Identifying Endogenous Binding Partners for Btf and Trap150

IDENTIFYING ENDOGENOUS BINDING PARTNERS FOR BTF AND TRAP150

A Thesis submitted in partial fulfillment of the

requirements for the degree of

Master of Science

JAYLEN BRAXTON HUDSON

B.S. UNIVERSITY OF DAYTON, 2018

2020

Wright State University

WRIGHT STATE UNIVERSITY GRADUATE SCHOOL

April 28, 2020

I HEREBY RECOMMEND THAT THE THESIS PREPARED UNDER MY SUPERVISION BY Jaylen Braxton Hudson ENTITLED Identifying Endogenous Binding Partners for Btf and TRAP150 BE ACCEPTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Master of Science.

______Paula A. Bubulya, Ph.D. Thesis Director

______David L. Goldstein, Ph.D. Chair, Department of Biological Sciences

Committee on Final Examination:

______Quan Zhong, Ph.D.

______Labib Rouhana, Ph.D.

______Barry Milligan, Ph.D. Interim Dean of the Graduate School

ABSTRACT

Hudson, Jaylen Braxton. M.S. Department of Biological Sciences, Wright State University, 2020. Identifying Endogenous Binding Partners of Btf and TRAP150.

Since being discovered in the early 1990s, the primary functions of classical serine- arginine rich (SR) proteins and non-classical SR-related proteins have been demonstrated in gene regulatory processes such as pre-mRNA processing, mRNA metabolism, and nuclear export of mRNAs. Bcl-2-associated transcription factor 1 (BCLAF1, also called Btf) and Thyroid hormone receptor associated protein 3 (THRAP3, also called TRAP150) are homologous non-classical

SR-like splicing factors. In this thesis, I have identified endogenous binding partners of Btf and

TRAP150. I used statistical analysis to select common versus distinct protein partners and compared previously documented subcellular localization of these identified proteins with studies for Btf/TRAP150. Previous studies indicate that Btf and TRAP150 have overlapping roles in maintaining cell cycle progression and DNA damage repair, but different roles in regulating subcellular mRNA distribution. Therefore, I hypothesized that Btf and TRAP150 interact with a common set of protein partners for pathways in which they share functions yet have distinct protein partners that impart unique functions. MS-based proteomic analysis of immunoprecipitated Btf and TRAP150 complexes revealed novel interactions with FXR1P and

FXR2P, autosomal paralogs of Fragile X mental retardation protein (FMRP). FXR1P is a common partner for both Btf and TRAP150 whereas FXR2P is distinct to TRAP150. Future studies will confirm interaction between these binding partners as well as others from the analysis. Also, siRNA depletion studies will investigate the mechanisms for how Btf/TRAP150 impact subcellular distribution and function of Fragile X proteins.

iii

Table of Contents

CHAPTER 1: SIGNIFICANCE AND BACKGROUND………………………………….…..1

1.2 Nuclear speckles…………………………………………………….…..14

1.3 SR and SR-related proteins…………………………………...……….19

1.4 Btf and TRAP150…………………………………………………………19

1.5 Subcellular mRNA distribution………………………………..………23

1.6 Regulation of DNA Damage Repair transcripts………..……..……24

CHAPTER 2: HYPOTHESIS AND EXPERIMENTAL AIMS……………………………..27

2.1 Specific Aims…………………………………………………….……….27

CHAPTER 3: MATERIALS AND METHODS…………………………………..………….31

3.1 Thawing/Freezing Cells………………………………………...……....31

3.2 Cell Culture…………………………………………………….………….31

3.3 Plating Cells………………………………………………………………32

3.4 Whole Cell Nuclear Extraction (WCNE) preparation ……...………32

3.5 Protein Quantification (Bradford Assay)………………….…………33

3.6 SDS-PAGE and Immunoblotting………………………………………33

3.7 Immunoprecipitation……………………………….……………………35

3.8 Silver Staining of SDS-PAGE gels……………………………………36

3.9 Mass Spectrometry and Proteomics…………….…………………...37

3.10 Immunofluorescence…………………………….…………………….37

CHAPTER 4: RESULTS OF IMMUNOPRECIPITATION AND MASS SPECTROMETRY…………………………………………………………………………….39

4.1 Isolation of Btf and TRAP150 complexes by Immunoprecipitation (IP)…………...... 39

4.2 Using silver staining to detect endogenous binding partners…..47

4.3 Mass spectrometry results and data analysis…………………...…54

CHAPTER 5: DISCUSSION………………………………………………………………….80

REFERENCES………………………………………………………………………….……..87

ABBREVIATIONS……………………………………………………………………..………95

List of Figures

Figure 1. A schematic diagram of co/post-transcriptional regulation of mRNAs….4

Figure 2: Pre-mRNA splicing in the absence of an SR protein RS domain…………8

Figure 3: Regulation of post-transcriptional gene expression and protein-protein

interaction by the EJC……………………………………………………………………….12

Figure 4: A schematic diagram of nuclear domains in the mammalian cell

nucleus………………………………………………………………………………………...15

Figure 5: Diversity of gene expression from the regulation of nuclear speckle

proteins………………………………………………………………………………………...17

Figure 6: Alignment data for Btf and TRAP150 showing sequence similarity……21

Figure 7: Immunoblot images for Btf and TRAP150 IPs (Replicate #1)………….…41

Figure 8: Immunoblot images for Btf and TRAP150 IPs (Replicate #2)…………….43

Figure 9: Immunoblot images for Btf and TRAP150 IPs (Replicate #3)………….…45

Figure 10: Silver stain image for Btf/TRAP150 (Replicate #1)………………....…….48

Figure 11: Silver stain images for Btf/TRAP150 (Replicate #2)………………………50

Figure 12: Silver stain images for Btf/TRAP150 (Replicate #3)………………….…..52

Figure 13: Linear regression analysis for Btf binding partners...…….……………..74

Figure 14: Linear regression analysis for TRAP150 binding partners……………..77

vii

List of Tables

Table 1: Category description for variables in MS-based proteomics analysis.…55

Table 2: Quantification of identified endogenous binding partners for Btf and

TRAP150……………………………………………………………………………….………58

Table 3: Common binding partners for Btf and TRAP150 w/ unique biological function………………………………………………………………………………………...62

Table 4: Biological pathways associated with Btf binding partners..……….……..65

Table 5: Biological pathways associated with TRAP150 binding partners………..68

Table 6: Correlation between variables for Btf and TRAP150………………………..72

viii

ACKNOWLEDGMENTS

I would like to recognize and express sincere gratitude to my advisor, Dr. Paula Bubulya,

for her supervision of my research. Her constant support, guidance, and encouragement in the

past two years has helped me grow as a graduate student in many aspects. I would also like to

thank my Thesis committee consisting of Dr. Quan Zhong and Dr. Labib Rouhana. Their insight

as professors in Molecular Genetics have allowed me to develop important skills during my

study.

In addition, I am thankful for the students in the Bubulya lab, Melissa Ward, Jacob Ward,

and Rawan H. Alqahtani for our productive discussions during weekly meetings. I appreciate the

strong relationships that we were able to build during my time at Wright State University. I

greatly appreciate The Ohio State University for accepting me into their Molecular, Cellular, and

Developmental Biology Program. I look forward to doing future research along with their

professors and Ph. D. students. Lastly, I would like to thank my parents, Beverly Hudson and

Tyrone Stepter for showing support, as well as my brother James Hudson. I am grateful for all of

my loved ones, especially Sinclaire Smith for her belief in my life goals.

This study was supported by NIH Award grant 2R15GM084407-03 and -04. The

Bubulya lab would like to thank the Campus Chemical Instrument Center Mass Spectrometry and Proteomics Facility at The Ohio State University. Dr. Liwen Zhang provided valuable contributions to our endogenous binding partner study. The Fusion Orbitrap Instrument at OSU

was supported by NIH Award Grant S10 OD018056. The lab also thanks Dr. Volker Bahn at

Wright State University for aiding with R statistical analysis during the study. Gene Ontology

studies were supported by grant U41 HG002273 from the National Human Genome Research

Institute.

CHAPTER 1: SIGNIFICANCE AND BACKGROUND

1.1 INTRODUCTION

Gene expression is a fundamental multistep process that regulates all aspects of the

eukaryotic cell. The final gene products, proteins and RNA, are essential for maintaining cell

structure, function, and homeostasis. The events of gene expression are represented in the mRNA

production cycle which consists of transcription, pre-mRNA splicing, mRNA export, translation, mRNA stability, and mRNA decay (Martinez, 2018). While transcription is the most highly regulated step during RNA processing, co-transcriptional events such as pre-mRNA splicing have the potential to affect biological processes outside of gene expression (Li, 2006).

Although early studies suggest that the stages of gene expression are independent, genomic analyses confirm a functional connection and coordination between events (Komili,

2008). Several transcription factors and RNA processing factors are coupled through similar binding patterns that help recruit RNA polymerase II (Komili, 2008; White, 2010). Before nuclear export, the mRNA transcript undergoes 5’ capping, splicing, and 3’polyadenylation modifications coordinated by the carboxy terminal domain (CTD) of RNA polymerase II (Figure

1; Ramirez-Clavijo, 2013). After mRNA processing, the mature molecule is exported from the nucleus to the cytoplasm where it is impacted by other regulatory pathways including mRNA decay/stabilization, mRNA localization, and mRNA translation (Figure 1; Le Hir et al., 2001;

Woodward, 2017; Martinez, 2018).

The CTD of RNA polymerase II is a hypophosphorylated complex that contains several sites within its tandem heptapeptide consensus sequence ‘YSPTSPS’ (Rosonina, 2002). CTD

Ser-5 phosphorylation by TFIIH kinase allows the large subunit to function in several critical

steps of mRNA synthesis (Hong et al., 2009). The CTD ensures efficient co-transcriptional

modification by recruiting and loading RNA processing enzymes onto nascent pre-mRNAs.

Moreover, the CTD is known to couple mRNA processing to transcription through the formation

of a 5’ cap and the 3’ cleavage of transcripts (McCracken et al., 1997). The CTD also loads splicing factors onto pre-mRNA and serves as an activator for splicing machinery to form the pre-splicesomal A complex (McCracken et al., 1997; Millhouse, 2005).

The CTD acts as a flexible binding scaffold for a series of enzymes during the addition of the 7-methylguanosine cap to the mRNA transcript (Phatnani, 2006). During 5’ capping, triphosphatase, guanylyltransferase, and methyltransferase bind to CTD repeats determined by different phosphorylation patterns (Phatnani, 2006; Martinez-Rucobo et al. 2015). 5’ capping takes place after approximately 25 RNA nucleotides are transcribed (Shatkin, 2000). First, triphosphatase removes the terminal 5’ phosphate from the nascent pre-mRNA. Next, guanylyltransferase creates the guanosine cap by transferring a GMP to the RNA diphosphate end, forming a 5’-5’ link between the GTP and RNA molecule (Shibagki et al., 1992). Last, methyltransferase adds a methyl group to the G residue at the N7 position, resulting in a mature mRNA cap structure (Mao et al., 1995). The 5’ methyl cap protects the mRNA transcript from enzymatic degradation and regulates ribosome binding during translation.

The addition of the 3’ poly (A)+ tail is another co-transcriptional mechanism that helps

protect mRNA molecules from degradation in the cytoplasm. Polyadenylation is characterized by

3 different events: cleavage, 3’adenine addition, and the termination of transcription (Mandel,

2008). During the process, Cleavage and Polyadenylation specificity factor (CPSF) and Cleavage

stimulation factor (CstF) complexes are transferred to the pre-mRNA by the CTD then

stimulated by poly (A) signal sequences within the mRNA (Barilla, 2001). CPSFs and CstFs cleave the 3’ signaling region which is followed by the recruitment of poly (A) polymerase.

Next, poly (A) polymerase uses ATP to add approximately 200 adenines to the newly cleaved

RNA to complete the poly (A)+ tail (Barilla, 2001; Mandel, 2008).

Figure 1: A schematic diagram of co/post-transcriptional regulation of mRNAs

Co-transcriptional and post-transcriptional events are essential for producing an export- ready mRNA that becomes subject to additional regulatory pathways in the cytoplasm (MBL

Life Science, 2017).

(MBL Life Science, 2017)

Splicing is catalyzed in the spliceosome, a large ribonucleoprotein (RNP) complex

composed of five snRNAs (U1,U2,U4, U5, and U6) where specific components are recruited to

the 5’ and 3’ splice sites (Wang, Z., 2008; Will, 2011). SR/SR-related splicing factors remove the intervening, non-coding sequences (introns) and leave the coding sequences (exons) for expression during RNA assembly. Alternative splicing increases the complexity of the genome as mRNA/protein isoforms are synthesized with different functions and structures (Zhou, 2013).

Both constitutive and alternative splicing are regulated by trans-acting factors that either promote or antagonize exon inclusion. SR/SR-related proteins and hnRNPs compete with each other to bind to four cis-acting sequences: exonic splicing enhancers (ESEs), intronic splicing enhancers

(ISEs), exonic splicing silencers (ESSs), and/or intron splicing silencers (ISSs) (Busch, 2012).

The resulting splicing pattern for a particular mRNA strand can ultimately determine the subcellular localization and function of a gene product.

Pre-mRNA splicing requires both snRNP and non-snRNP factors, however, may involve

RS domain-dependent and/or RS domain-independent mechanisms (Figure 2). Among the non- snRNP factors are SR proteins which bind to ESEs through their RNA recognition motif(s) to

promote spliceosome assembly as well as splice-site selection (Shepard, 2009). The RS Domain of SR proteins recruit the U2AF35 splicing factor to the 3’ upstream splice-site and/or the U1-

70K snRNP to the 5’ downstream splice-site (Figure 2A; Zhu et al., 2000). Next, a heterodimer composed of U2AF35 and U2AF65 recruits U2snRNP to the branch point followed by U4/U6 and U5 snRNPs (Zhu et al., 2000; Graveley et al., 2001). Upon spliceosome formation, the additional snRNPs create a lariat-like intermediate for intron excision and exon ligation. These indirect interactions are linked through the splicing co-activator Srm160 (Figure 2A; Zhu et al.,

2000). By a different mechanism, inhibitory proteins bound to ESSs can impede the activity of

SR proteins lacking RS Domains (Figure 2B; Zhu et al., 2000). As a result, the SR proteins are unable to target U2AF35 to the AG splice-site, however, function to antagonize other negative effects of inhibitory proteins. Splicing may still occur through non-specific protein interactions at the exon regions (Zhu et al., 2000).

Figure 2: Pre-mRNA splicing in the absence of an SR protein RS domain

Pre-mRNA splicing can involve both RS domain-dependent and RS domain-independent mechanisms. RS domains are extensively phosphorylated and function as splicing recruiters for the U1 snRNP and U2AF complex/U2 snRNP. However, the RS domain can be dispensable for splicing through a combination of these mechanisms (Zhu et al., 2000).

(Zhu et al., 2000)

As a result of pre-mRNA splicing in the nucleus, several proteins are deposited onto the transcript approximately 20-24 nucleotides upstream of exon-exon junctions (Le Hir et al.,

2000). Among these examples is the exon junction complex (EJC), a sequence-independent

binding platform that assembles spliced mRNAs into mature mRNPs (Le Hir et al., 2000). The

complex is composed of 4 core proteins, Magoh, elFA3, Y14, MLN51, and other auxiliary

factors (Wang et al., 2014; Tange et al., 2005). The EJC couples several RNA processing events

such as mRNA export, subcellular localization, translation, and nonsense-mediated decay by

traveling with the mRNA to the cytoplasm (Le Hir et al., 2001; Woodward, 2017).

While the structure of the exon junction complex (EJC) has been elucidated, studies

involving its assembly pathway and transcriptome regulation remain unknown. Recent studies on

the EJC interactome have revealed its location on spliced RNAs and association with other

mRNPs (Figure 3; Singh et al., 2012). The EJC interacts with SR and SR-like proteins to

regulate mRNA packaging, mRNA metabolism, and other events of gene expression (Figure 3).

In addition, Btf and TRAP150 show a more precise co-localization with EJC proteins than with

transcription or splicing factors (Varia et al., 2013).

Alternative splicing adds another layer of splicing regulation as SR/SR related proteins

promote mRNA transcript variation through different mechanisms. These mechanisms include

alternative promoter coupling to downstream exons, exon inclusion/exclusion, intron retention,

and/or alternative polyadenylation events (Zhou, 2013). The production mRNA isoforms depend

on several kinase signaling pathways that either directly or indirectly impact components of the

splicing machinery. Kinases such as SR protein kinases (SRPKs) and cdc2-like kinase

(CLK/STY) phosphorylate the RS domains of SR proteins, thereby regulating their assembly

10 into splicing complexes and subnuclear localization in nuclear speckles (Manley, 1996; Sacco-

Bubulya et al., 2002).

Figure 3: Regulation of post-transcriptional gene expression and protein-protein interaction by the EJC

Biochemical analysis of the cellular EJC interactome reveals a higher order mRNP structure. The diagram below depicts a new view of endogenous EJC proteins multimerizing with each other as well as with SR proteins. Solid black line: exonic RNA; Dashed black line: a generic intron; Color ovals: proteins enriched more > 10-fold in the EJC proteome listed in descending order of stoichiometry. Black ovals: undetected proteins known to bind to mRNA ends; Green spheres: bridging protein-protein interactions (Singh et al., 2012; sited by Cheedu et al., 2016).

(Singh et al., 2012)

1.2 Nuclear speckles

The mammalian cell nucleus contains several highly organized nuclear and subnuclear

compartments that have diverse functions in gene expression (Figure 4; Spector, 2001). Nuclear factors localize and shuttle between domains such as the nuclear speckles, paraspeckles, nucleoli,

Cajal bodies, promyelocytic leukemia bodies, and more (Figure 4; Spector, 2001; Lamond et al.

1998). SR and SR-like proteins occupy non-membrane bound nuclear speckles, also known as interchromatin granule clusters (IGGs). These nuclear domains act as storage and modification sites for pre-mRNA processing factors (Mintz, 2000; Spector 2011).

Nuclear speckles lack DNA and are largely devoid of transcriptional activity (Thiry,

1995; Cmarko et al., 1999). However, they contain polyadenylated RNA and pre-mRNA splicing factors that continuously cycle between transcription sites and the nuclear speckles (Politiz,

2006). This exchange is driven by a reversible phosphorylation mechanism that regulates the localization and stability of nuclear speckle proteins (Politiz, 2006; Sacco-Bubulya et al., 2002).

The nuclear speckle proteome is composed of approximately 180 identified proteins and 33 uncharacterized proteins (Saitoh, 2004). Nuclear speckle proteins help coordinate gene expression regulatory steps as well as outside biological pathways (Figure 5; Galganksi et al.,

2017). Therefore, understanding how nuclear speckle proteins interact with other components will expand our knowledge on the organization and function of SR-like splicing factors.

Figure 4: A schematic diagram of nuclear domains in the mammalian cell nucleus

The schematic model below shows a mammalian cell nucleus surrounded by diverse non- membrane bound domains. Some of these nuclear and subnuclear compartments include the nuclear speckles, nucleoli, Cleavage body, Cajal body, Gem, Polycomb Group (PcG) body, and promyelocytic leukemia (PML) body. These domains were characterized based on morphology, immunolocalization, structure, and function (Spector, 2001; Frege, 2015).

(Frege, 2015)

Figure 5: Diversity of gene expression from the regulation of nuclear speckle proteins

The following diagram depicts nuclear speckle proteins regulating various stages of gene expression through direct and indirect interactions. Nuclear speckle proteins can act as endogenous binding partners that function in pre-mRNA processing as well as other biological pathways. The depletion of nuclear speckle proteins can often result in human diseases including cancer and neurodegenerative syndrome (Galganski et al., 2017).

(Galganski et al., 2017)

1.3 SR and SR-related proteins

The SR family is a group of highly conserved RNA binding proteins with long

Serine/Arginine repeats. Classical SR proteins are characterized by having a C-terminal arginine-

serine rich RS domain and one or two N-terminal RNA recognition motifs (RRMs) (Long, 2009;

Shepard, 2009). The RS domain regulates subcellular localization as well as protein-protein interactions among SR proteins and other transcription machinery (Cazalla et al., 2002; Zhu et al., 2000). In contrast, the RRM is an essential RNA-binding domain that recruits splicing factors responsible for 5’ prime splice site recognition and spliceosome assembly (Maris, 2005).

Unlike the SR family, SR-related proteins have an RS domain located at their N-terminal region and lack distinct RNA binding motifs (Long, 2009). Previous studies confirm that SR-like

splicing factors have overlapping functions in mitotic chromosome alignment and DNA damage

repair, but distinct roles in subcellular localization of mRNA (Cheedu et al., 2016; Varia et al.,

2017; Vohhodina et al., 2017; Varia et al., 2013). Overall, SR and SR-related proteins help maintain overall gene expression in all metazoan species through their overlapping but distinct activities.

1.4 Btf and TRAP150

The SR-like factors, BCL2 Associated Factor 1 (BCLAF1, also called Btf) and Thyroid

Hormone Receptor Associated Protein 3 (THRAP3, also called TRAP150) are unique homologous transcription proteins that were identified in the nuclear speckle proteome. Btf primarily functions as a protein partner of the adenoviral bcl-2 homolog E1B 19K to promote

apoptosis and suppress transcription (Kasof et al., 1999). Previous studies analyzing postnatal

growth between Btf-deficient mice and wild-type mice also confirms the role of Btf in lung development, muscle development, and T-cell activation (Sarras, 2010). In contrast, TRAP150 was originally identified in the TRAP complex, a series of nuclear receptors that initiates transcription in a ligand-dependent manner (Ito et al., 1999). TRAP150 activates nuclear mRNA degradation in a domain separate from pre-mRNA splicing by interacting with the mRNA export factor TAP (Lee et al. 2010). TRAP150 also regulates biological processes such as signal- induced alternative splicing and DNA damage repair (Bell et al., 2012).

Btf and TRAP150 have distinct functions; however, both associate with components of the EJC to promote mRNA packaging into mRNPS (Varia et al., 2013). Nuclear speckle analysis confirms that Btf and TRAP150 share approximately 39% sequence identity and 66% similarity

(Figure 6; Potabathula et al., 2009; Cheedu et al., 2016). The SR-like factors also compensate for each other upon depletion in processes such as nuclear/cytoplasmic mRNA distribution and

DNA damage repair (Varia et al., 2013; Vohhodina et al., 2017). The overlapping function of these proteins in pre-mRNA splicing is clear; however, the extent of regulation in other cellular pathways remains uncertain. The main goal of my thesis is to identify all endogenous binding partners of Btf and TRAP150 to further investigate their subcellular localization and function.

This study may reveal binding partners in distinct complexes relevant to pre-mRNA splicing or pathways outside of pre-mRNA processing.

Figure 6: Alignment data for Btf and TRAP150 showing sequence similarity (also performed by Potabathula, D., 2009; Cheedu, D., 2016)

The figure below displays an updated BLAST search performed using a SIM tool on the

ExPasy Bioinformatics Resource Portal. A graphing program called LAWNVIEW was used to align the full-length amino acid sequences between Btf and TRAP150. Results confirm that Btf and TRAP150 have 39.5% identity in 945 overlapping residues. A colorimetric scale reveals several regions of similarity between their protein sequences.

(Alignment data and figure display were generated at www.expasy.org; also performed by Potabathula, D., 2009; Cheedu, D., 2016)

1.5 Subcellular mRNA distribution

Past research in the Bubulya lab supports that Btf and TRAP150 have overlapping

functions that are not completely redundant. Studies with U2OS 2-6-3 and rat BTM minigene

HeLa cell lines showed the degree of nuclear/cytoplasmic colocalization between the SR-like

proteins and several reporter proteins/RNAs, transcription factors, and RNA processing

machinery (Varia et al., 2013). The goal of this study was to highlight the distinct functions of

Btf and TRAP150 outside of pre-mRNA splicing such as nuclear export and mRNA retention.

Results confirmed that Btf/TRAP150 localize with the U1-70K splicing factor at small nuclear foci which is dependent on RNA polymerase II activity (Varia et al., 2013). Next, RNA-FISH was used to detect BTM minigene transcripts and confirm localization with Btf, TRAP150, and

SRSF1. The reporter transcripts co-localized with Btf and TRAP150 but not with the rtTA transcription activator that binds to DNA regulatory elements in the reporter gene promoter region (Varia et al., 2013).

Btf and TRAP150 showed precise overlap with the core EJC component Magoh in the active loci (Varia et al., 2013). Since EJC proteins are known to help promote mRNA assembly/ nuclear export, further studies explored the effects of siRNA knockdown on subcellular mRNA distribution of BTM transcripts and polyadenylated RNA (Le Hir et al., 2001). Quantitative RT-

PCR analysis showed that Btf depletion caused an increase in BTM transcripts as well as polyadenylated RNA in the cytoplasm (Varia et al., 2013). In contrast, TRAP150 knockdown increased BTM transcript levels in the nucleus and did not affect global mRNA distribution

(Varia et al., 2013). Based on these results, one possible hypothesis is that Btf and TRAP150 both participate in mRNA processing but have opposite effects on mRNA distribution. Another hypothesis is that Btf/TRAP150 co-localization with EJC factors is required for functions such as

mRNA export and/or stability. Depletion studies suggest that TRAP150 promotes nuclear export,

Btf promotes nuclear retention of mRNAs, and perhaps mRNP maturation results in the

formation of mutually exclusive complexes that ultimately regulate mRNA distribution.

Therefore, my follow-up study is designed to learn what unique nuclear complexes contain Btf versus TRAP150.

1.6 Regulation of DNA Damage Repair transcripts

Btf and TRAP150 indirectly regulate DNA damage response (DDR), cell-cycle arrest,

cell resistance, and more through a network of response proteins/signals (Ciccia, 2010). DDR is

characterized by two individual pathways that do not share overlapping targets; however, work

together to promote overall genomic stability. The ATM-Chk2 pathway primarily responds to

DNA double stranded breaks (DSB) whereas ATR-Chk1 improves DNA lesions and replication

stress (Matsuoka et al., 1998; Liu et al., 2000). Recent studies analyzed the distinct functions of

Btf and TRAP150 with DDR components such as BRCA1. To determine their regulation of cell

survival, U2OS and 293T cells were knocked down with Btf/TRAP150 siRNA then induced with

DNA damage treatments (Vohhodina et al., 2017). Ionizing radiation (IR) promotes double-

stranded DNA breaks whereas hydroxyurea (HU) stalls the replication forks during S phase

(Vohhodina et al., 2017).

Results showed TRAP150 depletion as opposed to Btf causes significant downregulation

of surviving cells following both treatments. FLAG-tagged TRAP150 constructs also confirmed

that deletion of the N-terminus containing the RNA binding RS domain and/or C-terminus containing no domains leads to only partial restoration of IR-induced damage cells (Vohhodina

et al., 2017). Therefore, both termini are necessary for efficient DNA damage restoration and cell

survival. Since Btf and TRAP150 compensate for each other in mRNA expression further studies

investigated the effects of co-depletion on cell sensitivity and DNA repair kinetics (Varia et al.,

2013; Vohhodina et al., 2017). Btf and TRAP150 co-depletion caused severe sensitivity to the

IR/HU treatments due to the loss of phosphorylation of 53BP1 and KAP1, key mediator proteins

in the DDR pathways (Vohhodina et al., 2017). These proteins as well as upstream ATM/ATR

kinases remained the same during individual Btf/TRAP150 depletion which supports the

mechanism of compensation between SR-like proteins (Varia et al., 2013; Vohhodina et al.,

2017).

Last, Btf and TRAP150 knockdown caused several defects in the splicing, processing,

and nuclear export of DDR transcripts. Results from qRT-PCR analysis showed that co/individual depletion decreases the total levels of fully spliced ATM transcripts in the nucleus.

This indicates that both Btf and TRAP150 function in ATM mRNA maturation. In contrast to the mRNA distribution study, TRAP150 depletion alone promotes a significant increase in nuclear retention of poly(A)+ RNA pool rather than Btf (Varia et al., 2013, Vohhodina et al., 2017). Co-

depletion led to an accumulation of ATM mRNA as well as unspliced DDR transcripts in the nucleus. These deficient transcripts include BRCA2, FANCL, FANCD2, and RAD51

(Vohhodina et al., 2017). These studies suggest that TRAP150 possesses a predominant function

in DNA damage repair through ATM mRNA nuclear export.

The goal of my project was to uncover all common and distinct endogenous binding

partners of Btf/TRAP150. The common binding partners are considered proteins with redundant

functions between Btf and TRAP150 such as roles in maintaining pre-mRNA splicing. Common

protein partners may also regulate biological pathways outside of mRNA processing along with

Btf/TRAP150. In contrast, distinct binding partners will likely have independent functions with either Btf or TRAP150 in a specific cellular pathway. These binding partners will reveal additional functions that distinguish the two SR-related proteins. The second goal was to use statistical analysis to categorize the mass spectrometry data and eliminate insignificant hits from the lists. I created an efficient system for selecting putative unique binding partners supported by the MS-based proteomics data.

CHAPTER 2: HYPOTHESES AND EXPERIMENTAL AIMS

2.1 Specific Aims

The purpose of this study is to identify the interacting binding partners of Btf/TRAP150

and assign their biological function. Some biological processes previously associated with Btf

and TRAP150 include nuclear export, mRNA trafficking, and mitotic alignment. I hypothesize

that isolating Btf and TRAP150 complexes will help elucidate their roles in these cellular

pathways as well as others. For example, if a binding partner for Btf promotes chromatin

assembly, we could explore if Btf knockdown alters the subcellular localization and expression of histones or chromatin assembly factors.

Studies in the Bubulya lab have already investigated the effects of Btf and TRAP150 knockdown on subcellular mRNA localization and cell cycle progression (Varia et al., 2013;

Varia et al., 2017). HumanExon10ST Array analysis revealed that Btf/TRAP150 depletion significantly alters gene expression indirectly through the regulation of mitotic checkpoint transcripts (Varia et al., 2017). This result validated six major regulator protein transcripts that

maintain upregulated abundance at unattached kinetochores (Aurora-B, CENP-F, BUB1, CDK1,

CDC20, and MAD2L1) (Varia et al., 2017). Since both Btf and TRAP150 depletion cause chromosome misalignment, Aurora-B was a transcript of high interest due to its role in kinetochore-microtubule attachment and chromosome segregation.

Studies concluded that the abundance of the Aurora B checkpoint transcript was altered following knockdown of Btf and TRAP150 (Varia et al., 2017). There is little known about the interacting protein partners of Btf and TRAP150. Therefore, my study is the first to directly compare Btf and TRAP150 binding partners to understand common versus distinct pre-mRNA

processing roles as well as other functions of these two SR-related proteins. Once the binding

partners are established, further research in cellular pathways can be performed to conclude

mechanistic roles.

The following specific aims were used to achieve my objectives:

Aim 1: Identify endogenous binding partners of Btf and TRAP150

a. Isolate Btf and TRAP150 complexes

b. Visualize targets along with their endogenous binding partners

Aim 2: Perform MS analysis and evaluate proteomics data

a. Complete mass spectrometry analysis of isolated complexes

b. Combine the mass spectrometry data to select common/distinct binding partners using

statistical analysis

c. Build a model for novel functional Btf and TRAP150 complexes

In Aim 1, the interacting partners of Btf and TRAP150 were identified to help us

understand the overlapping versus distinct functions of Btf and TRAP150. Btf and TRAP150

complexes were isolated from whole cell HeLa extract by immunoprecipitation (IP) with their

respective antibodies (WU10 and TRAP956). The immunoprecipitated complexes were resolved

in three biological replicates using SDS-PAGE and immunoblot assays. Silver staining analysis was important in visualizing shared or unique protein partners according to their molecular

weight on a similar 4 -20% acrylamide gel. The three IP replicates from separate extractions

were subsequently analyzed by mass spectrometry. The ‘extract + antibody’ and ‘no extract +

antibody’ fractions were reserved at 4°C for future comparisons.

To complete Aim 2a, replicate IP samples were sent for LC-MS/MS analysis at the

Campus Chemical Instrument Center (CCIC) Mass Spectrometry and Proteomics Facility at The

Ohio State University. Upon arrival, the samples were digested with trypsin and stored in acetic

acid at -80°C. The samples were fragmented in silico which measures peaks of charged

fragments within the peptide sequences. The sequences were compared to known/unknown

binding partners of Btf/TRAP150 and quantified using a proteomics database called MASCOT.

Next, a scaffold program was used to characterize the search results between common/distinct

proteins.

For Aim 2b, the endogenous binding partners were listed on an excel sheet to compare

the common versus distinct protein partners of Btf and TRAP150 as well as to identify their

known biological roles. In order to select common/distinct protein partners for future

biochemical and/or mechanistic studies, we created an effective system of categorization for the results. The three IP replicates for Btf and TRAP150 were combined into a new list using R program analysis. This list excludes protein partners that were not identified in all IP experiments as well as proteins that bind to Protein A agarose beads with only WU10 or

TRAP956 antibodies. The new binding partner lists were defined by the following categories: accession, family, member, score, mass, # of significant matches, # of significant sequences, and overall emPAI. FXR1P/FXR2P were the novel binding partners selected based on linear regression analysis and unique function.

For Aim 2c, I used information in the literature to understand the interactions of novel binding partners with Btf and/or TRAP150. For example, I sought to determine if Btf and/or

TRAP150 are localized in the same subcellular compartment as their respective binding partners,

FXR1P/FXR2P. I originally planned to perform reciprocal IPs in order to pull down the binding partners and confirm protein-protein interaction with Btf/TRAP150 antibodies. We also planned to knockdown Btf and TRAP150 with siRNA duplexes to identify binding partner mis- localization; however, university closure due to the COVID-19 pandemic interrupted these last two sets of experiments.

CHAPTER 3: MATERIALS AND METHODS

3.1 Thawing/ Freezing Cells

HeLa cells were stored cryogenically and suspended in freezing medium (95% fetal bovine serum (FBS), 5% dimethyl sulfoxide (DMSO; Cat. No D2650, Sigma, MO)). For extract preparation, HeLa cells were rapidly thawed in a 42 °C water bath for 1 minute then immediately transferred into a 15 ml conical tube along with 5 ml of Dulbecco’s Modified Eagle Medium

(DMEM) supplemented with 10% FBS and 1% penicillin/streptomycin. This step was performed quickly using sterile equipment to ensure cell survival. Next, the conical tube was centrifuged at

2,000 rpm for 2 minutes and the supernatant was removed by aspiration. Cells were resuspended in 10 ml of DMEM + 10% FBS + 1% pen/strep then plated into new 100 X 20 mm culture dishes. The plates were incubated in a 5% CO2 humidified chamber at 37°C for a 48-hour recovery period before their first passage.

3.2 Cell Culture

HeLa cells were cultured in DMEM + 10% FBS + 1% pen/strep until reaching 80-90% confluency. The normal doubling time for HeLa cells was 24 hours. The cells were passaged by washing the plates three times with 1X phosphate buffered saline (PBS; 137mM NaCl; 2.7mM

KCl; 4.3mM Na2HPO4; 1.47mM KH2PO4, pH 7.4). Next, the plates were incubated in 2 ml of trypsin-EDTA (Invitrogen, 0.25% trypsin) for approximately 2 minutes at 37°C. The enzymatic reaction was deactivated with 5 ml of DMEM and the cells were transferred to a 15 ml conical tube. The cells were centrifuged at 2000 rpm for 2 minutes. The supernatant was removed by aspiration and the pellet was resuspended in 10 ml of DMEM. HeLa cells were grown in 100 mm

X 20 mm culture dishes until confluency was reached for either whole cell nuclear extraction

(WCNE) or immunofluorescence (IF) experiments.

3.3 Plating Cells

Upon reaching 50-60% confluency, cells were prepared for an immunofluorescence experiment the following day. Cells were grown in small clusters as opposed to large masses for efficient microscope imaging. Glass coverslips were treated with a gelatin-coating solution

(0.1% gelatin in dH2O), washed with 70% ethanol, then gently passed through a flame for sterilization. The coverslips were placed in 6-well 35mm dishes for immunofluorescence. Lastly,

2 ml of HeLa cells from 50-60% confluent plates were trypsinized and suspended in DMEM +

10% FBS + 1% as described above. They were added to each well and incubated in a 5% CO2

humidified chamber at 37°C for a 24-hour recovery period before processing for

immunofluorescence.

3.4 Whole Cell Nuclear Extract (WCNE) preparation

Each batch of extract was produced from 30 culture dishes (100 X 20 mm) containing

HeLa cells at 90-100% confluency. Cells were washed two times with cold 1X PBS. Next, 1 ml

of cold PBS was added to each dish and cells were scraped with a rubber policeman. Cells were

transferred to 15 ml conical tubes (five plates of cells per tube). The scraping of cells in 1 ml 1X

PBS and transfer to conical tubes was repeated once more to capture all remaining cells in the

plates. Next, the cells were centrifuged at 4°C for 5 min at 2500 rpm. The pellet in each of the

six conical tubes was resuspended in 1 ml 1X PBS and transferred to 15 ml Eppendorf tubes,

then centrifuged at 2000 rpm for 5 minutes at 4°C. Afterwards, the supernatant was aspirated and

1 ml (200μl per plate) of lysis buffer (300mM NaCl, 100mM Tris pH-8, 0.2mM EDTA, 0.1%

NP40, 10% glycerol, protease inhibitor cocktail) was added to the remaining pellet. The tubes

were rotated for 45 minutes at 4°C then centrifuged at 14,000 rpm for 15 minutes at 4°C.

Supernatants were pooled into a single tube and 300μl aliquots were snap frozen in an ethanol/

dry ice solution. Extracts were stored at -80°C and later used for immunoprecipitation and/or

immunoblot analysis.

3.5 Protein Quantification (Bradford Assay)

Bradford assays were used to estimate the protein concentration in each cell extract to

ensure equal amounts of protein per experimental replicate (IPs, western blots). Increments of

bovine serum albumin (stock 10µg/µl) were diluted with dH2O and 2X Laemmli buffer (4%

SDS. 20% glycerol, 1.4M ꞵ-mercaptoethanol, 0.125 M Tris-HCl) to create standards of known protein concentrations (0, 1, 2, 5, 10, and 20 micrograms). Three replicate samples containing

2µl of protein extract were mixed with 498µl of dH2O. Next, 500µl of Coomassie Plus Protein

Assay Reagent (Thermo Scientific) was added to each sample for a total volume of 1 ml. The

samples were transferred to 1 ml cuvettes and measured for absorbance at 595nm using a

spectrophotometer. A graph of optical density versus µg protein displayed a best-fit line and was used to calculate protein concentrations of the unknown samples.

3.6 SDS-PAGE and Immunoblotting

HeLa WCNE was thawed on ice and diluted with additional WCNE buffers to equal 1 ml total volume having 100mM final NaCl concentration. This was performed by adding 600µl of

WCNE buffer containing 0mM NaCl plus 100µl of WCNE buffer containing 100mM NaCl.

Next, 50µl of 2X gel sample buffer (124 mM Tris HCl pH 6.8, 20% glycerol, 10% β- mercaptoethanol, 4% SDS, 0.05% bromophenol blue) were added to each reaction. The samples were boiled for 5 minutes at 110°C in a dry bath before being loaded into a 7% SDS polyacrylamide gel. A precision plus protein ladder (BioRad Catalog No. 161-0374) was also loaded to serve as a molecular mass marker. Each gel ran at 110V for approximately one hour in

1X running buffer (50 mM Tris-HCl, 192 mM glycine, 2% SDS). After gel electrophoresis, the proteins were transferred from the polyacrylamide running gel to a PVDF membrane. Protein transfer was performed at 200mA for 2 hours using a cold transfer buffer (25mM Tris HCl,

192mM glycine, 20% methanol, pH 8.3). Ponceau staining solution (0.5% Ponceau S, 1% acetic acid) was added to the PVDF membrane to confirm the protein lanes and mass markers. The membrane was blocked against non-specific bonding using 1X PBST (0.22M sodium phosphate monobasic, 1.2M sodium phosphate dibasic anhydrous, 137mM sodium chloride, and 1.0%

Tween) and 5% non-fat dry milk on a rotary shaker. The PVDF membranes were probed using the rabbit polyclonal antibodies TRAP956 and TRAP955 (for detecting TRAP150, diluted

1:1000; Bethyl Laboratories, Catalog numbers A300-956A and A300-955A) or WU10 (for detecting Btf, diluted 1:1000; Varia et al., 2013). Mouse anti-actin (1:1000) or rabbit anti-ꞵ- tubulin (1:1000) was applied to the lower portion of the membrane (<100 kDa) to confirm equal loading of protein in each lane. Primary antibodies were diluted in 1X PBST/5% nonfat dry milk and incubated for 1 hour. The membranes were washed 3 X 5 minutes with 1X PBST with agitation. Next, the secondary antibody, Goat anti-Rabbit IgG-HRP (1:25,000), was added to the

membrane along with 1X PBST/5% nonfat dry milk and incubated for 1 hour with agitation. The

membranes were washed 3 X 10 minutes with 1X PBST with agitation. The blots were

submersed in Thermo-Scientific Pierce ECL Western Blot reagents and imaged using a

chemiluminescence technique on a FUJI LAS-4000 Luminescent Image Analyzer (Fujifilm Life

Science USA, Stamford, CT).

3.7 Immunoprecipitation

Prior to immunoprecipitation, Protein A coated agarose beads were washed three times

and equilibrated overnight in 100mM NaCl WCNE buffer. Next, 10µl of equilibrated bead

volume was added to 1 ml of 100mM NaCl HeLa extract and rotated for 3 hours at 4oC. This

step pre-clears the HeLa extract of all ‘sticky’ proteins that non-specifically bind to the Protein A beads. The extract was centrifuged at 2000 rpm at 4oC and the supernatant was immediately used

for IP reactions. The remaining pellet was suspended in 100µl of 2X Laemmli buffer and

reserved for a ‘pre-cleared beads’ control during SDS-PAGE analysis. A 100µl aliquot of pre-

cleared extract was reserved for an ‘input’ control to determine levels of target proteins in the

WCNE. The remaining cell extract was divided into the following immunoprecipitation

reactions:

1.) extract + antibody: 300µl HeLa extract (100mM NaCl) + 5µl antibody

2.) antibody only: 300µl of 100mM NaCl extraction buffer + 5µl antibody

3.) extract only: 300µl HeLa extract (100mM NaCl)

The reactions were rotated for 90 minutes at 4oC then incubated overnight with 10µl of

protein A bead volume. The following day, the samples were centrifuged for 5 minutes at 4oC and 100µl of supernatant was removed from each tube to represent the unbound fractions. These samples were set aside in 100µl of 2X Laemmli buffer for subsequent SDS-PAGE analysis. All bead samples were washed 5 times with 300mM NaCl WCNE buffer. High salt concentration disrupts non-specific interactions between proteins and beads as well as antibodies. One final wash with 100mM NaCl WCNE buffer was performed to lower the salt concentration to a compatible state for SDS-PAGE. Afterwards, the bound fractions from the IP samples (the bead

pellets) were either suspended in 100µl of 2X Laemmli buffer for SDS-PAGE analysis or in cold

1X PBS for mass spectrometry analysis. Each replicate of immunoprecipitations was performed

twice for LC-MS/MS analysis in order to assess quality control prior to submitting samples.

3.8 Silver Staining of SDS-PAGE gels

Following electrophoresis, the gel was washed 2 X 5 minutes in dH2O, then incubated in

a fixative solution (60% dH2O, 30% ethanol, 10% acetic acid) for 2 X 15-minute intervals. The

gel was washed 2 X 5 minutes in a 10% ethanol solution then 2 X 5 minutes in dH2O. Next, the

gel was incubated in a sensitizer working solution for exactly 1 minute then rinsed twice with

dH2O. The gel was incubated in a Silver Stain Enhancer solution for 5 minutes then rinsed twice with dH2O. A final developer working solution was added to the gel for 2-3 minutes or until protein bands appeared. The staining period was stopped by adding 5% acetic acid solution directly to the gel. Prior to imaging, the gel was rinsed twice with dH2O then incubated under a

fume hood to remove extra liquid and allow for easier handling. The lanes on the gel were

analyzed for protein band position as well as putative unique versus common binding partners

for Btf/TRAP150.

3.9 Mass Spectrometry and Proteomics

The cell extracts were analyzed at the CCIC Mass Spectrometry and Proteomics Facility

at The Ohio State University. The IP samples were immediately digested with trypsin, dried, and

resuspended with 20µl of 50mM acetic acid. Next, 5µl of the sample was removed and injected

into a Thermo Fusion instrument for LC-MS/MS analysis. Data from the samples were searched against a UniProt human database called MASCOT to identify the interacting partners of Btf and

TRAP150. The results were summarized in a scaffold program for easy comparison then exported to an excel spreadsheet for further statistical analysis.

3.10 Immunofluorescence

HeLa cells seeded onto coverslips as described above were grown to 50-60% confluency in 6-well 35mm dishes containing 2 ml of DMEM + 10% FBS + 1% pen/strep. To process for immunofluorescence, media was aspirated and the coverslips with attached cells were washed twice with 1X PBS. The cells were fixed in 2% formaldehyde solution for 15 minutes at room temperature. Following three washes with 1X PBS, the cells were permeabilized with 0.2%

Triton X-100 in 1X PBS for 5 minutes at room temperature. Next, the cells were washed three times with 1X PBS containing 0.5% normal goat serum (PBS/NGS). The excess 1X PBS was removed from each coverslip with Whatman filter paper and the cells were incubated for 1 hour in 40μl of appropriate primary antibodies, anti-Btf (WU10; 1:2500) or anti-TRAP150 (1:1000,

Bethyl catalog number A300-956A), diluted in 1X PBS/0.5% NGS. The coverslips were

returned to their appropriate labeled wells then washed three times with PBS/0.5% NGS. The

cells were incubated in 40μl of secondary antibody solution per coverslip containing anti-rabbit

IgGs conjugated with Texas Red or FITC (1:500; Jackson Immunoresearch Laboratory). The plates were also counterstained with DAPI solution (10μg/ml) to label DNA regions. Last, coverslips were mounted onto glass slides using mounting medium (antifade poly- phenylenediamine) and sealed with clear nail polish. Samples were imaged using an inverted fluorescence microscope (Olympus IX83; n=1.3; 100x objective lens) and a low auto- fluorescence immersion oil microscope (Olympus Type F; n=1.518; Fisher Sci Cat

#NC0297589) for protein localization.

CHAPTER 4: RESULTS OF IMMUNOPRECIPITATION AND MASS

SPECTROMETRY

4.1 Isolation of Btf and TRAP150 complexes by Immunoprecipitation (IP)

Btf and TRAP150 complexes were isolated from HeLa whole cell nuclear extract. Three replicate IP experiments were created using a different extract preparation to study parallel measurements with random biological variation. Each biological replicate included 3 separate sets of IPs to reduce human error and perform follow up studies. The first samples were sent for mass spectrometry, the second samples were used for immunoblot/silver stain analysis, and the third set was reserved for future studies after obtaining MS results. Immunoprecipitation is a small-scale affinity purification of proteins using a specific antibody immobilized to magnetic particles or bead resin. In this experiment, we used Protein A Agarose beads which bind to rabbit

IgGs along with their protein targets. As the proteins of interest are isolated, other interacting partners may bind and can subsequently be identified in the purified fractions.

WU10 and TRAP956 polyclonal antibodies were added to the cell extract to immunoprecipitate Btf and TRAP150, respectively. Immunoprecipitates were subjected to 7%

SDS-PAGE followed by immunoblot analysis (see Figures 7, 8, and 9). Several control IP samples were included in the immunoblot analysis to ensure that our proteins of interest were efficiently isolated. A protein marker indicated standard molecular mass positions. We expected

Btf and TRAP150 to migrate at approximately 150kDa. Light/heavy chains of the IgG antibodies added during the IP step were present in bead eluates and observed at lower molecular masses.

Lane 2 included a ‘pre-cleared beads’ control set aside prior to the immunoprecipitation. The absence or low abundance of Btf and TRAP150 in this lane confirms that the target proteins only

bind to Protein A beads via their specific antibodies. Lane 3 contained an ‘input’ control which is

equivalent to 10% of the IP reaction in the entire sample.

Lanes 4, 6, and 8 represent the samples reserved from the bounded fraction of each IP

reaction. These reactions consisted of extract mixed with the specific target antibody (‘extract +

AB’; lane 4), specific target antibody without cell extract (‘no extract + AB’; lane 6), and

reactions without antibody (‘extract only’; lane 8). In contrast, lanes 5, 7, and 9 contain the

respective unbound samples from these IP reactions. Lane 4 of each figure displays the most

significant immunoprecipitate because it confirms Btf/TRAP150 isolated by their antibodies in

the bound fraction. All other bound/unbound samples serve as controls for non-specific binding

of HeLa proteins to agarose beads and for identifying the bands that correspond to IgG.

Immunoprecipitates containing cell extract mixed with specific target antibody and extract with

specific target antibody alone (lanes 4 and 6) from three biological IP replicates were sent for mass spectrometry and proteomics analysis at The Ohio State University.

Figure 7: Immunoblot images for Btf and TRAP150 IPs (Replicate #1)

The following immunoblot figures were from the first biological replicate IP experiment.

Btf/TRAP150 targets were isolated by their respective antibodies (WU10, lane 4 in Figure 7A;

anti-TRAP956, lane 4 in Figure 7B) whereas control IPs confirm no isolation with specific

antibodies alone (lane 6 in Figure 7A and 7B). Samples from both lanes were sent for further analysis. All lanes support the expectations for each control sample.

A. Btf

1 2 3 4 5 6 7 8 9

B. TRAP150

1 2 3 4 5 6 7 8 9

Figure 8: Immunoblot images for Btf and TRAP150 IPs (Replicate #2)

The following immunoblot figures were from the second biological replicate IP experiment. Btf/TRAP150 targets were isolated by their respective antibodies (WU10, lane 4 in

Figure 8A; anti-TRAP956, lane 4 in Figure 8B) whereas control IPs confirm no isolation with specific antibodies alone (lane 6 in Figure 8A and 8B). Samples from both lanes were sent for further analysis. All lanes except for lane 8 support expectations for the control samples (Figures

8A and 8B).

A. Btf

1 2 3 4 5 6 7 8 9

B. TRAP150

1 2 3 4 5 6 7 8 9

Figure 9: Immunoblot images for Btf and TRAP150 IPs (Replicate #3)

The following immunoblot figures were from the third biological replicate IP experiment.

Btf/TRAP150 targets were isolated by their respective antibodies (WU10, lane 4 in Figure 9A; anti-TRAP956, lane 4 in Figure 9B) whereas control IPs confirm no isolation with specific antibodies alone (lane 6 in Figure 9A and 9B). Samples from both lanes were sent for further analysis. All lanes support the expectations for each control sample.

A. Btf

1 2 3 4 5 6 7 8 9

B. TRAP150

1 2 3 4 5 6 7 8 9

4.2 Using silver staining to detect endogenous binding partners

Silver staining is a highly sensitive technique used to detect small amounts of proteins

and nucleic acids in an acrylamide gel. This method is useful for visualizing common/distinct

binding partners of Btf and TRAP150. It also allows us to confirm the presence of the WU10 and

TRAP956 antibodies as well as their targets. Similar to the western blot analysis,

immunoprecipitates were subjected to 4-20% SDS-PAGE. Next, the gels were fixed and developed until protein bands appeared for each bound fraction. In this experiment, the proteins bands corresponding to binding partners appeared faintly in all three replicate silver stain experiments (Figures 10, 11, and 12; extract + antibody). Immunoprecipitates without extract or antibody represent controls to visualize non-specific binding. Silver staining assays can be used to sequence individual bands of binding partners rather than performing mass spectrometry on the entire IP mixture. However, this technique was not used due to the LC-MS/MS results being conclusive from the on-bead digestion.

Figure 10: Silver stain image for Btf/TRAP150 (Replicate #1)

Image from silver stain assay performed on a 7% acrylamide gel after the first IP replicate experiment. Extract + antibody lanes were expected to detect protein bands for

Btf/TRAP150 targets as well as binding partners. Protein bands faintly appeared at larger molecular weights (150 kDa >) for Btf as opposed to TRAP150 (extract + antibody (Btf), lane 4 in Figure 10). Arrows indicate the molecular masses for the rabbit IgGs, WU10 and TRAP956.

Btf and TRAP150

1 2 3 4 5 6 7 8 9 10

Figure 11: Silver stain images for Btf/TRAP150 (Replicate #2)

Images from silver stain assay performed on a 4-20% acrylamide gel after the second IP replicate experiment. Extract + antibody lanes were expected to detect protein bands for

Btf/TRAP150 targets as well as binding partners. Arrows indicate the molecular masses for the rabbit IgGs, WU10 and TRAP956.

A. Btf

1 2 3 4 5 6

B. TRAP150

1 2 3 4 5 6

Figure 12: Silver stain images for Btf/TRAP150 (Replicate #3)

Images from silver stain assay performed on a 4-20% acrylamide gel after the third IP replicate experiment. Extract + antibody lanes were expected to detect protein bands for

Btf/TRAP150 targets as well as binding partners. Protein bands faintly appeared for Btf as opposed to TRAP150 (extract + antibody (Btf), lane 3 in Figure 12A). Arrows indicate the molecular masses for the rabbit IgGs, WU10 and TRAP956.

A. Btf 1 2 3 4 5

B. TRAP150 1 2 3 4 5

4.3 Mass spectrometry results and data analysis

Tandem mass spectrometry (MS/MS) is an analytical technique that breaks down ions

into fragments to measure their mass-to-charge ratio (Aebersold, 2003). The ions are characterized by chemical structure and quantified to reveal the proteins in an extract sample.

The IP samples containing ‘cell extract mixed with specific antibody’ and ‘extract with antibody alone’ were analyzed at the Campus Chemical Instrument Center (CCIC) Mass Spectrometry and

Proteomics Facility at The Ohio State University. Prior to arrival, replicate IP samples were

transferred in cold sterilized containers to reduce protein contaminants such as keratin and

polyethylene glycol (PEG) as well as promote accurate MS/MS sequencing. Next, the samples

were digested and fragmented in silico to produce protein patterns on a mass spectrum graph.

The patterns of binding partners were characterized by unique peaks corresponding to their

mass-to-charge ratio and searched against other peaks in the MASCOT Uniport human database.

MASCOT searching requires the following criteria to be met. First, the database must

contain the protein(s) of interest. The database must also account for cysteine modifications

within the peptide sequence of the protein(s). The trypsin digestion must be efficient without

non-specific cleavage. Lastly, the database must include the appropriate mass accuracy

parameters. The results of the binding partner search were combined using a Scaffold program

and listed on excel spreadsheets corresponding to each biological IP replicate. The binding

partners were organized by the following categories: accession, family, member, score, mass, #

of significant matches, # of significant sequences, and overall emPAI (see Table 1).

Table 1: Category description for variables in MS-based proteomics analysis

Binding partners identified in MS-based proteomics were categorized by several statistical variables. The following table includes the definition of each variable provided by the

CCIC Mass Spectrometry and Proteomics Facility at The Ohio State University.

Variables Description

Accession Protein name

Family A group of proteins that share a common evolutionary origin, usually reflected by their similar sequence, structure, and/or function.

Member Proteins further characterized by their sequence, structure, function, localization, synthesis, etc.

Score Protein score from MASCOT database search

Mass Protein molecular weight

# of significant matches # of unique spectra matched

# of significant sequences # of unique peptide sequences matched

Exponentially modified Overall quantification of total peptide matches. # of observed peptides protein abundance index divided by the number of observable peptides for each protein. (emPAI) Formula shown below:

Results from MS-based proteomics and the MASCOT database search were formatted

using R program analysis. Btf and TRAP150 binding partners from the three IP replicates were

aligned on an excel spreadsheet which revealed all proteins that were consistently identified in

each experiment. Protein partners that were only observed in one or two IP replicates were excluded from further studies due to their inconclusive statistical data. The remaining proteins

were merged into one list along with their average proteomics values using a query function.

Next, binding partners shared in the ‘no extract + antibody’ samples were also removed to rule

out all proteins that non-specifically bind to the WU10 and TRAP956 antibodies.

The newly generated lists from MS-based proteomics identified 207 endogenous binding

partners for Btf and 262 binding partners for TRAP150 (see Table 2A). As expected, the top

binding partners for Btf and TRAP150 consist of transcription factors, hnRNPs, splicing

machinery, nuclear export subunits, and other proteins that function in mRNA processing (Table

2B and 2C). The MS-based proteomics included 153 common interacting partners along with

their biological description for follow-up investigations of Btf and TRAP150 function (Table 3).

The list includes several unique proteins that participate in mRNA processing as well as other

biological pathways.

Table 2: Quantification of identified endogenous binding partners for Btf and

TRAP150

Results from MS-based proteomics identified 207 binding partners for Btf and 262

binding partners of TRAP150. The top protein partners participate in pre-mRNA splicing, mRNA packaging, nuclear export, and other stages of mRNA processing.

A. Venn diagram showing common/distinct binding partners

Btf TRAP150

54 153 109

B. Top 25 endogenous binding partners for Btf

Significant Significant Accession Family Member Score Mass matches sequences emPAI LMNA_HUMAN 1 1 472.24 74380 22.97333 16.4 1.340933 MYH9_HUMAN 86 1 413.9867 227646 20.52 15.36 1.0232 THOC2_HUMAN 3 1 390.72 184541 19.2 14.37333 1.008267 PNPH_HUMAN 15 1 367.56 32325 18.38667 13.90667 1.051067 TR150_HUMAN 2 1 354.3333 108658 17.78667 13.76 0.971467 BCLF1_HUMAN 2 2 331.1467 106173 16.54667 13.06667 0.931067 NPM_HUMAN 6 1 309.5467 32726 15.2 12.22667 0.894 HS71A_HUMAN 13 3 288.7333 70294 14.44 11.97333 0.756267 DDX17_HUMAN 17 1 284.4667 80906 14.24 11.82667 0.748933 DDX5_HUMAN 17 2 279.8133 69618 14.02667 11.66667 0.741067 GRP75_HUMAN 19 1 273.08 73920 13.77333 11.46667 0.722533 PDIP3_HUMAN 47 1 268.7333 46289 13.68 11.37333 0.752667 THOC1_HUMAN 31 1 269.4 76360 13.76 11.41333 0.747467 PLEC_HUMAN 50 1 266.8267 533462 13.52 11.14667 0.776667 NIPBL_HUMAN 8 1 258.52 317907 12.45333 10.05333 0.982 C1QBP_HUMAN 25 1 246.8133 31742 12 9.706667 0.978933 ECI2_HUMAN 11 1 242.84 43957 11.90667 9.693333 0.942 PABP1_HUMAN 32 1 235.8933 70854 11.62667 9.586667 0.934133 PABP4_HUMAN 32 2 231.9867 71080 11.46667 9.453333 0.9156 ANXA2_HUMAN 21 1 230.08 38808 11.33333 9.346667 0.962933 U5S1_HUMAN 18 1 223.7867 110336 11.16 9.226667 0.9104 U520_HUMAN 10 1 225.4267 246006 11.28 9.226667 0.9048 KI67_HUMAN 58 1 211.2 360698 10.61333 8.586667 0.898533 HNRH1_HUMAN 100 1 216.12 49484 11.10667 9.053333 0.9064 HNRPF_HUMAN 100 2 211.56 45985 10.93333 9 0.899733

C. Top 25 endogenous binding partners for Btf

Significant Significant Accession Family Member Score Mass matches sequences emPAI MYH9_HUMAN 4 1 615.2133 227646 27.64 18 2.28 TR150_HUMAN 1 1 531.7067 108658 24 15.77333 2.258667 HSP7C_HUMAN 18 1 490.96 71082 21.82667 14.6 2.280533 BIP_HUMAN 18 2 480.9733 72402 21.28 14.22667 2.262667 HS71A_HUMAN 18 3 463.4 70294 20.50667 13.8 2.209733 BCLF1_HUMAN 2 1 451.1733 106173 19.84 13.30667 2.178533 NPM_HUMAN 8 1 430.6267 32726 18.77333 12.73333 2.156267 THOC2_HUMAN 5 1 414.76 184541 18.04 12.49333 1.999467 ACTB_HUMAN 30 1 396.48 42052 17.09333 11.73333 2.008 ANXA2_HUMAN 12 1 370.9067 38808 16.25333 11.48 1.924267 DDX5_HUMAN 11 2 399.24 69618 18.38667 13.54667 1.834933 DDX17_HUMAN 11 1 393.3467 80906 18.16 13.2 1.818933 PABP1_HUMAN 15 1 395.24 70854 18.50667 13.56 1.7876 PABP4_HUMAN 15 2 384.56 71080 18.05333 13.18667 1.804 RL7A_HUMAN 43 1 373.96 30148 17.48 12.72 1.7992 GRP75_HUMAN 17 1 365.6933 73920 17.16 12.45333 1.710267 RLA2_HUMAN 44 1 363.7733 11658 17.09333 12.36 1.7196 RL18_HUMAN 85 1 367.48 21735 17.37333 12.57333 1.198 C1QBP_HUMAN 124 1 372.6 31742 17.57333 12.68 1.140533 HNRPK_HUMAN 7 1 374.4 51230 17.70667 12.73333 1.152 THOC6_HUMAN 57 1 363.04 38081 17.09333 12.48 1.120533 PTBP1_HUMAN 21 1 362.9067 57357 17.08 12.37333 1.177067 DDX21_HUMAN 16 1 357.88 87804 17.10667 12.53333 1.150933 RS3A_HUMAN 81 1 340.3867 30154 16.24 11.90667 1.1592 SFPQ_HUMAN 158 1 341.16 76216 16.45333 12.13333 1.109733

Table 3: Common binding partners for Btf and TRAP150 w/ unique biological

function

Among the identified proteins are 153 common binding partners that may have similar

functions in specific biological pathways with Btf and TRAP150. The highlighted rows indicate

Fragile X mental retardation proteins (FMRP) which are good candidates for future studies.

These protein partners have novel roles in the nuclear export, mRNA stability, and translational pathways.

Accession Description sp|Q9Y2W1|TR150_HUMAN TR150_HUMAN.Thyroid.hormone.receptor associated.protein.3 sp|Q9NYF8|BCLF1_HUMAN BCLF1_HUMAN.Bcl-2-associated.transcription.factor.1 sp|Q00839|HNRPU_HUMAN HNRPU_HUMAN.Heterogeneous.nuclear.ribonucleoprotein sp|P62888|RL30_HUMAN RL30_HUMAN.60S.ribosomal.protein.L30 sp|P62241|RS8_HUMAN RS8_HUMAN.40S.ribosomal.protein.S8 sp|P10412|H14_HUMAN H14_HUMAN.Histone.H1.4 sp|P19338|NUCL_HUMAN NUCL_HUMAN.Nucleolin sp|P27348|1433T_HUMAN 1433T_HUMAN.14-3-3.protein.theta sp|P26641|EF1G_HUMAN EF1G_HUMAN.Elongation.factor.1-gamma sp|Q9UKV3|ACINU_HUMAN ACINU_HUMAN.Apoptotic.chromatin.condensation.inducer.in.the.nucleus sp|P51114|FXR1_HUMAN FXR1_HUMAN.Fragile.X.mental.retardation.syndrome related.protein sp|O14950|ML12B_HUMAN ML12B_HUMAN.Myosin.regulatory.light.chain.12B sp|Q96FV9|THOC1_HUMAN THOC1_HUMAN.THO.complex.subunit.1 sp|P26368|U2AF2_HUMAN U2AF2_HUMAN.Splicing.factor.U2AF.65.kDa.subunit sp|O75152|ZC11A_HUMAN ZC11A_HUMAN.Zinc.finger.CCCH.domain-containing.protein sp|O43290|SNUT1_HUMAN SNUT1_HUMAN.U4/U6.U5.tri-snRNP-associated.protein.1 sp|Q7Z417|NUFP2_HUMAN NUFP2_HUMAN.Nuclear.fragile.X.mental.retardation-interacting.protein.2 sp|O00425|IF2B3_HUMAN IF2B3_HUMAN.Insulin-like.growth.factor.2.mRNA-binding.protein.3 sp|Q07955|SRSF1_HUMAN SRSF1_HUMAN.Serine/arginine-rich.splicing.factor.1 sp|Q9BTC0|DIDO1_HUMAN DIDO1_HUMAN.Death-inducer.obliterator.1 sp|P25205|MCM3_HUMAN MCM3_HUMAN.DNA.replication.licensing.factor.MCM3 sp|O14950|ML12B_HUMAN ML12B_HUMAN.Myosin.regulatory.light.chain.12B sp|Q9NR30|DDX21_HUMAN DDX21_HUMAN.Nucleolar.RNA.helicase.2 sp|Q06787|FMR1_HUMAN FMR1_HUMAN.Synaptic.functional.regulator.FMR1 sp|Q07065|CKAP4_HUMAN CKAP4_HUMAN.Cytoskeleton-associated.protein.4

Next, the endogenous binding partners of Btf and TRAP150 were classified by their primary biological pathway and function using a gene ontology (GO) search (Table 4 and 5). GO terms were assigned to each protein partner to create a computational model of biological systems. The GO resource calculates the number of proteins associated with each GO term as well as their fold enrichment and p-values. These statistical categories are more significant depending on the larger data sets of genes associated with binding partners.

GO fold enrichment measures the ratio of the total number of genes annotated to a biological term (background frequency) and the number of genes in the binding partner input list

(sample frequency) (Gene Ontology Consortium). Higher statistical values signify more genes associated with the biological process. The p-value represents the probability of seeing at least x number of genes out of the total n genes in the annotated list of specific GO terms (Gene

Ontology Consortium). The closer the p-value is to zero, the more significant the genes are associated with the GO term. Gene ontology analysis is a useful technique for selecting binding partners that are involved in cellular pathways during future studies.

Table 4: Biological pathways associated with Btf binding partners

Results from gene ontology analysis reveal the biological pathways impacted by

endogenous Btf binding partners (Gene Ontology Consortium). Fold enrichment measures the ratio of total associated genes in the biological process and genes in the protein partner input list.

Whereas the p-value calculates the probability of visualizing a specific number of genes in the total GO term list.

A. Circle graph representation of biological pathways for endogenous Btf

binding partners

Btf binding partners

Nuclear export

mRNA metabolic process

Subcellular localization of proteins Nitrogen compound metabolic process Ribosome biogenesis

Stress response

B. Complete list of biological pathways for binding partners distinct to Btf

Distinct to Btf binding partners Fold enrichment p value Gene expression 125 6.16 8.75E-04 mRNA export from nucleus 110 13.74 1.35E-07 mRNA metabolic process 103 14.91 7.80E-08 Subcellular localization of proteins 89 2.99 4.58E-17 Nitrogen compound metabolic process 63 1.96 1.59E-05 Ribosome biogenesis 53 16.33 1.54E-08 Stress response 50 1.39 7.13E-06 mRNA splicing, via spliceosome 43 14.49 6.51E-04 RNA splicing via transesterification reactions 43 14.34 4.68E-03 Apoptosis 29 1.85 1.51E-02 Chromosome organization 27 2.59 2.60E-05 Cell communication 26 1.35 2.76E-07 Cell cycle progression 22 1.79 2.14E-10 Cell population proliferation 20 1.22 1.51E-18 Protein phosphorylation 16 1.1 1.31E-06 Alternative splicing, via spliceosome 15 21.9 1.68E-13 Cytoplasmic translation 14 19.32 7.06E-06 Spliceosome complex assembly 11 19.79 1.53E-07 DNA damage repair 6 4.65 4.01E-11 Transcription by RNA polymerase II 5 1 5.88E-04 Chaperone-mediated protein folding 4 6.83 7.14E-08 T cell activation 4 1.22 1.76E-12 Heat shock response 3 2.8 1.14E-10 Cardiac muscle cell development 1 12.59 4.57E-04 Chromatin assembly 1 12.59 4.57E-04

Table 5: Biological pathways associated with TRAP150 binding partners

Results from gene ontology analysis reveal the biological pathways impacted by endogenous TRAP150 binding partners (Gene Ontology Consortium). Fold enrichment measures the ratio of total associated genes in the biological process and genes in the protein partner input list. Whereas the p-value calculates the probability of visualizing a specific number of genes in the total GO term list.

A. Circle graph representation of biological pathways for endogenous

TRAP150 binding partners

Subcellular localization of proteins Stress response

Translation

mRNA splicing, via spliceosome RNA splicing, via transesterification reactions Ribosome biogenesis

B. Complete list of biological pathways for binding partners distinct to

TRAP150

Distinct to TRAP150 binding partners Fold enrichment p value Gene expression 158 6.17 9.49E-88 Subcellular localization of proteins 88 4.28 4.22E-32 Stress response 71 1.56 1.06E-04 Translation 58 26.16 7.62E-48 mRNA splicing, via spliceosome 57 15.17 1.06E-46 RNA splicing, via transesterification reactions 57 14.92 1.29E-61 Ribosome biogenesis 47 12.55 4.47E-35 mRNA metabolic process 47 2.14 9.22E-07 Apoptosis 40 1.86 1.60E-04 Cell cycle progression 34 1.99 1.43E-04 rRNA processing 33 12.16 2.32E-24 DNA metabolic process 27 2.87 1.53E-06 mRNA 3'-end processing 21 12.02 8.60E-16 mRNA export from nucleus 21 15.19 1.29E-17 Chromosome organization 19 4.15 3.54E-07 DNA damage repair 17 2.63 3.73E-04 Spliceosome complex assembly 14 19.9 1.51E-13 RNA stabilization 13 19.9 1.14E-12 Rhythmic process 12 3.44 2.90E-04 RNA modification 11 5.4 1.12E-05 DNA duplex unwinding 10 7.44 2.03E-06 Telomere organization 9 7.16 8.80E-06 Mitotic spindle organization 8 4.19 8.91E-04 interleukin-12-mediated signaling pathway 8 13.84 3.21E-07 Alternative mRNA splicing, via spliceosome 7 32.77 1.21E-08

Pearson correlation and linear regression analysis offers an efficient system for selecting newly identified protein partners. Binding partners can be selected at random, however, using statistical measures helps support a strong interaction between proteins prior to future studies.

The Pearson correlation measures the strength of linear relationship between two continuous variables (Benesty, 2009). Correlation coefficients range between -1 and 1. A value of 1 suggests that two variables have a perfect positive association whereas -1 indicates a negative association

(Benesty, 2009). Correlation analysis is significant in studying MS-based proteomics results because it affirms which binding partner categories are relevant to each other. In this study, each statistical category was compared to the emPAI since the value quantifies total peptide abundance.

The ‘score’ group had the highest correlation coefficients with emPAI for Btf and

TRAP150 which signifies a strong positive relationship between variables (row entitled ‘emPAI’ in Table 6A and 6B). This group was followed by ‘significant matches’ and ‘significant sequences’ which were further analyzed through linear regression (row entitled ‘emPAI’ in Table

6A and 6B). Linear regression analysis of score, significant matches, and significant sequences vs emPAI identified binding partners of Btf/TRAP150 with outstanding statistical values in each category (see Figures 13 and 14). Proteins of interest with strong MS-based proteomics data including RNA binding proteins, nuclear export factors, ribosomal subunits, Fragile X proteins, and more were indicated by their values on the line graphs. The 3 figures from the Btf and

TRAP150 proteomics data also display several binding partners with similar low statistical values (lower left portion of graph; Figures 13 and 14). These protein partners will not be selected for future Btf/TRAP150 studies involving subcellular localization and function.

Table 6: Correlation between variables for Btf and TRAP150

Binding partners were categorized by the following: score, mass, # of significant matches, # of significant sequences, and overall emPAI. Correlation analysis shows which variables have the strongest linear relationship. Score, significant matches, and significant sequences have the highest coefficient value with emPAI which suggest that these categories are closely related (row entitled ‘emPAI’ in Table 6A and 6B).

A. Relation of statistical categories for Btf

B. Relation of statistical categories for TRAP150

Figure 13: Linear regression analysis for Btf binding partners

The following line graphs compare score, significant matches, and significant sequences

to emPAI which quantifies overall peptide abundance. Btf binding partners with outstanding values in each statistical category are represented by points towards the upper right portion of the graph. In contrast, binding partners with low values are grouped together towards the bottom left portion.

score vs emPAI for Btf y = 0.0021x + 0.2151 R² = 0.3385 10 9 8 7 6 5

emPAI 4 3 2 1 0 0 500 1000 1500 2000 2500 3000 Score

Figure 14: Linear regression analysis for TRAP150 binding partners

The following line graphs compare score, significant matches, and significant sequences to emPAI which quantifies overall peptide abundance. TRAP150 binding partners with outstanding values in each statistical category are represented by points towards the upper right portion of the graph. In contrast, binding partners with low values are grouped together towards the bottom left portion.

score vs emPAI for TRAP150 y = 0.0019x + 0.6162 R² = 0.0601 16 14 12 10 8

emPAI 6 4 2 0 0 200 400 600 800 1000 1200 1400 score

CHAPTER 5: DISCUSSION

My study of endogenous binding partners for Btf and TRAP150 is a good resource for

future research in protein-protein interaction. The lists created from MS-based proteomic

analysis of immunoprecipitates include several common/distinct binding partners of these non-

classical SR-like proteins. I have identified 207 endogenous protein partners for Btf and 262 for

TRAP150. As expected, many interacting proteins have putative roles in pre-mRNA splicing

whereas others function in pathways outside of pre-mRNA processing. I’ve also identified 153 common protein partners between Btf and TRAP150 which suggests that the proteins share functions in specific pathways. Future studies will shed light on the novel biochemical functions

of binding partners not previously attributed to Btf or TRAP150. Binding partners with distinct

functions will elucidate additional roles for Btf and TRAP150 whereas common partners will

confirm their overlapping roles in pre-mRNA processing as well as other pathways.

The goal of Aim 1 was to immunoprecipitate Btf and TRAP150 complexes and confirm

the targets along with their protein partners. Immunoblots from the three biological IP replicates

showed that Btf and TRAP150 targets were successfully isolated via their respective antibodies

(lane 4, Figures 7, 8, and 9). Several controls helped support our results by eliminating the

possibilities of non-specific binding to WU10, TRAP956, and the Protein A agarose beads. In

the second IP replicate, Btf and TRAP150 signal was unexpectedly identified in the bound

fraction of the ‘extract only’ reaction (lanes 8, Figures 8A and 8B). This was likely due to small

amounts of target protein directly sticking to the agarose beads. The non-specific interaction may

have also been due to a sticky protein indirectly binding with our targets in a complex. In the

third IP replicate, Btf signal was identified in the ‘pre-cleared beads’ control which usually

signifies non-specific binding to agarose beads (lane 2, Figure 9A). However, the target was not

80 completely removed as shown in the remaining immunoprecipitates. Lane 2 is consistent with the other lanes in the IP replicate which suggests a high abundance of protein in the cell extract

(Figure 9A).

Silver staining results for Btf and TRAP150 did not support our expectations. The goal of the experiment was to reveal common and distinct protein bands for binding partners according to their molecular weight. Silver staining assays were also included in case individual bands required MS sequencing rather than the entire IP complex mixture. However, the MS results were conclusive which made the need for individual sequencing less important. Binding partners were not consistently detected as strongly stained protein bands in the immunoprecipitates containing ‘extract + antibody’ (Figures 10, 11, and 12). Silver staining analysis showed several proteins in the ‘input’ samples but only WU10 and TRAP956 antibodies in the bound fractions.

Some distinct protein bands faintly appeared at larger molecular weights (150 kDa >) for the ‘Btf

+ antibody’ fractions as opposed to ‘TRAP150 + antibody’ (Btf, lane 4 in Figure 10; Btf, lane 3 in Figure 12A). One hypothesis is likely due to silver staining being a sensitive technique that requires longer exposure with developer reagents. Another reason could be that the silver staining kit used is less compatible with our IP studies. Previous studies in the Bubulya lab used a different silver staining method that highlighted common/distinct endogenous partners of Btf and TRAP150 (Cheedu et al., 2016).

The goals of Aim 2 included using MS-based proteomics to generate a complete list of endogenous binding partners for Btf and TRAP150. Our objectives also consisted of evaluating the MS data and creating an efficient system for selecting novel proteins in future studies. As expected, the lists were composed of nuclear speckle proteins, EJC components, ribosomal subunits, and other proteins that regulate pre-mRNA processing. The lists also confirmed that Btf

and TRAP150 are strong binding partners of each other (Tables 2 and 3). Among their notable

common protein partners were SRSF1, SRSF6, U2AF2, SNUT1, and SF3B2 which are all

critical in pre-mRNA splicing (Table 3). I also identified several unique binding partners that

function outside of pre-mRNA processing such as insulin signaling proteins, nuclear lamina

filament proteins, and chromatin condensation inducers (Table 3). Notable binding partners of

Btf include H14 (histone) and MCM5 (DNA replication factor). In contrast, distinct binding

partners of TRAP150 include MDC1 (DDR checkpoint protein) and BIP (endoplasmic reticulum

chaperone).

GO enrichment analysis quantified all biological pathways impacted by Btf and

TRAP150 binding partners. The pathways identified are consistent with the overlapping hits

from the MS-based proteomics lists. However, there were many cellular processes shown with

different amounts of binding partners, fold enrichment, and p-values. For example, the top

biological pathway for Btf is mRNA nuclear export which has 110 binding partners and a 13.74-

fold enrichment value. Whereas the top pathway for TRAP150 is protein localization with 88 binding partners and a 4.28-fold enrichment value. These results support previous studies which depict Btf depletion causing an increase of global polyadenylated RNA and BTM transcripts in

the cytoplasm (Varia et al., 2013). The gene ontology search helped further distinguish Btf and

TRAP150 binding partners by showing the extent of association in mRNA distribution and other

biological pathways.

I used correlation and linear regression analysis to categorize the proteomics results and

select common/distinct protein partners. Pearson correlation identified that score, significant

matches, and significant sequences have strong linear relationships with overall emPAI (row

entitled ‘emPAI’ in Table 6A and 6B). Also, linear regression studies depicted Btf and TRAP150

binding partners with high statistical values in each category (Figures 13 and 14). Although interacting partners can be selected at random, statistical analysis offers a time and cost-efficient method for exploring novel proteins. This system is useful in confirming peptide interaction between Btf/TRAP150 and binding partners prior to exploring subcellular localization/function.

Moreover, protein partners with strong statistical data are more likely to show spatial overlap in subcellular compartments with Btf and TRAP150. The binding partners may also become regulated by the SR-like proteins in specific cellular pathways.

Among the most interesting binding partners identified were the Fragile X mental retardation proteins FMR1, FXR1P, FXR2P, and NUFP2 due to their unique functions. Fragile X syndrome is an X-linked genetic disorder caused by a full silencing mutation of the FMR1 gene

(Turner, 1996). Symptoms include mental retardation, attention deficit, anxiety, anger impulse, and overall disruptive behavior. Changes in the FMR1 gene impede the synthesis of Fragile X mental retardation protein, an RNA binding factor involved in nuclear/cytoplasmic subcellular localization of mRNPs (Bardoni, 2001). In addition, lack of expression of FMRP affects the regulation of RNA splicing as well as translational control of mRNAs. FXR1P and FXR2P are autosomal paralogs of FMRP that play roles in muscle, heart, and neuronal development

(Tamanini et al., 2000).

Previous studies suggest that FXR1P and FXR2P isoforms have functional similarities in

RNA binding, polyribosome association, and nucleocytoplasmic localization but differences in

RNA metabolism (Tamanini et al., 2000). FXR1P is a common binding partner of Btf/TRAP150 that shuttles between cytoplasm and nucleoplasm (Table 3) (Tamanini et al., 2000; Ma et al.,

2014). In contrast, FXR2P binds to TRAP150 and shuttles between the cytoplasm and nucleolus

(Tamanini et al., 2000). Although previous research has studied interaction between Btf and

FXR1P, little is known about the complete relationship of all FMRP proteins. Btf and TRAP150 are homologous SR-related proteins that commonly bind to FMR1, FXR1P, and NUFP2, but not

FXR2P (Table 3). Therefore, future research will uncover the mechanisms behind the selective

interaction between similar proteins.

Yeast two-hybrid studies confirmed that Btf interacts and co-localizes with FXR1P in the

cytoplasm of rat vascular smooth muscle cells (VSMCs) (Ma et al., 2014). A full reading frame

sequence of human FXR1 was inserted into a pGBKT7 vector during a yeast two-hybrid screening. Next, the recombinant plasmid was transfected into an AH109 yeast strain to create a bait protein. A second pGADT7 was created from a cDNA library and transfected with the Y187 pre-transformed yeast strain. After mating between the two yeast hybrids, a β-galactosidase assay produced 10 positive colonies (Ma et al., 2014). One distinct colony expressed 100% identity to the Btf transcription factor.

Direct protein interaction between FXR1P and Btf was validated through co- immunoprecipitation in co-transfected VSMCs in vitro (Ma et al., 2014). Last, FXR1 and Btf were fused with GFP and RFP respectively to show precise co-localization in cytoplasmic regions near the nucleus (Ma et al., 2014). Whether the proteins directly interact or share a linking factor in a complex remains unknown. Although the Btf and FXR1P do not share structural similarities, they both play key roles in cell growth, cell proliferation, and repair (Ma et al., 2014). Therefore, future studies will also determine if Btf and/or TRAP150 regulate the function of FMRP isoforms in specific cellular pathways. We hypothesize that Btf/TRAP150 bind to members of the Fragile X mental retardation complex and overlap in subcellular compartments.

In conclusion, Btf and TRAP150 were immunoprecipitated from whole cell nuclear

extract via WU10 and TRAP956 antibodies. IP samples from three biological replicates were

analyzed in immunoblot and silver staining assays. Immunoblot analysis confirmed successful

Btf and TRAP150 isolation in immunoprecipitates containing ‘extract + antibody’ reactions

(Lane 4; Figures 7, 8, and 9). In contrast, silver staining assays did not support our expectations

for visualizing common/distinct protein bands of binding partners (Figures 10, 11, and 12).

Although silver staining results were inconsistent, mass spectrometry results confirmed several protein partners involved in pre-mRNA processing and more. I used MS-based proteomics to list

262 binding partners for Btf, 207 binding partners for TRAP150, and 153 of which were common protein partners (Tables 2 and 3). Among those binding partners, FXR1P and FXR2P were selected for further studies based on statistical analysis and previously documented subcellular localization of the FMRP proteins.

Previous studies in the Bubulya lab suggest that Btf and TRAP150 also have distinct

functions in nuclear/cytoplasmic localization. Therefore, future studies will explore subcellular

localization of Btf/TRAP150 with FXR1P and FXR2P as well as their regulation on FMRP

function. We expect FXR1P to co-localize with Btf and TRAP150 in regions surrounding the

nucleus due to their roles in nuclear export. In contrast, FXR2P is unique to TRAP150, therefore,

we expect that these proteins will have spatial overlap in other subcellular compartments. This

study will help us further investigate the interaction of Btf and TRAP150 with their identified

binding partners. Future research will use siRNA duplexes to knockdown Btf /TRAP150 and

confirm possible mis-localization of FMRP proteins. These depletion studies will depict whether

FXR1P and FXR2P rely on Btf and/or TRAP150 for proper compartmentalization. They will

also elucidate the possible roles of Btf and TRAP150 in the FMRP nuclear export/translational

pathways.

Our results will lead to future studies such as constructing YFP deletion mutants for Btf

and/or TRAP150. These mutants will lack specific regions such as the C-terminus, N-terminus, or RS domain. Expression of deletion constructs may identify regions where binding partners interact which will help us understand unique features of Btf and TRAP150. Lastly, HA and

FLAG tags can be fused to the C-terminus or N-terminus of Btf/TRAP150 prior to different immunoassays. These epitope tags can be useful for detecting interactions between binding partners without the presence of specific antibodies. Future experiments will confirm whether

Btf and TRAP150 directly bind to FMRP isoforms or exist in the same complex as the proteins.

REFERENCES

Aebersold, R., and Mann, M. Mass spectrometry-based proteomics. Nature. 2003. 422: 198-207.

Bardoni, B., Schenck, A., and Mandel, J. The Fragile X mental retardation protein. Brain Res.

2001. 56(3): 375-382.

Barilla, D., Lee, B. A., and Proudfoot, N. J. Cleavage/polyadenylation factor IA associates with the carboxyl-terminal domain of RNA polymerase II in saccharomyces cerevisiae. PNAS. 2001.

98(2): 445-450.

Bell, P., Lukashchuk, N., Wagner, S.A., Weinert, B.T., Olsen, J. V., Baskcomb, L. Mann, M.,

Jackson, S. P., and Choudhary, C. Proteomic investigations reveal a role for RNA processing factor THRAP3 in the DNA damage response. Mol Cell Biol. 2012. 46(2): 212-225.

Benesty, J., Chen, J., Huang, Y., and Cohen, I. Pearson Correlation Coefficient. Springer Topics in Signal Processing. 2009. 1-4.

Busch, A., and Hertel, K. J. Evolution of SR Protein and hnRNP Splicing Regulatory Factors.

Wiley Interdiscip Rev RNA. 2012. 3(1): 1-12.

Cartegni, L., Wang, J., Zhu, Z., Zhang, M., and Krainer, A. Pre-mRNA splicing in the absence of an SR protein RS domain. Cold Spring Harbor. 2001. ESEfinder.

Cazalla, D., Zhu, J., Manche, L., Huber, E., Krainer, A. R., and Caceres, J. F. Nuclear Export and

Retention Signals in the RS Domain of SR Proteins. Mol Cell Biol. 2002. 6871-6882.

Cazalla, D., Newton, K., and Caceres, J.. A Novel SR-Related Protein is Required for the Second

Step of Pre-mRNA Splicing. Mol Cell Biol. 2005. 2969-2980.

Ciccia, A., and Elledge, S. J. The DNA damage response: making it safe to play with knives. Mol

Cell Biol. 2010. 40(2): 179-204.

Cheedu, D. Regulation of Mitotic Progression by Btf and TRAP150. Browse all Theses and

Dissertations. 2016. 2064: 1-96.

Cmarko, D. Verschure, P. J., Martin, T. E., Dahmus, M. E., Krause, S., Fu, X. D., Van Driel, R., and Fakan, S. Ultrastructural analysis of transcription and splicing in the cell nucleus after bromo-UTP microinjection. Mol Cell Biol. 1999. 10: 211-223.

Frege, T., and Uversky, V. N. Intrinsically disordered proteins in the nucleus of human cells.

Biochem,Biophys Rep. 2015. 1: 33-51.

Galganski, L., Urbanek, M.O., and Krzyzosiak, W. J. Nuclear speckles: molecular organization,

biological function, and role in disease. Nucleic Acids Res. 2017. 45(18): 10350-10368.

Gene Ontology Consortium. Grant P41 HG002273. 1999-2020.

Graveley, B. R., Hertel, K. J., and Maniatis, T. The role of U2AF35 and U2AF65 in enhancer

dependent splicing. RNA. 2001. 7(6): 806-818.

Hong, S. W., Hong, S. M., Yoo, J. W., Lee, Y. C., Kim, S., Lis, J. T., and Lee, D.

Phosphorylation of the RNA polymerase II C-terminal domain by TFIIH kinase is not essential for transcription of Saccharomyces cerevisiae genome. PNAS. 2009. 106(34): 14276-14280.

Ito, M., Yuan, C. X., Malik, S., Gu, W., Fondell, J. D., Yamamura, S., Fu, Z. Y., Zhang, X., Qin,

J. and Roeder, R. G. Identity between TRAP and SMCC complexes indicates novel pathways for the function of nuclear receptors and diverse mammalian activators. Mol Cell Biol. 1999. 3: 361-

370.

Kasof, G. M., Goyal, L., and White, E. Btf, a novel death-promoting transcriptional repressor

that interacts with Bcl-2-related proteins. Mol Cell Biol. 1999. 19(6): 4390-4404.

Komili, S., and Silver, P. A. Coupling and coordination in gene expression processes: a systems

biology view. Nature. 2008. 9(1): 38-48.

Lamond, A. I., and Earnshaw, W. C. Structure and function in the nucleus. Science. 1998. 280:

547-553.

Lee, K. M., Hsu, I. W., and Tarn, W. Y. TRAP150 activates pre-mRNA splicing and promotes

nuclear mRNA degradation. Nucleic Acids Res. 2010. 38(10): 3340-3350.

Le Hir, H., Gatfield, D., Izaurralde, E., and Moore, M. J. The exon-exon junction complex

provides a binding platform for factors involved in mRNA export and nonsense-mediated mRNA

decay. EMBO J. 2001. 20(17): 4987- 4997.

Le Hir, H., Izaurralde, E., Maquat, L. E., and Moore, M. J. The spliceosome deposits multiple

proteins 20-24 nucleotides upstream of mRNA exon-exon junctions. EMBO J. 2000. 19(24):

6860-6869.

Li, X., and Manley, J. L. Cotranscriptional processes and their influence on genome stability.

Genes & Dev. 2006. 20: 1838-1847.

Lin, S., and Fu, X. SR proteins and related factors in alternative splicing. Adv Exp Med Biol.

2007. 623:107-122.

Liu, Q., Guntuku, S., Cui, X. S., Matsuoka, S., Cortez, D., Tamai, K., Luo, G., Carattini-Rivera,

S., DeMayo, F., Bradley, A., Donehower, L. A., and Elledge, S. J. CHK1 is an essential kinase

that is regulated by ATR and required for the G(2)/M DNA damage checkpoint. Genes & Dev.

2000. 14(12): 1448-1459.

Long, J. C., and Caceres, J. F. The SR protein family of splicing factors: master regulators of

gene expression. Biochem J. 2009. 417(1): 15-27.

Ma, Y., Wang, C., Li, B., Qin, L., Su, J., Yang, M., and He, S. Bcl-2-associated transcription factor 1 interacts with fragile X-related protein 1. Acta Biochimica et Biophysica Sinica. 2014.

46(2): 119-127.

Mandel, C. R., Bai, Y., and Tong, L. Protein factors in pre-mRNA 3’ end processing. Cell Mol

Life Sci. 2008. 65(7-8): 1099-1122.

Maniatis, T., and Reed, R. An extensive network of coupling among gene expression machines.

Nature. 2002. 416(6880): 499-506.

Manley, J. L., and Tacke, R. SR proteins and splicing control. Genes & Dev. 1996. 10(13): 1569-

1579.

Mao, X., Schwer, B., and Shuman, S. Yeast mRNA cap methyltransferase is a 50-kilodalton protein encoded by an essential gene. Mol Cell Biol. 1995. 15, 4167-4174.

Maris, C., Dominguez, C., and Allain, F. H. The RNA recognition motif, a plastic RNA-binding platform to regulate post-transcriptional gene expression. Mol Biol and Biophys. 2005. 272(9):

2118-2131.

Martinez, N., and Gilbert, W. Pre-mRNA modifications and their role in nuclear processing.

Quant Biol. 2018. 6(3): 210-227.

Martinez-Rucobo, F. W., Kohler, R., van de Waterbeedmd, M., Heck, A. J., Hemann, M.,

Herzog, F., Stark, H., and Cramer, P. Molecular Basis of Transcription-Coupled Pre-mRNA

Capping. Mol Cell. 2015. 58(6): 1079-1089.

Matsuoka, S., Huang, M., and Elledge, S. J. Linkage of ATM to cell cycle regulation by the

CHK2 protein kinase. Science. 1998. 282(5395): 1893-1897.

MBL Life Science. Post-transcriptional regulation mechanism. Med and Biol Lab. 2017.

McCracken, S., Fong, N., Yankulov, K., Ballantyne, S., Pan, G., Greenblatt, J., Patterson, S. D.,

Wickens, M., and Bentley, D. L. The C-terminal domain of RNA Polymerase II couples mRNA processing to transcription. Nature. 1997. 385: 357-361.

Millhouse, S., and Manley, J. L. The C-Terminal Domain of RNA Polymerase II Functions as a

Phosphorlation-Dependent Splicing Activator in a Heterologous Protein. Mol Cell Biol. 2005.

25(2): 533-544.

Mintz, P. L., and Spector, D. L. Compartmentalization of RNA processing factors with nuclear speckles. J Struct Biol. 2000. 129(2-3): 241-251.

Phatnani, H. P., and Greenleaf. A. L. Phosphorylation and functions of the RNA Polymerase II

CTD. Genes & Dev. 2006. 20: 2922-2936.

Politz, J. C., Tuft, R. A., Prasanth, K. V., Baudendistel, N., Fogarty, K.E. Lifshitz, L. M.,

Langowski, J. Spector, D. L., and Pederson, T. Rapid, diffusional shuttling of poly(A) RNA between nuclear speckles and the nucleoplasm. Mol Biol Cell. 2006. 17(3): 1239-1249.

Potabathula, D. Probing a Role for TRAP150 in Gene Regulation. Browse all Theses and

Dissertations. 2009. 1-57.

Ramirez-Clavijo S., and Montoya-Ortiz, G. Gene Expression and regulation. Autoimmunity:

From Bench to Bedside. 2013. Chapter 1.

Rosonina, E., and Blencowe, B. J., Gene Expression: The Close Coupling of Transcription and

Splicing. Cell. 2002. 12(9): 319-321.

Sacco-Bubulya, P. A., and Spector, D. L. Disassembly of interchromatin granule clusters alters

the coordination of transcription and pre-mRNA splicing. J Cell Biol. 2002. 156(3): 425-436.

Saitoh, N., Spahr, C., Patterson, S., Bubulya, P., Neuwald, A., and Spector, D., Proteomic

Analysis of Interchromatin Granule Clusters. Mol. Biol. Cell. 2004. 15:8.3876-3890.

Saldi, T., Cortazar, M. A., Sheridan, R. M., and Bentley D. L. Coupling of RNA polymerase II transcription elongation with pre-mRNA splicing. J Mol Biol. 2016. 428(12): 2623-2635.

Sarras, H. Azami, S. A., and McPherson, P. L. In Search of a Function for BCLAF1. The Sci

World. 2010. 10: 1450-1461.

Shatkin, A. J., and Manley, J. L. The ends of the affair: capping and polyadenylation. Nat Struct

Biol. 2000. 7(10): 838-842.

Shepard, P., and Klemens, H. The SR protein family. Genome Biol. 2009. 10(10): 242.

Shibagaki, Y., Itoh, N., Yamada, H., Nagata, S., and Mizumoto, K. mRNA capping enzyme,

isolation and characterization of the gene encoding mRNA guanylyltransferase subunit from

Saccharomyces cerevisiae. J Biol Chem. 1992. 267, 9521-9528.

Singh, G., Kucukural, A., Cenik, C., Leszyk, J. D., Shaffer, S. A., Weng, Z., and Moore, M.J.

The cellular EJC interactome reveals higher-order mRNP structure and an EJC-SR protein nexus. Cell. 2012. 151(4): 750-764.

Spector, D. L. Nuclear domains. J Cell Sci. 2001. 114(16): 2891-2893.

Spector, D. L., and Lamond, A. I. Nuclear Speckles. CSH Perspect Biol. 2011. 3: a000646.

Tamanini, F., Kirkpatrick, L. L., Schonkeren, J., Van Unen, L., Bontekoe, C., Bakker, C.,

Nelson, D. L., Galjaard, Hans., Oostra, B. A., and Hoogeveen, A.T. The fragile X-related

proteins FXR1P and FXR2P contain a functional nucleolar-targeting signal equivalent to the

HIV-1 regulatory proteins. Human Mol Gen, 9(10): 1487-1493.

Tange, T. O., Shibuya, T., Jurica, M. S., and Moore, M. J. Biochemical analysis of the EJC

reveals two new factors and a stable tetrameric protein core. RNA. 2005. 11(12): 1869-83.

Thiry, M. The interchromatin granules. Histol Histopathol. 1995. 10: 1035-1045.

Turner, G., Webb, T., Wake, S., and Robinson, H. Prevalence of fragile X syndrome. FRAXA

Pop Studies. 1996. 64(1): 1-240.

Varia, S., Cheedu, C., Markey, M., Torres-Shafer, K., Battini, V. P., Bubulya, A., and Bubulya,

P. Alignment of Mitotic Chromosomes in Human Cells involves SR-Like Splicing Factors Btf and TRAP150. Int J Mol Sci. 2017. 18(9): 1956.

Varia, S., Potabathhula, D., Deng, Z., Bubulya, A., and Bubulya, P. Btf and TRAP150 have distinct roles in regulating subcellular mRNA distribution. Nucleus. 2013. 4(3): 229-240.

Vohhodina, J., Barros, E. M., Savage, A.L., Liberante, F. G., Manti, L. Bankhead, P., Cosgrove,

N., Madden, A. F., Harkin, D. P., and Savage, K. I. The RNA processing factors THRAP3 and

BCLAF1 promote the DNA damage response through selective mRNA splicing and nuclear export. Nucleic Acids Res. 2017. 45(22): 12816-12833.

Wang, Y., Liu, J., Huang, B., Xu, Y., Li, J., Huang, L., Lin, J., Zhang, J., Min, Q., Yang, W., and

Wang, X. Mechanisms of alternative splicing and its regulation. Biomed Rep. 2015. 3(2): 152-

158.

Wang, Z., and Burge, C. B. Splicing regulation: From a parts list of regulatory elements to an integrated splicing code. RNA. 2008. 14(5): 802-813.

Wang, Z., Murigneux, V., and Le Hir, H. Transcriptome-wide modulation of splicing by the exon junction complex. Genome Biol. 2014. 15(12): 551.

White, R. J., and Sharrocks, A. D. Coordinated control of the gene expression machinery. Cell.

2010. 26(5): 214-220.

Will, C.L., and Luhrmann, R. Spliceosome Structure and Function. Cold Spring Harb Perspect

Biol. 2011. 3(7): a003707.

Woodward, L. A., Mabin, J. W., Gangras, P. Singh, G. The exon junction complex: a lifelong guardian of mRNA fate. Wiley Rev RNA. 2017. 8(3): 1002.

Zhou, Z., and Fu, X. Regulation of splicing by SR proteins and SR protein-specific kinases.

Chromosoma. 2013. 122. 191-207.

Zhu, J., and Krainer, A. Pre-mRNA splicing in the absence of an SR protein RS domain. Genes

& Dev. 2000. 14: 3166-3178.

ABBREVIATIONS

AB - antibody

ATM - ataxia-telangiectasia mutated

ATR - ataxia-Rad3-related

CLKs - CDC-like kinases

CPSF - Cleavage and Polyadenylation specificity factor

CstF - Cleavage stimulation factor

CTD – carboxyl terminal domain

DDR - DNA Damage Response

DMEM – Dulbecco’s Modified Eagle’s medium

DNA - Deoxyribonucleic acids

EJC - Exon Junction Complex

FMRP - Fragile X mental retardation protein

GFP - green fluorescent protein

GMP - Guanosine monophosphate

GTP - Guanosine triphosphate

hnRNPs - heterogeneous nuclear ribonucleoproteins

LC-MS/MS – liquid chromatography mass spectrometry

95 mRNPs – messenger ribonucleoproteins

MS – mass spectrometry

PBS – Phosphate buffered saline

RFP - red fluorescent protein

RNA - Ribonucleic acids snRNA - small nuclear ribonucleic acids snRNPs - small nuclear ribonucleoproteins

SR - Serine-arginine-rich

SRPKS - SR protein kinases

SRSF - Serine and Arginine rich splicing factor

U2AF - U2 auxiliary factor