Xenopus sperm proteins

(X and Y), and depletion of this chaperone completely abolishes early chromatin decondensing

in fertilized eggs (Philpott 1992). Injection of NPM2 alone is as efficient at chromatin

decondensing as whole egg extract, emphasizing its importance during early development

(Laskey 1978; Dingwall C 1987; Philpott 1992). NPM2 has also displayed an ability to

decondense chromatin of introduced somatic cell nuclei (Barry 1972; Lohka 1983), but its

explicit expression in mature oocytes and eggs (Frehlick 2006) limits its chromatin remodelling

capabilities to these types of cells.

Mentioned previously, NPM2s ability to disassemble paternal chromatin is highly

dependent on its level of phosphorylation. During oocyte maturation NPM2 becomes

(47)

immature oocyte to approximately 7-10 in the mature egg, stopping at the mid-blastula

transition stage (Vancurova 1995). With 5 subunits forming a pentamer, NPM2 has been estimated to gain up to 70-100 phosphates in total (Philpott 2000). This hyperphosphorylation

occurs at sites within the N-terminal core and C-terminal tail of the protein identified with mass

spectrometry analysis of Xenopus egg NPM2 and computational prediction analysis (Bañuelos

2007). In addition, Banuelos and colleagues (2007) determined that both the N- and C-terminal

domains of NPM2 must become phosphorylated to maximize NPM2s decondensing efficiency.

It is important to note that while oocyte derived NPM2 (low in ptms) is still able to decondense

sperm nuclei, the process occurs at such a slow rate that it would be physiologically incapable

of meeting the demands of zygote development (Leno 1996). Clearly, phosphorylation is critical

in determining NPM2s chromatin remodelling efficiency, and to date has been the most

influential modification on this group of proteins.

It has been suggested that the A2 tract of X. laevis NPM2 acts as a binding site to

Xenopus sperm proteins. Additionally, stoichiometric binding assays of X. laevis NPM2 and

protamines from Dicentrarchus labrax (sea bass) have determined that one mol of NPM2 is

present for every 2.5 mol of protamine (Prieto 2002).

In contrast to X. laevis, studies have shown that NPM2 alone in Mus musculus (the

common house mouse) does not appear to be sufficient for sperm DNA decondensation after

fertilization. Npm2-knockout studies conducted by Burns and colleagues (2003) determined

that eggs from females lacking NPM2 still displayed normal sperm DNA decondensation with no

(48)

could be playing a different role in mammalian oocytes, and may suggests that other NPM

family members could be compensating in mammals during early zygote development (Burns, Viveiros et al. 2003; McLay and Clarke 2003). However, the current research on this topic is

extremely limited, as only a handful of studies have just begun uncovering NPM2s role in

chromatin assembly and disassembly in mice and humans.

NPM2 and Nucleolus-Like Bodies

While NPM2s role in sperm chromatin decondensation and protamine removal in mice

has yet to be fully understood, it appears to exert some effects within the fertilized egg during

early embryogenesis (Inoue 2011). Specifically, NPM2 has been implemented in the formation

of nucleolus-like bodies (NLBs), and although these NLBs are poorly understood, they are

known to localize in the germinal vesicles of mammalian oocytes and diffuse into the oocyte

cytoplasm until reassembly occurs in the pronuclei (Flechon 1998; Ogushi 2008). Here, nucleoli

form around the NLBs (which once in the pronuclei are referred to as nucleoli like precursors).

When defects in mice nuclei were identified in the Npm2-knockout studies (Burns, Viveiros et

al. 2003), the link between NPM2 and NLBs began to surface. Seven years later, a 16 amino acid

lysine-rich motif in the C-terminus region of NPM2 was shown to directly regulate NLB

formation (Inoue 2010). Shortly after, NLBs were linked to sperm chromatin decondensation as

its depletion from oocytes results in highly condensed paternal chromatin (Inoue 2011).

Interestingly enough, NPM2 expression was also lost when NLBs were removed, and

(49)

NPM2 may be the NLB component responsible for chromatin remodelling. While some data

determined by Inoue and colleagues (2010, 2011) backs up findings by Burns (2003), their somewhat contradicting evidence concerning NPM2’s chromatin decondensing abilities

highlights the important need for further research.

NPM2 in Apoptosis

NPM2 in X. laevis has also been linked to chromatin decondensation during apoptosis. In

vertebrates, ooyctes that go unused eventually deplete through apoptotic cell death, which

includes chromatin remodelling. Phosphorylation of a tyrosine residue at position 124 in

Xenopus NPM2 has previously been linked to apoptotic chromatin condensation (Lu 2005), with

dephosphorylation of this residue inactivating NPM2 and subsequently allowing for apoptotic

chromatin condensation to occur. Therefore, NPM2 in Xenopus may act as a general

(50)

NPM3

Twenty years following the discovery of NPM2, the third member of the NPM family,

NPM3, was isolated and characterized in mouse (MacArthur and Shackleford 1997). Similar to

NPM1, NPM3 is ubiquitously expressed across various tissues (Shackleford 2001), and while the

exact roles of this protein are still being uncovered, NPM3 has been linked to chromatin

reorganization following fertilization (McLay 2003) and in the regulation of NPM1 activity

during ribosomal RNA genesis (Huang 2005). More recently, the human NPM3 gene has been

described as a 2.1 kb gene composed 10 exons that is alternatively spliced and predicted to

have at least 5 splice variants; 3 of these variants potentially encode proteins (Thierry-Mieg and Thierry-Mieg 2006).

NPM3 and Chromatin Remodelling

NPM3 is also expressed within mammalian oocytes, and research suggests that this

protein may be critical for proper sperm chromatin remodelling. In response, one group has

looked further into processes involved during sperm chromatin decondensation in mammals by

specifically targeting two oocyteproteins: Nucleosome assembly protein (or NAP-1) and NPM3

(McLay 2003). Here, researchers determined that microinjection of NPM3 and NAP-1 antisense

oligonucleotides together, or NPM3 antisense oligonucleotides alone, in mouse oocytes

abolished proper histone assembly (detected by immunohistone presence) and subsequently

hindered the ability of paternal chromatin to progress beyond a dispersed state. This finding

(51)

oocyte, a role known to be played by NPM2 in Xenopus. However, this finding is the only one of

its nature, underscoring the need for future experiments in order to determine the exact role of NPM family members present within mammalian eggs.

NPM3 and NPM1

Previous yeast-two hybrid screens of NPM1 interacting proteins identified NPM3 as a

major interacting partner (Huang 2005). A strong technique in the study of protein-protein

interactions, yeast-two hybrid screening allows for the identification of interacting domains

through the use of Saccharomyces cerevisiae. If interactions between two proteins occurs, a

reporter gene is transcriptionally activated and can allow for the growth of yeast on (usually

nutrient lacking) media (Young 1998). While a strong technique, our inability to access the yeast

two hybrid screen and the availability of more precise, less time consuming methods led us to

opt for other protein-protein interaction study experiments.

The same study by Huang and colleagues (2005) went on to investigate NPM3’s

influence on NPM1, where it demonstrated an ability to alter pre-rRNA synthesis and

processing; a function that NPM1 has previously been linked to (Yung, Busch et al. 1985). In

contrast with NPM2, NPM3 and NPM1 are both widely expressed in mammalian tissue

(Shackleford 2001). Also, it has recently been suggested that while NPM3 is unable to form a

homo-pentameric structure on its own, one monomer of NPM3 can form heterologous

(52)

Dissertation Outline

The intent of this dissertation is to discuss the work done to characterize the in vitro

interactions between NPM proteins and core histones and protamines. The interaction

between X. laevis NPM2 and chromosomal proteins has been previously examined, with more

recent studies turning their focus to mammalian NPM2 and core histone complex formation.

Although X. laevis NPM2 was first believed to interact with chromosomal proteins through its

acidic tracts (A1, A2 and A3); to date, only the A1 tract has been associated with chromosomal

binding capabilities (Salvany et al., 2004). In humans, it was recently suggested that the long A2

tract may in fact bind to histones (Dutta et al., 2001; Prieto et al., 2002; Arnan et al., 2003; Banuelos et al., 2003). The following chapter elaborates on these studies by investigating the

possible physical interaction sites between core histones and full length mouse NPM2. In

addition, an attempt to investigate possible binding sites between mouse NPM2 and mouse protamines P1 and P2 is carried out. Previous literature has failed to look at possible interaction

sites between mouse protamines and mouse NPM2, making experiments attempted here novel

in nature. Although efforts to isolate and purify mouse NPM3 were attempted to identify NPM3

interaction sites with chromosomal proteins, they were ultimately unsuccessful. This

dissertation will therefore focus mainly on mouse NPM2. As NPM2 is known to play a key role

in the early stages of fertilization in both Xenopus and mammals, our work here hopes to

further elucidate the role of mouse NPM2 in the context of a histone chaperone. Focus

particularly on the mammalian model will build on research that has mainly focused around the

(53)