v} ' u 1 , 'I' , ‘ .I" I’ "V‘K: ‘ ‘ FVZI ". . ' "- -. ‘ i .‘23 54.0": " , 7‘ A . A ‘ Q- ‘.L:' '. , -, :2.“ ,V. ~; _ ' ‘. . ‘ . > I; v. ”:7 . . J,- , Q19. s V :‘1 -1 =. ... -. ~ - _ .. V- ._ .u . W. " :8 ,‘1 . -:--. ‘ 2.43:?“ V“ '91:. 3 Q 1‘1“". ‘ ‘ w 51,." ‘i " 134E . ‘3‘. r :cm--.~-H=; " ' ‘t: - , . , .. . ~ ”4!. V x 33:: ' n . . $ . . u & .. . "A :- ‘u x'.‘ ‘ ' , 'I . ‘ ' - > w “- 4. . '4‘ ' 1‘ ‘ ' ' L*‘-t-_Q“.'1“V,".'?::>’d' u. '1 . .zy ,' . . . ~ - - .. $313“. . f} ‘ WWW“??? . “J ‘v 't" .j. x . .- ' : : i1"? ‘ '. - ‘ Vv r.,~"’fi§:§‘?z“ " 33:. ' V .5?! .. ‘ V .. _..._. ‘ .: ' :‘D d. 4". ' $1} k Jafiqip'a)‘ ‘3 II . ‘. . ,-. 'ffi4. ' . .( ' .‘ti ‘ . - 4 é V" ‘I ’I't‘ ' I ‘ '. ' HIS" " I; 2 V ' .' ‘ ‘ r q- “45’ v‘\ - » - 1 ’ _-V‘,-‘ ‘ ‘ . ‘ 0m .3 win .. - i” . . ‘ 1 ; El Y ‘~‘~ >. «(”V‘t 'v: .. VJ '0’. {‘8 V ' V “G“ ‘ 132%“? "i ‘5 ‘ I f) ' ‘1‘ - Q 3: ‘ ‘2 H- r 312%“; - v - h 0‘1“}; ' . V t 'I“ .,,.,,.a_ 4' h“ 0" v.4“ 3k. ,. _. WWI-7" .f .345? ”.3 "a: “$3 _.:..:.'. gr; 'h‘» K . .1 .x', a 1k: : 2;? «3.12:: gziaV. ;, A: nail in: e- z . g: \-‘- ’4. 11 ‘ ' _ _;,I 1 . . ‘ '1» F L; hi i? ’ V Y? : ‘ - f’. ' r41 VI. "‘I‘ "V (war: rug-:2"... \. ‘l fn. ‘2; “an: s'si'em; ,x g '3‘: as r WV: 13.5. ,9». Par .- 9 , '7 : t: ‘ $.{:.f:.::;l;'j~- u r' “it {'13 V , I; ‘ H = . -. f; ' J 'f 1}: ‘f (”l r»? ” z 2m ,, mi: -~: "Jus ‘ . .- 35..-”...., .... .. :0 ‘mw V. IN-“ *‘a 7...“... -V..... a wu- Mn...- 3': «a - | n .p w ~nv- X “WJ; ...V. F M#U"‘t "V Um. ‘ r3“? “‘5'", ' n" . we. Dunc Hm; yum.- > n .‘m. an.» . ~. Inwmtwv a. v n. v .7 nt- —. .. «~— 4-4”! .......,.V r M‘d- 9‘ . 117‘ 5.. . _, . '5 DIV/V ":«x - TX .4522: .c . mam». .. “w... 2 1“ 2.1:". V--‘V._.-.. .V. VI“ lrlfl . V «.71 ~ II: V v - any? mu... 19px a v-«u I v..- ,2; m. «we» 3:» .33 ‘ ‘V Z": . Link“ . - I Al I ‘I I V” ~— ; .. ‘9 PF.‘ ¢ u» “ ‘ . .w» I mom. n- "any" I.- .— Writ-t 3‘ v.- a, 'u. m nah—“3" 3.... O.” 0“» r . "L 4- flaw « * ‘ ‘3'? a. ”m a?" " I‘ll w- a».-. vol-n . w.-.-. .‘ m .m- n I fiw-l «to. "94th' "aw .. 333': u:— ... 3 w. .42. V... .. Jig g u. \' I "I" “Ia-317.7. up u.- ‘ n “—24. ”2353-. , "1.3.-“ . 3:. _ ‘3: 3C 1r- » LN.— n... A. --v 1w— m-v ' ‘ . . «la-rm . . an .4"..- M.m%. . ~ v m.“ 4 ,9... < “i. {E l 4,2; A. 14?} 555 3 CHIGAN STATE UNIVE ESR SIT IIHIHHI Ill Ii" llHl ll llllllllHllllllllHll 3 1293 00592 3721 This is to certify that the dissertation entitled LIBRARY Michigan State University Kinetics of Porcine Carotid Artery Brain Isoform Creatine Kinase In Situ and In Vitro presented by Joseph F. Clark has been accepted towards fulfillment of the requirements for Ph.D. Physiology degree in W%flm Major professor Datef/gflm- (if, /999 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date duo. DATE DUE DATE DUE DATE DUE f ULi‘EEJq'ZL 13.33 “T u. JUN—349.95 MSU Is An Affirmative Action/Equal Opportunity Institution KINETICS OF PORCINE CAROTID ARTERY BRAIN ISOFORM CREATINE KINASE IN SITU AND IN VITRO By Joseph Floyd Clark A DISSERTATION Submitted to Michigan State University in partial fullfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1990 bumoog ABSTRACT KINETICS OF PORCINE CAROTID ARTERY BRAIN ISOFORM CREATINE KINASE IN SITU AND IN VITRO By Joseph Floyd Clark This thesis contains two major parts. The first part consists of the development of a purification procedure for the BB isoenzyme of creatine kinase (CK) from porcine carotid arteries and determination of its kinetics in solution. The second part involves performing sa- turation transfer experiments on perfused carotid arteries to deter- mine the kinetics of CK in the tissue. The purification of BBCK was accomplished using the procedure described by Wang and Cushman for human brain. Using the Cardiotrac method from Corning the activity of CK was determined for the samples and isoenzyme fractionation. The purification produced 97% BBCK with less than 3% MMCK and no detectable MBCK. During the purification it was found that porcine carotid artery contains approximately a l:l:l activity ratio of the CK isoenzymes BB, MB, and MM. This isoenzyme ratio had not been reported in other smooth muscle tissues studied to date. BBCK had an activity of 84 units per milligram of protein. SDS- PAGE electrophoresis showed one band demonstrating no significant protein impurities and a monomeric molecular weight of 42,500 Daltons for the BB isoenzyme. Saturation transfer was done on resting, isolated porcine carotid arteries which were perfused with physiological salt solution with 0.1 mM Pi. 31P NMR spectra were taken for all measurements. Exchange was observed between the phosphate peak of phosphocreatine and TATP. Results indicate (n-6), kf - 0.19 t 0.04 and k, - 0.12 z 0.03. These results were used to determine that the reaction was near equilibrium (0.71 t 0.11 as compared to one for equilibrium) as determined by the net flux ratio (the ratio of forward and reverse flux). The ATPase rate was found to be significantly less then the CK rate. I would like to dedicate this thesis to each and every member of my family. They have been my foundation, my strength, and support through all the long years it has taken for me to achieve this degree. To my mother and father who got me started and kept me going. It is by their example of hard work and dedication that I have been able to perserver and finally succeed. To my brother Jim, who is always there when I need him and always will be. To Kathy and Pete for always keeping me on my toes and a smile on my face. To Mommom and Poppop who have been there and always will be there with smiling faces and friendly advice. They are the people who have set the foundation for a strong family background and support. And to all my relatives who are all such a large part of my life, offering love and friendship and support. It is to my entire family network that I dedicate this thesis, because it is from their support and kindness that I have found the strength and pride to continue and to succeed. This work is just as much a product of their labor as it is a labor of mine, because without them, I would have been lost long ago. To all of you, thank you. iv ACKNOWLEDGEMENTS I’d like to take this opportunity to thank all of the people who have helped me through the years in preparing this thesis. I’d like to thank Dr. Dillon, my mentor and advisor, for his patience, un- derstanding, and wealth of knowledge that I have gained from him. It is through his guidance and open-mindedness that has allowed me to grow and develop as a scientist. I would like to thank the members of my guidance committee for their assistance and expertise in the development of my thesis project and writing of my thesis. Thanks too, go to Dr. Gale Harris for his many hours of assistance in the development of the DANTE pulse sequence. I would like to thank Dr. Seth Hootman for his expert tutelage on electrophoresis and Dr. Steve Heidemann for the use and instruction of the densitometer. I would like to thank everyone at Banes Packing for their friendly and helpful services in obtaining the tissue samples from the hogs. Finally, I would like to thank Deborah Murphy for her moral support through all the trying times, and for her patience and understanding during the critical and emotional periods which she has endured with the prepara- tion of this thesis. Also, I would like to express my deepest gratitude for her assistance in the typing and preparation of the thesis. I couldn’t have done it without her. TABLE OF CONTENTS LIST OF TABLES LIST OF ABBREVIATIONS INTRODUCTION Background and Rationale Objectives BACKGIOUNDANDLII‘HIA’IUREREVIEW Creatine Kinase Introduction Phosphocreatine Background Creatine Kinase CK Background CK Functional Background Metabolic and Energetic Background for CK CK Physical Characteristics Creatine Kinase Shuttle Physicochemical Action of PC and CK Smooth Muscle Background Skinned Smooth Muscle Studies NMR And Creatine Kinase Background NMR Background DANTE Background BBCK Clinical Background Summary vi ix xi l4 14 18 20 23 26 31 33 36 37 39 41 43 45 mm 47 Tissue Collection 47 Pouring Chromatographic Column 47 Purification of BBCK 48 Electrophoresis Methods 1 53 Molecular Weight Determination 57 lsoenzyme Determination 57 Methods For Kinetic Activity In Solution 59 Determination of Solution Kinetics 60 Kinetic Calculations 61 NMR Methods 62 DANTE Methods 63 NMR solution Methods 68 Methods for T1 Collection 70 Calculation of T1 in the Absence of Exchange 74 NMR Kinetic Experiments 76 NMR Peak Integration 77 CST Methods 77 MST Methods 80 RESULTS 84 Purification Results 84 Solution Kinetics Results 97 NMR Results 112 CST Results 112 MST Results 116 DISCUSSION 126 Purification I 126 vii Multiple Isoenzymes 127 CK Elution Profile 128 Electrophoresis 132 Possible Protein Impurities 132 Activity of the Purified BBCK 133 Solution Kinetics 135 CST Experiments 136 MST Experiments 137 DANTE Discussion 140 Rate Constant Introduction 142 Kinetics Summary 143 Conclusions 144 Future Directions 145 APPENDICIES 147 Appendix I 147 Appendix II 148 Appendix III 149 LIST OF REFERENCES 150 viii 10. ll. 12. 13. LIST OF TABLES CK Kinetics Comparison with BGPA Comparison of Kinetic Constants CK Tissue Activity Table Enzyme Purification Procedure Biorad High Molecular Weight Standard Delays for Progressive Saturation Experiments MST Saturation Protocol Purification Steps of BBCK Summary of Solution Kinetics Magnitization Changes from CST Experiments Rate Constants for CK Reaction Determined by CST Experiments Ratios of CST Results Kinetic Data from MST Experiments on Porcine Carotid Arteries ix 13 17 25 51 55 72 82 96 110 113 114 115 118 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. LIST OF FIGURES Creatine Phosphate Shuttle Electrical Analog Model Nuclear Spin Trajectories of Magnetization with DANTE Chromatographic Column Fluorometric Measurement of CK Control and Purified Fluorometric Measurement of CK Fraction from Homogenate Fluorometric Measurement of CK Ethanol Precipitation Elution Profile Electrophoresis Gel Densitometer Trace Progress of Reaction Kinetics Plots with PC Added Kinetics Plots with ATP Added Tl Experiment Graph CST Experiment Spectra Stack Plot MST Experiment Solution Control MST Experiment Spectra Stack Plot Creatine Loading 27 32 40 42 49 86 87 88 90 93 95 99 103 107 111 121 123 124 125 PC ATP ADP AMP MMCK BBCK MBCK Cr CKB mito CK BGPA BGPAP FDNB ATPase N MR CST MST T1 LIST OF ABBREVIATIONS creatine kinase phosphocreatine adenosine triphosphate adenosine diphosphate adenosine monophosphate muscle CK homodimer brain CK homodimer muscle brain CK heterodimer creatine pseudo first order rate constant in forward direction pseudo first order rate constant in reverse direction muscle CK monomer brain CK monomer mitochondrial CK B-guanidinopropionate B-guanidinopropionate phosphate 1-fluoro-2,4-dinitrobenzene adenosine triphosphatase nuclear magnetic resonance conventional saturation transfer multisite saturation transfer spin lattice relaxation time xi Pi YATP Z DANTE PW EDTA ME TA tris MOPS MES ETOH SDS-PAGE TEMED TCA G-6-P NADP NADPH spin spin relaxation time inorganic phosphate gamma phosphate of ATP magnetization along the Z axis magnetization along the Y axis magnetization along the X axis magnetization in the xy plane magnetization along the Z axis of the magnetic field local magnetic field strength hertz units (u mole/min.) delays alternating with nutations for tailored excitation pulse width continuous wave ethylene diamine tetra-acetic acid 2-mercapto ethanol tris acetate tris-(hydroxymethyl) amino methane 3-[N-morphilino]propane sulfonic acid 2-(N morphilino) ethane sulfonic acid ethanol sodium dodecyl sulfate polyacrylamide gel electrophoresis N,N,N ’,N, ’-tetramethylethylenediamine tricarboxylic acid glucose-6-phosphate nicotinamide adenine dinucleotide phosphate nicotinamide adenine dinucleotide phosphate reduced form xii BSA HPLC FID PSS U 3 TBA G6PDH bovine serum albumin high performance liquid chromatography free induction decay physiological saline solution daltons (atomic mass unit) milimoles per liter moles per liter triethanolamine glucose-6-phosphate dehydrogenase xiii IN D TI N Background and Rationale Creatine kinase (CK) exists in four different forms; three cytosolic and one mitochondrial isoenzyme. The three cytosolic isoenzymes are dimers of the CKM and CKB monomers, producing MMCK, MBCK, and BBCK. The mitochondrial isoenzyme of CK is an octomer found on the outside of the inner mitochondrial membrane (56,70). The molecular weight of the cytosolic CK has a range of 80-85 kD in the dimeric form. Cytosolic MMCK is a homodimer composed of two CKM monomers and is found predominantly in skeletal muscle. BBCK is a homodimer composed of two CKB monomers and is found in brain and smooth muscle. MBCK is a heterodimer composed of a CKM and a CKB monomer. It is found predominantly in the heart. CK catalyzes the reaction of: ADP + PC + H+ 9 ATP + Cr. The reaction going from phosphocreatine (PC) to ATP will be referred to in this thesis as the forward reaction. The reaction of ATP to PC will be referred to as the reverse reaction. It is via this reaction that CK has an important role in buffering the concentration of ATP in the cell, by phosphorylating ADP with PC. Studies by Walliman and Eppenberger (170,171,173) have shown that CK is bound to the myofibrils in the heart and skeletal muscle. This binding was predominantly at the'M and Z lines of the myofibrils. They found that 5-10% of the activity of CK can be bound to the myofibrils of heart and skeletal muscle. Studies by Ventura-Clapier et al have determined that CK may play a role in the contractility of skinned heart myofibrils (161,162,163, 164,165). In the skinned heart myofibrils they found that CK can be rinsed off the myofibrils and that the rinsing of CK from the myofibrils is accompanied by a decrease in contractility. They defined this decrease in contractility as a decrease in twitch tension associated with an increase in rest tension. The increase in rest tension is reported to be caused by an increase in the number of rigor bridges. They have also found that after rinsing CK off, it can be reapplied to the myofibrils with a significant return of function. The return of function is seen as an increase in twitch tension and decrease in rest tension. Their conclusion from these studies is that CK plays a significant role in maintaining the normal contractile f uno ction of the heart myofibrils. In studies performed by Seradarian et a1, cultured heart muscle cells incubated with exogenous creatine have an increase in PC content (138,139,140,l41,l42). They also found that the increase in PC content is significantly less if oxidative phosphorylation is inhibited by addition of oligomycin. Conversely, increased ATP con- centration is seen if cultured heart muscle is incubated with adenosine. This increase in ATP was significantly less if CK was in- hibited by the addition of FDNB. Their conclusion from these studies was that PC synthesis is linked to oxidative phosphorlation and that ATP synthysis is linked to the activity of CK. Clinically CK is used as a diagnostic indicator of heart attack, stroke, trauma, ischemia and muscular dystrophy (3,50,66,147). The cytosolic CK is released from the tissue via an unknown mechanism and is found, with kinetic activity, in the blood stream of patients. Jacobus et al have found a decrease in activity for the heart mitochondrial CK during ischemia (68). It has not yet been determined what, if any, correlation there is between the release of cytosolic CK and a decrease in mitochondrial CK activity. Cytosolic CK released into the blood stream may be due to some con- trolled release or a general permeabilization of the membrane to allow the leakage of cellular contents. When CK is found in the blood stream however, other cell products are not always found (178). On the other hand, ischemia associated with a heart attack generally produces cell death and is associated with significant increase in serum CK activity (124,125). Thus there is conflicting evidence on the possibility of leakage of cellular contents producing the rise in serum CK activity as well as experimental evidence refuting that claim. The mechanism or stimuli for CK release from tissue is not known and is an area of active study. The role of the mitochondrial CK (found on the outside of the inner mitochondrial membrane) is to synthesize PC from ATP which has been produced by oxidative phosphorlation. PC is then hypothesized to diffuse out of the mitochondria and across the cytoplasm to sites of ATP hydrolysis and buffer the ATP concentration in the cyt0plasm via cytosolic CK. After donating its phosphate to ATP, creatine then will diffuse back to the mitochondria to get rephosphorylated by mitochondrial CK from ATP (8,9). Jacobus and Saks have found that when coupled to oxidative phosphorlation, mitochondrial CK has an in- creased affinity f or the ATPOEOCr complex of up to 10 fold (69). They conclude that this increased affinity may increase the propensity for the release of PC which will then diffuse to other sites in the cell to help buffer the ATP concentration. This comunication of PC between the mitochondria and the myofilaments is referred to as the creatine phosphate shuttle. The creatine phosphate shuttle hypothesis is supported by Bessman et a1 (9) but is ascribed to a physical-chemical mechanism by Meyer et al (96,97). These differences are discussed in greater detail in the Background Section of this thesis. However, when addressing the energetics of a cell the creatine phosphate shuttle and possible subcellular localization of cytosolic CK (which may be independent processes) are important considerations. Mitochondrial CK is found in varying amounts in tissues depending upon the mitochondrial content and oxidative capacity of the tissue. chngar et al have hypothizied that the total PC concentration is proportional to the total CK activity in the tissue (64,65). The dialogue between mitochondrial CK and cytosolic CK has drawn much attention in the heart and skeletal muscle but is poorly studied in vascular smooth muscle due to the very low mitochondrial content of vascular smooth muscle and reduced diffusion distance due to a rel- atively small cell diameter. The conclusion to be drawn from the discussion above is that CK plays in a significant role in the buffering of the ATP concentration in the cell as demonstrated by studies in the heart and skeletal muscle. CK is also important in the contractility of the tissue, and as a clinical diagnostic indicator. The studies discussed above have focused predominantly on the skeletal and heart muscle and thus on the MM and MB cytosolic isoenzymes of CK. To date, there are few studies of BBCK and its role in the buffering of ATP in smooth muscle or its role in contractility or other normative functions of the smooth muscle tissue. The question being asked in this thesis is what are the kinetics of porcine carotid artery CK in situ and BBCK in vitro. This question can be addressed by determining the kinetics of CK in the porcine carotid artery under resting conditions, to purify BBCK from the carotid artery, and to characterize that purified CK to determine its kinetics. This study will further the knowledge and un- derstanding of CK and vascular smooth muscle energetics which have been sparse compared to those of heart and skeletal muscle. Objectives The questions being asked in this thesis are: what are the kinetics of CK observable in the porcine carotid artery, using the NMR technique of saturation transfer, and what are they for BBCK which has been purified from porcine carotid artery? To be able to answer these questions, several objectives must be obtained. First, BBCK must be purified from the porcine carotid artery. Second, the purified BBCK must be characterized to obtain the enzyme kinetics. Third, the CK kinetics of the porcine carotid artery must be determined in the NMR using saturation transfer. Objective 1. To be able to study the kinetics of an enzyme it should be isolated to an adequate degree of purity for the studies to be performed. The first objective of this thesis is to purify the representative CK from porcine carotid artery. BBCK was assumed to be the predominant form of CK in smooth muscle; thus the purification procedure used in this thesis utilizes a technique for isolating the BBCK from the porcine carotid artery. Objective 2. After purification of the enzyme, the kinetics are measured in solution focusing on the NMR visible substrates of PC and ATP. The kinetics are measured under steady state, pseudo-first order conditions to determine K m and Vm“. Objective 3. NMR saturation transfer is performed to determine the kinetics of the CK reaction in the tissue. Saturation transfer can determine the pseudo first order rate constant of the NMR visible nuclei undergoing exchange. This assumes that all of the measurable ATP and PC are available for the CK reaction. Conversely, if metabolites are sequestered around or away from exchange sites (e.g., shuttling between mitochondria and myofilaments) errors in calculating the pseudo first order rate constants may occur. Each of the objectives above are designed to provide information regarding the physiology and /or biochemistry of vascular smooth muscle CK. The purification and characterization of BBCK from vascular smooth muscle has not been reported prior to this thesis. These are the focus of Objectives One and Two. There have been few reports on the kinetics of CK in smooth muscle using NMR saturation transfer, and none using vascular smooth muscle. This is the focus of Objective Three. The results of these three objectives will contribute to a better understanding of BBCK in vascular smooth muscle. Creatine Kinase Introduction Creatine kinase kinetics have been studied extensively in solution over the years (102,107,108,109). Creatine kinase catalizes the reac- tion: ADP + PC 9 ATP + Cr. The reaction of PC to ATP will be referred to as forward and ATP to PC as reverse. PC participates in no other known reactions and can only make ATP from ADP, and thus is considered an ATP buffer (28,169). When a tissue (e.g., skeletal or cardiac muscle) is stimulated, PC falls first to very low levels before ATP begins to fall (98,120, 122,146). Qualitatively, the same phenomenon is observed in smooth muscle but is poorly understood due to the limited extent of study (20,63,114). CK is found in many tissues (31,37,64,71,85,95,ll8, 122,155,158,175). It exists in four forms: BB, MM, MB and mitochondrial. BB is CK found in brain. MM is CK found in skeletal muscle. MB is CK found in the heart. Porcine carotid arteries, like other smooth muscles, contains a large BB isoform concentration (174). The MM, MB, and BB isoenzymes are found in the cell cytosol. They are dimers with a molecular weight between 80 and 85 kD. The mitochondrial CK is an octomer found on the outside of the inner mitochoncrial membrane. In general, the specific activity of CK in heart and skeletal muscle is greater than in smooth muscle or brain (31,65,1 72,179). Phosphocreatine Background Phosphocreatine (PC) is intimately involved in energy metabolism (96,98) and contains a phosphate with a high free energy of hydrolysis (84,108). It is present in mammalian cells and in amphibians (30). Arginine phosphate is found in lower life forms and is considered PC’s evolutionary precursor (1,2,30). Arginine kinase also has a 10 fold lower activity than CK (98). PC was once thought to be the energy source for muscle contraction (67). Later studies (22) found that ATP was the true energy currency for the muscle contraction. The role of PC was found to be as a buffer for ATP. As ATP is used to supply energy for muscle contraction and other cell functions, PC is used to rephosphorylate ADP to ATP. This reaction is accomplished enzymatically by creatine kinase. As discussed above, creatine kinase (CK) catalyzes the reaction: ADP + PC 0 ATP + Creatine. PC is not known to be involved in any other reactions except 10 degradation to creatinine. This degradation is a constant non-enzy- matic conversion of PC to creatinine. and is not considered to significantly effect the kinetics of PC to ATP (145,169). The buffering of ATP via the CK reaction has been demonstrated by several researchers using various techniques (l4,l9,22,38,48,64,77,97, 13l,l32,14l,146,l62,l79). Both solution techniques and NMR have been employed to examine CK kinetics. CK buffering has been demonstrated by substituting PC with analogs like b-guanidinopropionate phosphate (BGPAP) (26,40,41,42,43). The buffering of ATP by CK has been observed in vivo as well as in cell cultures (138.139.140.141). Several of these techniques will be discussed in greater detail below. Cain and Davies (22) first demonstrated the buffering of ATP by PC by inhibiting CK with l-f1uoro-2,4-dinitrobenzene (FDNB). When the CK in muscles was inhibited the muscle failed to contract after 3 to 4 twitches. Meyer (96) described the CK system as having energy storage capacitance for rat skeletal muscle. This capacitance was described as being directly proportional to the total creatine level. Fisher and Dillon (38) used NMR to measure the high energy phosph- ate in porcine carotid artery. They found that during hypoxia PC decreased to 16% of control while ATP only decreased to 81% of control. Initial changes in PC concentration is seen well before the change in ATP concentration. As the above studies have determined, PC serves as an energy 11 reservoir in muscle. In an effort to identify biochemical and pathological changes due to deficiencies in either of these compounds, Fitch (26,40,41,42,43) has fed rats a creatine analog B- guanidinopropionate. It has been shown that B—guanidinopropionate (BGPA) will competitively compete with creatine for entry into skele- tal muscle (43). As the bGPA is taken into the skeletal muscle it will get phosphorylated to B-guanidinopropionate phosphate (BGPAP) by CK. This process of uptake and phosphorylation produces a decrease in cell creatine concentration from 15 to 8 u moles/g as well as a decrease in cell PC from 22.5 to 1.6 u moles/g (63). In these experiments the rats are fed a diet of 1% BGPA for 6 weeks. It was also found that the BGPA diet will produce a decrease in ATP and g1ucose-6-phosphate concentrations of 6.77 to 3.22 and 0.74 to 0.34 H moles/ g respectively. It might be predicted that BGPAP would have the same high energy phosphate bond as PC and be used as a source of energy in the muscle. However, BGPAP does not behave as ef f iciently as PC to buffer the con- centration of ATP. From in vitro studies on skeletal muscle CK, BGPAP has a Vm" which is 0.18% of PC (26). The low affinity of CK for 8GPAP seen in vitro supports observations, in viva, of decreased tension and rapid fall in tension for rat skeletal muscle which was loaded with BGPAP (40). Meyer et al (96a) found in BGPA fed rats that force did not decline with stimulation. They report that PC is not essential for energy production under steady-state conditions but that Pi from PC hydrolysis may be important for maximally activating glycolysis and/or glycogenolysis. 12 Pathological changes have also been seen in BGPA fed animals such as retarded growth, abnormal muscle contractions and a decrease in the size of white skeletal muscle fibers (26,40). The physiological im- plications (if any) of these pathological changes have not been deter- mined. BGPA as a creatine analog can thus be used to effectively sub- stitute BGPAP for PC in the cell. However, BGPAP cannot be used by the cell to buffer ATP as efficiently as PC because CK has a much lower Vm" for bGPAP. Also, the Km for BGPAP is lower than PC further compromising CK efficiency. (see Table l) (26). 13 TABLE I N T I N IT O O substrate Km Vmax EPA 50 mM 0.21 u M/min/mg BGPAP 4.9 0.22 Cr 16.7 75.] PC 2.2 231 ‘Ref. Chevali and Fitch 1979, (26) experiments done with rabbit skeletal muscle MMCK. As can been seen in Table 1, CK favors the hydrolysis of PC and would not be able to utilize BGPAP as an effective substitute for PC. Thus BGPAP cannot buffer the ATP concentration as efficiently as PC. When PC is replaced with BGPAP the muscle f atigues quickly as well as demonstrates a decrease in twitch tension. Conclusions drawn from BGPAP data should be analyzed carefully because a finite amount of buffering will still be available via the residual PC in the cells as well as slow transfer from the BGPAP. Creatine Kinase W Phosphocreatine is made by the phosphorylation of creatine by cre- atine kinase using up 1 mole of ATP per mole of PC produced. This re- action also produces one proton. Therefore, the reaction can be written: PC+ADP+H+ 0 ATP + CR In mammals this reaction is catalyzed by the enzyme creatine kinase. CK exists as one of four isoenzymes. The form of CK found in the mitochondria is an octomer on the outside of the inner mitochondrial membrane of all tissues containing the CK system (69,89,129) and is referred to as mitochondrial CK. The other three isoforms are dimers. The BB isoenzyme is a homodimer of two CKB monomers. The BBCK form is the isoform found in brain (52,53,88,174). There is also an MM isoenzyme. This is a homodimer of the CKM monomer. The MM isoenzyme is the major form of CK found in skeletal muscle. There are small amounts of BB and MB present in the skeletal muscle of humans. BB and MB are also found in larger amounts early in skeletal muscle development (71,154). The MBCK isoenzyme is a heter- odimer of the M and B monomers. This isoenzyme is found predominantly in the heart muscle. The two other cytosolic isoenzymes are also 14 15 present in the heart (71). Chegwidden and Watts (25) compared the kinetic constants of CK from various sources and species. They found that the kinetics of CK varied from species to species and tissue to tissue (see Table 2). 9 E‘PC'ADP R 1x E'ATP ‘ Kinetic formula for two substrates, two products, on enzyme kinetics / / E model discussed by Chegwidden (25). In this model K‘ is the binding of creatine (Cr) to CK (E), Kb is the binding of ATP to E, K’b is the binding of ATP to the E -Cr complex, I('" is the binding of Cr to the E-ATP complex, K’p is the binding of ADP to the E °PC complex, K’q is the binding of PC to the E °ADP complex, Kp is the binding of PC to E and Kq is the binding of ADP to E. Using this model, Chegwidden examined the details of the enzyme substrate and enzyme product interactions of CK from various sources (Table 2). 16 l7 TABLEZ ri .f i i 11 nt Monkey Calf Rabbit Calf Kinetic Skeletal Skeletal Skeletal Calf Smooth Reaction Constant Muscle Muscle Muscle Brain Muscle *Cr+E Ka 154 53 15.6 29 2.20 ATP+E Kb 1.79 0.97 1.2 0.93 0.75 Cr+E °ATP K ’a 12.75 21 6.1 3.7 0.58 ATP+E °Cr K ’b 0.147 0.78 0.48 0.13 0.20 E+PC Kp 16.7 45 8.6 20 ND# E+ADP K8 0.177 0.17 0.17 0.12 ND E - ADP+PC K ’p 3.5 23 2.9 2.0 ND E °PC+ADP K ’8 0.039 0.094 0.05 0.01 ND ’Cr: Creatine #ND: Not Determined Ref. #25 From the data presented in Table 2 it can be seen that the CK kin- etics from different species vary. There are also differences in the kinetics of BBCK from calf depending upon whether the isoenzyme is from brain or smooth muscle. These data appear to support the hypo- thesis of Vaidya (158) in which the CK is modified post l8 translationally depending upon the tissue or species which they believe may alter the role of the CK. Focant (44,45,46) examined CK from smooth muscle. Smooth muscle BBCK had differing amino acid content, differing kinetics, and differ- ing physical characteristics than the BBCK from the brain. It was demonstrated that the ox stomach BBCK had a molecular weight of 83,500 Daltons while the ox brain had a molecular weight of 80,000 Daltons (44). However, an examination of the amino acid content demonstrated that the differences between the two enzymes could not be simply ex- plained by post translation modification of the protein. In another study (46) it was found that the kinetics of the smooth muscle BBCK was significantly different than the brain BBCK. CK Emanuel. Mamas As mentioned in the Rationale Section, Wallimann found CK bound to the M line of myofibrils, while Ventura-Clapier et a1 (160,162,163) found that a large part of the MBCK activity is reversibly bound to heart myofibrils. Their results suggest that there are specific binding sites of CK in the myofibrils other than at the M line. These binding sites may have a lower affinity and thus may have been rinsed off the myofibrils during Wallimann’s experiments. However, these ex- periments were not able to localize the binding sites in the myofilaments. In similar experiments using chicken heart, Ventura- Clapier et al found that the MM isoenzyme will also reversibly bind to the myofilaments (163). The heart is the only tissue where MBCK is 19 found in any significant quantities. The ratio in the heart of MM and MB can vary from species to species (154) (Table 3). The role of the various CK isoenzymes or their localization is not known. Vaidya et al (158) determined that human heart contained 5 variants of the MM isoenzyme. This was accomplished by use of isoelectric focusing to determine the individual pl’s of MMCK. Five variants with five distinct isoelectric points were found in the human heart. They also described different kinetic activities and attributed these observations to post translational modification of the MMCK. Mechanical and chemical skinning techniques have been employed to evaluate the binding and activity of CK in the heart (94,132,159,160, 162). These studies showed that the relaxation of rigor tension in skinned heart muscle fiber was intimately dependent upon the kinetic activity of CK. Ventura-Clapier et at (162) demonstrated that in the absence of ATP, at high (12mM) PC and low (250uM) MgADP, the relaxa- tion of skinned papillary muscle could be maintained in the presence of active CK. In the absence of CK, rigor tension would develop under identical conditions. Ventura-Clapier (163) also found that the addition of CK will relax rigor tension in myocardial tissue. These results demonstrate that CK plays a role in the mechanical properties of heart muscle by providing ATP as well as removing ADP from the active site of actomyosin ATPase and thus effecting the mechanics. WMWWMCK Seraydarian (138,140) has examined the correlation between PC and muscle function using cultured neonatal rat heart. It was found that muscle fatigue in frog sartorius correlated with a fall in PC while no significant fall in ATP concentration occurred. A concomitant rise in Pi is observed with the fall in PC. The increased Pi lowers the free energy of ATP hydrolysis within the cell. It is believed that this lower free energy contributes to the decrease in muscle tension. In- creasing creatine in the medium produced an increase in PC. However, the increased PC was not maintained if the inhibitor of oxidative phosphorylation, oligomycin, was added to the incubation medium (141,142). These data suggest that creatine stimulated PC synthesis is dependent upon oxidative phosphorylation. Seraydarian et al (139) increased ATP concentration by incubating heart culture cells in 50 uM adenosine and also observed an increase in the rate of spontaneous contractions. The PC concentration, however, remained unchanged. The concomitant increase in ATP con- centration with the increase in contraction rate was considered evidence that membrane excitability and intracellular ATP are correl- ated in the cultured heart cell. The addition of FDNB, the inhibitor of CK, prevented the net synthesis of ATP in the presence of adenosine. These results suggest a possible role of CK in the increa- se of the ATP concentration in the presence of adenosine. Bessman (8,10) used labeled Pi to study mitochondrial CK. It was 20 21 found that if 32P labeled Pi was added to mitochondrial suspensions, the Pi was preferentially incorporated into PC. They found that the specific activity of mitochondrial CK was less then the specific activity of the labeled 7 phosphate of the total pool of ATP. ATP generated by mitochondria was not labeled and must have been used to produce the unlabeled fraction of the PC. It was suggested from these data that oxidative phosphorylation supplies ATP to CK in the mitochondria without the ATP mixing with the extra mitochondrial pool of ATP. In a parallel set of experiments, Bessman inhibited ATPase and ATPsynthase activity with carbonylcyanide chlorophenylhydrazone and atractyloside respectively and used exogenous 32P labeled 7ATP to observe the exchange of 32P from ATP to form PC. Quantitatively there was a decrease in PC formation and all the PC formed was from ex- ogenous ATP. Bessman concluded from these studies that mitochondrial compartmentation of CK allows the formation of significantly more PC in the presence of oxidative phosphorylation than the bound CK can produce in the presence of exogenous ATP alone. Jacobus and Saks (68,69) described the coupling of mitochondrial CK to oxidative phosphorylation. This study demonstrated that when mitochondrial CK is coupled to oxidative phosphorylation the dissocia- tion constant of MgATP for CK decreased 10 fold for the ternary complex, E °MgATP °creatine. Thus, oxidative phosphorylation increases the apparent stability of the E °MgATP 'creatine complex and will decrease the release of MgATP into the cellular medium and 22 concomitantly increase the release of PC into the cellular medium. Iyengar (64) describes a general correlation between the amount of CK activity in a tissue to the PC content in the brain and skeletal muscle. Fisher and Dillon (38,39) found 0.5 mM PC in porcine carotid artery. This is much less then the 15 mM PC (14) found in heart muscle. Lang found human carotid artery to contain less than 1% of the heart CK activity. In spite of this relatively low concentration of PC and CK activity, the porcine carotid artery is able to maintain sustained contraction and efficiently buffer its ATP concentration (4,11,20,39,104,l67). Reiss and Kaye (123) described an increase in uterus CK induced by estrogen treatment. Estrogen will subsequently increase the phosphogen content in the uterus as well as other well established anabolic effects. This lends support to Iyengar’s finding of a rela- tionship between CK content and PC concentration. As can be seen in Table 3, smooth muscle predominantly contains the isoenzyme BBCK (64,65,71,154). The purified BBCK generally has lower activity than the MM and MB forms (64,66,83). NMR experiments done with bullfrog stomach muscle (179) showed CK to have a forward rate constant of 0.16 s". This value is about half the 0.3 s'1 that Ugurbil (155,157) found in the rat heart. Carotid arteries contain approximately 0.5-1% the cytosolic CK activity that heart contains (71,88,154). There is also a lower PC concentration (38,86). It might be expected from these observations that the smooth muscle would 23 not exhibit good buffering capacity for ATP. However, it has been demonstrated that smooth muscle can buffer its ATP concentration similar to the buffering found in other muscles (4,11,20,38,39,86, 104,167). Grossman (52) used monkey brain to purify BBCK and described con- f ormational modifications in the enzyme using fluorescence and f luore- scence polarization. Polarization changes of labeled BBCK due to ADP binding were measured to determine structural distances and conforma- tion. Grossman’s experiments found a propensity for conformational changes and concluded that the brain BBCK is a more flexible protein then muscle CK. Also it was observed that the active site is more open than the muscle CK and that the binding of ADP produces a more compact CK. This more compact CK changes the conformation of BBCK such that it resembles muscle CK, and from these data conformational modification was implicated in the regulation of monkey brain BBCK (52). In this way, the neurobiochemical role of BBCK was thought to be distinct from the metabolic functions of muscle CK. Grossman (53) suggested isoenzyme specific compartmentation which could be the result of different subunit arrangements. In these ex- periments rabbit brain and skeletal muscle was used to isolate the BB and MM isoenzymes respectively. These two homodimers were used to produce the MB heterodimer. In this study the energy transfer from one subunit to the next subunit between the reactive thiols of MMCK 24 was measured. The reactive thiols were found to be separated by a distance of 48.6-60.4 A, such that energy transfer between thiols had a low efficiency. The heterodimers were found to have a shorter site- site distance of 27-52 A. It was this difference in active site distance and energy transfer between the two isoenzymes which led to the suggestion of isoenzyme compartmentation. The difference in the conformation of the two isoenzymes, suggested by the observed dif f ere- nce in active site distance, might explain the lack of binding of the MB form to the M line. As discussed above there are different isoenzymes of CK and dif- f erences within isoenzymes. Several studies have identified multiple varients of the MM isoenzyme (36,149,150,151,158). It was found that skeletal muscle has multiple forms of MMCK and that these forms have differing p13. The different isoenzymes also demonstrated differing kinetics. Such heterogeneities within the isoenzymes of CK was attributed to post translational modification (158), but it is unclear if these physical characteristics have physiological significance. Table 3 IisuuLSanuue Skeletal Muscle Tongue Diaphragm Heart Right Atrium Brain Cerebrum Cerebellum Hypophysis Spinal Cord Stomach Gastrointestinal Tract Colon Ileum Bladder Aorta Carotid S3sLLEEBJELASZILZIIJLJusBLJi 1L0: fhkflflcni sagas}; 3281 100 860 96 3 225 90 s 140 94 4 800 52 46 402 78 22 90 87 11 27 23 3 6 140 3 125 4 161 3 1 35 2-7 3-5 1 39 7 2 56 2 100 100 100 100 91 97 96 96 89-93 54 42 Rail-e. 154 71 71 71. 88 154 88 88 88 88 71 88 154 154 71 71 88 25 WWW; A transport mechanism has been proposed by which PC from the mitochondria is transported to the cytoplasm (8,9,10,94,98,133). The method by which PC is transported from the mitochondria to energy uti- lizing structures in the cell has been referred to as the PC shuttle. The PC shuttle hypothesis is not, however, universally accepted. A discussion of the literature regarding the shuttle hypothesis should be examined with attention to the controversy involved with this proposed shuttle mechanism. However, due to the nature of this thesis consideration of the experimental observations and theories pertinent to the PC shuttle hypothesis is warranted and will be discussed. In muscle cells, as well as many other cells, the PC synthesized by mitochondrial CK is utilized by the cytosolic CK elsewhere in the cell (e.g., myofilaments). The mechanism by which PC gets to the myofila- ments is called the phosphocreatine shuttle. This shuttle also may act to signal an energy demand as well as delivering energy to the organelles as PC. The phosphocreatine shuttle acts as an intercom- munication process to signal demand for energy and the transport of energy produced as a response to that signal to the sites utilizing 26 atroeuonantxa TRANS‘ LOCASE acspxanroar CPK eonraarnrnr ,. \ 1.0? CHAIM 1 INTERMEHBRAHE SPACE MITOCHOHDRION /> m\ /— \\‘__* -) CR +'--CR < \ g... C? ---.,cr./ 1HTERVENING SPACE //:::finntLL::\\\\ 7 ATP ATP‘ canrxnrneur “55 bk\‘- ADP HYOFIERIL Figure 1. Creatine phosphate shuttle showing the site of synthesis at the inner mitochondrial membrane. CK resides on the outer surface of the inner mitochondrial membrane. The intervening space is the space within the cell between the mitochondria and the myofilaments. At the myofilaments, the CK is depicted as being bound to the A band of the myofibril. In this model the mitochondrial energy is transported as phosphocreatine (Bessman). CPK: creatine phosphate, CP: phosphocre- atine. 27 28 energy (10,112,133,152). ADP could be the signal for energy demand, but Jacobus and Saks (69) determined that creatine as a signal for energy demand would have a parallel effect as that of ADP to signal energy demand (Figure 1). During the contraction of a muscle, ATP is hydrolyzed and ADP is produced by CK phosphorylating ADP with PC, producing creatine. Cre- atine can then act at the mitochondria to stimulate oxidative phosphorylation (69,70, 129). This production of creatine, concom- itant with ADP, occurs at or near the site of contraction. The cre- atine has its signaling action at the mitochondria. Therefore the creatine must diffuse or be transported away from the myofibrils to the mitochondria. Meyer (98) determined that simple diffusion of the metabolites is an adequate mechanism for transport within cells. Creatine at the mitochondria gets phosphorylated by mitochondrial CK using ATP synthesized via oxidative phosphorylation (69). ATP donates its gamma phosphate and ADP is produced. ADP production is an immediate control stimulus at the mitochondria for respiration to produce more ATP (10,127). PC leaves the mitochondria and diffuses towards the myofilaments (69,98,127). The net flux of creatine is in the reverse direction. PC arrives at the myofilaments as a carrier of energy from the mitochondria. This energy is transferred from PC to ADP via 29 cytoplasmic CK to produce ATP and creatine as well as consuming one H+. The dialogue between the two isoenzymes of CK occurs within brain, heart, skeletal and smooth muscles (8,9). Each tissue contains 2 different isoenzymes (mitochondrial and cytoplasmic CK) to help effect the dialogue in the different locations. The shuttle of cre- atine and PC are what is exchanged in this dialogue. The creatine - PC system allows for communication of information and energy between the site of energy synthesis and the site of energy utilization. This system can allow the dialogue to be conducted between certain points in the cell which are capable of transducing the message. The 2 CK isoenzymes act as the transducers in this dialogue. Therefore a cell system such as Na+K+ ATPase which is coupled to cytosolic CK (130) would be able to buffer its energy supply with PC. A cell system not physically coupled to a CK isoenzyme would not be able to buffer its energy supply in this way. A cell system like. this could use free ATP from the mitochondria or PC buffering from soluble CK to meet its energy demands. Cell systems coupled to CK could then have perf erred buffering of their energy supply. Such systems in the cell are the myofilaments (160,161,170) and the Na+K+ ATPase (51,130). Peripheral components which utilize ATP, and may have a sudden increase in ATP demand, might be considered as excellent candidates for a coupled CK system. Meyer et a1 (98) discussed the PC shuttle in relation to storage of high energy phosphate that buffers changes in ATP levels. It was demonstrated that the properties of the CK reaction permit it to 30 function as the tranporter of high-energy phosphate within cells. The shuttling of CK metabolites is described as being analagous to f acili- tated diffusion. Applying Fick’s law of diffusion, metabolic condi- tions in the cell, and assuming CK to be at equilibrium, it was deter- mined that by diffusion alone the'majority of the diffusive flux of high-energy phosphate from site to site within mammalian muscle is carried by PC. The conclusion drawn here was that the transport and buffer aspects of the CK reaction are features of the same fundamental properties. The transport function and buffer function of CK both were viewed as resulting from the CK reaction being near equilibrium. Meyer et a1 (98) discussed the function of CK in muscle. The CK reaction was discussed as tending to lessen the dissipation of free energy by diffusion. The steady state calculations, regarding the spatial buffering of CK, for a hypothetical 30- um diameter myocardial cell were applied to this system. They found that the free energy drop was less then 1 kJ/mole even without the CK reaction. They con- cluded that due to the distances typically involved and small energy drop, it would be unlikely that CK would be required to support high- energy phosphate transport into myofibrils. This conclusion assumes that the mitochondria are packed around the myofibrils as in the case for the myofibrils in the rat heart. The above argument regarding spatial buffering of the CK reaction would be most important for large cells where the mitochondrial dis- tribution could be non-uniform and a significant ATPase rate exists. In small cells (e.g., smooth muscle) the diffusion distances are 31 sufficiently small such that the buffering by CK would not have a significant effect because no significant diffusion gradients for any of the metabolites would be established. In this case even without the CK reaction there would be little ATP gradient. WMfiEMCfi. Meyer (96) used an electrical analog model to represent respiratory control in muscle during submaximal rates of oxidation. In this model the CK reaction is described as capacitance. The PC concentration is analogous to stored energy charge on the capacitor. Resistance represents the number and properties of the mitochondria while current is analgous to the rate of oxidative phosphorylation (Figure 2). If this model is correct, the level of PC must be linearly related to the cytosolic free energy of ATP hydrolysis over the observed sub- maximal oxidative rates in the cell. Also the ATP synthesis by oxid- ative phosphorylation must be linearly related to the cytosolic free energy of ATP hydrolysis. It was determined that the above rela- tionship was linear over much of the range of phosphorylation which would be observed during moderate work. Thus the free energy of ATP hydrolysis is pr0portional to the level of PC and the energy storage capacitance of the CK system is linearly proportional to the total level of creatine. The model can thus make several predictions as long as the above mentioned limitations are observed. The first prediction is that the Figure 2. Electrical analog model for respiratory control of the muscle as described by Meyer (1988) for submaximal rates of oxidation. Vb represents the free energy potential in J/mole. Rm is resistance, which is a function of the number and properties of the mitochondria in the cell. Vo represents the cytosolic free energy of ATP hydrolysis. C is capacitance, which is due to the concentration of PC and the CK reac- tion. Icy represents the current due to the ATPase rate. , 32 33 time constant for PC changes should be identical at the beginning of and during the recovery from a step change in cyt0plasmic ATPase rate. Also, the apparent time constants should be independant of ATPase rate. Finally, steady-state oxygen consumption should be linear at a steady-state level of PC during ATPase rates which are below the maximum aerobic capacity. These predictions were tested experimentally using NMR measurements in the rat gastrocnemius. The results were consistent with the elec- trical analog model. It was found that the PC time constants changed independant of work rate and were similar at the onset vs. during re- covery after stimulation. Also, the relationship between steady state PC levels and the rate-force product was linear. The final conclusion of the study by Meyer was that the apparent first order behavior of PC concentration changes, observed in the rat gastrocnemius can be modeled using a general linear circuit analog. Smooth Muscle Background Paul et a1 (114,115,116) studied the contractibility of vascular smooth muscle and its coupling to energy metabolism in the cell. Using oxygen consumption as a measure of ATP utiliztion, the ATP cost of contraction was measured. By measuring 02 consumption and ATP syn- thesis it was found that the ratio of ATP synthesis to O utilization (ATP:O ratio) was approximately 3. The time course of energy utiliza- tion was measured during steady-state isometric contraction. Oxygen consumption, (ATP synthesis) did not correlate with the development 34 and maintenance of force. It was found that 02 consumption closely paralleled the velocity of contraction. This phenomenon was referred to as an increased energy utilization in the pre-steady-state condi- tion and the result is that maintained tension in vascular smooth muscle does not cost as much energetically as tension development. This energetic economy of maintained tension was estimated to be approximately a factor of 2 over tension development. The difference in contractile costs of vascular smooth muscle was related to the dif- f erences in smooth muscle contractile protein interaction compared to skeletal muscle. Dillon et a1 (34) first described the latch phenomenon in porcine carotid arteries. Latch describes the state of vascular smooth muscle where force is maintained with high energy economy. It was described as a state where the attached crossbridge is dephosphorylated but remains attached to the actin molecule. It was found that the shorte- ning velocity was. correlated to the level of myosin light chain phosphorylation. The phosphorylation of myosin light chains are Cal"2 dependant. Therefore, Ca+2 dependent myosin phosphorylation generates actomyosin activity and crossbridge cycling, while myosin light chain phosphatase can dephosphorylate attached crossbridges and reduce the rate of crossbridge detachment. The attached but noncycling cross bridge can maintain tension and is referred to as a latch bridge. This latch bridge can thus maintain tension with low energy cost to the tissue. The release of the latch bridge is reported to occur with a time constant greater then that seen in cycling crossbridges (24,54,55). 35 Hai and Murphy (54,55) discussed the phosphorylation of myosin light chains and the energetics of contraction. Their model is in- trinsically dependent upon the latch bridges having the same force generating capacity but a 5-f old slower rate of detachment then the phosphorylated and cycling crossbridges. Estimates of ATP turnover determined that the tension maintenance is very economical. In con- trast, the efficiency of the work done is considerably lower when com- pared to skeletal muscle. The low efficiency was due to a high ATP consumption rate for crossbridge phosphorylation. Therefore, the energetics of smooth muscle is economical for the maintenance of tension while not very efficient in generating tension. Butler et a1 (21) using rabbit portal vein performed pulse chase experiments to determine the nucleotide bound to myosin and the release rate under relaxed and activated conditions. The results suggested that the myosin exists primarily complexed with ADP. This complex predominates between ADP and myosin during active and relaxed states. The release of ADP in the relaxed smooth muscle was biex- ponential. The first exponential of ADP release contained approxim- ately 1/3 of the total ADP and was 5-10 times faster then the other release componant. It was suggested that these data indicate that there may be a 5 to 10 fold difference in the rates of cycling for different attached crossbridges in smooth muscle. This, they proposed may also account for the variable energy costs for force output. Skinned Smooth Muscle Studies Skinned skeletal and heart muscle has been used to study muscle metabolic and biochemical function as well as CK action (4,15,21,23, 95,130,132,159,160,161,162,163,164). Skinning techniques have been used to study smooth muscle (58,61,72,75,106,lll, 119,126,168) but the study of CK in smooth muscle is limited. Bose (15) using canine trachael smooth muscle found that rigor tension was induced when the CK inhibitor FDNB was added in the presence of iodoacetic acid. It was also noted that this change in tension was not significantly dif- f erent in the presence or absence of Ca”. Kargacin and Fay (73) determined that after saponin skinning smooth muscle will retain its ability to contract and that the Ca"’2 regulatory mechanism is still required (i.e., calmodulin action is still needed to induce shortening) (76). It was also found that the shortening velocity is decreased if ATP concentration is decreased. If CK and PC are added to the medium in the presence of a low ATP con- centration the rate of shortening is significatly increased. Similar findings have been found in the heart by Ventura-Clapier et a1 (162,165). Coupled with the observations of Bose (15), the results may be considered as evidence for CK having a role in the contractile mechanism of smooth muscle. Kossmann et a1 (81) examined the effect of skinning on smooth muscle preparations. They found that studies on skinned smooth muscle can be regarded with confidence because the act of demembranation does 36 37 not grossly disturb the composition and organization of the con- tractile apparatus. However, the storage and handling of the skinned smooth muscle preparation can alter the contractile response. It was observed that there can be significant protein loss from the muscle during sequential contraction and'relaxation cycles. Storage in glycerol will also lead to a significant loss of contractile proteins and contractility. These studies described a means for controlling and monitoring the integrity of skinned smooth muscle preparations. Kossmann et a1 believe that the skinning model is a valid research model, but that the results of skinning experiments need to be care- fully examined and interpreted (81). NMR and Creatine Kinase Background NMR has been used in chemistry, physiology and medicine as an analytical tool. NMR spectra give data on molecular structure, con- centration, orientation and interaction (5,27,47,57,62,101). The NMR technique of saturation transfer has been used to measure molecular exchange of the CK reaction in living tissue (l6,47,93,97,120,l44, 155). NMR is a non-invasive technique which makes it well suited for biological and medical applications. Saturation transfer can be used to determine the CK kinetics in living tissue and has been used to do so in several experiments (14,16,17,19,l79). The technique itself is discussed in greater detail in the Methods Section. One advantage of the technique is that it can be used to determine the kinetics of CK in the living tissue 38 and be able to alter conditions of the experiments and repeat experi- ments in the same tissue. Thus, each tissue is its own control (l44,l46,157,l79). It is also possible to make measurements on f unc- tioning organs while in the living organism (17). Bittl et al (14) found that the CK flux increased with oxygen con- sumption and cardiac performance. This relationship held over a wide range of cardiac performance. They suggested that the CK reaction was controlled by a substrate that also controlled oxidative phosphoryla- tion. This prediction is consistant with the discussion previously regarding creatine’s effect on oxidative phosphorylation (8,69). Ugurbil’s (156) treatment of the CK reaction as a three site exchange takes into account ATPase effects on ATP degradation (the reverse rate constant). It was determined experimentally that the CK reaction was in fact at equilibrium in heart, and that the continuous wave saturation transfer experiments, assuming a two site exchange, are prone to underestimating kr' Using a technique similar to Ugurbil’s MST, Spencer et al in 1988 (146) repeated experiments by Bittl et a1 (14). Spencer found that the fluxes were erroneous if a two site exchange is assumed. The results showed that CK was at equilibrium during conditions where pre- vious results gave significant differences in the net flux. These ex- periments also showed that the discrepancy resided in the determina- tion of the rate of ATP degradation or k1,. The value of kl. had been reported as being lower than it was when measured with MST. To date 39 there are no studies examining the MST and CST techniques in the CK kinetics of smooth muscle. NMR Background The longitudinal relaxation time (T1) of the exchanging nuclei effect the calculation of the kinetic rate constant. Therefore the T1 for each nuclei must be determined and taken into account when calculating the rate constant (18). To generate T1 the nuclei must first be in a low energy state dic- tated by the magnetic field. When atoms are in the presence of a strong homogeneous magnetic field, some of the nuclei align themselves with parallel and anti-parallel spins to the field. As long as the magnetic field is applied, the sum of the spins will form a net vector along with the Z axis of the magnetic field (MO). The nuclear spins will precess at the Larmor f reguency (0°. Once the nuclei are so arranged a radio frequency pulse is applied at the Larmor frequency along the X axis. This causes the magnetization vector, Mo, to tip away from the Z axis (Figure 3). The frequency of transmission for the nuclei is determined by the formula: 11 - 7 Bo/Zar. Where: 1: - the resonance frequency, 1 =- the gyromagnetic ratio, B0 = the local magnetic field strength. After a radiofrequency pulse, the nuclear spin is in a high energy state. This energy state is 1 Magnetic 1 field Qé‘ (n m--l; Figure 3. Modified from Gadian 1982 (47). A representation of the magnetic moment generated by the spin of a nucleus is seen in (a). The orientation that can be formed by the nucleus when in an applied magnet- ic field B0 is seen in (b). 40 41 represented by a net vector in the x-y plane given by the formula: Mxy =- M0 sin 9. Where: Mx is the magnetization in the x-y plane and 6 is the angle Y away from M0. The 90 degree pulse width is a pulse long enough to flip the net spin 90 degrees away from Mo. In this way, the flipping of the nuclei can continue through 360 degrees. As the nuclei begins to recover back towards the Z axis they recover with an exponential time constant referred to as the longitudinal or spin lattice relaxa- tion time. During the recovery, nuclei emit energy which can be re- ceived by rf coils tuned to that frequency (5,47,101). DANTE Background To be able to perform multisite saturation transfer (MST) two tailored pulse sequences must be applied simultaneously in the time ex- citations domain so as to result in two different saturating excitations in the frequency domain. Bittl et al (13) and Spencer et al (146) per- formed MST by using a synthesizer and harmonic oscillator. Ugurbil (156) generated MST using two computer-controlled synthesizers. The synthesizer generates saturation with a continuous pulse applied in the time domain to excite a specified resonance in the frequency domain. The continuous wave (CW) saturation (soft pulse) will have the effect of exciting a single resonance peak in the frequency domain without excit- ing neighboring resonances (Figure 4). In contrast a "hard pulse" is Figure 4. Trajectories of magnetization vectors computed by the Block equations as described by Morris and Freeman (1978). In this example the excitation train of twenty pulses were spaced two milliseconds apart and their sum totaled the 90° pulse. Trajectory [a] represents the magnetization changes which occur on resonance for this pulse train. Trajectories [b] through [1] represent trajectories which reside off re- sonance of trajectory [a] by increments of 2.5 Hz. The unlabled trajectories opposite of trajectory [a] represents vector effects with equidistant offset to the labeled trajectories. As can be seen from the [h] and [i] trajectories, an extended train of pulses would cause the off resonance trajectories to achieve a steady state approaching'the Z axis. 42 43 used to excite a region in the frequency domain. Morris and Freeman (100) described a pulse sequence utilizing a train of hard pulses which will generate a tailored excitation called DANTE. The details of this technique are discussed in the Methods Section. BBCK Clinical Background BBCK has been used clinically as a biomarker of neurological disease or damage and is used as a parameter for the diagnosis of certain cerebral carcinomas (29,166). It has been found in the blood stream during parturition, Caesarian section, and coronary bypass surgery (158,166). Normal serum BBCK values are below 1 U/liter. Vladutin et al (158), however, found that during bypass surgery BBCK in- creased to as much as 43 U/liter with a concomitant increase in MBCK. They found that the clearance of BBCK is faster than the clearance of MBCK. This study also noted that the isoenzymes can dissociate into monomors in vivo and reassociate. However, the possibility of BBCK orginating from reassociated B monomers or MBCK originating from M and B monomers was not investigated. The tissue origin of the BBCK was not determined in this study. BBCK has high activity in brain, uterus (especially during preg- nancy), gastrointestinal tract and vascular smooth muscle (33,65,123). Damage to these tissues has been reported to cause increased BBCK activity levels in the serum (29,49,166,180). The release of BBCK from 44 the brain is believed to be attenuated due to the blood brain barrier. Perturbation of the blood brain barrier, such as from trauma, however, may allow increased release of BBCK from the brain (88). The potential use of BBCK for clinical diagnostic purposes is con- founded by significant inactivation of BBCK while in the bloodstream (113). Thus assays using kinetic activity for BBCK fractionation can be inaccurate. Conversly, immunological assays for CK fractionation which react with the CKB monomer will cross react with MBCK and produce mis- leading results when determining BBCK (59,134). BBCK has been found bound to IgG’s in the human serum (88). Binding to IgG in the human serum will cause a change in the electrophorectic mobility of BBCK and again lead to potentially erroneous results (90). It has been observed that while in the serum, CK dimers will dissociate and reassociate. The monomers can reassociate with either monomer, so CKB may be recombining with CKM to form MBCK (74,134). If this were to occur, MBCK might be more likely to be detected in the serum because of its increased stability. BBCK is a significant fraction of the heart and skeletal muscle CK activity during early developmental stages (71,88,110). It also pers- ists as a large part of the adult skeletal muscle CK with muscular dystrophy (83,78). In patients with Duchene Muscular Dystrophy, BBCK comprises 10% of the CK activity in the afflicted muscle (83). Therefore, the BBCK which might be found in the serum of young or 45 muscular dystrophy patients could be originating from tissue normally not associated with BBCK. As discussed above the BBCK can originate from multiple organs, unlike MBCK which is predominantly found in the heart and in only trace amounts in the skeletal muscle (99). Thus BBCK may have clinical relevence as a diagnositc tool, but the results from BBCK measurements should be evaluated cautiously because the tissue of origin may be uncertain (32). The possibility of BBCK release from smooth muscle becomes increas- ingly convoluted in light of the multiple CK isoenzymes which are observed in smooth muslce (71,154). CK isoenzymes are known to have differing kinetics and different physical properties (45,46,53,74,83,90, 99). If a tissue contains multiple f arms of CK it is not known if these different f arms are localized in different places in the cell. In vitra studies have demonstrated kinetic differences between CK isoenzymes as well as different physical properties. Summary From the discussion above, it appears that much is known about MM and MBCK from heart and skeletal muscle from in viva and in vitra studies, and using multiple research techniques. Less attention, however, has been given to the study of BBCK in vascular smooth muscle. The purpose of this thesis is to determine the kinetics of CK observable in the porcine carotid artery using the NMR technique of saturation 46 transfer, and the kinetics of BBCK in solution which has been purified from porcine carotid artery. The purification of a single isoenzyme from the tissue is performed because it has not been determined what, if any, interaction the multiple isoenzymes undergo within a tissue. If multiple forms of cytosolic CK are present in the porcine carotid artery, saturation transfer results will reflect the kinetics of the available isoforms and/or of nuclei which are NMR visible. Performing these experiments contribute to better understanding of vascular smooth muscle energetics and BBCK kinetics. MEIHQQS Tissue Collection Porcine carotid arteries were obtained from pigs at the time of slaughter. Tissue was kept in ice cold physiological salt solution (PSS) after removing clots and flushing the lumen with PSS. PSS con- tained; 116 mM NaCl, 5.4 mM KCl, 25.3 mM NaHCO3, 1.1 mM NaHzPO4, 2.5 mM CaClz, 1.25 mM MgSO4, 0.1 mM EDTA and 15 mM glucose. Within 2 to 3 hours of collection arteries were randomly divided into two groups. The first group of arteries were placed in fresh PSS and stored over- night at 4°C for NMR kinetic experiments. The second group was put aside to be prepared for purification experiments. Pouring Chromatographic Column Several weeks prior to the purification experiments the chromato- graphic columns were prepared. Six Pyrex columns were purchased from Corning Glass Works (Corning, NY) with dimensions of 20 x 400 mm and fine glass frit at the bottom. Hydroxylapatite for the chromatography was purchased from Biorad (Richmond, CA), pre-suspended in a solution containing 0.01 M sodium phosphate and 0.02% NaN3. Packing of the columns was accomplished following the methods of Ault (6). The hydroxylapatitc solution was thoroughly mixed and approximately 30 ml 47 48 of solution was poured quickly into each column. Columns were then allowed to settle over night at 4°C. After settling, excess buffer was siphoned off and the pouring procedure repeated. This sequence was repeated until the columns were packed to a height of 25 cm of hydroxylapatitc. During this procedure the columns and solutions were kept at 4°C and the columns were not allowed to dry. After the columns were poured, they were allowed to settle further for several days. The columns were then fitted with disks placed at the top of the column of hydroxylapatitc. The disks were made of Q2 Filter Paper from Fisher which were cut to fit the internal diameter of the columns. On top of the filter paper was placed glass wool and finally a layer of glass beads (Figure 5). Once the packing was complete the columns were rinsed thoroughly with column buffer. Column buffer con- tained 10 mM tris acetate, 25 mM ME and 0.5 mM EDTA at a pH of 7.4. Rinsing was accomplished at 4°C and the columns were protected from drying. The columns were kept at all times at 4°C during experiments as well as between experiments. Purification of BBCK Freshly collected arteries were stripped of loose connective tissue and adventitia using the technique of Herlihy and Murphy (60). This technique involves removing the bulk of the connective tissue by blunt dissection and scissors leaving the artery and a thin protective sheath of connective tissue surrounding the artery. By anchoring the proximal aspect of the artery and grasping a small section of connec- tive tissue just distal to the anchor the remaining connective tissue GLASS BEADS 3.133.}? .172: __ G L A SS WO 0 L FILTER PAPER HYDROXYLAPATITE GLASS FRIT TYGON TUBING CLAMP Figure 5. Chromatographic column used for purification of creatine kinase modified from Ault (6). See text for packing instructions and dimensions. 49 50 can be pealed off distally. The remaining tissue has 96% vascular smooth muscle cells, the remainder being endothelial cells from the intima (60). Stripped arteries were fast frozen with clamps precooled in liquid nitrogen. Frozen arteries were stored at -85°C for up to several months to accumulate sufficient tissue mass for purification. On the day of the purification procedure, 100g of frozen tissue was quickly mixed with 400 ml of standard buffer at 4°C. Standard buffer cons- isted of: 50 mM Tris acetate (TA), 0.5 mM EDTA, 25 mM 2- mercaptoethanol (ME) and 10 mM KCl. The pH was brought up to 7.4 using 1 N NaOH. Unless otherwise stated, all procedures and solutions are at 4°C. The tissue and buffer was homogenized for 30 minutes in a Waring blender. Following the procedure described by Wang and Cushman (174), the homogenate was centrifuged at 37,000 g for 45 minutes. Modifications to Wang and Cushman’s method for the purification of BBCK from human brain were obtained from methods of Iyengar et a1 (65) where 10 mM KCl was used during the homogenization and the duration of homogenization increased to 30 minutes and repeated twice. While saving the pellet, the supernatant was precipitated with 1.57 volumes of 95% ethanol chilled to -10°C. The precipitated supernatant was centrifuged at 8,000 g for 20 minutes. While pooling this pellet with the other pellet from above, the supernatant was precipitated with 1.23 volumes of 95% ethanol. Using 2-3 times the volume of the pooled pellets, standard buffer was used to rehomogenize the pellets for 20 minutes. The rehomogenized pellets were centrifuged at 37,000 g for 45 minutes. The pellet was discarded and the supernatant precipitated Table 4 Bastian Tissue Collection Clean tissue Freeze & store tissue Homogenize Centrifuge @ 37,0003 Precipitate supernatant Centrifuge supernatant @ 8,000g Rehomogenize pooled pellets Centrifuge homogenate @37,000g Precipitate supernatant Centrifuge @ 8,000g Pool all supernatants Precipitate pooled supernatants Centrifuge pooled supernatants Load column with supernatant Rinse column Elute protein Concentrate protein 201111135313 At time of slaughter Herlihy & Murphy -85°C 30 min in standard buffer 45 min, Save pellet 1.57 v/v ETOH -10°C 20 min, Save pellet 30 min, in standard buffer 45 min, Discard pellet 1.23 v/v ETOH -10°C 20 min, Discard pellet Final ETOH content of 70% 20 min, Discard pellet Discard eluent Discard eluent Save eluent Fractional Dialysis 51 52 with 1.57 volumes of 95% ethanol and centrifuged at 8,000 g for 20 minutes. The pellet was discarded and the resulting supernatant was brought up to 70% ethanol with 1.23 volumes of 95% ethanol. All supernatants were pooled and centrifuged at 8,000 g for 20 minutes and the pellet discarded (Table 4). The supernatant was loaded on a 40 cm hydroxylapatitc column with a diameter of 2 cm packed previously to a height of 25 cm using hydroxylapatitc purchased from Biorad (as described above). The column had been rinsed thoroughly with column buffer containing 10 mM TA, 0.5 mM EDTA and 25 mM ME at a pH of 7.4 and kept at 4°C. Once the columns were loaded, they were rinsed with a volume of column buffer equivalent to the total volume of standard buffer used in preparation. Elution was performed with an elution solution of 10 mM KH2P04, 0.5 mM EDTA and 25 mM ME. The volume of elution solution was the same as that of the column buffer used to rinse the columns. Fractions were collected from the column every 7 mls. Fractions with approximately 17 ug of protein per ml or greater were pooled. Figure 9 (Results Section) shows the elution profile from a purification batch. Protein concentration was determined using the Biorad (Richmond CA) microassay procedure. The eluted enzyme was concentrated by fractional dialysis. Fractional dialysis consists of dialyizing against 70% ethanol and 25 mM ME. Following ethanol dialysis, the dialysis tubing is tied off at a smaller volume and dialyzed against standard dialysis solution containing 10 mM TA and 25 mM ME solution to remove the ethanol. If further concentration is desired, this process can be repeated. The 53 final protein concentration obtained by this method was 66 ug/ml. Electrophoresis Methods SDS-PAGE electrophoresis was performed to determine purity and molecular weight of the enzyme. The method of Lamelli (87) was followed. The plates for the electrophoresis were thoroughly rinsed with distilled water and dried in a drying oven prior to the pouring or assembly of the plates. They were assembled to make a 1 mm thick gel using plastic spacers at the sides, and sealed at the bottom and sides with a silicon tubing gasket. The plates were mounted together using 4 clips at the sides of the plates. Next the lower gel was pre- pared and poured. This included preparing the lower gel with 8 ml of a solution of 30% acrylamide with 0.8% bisacrylamide, 8 ml of a buffer containing 1.5 M tris and 0.4% SDS, pH 8.8, and 16 ml of distilled water. The above was filtered and polymerization initiated with the addition of 10 ul of TEMED and 365 pl of ammonium persulfate. The pouring of the lower gel was accomplished by quickly adding the gel solution between the plates. This was done using a Pasteur pipette. During the pouring, care was taken to prevent the formation of bubbles and to detect the presence of leaks. If a leak was observed, the plates were disassembled and re-assembled before continuing. To prevent drying of the lower gel an over lay solution of 11.25 ml of water, 3.25 ml of a solution containing 0.5 M tris and 0.4% SDS at pH 6.5, 15 ul TEMED and 45 ul of 10% ammonium persulfate was placed over the lower gel. The upper gel consisted of 8.6 ml water, with 2.25 ml of a solution containing 30% acrylamide and 0.8% bisacrylamide and 54 3.75 ml of a solution containing 0.5 M Tris and 0.4% SDS at a pH of 6.8. The polymerization was initiated with the addition of 15 ul TEMED and 45 pl 10% ammonium persulf ate. The overlay was removed by paper absorption and the upper gel poured using a technique similiar to that used for the lower gel. A teflon comb was placed in the top of the upper gel solution to f arm 16 lanes leaving greater than 1 cm of upper gel from the bottom of the lane to the top of the lower gel. The upper gel and comb was cleared of bubbles and allowed to polymer- ize for a minimum of 60 minutes. While the gel is polymerizing 1 liter of running buffer is pre- pared. Running buffer contains 0.025 M tris, 0.192 M glycine and 0.1% SDS. This can be stored as a stock solution at 10 times the above concentration at a pH of 8.3. To run gels the comb and gasket was removed and the plate placed in the holder. Sufficient running buffer was added to cover the gel’s top and bottom chambers. Bubbles were removed from above and below the gel’s surface to ensure even resistance across the gel. A total of 20 ul of sample was loaded into each lane using an Eppendorf pipette. Samples contained 10 ul of sample buffer composed of: 30 mM tris, 9% SDS, 15% glycerol, 5 mM ME and 0.05% bromphenol blue. Up to 10 ul of protein solution sample could be added to each well. Molecular weight standards contained the Biorad high molecular weight standard. Table 5 lists the contents of Biorad high molecular weight standard. 55 TABLE 5 1 1 h n Protein Molecular Weight (Daltons) Myosin (rabbit skeletal muscle) 200,000 B-galactosidase (E. coli) 116,250 Phosphorylase (rabbit muscle) 97,400 Bovine serum albumin 66,200 Ovalbumine (hen egg white) 45,000 Gels were run at 20 milliamps using a Gelmen power supply f or 3-4 hours. The power was turned off and the gel removed and placed in a staining solution containing 0.05% coomassie blue and 25% TCA. The gel was stained for 30 minutes with constant shaking. Destaining was accomplished by soaking the gel in a 10% methanol, 10% acetic acid solution for 10 hours. Gel drying was accomplished by using a Slaborel drying unit from Ann Arbor Plastics/Scientific Products (Ann Arbor, MI). This procedure used cellophane sheets which were soaked in the destaining solution and placed on either side of the gel. This sandwich setup is allowed to dry for 6-8 hours in a well ventilated warm and dry area. After drying the excess cellophane was cut away leaving the dried gel. Silver staining of the SDS-PAGE gels was performed using the 56 Gelcode silver stain kit from Pierce Chemical Company (Rockford, IL). The kit comes complete with 4 bottles containing the concentrated reagents. Reagent one contained the silver stain reagent. Reagent two contained the reducer aldehyde. Reagent three contained the reducer base. Reagent four contained the stabilizer base. The specific contents and concentrations of these reagents are proprietary. Reagent one was diluted 1:15 (v/v) with deionized water for use. Reagents two and three were diluted 1:7.5 and reagent four 1:45 for use. Prior to silver staining the gel was fixed with 50% ethanol and 5% acetic acid. This can be done to rehydrate a dried gel with a re- hydration time of 12 hours. The gel was washed with deionized water for three hours and stained with diluted reagent one for one hour. Reagents two and three were mixed immediately prior to use and the gel soaked for 10 minutes in them. The gel is rinsed 3 times, for 60 minutes each time, in the diluted reagent four and dried using the technique described above. Following electrophoresis the stained gel was analyzed with a densitometer to determine the relative purity of the enzyme. Each lane was scanned for the length of the lane over a width of 3.2 mm. Figure 11A (Results Section) is a scan taken from the lane containing the fraction taken from the crude homogenate. Figure 11B is taken from the lane containing purified BBCK. Figure 11C is taken from BBCK purchased from Sigma Chemical Company. Molecular Weight Determination Molecular weight of the monomeric BB creatine kinase was determined by plating the molecular weight vs. migration distance on 3 cycle semi log paper. The migration distance of the purified protein can thus be used to determine the monomeric molecular weight using interpolation. See Appendix III. lsoenzyme Determination Fractionation of the three cytosolic CK isoenzymes was performed using the Cardiotrac Fluorometric method by Corning (Corning NY). This method involves separation of the isoenzyme by agarose electrophoresis. Using agarose film supplied in the kit approximately 1 ul of sample is applied to the film in the preformed well. Elec- trophoresis is performed after each chamber is filled with running buffer which contains the Barbital buffer supplied in the cardiotrac kit containing a final concentration of 0.05 M sodium barbital with a pH of 8.6. The film is placed in a horizontal chamber with the sample well closest to the cathode. Power was turned on and ran for 20 minutes at 90 volts using a dedicated Corning Power supply (Corning, NY). When electrophoresis was complete the excess buffer was removed and pre- soaked substrate paper placed over the gel. The substrate paper was soaked for 30 minutes in a solution containing 1 ml of MES buffer added to l vial of substrate reagent. The MES contained a proprietary 57 58 concentration of morpholino ethane sulfonic acid at pH of 6.2. The lyophilized reagent vial contained; PC, ADP, glucose, hexokinase, NADP, glucose-6-phosphate dehydrogenase and AMP. Following the mixing procedures described the final concentrations of the reagents will be: 90mM PC, 12 mM ADP, 60 mM glucose, 9000 IU/liter hexokinase, 6 mM NADP, 7500 IU/liter glucose-6-phosphate dehydrogenase, and 15 mM AMP. The agarose gel and substrate paper were incubated for 30 minutes. Activity of CK was measured by the production of NADPH using the following series of reactions: PC + ADP + W «- Creatine + ATP ATP + Glucose 0 ADP + G-6-P G-6-P + NADP 0 NADPH + H3" + 6-phosphogluconate After drying, the production of NADPH is measured by a Beckmen Apprase Densitometer (Beckmen Instruments) which will scan the gel for f luore- scent changes occurring at 365 nanometers from the NADPH produced during the incubation. This procedure will give total CK for a given lane and activity vs. distance migrated. The fractional distances migrated are divided into 3 regions according to the migration char- acteristics of the isoenzymes to give the ratios of activity for CK isoenzymes in the sample. Methods For Kinetic Activity In Solution Protein concentrations of purified enzyme was determined using the Biorad dye binding assay. See Appendix I. This assay uses the prin- cipal of the Bradford assay (147), a colormetric measurement of Coomassie Brilliant blue. This method has preferential binding of protein and dye to f arm a blue complex with an extinction coefficient much greater than the free dye. The advantages of this technique are that few impurities or solvents effect the results and measurements can be made immediately. Specific activity of the enzyme was calculated by determining the rate of ATP production. Where one unit of activity is defined as the amount of enzyme which will convert 1 1.1 mole of ADP to ATP per minute under standard conditions. The standard conditions are 37°C and sa- turating conditions of substrate. The stock solution for kinetic experiments consisted of 50 mM tris, 10 mM MgC12, 25 mM ME 0.1% BSA at a pH of 7.5. The reaction was started at time zero by addition of substrate. If the forward reac- tion was being examined, PC was added to a solution containing 1 mM ADP in the presence of 33 ug/ml enzyme. To examine the reverse reac- tion the reaction was started by the addition of ATP to a solution of 20 mM Cr. All kinetic experiments were carried out at 23°C with a total volume of 1 ml. The reaction was stopped by addition of 0.5 ml of 3 N perchloric acid. One minute later, the pH was brought back up to approximately 8.0 with 0.5 ml of 3 N KOH and 1 M TEA. 59 60 The curve seen in Figure 12 (Results Section) is representative of the progress of the reaction under saturating conditions of substrate versus time. The time points seen in this curve are 0, 5, 20, 60 and 180 seconds. This reaction was started by the addition of 10 mM PC to a solution of 1 mM ADP and measuring ATP production. This curve was used to determine that the appropriate time to stop the reaction for the kinetic experiments was 20 seconds. ATP production was used to measure the forward reaction while dis- apprearance of ATP was used to measure the reverse reaction. ATP was measured spectrophotometrically using the coupled enzyme assay des- cribed above to produce NADPH. NADPH production was measured by com- bining 2 mls of assay stock solution, 0.5 m1 of sample, 25 ml of 50 mM NADP and 10 m1 of a solution containing 0.1 U/ml G6PDH and 0.1 U/ml of Hexokinase. The assay stock solution contained 50 mM tris, 1 mM MgC12, 1 mM glucose, 0.5 mM ME at a pH of 8.0. This reaction was allowed to come to equilibrium for 30 minutes and spectrophotrometric measurements made at 340 nm. See Appendix 11. Determination of Solution Kinetics Kinetics experiments on purified BBCK were performed in 1 ml of solution and stepped with 3N perchloric acid after 20 seconds. The determination of the forward kinetics was performed in a reaction solution containing 1 mM ADP, kinetic stock solution, and 33 ug/ml of BBCK. The reaction was initiated with the addition of varying amounts of PC. The PC concentrations used produced a final concentration in 61 the reaction system of 20, 10, 5, 2, and 1 mM and were all performed in duplicate. Following the additions of acid and base the solution was kept at 4°C until ATP measurements. ATP measurements were accom- plished within 24 hours of the kinetic experiments. The reverse kinetics were determined by using a solution containing the kinetic stock solution and 20 mM creatine as well as 33 ug/ml of enzyme. The reaction was started with the addition of varying con- centrations of ATP. The final concentrations of ATP were 60, 20, 10, 5, and 2 mM The progress of the reaction was determined by measuring the ATP concentration present in the kinetic solution and calculating the disappearance of ATP. Kinetic Calculations K m and Vnlux were determined from the Lineweaver-Burke, Hanes-Woolf, and Hofstee plots. To accomplish this, forward and reverse data were plotted by 1 /u vs 1/[S], [Sl/u vs [S], and u vs u/[S] respectively. A multiple determination of the Km and Vmum as described above produces a range of values for Km and Van“ which are reported in the Results Section. NMR Methods The arteries were kept at 4°C in PSS until the time of the experi- ment the following day. Two to three arteries, containing a total tissue weight of 2 grams, were cleaned, cut horizontally into rings, and placed into a 10 mm NMR tube. Arteries were superf used from the bottom of the NMR tube and the solution removed from above the tissue and recirculated at 37° C. The solution contained regular PSS except for the following changes: 0.1 mM NHZPO4 and 19 mM NaHCO3. The low NaH2P04 concentration decreases the contribution of perfusion Pi to the Pi signal observed from the tissue. The NaHCO3 concentration is decreased because the experiments were performed at 37°C using a bicarbonate buffered solution with continuous bubbling by 95% Oz, 5% C02. Perfusion was accomplished by a Harvard peristaltic pump with a flow rate of 10 ml/min. The Tygon tubing used was approximately 3 meters long because the perfusion apparatus will not operate in close proximity to the magnet due to the strong magnetic field. To prevent solution cooling while in the perfusion lines the Tygon tubing was jacketed with silastic tubing perfused with 37°C water from a Haake constant temperature pump. The infusion line was jacketed in this manner up to the bore of the magnet. The bore temperature was main- tained at 37°C with forced warm air. Siphoned solution was returned via unjacketed tubing to a 1 liter flask of PSS. PSS was continuously bubbled at 37°C with 95% 02, 5% coz. 62 63 The field was optimized by shimming on the 1H resonance at 400.131 MHz. The perfused tissue was allowed to equilibrate for 2-3 hours before any experimental measurements were obtained. During the equi- libration period pre-control spectra were continuously taken to ensure tissue stability and viability. Control spectra consisted of 200 summed scans collected 15 seconds appart using a 18 usecond pulse width (PW). DANTE Methods The specific methods described by Ugurbil and lngwall were not available for these experiments. Therefore, a different technique was used which employed both CW and DANTE saturation. Both hard (DANTE) and soft (CW) pulses are utilized simultaneously to generate the two tailored excitations. The DANTE pulse sequence described by Freeman and Morris generates selective excitation by a regular sequence of identical, short radio- frequency pulses. The pulse width (a) is very short and the nutation time (t) between the pulses is also short. This pulse sequence will produce excitation which is offset from the transmitter frequency by an amount: Au - 1/(a+t), 3 where a is the pulse width and t is the nutation time between pulses and Au is in Hz. The elicited excitation will occur at the center 64 frequency and at intervals of Au. The excitation produced by DANTE at the offset frequencies described by Au occurs by a cumulative effect of the short hard pulses. Before the first pulse, the net nuclear magnetization vector M is at its Boltzman equilibrium value, M0° During the first pulse the magnetic moment vectors of the nuclei are tipped away from their equilibrium value along a particular trajectory. The next pulse tips vectors of those nuclei at the Au resonance farther away from M0 along the same trajectory on the unit sphere of the nuclei. Vectors of nuclei which are not at the Au frequency are tipped away from Mo but along different trajectories. The differences in the trajectories depend upon the value of Au. Subsequent pulses cause the vectors for nuclei at the Au frequency to be tipped again along the same trajec- tory away from Mo while other vectors are tipped along their shifted trajectories (Figure 4 in Background Section). The continued appli- cation of the short, hard pulses will flip the on-resonance vectors along the same trajectory. Of f -resonance magnetization vectors will continue to have their trajectories shifted such that they will reach a steady state with their resultant vector near MO. In this way, nuclei which resonate at frequency Au will have their magnetization vectors flipped by the cumulative effect of the pulses, while of f -re- sonance magnetization will remain near Mo-thus producing a selective saturation at the on-resonance frequency. Because the DANTE sequence was developed to selectively excite a spin vector such that the net effect approximates a 90° or 180° pulse, 65 it had to be modified to produce a continuous-wave type of saturation. This was accomplished by making the number of pulses (11) very large such that the on-resonance vector is continously saturated. The objective of the MST experiment was to saturate one resonance of interest with DANTE while saturating another resonance with CW, and to do it such that the times required for equivalent saturation be equal. To demonstrate that the small alteration in the pulse train did not alter the magnetization effects shown to occur by Freeman and Morris, the Bloch equations were solved incorporating the changes discussed above. It is assumed that the magnetization can be represented by Mx, My and M2 where M2 is along the Mo axis. Initially the magnetization vectors will be along M2. The net magnetization is represented by its components in the x, y and z coordinate system as follows: Mz(t) = Mz'cosa - My'sina + Mo[l - exp(-t/I‘l)] My(t) . (My' cosa cosAot + Mz‘sina cosAmt - Mx'sinAmt)exp(-t flz) Mx(t) - (Mx'cosAmt + M 'cosa sinAtot + Mz'sina sinAmt)exp(-t/l‘2). Y Suppose a is applied and is small such that cosa=1, sinus a, and M2' 5M0. Then, 66 M20) - Mo. My(t) - (My'cosAtot + MoacosAot - Mx'sinAmt)exp(-t/T2), and Mx(t) - (Mx'cosAcot +My'sinA0t + Mo asinAmt)exp(-t/T2). Combine the terms Mx and My as follows: Mxy(t) - Mx(t) + iMy(t) =- (Mx'cosAmt + My'sin Amt + MO asinAwt)exp(-t /'1'2) + i(My'cosArnt + Mo acosAmt -M'sinAmt)exp(-t /T2) Which simplifies to: Mxy(t) =- Mx'(cosAmt - isinAot)exp(-t /T2) +iMy'(cosAmt - isinAmt)exp(-t /'1'2) +iMoa(cosAaJt - isinAcot)exp(-t/T2). =(Mx' + iMy' + iMoa)(cosAut - isinAwt)exp(-t /1' 2). Then by identifying Mxy . and by the definition of exp(-1'Aa)t), Mxy(t)-(Mxy' + iMoa)exp-iA¢0t exp(-t /l‘2). In a departure from the reasoning followed by Morris and Freeman (100) 10 11 12 13 67 which applied a cumulative 90° or 180° pulse, an extended series of small 11 pulses is applied. The sum of these pulses is greater than the 90° or 180° and is applied over an extended period of time similar to a CW saturating pulse. Mxy will be expressed by: Mxy(t) - iMo exp [-(t-t m)/r2] exp [-i Adt-tmfl 10:, exp [-(tm-tp/rz] cxpl-iAwm-mh +5" {am-1 exp[-(tm-tm-l )/l'2] exp[-iA01(tm-tm,1)]) +{cxm exp[-(tm-tm)/T2] exp [-iAth-tmfln 14 Expression 14 can be simplified to Mxy(t) - iMo Eon [exp-(tm-tnv‘l‘z] exp[-iAtu(tm-tn)] 15 Remembering that a is constant and small, Mxy(t) - 1M0 {ancxp [-(tm-tn)/T2] exp [-1 Album-tn) ]. 16 where Z is from n-l to m. Assume t > tm, and that t is large, then am CXDL-(t-thTz] cxpl-z‘Am-tmn z 61 exp [-(tm-tn)/T2] exp [-i Adam-tn) ], 17 which upon insertion into equation 16, yields 68 Mxy(t) - iMo 2a,, [exp-(t-tn)/T2] exp[-iA01(t-tn) ]. 18 Equation 18 demonstrates that the modified DANTE pulse sequence will sa- turate a resonance of interest with the same results as those obtained by Freeman and Morris (100). The modifications to the DANTE pulse sequence used in this thesis thus do not alter the results demonstrated by Freeman and Morris. Therefore, when this modified sequence is applied to a resonance for an extended period of time, the resonance will be saturated, as explained by the above solution of the Bloch equa- tions. A two site saturation experiment, which applies the above pulse sequence in conjunction with CW saturation, is thus possible. Therefore, Ugurbil’s MST experiments may be performed by using a modi- fied DANTE pulse sequence instead of dual CW saturation. . NMR Solution Methods Control experiments to demonstrate DANTE were performed on a solu- tion containing 2.5 mM ATP, 5.0 mM PC, 0.6 mM ADP, and 30.0 mM Mg+ at pH 7.6 with 10 mM MOPS (Figure 17 in Results Section). These concentra- tions are supra-normal for smooth muscle but higher phosphate concentra- tions allow rapid collection of spectra which also allows for quick interpretation of spectra and facilitates the process of optimizing pulse conditions. For all solution spectra 4 dummy scans were used and 16 scans for collection with 15 second cycle time between transients. For complete saturation, the DANTE pulse was generated by applying 69 50,000 (n) pulses with a pulse width (11) of 0.7 u seconds. The delay between pulses (t) was 0.00029 seconds or 290 1.1 seconds. This cycle time would generate a saturation at the center frequency and at a f re- quency 3440 Hz (18.78 ppm) away and continue to repeat at this frequency (Av) Au - 1/(¢x+t) 19 The pulse sequence described above will generate a train of pulses which will last 14.6 seconds in duration. During the pulse train the CW sa- turation can be accomplished. The duration of the entire saturation in- cluding DANTE and CW saturation is 15 seconds. The DANTE pulse sequence was applied to the PC peak with CW saturation to the 7ATP (Figure 178). This was done to demonstrate that DANTE and CW could be applied sim- ulatneously and compared to Figure 17A where no saturation was applied to any resonance. To determine if the DANTE and CW saturation had similar effects, both DANTE and CW were applied to PC for 3 seconds. The 3 second saturation was applied at the end of a 15 second saturation of 7ATP. To accomplish DANTE saturation for 3 seconds, a saturation was applied to the PC peak with 10,000 pulses using a pulse width of 0.7 11 seconds and a 290 mili second interpulse delay (Figure 17 E and F). The 3 second saturation from CW was accomplished by starting the CW satura- tion of PC after 40,000 pulses of DANTE saturation of 7ATP and then com- pleting 10,000 pulses of DANTE for the remaining 3 seconds (Figure 17E) producing incomplete saturation. In both spectra with the short 70 duration of saturation the saturation is incomplete to a similar extent. In Figure 17D PC is completely saturated with DANTE and 7ATP is not sa- turated. In this spectrum CW saturation was applied 410 Hz upfield of 7ATP (VATP and PC are 410 Hz apart). In Figure 17E 7ATP is completely saturated with DANTE saturation while PC remains unsaturated. CW sa- turation was applied 410 Hz down field of PC. Figures 17 C and D were performed to demonstrate that DANTE saturation could be selectively applied to either peak of interest and completely saturate said peak. Methods for T1 Collection For Tl experiments the arteries were positioned in the magnet as described above. They were unstimulated at 37°C, and perfused with regular PSS for the entire experiment. The tissue was allowed to equi- librate in the NMR for 2 to 3 hours. During this time control spectra were collected. After tissue stability was confirmed from the control spectra, the 90 degree pulse was determined. The 90 degree pulse was determined by estimating a 180 degree pulse and pulsing the tissue using pulses approximately one half the 180. By systematically changing the pulse width the 90 degree pulse was chosen by being the pulse width which generated the spectrum with the greatest peak height for 7 ATP. Once the 90 degree pulse was found the T1 exper- iments could be performed on the tissue using the technique of progressive saturation. 71 Progressive saturation was performed to determine the spin lattice relaxation time (T1) of the phosphorus nuclei in the presence of ex- change. This technique is frequently used f or the measurement of T1 in biological systems because T1 can be determined quickly (47). The dis- advantage to this technique is it requires a very accurate 90 degree pulse. The progressive saturation technique involves giving a series of five pulses and collecting the signal after the train of pulses. The duration between the pulses is kept constant until the desired signal to noise is obtained. Then the period of time between the pulses is changed and the process is repeated. The series of time intervals or delays used to obtain the T15 in this thesis is shown in Table 6. 72 TABLE 6 P 1v r i x r1 11 Series Number Delay sec. 1 .0001 2 .001 3 .01 4 .1 5 1.0 6 2.0 7 5.0 8 10.0 9 15.0 I An acquisition delay of 0.204 seconds is added to each of these times. As the delay is increased the recovery of the nucleus will become more complete. The extent of the recovery can be plotted as delay vs. peak intensity (Figure 15 in Results Section). This plot can be fit to the formula I = A + B exp (-t/Tl) 20 73 Where t stands for the delay and I is the peak intensity and A and B are constants which are fit to the data. This three parameter fit helps correct for small errors in the 90 degree pulse which would otherwise cause errors in T1. The T1 for each of the phosphorus nuclei can be determined from a single progressive saturation experiment. However, because of time con- straints when using a living tissue, the saturation transfer experiments must be performed on another tissue. Therefore, for rate constant calculations an average T1 value must be used and not the T1 of the in- dividual rate constant calculation. Using an averaged T1 for all the kinetic experiments assumes that T1 doesn’t change from experiment to experiment. Tl can be effected by several parameters. To reduce changes in T1 from experiment to experi- ment, several precautions have been observed in this thesis. Because field strength can effect T1 (57,101), the same NMR specrtometer was used for all T1 calculations and saturation transfer experiemtns. Changes in peak position and intensity can also effect T1, so unsatur- ated spectra were obtained at the end of the T1 experiments to demonstr- ate no significant changes in the tissues’ spectra. If a decrease in peak area of 10% or more was observed in PC peak area, the experiment was discarded. An inaccurate 90 degree pulse is known to cause changes in T1, so for every Tl experiment a 90 degree pulse was determined. Spectra for T1 experiments were transformed using line broadening 74 corresponding to 20 Hz. It is asssumed that these precautions will decrease changes in T1. The T1 experiments consisted of 360 summed FIDs collected over a spectral window of 10,000 Hz. The last spectrum collected had a delay of 15 seconds. This was transformed and compared to the last control spectrum before the experiment commenced. This comparison was needed to ensure there were no metabolic changes during the experiment which might have effected the high energy phosphate content and result in an in- accurate Tl. Experiments were discarded if changes of 10% or more in PC peak area were observed. The series of spectra were all transformed with identical process- ing. After transformation, the series of spectra could be used to calculate individual Tls of any peak. The calculation of T1 was per- formed by a best fit method as described in the previous section. Calculation of T1 in the Absence of Exchange The T1 in the absence of exchange can be calculated from equations in (97): l/r - 1/‘1'1PC+ k1 and 75 t MPC - Mp5 /(1 + leipC). r is the longitudinal rate constant measured in the presence of ex- change, TlPC is the longitudinal rate constant in the absence of ex- change, k1 is the pseudo f irst-order rate constant for the exchange between PC and VATP (k1 is k{ in this thesis), MPC is the steady-state PC magnetization when YATP is saturated, and MPC is the steady-state magnetization of PC when there is no saturation. r, MPC’ and MPC were measured experimentally. k1 and TlPC can then be calculated from the above independent equations, each having the same two unknowns. Similar equations can be used for 7ATP and Pi. During the estimation of the Tl‘yATP’ the exchange with Pi was assumed to be negligible at rest rel- ative to the exchange with PC. This assumption was subsequently found to be experimentally valid, since even under stimulated conditions the exchange between Pi and YATP is low relative to the CK exchange. For the Pi measurements, the saturated state occured when both PC and VATP were saturated, while the measurement of Pi magnetization in the absence of 7ATP saturation retained the PC saturation. The more rapid relaxa- tion of VATP means it will be least effected by exchange relaxation. Similarly, a low change in magnetization upon saturation results in little change in the T1 estimation from the exchangable condition. The calculated values of the rate constants and T1 values in the absence of exchange are presented in the Results Section. NMR Kinetic Experiments All NMR experiments were performed in a Bruker wide bore 9.4 Tesla NMR spectrometer. The signal was optimized by shimming on the 1H re- sonance at 400.131 MHz. 31P experiments were performed at 161.97 MHz. The perfused tissue was allowed to equilibrate for 2-3 hours before ex- periments were started. During this time, control unsaturated spectra were obtained. These spectra consisted of 200 summed scans collected 15 seconds apart using the 90 degree pulse. All data were stored in an Aspect 3000 computer interfaced with the spectrometer. Stored free in- duction decays (FIDs) were Fourier transformed using exponential line broadening corresponding to 35 Hz. The transformed spectra from the control experiments were compared to each other with regard to time. These spectra demonstrated the viability and stability of the tissue. The tissue was considered stable if the high energy phosphate spectra remained unchanged for 1 hour. The tissue was considered viable if the PC:ATP ratio was greater than 0.4. The estimated concentrations of ATP and PC present in the porcine carotid artery has been reported as 0.70 and 0.51 umol/g tissue wet weight (38). If either of these criteria were not met the tissue was removed and replaced with fresh tissue. 76 NMR Peak Integration Peak integration was accomplished by utilizing the standard in- tegration routine available on the DISNMR software supplied by Bruker Instruments with the Aspect 3000 computer. Integral regions were demarcated with the cursor using a representative spectrum from the series of experiments. Starting immediately downfield of the peak of interest the cursor position was marked. Next the cursor was moved im- mediately upfield of that same peak and the position marked. The cursor is then moved downfield of the next upfield peak and the marking process repeated. Once all the desired peaks have been demarcated the areas for the integral of each spectra can be obtained using the same demar- cations. Phasing and processing of every spectra within an experiment was repeated in the same fashion. During the experiments there was no significant shifting of any of the peak positions such that a peak of interest would fall outside of the demarcated region. Integrals for the peaks of interest were recorded and used to calculate the rate con- stants. CST Methods Conventional saturation transfer (CST) experiments were performed following procedures detailed by Brown, (18,19), and Meyer (97). The resonant frequency of interest was selected and saturated with a monochromatic radio frequency pulse. This results in absence of signal 77 78 at the saturating frequency following a 90 degree pulse. The cause for the absence of signal is a randomization of nuclear spin at that f re- quency. Saturated nuclei maintain the randomization of spin even if they undergo chemical exchange. Exchanged randomization from one nucleus to another during a chemical reaction will effect net magnetiza- tion at that resonance accordingly. The change in magnetization can be observed by a diminution of the second resonance. This decrease in magnetization at the second resonance is used to calculate the pseudo first order rate constants (kf and kr) of the CK kinetics between ATP and PC. When ATP is saturated, the forward rate constant (kf) is determined using the relaxation time (T1) in the absence of exchange for PC and the ratio of the change in magnetization of PC. In this way the kf for PC degradation can be determined: kf -(1/r1m)- (1 - M‘/M°). 21 Where: Tlm is the spin lattice relaxation time of PC in the absence of . o o I o a o a * exchange, Tl - Tlm/(M /M ), M - magnet1zation wrthout saturation, M = magnetization with saturation of PC. If PC is saturated, then the reverse rate constant (k1,) is determined. Control spectra are obtained by saturating off resonance of PC and 7ATP. The experimental protocol was to obtain unsaturated control specta 79 and determine the chemical shift distance between PC and YATP. This chemical shift distance was usually 410 Hz. Therefore, 410 Hz downfield of PC is saturated for the first experimental spectrum. PC and YATP are then saturated. Finally, 410 Hz upfield of YATP is saturated. The sa- turation downfield of PC is used for M° during kf calculations. The sa- turation upfield of VATP is used for M° during kr calculations. The flux can be determined from the product of k and the metabolite con- centration. The flux ratio is obtained by the product of the pseudo first order rate constant ratios and the peak area ratios. The flux ratio can be used to determine the equilibrium state of the reaction. A reaction at equilibrium will have a flux ratio of one. Spectra were obtained by saturating with a monochromatic continuous wave pulse from a computer controlled Bruker BSV3 synthesizer (Bruker Instruments, Billerica MA). The saturating resonance was sent through the same coil used for pulsing and collecting. The duration of saturation was 15 seconds. This time is sufficient to produce complete saturation of the peak of interest and full relaxation of other peaks between pulses. Each spectrum consisted of 360 summed FIDs. The saturating pulse re- mained on at all times during the experiment except during pulse and collection. Saturation consisted of 0.4 Watts, which was sufficient power to saturate the PC and VATP peaks. The pulse width for collection was 18 useconds which corresponded to the 90 degree pulse over a spectral window of 10,000 Hz. MST Methods To determine if the ATPase rate can be observed in the carotid multisite saturation transfer (MST) was performed. The carotids were loaded with creatine to increase the PC concentration in the cells and prevent PC depletion during stimulation. Creatine loading has been demonstrated to produce BOO-400% increase in intracellular PC concentra- tion (135). The loading is accomplished by soaking the arteries in 40 mM creatine overnight. The creatine loading solution contained 96 mM NaCl, 5.4 mM KCl, 25.3 mM NaHCO3, 1.1 mM NaH2P04, 2.5 mM CaClz, 1.2 mM MgSO4, 0.01 mM EDTA, 15 mM glucose, 40 mM creatine at 4°C. Using this loading procedure PC content increased by six fold in the carotids (Figure 19). To further increase ATPase activity, the arteries were stimulated with 75 mM KCl during the NMR experiments. The arteries were prepared for the experiments as described above. The perfusion solution however consisted of 75 mM KCl along with 40 mM creatine and 19 mM NaCl. The other crystalloid concentrations were the same as listed above. Because of ATPase activity the CK reaction might be considered as a three site exchange (see Background Section) thus multisite saturation transfer would be necessary to accurately determine CK kinetics. The experimental protocol was to prepare the tissue as described above to increase PC concentration and perf use with the potassium stimulating 80 81 PSS. Control spectra were collected for 2-3 hours to determine a steady state for the tissue with regard to potassium stimulation. Multisite saturation was accomplished using the combined continuous wave and modi- f ied DANTE technique described previously. The saturation protocol is listed in Table 7. 82 TABLE 7 W Collection # DANTE Saturation CW Saturation 1 PC Downfield of PC 2 PC Pi 3 PC 7ATP 4 PC Upfield of YATP 5 Downfield of PC Upfield of YATP To determine kf from an MST experiment the equation: k, - (a,/o)[(M;,-Mi," )/MI,] 22 was used. In this expression Mfr is the magnetization of ATP with PC sa- turated which is determined from Collection 1, Table 7. The parameter a t is defined by the relationship M9, Pf - 0[M7/MP: ] - kf/kr. The parameter «7 is equal to T11;1 plus the sum of the degradation rate 83 constants of 1ATP. 'I‘l.’,'1 is the spin-lattice relaxation time of YATP in the absence of any exchange. (1.), is the spin-lattice relaxation time of YATP while Pi and PC are saturated and can be calculated as described t a above for the CST experiments. Collection 2 is used to determine M the magnetization of 1ATP when Pi and PC are saturated. The Tl’s in the absence of exchange previously calculated were used for all MST kinetic caluclations. To determine kt, Collection 4 was made to determine the it 7 when YATP and PC are saturated. Collection 5 was obtained to confirm value for Mfrwhen PC is saturated, and Collection 3 was used for M tissue stability by demonstrating less then 10% change in the high- energy phosphate metabolite concentration from pre-control values. These measurements are made by defining ATP $Pi as kf and Pi $ATP as k1,. Purification Results To determine the isoenzyme activities present in the carotid artery, the f luorometric cardiotrac assay method by Corning (Corning, NY) washemployed. This method separates the isoenzymes with agarose electrophoresis and measures the activity of the individual isoenzymes using the coupled enzyme assay (discussed in Methods Section) with the production of NADPH measured fluorometrically. The migration by the enzyme closest to the anode is typical of the BB isoenzyme. The MMCK stays closest to origin at the cathode, and MBCK is observed between MM and BB. Figure 6A is a record of the fluorometric absorbance of a control fractionation with known amounts of MM, MB and BBCK. It shows a peak migrating with the characteristics of MMCK and trace amounts of MB and BBCK. Figure 6B is a record of the same control with the puri- f ied carotid BBCK added. The specific activity of the enzyme was determined to be 84 units/mg of protein. 96.9% of the activity added to the assay system came from the BBCK isoenzyme. While the remaining 3.1% came from MMCK and no detectable MBCK activity was observed. Figure 7 is a trace obtained from a fraction of the first homogen- ate. This represents the distribution of CK isoenzyme activities 84 85 Figure 6. Fluormetric measurementof CK isoenzymes using production of NADPH from the coupled enzyme assay. Figure 6A is a plot of NADPH production vs. migration of agarose (see text). This plot is from a control serum with known MM activity. It contains 86.7% of MM, 4.6% MB, and 8.6% BB respectively. Figure 6B is a plot from purified CK run along with the control. Peak BB demonstrates an increase in CK activity and migrates according to the criteria for BBCK. This plot contains 63.4% MM, 1.5% MB, and 35.2% BB. Subtracting Figure 6A from 68 activity gives 3.1% MM no significant MB and 96.9% BB activity. I: II. II .. ,‘IOIDIIIECI-'II.| >.—. .mZuhZ. w) .h<.—m¢ MIGRATION DISTANCE 86 87 RELATIVE INTENSITY MIGRATION DISTANCE Figure 7. This figure is generated by the same technique used in Figure 6 but is from the crude homogenate without added MMCK. It shows the ratio of isoenzymes coming from the porcine carotid arteries as 23.5% MM, 40.5% MB, and 36% BB. The specific activity for this plot was calculated to be 0.817 U/mg of protein. to!»---n-----~lv¢o ‘~l’~v~'- "~ 5 - " O"----Iv 88 O o .Mm~ W O RELATIVE INTENSITY MIGRATION DISTANCE Figure 8. This figure is the isoenzyme fractionation of CK taken after the first alcohol precipitation. This figure was generated using the agarose electrophoresis technique used to obtain Figure 6. The ratio of the peak areas for BB, MB and MMCK was 2:1:0.2 after the second alcohol precipitation. 89 Figure 9. This figure is the elution profile of protein concentration and kinetic activity taken during the elution of the protein off the hydroxylapatitc column. Where: a is protein concentration and O is specific activity of the enzyme eluted. Samples were taken at 7 ml in- criments off the column which had a flow rate of abour 14 ml/hour. \ 90 (WU/Ow) 11M 95 m2_,:c> 20:3,: OF Om GM 0% om %mm owm of” com 9: 03 cm cw o Owl - .. _. _ q A _ fi‘ * o, I \/ \ Q— ”l.\ :/.. /.I...1./ :(W.../... ..:\Omlo “0 a / I O O . / . O O 0100 /O/ \/ C. CO _ - o 000 O; \ _ om - / 80b _ a o O G _ o OO\ O m. _ m1 on m _ m U _ O\ O 8 _ Mu ow - /o m. u m. \ 0\ mm _ mm. 8 o " 235E _ cm 8 H m, c25< gngnm az< zo_._-<..,z-_z.dzoo 290% a 50% 29-5-: VS EEWZ 951$ mzamon Om (NJ/Em) NIELOEjd 91 present in the carotid artery. It shows that 23.5% of the CK activity comes from MMCK, 40.5% is from MBCK and 36.0% is from BBCK. From these data it can also be determined that the porcine carotid artery has a specific activity 0.81 U/mg of tissue protein. Figure 8 is a fractionation taken after the first alcohol pre- cipitation. It demonstrates a loss of MM and MB activity while main- taining BB activity. The resulting activity ratio is approximately 2:1:0.2 for BB, MB and MM respectively. Elution of the essentially purified protein off of the column was accomplished by using 10 mM phosphate buffer solution. This solution contains 10 mM NaH2P04, 25 mM ME and 0.5 mM EDTA. Figure 9 is a plot of the protein concentration (ug protein/ml) eluted vs. time and specific activity (U/liter) vs. time. Fractions containing 17 ug/ml protein or greater were pooled for concentration of the protein. The increase in protein concentration and activity paralleled each other. The elevated baseline for protein and activity is indicative of a loss of protein product. Such a loss will decrease the total yield of the procedure. However, no significant loss of specific activity or purity would be predicted. Figure 10 is an SDS-PAGE of the purified protein stained with silver stain. Lane A contains a fraction from the crude homogenate of the porcine carotid artery. Lane B contains the molecular weight 92 standard. Lane C contains the purified BBCK. Lane D contains puri- f ied BBCK from Sigma Chemical Co. Table 8 is a table of the results of the activity, yield, and isoenzyme ratio of three different fractions taken during the purifi- cation procedure. Fraction one is the crude homogenate. As can be seen, there is an approximately l:1:l activity ratio of the three cytosolic isoenzymes. Fraction two is taken after alcohol precipita- tion and fraction three is taken from the purified enzyme. Figure 11 was obtained by performing a densitometer analysis on the silver stained gels. Figure 11A is a tracing obtained for the lane containing a fraction taken from the crude homogenate. It demonstr- ates the presence of many soluble proteins from the carotid artery. Figure 11C was obtained from a fraction of the purified BBCK. It demonstrates one predominating band with two smaller bands of higher molecular weight. Figure 113 is taken from the lane containing BBCK from rabbit brain purchased from Sigma Chemical Company. It demonstr- ates a similar migration distance for BBCK and also similar migration distances for some impurities. Comparing Figure 11 B and C it can be seen that the BBCK from porcine carotid artery has a greater purifi- cation of the BBCK than that of rabbit brain. l 42.5 Figure 10. This figure is a representation of the electrophoresis of the purified BBCK. Lane 10A contains a fraction of the crude homogenate from the procine carotid artery. Lane 103 is the Biorad high molecular weight standard. Lane 10C is purified BBCK. Lane [DD is purified BBCK purchased for Sigma Chemical Company. 94 Figure 11. A, B and C were obtained by an Ultrascan XS scanning densitometer of the silver stained electrophoresis gels. The y axis is the absorbance reading taken from the densitometer. The x axis is the distance along the lane being scanned. The x axis corresponds increas- ing molecular weight found in the gel. Figure 11A is from the lane con- taining the crude homogenate of the carotid arteries. Figure 11B is from the lane containing purified BBCK purchased from Sigma Chemical Company. Figure 11C is from the purified BBCK from our experiments. INTENSITY L DENSITOMETER TRAQE.......... aorrom TOP 95 96 TABLE 8 E '[i . . S f BE :1; Fraction # Activity U/mg protein Ratio BBzMBzMM 0/oCK Activity 1 0.8 1:1:1 100 2 ll 2:1:0.2 12 3 84 1:0:0 8 Table 8 demonstrates that the alcohol precipitation produces a 88% decrease in total CK activity as can be seen in the differences in the percent CK activity between fraction 1 and 2. This loss of activity is coincident with the loss of MM and MB activity compared to the BB activity demonstrated by the change in the ratios. The specific activity of the purified protein produced in fraction 3 was 84 U/mg protein. Fluorometric analysis of the electrophoresis demonstrated that the purification procedure produced a significant purification in the BBCK observed and a concomitant decrease in other proteins. A comparison of Figures 11 B and C demonstrates that the purification procedure produces a greater fraction of BBCK than does the BBCK supplied by 97 Sigma. Solution Kinetic Results Figure 12A is the progress of the reaction plot determined by measuring percent ATP produced. It demonstrates sigmoidicity of the curve consistant with an allosteric enzyme with a presteady-state interval of about five seconds. Thus, velocity measurements were determined assuming a five second presteady-state interval. Figure 12 B and C are the velocity plots for the kinetic experi- ments used to produce Figures 13 and 14. Figure 12B was produced from additions of PC and Figure 12C was produced from additions of ATP. The data obtained from the solution kinetic experiments was used to generate Figures 13 and 14. Figure 13 A, B and C are the Lineweaver- Burke, Hanes-Woolf, and Hofstee plots for PC additions. From these plots Vm" and Km can be determined for CK under these solution condi- tions. It was determined that Vmu was 0.019, 0.017 and 0.017 umoles/sec respectively. The Km was found to be 1.32, 0.93, and 0.95 mM respectively. The values for the reverse reaction can be determin- ed from Figures 14 A, B and C. Again these are the Lineweaver-Burke, Hanes-Woolf and Hofstee plots produced by additions of ATP. Vmu was determined to be 0.29, 0.58 and 0.35 umoles/sec respectively. Km was found to be 19.4, 42.9 and 16.1 mM respectively (Table 9). 98 Figure 12A. Progress of the purified BBCK reaction as measured by the % of the ATP produced. Figure 128 is the velocity curve of ATP production from PC additions. Figure 12C is the velocity curve of ATP consumption from ATP additions. 99 A88 ME: om_ om_ om - _ a 11k ozfi mZO:-_00< 00 2000 0:950 .00 0-040 27.: 105 009 x 31880.0 ~0;\> 0m0.0 90.0 00.0 000.0 000.0 A q q 1 2.: 00.0 H Ex .oom\mv_orc: \1_0.0 H V8E> mZOE00< 00 2000 DECK/C LO 0.0-00 00-51.00: 000.0 0_0.0 0m0.0 <39$/SS|OL.UH) /\ 106 Figure 14 A. This is a Lineweaver Burke plot of the BBCK kinetics with regard to ATP. This plot demonstrated a vmax of 0.29 umoles/second and a Km of 19.4 mM for ATP. Figure 14 B is a Hanes-Woolf plot of BBCK kin- etics with regard to ATP additions. The Vm and Km determined from 8X this plot was 0.58 umoles/sec and 42.9 mM respectively. Figure 14 C is an Hofstee plot of BBCK kinetics with regard to ATP additions. The Vm" and Km determined from this plot was 0.34 umoles/sec and 16.1 mM respec- tively. Af12:;_e: mZ©:._00< «PE 2000 0:Oz<0 LO E041 014 d and {fill :spcc- 108 00_ 00m 00m 2E 00Nv H Ex .o@m\mo_oE3 00.0 H V5E> mZOE00< nfi< @9522 x000 0:,©z<0 00 F040 >>II 9-01 X (388) A/[div] 109 m: _, x A733 _‘.::> 0m00 A(:00 0_0.0 000.0 000.0 _ J _ 000.0 / O O 0070 00m.0 o / 000.0 1 /1 L 1 00¢.0 000.0 2...: 000— H Ex .uom\m@_o_-._-._3 90.0 H V8E> mZO_-:00< nfi< HmQCMZE .1000 0Coa<0 mo H000 mfifimLOI (DaS/SSWLUH) /\ 110 TABLE 9 Isms: 11m SLLMMAEX. PC Additions: ' Vmu‘ Km ‘1‘ R Lineweaver-Burke 0.019 1.32 0.977 Hanes«Woolf 0.017 0.93 1.000 Hofstee 0.017 0.95 0.919 ATP Additions: Lineweaver-Burke 0.29 19.3 0.905 Hanes-Woolf 0.58 42.9 0.898 Hofstee 0.35 16.1 0.489 "' umoles /sec ‘i’mM Table 9 summarizes the range of values for kinetic data. The re- sulting Km values for ATP has a relatively large range of 16.1 to 42.9 mM. It appears that the Km for ATP is much higher than the concentra- tion of ATP found in carotid artery (0.7 mM). The R values determined for each plot demonstrates values from 0.489 to 0.905 for the ATP additions while PC additions data had R values from 0.919 to 1.000. The low R value observed for the Hofstee plot makes it a questionable 111 Figure 15. This figure is the computer generated plot of delay vs peak intensity to determind the T1 of PC. The fitted curve was fit to the formula I - A + B exp(-t/l‘1). Where t stands for the delay between pulses and I is the peak intensity. The values of A and B are constants which are fit to the data. In this particular experiment the calculated value for T1 was 2.46 seconds. 112 estimator of Km and Vmu. The other methods should still be suffi- cient for estimating the kinetic data, however. The similar results and higher R values for PC additions produce strong estimates for their Km and Vm values. IX NMR Results The T1 experiments generated Tls for Pi, PC, and 7ATP (Figure 15). Though other peak's T1 is available from the experiments they are not needed for the kinetic calculations, so they were not determined. Tl values in the presence of exchange calculated for Pi, PC and VATP were 3.80 s 0.55, 2.85 e 0.53, and 1.05 2 0.11 s" respectively (n-S, SE). These vaules were used for calculating the T1 in absence of exchange and for the calculation of the forward and reverse rate constants for CST and MST experiments. CST Results The CST experiments performed produced the changes in magnetization listed in Table 10. 113 TABLE 10 M 12 in fr Ex rim nt Experiment number 1 2 3 4 5 PC area 15.5 25.9 34.8 28.9 13.4 PC‘ area 10.0 20.9 19.7 10.3 6.8 YATP area 32.9 49.1 79.3 44.8 35.6 VATP” area 31.2 47.3 68.3 34.8 30.6 " areas of PC when YATP is saturated. ** area of VATP when PC is saturated. 80.6 69.2 l‘ Values of peak areas seen in Table 10 are relative intensity values. They do not reflect actual units of concentration. The ratios of the peak areas do correspond to the ratio of metabolites in the tissue in these unsaturated spectra. The change in magnetization is indicative of molecular exchange from the saturated species. Table 11 lists the rate constants for each experiment and the mean of 0.17 z 0.04 3'1 (SE n-6) for kf and 0.12 s 0.03 s‘1 for kr (SE n-6). The in- dividual values found in Table 11 were calculated using the mean T1 f 01' 7ATP and PC in the absence of exchange of 1.2 and 5.2 seconds re- SPCCtively. These values were calculated using the solution discussed 114 in the Methods Section for estimating T1 in the absence of exchange. TABLE 11 Experiment number 1 2 3 4 5 6 Mean(SE) kf 0.11 0.05 0.15 0.35 0.19 0.19 01720.04 kr 0.05 0.03 0.14 0.24 0.14 0.14 01210.03 "' -1 seconds To determine if the CK reaction was at equilibrium it was necessary to determine wheather the flux ratios of the reactants and products were significantly different than one. This was accomplished by calculating the ratio of the rate constants and the ratio of the unsa- turated peak areas. The flux ratio is calculated from the expression: [kf x PC]/[kr x 7ATP]. Table 12 illustrates these ratios from the in- dividual experiments. 115 TABLE 12 W Experiment number 1 2 3 4 5 6 Mean(SE) Ratio of PC/‘VATP 0.47 0.53 0.44 0.64 0.38 0.37 04720.06 Ratio of kf/kr 2.28 1.44 1.09 1.45 1.37 1.35 15020.18 Flux Ratio 1.07 0.76 0.48 0.93 0.52 0.50 0.71 20.1 1 The mean flux ratio demonstrated in Table 12 is the product of peak area ratios and rate constant ratios. The value of 0.71 t 0.11 is less than one. A reaction with a flux ratio away from one is in- dicative of a reaction not at equilibrium. In the results above this could be due to an underestimation of the PC content in the cell. The small peak height of PC and the signal to noise associated with the spectrometer can produce an underestimation of the integrated area of PC (96b) and thus lead to a flux ratio less than one. MST Results Figure 17 shows spectra from the ATP-PC solution with multiple sa- turation sites. Figure 17A shows the control spectrum. In Figure 173, the DANTE pulse sequence was applied to the PC peak and CW sa- turation to the VATP. This was done to demonstrate that DANTE and CW can be applied simultaneously. In Figure 17C, VATP is saturated using DANTE, while CW is applied 410 Hz downfield of PC. In Figure 17D, PC is saturated by DANTE, and CW is applied 410 HZ upfield of 7ATP. To determine if the DANTE and CW saturation had similar effects, both were used to partially saturate PC during complete YATP saturation. CW saturation was applied to PC for three-seconds while DANTE was applied to YATP for 15 seconds, and then DANTE was applied to PC for three-seconds while CW was applied to 1ATP saturated for 15 seconds. The three second saturation of PC was applied during the final three seconds of the 1575econd saturation of YATP. The three second satura- tion using CW was accomplished by starting the CW saturation of PC after 40,000 pulses of DANTE saturation of YATP, and then completing 10,000 pulses of DANTE for the remaining three seconds (Figure 1713:). To accomplish the DANTE saturation for three-seconds, 10,000 pulses were applied to PC with a pulse width of 0.7 useconds and a 290 millisecond interpulse delay during the final three seconds of the 15- second YATP saturation by CW (Figure 17F). Tissue saturations, of non-creatine loaded unstimulated arteries, are shown in Figure 18 to demonstrate the ability to saturate multiple peaks in a tissue. Figure 18A shows a control spectrum where DANTE 116 117 and CW are applied 410 Hz downfield of PC and 410 Hz upfield of VATP, respectively. In Figure 18B, PC is completely saturated by DANTE and 1ATP is not saturated. In this spectrum, CW saturation was applied 410 Hz upfield of 7ATP. In Figure 18C, PC is saturated by DANTE and Pi is saturated by CW. The residual peak observed at the Pi resonance may be due to the Pi in the perfusion solution which enters the observable region of the spectrometer without being sufficiently irradiated to effect complete saturation. In Figure 18D, YATP is com- pletely saturated by CW saturation while PC remains saturated with DANTE. Figure 1813 has CW and DANTE saturation 410 Hz downfield of PC and 410 Hz upfield of 7ATP. Creatine loading produced an increase in the PC concentration of six fold (Figure 19). The results from the MST experiments, which were performed on the creatine loaded and potassium stimulated arter- ies, are summarized in Table 13. The data indicate that the reaction of ATP to Pi (kf) was 0.07 t 0.03 s'1 and 1tr of 0.005 t 0.004 s“. 118 Table 13 Experiment number 1 2 3 4 Mean (SE) kf’ 0.20 -0.02 -0.07 0.17 00720.08 k; 0.013 -0.002 0.007 0.0 0.005 20.004 " -1 seconds The mean values from Table 12 are not significantly different than zero. Thus no significant ATPase activity is measurable using MST under the described conditions. These data were unable to determine a usable value for the ATPase rate constants. This result is due to low ATPase activity in the porcine carotid artery. The slow ATPase rate may be below the resolution of the NMR spectrometer. MST experiments were not performed to determine the CK kinetics because the PC con- centration increases during creatine loading to such a concentration that small changes in the peak area could not be accurately measured. Absence of observable ATPase under stimulated conditions and the 119 presence of measurable CK activity under unstimulated conditions may obviate the need for a three site model in vascular smooth muscle. This concept is discussed in greater detail in the Discussion Section. 120 Figure 16. The 3lP spectra of porcine carotid arteries during a CST ex- periment. Spectra were collected with 400 summed FID’s. Figure 16A is a control spectrum with saturation upfield of 1ATP. Figure 16B is a spectrum with YATP saturated. Figure 16C is a spectrum with PC satur- ated. Figure 16D is a control spectrum with saturation downfield of PC. 121 MI’E pl O