xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004568
Photothermal studies of carboxymyoglobin
h [electronic resource] /
by Meagan Small.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Thesis (MS)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: Small ligand diffusion in heme proteins is not fully understood. To help better understand CO diffusion, three systems were investigated: L29H/F43H site-directed sperm whale myoglobin, horse heart myoglobin in a heavy water buffer, and calixresorcinarene. Binding of copper to calixresorcinarene was photophysically characterized to unravel transient binding of small molecules in heme-copper proteins. Copper binding was found to have a low dissociation constant of approximately 8.6 uM. Reaction profiles using photoacoustic calorimetry were constructed for the myoglobin systems. In deuterium oxide, ligand escape is not rate limited by water entry. Large enthalpy differences arise from the thermodynamic properties of deuterium oxide and the extensive hydrogen bonding network in myoglobin. In the mutant, CO rebinds primarily to the heme and is exothermic with a large volume contraction because of altered electrostatics within the binding pocket and higher water occupancy.
Advisor: Randy W. Larsen, Ph.D.
t USF Electronic Theses and Dissertations.
Photothermal Studies of Carboxymyoglobin by Meagan Small A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Randy W. Larsen, Ph.D. Brian Space, Ph.D. David Merkler, Ph.D. Martin Muschol, Ph.D. Date of Approval: July 15 2010 Keywords: myoglobin, thermodynamics, heme proteins, calixarenes, photoacoustic calorimetry Copyright 2010, Meagan Small
Acknowledgments I am grateful to Professor Randy W. Larsen for being a helpful and instructive advisor, under whose guidance I have learned a tremendous amount. Many thanks to Dr. Kirpal Bisht for generous donation of calixarene as well as Dr. Yi Lu (University of I llinois, Urbana Champaign) for the supply of engineered Cu B Mb
i TABLE OF CONTENTS LIST OF TABLES ii i LIST OF FIGURES i v LIST OF ACRONYMS v i ABSTRACT v i i CHAPTER ONE: TIME RESOLVED THERMODYNAMICS OF CO PHOTOLYSIS FROM CARBOXYMYOGLOBIN IN A D 2 O BUFFER 1 Introduction 1 Methods 10 Results/Discussion 12 Conclusion 26 References 27 CHAPTER TWO: TIME RESOLVED THERMODYNAMICS OF CO PHOTOLYSIS FROM AN ENGINEERED CU B MB 33 Introduction 33 Methods 38 Results/Discussion 40 Conclusion 56 References 57 CHAPTER THREE: PHOTOPHYSICAL CHARACTERIZATION OF A NOVEL RESORCINARENE BASED CAVITAND 60 Introduction 60 Methods 65 Steady State Absorption and Emission Measurements 66 Quantum Yield 67 Polarization 67 Job Plot 68 Fluorescence Lifetimes 68 Molecular Modeling and Electronic Structure Calculations 68 Results/Discussion 68 Conclusion 83 References 84
ii SUMMARY 91 APPENDIX 1 : PHOTOACOUSTIC CALORIMETRY 94 Theory 95 Instrumentation 99 References 101
iii LIST OF TABLES Table 1 .1 Thermodynamic profiles for D 2 OMb and hhMb. 1 6 Table 1. 2 Thermodynamic profiles for hhMb as reported in the literature. 16 Table 2.1 Thermodynamic profiles for swMb and CuBMb. 45 Table 2.2 Thermodynamic profiles for swMb as reported in the literature. 46 Table 2.3 The total solvent accessible volume of the four largest cavities within swMb and Cu B Mb. 53 Table 3.1 Quantum yield paramete rs for calixresorcinarene. 73 Table 3.2 Lifetimes of Cu(calixarene) complexes. 76
iv LIST OF FIGURES Figure 1. 1 A three dimensional representation of carbon monoxide bound horse heart myoglobin 2 Figure 1. 2 The heme group in myoglobin, iron protoporphyrin IX 3 Figure 1. 3 The events involving the ligand CO that occur upon photolysis of the Fe CO bond 4 Figure 1. 4 A schematic representation of the solvent responses and associated photothermal methods associated with the dissipation of heat. 8 Figure 1. 5 Th e overlaid UV/Vis spectra of CO bound horse heart myoglobin in H 2 O (black) and D 2 O (red) buffers 13 Figure 1. 6 The acoustic traces arising from CO photolysis of D 2 OMb (black) and the calibration compound Fe4SP (red). 13 Figure 1. 7 The fit (left) and aut ocorrelation (right) obtained from Simplex parameter estimation algorithm for one trial of D 2 OMb. 14 Figure 1. 8 A side by side comparison of the C p plots for D 2 OMb (left) and hhMb (right). 1 5 Figure 2.1 The four most commo n heme types, a, b, c, and o. 35 Figure 2.2 The overlaid UV/Vis spectra of CO bound swMb (black) and Cu B Mb (red) 41 Figure 2.3 The changes in absorbance as a function of time for Cu B Mb (left) and hhMb (right). 43 Figure 2.4 Transient absorption plots for Cu B Mb (ri ght) and hhMb (left) 4 3 Figure 2.5 The acoustic traces arising from CO photolysis of Cu B Mb (black) and the calibration compound Fe4SP (red). 44 Figure 2.6 A side by side comparison of the C p h plots for Cu B Mb (left) and swMb (right). 45
v F igure 2.7 An overlay of the heme and distal pocket of Cu B Mb (red) and swMb (blue) 51 Figure 2.8 A side by side comparison of the heme and distal pocket of swMb (left) and Cu B Mb (right). 52 Figure 3.1 (left) A line structure representation of the calixresorcinarene based cavitand 60 Figure 3.2 A 3D cartoon representation of a copper coordinated cavitand, in which an open coordination site is occupied by CO. 65 Figure 3.3 A side by side comparison of the cavitand monomer to 5OH H 2 BDC 71 Figure 3.4 (left) The steady state absorption (black) and emission (red) spectra of calixresorcinarene. 71 Figure 3. 5 The absorption peaks represented as lines (top) and peaks (bottom) based on electronic structure calculations, as performed i n HyperChem 8.0 using ZINDO S/CI with closed shell atoms (RHF). 72 Figure 3. 6 Steady state (left) absorption and (right) emission spectra for the titration of Cu(II)Cl 2 into ~ 30uM calixresorcinarene. 75 Figure 3. 7 0 Stern Volmer plot for the titration of Cu(II)Cl 2 into ~20 uM cavitand. 75 Figure 3. 8 The time resolved emission spectra for ~20 uM calixarene and Cu(II)Cl 2 75 Figure 3.9 Polarization as a function of [Cu(II)(OAc) 2 ] for ~ 120 uM cavitand. 78 Figure 3.10 Job plot for calixarene and Cu(II)Cl 2 80 Figure 3.11 The steady state I/I 0 Stern Volmer plot for the titration of Cu(II)Cl 2 into ~20 uM cavitand. 80 Figure 3.12 The fraction bound plot for calixresorcinarene, showing the relative change in the intensity as a function of [Cu(II)Cl 2 ].. 81 Figure 3.13 Geometry optimized Cu(calixresorcinarene). 83 Figure A1.1 (left) Jablonski Perrin diagram showing the various transitions that occur between energy levels and their associated processes 98 Figure A1. 2 The current design in place for photoacoustic calorimetry 99
vi LIST OF ACRONYMS ( alphabetical ) 5OH BDC 5 hydroxybenzenedicarboxylate Arg(N) arginine Cu B copper B center (heme copper oxidases) Cu B Mb L29H/F43H site directed mutated myoglobin D 2 O deuterium oxide EtOH ethanol Fe4SP iron tetrasulfanatophenyl porphyrin hhMb horse heart myoglobin HOMO highest occupied molecular orbital LUMO lowest unoccupied molecular orbital Mb myoglobin Nd:YAG neodymium ytterium aluminum garnet PAC photoacoustic calorimetry RHF restricted Hartree Fock swMb sperm whale myoglobin recombinant TA transient absorption UV/Vis ultraviolet/visible
vii PHOTOTHERMAL STUDIES OF CARBOXYMYOGLOBIN Meagan Small ABSTRACT S mall ligand diffusion in heme proteins is not fully understood. To help better understand CO diffusion three systems were investigated: L29H/F43H site directed sperm whale myoglobin horse heart myoglobin in a h e a v y w a t e r buff er, and calixresorcinarene. Binding of copper to calixresorcinarene was photophysically characterized to unravel transient binding of small molecules in heme copper proteins Copper binding was found to have a l o w d i s s o c i a t i o n c o n s t a n t o f a p p r o x i m a t e l y 8.6 uM Reaction profiles using photoacoustic calorimetry were constructed for the myoglobin systems I n d e u t e r i u m o x i d e l igand escape is not rat e limited by w a t e r e n t r y Large enthalpy differences arise from the thermodynamic p roperties o f d e u t e r i u m o x i d e a n d t h e extensive hydrogen bonding network i n myoglobin. I n t h e m u t a n t CO r e b i n d s p r i m a r i l y t o t h e h e m e a n d i s exothermic with a large volume contraction b e c a u s e o f a l t e r e d elec trostatics within the binding pocket a n d higher w a t e r o c c u p a n c y
1 CHAPTER 1 TIME RESOLVED THERMODYNAMICS OF CO PHOTOLYSIS FROM CARBOXYMYOGLOBIN IN A D 2 O BUFFER Introduction Myoglobin is a small, soluble heme protein that is r esponsible for oxygen storage and transport in muscle tissue. It is a single subunit containing 153 amino acid residues with a molecular weight of 17.8 kD. Myoglobin ( Figure 1. 1) was the first protein to have its 3 dimensional structure elucidated by x ray crystallography (Kendrew et al., 1958 and 1962 ). It is a member of the globin fami ly, which has a distinct globular alpha helical helices in myoglobin) containing a hydrophobic pocket for the heme (Aronson et. al, 1994) capability arises from the heme, which is an iron protoporphyrin IX ( Figur e 1. 2 ) linked to the protein backbone covalently via a proximal histidine residue (93) and electrostatically stabilized via approximately 90 interactions formed between the substituents of the heme and surrounding amino acid residues ( Antonini and Brunori, 1971 ). A variety of ligands, including O 2 CO, and CN can be coordinated to the iron in myoglobin, with oxygen being the most physiologically relevant (see Antonini and Brunori, 1971 for a thorough review of ligands). Iron exists in either the +2 (ferrous) or +3 (ferric) oxidation states in myoglobin. The +2 is the most biologically important state
2 as this is the only oxidation state to which O 2 coordinates. B inding of small molecules to the heme has been studied using a variety of methods, aimed at understanding the kinetics and equilibria of ligand binding A comprehensive review of these methods and important experimental data are also highlighted in Antonini and Brunori (1971). Here, the methods that are most relevant to the diffusion of ligands from the protein matrix will be discussed. Figure 1.1 A three dimensional representation of carbon monoxide bound horse heart myoglobin. Hydrogen atoms are omitted for clarity. Color codes: turquoise=carbon, red=oxygen, and blue=nitrogen. The heme is represented in pink, with the CO ligand shown as sphere s. The green ribbon shows the alpha helical
3 Traditionally flash photolysis of CO is the method employed to probe small molecule diffusion in heme proteins (Antonini and Brunori, 1971) CO is advantageous over the physiologically relevant O 2 for two reasons First CO has a strong, characteristic mid IR stretching frequency that can be monitored spectroscopically, and has been shown by Caughey and coworkers to be sensitive to the heme pocket environment ( e.g. Alben and Caughey, 1968; Caughey et al., 1969 ) S econdly, the Fe CO bond is photolytically cleaved with a quantum yield of unity for wavelengths within ultraviolet/visible scale (see Antonini and Brunori, 1971 and references therein), compared to O 2 with a qua ntum yield of 0.03 (Brunori et al., 1973; Gibson and Ainsworth, 1957 ). Flash photolysis of ligands allows the investigation of subsequent conformational relaxations and events that are associated with photolysis and diffusion from the protein matrix whic h are discussed below. Figure 1.2 The heme group in myoglobin, iron protoporphyrin IX.
4 Myoglobin has been studied extensively over the past 50 years, both to understand ligand diffusion and as a model heme protein. As such, s ignificant progress has been made in both respects. A wi de variety of t ime resolved spectroscopic ( Scott and Gibson, 1997; Dadusc et al., 2001; Franzen and Boxer, 1997; Mukai et al., 1998; Rector et al., 1998 ), computational ( Bossa et al., 2004; Elber and Karplus, 1990; Nutt and Meuwly, 2004 ), and x ray crystallographic ( Goldbeck et al., 2006; Hummer et al., 2004; Schmidt et al., 2005 ) methods have allowed for the detailed characterization of the protein. Still, the escape pathway of ligands from myoglobin, the simplest of heme protein s, is not fully und erstood Flash photolysis is a convenient technique to probe the dynamics of CO diffusion in myoglobin. In fact, much has been learned about the events that follow CO photodissociation, Figure 1. 3 It is known, for instance, that the Fe CO bond is cleav ed within ~350 fs (Martin et al., 1983). The vibrationally hot heme, furthermore, has been shown to heat the solvent within tens of picoseconds via thermal energy transfer between the van der Waals contacts and solvent (see Mizutani and Kitagawa 2001 and references therein). The escape of CO to the solvent occurs on a hundreds of nanoseconds time scale (Peters et al., 1991) and is presumably associated Figure 1.3 The events involving the ligand CO that occur upon photolysis of the Fe CO bond. In the first step, the incident photon (green to represent the 532 nm Nd YAG laser used here) cleaves the Fe CO bond, which is then transiently docked within the protein mat rix. The CO then escapes the protein in the second step.
5 with the entry of water into the distal pocket, as crystal structures of deoxymyoglobin show a wate r molecule hydrogen Schoenborn, 1981). Ligand diffusion from the protein matrix remains elusive in part because high resolution x ray crystal structures show that there is no defined channel for ligands to exit without conformational changes to expose pathways for the ligand to diffuse (Perutz and Mathews, 1966; Hummer et al., 2004). It has been proposed that these structural changes involve swinging of His64 in which horse heart myoglobin has been fou nd to exist in closed and open substates corresponding to the position of His64 ( see Tian et al., 1993 and references therein ; Mukai et al., 1998 ). Other studies suggest that the diffusion of small molecules involves hydro phobic pockets (x enon binding sit es) through which the ligand gains access to the solvent ( Elber and Karplus, 1990 ; Brunori and Gibson, 2001 ). The existence of xenon binding sites is based on crystal structures obtained by incubation of myoglobin with high pressure Xenon gas ( Shoenborn e t al., 1965; Tilton et al., 1984 ). CO escape from the protein matrix may involve one or both mechanisms The latter is plausible based on recent studies by Dantsker et al. (2005) and Bossa et al. (2004 and 2005) binding sites. Coupled with r ecent time resolved x ray crystallographic studies showing that CO does spend significant time in binding sites Xe1 and Xe4 prior to gaining access t o the solvent it suggests that these hydrophobic pockets may host the ligand as it makes its way out of the protein matrix ( Schotte et al., 2003 ; et al., 2004; Schmidt et al., 2005 ). The role of water in the distal pocket is still unclear, and its presence has several implications for ligand binding and diffusion. Entry of water and its significance in the
6 distal pocket were recently investigated by Goldbeck et al. (2006), in which time resolved absorption was used to further characterize water occ upancy in the distal pocket, rate limited by CO diffusion out of the protein matrix, occurring within several hundred nanoseconds after CO photodissociation (Goldbeck et al., 2006). CO association and equilibrium constants have been found to be dependent on the extent of water occupancy in the distal pocket with slower binding observed at higher water occupancies (Goldbeck et al., 2006 ). Distal pocket hydration also app ears to confer electrostatic stabilization that increases the affinity for O 2 binding versus CO binding, as suggested by Phillips Jr. ( and Olson, 1997; et al., 1999 ), in which equilibrium constants were found to be three orders of magnitude higher for O 2 i H bond. Such findings imply that water occupancy may play a pivotal role in discriminating O 2 compared to other non physiological diatomic molecules. Furthermore, distal pocket hydration may explain the necessity of xenon b inding cavities in myoglobin bas ed on O 2 rebinding studies by McNaughton et al. (2003) They proposed that the presence of these additional sites increases the likelihood of O 2 binding to the Fe 2+ because oxygen can equilibrate in these cavities and subsequently out compete water Chu et al. (2000 and references therein) suggest that hydrophobic pockets may have evolved for this purpose. Furthermore, the physiological role of water in the distal pocket has been investigated in the context of underst anding ligand escape. A recent molecular dynamics study (MD) by Bossa et al. (2005) indicated that presence of water play s a role in maintaining the His64 gate bond formed between H 2 It was found that upon wate r efflux and disruption of the hydrogen bond that His64 swung into an open
7 conformation (Bossa et al., 2005). Understanding the pivotal role that water may play in ligand escape is an energetic problem. It requires elucidating the effects of water on the reaction profile of CO diffusion more specifically each step in the mechanism Deuterium oxide or heavy water, is the simplest way to probe the effects of water on ligand diffusion via the magnitudes of changes observed in the thermodynamics. One common method for determining enthalpy and volume changes is with the Eyring equations (Eyring, 1935; Evans and Polan yi, 1935) in which plotting the rates as a function of the inverse of the temperature yields a line where the slope and the y ) as in Equation ( 1. 1 ). The activation volu me change is determined by taking the derivative of the rate with respect to pressure and plotting versus 1/T, as shown in Equation ( 1. 2 ). ( 1. 1 ) ( 1. 2 ) In this way, the total enthalpy and volume change for each step can be obtained by determining the rates in the forward and reverse directions. For the CO photolysis employed here, however, the Eyring equations are limited in their utility. The activatio n enthalpy and volume can only be obtained in the forward direction as CO flash photolysis is an irreversible process. Therefore, it is not possible to get at the total enthalpy and
8 volume changes for each step in the mechanism because the enthalpy and vo lume changes cannot be obtained in the reverse direction. Obtaining complete reaction profiles for processes involving CO flash photolysis, such as ligand diffusion, must rely on methods that can obtain the total enthalpy and volume changes for the interm ediates in the reaction. Thermal photonics or photothermal methods can provide this type of information as discussed briefly below (for more detail, see aforementioned reviews and Appendix 1 ). Thermal photonics or p hotothermal methods were first introdu ce d nearly 20 years ago and a number of good reviews exist discussing them in detail ( see reviews by Larsen Braslavsky and Heibel, 1992; Gensch and Viappiani, 2003; and references therein ). Their versatility resides in the physical premise of these methods. Thermal relaxation to the ground state from a photoexcited molecule induces heating of the solvent and a corresponding solvent re s ponse such as a pressure change or a change in the refractive index, Figure 1. 4. Pressure changes result in the generation of an Figure 1.4 A schematic representation of the solvent responses and associated photothermal methods associated with the dissipation of heat.
9 acoustic wave that is detected in a technique known as photoacoustic calorimetry, which will be the method utilized in this and subsequent studies. Changes in the refractive index are studied in methods such as transient grating, photothermal beam deflection, and thermal lensing, which will not be discussed here. These methods are advantageous for two reasons. T hey are fast; sandwiching the time scales that are most pertinent to biological processes (nanoseconds to milliseconds) Secondly, these methods do not rely on optical probes and therefore are not limited to systems that contain chromophores ( ). Instead, they rely on thermal relaxation and are therefore inherently thermodynamic. As such they have been employed in such diverse applications as protein folding, ligand binding, and electron transfer ( see aforementioned reviews and references therein ) Photoacoustic calorimetry has been previously applied to horse heart myoglobin to obtain complete thermodynamic profiles for CO photolysis and ligand diffusion (Westrick and Peters, 1990; Westrick and Norris, 1993; Angeloni and Feis, 2003) ; however this is the first known photothermal study of myoglobin in a heavy water buffer. Unraveling the role of the water molecule in ligand diffusion requires knowledge of the energetics affected by the presence of water in the distal pocket. The magnitudes of the enthalpy and volume changes can be determined using photoacoustic calorimetry for CO photolysis from myoglobin in a n H 2 O and D 2 O based buffer in which the differences in the energetics are a measure of the effects of water in ligand diffusion.
10 Methods Horse heart m yoglobin and deuterium oxide were obtained from Sigma Aldrich (St. Louis, MO) and used without further purification. Monobasic potassium phosphate (KH 2 PO 4 ) was obtained from Sigma Chemical Company (St. Louis, MO) and dibasic potassium phosphate (K 2 HPO 4 ) from Fisher Scientific (Fair Lawn, NJ) A 50 mM phosphate buf fer in D 2 O was prepared using the Henderson Hasselbach equation ( 3 ) pK a2 =6.865 for phosphate and was found to have a pH of 8 In addition, a 50 mM phosphate buffer in nanopure water was prepared in the standard way, pH~8 Protein samples were prepared b y the dilution of a stock solution in each buffer The CO bound form was L 50 mg/ml sodium dithionite (Sigma Aldrich Co., same buffer) to the de aerated sample, purged with Argon for ~5 minutes. Absorption spectra were obtai ned using a Shimadzu UV 2401 PC spectrophotometer. Fe (III) (4 sulfonatophenyl) porphine (Fe4SP, Frontier Scientific Inc., Logan, UT, same buffer) was used as the calorimetric calibration compound, with the OD of the reference being adjusted to within <0. 05 of the sample compound. ( 1. 3 ) Photoacoustic calorimetry (PAC) instrumentation has been reviewed in detail previously (Larsen 2007 ). D ata was obtained using a 1 cm x 1 cm quartz cuvette containing 2 m L sample in a Quantum Northwest variable temperature cell holder, controlled within 0.02 C. Contact between the Panametrics V103 piezoelectric
11 detector and the cuvette was facilitated by a thin layer of vacuum grease. CO photolysis is initiated by a 532 nm laser pulse from a frequency doubled Q switched Nd:YAG laser (Cont J per pulse). Voltages were amplified using a Panametrics ultrasonic pre amplifier (model 5662), digitized and recorded by a Picoscope 3205 (50 MHz) osci lloscope Acoustic data obtained was analyzed using the multiple temperature method, in which temperature dependent acoustic traces of both sample and re ference were obtained between 16 and 35 C. Sample and reference acoustic waves were generated by av eraging over 50 laser impulses. Plots of the acoustic waves for both D 2 OMb and hhMb revealed a frequency shift relative to Fe4SP indicating multiple kinetic events occurring within the resolvable time scale of the instrument, < 20 to 50 ns ) of these events were determined with an in house developed Simplex parameter estimation algorithm using a biphasic fit as is discussed in detail in Appendix 1. The amplitudes obtained for each phase fro m the Simplex algorithm were plotted as a function of C p 2 O. The data points were fit to a linear equation using OriginPro 8.0 that corresponds to Equation ( 4 ) with a y intercept that is a measure of the heat release d to the solvent (Q) and a slope that is conf ) of the process. ( 1. 4 )
12 where is the ratio of sample to reference acoustic waves, E h is the photon energy the quantum yield for the process Q is the amount of heat released to solvent C p is the heat capacity of the solvent (cal/g K) solvent density (g/mL) and coefficient of thermal expansion of solvent (K 1 ) The magnitudes of the enthalpies and volume changes were obtained by averaging the amplitudes obtained for the three data sets and plotting these amplitudes as a function of C p conf were obtained in the standard way with E h = 53.7325 kcal mol 1 199 1). Results and Discussion The steady state absorption spectra for horse heart myoglobin in a water based (hhMb) and deuterium oxide based (D 2 OMb) phosphate buffer are identical and shown in Figure 1.5 The UV/Vis spectra of deoxy myoglobins ( in D 2 O and H 2 O, not shown ) are characterized by a Soret peak at 426 nm, which is blue shifted in the carboxy form t o 423 nm. A smaller peak at 554 nm in deoxymyoglobin (of both types) is replaced by two peaks at 540 and 579 nm in the carbon monoxide bound forms Photoacoustic calorimetry was performed on hhMb and D 2 OMb as well as calibration compound Fe4SP (in respective buffers). The acoustic traces for one trial of D 2 OMb (and reference) are shown in Figure 1.6 A shift in frequency of the sample related to the frequency of the calibration compound is indicative of a multiple kinetic In this
13 case, the frequency shift was fit a bip hasic process. The first phase occurs faster than can be resolved by the instrument, <50 ns. On the basis of a Simplex parameter estimation Figure 1.6 The acoustic traces arising from CO photolysis of D 2 OMb (black) and the calibration compound Fe4SP (red). Figure 1.5 The overlaid UV/Vis spectra of C O b o u n d horse heart myoglobin in H 2 O (black) and D 2 O (red) buffers. Absorbances were normalized to 1 at the Soret peak (422 nm).
14 algorithm, the second slow phase was determined to be ~550 ns in hhMb and slightly slower at ~ 600 ns for D 2 OMb. These lifetimes represent the average of three trials. The fit and autocorrelation obtained from the parameter estimation algorithm for one trial of D 2 OMb is shown in Figure 1.7. An additional third fast phase has been observed at ~ 29 ns by Norris and Peters (1993) and ~ 80 ns by Angeloni and Feis (2003) that is not resolved here. This fast phase was shown to have a small amplitude (Norris and Peters, 1993) and is close to the time resolution of th e instrument (50 ns), which is presumably the reason the phase is not observed here. Figure 1.7 The fit (left) and autocorrelation (right) obtained from Simplex parameter estimation algorithm for one trial of D 2 OMb.
15 A side by side comparison of the C p plots is shown in Figure 1.8 The magnitudes of the enthalpy and volume changes for the initial prompt phase in hhMb were determined to be p p = 2 2 kcal mol 1 / 2.3 0.3 mL mol 1 while those for D 2 OMb were determined to be p p = 8 3 kcal mol 1 / 1.3 0.3 mL mol 1 where the subscript p refers to the prompt phase occurring <50 ns. The slow phase enthalpy and volume changes for hhMb and D 2 s s = 11 3 kcal mol 1 /6.4 0.6 mL mol 1 s s = 34 15 kcal mol 1 /10 1 mL mol 1 respectively. These values are summarized in Table 1. 1 T he observed magnitudes of the enthalpy and volume changes for hhMb are comparable to those observed in the aforementioned photothermal studies which are summarized in Table 1. 2 where the initial prompt phase and ~20 80 ns phases have been combined to compare directl y with the prompt phase observed in this photoacoustic study. Figure 1.8 A side by side comparison of the C p plots for D 2 OMb (left) and hhMb (right).
16 H p kcal/mol V p mL/mol s ns H s kcal/mol V s mL/mol H Total kcal/mol V Total mL/mol hhMb 2 2 2.3 0.3 550 11 3 6.4 0.6 13 4 4 1 0. 7 D 2 OMb 8 3 1.3 0.3 600 34 15 10 1 42 1 5 9 1 H p kcal/mol V p mL/mol s ns H s kcal/mol V s mL/mol H Total kcal/mol V Total mL/mol Westrick & Peters (1990) 7 2 1.7 0. 5 700 7 1 12.1 0.2 14 3 13.8 0.7 Norris & Peters (1993)** 6 4 1.9 0. 8 75 0 10 8 15 3 16 12 13 4 Angeloni & Feis (2003)** 12 1 3 .0 0. 4 8 00 6 7 19 1 18 8 16 1 hhMb 2 2 2.3 0.3 550 11 3 6.4 0.6 13 4 4 1 0. 7 The errors reported are relative, meaning they have been normalized to the quantum yield. The lifetimes reported for the slow phase are approximate and are shown for p and p values with our results, the fas t phases were combined in those cases with the f f = 3.8 0.7 kcal mol 1 / 2.7 0.2 mL mol 1 and Feis f f = 3.1 0.2 kcal mol 1 / 3.0 0.2 mL mol 1 See References for citations. p s are approximate. Table 1.2 Thermodynamic profiles for hhMb as reported in the literature. Table 1.1 Thermodynamic profiles for D 2 OMb and hhMb.
17 Contributions to the prompt phase enthalpy and volume changes are expected from the Fe CO bond cleavage, low spin to high spin transition of the heme iron, as well conformational changes occurring as the CO is transiently docked and the protein relaxes. for Fe CO bond cleavage and the spin state change ha ve been estimated by to be 17 kcal mol 1 and 1 5 mL mol 1 Subtracting the expected contributions to from the observed values would suggest that the conformation 15 kcal mol 1 and 17 mL mol 1 9 kcal mol 1 16 mL mol 1 for D 2 OMb Angeloni and Feis (2003) attribute th e negative volume change to distal pocket cavity contractions as a result of the Fe CO bond breakage, for which they utilize time resolved x It is proposed that the dissociated and transiently dock ed CO no longer electrostatically and/or sterically hinders the surrounding residues and therefore contraction is electrostatically and/ or sterically possible as suggested in Phillips et al. (1999) and Kachalova et al. (1999) respectively. The enthalpy c hange for Fe CO bond cleavage is endothermic by 17 kcal mol 1 only slightly endothermic in the prompt phase observed here for both types of Mb. The volume contraction has also been attributed to the disruption of the salt bridge (Peters et al., 1991) in which a volume contraction occurs upon breaking of the Lys45 heme 6 propionate salt bridge due to rearrangement of the solvent molecules around the exposed charges or electrostriction (van Eldik, 1989; Heremans, 1982). Photoacoustic studies don e by Peters et al. (1991) on myoglobin site directed mutants lacking the Lys45 salt 8 kcal mol 1 and 5 mL mol 1 respectively; a volume change supported by densitometric studies of amino acids
18 show ing electrostriction volume contractions ranging from 1 to 20 mL mol 1 (Kharakoz, 1989). (Belogortseva et al., 2007) of the heme propionate groups of myoglobin has shown that enthalpic and volume contributions from the disruption of the salt bridge are negligible which is supported by ligand rebinding studies suggesting that the absence of a charge at position 45 does not affect the energetic barrier for CO escape (Balasubramanian, 1993 ) Presumably, the structural changes involved in contraction of the distal pocket are exothermic in nature and thus compensate in part for the large endothermic requirements of cleaving the Fe CO bond. Following Peters et al. (1991), disruption of the Ly s45 values, enthalpy and volume changes resulting from conformational changes are expected to be 7 kcal mol 1 and 12 mL mol 1 (hhMb) and 1 kcal mol 1 and 11 mL mol 1 (D 2 OMb) 15 kcal mol 1 17 mL mol 1 9 kcal mol 1 16 mL mol 1 for D 2 OMb. In either case, the conformational changes associated with the prompt phas e are exothermic in nature and result in volume contraction. Furthermore, i t is expected that no isotopic effects be observed in this phase as the rate of water entry is not fast enough to influence the prompt phase. T he magnitudes of t he enthalpy changes for both D 2 OMb and hhMb are similar within error, while the differences in the volume change may be attributed to the properties of the two solvent s D 2 O and H 2 O D euterium oxide is 1.243 times more viscous than water ( Nmethy and Scher aga, 1964) Therefore, volume changes within the distal pocket would be expected to have a slightly smaller volume change in D 2 O than H 2 O as suggested by time resolved vibrational studies in
19 glycerol showing that conformational fluctuations are more hindered in viscous solvents (Kim and Lim, 2006) The slow phase can be attributed to ligand escape since CO diffusion is known to occur within hundreds of nanoseconds (Peters et al., 1991) at which point it is known that a water molecule enters based on crystal structures showing water bound to His64 (Phillips, 1980; Phillips and Schoenborn, 1981). The lifetime s for the slow phase observed here are similar for D 2 OMb and hhMb at ~ 600 ns and ~ 550 ns respectively suggesting that this phase is not rate li mited by the entry of water into the distal pocket or that the mechanism guiding ligand escape is similar in both D 2 O and H 2 O However, there are clearly differences in the magnitudes of the enthalpy and vo lume changes with s = 11 3 kcal mol 1 s = 6.4 0.6 mL mol 1 s = 34 15 kcal mol 1 s = 10 1 mL mol 1 f or D 2 OMb again see Table 1.1 In addition to the escape of CO to the solvent enthalpy and volume changes arise from any conformational changes associated with ligand escape, including reformation of the salt bridge and CO solvation. The volume changes occurring in the slow phase for hhMb have recently been explained in detail by Angeloni and Feis (2003). They observe a slow phase volume change of ~ 19 mL mol 1 that they propose is due to expansion from CO solvation as it enters the solvent and also contraction from electrostriction of charges arising from breaking the Lys45 heme propionate salt bridge This volume expansion is slightly larger than that observ ed here. However, the breaking of the salt bridge is proposed by Peters et al. (1991) to be broken during an earlier phase, rather than in the slow phase as in Angeloni and Feis (2003). Similarly, smaller volume changes (12 15 mL mol 1 ) have been observe d by Westrick and Peters (1990) as well as Norris and Peters (1993). In fact,
20 the volume changes observed here of ~ 6 mL mol 1 are in good agreement with transient s was observed to be 8 1 mL mol 1 In the slow phase, CO escape follows a sequence of events that open up a channel for the ligand to escape, most importantly His64 from a closed to open conformation to create a pathway to the solvent Though this may not be the only pathway for ligand escape, namely xenon binding cavities or conformational changes accommodating alternative channel opening (Huang and Boxer, 1994; Elber and Karplus, 1990), it certainly appears to be a dominant one (Scott et al., 2001) Therefore the overall volum e and enthalpy change must account for contributions from volume and enthalpy changes of CO escape, solvation, and conformational changes arising from the His64 Values for the partial molar volume of CO in water range from 29 37 mL mol 1 which is the vo lume expansion expected when CO leaves the protein matrix to the solvent ( Moore et al., 1982; Eley, 1939) Additionally, t he expected volume contribution due to water entry into the distal pocket depends on the percentage of water occupancy. Recent studies by Goldbeck et al. (2006) and Kachalova et al. (1999) have suggested that the occupancy of water in the distal pocket may not be as high as the full 84% occupancy originally proposed by Phillips and Schoenborn (1981) and Quillin et al.(1993). Kachalova et al. (1999) report occupancy of 60% due to multiple conformations of of His64. The volume contraction associated with water entry, therefore, is estimated to be between 10 to 14 mL mol 1 for occupancies of 0.6 to 0.8, respectively. CO and entry of water is expected to be ~ 2 1 mL mol 1 (this represents the average of the aforementioned values the range would be 15 to 27 mL mol 1 ). However, the
21 magnitudes of the volume changes for both hhMb and D 2 OMb are considerably smaller (by ~15 and 11 mL mol 1 respectively). Understanding the effects of D 2 O versus H 2 O necessitates a thorough explanation of these volume changes. The relative difference between the volume changes for D 2 OMb com pared to hhMb is small at ~4 mL mol 1 higher. Such a small difference suggests that the volume changes for the processes occurring in this slow phase are not significantly different between the solvents H 2 O and D 2 O. Additional structural changes within the protein, specifically the distal pocket and His64 pres umably accou nt for the remaining ~ 15 (hhMb) and 11 mL mol 1 (D 2 OMb) And t hough the exact origin of this volume contraction is unclear, Volume changes associated with the movement of His64 have be en estimated by Sakakura et al. (2002) based on transient grating studies of distal histidine mutants It was determined that the volume changes and kinetics of the second phase were highly dependent on the polarity of residue 64 proximately 10 to 12 mL mol 1 smaller in mutants containing polar residues (either His or Gln) versus nonpolar residues (Leu or Val). The volume changes are an approximation of the volume changes that are associated with His64 movement and suggest that an electrostatic interaction may guide the mo vement of residue 64, supporting results by Bossa et al. (2005) who found that the presence of CO in the distal pocket is strongly correlated with the opening of the His64 gate. Additionally, it has been proposed that the reformation of the Lys45 he me propionate salt bridge occurs on this time scale contributing to the enthalpy and volume changes, however it is not included here based on time resolved resonance raman studies
22 by Nakashima et al. (1998) suggesting that the salt bridge remains disrupted until CO recombination. Despite similarities in the volume changes, s ignificant slow phase enthalpy differences between the myoglobins are observed however, with D 2 OMb ~23 kcal mol 1 higher than hhMb. Since the similarity of the volume changes suggests that the same mechanism is guiding ligand diffusion and escape, the large difference in the enthalpy must arise from the solvent interaction with the protein and ligand. The events that accompany photodissociation and ligand escape in myoglobin have been explained. 3 kcal mol 1 ( Leung et al., 1987 ). The enthalpy of transferring ligands from H 2 O to D 2 O has also been calculated by Antonini and Brunori (1971) to be less than 1 kcal mol 1 which is negligible considering the large enthalpy change differences between the two myoglobins. Water entry has also been estimated by the temperature dependence of w ater entry/exit constants at 8 kcal mol 1 (Goldbeck et al., 2006). Thus together, the expected events would contribute 5 kcal mol 1 for the slow phase, while the observed enthalpy changes are 16 kcal mol 1 and 39 kcal mol 1 larger (more endothermic ) for hhMb and D 2 OMb respectively. Thus the energetics of ligand escape and the associated conformational changes are mor e endothermic i n D 2 O versus H 2 O presumably because of the nature of the solvent CO solvation in D 2 O versus H 2 O is not expected to c transfer of ligands from H 2 O to D 2 O is negligible (on the order of 1 kcal mol 1 or less) (Arnett and McKelvey, 1969).
23 Recent spectroscopic studies by Frau e nfelder et al. (2009) demonstrated that global protein motions (or alpha relaxations) are controlled by the viscosity of the solvent, while local internal motions (beta relaxations) are dominated by the hydration shell of the protein. Similarly, the conformation al changes involved in channel opening directly correlated to fluctuations in the solvent and thus highly dependent on solvent viscosity (Fenimore et al., 2002). As aforement ioned, the viscosity of D 2 O is approximately 1.243 times greater than that of H 2 O (Nmethy and Scheraga, 1964). Time resolved vibrational spectroscopy studies by Kim and Lim (2006) suggest that the conformational relaxations that occur within the protein are hindered in more viscous solvents. Phosphorescence studies by Cioni and Strambini (2002) determined that proteins are in fact more rigid in D 2 O versus H 2 O. Taken together, it is proposed here that the structural changes required for opening of a pathway in myoglobin, namely His64 gating, require more energy to overcome dynamic barriers imparted by the higher solvent viscosity in D 2 O. Though the magnitude of the enthalpy difference for this movement is unclear at this time, it is believed to be on the order of 2 3 kcal mol 1 based on Frau e relaxations (viz. those controlling local in ternal motions) decreased by ~ 2 kcal mol 1 as the hydration number was increased from 1 to 2, where hydration numbers have recently been shown to be slightly lower in H 2 O than D 2 O by Kielek and Marczak (2010) to be contributing to the enthalpy differences in this study, it is not unlikely that the activation enthalpy would increase with hydration value because global conformational changes are highly solvent dependent, thereby
24 increasing the energetic barrier o f global motions and making the overall enthalpy more endothermic. Conformational rearrangements within the slow phase associated with the opening of a channel for CO to escape involve reorganization of residues within the distal pocket, where the importance of the hydrogen bonding network has been investigated and proven based on altering of the heme propionates and also site directed mutagenesis ( see Belogortseva et al., 2007 ; Barrick, 2000; and references therein ) In studying geminate recombination in myoglobin mutants of residues 64 and 68, Sugimoto et al. (1998) fou nd that altering the hydrogen bonding network within the distal pocket increased the percentage of CO molecules that geminately rebound to the heme group via restriction of side chain resi due dynamics accompanying CO escape. In fact, it was determined tha t the identity of the hydroxyl group was the most distinguishing factor in the geminate recombination. Furthermore, the hydrogen bonding network appears to affect water within the distal pocket, as presented by Quillin et al. (1993) using crystal structur es of distal histidine mutants (His64), where the presence of water within the distal pocket was found to be significantly reduced when apolar residues were substituted. The implications of the hydrogen bonding network in the context of the energetics are discussed below. One of the most distinguishing and remarkable properties of water is the extensive hydrogen bonding network that it exhibits (Marchal, 2007). Moreover, D 2 O and H 2 O show slight differences in regards to the energetics of their hydrogen bonding networks. It is well known that the bond strength of an HO~ D bo nd is greater than that of an HO~ H (~ represents the hydrogen bond) bond by approximately 0.1 kcal mol 1 becaus e the
25 former has a lower zero point vibrational energy ( ). This value (0.1 to 0.2 kcal mol 1 ) is supported by spectroscopic s tudies by Dorster et al. (see Demmel et al., 1997) and calorimetry studies by Benjamin and Benson (1963). The former upheld the hydrogen bonding strength differences in myoglobin using infrared spectra of hydrated myoglobin to reveal that the hydrogen bond strength of H~ O D was significantly higher than H~O 1 corresponding to a 0.1 kcal mo l 1 difference. The importance of the hydrogen bonding network has been investigated in a number of other proteins as well, and was in general found that alpha helical proteins (such as myoglobin) are destabilized in D 2 O, while those dominated by beta sheets are stabilized in D 2 O (Cho et al., 2009 and references therein). Myoglobin has roughly 180 hydrogen bonds between the residue side chain groups and water (Doster and Settles, 1998 ) Given the large number of hydrogen bonds between the residue side chains and solvent and the 0.1 kcal mol 1 difference between the two types of hydrogen bonds, it is likely that the large enthalpy difference observed in the slow phase of D 2 OMb arises from the cumulative effect of using bulk H 2 O versus D 2 O as the solvent.
26 Conclusion This is the first known photothermal study to probe the energetics of water entry into the distal pocket via the use of heavy water. The presence of water hydrogen bonded to His64 has been implied in abetting the heme in discr iminating oxygen from other diatomic ligands, such as carbon monoxide, which shows a higher affinity for iron, as well as explaining the necessity for secondary hydrophobic cavities. The results presented suggest that the mechanism guiding ligand escape d oes not show an isotope effect. Ligand escape is not rate limited by the entry of water into the distal pocket as both show similar lifetimes for this phase. In addition, the magnitudes of the volume changes for both the prompt and slow phases do not sho w significant differences arising where CO diffusion from the protein matrix is significantly more endothermic in D 2 O versus H 2 O buffers. It is proposed here that the extensive hydrogen bonding network of myoglobin coupled with the unique thermodynamic properties of D 2 O and H 2 O is responsible for this observation
27 References 1. Alben, J. O.; Caughey W. S., An infrared study of bound carbon monoxide in the human red blood cell, isolated hemoglobin, and heme carbonyls. Biochemistry 1968, 7, (1), 175 183. 2. Angeloni, L.; Feis, A., Protein relaxation in the photodissociation of myoglobin CO complexes. Photochem. Photobiol. Sci. 2003, 2, 730 740. 3. Antonini, E.; Brunori, M., Hemoglobin and myoglobin in their reactions with ligands North Holland Publishing Company: Amsterdam, 1971. 4. Arnett, E. M.; McKelvey, D. R., Solvent isotope effects on nonreactin g solutes. In Solute solvent interactions Coetzee, J. F.; Ritchie, C. D., Eds. Marcel Dekker: New York, NY, 1969. 5. Aronson, H. E. G.; William E. Royer, J.; Hendrickson, W. A., Quantification of tertiary structural conservation despite primary sequence d rift in the globin fold. Protein Science 1994, 3, 1706 1711. 6. Austin, R. H.; Beeson, K. W.; Eisenstein, L.; Frauenfelder, H.; Gunsalus, I. C., Dynamics of ligand binding to myoglobin. Biochemistry 1975, 14, ( ), 5355 5373. 7. Balasubramanian, S.; Lambrig ht, D. G.; Marden, M. C.; Boxer, S. G., CO recombination to human myoglobin mutants in glycerol water solutions. Biochemistry 1993, 32, 2202 2212. 8. Barrick, D., Trans substitution of the proximal hydrogen bond in myoglobin: II. Energetics, functional con sequences and implications for hemoglobin allostery. Proteins: Structure, Function, and Genetics 2000, 39, 291 308. 9. heme propionate groups to the conformational dynamics associ ated with CO photodissociation from horse heart myoglobin. J. Inorg. Biochem. 2007, 101, 977 986. 10. Benjamin, L.; Benson, G. C., A deuterium isotope effect on the excess enthalpy of methanol water solutions J. Phys. Chem. B 1963, 67, 858 861. 11. Bossa, C.; Amadei, A.; Daidone, I.; Anselmi, M.; Vallone, B.; Brunori, M.; Nola, A. D., Molecular dynamics simulation of sperm whale myoglobin: effects of mutations and trapped CO on the structure and dynamics of cavities. Biophys. J. 2005, 89, 465 474. 12. Bossa C.; Anselmi, M.; Roccatano, D.; Amadei, A.; Vallone, B.; Brunori, M.; Nola, A. D., Extended Molecular Dynamics Simulation of the Carbon Monoxide Migration in Sperm Whale Myoglobin. Biophys. J. 2004, 86, 3855 3862. 13. Braslavsky, S. E.; Heibel, G. E., Ti me resolved photothermal and photoacoustics methods applied to photoinduced processes in solution. Chem. Rev. 1992, 92, 1381 1410. 14. Brunori, M.; Giacometti, G. M.; Antonini, E.; Wyman, J., Heme Proteins: Quantum Yield Determined by the Pulse Method. PNAS 1973, 70, (11), 3141 3144.
28 15. Brunori, M.; Gibson, Q. H., Cavities and packing defects in the structural dynamics of myoglobin. EMBO Reports 2001, 2, (8), 674 679. 16. Carver, T. E.; Rohlfs, R. J.; Olson, J. S.; Gibson, Q. H.; Blackmore, R. S.; Sprin ger, B. A.; Sligar, S. G., Analysis of the kinetic barriers for ligand binding to sperm whale myoglobin using site directed mutagenesis and laser photolysis techniques. J. Biol. Chem. 1990, 265, (32), 20007 20020. 17. Caughey, W. S.; Alben, J. O.; McCoy, S .; Boyer, S. H.; Charache, S.; Hathaway, P., Differences in the infrared stretching frequency of carbon monoxide bound to abnormal hemoglobins. Biochemistry 1969, 8, (1), 59 62. 18. Cho, Y.; Sagle, L. B.; Iimura, S.; Zhang, Y.; Kherb, J.; Chilkoti, A.; Sch oltz, J. turn structure is stabilized in D2O. J. Am. Chem. Soc. 2009, 131, 15188 15193. 19. Chu, K.; Vojtchovsk, J.; McMahon, B. H.; Sweet, R. M.; Berendzen, J.; Schlichting, I., Structure of a ligand binding inter mediate in wild type carbonmonoxy myoglobin. Nature 2000, 403, 921 923. 20. Cioni, P.; Strambini, G. B., Effect of heavy water on protein flexibility. Biophysical Journal 2002, 82, 3246 3253. 21. Dadusc, G.; Ogilvie, J. P.; Schulenberg, P.; Marvet, U.; Mil ler, R. J. D., Diffractive optics based heterodyne detected four wave mixing signals of protein motion: From "protein quakes" to ligand escape for myoglobin. PNAS 2001, 98, (11), 6110 6115. 22. Dantsker, D.; Roche, C.; Samuni, U.; Blouin, G.; Olson, J. S.; Friedman, J. M., The position 68(E11) side chain in myoglobin regulates ligand capture, bond formation with heme iron, and internal movement into the xenon cavities. J. Biol. Chem. 2005, 280, (46), 38740 38755. 23. Demmel, F.; Doster, W.; Petry, W.; Schul te, A., Vibrational frequency shifts as a probe of hydrogen bonds: thermal expansion and glass transition of myoglobin in mixed solvents. Eur. Biophys. J. 1997, 26, 327 335. 24. Doster, W.; Settles, M., The dynamical transition in protein: the role of hydr ogen bonds. Les Houches Lectures: Workshop on Hydration Processes in Biology: Theoretical and Experimental Approaches 1998 25. Elber, R.; Karplus, M., Enhanced Sampling in Molecular Dynamics: Use of the Time Dependent Hartree Approximation for a Simulatio n of Carbon Monoxide Diffusion through Myoglobin. J. Am. Chem. Soc. 1990, 112, 9161 9175. 26. Eley, D. D., On the solubility of gases. Part II. A comparison of organic solvent with water. Trans. Faraday Soc. 1939, 35, 1421 1432. 27. Evans, M. G.; Polanyi M., Some applications of the transition state method to the calculation of reaction velocities, especially in solution. Trans. Faraday Soc. 1935, 31, 875 894. 28. Eyring, H., The activated complex in chemical reactions. J. Chem. Phys. 1935, 3, 107 115. 29. Feis, A.; Angeloni, L., Photodissociation of the CO complex of horseradish peroxidase studied by laser induced optoacoustic spectroscopy. J. Phys. Chem. B 2001, 105, 2638 2643.
29 30. Fenimore, P. W.; Frauenfelder, H.; McMahon, B. H.; Parak, F. G., Slavin g: solvent fluctuations dominate protein dynamics and functions. PNAS 2002, 99, (25), 16047 16051. 31. Franzen, S.; Boxer, S. G., On the Origin of Heme Absorption Band Shifts and Associated Protein Structural Relaxation in Myoglobin following Flash Photoly sis. J. Biol. Chem. 1997, 272, (15, Issue of April 11), 9655 9660. 32. Frauenfelder, H.; Chen, G.; Berendzen, J.; Fenimore, P. W.; Jansson, H.; McMahon, B. H.; Stroe, I. R.; Swenson, J.; Young, R. D., A unified model of protein dynamics. PNAS 2009, 106, (1 3), 5129 5134. 33. Gensch, T.; Viappiani, C., Time resolved photothermal methods: accessing time resolved thermodynamics of photoinduced processes in chemistry and biology. Photochem. Photobiol. Sci. 2003, 2, 699 721. 34. Gibson, Q. H., An apparatus for fl ash photolysis and its applications to the reactions of myoglobin with gases. J. Physiol. 1956, 1956, (134), 112 122. 35. Gibson, Q. H.; Ainsworth, S., Photosensitivity of Hm Compounds. Nature 1957, 180, 1416 1417. 36. Goldbeck, R. A.; Bhaskaran, S.; Orte ga, C.; Mendoza, J. L.; Olson, J. S.; Soman, J.; Kliger, D. S.; Esquerra, R. M., Water and ligand entry in myoglobin: Assessing the speed and extent of heme pocket hydration after CO photodissociation. PNAS 2006, 103, (5), 1254 1259. 37. Heremans, K., High pressure effects on proteins and other biomolecules. Annu. Rev. Biophys. Bioeng. 1982, 11, 1 21. 38. Huang, X.; Boxer, S. G., Discovery of new ligand binding pathways in myoglobin by random mutagenesis. Nat. Struct. Biol. 1994, 1, 226 229. 39. Hummer, G.; Schotte, F.; Anfinrud, P. A., Unveiling functional protein motions with picosecond x ray crystallography and molecular dynamics simulations. PNAS 2004, 101, (43), 15330 15334. 40. Kachalova, G. S.; Popov, A. N.; Bartunik, H. D., A steric mechanism for inhibition of CO binding to heme proteins. Science 1999, 284, 473 476. 41. Kendrew, J. C., The structure of globular proteins. Comparative biochemistry and physiology 1962, 4, 249 252. 42. Kendrew, J. C.; Bodo, G.; Dintzis, H. M.; Parrish, R G.; Wyckoff, H.; Phillips, D. C., A three dimensional model of the myoglobin molecule obtained by x ray analysis. Nature 1958, 181, (4610), 662 666. 43. Kharakoz, D. P., Volumetric properties of proteins and their analogs in diluted water solutions: 1. P C. Biophys. Chem. 1989, 34, (2), 115 125. 44. Kielek, K.; Marczak, W., Hydration on non electrolytes in H2O and D2O investigated by Passynski's method. Int. J. Thermophys. 2010, 31, 77 84. 45. Kim, S.; Lim, M., Eff ects of solvent viscosity on conformational dynamics of heme pocket in myoglobin and hemoglobin. Bull. Korean Chem. Soc. 2006, 27, (11), 1825 1831. 46. Larsen, R. W., Lifetime distributions in photoacoustic calorimetry. In 54th Biophysical Society Annual M eeting, Abstract 2010. 47. to heme proteins. Coord. Chem. Rev. 2007, 251, 1101 1127.
30 48. Leung, W. P.; Cho, K. C.; Chau, S. K.; Choy, C. L., Measurement of the protein ligand bond energy of carboxymyoglobin by pulsed photoacoustic calorimetry. Chemical Physics Letters 1987, 141, (3), 220 224. 49. Marchal, Y., The hydrogen bond and the water molecule. The physics and chemistry of water, aqueous and bio media. Elsevier: Amsterdam, Netherlands, 2007. 50. Martin, J. L.; Migus, A.; Poyart, C.; LeCarpentier, Y.; Astier, R.; Antonetti, A., Femtosecond photolysis of CO ligated protoheme and hemoproteins: appearance of deoxy species with 350 fsec time constant. PNAS 1983, 80, 173 177. 51. McNaughton, L.; Hernndez, G.; LeMaster, D. M., Equilibrium O2 distribution in the Zn2+ protoporphyrin IX deoxymyoglobin mimic: application to oxygen migration pathway analysis. J. Am. Chem. Soc. 2003, 125, 3813 3820. 52. n, R. W., Thermodynamic profiles for CO photodissociation from heme model compounds: effect of proximal ligands. Inorganic Chemistry 2005, 44, 1006 1014. 53. Mizutani, Y.; Kitagawa, T., Ultrafast dynamics of myoglobin probed by time resolved resonance rama n spectroscopy. In The Chemical Record The Japan Chemical Journal Forum and John Wiley & Sons, Inc.: 2001; Vol. 1, pp 258 275. 54. Moore, J. C.; Battino, R.; Rettich, T. R.; Handa, Y. P.; Wilhelm, E., Partial molar volumes of "gases" at infinite dilution in water at 298.15 K. J. Chem. Eng. Data 1982, 27, 22 24. 55. Mukai, M.; Nakashima, S.; Oleson, J. S.; Kitagawa, T., Time Resolved UV Raman Detection of a Transient Open Form of the Ligand Pathway in Tyr64(E7) Myoglobin. J. Phys. Chem. B 1998, 102, 3624 3630. 56. Nakashima, S.; Kitagawa, T.; Olson, J. S., Time resolved resonance Raman study of intermediates generated after photodissociation of wild type and mutant co myoglobins. Chem. Phys. 1998, 228, 323 336. 57. Nmethy, G.; Scheraga, H. A., Structure of water and hydrophobic bonding in proteins. IV. The thermodynamic properties of liquid deuterium oxide. J. Chemical Physics 1964, 41, (3), 680 689. 58. Norris, C. L.; Peters, K. S., A photoacoustic calorimetry study of horse carboxymyoglobin o n the 10 nanosecond time scale. Biophys. J. 1993, 65, 1660 1665. 59. Nutt, D. R.; Meuwly, M., CO migration in native and mutant myoglobin: Atomistic simulations for the understanding of protein function. PNAS 2004, 101, (16), 5998 6002. 60. Perutz, M. F.; Mathews, F. S., An X ray study of azide methaemoglobin. J. Mol. Biol. 1966, 21, (1), 199 202. 61. Peters, K. S.; Watson, T.; Marr, K., Time Resolved Photoacoustic Calorimetry: A Study of Myoglobin and Rhodopsin. Annu. Rev. Biophys. Biophys. Chem. 1991, 20, 343 362. 62. Phillips Jr., G. N.; Olson, J. S., Myoglobin discriminates between O2, NO, and CO by electrostatic interactions with the bound ligand. J. Biol. Inorg. Chem. 1997, 2, 544 552.
31 63. Phillips Jr., G. N.; Teodoro, M. L.; Li, T.; Smith, B.; Olson, J. S., Bound CO is a molecular probe of electrostatic potential in the distal pocket of myoglobin. J. Phys. Chem. B 1999, 103, 8817 8829. 64. Phillips, S. E. V., Structure and refinement of oxymyoglobin at 1.6 resolution J. Mol. Biol. 1980, 142, (5), 531 554. 65. Phillips, S. E. V.; Schoenborn, B. P., Neutron diffraction reveals oxygen histidine hydrogen bond in oxymyoglobin Nature 1981, 292, 81 82. 66. Quillin, M. L.; Arduini R. M.; Olson, J. S.; George N. Phillips, J., High resolution crystal structures of distal histidine mutants of sperm whale myoglobin. J. Mol. Biol. 1993, 234, 140 155. 67. Rector, K. D.; Engholm, J. R.; Hill, J. R.; Myers, D. J.; Hu, R.; Boxer, S. G.; Dl ott, D. D.; Fayer, M. D., Dynamics of Myoglobin CO with the Proximal Histidine Removed: Vibrational Echo Experiments. J. Phys. Chem. B 1998, 102, 331 333. 68. Sakakura, M.; Morishima, I.; Terazima, M., Structural dynamics of distal histidine replaced mutan ts of myoglobin accompanied with the photodissociation reaction of the ligand. Biochemistry 2002, 41, 4837 4846. 69. Sakakura, M.; Morishima, M.; Terazima, M. J., The structural dynamics and ligand releasing process after photodissociation of sperm whale c arboxymyoglobin. J. Phys. Chem. B 2001, 105, 10424 10434. 70. Schmidt, M.; Nienhaus, K.; Pahl, R.; Krasselt, A.; Anderson, S.; Parak, F.; myoglobin: A time resolved crystallograp hic study on L29W MbCO. PNAS 2005, 102, (33), 11704 11709. 71. Schotte, F.; Soman, J.; Olson, J. S.; Wulff, M.; Anfinrud, P. A., Picosecond time resolved X ray crystallography: probing protein function in real time. J. Struct. Biol. 2005, 147, 235 246. 72. Scott, E. E.; Gibson, Q. H., Ligand Migration in Sperm Whale Myoglobin. Biochemistry 1997, 36, 11909 11917. 73. Scott, E. E.; Gibson, Q. H.; Olson, J. S., Mapping pathways for O2 entry into and exit from myoglobin. J. Biol. Chem. 2001, 276, 5177 5188. 74. Shoenborn, B. P.; Watson, H. C.; Kendrew, J. C., Binding of xenon to sperm whale myoglobin. Nature 1965, 207, 28 30. 75. C.; Schildkamp, W.; Wulff, M.; Moffat, K., Protein c onformational relaxation and ligand migration in myoglobin: a nanosecond to millisecond molecular movie from time resolved Laue X ray diffraction. Biochemistry 2001, 40, 13802 13815. 76. y.; Ursby, T.; Pradervand, C.; Ren, Z.; Adachi, S. i.; Schildkamp, W.; Bourgeois, D.; Wulff, M.; Moffat, K., Photolysis of the carbon monoxide complex of myoglobin: nanosecond time resolved crystallography. Science 1996, 274, (5293), 1726 1729. 77. Sugimoto, T.; Unno, M.; Shiro, Y.; Dou, Y.; Ikeda Saito M., Myoglobin mutants giving the largest geminate yield in CO rebinding in the nanosecond time domain. Biophysical Journal 1998, 75, 2188 2194.
32 78. Tian, W. D.; Sage, J. T.; Champion, P. M., Investigations of ligand association and dissociation rates in the "open" and "closed" states of myoglobin. J. Mol. Biol. 1993, 233, 155 166. 79. Tilton, R. F.; Kuntz, I. D.; Petsko, G. A., Cavities in proteins: structure of a metmyoglobin xenon complex solved to 1.9 . Biochemistry 1984, 23, 2849 2857. 81. van Eldik, R.; Asano, T.; le Noble, W. J., Activation and reaction volumes in solution. 2. Chem. Rev. 1989, 89, 549 688. 82. Wang, Y.; Larsson, D. S. D.; Spoel, D. v. d., Encapsulation of myoglobin in a cetyl trimethylammonium bromide micelle in vacuo: a simulation study. Biochemistry 2009, 48, 1006 1015. 83. Westrick, J. A.; Peters, K. S., A photoacoustic calorimetry study of horse myoglobin. Biophys. Chem. 1990, 37, 73 79.
33 CHAPTER 2 TIME RESOLVED THERMODYNAMICS OF CO PHOTOLYSIS FROM AN ENGINEERED C U B MB Introduction Metalloproteins comprise a broad class of proteins, which are capable of biological processes ranging from photosynthesis to electron transfer to oxygen storage and activation Though they share the common feature of having a single or multiple metallic centers, these centers vary widely in terms of the identity of the metal, metal oxidation state, coordination functionality, as well as combinations of ligands ( see Holm et al., 1996 for a tabular summary ) Altogether, the seemingly limitless c ombination of metal type and environment make metalloproteins capable of unique and versatile chemistries that are still relatively unclear. Even so, much has been learned about metalloproteins and enzymes within the past decade by Lu et al., 200 1 and issue; Holm et al., 1996) and metalloprotein chemistry is becoming more accessible to biomime tic model systems. Over half of known proteins contain metal centers, with over a third of all proteins being proposed to require metals for activity (Thomson and Gray, 1998; Holm et al., 1996). Therefore, understanding the role that metal centers play in proteins is vital to unravelin g the structure to function relationship.
34 R ecent advances in metalloproteins have focused on protein engineering and de novo design aim ed at probing this role (see reviews by Lu et al., 2001 as well as Holm et al., 1996 ). The de novo protein design of pro tein scaffolds is appealing because there is a commonality in protein scaffolds In a recent structural classification of approximately 38,000 PDB entries it was found that the proteins could be grouped into 1195 different structural categories (Scop, 2009) This suggests that nature utilizes similar protein scaffolds to design proteins with widely varying functions (Lu et al., 2001). Similarly, t he de novo design of active sites into existing protein scaffolds is also desirable because many existing protein scaffolds are robust and able to accommodate mutations without significant structural changes. Traditionally, this has been utilized to probe the relationship of structure to function based on the effects of specific mutations (Lu et al,. 2009) In metal binding site design, however, not only can mutations be introduced to existing active sites providing unique insights into the dependence of function on key structural elements; in addition, novel binding sites can be engineered for the purposes o f discovering new functionalities and ultimately probing the mechanisms underlying metalloprotein chemistry (Lu et al. 2001) Among the metalloproteins, heme proteins are the most versatile and largest group (Yeung and Li, 2008). Heme proteins are named for their heme active sites, a protoporphyrin IX or derivative ( Figure 1.2 and 2.1 ) and represent a diverse group of proteins: monooxygenases, oxidases, peroxidases, catalases, globins (Larsen and Many successfu l attempts have been made to introduce novel functionality into existing heme proteins by simply altering the heme environment including the covalently bound proximal and distal residues as well as residues within the
35 heme pockets (see review Lu et al., 2 001). Site directed mutagenesis studies by Sligar and Egeberg ( 1987 ) generated a cytochrome c type heme protein capable of peroxidase activity from a cytochrome b 5 type by simple mutation of the proximal residue. The introduction of heme degradation acti vity, furthermore, was observed in myoglobin after site directed mutagenesis of various residues within the heme pocket (Murakami et al., 1999; Hildebrand et al., 1996) and in cytochrome b 5 upon mutation of an axial histidine ( Rodrguez and Rivera, 1998 ). Proteins with heme active sites are ideal for this type of protein design because of their structural similarity, varying slightly in structure depending on the substitution along the macrocycle backbone. The structures of the four most common hemes a, b, c and o are shown in Figure 2.1 . Figure 2.1 The four most common heme types, a, b, c, and o.
36 The redesign of existing heme active sites within their native protein scaffolds is ideal because of the structural similarities between proteins and these key elements, as mentioned. One practical implication of this structural similarity is to utilize s impler, smaller metalloprotein scaffolds to mimic and ultimately recreate the activity of larger, more complex ones, namely heme proteins. Within the family of heme proteins, the globins (viz. hemoglobin and myoglobin) are one of the most biochemically, c omputationally, and spectroscopically characterized ( e.g. Antonini and Brunori, 1971) M yoglobin has been studied extensively as a model for larger, more complex heme proteins (Yeung and Lu, 2008) Recently, a site directed mutant of sperm whale myoglob in (Cu B Mb) was created in which Leu29 and Phe43 were replaced by two histidine residues (Sigman et al., 2000). The mutations create an environment of three histidine residues (His29, His43, and His64) that can be coordinated with a copper atom, a mock Cu B center, in close proximity to the heme. The insertion of a Cu B center into a myoglobin framework was found to introduce O 2 reduction activity similar to that observed in its biomimetic target, heme copper oxidase Similar studies using myoglobin as a pr otein framew ork have introduced catalase ( Adachi et al., 1993; Egeberg et al., 1993; Hildebrand et al., 1995 ) and peroxidase ( Adachi et al., 1993; Ozaki et al., 1996; Matsui et al., 1999) activities i nto myoglobin on the basis of site directed mutations h ighlighting the importance of proximal and distal residues in protein function. Hence, these mutants are vital in providing insight into how simple modifications can alter protein function, which will be used here for Cu B Mb to probe the effects of distal pocket mutations on CO escape.
37 The structure, kinetics and reactions of myoglobin were discussed in detail in Chapter 1 and hence the following serves as a summary in the context of mutations within the heme pocket. Ligand escape in myoglobin is believed to occur in multiple steps, during which the small ligand is transiently docked within the protein prior to gaining access to the solvent and has been shown to involve fluctuations that are presumably correlated with the creation of channels for the CO to escape. Within the distal pocket, the effects of mutating Leu29 and Phe43 to histidine residues introduce not only steric effects but electrostatic effects as well, see Figure 2. 8 and 2. 9 (in Results and Discussion) It was shown in the previous chapter the importance of the electrostatics specifically the hydrogen bonding network within myoglobin; hence mutations of this magnit ude additionally probe the nature of this effect. Because the methods of protein engineering presented here rely on existing protein scaffolds and (for the most part) existing active sites, it is important to detail the energetic implications of this type of protein design and engineering. In Cu B Mb in the absence of metal, which will be the focus of this chapter, the presence o f two larger, polar resid u es within the distal pocket is expected to affect the energetics of CO escape from the protein matrix. Using unique photothermal methods that have been highlighted in the previous chapter the energetic of CO escap e from engineer ed myoglobin and native myoglobin can be determined and should provide some insight into the effects of engineering a mock CuB center into a heme protein. The subsequent effects of copper coordination on CO diffusion are not probed here, though the ramifi cations of creating a heme copper oxidase like Cu B center warrants further study. Ultimately, a thorough understanding of the energetic s of protein
38 engineering will help in the development of proteins and biosynthetic models with novel functionalities and to unravel the role of structure to function in biological molecules. Methods Recombinant sperm whale myoglobin was obtained from Sigma Aldrich and used without further purification. Recombinant L29H/F43H myoglobin (hereafter Cu B Mb) was gratefully obtained from Dr Yi Lu (University of Illinois, Urbana Champaign ). Both proteins were dissolved in a 50 mM phosphate buffer, pH= 8 prior to each data acquisition to prevent protein degradation The CO bound form s were prepared as in Chapter 1. The con centration of Cu B Mb was determined using an extinction coefficient at 408 nm of 151 mM 1 cm 1 as provided by Dr. Lu. The techniques employed are similar to that of Chapter 1, hence will not be discussed in detail here with the exception of details unique to this data set. T emperature dependent acoustic traces of both sample and reference were obtained between 13 and 35 C by averaging over 10 single laser shots, in which the sample and reference were allowed to sit for 2 minutes between each laser impulse. Though this is unnecessary for the reference, as CO rebinds within the repetition rate of the laser (1 Hz), it was done to maintain consistency with the sample. The data reported is an average of three trials. CO photolysis and ligand escape were dete rmined to occur within the time resolution of the instrument. The amplitude of the phase for both sample and reference
39 was calculated by summing the absolute values of the magnitudes for the maximum and minimum from the acoustic traces for each temperatur e. determined as discussed in Chapter 1 (Methods) and Appendix 1. Transient absorption plots were obtained in two ways. First, the kinetics mode of a Shimadzu UV 2401PC spectrophotometer was used to monitor the change in the absorbanc e (namely growth) of the 422 nm band for Cu B Mb and hhMb to determine the time scale of CO rebinding. The CO was photolyzed by shining white light on the sample for approximately 30 seconds. Secondly, the changes in absorption as a function of time were o btained for both Cu B Mb and hhMb in which CO photolysis was initiated using a Continuum Leopard I frequency exc =532 nm, <20 ps pul se width, ~40 mW average power). The probe beam (a 150 W Xenon arc lamp, Thermo Oriel) was focused with a mirror and passed through the 0.2 cm side of a 0.2 cm x 1 cm cuvette containing the sample. The transmitted light was then passed through another mirror and detected using a PMT set at 440 nm (H6780, Hamamatsu), amplified (70710, Thermo Oriel), and dig itized (4 GHz transient digitizer, TDS 7404, Tektronix). and the in which single laser shots were used to initiate CO photolysis. The absorbance decays represent the average of 14 single shots, where readings were triggered from a photodiode The quantum yield of CO photolysis was determined using the pulsed method of Brunori et al. (1973) with transient absorption as described in using horse heart myoglobin as a referen ce and Equation (2.2 ), hhMb from Peters et al., 1991), = 73.4 mM 1 cm 1 for horse heart myoglobin 1 cm 1 as calculated from
40 the absorption spectra), c is the concent ration of the species, and A is the change in the absorbance at 440 nm (2.2) The absorption decay over time were fit to m onophasic (hhMb) and biphasic (Cu B Mb) exponential decays from OriginPro 8.0 and values were extracted from the fits: o ) and pre exponential factors (A 1 A 2 ) in Equation 2.3 and 2.4, respectively. (2.3) (2.4) Results and Discussion The steady state absorption spectra for sperm whale myoglobin (sw Mb) and L29H/F43H sperm whale myoglobin ( Cu B Mb) are identical and the spectra for CO bound Cu B Mb and swMb are shown in Figure 2. 2 The UV/Vis spectra of deoxy myoglobins (both deoxysw Mb and deoxy Cu B Mb) are char acterized by a Soret peak at 433 nm, which is blue shifted in the carboxy form to 421 nm. A smaller peak at 55 4 nm
41 in deoxymyoglobin (of both types) is replaced by two peaks at 540 and 579 nm in the carbon monoxide bound forms. It was found that the rebinding of CO to Cu B Mb occurs on a second time scale, as shown in Figure 2. 3 (left) while CO is known to rebind to horse heart myoglobin within milliseconds ( ) Figure 2. 4 ( right ) If the entire myoglobin population had CO bound to the heme in Cu B Mb, there would be no absorbance changes observed on the second time scale, as the band being observed corresponds to the CO bound myoglobin Soret. All of the CO is rebound in hhMb and hence there is no change in the absorbance of the CO band as is the ca se for horse heart myoglobin in Figure 2. 3 (right). Similarly, Figure 2. 4 shows the transient absorption for Cu B Mb and hh Mb indicating that the lifetime for CO rebinding is longer in Cu B Mb. CO rebinding to horse Figure 2.2 The overlaid UV/Vis spectra of CO bound swMb (black) and Cu B Mb (red).
42 heart myoglobin was found to be monophasi c, with a lifetime of 3.27 0.01 ms, while CO rebinding to Cu B 1 =2.01 0.03 2 = 18.1 0.3 ms. The quantum yield for CO photolysis in Cu B Mb was Photoacoustic calorimetry was performed on swMb and Cu B Mb as well as calibration compound Fe4SP. The acoustic traces for one trial of Cu B Mb (and reference) are shown in Figure 2. 5 The absence of a shift in frequency of the Cu B Mb related to the frequency of the calibration compound is indicative that the events associated with CO photolysis and escape are occurring within the time resolution of the instrument (< 50 ns). Photothermal studies of sperm whale myoglobin have been reported in the literature a nd the associated acoustic traces are not reported here (Peters et al., 1991; Angeloni and Feis, 2001). For recombinant sperm whale myoglobin, two phases are observed. The prompt phase occurs within the time resolution of the instrument (< 50 ns) and the second slow phase occurs with a lifetime of ~ 850 ns. Peters et al. (1991) attribute the slow phase to the escape of CO to the solvent, which they observe to occur with a lifetime of ~ 700 ns. The lifetime of the slow phase is slightly higher here, howe ver falls within the range of Peters et al. (1991) and the 940 ns lifetime observed by Feis and Angeloni (2001).
43 Figure 2.4 Transient absorption plots for Cu B Mb (right) and hhMb (left). The fits and indicated lifetimes were obtained using Equations 2.3 and 2.4 respectively. The transient absorption for Cu B Mb is on a 4 ms time scale, while that for hhMb is on a 2 ms time scale. Figure 2.3 The changes in absorbance as a function of time for Cu B Mb (left) and hhMb (right). The flat line of hhMb is indicative that all of the CO has rebound to the heme within the time scale of the measurement, hence there is no growth in the CO band at 422 nm. This is not the case for Cu B Mb (left), in which the in crease in the absorbance at 422 nm indicates that the CO is still rebinding on this time scale. Baselines were obtained in both cases, in which CO was not photolyzed.
44 A side by side comparison of the C p plots is shown in Figure 2. 6 The magnitudes of the total enthalpy and volume changes for swMb were determined to be Total Total = 6 6 kcal mol 1 /2 1 mL mol 1 while those for Cu B Mb were determined to be Total Total = 87 6 kcal mol 1 / 22 2 mL mol 1 These value s are summarized in Table 2.1. T he magnitudes of the enthalpy and volume changes for sw Mb are comparable to those observed in the aforementioned photothermal studies, which are summarized in Table 2 .2. Figure 2. 5 The acoustic traces arising from CO photolysis of Cu B Mb (black) and the calibration compound Fe4SP (red).
45 p s are approximate. Dashes indicate that the values are not applicable because a slow phase was not observed. p kcal/mol p mL/mol s ns s kcal/mol s mL/mol Total kcal/mo l Total mL/mol swMb 20 4 1.7 0.7 850 26 4 3.8 0.8 6 6 2 1 Cu B Mb 87 6 22 2 87 6 22 2 Table 2 .1 Thermodynamic profiles for swMb and Cu B Mb. Figure 2. 6 A side by side comparison of the C p plots for Cu B Mb (left) and swMb (right).
46 p kcal/mol p mL/mol s ns s kcal/mol s mL/mol Total kcal/mol Total mL/mol Feis (2003) 12 1 4.9 0.3 940 9 1 15.3 0.8 21 2 10 1 Peters (1990) 1 2 9.2 0.5 700 11 2 15 1 11 2 6 1 swMb 20 4 1.7 0.7 850 26 4 3.8 0.8 6 6 2 1 Cu B Mb 87 6 22 2 87 6 22 2 Following the analysis and discussion from Chapter 1, the contributions to the enthalpy and volume changes for the prompt phase in sperm whale myoglobin arise from Fe CO bond cleavage, heme low spin to high spin transition, and conformational changes. The magnitudes of the enthalpy and volume changes for Fe CO bond cleavage and spin state changes are: 1 1 2005). Thus, conformational changes associated with the prompt phase would result in 23 kcal mol 1 17 mL mol 1 for sw 104 kcal mol 1 37 mL mol 1 for Cu B Mb. Previously, the negative volume change was attributed to distal pocket volume contraction as the CO transiently migrates from the dist al pocket and no longer hinders surrounding residues (Angeloni and Feis, 2003 ; Phillips et al. 1999; Kachalova et al. 1999) The same is true for both Cu B Mb and swMb, as suggested by Table 2 .2 Thermodynamic profiles for swMb as reported in the literature. p s are approximate. Dashes indicate that the values are not applicable because a slow phase was not observed.
47 the volume contractions in both cases. for both types of myoglobin is significantly exothermic, d espite Fe CO bond cleavage being endothermic Cited literature values are also largely endothermic, however, which suggests that the conformational changes accompying volume contraction as well as protein rearrangements upon transient CO docking are signi ficantly exothermic. The breaking of the Arg45 heme propionate salt bridge in sperm whale myoglobin has been estimated by Peters et al. (1991) to result be 8 kcal mol 1 and 5 mL mol 1 respectively. Though this was proposed to be negligible in the case of horse heart myoglobin in the previous 2007), the salt bridge is expected to be stronger in swMb because Arg is capable of formin g three hydrogen bonds compared to two of Lys ( Vojtechovsk et al., 1999). Therefore, breaking of the salt bridge likely contributes a small amount to the exothermic enthalpy of the prompt phase. Presumably, the remaining enthalpy release in swMb ( 15 to 23 kcal mol 1 depending on the contribution from the salt bridge) arises from the conformational changes associated with the volume contraction, as was the case with hhMb and D 2 OMb of Chapter 1. In the slow phase, which appears to be combined in the prom pt phase in Cu B Mb, volume and enthalpy changes must account for CO escape, solvation, water entry and conformational changes arising from protein rearrangements associated with channel opening (namely, His64 gating). Following the previous chapter, the vo lume expansion expected when CO leaves the protein matrix to the solvent is ~ 33 mL mol 1 (Moore et al., 1982; Eley, 1939), while the volume contraction associated with water entry is estimated to be between 10 to 14 mL mol 1 for occupancies of 0.6 to 0.8, respectively.
48 between 15 and 27 mL mol 1 The Angeloni (2003) as well as Westrick et al. (1990) fall within this range as seen in Table 2.2 ; however, are ~ 10 mL mol 1 Using the observed and expected volume changes, the conformational changes associated with the 15 mL mol 1 (for swMb m easured here) and ~ 5 mL mol 1 (literature). It is unclear from where the 10 mL mol 1 difference in volume change arises, though may originate in part from the differences in buffer conditions, where the buffer used here is half of the ionic strength of that used in Westrick et al. (1990) and Feis and Angeloni (2003) and hence volume changes would be less restricted due to slight viscosity differences. The magnitudes of the enthalpy changes determined here for swMb, furthermore, are drastically different from those reported in the literature in that the slow phase enthalpy change is largely exothermic here, as opposed to endothermic in 26 kcal mol 1 versus ~ 10 kcal mol 1 respectively Clearly, other events must accoun t for the additional 36 kcal mol 1 heat release measured in this study. The magnitudes of the total enthalpy changes for Cu B Mb and sw Mb are significantly different also which suggests that the events associated with ligand diffusion are different in Cu B Mb. The events associated with CO photolysis and diffusion in Cu B Mb cannot be resolved in PAC, occurring < 50 ns; whereas, in swMb a ~ 800 ns can be resolved in addition to the < 50 ns phase. Furthermore, the enthalpy and volume changes in Cu B Mb are ~ 81 kcal mol 1 and ~ 23 mL mol 1 lower co rresponding to a larger, more end othermic volume contraction Photothermal studies ( e.g. Peters et
49 al., 1991 ) done on myoglobin mutants in general have not accounted for the quantum yield changes based on residue mutations. The quantum yield of Cu B Mb is significantly lower (12.4%) than swMb (96%), confirming that CO photolysis and escape is indeed altered in Cu B Mb. In fact, the high percentage of CO recombination to the heme implies that large barriers exist for CO escape, suggesting that the remaining 12% of CO molecules may not escape to the solvent at all. Instead, the CO molecules diffuse within the protein matrix that results in significantly altered values Cu B Mb is an L29H/F43H mutant of swMb, w here the introduction of the polar, aromatic histidine in place of the nonpolar, nonaromatic leucine and nonpolar phenylalanine is expected to alter the solvent accessible volume as well as the electrostatics within the distal pocket. The effects of intro ducing these mutations are shown in an overlay in Figure 2. 7 His64 as detailed in the previous chapter (Tian et al., 1993 and references therein; Mukai et al., 1998) which presumably occurs on the time scale of CO escape, ~ 800 ns. In Cu B Mb, His64 is swung toward the solvent in an open conformation while His43 is also swung outward from the distal cavity toward the solvent. n proposed to be the predominant (~80%) mechanism for CO escape (see Brunori et al., 2004 and references therein) the solvent and therefore CO escape would be faster. However, the percentage of mo lecules that do not rebind to the heme is low (~ 12%), suggesting that the energetic barrier to rebinding is significantly lower than CO escape and thus it is proposed that these CO molecules remain transiently docked in the matrix A predominant channel for CO escape has been proposed by Elber and Karplus (1990) to reside between residues
50 Val68 and Ile 107 (shown in Figure 2. 8 ) key residues in delineating the distal pocket in which it was determined that CO spends a significant portion of its time sampling this pocket The outward positions of His64 and His43 ultimately enlarge this AB/G cavity, presumably facilitating CO docking within it. A side by side comparison of the occluded pocket within swMb and Cu B Mb is shown in Figure 2. 8 Several fa ctors may be responsible for increased CO rebinding to the heme: steric or electrostatic effects. A study by Gibson et al. (1992) on CO recombination kinetics in Leu29 mutants found that larger residues at the 29 position resulted in lower geminate recomb ination. In addition, histidine by approximately 6 3 ( Creighton, 1993) or a partial molar volume difference of ~ 9 cm 3 mol 1 (Sirimulla, 2010) Together, these suggest that steric effects are not occurring here. In fact, the effect of decreased cavity sizes can be ruled out entirely based on surface topography calculations in which Cu B Mb was in general found to have larger cavities and pockets than swMb (Dundas et al., 2006), see Table 2.3. Therefore, the effect of altered electrostatics appears to be the most important factor in the decreased quantum yield, which is not unexpected as the heme pocket in Cu B Mb is more polar compared to swMb based on the presence of two additional h istidines. This is supported by a number of other studies showing increased geminate recombination when polar residues are present in the distal pocket (Carver et al., 1990; Gibson et al., 1992; Braunstein et al., 1988; Lambright et al., 1994; Sugimoto et al., 1998).
51 Figure 2. 7 An overlay of the heme and distal pocket of Cu B Mb (red) and swMb (blue). Residues delineating the distal pocket are labeled. The backbone and non distal pocket residues are also overlaid and use the following color scheme: green (carbon atoms of swMb), light blue (carbon atoms of Cu B Mb), red (oxygen), dark blue (nitrogen), white (hydrogen).
52 B Mb and swMb. Though this is the first known thermodynamic study of the double mutant L29H/F43H (Cu B Mb), some comparisons can be drawn from mutants reported in the literature. In L29 W mutants, iron motions and subsequent volume contractions were found to be smaller and overall the heme domed toward the distal pocket rather than the proximal pocket (Nienhaus et al., 2005) Furthermore, transient grating studies by Figure 2. 8 A side by side comparison of the heme and distal pocket of swMb (left) and Cu B Mb (right). Residues delinea ting the distal pocket are labeled. The backbone and non distal pocket residues are also overlaid and use the following color scheme: green (carbon atoms of swMb), light blue (carbon atoms of Cu B Mb), red (oxygen), dark blue (nitrogen), white (hydrogen).
53 Table 2.3 The total solvent accessible volume of the four largest cavities within swMb and Cu B Mb. Sakakura et al. (2002) on H64L, H64V, and H64Q mutants have shown volume differences ranging from 0 to 20 mL mol 1 for mu tants within the distal pocket. As the volume contraction for Cu B Mb is ~ 24 mL mol 1 greater than swMb, the magnitude of this volume change falls within the range of observed values. The enthalpy difference s, furthermore, are significantly different between Cu B for the events occurring within swMb and Cu B Mb are also different. A n umber of processes as explained in detail in Chapter 1 and briefly above must occur for CO to leave the protein matrix, including Fe CO bond cleavage, spin state change of the heme, as well as events that occur upon CO escape such as CO solvation. These processes are not dependent on the identity of residues within the distal pocket, hence it would appear that conformational or structural changes occurring within the protein matrix associated with these events is the origin of the 81 kcal mol 1 difference; namely, the structural rearrangements in Cu B Mb are more exothermic by 81 kcal mol 1 swMb Cu B Mb Volume ( 3 ) 275.54 399.36 In the CASTp calculation, the four largest cavities correspond to pocket ID 16 through 19, presumably corresponding with the four Xenon binding cavities on the basis of visualization.
54 There is no precedent in the literature for such a large enthalpic difference between swMb and the mutant Cu B Mb. Transient grating studies aforementioned of H64L, H64V, and H64Q rev 5 kcal mol 1 (Sakakura et al., 2002) while R45N and R45G site directed mutants of Westrick et al (1990) were found 1 difference). These mutations differ from the double mutant studied here, however. In both cases, the substitutions involve removing a charged residue and replacing it with a uncharged residue; that is, there is no net increase in polarity as in the Cu B M b Introduction of two polar, charged residues in place of two nonpolar, uncharged residues drastically changes the hydrogen bonding network and electrostatics within the pocket which is proposed to a ffect the extent of water binding within the heme pock et. It was discussed in Chapter 1 that the presence of water within the distal pocket is significantly reduced when apolar residues are present (Quillin et al., 1993). Moreover, in a site directed mutation study by Liong et al. (2001) it was found that s ubstitution of apolar residues with polar residues within the heme pocket results in the appearance of 3 4 crystallographic waters. Hence, the additional 81 kcal mol 1 released may arise in part from the heat release as additional waters enter the heme po cket also accounting for the larger volume contraction water entry was estimated by Goldbeck et al. (2006) to be exothermic by 8 kcal mol 1 and 10 mL mol 1 ; hence the presence of 3 to 4 additional water molecules would be expected to have an exothermic heat release of 24 to 32 kcal mol 1 associated with an ~ 30 to 40 mL mol 1 volume contraction Though this estimate leaves an additional heat release of 50 kcal mol 1 the presence of additional waters cannot be ruled out en tirely, as the site directed mutants in Liong et al. (2001) were single mutants
55 and Cu B Mb contains two additional polar residues. Furthermore, structural reorganization of the protein around the waters could presumably involve a volume expansion on the or der of 10 20 mL mol 1 observed in this study. The additional presence of water within the distal pocket also accounts for the increased CO rebinding rate, as CO access to the heme pocket involves competing with the water molecules. As mentioned, the rebinding rate of CO to Cu B Mb was found to be on the order of seconds, where CO rebinding to swMb occurs within milliseconds Therefore, it would appear that CO rebinding to the heme is significantly slower because of the increased number of waters that hinder CO access to the heme pocket.
56 Conclusion This is the first known photothermal study to probe the energetics of an L29H/F43H sperm whale myoglobin mutant, known as Cu B Mb. Developments in protein engineering, specifically altering existing metal binding sites and protein scaffolds to introduce novel functionality relies on a thorough understanding of the mechanistic implications of mutations. Here, the effect on CO esc ape was probed in a myoglobin mutant, whose active site pocket was redesigned as a potential heme copper protein mimic. Events associated with CO photolysis in Cu B Mb were found to occur within the time resolution of the instrument (< 50 ns) and be largely exothermic, with a large volume contraction. The results suggest that the mechanism guiding ligand escape is highly dependent on the electrostatic environment within the heme pocket, where the introduction of polarity via the two histidine mutations hin ders CO escape Additionally, the added polarity into the distal pocket is responsible for the increased water population and the subsequent large heat release associated with water entry as well as increased CO rebinding time
57 References 1. Adachi, S. i.; Nagano, S.; Ishimori, K.; Watanabe, Y.; Morishima, I., Roles of proximal ligand in heme proteins: replacement of proximal histidine of human myoglobin with cysteine and tyrosine by site directed mutagenesis as models for P 450, chloroperoxidase, and catalase. Biochemistry 1993, 32, 241 252. 2. Antonini, E.; Brunori, M., Hemoglobin and myoglobin in their reactions with ligands North Holland Publishing Company: Amsterdam, 1971. 3. Braunstein, D.; Ansari, A.; Berendzen, J.; C owen, B. R.; Egeberg, K. D.; Fraunfelder, H.; Hong, M. K.; Ormos, P.; Sauke, T. B.; Scholl, R.; Schulte, A.; Sligar, S. G.; Springer, B. A.; Steinbach, P. J.; Young, R. D., Ligand binding to synthetic mutant myoglobin (HisE7 -> Gly): role of the distal his tidine. Proc. Natl. Acad. Sci. USA 1988, 85, (8497 8501). 4. Brunori, M.; Bourgeois, D.; Vallone, B., The structural dynamics of myoglobin. J. Struct. Biol. 2004, 147, 223 234. 5. Brunori, M.; Giacometti, G. M.; Antonini, E.; Wyman, J., Heme proteins: quantum yield determined by the pulse method. PNAS 1973, 70, (11), 3141 3144. 6. Carver, T. E.; Rohlfs, R. J.; Olson, J. S.; Gibson, Q. H.; Blackmore, R. S.; Springer, B. A.; Sligar, S G., Analysis of the kinetic barriers for ligand binding to sperm whale myoglobin using site directed mutagenesis and laser photolysis techniques. J. Biol. Chem. 1990, 265, (32), 20007 20020. 7. Creighton, T. E., Proteins: structures and molecular propert ies Macmillan: New York, NY, 1993. 8. metal ion type, coordination number and the amino acid residues involve in the coordination. Acta Cryst. 2008, D64, 257 263. 9. Dundas J.; Ouyang, Z.; Tseng, J.; Binkowski, A.; Turpaz, Y.; Liang, J., CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucleic Acid Research 2006, 34, W116 W118. 10. Egeberg, K. D.; Springer, B. A.; Martinis, S. A.; Sligar, S. G.; Morikis, D.; Champion, P. M., Alteration of sperm whale myoglobin heme axial ligation by site directed mutagenesis. Biochemistry 1990, 29, 9783 9791. 11. Elber, R.; Karplus, M., Enhanced Sampling in Molecular Dynamics: Use of the Time Dependent Hartree Approximation for a Simulation of Carbon Monoxide Diffusion through Myoglobin. J. Am. Chem. Soc. 1990, 112, 9161 9175. 12. Gibson, Q. H.; Regan, R.; Elber, R.; Olson, J. S.; Carver, T. E., Distal pocket residues affect picosecond ligand recombination in myoglobin. J. Biol. Chem. 1992, 267, (31), 22022 22034. 13. Goldbeck, R. A.; Bhaskaran, S.; Ortega, C.; Mendoza, J. L.; Olson, J. S.; Soman, J.; Kliger, D. S.; Esquerra, R. M., Water and ligand entry in m yoglobin: Assessing the speed and extent of heme pocket hydration after CO photodissociation. PNAS 2006, 103, (5), 1254 1259.
58 14. Hildebrand, D. P.; Burk, D. L.; Maurus, R.; Ferrer, J. C.; Brayer, G. D.; Mauk, A. G., The proximal ligand variant His93Tyr of horse heart myoglobin. Biochemistry 1995, 34, 1997 2005. 15. Hildebrand, D. P.; Tang, H. l.; Luo, Y.; Hunter, C. L.; Smith, M.; Brayer, G. D.; Mauk, A. G., Efficient coupled oxidation of heme by an active site variant of horse heart myoglobin. J. Am. Chem Soc. 1996, 118, 12909 12915. 16. Holm, R. H.; Kennepohl, P.; Solomon, E. I., Structural and functional aspects of metal sites in biology. Chem. Rev. 1996, 96, (2239 2314). 17. Lambright, D. G.; Balasubramanian, S.; Decatur, S. M.; Boxer, S. G., Anatomy a nd dynamics of a ligand binding pathway in myoglobin: the roles of residues 45, 60, 64, and 68. Biochemistry 1994, 33, (18), 5518 5525. 18. to heme proteins. Coor. Chem. Rev. 2007 251, 1101 1127. 19. Liong, E. C.; Dou, Y.; Scott, E. E.; Olson, J. S.; Phillips Jr., G. N., Waterproofing the heme pocket. J. Biol. Chem. 2001, 276, (12). 20. Lu, Y.; Berry, S. M.; Pfister, T. D., Engineering novel metalloproteins: design of metal bindin g sites into native protein scaffolds. Chem. Rev. 2001, 101, 3047 3080. 21. Lu, Y.; Yeung, N.; Sieracki, N.; Marshall, N. M., Design of functional metalloproteins. Nature 2009, 460, (7257), 855 862. 22. Matsui, T.; Ozaki, S. i.; Liong, E.; Phillips Jr., G. N., Effects of the location of distal histidine in the reaction of myoglobin with hydrogen peroxide. J. Biol. Chem. 1999, 274, (5), 2838 2844. 23. Mukai, M.; Nakashima, S.; Oleson, J. S.; Kitagawa, T., Time Resolved UV Raman Detection of a Transient Open Form of the Ligand Pathway in Tyr64(E7) Myoglobin. J. Phys. Chem. B 1998, 102, 3624 3630. 24. Murakami, T.; Morishima, I.; Matsui, T.; Ozaki, S. i.; Hara, I.; Yang, H. J.; Watanabe, Y., Effects of the arrangment of a distal catalytic residue on regioselect ivity and reactivity in the coupled oxidation of sperm whale myoglobin mutatnts. J. Am. Chem. Soc. 1999, 121, 2007 2011. 25. Nienhaus, K.; Ostermann, A.; Nienhaus, G. U.; Parak, F. G.; Schmidt, M., Ligand migration and protein fluctuations in myoglobin mut ant L29W. Biochemistry 2005, 44, 5095 5105. 26. Ozaki, S. i.; Matsui, T.; Watanabe, Y., Conversion of myoglobin into a highly stereo specific peroxygenase by the L29H/H64L mutation. J. Am. Chem. Soc. 1996, 118, 9784 9785. 27. Rodrguez, J. C.; Rivera, M., Conversion of mitochondrial cytochrome b5 into a species capable of performing the efficient coupled oxidation of heme. Biochemistry 1998, 37, 13082 13090. 28. Scop Structural classification of proteins, 1.75 release. http://scop.mrc lmb.cam.ac.uk/scop/cou nt.html 29. Sigman, J. A.; Kwok, B. C.; Lu, Y., From myoglohbin to heme copper oxidase: design and engineering of a CuB center into sperm whale myoglobin. J. Am. Chem. Soc. 2000, 122, 8192 8196.
59 30. Sirimulla, S.; Lerma, M.; Herndon, W. C., Predictino of partial molar volumes of amino acids and small peptides: counting atoms versus topological indices. J. Chem. Inf. Model. 2010, 50, 194 204. 31. Sligar, S. G.; Egeberg, K. D., Alteration of heme axial ligands by site directed mutagenesis: a cytochrome becom es a catalytic demethylase. J. Am. Chem. Soc. 1987, 109, 7896 7897. 32. Sugimoto, T.; Unno, M.; Shiro, Y.; Dou, Y.; Ikeda Saito, M., Myoglobin mutants giving the largest geminate yield in CO rebinding in the nanosecond time domain. Biophysical Journal 1998 75, 2188 2194. 33. Thomson, A. J.; Gray, H. B., Bio inorganic chemistry. Curr. Op. Chem. Biol. 1998, 2, 155 158. 34. Tian W. D.; Sage, J. T.; Champion, P. M., Investigations of ligand association and dissociation rates in the "open" and "closed" states of myoglobin. J. Mol. Biol. 1993, 233, 155 166. 35. Vojtechovsk, J.; Chu, K.; Berendzen, J.; Sweet, R. M.; Schlichting, I. Crystal structure of myoglobin ligand complexes at near atomic resolution. Biophys. J. 1999, 77, 2153 2174. 36. Westrick, J. A.; Peters, K. S., A photoacoustic calorimetry study of horse myoglobin. Biophys. Chem. 1990, 37, 73 79. 37. Yeung, N.; Lu, Y., O ne heme, diverse functions: using biosynthetic myoglobin models to gain insights into heme copper oxidases and nitric oxide reductases. Chem Biodivers 2008, 5, (8), 1437 1454.
60 CHAPTER 3 PHOTOPHYSICAL CHARACTERIZATION OF A NOVEL RESORCINARENE BASED CAVITAND Introduction Calix[n ]arenes (Figure 3. 1) are a class of cyclic oligomer s whose backbone s are comprised of n repeating functionalized aromatic rings that are bridged to enclose a hydrophobic cavity. F irst synthesized by Zinke ( 1941 ) from the condensation of p tert butylphenol and formaldehyde their cyclic al structure was not reported with substantial structural evidence until 1952 (Zinke, et al.) T heir mostly widely used name calixarene, was introduced by Gutsche in 1978 (Gutsche and Muthukrishnan, 1978) because of the ir shape, which resembles a Greek vase known as a calix. Calixarenes can be bridged only on the upper rim of the aromatic rings or both the upper and lower rim to form a more conformationally rigid structure, known as a cavitand (Verboom, 2001). The name cavitand was introduced by Cram in 1982 (Moran, et. al., 1982) to describe a class of Cram showed that cavitands of varying functionality and sizes could be synthesized on depending on the alcohol and aldehyde precursors (Cram, et al., 1988). Those arising from reso rcinol are termed resorcinarene based cavitands which will be the focus of this photophysical characterization study.
61 The appeal of calixarenes and their rigidized derivatives cavitands resides in their highly tunable chemistry, namely the synthetic control over the cavity sizes and the array of functional groups that can be substituted along the upper and/or lower rims ( Parulekar, 2007; Gutsche 1983). The cavity sizes depend on the number of aromatic subunits, which can be found ranging from n=3 to 9 (Gutsche, 1985; Neri, 2001and references therein ) The hydrophobic cavities of calixarenes and cavitands have been exploited in a wide range of applications including host guest chemistry (Ihm 2004; Xu 1997; Lledo 2008; Moran 1982; Shinkai 1986 ), biomimetic complexes via metal ion coordination (Clainche 2000; Over 2009; Blanchard 1998; Rondelez 2002; Cacciapaglia 2006), and supramolecular assemb lies (Harrison, 2002; Fochi, 2001; At wood 1993). For ins tance, the novel functionalized resorcin  arene derived cavitand that will be studied here is functionalized with imidazole arms that present a potential site for copper ion Figure 3. 1 (left) A line structure representation of the calixresorcinarene based cavitand. In this structure, n=4 and R=C 6 H 13 (right) A 3D ball and stick representation of calixresorcinarene, with the R groups replaced with Hydrogen atoms.
62 coordination, and bridged along the up per and lower rims with oxygen and carbon respectively. descri be the use of artificial chemical systems to reproduce biological process es (see Breslow, 2009 and references therein). At the time, he was working on cyclodextrins, a sugar based macrocycle analogous to calixarenes, to mimic bifunctional enzyme catalysis ( Breslow, 2009, 1995, 1994) Today, t he field has grown to encompass areas such as bioinorganic and bioorganic chemistry Biomimetic chemistry is largely appealing, in part because a high degree of control can be exhibited over the chemical structure of the model system and a corresponding structure function relationship can be established that is difficult to obtain in vivo Also, biomimetic model systems have the potential to provide important mechanistic details that cannot be unraveled or even asc ertained in complicated biological systems (e.g. Wang, 1998; Schrder, 2009; Collman, 2008 ). However, t he design of enzyme mimics is difficult. First, the biomimetic model system must possess the same coordination environment as the intended target. Se condly, the model system must have some degree of flexibility. And thirdly, there must be a way to characterize the system, preferably spectroscopically. Gutsche (1989, 1988, and 1983) first recognized the applicability of calixarenes as enzyme mimics because of their relatively straightforward synthesis, highly versatile chemistry and their ability to coordinate metal s retaining open coordination sites for small molecules A variety of functional groups can be substituted onto the rims and calixarene cavity size s are tunable based on the number and types of aromatic monomers Calixarenes are known to have the conformational flexibility inherent in biological
63 systems (Gutsche, 1989; Rondelez, 2002), though their derived cavitands are considerably less so because of the upper and lower rim bridging of the aromatic rings (Verboom, 2001). In fact, metal coordination to calixarenes has been shown to induce conformational changes that are being used as a form of allosteric communication (Ohseto et. al, 199 5; Shinkai et. al, 1990; Iwamoto et. al, 1993). Calixarenes are also ideal for enzyme mimetic systems because of their inherent spectroscopy. Early in the history of calixarenes, characterization by Kmmerer revealed that they have characteristic absorption bands at 280 nm and 288 nm ( Kmmerer and Happel, 1980 and 1981; Kmmerer et al., 1981 ; Gutsche, 1989 and references therein ) However, overall little work has been done in characterizing metal complexation to cali xarenes on the basis of spectral changes of the calixarene itself ( for one example, see Prodi 2000) For the most part, calixarenes have been well characterized in terms of the photophysical properties of their fluorescent substituents for sensor base d applications (for a thorough review, see Kim and Quang, 2007 ) These studies have relied on spectroscopic changes of the functional g roups, not the calixarene, from anion or cation binding ( Schazmann et al., 2006; Kim et al., 2004; Choi et al., 2006; Ni shizawa et al., 1999 ; Megyesi and Biczk, 2010 ) or from metal ion complexation ( Beer, 1998; Ji et al., 2000 ) In fact, calixarene photophysical characterization has even been done in terms of spectral shifts of the metal bands themselves (Gutsche and Nam, 1988). Recently, Reinaud and coworkers have utilized functionalized calix[6 ]arenes coordinated with copper (I) as a novel way to mimic the mononuclear copper centers that are found in copper enzymes (Blanchard et al. 1998 (a,b); Rondelez et al., 200 0 and 2002 (a,b); Clainche et al. 2000). By functionalizing both the upper and lower rims with
64 imidazole and sulfonato groups respectively they were able to synthesize a water soluble three coordinate cuprous complex. In general, t hree coordinate copper (I) complexes are difficult to synthesize because of copper (I) disproportionation in solution as well as oxidati on by air (Sorrell et. al, 1982 ). However, with this case it was found that the imidazole groups surrounded the copper ion in such a wa y as to stabilize it in both air and water (Rondelez et al. 2002). Within the field of biomimetic chemistry, de novo protein design is rapidly becoming a test of what has been learned about structure function and mechanism thus far, see DeGrado, 2001 and issue for example. This has been discussed in detail in the previous chapter ; however, to summarize, much can be learned about m etalloproteins specifically heme copper oxidases (HCOs) in this case, via the engineering of novel catalytic site s into existing protein scaffolds The Cu B Mb model system of the previous chapter, for instance, introduces a novel copper binding site in to a myoglobin scaffold, thereby introducing new functionality. Namely, Cu B Mb is both an O 2 activator and carrier (Sigman et al., 2000). Because s mall diatomic molecules such as O 2 CN and CO transiently bind to the Cu B center in HCOs as well as in Cu B Mb o ne way to unravel the nature of the Cu CO bond in these system s is via model systems. A simpler model such as the copper coordinated resorcinarene based cavitand presented here is ideal for a number of reasons First, t he presence of four imidazole arms substituted on the upper rim is analogous to the presence of histidines, which are the physiological ligands of copper in heme copper oxidases and many other copper proteins. Second t he aforementioned stability conferred by imidazole functionalized calixarenes on the Cu(II)/Cu(I) oxidation states allows for the characterization of both cupric and cuprous
65 species. And finally, the coordination geometry of the copper in the resorcinarene based cavitand exposes open metal coordination sites for small molecules such as CO, CN and O 2 as in Figure 3. 2 With this motivation in mind, t he results presented here are the initial photophysical characterization of a novel resorcinarene based cavitand using the inherent spectroscopy of the cavitand alone, not backbone substituents Figure 3.2 A 3D cartoon representation of a copper coordinated cavitand, in which an open coordination site is occupied by CO.
66 Methods The resorcinarene based cavitand was obtained from Dr. Kirpal S. Bisht. All chemicals were used as supplied. Anhydrous c opper (II) chloride was purchased from MCB (Matheson, Coleman, and Bell; Norwood, OH) copper (II) acetate monohydrate from Sigma Aldrich (St. Louis, MO) 5 hydroxybenzene 1,3 dicarboxylate and ethanol from Alfa Aesar (Ward Hill, MA) Absorption spectra were obtained using a Shimadzu UV 2401 PC spectrophotometer. Steady state e mission spectra and fluorescence polarization measurements were obtained using an ISS PC1 (ISS, Inc., Champaign, IL) single photon counting spectrofluorome ter. All measurements were conducted in ethanol. Analysis and g raphs were obtained using OriginPro 8.0. Steady State Absorption and Emission Measurements The concentration of calixresorcinarene was calculated using the determined molar extinction coe fficient at 280 nm, as determine via a Beer Lambert plot, Figure 3.4. Titrations were done by the addition of a known amount of Cu(II)Cl 2 into a cuvette containing a predetermined concentration of calixarene with stea d y state a bsorption and emission spectra taken after each addition of Cu(II)Cl 2 Absorption spectra were corrected for the dilution; however this did not exceed 5%. Because Cu(II)Cl 2 has a broad absorption band, the contributions to the primary and secondary inner filter for the excitation wavelength and over the emission wavelength range (290 to 400 nm) respectively were adjusted according to Equation ( 3. 1 ). All emission spectra were corrected according to this equation prior to taking the integrated intensities.
67 ( 3. 1) This equation presumes that the primary and secondary inner filter effects are equally weighted and thus the OD used to correct for the inner filter is the average of the optical densities at the excitation and emission wavelengths. Averaging the optical densities is reasonabl e if the sample is under constant irradiation, half of the population of absorbers photons. The dissociation constant was obtained by plotting the relative change in integrated intensity versus the concentration of Cu(II)Cl 2 Fits were obtained according to Equation ( 3 2 ). ( 3. 2) Quantum Yield The quantum yield for calix resorcin arene was calculated using the comparative method of Williams, et al. (1983). This method requires taking the ratio of the integrated emission spectra for the cavitand to that of the reference compound 5 hydroxy benzene dicarboxylate (5OH BDC) as in Equation (3 ) The emission spectra were normalized according to (1) Because all spectra for sample and reference were obtained in EtOH, the refractive index is assumed to be the same.
68 ( 3. 3) Polarization Fluorescence polarization studies were done using copper(II) acetate. The polarization and anisotropies were obtained for stochiometric additions from 0 to 3 equivalents of Cu(OAc) 2 to a known concentration of calixarene. The absorption peak of Cu(OAc) 2 is blue shifted in refe rence to Cu(II)Cl 2 therefore this copper salt was chosen here in order to avoid inner filter effects observed with Cu(II)Cl 2 that cannot be corrected S amples were excited at 290 nm and polarization measurements obtained using a 1 mm slit width. Job Plot The binding sto ichiometry was determined via the method of invariate concentration ( Job plot ) in which the total concentration was fixed at 50 uM. First, a solution of 2 mL ~50 uM calixarene was prepared an d the absorption and emission spectra ob tained, then a serial dilution was performed with increasing amount of Cu(II)Cl 2 added so that [calixarene] + [ Cu(II)Cl 2 ] = 50 uM. The data was fit b y s o l v i n g f o r x i n t w o l i n e a r e q u a t i o n s s e t e q u a l t o e a c h o t h e r a t y F i t s w e r e o b t a i n e d u s i n g O r i g i n P r o 8 0 Fluorescence Lifetimes Fluorescence lifetimes were obtained by excitation of the calixresorcinarene (with and with out Cu(II)Cl 2 ) with a Continuum Leopard I frequency quadrupled
69 exc =266 nm, <20 ps pulse width, ~40 mW average power with a 20 Hz repetition rate), with the emission collected perpendi cular to the excitation beam. Molecular Modeling and Electronic Structure Calculations Calixresorcinarene and Cu(calixresorcinarene) were modeled using HyperChem 8.0 Geometry optimizations were performed using the semi empirical method PM3 and Polak Ribiere (conjugate gradient) algorithm. T he C 6 H 12 alkyl chains on the lower rim were replaced with hydrogen atoms for efficiency and are not expected to contribute to the overall spectroscopy or backbone geometry Atoms were treated as closed shell ( RHF ), with 0 charge placed on the cavitand and a charg e of 2 on the copper bound cavitand (spin multiplicity of 2) All calculations were set to an RMS gradient of 0.01 kcal 1 mol 1 Electronic structure calculations were performed using the semi empirical method ZINDO/S in HyperChem 8.0. As before, a toms were treated as closed shell (RHF) and transitions were calculated using single excited state CI for 10 occupied and unoccupied orbitals. Results and Discussion Calixreso r cinarene is a cyclic oligomer of benzene like derivatives, functionalized on the upper and lower rim with bridging oxygen and carbon atoms see Figure 3. 1 The photophysical properties, therefore, are expected to be similar to those of benzene monomers in solution based on the struct ure of the cavitand backbone alone 5
70 hydroxybenzene 1,3 dicarboxylate (5 OH H 2 BDC ) has previously been characterized in our laboratory (Larsen, 2007) and is analogous to the individual subunits of the resorcin  arene based cavitand as shown in Figure 3. 3 The UV/Vis and corresponding emission spectra for calixarene are shown in Figure 3.4, with the corresponding Beer Lambert plot. The calixarene is characterized by an 1 cm 1 ), with a small shoulder at 290 nm and an emission band at 313 nm. Electronic structure calculations, Figure 3.5, (under C 1 orbitals 196A to 200A. The absorption of calixarene in ethanol is similar to that of 1,4 max at 290 nm (Berlman, 1971). The ~ 33 n m Stokes shift is also comparable to the shift of dimethoxybenzene (~ 30 nm) and that of protonated 5OH H 2 BDC of ~43 nm (Berlman, 1971; Larsen, 2007). The spectroscopic similarities suggest that the absorption and emission bands reported here for the resorcinarene based cavitand arise from the photophysical properties of the individual subunits, not from the upper and/or lower rim functionality. Few studies have characterized calixarenes using the inherent absorptio n and emission of the aromatic backbone alone One by Prodi et al (2000) studied a calixarene crown ether that shows similar absorption and emission spectra to the resorcinarene based cavitand of this work, with an absorption band centered at 270 nm, small shoulder at 280 nm, and emission band centered at 310 nm, corresponding to a Soret shift of 40 nm. The similarity between the crown ether calixarene of Prodi et al.
71 (2000) and 5OH H 2 BDC presumably arises from the commonalit y of both functional groups to alk oxybenzene derivatives Figure 3. 3 A side by side comparison of the cavitand monomer to 5OH H 2 BDC. Note that the C 6 H 13 group has been omitted from the calixresorcinarene as well as carbon atoms joining individual monomers. Hydrogen atoms are omitted for clarity. Figure 3.4 (left) The steady state absorption ( b l a c k ) and emission ( r e d ) sp ectra of calixresorcinarene (right) The Beer Lambert plot for calixresorcinarene at 280 nm, as determined by steady state absorption spectra for a 5 concentrations of cavitand.
72 The quantum yield of the resorcinarene based cavitand is 0.011, which is also similar to that of 5OH BDC The magnitudes of the intensities obtained as summarized in Table 3.1, with the quantum yield determined using Equation (3). Compared to 1,4 an order of magnitude (Berlman, 1971) The di fference may be attributed to the solvent Figure 3.5 The absorption peaks represented as lines (top) and peaks (bottom) based on electronic structure calculations, as performed in Hype rChem 8.0 using ZINDO S/CI with closed shell atoms (RHF). Peaks under 250 nm were not considered as relevant bands, as these are not resolvable in the UV/Vis spectrum. Also, lines along the top of each represent forbidden transitions that are not observe d in the UV/Vis spectrum. The band in purple at 268.67 nm transition from MO 193 to 198, with an oscillator strength of 0.494, while that at 273.42 nm (not labeled) corresponds to an n to transition from MO 196 to 200, with an oscillator strength of 0.246.
73 Parameter Calixarene 5OH BDC Integrated Intensity 12278529.96 19999448.38 OD (at 280 nm) 0.283 0.369 0.011 0.014 polarities, but is more than likely due to concentration effects as suggested by Morris et al. (1976) in which the incident photons are reabsorbed and emitted leading to a higher quantum yield The concentration of 1,4 dimethoxybenzene in that study is approximately 150 times greater (4.7 mM) than the calixresorcin arene concentration here (0.027 mM) The quantum yield is also similar to fluorescence quantum yields obtained by Mallier et al. (1996) for calix[n]arenes, n=4,6, and 8, ranging from 0.05 to 0.10 which is slightly larger than the quantum yields reported here This suggests that fewer photons undergo nonradiative decay in the calixarenes of Mallier et al. (1996) and that the novel calixresorcinarene characterized presently behaves spectroscopically more like the individual monomers. Addition of Cu(II) Cl 2 into the resorcinarene based cavitand is complicated by primary and secondary inner filter effects due to the cupric broad absorption band in this region The increase in absorbance shown in Figure 3.6 (left) is due to the The values obtained from steady state absorption and emission spectra of calixarene and 5OH BDC and used in Equation (3) to calculate the quantum yield of calixarene in red. Table 3.1 Quantum yield parameters for calixresorcinarene
74 Cu(II)Cl 2 unchanged upon copper coordination Addition of Cu(II) Cl 2 to calixarene results in decreased fluorescence, presumably from the formation of a non fluorescent Cu(II)(calixarene) complex Figure 3.6 (right) This is further supported from the Stern o Figure 3.7 which indicates a static quenching mechanism as evidenced by the fact that the lifetimes do not decrease after all of the cavitand is copper bound, indicating that diffusional quenching is not predominant here The lifetimes were obtained ( Table 3.2 ) from first order exponential decays of the time re solved emission for the calixresorcinarene based cavitand as well as for the copper coordinated cavitand, shown in Figure 3.8 For the calixresorcinarene, the lifetime was determined to be 4.28 0.04 ns and decreased slightly for stochiometric add itions of copper (II) chloride up to 1.5 (to 3.59 0.05 ns at a 1:1 ratio) after which no change in the lifetime was observed This lifetime is consistent with methoxybenzene derivatives, as the lifetimes of 1,4 dimethoxybenzene and deprotonated 5OH BD C are 2.7 ns (in ethanol) and ~ 2 ns (methanol), respectively(Berlman, 1971; Larsen, 2007). The slightly longer lifetime of the calixresorcinarene based cavitand may be explained by small changes to the dipole moment that arise when the benzene subunits are conjugated together.
75 Figure 3.6 Steady state (left) absorption and (right) emission spectra for the titration of Cu(II)Cl 2 into ~ 30uM calixresorcinarene. The concentration of Cu(II)Cl 2 ranged from 0 to 150 uM. Figure 3.7 0 Stern Volmer plot for the titration of Cu(II)Cl 2 into ~20 uM cavitand. [Cu(II)Cl 2 ] ranged from 0 to 60 uM. The data was fit to a line and was found to have a y intercept= 1.05 0.06, and slope = 0.006 0.002 uM 1 The linearity of the data suggests that no diffusional quenching is occurring here. Figure 3.8 The time resolved emission spectra for ~20 uM calixarene and Cu(II)Cl 2 [Cu(II)Cl 2 ] ranged from 0 to 60 uM. Lifetimes, Table 3.2, were obtained using first order exponential deca ys: y = y 0 + A 1 *exp( (x x 0 )/ 1 ).
76 Cu(II)Cl 2 :cavitand Ratio Lifetime (ns) 0 4.28 0.04 0.5 4.09 0.06 1 3.59 0.05 1.5 3.14 0.05 2 3.25 0.05 2.5 3.19 0.05 3 3.28 0.06 Fluorescence polarization studies of the calixresorcinarene based cavitand show that the polarization does not change from the free ligand to bound complex, P= 0.027 0.004 for the free ligand, P= 0.026 0.009for the bound complex (at a 1:1 stochiometric ratio of Cu(II) (OAc) 2 to calixresorcinarene). In addition, the polarization is similar to that report ed for deprotonated [ 5OH BDC] 2 with P= 0.038 0.004 (Larsen, 2007). In general, polarization is dependent on the ratio of the fluorescent lifetime of the molecule to the rotational correlation time as shown in Equation (3.4) where the rotational correlation time is given by Equation (3.5) and is a measure of the r otational speed of the molecule, which is dependent on size and shape (Perrin, 1926 as referenced in Lakowicz, 2006) Thus, as suggested in Larsen (2007), the low polarization observed here is expected considering the long fluorescence lifetime of the molecule, ~ 4 ns. The lifetimes were obtained by the first exponential decay fit of the time resolved emission spectra from Figure 3.8 using OriginPro 8.0. Table 3.2 Lifetimes of Cu(calixarene) complexes
77 ( 3.4 ) Where P is polarization, P 0 the polarization t time f the fluorescence lifetime, and rot the rotational correlation time of the fluorophore. (3.5) rot V the molecular volume. Here, the molecule is presumed to be spherical. The decrease in the fluorescence lifetimes upon copper addition can be explained in terms of a decrease in the rotational correlation time of the complex, which is indicative of larger complex formation. Using the Perrin equation, in which the polarization ratio of C u(calixresorcinarene) to the free ligand is 1(P Cucalix /P calix = 1), the rotational and fluorescence lifetimes are related by Equation (3.6) ( 3. 6 ) Where the subscripts Cucalix and calix refer to the metal bound and free ligand respectively. The change in the rotational correlation time is additional evidence of the copper complexation of calixresorcinarene. Large errors in the polarization arise at higher concentrations of Cu(II)(OAc) 2 The polari zation as a function of Cu(II) (OAc) 2 to calixresorcinarene ratio is shown in
78 Figure 3.9 These errors are the result of inner filter effects, which unfortunately cannot be avoided d espite exciting the calixarene at a longer wavelength than lambda max and the use of Cu(II) (OAc) 2 that absorbs at lower wavelengths Q uenched emission intensities arising from the inner filter effects approach the resolution of the instrument and hence larger standard errors result. However, as the binding of copper to the calixresorcinarene based cavitand is ~ 1 1 5 to 1 (discussed below), the errors at concentrations above this ratio are presumably insignificant. Relative c hanges in the emission intensit y were plotted as a function of [ calixarene ]/[total], in which the total concentration is the sum of [calix(4)arene] and [Cu(II)Cl 2 ], Figure 3.10 The curve indicates a 1:1 5 bin ding stoichiometry of the calixarene to Cu (II) Similar studies of copper chelation to calixarene bearing a naphthyl 5 isoxazolyl based functional group show 1:1 binding, in ad dition to a quenched fluorescence by the addition of copper perchlorate (Senthilvelan, 2009). In this study, the Figure 3.9 Polarization as a function of [Cu(II)(OAc) 2 ] for ~ 120 uM cavitand.
79 dissociation constant in acetonitrile was determined to be 4.8 x 10 4 M (480 uM) which is directly comparable to the dissociation constant determined for copper binding to the novel resorcinarene based cavitand of the current study as discussed below. T h e h i g h e r b i n d i n g s t o c h i o m e t r y h e r e i n r e g a r d s t o c o p p e r b i n d i n g i s p r e s u m a b l y d u e t o t h e e r r o r i n e m i s s i o n i n t e n s i t i e s f r o m i n n e r f i l t e r e f f e c t s aforementioned The o Volmer plot shows that the quenching mechanism is from non fluorescent complex formation. When quenching is assumed to be primarily from non fluorescent complex formation (static quenching), the slope of the line for the I O /I versus [CuCl 2 ] plot is the association constant, Equation (3.7) ( 3.7 ) Figure 3.11 shows the steady state Stern Volmer plot and the associated linear fit. The slope was determined to be K A = 0.07 0.01 uM 1 (7 x 10 4 1 x 10 4 M 1 ), corresponding to a dissociation constant of 14 2 uM. The assumption of a static quenching mechanism yields a dissociation constant that agrees fairly well with that determined by e primary mode of quenching present here. Again, this is further evidenced by the fact that higher concentrations of Cu(II)Cl 2 do not affect the fluorescence lifetimes as would be the case in diffusional quenching.
80 The binding affinity of the cupric ion to calix[ 4]resorcinarene is based on steady state emission changes. The K D is determined to be 8.6 0.2 uM using Equation (2) Figure 3.12 Equation ( 2 ) (see Methods section) represents a modified Hill plot that indicates a slight degree of cooperativity with n= 1.63 0.06 However, when fit to n=1, the K D does not change significantly (K D = 8.0 0.7 uM). Figure 3.12 also indicates that quenching of the calixarene is ~90% complete at a ratio of 1:1 Cu:calixarene. The dissociation constant s obtained from the modified Hill plot as well as from the steady state Stern Volmer plot are similar supporting the low dissociation constant. Figure 3.10 Job plot for calixar ene and Cu(II)Cl 2 The total concentration ( calixarene + Cu(II)Cl 2 )was kept fixed at 50 uM. The data was fit b y s o l v i n g f o r x i n t w o l i n e a r e q u a t i o n s s e t e q u a l t o e a c h o t h e r a t y X w a s d e t e r m i n e d t o b e 0.5 9 0.0 9 This indicates 1:1 binding. Figure 3. 11 The steady state I/I 0 Stern Volmer plot for the titration of Cu(II)Cl 2 into ~20 uM cavitand. [Cu(II)Cl 2 ] ranged from 0 to 60 uM. The data was fit to a line and was found to have a y intercept= 1.6 0.4, slope = 0.07 0.01 uM 1 This can be related to the dissociation constant via the following: I/I o = 1 + K S [Q], with K s = K A = 0.07 0.01 uM 1 giving a K D =14 2 uM
81 The aforementioned study of naphthyl 5 isoxazolyl functionalized calix arene showed a copper (II) dissociation constant of K D = 480 uM (4.8 x 10 4 M) (Senthilvelan, 2009). In this study, the copper (II) was found to be hexa coordinate with the nitrogen and oxygen from each of the two naphthyl 5 isoxazolyl groups and the two phenolic oxygens along the lower rim. The presence of a MLCT band characteristic of a cuprous species showed that the phenolic hydroxyl groups reduced the copper (II) ion and locked the copper (I) ion in place. Other studies have indicated a similar effect, in which the relatively air unstable cuprous ion can be stabilized by the presence of nitrogen bearing ligands ( Rondelez et al., 2002 ). Though the reduction of the copper (II) is not proposed here, there is precedence for the stabilization of the cuprous ion and therefore, the Figure 3.12 The fraction bound plot for calixresorcinarene, showing the relative change in the intensity as a function of [Cu(II)Cl 2 ]. The data was fit to Equation (2), see Methods, in which the K D was found to be 8.6 0.02 uM, n= 1.63 0.06, and max = 0.797 0.004.
82 resorcinarene based cavitand is a potential spectroscopic probe for copper complexation. Copper (II) may exist in a variety of geometries and coordinations including square planar, trigonal bipyrami dal, and octahedral, with the most common coordination of the cupric ion being a distorted tetrahedral, in which the four bonds are in 1 plane (Wilks, 1984). Calixresorcinarene is a tetradentate ligand, with the pi electrons of the four nitrogen s in th e methyl imidazole bearing groups representing sites of coordination for the copper (II) ion. Molecular modeling of Cu(calixresorcinarene) affirms that the four methylimidazole groups coordinate the cupric ion in a tetrahedron geometry, with the methyl imidazole groups in one plane and the copper slightly below plane Figure 3.13 Even though t he possibility of a fifth ligand cannot be ruled out here the existence of a sixth ligand is prohibited on the basis of cavitand. A fifth ligand, if it does indeed exist, however, does not appear to change the spectroscopic properties of the Cu(calixresorcinarene) as the steady state absorption and emission spectra are similar for the titration of Cu(II)(OAc) 2 and Cu(II )Cl 2 into calixresorcinarene and the counterions are presumably the coordinating ligands. This would suggest that the copper exists in four coordinate geometry.
83 Figure 3.13 Geometry optimized Cu(calixresorcinarene). Optimization was done using HyperChem 8.0, with a semi empirical PM3 force field, treating the atoms as closed shell (RHF). The total charge on the complex was +2, representing the cupric ion (spin multiplic ity of 2). Geometry optimizations were performed using the Polak Ribiere (conjugate gradient) algorithm with a RMS gradient of 0.01 kcal mol 1 1 The C 6 H 12 alkyl chains were replaced with hydrogen atoms for efficiency.
84 Conclusion Calixresorcinarene is a cyclic oligomer of functionalized alkoxybenzene monomers, whose individual photophysical properties are responsible for the spectroscopic properties of the cavitand * transitions at 290 nm and 280 nm, respectively, characteristic of alkoxybenzene The cupric ion is found to bind the novel calixresorcinarene based cavitand tightly in a 1:1 stochiometry, with a dissociation constant of ~ 8 uM. The four nitrogen bearing groups on the upper rim of the cavitand represent the four ligands of the copper ion, coordinating in a semi flattened tetrahedron geometry. The copper complexation of calixarenes represents a novel way to model copper sites in biological molecules because of their versatility and ability to coordinate metals retaining open ligand coordination sites
85 References 1. Amrhein, P.; Wash, P. L.; Shivanyuk, A.; Jr., J. R., Metal Ligation Regulates Conformational Equilibria and Binding Properties of Cavitands. Organic Letters 2002, 4, (3), 319 321. 2. Atwood, J. L.; Orr, G. W.; Juneja, R. K.; Bott, S. G.; Hamada, F., Supramolecular assemblies based on calixarenes. Pure & Appl. Chem. 1993, 65, (7), 1471 1476. 3. Beer, P. D., Transition Metal Receptor Systems for the Selective Recognition and Sensing of Anionic Guest Species. Acc. Chem. Res. 1998, 31, 71 80. 4. Berlman, I. B., Handbook of Fluorescence Spectra of Aromatic Molecules 2nd ed.; Academic Press: New York and London, 1971. 5. Blanchard, S.; Clainche, L. L.; Rager, M. N.; Chansou, B.; Tuchagues, J. P.; Duprat, A. F.; Mest, Y. L.; Re i naud, O., Calixarene Based Copper(I) Complexes as Models for Monocopper Sites in Enzymes. Angew. Chem. Int. Ed. 1998, 37, (19) 2732 2735. 6. Blanchard, S.; Rager, M. N.; Duprat, A. F.; Renaud, O., A novel calixarene based mononuclear copper(I) complex that exhibits chirality at low temperature. New J. Chem. 1998 1143 1146. 7. Bhmer, V., Calixarenes, Macrocycles with (Almost ) Unlimited Possibilities. Angew. Chem. Int. Ed. Engl. 1995, 34, 713 745. 8. Bonomo, R. P.; D'Alessandro, F.; Grasso, G.; Impellizzeri, G.; Pappalardo, G.; Rizzarelli, E.; Tabb, G., Carcinine cyclodextrin derivatives as scavenger entities of OH radicals and SOD like properties of their copper(II) complexes. Inorganica Chimica Acta 2008, 361, 1705 1714. 9. Breslow, R., Biomimetic Chemistry. Pure & Appl. Chem. 1994, 66, (8), 1573 1582. 10. Breslow, R., Biomimetic Chemistry and Artificial Enzymes: Catalysis by Design. Acc. Chem. Res. 1995, 28, 146 153. 11. Breslow, R., Biomimetic Chemistry: Biology as an Inspiration. J. Biol. Chem. 2009, 284, (3), 1337 1342. 12. Cacciapaglia, R.; Casnati, A.; Mandolini, L.; Reinhoudt, D. N.; Salvio, R.; Sartori, A.; Ungaro, R., Catalysis of Diribonucleoside Monophosphate Cleavage by Water Soluble Copper(II) Complexes of Calixarene Based Nitrogen Ligands. J. Am. Chem. Soc. 2006, 128, 12322 12330. 13. Cao, Y. D.; Zheng, Q. Y.; Chen, C. F.; Hu, H. M.; Huang, Z. T., Synthesis of a novel three coordinate copper(I) complex: a structural mimic of the reduced form of type 1 site in copper protein. Inorganica Chimica Acta 2004, 357, 316 320. 14. Choi, J. K.; Kim, S. H.; Yoon, J.; Lee, K. H.; Bartsch, R. A.; Seung Kim, J., A PCT Base d, Pyrene Armed Calixcrown Fluoroionophore. J. Org. Chem 2006, 71, 8011 8015. 15. Clainche, L. L.; Giorgi, M.; Renaud, O., Novel Biomimetic Calixarene Based Copper(II) Complexes. Inorg. Chem. 2000, 39, 3436 3437.
86 16. Collman, J. P.; Decrau, R. A., F unctional biomimetic models for the active site in the respiratory enzyme cytochrome c oxidase. Chem. Commun. 2008 5065 5076. 17. Cram, D. J.; Karbach, S.; Kim, H. E.; Knobler, C. B.; Maverick, E. F.; Ericson, J. L.; Helgeson, R. C., Host Guest Complexati on. 46. Cavitands as Open Molecular Vessels Form Solvates. J. Am. Chem. Soc. 1988, 110, (7), 2229 2237. 18. DeGrado, W. F., Introduction: Protein Design. Chem. Rev. 2001, 101, (10), 3025 2026. 19. orrelation between the metal ion type, coordination number and the amino acid residues involved in the coordination Acta Cryst. 2008, D64, 257 263. 20. Fochi, F.; Jacopozzi, P.; Wegelius, E.; Rissanen, K.; Cozzini, P.; Marastoni, E.; Fisicaro, E.; Manini, P.; Fokkens, R.; Dalcanale, E., Self Assembly and Anion Encapsulation Properties of Cavitand Based Coordination Cages. J. Am. Chem. Soc. 2001, 123, 7539 7552. 21. Fox, O. D.; Dalley, N. K.; Harrison, R. G., A Metal Assembled, pH Dependent, Resorcinarene Based Cage Molecule. J. Am. Chem. Soc. 1998, 1998, (120), 7111 7112. 22. Gembus, A.; Corzilius, B.; Eichel, R. A.; Dinse, K. P.; Immel, S.; Stumm, D.; Fla uaus, M.; Plenio, H., Electron Paramagnetic Resonance Structure Investigation of Copper Complexation in a Hemicarcerand. J. Phys. Chem B 2006, 110, 15012 15020. 23. Gustche, C. D.; Nam, K. C., Calixarenes. 22. Synthesis, Properties, and Metal Complexation of Aminocalixarenes. J. Am. Chem. Soc. 1988, 110, 6153 6162. 24. Gutsche, C. D., Calixarenes. Acc. Chem. Res. 1983, 16, 161 170. 25. Gutsche, C. D., Calixarenes The Royal Society of Chemistry: St. Louis, U.S.A, 1989; Vol. 26. Gutsche, C. D.; Bauer, L. J., Calixarenes. 13. The conformational properties of calixarenes, calixarenes, calixarenes, and oxacalixarenes. J. Am. Chem. Soc. 1985, 107, 6052 6059. 27. Gutsche, C. D.; Muthukrishnan, R., Calixarenes. 1. Analysis of the Product Mixtures Pr oduced by the Base Catalyzed Condensation of Formaldehyde with Para Substituted Phenols. J. Org. Chem 1978, 43, (25), 4905 4906. 28. Harrison, R. G.; Fox, O. D.; Meng, M. O.; Dalley, N. K.; Barbour, L. J., Cation Control of Pore and Channel Size in Cage Based Metal Organic Porous Material. Inorg. Chem. 2002, 41, 838 843. 29. Hooley, R. J.; Anda, H. J. V.; Jr., J. R., Cavitands with Revo lving Doors Regulate Binding Selectivities and Rates in Water. J. Am. Chem. Soc. 2006, 128, 3894 3895. 30. Hooley, R. J.; Shenoy, S. R.; Jr., J. R., Electronic and Steric Effects in Binding of Deep Cavitands. Organic Letters 2008, 10, (23), 5397 5400. 31. Ihm, C.; Park, Y.; Paek, K., Small cavitands specifically binding a water molecule. Organic Letters 2004, 6, (3), 369 372. 32. uence of Solvent Polarity and Metal Ions. Tetrahedron 1993, 49, (44), 9937 9946.
87 33. Izzet, G.; Zeng, X.; Over, D.; Douziech, B.; Zeitouny, J.; Giorgi, M.; Jabin, I.; Mest, Y. L.; Reinaud, O., First Insights into the Electronic Properties of a Cu(II) Cente r Embedded in the PN 3 Cap of a Calixarene Based Ligand. Inorg. Chem. 2007, 46, 375 377. 34. Ji, H. F.; Dabestani, R.; Brown, G. M., A Supramolecular Fluorescent Probe, Activated by Protons To Detect Cesium and Potassium Ions, Mimics the Function of a Lo gic Gate. J. Am. Chem. Soc. 2000, 122, 9306 9307. 35. Kmmerer, H.; Happel, G., Stepwise Synthesis of a Cycloheptamer from Para Cresol, 4 Tert Butylphenol and Formaldehyde Comparison with a Phenolic Heptanuclear Chain Oligomer. Makromolekulare Chemie Mac romolecular Chemistry and Physics 1980, 181, (10), 2049 2062. 36. Kmmerer, H.; Happel, G., Stepwise Syntheses of O,O' Methylene Bridged Cyclohexamers with Para Cresol and 4 Tert Butyl Phenol Units Comparison with Similarly Structured Hexanuclear Chain O ligomers. Monatshefte Fur Chemie 1981, 112, (6 7), 759 768. 37. Kammerer, H.; Happel, G.; Mathiasch, B., Stepwise Syntheses and Properties of Some Cyclopentamers of Methylene Bridged 5 Alkyl 2 Hydroxy 1,3 Phenylene Units. Makromolekulare Chemie Macromolecu lar Chemistry and Physics 1981, 182, (6), 1685 1694. 38. Kim, J. S.; Quang, D. T., Calixarene Derived Fluorescent Probes. Chem. Rev. 2007, 107, 3780 3799. 39. Kim, S. K.; Lee, S. H.; Lee, J. Y.; Lee, J. Y.; Bartsch, R. S.; Kim, J. S., An Excimer Based, Bin uclear, On Off Switchable Calixcrown Chemosensor. J. Am. Chem. Soc. 2004, 126, 16499 16506. 40. Kitajima, N.; Moro oka, Y., Copper Dioxygen Complexes. Inorganic and Bioinorganic Perspectives. Chem. Rev. 1994, 94, ( ), 737 757. 41. Lakowicz, J. R., Principles of Fluorescence Spectroscopy 3rd ed.; Springer: New York, NY, 2006. 42. Larsen, R. W.; McManus, G. J.; IV, J. J. P.; Rivera Otero, E.; Zaworotko, M. J., Spectroscopic Characterization of Hydroxylated Nanoballs in Methanol. Inorg. Chem. 2007, 46 5904 5910. 43. heme proteins. Coord. Chem. Rev. 2007, 251, 1101 1127. 44. Lenthall, J. T.; Steed, J. W., Organometallic cavitands: Cation anion binding via metallation. Coord. Chem. Rev. 2007 45. Linton, B.; Hamilton, A. D., Formation of Artificial Receptors by Metal Templated Self Assembly. Chem. Rev. 1997, 97, 1669 1680. 46. Lled, A.; Hooley, R. J.; Jr., J. R., Recognition of Guests by Water Stabilized Cavitand Hosts. Organic Letters 2008, 10, (17), 3669 3671. 47. Lu, Y.; Berry, S. M.; Pfister, T. D., Engineering Novel Metalloproteins: Design of Metal Binding Sites into Native Protein Scaffolds. Chem. Rev. 2001, 101, 3047 3080. 48. Mallier, P.; Gudefin A.; E hlinger N.; Perrin, M., Photodegradation of Calixarenes. Dyes and Pigments 1996, 31, (1), 13 17.
88 49. Megyesi, M.; Biczk, L., Considerable Change of Fluorescence Properties upon Multiple Binding of Coralyne to 4 Sulfanatocalixarenes. J. Phys. Chem B 2010, 114, 2814 2819. 50. Mtivier, R.; Leray, I.; Valeur, B., Photophysics of calixarenes bearing two or four dansyl fluorophores: charge, proton and energy transfers. Photochem. Photobiol. Sci. 2004, 3, 374 380. 51. Moran, J. R.; Karbach, S.; Cram, D. J., Cavitands: Synthetic Molecular Vessels. J. Am. Chem. Soc. 1982, 104, 5826 5828. 52. Morris, J. V.; Mahaney, M. A.; Huber, J. R., Fluorescence Quantum Yield Determinations. 9,10 Diphenylanthracene as a Reference Standard in Different Solvents. Journal of Physical Chemistry 1976, 80, (9), 969 974. 53. Namor, A. F. D. d.; Cleverley, R. M.; Zapata Ormachea, M. L., Thermodynamics of Calixarene Chemistry. Chem. Rev. 1998, 98, 2495 2525. 54. Neri, P.; Consoli, G. M. L.; Cunsolo, F.; Geraci, C.; Piattelli, M., Chemistry of Larger Calix[n]arenes (n=7,8,9). In Calixarenes 2001 Asfari, Z., Ed. Kluwer Academic Publishers: Dordrecht, Netherlands, 2001; pp 89 109. 55. Nishizawa, S.; Kato, Y.; Terama e, N., Fluorescence Sensing of Anions via Intramolecular Excimer Formation in a Pyrophosphate Induced Self Assembly of a Pyrene Functionalized Guanidinium Receptor. J. Am. Chem. Soc. 1999, 121, 9463 9464. 56. Ohseto, F.; Yamamoto, H.; Matsumoto, H.; Shinka i, S., Allosteric Communication between the Metal binding Lower Rim and the Sugar binding Upper Rim on a Calixcrown Platform. Tetrahedron Letters 1995, 36, (38), 6911 6914. 57. Over, D.; Lande, A. d. l.; Zeng, X.; Parisel, O.; Renaud, O., Replacement of a Nitrogen by a Phosphorus Donor in Biomimetic Copper Complexes: a Surprising and Informative Case Study with Calixarene Based Cryptands. Inorg. Chem. 2009, 48, 4317 4330. 58. Parulekar, S. N.; Muppalla, K.; Fronczek, F. R.; Bisht, K. S., Synthesis of resorcinarene cavitands by ring closing metathesis. Chem. Commun. 2007 4901 4903. 59. Perrin, F., Polarisation de la lumire de fluorescence. Vie moyenne des molcules dans l'etat excit. Jour de Phys, Vleme Srie 1926, 7, 390 401. 60. Prodi, L.; Boll etta, F.; Montalti, M.; Zaccheroni, N.; Casnati, A.; Sansone, F.; Ungaro, R., Photophysics of 1,3 alternate calixarene crowns and of their metal ion complexes: evidence for cation New J. Chem. 2000, 24, 155 158. 61. Psychogios, N.; Regnouf de Vains, J. B., A new water soluble calixarene podand incorporating p sulphonate groups and 2,2' bipyridine chelating units. Tetrahedron Letters 2002, 43, 77 80. 62. Purse, B. W.; Jr., J. R., Functional cavitands: Chemical react ivity in structured environments. PNAS 2005, 102, (31), 10777 10782. 63. Rondelez, Y.; Bertho, G.; Renaud, O., The First Water Soluble Copper(I) Calixarene Complex Presenting a Hydrophobic Ligand Binding Pocket: A Remarkable Model for Active Sites in Me talloenzymes. Angew. Chem. 2002, 114, (6), 1086 1088.
89 64. Rondelez, Y.; Rager, M. N.; Duprat, A.; Renaud, O., Calixarene Based Cuprous "Funnel Complexes": A Mimic for the Substrate Access Channel to Metalloenzyme Active Sites. J. Am. Chem. Soc. 2002, 12 4, (7), 1334 1340. 65. Rondelez, Y.; Snque, O.; Rager, M. N.; Duprat, A. F.; Renaud, O., Biomimetic Copper(I) CO Complexes: A Structural and Dynamic Study of a Calixarene Based Supramolecular System. Chem. Eur. J. 2000, 6, (22), 4218 4226. 66. Sanders J. K. M., Supramolecular Catalysis in Transition. Chem. Eur. J. 1998, 4, (8), 1378 1383. 67. Schazmann, B.; Alhashimy, N.; Diamond, D., Chloride Selective Calixarene Optical Sensor Combining Urea Functionality with Pyrene Excimer Transduction. J. Am. Chem. Soc. 2006, 128, 8607 8614. 68. Schrder, K.; Enthaler, S.; Bitterlich, B.; Schulz, T.; Spannenberg, A.; Tse, M. K.; Junge, K.; Beller, M., Design of and Mechanistic Studies on a Biomimetic Iron Imidazole Catalyst System for Epoxidation Olefins with H ydrogen Peroxide. Chem. Eur. J. 2009, 15, 5471 5481. 69. Snque, O.; Campion, M.; Douziech, B.; Giorgi, M.; Rivire, E.; Journaux, Y.; Mest, Y. L.; Renaud, O., Supramolecular Assembly with Calixarene and Copper Ions Formation of a Novel Tetranuclear C ore Exhibiting Unusual Redox Properties and Catecholase Activity. Eur. J. Inorg. Chem. 2002 2007 2014. 70. Senthilvelan, A.; Ho, I. T.; Chang, K. C.; Lee, G. H.; Liu, Y. H.; Chung, W. S., cooperative r ecognition of a copper cation and anion by a c alixa rene substituted at the lower r im by a a mino u nsaturate d k etone. Chem. Eur. J. 2009, 15, 6152 6160. 71. partial c isomerism in t etramethoxy p t butylcalixarene. novel s olv ent effects an d metal template e ffects. Chem. Lett. 1990, 19, (8), 1263 1266. 72. Shinkai, S.; Mori, S.; Koreishi, H.; Tsubaki, T.; Manabe, O., hexasulfonated c alixare ne derivatives: A new class of c a talysts, surfactants, and host m olecules. J. Am. Chem. Soc. 1986, 1 08, 2409 2416. 73. Sigman, J. A.; Kwok, B. C.; Lu, Y., From myoglobin to heme copper oxidase: design and e ngineering of a CuB Center into sperm whale m yoglobin. J. Am. Chem. Soc. 2000, 122, (34), 8192 8196. 74. Sliwa, W.; Deska, M., Cavitands (Review). Che mistry of Heterocyclic Compounds 2002, 38, (6), 646 667. 75. Sorrell, T. N.; Malachowski, M. R.; Jameson, D. L., Synthesis, structure, and reactivity of a b inucl ear three coordinate copper(I) c omplex. Inorg. Chem. 1982, 21, 3250 3252. 76. Sorrell, T. N.; Pigge, F. C.; White, P. S., C alixresorcinarenes as ligands: s y nthesis and c haracterizatio n of transition metal cavitand c omplexes. Inorg. Chem. 1994, 33, 632 635. 77. Verboom, W., Cavitands. In Calixarenes 2001 Asfari, Z., Ed. Kluwer Academic Publishers: Dordrecht, Netherlands, 2001; pp 181 198. 78. Vysotsky, M. O.; Mogck, O.; Rudzevich, Y.; Shivanyuk, A.; Bhmer, V.; Brody, M. S.; Cho, Y. L.; Rudkevic h, D. M.; Jr., J. R., enhanced t hermody namic and kinetic stability of c alixa rene dimers locked in the c one c onformation. J. Org. Chem 2004, 69, 6115 6120.
90 79. Wang, Y.; DuBois, J. L.; Hedman, B.; Hodgson, K. O.; Stack, T. D. P., Catal ytic galactose oxidase models: b iom imetic Cu(II) phenoxyl radical r eactivity. Science 1998, 279, 537 540. 80. Wells, A. F., Structural Inorganic Chemistry 5th ed.; Oxford University Press: New York, New York, 1984. 81. Williams, A. T. R.; Winfield, S. A.; Miller, J. N., Relative fluorescence quantum yields using a computer controlled luminescence spectrometer. Analyst 1983, 10 8, 1067 1071. 82. Witte, T.; Decker, B.; Mattay, J.; Huber, K., Formation of branched calixarene aggregates a time r e solved static Light scattering s tudy. J. Am. Chem. Soc. 2004, 126, 9276 9282. 83. Xie, D.; Gutsche, C. D., Synthesis and r eactivity of cali xarene based copper c omplexes. J. Org. Chem 1998, 63, 9270 9278. 84. Xu, W.; Vittal, J. J.; Puddephatt, R. J., Anionic calixarene complexes of c op per(I) and silver(I) as cation r eceptors. Inorg. Chem. 1997, 36, 86 94. 85. Zinke, A.; Kretz, R.; Leggewie, E.; Hssinger, K., Monatsch Chem. 1952, 83. 86. Zinke, A.; Ziegler, E., Ber. 1941, 74.
91 SUMMARY Despite significant progress in understanding heme proteins, small ligand diffusion from the heme is still not fully understood. Myoglobin is a small, soluble heme protein that has been utilized as a model for larger, more complex heme proteins because of the commonality that all heme proteins have in terms of their metal site structures. Myoglobin carries oxygen in muscle cells and has been extensively characterized using a variety of methods. Much has been learned about ligand escape in myoglobin. For instance, it is well known that protein fluctuations are involved in exposing a pathway from the heme to the solvent. Though the exact mechanism of the protein fluctuations are under debate, namely the relative importance of His64 gati ng or hydrophobic xenon binding cavities, there is no doubt that these processes play a pivotal role. In this study, probing the role of water entry into the protein distal pocket was explored by studying CO escape in a heavy water buffer. The dynamics o f water entry would be expected to differ from that of normal water buffer if water entry was indeed a rate determining step in the diffusion of CO from the protein matrix. It was found that this is not the case as the lifetimes for CO escape in both case s were similar; however, that CO escape in D 2 O was significantly more endothermic as a result of the hydrogen bonding environment differences in the two buffers. The results point to the importance of His64 in the escape of CO, as this residue is highly d ependent on the electros tatics within the distal pocket and whose swinging is a likely source of the energetic differences.
92 Structure/function studies of site directed myoglobin mutants has been a method to probe the role of specific residues within the he me pocket. In so doing, it has been discovered that simple modifications to existing protein scaffolds can have dramatic learned about heme proteins via the introduction of novel functionality into ex iting protein frameworks through site directed mutations. The second model system, an L29H/F43H site directed engineered sperm whale myoglobin (Cu B Mb), represents such a protein, in which the double mutant when bound with copper is an ideal model system f or heme copper proteins. It is important in these model systems to understand the effects of mutations, specifically ones known to introduce novel functionality, as a potential for generating better model proteins. Accordingly, it was found that the L29H and F43H mutations enlarge the heme pocket cavity, giving the CO a more direct pathway to the solvent that decreases the lifetime for CO escape. The results here too point toward the importance of His64 in the mechanism of ligand escape, as the residue i s found open Furthermore, the importance of the electrostatic environment within the heme pocket is also reinforced as the increased polarity of the pocket is responsible for additional waters to be present in the pocket. The third model system, a novel calixresorcinarene represents the pinnacle of what metalloprotein engineering aims to do, the rational design of novel metal binding sites for specific functionalities With this aim in mind, copper binding to calixresorcinarene was photophysically characterized as a potential model for the transient binding of small molecules in heme copper proteins. Calixresorcinarene is found to bind copper (II) in a four coordinate flattened tetrahedron geometry, similar to the that in type II copper proteins in a 1:1 binding stoichiometry. Its bulk photophysical
93 properties are analogous to those of its monomers, alkoxybenzene, suggesting that the calixresorcinarene as a whole does not change the electronic and photophysical properties upon complexation an d therefore represents an ideal active site mimic. As has been demonstrated, much progress has been toward the ultimate goal of elucidating the mechanism underlying ligand diffusion in heme proteins. Similarly, great strides have been made in protein engi neering in the de novo design of proteins and active sites. Though the work presented here does not claim to do either, the results have hopefully contributed to the realization of these aims in some insightful way.
94 APPENDIX 1 PHOTOACOUSTIC CALORIMETRY
95 Theory In photoacoustic calorimetry (PAC), the acoustic wave generated is a convolution of the first order exponential decays of all the resolvable kinetic events occurring with an instrument response function, Equation (A1.1). Kinetic events occurring within 50 ns to can be resolved as this is the time window of the instrument. (A1.1 ) F(t) represents the instrument response function, i is the amplitude of that phase (a i is the lifetime of that phase. For reactions involving multiple kinetic events, the individual amplitudes and lifetimes are extracted u tili zing a Simplex parameter estimation algorithm developed in the lab oratory (e.g. see Larsen, 2010) convolution of a calibration compound acoustic trace (represents F(t) ) with first order exponential decays of the even ts, where initial estimates of the lifetimes and amplitudes are utilized. The resulting convoluted acoustic wave is compared to that of the sample process is repeated acoustic wave are determined to be close in frequency and amplitude. In general, the decay of a photoexcited molecule to the ground state can occur via nonradiative decay or fluorescence or un dergo intersystem crossing and subsequent
96 phosphorescence, as summarized in the Jablonski Perrin diagram, Figure A1.1 (left). Solvent volume changes in photoacoustic calorimetry have a thermal and non thermal origin, Equation (A1.2), corresponding to hea t dissipated via nonradiative decay and heat released/consumed from any photochemistry (e.g. spin state changes, conformational changes, electrostriction, bond cleavage), Figure A1.1 (right). Resolving the magnitude of the volume change from the photochem istry involves the use of a calibration compound that must have no photochemistry occurring, in which the volume change has only thermal origins. The solvent volume response to heat is related to the isothermal 1 ), density of capacity (C p in cal g 1 K 1 ). Thus the signal of the calibration compound (R) can be expressed as in Equation (A1.3), where the heat (Q) for the reference is equivalent to the energy of the incident photon, E set up, and E a is the number of Einsteins absorbed. ( A1.2 ) (A1.3) The signal from the sample contains both thermal and nonthermal contributions to i,thermal i,photochemistry respectively) and is expressed i is the quantum yield of the process.
97 ( A1.4) The subscript denotes the phase of the kinetic event, with each phase contributing to the total volume change via individual nonthermal (photochemical) volume changes. Substitution of the thermal volume change expression, as bef ore, yields Equation (A1.5). (A1.5 ) The calibration compound is used to quantify the heat and volume changes that are separate from the photochemistry. A ratio of the sample and reference signals eliminates terms that are dependent on the instrument (viz. a ). This ratio is obtained either from the amplitude determined in the Simplex parameter estimation ( multiphasic processes) or from a ratio of the acoustic wave amplitudes (monophasic processes). Plotting the amplitude as a function of C p the heat (Q) and the change in volume due to photochemistry as the y intercept and slope, respectively; where the C p ( A1.6 ) The change in the volume from pho tochemistry is obtained by dividing the slope by the quantum yield for all phases. For the initial (prompt) phase occurring within the time resolution of the instrument, the magnitude of the enthalpy change is equivalent to
98 the difference of the incident photon energy (E ) and the heat energy released (Q) scaled phase is the opposite sign of the heat released scaled by the quantum yield, (A1.8). ( A1.7 ) ( A1.8 ) Figure A1.1 (left) Jablonski Perrin diagram showing the various transitions that occur between energy levels and their associated processes. (right) A modified Jablonski Perrin diagram showing the transitions specific to photoacou stic calorimetry (highlighted in yellow). The lettering is as follows: A=absorption, F=fluorescence, IC=internal conversion, NR=non radiative relaxation, ISC=intersystem crossing, P=phosphorescence, S o = ground state, S 1,2,3 = first, second, third excited singlet state, T 1 =first excited triplet state.
99 Instrumentation The instrumentation of photoacoustic calorimetry (PAC) utilized in our laboratory reader is referred to Gensch and Viappiani (2003), Peters et al. (1991), and Braslvasky and Heibel (1992). Figure A1.2 shows the design currently in use here. The photochemistry is initiated using the second (532 nm) or third harmonic (355 nm) of a Q switched Nd:YAG laser (Co ntinuum Minilite I) with a 1 Hz pulse rate. The pulse length prevent multiphoton absorption. Samples are placed in a 1 cm by 1 cm cuvette, where contact between the cuvette and piezoelectric detector is made via a thin layer of vacuum grease. Pressure waves Figure A1.2 The current design in place for photoacoustic calorimetr y. The inset shows a close up of the cuvette holder and the piezoelectric crystal.
100 generated by the deposition of heat from photochemistry are converted to voltages via the compression of a piezoelectric crystal, as seen in A 1.1 inset. Voltages were amplified using a Panametrics ultrasonic pre amplifier (model 5662), digitized and reco rded by a Picoscope 3205 (50 MHz) oscilloscope Acoustic waves are recorded on a computer for further use. In general, absorbances are kept between 0.1 and 0.6.
101 References 1. Braslavsky, S. E.; Heibel, G. E., Time resolved photothermal and photoacoustics methods applied to photoinduced processes in solution. Chem. Rev. 1992, 92, 1381 1410. 2. Gensch, T.; Viappiani, C., Time resolved photothermal methods: accessing time resolved thermodynamics of photoinduced processes in chemistry and biology. Photochem. Photobiol. Sci. 2003, 2, 699 721. 3. Larsen, R. W., Lifetime distributions in photoacoustic calorimetry. In 54th Biophysical Society Annual Meeting, Abstract 2010. 4. to heme proteins. Coord. Chem. Rev. 2007, 251, 1101 1127. 5. Peters, K. S.; Watson, T.; Marr, K., Time Resolved Photoacoustic Calorimetry: A Study of Myoglobin and Rhodopsin. Annu. Rev. Biophys. Biophys. Chem. 1991, 20, 343 362.
ABOUT THE AUTHOR Meagan Small was born in Tampa, FL and received a B.S. Degree in Chemistry and Biology from the University of South Florida in 2007. During her undergraduate studies, she received both a USF Honors Scholarship and USF Presidential Scholarship and earned t he ACS Analytical Chemistry award in 2006. She was also awarded the USF Graduate Fellowship in Fall 2007 and Spring 2008 upon the start of her graduate career. From 2007 to the present, she has been a teaching and research assistant in the Department of Chemistry at University of South Florida, teaching such classes as General Chemistry lab as well as Physical Chemistry lab and weekly Physical Chemistry recitation sessions.