STRUCTURAL INVESTIGATION OF RNA-RNA AND RNA-PROTEIN. INTERACTIONS ...... Phase-Modulated CLEAN chemical exchange pulse sequence. COSY .... characterizing important interactions necessary for splicing reactions are therefore ..... The third question in spliceosomal assembly involved determining if p14.
THE FLORIDA STATE UNIVERSITY COLLEGE OF ARTS & SCIENCES STRUCTURAL INVESTIGATION OF RNA-RNA AND RNA-PROTEIN INTERACTIONS INVOLVING THE pre-mRNA BRANCH SITE REGION OF THE FUNCTIONAL CORE OF THE SPLICEOSOME
By KERSTEN T. SCHROEDER
A Dissertation submitted to the Department of Chemistry & Biochemistry in partial fulfillment of the requirements for the degree of Doctor of Philosophy
Degree Awarded: Semester Summer, 2006
The members of the Committee approve the Dissertation of Kersten T. Schroeder defended on June 28, 2006. ______________________________ Nancy L. Greenbaum Professor Directing Dissertation ______________________________ Debra A. Fadool Outside Committee Member ______________________________ Hong Li Committee Member ______________________________ Oliver Steinbock Committee Member
Approved: ____________________________________________ Naresh Dalal, Department Chair, Department of Chemistry & Biochemistry ____________________________________________ Joseph Travis, Dean, College of Arts & Sciences
The Office of Graduate Studies has verified and approved the above named committee members.
ii
I dedicate my dissertation to my family and friends. They have all been supportive of me through this whole process.
iii
ACKNOWLEDGEMENTS I would like to thank my family and friends for support throughout my academic career from San Pablo Elementary to Neptune Beach Elementary to the Jacksonville Beach Sixth Grade Center to Fletcher Middle School to Fletcher High School to University of Florida and at my final educational stop of The Florida State University. I would like to especially thank Dr. Nancy Greenbaum for her guidance and support throughout my graduate school career. She helped me to solve and work through problems with our many enlightening discussions. I would like to thank Dr. Hong Li, Dr. Oliver Steinbock, and Dr. Debra Fadool for serving as committee members and as mentors. I would like to give a special thank you to Dr. Timothy Logan for all the time and effort he put in helping me learn NMR. I would also like to thank Dr. Logan for our many discussions about various scientific questions. I would like to thank Dr. Joseph Vaughn, Dr. Tom Gedris and Steven Freitag for their help and support in using the NMR facilities in the Nuclear Magnetic Resonance Laboratory located in the Department of Chemistry & Biochemistry. I would like to thank Dr. Jack Skalicky and Dr. Fengli Zhang for their help and support in use of the NMR facilities at the National High Magnetic Field Laboratory. I would like to thank Hank Hendricks and Dr. Umesh Goli for their help and support in use of the facilities in the Biochemical Analysis and Synthesis Service (BASS) Laboratory in the Chemistry & Biochemistry Department. I would like to thank Dr. Ewa Bienkiewicz and Dr. Claudius Mundoma for their help and support in use of the facilities in the Physical Biochemistry Facility in the Institute of Molecular Biology. I would also like to thank Dr. Scott Showalter for his useful suggestions and many discussions. I would like to thank the past and present graduate students and technicians in the Greenbaum lab for their support and eye opening discussions
iv
about science and life. There have been numerous people in the lab that have helped me and they include, but are not limited to: Dr. Meredith Newby Lambert, LauraJane Phelps, Dr. Terrie Luong Moore, Dr. Jon Epperson, Dr. Claudius Mundoma, Janice Dodge, Alex Porges, Darui Xu, Faqing Yuan, Amy Bryant, Dr. Jörg Schlatterer, Luis Guerrero, Milena Popović, Joycelynn Nelson, Charmaine O’brien, Erika Stewart, Theodore Aquino, Christopher Staudinger, Karen Cherkis, Robert Brooks, Rebekah Welch, and Mary Pfost. I would especially like to thank Amy Bryant for her dedication, support and management of the lab. I would like to thank Luis Guerrero for his help and determination in helping make the protein soluble at higher concentrations. I would also like to thank past group members Meredith Newby-Lambert and LauraJane Phelps for their support and help in the beginning of my graduate school career. Finally, I would like to thank a special member of the group, Milena Popović, for her love and support in seeing me to the end of my difficult project.
v
TABLE OF CONTENTS
LIST OF TABLES .................................................................................. LIST OF FIGURES................................................................................. LIST OF ABREVATIONS AND SYMBOLS ........................................... ABSTRACT ..........................................................................................
ix x xiv xviii
1. INTRODUCTION...............................................................................
1
1.1 Introduction .............................................................................. 1.2 DNA and RNA .......................................................................... 1.3 RNA as catalyst........................................................................ 1.4 Proteins and RNA binding Proteins .......................................... 1.5 RNAs and proteins in the spliceosome.....................................
1 2 4 8 11
2. MATERIALS & METHODS................................................................
23
2.1 General methods for preparations of RNA strands .................. 2.2 RNA duplexes for NMR in supercooled water .......................... 2.2.1 Design and synthesis of samples..................................... 2.2.2 Sample preparation.......................................................... 2.2.3 Data collection using NMR spectroscopy......................... 2.2.4 Assignment of RNA resonance peaks by homonuclear NMR 2.3 Design of molecules in U2 snRNP studies ............................... 2.4 RNA samples in U2 snRNP studies ......................................... 2.4.1 RNA labeling with 32P radioactivity................................... 2.5 Expression and Purification of proteins p14 and SF3b155....... 2.5.1 Expression of p14 and SF3b155 proteins. ....................... 2.5.2 Identification of proteins by SDS-PAGE........................... 2.5.3 Purification of p14 proteins .............................................. 2.5.4 Purification of SF3b155 proteins ...................................... 2.5.5 Determination of monomeric proteins by DLS ................. 2.6 Gel Retardation Studies ........................................................... 2.7 Fluorescence spectroscopy...................................................... 2.7.1 Sample preparation......................................................... 2.7.2 Data collection ................................................................ 2.8 Circular Dichroism spectroscopy .............................................. 2.8.1 Sample preparation......................................................... 2.8.2 Data collection ................................................................ vi
23 23 23 24 24 25 25 26 26 26 26 27 27 29 29 30 32 32 32 34 34 34
2.9 Nitrocellulose Membrane Retention studies ............................. 2.9.1 Sample preparation......................................................... 2.9.2 Data collection for filter binding studies........................... 2.10 Isothermal Titration Calorimetry ............................................. 2.11 Nuclear Magnetic Resonance spectroscopy ......................... 2.11.1 Basics of NMR .............................................................. 2.11.2 Sample preparation....................................................... 2.11.3 Data collection .............................................................
36 36 36 37 38 38 38 38
3. STUDY of RNA DUPLEXES BY NMR IN SUPERCOOLED WATER
55
3.1 Introduction .............................................................................. 3.2 Results and Discussion ............................................................ 3.2.1 Pseudouridine monophosphate ....................................... 3.2.2 Complementary duplex .................................................... 3.2.3 Branch site duplex ........................................................... 3.2.4 Thermal coefficients......................................................... 3.2.5 Chemical Exchange ......................................................... 3.2.6 Buffer Effects ................................................................... 3.3 Conclusions..............................................................................
55 58 58 58 63 65 67 69 73
4. EQUILIBRIUM INTERACTION OF p14 WITH BRANCH SITE RNA 4.1 Introduction .............................................................................. 4.2 Results .................................................................................... 4.2.1 Analysis of p14................................................................. 4.2.2 Interaction of RNA and p14 by nitrocellulose membrane retention ................................... 4.2.3 Equilibrium studies of p14 and RNA interactions using NMR 4.2.4 Equilibrium studies of p14 with the pre-mRNA branch site duplex using circular dichroism (CD). ............................. 4.2.5 Structural studies of 15N-labeled p14 by NMR spectroscopy 4.2.6 Structural studies of 2H-13C-15N-labeled p14 by NMR ...... 4.2.7 Structural studies of 2H-13C-15N-labeled p14 with addition of branch site RNA by NMR spectroscopy...................... 4.3 Discussion and Conclusion ......................................................
74 74 76 76 81 83 88 90 91 98 101
5. INTERACTION OF SF3b155 WITH p14 AND THE BRANCH SITE RNA
103
5.1 Introduction .............................................................................. 5.2 Results of experiments............................................................. 5.2.1 Analysis of SF3b155 ........................................................
103 104 104
vii
5.2.2 Interaction of RNA, p14, and SF3b155 observed by gel .. 5.2.3 Interaction of RNA and p14 with SF3b155 by nitrocellulose membrane retention....................................................... 5.2.4 Equilibrium studies of p14, intron RNA, and SF3b155 using circular dichroism ................................................. 5.2.5 Equilibrium studies of p14 and SF3b155 interaction using fluorescence .......................................................... 5.2.6 Equilibrium studies of p14 and SF3b155 interaction using ITC 5.2.7 Structural studies of SF3b155 by NMR spectroscopy...... 5.2.8 Structural studies of SF3b155 with p14 by NMR spectroscopy 5.2.9 NMR studies of SF3b155(1-462) ..................................... 5.3 Discussion................................................................................ 5.3.1 Analysis of SF3b155 ........................................................ 5.3.2 Equilibrium studies of p14-SF3b155 binding to the branch site RNA........................................................................ 5.3.3 Equilibrium studies of p14 and SF3b155 ......................... 5.4 Conclusions.............................................................................. 6. DISCUSSION ...................................................................................
108 112 112 118 123 126 126 127 133 133 133 134 136 138
6.1 Introduction .............................................................................. 138 6.2 Summary of Results ................................................................. 139 6.3 Future Studies .......................................................................... 141 6.3.1 Future studies involving NMR in supercooled studies...... 141 6.3.2 Electrostatic studies of the pre-mRNA branch site duplex and p14 ............................................................ 142 6.3.3 Studies of SF3b155 to the full-length intron strand without p14 144 6.3.4 Mutation studies of p14 to determine amino acid residues of p14 involved in binding of the branch site duplex ........... 144 6.3.5 Mutation studies of SF3b155 to both p14 and full-length intron to determine which amino acid residues of SF3b155 are involved in binding to both p14 and the full-length intron .................... 146 6.4 Conclusions.............................................................................. 147 APPENDICES .................................................................................... A DNA and Amino Acid Sequences of proteins ………………….. B Structure and information about selected buffers .....................
149 149 158
REFERENCES ....................................................................................
163
BIOGRAPHICAL SKETCH ....................................................................
171
viii
LIST OF TABLES
Table 4.1 Buffer conditions of p14..........................................................
79
Table 4.2 Dissociation constant of p14 to RNA ......................................
82
Table 4.3 Binding affinities of p14 to various spliceosomal components
81
Table 5.1 Binding affinities of p14-SF3b155(199-462) to branch site duplex
113
ix
LIST OF FIGURES Figure 1.1 Nucleobases, nucleosides, and nucleotides .........................
3
Figure 1.2 Uridine and Pseudouridine monophosphate .........................
5
Figure 1.3 Secondary structures of naturally occurring ribozymes.........
7
Figure 1.4 Comparison of splicing mechanisms .....................................
9
Figure 1.5 Structures and symbols of the 20 amino acids......................
10
Figure 1.6 The cycle of the spliceosome ................................................
12
Figure 1.7 Splicing mechanism ..............................................................
14
Figure 1.8 Cartoon of U2 snRNP ...........................................................
15
Figure 1.9 Structure of SF3b complex and model of U11/U12 di-snRNP
19
Figure 1.10 Crystal structure and homology model of p14 ....................
20
Figure 2.1 Visualization of proteins by SDS-PAGE ...............................
28
Figure 2.2 DLS graph of lysozyme ........................................................
31
Figure 2.3 Jablonski Diagram ...............................................................
33
Figure 2.4 CD spectra of poly-L-lysine ..................................................
35
Figure 2.5 Hydrogen nuclear spin .........................................................
40
Figure 2.6 Jump-return echo scheme and pulse sequence ...................
41
Figure 2.7 NOESY scheme and pulse sequence .................................
42
x
Figure 2.8 TOCSY scheme and pulse sequence ..................................
43
Figure 2.9 COSY scheme and pulse sequence ...................................
44
Figure 2.10 HSQC scheme and pulse sequence .................................
45
Figure 2.11 HNCA scheme and pulse sequence ..................................
46
Figure 2.12 HNCO scheme and pulse sequence .................................
47
Figure 2.13 HNCACB scheme and pulse sequence ..............................
48
Figure 2.14 1H-15N-NOESY-HSQC scheme and pulse sequence ........
49
Figure 2.151H-15N-TOCSY-HSQC scheme and pulse sequence ..........
50
Figure 2.16 CBCA(CO)NH scheme and pulse sequence .....................
51
Figure 2.17 HN(CO)CA scheme and pulse sequence ...........................
52
Figure 2.18 HN(CA)CO scheme and pulse sequence ...........................
53
Figure 2.19 HCCH-TOCSY scheme and pulse sequence .....................
54
Figure 3.1 Sequences of the two RNA duplexes ...................................
57
Figure 3.2 Imino proton spectra of ψ-monophosphate .........................
59
Figure 3.3 Chemical shift versus temperature of ψMP .........................
60
Figure 3.4 Imino proton spectra of ψcomp ..............................................
62
Figure 3.5 Imino proton spectra of ψBP ................................................
64
Figure 3.6 Thermal coefficients for ψcomp and ψBP ...............................
66
Figure 3.7 The chemical exchange rate verse temperature for ψBP .....
68
Figure 3.8 HSQC spectra of RNA duplexes with 15N-labeled U and ψ ..
70
Figure 3.9 Calculations of Area/Volume for different NMR tubes ...........
72
xi
Figure 4.1 Probability of unstructured residues of p14 ..........................
78
Figure 4.2 DLS of p14 ...........................................................................
80
Figure 4.3 Representative digrams of possible secondary structures of RNA 84 Figure 4.4 Imino proton spectra of pre-mRNA with and w/o p14 ...........
86
Figure 4.5 NMR spectra of ψ-modified branch site duplex with and w/o p14
87
Figure 4.6 CD spectra of p14 with intron RNA .....................................
89
Figure 4.7 MR spectra of p14 with addition of intron RNA and SF3b155
92
Figure 4.8 NMR spectra of 2H-15N-labeled p14 .....................................
94
Figure 4.9 HSQC spectra of 2H-13C-15N-labeled p14 .............................
95
Figure 4.10 TROSY spectra of 2H-13C-15N-labeled p14 .........................
96
Figure 4.11 1H-13C-HSQC with 2H decoupling of 2H-13C-15N-labeled p14
97
Figure 4.12 1H15N-HSQC spectrum of 2H-13C-15N-labeled p14 with intron
99
Figure 4.13 1H15N-HSQC spectrum of 2H-13C-15N-labeled p14 with branch site RNA................................................................
100
Figure 5.1 Probability of unstructured residues of full-length SF3b155 .
106
Figure 5.2 Probability of unstructured residues of SF3b155(1-462) ......
107
Figure 5.3 Representative gels of p14 to RNA and SF3b155 ................
109
Figure 5.4 Representative gel of p14 to RNA, tRNA and SF3b155 .......
110
Figure 5.5 CD spectra of p14 with intron RNA and SF3b155(199-462) .
114
Figure 5.6 CD spectra of p14 with SF3b155(199-462) at far UV ...........
116
Figure 5.7 CD spectra of p14 with SF3b155(199-462) at near UV ........
117
Figure 5.8 Emission spectra of SF3b155(199-462) ...............................
120
xii
Figure 5.9 Emission spectra of SF3b155(255-462) excited at 280nm ...
121
Figure 5.10 Emission spectra of SF3b155(1-462) excited at 280nm ....
122
Figure 5.11 Representative ITC isotherm of SF3b155(199-462) to p14.
124
Figure 5.12 Representative ITC isotherm of SF3b155(1-462) to p14 ....
125
Figure 5.13 Comparison HSQC spectrum of SF3b155(255-462) and SF3b155(199-462).......................................................
128
Figure 5.14 HSQC spectrum of SF3b155(255-462) with and w/o p14 ..
129
Figure 5.15 HSQC spectrum of SF3b155(199-462) with and w/o p14 ..
130
Figure 5.16 Cropped 1H-15N-HSQC spectrum of SF3b155(199-462) .....
131
Figure 5.17 1H-15N-HSQC spectrum of 15N-labeled SF3b155(1-462) ....
132
Figure 6.1 Surface electrostatic potential maps of p14 and ψ-modified branch site duplex................................................................
143
xiii
LIST OF ABREVATIONS AND SYMBOLS ψ ψBP ψcomp ψMP µg µL µM ∆G ∆H ∆S τmix 1 H 2 H 13 C 15 N 32 P 1D 2D 2’-OH 3D 3’SS 5’SS Å A/D A/V ATP A.U. BES BPTI BO °C CD Cryo-EM D2O DEPC DNA dsRBD dsRBM dsRRM
pseudouridine, modified RNA nucleoside ψ-modified pre-mRNA branch site duplex ψ-modified complementary duplex pseudouridine monophosphate 10-6 grams 10-6 liters 10-6 molar (moles per liter) change in Gibbs free energy (∆G = ∆H - T∆S) change in enthalpy change in entropy mixing time used in an NMR experiment hydrogen, containing one proton and no neutrons deuterium (hydrogen with one neutron and one proton) carbon-13 nitrogen-15 radioactive phosphorous-32 one-dimensional two-dimensional ribose functional group at the 2’ position, nucleophile in ribozyme three-dimensional 3’-splice site 5’-splice site Å ngstrom, 10-10 meters analog/digital area divided by volume adenosine triphosphate arbitrary units N,N-Bis-(2-hydroxyethyl)-2-aminoethanesulfonic acid, buffer Bovine Pancreatic Trypsin Inhibitor static magnetic field degrees Celsius circular dichroism electron cryo-microscopy deuterium oxide diethyl pyrocarbonate deoxyribonucleic acid double stranded RNA-binding domain (proteins) double stranded RNA-binding motifs (proteins) double stranded RNA Recognition Motif (proteins) xiv
dsRNA DTT DtxR ε E. coli EDTA Far-UV GST H3BO3 HEAT HEPES HPLC hνEM hνEX Hz IC ID IPTG ITC kcal KD kDa kex LB MBq mg mL MHz mm mM MOPS mRNA Near-UV NHMFL Ni2+ NH NH2 nm NMR NOE O.D. PAGE
double-stranded RNA dithiothreitol Diphtheria toxin repressor (used as a control protein) extinction coefficient (M*cm)-1;used in Beer’s Law (A=εbc) Escherichia coli ethylene diamine tetraacetic acid far-ultraviolet, from 190-250 nm glutathione S-transferase boric acid Huntingtin, Elongation factor 3, protein phosphatase 2A, TOR1 4-(2-Hydroxyethyl) piperazine-1-ethanesulfonic acid, buffer high pressure liquid chromatography energy of the emitted photon energy of the excited photon unit that equals per second (s) intensity of cross peak intensity of diagonal peak isopropyl-β-thiogalactoside isothermal titration calorimetry kilocalories (1 kilocalorie = 4, 184 Joules) dissociation constant (a measure of binding affinity) 103 Daltons (Dalton=atomic mass unit) rate of chemical exchange Luria broth 106 Becquerel (unit of radioactivity) 10-3 grams 10-3 liters 106 Hz (Hz=per second) 10-3 meters 10-3 molar (moles per liter) 3-(N-morpholino)-propanesulfonic acid buffer messenger ribonucleic acid near-ultraviolet, from 250-320 nm National High Magnetic Field Laboratory nickel cation amide group amino group 10-9 meters nuclear magnetic resonance nuclear Overhauser enhancement outer diameter polyacrylamide gel electrophoresis
xv
pET pI pmol PMSF PP2A pre-mRNA ppb ppm psi r R:P Rh RBD rRNA RNA RNAase RNase P RNP RPM RRM S. cerevisiae SDS SF snRNA snRNP snoRNA T TCEP tRNA Tris U/µL UMP UV ν1/2 WC
DNA cloning vector isoelectric point 10-12 moles phenyl methyl sulfonyl fluoride protein phosphatase 2A precursor messenger RNA parts per billion parts per million pounds per square inch radius of a molecule RNA-protein complex hydrodynamic radius RNA binding domain (protein) ribosomal RNA ribonucleic acid ribonuclease ribonuclease P ribonucleoprotein revolutions per minute RNA recognition motif (protein) Saccharomyces cerevisiae (yeast) sodium dodecyl sulfate splicing factor small nuclear RNA, the RNA components of the spliceosome small nuclear ribonucleoprotein particle small nucleolar RNA temperature Tris(2-carboxyethyl)phosphine transfer RNA Tris (hydroxymethyl) aminomethane, buffer units/10-6 liters uridine monophosphate ultraviolent light line width at half height Watson-Crick base pair
RNA Nucleosides A C G U
Adenosine Cytidine Guanine Uridine
xvi
Twenty Common Amino Acid Residues A, Ala alanine C, Cys cysteine D, Asp aspartate or aspartic acid E, Glu glutamate or glutamic acid F, Phe phenylalanine G, Gly glycine H, His histidine I, Ile isoleucine K, Lys lysine L, Leu leucine M, Met methionine N, Asn asparagine P, Pro proline Q, Gln glutamine R, Arg arginine S, Ser serine T, Thr threonine V, Val valine W, Trp tryptophan Y, Tyr tyrosine NMR Pulse Sequences CBCA(CO)NH named for order of magnetization transfer: H-N-Cα-Cβ CLEANEX-PM Phase-Modulated CLEAN chemical exchange pulse sequence COSY Correlation spectroscopy CRINEPT cross-correlated relaxation-enhanced polarization transfer HBHA(CO)NH named for order of magnetization transfer: H-N-Hα-Hβ HNCA named for order of magnetization transfer: H-N-Cα HN(CA)CO named for order of magnetization transfer: H-N-CO HN(C)N named for order of magnetization transfer: H-N(i)-N(i-1) HNCACB named for order of magnetization transfer: H-N-Cα-Cβ HNCO named for order of magnetization transfer: H-N-CO HN(CO)CA named for order of magnetization transfer: H-N-Cα HN(CO)CACB named for order of magnetization transfer: H-N-Cα-Cβ HSQC Heteronuclear Single Quantum Coherence INEPT Insensitive Nuclei Enhanced by Polarization transfer 15 NH-NOESY N-edited NOESY 15 NH-TOCSY N-edited TOCSY NOESY Nuclear Overhauser Enhancement spectroscopy TOCSY Total Correlated spectroscopy TROSY Transverse Relaxation Optimized spectroscopy
xvii
ABSTRACT The removal of introns from precursor-messenger (pre-m)RNA in eukaryotes is mediated by the spliceosome, a dynamic supramolecular complex comprising five small nuclear (sn)RNAs and numerous proteins, organized into small ribonucleoprotein (snRNP) particles. At the functional center of the spliceosome is the pairing of a highly conserved region of the RNA intron and the U2 snRNA, called the pre-mRNA branch site duplex, and two U2 snRNP proteins, p14 and SF3b155. The goal of the research described in this dissertation was to characterize the interaction between p14 and the branch site RNA, and determine whether SF3b155 impacts on that interaction. In order to refine the structure of the pairing between the intron and pseudouridine (ψ)-modified U2 snRNA (M.I. Newby and N.L. Greenbaum, 2001 RNA 7; 833-845), Nuclear Magnetic Resonance (NMR) experiments were performed in supercooled water in order to decrease the temperature-dependent exchange of protons in RNA duplexes. NMR spectra of aqueous samples of RNA in bundles of narrow capillaries acquired at temperatures as low as -18 °C reveal resonances of exchangeable protons not seen at higher temperatures. In particular, we detected the imino protons of terminal base pairs and the imino proton of a non-base paired ψ in the pre-mRNA branch site helix. Analysis of the temperature dependence of chemical shift changes (thermal coefficients) for imino protons corroborated hydrogen bonding patterns observed in the NMR derived structural model of the branch site helix. The ability to observe non-base paired imino protons of RNA loop and bulge regions is of significant value in refining the structure of RNA motifs containing non-base paired regions. Also, detection of the imino proton of the non-base paired ψ will help identify this resonance when p14 is added. As measures of affinity between p14 and branch site RNA, membrane filtration and gel electrophoresis methods indicated that the U2 snRNP protein p14 binds a single-stranded RNA representing the pre-mRNA intron in solution; however, affinity is xviii
significantly greater when the intron is paired with a U2 snRNA strand modified by ψ in its phylogenetically conserved location opposite the branch site residue. The binding between p14 and its cognate RNA is enhanced further upon addition of a segment of the neighboring protein SF3b155 that contacts p14 but not the RNA. NMR spectra reveal that p14 undergoes significant changes upon complex formation with both the pre-mRNA branch site duplex and SF3b155, in contrast, the RNA does not undergo any marked change, and changes in SF3b155 are limited to specific regions. These data imply that the three components (pre-mRNA branch site duplex, p14, and SF3b155) interact with each other under equilibrium conditions, and that SF3b155 facilitates the p14-RNA interaction. These results demonstrate, for the first time, specific recognition of a double-stranded RNA motif by an RNA Recognition Motif (RRM) protein, and suggest a cooperative mechanism for assembly of the pre-mRNA, U2 snRNA, p14 and SF3b155.
xix
CHAPTER 1 INTRODUCTION
DNA
RNA
Protein
1.1 Introduction Gene expression and its regulation are essential processes for all living organisms. Transcription of deoxyribonucleic acid (DNA) into messenger ribonucleic acid (mRNA) and translation of mRNA into proteins are integral steps in gene expression. Before mRNA is translated into proteins, introns (regions of mRNA that do not code for proteins) are removed and the flanking exons (protein coding regions) are ligated in a process analogous to the editing of movie film. This process is called precursor (pre)-mRNA splicing and occurs in the nucleus of the cell. Understanding the splicing mechanism can help one determine how defects in pre-mRNA splicing lead to certain diseases, including several cancers, heart diseases, cardiomyopathies and neuropathies. Research aimed at characterizing important interactions necessary for splicing reactions are therefore vital to the medical endeavor to find cures. The goal of this dissertation research is to understand the molecular basis for recognition associated with the assembly of several components involved in the splicing mechanism.
1
1.2 DNA and RNA In 1869, DNA was discovered by Johann Friedrich Meischer when he
found that nucleic acids, or long-chain polymers of nucleotides, from the nuclei of white blood cells were made up of sugar, phosphoric acid, and several
nitrogenous bases (Meischer, 1871; Dahm, 2005). Later, it was determined that the sugar in nucleic acid can be either a ribose (as in RNA) or deoxyribose (as in DNA). RNA is a nucleic acid that is composed of a phosphate and ribose sugar backbone. A fundamental difference between RNA and DNA is that RNA has a ribose sugar ring and its 2’-hydroxyl (-OH) is an important chemical and structural feature of RNA, whereas the sugar for DNA is a deoxyribose, which has a hydrogen in place of the 2’-OH. The four RNA nitrogenous bases are called: adenine, guanine, cytosine and uracil (Figure 1.1). Nucleotides (base+sugar+phosphate) form polymers, known as polynucleotides, which can form higher order structure and may control cellular processes. The first three-dimensional structure of DNA was the double helix model, derived from fiber diffraction data (Watson & Crick, 1953). Until the last few decades, very little was known about the structural characteristics of RNA; the first RNA structure determined was of transfer (t)RNA in 1974 (Kim et al., 1974). Nucleic acids are essential for transmission, expression and conservation of genetic information. In 1944, Oswald T. Avery showed that DNA was the chemical basis for specific transmission of heritable information and not proteins, as once believed (Avery, 1944). In a 1957 address, Francis Crick postulated that information was transmitted from DNA and RNA to proteins (Crick, 1958). Since the address by Crick, the role of DNA as the genetic template has been abundantly demonstrated. However, only in the second part of the 20th century has the versatility of RNA emerged. It was first thought that RNA served solely as an intermediary molecule in genetic expression. It was subsequently found that tRNA serves as the adaptor molecule in translation of mRNA into proteins, and that the major component by mass is ribosomal RNA (rRNA) (although it was not known until much later that the RNA component, and not the protein, performed
2
Nucleobase
Nucleoside
Nucleotide
Adenosine A
Adenosine monophosphate AMP
Guanosine G
Guanosine monophosphate GMP
Uridine U
Uridine monophosphate UMP
Cytidine C
Cytidine monophosphate CMP
Adenine
Guanine
Uracil
Cytosine
Figure 1.1 Nucleobases, nucleosides, and nucleotides. Nucleobases consist of heterocyclic aromatic rings with nitrogenous (N) atoms. A heterocyclic, aromatic, 6-membered ring (1,3-N substituted) is called a pyrimidine. A bicyclic organic compound with 4 nitrogen atoms at positions 1, 3, 7 and 9 (7H-imidazol[4,5d]pyrimidin) is called a purine. Nucleosides consist of a nucleobase covalently bonded to a carbon of a ribose, a five carbon sugar called pentose. Nucleotides consist of a heterocyclic nucleobase, a pentose sugar, and a phosphate or a polyphosphate group.
3
the chemistry of making proteins). Today, it is known that RNA molecules can be catalysts and serve as regulatory molecules in many biological processes. Among the most relevant in this dissertation are small nuclear (sn)RNAs, which are short RNA transcripts that associate with proteins to form small nuclear ribonucleoprotein particles (snRNPs) and participate in RNA processing. There are a number of post-transcriptional modifications to RNA transcripts. Enzymatically catalyzed post-transcriptional modifications can be made to RNA include capping of the 5’ end with a 7-methyl-guanosine (added immediately to mRNA after the start of transcription) and polyadenylation of the 3’ end of the mRNA. Post-transcriptional base modifications to individual nucleotides include methylation of RNA bases and pseudouridylation of uridine to pseudouridine (ψ). Pseudouridine (Figure 1.2) is a rotational isomer of uridine (Figure 1.2) and is attached to its ribose through carbon atom number five. Each of these post-transcriptional modifications in RNA have different functional and structural roles in the cell.
1.3 RNA as a catalyst In 1977, Richard Roberts, working at the Cold Spring Harbor Laboratory, and Phillip Sharp, working at the Massachusetts Institute of Technology found that an individual gene contains several DNA segments separated by extraneous (non-coding) DNA called introns (or discontinuous genetic information) (Berget et al., 1977; Chow et al., 1977; Gelinas et al., 1977; Gelinas & Roberts, 1977). Roberts and Sharp received the 1993 Nobel Prize in Physiology or Medicine for discovering introns and the mechanisms of gene-splicing. The catalytic ability of RNA was discovered in 1982 by Thomas Cech’s group from the University of Colorado. They found that rRNA in the ciliate protozoan Tetrahymena thermophila could cleave and splice itself without the use of external proteins or energy source (Kruger et al., 1982). Shortly afterwards, Sidney Altman's group at Yale University independently found catalytic RNA in an enzyme called ribonuclease P (RNase P) in the bacteria Escherichia coli (Reed et al., 1982). It was a revolutionary finding that RNA
4
ψMP
UMP
Figure 1.2 Uridine and pseudouridine monophosphate. Uridine (U) is a nucleobase base that is attached to its ribose by nitrogen atom number 1 (made with MDL ISIS/Draw 2.5 software, San Leandro, CA). Pseudouridine (ψ) is a rotational isomer of uridine and is attached to its ribose by carbon atom number 5.
5
could act as an enzyme, now referred to as a ribozyme, resulting in Cech and Altman sharing the 1989 Nobel Prize in Chemistry. More ribozymes have been discovered since Cech and Altman published their first papers. Another four small ribozymes discovered were the hammerhead (Forster & Symons, 1987), hairpin (Buzayan et al., 1986), hepatitis delta virus (Kuo et al., 1988), and Neurospora VS (Saville & Collins, 1990) ribozymes (Figure 1.3). Each small ribozyme can cleave the phosphodiester backbone of RNA in order to free a 5’-OH and form a 2’-3’-cyclic phosphodiester bond, which then spontaneously generates either a 2’ or 3’ phosphate. These leaving groups are important because unlike cleavage of the backbone through hydrolysis, these ribozymes cleave in a reversible fashion, and are able to ligate the products in order to form another phosphodiester bond. Two large ribozymes have been discovered called group I self-splicing introns (Tanner & Cech, 1985) and group II self-splicing introns (Peebles et al., 1986; van der Veen et al., 1986). Group I introns from within the intervening sequences of rRNA have been discovered in the nuclear genome of Tetrahymena thermophila; in the mitochondrial genome of certain fungi, yeast, and some chloroplast genomes; and in the nuclear genome of Physarum polycephalum. Group II introns from within the intervening sequences of mRNA have been discovered in the mitochondrial genome of yeast, fungi and some higher eukaryotes in addition to some chloroplast genomes. The splicing mechanisms for group I and group II introns are not identical; in fact, cleavage is facilitated through two very different types of reactions. The group I intron undergoes trans-splicing where the nucleophile for the first splicing reaction comes from the 3’-OH of an exogenous (outside of system) guanosine. The group II intron carries out cis-splicing where the nucleophile comes from a 2’-OH of an endogenous (same RNA strand) adenosine within the intron being removed and the exons being ligated together (Figure 1.4). When group II introns were discovered, it was observed that the products of splicing were similar to those of pre-mRNA splicing in the spliceosome and later determined to have the same splicing mechanism (explained in section 1.5)(Figure 1.4).
6
(plant viral satellite RNAs)
(satellite RNA of tobacco ringspot virus)
(hepatitis delta virus)
Figure 1.3 Secondary structures of naturally occurring ribozymes. The green shows the ribozyme region and the black shows the intron region substrate or exon region (Takagi et al., 2001). These ribozymes are important because it shows that RNA can act as an enzyme and catalyze its own reactions without the aid of proteins.
7
1.4 Proteins and RNA binding Proteins The first proteins discovered were egg albumin, serum albumin, and serum fibrin by Gerardus Johannes Mulder in 1839 (Mulder, 1839), but Jöns Jakob Berzelius is credited with coining the phrase “proteios”, now called proteins (Oesper, 1966). Proteins are biological molecules translated from mature mRNA. There are 20 different amino acids (Figure 1.5) that can be arranged into polymers (polypeptides) of up to thousands of amino acids long, and their potential for variety is extraordinary. The diversity of proteins allows for various functions, such as specific enzymes that compose a cell's metabolism. For example, Escherichia coli bacterium, one of the simplest biological organisms, has over a thousand different proteins working at various times to catalyze the necessary biological reactions to sustain life. Myoglobin and hemoglobin were the first three-dimensional protein structures solved by X-ray crystallography by Max Perutz and John Cowdery Kendrew in 1957 (Kendrew & Perutz, 1957), which led to their receiving a Nobel Prize in Chemistry in 1962. Proteinase IIa inhibitor from bull seminal plasma was the first three-dimensional protein structure determined by NMR spectroscopy, reported by Kurt Wüthrich and co-workers in 1985 (Williamson et al., 1985). Currently, there have been more than 33,000 protein structures solved, approximately 460 RNA structures solved, and approximately 1475 proteinnucleic acid structures solved according to http://www.pdb.org. The study of both RNAs and proteins has led to the discovery of many interesting protein domains and families. In particular, proteins that contain RNA Recognition Motifs (RRMs), also known as the RNA binding domains (RBDs) of ribonucleoprotein domains (RNPs), have been discovered to bind RNA and, in some instances, other proteins. Proteins that contain an RRM have a consensus region of 85-90 amino acid residues with a globular domain consisting of two αhelices and four β-strands with an order of β1-α1-β2-β3-α2-β4 where a conserved hexameric region (called RNP2) is located in β1 strand and a conserved octameric region (called RNP1) is located in β3 strand (Nagai et al., 1990). It has been reported that proteins containing an RRM only bind single-stranded RNA
8
Figure 1.4 Comparison of splicing mechanisms. The image above shows a cartoon of splicing of the intron of the pre-mRNA intron, group I self-splicing intron, and group II self-splicing intron with a comparison of the first and second transfer reactions. This is important because it shows the nucleophile for nuclear pre-mRNA and group II are from within the same sequence and shows analogous splicing. However, the group I intron shows splicing begins with a 3’-OH of an exogenous guanosine.
9
Alanine (Ala/A)
Cysteine (Cys/C)
Histidine (His/H)
Methionine (Met/M)
Threonine (Thr/T)
Arginine (Arg/R)
Asparagine (Asn/N) Aspartic acid (Asp/D)
Glutamic Acid (Glu/E) Glutamine (Gln/Q)
Isoleucine (Ile/I)
Phenylalanine (Phe/F)
Tryptophan (Trp/W)
Glycine (Gly/G)
Leucine (Leu/L)
Lysine (Lys/K)
Proline (Pro/P)
Serine (Ser/S)
Tyrosine (Tyr/Y)
Valine (Val/V)
Figure 1.5 Structures and symbols for the 20 amino acids. Each can be encoded for in protein synthesis by the standard genetic code.
10
(Birney et al., 1993; Hall, 2002). The reported binding constants of RRM proteins reflect relatively low affinity. However, the binding is usually fairly specific (Burd & Dreyfuss, 1994; Hall, 2002). Proteins containing double stranded RNA-binding domains (dsRBD), which consist of double stranded RNA-binding motifs (dsRBMs), interact with double-stranded (ds)RNA (St Johnston et al., 1992). These dsRBD proteins contain two tandem dsRBMs, where each dsRBM has a conserved stretch of 6575 amino acids with a structural fold of α1-β1-β2-β3-α2 (St Johnston et al., 1992). It has been observed that many dsRBM proteins interact with RNA strands that contain non-Watson-Crick structural features such as bulges and loops (Zheng & Bevilacqua, 2000). RRMs and dsRRMs have each been discovered in a number of spliceosomal proteins. Furthermore, there are some spliceosomal proteins that contain both RRMs and dsRRMS as part of their structural folds. One protein, p14, studied in this dissertation contains a sequence consistent with that of a single RRM, but evidence shown here supports its specific recognition of a double-stranded RNA. This would suggest that p14 would contain a dsRRM, but analysis of the amino acid sequence of p14 is consistent with that of a single RRM. Therefore, it appears that this protein behaves like a dsRRM, but has an amino acid sequence typical of an RRM.
1.5 RNAs and proteins in the spliceosome Five small nuclear RNAs (U1, U2, U4, U5 and U6 snRNAs) and ~100 proteins comprise the spliceosome, a supramolecular RNA-protein assembly (Figure 1.6) (Konarska & Sharp, 1987; Michaud & Reed, 1991; Jamison et al., 1992; Burge, 1999). Of these components, U2 snRNA and U6 snRNA contain the most phylogenetically conserved sequences and are the only two snRNAs believed to be involved in catalysis of splicing of pre-mRNA (Burge, 1999). Assemblies of snRNAs and protein components that surround the pre-mRNA are called small nuclear ribonucleoprotein particles (snRNPs) (Lerner & Steitz, 1979). U1, U2 snRNPs and the U4/U5/U6 tri-snRNP make up the five snRNAs and
11
Figure 1.6 The cyclic nature of spliceosome assembly and activity. Schematic representation of the spliceosome cycle in terms of the role of snRNPs in the premRNA splicing process. Pre-mRNA (at the top of the schematic), containing two exons separated by an intron, enters splicing complexes with snRNPs and exits as mRNA (bottom left) and excised lariat intron (far left center). The individual snRNPs are U1, U2, U4, U5, and U6 and are represented in the CC, A, Bl, B2, Cl, C2, and I complexes within the splicing pathway that have been distinguished biochemically and/or genetically. 5’ SS, 3’ SS, bs, and Py correspond to the 5’ and 3’ splice sites, the branch site, and the polypyrimidine tract, respectively. [edited from Molecular Biology 2nd (Weaver, 2002)].
12
multi-protein components involved in the dynamic assembly of the spliceosome. Assembly of the competent spliceosome involves highly specific and ordered RNA-RNA and RNA-protein interactions around the pre-mRNA substrate in order for the splicing mechanism to be carried out. The first step of splicing requires the presence of the U2 snRNP, among other spliceosomal components. It is characterized by a nucleophilic attack of the 2’-OH of a specific adenosine(A) of the intron at the 5’ splice site, located directly 3’ of exon 1 (Figure 1.7) (Moore & Sharp, 1992, 1993). It was also determined that the second step of splicing occurs when the 3’-OH terminus of exon 1 (that on the 5’ end) attacks the phosphodiester bond between the 3’ end of the intron and exon 2 (which is on the 3’-end) (Moore & Sharp, 1992, 1993)(Figure 1.4). Exons 1 and 2 are ligated together and the intron is released in lariat form along with the U2 snRNA, U5 snRNA and U6 snRNA (Chiara & Reed, 1995). The adenosine at the center of the lariat structure forms phosphodiester linkages through its 2’ and 3’ hydroxyl groups; it is because of this branched product that this region has been called the “branch site”. This finding was ground-breaking and especially important because it had previously been thought that the catalytic activity of intronic splicing in the spliceosome was carried out by the surrounding proteins (Lerner et al., 1980). Now, there is evidence that the U2 and U6 snRNAs can achieve pre-mRNA splicing without the presence of proteins (Valadkhan & Manley, 2001, 2003), suggesting that the spliceosome is a ribozyme. It is important to note that during splicing, the premRNA undergoes two SN2 type reactions, where both steps are transesterification reactions. These transesterification reactions are characterized by the use of two different hydroxyl groups from RNA nucleotides. The loss of these hydroxyl groups or the splice regions where the hydroxyl groups attack prevents splicing, protein synthesis, or even more deleterious, result in synthesis of a non-functional protein that can cause diseases or certain cancers. A cartoon rendition of the U2 snRNP (Figure 1.8) shows the pairing of the intron-U2 snRNA and the unpaired branch site A (Ruskin et al., 1984) and the presumed positions of p14 and SF3b155 near the branch site adenosine.
13
Figure 1.7 Two schematic views of the splicing mechanism. The figure on the right shows the first step of splicing, which is initiated by the 2’-OH of the branch site adenosine acting as the nucleophile that attacks the phosphate group at the 5’ splice site and forms a lariat structure. The second step of splicing is initiated by the free 3’-OH of the 5’ splice site attacking at the 3’ splice site. The figure on the left shows the mechanism of both steps of splicing with the stereochemistry of each step shown beside it.
14
5’-splice site
branch site region polypyrimidine (Y) tract
3’-splice site
Figure 1.8 A cartoon rendition representation of the U2 snRNP (top) showing the pairing of the intron-U2 snRNA and the proposed location of p14 and SF3b155 binding to the pre-mRNA(adapted from (Will et al., 2001). A representation of the pre-mRNA branch site region indicating the seven base pairs between the intron RNA and the U2 snRNA at the functional core of the spliceosome (bottom). The pre-mRNA region between the 5’-exon (exon 1) and 3’-exon (exon 2) consists of the 5’-splice site, the nucleotides between the 5’splice site and the branch site region, the branch site region, the polypyrimidine tract, and the 3’-splice site.
15
It has recently been shown that a highly conserved pseudouridine (ψ) residue in the U2 snRNA opposite of the adenosine that is adjacent to the branch site A is an integral component in pre-mRNA branch site structure because only in the presence of ψ is the A in an extrahelical position (Newby & Greenbaum, 2001, 2002b, a). In addition to providing the nucleophile for the first step of splicing, recognition of the pre-mRNA helix is thought to be involved in the assembly of the U2 snRNP (see ahead). The U2 snRNP protein p14 has been shown to cross-link to the branch site A, observed in the Sharp laboratory (MacMillan et al., 1994). Another U2 snRNP protein SF3b155 has been shown to cross-link to p14 and has been depicted surrounding p14 (Will et al., 2001; Golas et al., 2003, 2005). In addition to their role in chemistry, specific features of RNA are involved in assembly of the U2 snRNP. The first step begins when the single stranded intron is identified by the branch point binding protein (Abovich & Rosbash, 1997; Berglund et al., 1997), this protein is then displaced after pairing of the intron with the U2 snRNA. Several specific and non-specific snRNP proteins have been shown to form UV-induced cross-links to single-stranded regions surrounding the branch site and/or to other proteins in the region. One of the U2 snRNP proteins is U2AF65 (called Mud2p in yeast)(Abovich et al., 1994), which binds to the polypyrimidine tract downstream of the branch site A. It has also been shown that subunits of splicing factors SF3a and SF3b associate with a 20-nucleotide stretch upstream of the branch site A (Kramer et al., 1999). A component of SF3b called SF3b155 (~155 kDa protein) has been shown to form UV-induced cross-links to single stranded RNA bases of the intron positioned +5 and –6 from the branch site adenosine (Will et al., 2001). Also, SF3b155 forms UV-induced cross-links to the intron immediately adjacent to U2AF, suggesting SF3b155 involvement in recruiting U2 snRNP to the branch site (Gozani et al., 1998). Another smaller component of the SF3b, p14, has been shown to form a covalent UV-induced cross-link through a linker molecule tethered to the branch site A in the context of an early spliceosomal assembly called complex A, and persists through at least the first step of splicing (MacMillan et al., 1994; Will et
16
al., 2001). Phylogenetic studies have shown that p14 is highly conserved among species (Will et al., 2001) and knockout studies of the yeast p14 analog, Snu17p, finds that the protein is not essential for survival. However, the yeast cell devoid of Snu17p has hindered growth and has a defect in pre-mRNA splicing (Gottschalk et al., 2001). The co-precipitation of Snu17p with pre-mRNA suggests that it is involved in the first step of splicing (Gottschalk et al., 2001). This suggests that specific recognition occurs between p14 and the branch site A, but a direct interaction in aqueous solution between p14 and the branch site RNA duplex in the absence of other proteins has not been shown. The spliceosomal protein p14 is a 14.6 kDa protein that is one of many proteins comprising the SF3b component of the U2 snRNP complex. The p14 contains 125 amino acid residues and is known to contain an RRM. The amino acid sequence of p14 is shown below with the underlined amino acid residues representing the RRM region: MGMQAAKRAN IRLPPEVNRI LYIRNLPYKI TAEEMYDIFG KYGPIRQIRV GNTPETRGTA YVVYEDIFDA KNACDHLSGF NVCNRYLVVL YYNANRAFQK MDTKKKEEQL KLLKEKYGIN TDPPK. The persistence of the p14 interaction through at least the first step of splicing suggests that p14 may participate in the assembly of spliceosomal proteins and/or a more direct functional role in the splicing process. Therefore, direct recognition of the pre-mRNA branch site region by p14 could be used in recruiting other elements in assembly of the U2 snRNP, such as SF3b155. The role of SF3b155 may be to recruit other proteins to the branch site region or to help stabilize the RNA-protein interactions in the branch site region. The Query laboratory has shown through glutathione S-transferase (GST)tagged GST-pull down experiments and co-precipitation that SF3b155 forms a stable complex with p14 (Will et al., 2001). Studies with fragments of SF3b155 lacking the amino-terminus and 22 tandem repeats that specific segments in SF3b155 interact with p14. It is believed that the regions containing RWDETP and TPGH repeats of SF3b155 have a role in binding to p14. Electron cryomicroscopy (cryo-EM) of a ~450 kDa SF3b complex displays a structure at ~10 Å resolution (Golas et al., 2003) where p14 is located at the catalytic core of the 17
complex (Figure1.9). The structure also shows that protein SF3b155 has its 22 tandem helical repeats in the outer shell of the complex surrounding p14. This structure shows the RRM of p14 in the catalytic core, which implicates a recognition role for p14 in the first step of splicing in the spliceosome. A structure determined by electron cryomicroscopy (cryo-EM) of the U11/U12 snRNP of the atac spliceosome (Figure 1.9) (Golas et al., 2005) may indicate a groove formed by the SF3b complex suggesting interaction between SF3b155, p14 and the premRNA (Golas et al., 2003, 2005). However, direct recognition between the protein and RNA has not been shown. A report of a crystal structure of p14 (Schellenberg et al., 2006) stated that the binding interface between p14 and the branch site region is on the parallel β1β3 sheet. In separate experiments, these authors showed that Y22, which is part of the β1-strand, formed a cross-link to the branch site A of a single-stranded intron strand (Schellenberg et al., 2006) (Figure 1.10). A homology model of p14 using some NMR chemical shifts (Spadaccini et al., 2006) predicted that the β3', β3" hairpin of p14 interacts with the branch site region, which is in direct conflict with the crystal structure of p14 (Figure 1.10). These two structures yield yet another view of the binding interface between pre-mRNA and the U2 snRNPassociated proteins. Also, predictions that the canonical RNA binding surface, one of the β sheets, appears to be the surface that interfaces with the p14binding domain of SF3b155 (Schellenberg et al., 2006; Spadaccini et al., 2006), suggesting that its contacts with RNA must be somewhere else. However, this information is partially contradicted by evidence from Schellenberg et al. (2006) that shows that contacts with the branch site A occurs via Y22, which is part of the β1 strand. These two different views of p14 predict where the pre-mRNA is located, but neither one show a solution structure of p14 that includes the premRNA. However, these data suggest that one thing is certain about p14; p14 is not a typical RRM protein. The studies presented in this dissertation investigate recognition/ interaction among spliceosomal components. First, I have helped refine the structure of the ψ-dependent branch site helix by employing a new technique to
18
Figure 1.9 (left) Cryo-EM structure of a ~450kDa SF3b complex at ~10 Å resolution (Golas et al., 2003). In yellow is p14 at the center of the SF3b complex, in rainbow colors is SF3b155 surrounding p14 and in green is SF3b49 showing it on the periphery of the complex. (right) 3D model of the postulated path of the pre-mRNA on the surface of the U11/U12 di-snRNP (Golas et al., 2005). The model indicates an interpretation of the cryo-EM structure of the U11/U12 di-snRNP complex using previously published data. Comparison of the two cryo-EM structures shows two different views of where p14 can be located during spliceosomal assembly.
19
SF3b155 (373-415)
Regions predicted to bind pre-mRNA RNP2 contains β1-strand which includes Y22
RNP1 contains β3-strand
Figure 1.10 (left) Crystal structure of a spliceosomal protein p14 (Schellenberg et al., 2006). The 2.5 Å crystal structure of the p14·SF3b155 complex has a structural fold of β1-α1-β2-β3-α2-β4-α3-α4 with the C-terminal helices of p14 and the SF3b155 peptide shown in yellow. RNP1 is shown in red and RNP2 is shown in blue. (right) Homology structure of a spliceosomal protein p14. This model of p14 is based on limited NMR chemical shifts using the U1A RRM protein for sequence alignment (Spadaccini et al., 2006) and using a threading program. Comparison of the two structural models shows two different locations for the binding of p14 to the pre-mRNA. The crystal structure on the left predicts that binding should be in the RNP1 on the β1 strand. However, the β1-β3 sheets in the homology model of p14 is not near the β3', β3" hairpin of p14 which is hypothesized to interact with the pre-mRNA.
20
RNA by using NMR spectroscopy in supercooled water to achieve temperatures well below zero degrees. I chose to study the pre-mRNA branch site helix of the yeast spliceosome because the presence of a conserved ψ induces a strikingly different structure than observed with uridine in the same location in the branch site helix (Newby & Greenbaum, 2001, 2002b). Structural models of the branch site duplex suggested that ψ did not appear to form a base pair with the opposing adenine (A23, adjacent to the bulged base, A24), and no resonance attributable to ψ N3H was visible at ambient temperatures. Identification of this resonance would contribute important information about the environment of this imino proton of pseudouridine in the branch site helix with respect to surrounding bases. Therefore, applying an innovative technique using NMR spectroscopy in supercooled water, we were able to study the once invisible ψ N3H in the branch site duplex by going to temperatures well below zero and studying the behavior of this pseudouridine imino proton. The second set of studies addresses whether the branch site duplex interacts with p14 in solution and whether p14 binds to either the single-stranded pre-mRNA intron or the double-stranded U2 snRNA-intron duplex. We also tested whether the pseudouridine-modified U2 snRNA-intron duplex as compared to the unmodified U2 snRNA-intron duplex encourages higher affinity to p14 in solution. To test these interactions with p14 along with the addition of SF3b155, we have used a variety of techniques ranging from gel retardation experiments, nitrocellulose membrane retention experiments, circular dichroism (CD) spectroscopy, and NMR spectroscopy. I chose to study the interaction of p14 and the branch site region because I wanted to understand what role p14 has in the branch site region. The third question in spliceosomal assembly involved determining if p14 and SF3b155 interact with each other in solution, and if so, characterizing the binding affinity between these two dynamic proteins at the functional core of the spliceosome. We used CD spectroscopy, fluorescence, NMR spectroscopy and isothermal titration calorimetry (ITC) to determine if an interaction occurred
21
between p14 and SF3b155, and ITC was utilized to determine the binding affinity between the two spliceosomal proteins. Questions addressed in these experiments include which of the biomolecules interact with each other and whether there is a change in conformation upon interaction. Preliminary data suggest only minor changes in RNA structure, but major structural changes in p14, when these molecules interact in solution. These data and other biochemical data are consistent with a direct interaction between the double-stranded RNA of the branch site region and p14. We have demonstrated that p14 and SF3b155 bind weakly to each other in solution. I was also able to demonstrate that SF3b155 does not bind to the short intron strand alone, but collaborating with others in the laboratory, we were able to show that SF3b155 can bind to the full-length intron without p14. I will present a model of the RNA-RNA and RNA-protein interactions and propose that p14 fills the widened major groove of the pseudouridine-modified branch site helix. We will suggest that SF3b155 interacts directly with p14 and surrounds p14 by loosely anchoring itself to the pre-mRNA intron to the region between the 5’splice site and the branch site region, and also to the region called the polypyrimidine tract, which resides between the branch site region and the 3’splice site.
22
CHAPTER 2 MATERIALS & METHODS
2.1 General Methods for preparations of RNA strands for biochemical studies. First, all solutions that will be used with RNA must be treated in order to be free of Ribonuclease (RNase), an enzyme which degrades RNA. In order to make solutions RNase-free, we treated the water with diethyl pyrocarbonate (DEPC) overnight and then autoclaved the water to degrade the organic compound. RNA oligomers used for biochemical and NMR studies reported in this dissertation were synthesized commercially by Dharmacon, Inc (Lafayette, CO) and designed to represent biological sequences contained in the spliceosome. The synthesized RNA molecules were deprotected using company protocol and analyzed by gel and were usually ready as received. If needed, the RNA samples were purified by gel electrophoresis, extracted from the gel using standard protocols and then washed by ethanol precipitation. An exception to using synthesized RNAs were experiments that required longer RNA oligmers. In this case, longer RNA oligomers were prepared by standard in vitro transcription techniques.
2.2 RNA duplexes for NMR in supercooled water 2.2.1 Design and synthesis of sa mples The three solutions of RNA we used were the pseudouridine monophosphate (ψMP), a complementary RNA duplex (ψcomp, 5’-GGUGψAGUA3’ vs 5’-UACUACACC-3’; shown in Figure 1, top), and an RNA duplex with a bulged adenosine, representing a minimal form of the pre-mRNA branch site from Saccharomyces cerevisiae (ψBP, 5’-GGUGψAGUA-3’ vs 5’UACUAACACC-3’; bulged A is underlined; Figure 1, bottom)(Newby & Greenbaum, 2001). Pseudouridine monophosphate was purchased from Berry & Associates (Dexter, MI, #PYA11080). RNA oligomers were purchased from
23
Dharmacon Research (Boulder, CO) and deprotected according to company protocol (Wincott et al., 1995). Samples were precipitated with ethanol, partially desalted by three washes of the pelleted RNA with 80% ethanol, and lyophilized to dryness. The RNA pellets were then resuspended in DEPC-treated water, and strand concentrations were calculated from the absorbance at 260 nm. Equimolar amounts of the strands were combined to obtain duplexes of the ψcomp and ψBP and verified by mobility on a non-denaturing gel. The duplexes were lyophilized to dryness and resuspended in 250 µL of NMR buffer consisting of 10 mM sodium phosphate, pH 6.4, 50 mM sodium chloride, and 0.1 mM ethylenediaminetetraacetate (EDTA) in 90%H20/ 10% D2O (99.96%; Cambridge Isotope Laboratories, Cambridge, MA). Final concentration of RNA strands in NMR samples was approximately ~2.7 mM for ψcomp and ~2.5 mM for ψBP. NMR assignments of imino protons of ψBP were reported previously (Newby & Greenbaum, 2001, 2002a, b).
2.2.2 Sample Preparation Open-ended glass capillaries of 1.0 mm O.D. were purchased from Fisher Scientific (Hampton, NH, #34500-99). The capillaries were prepared by soaking in 100% methanol overnight and then dried at 170 °C. Solutions of RNA were centrifuged at 14,000 rpm for 30 minutes in order to remove particles that could nucleate ice crystals. The samples were then taken up by capillary action with approximately 27 µL in each capillary. The ends of the open capillaries were flamed sealed and the capillaries were placed in a 5 mm NMR tube (Poppe & van Halbeek, 1994; Skalicky et al., 2000; Skalicky et al., 2001). Bundles of 9-10 capillaries have a filling factor of 30-40% to yield an effective concentration of ~1 mM for the two RNA duplexes; ψMP, with a starting concentration of 10 mM in six capillaries, had an effective concentration of ~2.5 mM.
2.2.3 Data Collection using NMR spectroscopy NMR spectra were acquired on a 720 MHz and 600 MHz Varian Unity Plus spectrometers (National High Magnetic Field Laboratory, Tallahassee, FL). 24
Samples were cooled at a rate of -1.5 to -2.5 °C/hour with an increase in rate as the temperature decreased. The jump-return echo pulse sequence (Sklenar & Bax, 1987) was used to array the temperature decrease in order to observe the imino protons at different temperatures and freezing of one or more capillaries was monitored by a proportional decrease in signal intensity; however, the breaking of capillary tubes was not observed. We used 64 steady state scans prior to each acquisition. In order to improve signal quality at lower temperatures, we doubled the number of acquisitions for each incremental temperature decrease of 2.5 °C: 128 at -5 °C, 512 scans for -10 °C, and up to 4096 scans at -17.5 °C. Processing of NMR data and simulation line-widths at half height were accomplished using Felix 2.3 (Biosyn).
2.2.4 Assignment of RNA resonance peaks by homonuclear NMR In order to make resonance assignments of RNA, homonuclear NOESY, TOCSY, and COSY experiments are performed. Each resonance (chemical shift) in a 1D-NMR spectrum corresponds with one hydrogen atom (proton) in the RNA molecule. In 2D-homonuclear NMR spectrum, each cross-peak represents the magnetization between one proton and another proton. We assigned distances of interactions based upon relative intensity of NOE cross peaks acquired at different mixing times to assess NOE buildup rates. In the case of the branch site duplex, all, but the ψN3H and the terminal protons, chemical shifts were already assigned (Newby & Greenbaum, 2001), so the sequential assignment of the new peaks was relatively straight forward (like connecting the dots) and in the case of ψN3H, it was by process of elimination.
2.3 Design of molecules in U2 snRNP studies Molecules used for this study were designed to represent native sequences of the U2 snRNP proteins in humans. The clones of p14 were a gift from Andy Berglund at University of Oregon (Eugene, Oregon). Clones of p14 and SF3b155 were gifts from Charles Query at Albert Einstein College of
25
Medicine (New York, New York). The different proteins were sub-cloned into either pET15b or pET11a using standard cloning methods.
2.4 RNA samples in U2 snRNP studies RNA sequences purchased correspond to 26-mer and 29-mer intron strands 5’-GGGCCUGGCUUUUUUUACUGACACUU-3’ and 5’-GCUCGUCAGUCUCGAGGGUACUGACACUU-3’, respectively, and 22-mer U2 snRNA strands 5’-AAGAUCAAGUGUAGUAUCUGCC-3’ and 5’-AAGAUCAAGUGΨAGUAUCUGCC-3’. γ-32P-ATP was purchased from MP Biomedical (Irvine, CA). T4 polynucleotide kinase (10 U/µL) was obtained from New England BioLabs (Ipswich, MA). Other chemicals and biochemicals were purchased from Sigma Aldrich (St. Louis, Mo) and Fisher Scientific (Hampton, NH), and used as received. Detection of 32P-labeled RNA was performed by using a phosphor screen and analyzed with a phosphorimager Storm 850 from Molecular Dynamics (Sunny Vale, CA).
2.4.1 Labeling of RNA with radioactivity For gel retardation studies and nitrocellulose membrane retention studies, intron strands were 5’-endlabeled with γ-32P -ATP. RNA (~1 pmol) was dissolved in Buffer A (50 mM tris (hydroxymethyl)aminomethane (Tris)-HCl, 10 mM MgCl2, 5 mM dithiothreitol (DTT), 0.1 mM spermidine, 0.1 mM EDTA, pH 7.6) and incubated with γ-32P -ATP (1 µL, 6.1 MBq) and T4 polynucleotide kinase (10 U) at 37 °C for 2 hours. End-labeled product was purified by denaturating 15% polyacrylamide gel electrophoresis (PAGE). 32P -labeled RNA samples were combined with unlabeled RNA of the same sequence in order to standardize RNA concentrations for binding assays.
2.5 Expression and purification of proteins p14 and SF3b155 2.5.1 Expression of p14 and SF3b155 proteins Liquid cultures of Escherichia coli BL21 (DE3) cells were grown by autoinduction (Grabski, 2003) at 37 °C and shaken for twenty-four hours. Cells were
26
then harvested by centrifugation in a JLA-8.100 rotor (Beckmann Coulter, Fullerton, CA) at 6,000 rpm, 4 °C for 15 minutes, resuspended in lysis buffer (50mM Tris-HCl pH 8.0,), and then lysed using a microfluidizer (Microfluidics Corporation, Newton, MA) at 12,000 psi according to the manufacturer’s instructions. The lysate was clarified by centrifugation in a JA-25.5 rotor (Beckmann Coulter, Fullerton, CA) at 20,000 rpm, 4 °C for 30 minutes.
2.5.2 Identification of proteins by SDS-PAGE p14 has a molecular weight of 14.6 kDa. SFb155 (1-462) has a molecular weight of 50.4 kDa, SFb155 (255-462) has a molecular weight of 22.3 kDa, SFb155 (199-462) has a molecular weight of 28.3 kDa, and SFb155 (1-254) has a molecular weight of 28.2 kDa (all determined by a molecular weight calculator called ExPASy ProtParam located at http://us.expasy.org/tools/protparam.html). Identification of the proteins was accomplished using a SDS-PAGE gel and with molecular weight ladder loaded in a control lane. The band from the electrophoresed protein was compared to the SDS- PAGE low range ladder from Bio-Rad (Hercules, CA) to allow for accurate identification of the protein by molecular weight (Figure 2.1).
2.5.3 Purification of p14 proteins To purify p14, A nickel column (HisTrap™ HP) was pre-equilibrated with buffer A (50 mM Tris-HCl, pH 8.0, 500 mM NaCl, and 10 mM imidazole). Clarified lysate was loaded on a freshly cleaned and charged nickel column (preequilibrated Buffer A: 50mM tris-HCl, pH 8.5, 500 mM NaCl, 10 mM imidazole) for purification. On the nickel column, we employed a step gradient ranging from 0-30% buffer B (50mM tris-HCl, pH 8.5, 500mM NaCl, and 1M imidazole) to elute the protein. Fractions were then analyzed by 15 % SDS-PAGE, and fractions containing the desired protein were pooled and dialyzed in thrombin cleavage buffer (20mM tris-HCl, pH 8.5, 200mM NaCl) overnight at 4°C. Purified p14 was then dialyzed into NMR buffer (10mM BES [N,N-Bis-(2-hydroxyethyl)-2-
27
A
B
C
Figure 2.1 Visualization of proteins by SDS-PAGE. (A) The SDS-PAGE on the left shows purified p14 at the correct molecular weight of approximately 14.6 kDa from fractions collected by FPLC. (B) The SDS-PAGE in the middle shows purified SF3b155(199-462) from fractions collected by FPLC in the second lane at a molecular weight of approximately 30 kDa. (C) The SDS-PAGE on the left shows purified SF3b155(1-462) from fractions collected by FPLC in the first lane at a molecular weight of approximately 50 kDa. All separating gels contained 15% polyacrylamide and after electrophoresis were stained with coomassie-Blue R250 for visualization.
28
aminoethanesulfonic Acid] or HEPES [4-(2-Hydroxyethyl) piperazine-1ethanesulfonic acid], 50 mM H3BO3, 400 mM NaCl, pH 6.5). After overnight dialysis, p14 was then concentrated with a Millipore© Amicon stirred cell apparatus (Billerica, MA) to 500 µM.
2.5.4 Purification of SF3b155 proteins To purify SF3b155 proteins, an anion exchange column (HiTrap™ Capto™ Q) was first pre-equilibrated with buffer A (50 mM Tris, pH 8.0), and clarified lysate was then loaded onto the column. Fractions were eluted using a 2.5% step gradient of buffer B (50 mM tris, 1 M NaCl, pH 8.0), and then identified by 15 % SDS-PAGE. Fractions found to contain SF3b155 proteins were pooled together and dialyzed in buffer A (10 mM BES, 50 mM H3BO3, pH 5.3). The dialyzed protein was then loaded onto a cation exchange column (HiTrap™ SP HP) and eluted using a step gradient ranging from 0-25% in 2.5% increments of buffer B (10 mM BES, 50mM H3BO3, 1.0 M NaCl, pH 5.3). The collected fractions were identified by 15% SDS PAGE, and fractions containing the SF3b155 protein were dialyzed into NMR buffer (10mM BES or HEPES, 50 mM H3BO3, 400 mM NaCl, pH 6.5). After overnight dialysis of the SF3b155 protein, the diasylate was then concentrated with a Millipore© Amicon stirred cell apparatus (Billerica, MA) to 500 µM. All chromatography columns were purchased from Amersham Biosciences (Piscataway, NJ).
2.5.5 Determination of monomeric proteins by Dynamic Light Scattering Dynamic Light Scattering (DLS) was used to determine the oligomeric state of p14 in solution. In this case, DLS provides information about the size (hydrodynamic radius), estimated molecular weight and distribution of macromolecules in solution. The DynaPro-MS/X DLS (Wyatt Technology, Santa Barbara, CA) apparatus detects molecular size ranging from 0.5 nm to 1µm in hydrodynamic radius. For this assay, our samples consisted of 50 µL of 2 mg/mL of lysozyme and 4 mg/mL of p14 in 10 mM BES, 50 mM H3BO3, 400 mM NaCl at pH 6.5 and 40 scans were collected over the course of the DLS experiment. A
29
representative DLS data graph of lysozyme (Figure 2.2) indicates the presence of two peaks. The first peak at ~0.05 nm is attributed to the solvent peak and second peak at ~2.0 nm is contributed to monomeric lysozyme. The peak attributed to monomeric lysozyme had a hydrodynamic radius of 1.81 nm with an estimated molecular weight of 13.06 kDa, which is slight lower than the actual weight of 14.3 kDa. p14 with an estimated molecular weight of 17.44 kDa, which is slightly higher than the calculated molecular weight of the 14.6 kDa p14. In Chapter 4, dynamic light scattering (DLS) data will be presented that shows p14 to be monomeric with a hydrodynamic radius of 1.99 nm.
2.6 Gel Retardation Studies Agarose gel electrophoresis and PAGE were used to characterize protein binding at the intron branch site. Horizontal agarose gels were first utilized due to the large difference in isoelectric points (pI) predicted for p14 and SF3b155. Using ExPASy ProtParam to calculate apparent pIs, (located at http://us.expasy.org/tools/protparam.html), we found that p14 had a pI of 9.41 and SF3b155 had a pI range between 5.63 to 6.39. Therefore, the p14 will run towards the anode (negative electrode), while SF3b155 will travel towards the cathode (positive electrode) during electrophoresis at pH 6.5. We poured 3 mm thick 1.88% horizontal agarose gels placing the comb in the center of the gel to create wells. With this approach, we were able to capture either protein on the gel, regardless of the direction in which the protein runs. This alleviated the problem of visualizing both protein bands on a gel simultaneously, however, the required thickness of the agarose gel made it difficult to calculate apparent binding constants due to a potential gradient of radioactive material within the gel profile and smearing product bands. To alleviate these problems, we poured 5% PAGE gels buffered with Tris-MOPS. This allowed the gel to capture p14, but poor resolution and “smearing” of the gel bands became problematic with this low percentage gel approach. The results of these two gel strategies produced only qualitative data and another technique was needed for quantitative data analysis.
30
lysozyme
solvent
Figure 2.2 Representative DLS graph of percent intensity verses hydrodynamic radius (Rh) shows two peaks. Sample used was 2 mg/mL lysozyme in 10 mM BES, 50mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer. The first peak at ~0.05 nm is attributed to the solvent peak and the second peak at 1.81 nm is attributed to monomeric lysozyme with an estimated molecular weight of 13.06 kDa, which is slightly lower than the actual molecular weight of the 14.3 kDa.
31
2.7 Fluorescence Spectroscopy 2.7.1 Sample preparation Samples of proteins and RNA for fluorescence studies were prepared in the 10 mM BES, 50mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer. The concentration of RNA was determined by taking an absorbance at 260 nm and using a conversion factor of 1.0 A.U. ≈ 40 µg/mL for single stranded RNA. The concentration of each protein was determined by taking an absorbance at 280 nm and using Beer’s Law and the extinction coefficient specific to that protein (Appendix A). The RNA and protein were diluted to between 4 µM and 10 µM in the 10 mM BES, 50mM H3BO3, 400 mM NaCl @ pH 6.5 buffer with a final volume of 150 µL in a 4 mm square fluorescence cuvette from Starna Cell, Inc (Atascadero, CA).
2.7.2 Data Collection After the samples were prepared, experiments were performed using a Varian Cary Eclipse (Palo Alto, CA) fluorescence spectrophotometer with a xenon flash lamp. Measurements were taken at a rate of 80 points/minute. The SF3b155 proteins at concentrations of 6 µM in 10 mM HEPES, 50 mM H3BO3, 400 mM NaCl @ pH 6.5 were excited at 280 nm with a 5 nm slit width and emission spectra were collected from 300 nm to 600 nm. Then, various amounts of p14, RNA, and DNA were added to 6 µM of a fragment of SF3b155 in 10 mM HEPES, 50 mM H3BO3, 400 mM NaCl @ pH 6.5. The fluorescence intensity was measured in arbitrary units (A.U.). The data was plotted and analyzed using SigmaPlot 8.0 (Point Richmond, CA). Figure 2.3 shows a Jablonski Diagram illustrating the three steps involved in fluorescence. Excitation (S0-S1’) results from absorption of a photon of energy hνEX supplied by an external source by the fluorophore, creating an excited electronic singlet state (S1’). Excited-state lifetime (S1’-S1) lasts for a finite time, during which the fluorophore undergoes conformational changes and interacts with the environment. Fluorescence emission (S1-S0) results from a photon of energy hνEM being emitted, returning the fluorophore to its ground state S0.
32
• 1: Excitation • 2: Excited-State Lifetime • 3: Fluorescence Emission
Figure 2.3 Jablonski Diagram. The diagram above illustrates the processes involved in excited electronic singlet state by optical absorption and subsequent emission of fluorescence. Fluorescence is the result of a three-stage process, excitation, excited-state lifetime, and fluorescence emission that occurs in fluorophores. Excitation (S0-S1’) results from a photon of energy hνEX supplied by an external source absorbed by the fluorophore, creating an excited electronic singlet state (S1’). Excited-state lifetime (S1’-S1) lasts for a finite time, during which the fluorophore undergoes conformational changes and is also subject to interaction with the environment. Fluorescence emission (S1-S0) results from a photon of energy hνEM being emitted, returning the fluorophore to its ground state S0. The difference in energy (hνEX – hνEM) between before excitation and after emission is the recorded intensity.
33
2.8 Circular Dichroism (CD) Spectroscopy 2.8.1 Sample preparation Samples of proteins and RNA were prepared in a buffer of 10 mM BES, 50 mM H3BO3, and 400 mM NaCl @ pH 6.5. The concentration of RNA was determined by measurement of absorbance at 260 nm and using a conversion factor of 1.0 A.U. ≈ 40 µg/mL for single stranded RNA. The concentration of each protein was determined by taking an absorbance at 280 nm and using the extinction coefficient specific to that protein (Appendix A) in combination with Beer’s Law. The RNA and protein were diluted to between 10 µM and 25 µM for far UV and between 70 µM and 100 µM for near UV in the 10 mM BES, 50mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer with a final volume of 200 µL in a rectangular cuvette with a pathlength of 0.1 cm.
2.8.2 Data collection An AVIV 202 (Wyatt Technology, Santa Barbara, CA) CD spectrometer was used to measure the differential absorbance of left and right handed circularly polarized light. The data were collected between wavelengths ranging from 190 nm to 400 nm at 4 °C using a Peltier temperature controlled system. The data were plotted and analyzed using SigmaPlot 8.0 (Point Richmond, CA) and normalized using the protein concentration. Figure 2.4 shows CD spectra of poly-L-lysine with the conformations of αhelix, β-strand, and random coil (Greenfield & Fasman, 1969). The spectrum of poly-L-lysine in an α-helix conformation shows two minima at ~208 nm and ~222 nm and a maximum at ~190 nm. The spectrum of poly-L-lysine in a β-strand conformation shows a minimum at ~218 nm and a maximum at ~195 nm. The spectrum of poly-L-lysine in a random coil conformation shows a minimum at ~195 nm and a maximum at ~220 nm.
34
Figure 2.4 CD spectra of poly-L-lysine at different conformations in aqueous solution (adapted from (Greenfield & Fasman, 1969).
35
2.9 Nitrocellulose Membrane Retention Studies 2.9.1 Sample preparation RNA samples were 5’-endlabeled with γ-32P -ATP. RNA (~1 pmol) was dissolved in Buffer A (50 mM tris (hydroxymethyl)aminomethane (Tris)-HCl, 10 mM MgCl2, 5 mM dithiothreitol (DTT), 0.1 mM spermidine, 0.1 mM EDTA, pH 7.6) and incubated with γ-32P -ATP (1 µL, 6.1 MBq) and T4 polynucleotide kinase (10 U) at 37 °C for 2 hours. End-labeled product was purified by 15% denaturing PAGE. 32P -labeled RNA samples were mixed with 3 µM unlabeled RNA of the same sequence in order to standardize RNA concentrations. Samples of proteins and RNA were prepared in the 10 mM BES, 50 mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer with 5% glycerol (C3H8O3). The concentration of RNA was determined by taking an absorbance at 260 nm and using a conversion factor of 1.0 A.U. ≈ 40 µg/mL for single stranded RNA. The concentration of each protein was determined by taking an absorbance at 280 nm and using Beer’s Law with the extinction coefficient specific to that protein (Appendix A). The concentration of RNA used was 0.5 µM and the protein concentration ranged from 0.5 µM to 50 µM.
2.9.2 Data collection for filter binding studies. Filter binding studies were performed with a Minifold I Dot-Blot System (96-Well) purchased from Whatman (Florham Park, NJ)(Hall & Stump, 1992). Once the system was assembled, blot paper (Whatman, Florham Park, NJ) and either 0.2 µm or 0.45 µm nitrocellulose membrane was used (Whatman, Florham Park, NJ). Some experiments used a HybondTM-N+ membrane (Amersham, Piscataway, NJ) between the nitrocellulose membrane and blot paper for a more direct count of unbound RNA. The wells containing the nitrocellulose membrane were washed three times with 400 µL of BES buffer (described above). The binding reactions were then applied to each well; total reaction volume was 3 µL. Next, 200 µL of BES buffer was then washed over each well and the membrane dried by vacuum filtration. Lastly, as a count of total RNA concentration, 1 µL of 32
P-labeled RNA was air dried onto the filter. After overnight exposure on the 36
Molecular Dynamics Phosphor Screen, 32P-labeled RNA was detected by using the phosphorimager Storm 850 from Molecular Dynamics (Sunny Vale, CA) and volumes were determined using Molecular Dynmaics ImageQuant v1.1 software (Sunnyvale, CA).
2.10 Isothermal Titration Calorimetry Isothermal titration calorimetry (ITC) is used to determine the thermodynamic properties of various molecular interactions, such as protein interactions with another proteins, DNA, RNA or metals. This technique measures the heat, either absorbed or released upon binding, and gives the information for determining binding constants (Kd), enthalpy (∆H), entropy (∆S), and stoichiometry of the reaction. ITC studies were conducted on a Microcal VP-ITC ultra-sensitive isothermal titration microcalorimeter. The VP-ITC unit has a precise temperature control set at 5 °C or 10 °C to directly measure heat evolved or absorbed in liquid samples as a result of injection of reactant (Doyle, 1999). The reference cell was filled with Millipore ultrapure deionized water. All samples were extensively dialyzed in 10 mM BES with 50 mM boric acid, 400 mM NaCl @ pH 6.5 with 20 mM Tris(2-carboxyethyl)phosphine (TCEP). Samples were filtered and degassed by gently stirring the sample under vacuum for 5 minutes using the Thermo-vac (a vacuum thermostating system) prior to introduction to the calorimeter. The experiment consists of 40 injections of 6 µL of SF3b155 (concentrations 100 µM to 500 µM) at room temperature was titrated into p14 (concentrations 5 µM to 25 µM) over 12 s by a syringe spinning at 270 rpm, with a 240 second equilibration time between injections. The instrument was controlled by Microcal Observer software comprising a 16-bit A/D converter board for data acquisition and a second interface board for calorimetric control. Data from control experiments in which the titrant was injected into buffer were subtracted from all protein-protein data to correct for heats of dilution. A simple binding model described by a ligand binding to a single set of n identical, non-interacting sites on a macromolecule (Doyle, 1999) was used to obtain the binding constant,
37
enthalpic (∆H), and entropic (∆S) terms contributing to the Gibbs free energy (∆G) of association.
2.11 Nuclear Magnetic Resonance Spectroscopy 2.11.1 Basics of NMR Figure 2.5 shows hydrogen nuclear spin (top graph) in the absence of an external magnetic field (left) and in the presence of an applied magnetic field (right). The energy differences between the hydrogen nuclear spin states indicated the magnetic field. The cartoon in the bottom of Figure 2.5 shows hydrogen nuclei (protons) when a magnetic field is not present (left). When a magnetic field (BO) is present (Figure 2.5 bottom, right), the spins align with the magnetic field.
2.11.2 Sample preparation Samples of RNA and protein were prepared in either 10 mM HEPES or 10 mM BES with 50 mM boric acid, 400 mM NaCl at pH 6.5. The protein concentrations were measured at 280 nm, and the concentrations were calculated using Beer’s Law and the specific extinction coefficient to that protein (Appendix A). The concentration of RNA was determined by taking an absorbance at 260 nm and using a conversion factor of 1.0 A.U. ≈ 40 µg/mL for single stranded RNA. The protein and RNA concentrations for each NMR experiment were between 300 µM and 500 µM, except for the experiments with RNA in capillary tubes, as described.
2.11.3 Data collection NMR spectra were acquired with the following instruments: a 720-MHz Varian Unity Plus spectrometer (National High Magnetic field Laboratory (NHMFL), Tallahassee, FL), a 600-MHz Varian Unity Plus spectrometer (National High Magnetic field Laboratory (NHMFL), Tallahassee, FL), and a 500-MHz Varian spectrometer (FSU Department of Chemistry & Biochemistry, Tallahassee, FL). For one-dimensional NMR experiments, we used a jump-
38
return echo pulse sequence (Figure 2.6)(Sklenar & Bax, 1987). For HSQC (Figure 2.10) (Kay, 1992; Kay et al., 1992) and TROSY(Pervushin et al., 1997; Meissner et al., 1998; Pervushin et al., 1998a) experiments, we used standard Varian NMR pulse sequences using gradient solvent suppression. For both HSQC and TROSY, the experiments were run at either 4 °C for 15N-labeled proteins and 10 °C for 2H15N-labeled and 2H13C15N-labeled proteins with either 16 scans, 32 scans or 64 scans (for best resolution) with 256 increments. NMR experiments to assist in sequence-specific assignment of amides of 2
13
H- C-15N-labeled p14 include: HNCA (Figure 2.11) (Yamazaki et al., 1994;
Pervushin et al., 1998b; Yang & Kay, 1999), HNCO (Figure 2.12) (Pervushin et al., 1998b; Yang & Kay, 1999), HNCACB (Figure 2.13 ) (Shan et al., 1996; Pervushin et al., 1998b; Yang & Kay, 1999), NH-NOESY (Figure 2.14) (Silver et al., 1984), NH-TOCSY (Figure 2.15) (Silver et al., 1984), HN(CO)CACB (Figure 2.16) (Shan et al., 1996; Pervushin et al., 1998b; Yang & Kay, 1999), HN(CO)CA (Figure 2.17) (Yamazaki et al., 1994; Pervushin et al., 1998b; Yang & Kay, 1999), and HN(CA)CO (Figure 2.18) (Wittekind et al., 1992; Kay et al., 1994; Muhandiram & Kay, 1994). Side chains were assigned using HC(CO)NHTOCSY(Logan et al., 1992; Montelione et al., 1992; Grzesiek et al., 1993; Logan et al., 1993), HBHA(CO)NH (Swapna & Montelione, 1999), and HCCH-TOCSY (Figure 2.19) (Bax et al., 1990; Kay et al., 1993) experiments. Data were processed with software NMRpipe (Delaglio et al., 1995) or SpinWorks 2.5.4 (Marat, 2005) and analyzed using software NMRView (Johnson & Blevins, 1994) or Sparky (Kneller, 1989). Chemical-shift assignments will be obtained from standard three-dimensional triple resonance experiments (Sattler et al., 1999). For NMR-binding experiments, unlabeled RNA strands or proteins were added to 15
N-labeled p14 or 15N-labeled-SF3b155 fragments, and chemical shift changes
were monitored by two-dimensional 1H-15N HSQC correlation spectra.
39
Figure 2.5 The top figure shows hydrogen nuclear spin in the absence of an external magnetic field (left) and in the presence of an applied magnetic field (right). The energy differences between the hydrogen nuclear spin states are indicated by the magnetic field. The bottom figure shows hydrogen nuclei (protons) when a magnetic field is not present (left) and when a magnetic field (BO) is present (right).
40
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed.
ACQ 1
H
Figure 2.6 Jump-return (1-1) echo scheme and pulse sequence, used for generating one-dimensional spectrum. Where 90° is the pulse width, T is a constant delay time, HS is the homospoil pulse, ∆ is a delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, and ACQ is the acquisition time. In order to increase the suppression per individual scan, short homospoil pulses (HS) can be applied during the delays, ∆. (adapted from (Sklenar & Bax, 1987)).
41
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed.
ACQ 1
H
Figure 2.7 NOESY scheme and pulse sequence. Nuclear Overhauser Enhancement spectroscopy (NOESY) is used for generating two-dimensional spectrum. Where 90° is the pulse width, t1 is the evolution time, τmix is the mixing time, and ACQ is the acquisition time. (adapted from (Silver et al., 1984)).
42
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Green Atoms are involved in the magnetization transfer, but they are not observed in the final spectra.
τmix ACQ 1
H
Figure 2.8 TOCSY scheme and pulse sequence. Total Correlated Spectroscopy (TOCSY) is used for generating two-dimensional spectrum. Where 90° is the pulse width, t1 is the evolution time, τmix is the mixing time, and ACQ is the acquisition time. (adapted from (Silver et al., 1984)).
43
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Green Atoms are involved in the magnetization transfer, but they are not observed in the final spectra.
Figure 2.9 COSY scheme and pulse sequence. Correlation spectroscopy (COSY) is used for generating two-dimensional spectrum. Where 90° is the pulse width, t1 is the evolution time, and ACQ is the acquisition time. (adapted from (Rance, 1983)).
44
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.10 HSQC scheme and pulse sequence. Heteronuclear Single Quantum Coherence (HSQC) is used for generating two-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, t1 is the evolution time, GARP is a decoupling sequence used to decouple 15 N from 1H, and t2 is the acquisition time. (adapted from (Sattler et al., 1999)).
45
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.11 HNCA scheme and pulse sequence. Named for order of magnetization transfer, H-N-Cα, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, T is a constant delay time, Φ is equal to the phase cycled along the x, y, x, and -y axis, τ is a delay time, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
46
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.12 HNCO scheme and pulse sequence. Named for order of magnetization transfer, H-N-CO, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, T is a constant delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, τ is a delay time, δ is a delay time, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
47
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.13 HNCACB scheme and pulse sequence. Named for order of magnetization transfer, H-N-Cα-Cβ, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, Φ and ψ is equal to the phase cycled along the x, y, -x, and -y axis, DIPSI is an isotropic mixing time, GARP is a decoupling sequence used to decouple 15N from 1H, τ is a delay time, ∆ is a delay time, δ is a delay time, Gz is the gradient applied to the z-axis, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
48
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.14 1H-15N-NOESY-HSQC scheme and pulse sequence used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, ∆ is a delay time, ε is a delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, τmix is the mixing time, t1 is the evolution time, t2 is the evolution time, cpd is a decoupling sequence used to decouple 15N from 1H, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
49
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.15 1H-15N-TOCSY-HSQC scheme and pulse sequence used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, ∆ is a delay time, ε is a delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, MLEV is the isotropic mixing time, t1 is the evolution time, t2 is the evolution time, GARP is a decoupling sequence used to decouple 13C from 1H, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
50
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Green Atoms are involved in the magnetization transfer, but they are not observed in the final spectra. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.16 CBCA(CO)NH scheme and pulse sequence. Named for order of magnetization transfer, H-N-Cα-Cβ, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, T is a constant delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, BS is a Bloch-Siegert compensating pulse, DIPSI is an isotropic mixing time, GARP is a decoupling sequence used to decouple 15N from 1 H, t1 is the evolution time, t2 is the evolution time, t3 is the acquisition time, and α, β, δ, ζ, υ, γ, η are delay times. (adapted from (Sattler et al., 1999)).
51
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Green Atoms are involved in the magnetization transfer, but they are not observed in the final spectra. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.17 HN(CO)CA scheme and pulse sequence. Named for order of magnetization transfer, H-N-Cα, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, τ is a delay time, δ is a delay time, ε is a delay time, Φ is equal to the phase cycled along the x, y, -x, and -y axis, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
52
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Green Atoms are involved in the magnetization transfer, but they are not observed in the final spectra. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.18 HN(CA)CO scheme and pulse sequence. Named for order of magnetization transfer, H-N-CO, used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, Φ and ψ is equal to the phase cycled along the x, y, -x, and -y axis, BSP is a Bloch-Siegert compensating pulse, DIPSI is an isotropic mixing time, GARP is a decoupling sequence used to decouple 15N from 1H, τ is a delay time, ∆ is a delay time, δ is a delay time, Gz is the gradient applied to the z-axis, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
53
Color Key: Red Atoms are those whose chemical shifts are going to be measured or observed. Blue Arrows depict the direction of the magnetization transfer.
Figure 2.19 HCCH-TOCSY scheme and pulse sequence used for generating three-dimensional spectrum. Where a single block line is a 90° pulse width, a double block line is a 180° pulse width, Φ ψ is equal to the phase cycled along the x, y, -x, and -y axis, ∆ is a delay time, γ is a delay time, ε is a delay time, DIPSI is an isotropic mixing time, GARP is a decoupling sequence used to decouple 13C from 1H, t1 is the evolution time, t2 is the evolution time, and t3 is the acquisition time. (adapted from (Sattler et al., 1999)).
54
CHAPTER 3 REFINEMENT OF THE BRANCH SITE HELIX STRUCTURE BY NMR STUDIES OF RNA DUPLEXES IN SUPERCOOLED WATER
3.1 Introduction Determination of nucleic acid structures by NMR methodology relies upon numerous inter- and intra-molecule distances and angular constraints. Interactions involving exchangeable imino and amino protons provide important information about hydrogen bonds and stabilizing interactions with water molecules, particularly for analysis of unusual structural motifs of folded RNA molecules. A sizable fraction of labile protons in RNA (-NH, -NH2, and -2’OH), however, exchange rapidly with solvent at T > 0 °C. As a result, they are exchange broadened beyond detection, thus escaping use in structural studies. Sufficiently decreased temperature slows chemical exchange between these protons and water, allowing for their direct detection and recruitment for structural study. Borer and colleagues (Kerwood et al., 2001) detected resonance peaks of imino protons of terminal nucleotides at -6 °C that were not observed at 0 °C. In order to monitor non-base paired imino protons, however, a method to achieve temperatures as low as -20 °C without freezing is desirable. Exploiting the empirical observation that the freezing point of water decreases proportionally with volume (Angell, 1982), Poppe and van Halbeek (Poppe & van Halbeek, 1994) studied sucrose in glass capillary tubes at -17 °C, allowing measurement of -OH proton chemical shifts and 3JHH couplings(Poppe & van Halbeek, 1994). The capillary technique was extended to Bovine Pancreatic Trypsin Inhibitor (BPTI), ubiquitin, dATP and dGTP in supercooled water at about -18 °C (Skalicky et al., 2000; Skalicky et al., 2001) In this study, we have investigated the structural role of exchangeable protons of pseudouridine (ψ), a rotational isomer of uridine attached to its ribose
55
through C5, in RNA duplexes. ψ has two imino nitrogen atoms, ψ N1H and ψ N3H, both of which are protonated at physiological pH (Hall & McLaughlin, 1992). The presence of ψ in RNA helices has been shown to increase thermal stability without altering structure (Davis & Poulter, 1991; Hall & McLaughlin, 1992; Arnez & Steitz, 1994; Kintanar et al., 1994; Durant & Davis, 1999; Yarian et al., 1999). This is postulated to be the result of a water-mediated hydrogen bond involving the ψ N1H (Arnez & Steitz, 1994; Newby & Greenbaum, 2002a) and/or improved base stacking (Yarian et al., 1999; Chui et al., 2002). In the case of the premRNA branch site helix of the yeast spliceosome, the presence of a conserved ψ induces a strikingly different structure than that observed with uridine (Newby & Greenbaum, 2001, 2002b). Structural models of the branch site duplex suggested that ψ did not appear to form a base pair with the opposing adenine (A23, adjacent to the bulged base, A24), and no resonance attributable to ψ N3H was visible (Figure 3.1). Identification of this resonance would contribute important information about the environment of this imino proton with respect to surrounding bases. I chose to study the pre-mRNA branch site helix of the yeast spliceosome because the presence of a conserved ψ induces a strikingly different structure than when ψ is observed in a complementary duplex. Therefore, the rationale behind these studies was to determine the location of ψ N3H because we wanted to refine the structure of the branch site duplex and in doing so, understand the role of the conserved ψ in the branch site duplex. In order to slow the exchange of imino protons, we have acquired NMR spectra of two RNA duplexes containing ψ (Figure 3.1) in aqueous solutions at supercooled temperatures by implementation of the capillary technique (Poppe & van Halbeek, 1994). In addition to detecting imino protons corresponding to terminal base pairs of an RNA duplex, we observed the upfield shifted N3H imino proton of a non-base paired ψ in the branch site duplex. Determining the role of ψ in stabilizing RNA structures may explain its phylogenetic conservation in the branch site helix and elsewhere in structural RNA molecules. Also, as the temperature decreased below zero, we observed that the chemical exchange
56
Figure 3.1 Sequences of the two RNA duplexes ψcomp (top) and ψBP (bottom) (Schroeder et al., 2005). ψBP represents short sequences from the pairing of U2 snRNA (top strand) and the intron (bottom strand) from the yeast S. cerevisiae. The numbering scheme was that used in previous structure determination studies (Newby & Greenbaum, 2001, 2002b, a). The position of ψ (ψ6 in this sequence) corresponds to a phylogenetically conserved ψ residue in U2 snRNA (ψ35 in yeast). ψcomp is similar to the branch site duplex, but without the bulged adenosine (A).
57
rate of certain imino protons in an RNA duplex increased instead of slowing down the chemical exchange between these protons and water.
3.2 Results and Discussion 3.2.1 Pseudouridine monophosphate. We first acquired NMR spectra of pseudouridine monophosphate (ψMP) in order to characterize the resonances of its two imino protons, ψ N1H and ψ N3H, in a non-base paired environment. Spectra of imino protons (including line widths) were identical for samples in capillaries and as a bulk sample at 0 and 5 °C (data not shown). The ψ N1H and ψ N3H proton chemical shifts were 10.6 and 10.8 ppm and line-widths at halfheight (ν1/2) were ~3.8 and ~5.1 Hz at 0 °C, respectively (Figure 3.2). Upon cooling to -15 °C, ν1/2 decreased monotonically (~2.0 and ~2.7 Hz for N1H and N3H, respectively), consistent with line narrowing associated with slower solvent exchange. ν1/2 of N1H was slightly larger than that of N3H, presumably as a result of dipolar broadening from the proximal H6 proton of ψ (~ 2.5 and 4.8 Å between H6 and N1H and N3H, respectively). We also observed a slight temperature dependence of chemical shift (thermal coefficient) for the imino protons. Positive thermal coefficients (i.e. more upfield chemical shift as a function of decreased temperature) correlate with rapid exchange, whereas negative thermal coefficients correlate with slower exchange, and are therefore consistent with hydrogen bonding (Nonin et al., 1995). We plotted the chemical shift with respect to temperature for each imino proton. The thermal coefficients for ψ N1H and ψ N3H were 6.1 and 5.3 ppb/ °C, respectively, consistent with nonbase paired environment (Figure 3.3). 3.2.2 Complementary duplex. By comparison, NMR spectra of a complementary duplex (ψcomp, Figure 3.1) indicate that all imino protons, including N3H of ψ, participate in Watson-Crick base pairs (Figure 3.3). ψ N3H in the anti conformation forms a hydrogen bond with the opposing adenine N1, reflected in the downfield shift of the imino proton resonance to ~13.1 ppm (Figure 3.3)(Hall & McLaughlin, 1992; Durant & Davis, 1999). Further 58
Figure 3.2 Imino proton spectra of ψ-monophosphate (ψMP) acquired at different temperatures between 5 °C and -15 °C (Schroeder et al., 2005). ψMP (10 mM in 10 mM sodium phosphate, pH 6.4, 50 mM NaCl, 0.1 mM EDTA, in 90% H20/10% D2O) was taken up into six capillary tubes, which were placed in a 5 mm NMR tube (fill factor ~0.25). Data were collected using a jump-return-echo pulse sequence (Sklenar & Bax, 1987) on a Varian 720 MHz and 600 MHz spectrometer at the National High Magnetic Field Laboratory, Tallahassee, FL.
59
Figure 3.3 Chemical shift recorded with respect to temperature at which the sample was recorded. The slopes derived from these plots were used to calculate the thermal coefficients for ψ N1H and ψ N3H of ψ-monophosphate (ψMP). The thermal coefficients for ψ N1H and ψ N3H were calculated to be 6.1 and 5.3 ppb/ °C, respectively. The positive slopes are consistent with non-base paired environment (Nonin et al., 1995).
60
confirmation of this assignment came from comparison of chemical shifts of a similar duplex in which ψ were replaced by uridine (U). We assigned the resonance at 10.6 ppm to ψ N1H based upon the chemical shift for this imino proton in ψMP and from observation of an NOE between it and the H6 proton (~2.5 Å away) (Hall & McLaughlin, 1992; Newby & Greenbaum, 2001). Previous studies from our laboratory have shown that ψ N1H, which is on the major groove edge of the base when in an anti conformation, is visible in a one-dimensional NMR spectrum, and exhibits an NOE to its own H6 (Newby & Greenbaum, 2001). In order to determine how this proton is protected from rapid exchange with solvent, our previous studies made use of a CLEANEX-PM pulse sequence (Hwang et al., 1997) to characterize the interaction between ψ N1H and water molecules (Newby & Greenbaum, 2002a). For protons undergoing chemical exchange, this experiment determines whether there is a component of cross relaxation. Appearance of a negative peak at the resonance location of ψ N1H indicated that this proton undergoes cross-relaxation with water. The CLEANEXPM spectrum revealed a strong negative peak at the resonance location of ψ N1H (but no other imino proton), indicating that the interaction with water is characterized by a significant component of cross relaxation. These observations are consistent with existence of ψ N1H participating in a water mediated hydrogen bond with a phosphate oxygen atom of the same or a neighboring nucleotide, as predicted by Durant &Davis (1999), and visualized in a crystal structure of tRNA containing pseudouridine (Arnez & Steitz, 1994). At about -10 °C, a broad peak appeared at ~12.2 ppm with a shoulder at ~12.5 ppm. All imino protons of ψcomp had previously been assigned except for those belonging to terminal base pairs (Figure 3.4); therefore, these new peaks were attributed to the imino protons of terminal residues. In contrast with the narrowing of peaks of the monophosphate at lower temperatures, broadening of peaks of the complementary duplex was observed below 0 °C (line widths increased gradually from ~50 Hz to ~130 Hz). Interestingly, it is also worth noting that the line widths for the non-exchangeable protons in ψcomp increase
61
*
*
Figure 3.4 Imino proton spectra of ψcomp acquired at temperatures between 5 °C and -15 °C (Schroeder et al., 2005). ψcomp (2.7 mM in 10 mM sodium phosphate, pH 6.4, 50 mM NaCl, 0.1 mM EDTA, in 90% H20/10% D2O) were taken up into ten capillary tubes, which were placed in a 5 mm NMR tube (fill factor ~0.4). Data were collected using a jump-return-echo pulse sequence (Sklenar & Bax, 1987) with the number of scans increasing as the temperature decreases on a Varian 720 MHz and 600 MHz spectrometer at the National High Magnetic Field Laboratory, Tallahassee, FL. As the temperature decreases, two peaks begin to emerge (*), which correspond to the G2 and U19 terminal protons, and the linewidths at half height increase.
62
from ~5 Hz to ~10 Hz over the same temperature range. The exact reasons for this behavior may be a complex combination of slower tumbling of the larger molecule as a result of increased solvent viscosity, salt and buffer effects, or may represent the beginning of cold denaturation. 3.2.3 Branch site duplex. We then performed similar experiments on a minimal pre-mRNA branch site duplex of S. cerevisiae (ψBP), which represents the pairing between a short consensus region of the U2 snRNA and the intron (Figure 3.1). In contrast with the complementary duplex, presence of a ψ residue in the U2 snRNA strand of ψBP (top strand in Figure 3.1) in its conserved position (Yu et al., 1998; Ma et al., 2003) results in a very different conformation than in its unmodified counterpart (Newby & Greenbaum, 2001, 2002b). In the novel motif, the unpaired adenosine is extruded from the helix and forms a base triple with the minor groove edge of A7 in the A7-U22 base pair. The 2’OH of this extra-helical adenosine is the nucleophile in the first cleavage step of splicing. The structural motif preferred in the presence of ψ35 may explain the high degree of phylogenetic conservation of this modified base in this location. In ψBP, the chemical shift of ψN1H is ~10.5 ppm (~10.6 ppm in ψMP and ψcomp). As was observed in ψcomp, previous NMR investigation of the interaction of ψ N1H of ψBP indicated that this proton undergoes cross-relaxation with water molecule(s) in the major groove of the duplex (Newby & Greenbaum, 2002a). Unlike the case with ψcomp, proton spectra of ψBP revealed no resonance attributable to ψ N3H, and structural models did not indicate formation of a base pair between ψ and the opposing adenine (A23, adjacent to the branch site base A24). In order to identify the resonance location of the absent (and presumably exchange broadened) ψ N3H in the branch site duplex at 0-5 °C, we acquired spectra of ψBP in supercooled water (Figure 3.5). As in spectra of the complementary duplex, a broad new peak emerged at ~12.2 ppm below -5 °C, corresponding to G2 terminal base pair. The imino proton of the other terminal base pair may be degenerate with a resonance at ~13.3 ppm. Unique to the
63
*
*
Figure 3.5 Imino proton spectra of ψBP acquired at different temperatures between 5 °C and –17.5 °C (Schroeder et al., 2005). ψBP (2.5 mM in 10 mM sodium phosphate, pH 6.4, 50 mM NaCl, 0.1 mM EDTA, in 90% H20/10% D2O) was taken up into nine capillary tubes, which were placed in a 5 mm NMR tube (fill factor ~0.3). Data were collected using a jump-return-echo pulse sequence (Sklenar & Bax, 1987) on a Varian 720 MHz and 600 MHz spectrometer at the National High Magnetic Field Laboratory, Tallahassee, FL. As the temperature decreases, the two peaks begin to emerge (*), which correspond to ψ N3H and the G2 terminal proton, and the line-widths at half height increase.
64
branch site duplex, a broad peak emerged at ~11.2 ppm at -15 °C. Because assignments of all other imino protons of the branch site helix were made by systematic comparison with other duplexes (Newby & Greenbaum, 2002b), we identified this upfield shifted resonance as that of ψ N3H. No NOEs were observed from this broad peak. The ψ N3H chemical shift was close to that of the unpaired N3H of ψMP, which is not base paired (Figure 3.2), and very different from that of the Watson-Crick paired ψ of ψcomp (Figure 3.4). The upfield location of ψ N3H of ψBP (Figure 3.5) further supports our original conclusion that ψ6 does not form a canonical base pair with A23 (Newby & Greenbaum, 2002b). This also supports the hypothesis that the spliceosomal protein, p14, can interact with the widened major groove when pseudouridine is present in the branch site duplex. Also, as compared with ψcomp, the line widths of ψBP began to broaden below 0 °C, increasing from ~50 Hz to ~200 Hz. Interestingly, it is also worth noting that the line widths for the non-exchangeable protons in ψBP barely increase from ~5 Hz to ~10 Hz in the same temperature range. In the case of A24 2H, its line width increases from ~7 Hz to ~16 Hz. This difference may reflect slower tumbling of the larger molecule as solvent viscosity increased, salt and buffer effects, or may represent the beginning of cold denaturation. 3.2.4 Thermal coefficients. We noted that chemical shifts of each imino proton consistently increased or decreased slightly with a decrease in temperature. We therefore generated a thermal coefficient for each imino proton resonance by plotting the chemical shifts as a function of temperature (Figure 3.6). For ψcomp, negative thermal coefficients (more downfield shifts with respect to decreased temperature) were observed for all protons except the terminal imino proton (Figure 3.6, top). The thermal coefficient for ψN1H and ψN3H were -6.2 ppb/ °C and -0.7 ppb/ °C, respectively, suggesting that both form hydrogen bonds (Newby & Greenbaum, 2001, 2002b, a). The thermal coefficient for G2 was positive, consistent with rapid exchange expected for a solvent-exposed terminal base pair.
65
Figure 3.6 Thermal coefficients for ψcomp (top) and ψBP (bottom) (Schroeder et al., 2005). The chemical shift for each proton was plotted as a function of temperature at which the spectrum was acquired. The value, in ppb/ °C, for each imino proton is shown next to its position in the sequence. Positive thermal coefficients correlate with rapid exchange and negative thermal coefficients correlate with hydrogen bonding .
66
In ψBP, protons with negative thermal coefficients belonged to G8, G5, G3, U4 and ψ N1H, and those with positive thermal coefficients were G2, U22 and ψ N3H (Figure 3.6, bottom). The negative thermal coefficient for ψ N1H (-3.7 ppb/ °C) reinforced the conclusion that this imino proton is involved in a hydrogen bond, as suggested by the results of earlier NMR studies (Newby & Greenbaum, 2002a). The thermal coefficient of U22 was 1.03 ppb/°C and that of ψ N3H was 0.08 ppb/ °C. Although U22 forms a Watson-Crick base pair with A7, its involvement in the base triple with the branch site A and its position adjacent to the unpaired ψ exposes U22 N3H to solvent, which apparently increases its exchange rate. For ψ N3H, this positive value is in accord with rapid exchange, suggesting the proton is not involved in a hydrogen bond. This observation, combined with its upfield shifted position and lack of NOEs to surrounding residues, especially the adjacent A2H, is entirely consistent with our original conclusion about this ψ’s non-base paired status (Newby & Greenbaum, 2002a) and assists in further refinement of the ψBP structure. 3.2.5 Chemical Exchange. We measured the proton rate of chemical exchange with respect to temperature for each imino proton in the branch site duplex (ψBP). We collected a number of two-dimensional NOESY spectra of ψBP and from each spectrum determined the intensity of each cross peak (IC) and intensity of each diagonal peak (ID). We calculated the chemical exchange using the following equation: kex=IC/(ID*τmix) where kex is the chemical exchange and τmix is the mixing time of the NOESY experiment. The results indicate that the rate of chemical exchange with respect to temperature (Figure 3.7) for each imino proton in ψBP increases below 0 °C. As expected, the rate of chemical exchange decreases from 10 °C to 0 °C for each imino proton. All the imino protons have a chemical shift between 0.005 s-1 and 0.01 s-1 at 0 °C. However, at -15 °C, U4 and G8 increased to 0.015 s-1; G3, G5, and ψ6 N1H increased to 0.02 s-1; and U22 increased to 0.09 s-1. This increase is consistent with the line-widths at half height increasing as the temperature
67
Figure 3.7 The chemical exchange rate verse temperature for ψBP. The chemical exchange for each proton was plotted as a function of temperature at which the spectrum was acquired. The chemical exchange shows a minimum at 0 °C for each imino proton except ψ6 N3H where the peaks were too broad.
68
decreases below zero degrees. We can rule out that the cause of increasing both the line-widths and rate of chemical exchange is freezing of the capillaries because we do not observe frozen capillaries and still observe chemical shifts. We are currently trying to determine the cause of this phenomenon. The processes that may contribute to this effect could be: 1) slower tumbling of the larger molecule as a result of increased solvent viscosity; 2) salt and buffer effects; or 3) the beginning of cold denaturation. 3.2.6 Buffer Effects. To investigate why the line-widths at half height and the rate of chemical exchange increased as we decreased the temperature below 0°C for the RNA duplexes, we will use 15N-labeled uridine (U) and pseudouridine (ψ) to measure the effects of salt and buffer to determine if this changes the pKa of the system. If there is a calculated pKa change, it means that the local chemical environment of the RNA changes as the temperature decreases. In collaboration with J.P. Desaulniers from Dr. Christine Chow’s laboratory at Wayne State University (Detroit, MI), we are planning to use 15N-labeled U and ψ in RNA duplexes to further study the effects of performing NMR experiments in supercooled water. We have acquired 1H-15N HSQC spectra of a complementary duplex with [3-15N]-labeled U. The HSQC spectrum below shows preliminary data from the complementary duplex with [3-15N]-labeled U (Figure 3.8). These data indicate that we will be able to monitor the chemical shift changes of the 15
N-labeled U or 15N-labeled ψ at position 6 in the RNA duplexes previously
studied at supercooled temperatures. Also, we have begun investigating why line widths broaden and the chemical exchange rates increase as temperatures fall below 0 °C. We have begun using 15N-labeled uridine and pseudouridine to measure the effects of salt and buffer to determine if this changes the pKa of the system. As of now, this is beyond the scope of this dissertation because the underlying physical explanation for the observed increase in chemical exchange rate as the temperature is decreased may be more complicated than the contribution from salt and buffer effects. This phenomenon may be the beginnings of cold
69
Figure 3.8 HSQC spectra of RNA duplexes with 15N-labeled U and ψ. The top spectrum shows a 1 mM complementary duplex (top), in 10 capillary tubes with [3-15N]-labeled U at position 6, acquired at 4°C on a 500 MHz Varian spectrometer. The 1H-15N HSQC experiment reveals the imino proton cross-peak associated with an [3-15N]-labeled U in a complementary duplex. The bottom spectrum shows a 1 mM complementary duplex (top), with [1,3-15N]-labeled ψ at position 6, acquired at 4°C on a 500 MHz Varian spectrometer. The 1H-15N HSQC experiment reveals the imino proton cross-peaks associated with an [1,315 N]-labeled ψ in a complementary duplex.
70
denaturation (Li et al., 1995; Narlikar & Herschlag, 1996) , which has not been as thoroughly studied as thermal denaturation in RNA. Cold denaturation is usually associated with deterioration of the hydrophobic effect caused by the temperature-dependent structure of an aqueous solution. The first direct evidence of cold denaturation in RNA was observed studying the hammerhead ribozyme in methanol solution by circular dichroism (Mikulecky & Feig, 2002) and isothermal titration calorimetry (Mikulecky & Feig, 2004). It has been stated previously that the docking of a helix involved in the group I intron showed local unfolding at reduced temperatures, which could be “cold denaturation” (Li et al., 1995; Narlikar & Herschlag, 1996). However, the study of cold denaturation of biomolecules (in this case only proteins) by NMR has only begun recently (Babu et al., 2004; Szyperski et al., 2006). In the future, however, it may be determined that the RNA with an incorporated ψ undergoes cold denaturation as the temperature decreases. An intriguing aspect associated with performing NMR in supercooled water is that the samples do not freeze until about -20 °C. It has been described that this is possible because there is an increase in the surface area to volume ratio in the capillary tubes. Using standard NMR tubes with 500 µL or Shigemi® (Allison Park, PA) NMR tubes with 300 µL, the ratio of surface area to volume is 3 cm-1 (Figure 3.9), but using capillary tubes increase the surface area to volume ratio increases to 20 cm-1 (Figure 3.9). However, according to the ideal gas law, if the volume and pressure remain constant then the temperature cannot change. Since this cannot be the case as the temperature is decreased and the volume does not change, then the pressure may have some role in allowing solutions in capillary tubes to go to supercooled temperatures. Thus, future experiments will test whether pressure is involved in keeping solutions from freezing as the temperature is decreased.
71
5 mm NMR Tube
Shigemi Tube
Capillary Tube
Figure 3.9 Calculations of Area/Volume (A/V) of a 5 mm NMR tube with 500 µL, a 5 mm Shigemi® tube with 300 µL, and a 0.1 cm capillary tube with 27 µL. Both of the 5 mm tubes have an A/V of 3 cm-1. The capillary tube has an A/V of 20 cm-1, which has a larger increase in the surface area versus volume than the NMR tube and Shigemi® tube and may contribute into allowing the RNA in aqueous solution to go to supercooled temperatures.
72
3.3 Conclusions In conclusion, NMR data at low temperatures allow us to identify imino protons, such as the ψ N3H in ψBP and terminal protons, which are undetected at higher temperatures due to rapid exchange. This technique has been helpful in refinement of the structure of ψBP, and may be in the future for other non-base paired regions of novel RNA motifs. It may extend the temperature range over which chemical shift data can be acquired. Applying the capillary technique appears to be promising for NMR in supercooled water using other interesting biomolecules.
73
CHAPTER 4 EQUILIBRIUM INTERACTION OF p14 WITH BRANCH SITE RNA
4.1 Introduction Assembly of the U2 snRNP involves the pre-mRNA, U2 snRNA and numerous protein components. Recognition of the U2 snRNA with a conserved region of the pre-mRNA and a protein component of U2 snRNP, called SF3b p14 (referred to as p14) is of vital importance for the assembly of the functional core of the spliceosome. p14 has been shown to cross-link to the branch site A through a benzophenone moiety tethered to the branch site A and persist through the first step of splicing (MacMillan et al., 1994; Query et al., 1996; Query et al., 1997; Will et al., 2001). This 14.6 kDa protein (125 amino acid residues) is one of seven proteins comprising the SF3b component of the U2 snRNP complex (Will et al., 2002). Approximately 86 of the 125 amino acid residues of p14 are similar to an RNA recognition motif (RRM) comprising two α-helices and four β-strands ordered in β1-α1-β2-β3-α2-β4 (Nagai et al., 1990). Proteins containing an RRM have previously been reported to bind single stranded RNA (Birney et al., 1993; Hall, 2002). The affinity of RRM proteins for their target RNAs is generally low (Kd in the range of 10-6 M), but binding is relatively specific (Burd & Dreyfuss, 1994; Hall, 2002). Phylogenetic studies have shown that p14 is highly conserved among eukaryotes (Will et al., 2001). Knockout studies of yeast p14 homolog (Snu17p) find that the protein is not essential in vivo. However, yeast cells without Snu17p exhibited hindered growth and a defect in pre-mRNA splicing (Gottschalk et al., 2001). The co-precipitation of Snu17p with pre-mRNA suggests that it is involved in the first step of splicing (Gottschalk et al., 2001). These results are consistent with the reported interaction (via cross-linking data) of the branch site RNA with p14 persisting through at least the first step of splicing (MacMillan et al., 1994; Will et al., 2001). Overall, these findings suggest that p14 may be
74
instrumental in the assembly of functionally important elements of the functional core of the spliceosome and/or in the splicing process. Several structures of p14 have been reported in the literature recently. A crystal structure of p14 with a small segment of SF3b155 shows that p14 has an anti-parallel β1-β3 sheet at its core and that the short β-strand of SF3b155 interacts with β3 and α4 of the p14 (Schellenberg et al., 2006). Schellenberg et. al. also showed that Y22 of p14, which is part of the β1 strand, induced a covalent cross-link to the branch site A (Schellenberg et al., 2006). Using NMR chemical shifts, a homology model of p14 was proposed, which shows β3', β3" hairpin of p14 interaction with the branch site region (Spadaccini et al., 2006). This contradicts the conclusion presented by Schellenberg et al. (2006) and other proteins containing an RRM that have been shown to bind RNA through their β1β3 anti-parallel sheet (Gorlach et al., 1992). Moreover, both models of p14 suggest that the β1-β3 anti-parallel strand appears to be the surface that interfaces with the p14-binding domain of SF3b155 (Schellenberg et al., 2006; Spadaccini et al., 2006), suggesting that it contacts the RNA elsewhere. Therefore, there are many unsolved questions about what role p14 has at the spliceosomal core. Previous studies do not address the question of whether p14 specifically recognizes branch site RNA in solution as opposed to being positioned nonspecifically near RNA in the context of the assembled U2 snRNP. This is an important question because it addresses the function of this RRM protein during spliceosome assembly. In the case of specific recognition, the question that remains is whether p14 prefers single- or double-stranded RNA; if doublestranded, does it recognize the ψ-dependent branch site motif over the unmodified branch site motif? In order to determine whether p14 recognizes the branch site RNA directly in solution, we have investigated the equilibrium binding of p14 to the pre-mRNA branch site. We have found that p14 binds to the intron strand alone in solution, that binding is enhanced when the intron is paired with the U2 snRNA, and that binding is most specific and of greatest affinity in the presence of ψ in U2snRNA
75
in its phylogenetically conserved position. These data are in marked contrast to the generally held view that RRM proteins recognize only single-stranded RNA (Birney et al., 1993; Hall, 2002). We propose that specific recognition of the branch site RNA by p14 is an important step in the spliceosome assembly. 4.2 Results 4.2.1 Analysis of p14. SF3b p14 is a 125 amino acid residue spliceosomal protein with a molecular weight of 14.6 kDa (calculated from ExPASy ProtParam, http://us.expasy.org/tools/protparam.html). MGMQAAKRAN IRLPPEVNRI LYIRNLPYKI TAEEMYDIFG KYGPIRQIRV GNTPETRGTA YVVYEDIFDA KNACDHLSGF NVCNRYLVVLYYNANRAFQK MDTKKKEEQL KLLKEKYGIN TDPPK Network Protein Sequence Analysis (NPS@) predicts a secondary structure motif of β1-α1-β2-α2-α3-β3 (β and α motifs are color coded to match amino acid residues in sequence above). The residues 10-95 (underlined in the sequence above) correspond to those recognized as belonging to the RRM family, which typically consists of two α-helices and four β-strands with an order of β1-α1-β2-β3α2-β4 (Nagai et al., 1990; Birney et al., 1993). The full-length protein has a calculated isoelectric point (pI) of 9.4. We have not been able to express and purify the truncated versions: p14(10-95), p14(1-105), and p14(10-125); suggesting that both the N-terminus and C-terminus are needed to stabilize p14. In order to analyze the tendency of regions of p14 to be unstructured, we used PONDR® (Romero et al., 1997a; Romero et al., 1997b; Li et al., 1999; Romero et al., 2001) (http://www.pondr.com/) and applied this algorithm to the primary sequence of p14. Calculated by these algorithms, a probability of 0.5 or greater indicates significant disorder The results from three PONDR® algorithms, VLXT, XL1_XT, and VL3, show the unstructured nature of p14 by predictions of unstructured residues at both the N- and C-terminus (Figure 4.1). Even though both the N- and C- terminus of p14 are predicted to be unstructured
76
and since we cannot make truncated versions of p14, we concluded that both the N- and C- terminal regions must be present for protein viability. Initial observations of p14 revealed that it was precipitated at room temperature in phosphate buffer. We performed a buffer screen in order to determine the optimal buffer conditions for p14. The solubility of p14 was assessed using different buffers, salt concentrations, pH values, and temperature conditions by looking for a visible precipitate. 15 mL of p14 were dialyzed in one liter of buffer at 4 °C overnight and checked for visible precipitation. Following overnight dialysis, samples that did not visibly precipitate were concentrated to 300-500 µM at 4 °C and the level of aggregation was again assessed by observation and measuring concentration from the extinction coefficient of p14 (calculated at www.expasy.org). From these studies, we determined that only four buffers fostered a concentration of 200 µM or greater without precipitation (Table 4.1) and that p14 is only stable at temperatures below 10 °C. After desirable solubility conditions for p14 were determined, we tested whether p14 was a monomer in particular buffers by dynamic light scattering (DLS). As shown in Figure 4.2, it was determined by DLS that p14 is a monomer in 10 mM BES (or 10 mM HEPES), 50 mM H3BO3 and 400 mM NaCl at pH 6.5. The hydrodynamic radius of p14 was measured to be 1.99 nm, with an estimated molecular weight of 17.44 kDa, which is slightly higher than the calculated molecular weight of the 14.6 kDa p14. Since, it was advantageous to work at lower concentration, it was determined that p14 was stable in 5 mM BES, 10 mM H3BO3, and 100 mM NaCl at pH 6.5 for low concentrations (below 200 µM) of p14.
77
Figure 4.1. Probability of unstructured residues of p14. Shows the probability of unstructured residues of p14 by the algorithm, PONDR®(Romero et al., 1997a; Romero et al., 1997b; Li et al., 1999; Romero et al., 2001) (http://www.pondr.com/). VLXT, XL1_XT and VL3 are different algorithms used by PONDR®. A probability of 0.5 or greater indicates significant disorder and both the N-terminus and C-terminus are predicted to be very disordered. The different curves use different algorithms, but all make the same qualitative point. Therefore, we consider it likely that both the N- and C- terminal regions must be present for protein viability.
78
Table 4.1 Buffer conditions of p14. Those buffers with one asterisk (*) allowed p14 to achieve concentrations of greater than 0.2 mM for two days at 4 °C. The buffers with two asterisks (**) enabled p14 to be stable at concentrations above 0.2 mM for a few weeks at 4 °C.
Buffers [buffer] [salt] *BES 10 mM 400 mM *Boric Acid 10 mM 400 mM (H3BO3) CAPS 10 mM 400 mM Citric Acid 10 mM 400 mM *HEPES 10 mM 400 mM phosphate 10 mM 200 mM PIPes 10 mM 200 mM ACES 10 mM 400 mM MES 10 mM 400 mM MOPS 10 mM 400 mM salt water 400 mM ADA 10 mM 400 mM HCO3¯ 10 mM 400 mM L-Arg/LGlu/ H3BO3 50 mM 400 mM 10 mM/ **BES/ 50 mM 400 mM H3BO3 **HEPES/ 10 mM/ 50 mM 400 mM H3BO3
pH 6.5
highest [p14] precipitated 0.4 mM no
6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 5.5
0.3 mM 0.4 mM 0.2 mM -
no yes yes no yes yes yes yes yes yes yes no at low pH
6.5
-
yes
6.5
0.4 mM
no
6.5
0.5 mM
no
79
p14
solvent
Figure 4.2 DLS graph of percent intensity verses hydrodynamic radius (Rh) shows two peaks. Sample used was 4 mg/mL p14 in 10 mM BES, 50mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer. The first peak at ~0.05 nm is attributed to the solvent peak and the peak at 1.99 nm is attributed to monomeric p14 with an estimated molecular weight of 17.44 kDa, which is slightly higher than the calculated molecular weight of the 14.6 kDa p14.
80
4.2.2 Interaction of RNA and p14 by nitrocellulose membrane retention. We used nitrocellulose membrane retention (Hall & Stump, 1992; Wong & Lohman, 1993) to determine binding constants of intron RNA to p14 in the absence and presence of U2 snRNA. In order to measure the Kd of p14 and RNA, we used nitrocellulose membrane filter binding studies and monitored binding of 32P-labeled intron and p14 with and without U2 snRNA (unmodified and ψ-modified). Each experiment was repeated three to five times and was done by mixing 0.5 µM of 32P-labeled intron, 0.5 µM of U2 snRNA (in appropriate experiment) increasing concentrations of p14 (0.5 µM to 50 µM). The fraction of RNA bound was calculated using the average counts from each reaction divided by the total unbound RNA. The dissociation constant (Kd) was calculated by using the following equation: Kd = ([RNA]-[RNA:Protein])([Protein]-[RNA:Protein])/[RNA:Protein complex] First we performed experiments in high salt because the protein was stable at higher concentrations and NMR experiments were performed in 400 mM NaCl. Then, after it was determined that p14 was stable at lower salt concentration, we used the buffer with lower salt. Under conditions of high salt, required for some NMR experiments, we measured a Kd of 271 ± 30 µM for p14 to single-stranded intron, 239 ± 30 µM for p14 to the double-stranded branch site with unmodified U2 snRNA, and 217 ± 25 µM for p14 to the branch site duplex with ψ-modified U2 snRNA (Table 4.2). These values show an uncharacteristically low affinity of RRM proteins for RNA. However the preference for the ψ-modified branch site duplex over the single-stranded intron or unmodified branch site duplex was considered remarkable. These data were corroborated by gel retardation studies that will be presented in the next chapter. Also, a possible source of error was that the RNA was dried onto the nitrocellulose membrane in order to obtain a count for unbound RNA and not measured directly in the sample well using a HybondTM-N+ membrane (Amersham, Piscataway, NJ).
81
Table 4.2 Dissociation constant (Kd) of p14 to single-stranded intron and double-stranded branch site RNA components under different buffer and salt conditions by nitrocellulose membrane retention. Note there is no significant difference between the dissociation constants in 100 mM salt, but the larger experimental errors are due to not using a HybondTM-N+ membrane (Amersham, Piscataway, NJ) in order to directly measure the unbound RNA.
Kd for p14 to: Conditions at pH 6.5: 10 mM BES, 50 mM H3BO3, 400 mM NaCl 10 mM BES, 50 mM H3BO3, 100 mM NaCl 10 mM BES, 50 mM H3BO3, 100 mM KCl 5 mM BES, 10 mM H3BO3, 100 mM NaCl
intron
intron/U2
intron/U2 (ψ)
271 ± 30 µM
239 ± 30 µM
217 ± 25 µM
37 ± 35 µM
-
-
40 ± 35 µM
-
-
35 ± 10 µM
28 ± 9 µM
19 ± 9 µM
82
We then tested binding equilibria under more physiological conditions, using 5 mM BES, 10 mM H3BO3, and 100 mM NaCl at pH 6.5 and a HybondTMN+ membrane (Amersham, Piscataway, NJ) to obtain a more accurate count of unbound RNA. With these changes, we obtained a very reliable profile of binding between protein and the RNA ligands. The interaction of p14 and the singlestranded intron had a Kd of 35 ± 10 µM, p14 bound to the double-stranded branch site with unmodified U2 snRNA with a Kd of 28 ± 9 µM, and, intriguingly, p14 now interacted with the branch site duplex with ψ-modified U2 snRNA with a Kd of 18 ± 9 µM for p14, i.e. an affinity twice that of the single-stranded intron (Table 4.2). Since the high salt concentration does affect the binding of p14 to the branch site RNA, these results may be indicative of a certain degree of electrostatic interaction between the two. Measurements made in different monovalent salts indicated no significant difference between NaCl and KCl (Table 4.2).
4.2.3 Equilibrium studies of p14 and RNA interactions using NMR spectroscopy. We wanted to obtain structural information about the branch site RNA with the addition of p14. The goal of these experiments is to characterize the changes in RNA when p14 is added and the next section will allow us to characterize the changes of p14 when the RNA ligands are added. Although we designed the RNA sample to mimic a single-stranded intron with no secondary structure, we submitted the sequence to mFold (www.idtdna.com) to determine if there is any predicted secondary structure. The algorithm uses nearest neighbor thermodynamic approximation and predicted two possible conformations (Figure 4.3). The two conformations form unusual structures and non-Watson-Crick base pairs. One is predicted to have 4 base pairs with a ∆G= -3.6 kcal/mol and the other is predicted to have six non-continuous base pairs with a ∆G= -2.1 kcal/mol.
83
∆G= -3.6 kcal/mole
∆G= -2.1 kcal/mole
Figure 4.3 Representative diagrams of possible secondary structures of the 26 nucleotide intron. Using nearest neighbor thermodynamic rules, mFold (www.idtdna.com) predicts two structures with a ∆G < -2 kcal/mole for the following sequence 5’- GGGCCUGGCUUUUUUUACUGACACUU-3’ that represents the pre-mRNA intron strand.
84
In order to observe changes in RNA conformation upon binding of p14 without spectral interference of the protein, we monitored NMR spectra of the imino proton region (10 -15 ppm). One-dimensional NMR spectra of the singlestranded intron RNA (Figure 4.4A) and the branch site duplex (Figure 4.5A) were collected on the 500 MHz instrument at 4°C. The 1H-NMR spectrum of 300 µM single-stranded intron RNA shows seven resonances in the imino region at ~10.2, ~10.7, ~11.1, ~12.0, ~12.9, ~13.3, and ~13.5 ppm (Figure 4.4A) suggesting that the RNA does fold, perhaps to form structures such as those shown in Figure 4.3. In general, resonances between 12 and 15 ppm are typical of imino protons involved in Watson-Crick base pairs and resonances between 10 and 12 ppm are typical of imino protons involved in non-Watson-Crick base pairs. However, upon adding stoichiometric amounts of p14 (Figure 4.4B) five resonances appeared at different locations or are undetectable. This indicates a change of the intron RNA conformation upon addition of p14 due to the changes in all imino protons except for a broad peak at ~10.7 ppm and another peak at ~13.5 ppm. There are two possible explanations for these small resonance peaks. One explanation is that the imino protons of RNA are protected by forming hydrogen bonds with the protein, which is probably unlikely because this would indicate that p14 is binding to the Watson-Crick face of the RNA. Since the preferred binding target is double-stranded RNA meaning the Watson-Crick face is not accessible for binding to p14 and assuming that p14 would recognize single- and double-stranded RNA in the same mode, we propose that p14 binds the negatively charged backbone or major groove of the branch site RNA. Alternatively, the peaks are remnants of RNA base pairs that were broken when p14 was added. The specific binding of p14 to the RNA will be studied by NMR using a 15N-labeled RNA transcript representing the pre-mRNA. We then obtained spectra of the imino region of the ψ-modified branch site displaying chemical shifts of: ψ6 N1H at 10.4 ppm, G4b at 11.1 ppm, G11b at 11.8 ppm, G6b 12.0 ppm, G5 at 12.7 ppm, U18b at 12.9 ppm, U22b at 13.1 ppm, G3 at 13.1 ppm, G8 13.5 ppm, U28b at 13.5 ppm, U4 at 13.7 ppm, U19b at 14.2 ppm, and U9 at 14.4 ppm (Figure 4.5B). The chemical shifts of the
85
A intron RNA
B
intron RNA +p14
Figure 4.4 Imino proton spectra of (A) 26 mer pre-mRNA intron and (B) intron with p14. Collected at 4 °C acquired on the 720-MHz Varian Unity Plus spectrometer (National High Magnetic field Laboratory (NHMFL), Tallahassee, FL) with 256 scans. Addition of the protein results in significant changes to the RNA. When the RNA was alone, it displayed signs of self-folding or pairing. However, in the presence of p14, far fewer resonances were visible, suggesting that the residual structure that was formed disappeared.
86
A
B
Figure 4.5 Imino proton spectra of (A) ψ-modified branch site duplex (pre-mRNA intron-U2 snRNA) (B) ψ-modified branch site duplex with p14. Collected at 4 °C acquired on the 720-MHz Varian Unity Plus spectrometer (National High Magnetic field Laboratory (NHMFL), Tallahassee, FL) with 256 scans. The intron strand representing the intron is labeled with “b” to distinguish it from the strand representing U2 snRNA. The peaks were assigned with the help of a short branch site duplex assigned previously (Newby & Greenbaum, 2001) and with the help of a longer duplex representing the branch site (Popović et al., unpublished data). The spectrum with the duplex and p14 (B) shows that some peaks get broader when p14 is added, which is consistent slower tumbling of the molecule indicating binding between the two. Upon binding of p14 to the branch site duplex, the increase in intensity of U22b and the appearance of ψ6 N3H provide evidence of stabilization of the branch site region near the branch site A.
87
imino protons in the branch site duplex in this study are very similar to the chemical shifts of the imino protons in the branch site duplex from previous studies in our laboratory (Newby & Greenbaum, 2001, 2002b). Even though the strands are longer, the duplex pairing is the same as in the duplex Newby and Greenbaum used (2001, 2002b), with the only difference being that the 5’-end of the intron folds or self-pairs. Upon addition of equimolar amount of p14 (300 µM) (Figure 4.5B), no chemical shift changes were observed in spectra indicative of no major structural change of the ψ-modified pre-mRNA branch site duplex. However, small changes and shifts in resonance locations were noted, as well as broader line-widths of imino protons of bases in the boxed region around the branch site A (inset in Figure 4.5). Specifically, stabilization of U22b at 13.1 ppm, shifts of G8 to 13.4 ppm and U4 to 13.8 ppm, and the appearance of the imino proton of the unpaired ψ6 N3H at 10.8 ppm, which is normally exchange broadened beyond detection. Both imino spectra of intron RNA-p14 (Figure 4.4B) and the branch site duplex-p14 (Figure 4.5B) indicate that p14 binds the RNA under equilibrium conditions and there is no global change of the branch site duplex upon addition of p14. Therefore, the structural motif of the branch site duplex does not change, but is stabilized in the context of p14.
4.2.4 Equilibrium studies of p14 with the pre-mRNA branch site duplex using circular dichroism (CD). In order to obtain information about changes in secondary structure of p14 and the pre-assembled ψ-modified branch site duplex, CD experiments in the far UV region were performed at 4 °C (Figure 4.6). Each molecule was prepared to the final concentration of 10 µM (200 µL) in 10 mM BES, 50 mM H3BO3, and 400 mM NaCl @ pH 6.5 buffer. The branch site duplex was scanned alone and displayed characteristics of a double stranded RNA with maxima at ~270 nm and ~218 nm and a minimum at ~240 nm. Then, p14 was scanned alone and had minimum at 212 nm, which is a mixture of random coil and α-helix. Then, p14 was added to the branch site duplex in the same cuvette and after being scanned displayed a minimum at 225 nm
88
Figure 4.6 CD spectra of p14 with an RNA duplex. The red line indicates the spectrum of p14 alone. The green line indicates the spectrum of RNA duplex (shown in the corner). The blue line indicates the spectrum of the mixture of p14 and the pre-RNA duplex. The black line shows the sum the spectrum of p14 (red) and the RNA duplex (green) indicating what the spectrum would be like if there was no interaction. These spectrum show us that the secondary structure of the ψ-modified branch site RNA duplex does not change because there is only a slight increase in ellipticity at 270 nm, but that p14 may change due to the differences between the addition of the two traces and the trace of the two molecules together in the same cuvette. Therefore, since there is a difference between the blue line and black line, this could mean there is some level of interaction between the p14 and the pre-mRNA duplex.
89
and a maximum at approximately 270 nm. The trace of p14 and branch site duplex was very different than the added traces of the individual p14 and branch site duplex (Figure 4.6). These data suggest that the secondary structure of the ψ-modified branch site duplex does not change when p14 is added because the region attributed to an RNA A-type helix (the region between 255 nm and 285 nm) does not drastically shift and the slight increase in ellipticity indicates that the RNA complex is becoming more stable. The protein alone as predicted by CONTIN (Bobba et al., 1990; Lobley et al., 2002), shows that p14 alone has approximately 5 % α-helical character, 35 % β-sheet character, and 60 % random coil character. Because of the different spectral properties of RNA and protein between 210 nm and 225 nm, we cannot predict how much of the secondary change is due to p14. However, it can be concluded that since the difference between the trace of p14 and the branch site duplex is different than the added traces of the two molecules alone, the two molecules are interacting under equilibrium conditions. 4.2.5 Structural studies of 15N-labeled p14 by NMR spectroscopy. In order to determine the structural features of p14, we performed NMR studies of p14 in the absence and presence of both intron RNA and SF3b155. A 1H-15N Heteronuclear Single Quantum Coherence (HSQC)(Kay, 1992; Kay et al., 1992) experiment used to assess structural features by identifying individual spin systems. A 1H-15N HSQC provides correlation between a 15N-labeled nitrogen atom and its attached proton. If the protein is properly folded, this experiment will provide one resonance per individual spin systems because the protein backbone has only one amide (HN) group. There are two exceptions: 1) proline (P) residues, the amide nitrogen of which is not protonated; and 2) some Nterminal amide residues, which undergo rapid exchange with the solvent, and are therefore not observable. One factor that can limit the ability to identify all amide backbone resonances is spectral overlap, in which more than one resonance has the same 1H-15N chemical shift. If the protein has a majority of the proton resonances overlapped in the region between 8 to 8.5 ppm and it is difficult to
90
discern individual resonances, then it is an indication that the protein is unfolded or unstructured. If the spectrum displays collapsed peaks in a ±1 ppm range of 8.5 ppm, then it is an indication that the protein has a highly helical nature. However, if the spectrum displays proton resonances in the range from 6.5 to 10.5 ppm, then it is an indication that the protein is folded and usually has some β-sheets present in the protein structure. Also, amino (-NH2) nitrogen atoms of protein side chains usually resonate near ~112 ppm in the 15N-dimension and the amino protons resonate between 6.8 and 7.6 ppm in the 1H-dimension with glutamine and asparagine residues between 6.8 and 7.2 ppm in the 1Hdimension. First, a 1H-15N HSQC spectrum of a 15N-labeled p14 revealed ~50% of the anticipated 125 cross peaks and limited spectral dispersion, suggesting that p14 is partially unstructured (Figure 4.7). Addition of intron RNA to 15N-labeled p14 yielded ~65% of the possible 125 amide cross peaks and additional spectral dispersion in the α-helical region (Figure 4.7). Addition of SF3b155(255-462), previously shown to bind p14, to 15N-labeled p14 resulted in the appearance of ~60% of the possible 125 amide cross peaks and different cross-peaks as compared to with RNA(Figure 4.7). Finally, addition of both intron RNA and SF3b155 (255-462) to 15N-labeled p14 at a 1:1:1 ratio resulted in detection of ~75% of the possible 125 amide cross peaks of p14 and shifting of a number of the original p14 cross peaks (Figure 4.7). These data provide evidence that p14 undergoes major conformational change upon binding of SF3b155 and its cognate RNA. However, the spectral dispersion is not sufficient for comparison studies, but we can conclude from these data that p14 is unstructured and does change when one adds both an intron strand and SF3b155 separately or together. Therefore, we need to find better conditions for the protein in order to improve spectral dispersion. 4.2.6 Structural studies of 2H-13C-15N-labeled p14 by NMR spectroscopy. One of the problems encountered in NMR spectroscopy of unlabeled and 15Nlabeled proteins is spectral overlap because of rapid exchange.
91
A
B
C
D
Figure 4.7 NMR spectra of p14 with addition of intron RNA and SF3b155(255462). (A) 1H15N-HSQC of 15N-labeled p14. (B) 1H15N-HSQC of 15N-labeled p14 alone (black) with intron RNA (green). (C) 1H15N-HSQC of 15N-labeled p14 alone (black) with SF3b155(255-462)(green). (D) 1H15N-HSQC of 15N-labeled p14 alone (black) with intron RNA and SF3b155(255-462) (green). All samples were 300 µM and collected at 4 °C on a 720 MHz Varian spectrometer (NHMFL). However, the spectra above show that p14 is unstructured, but the side chain region shows a slight shift in spectra B and D. Also, the amide backbone region between 8 and 8.5 ppm in spectra B, C, and D show slight shifts suggesting there is a change in structure when binding occurs. 92
In order to slow down the chemical exchange and possibly improve spectral dispersion, p14 was grown in deuterated media in order to have either a 2H-13C15
N-labeled protein or 2H-15N-labeled protein. This will allow us to perform more
NMR experiments, such as triply labeled experiments and take advantage of the deuterium on the triply-labeled protein. The 1H-15N HSQC spectrum of 2H-15Nlabeled p14 (Figure 4.8) reveals ~50% of the amide cross peaks and has greater spectral dispersion than with the 15N-labeled p14. By comparison, the 1H-15N HSQC spectrum of 2H-13C-15N-labeled p14 (Figure 4.9) reveals ~80% of the amide cross peaks and has greater spectral dispersion than with the 15N-labeled p14. The 2H-13C-15N-labeled p14 produces a better 1H-15N HSQC spectrum because deuteration of p14 slows down the chemical exchange of the amide protons with the solvent and allows for decreased line-widths. The reduction of the line-widths allows for sharper cross-peaks, which helps in distinguishing resonances that may have been previously overlapped. Since the 1H-15N HSQC spectrum of the 2H-13C-15N-labeled p14 looked promising, a 1H-15N-Transverse-Relaxation Optimized Spectroscopy (TROSY)HSQC experiment was performed. A 1H-15N-TROSY-HSQC experiment detects the narrowest component of the four non-decoupled HSQC components by canceling of transverse relaxation caused by chemical shift anisotropy (CSA) and by dipole-dipole coupling due to a magnetic field (Pervushin et al., 1997). The difference between an HSQC and a TROSY-HSQC is that an HSQC produces spectra with two doublets cross-peaks collapsed to a singlet by decoupling in both the 1H and 15N dimension. TROSY-HSQC does not use decoupling, but uses a selective pulse to observe the most relaxing component of the two doublets of a protein amide. A 1H-15N-TROSY-HSQC spectrum of 2H-13C-15N-labeled p14 (Figure 4.10) yielded ~80% of the cross-peaks, but did not show side chain resonances of amino acid residues. A 1H-13C-HSQC with 2H decoupling of 2H-13C-15N-labeled p14 acquired at 10 °C showed very good dispersion of the protons attached to 13
C-labeled atoms (Figure 4.11). These data suggested that three-dimensional
NMR data collection of p14 in solution is possible. Therefore, future experiments
93
Figure 4.8 1H15N-HSQC of 2H-15N-labeled p14 at 10 °C on a 720 MHz Varian spectrometer (NHMFL). The relatively narrow proton chemical shift dispersion and the ~60 of the anticipated 125 cross-peaks of p14 indicated that p14 is partially unstructured under these conditions.
94
G43
Y22
Figure 4.9 1H-15N-HSQC of 2H-13C-15N-labeled p14 at 10 °C on a 900 MHz Bruker spectrometer (NHMFL). There are several reasons that this spectrum is better improved over Figure 4.7 and 4.8. One contributing factor is sharper linewidths, the result of deuteration of p14, which slows down chemical exchange. There is considerably greater proton chemical shift dispersion in this figure as compared with Figure 4.7 and 4.8, suggesting that p14 is significantly more structured. Although it is not clear why this occurs, two possible reasons are the altered growth conditions of the triply labeled protein and changes in the chemical environment upon replacement of hydrogen atoms with deuterons, which may help make p14 more stable. This 1H-15N HSQC spectrum of 2H-13C15 N-labeled p14 reveals ~90 of the 125 amide cross peaks and has greater spectral dispersion than with the 15N-labeled p14 and 2H-15N-labeled p14. Y22 and G43 were identified using the chemical shifts of p14 published by Spadaccini et al. (2006).
95
Figure 4.10 1H-15N-TROSY-HSQC of 2H-13C-15N-labeled p14 at 10 °C on a 900 MHz Bruker spectrometer (NHMFL). This spectrum of 2H-13C-15N-labeled p14 reveals ~70-75% of the amide cross peaks. This spectrum shows smaller, sharper cross-peaks than the HSQC, but in this spectrum we do not observe the side chain cross-peaks.
96
Figure 4.11 1H-13C-HSQC with 2H decoupling of 2H-13C-15N-labeled p14 acquired at 10 °C on the 720 MHz with 32 scans and 512 increments. This shows very good dispersion of the protons attached to 13C-labeled atoms and is less noisy than when 1H-13C-HSQC is performed without 2H decoupling. These data suggest that three-dimensional NMR data can be acquired.
97
will embark upon collecting three-dimensional data sets in order to make chemical shift assignments of p14, which can be used with other restraints to solve a solution structure of p14. 4.2.7 Structural studies of 2H-13C-15N-labeled p14 with addition of branch site RNA by NMR spectroscopy. The next step was to measure spectral changes of p14 upon adding the branch site RNA. First, we added equimolar amounts of unlabeled 26-nucleotide intron to the 2H-13C-15N-labeled p14 and collected a 1H-15N-HSQC (Figure 4.12). We observed 110 amide cross-peaks of p14 (of the possible 125) upon addition of single-stranded RNA, as well as change of position of some of the existing cross-peaks, suggesting a partial increase in the structure of p14 when the single-stranded intron was added. These data are consistent with the binding studies that indicate the intron and p14 interact (with a Kd ~35 µM) under equilibrium conditions. The final goal was to measure the spectral changes of p14 upon addition of the ψ-modified branch site duplex, characterized by a Kd of 19 µM (section 4.2.2). An HSQC of 2H-13C-15N-labeled p14 following addition of equimolar amount of unlabeled RNA duplex was collected (Figure 4.13). We observed all 125 amide cross-peaks of p14 with the addition of ψ-modified branch site and noted that some cross-peaks moved (as compared with the protein alone and the protein complexed with the single-stranded intron RNA), suggesting that there is a further increase in structure of p14 upon addition of the ψ-modified branch site duplex. These data corroborate our previous studies suggesting that p14 preferentially recognizes the ψ-modified branch site RNA and undergoes an induced fit upon binding to it. This is an important finding because other RRM proteins (the class of which p14 is a member of) typically recognize single-stranded RNA. The finding that p14 recognizes the ψ-dependent branch site helix has fascinating implications in spliceosome assembly. These NMR spectra display dramatic changes in isotopically labeled p14 upon binding of the single-stranded intron and the ψ-modified branch site helix, indicating that major structural
98
Figure 4.12 In a 1H15N-HSQC spectrum of 2H-13C-15N-labeled p14 (red), 90 amide cross-peaks are visible, out of a possible 125, suggesting that the protein is partially unstructured in solution. Upon addition of an unlabeled 26 nucleotide RNA sequence representing the intron strand to labeled p14 (green), another 20 cross-peaks appear (87% of total now visible), and 15-20 of the existing peaks are shifted, suggesting an increase in, as well as some changes in, the structure of p14 (shown best in insert). Samples included 300 µM protein and RNA. Each experiment was collected at 10 °C and acquired on a 720-MHz Varian Unity Plus spectrometer with 64 scans and 256 increments.
99
Figure 4.13 The 1H15N-HSQC spectrum of uncomplexed 2H-13C-15N-labeled p14, with 90 amide cross-peaks out of a possible 125, is shown in red. A change upon addition of an equimolar amount of an unlabeled RNA duplex representing the ψ-modified U2 snRNA-intron branch site helix to labeled p14, and 125 amide cross-peaks (blue) were observed, and a number of cross-peaks had shifted locations in comparison with the previous spectra (insets). Each experiment was collected with 300 µM samples at 10 °C and acquired on the 720-MHz Varian Unity Plus spectrometer with 64 scans and 256 increments.
100
rearrangement occurs upon both single- and double-stranded branch site RNA is added.
4.3 Discussion and Conclusion These findings indicate that p14 has a higher affinity for the ψ-modified branch site duplex than for the single-stranded intron and the unmodified duplex. Our observation that p14 binds more tightly to the double stranded branch site RNA is in direct opposition to previously published data (Birney et al., 1993; Hall, 2002). Most significantly, the finding that p14 binds to the ψ-modified U2 snRNAintron with the highest affinity is very exciting because it may yield insight into why the ψ is conserved in the juxtaposed position to the branch site A. We observed no major change in the RNA duplex after analysis of our NMR spectra of the branch site RNA, which would suggest no major structural change upon interaction with p14. Stabilization of U22b and the appearance of the imino proton of the unpaired ψ6 N3H indicate that there is interaction between p14 and both ψ6 and U22b and more than likely the A24b, in the boxed region of the branch site duplex (Figure 4.5). However, our NMR data and nitrocellulose membrane retention data suggest that p14 does not bind the U2 snRNA strand. While the dissociation constant of p14 with the ψ-modified branch site duplex is relatively low and specific because it does not bind tRNA or DNA, the low affinity of p14 to its RNA target is consistent with other RRM proteins (Burd & Dreyfuss, 1994; Hall, 2002). We have determined that regions of p14 directly contact the ψ-modified branch site duplex. Determining the specific amino acid residues of p14 that bind the branch site duplex was beyond the scope of these studies. However, it has been proposed that Y22 of p14 interacts with the branch site A (Schellenberg et al., 2006). It has also been proposed that the β3', β3" hairpin of p14 interacts with the branch site region (Spadaccini et al., 2006). A more detailed analysis of the interaction between p14 and the branch site region will need to be conducted in order to pinpoint the exact amino acids involved in interaction of p14 with the branch site region. However, the hypothesis that the ψ-modified U2 snRNA 101
needs to pair the pre-mRNA should help shed some light on the binding mode of p14 to the branch site duplex. The most exciting aspect of my dissertation is that the addition of singlestranded intron or double-stranded branch site RNA shows an increase in structure of p14. Since p14 is a good candidate for structural studies because of size (less than 15 kDa), it was beyond the scope of this dissertation to solve the structure of p14 in solution. Like many other RRM proteins, p14 is notoriously difficult to solubilize (e.g. Schellenberg et al., 2006; which explains why structural studies have lagged behind many other proteins). Even though making sequential assignments was beyond the scope of this dissertation, the most significant discovery was that not only was this the first time that a protein containing an RRM was shown to bind double-stranded RNA, but that NMR spectrum of p14 bound to the branch site duplex shows that p14 undergoes a structural change upon binding. In conclusion, our data provide evidence that p14 binds specifically to the branch site RNA, demonstrating the greatest affinity for the double-stranded ψmodified U2 snRNA-intron duplex. This is the first time that a protein containing an RNA Recognition Motif was observed to bind double-stranded RNA. When binding occurs, it was shown that the chemical shifts of the branch site duplex are only perturbed in the region near the branch site A, whereas the other imino protons do not shift suggesting that the structure of the branch site duplex may only change slightly near the binding region. These data also suggest that p14 undergoes a structural change when its cognate RNA is added. This is consistent with data from other proteins that undergo structural change upon binding. This study may be aided by future studies which include: mutations of p14, residual dipolar couplings using longer representative RNA strands of the branch site duplex, and site-specific labeling of certain residues in p14. Y22 of p14 may be a good target for mutation studies and site-specific labeling because it has been shown to cross-link to the branch site A of the pre-mRNA (Schellenberg et al., 2006).
102
CHAPTER 5 EQUILIBRUM INTERACTION OF SF3b155 WITH p14 AND THE BRANCH SITE RNA
5.1 Introduction Assembly of the functional core of the eukaryotic spliceosome involves recognition among U2 snRNA, U6 snRNA, the pre-mRNA intron, and numerous U2 snRNP proteins. The U2 snRNP comprises two main splicing factor (SF) complexes called SF3a and SF3b. A subunit of these splicing complexes called SF3b155 has been shown to associate (via cross-linking interactions) with nucleotides +5 and -6 of the intron (positions with respect to the branch site adenosine (A)) (Gozani et al., 1998). Photo-cross-linking studies have shown p14 to be the only U2 snRNP protein in the immediate vicinity of the branch site A (MacMillan et al., 1994), whereas segments of SF3b155 interact with p14 (Will et al., 2001) and flanking regions of the intron (Gozani et al., 1998). Cryoelectron microscopy studies have suggested that p14 is surrounded and/or contacted by SF3b155 in the U2 snRNP functional core (Golas et al., 2003, 2005). The recent co-crystal structure of p14 with a short fragment of SF3b155 (residues 373-415) indicates that residues 410-415 of SF3b155 may interact with the β3-strand, which is contained in RNP1, or the C-terminus of the p14 (Schellenberg et al., 2006). This means that SF3b155 and the branch site RNA are predicted to bind the same region of p14. Spadaccini et. al. (2006) measured the interaction of SF3b155(282-424) and p14 by isothermal titration calorimetry (ITC). They reported that the interaction is exothermic and has a Kd of 9.6 nm (Spadaccini et al., 2006), which is very strong binding. Also, Spadaccini et al (2006) collected NMR spectra in order to define residues within SF3b155 that participate in the binding interface with p14. Their spectrum of 15N-labeled SF3b155(381-424) shows helical conformation upon addition of p14 for residues 410-414 and 419-423 of SF3b155 (Spadaccini et al., 2006).
103
We propose that direct recognition of the branch site by p14 is an important step in the assembly of the functional core of the spliceosome. Equilibrium studies presented in the last chapter indicate specific recognition and binding of the ψ-modified branch site helix by p14. We will now test whether that binding is further enhanced when a segment of SF3b155 known to bind p14 is added to the p14-RNA complex. The interaction between p14 and segments of SF3b155 will be examined by gel retardation, nitrocellulose membrane retention, circular dichroism (CD), fluorescence, ITC and NMR. Studies presented here indicate that the addition of SF3b155 enhances binding of p14 to the branch site RNA. CD spectra indicate that the interaction between p14 and SF3b155 (residues 199-462) is accompanied by an overall structural change in the proteins, although it is not possible to discern which protein exhibits specific changes. Isothermal titration calorimetry (ITC) studies show that the binding affinity between p14 and SF3b155(199-462) is weak and endothermic and the binding between p14 and SF3b155(1-462) is moderate and exothermic. NMR studies show a global change in p14 and what appears to be only selected regions of SF3b155.
5.2 Results 5.2.1 Analysis of SF3b155. The full-length SF3b155 protein has 1304 amino acid residues, but we are studying truncated versions of SF3b155 from which the region containing the 22 tandem PP2A-like repeats or the HEAT motif on the Cterminal end has been deleted. SF3b155(1-462) is the first 462 amino acid residues of SF3b155. The molecular weight of SF3b155(1-462) is 50.4 kDa (calculated by ExPASy, http://us.expasy.org/tools/protparam.html) with the following sequence: MAKIAKTHED IEAQIREIQG KKAALDEAQG VGLDSTGYYD QEIYGGSDSR FAGYVTSIAA TELEDDDDDY SSSTSLLGQK KPGYHAPVAL LNDIPQSTEQ YDPFAEHRPP KIADREDEYK KHRRTMIISP ERLDPFADGG KTPDPKMNVR TYMDVMREQH LTKEEREIRQ QLAEKAKAGE LKVVNGAAAS QPPSKRKRRW
104
DQTADQTPGA TPKKLSSWDQ AETPGHTPSL RWDETPGRAK GSETPGATPG SKIWDPTPSH TPAGAATPGR GDTPGHATPG HGGATSSARK NRWDETPKTE RDTPGHGSGW AETPRTDRGG DSIGETPTPG ASKRKSRWDE TPASQMGGST PVLTPGKTPI GTPAMNMATP TPGHIMSMTP EQLQAWRWER EIDERNRPLS DEELDAMFPE GYKVLPPPAG YVPIRTPARK LTATPTPLGG MTGFHMQTED RTMKSVNDQP SG Network Protein Sequence Analysis (NPS@) predicts a secondary structure motif of α1-β1-β2-β3-α2-β4-α3-β5-α4-α5-β6-β7-α6-α7-β8-β9-α8-β10-α9-α10-β11-α12 (β (blue) and α (red) motifs color-coded to match amino acid residues in sequence above). We have been successful in expressing and purifying SF3b155(1-462) and shorter versions thereof, SF3b155(199-462), and SF3b155(25-462). In order to predict the relative structured vs. unstructured regions of SF3b155, we used an algorithm called PONDR® (Romero et al., 1997a; Romero et al., 1997b; Li et al., 1999; Romero et al., 2001)(http://www.pondr.com) in order to predict the unstructured residues. By this approach, a probability of 0.5 or greater indicates significant disorder of certain amino acid residues. We first applied PONDR® to the full-length SF3b155 (Figure 5.1) and it predicted that the N-terminus of the full-length SF3b155 has an unstructured nature and the C-terminus is predicted to have a structured nature. PONDR® was also used to predict the unstructured residues of SF3b155(1-462) in the presence (Figure 5.2A) and absence (Figure 5.2B) of the C-terminus of the full-length SF3b155. Both the N-terminus and Cterminus of the truncated SF3b155 proteins from amino acids 1-462, 199-462 (data not shown) and 255-462(data not shown) are predicted to be very disordered. The different curves use different algorithms, but all make the same qualitative point that the N-terminus of full-length SF3b155 has unstructured character. These data indicate that working with a truncated SF3b155 is a good representation of the full-length SF3b155.
105
Probability of unstructured residues
1.0 VLXT XL1_XT VL3 0.8
0.6
0.4
0.2
0.0 0
200
400
600
800
1000
1200
Residue Number
Figure 5.1 Probability of unstructured residues of full-length SF3b155. VLXT, XL1_XT and VL3 are different algorithms used by PONDR®. A probability of 0.5 or greater indicates significant disorder and the N-terminus is predicted to be very disordered. This figure shows that the full length SF3b155 is predicted to have a fairly unstructured region in the first 500 amino acid residues, but the Cterminus, which contains the 22 tandem repeats, is predicted to be structured.
106
Probability of unstructured residues
A
1.0
VLXT XL1_XT VL3
0.8
0.6
0.4
0.2
0.0 0
Probability of unstructured residues
B
100
200
300
100
200
300
400
1.0
0.8
0.6
0.4
0.2
0.0
Residue Number
400 Num vs VLXT Num vs XL1_XT Num vs VL3
Figure 5.2 Probability of unstructured residues of SF3b155(1-462). VLXT, XL1_XT and VL3 are different algorithms used by PONDR®. A probability of 0.5 or greater indicates significant disorder. Top, predicted unstructured regions of SF3b155(1-462) in the context of full length SF3b155. Bottom, predicted unstructured region of SF3b155(1-462) alone.
107
5.2.2 Interaction of RNA, p14, and SF3b155 observed by gel electrophoresis. Equilibrium binding of p14 to single- or double-stranded RNA was first assessed by retardation of 32P-labeled RNA on a horizontal agarose gel (Figure 5.3A, Figure 5.4). A horizontal gel was used because the RNA and SF3b155, which has a pI of ~6 (calculated using http://us.expasy.org), are acidic molecules and migrate toward the anode. However, since p14 is a markedly basic protein (the unfolded protein has a pI of ~9.4 calculated using http://us.expasy.org), p14 migrates toward the cathode. In order to accommodate the proteins’ markedly different migration behavior on a gel as a result of their different properties, we prepared horizontal agarose gels placing the comb in the middle. The intensity of bands of radio-labeled RNA was monitored by phosphorimager. Single-stranded 32P-labeled intron migrated from the center of the gel to halfway on the gel towards the anode. In lanes in which 10 µM p14 was added, a weak new band was observed near the well. There was no change in the shifted band when tRNA or nonspecific protein (DtxR) was included in excess, implying that binding between p14 and the intron was specific. In order to estimate affinity of the p14-intron pairing, we measured band radioactivity in assays in which both p14 and/or unlabeled tRNA were titrated into 32
P-labeled intron to alter mole fractions. A Kd was calculated according to the
equation: Kd = [p14][intron]/[p14-intron] Under conditions in which 0.5 µM to 50 µM p14 was added to 0.5 µM 32P-labeled intron (Figure 5.3A), the Kd was calculated to be ~130 µM. This apparent binding affinity is approximately four-fold less than the biding affinity calculated by nitrocellulose membrane retention (Kd of ~35 µM) in Chapter 4.2.2. We consider this discrepancy to probably be an artifact due to the thickness of the gel. Therefore, this may cause an underestimated in reading of the radioactivity. We are also concerned about the possibility that some radioactivity may diffuse into the tank buffer because the shifted band stays near the well. Either of these affects may lead to erroneous results, in this cause an underestimate results.
108
A
B *
* intron unmod U2 mod U2 p14 Sf3b155
0.5 0.5 25 25
0.5 0.5 25 25
RNAprotein complex
Unbound RNA
* indicates 32P-labeled All conc. µM
Representation of *intron RNA
12-100 nt space
BP region
Figure 5.3 Representative gels of p14 to RNA and SF3b155. (A) This horizontal 1.88% agarose gel (thickness: 3mm) shows the retardation of 32P-labeled intron RNA in the presence of p14, SF3b155(199-462) and U2 snRNA. The results indicate that p14 binds intron RNA in the presence of SF3b155(199-462), but more tightly in the presence of U2 snRNA and even more tightly in the presence of SF3b155(199-462). The results also indicate that SF3b155(199-462) does not bind to the intron RNA that does not have a polypyrimidine track. Each reaction (with all concentrations mentioned above in µM) was carried out in 10 mM BES, 50 mM H3BO3, and 400 mM NaCl with 5% glycerol @ pH 6.5. (B) This 5% TRISMOPS PAGE shows that p14 in the presence of SF3b155(199-462) binds more tightly to the ψ-modified branch site duplex than to the unmodified (no ψ present) branch site duplex. Each reaction (with all concentrations mentioned above in µM) was carried out in 10 mM BES, 50 mM H3BO3, and 400 mM NaCl with 5% glycerol @ pH 6.5.
109
32
0.5µ
0.5µM
0.5µM
0.5µM
0.5µM
0.5µM
0.5µM
0.5µM
p14
-
25µM
25µM
25µM
-
25µM
25µM
25µM
SF3b155
-
-
25µM
25µM
-
-
25µM
25µM
U2 snRNA
-
-
-
-
0.5µM
0.5µM
0.5µM
0.5µM
tRNA
-
-
-
1.0µM
-
-
-
1.0µM
P-intron
RNAprotein complex
Unbound RNA
Figure 5.4 Representative gel of p14 to RNA, tRNA and SF3b155. A horizontal 1.88% agarose gel (thickness: 3mm) showing the retardation of 32P-labeled intron RNA in the presence of p14, SF3b155(199-462) and U2 snRNA. The results indicate that p14 binds intron RNA, but more tightly in the presence of U2 snRNA and even more tightly in the presence of SF3b155(199-462). The presence of tRNA in the reaction does not seem to have an affect on binding of p14 to intron RNA. Each reaction (with all concentrations mentioned above in µM) was carried out in 10 mM BES, 50 mM H3BO3, and 400 mM NaCl with 5% glycerol @ pH 6.5.
110
Therefore, as will be shown in the next gels, these qualitative results show the same trend as nitrocellulose membrane retention, so we will continue to report these qualitative data because the error is consistent with all experiments on the horizontal agarose gels. We repeated these experiments with the double-stranded ψ-modified branch site duplex and the shifted band was more intense that with the singlestranded intron. The Kd of p14 to the ψ-modified branch site RNA was ~110 µM. Once again the absolute value is considerable greater than that measured by nitrocellulose membrane retention (Kd ~19 µM); however it is consistent with the pattern that p14 binds the double-stranded more with more affinity than singlestranded RNA. We chose to use SF3b155(199-462) on the assumption that it did not bind the intron strand alone. We assessed the effect of segments of SF3b155 on RNA and found that there was no shifted band when SF3b155(199-462) was mixed with both single- and double-stranded RNA. Thus, indicating that there is no direct interaction between SF3b155(199-462) and the RNA strand. In order to determine whether the presence of SF3b155(199-462) alters the affinity of p14 for its target RNA, we repeated the above horizontal gel experiments in the presence of SF3b155(199-462). The addition of equimolar amounts of SF3b155(199-462) and p14 to 32P-labeled intron RNA yields higher intensity bands near the gel well than with only p14 and 32P-labeled intron RNA (Figure 5.3 and 5.4). The Kd of SF3b155(199-462) and p14 for the singlestranded intron was ~100 µM. These data suggest binding of p14 to the intron RNA is enhanced in the presence of SF3b155(199-462). A similar result was achieved with the ψ-modified branch site helix in the presence of SF3b155(199462), which was measured to have a Kd of ~ 30 µM. In a separate set of experiments, we used a 5% TRIS-MOPS nondenaturing polyacrylamide gel (PAGE)(as opposed to the horizontal agarose gels previously described), to test specifically test the impact of the conserved ψ modification of RNA on binding to the p14-SF3b155(199-462) complex. From resulting data, we calculated Kd the dissociation constant of p14-SF3b155(199-
111
462) for the ψ-modified branch site duplex was measured to be ~100 µM as compared to ~200 µM of p14-SF3b155(199-462) to the unmodified branch site duplex (Figure 5.3B). This tells us that the presence of the ψ- modified branch site motif is an important structural determinant in assembly of this. We know that the PAGE gels suffer from the see problems as the agarose gels. However, we see that the qualitative results are clear. Overall, the data collected from gels lead us to believe that p14 prefers the ψ-modified branch site duplex over the intron RNA and unmodified branch site duplex and that SF3b155 enhances binding of p14 to its cognate RNA.
5.2.3 Interaction of RNA and p14 with SF3b155 by nitrocellulose membrane retention. We used nitrocellulose membrane retention assays to determine binding constants of double-stranded branch site duplex in presence of SF3b155. We performed the experiments as described in Material & Methods and Chapter 4. Each experiment was repeated three to five times and by mixing 0.5 µM of 32
P-labeled intron and U2 snRNA to increasing concentrations of p14 and
SF3b155 from 0.5 µM to 50 µM. The fraction of RNA bound was calculated and the Kd was calculated as described previously. In Chapter 4, we reported that the Kd of p14 for the ψ-modified branch site helix was 19 µM; upon addition of SF3b155(199-462) at the same concentration as p14 in low salt BES buffer, the Kd was calculated at 15 ± 5 µM (Table 5.1). The same experiment was performed in high salt BES buffer, in which the Kd for p14 to the RNA helix was 217 µM. Upon addition of SF3b155(199-462), the Kd was 166 ± 32 µM in (Table 5.1). These data fully support the conclusion that presence of this segment of SF3b155 enhances the affinity between p14 and its RNA target.
5.2.4 Equilibrium studies of p14, intron RNA, and SF3b155 using circular dichroism. To determine whether SF3b155(199-462) enhances binding between p14 to the intron RNA, circular dichroism experiments at far UV wavelengths were performed at 4 °C (Figure 5.5). Each molecule was prepared with a concentration of 10 µM in 200 µL buffer (10 mM BES, 50 mM H3BO3, and 112
Table 5.1 Binding affinities (Kd) of p14-SF3b155(199-462) to the ψ-modified branch site duplex in both high salt (400 mM) and low salt (100mM NaCl) BES buffer at pH 6.5 by nitrocellulose membrane retention.
Complex
Kd
*intron//U2(Ψ)/p14/SF3b155(199-462) in high salt *intron//U2(Ψ)/p14/SF3b155(199-462) in low salt
113
166 ± 32 µM 15 ± 5 µM
Figure 5.5 CD spectra of p14 with intron RNA and SF3b155(199-462). The green line indicates the spectrum of the mixture of p14 and intron RNA. The blue line indicates the spectrum of p14 and SF3b155(199-462). The red line indicates the spectrum of intron RNA, p14, and SF3b155(199-462). The black line indicates what p14 alone, intron RNA alone, and SF3b155(199-462) alone would be if there was no interaction. Since there is a difference between the red line and black line, this means there is a degree of interaction between the p14, intron RNA, and SF3b155(199-462).
114
400 mM NaCl @ pH 6.5). The spectra of combination of p14, SF3b155 and intron RNA display different traces when p14 and intron RNA are in the same cuvette, p14 and SF3b155(199-462) are in the same cuvette, and p14, intron RNA and SF3b155(199-462) are in the same cuvette. The difference between the trace of intron RNA + p14+ SF3b155(199-462) and the trace of the addition of p14 + intron RNA + SF3b155(199-462) traces is a 33% decrease in absorbance. This suggests that there is an interaction between the three molecules, but other experiments and techniques need to be performed in order to understand their relationship to each other. In order to obtain information about changes in secondary structure of p14 and SF3b155 under equilibrium conditions, circular dichroism experiments were performed (Figure 5.6 and 5.7). Both p14 (amino acid residues 1-125) and SF3b155(199-462) were scanned separately at both far UV (190-260 nm) and near UV (250-320 nm) wavelengths. The CD spectrum of p14 at far UV (Figure 5.6) had minima at 203 nm, 212 nm and 222 nm, which is characteristic of a mixture of random coil, α-helical, and β-sheet character. Analysis of CD spectrum of p14 by CONTIN (Bobba et al., 1990; Lobley et al., 2002), predicts that p14 alone has approximately 5 % α-helical character, 35 % β-sheet character, and 60 % random coil character. As expected, the CD spectrum at near UV for p14 exhibited no secondary structure because p14 has very few aromatic amino acids (no tryptophan residues, four phenylalanine residues and ten tyrosine residues). The CD spectrum of SF3b155(199-462) at far UV had minima at 202 nm, 208 nm and 216 nm, characteristic of random coil, α-helical, and β-sheet character. Analysis of CD spectrum of SF3b155(199-462) by CONTIN (Bobba et al., 1990; Lobley et al., 2002), predicts that SF3b155(199462) alone has approximately 10 % α-helical character, 40 % β-sheet character, and 50% random coil character. However, in the near UV region of the spectra, the CD spectrum of SF3b155(199-462) had a maximum at 260 nm. There are nine tryptophan residues, two phenylalanine residues, and two tyrosine residues and their chemical environment could be monitored at 260 nm because this is the region where aromatic amino acid residues exhibit signal.
115
Figure 5.6 CD spectra of p14 with SF3b155(199-462) at far UV. The blue line indicates the spectrum of p14. The red line indicates the spectrum of SF3b155(199-462). The green line indicates the mixture of p14 and SF3b1559199-462). The black line indicates what p14 (blue) and SF3b155(199462)(red) would be if there was no interaction. Since there is a difference between the green line and black line, this means there is between p14 and SF3b155(199-462). We note that there is an increase in α-helical content from 5% to 7%. However, these measurements in the far UV can not distinguish which protein(s) are contributing to the change in spectrum.
116
Figure 5.7 CD spectra of p14 with SF3b155(199-462) at near UV. The blue line indicates the spectrum of p14. The red line indicates the spectrum of SF3b155(199-462). The green line indicates the mixture of p14 and SF3b1559199-462). The black line indicates what p14 (blue) and SF3b155(199462)(red) would be if there was no interaction. Since there is a difference between the green line and black line, this means there is some level of interaction between p14 and SF3b155(199-462). At ~285 nm when p14 is added to SF3b155(199-462), a minimum emerges when p14 is added to SF3b155(199462). The minimum at 285 nm may be indicative of the tryptophan residues in SF3b155(199-462) when p14 is added suggesting that the environment around the tryptophan residues changes.
117
When p14 and SF3b155(199-462) are added in equimolar concentrations in the same cuvette, the spectra in both far UV (Figure 5.6) and near UV (Figure 5.8) region of the spectrum are different than from the predicted traces of the sum of the individual traces of p14 and SF3b155(199-462). The difference between the two traces was a 45 % reduction in ellipticity at far UV and a 30 % decrease in molar ellipticity at near UV than what was expected based on a summation of their traces. Analysis of CD spectrum of p14 and SF3b155(199462) together by CONTIN (Bobba et al., 1990; Lobley et al., 2002), predicts that the two proteins together have approximately 7% α-helical character, 38% βsheet character, and 55% random coil character. The near UV spectrum (Figure 5.7) exhibited a minimum at 285 nm upon addition of p14 to SF3b155(199-462), which suggests the region of SF3b155 that contains the tryptophan residues undergoes a change in environment. More specifically, a net protonation and conjugation state of the indole group that is correlated with its degree of protection or exposure to solvent. Combined with NMR data shown later in Figure 5.15, we speculate that the net change is towards a higher state of protonation. We anticipate that this is caused by solvent exposure. The change in environment of the tryptophan residues will be studied by fluorescence and NMR spectroscopy. Since there is a difference between the traces of the two proteins in the same cuvette than the trace of the sum of the two individual proteins and a new minimum at near UV, these data suggest that there is an interaction between p14 and SF3b155(199-462) in solution and the secondary structure of the two proteins change.
5.2.5 Equilibrium studies of p14 and SF3b155 interaction using fluorescence. To determine whether structural rearrangement of SF3b155(199462) occurs when p14 is added, we took advantage of the fact that SFb155(199462) has 9 tryptophan residues and performed fluorescence to follow this interaction (Figure 5.8). We excited SF3b155(199-462) at 280 nm and scanned the emissions from 300 to 500 nm with a maximum at ~350 nm. Fluorescence was also performed on p14 and negative control protein (DtxR) and each
118
displayed practically no fluorescence, which was expected because neither protein has tryptophan residues. When p14 was added to SF3b155(199-462) and excited at 280 nm, the fluorescence intensity increased by ~66% and showed a slight blue shift, which indicates that structural rearrangement of SF3b155(199-462) changes the environment of the tryptophan residues in solution, presumably by exposure to solvent. However, to determine if SF3b155(199-462) specifically binds to p14, we added a control protein to SF3b155(199-462) and excited it at 280 nm. The emission spectrum of SF3b155(199-462) and DtxR showed approximately the same intensity as SF3b155(199-462) alone suggesting that the control protein does not bind SF3b155(199-462), indicating that SF3b155(199-462) specifically binds p14. To determine whether the structural rearrangement of SF3b155(255-462) and SF3b155(1-462) is the same as SFb155(199-462) when p14 is added, we took advantage of the fact that SF3b155(255-462) has 5 tryptophan residues and SFb155(1-462) has 9 tryptophan residues and performed fluorescence to follow this interaction. SFb155(255-462) (Figure 5.9) and SF3b155(1-462) (Figure 5.10) were excited at 280 nm and the emission spectrum displayed a maximum at ~350 nm with an intensity of ~11 A.U and ~30 A.U., respectively. When p14 was added to SF3b155(255-462), there was a 33% increase in fluorescence. Also, when the control protein was added, there was no significant decrease in the SF3b155(255-462) fluorescence spectra (Figure 5.9). This suggests that the region where tryptophan residues are located in SF3b155(255-462) undergoes structural rearrangement when p14 is added and more than likely the region around the tryptophan is solvent exposed. In contrast, when p14 and the control protein were added to SF3b155(1462), there was no significant decrease in intensity of the fluorescence spectra (Figure 5.10). However, this does not mean that SF3b155(1-462) does not bind p14, but it suggests that the tryptophan residues do not become exposed when p14 is added. Structural rearrangement of SF3b155(1-462) may still occur, but the N-terminal amino acid residues of SF3b155(1-462) may shield the tryptophan residues from the solvent, so there may not necessarily be any difference in the
119
25 p14 SF3b155(199-462) SF3b155(199-462)+p14 SF3155(199-462) +DtxR DtxR
20
A.U.
15
10
5
0 300
350
400
450
500
wavelength (nm)
Figure 5.8 Emission spectra of SF3b155(199-462) excited at 280nm. The red line is SF3b155(199-462) alone. The blue line is SF3b155(199-462) with p14 and shows a 66% increase in fluorescence intensity and a slight blue shift. The yellow line is p14 and the black line is DtxR, which both show hardly any fluorescence because neither one has intrinsic tryptophan residues. The green line is SF3b155(199-462) with DtxR and it shows hardly any change in fluorescence intensity as expected because the two proteins are not known to interact. These spectrum show that there is an increase in fluorescence of SF3b155(199-462) and it is proposed that this is due to exposure of the tryptophan residues to solvent.
120
18 p14 SF3b155(255-462) SF3b155(255-462)+p14 SF3b155(255-462)+DtxR DtxR
16 14
A.U.
12 10 8 6 4 2 0 300
350
400
450
500
wavelength (nm) Figure 5.9 Emission spectra of SF3b155(255-462) excited at 280nm. The red line is SF3b155(255-462) alone. The blue line is SF3b155(255-462) with p14 and shows a 33% increase in fluorescence intensity and a slight blue shift. The yellow line is p14 and the black line is DtxR, which both show hardly any fluorescence because neither one has intrinsic tryptophan residues. The green line is the negative control and it shows hardly any change in fluorescence intensity as expected because the two proteins are not known to interact. These spectrum show that there is an increase in fluorescence of SF3b155(255-462), but a smaller increase than with SF3b155(199-462). It is proposed that this is due to exposure of the tryptophan residues to solvent by presumably structural rearrangement of SF3b155 upon binding.
121
35 p14 SF3b155(1-462) SF3b155(1-462) +p14 SF3b155(1-462) +DtxR DtxR
30 25
A.U.
20 15 10 5 0 300
350
400
450
500
wavelength (nm) Figure 5.10 Emission spectra of SF3b155(1-462) excited at 280nm. The red line is SF3b155(1-462) alone. The blue line is SF3b155(1-462) with p14 and there is a slight change in fluorescence intensity. The yellow line is p14 and the black line is DtxR , which both show hardly any fluorescence because neither one has intrinsic tryptophan residues and the green line the negative control between SF3b155(1-462) with DtxR. It is proposed that since there is no increase in fluorescence intensity, it is not because the two proteins are not binding. However, it is proposed that the N-terminus helps stabilize the C-terminal region of SF3b155(1-462) and therefore stabilizing the regions around the nine tryptophan residues.
122
tryptophan fluorescence intensity. To study the role of tryptophan residues in SF3b155, additional fluorescence and NMR studies will need to be performed using different fragments of SF3b155 in the presence of both p14 and different length segments of the intron RNA strand.
5.2.6 Equilibrium studies of p14 and SF3b155 interaction using isothermal titration calorimetry (ITC). ITC was used to determine the thermodynamic properties of fragments of SF3b155 titrated into p14. ITC measures the heat, either absorbed or released upon binding, and from this data we were able to determine a dissociation constant (Kd) between SF3b155 and p14. First, the high salt BES buffer (10 mM BES, 50 mM H3BO3, 400 mM NaCl, 20 mM TCEP at pH 6.5) was titrated into the high salt BES buffer and showed no change at 10 °C. Next, as a control, SF3b155(199-462) in high salt BES buffer was titrated into high salt BES buffer and there was no change. Titration of SF3b155(199462) into p14 at 10 °C in high salt BES buffer and subtraction of buffer-buffer titration (Figure 5.11) resulted in a Kd ~120 µM, ∆H is ~81.8 kcal/mol and ∆S is ~ 312 e.u. (reproducible over 3 trials). The positive peaks in the top portion of Figure 5.11 indicate that the interaction is endothermic, as opposed to the exothermic reaction of p14 to SF3b155(282-424) as reported by Spadaccini et. al. (2006). These data could be different than what was reported by Spadaccini et. al. (2006) for a few reasons such as; different buffers were used, their studies were at 25 °C where our studies have shown p14 aggregates, and p14 was titrated into SF3b155. These differences could attribute to why there is a difference between the two studies. Also, ITC studies were repeated with SF3b155(199-462) to p14 in low salt BES buffer (5 mM BES, 10 mM H3BO3, 100 mM NaCl, 20 mM TCEP at pH 6.5 ) at 10°C and yielded the same results as in high salt buffer. Moreover, these ITC data collected at 10°C indicate there is low binding affinity between SF3b155(199-462) and p14 under these conditions. ITC studies were then performed with SF3b155(1-462) titrated into p14 in low salt BES buffer at 10°C (Figure 5.12). These studies resulted in a Kd of ~1 .3 µM, a ∆H of -12.4 kcal/mol and a ∆S of -16.8 e.u. and indicated binding between
123
Figure 5.11 Representative ITC isotherm of SF3b155(199-462) to p14. All titrations were measured at 10 °C with SF3b155(199-462) titrated into p14. Using a one site binding fit, the Kd is ~120 µM, ∆H is ~81.8 kcal/mol and ∆S is ~ 312 e.u. These data suggest that the binding between p14 and SF3b155(199462) is weak and endothermic, meaning that energy is required in order for binding to occur under these conditions.
124
Figure 5.12 Representative ITC isotherm of SF3b155(1-462) to p14. All titrations were measured at 10 °C with SF3b155(1-462) titrated into p14. Using a one site binding fit, the Kd is ~ 1.3 µM, ∆H is approximately -12.4 kcal/mol and ∆S is approximately -16.8 e.u. These data suggest that the binding between p14 and SF3b155(199-462) is weak and exothermic, which means that energy is not required for binding between p14 and SF3b155(1-462).
125
SF3b155(1-462) to p14 was exothermic. This tells us that the N-terminal region of SF3b155 interacts with p14 and the energertics is different than with the shortened segments. This is strikingly different than what we collected with SF3b155(199-462), but this could indicate that SF3b155 needs its N-terminal region for proper folding to occur. Therefore, when SF3b155 is properly folded, it does not require energy to bind to p14, which is what is expected in vivo. 5.2.7 Structural studies of SF3b155 by NMR spectroscopy. In order to obtain spectroscopic information of SF3b155 that can be used to derive the degree of order and of certain regions (e.g. those containing tryptophan residues), as well as changes upon binding p14, we have performed NMR studies of an isotopically labeled SF33b155. Two fragments of SF3b155 have been studied containing amino acid residues from 255-462 and amino acid residues from 199-462 in the absence and presence of p14. 1H-15N HSQC spectra of 15N-labeled SF3b155(255-462) (black peaks in Figure 5.13) and 15N-labeled SF3b155(199462) (red peaks in Figure 5.13) display characteristics of unstructured proteins, since the majority of the resonances are between 8 and 8.6 ppm. The additional 56 amino acids of SF3b155(199-462) have the same spectral properties as SF3b155(255-462) as shown in Figure 5.13 where the two spectra are superimposed. Therefore, the two fragments of SF3b155 have the same unstructured nature as evident from the similarity between the spectra of SF3b155(255-462) and SF3b155(199-462) (Figure 5.13).
5.2.8 Structural studies of SF3b155 with p14 by NMR spectroscopy. Addition of unlabeled p14 to 15N-labeled SF3b155(255-462) results in the appearance of approximately ten new cross peaks in a 1H-15N HSQC spectrum, but no change in existing cross peaks (Figure 5.14). The down-field 15N-shifts correspond to increase α-helical formation in the C-terminal region of the protein. Also, there was no change in cross-peaks corresponding to the tryptophan residues (1H-10 ppm and 15N-130 ppm) upon addition of p14, which is consistent with the spectrum of SF3b155(282-424) plus p14 presented by Spadaccini et. al.
126
(2006). In contrast, 1H-15N HSQC spectrum of 15N-labeled SF3b155(199-462) (Figure 5.15 with unlabeled p14 reveals ~15 additional residues (Figure 5.16) and a change of 4 cross-peaks corresponding to the tryptophan residues upon addition of p14 (Figure 5.16). The cross-peaks corresponding to the tryptophan residues disperse in both the 1H and 15N dimensions, indicating they are interacting. Also, emergence of new cross-peaks in the side chain region and Cterminal region when p14 is added to SF3b155(199-462)(Figure 5.16) suggests that SF3b155 is becoming more structured. These data suggest that there is structural rearrangement of SF3b155(199-462) when p14 is added and corroborate the data collected with circular dichroism and fluorescence experiments suggesting that the tryptophan residues could be exposed to solvent when p14 is added. NMR studies of the two fragments of SF3b155 provide evidence that both fragments bind p14 under equilibrium conditions. However, it will not be clear whether there is change in structure due to the fact that we are using a truncated portion of SF3b155 and not the entire N-terminus of SF3b155. In order to determine whether the N-terminus of SF3b155 helps stabilize the short fragments of SF3b155, NMR studies need to be conducted. 5.2.9 NMR studies of SF3b155(1-462). In order to determine if the structural features of the N-terminus of SF3b155 can be obtained, we have performed NMR studies of SF3b155(1-462). 1H-15N HSQC spectrum of 15N-labeled SF3b155(1-462) (Figure 5.17) displayed not even half of the anticipated resonances and a narrow proton range. The spectrum also showed good dispersion in the 15N dimension and the side chain chemical shifts collapsed, so altogether these characteristics are consistent with an unstructured protein. The difficulty in collecting useful NMR spectrum was expected because SF3b155(1462) is a 50.4 kDa protein and in order to perform structural studies by NMR spectroscopy, specific isotopic labeling techniques will need to be used and use NMR techniques developed for molecules up to ~100kDa, such as TROSY and cross-correlated relaxation-enhanced polarization transfer (CRINEPT)
127
Figure 5.13 Comparison 1H-15N-HSQC spectrum of15N-labeled SF3b155(255462)(black) and 15N-labeled SF3b155(199-462)(red) at 4°C acquired on the Varian 720 MHz spectrometer at the National High Magnetic Field Laboratory (NHMFL, Tallahassee, FL). The 1H-15N-HSQC experiment was acquired with 64 scans and 256 increments. The spectrum of both SF3b155(255-462) and SF3b155(199-462) have characteristics of an unstructured protein. Both proteins show resonances attributed to tryptophan residues (between 10 to 10.4 ppm of 1 H), glycine residues (between 8.2 to 8.6 ppm of 1H and 108 to 112 ppm of 15N) and amino groups of amino acid side chains (between 6.8 and 7.6 ppm of 1H). In the region that correlates to the tryptophan residues, SF3b155(255-462) has four cross-peaks and SF3b155(199-462) has five cross-peaks.
128
Figure 5.14 1H-15N-HSQC spectrum of 15N-labeled SF3b155(255-462)(black) and 15N-labeled SF3b155(255-462) with equimolar p14 (red) at 4°C acquired on the Varian 720 MHz spectrometer at the National High Magnetic Field Laboratory (NHMFL, Tallahassee, FL) with 64 scans and 256 increments. The spectrum of SF3b155(255-462) has characteristics of an unstructured protein. However, when p14 is added to SF3b155(255-462), four peaks appear in the region where amino (-NH2) protons resonant and ten peaks appear in the region related to the C-terminal end of the unfolded protein (between 7.8 ad 8.2 ppm of 1H), indicative of increase in α-helical formation.
129
Figure 5.15 1H-15N-HSQC spectrum of 15N-labeled SF3b155(199-462)(black) and 15N-labeled SF3b155(199-462) with equimolar p14 (red) at 4°C acquired on the Varian 720 MHz spectrometer at the National High Magnetic Field Laboratory (NHMFL, Tallahassee, FL). The 1H-15N-HSQC experiment was acquired with 64 scans and 256 increments. The spectrum of SF3b155(199-462) has characteristics of an unstructured protein. However, when p14 is added to SF3b155(255-462), eight peaks appear in the region where amino (-NH2) protons resonant, twelve peaks from the region indicative of α-helical formation in the Cterminal end of an unfolded protein (7-8 ppm), and five more peaks appear in the tryptophan region (10-11 ppm). This indicates that the tryptophan residues may undergo a change in chemical environment when p14 is added and SF3b155(199-462) becomes slightly more structured in the presence of p14.
130
A
B
C
Figure 5.16 Cropped 1H-15N- HSQC spectrum of 15N-labeled SF3b155(199-462) (300 µM sample in 10 mM BES, 50 mM H3BO3, 400 mM NaCl @ pH6.5) collected at 4°C on the 720MHz spectrometer using 64 scans and 256 increments. (A) Enlarged view of spectrum showing emergence of new crosspeaks in the side chain region when SF3b155(199-462) has p14 added (shown in red). (B) Enlarged view of spectrum showing emergence of new cross-peaks in the tryptophan region when p14 is added to 15N-labeled SF3b155(199-462) (shown in red). (C) Enlarged view of spectrum showing emergence of new cross-peaks in the region most likely to be the C-terminal NH2 of unfolded protein region when p14 is added to 15N-labeled SF3b155(199-462) (shown in red).
131
Figure 5.17 1H-15N-HSQC spectrum of 15N-labeled SF3b155(1-462) at 4°C acquired on the Varian 720 MHz spectrometer at the National High Magnetic Field Laboratory (NHMFL, Tallahassee, FL). The 1H-15N-HSQC experiment was acquired with 64 scans and 256 increments. The spectrum of SF3b155(1-462) has characteristics of a highly unstructured protein and only displays 10% of the expected 462 cross-peaks of SF3b155(1-462).
132
(Riek et al., 2000). Therefore, the N-terminus of SF3b155 will have to be grown in deuterated minimal media in order to slow down the chemical exchange of the protein with solvent. Also, 13C-labeling of specific amino acid residues of SF3b155 may alleviate spectral overlap and could help follow the interaction of SF3b155 to both p14. Besides working on the interaction between p14 and SFb155, I have been working with Karen Cherkis and Amy Bryant on the interaction of SF3b155 (when p14 is not present) to the full-length intron, an RNA strand that represents the region between exon 1 and exon 2. We have found that the binding between SF3b155(1-462) to the full intron binds weakly and nonspecifically. The data could be aided by site specific labeling of SF3b155(1-462) in order to determine the amino acid residues that interact with the intron upstream and down stream of the branch site adenosine.
5.3 Discussion 5.3.1 Analysis of SF3b155 SF3b155 is the largest SF3b protein and has a molecular weight of approximately 145 kDa. Previous studies have shown that the amino acids residues in the N-terminus region of SF3b155, amino acid residues from 1 to 462, were responsible for interaction with p14 and the pre-mRNA (Will et al., 2001). Therefore, in order to study SF3b155 interaction with p14, shortened fragments have been studied. Sequence analysis indicates that the first 462 amino acids residues are predicted to be unstructured. The probability of unstructured residues of SF3b155(1-462), SF3b155(199-462), and SF3b155(255-462) shows that there is a slight difference between the fragment of SF3b155 alone and that of the same region in the full-length SF3b155. By inspection, the differences are not large enough to cause reason for concern in using these shortened fragments for studying binding interactions. However, it did not go unnoticed that different fragments of SF3b155 may adopt different structures when alone in solution, than if it was involved in the full-length protein, and it may act differently when it binds p14.
133
5.3.2 Equilibrium studies of p14-SF3b155 binding to the branch site RNA These studies establish for the first time that when SF3b155 is present, p14 binds with higher affinity to the double stranded ψ-modified branch site duplex. This is important because it suggest that SF3b155 facilitates the interaction between p14 and its RNA target. Therefore, the cooperative behavior of the two proteins may have some significance for optimum binding between p14 and the branch site region. This may yield insight into the mechanism by which numerous proteins contribute to the recognition of RNA during spliceosome assembly. Therefore, our studies (published here and unpublished results) and previous studies (Query et al., 1994; Query et al., 1996; Query et al., 1997; Gottschalk et al., 2001; Will et al., 2001; Golas et al., 2003, 2005; Schellenberg et al., 2006; Spadaccini et al., 2006) indicate that recognition of the branch site RNA by p14, and its enhancement by SF3b155, must be relatively important in assembly of the spliceosome.
5.3.3 Equilibrium studies of p14 and SF3b155 The studies of p14 and SF3b155 by CD spectroscopy, ITC, fluorescence, and NMR indicate the binding between p14 and SF3b155 in solution indeed occurs. By CD, fluorescence and NMR we were able to show that the environment around the tryptophan residues in the shorten fragments of SF3b1555 changes when p14 is added, suggesting that the tryptophan residues could be exposed to solvent or that binding may occur in the vicinity of one of the tryptophan residues. We propose that since the tryptophan residues become more distinct when p14 is added, this suggests that the tryptophan residues are possibly becoming solvent exposed. It was also determined by CD that the secondary structure of the two proteins together is a mixture of α-helix, β-sheet, and random coil characteristics. Since there is a difference by CD between the traces of the two proteins in the same cuvette than the trace of the sum of the two individual proteins, this difference indicates that there is an interaction between the two proteins.
134
The titration of SF3b155(199-462) into p14 at 10 °C yielded strikingly different ITC data than reported by Spadaccini et al (2006) at 25 °C. Since p14, like many RRM proteins, is notoriously difficult to solubilize and unstable at higher temperature (e.g. our studies and Schellenberg et al., 2006), there was cause for confusion by several points in the Spadaccini et al. (2206) paper. Their ITC studies were performed with p14 as the titrant (at high concentrations) at 25 °C and the using a phosphate buffer. The buffer screens we performed found that high concentrations of p14 were not achievable in phosphate buffer. They also report a Kd of p14 binding to SF3b155(282-424) to be 9.6 nM (but show no ITC isotherms), but this would suggest that the binding is very strong and similar to that of a protein-drug interaction, and therefore unrealistic for the limited contact between these proteins identified in recent studies (Spadaccini et al., 2006; Schellenberg et al., 2006). In contrast, our ITC studies on this system were performed by titrating SF3b155(199-462) and SF3b155(1-462) into p14 at 10 °C, and resulted in a Kd of ~120 µM and ~1.3 µM. Interestingly, our data indicate that the interaction is endothermic for SF3b155(199-462) to p14, as opposed to exothermic for SF3b155(1-462) to p14. These studies indicate that binding occurs between SF3b155 and p14, but the mode of binding may differ depending on the length of SF3b155, which could indicate that the N-terminal region of SF3b155(1-462) helps stabilize the amino acid residues from 199 to 462 of SF3b155. We will repeat the ITC studies in buffer conditions with lower salt and expect to achieve a more reasonable binding constant under these new buffer conditions. For comparison to other spliceosomal protein, it was determined that the interaction between SF3b155(190-344) and U2AF65(375-475) by ITC has a binding constant of approximately 3-5 µM (Thickman et al., 2006), which is considered a realistic binding affinity for the interaction between proteins involved in the pre-mRNA splicing process. The HSQC spectra of SF3b155 in the absence and presence of p14 show small spectral differences. These changes indicate that binding between p14 and SF3b155 does occur and the two form a stable complex. Since the spectral region where tryptophan residues resonant change, these studies may be aided
135
by mutation studies of tryptophan residues in order to determine if any tryptophan residues in SF3b155 are in the vicinity of where binding between p14 and SF3b155 occur. However, from the lack of major spectral differences between SF3b155 alone and in the presence of p14, we can predict that there is only slight structural rearrangement of SF3b155 when p14 is added. Altogether, CD data suggest that when SF3b155 is added to p14 there is an increase in α-helix and β-sheet character. Fluorescence data suggest that the tryptophan residues of SF3b155 could become solvent exposed in SF3b155(199462) and SF3b155(255-462) when p14 is added, but there is not a change in fluorescence when p14 is added to the N-terminal region of SF3b155(1-462). In conjunction with the ITC data, this suggest that the N-terminal region of SF3b155 may be needed in order to help stabilize the C-terminal region of SF3b155(1462). This may also explain why the interaction between SF3b155(199-462) to p14 is endothermic and the interaction between SF3b155(1-462) to p14 is exothermic. NMR studies suggest that the SF3b155(199-462) undergoes structural rearrangement when p14 is added and more resonances attributed to tryptophan residues are observed. This observation corroborates our CD and fluorescence, which indicate that tryptophan residues of the C-terminal region of SF3b155 (amino acids 199 to 462) could become solvent exposed. However, more NMR studies will need to be performed in order to corroborate our proposal that the N-terminal region of SF3b155 helps stabilize the C-terminal region of SF3b155(1-462).
5.4 Conclusions These biochemical data show that p14 interacts with SF3b155 in solution and that SF3b155 enhances binding between p14 and the branch site RNA. These data also provide evidence that p14 binds SF3b155 and that SF3b155 may undergo a conformational change upon binding, since there is a change in environment of the tryptophan residues in SF3b155(199-462). We also have evidence that SF3b155 binds p14 under equimolar conditions with low binding affinity. This allows for interaction between p14 and SF3b155, but also allows for
136
the flexibility to change conformation and bind other spliceosomal components, such as the regions upstream and downstream of the branch site A on the premRNA. We have begun to look at the amino acid residues of SF3b155 that bind the pre-mRNA in the absence of p14. Studying both the interaction of SF3b155 to p14 and pre-mRNA will be aided by site-specific labeling of SF3b155 for NMR experiments and making certain mutation of SF3b155, such as certain tryptophan residues to alanine residues. Determining the specific amino acid residues of SF3b155 that interact with p14 and the pre-mRNA may yield insight into these interactions and the assembly of the functional core of the spliceosome.
137
CHAPTER 6 DISCUSSION
6.1 Introduction Research described in this dissertation focused on fundamental RNA-RNA and RNA-protein interactions at the functional core of the spliceosome that include the pairing of the U2 snRNA with the branch site region of the pre-mRNA, U2 snRNP protein p14 interacting with the double-stranded branch site region, and p14 interacting with SF3b155. The RNA-RNA interaction of the ψ-modified branch site structure features an extrahelical branch site nucleotide that is anchored by a base triple and distinct electrostatic pattern (Newby & Greenbaum, 2001, 2002b); therefore supporting a model of recognition on the basis of both shape and charge (Xu et al., 2005). However, the molecular and atomic details of the RNA interactions with p14 and SF3b155 remain unknown. A structural model of the U11/U12 snRNP of the atac spliceosome, analogous to the U2 snRNP, determined by cryo-EM displays a shallow gulley in the protein surface of the SF3b complex, lined with regions of proteins shown to form cross-links with elements of the precursor mRNA intron, suggesting that this is the “track” on the pre-mRNA and associated snRNAs move through the spliceosome (Golas et al., 2005). In particular, a 14.6 kDa component of the SF3b complex, called p14, forms a covalent UV-induced cross-link to a derivatized branch site A during assembly of complex A (MacMillan et al., 1994; Query et al., 1996; Query et al., 1997; Will et al., 2001). SF3b155 has been shown to contact p14 (Gozani et al., 1998) as well as to cross-link to nucleotides of the intron upstream an downstream from the branch site A (Gozani et al., 1998). However, it is not known whether there is direct recognition between p14 and RNA, or whether these contacts are specific or nonspecific. This question is of central importance because, if interaction is specific, it may have critical implications in providing the environment for assembly and folding of the RNA
138
spliceosomal core prior to splicing. In order to determine the molecular determinants of the spliceosomal assembly, we have refined the structural view of the branch site duplex, determined equilibrium binding of p14 to the pre-mRNA branch site in solution, and determined the impact of the p14-binding segment of SF3b155 on its affinity for p14 at the functional core of pre-mRNA splicing apparatus.
6.2 Summary of Results Acquiring NMR spectra of RNA samples in supercooled water allows us to identify exchanged broadened imino proton resonances by slowing down the rapid exchange that occurs at higher temperatures. This enabled us to search for imino protons we were not able to detect previously in a complementary duplex or a duplex representing the branch site region. At low temperatures, we were able to identify the terminal protons in a complementary duplex and the branch site duplex. In addition, NMR experiments in supercooled water helped us identify the ψ N3H in ψ-modified branch site duplex, which was undetectable at higher temperatures due to rapid exchange. This was also observed by structural models, which did not indicate formation of a base pair between ψ N3H and the opposing adenine (A23, adjacent to the branch site base A24). This technique has been helpful in refinement of the structure of ψ-modified branch site duplex (Newby & Greenbaum, 2001, 2002b) because identification of ψ N3H contributed important information about the environment of this imino proton with respect to surrounding bases. Also, identification of the ψ N3H in ψ-modified branch site region helped us in identify it in subsequent experiments when p14 was added to a longer RNA strand of the pre-mRNA and ψ-modified U2 snRNA. An unexpected result of using the capillary technique for NMR experiments in supercooled water was that the line-width at half-height increased as the temperature decreased, which was a consequence of the chemical exchange of the RNA duplexes increasing as the temperatures decreased. This could be caused by buffer effects, pH, or the beginnings of cold denaturation. We propose that this is the beginnings of cold denaturation because there was
139
no affect on the amino protons and the chemical exchange rate increase of the imino protons could mean that base pairing is slightly perturbed, but not completely disrupted. Therefore, our hypothesis is that we are observing the beginnings of cold denaturation in RNA duplexes under supercooled aqueous conditions. When adding p14 to branch site RNAs, our data provide evidence that p14 binds the single-stranded pre-mRNA in solution at moderate affinity (Kd ~35 µM). It also shows that p14 binds the ψ-modified branch site duplex at twice the affinity (Kd ~19 µM) than the single-stranded RNA. This later finding is important because we demonstrate for the first time that a protein containing an RNA Recognition Motif binds double-stranded RNA. Using NMR, we observed only small chemical shift changes and a narrower of resonances near the branch site A, suggesting that the structure of the branch site duplex may be stabilized when p14 binds this region of the branch site RNA. These data suggest that p14 undergoes structural change when the single-stranded pre-mRNA or the doublestranded branch site duplex is added to p14. This finding is consistent with other proteins that contain an RRM, which undergo structural change upon binding to its cognate RNA. Also, these studies show that the binding of p14 to the branch site RNA is salt dependent, meaning that very high concentrations (~400 mM NaCl) of salt disrupt the binding of p14 to the RNA. We propose that the branch site duplex structure does not change when p14 is added, but that p14 undergoes a conformational change upon binding and that p14 provides the scaffolding for positioning of the branch site duplex and the nucleophile for the attack during the first step of splicing. Addition of SF3b155(199-462), a segment that binds only to p14 and not the RNA, interacts with p14 with low affinity (Kd ~120 µM) and enhances binding between p14 and the branch site RNA. However, SF3b155(1-462) interacts with p14 with moderate affinity (Kd ~1 µM). These data also provide evidence that SF3b155(199-462) undergoes a conformational change upon binding of p14, which is evident by a change in environment of tryptophan residues in SF3b155(199-462) shown by circular dichroism, fluorescence, and NMR
140
spectroscopy data. This conformational change may allow for interactions between p14 and SF3b155(199-462) to occur. It may also allow for flexibility of SF3b155(199-462) to undergo conformational changes in order to bind other spliceosomal components, such as the branch site RNA complex that includes the pre-mRNA and U2 snRNA. Also, since we do not see similar extensive spectroscopic changes in SF3b155(1-462) upon binding p14 as compared to SF3b155(199-462), it can be concluded that the smaller segments of SF3b155 are not structural autonomous. We speculate that this region of the SF3b complex is assembled through numerous moderate-affinity, but specific contacts between the U2 snRNA and the intron RNA, p14 and the double-stranded branch site duplex, SF3b155, p14 and the branch site duplex. Therefore, we postulate that SF3b155 mediated by p14-RNA affinity is a result of structure formation. These interactions allow for plasticity of the molecules in order to sustain dynamic assembly of the spliceosome’s catalytic core each time these components are needed for premRNA splicing to occur.
6.3 Future Studies 6.3.1 Future studies involving NMR in supercooled studies. First we will investigate the reasons chemical exchange rate of the imino protons increases as the temperature decreases. We will study whether the effect are caused by buffer effects, pH, or the beginnings of cold denaturation. We have begun to look into this by preparing RNA duplexes with 15N-labeled uridine and 15N-labeled pseudouridine in a complementary duplex in collaboration with J.P. Desaulniers and Dr. Christine Chow. We will take advantage of the 15N-labeled RNA and use it to study the pKa at temperatures below from 0 °C to approximately -20°C. Secondly, we will also study what allows the RNA samples to decrease to supercooled temperatures without freezing. It has been previously stated that the freezing temperature decreases as its relationship with surface area decreases. We will test whether pressure of the RNA duplex in the capillary tube has a role by comparing a DNA or RNA duplex in a set of capillaries closed at
141
both ends, closed at only one end, and with both ends open. This set of experiments will give us information about the relationship between temperature (T), pressure(P), and volume(V) and their contribution to this phenomenon, in order to determine whether the solutions remained unfrozen at supercooled temperatures because of increase in surface area or if it is caused by a change in pressure or possibly a combination of both. We will use the Ideal Gas law [PV=nRT] and the real gas law [(P+(n2a/V2))(V-b)=nRT] where a (units=(atm*L2)/mol2) and b (units=L/mol) are van der Waals constants for these calculations.
6.3.2 Electrostatic studies of the pre-mRNA branch site duplex and p14. Using computational techniques, we will study the electrostatic potential of p14 and whether it is possible that there is an electrostatic interaction with the branch site duplex. We will work in collaboration with Darui Xu and Dr. Marcia Fenley using a method that combines a hybrid boundary element and finite difference nonlinear Poisson–Boltzmann approach (Boschitsch & Fenley, 2004). Darui Xu, working with Drs. Fenley and Greenbaum, has already characterized the electrostatic surface potential of the unmodified and ψ-modified branch site helices (PDB accession codes 1LMV and 1LPW, respectively)(Xu et al., 2005). In Figure 6.1 (right), the model shows an electronegative “hot-spot” at the region near the 2’-OH of branch site A and negatively charged regions in the minor groove, major groove, and the phosphate backbone. They have now started calculations using the PDB accession code 2F9D from the crystal structure of p14 (Schellenberg et al., 2006)(Figure 6.1, left). Preliminary calculations show a basic patch (positively charged) in the anti-parallel β1-β3 strand, which contains Y22 of p14. The intriguing observation is that there may be electrostatic complementarity in addition to shape complementarity of p14 to the branch site duplex. Future experiments will provide evidence of overall polar and non-polar solvent accessible surface areas as well as refine the surface electrostatic regions of p14 and how they may dock with the branch site duplex.
142
Minor groove Branch site A
Major groove
Anti-parallel β1-β3 sheet that contains Y22 on theβ1 sheet
Highly basic patch
Highly negatively charged region
2’-OH of branch site A
Figure 6.1 Surface electrostatic potential maps of p14 and ψ-modified branch site duplex. Left, shows the electrostatic potential map of p14 showing anti-parallel β1-β3 sheet , which is the proposed RNA recognition site (Schellenberg et al., 2006). Right, shows the electrostatic potential map of ψmodified branch site (Xu et al., 2005) showing the widened major groove and the branch site A and its 2’-OH locations. The color mapping of the electrostatic potential was scaled as follows: yellow (most negative), followed by red, white (neutral), blue, and green (most positive).
143
6.3.3 Studies of SF3b155 to the full-length intron strand without p14. These studies will expand on research that Karen Cherkis and Amy Bryant have been working on in order to determine regions of SF3b155(1-462) which binds the full-intron without the presence of p14. Preliminary studies using nitrocellulose membrane retention shows that the interaction between SF3b155(1-462) and p14 is weak and non-specific. Our preliminary studies have also shown that SF3b155(199-462) does not bind the full-length intron RNA. However, this region of SF3b155 has been shown in this dissertation to enhance the binding of p14 to the ψ-modified branch site. Also, these particular studies as well as studying the interaction between p14 and SF3b155 could be aided by 13
C- and/or 15N- site-specific labeling of certain amino acid residues of SF3b155
and following the interaction by the change in chemical shifts obtained using NMR experiments. These studies should yield insight into the recognition of the branch site region by SF3b155 and may clarify how splicing can occur when p14 is not present.
6.3.4 Mutation studies of p14 to determine amino acid residues of p14 involved in binding of the branch site duplex. To test the Schellenberg et. al. (2006) hypothesis that tyrosine (Y) at position 22 of p14 binds to pre-mRNA, we propose several mutations of tyrosine, located in position 22 of p14, such as: 1) Y22W because tryptophan (W) and tyrosine are similar in structure because they share a conjugate ring and it also adds a fluorescent property in order to help in testing binding with RNA. 2) Y22F because they are similar in structure but phenylalanine (F) lacks the -OH group which may be a possible source of binding. 3) Y22S because serine (S) has no structural similarities with Y, but they are both polar; binding would be discouraged on the basis of a size requirement for the specific site. 4) Y22A because alanine (A) has no structure similarities with Y and also Y is polar and A is non-polar, which would discourage binding on the basis of
144
structure dissimilarities and polarity, if Y does indeed bind directly with the branch site RNA. To test the Spadaccini et. al. (2006) hypothesis that the β3'-β3'' hairpin of p14 binds to pre-mRNA, we propose the following mutations of p14 from one amino acid residue to another: 1) N81A because alanine has no structure similarities to asparagine (N) and N is a polar amino acid and A is a non-polar which should discourage any binding if N does indeed binding directly with pre-mRNA. 2) N81S because the polarity between asparagine and serine is not changed, so lack of binding would suggest an interaction other than electrostatics. 3) N81F because asparagine is a smaller polar group and phenylalanine is a bulky non-polar group, which would determine electrostatic and steric interactions that may occur between p14 and RNA. 4) N81Y because asparagine, a smaller polar group, would be changed to the bulky polar group of tyrosine to detect if size of polar group determines binding to the major grove of the branch site duplex. 5) V82A because valine (V) has structure similarities to alanine but A is smaller so we would be able to determine if size is a requirement of the non-polar group for binding to any part of the RNA and/or protein. 6) V82P because proline (P) has no structural similarities to valine and its inclusion may stimulate formation of sharp turns in the β-β hairpin, which should severely disrupt binding. 7) R85E because arginine (R) is changed from a positively charged amino acid residue to glutamic acid (E), a negatively charged amino acid residue, in order to test whether specific charge is involved and required in binding. 8) R85A because arginine is mutated from a positively charged, hydrophilic, amino acid residue to alanine, a non-polar, hydrophobic, amino acid residue, in order to test if p14 is binding by a dipole-dipole interaction.
145
9) Y86W because tyrosine and tryptophan have similar structures because they share a conjugate ring and it also adds a fluorescent properties to test binding with RNA. 10) Y86F because F lacks the -OH group of tyrosine, which may a possible source of binding. 11) Y86S because serine (S) has no structural similarities with Y, but they are both polar, so binding would be discouraged based of size. 12) Y86A because alanine and tyrosine have no structure or electrostatic similarities, which should discourage binding if Y does indeed bind directly with the branch site RNA. To test the Spadaccini et. al. (2006) hypothesis of the possible αC domain of p14 binds to pre-mRNA: we would propose the following mutations of p14: 1) Q99E because glutamine (Q) is a polar, hydrophilic amino acid residue and glutamic acid, a hydrophilic, positively charged amino acid residue to test whether the lack of charge on Q promotes binding to the pre-mRNA. 2) Q99A because glutamine is a polar, hydrophilic amino acid residue and alanine, a hydrophilic, non-polar amino acid residue, in order to test electrostatic interactions between p14 and the branch site RNA. 3) K100E because lysine (K) is a positively charged, hydrophilic amino acid residue to glutamic acid, a hydrophilic, negatively charged amino acid residue, in order to test whether the positive charge on K promotes binding of p14 to the branch site RNA. 4) K100A because lysine is a positively charged, hydrophilic and relatively large amino acid residue to alanine, a smaller, non polar, hydrophobic, amino acid residue, in order to test electrostatic interactions and requirements of binding between p14 and the branch site RNA.
6.3.5 Mutation studies of SF3b155 to both p14 and full-length intron to determine which amino acid residues of SF3b155 are involved in binding to both p14 and the full-length intron. To test whether the RW region of SF3b155 binds p14, we propose to introduce mutations of the following:
146
1) R198E, R199E, R231E, R292E, R337E, R387E, and R390E because these arginine residues are near tryptophan and because changing from a positively charged amino acid residue to glutamic acid (E), a negatively charged amino acid residue, this would test whether specific charge is involved and required in binding of SF3b155 to either p14 and the pre-mRNA branch site RNA. 2) R198, R199, R231, R292, R337, R387, and R390 because these arginine residues are near tryptophan and because changing a positively charged, hydrophilic, amino acid residue to alanine, a non-polar, hydrophobic, amino acid residue, could test if SF3b155 is binding by a dipole-dipole interaction. 3) W200A, W218A, W232A, W293A, W338A, W386A, W388A, W254A, and W310A because tryptophan and alanine have no structural similarities and if one of the tryptophan residues or the arginine adjacent to it, binds either p14 of the branch site RNA, then disrupting fluorescent properties of SF3b155 would be a good marker. 4) W200Y, W218Y, W232Y, W293Y, W338Y, W386Y, W388Y, W254Y,and W310Y because tryptophan and tyrosine have similar structures because they share a conjugate ring and it also lose a fluorescent properties, which could be a good indicator in fluorescence experiments.
6.4 Conclusions In conclusion, our studies have lead to a new hypothesis on how the molecular components at the functional core of the spliceosome interact with each other. Our studies have confirmed that the incorporation of pseudouridine in the branch site duplex helps stabilize the pairing of the pre-mRNA and the U2 snRNA and may explain why pseudouridine is conserved in the position juxtaposed to the branch site adenosine. We have also shown for the first time that p14 prefers double-stranded RNA, which implies that p14 is an unusual protein of the RRM family. We have shown that p14 undergoes conformational change when it binds to both its cognate single- and double-stranded RNA. We have shown that SF3b155 enhances the binding between p14 and the double-
147
stranded branch site duplex. We have also shown evidence that SF3b155 undergoes conformational change when it binds to p14, presumably so SF3b155 can also bind the pre-mRNA in the region between the 5’-splice site and the branch site region and the polypyrimidine tract. This interaction of SF3b155 to the pre-mRNA may also stabilize the branch site duplex without p14 or confirm our hypothesis that the presence of p14 is necessary to provide the scaffold for the branch site duplex in order to properly position the necessary components for the first step of splicing. In the process of assembly of U2 snRNP, we anticipate that the RNA complex is recognized by the pre-assembled SF3b proteins. This finding is consistent with the model presented by the cryo-EM structure of U11/U12 snRNP (Golas et al., 2005) where the RNA may recognize the groove formed by the protein complex, but we would like to extend the model by predicting where p14 binds the RNA. We propose that p14 will bind to the kinked backbone and minor groove of the branch site duplex in order to keep the 2’-OH of the branch site adenosine available for its role in the first step of pre-mRNA splicing.
148
APPENDIX A
DNA and Amino Acid Sequences of proteins
149
U2 snRNP protein p14 Homo sapiens p14 (ACCESSION # AF401310, P59708) [gi:15278117] Molecular weight: 14,570.8 Daltons Theoretical pI: 9.41 Extinction coefficient: 15,188 (M*cm)-1 DNA sequence of full length p14 1 atggcgatgc aagcggccaa gagggcgaac attcgacttc cacctgaagt aaatcggata 61 ttgtatataa gaaatttgcc atacaaaatc acagctgaag aaatgtatga tatatttggg 121 aaatatggac ctattcgtca aatcagagtg gggaacacac ctgaaactag aggaacagct 181 tatgtggtct atgaggacat ctttgatgcc aagaatgcat gtgatcacct atcgggattc 241 aatgtttgta acagatacct tgtggttttg tactataatg ccaacagggc atttcagaag 301 atggacacaa agaagaagga ggaacagttg aagcttctca aggagaaata tggcatcaac 361 acagatcctc ccaagtga Protein sequence of p14 1 MGMQAAKRAN IRLPPEVNRI LYIRNLPYKI TAEEMYDIFG KYGPIRQIRV 50 51 GNTPETRGTA YVVYEDIFDA KNACDHLSGF NVCNRYLVVL 80 81 YYNANRAFQK MDTKKKEEQL KLLKEKYGIN TDPPK 125
150
Full length SF3b155 Homo sapiens splicing factor 3b, subunit 1, 155kDa (SF3b1); spliceosomeassociated factor 155 [Homo sapiens] (ACCESSION # NM_012433, NM_001005526, NP_036565) [gi 54112116; 6912654] Molecular weight: 145815.3 Daltons Theoretical pI: 6.58 Extinction coefficient: 151,200 (M*cm)-1 DNA sequence of full length SF3b155 1 atggcgaaga tcgccaagac tcacgaagat attgaagcac agattcgaga aattcaaggc 61 aagaaggcag ctcttgatga agctcaagga gtgggcctcg attctacagg ttattatgac 121 caggaaattt atggtggaag tgacagcaga tttgctggat acgtgacatc aattgctgca 181 actgaacttg aagatgatga cgatgactat tcatcatcta cgagtttgct tggtcagaag 241 aagccaggat atcatgcccc tgtggcattg cttaatgata taccacagtc aacagaacag 301 tatgatccat ttgctgagca cagacctcca aagattgcag accgggaaga tgaatacaaa 361 aagcataggc ggaccatgat aatttcccca gagcgtcttg atccttttgc agatggaggg 421 aagacccctg atcctaaaat gaatgttagg acttacatgg atgtaatgcg agaacaacac 481 ttgactaaag aagaacgaga aattaggcaa cagctagcag aaaaagctaa agctggagaa 541 ctaaaagtcg tcaatggagc agcagcgtcc cagcctccat caaaacgaaa acggcgttgg 601 gatcaaacag ctgatcagac tcctggtgcc actcccaaaa aactatcaag ttgggatcag 661 gcagagaccc ctgggcatac tccttcctta agatgggatg agacaccagg tcgtgcaaag 721 ggaagcgaga ctcctggagc aaccccaggc tcaaaaatat gggatcctac acctagccac 781 acaccagcgg gagctgctac tcctggacga ggtgatacac caggccatgc gacaccaggc 841 catggaggcg caacttccag tgctcgtaaa aacagatggg atgaaacccc caaaacagag 901 agagatactc ctgggcatgg aagtggatgg gctgagactc ctcgaacaga tcgaggtgga 961 gattctattg gtgaaacacc gactcctgga gccagtaaaa gaaaatcacg gtgggatgaa 1021 acaccagcta gtcagatggg tggaagcact ccagttctga cccctggaaa gacaccaatt 1081 ggcacaccag ccatgaacat ggctacccct actccaggtc acataatgag tatgactcct 1141 gaacagcttc aggcttggcg gtgggaaaga gaaattgatg agagaaatcg cccactttct 1201 gatgaggaat tagatgctat gttcccagaa ggatataagg tacttcctcc tccagctggt 1261 tatgttccta ttcgaactcc agctcgaaag ctgacagcta ctccaacacc tttgggtggt 1321 atgactggtt tccacatgca aactgaagat cgaactatga aaagtgttaa tgaccagcca 1381 tctggaaatc ttccattttt aaaacctgat gatattcaat actttgataa actattggtt 1441 gatgttgatg aatcaacact tagtccagaa gagcaaaaag agagaaaaat aatgaagttg 1501 cttttaaaaa ttaagaatgg aacaccacca atgagaaagg ctgcattgcg tcagattact 1561 gataaagctc gtgaatttgg agctggtcct ttgtttaatc agattcttcc tctgctgatg 1621 tctcctacac ttgaggatca agagcgtcat ttacttgtga aagttattga taggatactg 1681 tacaaacttg atgacttagt tcgtccatat gtgcataaga tcctcgtggt cattgaaccg 1741 ctattgattg atgaagatta ctatgctaga gtggaaggcc tagagatcat ttctaatttg 1801 gcaaaggctg ctggtctggc tactatgatc tctaccatga gacctgatat agataacatg 1861 gatgagtatg tccgtaacac aacagctaga gcttttgctg ttgtagcctc tgccctgggc 1921 attccttctt tattgccctt cttaaaagct gtgtgcaaaa gcaagaagtc ctggcaagcg 1981 agacacactg gtattaagat tgtacaacag atagctattc ttatgggctg tgccatcttg 2041 ccacatctta gaagtttagt tgaaatcatt gaacatggtc ttgtggatga gcagcagaaa 2101 gttcggacca tcagtgcttt ggccattgct gccttggctg aagcagcaac tccttatggt
151
DNA sequence of full length SF3b155 continued 2161 atcgaatctt ttgattctgt gttaaagcct ttatggaagg gtatccgcca acacagagga 2221 aagggtttgg ctgctttctt gaaggctatt gggtatctta ttcctcttat ggatgcagaa 2281 tatgccaact actatactag agaagtgatg ttaatcctta ttcgagaatt ccagtctcct 2341 gatgaggaaa tgaaaaaaat tgtgctgaag gtggtaaaac agtgttgtgg gacagatggt 2401 gtagaagcaa actacattaa aacagagatt cttcctccct tttttaaaca cttctggcag 2461 cacaggatgg ctttggatag aagaaattac cgacagttag ttgatactac tgtggagttg 2521 gcaaacaaag taggtgcagc agaaattata tccaggattg tggatgatct gaaagatgaa 2581 gccgaacagt acagaaaaat ggtgatggag acaattgaga aaattatggg caatttggga 2641 gcagcagata ttgatcataa acttgaagaa caactgattg atggtattct ttatgctttc 2701 caagaacaga ctacagagga ctcagtaatg ttgaacggct ttggcacagt ggttaatgct 2761 cttggcaaac gagtcaaacc atacttgcct cagatctgtg gtacagtttt gtggcgttta 2821 aataacaaat ctgctaaagt taggcaacag gcagctgact tgatttctcg aactgctgtt 2881 gtcatgaaga cttgtcaaga ggaaaaattg atgggacact tgggtgttgt attgtatgag 2941 tatttgggtg aagagtaccc tgaagtattg ggcagcattc ttggagcact gaaggccatt 3001 gtaaatgtca taggtatgca taagatgact ccaccaatta aagatctgct gcctagactc 3061 acccccatct taaagaacag acatgaaaaa gtacaagaga attgtattga tcttgttggt 3121 cgtattgctg acaggggagc tgaatatgta tctgcaagag agtggatgag gatttgcttt 3181 gagcttttag agctcttaaa agcccacaaa aaggctattc gtagagccac agtcaacaca 3241 tttggttata ttgcaaaggc cattggccct catgatgtat tggctacact tctgaacaac 3301 ctcaaagttc aagaaaggca gaacagagtt tgtaccactg tagcaatagc tattgttgca 3361 gaaacatgtt caccctttac agtactccct gccttaatga atgaatacag agttcctgaa 3421 ctgaatgttc aaaatggagt gttaaaatcg ctttccttct tgtttgaata tattggtgaa 3481 atgggaaaag actacattta tgccgtaaca ccgttacttg aagatgcttt aatggataga 3541 gaccttgtac acagacagac ggctagtgca gtggtacagc acatgtcact tggggtttat 3601 ggatttggtt gtgaagattc gctgaatcac ttgttgaact atgtatggcc caatgtattt 3661 gagacatctc ctcatgtaat tcaggcagtt atgggagccc tagagggcct gagagttgct 3721 attggaccat gtagaatgtt gcaatattgt ttacagggtc tgtttcaccc agcccggaaa 3781 gtcagagatg tatattggaa aatttacaac tccatctaca ttggttccca ggacgctctc 3841 atagcacatt acccaagaat ctacaacgat gataagaaca cctatattcg ttatgaactt 3901 gactatatct tataatttta ttgtttattt tgtgtttaat gcacagctac ttcacacctt 3961 aaacttgctt tgatttggtg atgtaaactt ttaaacattg cagttcagtg tagaactggt 4021 catagaggaa gagctagaaa tccagtagca tgatttttaa ataacctgtc tttgtttttg 4081 atgttaaaca gtaaatgcca gtagtgacca agaacacagt gattatatac actatactgg 4141 agggatttca tttttaattc atctttatga agatttagaa ctcattcctt gtgtttaaag 4201 ggaatgttta attgagaaat aaacatttgt gtacaaaatg ctaaaaaaaa aaaaaaaaa
152
Protein sequence of full length SF3b155 1 MAKIAKTHED IEAQIREIQG KKAALDEAQG VGLDSTGYYD 40 41 QEIYGGSDSR FAGYVTSIAA TELEDDDDDY SSSTSLLGQK 80 81 KPGYHAPVAL LNDIPQSTEQ YDPFAEHRPP KIADREDEYK 120 121 KHRRTMIISP ERLDPFADGG KTPDPKMNVR TYMDVMREQH 160 161 LTKEEREIRQ QLAEKAKAGE LKVVNGAAAS QPPSKRKRRW 200 201 DQTADQTPGA TPKKLSSWDQ AETPGHTPSL RWDETPGRAK 240 241 GSETPGATPG SKIWDPTPSH TPAGAATPGR GDTPGHATPG 280 281 HGGATSSARK NRWDETPKTE RDTPGHGSGW AETPRTDRGG 320 321 DSIGETPTPG ASKRKSRWDE TPASQMGGST PVLTPGKTPI 360 361 GTPAMNMATP TPGHIMSMTP EQLQAWRWER EIDERNRPLS 400 401 DEELDAMFPE GYKVLPPPAG YVPIRTPARK LTATPTPLGG 440 441 MTGFHMQTED RTMKSVNDQP SGNLPFLKPD DIQYFDKLLV 480 481 DVDESTLSPE EQKERKIMKL LLKIKNGTPP MRKAALRQIT 520 521 DKAREFGAGP LFNQILPLLM SPTLEDQERH LLVKVIDRIL 560 561 YKLDDLVRPY VHKILVVIEP LLIDEDYYAR VEGLEIISNL 600 601 AKAAGLATMI STMRPDIDNM DEYVRNTTAR AFAVVASALG 640 641 IPSLLPFLKA VCKSKKSWQA RHTGIKIVQQ IAILMGCAIL 680 681 PHLRSLVEII EHGLVDEQQK VRTISALAIA ALAEAATPYG 720 721 IESFDSVLKP LWKGIRQHRG KGLAAFLKAI GYLIPLMDAE 760 761 YANYYTREVM LILIREFQSP DEEMKKIVLK VVKQCCGTDG 800 801 VEANYIKTEI LPPFFKHFWQ HRMALDRRNY RQLVDTTVEL 840 841 ANKVGAAEII SRIVDDLKDE AEQYRKMVME TIEKIMGNLG 880 881 AADIDHKLEE QLIDGILYAF QEQTTEDSVM LNGFGTVVNA 920 921 LGKRVKPYLP QICGTVLWRL NNKSAKVRQQ AADLISRTAV 960 961 VMKTCQEEKL MGHLGVVLYE YLGEEYPEVL GSILGALKAI 1000 1001 VNVIGMHKMT PPIKDLLPRL TPILKNRHEK VQENCIDLVG 1040 1041 RIADRGAEYV SAREWMRICF ELLELLKAHK KAIRRATVNT 1080 1081 FGYIAKAIGP HDVLATLLNN LKVQERQNRV CTTVAIAIVA 1120 1121 ETCSPFTVLP ALMNEYRVPE LNVQNGVLKS LSFLFEYIGE 1160 1161 MGKDYIYAVT PLLEDALMDR DLVHRQTASA VVQHMSLGVY 1200 1201 GFGCEDSLNH LLNYVWPNVF ETSPHVIQAV MGALEGLRVA 1240 1241 IGPCRMLQYC LQGLFHPARK VRDVYWKIYN SIYIGSQDAL 1280 1281 IAHYPRIYND DKNTYIRYEL DYIL 1304
153
SF3b155(1-462) Homo sapiens splicing factor 3b, subunit 1, 155kDa (SF3b1) SF3b155 (ACCESSION # NM_012433, NM_001005526 ) [gi 54112116:] Molecular weight: 50428.8 Daltons Theoretical pI: 5.91 Extinction coefficient: 67,005 (M*cm)-1 DNA sequence of SF3b155(1-462) 1 atggcgaaga tcgccaagac tcacgaagat attgaagcac agattcgaga aattcaaggc 61 aagaaggcag ctcttgatga agctcaagga gtgggcctcg attctacagg ttattatgac 121 caggaaattt atggtggaag tgacagcaga tttgctggat acgtgacatc aattgctgca 181 actgaacttg aagatgatga cgatgactat tcatcatcta cgagtttgct tggtcagaag 241 aagccaggat atcatgcccc tgtggcattg cttaatgata taccacagtc aacagaacag 301 tatgatccat ttgctgagca cagacctcca aagattgcag accgggaaga tgaatacaaa 361 aagcataggc ggaccatgat aatttcccca gagcgtcttg atccttttgc agatggaggg 421 aagacccctg atcctaaaat gaatgttagg acttacatgg atgtaatgcg agaacaacac 481 ttgactaaag aagaacgaga aattaggcaa cagctagcag aaaaagctaa agctggagaa 541 ctaaaagtcg tcaatggagc agcagcgtcc cagcctccat caaaacgaaa acggcgttgg 601 gatcaaacag ctgatcagac tcctggtgcc actcccaaaa aactatcaag ttgggatcag 661 gcagagaccc ctgggcatac tccttcctta agatgggatg agacaccagg tcgtgcaaag 721 ggaagcgaga ctcctggagc aaccccaggc tcaaaaatat gggatcctac acctagccac 781 acaccagcgg gagctgctac tcctggacga ggtgatacac caggccatgc gacaccaggc 841 catggaggcg caacttccag tgctcgtaaa aacagatggg atgaaacccc caaaacagag 901 agagatactc ctgggcatgg aagtggatgg gctgagactc ctcgaacaga tcgaggtgga 961 gattctattg gtgaaacacc gactcctgga gccagtaaaa gaaaatcacg gtgggatgaa 1021 acaccagcta gtcagatggg tggaagcact ccagttctga cccctggaaa gacaccaatt 1081 ggcacaccag ccatgaacat ggctacccct actccaggtc acataatgag tatgactcct 1141 gaacagcttc aggcttggcg gtgggaaaga gaaattgatg agagaaatcg cccactttct 1201 gatgaggaat tagatgctat gttcccagaa ggatataagg tacttcctcc tccagctggt 1261 tatgttccta ttcgaactcc agctcgaaag ctgacagcta ctccaacacc tttgggtggt 1321 atgactggtt tccacatgca aactgaagat cgaactatga aaagtgttaa tgaccagcca 1381 tctgga Protein sequence of SF3b155(1-462) 1 MAKIAKTHED IEAQIREIQG KKAALDEAQG VGLDSTGYYD 40 41 QEIYGGSDSR FAGYVTSIAA TELEDDDDDY SSSTSLLGQK 80 81 KPGYHAPVAL LNDIPQSTEQ YDPFAEHRPP KIADREDEYK 120 121 KHRRTMIISP ERLDPFADGG KTPDPKMNVR TYMDVMREQH 160 161 LTKEEREIRQ QLAEKAKAGE LKVVNGAAAS QPPSKRKRRW 200 201 DQTADQTPGA TPKKLSSWDQ AETPGHTPSL RWDETPGRAK 240 241 GSETPGATPG SKIWDPTPSH TPAGAATPGR GDTPGHATPG 280 281 HGGATSSARK NRWDETPKTE RDTPGHGSGW AETPRTDRGG 320 321 DSIGETPTPG ASKRKSRWDE TPASQMGGST PVLTPGKTPI 360 361 GTPAMNMATP TPGHIMSMTP EQLQAWRWER EIDERNRPLS 400 401 DEELDAMFPE GYKVLPPPAG YVPIRTPARK LTATPTPLGG 440 441 MTGFHMQTED RTMKSVNDQP SG 462 154
SF3b155(199-462) Homo sapiens splicing factor 3b, subunit 1, 155kDa (SF3b1) SF3b155 (ACCESSION # NM_012433, NM_001005526 ) [gi 54112116:] Molecular weight: 28,383.3 Daltons Theoretical pI: 6.39 Extinction coefficient: 53,634 (M*cm)-1 DNA sequence of SF3b155(199-462) 1atgcgttggg atcaaacagc tgatcagact cctggtgcca ctcccaaaaa actatcaagt 61 tgggatcagg cagagacccc tgggcatact ccttccttaa gatgggatga gacaccaggt 121 cgtgcaaagg gaagcgagac tcctggagca accccaggct caaaaatatg ggatcctaca 181 cctagccaca caccagcggg agctgctact cctggacgag gtgatacacc aggccatgcg 241 acaccaggcc atggaggcgc aacttccagt gctcgtaaaa acagatggga tgaaaccccc 301 aaaacagaga gagatactcc tgggcatgga agtggatggg ctgagactcc tcgaacagat 361 cgaggtggag attctattgg tgaaacaccg actcctggag ccagtaaaag aaaatcacgg 421 tgggatgaaa caccagctag tcagatgggt ggaagcactc cagttctgac ccctggaaag 481 acaccaattg gcacaccagc catgaacatg gctaccccta ctccaggtca cataatgagt 541 atgactcctg aacagcttca ggcttggcgg tgggaaagag aaattgatga gagaaatcgc 601 ccactttctg atgaggaatt agatgctatg ttcccagaag gatataaggt acttcctcct 661 ccagctggtt atgttcctat tcgaactcca gctcgaaagc tgacagctac tccaacacct 721 ttgggtggta tgactggttt ccacatgcaa actgaagatc gaactatgaa aagtgttaat 781 gaccagccat ctgga Protein sequence of SF3b155(199-462) 198 MRWDQTADQT PGATPKKLSS WDQAETPGHT PSLRWDETPG 237 238 RAKGSETPGA TPGSKIWDPT PSHTPAGAAT PGRGDTPGHA 277 278 TPGHGGATSS ARKNRWDETP KTERDTPGHG SGWAETPRTD 317 318 RGGDSIGETP TPGASKRKSR WDETPASQMG GSTPVLTPGK 357 358 TPIGTPAMNM ATPTPGHIMS MTPEQLQAWR WEREIDERNR 397 398 PLSDEELDAM FPEGYKVLPP PAGYVPIRTP ARKLTATPTP 437 438 LGGMTGFHMQ TEDRTMKSVN DQPSG 462
155
SF3b155(255-462) Homo sapiens splicing factor 3b, subunit 1, 155kDa (SF3b1) SF3b155 (ACCESSION # NM_012433, NM_001005526 ) [gi 54112116:] Molecular weight: 22,195.5 Daltons Theoretical pI: 6.34 Extinction coefficient: 31,234 (M*cm)-1 DNA sequence of SF3b155(255-462) 1 atggatccta cacctagcca cacaccagcg ggagctgcta ctcctggacg aggtgataca 61 ccaggccatg cgacaccagg ccatggaggc gcaacttcca gtgctcgtaa aaacagatgg 121 gatgaaaccc ccaaaacaga gagagatact cctgggcatg gaagtggatg ggctgagact 181 cctcgaacag atcgaggtgg agattctatt ggtgaaacac cgactcctgg agccagtaaa 241 agaaaatcac ggtgggatga aacaccagct agtcagatgg gtggaagcac tccagttctg 301 acccctggaa agacaccaat tggcacacca gccatgaaca tggctacccc tactccaggt 361 cacataatga gtatgactcc tgaacagctt caggcttggc ggtgggaaag agaaattgat 421 gagagaaatc gcccactttc tgatgaggaa ttagatgcta tgttcccaga aggatataag 481 gtacttcctc ctccagctgg ttatgttcct attcgaactc cagctcgaaa gctgacagct 541 actccaacac ctttgggtgg tatgactggt ttccacatgc aaactgaaga tcgaactatg 601 aaaagtgtta atgaccagcc atctgga Protein sequence of SF3b155(255-462) 254 MDPTPSHTPA GAATPGRGDT PGHATPGHGG ATSSARKNRW 293 294 DETPKTERDT PGHGSGWAET PRTDRGGDSI GETPTPGASK 333 334 RKSRWDETPA SQMGGSTPVL TPGKTPIGTP AMNMATPTPG 373 374 HIMSMTPEQL QAWRWEREID ERNRPLSDEE LDAMFPEGYK 413 414 VLPPPAGYVP IRTPARKLTA TPTPLGGMTG FHMQTEDRTM 453 454 KSVNDQPSG 462
156
SF3b155(1-254) Homo sapiens splicing factor 3b, subunit 1, 155kDa (SF3b1) SF3b155 (ACCESSION # NM_012433, NM_001005526 ) [gi 54112116:] Molecular weight: 28,251.2 Daltons Theoretical pI: 5.63 Extinction coefficient: 34,280 (M*cm)-1 DNA sequence of SF3b155(1-254) 1 atggcgaaga tcgccaagac tcacgaagat attgaagcac agattcgaga aattcaaggc 61 aagaaggcag ctcttgatga agctcaagga gtgggcctcg attctacagg ttattatgac 121 caggaaattt atggtggaag tgacagcaga tttgctggat acgtgacatc aattgctgca 181 actgaacttg aagatgatga cgatgactat tcatcatcta cgagtttgct tggtcagaag 241 aagccaggat atcatgcccc tgtggcattg cttaatgata taccacagtc aacagaacag 301 tatgatccat ttgctgagca cagacctcca aagattgcag accgggaaga tgaatacaaa 361 aagcataggc ggaccatgat aatttcccca gagcgtcttg atccttttgc agatggaggg 421 aagacccctg atcctaaaat gaatgttagg acttacatgg atgtaatgcg agaacaacac 481 ttgactaaag aagaacgaga aattaggcaa cagctagcag aaaaagctaa agctggagaa 541 ctaaaagtcg tcaatggagc agcagcgtcc cagcctccat caaaacgaaa acggcgttgg 601 gatcaaacag ctgatcagac tcctggtgcc actcccaaaa aactatcaag ttgggatcag 661 gcagagaccc ctgggcatac tccttcctta agatgggatg agacaccagg tcgtgcaaag 721 ggaagcgaga ctcctggagc aaccccaggc tcaaaaatat gg
Protein sequence of SF3b155(1-254) 1 MAKIAKTHED IEAQIREIQG KKAALDEAQG VGLDSTGYYD 40 41 QEIYGGSDSR FAGYVTSIAA TELEDDDDDY SSSTSLLGQK 80 81 KPGYHAPVAL LNDIPQSTEQ YDPFAEHRPP KIADREDEYK 120 121 KHRRTMIISP ERLDPFADGG KTPDPKMNVR TYMDVMREQH 160 161 LTKEEREIRQ QLAEKAKAGE LKVVNGAAAS QPPSKRKRRW 200 201 DQTADQTPGA TPKKLSSWDQ AETPGHTPSL RWDETPGRAK 240 241 GSETPGATPG SKIW 254
157
APPENDIX B
Structure and information about selected buffers
158
Acetate Buffer, also known as acetic acid
pKa=4.76, MW60.05
Notes: Acetic acid is a commonly used buffer at low pH values and is quite insensitive to temperature changes. Supplied as glacial acetic acid, which is very hygroscopic, and thus will absorb water and dilute itself unless tightly stoppered. BES, also called N,N-Bis-(2-hydroxyethyl)-2-aminoethanesulfonic Acid Notes: BES is a simple buffer that can also be Obtained as the sodium salt and which therefore can be made by mixing solutions of the conjugate acid and base, of by titrating a solution of the acid with strong base. The pKa=7.26 has been back corrected to the thermodynamic value, and will therefore be different to the working pKa’ (7.15) value cited in the literature BES has a very low ultraviolent aborbance at 260-280nm.
pKa=7.26, MW=213.3
Boric acid, also called boracic acid or orthoboric acid, is a mild acid often used as an antiseptic, insecticide, flame retardant, and as a precursor of other chemical compounds. It exists in the form of colorless crystals or a white powder and dissolves in water. It has the chemical formula H3BO3, sometimes written B(OH)3. pKa=9.23; MW=61.8 Notes: Boric Acid is not commonly used as a stand-alone buffer, but it finds application in Tris-Borate for electrophoresis in agarose gels. The ability of borate ions to bind to carbohydrates creates a ‘borate front’ that can assist in the electrophoresis separation. Boric Acid is used as a primary pH standard DEPC, also known as diethyl pyrocarbonate Notes: Moisture sensitive. DEPC derivitizes Histidine residues and is therefore an effective method to inactivate nucleases including RNAse. DEPC is irritating to the eyes, skin, and mucous membranes. It is a suspected carcinogen. It hydrolitically decomposes to CO2 & ethanol and
159
MW=162.1
may decompose to urethane (a possible carcinogen) in the presence of ammonia. MW=154.2
DTT, also called Dithiothreitol Notes: Protective agent for sulfhydryl groups (SH) groups. Reducing agent for disulfides in proteins and enzymes, biochemical research. DTT promotes proteins to unfold and maximize bonding. EDTA, also known as ethylenediaminetetraacetic acid
MW=292.2
Notes: Sequestrant of metal and metal deactivator. Also, know as a metal chelating agent.
Glycine, also called Aminoacetic acid
MW=60.05
Notes: Glycine is commonly used in electrophoresis systems. There is really little justification for using it as a general purpose buffer. As an amino acid, it has the potential to be a carbon and nitrogen source for microbial growth, it is a reactive primary amine, and it will find its way into samples that are destined for peptides sequencing, giving a glycine peak even when no glycine is expected. HEPES, also called 4-(2-Hydroxyethyl) piperazine-1-ethanesulfonic acid Notes: HEPES has a low absorbance in the ultraviolent, and is suitable for a wide range of applications. It has been used with great success in cell culture media. Buffers can be made by titration of the acid, or by mixture of the acid with the sodium of potassium salts.
160
pKa=7.66, MW=238.3
MOPS, also called 3-(N-Morpholino)-propanesulfonic acid Notes: A zwitterionic buffer is a structural analog to pKa=7.2MW=209.3 MES. It is a midrange pKa (7.2) with maximum water solubility and minimum solubility in all other solvents. It has minimal salt effects and minimal change in pKa with temperature. It is chemical and enzymatic stability. It has a minimal absorption in visible or UV range. PMSF, also called Phenylmethylsulfonyl fluoride Notes: Helps prevent protein degradation. Inhibits carboxypeptidase Y, chymotrypsin, factor Xa, papain, plasmin, proteinase K, subtilisin, thrombin, and trypsin. Non-inhibitory to cholineesterase. Less toxic than diisopropylfluorophosphate. Irreversible inhibitor of serine proteases.
MW=174.2
Toxic: Must be prepared fresh and added at several steps during sample preparation. Half-life of 1 hour, at pH 7.5. Stock solutions of 200 mM (100 mg/2.87 ml) in anhydrous solvents (MeOH, EtOH) are stable for at least 9 months at 4°C. Dilute 1:200 for a1 µM working solution.
TCEP, also called Tris(2carboxyethyl)phosphine Notes: Reduces organic disulfides to thiols rapidly and quantitatively in water. Selective for disulfides, does not react with other functional groups on proteins. More complete reductions than dithiothreitol ( DTT ) Cleland's reagent. Dilute (1mM) TCEP solutions react rapidly at room temperature. Reagent is odorless, non volatile and stable at room temperature.
161
MW=286.6
TRIS, also called Tris(hydroxymethyl)-aminomethane and Trizma Notes: A ubiquitious buffer, that has several problems, including a high temperatures sensitivity, reactivity as a primary amine, the need for Tris competent pH electrodes and some undesirable effects on some biological systems. Its continued use may be more to do with familiarity and to published recipes than to scientific justification. Urea, also called Carbamimidic acid
pKa=8.06, MW=121.14
pKa1=2.35, pKa2=9.78, MW=75.1
Laboratory use: Urea is a powerful protein denaturant. This property can be exploited to increase the solubility of some proteins. For this application it is used in concentrations up to 10M. Notes: It was the first organic compound to be artificially synthesized from inorganic starting materials. Urea is essentially a waste product: it has no physiological function. Humans produce a little uric acid as a result of purine breakdown. Indeed, excess uric acid production can lead to a type of arthritis known as gout. Used as an additive ingredient in cigarettes designed to enhance flavor.
162
REFERENCES Abovich N, Liao XC, Rosbash M. 1994. The yeast MUD2 protein: an interaction with PRP11 defines a bridge between commitment complexes and U2 snRNP addition. Genes Dev 8:843-854. Abovich N, Rosbash M. 1997. Cross-intron bridging interactions in the yeast commitment complex are conserved in mammals. Cell 89:403-412. Angell CA. 1982. In Water: a Comprehensive Treatise. New York: Plenum Press. Arnez JG, Steitz TA. 1994. Crystal structure of unmodified tRNA(Gln) complexed with glutaminyl-tRNA synthetase and ATP suggests a possible role for pseudo-uridines in stabilization of RNA structure. Biochemistry-Us 33:7560-7567. Avery OT, MacLeod, C.M., McCarty, M. 1944. Studies on the chemical nature of the substance inducing transformation of pneumonococcal types: induction of transformation by a deoxyribonucleic acid fraction isolated from pneumococcus type III. Journal of Experimental Medicine 79:137158. Babu CR, Hilser VJ, Wand AJ. 2004. Direct access to the cooperative substructure of proteins and the protein ensemble via cold denaturation. Nat Struct Mol Biol 11:352-357. Bax A, Clore GM, Gronenborn AM. 1990. H-1-H-1 Correlation Via Isotropic Mixing of C-13 Magnetization, a New 3-Dimensional Approach for Assigning H-1 and C-13 Spectra of C-13-Enriched Proteins. J Magnet Reson 88:425-431. Berget SM, Moore C, Sharp PA. 1977. Spliced segments at the 5' terminus of adenovirus 2 late mRNA. Proc Natl Acad Sci USA 74:3171-3175. Berglund JA, Chua K, Abovich N, Reed R, Rosbash M. 1997. The splicing factor BBP interacts specifically with the pre-mRNA branchpoint sequence UACUAAC. Cell 89:781-787. Birney E, Kumar S, Krainer AR. 1993. Analysis of the RNA-recognition motif and RS and RGG domains: conservation in metazoan pre-mRNA splicing factors. Nucleic Acids Res 21:5803-5816. Bobba A, Cavatorta P, Attimonelli M, Riccio P, Masotti L, Quagliariello E. 1990. Estimation of protein secondary structure from circular dichroism spectra: a critical examination of the CONTIN program. Protein Seq Data Anal 3:710. Boschitsch AH, Fenley MO. 2004. Hybrid boundary element and finite difference method for solving the nonlinear Poisson-Boltzmann equation. J Comput Chem 25:935-955. Burd CG, Dreyfuss G. 1994. RNA binding specificity of hnRNP A1: significance of hnRNP A1 high-affinity binding sites in pre-mRNA splicing. EMBO J 13:1197-1204. Burge CB, Tuschl, T., Sharp, P.A. 1999. Splicing of precursors to mRNAs by the spliceosomes. In: Gesteland, ed. The RNA World Second Edition. New York: Cold Spring Harbor Laboratory Press. pp 525-560.
163
Buzayan JM, Hampel A, Bruening G. 1986. Nucleotide sequence and newly formed phosphodiester bond of spontaneously ligated satellite tobacco ringspot virus RNA. Nucleic Acids Res 14:9729-9743. Chiara MD, Reed R. 1995. A 2-Step Mechanism for 5' and 3' Splice-Site Pairing. Nature 375:510-513. Chow LT, Gelinas RE, Broker TR, Roberts RJ. 1977. An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA. Cell 12:1-8. Chui HM, Desaulniers JP, Scaringe SA, Chow CS. 2002. Synthesis of helix 69 of Escherichia coli 23S rRNA containing its natural modified nucleosides, m(3)Psi and Psi. J Org Chem 67:8847-8854. Crick FH. 1958. On protein synthesis. Symp Soc Exp Biol 12:138-163. Dahm R. 2005. Friedrich Miescher and the discovery of DNA. Dev Biol 278:274288. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, Bax A. 1995. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6:277-293. Doyle ML. 1999. Titration Microcalorimetry. In Current Protocols in Protein Science. New York: John Wiley & Sons. Inc. pp 20.24.21-20.24.24. Durant PC, Davis DR. 1999. Stabilization of the anticodon stem-loop of tRNALys,3 by an A+-C base-pair and by pseudouridine. J Mol Biol 285:115-131. Forster AC, Symons RH. 1987. Self-cleavage of plus and minus RNAs of a virusoid and a structural model for the active sites. Cell 49:211-220. Gelinas RE, Chow LT, Roberts RJ, Broker TR, Klessig DF. 1977. The structure of late adenovirus type 2 messenger RNAs. Brookhaven Symp Biol:345347. Gelinas RE, Roberts RJ. 1977. One predominant 5'-undecanucleotide in adenovirus 2 late messenger RNAs. Cell 11:533-544. Golas MM, Sander B, Will CL, Luhrmann R, Stark H. 2003. Molecular architecture of the multiprotein splicing factor SF3b. Science 300:980-984. Golas MM, Sander B, Will CL, Luhrmann R, Stark H. 2005. Major conformational change in the complex SF3b upon integration into the spliceosomal U11/U12 di-snRNP as revealed by electron cryomicroscopy. Molecular Cell 17:869-883. Gorlach M, Wittekind M, Beckman RA, Mueller L, Dreyfuss G. 1992. Interaction of the RNA-binding domain of the hnRNP C proteins with RNA. EMBO J 11:3289-3295. Gottschalk A, Bartels C, Neubauer G, Luhrmann R, Fabrizio P. 2001. A novel yeast U2 snRNP protein, Snu17p, is required for the first catalytic step of splicing and for progression of spliceosome assembly. Molecular and Cellular Biology 21:3037-3046. Gozani O, Potashkin J, Reed R. 1998. A potential role for U2AF-SAP 155 interactions in recruiting U2 snRNP to the branch site. Molecular and Cellular Biology 18:4752-4760.
164
Grabski A, Mehler, M., & Drott, D. 2003. Unattended high-density cell growth and induction of protein expression with the Overnight Express Autoinduction System. inNovation 17:3-6. Greenfield N, Fasman GD. 1969. Computed circular dichroism spectra for the evaluation of protein conformation. Biochemistry-Us 8:4108-4116. Grzesiek S, Anglister J, Bax A. 1993. Correlation of Backbone Amide and Aliphatic Side-Chain Resonances in C-13/N-15-Enriched Proteins by Isotropic Mixing of C-13 Magnetization. J Magn Reson Ser B 101:114-119. Hall KB. 2002. RNA-protein interactions. Curr Opin Struct Biol 12:283-288. Hall KB, McLaughlin LW. 1992. Properties of pseudouridine N1 imino protons located in the major groove of an A-form RNA duplex. Nucleic Acids Res 20:1883-1889. Hall KB, Stump WT. 1992. Interaction of N-terminal domain of U1A protein with an RNA stem/loop. Nucleic Acids Res 20:4283-4290. Hwang TL, Mori S, Shaka AJ, vanZijl PCM. 1997. Application of phasemodulated CLEAN chemical EXchange spectroscopy (CLEANEX-PM) to detect water-protein proton exchange and intermolecular NOEs. J Am Chem Soc 119:6203-6204. Jamison SF, Crow A, Garcia-Blanco MA. 1992. The spliceosome assembly pathway in mammalian extracts. Mol Cell Biol 12:4279-4287. Johnson BA, Blevins RA. 1994. NMRView: A computer program for the visualization and analysis of NMR data. J Biomolecular NMR 4:603-614. Kay LE, Keifer, P., and Saarinen T. 1992. Pure absorption gradient-enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity. Journal of American Chemical Society 114:10663-10665. Kay LE, Nicholson LK, Delaglio F, Bax A, Torchia DA. 1992. Pulse Sequences for Removal of the Effects of Cross-Correlation between Dipolar and Chemical-Shift Anisotropy Relaxation Mechanism on the Measurement of Heteronuclear T1 and T2 Values in Proteins. J Magn Reson 97:359-375. Kay LE, Xu GY, Singer AU, Muhandiram DR, Formankay JD. 1993. A GradientEnhanced HCCH TOCSY Experiment for Recording Side-Chain H-1 and C-13 Correlations in H2O Samples of Proteins. J Magn Reson Ser B 101:333-337. Kay LE, Xu GY, Yamazaki T. 1994. Enhanced-Sensitivity Triple-Resonance Spectroscopy with Minimal H2O Saturation. J Magn Reson Ser A 109:129-133. Kendrew JC, Perutz MF. 1957. X-ray studies of compounds of biological interest. Annu Rev Biochem 26:327-372. Kerwood DJ, Cavaluzzi MJ, Borer PN. 2001. Structure of SL4 RNA from the HIV1 packaging signal. Biochemistry-Us 40:14518-14529. Kim SH, Suddath FL, Quigley GJ, McPherson A, Sussman JL, Wang AH, Seeman NC, Rich A. 1974. Three-dimensional tertiary structure of yeast phenylalanine transfer RNA. Science 185:435-440. Kneller TDGaDG. 1989. SPARKY 3. University of California, San Francisco. Konarska MM, Sharp PA. 1987. Interactions between small nuclear ribonucleoprotein particles in formation of spliceosomes. Cell 49:763-774.
165
Kramer A, Gruter P, Groning K, Kastner B. 1999. Combined biochemical and electron microscopic analyses reveal the architecture of the mammalian U2 snRNP. Journal of Cell Biology 145:1355-1368. Kruger K, Grabowski PJ, Zaug AJ, Sands J, Gottschling DE, Cech TR. 1982. Self-splicing RNA: autoexcision and autocyclization of the ribosomal RNA intervening sequence of Tetrahymena. Cell 31:147-157. Kuo MY, Sharmeen L, Dinter-Gottlieb G, Taylor J. 1988. Characterization of selfcleaving RNA sequences on the genome and antigenome of human hepatitis delta virus. J Virol 62:4439-4444. Lerner MR, Boyle JA, Mount SM, Wolin SL, Steitz JA. 1980. Are snRNPs involved in splicing? Nature 283:220-224. Lerner MR, Steitz JA. 1979. Antibodies to small nuclear RNAs complexed with proteins are produced by patients with systemic lupus erythematosus. Proc Natl Acad Sci U S A 76:5495-5499. Li X, Romero P, Rani M, Dunker AK, Obradovic Z. 1999. Predicting protein disorder for N-, C-, and internal regions. Genome Informatics 10:30-40. Li Y, Bevilacqua PC, Mathews D, Turner DH. 1995. Thermodynamic and activation parameters for binding of a pyrene-labeled substrate by the Tetrahymena ribozyme: docking is not diffusion-controlled and is driven by a favorable entropy change. Biochemistry-Us 34:14394-14399. Lobley A, Whitmore L, Wallace BA. 2002. DICHROWEB: An Interactive Website for the Analysis of Protein Secondary Structure from Circular Dichroism Data. Bioinformatics 18:211-212. Logan TM, Olejniczak ET, Xu RX, Fesik SW. 1992. Side-Chain and Backbone Assignments in Isotopically Labeled Proteins from 2 Heteronuclear Triple Resonance Experiments. Febs Lett 314:413-418. Logan TM, Olejniczak ET, Xu RX, Fesik SW. 1993. A General-Method for Assigning NMR-Spectra of Denatured Proteins Using 3D HC(CO)NHTOCSY Triple Resonance Experiments. Journal of Biomolecular NMR 3:225-231. Ma X, Zhao X, Yu YT. 2003. Pseudouridylation (Psi) of U2 snRNA in S. cerevisiae is catalyzed by an RNA-independent mechanism. EMBO J 22:1889-1897. MacMillan AM, Query CC, Allerson CR, Chen S, Verdine GL, Sharp PA. 1994. Dynamic association of proteins with the pre-mRNA branch region. Genes & Development 8:3008-3020. Marat K. 2005. SpinWorks. University of Manitoba. Meischer F. 1871. Ueber die chemische Zusammensetzung der Eiterzellen. MedChem Unters 4:441– 460. Meissner A, Schulte-Herbruggen T, Briand J, Sorensen OW. 1998. Double spinstate-selective coherence transfer. Application for two-dimensional selection of multiplet components with long transverse relaxation times. Mol Phys 95:1137-1142. Michaud S, Reed R. 1991. An ATP-independent complex commits pre-mRNA to the mammalian spliceosome assembly pathway. Genes Dev 5:2534-2546.
166
Mikulecky PJ, Feig AL. 2002. Cold denaturation of the hammerhead ribozyme. J Am Chem Soc 124:890-891. Mikulecky PJ, Feig AL. 2004. Heat capacity changes in RNA folding: application of perturbation theory to hammerhead ribozyme cold denaturation. Nucleic Acids Res 32:3967-3976. Montelione GT, Wuthrich K, Burgess AW, Nice EC, Wagner G, Gibson KD, Scheraga HA. 1992. Solution Structure of Murine Epidermal GrowthFactor Determined by NMR-Spectroscopy and Refined by Energy Minimization with Restraints. Biochemistry-Us 31:236-249. Moore MJ, Sharp PA. 1992. Site-specific modification of pre-mRNA: the 2'hydroxyl groups at the splice sites. Science 256:992-997. Moore MJ, Sharp PA. 1993. Evidence for two active sites in the spliceosome provided by stereochemistry of pre-mRNA splicing. Nature 365:364-368. Muhandiram DR, Kay LE. 1994. Gradient-Enhanced Triple-Resonance 3Dimensional NMR Experiments with Improved Sensitivity. J Magn Reson Ser B 103:203-216. Mulder GJ. 1839. Uber die Zusammensetzung einiger thierische Substanzen. J Prakt Chem 16:129-152. Nagai K, Oubridge C, Jessen TH, Li J, Evans PR. 1990. Crystal structure of the RNA-binding domain of the U1 small nuclear ribonucleoprotein A. Nature 348:515-520. Narlikar GJ, Herschlag D. 1996. Isolation of a local tertiary folding transition in the context of a globally folded RNA. Nat Struct Biol 3:701-710. Newby MI, Greenbaum NL. 2001. A conserved pseudouridine modification in eukaryotic U2 snRNA induces a change in branch-site architecture. RNA 7:833-845. Newby MI, Greenbaum NL. 2002a. Investigation of Overhauser effects between pseudouridine and water protons in RNA helices. Proc Natl Acad Sci U S A 99:12697-12702. Newby MI, Greenbaum NL. 2002b. Sculpting of the spliceosomal branch site recognition motif by a conserved pseudouridine. Nature Structural Biology 9:958-965. Nonin S, Leroy JL, Gueron M. 1995. Terminal base pairs of oligodeoxynucleotides: imino proton exchange and fraying. BiochemistryUs 34:10652-10659. Oesper RE. 1966. Jorpes,Je - Jac Berzelius - His Life and Work. J Chem Educ 43:A1100. Peebles CL, Perlman PS, Mecklenburg KL, Petrillo ML, Tabor JH, Jarrell KA, Cheng HL. 1986. A self-splicing RNA excises an intron lariat. Cell 44:213223. Pervushin K, Riek R, Wider G, Wuthrich K. 1997. Attenuated T-2 relaxation by mutual cancellation of dipole-dipole coupling and chemical shift anisotropy indicates an avenue to NMR structures of very large biological macromolecules in solution. P Natl Acad Sci USA 94:12366-12371.
167
Pervushin K, Riek R, Wider G, Wuthrich K. 1998a. Transverse relaxationoptimized spectroscopy (TROSY) for NMR studies of aromatic spin systems in C-13-labeled proteins. J Am Chem Soc 120:6394-6400. Pervushin KV, Wider G, Wuthrich K. 1998b. Single transition-to-single transition polarization transfer (ST2-PT) in [N-15,H-1]-TROSY. Journal of Biomolecular NMR 12:345-348. Poppe L, van Halbeek H. 1994. NMR spectroscopy of hydroxyl protons in supercooled carbohydrates. Nat Struct Biol 1:215-216. Query CC, McCaw PS, Sharp PA. 1997. A minimal spliceosomal complex A recognizes the branch site and polypyrimidine tract. Molecular and Cellular Biology 17:2944-2953. Query CC, Moore MJ, Sharp PA. 1994. Branch nucleophile selection in premRNA splicing: evidence for the bulged duplex model. Genes & Development 8:587-597. Query CC, Strobel SA, Sharp PA. 1996. Three recognition events at the branchsite adenine. EMBO J 15:1392-1402. Rance MS, O. W.; Bodenhausen, G.; Wagner, G.; Ernst, R. R.; Wüthrich, K. 1983. Improved spectral resolution in COSY NMR spectra of proteins via double quantum filtering. Biochem Biophys Res Commun 117:479-485. Reed RE, Baer MF, Guerrier-Takada C, Donis-Keller H, Altman S. 1982. Nucleotide sequence of the gene encoding the RNA subunit (M1 RNA) of ribonuclease P from Escherichia coli. Cell 30:627-636. Riek R, Pervushin K, Wuthrich K. 2000. TROSY and CRINEPT: NMR with large molecular and supramolecular structures in solution. Trends Biochem Sci 25:462-468. Romero P, Obradovic Z, Dunker AK. 1997a. Sequence data analysis for long disordered regions prediction in the calcineurin family. Genome Informatics 8:110-124. Romero P, Obradovic Z, Kissinger CR, Villafranca JE, Dunker AK. 1997b. Identifying Disordered Regions in Proteins from Amino Acid Sequences. Proc IEEE International Conference on Neural Networks:90-95. Romero P, Obradovic Z, Li X, Garner E, Brown C, Dunker AK. 2001. Sequence complexity of disordered protein. Struct Funct Gen 42:38-48. Ruskin B, Krainer AR, Maniatis T, Green MR. 1984. Excision of an intact intron as a novel lariat structure during pre-mRNA splicing in vitro. Cell 38:317331. Sattler M, Schleucher J, Griesinger C. 1999. Heteronuclear multidimensional NMR experiments for the structure determination of proteins in solution employing pulsed field gradients. Prog Nucl Mag Res Sp 34:93-158. Saville BJ, Collins RA. 1990. A site-specific self-cleavage reaction performed by a novel RNA in Neurospora mitochondria. Cell 61:685-696. Schellenberg MJ, Edwards RA, Ritchie DB, Kent OA, Golas MM, Stark H, Luhrmann R, Glover JN, MacMillan AM. 2006. Crystal structure of a core spliceosomal protein interface. Proc Natl Acad Sci U S A 103:1266-1271.
168
Schroeder KT, Skalicky JJ, Greenbaum NL. 2005. NMR spectroscopy of RNA duplexes containing pseudouridine in supercooled water. RNA 11:10121016. Shan X, Gardner KH, Muhandiram DR, Rao NS, Arrowsmith CH, Kay LE. 1996. Assignment of N-15, C-13(alpha), C-13(beta), and HN resonances in an N-15, C-13, H-2 labeled 64 kDa trp repressor-operator complex using triple-resonance NMR spectroscopy and H-2-decoupling. J Am Chem Soc 118:6570-6579. Silver MS, Joseph RI, Hoult DI. 1984. Highly Selective Pi/2 and Pi-Pulse Generation. J Magn Reson 59:347-351. Skalicky JJ, Mills JL, Sharma S, Szyperski T. 2001. Aromatic ring-flipping in supercooled water: implications for NMR-based structural biology of proteins. J Am Chem Soc 123:388-397. Skalicky JJ, Sukumaran DK, Mills JL, Szyperski T. 2000. Toward Structural Biology in Supercooled Water. J Am Chem Soc 122:3230-3231. Sklenar V, Bax A. 1987. Spin-Echo Water Suppression for the Generation of Pure-Phase Two-Dimensional NMR-Spectra. J Magn Reson 74:469-479. Spadaccini R, Reidt U, Dybkov O, Will C, Frank R, Stier G, Corsini L, Wahl MC, Luhrmann R, Sattler M. 2006. Biochemical and NMR analyses of an SF3b155-p14-U2AF-RNA interaction network involved in branch point definition during pre-mRNA splicing. RNA 12:410-425. St Johnston D, Brown NH, Gall JG, Jantsch M. 1992. A conserved doublestranded RNA-binding domain. Proc Natl Acad Sci U S A 89:1097910983. Swapna GVT, Montelione GT. 1999. Sensitivity-enhanced sim-CT HMQC PFGHBHA(CO)NH and PFG-CBCA(CO)NH triple-resonance experiments. J Magn Reson 137:437-442. Szyperski T, Mills JL, Perl D, Balbach J. 2006. Combined NMR-observation of cold denaturation in supercooled water and heat denaturation enables accurate measurement of DeltaC (p) of protein unfolding. Eur Biophys J 35:363-366. Takagi Y, Warashina M, Stec WJ, Yoshinari K, Taira K. 2001. Recent advances in the elucidation of the mechanisms of action of ribozymes. Nucleic Acids Res 29:1815-1834. Tanner NK, Cech TR. 1985. Self-catalyzed cyclization of the intervening sequence RNA of Tetrahymena: inhibition by methidiumpropyl.EDTA and localization of the major dye binding sites. Nucleic Acids Res 13:77597779. Thickman KR, Swenson MC, Kabogo JM, Gryczynski Z, Kielkopf CL. 2006. Multiple U2AF65 binding sites within SF3b155: thermodynamic and spectroscopic characterization of protein-protein interactions among premRNA splicing factors. J Mol Biol 356:664-683. Valadkhan S, Manley JL. 2001. Splicing-related catalysis by protein-free snRNAs. Nature 413:701-707. Valadkhan S, Manley JL. 2003. Characterization of the catalytic activity of U2 and U6 snRNAs. RNA 9:892-904.
169
van der Veen R, Arnberg AC, van der Horst G, Bonen L, Tabak HF, Grivell LA. 1986. Excised group II introns in yeast mitochondria are lariats and can be formed by self-splicing in vitro. Cell 44:225-234. Watson JD, Crick FH. 1953. Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid. Nature 171:737-738. Weaver R. 2002. Molecular Biology, 2nd edition. Will CL, Schneider C, MacMillan AM, Katopodis NF, Neubauer G, Wilm M, Luhrmann R, Query CC. 2001. A novel U2 and U11/U12 snRNP protein that associates with the pre-mRNA branch site. EMBO J 20:4536-4546. Will CL, Urlaub H, Achsel T, Gentzel M, Wilm M, Luhrmann R. 2002. Characterization of novel SF3b and 17S U2 snRNP proteins, including a human Prp5p homologue and an SF3b DEAD-box protein. EMBO J 21:4978-4988. Williamson MP, Havel TF, Wuthrich K. 1985. Solution conformation of proteinase inhibitor IIA from bull seminal plasma by 1H nuclear magnetic resonance and distance geometry. J Mol Biol 182:295-315. Wincott F, DiRenzo A, Shaffer C, Grimm S, Tracz D, Workman C, Sweedler D, Gonzalez C, Scaringe S, Usman N. 1995. Synthesis, deprotection, analysis and purification of RNA and ribozymes. Nucleic Acids Res 23:2677-2684. Wittekind M, Gorlach M, Friedrichs M, Dreyfuss G, Mueller L. 1992. H-1, C-13, and N-15 NMR Assignments and Global Folding Pattern of the RNABinding Domain of the Human Hnrnp C-Proteins. Biochemistry-Us 31:6254-6265. Wong I, Lohman TM. 1993. A double-filter method for nitrocellulose-filter binding: application to protein-nucleic acid interactions. Proc Natl Acad Sci U S A 90:5428-5432. Xu D, Greenbaum NL, Fenley MO. 2005. Recognition of the spliceosomal branch site RNA helix on the basis of surface and electrostatic features. Nucleic Acids Research 33:1154-1161. Yamazaki T, Lee W, Arrowsmith CH, Muhandiram DR, Kay LE. 1994. A Suite of Triple-Resonance NMR Experiments for the Backbone Assignment of N15, C-13, H-2 Labeled Proteins with High-Sensitivity. J Am Chem Soc 116:11655-11666. Yang DW, Kay LE. 1999. Improved (HN)-H-1-detected triple resonance TROSYbased experiments. Journal of Biomolecular NMR 13:3-10. Yarian CS, Basti MM, Cain RJ, Ansari G, Guenther RH, Sochacka E, Czerwinska G, Malkiewicz A, Agris PF. 1999. Structural and functional roles of the N1and N3-protons of psi at tRNA's position 39. Nucleic Acids Res 27:35433549. Yu YT, Shu MD, Steitz JA. 1998. Modifications of U2 snRNA are required for snRNP assembly and pre-mRNA splicing. Embo J 17:5783-5795. Zheng X, Bevilacqua PC. 2000. Straightening of bulged RNA by the doublestranded RNA-binding domain from the protein kinase PKR. Proc Natl Acad Sci U S A 97:14162-14167.
170
BIOGRAPHICAL SKETCH Kersten Schroeder was born on Thursday February 2, 1978 in Chicago, Illinois at Rush-Presbyterian-St. Luke's Medical Center (1653 West Congress Parkway Chicago, IL 60612). He was born to Debra Eliane Schroeder and David Wayne Schroeder on that snowy February morning at 8:44 am. His family moved to Jacksonville, FL in July of 1982. He started kindergarten as a pudgy little kid and started playing recreational soccer and took a liking to it right away in August of 1983. In the summer after 1St grade, he helped his mom raise money for the Muscular Dystrophy Association and decided then that he wanted to help find a way to cure this disease and other diseases. Education • • •
1999-present: Ph.D. Candidate in Biochemistry. Florida State University, Tallahassee, FL. 1996-1999: B.S., Microbiology. University of Florida, Gainesville, FL. 1992-1996: H.S. Diploma with honors. Fletcher High School, Neptune Beach, FL.
Employment • • • • • •
1999-present: Research/Teaching Assistant. Florida State University, Tallahassee, FL. 1998-1999: Biochemistry Research Assistant. University of Florida, Gainesville, FL. 1997-1999: Organic Laboratory Assistant. University of Florida, Gainesville, FL. 1996-1997: Cashier and cook. Fazoli’s, Gainesville, FL. 1994-1996: Cashier and cook. Bono’s Bar-B-Q, Neptune Beach, FL and Gainesville, FL. 1992-1995: Host, fry cook, and dishwasher: Sun Dog Diner, Neptune Beach, FL.
Publications •
• •
•
Kersten T. Schroeder, Amy N. Bryant, Jörg Schlatterer, Luis Guerrero, Christopher X. Staudinger, and Nancy L. Greenbaum. Specific recognition of the ψ-dependent premRNA branch site helix by U2 snRNP protein p14. In Preparation. Kersten T. Schroeder, Jack J. Skalicky and Nancy L. Greenbaum. 2005. NMR spectroscopy of RNA duplexes containing pseudouridine in supercooled water. RNA 11:1012-1016. Ying Xiong, Kersten T. Schroeder, Nancy L. Greenbaum, Christopher L. Hendrickson and Alan G. Marshall. 2004. Improved mass analysis of oligoribonucleotides by 13C,15N double depletion and electrospray ionization FT-ICR mass spectrometry. Analytical Chemistry 76:1804-1809. Sonia Rodriguez, Kersten T. Schroeder, Margaret M. Kayser and Jon D. Stewart. 2000. Asymmetric synthesis of beta-hydroxy esters and alpha-alkyl-beta-hydroxy esters by recombinant Escherichia coli expressing enzymes from baker's yeast. Journal of Organic Chemistry 65: 2586-2587.
Awards • • • • • • •
2005-2006: Joseph M. Schor Fellowship in Biochemistry. 2005: Who’s Who among Students in American Universities & Colleges. 2005: SGA Student Senator of the Month (September 2005). 2004: Florida State University Student Seminole Award. 2004: Florida State University Congress of Graduate Students Service Award. 2001-2002: National Science Foundation Research Training Fellowship. 1999-2001: Hoffman Teaching Award. Florida State University, Tallahassee, FL
171
Presentations at Scientific Meetings Oral Presentations • RNA Society Meeting, May 2005, Banff, Alberta, Canada. “Interaction of U2 snRNP protein p14 with the pre-mRNA branch site.” • International Society of Magnetic Resonance (ISMAR) Meeting, October 2004, Jacksonville, FL. “Solution NMR spectroscopy in supercooled water of RNA helices containing pseudouridine.” • Southeast Regional Magnetic Resonance Conference (SEMRC), October 2003, Tallahassee, FL. “Structural role of protons of pseudouridine in RNA duplexes as measured by NMR experiments in supercooled water.” • Florida Chapter of American Chemical Society (FAME) Meeting, May 2003, Orlando, FL. “NMR experiments of RNA duplexes in supercooled water.” Poster Presentations • International Society of Magnetic Resonance (ISMAR) Meeting, October 2004, Jacksonville, FL. “Interactions of Spliceosomal Proteins SF3b p14 and SF3b155 with the Pre-mRNA Branch Site.” • RNA Society Meeting, June 2004, Madison, WI. “Role of pseudouridine imino protons in stabilization of the spliceosomal branch site helix as observed by NMR spectroscopy in supercooled water.” • Florida Chapter of American Chemical Society (FAME) Meeting, May 2004, Orlando, FL. “Interaction of the pre-mRNA branch site with SF3b p14 and SF3b155.” • Florida Chapter of American Chemical Society (FAME) Meeting, May 2004, Orlando, FL. “Role of pseudouridine imino proton in stabilization of the spliceosomal branch site helix as observed by NMR spectroscopy in supercooled water.” • RNA Society Meeting, June 2002, Madison, WI. “Interaction of the p14 with the pre-mRNA intron branch site.” • Florida Chapter of American Chemical Society (FAME) Meeting, May 2001, Orlando, FL. “Probing the structural features of Domain 6 of the Group II intron.” • Florida Chapter of American Chemical Society (FAME) Meeting, May 1999, Orlando, FL. “Designing E. coli expression systems for Baker's yeast reductases and determining their substrate specificities.”
Service Scientific • • • • • • • •
2005: Panel Member, GradQuest: Graduate School Open House. Florida State University, Tallahassee, FL. 2005-present: Capitol Hill Day Participant for Biomedical Research Advocacy, Congressional Liaison Committee (CLC) of the Joint Steering Committee for Public Policy (JSC). Washington D.C. 2005-present: Member, Department of Chemistry & Biochemistry Student-Faculty Relations Committee. Florida State University, Tallahassee, FL. 2004-2005: Chemistry Tutor for the Women in Mathematics, Science and Engineering Program. Florida State University, Tallahassee, FL. 2001-2006: Volunteer for National High Magnetic Field Laboratory Open House. Tallahassee, FL. 2000-present: Undergraduate Chemistry Tutor. Florida State University, Tallahassee, FL. 2000-present: Mentor of Undergraduate Students in the Laboratory. Florida State University, Tallahassee, FL. 2000-2003: Volunteer Regional Science Fair Judge.
Teaching •
Spring 2006: Teaching Assistant, Recitation. General Chemistry I, CHM1045C (C. Jackson).
172
• • • • • • •
Spring 2003: Teaching Assistant, Recitation. General Chemistry I, CHM1045C (S. Dillon). Fall 2002: Teaching Assistant, Recitation. General Chemistry I, CHM1045 (A. Stiegman). Spring 2000: Teaching Assistant, Laboratory. General Biochemistry Laboratory, BCH4053L (R. Rill). Spring 2000: Teaching Assistant, Grader. General Biochemistry II, BCH4054 (Q. Sang). Spring 2000: Teaching Assistant, Grader. General Biochemistry I, BCH4053 (T. Logan). Spring 2000: Teaching Assistant, Grader. General Chemistry, CHM1020 (R. Light). Fall 1999: Teaching Assistant, Laboratory. General Chemistry I Laboratory, CHM1045L (B. Pulliam).
Student Government and Clubs • • • • • • • • • • • • • • • • • • • • • • • • •
2006: Member, Golden Opportunity Committee for Council of Honor Societies. Florida State University, Tallahassee, FL. 2005-present: Member, Garnet and Gold Key. FSU’s Premiere Leadership Honor Society. Florida State University, Tallahassee, FL. 2005-present: Member, Omicron Delta Kappa Society. National Leadership Honor Society. Florida State University, Tallahassee, FL. 2005: Member, SGA Student Senate-Congress of Graduate Students Mediation Committee. Florida State University, Tallahassee, FL. 2005: Voting Member of Student Health Insurance Symposium. Florida State University, Tallahassee, FL. 2005-present: Board Member, Agin’ Gator Soccer Society. Gainesville, FL. 2004-present: Member, Seminole Torchbearers. FSU, Tallahassee, FL. 2004-2005: SGA Graduate Studies Senator, 57th Student Senate. FSU, Tallahassee, FL. 2004-2005: Representative, 13th Congress of Graduate Students. FSU, Tallahassee, FL. 2004: Student Member: FSU SACS Accreditation Committee: Quality Enhancement Plan. FSU, Tallahassee, FL. th 2003-2004: Representative, 12 Congress of Graduate Students. FSU, Tallahassee, FL. th 2003-2004: Financial Officer for 12 Congress of Graduate Students. FSU, Tallahassee, FL. th 2002-2003: Representative, 11 Congress of Graduate Students. FSU, Tallahassee, FL. th 2002-2003: Financial Officer for 11 Congress of Graduate Students. FSU, Tallahassee, FL. th 2001-2002: Representative, 10 Congress of Graduate Students. FSU, Tallahassee, FL. th 2001-2002: Financial Officer for 10 Congress of Graduate Students. FSU, Tallahassee, FL. th 2000-2001: Representative, 9 Congress of Graduate Students. FSU, Tallahassee, FL. th 2000-2001: Financial Officer for 9 Congress of Graduate Students. FSU, Tallahassee, FL. th 1999-2000: Representative, 8 Congress of Graduate Students. FSU, Tallahassee, FL. 1999-2000: Financial Officer for 8th Congress of Graduate Students. FSU, Tallahassee, FL. 1999-present: Member, Agin’ Gator Soccer Society. Gainesville, FL. 1998-1999: President of the UF Men’s Soccer Club. University of Florida, Gainesville, FL. 1998-1999: Budget Chairman for Sports Club Council. University of Florida, Gainesville, FL. 1997-1999: Member, Sports Club Council. University of Florida, Gainesville, FL. 1996-1999: Member, UF Men’s Soccer Club. University of Florida. Gainesville, FL.
Community Service • • • • • • • • • •
2004: Captain of “The Green Team” in FSU Relay for Life. Tallahassee, FL. 2001-2003: Soccer Coach for travel soccer, Florida Youth Soccer Association. Tallahassee United Futból Club, Tallahassee, FL. 2001-present: State Soccer Referee, Grade 5. Florida State Referees. 2001-present: College Soccer Referee. National Intercollegiate Soccer Officials Association. 1999-2001: State Soccer Referee, Grade 6. Florida State Referees. 1997-1999: High School Soccer Coach. Oak Hall School, Gainesville, FL. 1996-1999: Soccer Referee, Grade 7. Florida State Referees. 1994-1996: Soccer Coach for recreational soccer, Beach’s Association Soccer League. First Coast United Soccer Club, Jacksonville Beach, FL. 1991-1996: Soccer Referee, Grade 8. Florida State Referees. 1990-1996: Volunteer in community cleanup, soup kitchens and building homes. Habitat for Humanities of Jacksonville. Beach United Methodist Ministry. Jacksonville Beach, FL.
173
