lfllllllllllzlllgllljlllllwlllllflllllllllllllzllljlzll 1'“, /-. This is to certify that the thesis entitled THE ROLE OF MONOVALENT CATIONS IN THE ACTIVATION 0F TRYPTOPHANASE: A SPECTRO- SCOPIC AND KINETIC STUDY presented by Barbara Kennedy has been accepted towards fulfillment of the requirements for Ph-D, degree in Biophysics M. A. El-Bayoumi wafliflkaw Major professor Date_QQ.LQb_eL9_._l9_Z9__ 0-7 639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. m EDIE OF MONOVLIENT CATIONS IN THE ACTIVATION 0F TRD‘I‘OPHANASEI A SPECTROSCOPIC AND KIIETIC STUDY By Berber: Kennedy A WITCH Submitted to Michigan State University in partial Mimnent or the requirements for the degree of MR OF PHILOSOPHY Department of Biophysics 1979 ABSTRACT THE ROLE OF MONOVAIENT CATIONS IN THE ACTIVATION 0F TRIPTOPHANASE: A SPECTROSCOPIC AND KINETIC STUDY By BarbaraKermedy Tryptophanase is a pyridml- '-phosphate enzyme which requires ac- tinting nonovalent cations in order to catalyze the 0‘ . fl -elinination reaction Intryptophani'fizo 7.2:? indole'0-p3rrowate-I'NH3 Different effects on the absorption spectrum of holotryptophanase are known for potassim, in which case there are two pH dependent bands at 337mandh20m, andfor sodim, inwhichcase thereisonlyahZOm band. In this study we have examined the effect of monovalent cations and tetranetlvlannonim and phosphate ions on the absorption spectrum of holotryptophanase. Amoniun. potassium. and rubidium strongly activate theenzyneandaleoprenotetheappeerance ofthe 337mband, whereas in the case of lithium. cesium, or sodium the 1+20 m band predominates andtheensyneactivityislow. 'nle337nnbandisdecreasedinbntfers containing tetranethylamonim ion and enhanced in buffers containing phosphate ion. Couparison or the fluorescence spectrum of apotryptophanase. A ex: 280 m, with the fluorescence spectrum of holotryptophanase, Inc 280 um, leads to the conclusion that the 337 m absorbing form of bound pyri-‘ clunky-phosphate is excited by nonraddative energy transfer from the tryptophan residues of the protein and in turn emits at about 14-00 m. a 510 um emission arises from the 420 an absorbing fans of bound pyri- dml-5'-phosphate which is created in the excited state by an intermo- lecularprotontransfer. Departures oftheratio ofthehOOnntolem emissions free the ratio of the 337 m to l+5220 nu absorptions nay indi- cate specific effects of nonovalent cations on the excited state proton transfer reaction. Analysis of fluorescence decay data in order to prove this mechanism is complicated by the presence of short, close-together lifetimes. In rapid-scanning stopped flow kinetics experiments involving changes of pH or nonovalent cations. we have observed a slow process. accords in chiration. in which a #20 m absorber disappears and a 337 m absorber appears, or vice versa. We hve interpreted this observation as a protonation-deprotonation reaction coupled to conformational change in the protein. A still slower process may involve the interconversion of the holoensyne and the apoenzyne. ACKNOWIEMI‘ENTS I would like to thank Dr. H. Ashraf El-Bayoumi for his friendship and guidance during the course of this work. Drs. Clarence H. Suelter and James L. Dye for their support of this project. and nw graduate student colleagues in the enzyme group for their help and advice. I also wish to express my appreciation to Dr. Estelle J. McGroarty for six years of friendship and assistance. I am especially grateful to Robin Fink for helping me to complete this work and for illustrating this dissertation. For encouragement, thanks to Nina, Vicki. Ann. Billie. Barbara, Shermila, Soheir. Jocelyne. Denise. Rene. Cindy, Myra, and Rachael. ii Chapter 1 . Chapter 2 . Chapter 3. Chapter u e mph! 5 e TABLE OF CONTENTS Absorption and emission properties of 36 compounds......................................... A. Introduction........'.............................. B. Absorption properties............................. C. Mission properties............................... Structure andfunction of B6 enzymes................. A. General properties of 36 enzymes.................. B. Aspartate aminotransferase........................ C. Glycogen phosphorylasm........................... D. ‘l‘ryptophanase..................................... An Impothesis regarding the role of monovalent cations intryptophanase.................. Results anddiscussion............................... A. Absorptionandemission studies................... B. Fluorescence lifetime studies..................... C. Kinetic studies................................... Materialsandmethods................................ A. Growth of g. £13.,B/1t7e.......................... B. Preparation of the Sepharose column............... C. Purification of tryptophanase..................... D. Buffers for tryptophanase......................... B. Preparation of tetramethylannnonium hydroxide. . . . . . iii 13 18 18 22 29 1+0 45 as 66 81 81 83 87 89 Fe W ‘8“,' Of WtoPMmsoeeeeeeeeeeeeeeeeeeeeee G. Spectral measurements.............................. H. Rapid-scanning stopped flow kinetics measurements....................................... Chapter 6. Future work........................................... Bibliography..uuu.uuuu...o.........o...................... iv 39 89 ‘8 Table 1. Table 2. Table 3. Table llv. Table 5. HST OFTABLES Activating constants for monovalent cations”. . . . . . Concentrations of cations.......................... lamp shift corrections............................. Decay parameters................................... may Wu]?! with 1‘“? Weeeeeeeeeeeeeeeee Figure 1. “Eur. 2e Figure 3. Figure it. Figure 5. Figure 6e hm7._ Figure 8. Figure 9. Figure 10. Figure 11. Figure 12. Figure 13. Figure 11+. Figure 15. Figure 16. Figure 17. Figure 18. Figure 19. LIST OF FIGURES B6 compotmdsl...................................... Ionic and tautomeric forms of a B6 Quinonoid structure of a 36 Schiff's bese3......... ylvdron-Dr-forwlpyridine Schiff's baseé.......... Pyridoxal-P aldehyde and hydrate“ Absorption spectra of pyridonl-P as a function of p311.............................. Absorption spectra of pyridonl-P lrydrateu........ Absorption spectra of pyridoxal-P aldellyden....... Ground state (left) and excited state“............ Fluorescence spectra of pyridonl-P Whis...“ Fluorescence spectra of pyridonl—P aldehydels. . . . Excited state proton transfer scheme19............ Reactions of B6 ennui}......................... Schiff's base with substrate amino acidzz......... QuinOnoid intermediatezz Protomtion at the fowl carbonzz................ Products of transminaticnzz 362 and l+26 nm forms of aspartate mwenuzueeeeeOeeeeeeeeeeeeeeeeeeeeeeeeee Pyridoxal-P Schiff's base in solution and enzymezu vi come-u 10 11 11 12 11+ 11+ 16 18 19 20 21 21 22 Figure 20. Figure 21. Figure 22. Figure 23. Figure 2“. Figure 25. Figure 26. Figure 27. Figure 28. F181” 29. F18“. 30. Figure 31. Figure 32. F18“. 33. Figure 3“. Figure 35. Figure 36. Figure 37. Figure 38. Figure 39. Figure ‘40. Figure 41. Reactants for nucledphilic additionzu Form]. urban bomd ummmeeeeeeeeeeeeee Active site of aspartate aminotransf‘eraeel7 Pyridonl-P Schiff's base micelle7’8.............. Absorption spectra of tryptophamse asa function of p529............................. Absorption spectra of tryptophanase as a function of cation29......................... Mechanism of action of tryptophanasozg............ Possible 337 nuforms............................. Analogous coenzyme forms.......................... Absorption and difference absorption spectra. . . . . . Pyridoxal-P binding reaction...................... Fluorescence spectrum of apotryptophanase......... Absorption spectra of tryptophanase in ‘DJA........ Tryptophanase absorption spectra.................. Absorption spectrum of holotryptophanase- L-ethionineo...................................... Fluorescence spectrum of holotryptophamse. . . . . . . . Absorption spectra in ammonites, potassium, rubidium............................... Absorption spectra in sodium. lithium, cesium..."............................. Absorption spectra inammemm.................... Absorption spectra inpotassium................... Absorption spectra in rubidim.................... Fluorescence spectrum of buffer................... Vii 25 27 28 31 33 1&1 l+2 so: 51 53 55 55 57 57 59 Figure M. Figure 1+5. Figure 1#6. Figure 1&7. Figure l+8. Figure '49. Figure 50. Fish" 51. Figure 52. Firm 53. Figure 9+. F18”. 55. Figure 56. Figure 5?. Fluorescence spectrum of tryptophanase inammum....................................... Fluorescence spectrum of tryptophanase inpotassium...................................... Fluorescence spectrum of tryptophanase in rubidium....................................... Fluorescence spectrum of tryptophanase in lithium........................................ Fluorescence spectrum of tryptcphanase in cesium......................................... Fluorescence spectrmn of tryptophamse in sodim......................................... Fluorescence spectra of tryptophanase inpotassium..................................... Fluorescence decay of 9-cyanoanthracene........... Fluorescence decay of tryptophamse, #20 nm....... Fluorescence decay of tryptophanase, 510 m....... pudrop Imdpfi imp” Absorbance-wavelength-time surface................ First and last spectra............................ Time out at 337 m................................ film. cut ‘t “20 mIOOOO...OOOOOOOOOOCOOCOOOOOOOOOO Pyridonl-P binding to apotryptophamse........... 59 6O 60 61 61 69 71 71 78 78 79 79 80 CHAPTER 1 ABSORPTION.AND EMISSION PROPERTIES OF 86 COMPOUNDS A. INTRODUCTION Vitamin B6 is the name given to pyridoxal-5'-phosphate (pyridoxal- P) and related compounds including pyridoxal , pyridoxol , and pyridoxa- mine. The biologically active form of Vitamin B6 is pyridoxal-P which is bound to a variety of enzymes as a Schiff‘s base with an e-amino group of a lysine residue. 1 ,O / .I H H \ a... pyridoxal-P H NH}:r H / ' H H / I \ N \ - on - H .1. u .1. Scfigfes his: pyridoxol pyridoxamine Figure 1. B6 compounds 1 1 2 The spectroscopic characteristics of B6 enzymes are essentially those of the pyridoxal-P Schiff's base in the environment provided by the active site of the enzyme. The pyridoxal-P Schiff's base exists in several ionic forms as well as displaying prototropic tautomerism in which a proton can move between the pyridine nitrogen and the imine ni- trogen. Such prototrOpic equilibria involving heteroatoms usually have activation energy that is low enough for the rates of the tautomeric reactions to be fast and for the tautomers to be in mobile equilibrium in solution. Furthermore. because of these low energy differences. tau- tomeric equilibrium constants are very sensitive to environmental ef- fects. in.particular to solvent, concentration. and temperature. There- fore. when analyzed with these facts in mind. the spectra of B6 enzymes may yield information about conditions and changes of conditions at the active site of the enzyme. The B6 coenzyme is thought to participate in the mechanism of cats- lysis in an intimate way. including the formation of a covalent linkage with the substrate amino acid to give other pyridoxal-P Sciff's bases at the active site of the enzyme. In their review article on heteroaromatic tautomerismz El Guero et a1. emphasize the importance of depicting reac- tion mechanisms in their correct ionic and tautomeric forms. The spectra of BB enzymes and the products of their reactions with substrates. inhi- bitors. and other molecules can.provide information which can assist in assigning correct ionic and tautomeric forms in the mechanisms of action of these enzymes. B..ABSORPTION PROPERTIES Johnson and Hetzler3 review the spectral data on B6 compounds and discuss the methodology for resolving the spectra of the various ionic and tautomeric forms. They discuss the B6 spectra in terms of the three 1r-Jf*transitions of benzene. Spectral variations of the structures in Figure 2 can be explained by considering the energies of locally excited benzene states (L.E.) and charge transfer states (C.T.). In the latter, the OH and 0- groups act as the electron donor and the azomethine group acts as the electron acceptor.“ One must bear in mind that the descrip- tion of a state as being C.T. or L.E. is correct only in the zeroth or- der approximation and that interactions of C.T. and L.E. states must be considered in order to account for the energies and intensities of the observed absorption bands. III. 420 nm Figure 2. Ionic and tautomeric forms of a 36 Schiff's baso3 l4. One may then consider the first absorption band in these three com- pounds to arise from a transition to a C.T. state mixed with the lowest L.E. state of benzene. The larger ionization energy of OH as compared to 0- would explain the shorter wavelength of the band of compound I as compared to compound II. The larger electron affinity of a protonated azomethine group would explain the longer wavelength absorption of com- pound III. Using valence bond theory language one would state that the quinonoid structure (Figure 3) has a major contribution in compound III, while in compound II it has a smaller contribution and in compound I it has practically no contribution. Figure 3. Quinonoid structure of a. Bo Schiff's boso3 A second factor that may affect the position of the first absorp- tion band is a rotation about the bond between the phenyl ring and the azomethine group. In general. the more planar the system the greater will be the amount of conjugation and the longer the wavelength of the absorption mazdmm. Johnson and hetzler3 use this possibility in 5 explaining the anomalous values for the absorption maximum of the de- mpyridonl-Jsucine Sciff's base. The pK‘ values are also anomalous for this compounds 11.? and 6.5 for the imine and pyridine nitrogens. respectively, as compared to more typical values of 10.5 and 5.9. In a recent abstract, Metaler et a1.5 report that they have mea- sured the electronic absorption spectra of the individual ionic and tautoneric forms of Schiff's bases of pyridonl-P, as well as deter- mining the pK‘ values. Spectra were resolved by asstming lognorml dis- tribution curves. A very complete assignnent of spectral maxim to the molecular species of B6 Sohiff's bases with methyl valinate am! n-butylamine is given by Karube and Hatsushim.6 Figure '4 sunmarises the assigxlnents for the longest wavelength band of the Schiff's base of 3-lvdroxy-lI-formyl- W. The spectra of the corresponding fame for pyridoxal-P are at longer wavelengths by less than 10 m. It is particularly interesting to note that the Schiff's base of pyridoatal-P with mettwl valinate in neu- tral methanol absorbs at 3&0 run, with a shorter wavelength maximum at 25} m. Presumably these absorptions arise from a form like compound II of Figure 1+. he Schiff’s base of pyridonl-P with n-butylamine absorbs at 418, 336, 288, and 272 nm. In this case. there is an equilibrium be- tween forms like II and III in Figure it. Also. more polar solvents favor the more polar form III. In neutral water, however, a dipolar form IV would be expected to predominate but it is not stable. file absomtion properties of the Schiff's base formed be‘lmeen n-dodecylamine and pyridoxal-P are studied by Gani et al.7’8 N-dodecyl- anine was shown to form mixed micelles with CetMe NBr with the pyridonl- 3 P bound in 1:1 stoichionetry to the n-dodecylamine. The complex absorbs F3 5‘ II. /N V, /N\H N/ N’ 325m 365m B a 3 III. /N\H+ VI. ,N\H+ VIII. /N\H+ \ " - \ \ H I , I , | , N N N H+ H+ h20nn I+2511”! 365nm F3 I IV. /N VII. /N ‘ \ " ' \ OH / / a. :3. 335nm 330nm Figure lb. 3-hydroxy-h-fornwlpyridine Schiff's base6 7 at 333 nm and 250 nm with a minor peak at #15 m. Gani et a1. attribute the spectrum.to an enolimine structure embedded in.a hydrophobic envi- ronment. Ryridczal Schiff's bases of poly'(IrHfl(amino acids) are studied'by Dentini et e1.9 At pH 7 in water/metlnnol solutions the pyridoawlidene- imine group absorbs at #20. 335. 253. and 210 nm. with a blue shift in the 335 um.band with increasing ratio of water to methanol. The authors suggest a Schiff's base linkage to one amino acid residue together with a partially electrostatic hydrogen bond between the nearest an; group and the phenolic oxygen. Conformational analysis of Schiff's bases of pyridoxa1.with.alanine. vnline, leucine. and phewlalanine are carried out by Weintraub et «1.10 All four Schiff's bases had an energy'minimum.corresponding to a planar conformation'with the carbonyl carbon.13(-carbon, and nitrogen.in the plane of the pyridine ring. In.addition, phenylalanine. alanine. and leuoine had an energy minimum.in which the C,‘-N bond was rotated so thattheH, a, or OOO'groupwusperpendimflartotheplane ofthepyri- dine ring. The population of the latter conformer ranged from 1% for phenylalanine to 17‘% for alanine. Morosov et e1.“ utilise the method of least squares in order to calculate the absorption spectra.ct the individual forms of pyridoxal-P. Pyridcxal-P is assumed to exist in three ionic forms of each an alde- hyde and a hydrate. the latter of which is fcrmed.by a hydrogen ion.ca- talysed nuclecphilic attachment reaction. The reactivity of'the carbonyl group is a maxim when the pyridine nitrogen is protcnated so that the equilibrium shifts towards the aldehyde forms as the pH increases. Sig- nificant amounts of the appropriate ionic forms of the hydrate and of / 0" \N ' / I — / \N / \N + H / H \N ' H+ HO H / _ HO OH \N l HO / ' O— / \N HO H \N H+ / I OH \ {31+ ‘ Figure 5. Pyridoxal-P aldehyde and hydrate11 9 thealdehydeareshowntoenstatneutralpfi. Ionic strengthofthe solution does not change the spectra of the forms but with increasing ionic strength the hydrate/aldehyde ratio is increased. Ionization of the phosphate group seems to have little effect on the absorption spec- tra of the foms of pyridoxal-P. Morosov et al.'s spectra are shown in Figures 6, 7. and 8. Harris et al. 1‘2 evaluate the electronic absorption spectra of the individual forms of pyridcaral-P by a spectral resolution using lognorml curves. In order to obtain a good fit of their data they require an additional form to be present. an uncharged tautomer of the dipolar alde- lude form. Apparent pK‘ values of 3.62 and 8.33 are reported for the phenolic group and the pyridine nitrogen. respectively. Sav'in et e1.13 calculate the 1r-electronic structures end absorp- tion spectra of 36 cospounds by the semiempirical Pariser-Parr-Pople me- thod. SubstituentstctheparentcompoundB-lwdrcmypyridinehaveaweak effect on the absorption spectrum with the exception of the carbowl group substituted in the it position. In the last case. it is determined that the longest wavelength band is associated. in the lowest approxima- tion, with the transition of the electron from the upper filled orbital of the ring to the antibonding orbital of the carboml grow). In all cases upon transition to the lowest excited singlet state there occurs an increase of the T-electronic charge on the pyridine nitogen and a large decrease on the phenolic omen (Figure 9). 0n the basis of a more elaborate calculation. Bashulina et al. in report that upon excitation to the lowest excited singlet state a transfer of 7r-electronic charge amounting to 0.3 to 0.5 electron charges takes place from the pyridine ringandthephenolicgrouptothecarborwlgroup. 10 3799 el... 3.0 Ollie §w I QN I Ha ma mo noauossm a no mnamxopwnha mo snaoodm :ofluduomn< .w madman '-'I. 11 I 4 II c99°.°.°. -:ouns-m>0 11 Cation e—-—e Dipolar ionH Anion e-— -e 0'! 6" Cation e—o Dipolar ion n—-—n Anion 0-- no 20 Rum Figure 8. Absorption spectra of pyridoxal-P aldehyde11 1.h71 0. 39 1.016 1.227 Figure 9. Ground state (left) and excited statelu 13 Ce MSSION PROPERTIES Morosov et al.15 present the fluorescence spectra of the different forms of pyridoxal-P as resolved by the method of least squares. he hy- drate cation has an excited state ionization associated with a sharp re- duction in the px" of the phenolic groups emission is then observed from the dipolar ion. Shifts to shorter wavelengths of about 20 nm are attributed to the ionization of the phosphate hydrcmls and the pyridine nitrogen. Stokes shifts are reported to be 55 to '70 nm. In the aldehyde. dissociation of the pyridine nitrogen and the phosphate hydroxyls lead instead to a shift to longer wavelengths of about 40 run. he Stokes shifts are 115 to 150 nm. he latter fact is ex- plained by the authors by the formation of an intramolecular ludrogen bond in the excited state at all pH values. Bridges et al.16 the wave number of the longest wavelength of the excitation maximum and defim the Stokes shift as the diffennce bettmen that of the shortest wavelength of the fluorescence maximum and they state that the Stokes shift is a measure of the energy required to raise the compound to the first excited state. hey report Stokes shifts of 9.36. 5.86. 5.50. end 5.76 kit for the cation. dipolar ion, neutral form. and anion. respectively. In addition they give values of the Stokes shift for m 36 compounds and they conclude that for cations ionising intheexcitedstate one shouldexpectBtoiOkKas comparedtohto 6 1:1 for other forms not exhibiting excited state ionisations. From values for absorption and emission maxim of pyridoxal-P given by Morosov et al..” one can estimate the Stokes shifts for the pyridoxal- P aldehyde to be 7.2. 5.6. and 7.1 H for the cation. dipolar ion, and 11+ 5) m f 0 I \ __ pH 0.34.0 I \\ 2:3 320 I l 7 \ I \°\ 05‘- l “a I X\ 0/ \ I \o I, \‘x 1° 0" 7355“ Mm l l I 4-00 500 600 700 Mun Figure 11 .- Fluorescence spectra of pyridoxal-P aldehyde” 15 anion, respectively. Based on the results of Bridges et al.. excited state proton transfer cannot be inferred from these values alone. In substituted benzene molecules it is known that a large transfer of‘n'- electronic charge toward the carbonyl groups and away from the phenolic groups occurs upon excitation (a fact that is responsible for the in- crease in the pK: of the carbonyl group and the decrease in the pKa* of the phenolic group that sometimes leads to proton transfer in the excited state). It may be that the interaction with the surrounding medium is changed sufficiently in passing to the excited state to ex- plain the magnitude of the Stokes shifts that are observed. Fluorescence lifetimes of less than one nanosecond and very low quantum yields are reported.by Morosov et al.15 They argue that these observations support the excited state intramolecular hydrogen bonding hypothesis. Arrio-Dupont studies the Schiff's base of pyridoxal-P and n-butyl- amine.17 This compound absorbs at 340 and #08 nm in water and emits at #00 and 505 nm corresponding to the two absorptions. She measured the pic"t of the imine nitrogen to be 11.5 and calculated the pKa* by Heller's method18 to be 12.“. Protonation of the pyridine nitrogen does not change the absorption specrtrum but fluorescence intensity is enhanced twentyfold. In a pyridoxal-valine Schiff's base the pKa of the imine nitrogen was determined to be 8.5 with the pyridine nitrogen protonated as compared to 10.h if it is not protonated. In nonaqueous solvents the 3h0 nm absorbing species shows a second emission at 525 nm; the Stokes shift is greater than 10 kK.19 Arrio- Dupont proposes an excited state proton transfer scheme shown in Figure 12 to explain the two emissions. In nonpolar solvents the equilibrium 16 F.‘ R R ' I / N , N / OH I / H \ I ' +h‘N/ N ‘N I 1* I — - # R I ‘f x” \Hl' , N ‘14-? / " / " . I ——9 I +4“); II II" II Figure 12. Excited state proton transfer scheme19 17 is displaced toward 11" whereas in H accepting solvents the equilibrium is shifted toward I'm- solvent. Additional proof for the proton transfer mechanism is given by Veinberg et d.20.21 salicylaldehyde and o-methylsalicylaldehyde. The 535 nm emission ob- served in 001” in the case of salicylaldelvde is not observed in the case of o-mettwlsalicylaldehyde because the presence of a mettvl group instead of a proton makes proton transfer impossible. in a study of Schiff's bases of hemlamine with CHAPTER 2 STRUCTUE AND FUNCTION OF 86 ENZYZ'IES A. GENERAL PROPERTIES OF B6 ENZYI-iES Pyridoxal-P is the coenzyme of a number of enzymes involved in the 1 reactions of amino acid metabolism. Several classes of reactions cata- lyzed by these enzymes are shown in Figure 13. Racemization [fit-{3+ 'Ocvo H -(5; ‘ R 6T2- H-e Ca- R -0 /C\ 0 N H3+ Transamination , I R O ’ / 3 0- _ -' '- Decarboxylation to «LP -elimination .. NH; NH’ R H 0 H3 RJIO #4 A") 9h 0" 0‘ 4 Figure 13. Reactions of 36 enzymes‘ 18 19 he coenzyme is linked covalently to the enzyme by a condensation reaction of the pyridoxal-P carborwl group with the E. -amino group of a lysine residue in the enzyme active sites mmz + R'Hc=0 3:2 RNSGER' 4- H20 A general mechanism for pyridoxal-P catalyzed reactions has been proposed independently by Snell and by Braunstein and presented con- cisely by Pullman end Pullman.22 For purposes of their presentation. the ionic and tautomeric forms of the pyridonl-P are chosen arbitrarily. In addition, the effects of metal cations and of the 5' phosphate group are disregarded. In the first step, the enzyme bound Schiff's base (or imine or azcmethine) is converted to a Schiff's base with a substrate amino acid by a transimination reaction. ot-hwdroson :H "' d-carbon F3) \0 fonwlcarbon ; /N: / | OH \ {3+ 22 Figure 11+. Schiff's base with substrate amino acid 20 The second step requires labilization of the H. 000-. or R group attached to the e(-carbon. he driving force for this reaction is found in the gain of resonance energy upon formation of the quinonoid inter- mediate. shown in Figure 15 for the case of labilization of the -hydro- gen- 22 Figure 15. Quinonoid intermediate Following the labilization step is a protonation at the formyl car- bon in the case of transamination as shown in Figure 16 or a protonation at the K-carbon in the case of racimization. decarboxylation. or 01.5- elimination. Finally the C==N bond is hydrolyzed. he products formed at this stage may undergo further reaction and are shown in Figure 17 for the transamination reaction. Pullman and Pullman22 state: "... as concerns the metal ions which participate in the nonenzymic reactions. they are considered as models of the functions played by the apoenzymes of pyridoxal-phosphate proteins. heir activating action is postulated to occur through the formation of reactive chelated intermediates. . .the formation of metal chelates may influence favorably the reaction scheme by an appropriate direct elec- tronic influence on the charge distribution and by its contribution to the maintenance of coplanarity behveen the pyrimidine ring of the pyri- doxal and the external groups. his last effect has the double result of stabilizing the system and facilitating the Tr-electronic displace- ments involved in the mechanism of action of the coenzyme. " 21 22 Figure 16. Protonation.at.the formyl carbon N H; R ® / I H O M \N O" H + pyridcxamine .t-keto acid 22 Figure 17. Products of transamination B. ASPARTA‘E WM Aspartate aminotransferase catalyzes the reaction1 L-aspartic acid 4- d-ketoglutaric acid :2 oxaloacetic acid + L-glutamic acid he absorption spectrum of aspartate aminotransferase has a major peak near 280 nm corresponding to the absorption of the aromtic amino acid 2"Peelrsetaozenduzonnereettra.bht-.edto residues of the protein. the absorption of the coenzyme pyridcxal-P which is bound to a lysine residue of the protein in a Schiff's base linkage. he #26 m form pre- domlzetesetlowpnendls convertedto theaozmromesthepals raised. with a pK‘ of 6.2 for the transition. Ivanov end Isrpelskyz" cite evidence from the fluorescence studies of I“ase.'l.1l.a68 coenzyme. with protonated pyridine nitrogen and deprotomted imine ni- that suggests a dipolar form for the 362 nm form of the trogen. The #26 nm form appears upon protonation at the imine nitrogen. / <9 / I 0" \N N H-t' H+ Figure 18. 362 and #26 nm forms of aspartate aminotransferasezu 23 According to Ivanov and KarpeisRyZh the coenzyme is attached non- covalently to the enzyme at the phosphate group and flue methyl group. as well as the pyridine nitrogen being Imirogen bonded to a tyrosine residue of the protein. his last attachment is inferred from the pre- sence of a negative band with extremum at 295-300 nm in the circular di- chroism spectrum of holoaspartate aminotransferase which is absent in the spectrum of the apoenzyme. he authors claim that they can account for the position and shape of this circular dichroism band by assuming that optical activity is induced in an ionized tyrosine residue by the coenzyme. Furthennore. they report that when pyridoxal-P N-oacide. a pyridoxal-P analog which is incapable of accepting a proton at its ring nitrogen. is used to reconstitute the holoenzyme the circular dichroism spectrum is characteristic of an unionized tyrosine residue. he binding to the pyridine nitrogen is considered by Ivanov and Karpeislq to be responsible for lowering the pK‘ of the imine nitrogen from a solution value of 10. 5 to approximately 8 . An additional pKa lowering effect to the observed value of 6.2 is attributed to the pre- sence of a positively charged group 1+ in the active site. In Figure 19 are shown the possible ionic and tautomeric forms of the pyridoxal-P Schiff's base in solution and under the particular conditions existing in the active site of aspartate aminotransferase. At the lower end of the range of pH values at which the enzyme is active ( pH 6.0-9.5 ). it is clear that an N-protonated amino acid sub- strate interacts with the dipolar form of the coenzyme. In a model Schiff's base reaction. however. the reaction rate goes through a maxi.- mum as a function of pH. indicating that the most favorable reactants for the nucleophilic addition reaction are a deprotonated amino acid Figure 19. Pyridonl-P Schiff's base in solution and enzyme 25. and a protonated carborvl m.” >c=on++ 31152 2:: >C=NR 4- H20 Ivanov and Karpeislq reconcile this fact with the conditions of the ensyme reaction by proposing a reaction step in which there occurs de- protonation of the amino acid in favor of. the imine nitrogen of the co- enswne. his step is nude possible by neutralization of the positively charged group 1" by the carboxylic group of the substrate; the effect ofthisgroupinloweringthepK‘oftheininenitrogenisthuselimi- nated and the value of the pK‘‘ returns from 6.2 to approximately 8. Un- dertheinfluence ofX+ther‘ofthe substrateaminogroupislowered from 9.8 to about 7.7. nus. immediately before the step of nucleophilic addition the most favorable reactants have been prepared. 21+ Figure 20. Reactants for nuclecphilic addition 26 Ivanov and Karpeisky argue that .electrostatic forces orient the substrate amino acid at the active site in such a way that the lone pair of the substrate amino acid nitrogen is aligned with the pa orbital of the forwl carbon atom with a distance of no less than the sum of the van der Waals' radii of the two atoms. or 3.5 X. A sufficient con- dition for reducing this distance to the length of 1.5 R necessary for a covalent single bond can be achieved by displacement of the coenayme molecule . 'lhe conformational mobility required for this movement is made possible because the protonation of the imim nitrogen causes a dropinthepK‘ofthepyridimnitrogentobetweené.Oand6.5. me hydrogen bond with the tyrosine residue is disrupted and. once freed of this anchor. the coensyme canrotatebyanangle crsbcutuo" aboutan axis passing through its methyl and phosphate groups. The rotation is 'loclned' into place by formation of the covalent bond between the amino acid nitrogen and the formyl carbon. At this stage in the reaction. the forwl carbon is bonded tetra- hedrally as shown in Figure 21. In the next step. a proton is trans- ferred from the substrate nitrogen to the nitrogen of the E-emino group of the lysine residue of the protein and the bond to the latter group is disrupted. Imediately following the breaking of the bond. this group is still able to accept a proton and the authors claim tint the proton of the d-carbon is thereby abstracted. with formation of the quinonoid intermediate. his idea is consistent with thefact that as- partate aminotransferase displays a pH independent Vm in the pH range 6.0-9. 5 and may be explained if protonation-depretonation steps are as- sisted by internal protein residues so that no protons are released to or taken up from the medium at am stage of the reaction. 2? Figure 21. Fowl carbon bonded tetrahedrallyzu 28 Metzler et al.5 analyze the absorption spectrum of aspartate amino- transferase by fitting individual ionic and tautomeric spectra with log- normal distribution curves. he pyridoxal-P Schiff's bases on the en- zyme have peaks shifted bathochromically by about 1100 can‘1 from the positions of the peaks of the model compounds. Metzler et al. suggest the presence of a nearby positive charge or different conformations. Vibrational structure of the absorption bands suggests that the pyri- dine nitrogen is protonated. Arrio-mpcnt17 reports the fluore sconce spectrum of aspartate ami- notransferase. he 360 nm absorbing form displays an emission at #30 nm. while the 430 nm absorbing form emits at 520 nm. 0n the basis of compa- rison with the fluorescence properties of model compounds. Arrio-Dupont proposes the structures shown in Figure 22. Figure 22. Active site of aspartate aminotransferasej'7 29 C . GLYCCX'IEN PHOSPHORZLASE In the pH range 5-9. glycogen phosphorylase catalyzes the following reactionsl phosphate + amylose(n + 2) (—2"... glucose-l-P + amylose(n + 1) Pyridonl—P is required for activity and is linked to an enzyme lysine residue. but the biochemical role of the coenzyme is unclear. In addition to a 280 nm protein absorption. glycogen phosphorylase exhibits a major absorption band at 333 nm and a minor absorption band at #25 no.1? he emission spectrum of glycogen phosphorylase has two bands. one at 335 nm due to the aromatic amino acid residues in the pro- tein and one at 535 nm due to pyridonl-P.17 Based on the study of pyridonl-P Schiff's bases in nonaqueous sol- vents. ArriooDupontj'9 concludes that the 333 nm absorption is due to an enolimine fan of the pyridcxal-P Schiff's base in a hydrophobic environ- ment. Upon excitation of this form. however. fluorescence is found to occur entirely from the ketoenamine form which is reached by proton transfer across the chelate hydrogen bond in the excited state. as shown previously in Figure 12. Shaltiel and Cortijo26 observe that the 1&2 5 nm absorption increases with the addition of urea at neutral pH. Apparently the effect of dena- turingtheproteinistoexposethepyridoxal-P Schiff'sbasetothe more polar aqueous environment that favors the ketoenamine form. Jones and Gowgill27 suggest that upon excitation at 280 nm energ transfer occurs from the tryptophan residues of the protein to the 335 an absorbing form of the coenzyme which then emits at 515 nm. As 30 evidence. they report the observation of a peak in the excitation spectrum of the protein-bound pyridoxal-P at 280 run that is not present in the excitation of free pyridoxal-P. A problem with this conclusion. however. is the fact that pyridoxal-P Schiff's bases absorbing at about #20 nm generally have a higher energy absorption maximum near 280 rm. Cortijo et al.2'8 explore the question of energy transfer more carefully. he quantum yield of the apoenzyme was determined to be Qo= 0.18 and that of the holoenzyme to be QT: 0.12. From these values. assuming the difference to be purely a result of energy transfer. the efficiency of energy transfer was calculated: T=(1-QT/Qo)=0.33 Since this transfer may occur from as new as twelve different trypto- phan residues. the authors do not try to interpret this parameter in terms of a distance. but they do suggest that transfer efficiency may be a rather sensitive measure of conformatioml change. Shimomura and mm” cite studies that show that pyridoxal-P is buried inside the protein with the 5'-phosphate group near the substrate site. They report a maximm in absorption at 251 nm in addition to the one at 335 nm in the difference spectrum hemeen the holoenzyme and the apoenzyme. hese absorptions are characteristic of an enolimine form of the pyridoxal-P Schiff's base. A micellar experimental model for the pyridoxal-P site of glycogen phosphorylase is proposed by Gani et al.7'8 based on their studies of pyridoxal-P bound to mixed micelles of long chain allcylamines. The ab- sorption spectrum of a mixed micelle of hexadecyltrimethylammonimn bro- mide and n-dodecylamine to which pyridoxal—P binds stoichiometrically 31 to the latter component displays absorption maxima at 333 nm and 250 nm. The fluorescence spectrum displays a maximum at 550 nm upon excitation at 335 or 415 nm. Four possible structures absorbing at 325-335 nm are shown in Figure 23. The authors favor structure G in which the pyri- doxal-P is embedded in a hydrophobic microenvironment but in which there is an abrupt drop in hydrophobicity in the vicinity of the coen- -zyme that allows hydrophilic substances to approach closely. Figure 23. Pyridoxal-P Schiff's base micelle7’8 D. TRIPTOPHANASE A recent review by shell29 details the properties of tryptophanase. which catalyzes the d . P -elimination reaction L-tryptophan + 1120 2 indole + pyruvate + N113 as well as several related reactions. Tryptophanase is a tetrameric pro- tein with four identical subunits each of molecular weight 55,000. which binds one pyridoxal-P molecule per subunit. A pH titration of the enzyme from g. 921; in imidazole-H01 plus 0.1 M KCl and 0.1 M potassium carbonate buffers shows two pH dependent ab- sorption maxima at 337 and 1420 nm with an apparent PKa of 7.1». In the presence of sodium or imidazole alone. only the 420 nm band is observed. hese absorption spectra are shown in Figures 24 and 25. .4 a: ABSOR BANCE \JAVE LENGTH (roe) Figure 2“. Absorption spectra of tryptophanase as a function of p329 33 ABSOR BANCE \JAVELENGTH (ma) Figure 25. Absorption spectra of tryptophamse as a function of cation29 In tryptophanase. pyridoxal-P is present in an azomethine linkage with the E -amino group of a lysine residue. he mechanism of action of tryptophanase is given by Snell as shown in Figure 26.29 Briefly. the substrate amino acid forms a Schiff's base with the coenzyme pyridoxal-P and a proton is labilized at the ok-carbon. he indole group leaves. followed by the separation of an aminoacrylate which then is degraded to the products pyruvute and ammonia . Tryptophanase exhibits a requirement for monovalent cations. In his discussion of enzymes activated by monovalent cations. Suelter31 points out that many enzymes are activated by potassium. rubidium. and ammonium but not by lithium or sodium. He suggests that monovalent cations may participate in the enzyme-substrate complex or assist in the control of enzyme conformation. or both. A potential keto-enol tautomer might be / 1:! ' /l 0 ® I I \N+ a H A HM \RH NH3'\ }NH3 R {I 29 Figure 26. Mechanism of action of tryptophanase 35 involved in binding the monovalent cations. Snell29 indicates that a large conformational change occurs in the interconversion of holo- and apotryptophanase. he fact that sedimenta- tion rate increased much more in the presence of potassium than in the presence of sodium indicates that potassium promotes a 'tightening' of the enzyme structure or a major change in shape. or both. he role of monovalent cations in promoting tight coenzyme binding to the protein is compared to cation activity as a cofactor by Toraya et al.32 Holoenzyme solutions containing 0.1 M of the cation as its chloride were resolved of the coenzmee pyridcxal-P by gel filtration. Ability to hold the pyridonl-P and cofactor activities were high for potassium. rubidium. and ammonium and low for sodium. lithium. and ce- sium. he affinity of the sodium enzyme for pyridoxal-P was shown to be much lower (Kn) 311M!) than that of the potassium enzyme (ED 8 1.81am. A possible defect in the preceding study is the fact that the in- vestigators may have failed to use saturating concentrations of the mo- novalent cations. 1h. values reported by Suelter and 311.1133 for the ac- tivating constants for the various cations .t 25° c obtained in 0.6 ml! S-orthonitrophervlcysteine are given in Table 1. hese authors assert that each cation activates and that the Km for SOPC is not significantly affected by cation with the exception of sodium. whereas Vm changes and can be taken as a measure of cofactor activity. Failure to observe activity in previous work may therefore be a result of too little cation in the assay mixture. Presumbly this fact would affect the KB values reported by Toraya et al. as well. Suelter and Snell” discuss the interaction of the cation with the enzyme in terms of the difference in free energ of hydration of the 36 cation and the free energy of interaction of the naked cation with its binding site on the enzyme surface. hese differences are in the order potassium > rubidium) cesium > sodium > lithium. harther. they point out that cations between 1.3 X and 1.5 R in ionic radius are the best activators. Table 1. Activating constants nor monovalent cations33 Cation KD(mH) Lithium 54 iv 11.6 Sodium M e 0.06 Potassium 1.144 vb 0.06 hallium(I) 0.95 '* 0.1 Amonium 0.23 e 0.01 lhlbidium 3.5 "' 0.3 Cesium 115.6 vb 2.6 The complex betaleen holoenzyme and the dead-end inhibitor L-ethio- nine in the presence of activating cation develops a strong 508 an ab- sorption band. whereas there is only a small spectral change in the ab- sence of cation. Circular dichroism results indicate that Inethionine is bound to the enzyme in either case. hus. Suelter and Snell are able to conclude that monovalent cations are necessary for converting the en- zyme-L-ethionine complex to the 508 an absorbing form in which the proton on theok-carbon of L-ethionine has been labilized. Hggberg-Raibaud et al.? describe reactivation of their apoenzyme stocks by vigorous treatment with dithiothreitol in the presence of am- monium sulfate. an important technique experimentally. hey study the 3? activation of the enzyme by potassium and ammonium in the presence of saturating concentrations of pyridoxal-P and found sigmoidal curves with Hill coefficient of 2.7 for potassium and 1.5 for ammonimn. A concentra- tion of 0.1 M potassium wus adequate to saturate the enzyme; at 0.4 M potassium. about 20$ inactivation was observed. Hé'gberg-Raibaud et al. favor an interpretation of their results as a cooperative interaction between protomers in which the binding of potassium to one protomer leads to an increased affinity by the other protomers. At saturating potassium concentrations. the binding of pyridoxnl-P is followed by the appearance of activity in the apoenzyme. In phosphate buffer. there are two kinetically distinct steps with rate constants 0.49 min-1 1 and 0.095 nin‘ . a model of pro-existing equilibrium between two apoenzyme conformations. only one of which binds pyridonl-P. is discarded on the basis of a linear Scatchard plot in favor of an anti- oooperative model. A conformational change upon binding might be used to account for the slow dissociation kinetics observed for the holoen- zyme. Because phosphate ion is known to inhibit reconstitution of hole- aspartate aminotransferase. a Tricine-KOH buffer system was also used with results similar to those for phosphate buffer. Finally. Hggberg-Raibaud et al. measure the KB for binding pyri- doxal-P to apotryptophanase in the presence of potassium to be 0.1+,nh. Earlier reports of KD are criticized for not allowing sufficient time for complete reconstitution of the holoenzyme. Fenske and Defies»35 report the fluorescence spectrum of tryptopha- nase from B. &. Upon excitation at 280 nm. there is an emission at 350 m for the apoenzyme and emissions at 350 and 510 nm for the hole- enzyme. Resonance energy transfer from the tryptophan residues of the 38 protein to the pyridoxal-P is suggested as an explanation for the 510 nm emission since free pyridoxal-P does not absorb at 280 us. However. it should be remembered that the enzyme bound Schiff's base may well absorb at 280 nm. lean and DelolossB6 report hyperbolic kinetics from a spectrophotome- tric stuck of the binding of pyridonl-P to the apoenzyme. with a KD value of 1.6 H. A similar value was obtained by studying the quenching of apoenzyme fluorescence upon binding of pyridoxal-P. An average maxi- mum of 164$ of apoenzyme fluorescence could be quenched. he effect of pH on the fluorescence of holotryptophanase is also reported by Isom and Delbes.36 A form excited at 3+0 nm and emitting at Wm grows fromp37.1 topH9.0. followedbyadecrease atpH9.3: a form excited at 1&20 m and emitting at 510 nm grows from pH 9.3 to pH 7.1. However. since these investigators were working at 0.75 mg/ml. or 1hr}! in bound pyridoxal-P (1m x 52,000 per pyridoxal-P binding site). and the 100 PM free pyridoxal-P in solution has forms absorbing at 330 and 390 nm and emitting near l+00 and 500 nm. respectively. it is diffi- cult to have much confidence that they are seeing coenzyme fluorescence alone. A computer analysis of the changes in the absorption spectrum of native holotryptophanase with pH gives absorption spectra of the proto- mted and deprotonated forms.36 In 0.05 M potassium phosphate buffer containing 100,514 pyridoxal-P the authors report a pK‘ of 7.89 iv 0.019. However. the concentration of potassium is not high enough to saturate the enzyme if Suelter and Snell's33 values for the E. 92g enzyme apply. Furthermore. Isom and Moss do not mention using sulfhydryl compounds to protect their enzyme as is necessary with tryptophanase from E_‘.. ggli. 39 In a conzpanion paper. Isom and Dehoss37 report that the KD value for pyridoxal-P binding to apotryptophanase in the presence of sodium is tenfold higher than in the presence of potassium. hey also studied the binding of ANS to the apoenzyme: ANS fluorescence was enhanced and the emission maximum shifted as is characteristic of ANS in a hydro- phobic environment. Since they were also able to show that ANS is a competitive inhibitor of coenzyme binding. they were able to conclude that a hydrophobic region exists at the active site of tryptophanase. CHAPTER 3 AN HIPOTHESIS REGARDIN} ‘HIE ROLE OF HOMVAIENT GATIONS IN TRXP‘IDPHANASE here are at least trio forms of holotryptophanase including a form that absorbs at #20 nm and a form that absorbs at 337 mm: the latter form is observed only in the presence of activating monovalent cations and at sufficiently high pH values. Horino .nd Snell38 identify the 337 m absorbing form as a nonhydrogen bonded azomethine. as shown in Figure 2?. hey do not account for the fact that the same structure in the enzyme aspartate aminotransferase absorbs at the longer wavelength of 360 um. Snell and Dinari39 point out that there are a variety of ionic ard tautomeric forms possible including those differing by a pro- ton on the pyridine nitrogen ard they therefore conclude that the exact structure is not certain. Snell29 draws another structure for the 337 nm form. also shown in Figure 27. and he suggests that this hydrogen bonded structure could be favored in the hydrophobic environment of the enzyme active site as in the case of glycogen phosphorylase where this structure is present ard absorbs at 333 mu. However. unlike phosphorylase. holo- tryptopharnse can be reduced and inactivated by sodium borohydride. hue. no structure for the 337 no form is consistent with all of the available intonation. Since the first steps in the mechanism of chfi-elimination are 1&0 41 Figure 27. Possible 337 m forms apparently identical to those of transamination. it would be reasonable for coenzyme attachment to be the same in both these classes of enzymes. Ivanov and Karpeishyz“ identify . dipolar form of the coenzyme in aspar- tate ‘aminotransferase that is stabilized by the presence of a positive charge from the enzyme. By analogy. the role of the monovalent cation required by tryptophanase can be explained. Specificity for certain cations would be a result of the need to fit into a protein 'pocket' that holds the cation in place near the phenolic group. In particular. the cation would have the function of increasing the pKa of the pyridine nitrogen ard decreasing the pK‘ of the imine nitrogen. Structures of the coenzyme in the active sites of aspartate aminotransferase and tryp- tophanase according to this hypothesis are shown in Figure 28. Provided this hypothesis is correct. wig does this form absorb at 337 m in tryptophanase instead of at 360 m as in aspartate aminotrans- ferase? Referring to Figure 2. one can expect a variation in spectral wavelength from 367 nm (Structure II) to 322 :3 (Structure I) as a 42 2 l ‘9 GO. ® /I ‘- \ \ 3+ 3+ 73 2 Figure 28. Analogous coenzyme forms positive charge approaches the 0" group. he absorption modem is at longer wavelengths the more the electron on the oxygen is delocalized into the ring. which in general depends on the state of protonation of the phenolic oxygen and the imine nitrogen-in the model compounds. In the enzymes under discussion. one can think of other positive charges playing a role similar to that of the proton. From the spectral maxima it appears that in tryptophanase the monovalent cation is more effective in localizing the electron on the phenolic oxygen than is the positively charged residue in aspartate aminotransferase. but less so than a proton. Another effect that can give rise to spectral shifts is for-the protein to hold the pyridcmal-P Schiff's base in different conformtions about the bond between the pyridine ring and the azomethine carbon atom. Since protontion at the pyridine nitrogen has little effect on the absorption maximum of B6 enolilines ( see Figure 1+ ). there is another fern-that absorbs near 337 nm differing from the one discussed above by a proton. he l4'20 nm absorber is similar to Structure III. Figure 2. #3 represented in its major resomnce form in Figure 3. Again. there are two possible forms absorbing near #20 nm differing in state of proton- tion of the pyridine nitrogen. Protomtion of the pyridine nitrogen is promoted by the presence of activating monovalent cations. Another implication of Ivanov ard Karpeisky's model of aspartate aminotransferase is the necessity for rotation of the coenzyme in going from a form in which the pyridine nitrogen is protonated to one in which it is deprotonated (hydrogen bond to protein tyrosine residue disrupted). In tryptophanase. the presence of activating monovalent cations would affect the position of the equilibrimn between these forms. Going one step further. it is easy to think of this proton transfer-coenzyme rota- tion as being coupled to a change in protein conformation. If the con- formtional change is slow. there is a possibility of observing slow spectral interconversions after an abrupt change in pH or monovalent cation. An interesting point that my be important in interpreting the lu- minescence properties and action of light on the enzyme has been raised by Ivanov and Karpeisky?“ hey discuss the possibility of an excited state proton trander reaction from a protein residue to the imine nitro- gen in which the excited state derived from the 360 m absorber is con- verted into the excited state derived from the #30 nm absorber: these emit at #30 m and 505 nm. respectively. If this reaction is carried out to an extent within the lifetime of the excited state. excitation at 360 nm might result in an emission at 505 nm. Although this is not the case in aspartate aminotransferase. another enzyme active site might have a greater availability of protons and thus enable one to observe _ the phenomenon of intermolecular excited state proton transfer. u. Intramolecular excited state proton transfer has been suggested as an explamtion for the long wavelength emission of glycogen phosphorylase?1 CHAPTER # RESULTS AND DISCUSSION A. ABfilRPTION AND EMISSION STUDIES Several studies of the absorption spectra of tryptophamse urder verious conditions were reported (Chapter II). In the present work. a study of the absorption characteristics of the enzyme was undertaken in order to reproduce and extend previous results as a basis for the study of the fluorescence properties of tryptophanase and the interpre- tation of rapid scanning stopped flow kinetics data. Abgogption mctrum 2; W. he absorption spectrum of apotryptophanase in 25 ml! K-EPPS. pH 8.0. was measured against a blank of them buffer and the results are shown in Figure 29 (for details of buffer preparation. see Materials ard Methods). he peak-at 278 m can be explained by the well known absorption characteristics of the aromatic amino acid residues of the protein. Differeng mgtrum between m 993 ‘f_r_e_c_ EdoEI-P. he previous enzyme sample was diluted 1:10 with K-EPPS containing 100,.11 pyridoxal- P ard blanked against the same proportions of K-EPPS and K-EPPS con- taining 100p}! pyridoxal-P. The absorption spectrum was measured after 50 min. at 22°C and this spectrum is also shown in Figure 29 corrected for the dilution. After subtraction of the apoenzyme absorption. the difference spectrum between bound ard free pyridoxal-P has maxima at #5 tryptophanase 4- pyridoxal- -..--. —- tryptophanase P . . . . . . . difference spectrum ABSORBANCE 250 300 340 420 WAVELENGTH Figure 29. Absorption and difference absorption spectra #30. 380. 360. 337. and 325 nu. his difference spectrum represents the difference in absorption betwun the 337 and '+20 m forms of bound pyri- deal-P and the numerous a... hummus in the visible region of the spectrum for free pyridoxal-P. he #30 m peak represents a change in absorption as a result of the reaction shown in Figure 30. which can also account for the peaks at 325. 360. and 380 run. he peak at 337 m is due to the fact that bound pyridonl-P also has a 337 m form. Figure 30. Pyridonl-P binding reaction “7 he difference in the heights of the 280 nm peaks in the spectra of the apoenzyme and apoenzyme to which pyridonl-P has been added can be attributed to the absorption of the bound pyridoxal-P at 280 nm. fiodel studies indicate that the l+20 nm form of the pyridonl-P Schiff's base absorbs strongly at this wavelength. whereas free pyridoxal-P dOOC note fluorescence spectrum of; apgtgntophanase. he fluorescence spectrum of apotryptophanase in K-EPPS excited at 280 m was obtained. as shown in Figure 31. here is one emission maximum occurring at about 360 nm. corresponding to the emission of the tryptophan residues in the protein. Abggption mgtrum 9.; holotmtgphanase in _tl_1_e_ m _o_f_ tetrame- W. In an initial study of holotryptophanase absorption. an attempt was made to reproduce the results of Suelter and 831011.33 Holotryptophanase in activation buffer was passed over Sephadex G-25 fine (1322 cm). eluted with m-EPPS buffer. and collected in 0.5 m1 fractions; the fractions containing the highest concentrations of tryp- tophanase were pooled. In the absorptionspectrum of the sample. the overall intensity due to bound pyridoxal-P is low relative to the height of the protein peak as compared to the spectra of Suelter and Snell. his appears to be due to a resolution of pyridoxal-P from the holoenzyme in the process of preparing the enzyme on Sephadex gels. his phenomenon is discussed in the work of Toraya et al.32 Upon ad- dition of 0.15 M KCl and 10 ml! L—ethionine. a 508 an peak developed as reported by Suelter and Snell. However. the intensity of this peak re- lative to the enzyme peak also was much diminished. Sims the pyridozal-P - apoenzyme complex is quite labile in the O IOO o INTENSITY l J J 300 400 500 , 600 \JAVELEN 6TH ' Figure 31 . Fluorescence spectrum of apotryptophanase ABSORBANCE WAVE LENGTH Figure 32. Absorption spectra of tryptophanase in m #9 absence of monovalent cations. it is necessary to incorporate adequate amotmts of pyridoxal-P in samples in order to maintain a high degree of binding. As an estimate. one might assume a KB of “PM based on the value given by Isom and Dehoss37 for tryptophanase in the presence of sodium. a cation which does not activate except at very high concen- trations. Consider a simple mass action model in which the free pyri- doxal-P concentration is fixed by means of equilibrium dialysis and in which E represents enzyme pyridoxal-P binding sites. E + pyridoxal-P :2 E-pyridonl-P E e pyridonl-P +__ a K Edpyridoxal-P D Using the conservation equation total = E + E-pyridoxal-P one can derive the equation . E-pyridoxal-P KB -1 Fractioml saturation = w = 1 + Etotal p_yridoxal-P Notice that this result is independent of enzyme concentration. By sub- stitution into the last equation. one can calculate the pyridoxal-P concentration necesur-y for 95 i saturation to be 76/44. his value is to be cow to a concentration of 18,114 in pyridozal-P binding sites for a 1 mg/ml tryptophanase solution. Shelter and shell33 dialysed their tryptophanase m1. against m- EPPS containing only 8rd! pyridoxal-P. Based on the previous discussion. 50 it would appear that their enzyme was not completely saturated with pyridonl-P. In Figure 32 are shown spectra of tryptophamse in which 8 pl! and 100p! concentrations of pyridoxal-P are fixed by dialysis: in the former case. the enzyme is only somewhat more than half saturated. Absorption metrum 9; W in Lb; presence g goghate. A comparison of the absorption spectrum of tryptophanase in the pre- seme of potassium given in the preceding section with the spectrum of tryptophanase in the presence of potassium in Figure 25 showa a. dimi- nished 337 mn/ uzo he ratio in the former case. herefore it was decided to study the absorption spectra of tryptophanase in various buffer sys- tems. In Figure 33 are shown the absorption spectra of tryptophamse in potassium phosphate. in phosphate. and nit-phosphate including potas- sium chloride. As conquered to the presence of phosphate ion. the pre- sence of tetramethylammomum ion inhibits the formation of the 337 nm hand. his statement is consistent with the observation of Suelter and cur-1133 that tetramethylanonium ion inhibits the tryptophanase reed- tion. ammo mm 2; .the. W - “than mar. 10 uh L-ethionine was added to the potassium phosphate sample and the reference solution and the absorption spectrum recorder. he spectrum is shown in Figure 3+ and may represent the correct enzyme - L-ethionine complex absorbance relative to the absorbance of the aromtic amino acid residues since the enzyme is completely saturated with pyridonl-P. he L-ethionine - holoenzyme complex in the presence of activating mono- valent cations is able to eliminate the proton attached to the ak-car- bon to give rise to a quinonoid form. he long system of conjugation resembles that encountered in merocyanine dyes. whose spectra are also 2 ABSORBANCE 8 (fl ._.l__.l_l___L l 340 380 420 460 /"\ K phosphate \\ ----- — m phosphate ‘ -—-—--TMA phosphate + KCl WAVE LEN 6TH Figure 33. Tryptophanase absorption spectra 0.6 r AB SORBANCE WAVE L EN 6TH Figure 3“. Absorption spectrum of holotryptophanase-L-ethionine quite similar.“ hey display a definite vibratioml structure with a high frequency shoulder and some less pronounced structure at still higher frequencies. he separation of the vibratioml peaks. 1000 to 1100 ce'l. corresponds to the frequenct of a c-c stretch. Wand—trustw- In Figur- 35 is shown tho fluorescence spectm of tryptophanase in potassium phosphate buffer containing 10’s.! pyridoxal-P. excited at 280 me. he free pyridoxal-P emits very little when excited at 280 nm so there is little contribu- tion to this spectrum from free pyridonl-P. he main features of the holotryptophanase spectrum are a 350 nm peak. presumably corresponding to the emission of the tryptophan residues of the protein. a shoulder at about l+00 nm. and another emission at 510 um. I m mm 2: 9.0. We: -______Ir-°thi°nim 2M1. a totally new result is the fluorescence of the holoenzyme - 53 6 O N INTENSITY I J 300 400 500 600 WAVELENGTH Figure 35. Fluorescence spectrum of holotryptcphanase L-ethionine complex when excited at its maximum absorption wavelength of 508 nm. he fluorescence spectrum is a mirror image of the absorp- tion. with emission minimum at 525 II. 'Ab_r.o_1:p__ntio Lem 9!. W in its. p..__r°s°nco 22 2. 2E2: 2!. mg m. Concentrations of monovalent cations as their chlo- ride salts were chosen to be at least 10 KD33 and. for further conveni- ence in comparing experimental results. buffer and cation concentrations were chosen to agree with a study of pK‘ values for the conversion of the 337 form to the 1120 nm form.“2 Tryptophanase was prepared in m- EPPS. pH 8.0. containing 100,“)! pyridoxal-P: the concentrations of the cations are given in Table 2. Results are shown in Figures 36 and 37. Another set of such spectra were obtained for tryptophanase in m- CBS. pH 9.0. containing 100’“! pyridoxal-P and the cation concentra- tions given in Table 2.. Solutions containing sodium. lithium. and Table 2. Concentrations of cations Salt RbCl CsCl NHuCl LiCl KCl NaCl Concentrationfli) 0.5 0.5 0.025 0.6 0.1 1.0 55 OJ -—— ammonium Lu O" __...potassium U ........- rubidium I! <( co 9; 005- 90 m "( lllLllJllllllLlJll 300 350 400 450 500 WAVELENGTH Figure 36. Absorption spectra in amonium. potassium. rubidium A850 RBANCE lJJllllllillllllllJ 300 350 400 450 500 \JAVEL ENGTH Figure 37. Absorption spectra in sodium, lithium. cesium 56 cesium lost enzyme activity during overnight dialysis. he absorption spectra of holotryptophanase in the presence of ammonium. potassium. and rubidima at the twa pH values are shown in Figures 38. 39. and 1+0. respectively. Certain features of these spectra are immediately apparent. lithium. cesium. and sodium the cations shown by Suelter and Snell33 to have only a weak effect in activating the enzyme. exhibit a low ratio of the 337 m band to the #20 nm band. Cations which activate well. ammo- nium. potassium. and rubidium. have an enhanced 337 an absorption and en increased 337/420 nm ratios the long mvelength peak in the pre- sence of ammonium is red shifted by about 3 nm. In the case of sodium. overall intensity indicates that the enzyme may not be fully saturated ABSORBANCE 111111111111111111 300 350 400 450 500 MMNELENGTH Figure 38. Absorption spectra in ammonium 0.15‘ o I I I I ’8: 2° C ABSORBANCE Q 0 (n J l L 1 14 hi L J 1 300 S's‘o 400 450 500 WAVELENGTH Figure 39. Absorption spectra in potassium \‘ Lilllllll'llAJlLIIL 300 350 400 450 500 WAVELENGTH Figure 1+0. Absorption spectra in rubidium 58 with pyridoxal-P. The lithium spectrum is also anomalous in that the overall intensity appears to be higher than for the other cations and the long wavelength band is blue shifted by about 3 nm. In going to pH 9.0. the better activating cations were able to pre- serve enzyme activity through the period of dialysis. The pH 9.0 spectra display increased 337 nm bands and enhanced 337/#20 nm ratios. The large increase in the case of potassium indicates that the pH 8 to pH 9 region is in a steep part of the titration curve. or near the pKa' This observa- tion.agrees with the pKa value of about 8.5 measured in this buffer.“2 Fluoreggnge Ectra 93; holotmtophanase _ig the. presence 9; a_ $332.3. 2;; monovalent cations. Fluorescence spectra were obtained of tryptophanase in 'IMA-EPPS. pH 8.0. containing 100 lih pyridoxal-P and the concentra- tions of salts given in Table 2. A typical fluorescence spectrum of the buffer alone excited at 280 nm is shown in Figure #1. The buffer showed large emissions when excited at 3uo nm or 420 nm due to the contribution of free pyridoxal-P. In Figure: 1&2, 1&3. 41;. 1+5, 1+6, and h? are shown the fluorescence spectra obtained for tryptophanase in the presence of the various cations. excited at 280 nm. In Figure #8 is shown the fluore- scence spectrum of tryptophanase in ThA-CHES, pH 9.0, containing 100 M pyridoxal-P ans potassium as given in Table 2 along with the pH 8.0 spectrum. These spectra are drawn to the same scale as Figure 31 in which the apoenzyme fluorescence intensity is normalized to 100%. The most striking feature of the fluorescence spectrum of holotryp- tephanase excited at 280 nm is the existence of three emissions. A 350 nm emission is due to the aromatic amino acid residues of the pro- tein.and is also present in apotryptophanase. Emissions at 400 nm and 510 nm originate from the 337 and #20 nm forms of bound pyridoxal-P. 59 1007f- ). E: U) E E Z I E b l 300 400 500 WAVE LEN6TH Figure #1. Fluorescence spectrum of buffer IOOZr ). ’z ‘2 Lu E I I (Kb 4x) 5&3 5mg WAVELENGTH Figure #2. Fluorescence spectrum of tryptophanase in ammonium 60 IOO‘/.r I N TENS! TY 1 _r l J 300 400 500 600 WAVE LE N GT H Figure #3. Fluorescence spectrum of tryptophanase in potassium 100%: I NTENSITY l I j 300 400 500 600 WAVELENGTH Figure ##. Fluorescence spectrum of tryptophamse in rubidium 61 I007.- INTENSITY 300 460 WAVE L EN 6TH Figure #5. Fluorescence spectrum of tryptophanase in lithium Ioo‘ZI' INTENSITY l I 300 400 500 500 Figure #6. Fluorescence spectrum of tryptophanase in cesium IOOLF INTENSITY I _.I 300 400 500 00 WAVELENGTH Figure #7. Fluorescence spectrum of tryptophanase in sodium IooZI‘ : --— p3 9.0 g) ——-——PH 8.0 Ii' 3; | \\~ L 300 400 500 600 WAVELENGTH Figure #8. Fluorescence spectra of tryptophanase in potassium 63 respectively. It is difficult to demonstrate the #00 nm emission direct- ly by exciting at 337 nm because of the emission of the free pyridoxal-P in equilibrium with the enzyme. but its existence can be demonstrated by overlaying the fluorescence spectrum of the apoenzyme. Figure 31. and that of the holoenzyme in potassium phosphate. Figure 35. both at pH 8.0. One can readily see that in the latter spectrum there is addi- tioml intensity on the long wavelength side of the 350 nm peak. he holoenzyme in potassium phosphate exhibits a large 33? rm: absorption at this pH. to which this #00 nm emission can be attributed. Another instance of this additional intensity my be seen in Figure #8. comparing holoenzyme in the presence of potassium at pH 8.0 and pH 9.0. he increase in fluorescence intensity at #00 an can be correlated with the increase in the 337 an band in the absorption spectrum which occurs upon increasing the pH. here remains the question of why the #00 me and 510 nm emissions are excited when the holoenzyme is excited at 280 nm. Several possibili- ties axist. Clearly. the emission of the apoenzyme is quenched upon bind- ing of pyridoxal-P. Since the 350 ran from the protein tryptophan resi dues strongly overlaps the absorption of the 337 nm form of bound pyri- doxal-P. one might propose nonradiative energy transfer with tryptophan as donor and the 337 nm form of pyridonl-P as acceptor. emission being observed from the latter. as an explanation for the #00 an emission. Direct excitation of the 337 nm form at 280 nm is also possible but model compounds absorbing near 337 nm tend to absorb weakly at 280 nm. with a high energy absorption mximum near 250 nm. In the case of the 510 nm emission. the corresponding #20 nm ab- sorption does not overlap the emission of tryptophan. There is an 6# overlap with the #00 nm emission arising from 337 nm bound pyridoxal-P but these forms occur on different protein subunits and distances might well be too great for efficient transfer of energy. In model compounds. the forms absorbing near #20 me often have a 280 rm absorption maximum so that the 510 nm emission might be a result of direct excitation of the #20 nm form of bound pyridoxal-P. However, the 510 nm emission is also observed when the holoenzyme is excited in the 310-380 nm region where the #20 nm form absorbs very little. For this reason. and based on the photophysical behavior of orthohydroxyaldehydes and orthohydron- azonethinesia. a possible explanation for at least part of the 510 nm emission is proton transfer in the excited state. When the 337 nm form is excited. directly or via emery transfer. a proton can be accepted by the imine nitrogen of the enzyme bound Schiff's base. resulting in inter- conversion to the excited #20 m form. If the rate of this process is fast enough for the reaction to occur appreciably during the lifetime of the excited state. emission can be observed at 510 nm. Proof of this hy- pothesis could be obtained by an analysis of fluorescence lifetime data: this topic will be discussed in the section of this chapter devoted to fluorescene lifetime studies. Examining the fluorescence spectra as a function of monovalent ca- tion. one notices the similarity between the spectra in the presence of potassium and of rubidium. However. there is a significant increase in the 510 nm emission in the case of amonium. the other cation that acti- vates well. his is the opposite effect from what might be expected on the basis of the 337] #20 nm ratio which is greater in the presence of ammonium than in the presence of the other two cations. Therefore. it is possible tint ammonium acts to shift the equilibrium in the excited 65 state toward the proton transferred species and consequently toward the 510 an emission: however. it is also possible that there exists some change in the emitting species since there is also a shift in absorption spectrum. In the case of the fluorescence spectra of the poorly activating cations lithium. cesium. and sodium. one sees a blue shift of the 350 m band that can be attributed to the loss of the #00 nm contribution. his fact is consistent with the disappearance of the 337 m band in the ab- sorption spectra. In addition. the sodium sample shows a very high 350 m peak which may be explained if part of the enzyme is in the apoenzyme form with a higher quantum yield of fluorescence: this diminished bind- ing of pyridoxal-P also explains the decrease of the 510 an emission which is observed with sodium. he cesium sample shells an increased 510 nm emission as expected on the basis of the increased #20 nm absorption. As in the case of absorp- tion. the fluorescence spectrum of tryptophanase in the presence of li- thium is anomalous. he decrease in the 510 nm band might be explained by some characteristic of the cation which decreases proton transfer in the excited state. Alternatively. there my be a difference in the ground state since the absorption spectum shows changes in maximum wavelength of absorption. 66 B. FLUOIESCENCE um STUDIES he fluorescence lifetimeq" is an important photophysical parameter whose measurement may be accomplished by single photon counting tech- niques. Iewis et al.“3 discuss the manner in which the true decay function G)t) can be obtained from experimental data by solution of the integral equation 1: D(t) ' f I(t- I) (We) d) e where D(t) is the experimental decay curve and I(t) is the experimental lamp curve measured under identical conditions. In general. extracting lifetimes of less than 10 nsec. requires careful attention to the mathe- matical method. In addition. I(t) is wavelength dependent due to varia- tions in the lamp time characteristics and the spread of photomultiplier transit times. hese effects may introduce complications into analysis of decays of less than about 2 nsec. Ware et al.“ review various deconvolution methods and point out the importance for certain applications of being able to recover the true decay law without 5 2129.12 assumptions of its functional form. he method that they advocate is to represent C(t) as a general sum of ex- ponentials n _t/ X k C(t) = Z ak e K" he problem is linearized by preselection of the B’k's. varying only the ak's. he number of exponentials n is chosen sufficiently large that G(t) is flexible enough to fit a wide variety of decays. with no physi- cal significance being attributed to the xk's or a 's. Once G(t) has 1: 67 been obtained. it is possible to compare this function to various decay laws. Ware et al. have implemented the preceding analysis along with a least squares analysis of G(t) according to a one or two exponential decay law in a computer program which is used routinely in this labo- ratory after being obtained from Dr. William Ware. Ware et al.“ state that the criterion of success for their method is the smoothness of G(t) and also the sum of the squares of the residu- als between experimental points and points generated by convolution of G(t) with the experimental lamp curve. Reasons for a poor fit may in- clude poor choice of the X k's. poor data. or too short a lifetime. In addition. the integral equation itself may fail. In a recent paper. Lyke and More“5 discuss deconvolution of very short fluorescence lifetimes. hta analysis is carried out by nonlinear least squares assuming a sum of exponentials for G(t). Computer pro- grams implementing this method have been obtained from Dr. William Ware and Dr. Luiwig Brand and modified for use in this laboratory. he analysis of fluorescence decays by the method of nonlinear least squares is developed extensively by Grinvald and Steinberg.“6 hey argue. after Knight and Salinger“? that the preferred method of de- convolution assumes a functional form for G(t) with adjustable parame- ters. often a sum of exponentials. Goodness of fit is Judged by the sum of the squares of the weighted residuals. with the correct weights being given by a Poisson distribution in the case of single photon counting data. A feature of the nonlinear least squares program of Dr. Ludwig Brand is the use of a lamp shift correction. discussed by Gafni et al.“8 he energy dependence of the photomultiplier transit time introduces a 68 distortion in the profile of the lamp which these authors claim may be approximated by a relative time shift between the experimental data and lamp curves.IIn this-laboratory the lamp curve was shifted by a spline function interpolation program supplied by homas H. Pierce. Other methods of determining a proper lamp profile have been pro- posed. Britten and LockwoodM9 make use of two reference compounds. A and B. he equation dI (t) I'(t) = I(t) +11 .8:- is used to calculate the corrected lamp curve I'(t) from a range of possible values of q". the fluorescence lifetime of compound A. he I'(t) that best calculates the lifetime of standard B as Judged by the residuals is then used in analyzing lifetime data of the sample of inte- rest. Wong and Halpernso describe a method of attenuating the tail region of the experimental lamp curve in order to better analyze short life- times. A ratio correction technique for the variation of the photomultipli- er time response with wavelength is presented by Rayner et al.51 hese investigators propose alternate collection of the experimental data and lamp curves with the decay curves of a fluorescence standard selected for high fluorescence yield. short lifetime. and fluorescence spectrum spanning the excitation and emission wavelengths of the sample. he de- cay curves for the standard are measured at both of these wavelengths and mathematical analysis yields corrected data and lamp curves to which deconvolution procedures may be applied. McKinnon et al. 53 report computational performance tests of seven deconvolution methods. he method of Ware et al.“ was rejected as being 69 unable to tolerate eyen low levels of noise added to simulated double exponential decay data. The method of Grinvald and Steinbergué was Judged best in its ability to obtain correct values in spite of noise and closely spaced lifetimes. The fluorescence decay curve of 9-cyanoanthracene in cyclohexane is suggested by Gani et al.“8 as a test of the deconvolution procedure for nonlinear least squares amlysis. In the present work, the best fit of the 9—cyanoanthracene lifetime is used to determine a value for the lamp shift correction. In Figure 149 is shown a plot of the fluorescence decay of 9-cyanoanthracene observed at l+50 nm upon excitation at 3+0 run, together with the curve obtained upon convoluting the lamp with the best fit single exponential. IO‘ r COUNTS J I I 1 1_ 20 4o NANOSECONDS Figure 1&9. Fluorescence decay of 9-cyanoanthracene 70 he experimental lamp curve was interpolated and shifted various amounts between zero and 130 channels (0.383 usec./channel) and the values of the lifetime '7'. preexponential a. and izmrecalcuhtod Results are tabulated in Table 3. Table 3. lamp shift corrections Imp shirt . 'r “13' 0.000 0.0545 11.115 211.35 0.200 0.0565 11.10 5.866 0.285 0.0572 11.00 3.856 0.290 0.0572 10.99 3.815 0.295 0.0572 10.98 3.852 0.300 0.0573 10.93 3.879 0.325 0.0575 10.89 4.189 0.383 0.0578 10.81 6.200 0.766 0.0591 10.40 51.14 he fluorescence decay curve for holotryptophanase in m-EPPS. pH 8.0. containing 0.15 M KCl and excited in the 320-380 nm region was measured at #20 nm and 510 nm as shown in Figures 50 and 51. Attempts to deconvolute these data by the method of Ware et al.“ failed because of oscillations in G(t). In Tables it and 5 are tabulated the values of preexponential factors and lifetimes for a nonlinear least squares analysis of the 510 m data. he exponentials are asstuned and the calculations are performed without lamp shift correction and with a lamp shift correction of 0.275 nsec; in the latter. there is a large improvement inizand large changes in the parameters. As discussed previously. these data are something of a worst case 71 I0“- I05 _. U) "z‘ :3 4.. O :0 0 J L L l 20 4O NANOSECONDS Figure 50. Fluorescence decay of tryptophanase. l+20 m 6 IO f 105'- 0) 1— % 4L. OIO . O l 1 L l a] 20 4O NANOSECONDS Figure 51. Fluorescence decay of tryptophanase. 510 nm 72 Table 4. Decay parameters ‘1 ‘2 0.111 2.71 0.002 8.81 9.0a 0.087 2.78 0.001 10.7 8.02 Table 5. Decay parameters with lamp shifted ‘1 ‘2 0.089 1.29 0.058 3.111» 8.35 0.095 2.08 0.011 5.66 5.01 for this method of analysis because of the short. close together life- times and these results cannot be considered to be definitive. However. it can be stated that there are probably two. components to the decay withcnelifetime cf1t02nsec. andasecondcf3to6nsec. Analysis of the l+20 nm decay is further complicated by the pre- sence of scattered light. he data can be fitted with the following expression G(t) = 2.02 oxp(-t./0.321) + 0.170 exp(-t/2.10) + 0.03“ ozp(-t/7.oo) he first component of the decay is due to scattered light whose life- time departs from zero because of the lamp shift. An attempt to correct for lamp shift in this amlysis led to a divergence of the method as the short lifetime approached zero. Short lifetimes for the bound forms of pyridcnl-P are not unex- pected considering the short lifetimes of free pyridcxal-P. An improve- ment in the results presented here might be brought about by the method of hyner et alélwhich. however. uses a more complicated system of 601- lecting data than is presently available in this laboratory. Better 73 lifetime data is highly desirable in order to use lifetimes as para- meters for the investigation of changes in the enzyme due to buffers. monovalent cations. etc. Analysis of lifetime data could also give a direct proof of the existence of excited state proton transfer in pyridoxal-P enzyme. the possibility of which has been argued from the results of studies of model compounds up to m. 7’4 C. KIMC STUDIES Kinetic studies on tryptophanase were performed in collaboration with a group of investigators studying the structural and functional as- pects of enzymes as deduced by static. dynamic. and theoretical methods. other members of the group during the course of this study have included D.S. June and 0.11. Suelter. Biochemistry Department: T.H. Pierce. R. Cochran. S.V. Elias. F. Halaka. I. Behhaham-Nejad. F.H. Horne. and J.L. Dye. Chemistry Department; and ILA. El-Baycumi. Biophysics and Chemistry Departments. It is known that there is a conformatioml difference between holo- tryptophamse in the presence of sodium and hclctryptcphanase in the presence of potassium.29 Such a conformational change might be associa- ted with coenzyme rotation as is discussed in connection with the model developed in (hapter 3. Since decreasing the pH and removing activating monovalent cations appear to have similar effects on the absorption spectrum of the enzyme. one might also expect changes in conformation with pH. All of these considerations. as well as analysis of the enzyme reaction mechanism]?2 suggested the desirability of observing the effects on the absorption spectrum of sudden changes in pH or monovalent cation. June et al.53 performed three rapid scanning stopped flow studies of the effect on holctryptophanase of a change in pH or potassium con- oentration. In each of these experiments there was a clear observation of a slow process in which a 1120 m absorbing form disappeared and a 337 us form appeared. or vice versa. In the pH drop experiment (pH 8.53 to pH 6.72). interconversion of the 337 nm and 1.20 nm absorbances 911... a first order decay for the 75 337 m band and first order growth of the #20 m band follewing a rapid drop, of pH. Initial (abrupt) changes occur within the mixing time of the instrument (about 6. 5 msec. ). as evidenced by differences between the initial enzyme spectrum and a spectrum which has been syn- thesized to represent the enzyme Just after mixing. Upon raising the pH from 7.38 to 9.30. the l+20 nm and 337 nm forms also appear to interconvert at a similar rate. here are some very fast changes as well as a very slow process in which absorbance decreases at 325 and lH25 nm and increases at 360 run. he fact that these last spec- tral changes resemble a difference spectrum between bound and free pyri- doxal-P forms. a process that might well occur if pyridoxal-P binding were pH dependent. suggests that holoenzyme-apoenzyme interconversion may account for all or part of the very slow process. he third experiment involved the exchange of potassium for sodium in a sanple of holoenzyme. Since KD for sodium binding to the enzyme is about 1&0 m and for potassium is 1.4 nil. cation exchange can be effected by mixing equal volumes of identical concentrations. as demonstrated by a calculation of reaction velocity attaimble in this situation. Using the equation for competitive inhibition (as suggested by 0.11. Suelter) ' + [+3 {_=-}+ VKEK+1 (1+x:+) where v is the expected velocity. V is the maximum velocity (observed in the presence of K+ alone. KM'I- is the dissociation constant for 14+, and [ H+lis the concentration of the cation M+. one can calculate the value v = 0.9% V for 50 ml! final concentration of each cation. his can be achieved by making up the enzyme in 100 mM sodium and mixing it with 76 r I rfi T I r i ' I '11171 02;- . L 1 l 00 j to 00 ‘5 a“a i i e i O r C __‘o e 3 ‘ A 1 0 [- 0" . . I U ‘ao a l 80.”. . a asa... 1 4° :. . .0. .005 .‘ 5.5 i" ° 3 3° 0 I g +.e.. . Oo 00 0‘ ' ‘ ‘ o O 0' L- O ‘ i ‘..‘ ° 0.01:4 1 1 1 1 l 1 L 1 1 1 1L1 L4, 320 340 360 380 400 420 440 Wavelength (nm) Tryptophanase spectra (path length 1.85 cm) corrected for free pyridoxal-P absorption. observed in the pH drop experiment: zero time. 0: first spectrum after mixing (1 = 6-12 ms). 0; spectrum after comple- tion of the first-order process (I = 8 s). A. ’3 U —-0.00 5 '9 -0.04 g .0 -0.08 § wovekflB'h (”fl Absorbance difference-wavelength—time surface observed in the pH jump experiment. The spectra of free pyridoxa|~P and of trypto- phanase at zero time (A000) have been substractcd to give these difference spectra. Figure 52. pH drop and pH jump 53 77 a solution containing 100 mh potassium. A concentration of 80,“' M pyridonl-P in the enzyme sample was near the mnmum value feasible for use in the stopped flow apparatus. As was shown earlier. tryptOphamse in 100,04 pyridoxal-P my not be completely saturated with coenzyme. herefore. effects of additional pyridcxal-P binding to the enzyme when potassium is added are likely to be observed if they occur on a time scale similar to the other pro- cesses. i.e.. interconversion of free and bound forms cannot be cleanly separated from interconversion of bound forms. his 'pctassium jump' experiment clearly showed the slow 420 nm to 337 nm interconversion with approximately the same rate as in the pH change experiments. Figure53 is an absorbance-wavelength-time surface generated by computer for this experiment. Figure 5% shows the first spectrum taken after mixing and the last spectrum taken 68 see later. Figures 55 and 56 show the time cuts of the absorbance at 337 m and #20 m. respectively. hese changes follow rate laws more complex than first order. A preliminary investigation of the rate of pyridoxal-P binding to the apoenzyme is shown in Figure 57. hese data demonstrate that the rate of binding in the buffers used for pH change stopped flow experi- is much faster than the rate in potassium phosphate buffer that has been used to study birding.3u' his result confirms the idea that binding ef- fects may well be a factor in experiments in which pH or monovalent ca- tion is changed. ‘ It is important to add the fact that much closely related research on tryptophanase has been carried out by members of the enzyme group since the time I concluded my work and began writing this dissertation. ABSORBANCE 78 Figure 53. Absorbance-wavelength-time surface First spectrum __ _- last spectrum (68 sec) l l 340 380 420 WAVELENGTH Figure 54. First and last spectra ABSORBANCE ABSORBANCE 79 l 4 IO 20 SECONDS Figure 55. Time out at 337 mm "afikflu l I l I L 20 40 S ECON DS Figure 56. Time out at 420 nm 4325—. ABSORBANCE 80 MINUTES Figure 57. Pyridonl-P binding to apctryptophanase CHAP'IER5 mums AND 1031110136 A. 0mm or g. 00LI B/1t7a 3;. £2}; B/1t7a is a mutant of E. a]; strain B which is constitutive for tryptophanase. Cultures were maintained on plates ard slants of minimal agar. M 9.2!! 32.9 2;. KKZPOu 0.9 8. K290“ 2.2 g. 013.9280“ 0.3 g. bactoagar 5.1! g. After autoclaving. add separately autoclaved inlole (1 mg/ml) 3 ml. glucose (30 mg/ml) 10 ml. and filter sterilized minerals. 1 ml. M. .122 E. 0.6122320 0.03 e. Hssou'flzo 3 e. FOSOu'7HZO 0.03 8. In order to grow starting materials for the preparation of tryptophanase. a 10 ml. culture of nutrient medium was innoculated from a single colony obtained from a culture streaked on minimal agar. 81 82 mtrient media . Ll. casein ivdrolysate . 10 g. KHZPOu 7.5 g. yeast extract 0.1 g. K2904 16.5 8. bactotryptone 10 g. (mh)230h 1 g. After autoclaving. add separately autoclaved indole (1 mg/ml) 25 ml. and filter sterilized minerals 1 ml. L-tryptophan (10 lug/Ml) 3.11mi. mugs. 199, 9;. CaCIz'ZEzO 0.03 g. ZnSOu'7820 0.01 g. rosou'7azo 0.03 g. ousou'sazo 0.01 g. wou'mzo 3 g. Nam. 2.5 g. HnSOu'7820 0.01 g. Successive innoculaticns of the entire culture were made into 100 ml. . 5 1. and 100 1. volumes. with growth times of approximately 12 hr. into stationary phase. he 100 1. culture was harvested by centrifugation at an optical density at 660 nm of h.% with a yield of 197.7 g of wet weight of cells. which was frozen in 30 g portions for up to three months before being subjected to purification. Cells were grown with 100 rpm mixing to prevent sedimentation and a slow rate of aeration. A typical yield of E. 29;; in an anaerobic culture is about 150 g . whereas Mr aerobic conditions the yield may be more like 450 g. he nearly anaerobic growth conditions are apparently important for the maximum production of tryptophanase and we have generally found that the higher the total 33 cell wet weight harvested. the lower amount of tryptophanase per 30 g. of starting material. B. PEEPARATION OF THE SEPHAROSE 00me Ligands are coupled to Sepharose by cyanogen bromide in a procedure adapted from that of Shaltiel and Er-elr 1. Dissolve 15 g. of cyanogen bromide in a small volume of p-dioxane. 2. Dissolve 75 g. of 1.7-diamincheptane in 300 ml. of 520: adjust pH to approximately 9.5. 3. Wash 150 g. of Sepharose 1&8 with 5 volumes of H20. Resuspend in a total volume of 300 ml. it. Add the cyanogen bromide to the Sepharose with stirring. Maintain the pH at 11. 5. After 8 min.. stop the reaction by the addition of ice. 6. Wash with 1 l. of £120. Resuspend in 300 ml. 0.1 M 11.13003. 7. Immediately add the 1.7-diamimheptane with stirring: adjust the P3 to 9.5. 8. Stir overnight at room temperature. 9. Wash with 320. 0.1 H NaHCOB. 0.05 M NaOH. H20. 0.1 14 03300011, and 320. he Sepharose column is stored in the cold in the presence of 0.02 $ sodium azide as a preservative between purifications of the enzyme. It is regenerated with washes of NaOH. 820. HCl.. and H20. followed by adjusting the pH to 7.0". An earlier version of this coltnnn was made with only half as much ligand as above. he column 8h» failed to purify tryptophanase properly and this fact explains the report by Suelter et al. that it was necessary to add a IEAE column to the purification scheme. Also. we have observed the column to fail from time to time and in an atteth to protect the Sepharose we have made the regeneration procedures more gentle by decreasing the acid and base concentrations from 0.5 h to 0.2 3. Column material that has failed can be observed to stick to glass or even to form lumps. C. PURIFICATION OF WPEANASE he purification procedure is adapted from the procedure of Suelter et al."’0 £22.19 30 g. wet weight of cells of Q. 3211.; B/1t7a were suspended in 150 ml. of cold Buffer A and treated on ice by sonication for three 3 min. cooling periods bettreen. Cell debris was removed by centrifugation. m he supernatant from Step 1 was diluted with Buffer A to a protein concentration of 10 mg/ml. as determined by the method of Warburg and Christian. and adjusted to pH 6.0 with 10 $ acetic acid.. Nucleic acids were precipitated by the addition of onefifth volume of 2 $ protamine sulfate. followed by stirring at room temperarure and centrifugation. m he supernatant solution from Step 2 was adjusted to pH 7.0 with 10 at NHnOH. Solid (113“)2301, (211.8 g/100 ml) was added gradually with stirring over a 30 min period on ice. with the pH maintained at 7.0 with ”mums. After removing the precipitate by centrifugation. solid 0130280“ (15.6 g/ml) was added in the same fashion. 85 The precipitate was collected by centrifugation. dissolved in a minimum volume of Buffer B. and adjusted to one-third the volume of the supernatant solution from Step 2. One-hundred ml. portions were heated to 65°C in a 72°C water bath and kept at this temperature for 5 min. The denatured protein was removed by centrifugaticn. Step 4. The supernatant solution from Step 3 was made 60 % saturated in (Nan)zs°ti by addition of 25 g/ 100 ml gradually with stirring over a 30 min period on ice. The precipitate obtained 'by centrifugation was dissolved in a minimum volume of Buffer C and dialyzed against two changes of 1 l. of Buffer C overnight.. Step 2. The dialyzed solution from Step 4 was applied at room temperature to a Seph-C‘7-NH2 column (2.5xfl0 cm) equilibrated with Buffer C. The column.was washed with Buffer C to remove non- adsorbed.protein. .At this point. monitoring at 280 nm and collection of fractions of the effluent was begun in order to be able to save the enzyme should it fail to stick to the column. Tryptophanase was eluted with a 1000 ml linear gradient composed of 500 m1 of Buffer C and 500 ml of Buffer C containing 0.2.M.NHuCl. Tryptophanase elutes at 0.13-0.15 M NhnCl. §£gg_§: {Active fractions from the Sepharose column were pooled and dialyzed overnight against Buffer D. If apoenzyme was desired, the enzyme solution was made 10 mH in penicillamine before dialysis.. The precipitated material in the dialysos bag is centrifuged and resuspended in a minimum volume of Buffer D. Tryptophanase may be stored in.this manner in the refrigerator for several months: additio- nal sulfhydryl compounds may be added from time to time. 86 Oneoanieolateasmuchas 300mgof tryptophanasebythe preceding procedure. although the dialyzed solution from Step l4- neededtobesplitandhalfandtheSepharose columnruntwice. his high yield was attributed to a particularly good batch of E. a; B/1t7a. 0n the other hand. a poor batch of bacteria and a failing column have yielded as little as 30 mg of tryptophanase. reproducibly. upon following the same. procedure. Ms: A. 500 ml. potassium phasphate. pH 7.0. 1 H 50 ml Ha m, 0.1 ll 10 ml dithiothreitol. 0.1 l 1 ml are; a.._200 a; potassium phosphate. pH 7.8, 1 H 20 ml Ha mm. 0.1 I! it ml pyridonl- . 20 i4 2 ml (Manson. solid 22 g dithiothreitol. 0.1 l! 2 ml M2 9' Ll. ammonium phosphate. pH 7.0. 1 H 25 ml Bk EMA. 0.1 H 10 ml glycerol. pure 100 ml dithiothreitol. 0.1 H 10 ml m .D. potassium phosphate, pH 7.0. 1 M 100 ml 87 u. m. 0.1 u 10 ml (““230”, solid 662 g fi-mercaptoethanol. pure 0.653m1 320 900 ml Additional (NHu)ZSOu is added in the amount 66.2 g/ 100 ml of the volume in the dialysis bag (90 fl saturation). D. MRS FOR 13!?um Luna" gin—1111:: 3.0.9. 9.1..- potassium phosphate, pH 8.0. 1 H _ 5 .1 KCl. solid 0.75 3 Bk EDTA. 0.1 M . 2 ml dithiothreitol. 0.1 H 1 ml pyridexal-P. 20 ml! 0.5 m1 glycerol. pure 20 ml g5, g; m-EPPS buffer, 599 ml. N-Z-Ivdroxyethylpiperasine propane sulfomc acid 3. 15 g EDTA. 0.1 H. titrated to pH 8.0 with tetramethyl- ammonium hydroxide (mt-on) 0.5 ml dithiothreitol. 0.1 H 5 ml Titrate to pH 8.0 with aux-on 88 a: a! W mass. in mi. phosphoric acid. concentrated m, 0.1 H, titrated to pH 8.0 with m-on dithiothreitol. 0.1 H Titrate to pH 8.0 with m-on a: an. m mass. :29 mi N—Z-hpdroxyettvlpiperaaine propanesulfonic acid nun, titrated to pH 8.0 with m-on dithiothreitel. 0.1 M ‘61. solid Titrate to pH 8.0 with fill-OH 9L1. ! 22.—£24....“ um M59. m, 59.9. 9.1." potassium phosphate. pH 8.0, 1 u m, 0.1 n, titrated to pH 8.0. 1 u dithiothreitol. 0.1 H (some preparation of sodium phosphate buffer) 194.__§m-EPP mam tio . $92.19.- N—Z-tvdroiqethylpiperayine propanesulfonic acid m. solid dithiothreitol. 0.1 H cation Titrate to pH 8.0 with m-oa 1A2 g. 0.5 m1. 5 ml- 3.15 s. 0.5 m1. 5 n1. 1% s- 50 311- 0.5 ml. 5 ml. 0.63 g. 0.029 a. 1 m1. (table 2) 89 im__M-cm8c_oo_tsiaisaams.ioam. , cyclohexylaminoethamsulfonic acid 0.10% g. m. solid 0.029 8. dithiothreitol. 0.1 M 1 ml. cation (Table 2) Titrate to pH 9.0 with TEA-OH E. PIEPARATION 0F TETBAMEWONIU‘I HIDRDIUE (ma-OH) 1. Wash mm 128. 20-50 mesh. with 0.1 )4 K03 2. Pour column and wash with 1120 untill effluent pH is 7. 3. Dissolve “.38 g. of recrystallized MC]. in 1+0 ml of 320. ll». Apply nil-GI. to column. Wash with 820 and monitor the pH of the effluent. 5. When effluent pH turns basic. collect 80 ml. The resulting nil-OH is about 0.5 M in concentration. It should be used immediately or stored frozen until use. I. ENZIME ASSAY OF TRIP'IOPHLNASB 1he pseudosubstrate S—o-nitrophewl—L-cysteine (SOPC) was used for the spectrophotometric assay of tryptophanase activity. Suelter et al.u0have shown that SOPC under goes «hp -eliminaticn to give an o-nitrcphenolate ion and that the reaction can be followed at 370 nm with a 48 = 1860 liters mold cut-1. ‘Ihe assay procedure involves preparing a cuvette containing 0.5 ml volumes of 0.1 M potassium phosphate buffer. pH 8.0, and 1.2 ml! SOPC at 90 3°eC. lypically the reaction is started \with a few microliters of enzyme solution. 'Ihe initial velocity is determined from the steepest part of the absorbance versus time curve and activity is calculated from the eqmtions = (AbsorbenceBYo/gziy) x dilution 1 . activity] ml “Die specific activity as calculated by dividing the activity] ml by the protein concentration in mg/ml. Suelter et al. report 6 = 0.795 as nil-1 no ell-1 for use in determining protein. G. WW Absorption spectra were recorded using a Cary 15 spectrophoto- Inter. Enzyme assays were conducted on a Beckman DU modified with a Gilfcrd Model 220 optical density converter. Moat emission spectra were obtained with an Aminco-Keirs spectroquoreIIeter with a high pressure Xenon arc lamp and modified with an M 9781 R photomultiplier tube. 'Ilie emission spectrum of the holntryptophanase-Ipethiordne complex was obtained with a compomnt system for higher resolution. The excitation wavelength was selected by a Bausch and Lomb 10 cm grating blazed at 3000 R in a husch and Lamb 500 mm monochromato‘r. A 'I'EW sample holder with collimating lenses was used for the right angle alignment. line emission monochromatcr was a 750 mm Curry-lunar spectrometer (SPEX 1711-11) which utilized a 10 cm Bausch and Lamb grating blazed 91 at 5000 R. Emission was detected with an m: 9558 cu photomultiplier tube operated on a Fluke 14-123 power supply. The system utilized a PAR H—3 lock-in amplifier with reference provided by a light chopper. The amplified emission was displayed on a strip chart recorder. Nanosecond time resolved fluorescence decay measurements were ob- tained with a single photon counting apparatus. The nanosecond flash lamp is a thyratron-gated lamp with a typical pulse rate of 35 KHz. The lamp is detected with a 1P28 photomultiplier tube and the resulting pulse. after discrimination against low level noise (ORTEC Model #36) is used to start the time to amplitude converter (TAG. OR‘I‘EC Model #57). Fluorescence photons are detected by a 56DUVP photomultiplier tibe and the resulting pulse is used to stop the TAC. lamp pulse and photon counting rates are measured with HONSANTO Model 150A digital counters. Time Jitter is eliminated by using an OR- EC Model #63 discriminator. The calibration of the TAC is performed with the ORTEC Model 162 time calibrator. Output of the system is fed to a NUCLEAR DATA Model 1100 multi- channel analyzer. ‘l‘he contents of the multichannel analyzer memory can be displayed on an oscilloscope (HEWIETT-PACKLRD Model 1308B) or can be fed to a teletype equipped with tape punch. Communication with the MSU CDC 6500 computer is accomplished by teletype through an acoustic coup- ler and a digital plotter (HEWLETT-PACKARD Model 7200A). 92 n. RAPID-SCANNIN} 3'10me 'mw KINETICS MEASUREMENTS The instrument used for rapid-scanning stepped flow kinetics stu- dies was specifically designed for enzyme studies by H056. based on an earlier system of Papadakis et al.57 Repetitive scan rates of up to 150 scans per second allow observation of processes occurring on a mil- lisecond time scale after an initial dead time of about 6.5 msec. The system is constructed entirely of inert nterials and requires only small volumes (typically about 12 ml) in order to protect and conserve precious biological samples. “his light source is a 1000 W Xenon arc lamp. The system utilizes a seaming monochromator with scan rates of 3 to 150 spectra per second. Sample and reference sigmls proceed through a beam splitting fiber op- tics line to matched photomultiplier tubes. with the photomultiplier outputs being amplified by a logarithmic amplifier. nu sampling system uses a sample-and—hold amplifier and a 12-bit A to D converter. The system is interfaced to a IEC PIP-8/I commuter. Calibration procedures for the system involve collection of the spectra for neutral density filters in order to determine absorbance values. The spectra of holmium oxide and didymium glass are also col- lected in order to calibrate the wavelength intervals. Calibration data is suhnitted for analysis to programs on the MSU CDC 6500 computer. In addition to the software for the control of the acquisition of data. there is available on the PIP-Bl]: software for the analysis of data that can be used during the experimental session or at a later time. Spectra can be displayed. averaged. or expanded and solvent or infinity spectra subtracted. Time cuts at particular wavelengths can be 93 implemented. Finally. data canbe transferredtotheMSUCDC6500 com- puter for calibration procedures and a powerful battery of nonlinear regression programs. KINFIT. A compact and elegant display of the re- sults of a rapid-scanning stopped flow 'push' is the absorbance-wave- length-time surface. one of which is shown in Figure 53. CHAPTER 6 FUTURE WORK As discussed in Chapter 1. B6 compounds often have numerous ionic and tautomeric forms. Enzyme bound Schiff's bases are no exception and this fact explains the large amount of computer analysis of the absorp- tion spectra of 36 compounds in the literature. Johnson and Metzler3. for enmple. discuss the methodology for resolving enzyme spectra into the spectra of the individual ionic and tautomeric forms and present such a spectral resolution for aspartate aminotransferase. In general. such methods involve a nonlinear least squares fitting of the indivi- dual spectra which best reconstitute the overall spectrum. he absorption spectrum of holotryptophamse could be amlyzed by the method described by Johnson and btzlera. Although objections might be raised to the use of lognormal distributions. the method has the ad- vantage that programs are available. Certain features of the spectra could be used to extract information about the ionic and tautomeric forms present in the enzyme. Spectral forms thus generated might also be useful as inputs into a more sophisticated analysis. One such analysis is the analysis of kinetic data by the method of principal components.58 Ionic and tautomeric forms of the enzyme as well as reaction intermediates may be obtained by this method. The ex- periments in which pH or monovalent cation is changed are particularly 9“ 95 good candidates .for this analysis because of their relative simplicity. Similarly. the fluorescence spectrum of heletryptophanase is com- posed ef the spectra of several ionic and tautomeric forms of the en- zyme bound Schiff's base and free pyridexal-P which is necessarily in equilibrium with the enzyme. Several approaches to the analysis of multicempenent fluorescence spectra are reported in the literature. Weber60 discusses a matrix method for the resolution of fluore- scence spectra. he necessary data are elements of an array of values of the fluorescence intensity for pairs of excitation and emission wavelengths. mathematical analysis of which results in the number of components present in a mixture. Warner et al.61"62 implement the collection of the 'emission-exci- tation matrix' by means of a video fluerimeter which uses a low light level television sensor. he spectra of the individml components are fitted by a least squares method employing known spectra or a principal component analysis. O'Harer and Parks63 discuss the technique of selective modulation in which the excitation wavelength is modulated and the corresponding emission is selected by means of a lock-in amplifier. This meted is de- monstrated by the authors for mixtures of up to three components. wshl and Brechenahés suggest that an analysis of the fluorescence spectrum of a mixture can be accomplished by analysis of fluorescence lifetime data. If each of the components is a single exponential emit- ter. the mixture will have a decay law with a term for each component of the mixture. The coefficient of each term will be proportional to the concentration of that component and the quantum yield of fluore- scence for that component. 96 Another area for further study is excited state proton transfer. Usually this phenomenon is investigated by experiments in which the solvent or temperature is varied and the rate of proton transfer is measured. These experiments would give ambiguous results in the case of a small molecule embedded in a protein which itself would respond to these changes of parameters and thus cause changes in the environ- ment of the small molecule in an undetermined manner. One could not directly prove the excited state proton transfer hypothesis for hole- tryptophanase by such experiments. A possible method of proof is suggested by Stryer66. who discusses a deuterium isotope effect on the emission of compounds in which excited state proton transfer is known to occur. He interprets the change in spectral shapes for molecules in deuterated media as resulting from changes in rates of proton transfer due to the effect of the deuteron participating in the reaction: changing the rate changes relative pro- portions of the original form and the proton transferred form and hence the emission. The absence of a deuterium isotope effect cannot be taken as conclusive evidence that there is no excited state proton transfer because even deuteron rates may be fast enough so that equilibrium is established within the lifetime of the excited state. Also. the proton transfer my be internal to the protein. i.e.. assisted by a protein residue . and therefore independent of the presence of deuterium in the exterml medium. Another method of looking at this question is suggested by Leken et al.67 who advocate the use of excited state proton transfer as a bio- logical probe. Rates of excited state proton transfer may vary with en- vironment in which a molecule finds itself and therefore. for a small 9? molecule embedded in a protein. these rates may be used to detect changes of protein conformation. Conformational analysis of pyridexal-P amino acid Sciff 's bases 1° could be extended to the calculations as performed by Weintraub et al. of properties of the absorption spectra as a function of cenfemtienal parameters. Also. a positive charge representing a monovalent cation could be incorporated into such calculations. Detailed knowledge of the effects on the absorption spectrum of conformation and a positive charge could be used to answer such questions as whether the enzyme holds the coenzyme in one or more preferred conformations and which positions for the monovalent cation are consistent with spectral data. Such information might assist in attempts to model species present at the active site of the enzyme and to unravel the details of the cata- lytic act. BIBLIOGRAPHY 1. Mahler. 3.11.. and Cerdee. E.H.. Biological Chemigtgz. 2nd ed.. Harper and Row Publishers. New York. 1971. 2. Elguere.J.. Marzin.C.. Katritzky. A.R.. and Linda.P.. The. Tautemerism 9; kteremleg. Academic Press. New York. 1976. 3. Johnson. R.J.. and Metzler. 13.3.. Methods in Enzymelegy 19;. 433(1970). u. El-Bayeumi. M.A.. El-Asser.M.. and Abdel-Halim. F.. J. Amer. Chem. 30°- 23» 586. 5900971). 5. Hetzler. D.E.. Cahill.A.. Harris. C.M.. Likes. J.. and Tang. 3.. Fed. Prue. 3.5.: 8550-977). 6. Karube. L. and Matsushima. 1.. (hem. Pharm. Bull. 2 . 2568(1977). 7. Kupfer. A.. Gani. V.. ans Shaltiel. 8.. Biochem. Biophys. Res. Comm. 22, 100M197”. 8. Gani. V.. Kupfer. A.. and Shaltiel. 5.. Biochemistry 11. 129+(1978). 9. Dentini. H.. Perretti. G.. and Savino. M.. Biopolymers 12. 897(1978). 10. Weintraub. H.J.R.. Tsai. M.-D.. Byrn. S.R.. Chang. C.-J.. and Floss. H.J.. Int. J. Quantum Chem: Quantum Biol. Symp. No. 3. 99(1976). 11. Heresev. Iu.V.. Bazhulina. N.P.. Cheraskasina. L.P.. and Karpeiskii. n. 1a.. Biefizika 1g. 397(1967). 12. Harris. C.M.. Johnson. R.J.. and Metzler. D.E.. Biochim. Biophys. Acta 1:21. 181(1976). 13. Savin. F.A.. Bazhulina. N.P.. and Merosev. Yu.V.. Molekulyarnaya Biologiya z. 67u(19_73). 11+. 15. 16. 17. 18. 19. 20. 21. 26. 27. 28. 29. 31. 99 Bashulim. N.P.. Kirpichnikov. M.P.. Morosov, Yu.M.. Savin. F.A.. Simvim. I..B.. and Florentiev, v.L.. Mol. Photoehemé. 367(197u). Horosov. Iu.V.. Bashulina. N.P.. Cherakashina. L.P.. and Karpeislcii. M.Ia.. Biofizika ;2_, 773(1967). Bridges. J.W.. Davies. D.S.. and Williams. R.T.. Biochem. J. 2g, “51(1966). Arrio-Dupont. 14.. Photochem. Photobiol. 1;. 297(1970). Weller. 1.. 2.12m; }_n_ Reaction Kinetic; (6. Porter. ed. ). Vol. II. Pergamom Press. London. 1961. Arrio-Dupont. M.. Biochem. Biophys. Res. Comm. &. 653(1971). Veinherg. S.. Steinberg. I.Z.. and Shaltiel. S.. Metab. Intercomers. Enzymes. Int. Symp.. nth, nu(1975). Veinher. 3.. Shaltiel. 8.. and Steinberg. I.Z.. Israel J. Chem. ,1_2__. #214197“). Pullman. B.. and Pullman. A.. Minn Biochemism. Interscience Publishers. New York. 1963. Roberts. J.D.. and Caserio. 11.0.. B322 W 2; 9539;; _Ch__e_- m, 14.1. Benjamin. Inc.. l-lenlo Park. 1964. Ivanov. V.I.. and Karpeisky. H.Ia.. Advances in Enzymology 33. 21 (1969). Shaltiel. 8.. and Cor‘tijo. 14.. Biochem. Biophys. Res. Comm. 9;. 594 (1970). . Jones. D.C.. and Cowgin, R.w., Biochemistry £9, 1.276(1971). Cortijo. 14.. Steinberg. I.Z.. and Shaltiel. 3.. J. Biol. Chem. 226;. 933(1971). Snell. 3.3.. Advances in Enzymology 93. 287(1975). Suelter. C.H.. Science 16g. 789(1970). 32. 33. 3“. 35. 36. 37. 38. 39. 40. 1+1 . 1+2. 1+3. 165. 1+7. “-9. 50. 51 . 100 Toraya. T.. Nihira. ‘13.. and M111. 5.. Eur. J. Biochem. 62. 1419 (1976). Suelter. C.H.. end Snell. 3.3.. J. Biol. Chem. 22. 1852(1977). Hogberg-Ihibaud. A.. Reibeud. 0.. and Goldberg. 14.. J. Biol. Chem. 22;. 3352(1975). Fenske. J.D.. end DeMoss. 3.1).. J. Biol. Chem. 252, 75910975). Isom. 3.0.. and DeHoss. R.D.. Biochemistry 11:. h291(1975). Isom. 3.0.. end DeHoss. 3.1)., Biochemistry gt, 1298(1975). Merino. I.. end Snell. 2.3.. J. Biol. Chem. .232, 2800(1967). Snell. E.E.. and Bum-1. s.J., gem. vol. 2, 3rd ed.. P.D. Boyer. ed.. Academic Press. New York. 1970. Suelter. C.H.. weng. J., end Snell. E.E.. Anal. Biochem. 6, 221 (1976). Echo. E.G.. Spectroohim. Acts 1;. 192(1958). June. D.S.. personal communication. Louis. C.. Ware. W.R.. Doemerw. L.J.. and Nemzek. T.L.. Rev. Sci. Inetr. £14.. 107(1973). Ware. W.R.. Doemerw. L.J.. and Nemzek. T.L.. J. Phys. Chem. ZZ- 2038(1973). lake. 12.1... end Ware. w.R.. Rev. Sci. Instr. fig. 320(1977). Grinvald. A.. end Steinberg. I.Z.. Anal. Biochem. 12. 583(1974). Knight. 1.33.. end Selinger. B.n‘.. Spectrochim. Acta 21;, 1223 (1971). Gatni. A.. Modlin. 3.1... end Brand. 1... Bioptm. J. 15,. 263(1975). Britten. A.. and Lockwood. 6.. Biol. Photochem. 1. 79(1976). Wong. D.K.. and Halpern. A.l~i.. Photochem. Photobiol. 2i. 609(1976). Rayner. B.. McKinnon. A.E.. Szabo. A.G.. end Beckett. 9.1.. Can. 58. 59. 60. 61 . 63. 65. 101 J. Chou. fl. 321+6(1976). McKinnon. A.E.. Szabo. A.G.. and Miller. D.R.. J. Phys. Chem. 2. 156M197”. June. D.S.. Kennedy. B.. Pierce. '1‘.H.. Elias. S.. Halaka. F.. Beh- hahani-Nejad. 1.. El Bayoumi. A.. Suelter. 0.3.. and Dye. J.L.. J. Amer. Chem. Soc. .121, 2218(1979). Slnltiel. 8.. end Er-el. 2., Proc. Nat. Acad. Sci.(U.S.) 29. 778 (1973). Ho. Cir-B.. Diseertation.~.1fi.chigarr8tate University. 1976. Papadalc'ls. N.. Coolen. R.B.. and Dye. J.L.. Anal. Chem. 21. 1611-11 (1975). Cochran. R.N.. Dissertation. Michigan State University. 1977. Coolen. R.B.. Papadakis. N.. Avery. J., Enke. 0.6.. and Dye. J.L.. Anal. Chem. ‘12. 16h9(1975). Weber. 6.. Nature 133. 27(1961). Warner. 1.11.. Callie. J.B.. mvidson. 13.1%.. and Christian. G.D.. Clin. Chem. 22. 1u83(1976). O'Haver. r.c.. end Parks. w.n.. Anal. Chem. g. 1886(1974). wenl. Ph.. end Auchet. J.C.. Biochim. Biophys. Acta 2_8§.. 99(1972). Wahl. Ph.. and Brochon. J.C.. in C. Sadron. ed.. mm 9}; Conformational M1 in Biolog; cal Macromolecules. D. Reidel Publishing Co.. Dordrecht-Honand. 1973. Stryer. L.. J. Amer. Chem. Soc. _8_§. 5708(1966). Loken. 11.. Hayes. J., Gohlke. J., and Brand. L.. Biochemistry 13;. "779(1972). Faeella. P.. Turano. C.. Giartozio. A.. and Hammady. I.. Ital. J. Biochem. 1_o. 174(1961). nICHIan STATE UNIV. LIBRQRIES [NI “11W W “I WI "ll IN ll WNW WW \IIHI 31293100643232