Role of RNA editing of Flna in platelets

Platelets are one of the major types of blood cells, and help form clots to stop bleeding. However, platelet mediated clot formation can also cause health problems including stroke, pulmonary embolism, and complications of cancer. Platelets are anucleate, yet contain thousands of functional mRNAs. Various mechanisms of posttranscriptional modification of RNA in platelets, that also alter platelet function, have been described. We predicted that ADAR-mediated RNA editing, which has been described in other cells, might be a novel mechanism of post-transcriptional modification in platelets. Using RNA-seq analysis we identified several edited transcripts in platelets. PCR amplification and Sanger sequencing were used to validate two highly edited RNA sites in platelets: coatomer subunit A (COPA) and filamin A (FLNA). Both platelets and megakaryocytes contain ADAR1 and ADAR2 transcripts, which code for the enzymes responsible for RNA editing. Western blots indicated the presence of ADAR2 protein and a high relative abundance of the cytosolic isoform of ADAR1 protein in platelets. Surprisingly, little editing was observed in megakaryocytes, the cells from which platelets are generated. Several compounds known to induce editing did not stimulate editing in the megakaryocytes in vitro. To begin to address the functional relevance of FLNA editing in platelets, we designed a wild type (unedited) and mutant (edited) FLNA construct, and performed cell culture experiments. Overexpression experiments in HeLa cell lines showed no effect of editing on FLNA protein expression levels. Visualization by confocal microscopy also indicated no apparent differences in the localization of edited versus unedited FLNA protein. Thus mRNA, including FLNA, is robustly edited in platelets, however the functional relevance of editing in platelets is still uncertain.

ROLE OF RNA EDITING OF FLNA IN PLATELETS by Amanda Khoury A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment of the Requirements for the Honors Degree in Bachelor of Arts In Biology Approved: ____________________ Dr. Jesse Rowley Supervisor ____________________ Dr. Neil Vickers Chair, Department of Biology ____________________ Dr. Darryl Kropf Department Honors Advisor ____________________ Dr. Sylvia D. Torti Dean, Honors College May 2014     ABSTRACT Platelets are one of the major types of blood cells, and help form clots to stop bleeding. However, platelet mediated clot formation can also cause health problems including stroke, pulmonary embolism, and complications of cancer. Platelets are anucleate, yet contain thousands of functional mRNAs. Various mechanisms of posttranscriptional modification of RNA in platelets, that also alter platelet function, have been described. We predicted that ADAR-mediated RNA editing, which has been described in other cells, might be a novel mechanism of post-transcriptional modification in platelets. Using RNA-seq analysis we identified several edited transcripts in platelets. PCR amplification and Sanger sequencing were used to validate two highly edited RNA sites in platelets: coatomer subunit A (COPA) and filamin A (FLNA). Both platelets and megakaryocytes contain ADAR1 and ADAR2 transcripts, which code for the enzymes responsible for RNA editing. Western blots indicated the presence of ADAR2 protein and a high relative abundance of the cytosolic isoform of ADAR1 protein in platelets. Surprisingly, little editing was observed in megakaryocytes, the cells from which platelets are generated. Several compounds known to induce editing did not stimulate editing in the megakaryocytes in vitro. To begin to address the functional relevance of FLNA editing in platelets, we designed a wild type (unedited) and mutant (edited) FLNA construct, and performed cell culture experiments. Overexpression experiments in HeLa cell lines showed no effect of editing on FLNA protein expression levels. Visualization by confocal microscopy also indicated no apparent differences in the localization of edited versus unedited FLNA protein. Thus mRNA, including FLNA, is robustly edited in platelets, however the functional relevance of editing in platelets is still uncertain.       ii   TABLE OF CONTENTS ABSTRACT ii INTRODUCTION 1 METHODS 8 RESULTS 12 DISCUSSION 24 REFERENCES 29       iii   1 INTRODUCTION Mulitpotent hematopoietic stem cells found in the bone marrow give rise to myeloid and lymphoid cell types found in the blood [12]. Specifically, hematopoietic stem cells can develop into two precursor cell types, burst-forming cells and colonyforming cells, both of which express the CD34 antigen. Maturation of these cells types lead to precursor megakaryocytes (megs), which eventually develop into mature megakaryocytes, a center for platelet biogenesis [1]. Efficient formation of megakaryocytes, or thrombopoiesis, is known to require only one cytokine called thrombopoietin (TPO). Although mature megakaryocytes are primarily found in the bone marrow, they can also be found in the lungs and in peripheral blood [2]. The structure of a megakaryocyte, especially its cytoplasm and membrane system, is fitted for efficient platelet biogenesis. Before a megakaryocyte is able to release platelets, it expands considerably by multiple cycles of endomitosis (process that undergoes most aspects of mitosis except telophase and cytokinesis), regulated by TPO. This process amplifies the DNA as much as 64-fold and fills the cell with a high concentration of ribosomes allowing for production of platelet specific proteins [3]. Other modifications, such as assembly of mitochondria and dense granules, also occur during maturation so that by complete maturation, megakaryocytes are prepared with all the machinery required for the costly task of platelet biogenesis [2]. The formation of platelets from megakaryocytes is not completely understood and requires multiple transcription factors working in concert. The process begins with an extension of the cytoplasm into long, branched proplatelets from the megakaryocyte. The ends of these proplatelets swell, as it is the primary site of platelet assembly and release       2 [2]. After the megakaryocytes are completely assembled into a network of proplatelets, retraction occurs, releasing portions of the proplatelets from the main cell body, which mirrors release of platelets (Figure 1). The mechanism by which the platelets are released is not known. Each megakaryocyte is estimated to generate thousands of platelets [4]. As remnants of the megakaryocyte, the platelets are supplied with organelles and molecules from the megakaryocyte, including mRNA [25]. Figure 1: Anatomy of a proplatelet [2]. This figure illustrates a megakaryocyte in the process of producing proplatelet extensions (shaft) and swellings, where the majority of platelet biogenesis occurs. Adopted by Patel SR, 2005. The mature platelets are small, diskshaped, anucleate cells with an average life span of only five to nine days. Platelets circulate in the blood and are vital for hemostasis by their ability to initiate blood clots. Platelet adhesion and activation is triggered at sites of vascular wall injury and requires the interaction of platelets with molecules of the subendothelial extracellular matrix [5]. Platelets are known to interact with macromolecules including collagen, von Willebrand factor [7] and fibrinogen. Platelet integrin alpha2B (IαIIB) and integrin beta3 (Iβ3) interact with fibrinogen to facilitate platelet aggregation and act as cell-signaling molecules during platelet activation [36]. Activation of integrins strengthens the interaction of platelets with the blood vessel wall [6]. Firm adhesion of the platelets to the injured vessel wall establishes a platelet monolayer [5]. Initial adhesion and activation triggers release of mediators from platelets that are known to interact via G protein-       3 coupled receptors, which can further amplify their own formation and release, resulting in a growing thrombus [5]. The growing thrombus activates clotting factors that will lead to the conversion of a soluble protein, fibrinogen, into a cross-linking protein called fibrin to form the blood clot [31]. A key molecule, and pertinent to this thesis, that facilitates platelet formation and platelet activation is FLNA. FLNA is found in both megakaryocytes and platelets. A 280kDa actin-binding protein, FLNA interacts with other proteins to establish a molecular scaffold, and to promote cell motility and signaling [27]. FLNA contains an N-terminal actin-binding domain followed by 24 Ig-like repeats, and a C-terminal domain required for dimerization. In humans, FLNA has been shown to interact with SYK [6], glycoprotein Ibα (GPIBα), the von Willebrand factor, and integrin αIIbβ (leading to activation). FLNA allows for correct platelet adhesion and plasma membrane stability during platelet activation [29]. The binding of FLNA to actin is important for regulating cytoskeleton dynamics. For example, research shows that M2 melanoma cells lacking FLNA have unstable surfaces and irregular bulging of the plasma membrane [7]. Further, gene knockout studies have demonstrated that FLNA is required for production and proper function of platelets [28]. Although some clots are beneficial, some can be extremely harmful. Blood clots can cause many health problems including stroke, heart attack, pulmonary embolism, and complications of cancer [9]. According to the American Heart Association, thrombi (clots) are a leading cause of death and disability in the United States [37]. There are also many known diseases associated with platelet malfunction. Bernard-Soulier syndrome is characterized by increased bleeding time due to impaired platelet adhesion because of the       4 absent glycoprotein that acts as the Von Willebrand factor receptor on the platelet surface [10]. Another disease called Gray Platelet Syndrome results in an increased risk of bleeding. This is because proteins required for activation of platelets are incorrectly packaged [11]. Although the role of platelets in adverse thrombus formation is well appreciated [9], the variable effectiveness of antiplatelet therapies [8] indicate our lack of understanding of how platelets function. To more effectively combat the health burden of platelet related disease, more studies are needed to elucidate the underlying mechanisms governing platelet function. Various mechanisms of post-transcriptional modification of RNA in platelets, such as splicing of pre-mRNA into mRNA within the cytoplasm of platelets, that also alter platelet function, have been described [13]. We predicted that ADAR-mediated RNA editing, which has been described in other cell types, might be a novel mechanism of post-transcriptional modification in megakaryocytes and platelets. RNA editing is a post-transcriptional process that causes a chemical change in the base makeup of transcripts by various enzymes. There are two major types of RNA editing. The first type, which is specific but not very common in mammals, is C-to-U editing by the APOBEC family of enzymes [15]. However, the most common type of RNA editing in mammals is A-to-I editing, where adenosine is deaminated to produce inosine by adenosine deaminases acting on RNA (ADAR) enzymes [14]. There are three major ADAR gene families (ADAR1-3) identified in vertebrates. ADAR1 and ADAR2 can be found in multiple tissue types in the body, while ADAR3 is localized specifically in the brain. Further, ADAR1 is known to have two isoforms: cytoplasmic isoform (150kD) and nuclear form (110kD), therefore editing may occur in       5 either the nucleus or the cytosol [16]. ADARs recognize double-stranded RNA (both intra- and intermolecular dsRNA) as their key substrate [18]. All three ADAR gene families have common domain structures, notably one to three repeats of the dsRNA-binding domain (dsRBD) that is required for dsRNA binding (Figure 2). The gene families also have distinct features that are conserved domains across species [17]. Figure 2: Mechanisms of ADAR Editing [19]. ADAR-mediated RNA editing recognizes and binds double-stranded RNA via its double stranded binding domain, so that the target nucleotide is located near ADARs catalytic domain which leads to an A to I conversion. Adopted by Hamilton CE, 2010 A to I editing by ADARs can occur at apparently random sites within double stranded RNA regions. On the other hand, editing also occurs at conserved specific sites. Specific editing requires imperfect double stranded RNA pairing between the exon sequence around the editing site and another downstream editing complementary sequence (ECS), which is usually an intronic sequence. Thus, mismatched bases result in bulges and loops that ADARs recognize. ADAR catalyzes hydrolytic deamination of adenosine to inosine at these sites [17]. Interestingly, some edited sites are favorably edited only by ADAR1 or ADAR2, which is most likely attributed to the location of their dsRBDs. ADAR1 and ADAR2 require homodimerization, thus positioning specific residues near ADAR’s catalytic domain, while ADAR3 does not seem to require homodimerization [17]. Inosines contained within mRNA are recognized as guanosines by the ribosome during translation. Therefore the A-I edit in coding sequences can lead to amino acid changes, and potential changes in protein function [16]. Because RNA editing of specific       6 loci is not complete, there are usually two proteins co-expressed in the same cell: the edited and unedited versions, which can be regulated in a cell-type specific or timedependent manner [16]. ADAR-mediated RNA editing sites are more frequent outside of the coding regions including Alu repeats, 5’UTRs, 3’UTRs and intronic sequences. The A-I conversion has many studied physiological effects including creating alternative splice sites, altering the amino acid sequence, effects on transport, as well as silencing of retrotransposons [16]. RNA editing is also known to play a major role of miRNA biogenesis, by allowing for a large increase of diversity and potential targets [23]. Research in cell culture and model organisms continues to delineate the physiologic relevance of ADAR mediated RNA editing. For example, A-to-I editing significantly affects the function of neurotransmitter receptors in the central nervous system of mammals, flies and worms. One of the most studied receptors is the glutamate receptor subunit GluR-2. The GluR-2 receptor undergoes over 99.9% editing at one position leading to a Q to R amino acid change, which is vital to maintain low Ca2+ permeability of the channel [20]. Reduced editing of the glutamate receptors causes increased permeability of Ca2+. Transgenic mice with even slightly reduced GluR-2 editing suffer sever epileptic seizures and die within two weeks [21]. ADAR-mediated RNA editing also plays an important role in suppressing interferon signaling, and thereby block premature apoptosis in hematopoiesis [22]. ADAR1 mouse knockouts therefore suffer from liver and bone marrow hematopoietic defects and typically die at their embryonic stage [19]. Other studies link RNA editing with cancer. Analysis of Alu sequences indicate that brain tumors undergo significantly less editing compared to “normal” brains. This is true       7 across other tissues as well, including prostate, lung, kidney and testis tumors. Interestingly, FLNA, one of the proteins of interest in our experiments, is also significantly less edited in brain tumors [24]. Although the function of RNA editing has been studied in other cell types, very little is known about RNA editing in platelets. A preliminary bioinformatics analysis of our previously published human and mouse platelets RNA-seq transcriptome data sets [26] suggested several transcripts are edited in platelets. In particular, COPA and FLNA are potentially edited in platelets. This thesis examines ADAR enzyme expression in platelets and the presence, origin and the functional significance of potential RNA edited sites in platelets. We find that ADAR1 and ADAR2 transcripts are found in both megakaryocytes and platelets. We verified the presence of ADAR1 and ADAR2 proteins in platelets and the presence of ADAR2 protein in megakaryocytes. We verified by Sanger sequencing the robust editing of COPA and FLNA in platelets. Interestingly, we see little editing in megakaryocytes, and various compounds predicted to induce editing did not induce editing of FLNA or COPA in megakaryocytes. In order to address the functional significance of editing, we performed overexpression experiments in HeLa cells and observed the effects on protein levels and localization.       8 METHODS RNA Isolation, PCR and Sanger Sequencing CD45 magnetic bead (to reduce leukocytes, Miltenyl Biotec) depleted Platelets, Cd34 megakaryocytes and Meg-01 cells lysed in 1ml of TRIzol® (Invitrogen™, Ambion®, Life Technologies) were obtained. RNA was isolated using 20% chloroform extraction precipitated with isopropanol, and washed with ethanol. RNA was treated with Turbo DNAse (Ambion®, Life Technologies) to remove contaminating DNA. The RNA was re-precipitated overnight using Sodium Acetate:Ethanol, Reverse transcription of the RNA was performed using Superscript III (Invitrogen™, Life Technologies). Primers flanking COPA and FLNA were designed using primer3 software [32]. Sequences of primers used are: FLNAforward: GTGTGGCTTATGTGGTCCAGPCR, FLNAreverse: ATGAAGCGCACAGCATACTT, COPAforward: GATCAGACCATCCGAGTGTG, and COPAreverse: GAAGGCAGCCCAGTTTACTC amplification of the cDNA was performed using pfx polymerase (Platinum®, Life Technologies) PCR products were cut out of a 1.5% agarose gel and purified using a DNA gel purification kit (Qiagen). Samples were sequenced by the University of Utah Sanger sequencing Core lab. Isolation of Bone Marrow Megakaryocytes Bone marrow aspirates were purchased from Lonza. Megakaryocytes were isolated using density gradient centrifugation followed by magnetic bead selection following the procedure found in Tanaka et. al. [30]. In some instances, human bone marrow megakaryocytes were further purified using a laser capture technique.       9 Western Blot Analysis Platelets or megakaryocytes were lysed with either lamelli buffer or RIPA buffer containing EDTA-free protease inhibitor (Roche). For samples originally lysed with RIPA, total protein was quantified by BCA assay (Pierce™), lamelli was added in a 1:1 ratio and samples were boiled for 5 min. Equal concentration of platelet or megakaryocyte lysate was loaded on an 8% SDS-page gel. The samples were run at 100 volts for about 1.5 hours. Proteins were transferred onto a nitrocellulose membrane in 1X Transfer Buffer (10% methanol) for one hour at 4˚C. The membranes were then blocked for two hours in 5% milk powder in 0.1% TBST Blocking Buffer (B.B.) at room temperature. ADAR1 antibody (TBZ #174 kindly provided by Dr. Brenda Bass) and ADAR2 antibody (Sigma Lot A38826) were used at 1:10,000 in B.B. overnight. Membranes were washed in 0.1% TBST, and the secondary antibody, HRP anti-rabbit (Cell Signaling #7074S) in 1:2000 B.B. was added for one hour at room temperature. Finally, the membranes were washed in TBST and developed using 1:1 ECL Blotting Detection reagents (Amersham). A western blot was also performed on anti-MYC samples following the same protocol. Anti-MYC primary antibody (Santa Cruz 9E10) was used in a 1:100 dilution.       10 Inducing RNA Editing Cd34 megakaryocytes at 2.8 million cells/ml media were obtained in a 6-well tissue culture plate. Various compounds, previously known to induce RNA editing were chosen: Tg003, TNF-alpha, cycloheximide, and valproic acid (VPA). Tg003 was used at 100µM, TNF-alpha at 5ng/ml, cycloheximide at 100µg/ml, VPA at 400µg/ml. Plasma in a 1:1 or 1:10 dilution was also tested. Immunoprecipiation of FLNA and c-MYC tagged FLNA 7.5 billion bead-depleted platelets were obtained from a healthy donor. Platelets were washed with PSG (pipes saline glucose) containing prostaglandin E1 (PGE) then lysed with 4ml of RIPA buffer (with Roche EDTA-free protease inhibitors). Myc tagged FLNA was precipitated using a c-Myc Tag IP/Co-IP Kit (Thermo Scientific) following the elution protocol #2 as described in the manual. HeLa Cell Transfection C-terminal c-myc tagged full length mutant FLNA (MUT) (A6998G; to mimic the edited site) and full length wild type (WT) FLNA (unedited site) plasmid constructs were created by PCR mutagenesis and standard plasmid cloning techniques. HeLa cells at 90% confluency were transfected with 4 µg of MUT or WT plasmid diluted in 500µl of Optimem/Lipofectamine 2000 (Invitrogen™) according to manufacturer protocol. Transfected cells were processed 24 hours after transfection by confocal microscopy or western blot analysis.       11 Microscopy To visualize by confocal microscopy, transfected HeLa cells were plated on 6-well Fibronectin coated plate overnight to confluency. The next day, the cells were fixed by adding 4% paraformaldehyde (PFA) to a final concentration of 2%. The cells were washed with PBS then lysed with 0.1% Triton x100 for 5 min. The cells were washed again with PBS and blocked for 1 hour at room temperature in 10% goat serum. AntiMyc Primary antibody (Santa Cruz Biotech 9E10) was added in 1:100 dilution. These were incubated for 1 hour at room temperature then washed with PBS. Goat anti-mouse 546-conjugated secondary antibody was added to the samples. The cells were incubated for 1 hour in the dark then washed three times with PBS. Cells were incubated with phalloidin 488 for 20 min in the dark then washed with PBS. Cells were imaged with high-resolution confocal laser immunofluorescence microscopy (Olympus FluoView 1000).       12 RESULTS RNA-seq analysis and Sanger sequencing locate edited sites in platelets RNA-seq data from platelets [33] was analyzed for A-G mismatches between reference DNA sequence and RNA reads. Potential edited sites were further filtered by eliminating mismatches found in single nucleotide polymorphism (SNP) database (1000 Genomes) [34]. Edit sites were annotated with ANNOVAR [35] for their potential affect on splicing or coding sequence changes. Table 1 summarizes the results of potential edited sites in platelets. Gene Chrom. Fcgr2a chr1 copA Fam82B TNFAIP8L1 Apobec3c FLNA HLA DPB1 B HLA DPB1 A ZNF385A chr1 chr8 chr19 chr22 chrX chr6 chr6 chr12 Position Donor DNA Sanger cDNA Sanger donor1 donor2 donor3 . A A . . . donor1 donor2 donor3 A A A A/G A/G A/G donor1 donor2 donor3 A A A . . G donor1 donor2 donor3 A A A . . A donor1 donor2 donor3 . A/G . . . . 161475775 160302244 87516482 4654601 39414912 153579950 donor1 donor2 donor3 A A A A/G A/G A/G donor1 donor2 donor3 . . G . . . donor1 donor2 donor3 A A A A A A donor1 donor2 donor3 . . A . . . 33048694 33054457 54767903 SNP In 1000g? N SNP/edit/ Ref? N Edit N matches SLC2A3 N Ref N SNP N Edit N SNP N REF N ? Annovar: Change to Protein? ? COPA:NM_004371:exon6:c.A490G:p.I164V,C OPA:NM_001098398:exon6:c.A490G:p.I164V FLNA:NM_001110556:exon43:c.A7022G:p.Q 2341R,FLNA:NM_001456:exon42:c.A6998G: p.Q2333R Table 1: Potential RNA editing sites in platelets. The first three columns indicate the name of the gene with the potential edited site, the chromosome the gene is found on, and the position of the edited nucleotide, respectively. The fourth column indicates the use of three different human blood donors. The fifth and six columns show the results of Sanger Sequencing of DNA and cDNA (reverse transcribed RNA), respectively. The 7th column indicates that none of the shown genes matched to the SNP 1000g database (N=No). The next column indicates that COPA and FLNA were the only two genes that were validated as edited sites. Finally, the last column indicates the potential change of amino acid due to the edited site using the Annovar database.         13 We performed Sanger sequencing on PCR amplified DNA and cDNA (RNA) from up to three different donors to test for editing of several edited sites. To do this, we designed PCR primers flanking the predicted editing site. Different primers were used for the RNA and DNA for the edited sites found near splice junctions. DNA from the monolayer cells and RNA from platelets were isolated from the same individual. The edited region was amplified by PCR, run on an agarose electrophoresis gel, and bands of the correct size were cut out and purified. The products were sequenced by Sanger sequencing. A summary of the sequencing results is found in Table 1. Two of the sites (APOBEC3C, HLADPB1B) were indeed SNPs that are not present in the SNP database. Two sites (TNFAIP8L1, HLADPB1A) did not show evidence of editing by sanger sequencing. For three of the sites we were unable to specifically amplify the region from platelet cDNA (FCGR2A, ZNF385A, FAM82B). We verified two genes that undergo high level of ADAR-mediated RNA editing in platelets: filamin A (FLNA) and coatomer subunit alpha (COPA). Both of these edit sites potentially result in altered amino acid incorporation. FLNA results in a Q2333R mutation and COPA in an I164V mutation (Table 1). Figure 3 shows the Sanger sequencing results for FLNA and COPA. The shaded region in the figure highlights the nucleotide predicted by the RNA-seq data to undergo editing. Comparison of the DNA and RNA sequencing traces of FLNA indicate significant editing, as the frequency of the “A” nucleotide is almost the same as the frequency of the “G” nucleotide. Similarly, comparison of sequencing traces of DNA and RNA of COPA, also indicate significant editing. In fact, in COPA the edited nucleotide is more frequent than the unedited nucleotide.       14 C   A   B   D   Figure 3: Verification by Sanger Sequencing of editing of FLNA and COPA in Platelets. Each panel (A-D) shows sequencing traces of the DNA or the RNA from the regions flanking the editing site (highlighted in blue). (A) DNA sequence trace of FLNA. (B) RNA sequence trace of FLNA. (C) DNA sequence trace of COPA. (D) RNA sequence trace of COPA. Note that only Adenosine (green) is detected at the edit site in the DNA, while in the RNA there exists both Adenosine (Green) and Guanosine (black). Because Sanger sequencing results matched RNA-seq predictions for FLNA and COPA, we reasoned that RNA-seq of these sites could robustly detect differences in editing between samples. We therefore compared RNA-seq data of FLNA and COPA in platelets to RNA-seq reads from polymorphonuclear leukocytes (PMNs), another abundant blood cell type that arises from myeloid cell line. As shown in Figure 4 significant editing is seen in platelets while no editing is seen in PMNs (Figure 4C and 4D). This result indicates specific editing of COPA and FLNA in platelets and not in PMNs.       15 A   B   C   D   Figure 4: RNA-seq read coverage data of platelets and PMNs. Each panel (A-D) examines the read coverage around the proposed edited site, which is bound by dashed lines. The data is shown for the antiparallel strand and mismatches to the reference sequence are shown in blue. (A) RNA-seq of FLNA in platelets. (B) RNA-seq of COPA in platelets. (C-D) Shows a comparison of RNA-seq of PMNs (above black line) and platelets (below black line). (C) RNA-seq of FLNA. (D) RNA-seq of COPA.   Less editing of COPA and FLNA in megakaryocytes detected by Sanger sequencing Surprisingly, the Sanger sequencing data of megakaryocytes indicated that there was little RNA editing in megakaryocytes compared to the platelets. Due to the nature and difficulty of isolation of megakaryocytes from human bone marrow, we first examined the edited site in Meg-01 cells (an immortal human megakaryoblastic leukemia cell line). No editing of FLNA or COPA was seen in this cell type (Figure 5).       16 A B Figure 5: Sanger Sequencing of Meg-01 RNA. Each panel (A-B) shows Sanger sequencing data of region flanking target site (highlighted in blue). Green curve indicates presence of Adenosine (unedited) at target site. (A) Sequencing data of FLNA indicates no editing. (B) Sequencing data of COPA indicates no editing.   We then examined cord blood CD34+ hematopoietic stem cell derived primary megakaryocytes at different time points after culture to test whether editing occurs late in megakaryocyte maturation. In our hands, based on megakaryocyte marker expression, CD34+ stem cells differentiate into megakaryocytes by day 14 of culture in TPO (data not shown). Figure 6 shows sequencing data obtained from CD34 megakaryocytes at days 4, 8, and 14. A FLNA does not show an increase of RNA editing even B in mature megakaryocytes, while COPA shows C a minor increase of editing by day 14.     Figure 6: CD34+ Megakaryocytes RNA Sanger sequencing. Sanger sequencing of FLNA (left) and COPA (right) RNA expressed in Cd34 megakaryocytes. Target site is highlighted in blue and shows the presence of Adenosine (green) and a smaller curve of Guanosine (black). Cd34 megakaryocytes at (A) day 4 (B) day 8 or (C) day 14 of culture.   17 Because we could not detect significant editing in MEG-01 or cultured CD34+ derived megakaryocytes, we attempted to isolate native megakaryocytes from human bone marrow using density centrifugation and laser capture microscopy. Figure 7 shows an example of a megakaryocyte that was isolated by laser capture. Analysis of Sanger sequencing data in Figure 8 shows that the megakaryocytes isolated from the human bone marrow indicate partial RNA editing of the target nucleotide in both COPA and FLNA, although the Figure 7: Bone Marrow Isolation. This image is an example of a laser-captured megakaryocyte. The halo around the cell indicates where the megakaryocyte was cut out of the membrane. Platelet contaminants surround the megakaryocyte. extent of editing is not as pronounced as in the platelets. It is important note that we could not completely separate megakaryocytes from platelets even by laser capture microscopy, as seen in Figure 7. Therefore, the observed partial editing may be influenced by contamination of platelets or red blood cells. A B Figure 8: Sanger Sequencing of Megakaryocytes Isolated from Human Bone Marrow. Each panel (A-B) shows Sanger sequencing data of region flanking target site (highlighted in blue). Green curve indicates presence of Adenosine while black curve indicates Guanosine. (A) RNA sequencing data of FLNA shows partial editing. (B) RNA sequencing data of COPA show editing at target site.         18 RNA-seq confirms presence of ADAR1 and ADAR2 transcripts in platelets and megakaryocytes To identify the source of editing observed in megakaryocytes (megs) and platelets we mined RNA-seq data from platelets, CD34+ derived megs, and bone marrow megs for ADAR1 and ADAR1B (ADAR2) transcript expression. As shown in Figure 9 and Figure 10, platelets, CD34+ derived megakaryocytes, and bone marrow megs contain both ADAR1 and ADAR2 transcripts. Although RNA-seq measurements between different cell types cannot be directly compared, the RNA-seq data strongly suggest that there is higher amounts of ADAR1 transcripts than ADAR2 in all three cell types, and there is significantly higher levels of RNA transcripts of both of ADAR1 and ADAR2 in the CD34+ derived megs and bone marrow megs relative to the platelets. B A C Figure 9: RNA-seq expression of ADAR1. All three panels (A-C) are read coverage maps that allow visualization of expression levels of ADAR1 transcripts. (A) RNA-seq in platelets with scale: 0-20. (B) RNA-seq of bone marrow megakaryocytes with scale:0-125. (C) RNA-seq of Cd34 megakaryocytes with scale: 0-120.         19 A B C Figure 10: RNA-seq data of ADAR2. All three panels (A-C) are read coverage maps that allow visualization of expression levels of ADAR2 transcripts. (A) RNA-seq in platelets with scale: 05.5. (B) RNA-seq of bone marrow megakaryocytes with scale: 0-35. (C) RNA-seq of Cd34 megakaryocytes with scale: 0-13. Presence of ADAR1 and ADAR2 protein confirmed in platelets, and ADAR2 proteins confirmed in megakaryocytes. Western blot analysis of ADAR1 and ADAR2 protein expression were performed on platelet lysate as well as CD34+ megakaryocyte lysate. We could not obtain sufficient material to perform western blots on bone marrow megakaryocytes. Figure 11A shows that ADAR1 protein is present in platelets, and that there is a high relative abundance of the cytosolic isoform of ADAR1. We have not yet been able to detect ADAR1 in CD34+ megakaryocytes, although detection of this protein in megakaryocytes may require further optimization. Figure 11B shows that ADAR2 protein is present in both platelets and megakaryocytes.       20 A B 230 230 130 130 95 95 Figure 11: Western Blots for ADAR Proteins. Western blot analysis of (A) ADAR1 in platelets and (B) ADAR2 in platelets and CD34+ megakaryocytes. A)Western blot with ADAR1 antibody shows presence of cytosolic isoform (130kD) of ADAR1 in three samples of platelet lysate. (B) Western Blot with ADAR2 antibody shows presence of ADAR2 in megakaryocytes lysed in Ripa Buffer, Lamelli, or Platelets in lamelli (left to right). RNA editing of COPA and FLNA is not induced by compounds predicted to induce editing Because we did not observe editing in CD34+ derived megakaryocytes, we wondered if editing could be induced. We therefore attempted to induce RNA editing in CD34+ derived megs by culturing the cells in the presence of various compounds. The compounds that were used were thought to induce editing because of our understanding of megakaryocytes and platelet biology. The rationale for why we picked each compound is found in the discussion, and the doses used were within range of the demonstrated activity for each of the compounds. After culturing the cells for 24 hours with the various compounds, RNA was isolated, reverse transcribed, amplified and sequenced. Table 2 shows the sequence data for FLNA and COPA. None of the compounds were able to induce editing, as Sanger sequencing detected either very little editing or no editing at all.       21 Table 2: Induce RNA Editing   Sequence Data (COPA) Concentration Tg003 100 µM No Valproic Acid (VPA) 400 µg/ml No Cycloheximide 100 µg/ml Not Done Plasma 1:2 Not Done Plasma 1:10 TNF-α     Sequence Data (FLNA) Compound 200 µg/ml Editing? No No No Not Done No   22 RNA editing of target sequence of FLNA does not affect expression levels of FLNA protein To further test for the functional MYC FLNA (280kD) M W C M W C significance of RNA editing of FLNA, we designed two MYC- Figure 12: Protein Expression. Western Blot of total HeLa cell lysates (right three lanes) or MYC Immunoprecipitated Hela lysates (left three lanes). Hela cells were transfected with MUTFLNA (M), WT-FLNA (W) or Mock transfected (C) prior to lysis. tagged constructs: WT (unable to be unedited because lacks introns) and MUT (A>G to mimic a completely edited site) FLNA. These overexpressed constructs were then transfected into HeLa cells. To test whether the edited site affects protein expression in the HeLa cell, we performed a western blot on the WT, MUT and MOCK (empty vector) transfected cells with a primary antibody that binds the MYC-tag in our constructs. Figure 12 shows that we were able to successfully transfect the HeLa cells, as we can see the FLNA protein around the expected 280kD marker. However, we do not observe a significant difference in the level of protein expression between the MUT and WT transfected cells, indicating that RNA editing of FLNA may not affect protein expression in HeLa cells. RNA editing of FLNA does not affect localization of FLNA and Actin in HeLa cells. Finally, because FLNA is known to play a role in actin binding and cytoskeletal dynamics of activated platelets, we decided to test whether the edited site in FLNA affects localization of FLNA and/or actin. HeLa cells were transfected with the myc tagged WT and MUT constructs. Immunofluorescence microscopy with stain against actin (green) and antibody against MYC (red) were used and the cells were visualized       23 with a confocal microscope. Figure 13A and 13B are photographs taken with the microscope of the WT and MUT transfected HeLa cells, respectively. Fields were captured where a transfected cell (red and green) could be visualized next to an untransfected cell (green only). As represented by Figure 13, we did not observe a significant difference in the localization of the mutant (edited) FLNA compared to wild type FLNA. We also did not observe a difference in actin localization in WT versus MUT transfected cells. A B Figure 13: Confocal Microscopy of HeLa Cells transfected with MUT or WT FLNA. Shown are representative fields from confocal imaging of transfected HeLa cells. Actin stained with phalloidin is shown in green, while MYC-tagged FLNA is stained red with anti-MYC antibodies. (A) Shows WT FLNA (unedited) transfected HeLa cells. Panel A shows an image of only the actin stain, only the MYC stain, and then an overlay of both (from left to right). Not that one cell is transfected and one is not. (B) Shows MUT FLNA (edited) transfected HeLa Cells. Again, Panel B shows an image of only the actin stain, only the MYC stain, and then an overlay of both (from left to right). Note that one cell is       24 DISCUSSION Platelets, which are generated from megakaryocyte myeloid cells, are one type of blood cell that circulates in the body [1]. Correct function of platelets is necessary to stop bleeding at sites of injury [5]. However, uncontrolled platelet activation, or incorrect platelet formation has been shown to lead to many human diseases [9, 10, 11]. Although research has progressed, we still don’t have a deep understanding of platelet biogenesis or platelet modifications. This is exemplified in the lack of effective treatments for platelet related diseases. This project focuses on RNA editing in platelets, one type of posttranscriptional modification that can possibly affect proper platelet function. RNA editing has previously been shown in other studies to have a significant role in the proper function of various cell types including nerve cells, brain tumors, and liver cells [19, 20, 22, 24]. The most common type of RNA editing in mammals is ADARmediated RNA editing, which results in an Adenosine to Inosine conversion [15]. Prior knowledge of RNA editing and its functional role in other cell types, lead us to believe that it may have a functional role in platelets as well. By examining RNA-seq data from the human transcriptome we were able to find two potential targets of ADAR editing in platelets: COPA and FLNA. This is the first time any edited sites have been examined in platelets, and the physiological implication is not known. In order to verify the edited sites, we used PCR and Sanger sequencing to amplify and examine the target sequence. The DNA sequence of the edit sites in FLNA and COPA contained only one chromatogram peak (Adenosine). The sequenced RNA, of both COPA and FLNA, contained two chromatogram peaks at the edited site. This indicates that the nucleotide at that site is variable, and is either an Adenosine (A) or a       25 Guanosine (G). Because we only observed the variable nucleotide at the RNA level, the modification must have occurred at the transcript level. Further, the nucleotide varies between A and G, which is indicative of ADAR-mediated RNA editing, as the inserted Inosine by ADAR is read as a Guanosine by the polymerase machinery. These two findings confirm the two edited sites in COPA and FLNA. RNA-seq analysis correlated with Sanger sequencing data, allowing us to compare editing in different cell types using RNA-seq. Interestingly, we did not observe editing of FLNA or COPA in PMN’s, a closely related myeloid cell. During platelet biogenesis, mRNA from the megakaryocytes is passed on to the platelets. We would therefore predict that the mRNA in both platelets and megakaryocytes would have the same sequence. However, PCR and Sanger sequencing of Meg-01 cells, an immortalized megakaryocyte cell line showed little editing. CD34+ stem cell derived megakaryocytes also showed little editing, although very slight editing was observed in late stage megakaryocytes, suggesting that editing may begin to occur late in megakaryocyte development. As discussed in the background, megakaryocytes undergo various processes and modifications as they mature into a platelet biogenesis center [1,2,3]. RNA editing could potentially be one of these processes that are activated in the formation of proplatelets. Using our knowledge of megakaryocyte and platelet biology we tried to induce editing of FLNA and COPA in CD34+ derived megakaryocytes using various compounds (Table 2). The first compound we used, TG003, is a compound that inhibits splicing [39]. TG003 has not been seen to induce editing in previous experiments. However, by inhibiting splicing in the CD34+ megakaryocytes, we hypothesized that       26 TG003 would give the cell more time to form the intron/exon dsRNA required for recognition by ADAR. Cycloheximide, a protein synthesis inhibitor, also inhibits nonsense-mediated decay (NMD) of unspliced transcripts that escape the nucleus [40]. There is evidence that platelets retain unspliced cytoplasmic RNA [13], however we hypothesized that CD34+ megakaryocytes may have a functioning NMD mechanism whereas platelets do not. With functioning NMD, CD34+ megs, as opposed to platelets, may not contain unspliced messages that can form the dsRNA intron/exon pairing. We predicted that blocking of NMD by cycloheximide might induce editing by allowing the formation of intron/exon pairing. Valproic acid is a compound that induces differentiation of cultured megakaryocytes to express markers of native megs [41]. We also tried to induce editing by growing the CD34+ megakaryocytes in the presence of plasma. A major difference between platelets, bone marrow megs, and our cultured megakaryocytes is the absence of plasma within CD34+ derived megakaryocyte cultures. Using plasma, we attempted to replicate the platelet environment. TNF-alpha, an inflammatory stressor is a known inducer of ADAR-mediated RNA editing [38]. Despite our strong biological rationale for the use of each compound, none of the above treatments induced editing in cultured megakaryocytes. Finally, we tried to induce editing, using interferon-alpha, a well-known RNA editing inducer [42], but were unable to see any editing (data not shown). Bone megakaryocytes demonstrated higher levels of RNA editing than the Meg01 cells or CD34+ derived megakaryocytes (Figure 8). A caveat to this result is that the bone marrow isolation may not have been efficient at obtaining purified megakaryocytes. The isolation may be contaminated with red blood cells or platelets. To confirm the       27 results, further experiments are requires in order to optimize isolation techniques in order to ensure no platelet contamination. ADAR enzymes are necessary in order to obtain the A to I conversion [14]. Therefore, one possible reason to explain the decrease or lack of editing in megakaryocytes could be attributed to the absence of ADAR transcript and protein in megakaryocytes. However, the presence of ADAR1 and ADAR2 transcripts in RNA-seq analysis of megakaryocytes and platelets does not support this hypothesis. We also detected ADAR1 and ADAR2 proteins in platelets, and ADAR2 in megakaryocytes (ADAR1 still not confirmed) in our western blots. Therefore the lack of editing in megakaryocytes is likely not due to the absence of ADARs. Why megakaryocytes display little editing is still a mystery. We examined the potential functional role of RNA editing in platelets. We hypothesized that editing might induce a change in protein expression levels of FLNA. By overexpressing edited versus unedited FLNA transcripts in HeLa cells and comparing protein levels by western blot, we saw no significant effect on expression levels. FLNA plays a role in actin binding and rearrangement during platelet activation. RNA editing of FLNA could potentially be necessary for proper binding of actin and maintaining cell structure. The edited site of FLNA potentially results in an amino acid change from glutamine (nonpolar) to arginine (positively charged). The change from a nonpolar amino acid to a charged amino acid could alter interactions of FLNA with actin. However, we did not observe overt differences in cytoskeletal arrangement or FLNA localization when comparing edited versus unedited FLNA. A major caveat to these experiments is the use of HeLa and not platelets for overexpression because transfection of anucleate platelets is       28 technically unfeasible. Thus mRNA, including FLNA mRNA, is robustly edited in platelets, however the functional relevance of editing in platelets is still uncertain. One possible effect of FLNA editing could be alteration of FLNA phosphorylation. The edited site in FLNA occurs near two phosphorylation sites. Changes in phosphorylation could have downstream signaling effects. The edited site might also effect on splicing of the FLNA transcript. The edited site occurs near a splice site and RNA editing in other cell types has been seen to effect splicing [13]. However, this is probably not an effect in platelets, as we do not see different size of FLNA protein in megakaryocytes versus platelets. Another experiment can examine the edited sites effect on binding partners. FLNA is a major binding protein and binds proteins required in correct platelet function. The edited site may be required for correct binding. Animal models with specific knockout of ADARs in platelets will help address these questions in the future.       29 REFERENCES [1] Briddell RA, Brandt JE, Straneva JE, Srour EF, Hoffman R. Characterization of the human burst-forming unit-megakaryocyte. Blood. 1989;74(1):145-51. [2] Patel SR, Hartwig JH, Italiano JE. The biogenesis of platelets from megakaryocyte proplatelets. J Clin Invest. 2005;115(12):3348-54. [3] Long MW, Williams N, Ebbe S. Immature megakaryocytes in the mouse: physical characteristics, cell cycle status, and in vitro responsiveness to thrombopoietic stimulatory factor. Blood. 1982;59(3):569-75. [4] Italiano JE, Lecine P, Shivdasani RA, Hartwig JH. Blood platelets are assembled principally at the ends of proplatelet processes produced by differentiated megakaryocytes. J Cell Biol. 1999;147:1299–1312. [5] Offermanns S. Activation of Platelet Function Through G Protein-Coupled Receptors. Circulation Research. 2006;99(12):1293-1304. [6] Falet H, Pollitt AY, Begonja AJ, et al. A novel interaction between FlnA and Syk regulates platelet ITAM-mediated receptor signaling and function. J Exp Med. 2010;207(9):1967-79. [7] Jurak Begonja A, Hoffmeister KM, Hartwig JH, Falet H. FlnA-null megakaryocytes prematurely release large and fragile platelets that circulate poorly.Blood. 2011 Aug 25;118(8):2285-95. Epub 2011 Jun 7. [8] Bhatt DL. What makes platelets angry: diabetes, fibrinogen, obesity, and impaired response to antiplatelet therapy? Journal of the American College of Cardiology. 2008;52:1060-1. [9] Trip MD, Cats VM, Capelle FJ van, Vreeken J. Platelet hyperreactivity and prognosis in survivors of myocardial infarction. The New England journal of medicine. 1990;322:1549-54. [10] Balduini CL, Iolascon A, Savoia A. Inherited thrombocytopenias: from genes to therapy. Haematologica. 2002;87(8):860-80. [11] Cramer EM, Vainchenker W, Vinci G, Guichard J, Breton-gorius J. Gray platelet syndrome: immunoelectron microscopic localization of fibrinogen and von Willebrand factor in platelets and megakaryocytes. Blood. 1985;66(6):1309 [12] Michelson, Alan. Platelets . 2nd ed. . Burlington: Elsevier, 2007. 79-82. Print.       30 [13] Zimmerman GA, Weyrich AS. Signal-dependent protein synthesis by activated platelets: new pathways to altered phenotype and function. Arterioscler Thromb Vasc Biol. 2008;28(3):s17-24. [14] Park E, Williams B, Wold BJ, Mortazavi A. RNA editing in the human ENCODE RNA-seq data. Genome Res. 2012;22(9):1626-33. [15] Gerber AP, Keller W. RNA editing by base deamination: more enzymes, more targets, new mysteries. Trends Biochem Sci. 2001;26(6):376-84. [16] Maas S, Kawahara Y, Tamburro KM, Nishikura K. A-to-I RNA editing and human disease. RNA Biol. 2006;3(1):1-9. [17] Nishikura K. Functions and regulation of RNA editing by ADAR deaminases. Annu Rev Biochem. 2010;79:321-49. [18] Nishikura K, Yoo C, Kim U, et al. Substrate specificity of the dsRNA unwinding/modifying activity. EMBO J. 1991;10(11):3523-32. [19] Hamilton CE, Papavasiliou FN, Rosenberg BR. Diverse functions for DNA and RNA editing in the immune system. RNA Biol.2010;7(2):220-8. [20] Seeburg PH, Higuchi M, Sprengel R. RNA editing of brain glutamate receptor channels: mechanism and physiology. Brain Res Brain Res Rev. 1998;26:217–29. [21] Brusa R, et al. Early-onset epilepsy and postnatal lethality associated with an editing-deficient GluR-B allele in mice. Science. 1995;270:1677–80. [22] Iizasa H, Nishikura K. 2009. A new function for the RNA-editing enzyme ADAR. Nat Immunol 10: 16–18. [23] Blow MJ, Grocock RJ, Van dongen S, et al. RNA editing of human microRNAs. Genome Biol. 2006;7(4):R27. [24] Paz N, Levanon EY, Amariglio N, Heimberger AB, Ram Z, Constantini S, Barbash ZS, Adamsky K, Safran M, Hirschberg A, Krupsky M, Ben-Dov I, Cazacu S, Mikkelsen T, Brodie C, Eisenberg E, Rechavi G. Altered adenosine-to-inosine RNA editing in human cancer. Genome Res. 2007;14:1586–1595. doi: 10.1101/gr.6493107. [25] Deutsch VR, Tomer A. Megakaryocyte development and platelet production. Br J Haematol. 2006;134(5):453-66.       31 [26] Rowley JW, Oler AJ, Tolley ND, Hunter BN, Low EN, Nix DA, Yost CC, Zimmerman GA, Weyrich AS. Genome-wide RNA-seq analysis of human and mouse platelet transcriptomes. Blood. 2011 Oct 6;118(14):e101-11. Epub 2011 May 19. [27] Riedmann EM, Schopoff S, Hartner JC, Jantsch MF. Specificity of ADAR-mediated RNA editing in newly identified targets. RNA. 2008;14(6):1110-8. [28] Michelson, Alan. Platelets . 2nd ed. . Burlington: Elsevier, 2007. 79-82. Print. [29] Berrou E, Adam F, Lebret M, et al. Heterogeneity of platelet functional alterations in patients with filamin A mutations. Arterioscler Thromb Vasc Biol. 2013;33(1):e11-8. [30] Tanaka H, Ishida Y, Kaneko T, Matsumoto N. Isolation of human megakaryocytes by immunomagnetic beads. Br J Haematol. 1989 Sep;73(1):18-22. [31] Muszbek L, Bagoly Z, Bereczky Z, Katona E. The involvement of blood coagulation factor XIII in fibrinolysis and thrombosis. Cardiovasc Hematol Agents Med Chem. 2008;6(3):190-205. [32] Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M and Rozen SG. Primer3--new capabilities and interfaces. Nucleic Acids Res. 2012 Aug 1;40(15):e115. [33] Rowley JW, Oler AJ, Tolley ND, et al. Genome-wide RNA-seq analysis of human and mouse platelet transcriptomes. Blood. 2011;118(14):e101-11. [34] An integrated map of genetic variation from 1,092 human genomes, McVean et Al, Nature 491, 56–65 (01 November 2012) doi:10.1038/nature11632. [35] Wang K, Li M, Hakonarson H. ANNOVAR: Functional annotation of genetic variants from next-generation sequencing data, Nucleic Acids Research, 38:e164, 2010 [36] Kasirer-friede A, Kang J, Kahner B, Ye F, Ginsberg MH, Shattil SJ. ADAP interactions with talin and kindlin promote platelet integrin αIIbβ3 activation and stable fibrinogen binding. Blood. 2014; [37] "Heartorg Home Page." American Heart Association. N.p., n.d. Web. 20 Apr. 2014. [38] Meltzer M, Long K, Nie Y, Gupta M, Yang J, Montano M. The RNA editor gene ADAR1 is induced in myoblasts by inflammatory ligands and buffers stress response. Clin Transl Sci. 2010;3(3):73-80.       32 [39] Muraki M, Ohkawara B, Hosoya T, et al. Manipulation of alternative splicing by a newly developed inhibitor of Clks. J Biol Chem. 2004;279(23):24246-54. [40] Hillman RT, Green RE, Brenner SE. An unappreciated role for RNA surveillance. Genome Biol. 2004;5(2):R8. [41] Schweinfurth N, Hohmann S, Deuschle M, Lederbogen F, Schloss P. Valproic acid and all trans retinoic acid differentially induce megakaryopoiesis and platelet-like particle formation from the megakaryoblastic cell line MEG-01. Platelets. 2010;21(8):648-57. [42] Hartwig D, Schütte C, Warnecke J, et al. The large form of ADAR 1 is responsible for enhanced hepatitis delta virus RNA editing in interferon-alpha-stimulated host cells. J Viral Hepat. 2006;13(3):150-7.