Understanding and Engineering the Collagen Triple Helix

UNDERSTANDING AND ENGINEERING THE COLLAGEN TRIPLE HELIX Matthew Donald Shoulders Under the supervision of Professor Ronald T. Raines At the University of Wisconsin-Madison This thesis presents a hypothesis-driven approach to collagen research that integrates the power of organic chemistry with the tools of biophysics to enhance our understanding of proline conformation and collagen structure and stability. Collagen is an important structural protein and the most abundant protein in animals. The amino acid proline contributes greatly to the structure and stability of this essential protein. My efforts toward designing new means to control proline conformation, in tandem with applying those studies for designing collagen mimics with unique properties and exploring the physicochemical basis of the structure and stability of collagen, are presented. The findings reported here have broad implications both for collagen and for protein engineering efforts focused on many other natural protein structures. Chapter 1 reviews recent findings in the collagen field and brings the reader up-to-date with a modem understanding of collagen triple-helix structure and stability. In Chapter 2, I recount the preparation of a new class of hyperstable collagen triple helices endowed with stability by steric effects rather than stereoelectronic effects. I explore the conformational preferences of 4methylprolines and incorporate them in collagen-related peptides to test the efficacy of steric effects for stabilizing the triple helix. Steric effects induced by proline 4-methylation reiterate previously discovered stereoelectronic effects on the collagen triple helix. Such fundamental interplay between steric and stereoelectronic effects was previously unknown in proteins and 11 provides a new means to modulate conformational stability. Notably, these steric effects on triple-helix stability are additive because the methyl groups protrude radially from the folded triple helix (stereoelectronic effects on the triple helix are not additive). This finding suggests the possibility of judicious integration of steric and stereoelectronic effects to generate extraordinarily hyperstable triple helices. Chapter 3 describes an experiment designed to understand the ongm of the stability conferred on the collagen triple helix by fluorination. The hyperstability of triple helices containing (2S,4R)-4-fluoroproline (Flp) has been attributed by some researchers to a hydrophobic effect rather than a stereoelectronic effect. I tested this hypothesis by replacing Hyp with (2S)-4,4-difluoroproline (Dfp) in collagen-related polypeptides. Dfp retains the hydrophobicity of Flp, but lacks the ability of Flp to define proline ring conformation. Unlike Flp, Dfp does not endow triple helices with elevated stability, indicating that the hyperstability conferred by Flp is not due to the hydrophobic effect. Since the hyperstability conferred on the collagen triple helix by fluorination is not due to the hydrophobic effect, in Chapter 4 I recount my efforts toward understanding the origin of that stabilization, which I ultimately demonstrate is attributable to preorganization. In the peptides reported in Chapter 4, pre organization is potentially achieved by installing proline side-chain substituents that impose stereoelectronic and steric effects, which restrict peptide main-chain torsion angles. Specifically, replacing proline residues in (ProProGlY)7 collagen strands with 4-fluoroproline and 4-methylproline diastereomers leads to the most stable known triple helices, having Tm values that are increased by >50 °C relative to (ProProGlY)7 triple helices. Differential scanning calorimetry data dictate an entropic basis to the hyperstability. X-Ray structural data at a resolution of 1.21 A reveal a prototypical triple helix with imperceptible deviations to its main III chain. These results show, for the first time, that conformationally constrained proline derivatives stabilize the triple helix via preorganization. The results presented in Chapter 4 will guide the application of 4-substituted proline derivatives to problems in protein engineering. Numerous researchers have explored the effects of the incorporation of 4-fluoroproline residues in the collagen triple helix. Unambiguous analyses of electron-withdrawing groups larger than fluorine on triple-helix structure and stability are less accessible. Chapter 5 reports the synthesis of 4-chloroproline diastereomers, explores their conformational properties, and reexamines all the results for 4-fluoroproline residues in collagen using the 4-chloroprolines in analogous experiments. Importantly, I demonstrate that a deleterious steric effect is responsible for the observation that stereoelectronic effects on triple-helix stability are not additive. These results are guiding continued efforts toward self-assembled collagen materials, which I report in Appendix B. Researchers have long been puzzled by the observation that, unlike other proline derivatives with apparently similar conformational preferences, (2S,4S)-4-hydroxyproline (hyp) strongly destabilizes the collagen triple helix. In Chapter 6, I show that hyp destabilizes the collagen triple helix by forming a transannular hydrogen bond that wrongly pre organizes hyp residues for triplehelix formation and disrupts the essential interstrand hydrogen bond in the triple helix. Eliminating the hydrogen bond by the O-methylation ofhyp residues significantly rescues triplehelix stability. Surprisingly, recent research has shown that CY-exo puckered proline derivatives can stabilize the collagen triple helix when placed in the Xaa and Yaa position of collagen-related peptides. These findings are in stark contrast to much existing data, which shows that CY -exo puckered proline derivatives usually destabilize the triple helix in the Xaa position. Chapter 7 reports my lV discovery that this anomalous stability is due to stabilizing interstrand dipole--dipole interactions that pay the energetic penalty for incorporation of an incorrectly puckered proline derivative in the Xaa position of the collagen triple helix. Chapter 8 recounts the incorporation of a reactive functional group in collagen triple helices. (2S)-4-Ketoproline (Kep) has unique conformational preferences among 4-substituted proline derivatives. I show that Kep slightly destabilizes the triple helix when substituted for proline in the Xaa position. I describe our efforts toward generating covalently linked triple helices by introducing nucleophilic amino acid residues near Kep residues in collagen-related peptides. In Chapter 9, I summarize my research and briefly discuss future directions for the collagen field. Two appendices recount my efforts toward understanding the impact of CY-substituents on the prolyl peptide bond isomerization equilibrium constant and progress toward self-assembled collagen biomaterials. v ACKNOWLEDGEMENTS I can express only gratitude and astonishment when I look back at how much I have learned and how much I have changed during graduate school. On the rare occasions in the last eight months that I have managed to catch a breath and think about the past five years, I find that my memories of the University of Wisconsin are sharply colored by the friends I made, the teachers I studied under, the colleagues I toiled with, and the science I learned. I come away from this experience with a better understanding of life and a deeper knowledge of myself. I have learned how to do science, and I hope I have learned about science. Looking back, I must say I have enjoyed (nearly) every minute of it. The most important aspect of a PhD in organic chemistry is the research. Research projects like mine cannot be initiated and accomplished by a novice graduate student without the input and advice of mentors and colleagues. First and foremost, I acknowledge the person who has contributed most to my education at the UW-my advisor, Professor Ronald Raines. Ron provides a great deal of independence to all his lab members-an independence I deeply valued. It allowed me to explore many of my own ideas, which was always the most rewarding of experiences even when the experiments proved failures (as happened more than I would like to admit). Ron has assembled a talented group of graduate and postdoctoral researchers and created an excellent working environment. I am grateful that he invited me into his lab to do the science described in this thesis. Of course, the list of colleagues with whom I have interacted and who have guided my scientific and professional growth is far too long to mention everyone by name. My fellow Raines lab members and organic chemistry graduate students are talented, unique people and I have learned something from each of them-I list only a few here. Frank Kotch was a VI postdoctoral researcher in our lab during my early years at the UW-he is the nicest guy I have ever met and mentored me throughout our years working together. My friend Jeet Kalia (GDGL, man) kept me sane and contributed to my work in untold hundreds of lunch, dinner, and terrace conversations-I cannot thank him enough. Amit Choudhary shared a bay with me for much of my PhD, and pushed me both intellectually and personally to be a better and more disciplined scientist. Eddie Myers also drove me forward and kept me motivated to pursue scientific research and at the same time stay true to myself. Daniel Gottlieb fattened me up at The Library over the course of a year or so, but it was good times and good drinks. Kimberli Kamer, a UW undergraduate student, worked alongside me for three semesters. She is a talented and smart student, and co-authored what will eventually turn into a couple papers with me. I was lucky to work with her and I look forward to watching her career develop. I joined the Raines lab with my colleague (and roommate for four years), Joseph Binder. I remember chatting with him during first year about how much fun it would be to compete with each other in the same lab--and indeed it was. Other Raines lab colleagues, including Jia-Chemg Horng, Matt Soellner, Annie Tam, Luke Lavis, Sayani Chattapodhyay, Mike Palte, Greg Ellis, Tom Rutkoski, Greg Jakubczak, Christine Bradford, and Benjamin Caes, contributed to my personal and professional development and to my chemical education in diverse ways. I was fortunate to work with two talented collaborators-Dr. Kenneth Satyshur and Professor Katrina Forest-on the crystal structure solution that is so important for the conclusions of chapter 4 of this thesis. Ken's persistence in freezing my (often ugly) crystals, collecting data, and working toward a solution was most admirable. The eventual success I had growing crystals that provided high-resolution data is due in large part to the knowledge garnered from his years of experience in crystallography, which he so willingly shared with me. Katy likewise kept me VB motivated and excited as we worked toward completion of this project, and she actually collected some of the data that led to a structure solution herself. My thanks are due to both of them and it was a pleasure to work with them. I also was privileged to collaborate with Professor Grant Krow of Temple University on a couple of projects, one of which appears as an appendix to this thesis. I have developed great respect for his scientific skill and personal qualities, as well as his drive to simply follow the science where it leads. Both Grant and Katy are members of my thesis committee, and I thank them and my other committee members (Professors Bob McMahon and Sam Gellman) for their many contributions to my research. I would be remiss did I not acknowledge the mentors that contributed most to the earliest stages of my education as an organic chemist. Professor Felicia Etzkorn (Virginia Tech) played a particularly important role in my development and career choices and has always supported me as a scientist. I thank her for those contributions. Professor Harry Dorn (Virginia Tech) provided my first opportunity to do research in an organic lab when I was a sophomore chemistry major. Other advisors who granted me the invaluable opportunity to work in their labs and gain exposure to new types of research included Professor David Kingston (Virginia Tech), Dr. Julio Camarero (Lawrence Livermore National Laboratory), and Dr. R. Shane Addleman (Pacific Northwest National Laboratory). All have influenced me both personally and scientifically. Aside from the synthetic portions of my research, much of my time has been spent in university core facilities. The folks that run those facilities deserve far more credit than they ever receive for their contributions to science at the UW. Most importantly, I must acknowledge the many contributions of Dr. Darrell "BIFmaster" McCaslin to my education and my research. If there is a single person other than my advisor that was essential for the completion of my thesis research, it is Darrell. If I had a dollar for every minute we spent discussing science, biophysical V III measurements, and my own always either challenging or bewildering experiments and results, I would be a very rich man. Darrell's broad theoretical and practical expertise in biophysical techniques is unmatched. It is strange to see most of my research tied up in a nice package in this thesis The BIFmaster is the one person with an appreciation for all the pain and suffering that went into obtaining those pretty graphs and measurements-because he combed over so many of the preliminary results with me. Dr. Gary Case at the peptide synthesis lab was an invaluable resource as I attempted the syntheses of the ~ 100 different peptides necessary for my research (only some of which appear in the succeeding pages). Dr. Ilia Guzei's small molecule crystallography contributions to my research are individually acknowledged in each chapter of this thesis. Drs. Charlie Frye and Mark Anderson, at the chemistry and biochemistry NMR facilities, were always available and willing to help with experiments at a moment's notice. Last, but far from least, I must thank my family and friends. My girlfriend is a theoretical chemistry graduate student from my year. Yu-Shan and I met at the UW when I sat down next to her in August 2004 at an orientation function and said "Well, I hear you are the smartest student in our year." The half-spoken challenge in those words eventually evolved into a treasured relationship that is now going on three years. I am in no position to evaluate whether she turned out to be the best in our class (although my opinion is that she did), but she certainly turned out both smarter and wiser than I did. It is very rewarding to share the graduate school experience with someone who appreciates and enjoys discussing science, understands the various travails and pleasures, and is accepting of the work weeks that often morph into 80-90 hours in length. Not least, the thermodynamic analysis reported in chapter 4 of this thesis owes much to her contributions, and she wrote the Fortran program we used to fit the calorimetry data in that chapter. For putting up with me, supporting me, and loving me I cannot thank her enough. IX Friends from outside the Raines lab, including Sujan Shekhawat, Keith and Christina Zomchek, and John Hottle, were with me every step of the way-Thanks, y'all. Finally, my family was always available to talk about graduate school and support my interest in science. My dad, Craig, provided much guidance in navigating the dark, still waters of academe and is a resource I know I will continue to draw on throughout my career. My mom, Nancy, was always there to provide a listening ear. My brothers and sisters, Joni, Luke, Caleb, Joshua, and Heidi are a big part of my life, and I thank them for their love and friendship. The unconditional love and support of my grandparents has always served to prop me up. As I follow in the footsteps of my maternal grandfather, chemistry professor Donald Bettinger, I hope I can make them all proud. My research was funded by graduate fellowships from the US Department of Homeland Security and the American Chemical Society, Division of Medicinal Chemistry, as well as support from the National Institutes of Health. To those entities, and to the taxpayers and society members that support them, I am deeply grateful. x TABLE OF CONTENTS Abstract. ............................................................................................................ .i Acknowledgements ............................................................................................ v Table of