I I_I_I_I=I=l 7 O 0 to .2:- .l EN [1,: < d. Mfrhlflflwhfithwflfllgnl: {WUnMaJ‘iu-Rx; 91:. c. an, :23». .. Efifi: emu... ., V. g; _ 3%? V. 3,”... {Wadi a? 35% V V..wsr.V ..§.zz§?§, .. . . graviduw: . v :5 r . . . X l n. . E. .3 7 . , 1.3!? km“: x .1... , Ii! 57 IRA—3:13)}... ,, V... . . , 253% £55, - V 5. . a. ixquai... :334Q5?! . ,V , a. ,, i. rsnhmunaa 54,.Ease...n,5i> .. Lafixauwmwnwmii am... has)... .3. 0).. 2.11.3 .55.: n 5.33% .,...»3 . . 55...... , 3M: 3.. i 2 .2. . ) 3 V .1 I4! Iii]! V I‘v . fiugflgwgufififitgq 23%;?! 5.3.12. .1 A er 531751.15; 513.1... II. .529: in; .5 79.1%” 4 35.2.1232! . gmvfifiuvfigfigg . tragisrsl ‘59.. s ’3 :730353. :1, J) 51...; .5 s . .Ixcogfllflzl .t.fiJ-4:3flflu§.5: . 4 «Ski; V 295! .. 2"! . . 2a.... b.4972!!! MK. 1.. 3 KW... III)». , is}... 5.1.3. ) .,..V.,.........4f.§uflm pathway...“ .?5 =2... ‘ A: )3: 35.33.32 3... h : . 3.33% .Hfi.%.m_iu.w_.,bwusfi.é Jasfirfi.5?m£;sw§§,§ NV Va: 30.. 526.5. . 7.3%: kg, .5 354.313}. a; .VV. :3 mafia! ,fiasxgzifiwfifi, a, , ,, \r; 2. fig! . éfikfl. H3 : .54. .5 , , . v I . f 3 . . :13}. V 90))? 1.2.17! V )3figfxswvn fix fingwax‘aflgflgumumbz A . . a x. . .93? dang"... aw}? §?fi§§a$.s:§:c:hai M 5.; yam? {mug h.w5»§i£~. .M92§_ 55?...“ fiwflfififlgfglif? 3.22.; . s. . a...” V»? $3.15. a. j a .. 35.32.}...2ficmi . 73%. 1%.... .m:%;§.>fim% hzéairwfimfglfigfi flaws?» .fiflsihiumwwk a... 35%.. 35.5% § 1,», .192 «gigs... ,3: figfil _ Bang? 3%... ;%H. s V x. Vs... f: . £7113? r“. h I {fit}. it! I it“ . a 3543:.»3 fithhuwwmava. igggkx,‘ Jinn... .V s. :5 V s... fi.§:¢i—mmwsgi aangWVbhzfiflszW37g EJQJ. .Vi‘ . 5.55, .rfi. . V ,3... .V. ”S... znffiéaflfinq.§.§z§§lfi . , V . V . .V 4??fimpi.§a 2 3.5.3. , 5&1 fl ‘i’ :”)5‘ 54.33 . aggvsgnnggq 5.... a. 131- .t _ .357 .5 a. : c 3 a VI 3.. a... 23.9. “V V .V $33. ,aaafiw‘iuwsl ”“233me ‘ n; K)»; §m¥!éhfiuw.§§gV . as. .FV “£3... Vishfi. .iH Hz. 5.38.??? an; §2.3.§% 3m”. mfifiggs 59%. gig; .. .. . L II, . , a... l 1:4 A . , $34.5 raw rag war»??? 1 .53. (17.5? isflfu. . £3 $7 #w:.§§§w%§ 3 «ham fifri: sizafi... , (51 3. ,, 2313.? .. 3.; V.. :35: V $54.8 éyhuvafiaia} it a. .. z: 3.1!? (111.1 . . V .. in. r V .7... > xii-n. «f. :3 V ,é(_.5asf§wfiiii\i 121757.54 V . n:- .5%Qif L: . .V 33$ eflwn , ._ §:.§x~.¥ 2. “25531.29... .. . 311.5... VV .V , 53.... 1&3», Es..." s... . , +1 .4 n. t...v. . v -. frvfi’wmcwrnrfiugé. . fimr‘flr 2 HI ,, A h: . . 3.8.5. .3 ,. 2— .. €1.15... .1 an; 5 .uctk 45 V .6 . ”3.... . In. ,2 . 1'1?! $5117.71” u. _¢thu1fiflszt..§ Lum- Qttflflwmnm. 1.. .15.. . , , 2.5. 3 Asa. :1 11L an 4551523,.” .2 Vgraa 3519,; i 3.31.5... 1?:953?! A 23:. .t. a anuai I .1}. $411.: a. ; . .353 7 it... i Z 1.! . 5m"? rfive axfiffimchpvéagq ran. . .1- . . 3.. V A. 32. 1.45... . Eng . V xx)??? in: 35.1 *1 J5 V .r ‘3 I .a :LflM-fihn 7 3593. 1.33. 1 I"! .v r .1 lily-mun"... .5 “w... V )1? l. .wi , .- rV.?r?‘.Vm3nV. 7. g g) . . 11X... ,2 voyage-Inna: . «(Mum-1.3.5.. 4151).“..32): V :31" l J} I. ., "Hunks... I .1351. s25, 33. .. . g. r 4 v. .. ~»(n..}I V5.3. x3250):... (45... q r in?! a... . rwxfdgg; . 3.. V . .t. . .. c . v . . .63; . .fimwawga? “an. ... $93.1 .1. n i .2 ammwriwffi ; . its frank .1 . ”5%....an Luvn’ixnufiaufiwfiafi fig. .98. fivfifid. .. ;.k5~h§.¢&wwxbfi V 1 , a . 15. a: «13.1. , . _ ”(‘34 v t... .\.. 3.5. 1 “v. 7.3.7“. zgftw$~u V.f_rt.rltfiV. m.” $7.411. _ 17;.‘1 .v. firm): . . J: an a in. . e£.rtcrnfitvmmr 41hflwnmnré ._. r :1... $2.7 . . V e35. :3... u. ’3... , .. $4.... an, «331.. . .V . 7&1...’ r: 2 Hi?! Ruins . . ago. .1! V I. 1a.. 5%.»; 3...: r :2 I... V V .1. .. 3.33.: 39 1; J2... .Lahz... : Icahn. .. 35;...» , mega til rt». ..V . .... . rV . 7% $.14. (known. I. x ,3... .... . figiamaz. 2 V 3.1,. . V 3.0 V 3. £11? :33: 1%..» 1 7.6:! . it. as . . he. tug rt. V adv! V . .1 “5251 . 3311, .m¥.§a5§a$ ”1:: ikhfifig .V 1 .. .. v23 V xii, .. ,KIJI...7..,..zi.1rtiera 7:. i: ..V . {51.1,}: f. . V uflirvxrnnuu . . J. . I‘ll . V. 1‘34!» ,.V.V . ”01.. .4 V . 2.1.. ”13.4.? .7. r1flllrf...» . . , . ,V . «an... 7.2:. .77... L . , V J. , Mr! V 53. ”in”? V .itflfi, 1542;. . . Hfinrxuwuw U1? , . . .V V tr. 3.5.??? .. V , .. . (7229;7le 7.1 . v.45... :1-» . tyrant: V r... V _ , . , , .1. V 9. V . .. . . H117". u’bfi’i’dflb : A . :57?!th :.....r.h.w7!..¢n...i: LPI... . . I. n 13.»; ,3}.N..£!.73. . r :~K..c..$76|v..r:»l5 .39“ .421... .. V 11.. «96:3. 1. .xintflanhnuu... . . A i .. 71. . V Vivfl: r. . . . . l- V . . .LLVI. : ‘ .liifinfiflfifi {2.ng 1....Iz....tz.h... ”Agni: FL}: in: if 37...! a... a. 4..) IL V ‘L LL! .r ,n 7.5.3.“? I .1" . . . I; 7. t 32%.}. .qfimmxrz 41. .J, Lnfiwruxmwfiwhpfifiif Prank: r a: V. 393%,} if V txufihuunlmx $¥fl “.9 . I: H831;9%“;qu +4 , ..§.J....LV:JIUJ£IIV;L. I’PLQV ‘ 751.....ltuuffix A . . , brillrivh .I‘ it..1£.f.l.7.: {:57}!!!ch r51! 1;}.77ié1. guru?!“ .i .1 .f‘ ) \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ 1239‘ *HEBil This is to certify that the thesis entitled Adenosine—, AMP-, ADP, and ATP—Selective Electrodes presented by Inna Deng has been accepted towards fulfillment of the requirements for Ph .D 0 degree in Chemi gtry Major professor Date July 31, 1979 0-7639 OVERDUE FINES ARE 25¢ PER DAY _ PER ITEM Return to book drop -to remmre this checkout from your record. ADENOSINE-, AMP-. ADP: and ATP-SELECTIVE ELECTRODES Bu Inna Deng A DISSERTATION Submitted to Michigan State University in partial PulFillment 0? the requirements For the degree 0F DOCTOR OF PHILOSOPHY Department 0F Chemistry 1979 a x \\\\i \w /4V/fl/J»L ABSTRACT ADENoerE~. AMP-l ADP, and ATP-SELECTIVE ELECTRODES Bu Inna Deng Combining immobilized enzymes with potentiometric electrodes is an advantageous technique For the use oF enzyme-catalyzed reactions in analytical determinations. In this research. a system of Four such enzyme electrodes was developed. Each electrode was constructed by coupling a layer which contained one or more immobilized enzymes to the ion—selective membrane of an ammonia-sensing elec- trode. The system consists 0F: 1) an adenosine electrode which is selective For adenosine, and uses the enzyme adenosine deaminase; 2) an adenosine monophosphate (AMP) electrode which is selective For AMP and adenosine, and uses the enzymes adenosine deaminase and alkaline phos- phatase; 3) an adenosine diphosphate (ADP) electrode which is selective For ADP, AMP, and adenosine: and uses the enzymes adenosine deaminase. alkaline phosphatase, and myokinase; and 4) an adenosine triphosphate (ATP) elec— trode which is selective For ATP, ADP. AMP. and adenosine. and uses the enzymes adenosine deaminase; alkaline ph05* phatase, and potato apyrase. The electrodes are quite sensitive; and can detect adenosine and AMP at the 1 uM level; and ADP and ATP at the 10 UN level. They demonstrate that stepwise enzyme reactions can be employed in enzyme electrodes with a relatively small increase in detection limit and little loss 0? sensitivity. The electrodes were tound to perForm well in solutions at pH 9 and 37°C. Four methods of enzyme immobilization were developed For use with enzyme electrodes. The First method used entrapment 0F enzymes between a cellophane membrane and the gas-permeable membrane 0? the ammonia sensor. An inert protein was entrapped with the enzymes to improve their stability, and a diFFusion step was added to remove impurities Prom the immobilized enzymes before uSe. The second method involved the creation 0F an in— soluble enzyme-and-protein polymer gel. A thin slice 0F gel was mounted on an ammonia sensor to create an AMP electrode.. While the detection limit 0F the electrode was high: this method may provide a practical way to produce large numbers 0? enzyme membranes commercially. The third method combined the entrapment procedure 0? the First method with copolymerization by glutaraldehyde. The resulting electrode showed poor sensitivity and re- sponse. In the Fourth method, glutaraldehyde was used both to bind the enzymes to the ammonia sensor membrane, and to copolymerize the enzymes. The resulting electrode had good sensitivity and a Fast response rate. ACKNOWLEDGMENTS My sincere thanks and appreciation go to Prof. Chris G. Enke: my research director. I am also very grateful to Dr. Stanley R. Crouch; who served as second reader. I would like to thank ProF. Willis A. Wood a? the Dept. 0? Biochemistry For several valuable suggestions. I am grateful to my dear Friend Francine Kloc For helping me with many aspects of my studies. My husband has my thanks and appreciation tor his great eFFort in editing and typ- ing this thesis. ii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER I. INTRODUCTION CHAPTER II. ENZYME ELECTRODES A. INTRODUCTION B. PRINCIPLES OF ASSAY BY ENZYME ELECTRODES C. TYPES OF ENZYME ELECTRODES Potentiometric Enzyme Electrodes Amperometric Enzyme Electrodes CHAPTER III. DETERMINATION OF ADENOSINE. AMP. ADP. AND ATP BY ENZYMATIC METHODS A. METHODS OF ENZYMATIC ASSAY Determination 09 Adenosine and Adenosine Monophosphate Based on Adenosine Deaminase Determination 09 5’-AMP (5’~ Adenylic Acid. ADP. and ATP Based on S’-Adenylic Acid Deaminase Enzymatic Assay For AMP. ADP. and ATP by Fluorescence Assay 0F.ATP Based on Luminescence Response of Luciferase Reaction iii viii ix 11 17 18 18 19 OUTLINE OF THE RESEARCH CONDUCTED AND RESULTS OBTAINED Development of the Enzyme Electrode System Further Development oF Enzyme Immobilization Methods CHAPTER IV. METHODS OF IMMOBILIZING ENZYMES THE IMMOBILIZATION OF ENZYMES 1. Absorption 2. Adsorption 3. Cross-Linking 4. Adsorption and Cross—Linking 5. Ion-Exchange Techniques 6. Entrapment 7. Copolymerization 8. Covalent Attachment Characteristics 0? Immobilized Enzymes Stirring. DiFFusion. and Steric EFFects ENZYME IMMOBILIZATION METHODS DEVELOPED IN THIS RESEARCH Method 1: Entrapment 0F Enzymes and Inert Protein Between a Cellophane Membrane and a Gas- Permeable Membrane Method 2: Copolymerization oF Enzymes and Albumin with Glutaraldehyde Method 3: Entrapment oF Enzymes that are Copolymerized with Glutaraldehyde iv 26 34 34 35 35 37 37 38 39 39 41 43 45 45 49 54 Method 4: Copolymerization 0F Cross-Linked Enzymes to a GasrPermeable Membrane by Glutaraldehyde CHAPTER V. ADENOSINE ELECTRODE A. INTRODUCTION B. EXPERIMENTAL Apparatus. Reagents. Procedure. C. RESULTS AND DISCUSION Calibration Curve. Effect of pH On Electrode Sensitivity. EFFect 0? Temperature on Electrode Sensitivity. Response Time. Selectivity Study. Sensitivity and Stability. CHAPTER VI. AMP ELECTRODE A. INTRODUCTION B. EXPERIMENTAL 1 Apparatus. Reagents. Procedure. C. RESULTS AND DISCUSION Calibration Curves. 55 6O 60 61 61 62 62 64 64 66 76 76 77 77 78 79 BO 80 Effect of pH on Electrode Sensitivity. Response Time. Selectivity Study. CHAPTER VII. ADP ELECTRODE A. INTRODUCTION B. EXPERIMENTAL Apparatus. Reagents. Procedure. C. RESULTS AND DISCUSION Calibration Curve. Effect of pH on Electrode Sensitivity. Selectivity Study. CHAPTER VIII. ATP ELECTRODE A. INTRODUCTION B. EXPERIMENTAL Apparatus. Reagents. Procedure. C. RESULTS AND DISCUSION Calibration Curves. Effect of pH on Electrode Sensitivity. Selectivity Study. vi 83 83 S6 88 BB 90 9O 91 92 93 93 93 95 98 9B 100 100 100 101 102 102 104 106 CHAPTER IX. CONCLUSIONS AND FUTURE PROSPECTS 108 LIST OF REFERENCES 112 vii Table Table Table Table Table Table ~Table Table Table pa LIST OF TABLES Potentiometric Enzyme Probes. Amperometric Enzyme Probes. Comparison for Emmobilized Enzyme Electrodes. Comparison of Techniques for Immobilizing Enzymes. Selectivity Studies for Adenosine Electrode. Long-term Behavior of Adenosine Electrode. Selectivity Studies for AMP Electrode II. Selectivity Studies for ADP Electrode. Selectivity Studies for AMP Electrode II. viii 16 4O 74 75 87 97 107 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure TU LIST OF FIGURES Attachment of Glutaraldehyde/to Enzyme. Enzyme Electrodes and Enzyme Polymer Gels. pH Profile of AMP Electrode Prepared by Method 3. Schematic Diagram of Adenosine Electrode. Calibration Curve for Adenosine Electrode. pH Profile of Adenosine Deaminase. Adenosine Electrode Response as Function of pH. Calibration Curves for Adenosine Electrode as Function of pH and Temperature. Effect of Temperature on Response of Adenosine Electrode. Calibration Curve for AMP Electrode I. Calibration Curve for AMP Electrode II. Response of AMP Electrode I as Function of pH. Response of AMP Electrode II as Function of pH. Calibration Curve for ADP Electrode. ix 51 56 57 63 65 67 68 7O 71 81 84 85 94 / Figure 15. Figure 16. Figure 17. Response of ADP Electrode as Function of pH. Calibration Curve for ATP Electrodes I and II. Response of ATP Electrode II as Function of pH. 96 103 105 CHAPTER I INTRODUCTION Enzymes are now playing more and more important roles in modern analytical chemistry and industrial processes. Many reactions catalyzed by enzymes have been used for the analytical determination of substrates. activators. in- hibitors. or the enzyme itself. Techniques which use enzyme-catalyzed reactions are gradually replacing those of classical wet chemistry. and are particularly evident in the clinical laboratory. One goal of both the chemical and biological scien- tist is to develop faster. simpler. more accurate. and less expensive clinical and chemical tests. This goal can be approached by using potentiometric membrane electrodes with immobilized enzymes. The immobilized enzymes can be reused hundreds or thousands of times. and. if stored properly. can remain functional for several weeks or months. The electrodes. which are often ion-selective electrodes. provide a simple and fast measurement tech- nique. By reusing the immobilzed enzyme. of course. the cost is greatly reduced. Further. the sample solutions remain essentially uncontaminated. and can be used for other tests. Thus. in many aspects. immobilized enzyme PJ electrodes offer an advantageous approach for many analyses. In this thesis. the development of four enzyme electrodes for adenosine. adenosine mdnophosphate (AMP). adenosine diphosphate (ADP). and adenosine triphosphate (ATP) is described. An ammonia electrode and a gas- permeable membrane are used to measure the ammonia evolved from the enzyme reactions. One or more enzymes are im- mobilized on the electrode between a gas-permeable mem- brane and a dialysis membrane. Within a limited range. the logarithm of the concentration of the substrate varies linearly with the electrode potential. This allows an accurate determination of the substrate concentration. The electrodes are extremely sensitive. and can detect adenosine and AMP at the 1 uM level. and ADP and ATP at the 10 UN level. This thesis is divided into nine chapters. Each chapter contains an independent sub-subJect or sub-proJect of the entire electrode prOJect. Chapter II is a broad review and discussion of enzyme electrodes. Various methods for determining adenosine. AMP. ADP. and ATP. including enzymatic methods. are de— scribed in Chapter III. An outline of the research lead— ing to development of the four enzyme electrodes for these four substances is also included in Chapter III. The various methods of immobilizing enzymes and the techniques developed for the enzyme electrodes used in this research are described in Chapter IV. The four enzyme electrodes (adenosine. AMP. ADP. and ATP) developed in this research are characterized in Chapters V. VI. VII. and VIII. The conclusions and future propects of this enzyme electrode prOJect are presented in Chapter IX. CHAPTER II ENZYME ELECTRODES A. INTRODUCTION During the past few years. membrane bioprobe elec— trodes have rapidly developed and become valuable tools for analytical and biomedical problems. Their papularity is due to their convenience as well as their accuracy in bioassays. One of the current directions in bioelectrochemistry has its basis in ion—selective membrane electrodes which use glass. liquid. and crystal membrane materials. For special analytical sensing purposes these basic membranes can be coated with immobilized enzymes or proteins. This can result in sensitivity to particular chemical or bio- logical constituents in chemical solutions or blood and body fluid samples. Such sensors can then be used as probes for analytical assays. and even for fundamental studies of processes in a single cell. An enzyme electrode is devised by coupling an enzyme layer with an ion-selective or gas-selective electrode. The enzyme layer is immobilized between the primary mem- brane sensor and the sample solution. Enzyme electrodes have been successfully applied to the assays of urea. glucose. amino acids. cholesterol. phosphate. and other species. Some electrodes are already being evaluated in clinical laboratories for use in measuring important body fluid constituents to aid medical diagnosis. B. PRINCIPLES OF ASSAY BY ENZYME ELECTRODES In brief. an enzyme electrode is constructed by im— mobilizing an enzyme layer between the electrode sensor and the sample system. Therefore. when the electrode is placed in contact with a sample solution. substrates and other chemicals in the solution diffuse into the enzyme layer. where the substrates react with the enzyme. The selective electrode used has to be sensitive to one of the substrates or products. The electrode then measures the depletion of the substrate or the formation of the product which results from the enzyme reaction. A linear rela- tionship between the logarithm of the substrate or product concentration and the electrode voltage or current re— sponse is established over a limited concentration range. For the potentiometric measurement. the linear relation— ship obeys the well-known Nernst equation [1]: E = constant + (RT/nF) 1n aM where E is the measured EMF; R the gas constant; T the absolute temperature; and F. n. and aM are the Faraday. ionic charge. and activity of the chemical to be measured. For a mixture of ions. the electrode potential of an en~ zyme electrode can be expressed by the following equation if the product of the enzyme catalyzed reaction is a mono- valent cation: E = EO + (RT/F) lnIap + ktai)1/Zil where E. E0. R. T. and F have the usual meaning. k is the selectivity ratio. ai and 21 are the activity and charge of the ith interfering ion. and ap is the activity of the monovalent cation produced in the reaction catalyzed by the enzyme. The oxygen electrode is the only amperometric elec- trode used as an enzyme electrode. Oxygen electrodes can be made in the laboratory. or obtained commercially. The oxygen electrode is made up of a platinum cathode and a silver anode. An enzyme electrode can be constructed by coupling an oxygen electrode with a layer of an enzyme which reacts to either consume or produce oxygen. Hhen oxygen is in non-rate limiting excess and the substrate concentration is well below the apparent Km for the im- mobilized enzyme. a linear relationship between substrate concentration and change in oxygen pressure is estab— lished E2]. The substrate concentration can thus be de- termined. .Though few enzyme electrodes employing oxygen electrodes are reported. some thirty enzymes which use oxygen and produce hydrogen peroxide can be used in constructing enzyme electrodes [3]. C. TYPES OF ENZYME ELECTRODES Enzyme electrodes are commonly categorized according to whether the type of detector employed is potentiometric or amperometric. The reactants or products are qUantified by potentiometry and amperometry. The behavior of the amperometric devices substantially differs from the po- tentiometric.due to the electrochemical consumption of the product at the sensor surface of the amperometric devices. Potentiometric Enzyme Electrodes There are two kinds of potentiometric enzyme elec- trodes: one uses an ion-selective electrodes the other uses a gas-sensing electrode. Ammonia-gas- and COz-gas- sensing electrodes have been developed and are frequently used for enzyme reactions which have ammonia or carbon dioxide as products. Since the gas-permeable membrane ex- cludes the passage of interferring monovalent cations. the electrode is thus able to provide excellent selectivity. The first potentiometric enzyme electrode. developed by Guilbault [43. was a urea electrode. It was prepared by entrapping urease in a polyacrylamide matrix which was then placed over a cation-selective electrode. Urease hydrolyzes urea to ammonia and carbon dioxide. The rate at which ammonia is produced from the urea diffusing into the membrane is proportional to the quantity of urea pre- sent for urea concentrations between 1.0 and 30.0 mg per 100 ml solution. However. the electrode shows consider- able response to sodium and potassium ions. which are present in large quantities inblood. This necessitated an ion-exchange pretreatment of biological fluid before measurement with the electrode. There is a linear relationship between the product concentration at the sensor surface and the bulk substrate concentration [5]. However. a sensor such as an ion- selective electrode will respond to any material which is converted to the product. and also to the bulk concentra— tion of the product. especially when it is present in considerable quantity [5]. A paper by Larry Bowers reviews and tabulates all the potentiometric enzyme probes reported up to 1976 [6]. Since then. two new potentiometric enzyme probes have been developed ~— a lactate sensor and a 5’-AMP sensor. These have been added to the end of Bowers’s table (Table 1). The advantages of the potentiometric membrane probe are that: 1) it is easy to construct and perform simple processes; 2) the result is reliable; 3) the cost is ON mu emu NH on Nu m« v” emu NH An .kmm newumnmam Loom «emcawmoa cwuwdwsoeeg mam Anczv memucmou can memucm :vom now>a¢um~om acmmumoxw «mcmcneme wcmcao~fiou 3cm ucmsaams coAmzvvav men cmmzamn umnfiaanoeefi wemncw mvfi>fiouw~wm acmuuwuxm «mcmcneoe coamzseuo own 00 muaawean noxcflu wemncw umcoamws 30am «mutated; anemone: o>um enema Cu m~m>wn moAxOAU consmo . oucmsossmacu +2 wmcocmmc 30am «oucmcwhsmvCA +2 . mz + heap mucoamms mac” .wucmsuecmacn hoaxOAo consmu mmcoamwc cum on «accessovcwa om>05c5a .mvuuuauuooocams Loom «z.fuon on maacmmcau mvmeLma eunucoms saw: 90mm; :9 ceaoznom Cu acwmmsm on vane wcwuoH muesvumnm mamncw mos: gag: vmaoauo» ommcwmsm no nos: yo coaomELOE VON: mfimfli UEINCO «A: cg cos age onv mucommmc 30am new) mcmmmmomc acmeammsamcm wmcmcoxmtcoH NW.N soundness .wucmcmssmacA +2 .+wz madamuc\macoseoo cooemu mange w>mvuwmwm macosec m>avoo~wm mgcosec o>aeuwumm mofixogo concmo w>uauwmmm saucoeeo swans; accounamlcwaumcoz w>aaom~wm ceaomo o>wauonmm mthOAu concmo moccaumau In m>uauonam_cvoao m>uouw~mm tau weapoH o>aauo~oo acaaoo u>uvuwnwm ca“ 20 m>uauw~wm ceavmu o>aauoawm cauvmo smoavwcocb oAHQU @8U30m hmfifimv wmmcmeLUmcoo menace; montaemmo canmcmuc momma: momma: momma: ommmc: mwmumxonemumu magmasmolw mmmCAHanwcmm mmmcuemanao oncomxo wmouano maneummcmamc wmmmc: mmmofimouauml wmmomxo owum cement: wmmnmxo taco causal; wamucw manage memncw unsawEOAocovom oemaumw mzc mos: was: was: mus: wcfimOLmAIw caaaaoacme mcaemvzfio omouauo mcflmmsmawc UONCHWL¢ ceamommec mkum ocwemla unwom ocfiemlu wvmfimcc .u wmnmh 10 low; and 4) the sample is appreciably uncontaminated. and can be reused. The limitations are that: 1) the chosen enzymatic reaction must generate or consume a species for which an ion-selective electrode exists; 2) some electrodes have limited selectivity and are therefore susceptible to ionic interferences [633 3) the gas diffusion electrodes. such as ammonia sensors and carbon dioxide sensors. respond slowly. and the membrane will clog in biological matrices; and 4) the diffusion of substrates and products into and out of the enzyme layer is slow. An excellent example for limitation 2) is the urEa sensor. which employs a cation- selective electrode to detect ammonium ions. The cation- selective electrode is also sensitive to the sodium and potassium ions which exist in considerable quantity in body fluid [7]. An improvement was made by Anfalt et a1. by using a gaseous diffusion membrane covalently bound with urease [8]. Thus. only free ammonia is determined. and sodium and potasium ions do not interfere. The latest electrode design improvement is the air gap electrode [9] -- it provides the advantage that the membrane will not clog in biological matrices. is H 11 Amperometric Enzyme Electodes An amperometric enzyme electrode is constructed by coupling an enzyme layer with an amperometric electrode. The theory is based on the relationship between the cur- rent and the rate of generation of electroactive species in the membrane phase. Historically. the first enzyme electrode developed was an amperometric enzyme electrode. It was a glucose sensor. which employed an oxygen elec- trode [2]. At glucose concentrations below Km for the entrapped enzyme. Updike and Hicks found a linear rela— tionship between glucose concentration and the measured oxygen depletion rate. Elegant mathematical treatments of mass—transport" coupled enzyme reactions in the membranes of amperometric electrodes have been developed. Mall and Maloy have derived equations which define the relationship of the current response and the substrate concentration [10]. The equation shows that the current may be limited by either the rate of the enzyme reaction or by the rate of diffusion. when the catalytic rate is limiting and the bulk concentration is low. i.e.. [S] {< Km. then nFAVX = ZKm ° {slbulk where i is the steady state current. V and 7 represent the enzyme concentration per unit volume of gel (mole/cc) and 12 the thickness of the enzyme membrane (cm). and n. F. and A have their usual electrochemical significance. Notice that there is no dependence on any diffusion coefficients. Hhen NZXZ becomes larger than unity [10]. the cal- culations indicate that the current is limited by dif- fusion only. i.e.. it is independent of V. the enzyme concentration. Thus. at high enzyme concentration and low substrate concentration. the steady-state current is limited by diffusion and is given by: i = nFADSCSJbulk / X where 08 and X represent the diffusion constant of'S and the thickness of the enzyme membrane (cm). As can be seen. the current response is inversely proportional to the membrane thickness. At sufficiently high substrate concentration. regardless of the amount of enzyme. the current becomes independent of the substrate concentra— tion. The current is also independent of the substrate concentration when substrate concentration is sufficiently low. The recent review paper by Larry Bowers includes a tabulation of most of the amperometric enzyme probes re~ ported up to 1976 [6]. They are listed in Table 2. Though only a few amperometric enzyme probes have been developed. many others may be developed in the future 13 ~D\me c I o mucus smac«u «oomvszw . Atwom om muesaumuw am cauvmncwo cmmmzo magammoz unfit an mmmuflsz Owen. momma: magmasmvlu .mcamammUIq mmmoaxo .wCMuzmmIJ .mN Uuum czaemlu .UCacmHmHmcmzutu ~U\me cu I o ,wN mg once; smwcau «mucmsmsswucfi uocmzvmz mmmUaxo accou~c Hocmcaw ammonxo oucmsmesmocu amounnm .ommamzm hm mmooaum .mowummxo an coneanmch moccaomam cant um :mozc acfinmxac mamcmmocm momxome cmmOcumz mo wannabe aficomEOsmmem a ma .mN aumsuu .uow Nu v mega mucoamwm muesaumam an ammoaxo amoua~u amououu woesvuwnw mcmsneme cmmmxo . N cmmmxo xcmno em cmmaxo yo unavouumn new: mousvuwnm am wmmvfixo amouznm mmouono .som manzmws\macmseoo pmuatmcmsb oemucm mamamcc .onou usozom soapy monasm memncm ausamaoLoaec .N wanmh 14 by coupling oxygen consuming or producing enzymes and oxy- gen probes. Clark identified some 30 enzymes which use oxygen and produce hydrogen peroxide [3]. If these en- zymes were used with polarographic platinum electrodes. Clark estimated that the following compounds could be detected: Acetaldehyde. D-mannose D-alanine. methanol Aliphatic nitro compounds. L-methionine D-aspartate. 6-methyl-D~glucose Benzaldehyde. N-methyl-L-amino acids Diamines. NAD ‘ 2-Dioxy-D-glucose. NADH Ethanol. oxalate Formaldehyde. L-phenylalanine L-galactonolactone. D-proline D-galactose. purine P-D-glucose. pyridoxamine phosphate D-glutamate. pyruvate Glycollate. saccosine L-gulono-X-lactone. spermine Hypoxanthine. sulfite D'lactate. tyramine ' L-lactate. urate Lactose. D-valine (+lMandalate. xanthine Though not all the above compounds are of interest to the clinical and analytical chemist. the list indicates the wide range of compounds potentially detectable with par- ticular types of enzyme electrodes. [The advantages of amperometric membrane probes are their simplicity. reliability. and low cost. Further. they avoid. to some extent. the limited dynamic range and slow response of potentiometric devices [10]. One dis- advantage is that the amperometric sensor may be quite sensitive to other electroactive species. In serum. a 15 number of reducing agents. such as uric acid. glutathione. and cysteine. could interfere with the quantification of peroxide. A comparison of potentiometric and amperometric enzyme probes is presented in Table 3. as originally re— ported by Larry Bowers [63. It is likely that the combination of the selectivity of enzymes and the operational simplicity of using an electrode sensor will bring about more widespread use of enzyme sensors as tools of analytical chemistry. 16 Table 3. Comparison of Immobilized Enzyme Electrodes (from Bowers [6]). Type Advantages Limitations Potentio- Simplicity of op- Slow response. memory metric probe Amperometric probe eration. easy to make. uses small amounts of enzyme Simplicity of op- eration. easy to make. wider linear range. more en- zymes produce or comsume oxygen or hydrogen peroxide. uses little enzyme effects. relatively few enzymes compat— ible with operation (ammonia. carbon di— oxide). sensitive to inhibitors and acti— vators. restricted linear range. incom- plete conversion Slow response. memory effects. elecroactive interferences. sen- sitivity to materials which consume hydro— gen peroxide. incom- plete conversion ~...rv ; 24s; l-rc-c-r .- _. CHAPTER III DETERMINATION OF ADENOSINE. AMP. ADP. AND ATP BY ENZYMATIC METHODS Adenosine and its derivatives all have key roles in biological functions. ATP is the keystone of all cellular activity. the ATP-ADP system functions as the energy car- rier in all living cells. AMP is the building block of DNA and RNA. In addition. adenosine is the precursor of three important coenzymes: flavin adenine dinucleotide (FAD). nicotinamide adenine dinucleotide (NAD). and nico- tinamide adenine dinucleotide phosphate (NADP). Thus. improved methods for determining adenosine. AMP. ADP. and ATP have great significance. Section A of this chapter describes the enzymatic methods of assay which have been reported. including the chemiluminescence method of firefly luciferin. Section B presents an outline of the research performed in this project. highlighting the methods which were developed. 17 18 A. METHODS OF ENZYMATIC ASSAY Determination of Adenosine and Adenosine Monophosphate Based on Adenosine Deaminase The adenosine deaminase method was first described by Kaplan in 1947 [313 and then modified by Kornberg and Prieer £32]. The deaminase from the intestine is specific for adenosine and will not deaminate other adenine de- rivatives. Kalckar has separated the deaminase from po- tent phosphatase present in the intestine. Kornberg and Prieer have described a method in which separation of phosphatase is not essential for the assay of adenosine. the phosphatase being inhibited by addition of phosphate. Basis of Assay for Adenosine. Deamination of adenosine is followed by measurement of the decrease in optical density at 265 nm. This is because deamination of adenosine is accompanied by a shift in the ultraviolet absorption spectrum. At 265 nm the molar absorptivity of inosine is only 40% that of adeno- sine. The reaction follows the equation: adenosine (adenine riboside) + H20'——-—9 inosine (hypoxanthine riboside) + NH3. Kornberg and Prieer'w used a phosphate buffer (0.05 M) to inhibit the dephosphorylation carried on by l9 phosphatase in adenosine deaminase preparations [323. This was not necessary for Kaplan’s method. since a phosphatase-free deaminase was prepared. Basis for Assay of Adenosine Monophosphatea The phosphate group of adenosine monophosphate (either 2’- or 3’- or 5’-monophosphate) is split off by phosphatase and the adenosine liberated is then sus- ceptible to deamination by adenosine deaminase. The ab- sorbance changes which take place at 265 nm thus vary with the amount of adenosine monophosphate (either 2’- or 3’- or 5’-monophosphate). The reactions follow the equations: AMP —Phosphatase > adenosine + Pi adenosine . deaminase . . adenoszne —> 1noszne + NH3. Determination of 5’-AHP (5’-Adenylic Acid). ADP. and ATP Based on 5’-Adenylic Acid Deaminase. Kaplan developed this method. which uses 5’-adenylic acid deaminase. in 1947 [31]. The enzyme. also named Schmidt’s deaminase. is purified muscle deaminase [33]. The purified enzyme is absolutely specific for 5’-AMP. having no action on adenine. adenosine. 2’-AMP. 3’—AMP. ADP. ATP. NADHI OT‘ NADPH. Basis for Assays of 5’-AMP. ADP. and ATP. The method is based on the stepwise conversion of ATP and ADP to 5’-AMP. which is subsequently converted to 5’-IMP by 5’-adenylic acid deaminase. This reaction can be followed spectrophotometrically at 265 nm. At this wavelength. the molar absorptivity of 5’-IMP is only 40% that of 5’-AMP. The reactions which take place are: AMP -——-) IMP (1), ADP -——-) 1/2 AMP + l/2 ATP (2). ATP -—-—9’AMP + 2Pi (3). Reaction (1) is the deamination of 5’-AMP to 5’-IMP. cata- lyzed by 5’-AMP deaminase C31. 33]. In Reaction (2). the enzyme adenylate kinase. also called myokinase E84. 353. causes a conversion of ADP into ATP and 5’-AMP. Re- action (3) is catalyzed by potato apyrase E36. 37]. The assay starts with the addition of deaminase to determine the possible amount of 5’-AMP present. Myo- kinase is then added. whereby 50% of the ADP is converted into 5’-AMP. which in turn is deaminated to form 5’-IMP. The subsequent addition of potato apyrase brings about dephosphorylation of ATP to form 5’-AMP. which is convert- ed to 5’-IMP. The presence'of the 5’-AMP deaminase makes Reactions (2) and (3) go to completion. The processes can 21 be monitored by observing the stepwise decrease in ab- sorbance at 265 nm. The three successive absorbance changes measured at 265 nm during the whole procedure indicate the concentra- tion of 5’-AMP. 1/2 ADP. and finally 1/2 ADP + ATP. Ac- cording to Kalckar [31]. conversion of 1 mg of 5’-AMP per milliliter to 5’-IMP causes a decrease in extinction of 0.019. From this. the amount of the three separate ade- nine nucleotides in the mixture may be calculated. Enzymatic Assay for AMP. ADP. and ATP by Fluorescence This method was first described by Bucher [38]. Its principle is that reduced pyridine nucleotide is readily detected by its fluorescence. excited at 340 nm. Coupled enzymatic methods. based on the method of Bucher [38]. were developed [393 to measure changes in the concentra- tions of AMP. ADP. and ATP. Basis for Assay of ATP. This method involves quantitative phosphorylation of glucose by ATP with hexokinase. and simultaneous measure— ment of NADPH reduction in the presence of glucose 6~ phosphate dehydrogenase. The reactions which take place are: ATP + glucose ——-¥> ADP + glucose 6-phosphate (1). glucose 6-phosphate + NADP+ -—-—9 ‘ 6-phosphogluconate + NADPH + H+ (2.. Reaction (1) is the phosphorylation of glucose. catalyzed by hexokinase in the presence of magnesium ions as the cofactor. In Reaction (2). the enzyme glucose 6-phosphate dehydrogenase causes a reduction of NADP+ to NADPH. The increase in fluorescence associated with NADP+ reduction is proportional to the concentration of ATP. Basis for Assay of ADP. This method involves quantitative dephosphorylation of phosphoenolpyruvate by ADP with pyruvate kinase and simultaneous measurement of NADH oxidation in the presence of lactate dehydrogenase. The reactions which take place are: phosphoenolpyruvate + ADP ;:::3 pyruvate + ATP (1). pyruvate + NADH + H+ ;:::: lactate + NAD+ (2.. Reaction (1) is the dephosphorylation of phosphoenol— pyruvate. catalyzed by pyruvate kinase in the presence of magnesium ions. In Reaction (2) the enzyme glucose 6- phosphate dehydrogenase causes the oxidation of NADH to NAD+. The decrease in fluorescence associated with NADH oxidation is proportional to the concentration of ADP. Basis for Assay of AMP. This method involves quantitative conversion of AMP into ADP with myokinase in the presence of ATP. The same sample used for ADP analysis can be used for determining AMP concentration by adding ATP and myokinase. The re- action is: AMP + ATP ;:::2 2ADP. The reaction is the phosphorylation of AMP catalyzed by myokinase in the presence of magnesium ions. The ADP formed is then subJected to pyruvate kinase and lactate dehydrogenase (Reactions (1) and (2) in determining ADP). The extent of the decrease in fluorescence associated with ADP formation by the myokinase reaction is equal to twice the concentration of AMP in the sample. Assay of ATP Based on Luminescence Response of Luciferase Reaction This method was first developed thirty years ago by McElroy [40]. who found that luminescence in fireflies requires ATP. in detail by Hasting £41]. Luminescence biochemistry has been reviewed TU 4: Basis for the Assay. This method of ATP analysis is based on the linear luminescence response of firefly extracts to added ATP when all other factors are present in excess [40. 42. 43]. Light is usually measured with a photomultiplier. In vitro light production by firefly lantern extract has been shown to depend upon the presence of luciferin. the enzyme luciferase. oxygen. magnesium ions. and ATP [40. 423. The reaction steps which take place are: ++ M LH2+ATP+E<_-—g—_:E—LH2—AMP+PP1 (1). E - LH2 - AMP + 02—95 - L - 0* + Product (2.. E—L—o* .;:e-L-o-+hr (:3). Reaction (1) shows the initial reaction. which leads to light emission. Formation of an enzyme-lyciferin- adenosine monophosphate complex (luciferyl-adenylate) and inorganic pyrophosphate (PPi) is catalyzed by luciferase in the presence of magnesuim ions. In Reaction (2). the complex is rapidly oxidizied to oxyluciferyl-adenylate. in an excited state. followed by release of a quantum of light in Reaction (3). when firefly lantern extract is mixed with ATP. there is an initial burst of light which rapidly declines to a low but uniform level of lumines- cence. The amount of light produced is directly pro~ portional to the amount of ATP. 25 Side Reactions and Interferences. Firefly luminescence is a sensitive method for ATP analysis. However. in addition to the disadvantage of being tedious. side reactions and interferences also cause inaccurate measurements. The presence of transphosphorylase enzymes in the crude extract commercially available results in light emission in the presence of molecules other than ATP which contain high-energy phosphate bonds. Both cytidine-5— triphosphate (CTP) and inosine-S-triphosphate (ITP) stimu- late light production equivalent to the ATP response. ADP stimulates less than 1% of the light production of ATP. The ionic composition of the reaction medium also affects light emission. as shown by Aledort et al. [44]. who demonstrated that emission rates were reduced linearly with increasing concentrations of several cations. Equi- molar concentrations of those cations investigated in- hibited light emission in the order of Ca++ . K+ } Na+ . Rb+ > Li+ } Choline. Furthermore. calcium in a concentration of 97 mM com- pletely suppressed light emission in the ATP range of 1.2 - 2.2 X 10_7 M. Heavy metals also reduce light in— tensity and yield false ATP results by inhibiting the enzyme activity. 26 B. OUTLINE OF THE RESEARCH CONDUCTED AND RESULTS OBTAINED Development of the Enzyme Electrode System All the present enzymatic methods require a homo- geneous and colorless sample solution. Moreover. they are all very tedious to perform. and the results are affected by many factors. The enzyme electrode method not only provides simple and fast performance. but may also allow determination in nonhomogeneous and contaminated solutions which cannot be analyzed optically. Thus. this method has several potential advantages over the assays described. The original goal of this prOJect was to produce an ATP electrode. It was planned to use an extract of en- zymes from'mitochondria. These enzymes would catalyze the respiratory reaction of ATP: ATP + NAD + OH“ ——> ADP + Pi + NADH + 1/2 02. The oxygen liberated would be measured by an oxygen electrode. To test the usefulness of the mitochondria extract for determining ATP. we combined the extract with various amounts of ATP in a closed reaction apparatus. The oxygen produced was measured by an oxygen electrode. However. the levels of oxygen measured were Judged to be too low for this method to be of practical use for performing assays. It was therefore decided to try another enzyme system for the ATP electrode. The next step was to construct electrodes for 5’-ATP. 5’-ADP. and 5’-AMP. based on the enzyme 5’PAMP deaminase. The enzymes and reactions used are those employed in the 5’~AMP deaminase assay discussed in Section A of this chapter. Such an electrode for 5’-AMP had been reported by Papastathopoulos and Rechnitz [23]. Their electrode used a solution of 5’-AMP deaminase trapped between a cellophane membrane and the gas-permeable membrane of an ammonia electrode. When the AMP. ADP. and ATP electrodes were made. it was decided to add a dialysis step to the procedure for removing impurities from the enzyme solution. This is not usually done in the procedures reported in the literature. but is important for the reasons discussed below (Chap— ter IV. Section B. Method 1). The three enzyme electrodes produced showed good linearity. but had poor sensitivity L. and had detection limits above 10' M. The response times were also quite long. To overcome these problems. a search was made for a more suitable enzyme system. There are several desirable characteristics for an enzyme to be used with an elec~ trode. They include: 1) selectivity to the desired substrate. 2) availability at low cost in pure form at .. WI —‘Tfi ”we 4 4 A flflv‘ high concentrations. and 8) high activity in the pH range at which the electrode is most sensitive. Based on these criteria. the enzyme adenosine de- aminase appeared promising for use in an adenosine elec- trode. (For a discussion of how the activity of adenosine deaminase in basic solutions aids the response of the adenosine electrode. see Chapter V.) This enzyme was used in developing four enzyme electrodes. which can determine ATP. ADP. AMP. and adenosine through a simple process. and which give excellent sensitivity and selectivity. The im- mobilization method used was similar to that used for con- structing the three electrodes based on 5’~AMP deaminase. It differs from the other entrapment methods reported in that it uses the inert protein albumin to stabilize the enzyme. A detailed description of the method and reasons for adding albumin are given below (Chapter IV. Section B. Method 1). Basis for the Assays. This method is based on construction of a system consisting of adenosine-. AMP-. ADP-. and ATP-selective electrodes. In each case. an enzyme layer is coupled with an ammonia—gas-sensing membrane electrode. The enzymatic method is based on the stepwise conversion of ATP. ADP. and AMP to adenosine. The adenosine is subsequently de- aminated by adenosine deaminase. a reaction which can be 29 followed by an ammonia gas sensor. The reactions which take place are: adenosine + H20 inosine + NH3 7 (1). AMP adenosine (2). ADP. 1/2 AMP + 1/2 ATP . (3). ATP AMP + 2Pi (4). Reactions (1). (2). (3). and (4) are catalyzed by adenosine deaminase. alkaline phosphatase. myokinase. and potato apyrase. respectivegy. The details are described in Section A of this chapter. The enzymes used in the construction of each sensor are given in the following table: glgctrode for: enzymes gig; adenosine adenosine deaminase AMP phosphatase and adenosine deaminase ADP myokinase. phosphatase. and adenosine deaminase ATP ‘ apyrase. phosphatase. and adenosine deaminase The use of more than one enzyme for an electrode is un- common. Only one electrode using two enzymes has been re— ported C273. and no electrodes using three or more enzymes appear in the literature. 30 It was discovered that these electrodes show enhanced responses at 370C. while some work has been done on the effect of temperature on enzyme electrode response [11]. the enzyme electrodes reported in the literature were all used at lower temperatures. All the electrodes were highly sensitive and de- veloped linear responses to their substrates over a wide range of concentrations. Therefore. the concentrations of adenosine. AMP. ADP. and ATP can be determined. Further details are given in Chapters V. VI. VII. and VIII. Further Development of Enzyme Immobilization Methods While the enzyme immobilization method used (called "Method 1" in Chapter IV) produces suitable electrodes. it has the disadvantage that membranes must be manufactured one at a time. Tran-Minh and Brown [45] had created a large membrane by copolymerizing enzymes and albumin with glutaraldehyde. Pieces of the membrane could be cut off and used for electrodes. while the remaining membrane was refrigerated and stored for later use. An attempted was made to use their method to form a membrane for the AMP electrode. However. the membrane produced proved to be too fragile to be used. It was therefore decided to attempt a new modification of the copolymerization method. By experimentation. the 31 quantities of glutaraldehyde and albumin necessary to produce a polymer gel were established. Individual mem- branes could then be sliced from the gel. Equipment was not available for slicing the gel as thinly as desired. and the sensitivity of an AMP electrode made with a thick membrane was low. However. this method appears promising for commercial applications. It is described in detail below (Chapter IV. Section B. Method 2). In an attempt to improve the long-term stability of the enzyme electrodes. another new method was developed. Basically. the method is to create an enzyme electrode by entrapment. and then copolymerize the enzyme solution by soaking the electrode in a glutaraldehyde solution. An AMP electrode was created by this technique. but its sensitivity and response were poor (see Chapter IV. Sec— tion B. Method 3). Further work in improving the response and stability of the electrodes led to the development of one more new method for immobilizing enzymes. The enzymes are cross- linked to the gas-sensing electrode and copolymerized by the polymerizing agent glutaraldehyde. An AMP electrode was constructed by this method. It showed very good sensitivity and a fast response rate. The method is discussed in Chapter IV. Section B. Method 4. and the characteristics of the AMP electrode constructed by this method are reported in Chapter VI. lgrmgm.r‘-r.-'—. . 44 4 CHAPTER IV METHODS OF IMMOBILIZING ENZYMES The development of the immobilized enzyme is pre- sented in this chapter. In Section A. the methods of immobilizing enzymes are reviewed and the characteristics of immobilized enzymes are discussed. The experiments and results leading to the methods of immobilizing enzymes de- veloped in this research are presented in Section B. In analytical chemistry. enzymatic assays are at~ tracting more and more interest because enzymes are highly specific toward their substrates. Nevertheless. soluble enzyme reagents have many undesirable properties which limit their usefulness in enzymatic assays. In aqueous solution. enzymatic activity is usually lost fairly rapid- ly. Sometimes air oxydizes the soluble enzyme or destroys its tertiary (”folded") structure at the air-water inter— face. Furthermore. the enzyme can be readily destroyed by heat or microbes [46]. Finally. the enzyme reagent. which is expensive. must be discarded after a single use. The interest in immobilized enzymes occurs because immobilized enzymes are not consumed. Another great advantage is that they lose little activity with use or time. A further interesting advantage is that. in some cases. the chemical properties of the enzymes can be 24.. l gr.- 1 .Av‘jsqxwe-a-e .- changed by immobilization. Through many immobilizing methods. enzyme molecules can be chemically bound to each other. or to a support. The resulting enzyme derivative is thus a different chemical species. Its catalytic pro— perties. such as optimum pH. resistance to heat denature- tion. and substrate specificity} may be quite different from the free enzyme [46. 473. An immobilized enzyme is usually stable for several weeks. or even months. For these reasons. the soluble enzymes now used in industrial or clinical laborartory procedures could someday be re- placed by more economical immobilized enzyme systems. 34 A. THE IMMOBILIZATION OF ENZYMES A Review of the Methods of Immobilizing Enzymes Enzyme immobilization can be traced back at least sixty years to the work of Nelson and Griffin [48]. Nevertheless. interest in them has not been widespread until the last ten years. Attention first arose in the 1950’s when Grubher and Schlecth immobilized several en- zymes on polyaminopolystyrene and on a chlorinated resin Amberlite XE—64 by covalent attachment E49. 50]. Since then. many methods have been developed for immobilizing enzymes. Several reviews [6. 46. 51-56] have covered the topic in detail. The methods can be classified into eight categories: 1) absorption. 2) adsorption. 3) cross-linking. 4) ad- sorption and cross-linking. 5) ion-exchange techniques. A 6) entrapment. 7) copolymerization. and 8) covalent at—' tachment £51]. 1. Absorption Among the methods of immobilizing enzymes. the ab- sorption method is the simplest C51]. Absorption methods have been reviewed in depth by McLaren and Parker [57]. The method usually avoids the use of harsh reagents and conditions. which might denature the enzyme. Its dis— advantage is that the enzymes are desorbed more easily 35 than with other methods. especially in the presence of substrates E58. 59]. 2. Adsorption The adsorption of enzymes onto surfaces is dependent on many factors. including pH. ionic strength. tempera- ture. enzyme concentration. and type of solvent used [60]. If the enzyme molecules are simply absorbed physically within the carrier matrix. then by prolonged washings with buffer or salt solutions. the enzymes may become detached from the carrier. The factors which affect the stability and activity of the adsorbed enzyme are the assay and storage conditions. the substrate concentrations. and the ionic strength of storage solution. as well as those conditions which affect the adsorption phenomenon it— self [51]. Commonly used adsorbents include: alumina. carbon. celluloses. clays. hydroxyapatite. and glasses. including controlled~pore glass. Among these. glass is the best support. Enzymes adsorbed to glass surfaces have ex- tremely long half-lives and can be reused without activity losses E61. 62]. 3. Cross-Linking Enzymes can be immobilized by cross-linking the enzyme molecules into insoluble matrices by use of 36 bifunctional reagents. The reaction produces covalent bonds with intermolecular cross-links between the reagent and the enzyme. Gaffield et al. have prepared an en- zymatically active. water soluble derivative upon reaction of papain and ~chymotrypsin with glutaraldehyde [633. They found that greater enzymatic activities were retained when the active centers of the enzymes were protected. In a similar fashion. Tran~Minh et al. cross-linked the en— zymes and inert protein with glutaraldehyde [453. The multifunctional agents most commonly used are glutaraldehyde. diazobenzidine. and its derivatives. Other lesser used agents include carpodiimides. diiso— thiocyanates. diisocyanates. and disulfonic acids [51]. Reactive groups in the enzyme include terminal amino and carboxyl groups and the substituents of some amino acid residues such as arginine (guanidyl substituent). lysin (amino). hystidine (imidazole). cysteine (sulphydryl). serine (hydroxyl). tyrosine (phenol). aspartic acid (carboxyl). and glutamic acid (carboxyl) [53]. There are many approaches to protecting the active sites of enzymes when cross-linking is performed. The presence of a species. such as a substrate. which selec- tively binds to the active site may enhance the activity of the immobilized enzyme by stabilizing it in its active conformation [64]. Performing the immobilization in the presence of competitive inhibitors also stabilizes 37 enzymes [65]. Binding to the active site in trypsin can be avoided E66. 673 by First polymerizing the enzyme with the N—carboxylanhydride of L-tyrosine. The polytyrosyl chain thus Formed supplies an alternative site to the essential residues in the enzyme. The activity of the cross-linked enzyme depends on many tactors. including concentration of enzyme. reagent used. pH. ionic strength. and the number a? cross-links produced. The overall apparent activity at the derivative also depends on the size at the substrate molecules. Gen— erally. substrates with high molecular weights cannot come in contact with enzymes in the interstices 0F the immobi- lized enzyme supports [51]. 4. Adsorption and Crossvlinking This technique allows an enzyme to be adsorbed on a support and then cross-linked. The process is generally rapid. simple. and produces a product with relatively good stability. The method was First developed by Haynes C473. Enzymes have been absorbed on colloidal silica and then cross-linked with glutaraldehyde. 5. Ion-Exchange Techniques Many enzymes will bond to ion—exchange resins without significant loss oF activity. ThereFore. these enzymes can be immobilized through ion-exchange techniques. The 38 immobilized enzyme remains attached and active unless the pH or ionic conditions are changed to cause elusion of the enzyme. This approach has been used successfully For im- mobilizing L-aminoacylase on DEAE-Sephadex. This product is now commercially available for producing L-methionine at a rate greater than 20.000 kilograms per month [513. b. Entrapment Enzymes can be immobilized be entrapping the enzyme within cross—linked. water insoluble polymers. usually by cross—linking the polymer in the presence of the enzyme. The large enzyme is entrapped in the polymer lattice which is small enough to prevent the enzyme From diFFusing out. but big enough to allow small substrate molecules to diF- Fuse into the polymer [51].! Materials used For entrapment include polyacrylamides. silicone rubber. silica gel. and starch. The ion-exchange method has the advantage that. apart From changes brought about by secondary interactions be- tween the enzyme and polymer. the enzyme is not perturbed. and thus loses little or no activity. Because 0? diF- Fusion control and steric hinderance. this method is less Favorable tor enzymes having large substrates. e.g.. ribo- nuclease.<$ and:8ramylase. trypsin. and chymotrypsin. 39 7. Copolymerization Another popular method is to copolymerize an enzyme by forming covalent bonds between its molecules. Either neutral or charged derivatives of copolymerized enzymes can be prepared. Maleic anhydride and ethylene are the most commonly used reagents for this method. Because of steric effects. the enzyme derivatives show little or no activity for large macromolecular substrates. For enzymes with such substrates. copolymerization. along with entrap— ment and macrocapsulation. is not recommended. 8. Covalent Attachment Covalent attachment of enzymes to water—insoluble carriers is the most commonly used method for immobilizing enzymes. It is versatile and produces immobilized enzymes with excellent stability. Numerous techniques have been developed. Carboxymethylcellulose azide. azo compounds. isocyanates and isothiocyanates. carbodiimides. cyanogen biamide. and glutaraldehyde are the most commonly used reagents. A comparison of four techniques for immobilizing enzymes. as originally reported by Larry Bowers [63. is presented in Table 4. ~— 4O Table 4. Comparison of Techniques for Immobilizing Enzymes (from Bowers C63). Method Advantages Limitations m_--——--’-—-_—_—_-~~—-----—‘~—_~‘-‘---“ —— Adsorption Protein- cross-linking Entrapment Covalent bonding Chemically simple; can give high ini- tial yield; widely applicable Can make membranes Chemically simple; high initial yield; can make membranes; widely applicable; can be lyophilized Very flexible ap- proach can achieve good flow proper- ties (glass sup- port); no leakage of protein Rigid control of con- ditions to prevent desorption Poor flow properties; low reactivity for high MN materials Poor flow properties. low reactivity for high MW materials; will leach protein slowly Low initial yield of enzyme activity 41 Characteristics of Immobilized Enzymes The behavior of the enzyme usually changes in many aspects upon immobilization. most commonly in its pH pro— file and its kinetics and stability. pH Profile. Immobilization can exhibit a carrier-dependent effect on the pH—activity profile. which is especially sensitive to the charge of the carrier. Goldstein et al. reported that the pH-activity profile of the a water-insoluble derivative of trypsin (IMET) at low ionic stregth shifted +2.5 pH units from that of the free enzyme [68]. At high ionic strength. the pH-activity curve of IMET shifted toward more acidic pH values. If a carrier is negatively charged. then a high concentration of positively charged ions. most commonly H . will accumulate at the boundary layer between the carrier and the surrounding solution. The accumulation of hydrogen ions will cause the pH at the carrier surface to be lower than that of bulk solution. in this case. the optimal pH in the bulk solution for the immobilized enzyme may be increased. since the immobilized enzyme actually sees a pH below that of the bulk solution. If the carrier is negatively charged. the opposite may occur. Through the choice of the carrier. it is possible 42 to give the immobilized enzyme different electrostatic fields. and thus select the desired pH-activity profile. Kinetics. A shift in the Michaelic constant (Km) sometimes ac- companies the immobilization of the enzyme. when a change is observed. it is generally an increase. The increase is usually related to the substrate charge and/or carrier diffusion effects. but in some cases. it is caused by tertiary changes in enzyme configuration [51]. The ionic strength of the solution affects the Km in many cases. Stability. The thermal stability of the enzyme is often enhanced by immobilization £51. 53]. However. enzymes which show excellent thermal stability do not necessarily show ex- cellent operational stability. This is because the opera— tional stability is a function of carrier durability. organic inhibitor concentration. and heavy metal inhibitor concentration. as well as thermal stability. Neetal [51] found that the clogging of the carrier can also hinder the stability of the immobilized enzyme. 43 Stirring. Diffusion. and Steric Effects The rates of reactions catalyzed by immobilized enzymes are affected by factors such as stirring. dif- fusion. and steric effects [53]. The rates of reactions catalyzed by polymer—bound enzymes are markedly increased by rapid stirring of their suspensions. Stirring reduces clumping and settling of the conJugates. and increases reaction rates that are diffusion controlled. Hornby et al. [69. 533 have reported the mathematical relationship of diffusion rates to the Michaelis—Menton constant of the enzyme. According to their report. the diffusion term gives an increasing Km (usually decreased rate). Km in- creases with inceasing thickness of the diffusion layer and with increasing maximum reaction rate of the free enzyme. Steric effects are intimately related to diffusion } control. They lessen the activities of inclusion con- Jugates of enzymes whose substrates have high molecular weight. For this reason. hydrolysis reactions catalyzed I by proteolytic enzymes are slower for large molecules. such as protein substrates. than for small molecules. such as small esters and amides. The phenomenon is noticed when an increase in the enzyme content of the immobilized layer from an already high level causes a noticeable re~ duction in the enzyme activity for substrates with high molecular weights. Adding the enzyme slowly to a stirred 44 suspension of polymer will cause fewer cross-linkings to form. 45 B. ENZYME IMMOBILIZATIDN METHODS DEVELOPED IN THIS RESEARCH The methods for immobilizing enzymes developed or modified in this research prOJect are described in this section. Four different methods for immobilizing enzymes onto an electrode were used in this research. The first method uses physical entrapment. the second uses CTDSS“ linking. the third uses a combination of cross-linking and physical entrapment. and the fourth uses cross-linking and covalent attachment. These methods and the details of their procedures are described individually. in the order in which they were developed. Method 1: Entrapment of Enzymes and Inert Protein Between a Cellophane Membrane and a Gas Permeable Membrane Method 1 was adapted for the enzyme electrode pro- Ject from a technique developed by Papastathopoulos and Rechnitz C23. 70]. Our goal was to obtain a simple. reliable. and economical method by which one or more enzymes could routinely be immobilized onto an ammonia- gas-permeable electrode. In this method. the enzymes and an inert protein are trapped between a gas-permeable mem- brane and a cellophane membrane. The inert protein is added to stabilize the enzymes. The presence of EDTA. certain metal ions (e.g.. the 46 cofactors of the enzyme). or an inert protein usually stabilizes enzymes in dilution. Because of its high molecular weight. the inert protein albumin can be en- trapped with enzymes within a dialysis membrane and will not diffuse out. Albumin is economical and readily available. The use of an inert protein has another advantage. Proteins usually become bound to. and/or trapped in. the materials commonly used as supports. When this happens to an enzyme. it causes a loss in activity since its active sites are bound to the support. or since trapped or bound enzymes often lose the flexibility necessary for the enzyme-substrate reaction. However. if an inert protein is present at a concentration much higher than the con— centration of the enzyme. most of the molecules which be- come bound or trapped will be those of the inert protein. This minimizes the loss in enzyme activity. Once the enzyme electrode is constructed. its enzyme membrane must be soaked in a buffer for at least one hour. The buffer must be stirred throughout this time. while the electrode is soaking. the used buffer should oc- casionally be discarded. and replaced by a fresh buffer. This should be done at least twice. This dialysis process allows impurities. especially ammonium ions. to diffuse out of the immobilized enzyme layer. This is necessary since many enzyme and protein preparations are obtained 47 using ammonium sulfate by the salt precipitation method. and thus they often contain high concentrations of am* monium ions. The buffer pH is also an important factor in the stability of an enzyme. The pH of the buffer in which the enzyme electrode is stored should be the pH at which the enzyme is most stable. usually the pH at which the enzyme is most active. Storage in a buffer with improper pH can easily denature the enzyme. The enzyme electrode should; also be kept refrigerated whenever possible. to prevent thermal denaturation. This method for immobilizing enzymes avoids com- plicated procedures. It uses no harsh reagents which can denature the enzyme. It also requires no covalent bonds be made to the enzymes as these can lower the enzyme ac- tivity. either by making active sites unavailable to the substrate. or by reducing the flexibility of the enzyme molecules. Only a few microliters of enzyme solution are needed for each enzyme electrode. The following are detailed directions for the pro— cedure developed: 1. Precut some circular cellophane membranes with size slightly larger than the gas-permeable membrane. go 48 Using a pair of clean tweezers. pick up a gas- permeable membrane. Place the desired amount of an albumin and enzyme preparation onto the shiny side of the gas membrane using a syringe or a pipet. Disperse the solution uniformly over the shiny sur- face of the gas-permeable membrane. This can be done using a piece of capillary tubing whose ends have been sealed in a flame. Using a pair of tweezers. pick up a cellophane mem- brane and carefully fit it on the enzyme-wetted side of the gas membrane. Install the two-layer membrane into the bottom cap of the ammonia electrode and assemble the electrode. Soak the enzyme membrane for at least one hour at 4 C in a continually stirred buffer of the appropriate pH. During this time. the buffer should be replaced with fresh buffer at least twice. 49 Method 2: Copolymerization of Enzymes and Albumin with Glutaraldehyde while Method 1 works well for producing single enzyme membranes. it does not allow a large number of membranes to be created at the same time. Tran-Minh and Brown [453 have reported a procedure for preparing relatively large enzyme membranes using enzymes and albumin. and using glutaraldehyde as a copolymerizing agent. The membranes could be stored in a refigerator and pieces could be cut off. as needed. for use in electrodes. We attempted to use their method in this research. but the membranes produced proved to be too fragile. A new method for immobilzing enzymes was therefore de~ veloped. The goal was to obtain an a polymer gel formed from enzymes and proteins. A membrane could be obtained from such a gel by slicing a thin piece from the gel and fitting it into the electrode. The rest of the gel could be refrigerated and stored for later use. This process appears promising for commercial production of an enzyme 0 membrane. In the method developed. we are able to obtain an insoluble enzyme-and-protein polymer gel inside a glass tube by copolymerizing enzymes and albumin with glutar- aldehyde. An advantage is that the insoluble enzymes are usually more resistant to atmospheric oxidation. heat. and microbial attacks [46]. However. part of the enzyme’s 50 activity is sacrificed since some active site are bound and the flexibility of the enzyme molecules is reduced. In addition. steric effects may be increased (see Sec- tion A). A 25% glutaraldehyde solution is used as a COPOIUMETT izing agent. In order to minimize the number of active sites bound by glutaraldehyde. this solution is saturated with the substrates of the enzymes to be polymerized. The presence of the substrates also decreases the formation of some unnecessary cross-links. Reducing the number of cross-links increases the average pore size. thus easing the passage of substrates in the gel. Since ATP. ADP. AMP. and adenosine are relatively large in size. extensive cross-linking should be avoided. An inert protein solution. containing 1.5% albumin. is used as proteic feed [45]. A high total protein con— centration in the polymer usually results in more cross- linking which creates a stronger polymer. To obtain optimal conditions for forming the most useful polymer gel. we tested various conditions. The following are detailed directions for producing the polymer: 1. Precut some 10 cm long glass tubing with an inside diameter slightly larger than the gas-permeable membrane. 51 Hz-N-Enzvme H\ Mcéo oec \H l 4 N- Enzyme l. CH Figure 1. Attachment of Glutaraldehyde to Enzyme. Wrap one end of a tube tightly with parafilm. Then insert this end into a rubber stopper. Using the stopper as a stand. position the tube so that the open end faces upward. Pipet the desire amount (between 0.40 and 0.96 ml) of 10% albumin solution into the tube. Add deionized water to bring the contents of the tube to 0.96 ml. Add 0.04 ml of 25% glutaraldehyde solution. (When making a polymerized enzyme gel. this solution should be saturated with the subtrates of the enzymes used.) Cover the open end of the tube with parafilm and. holding down the film tightly with the thumb. turn the tube over and back at least ten times to insure complete mixing. Tap the tube to dislodge any air bubbles. Carefully place a small volume of water on top of the solution. 53 9. Set the tube aside for one or two hours until a yellow polymer is formed. 10. Remove the gel from the tube as follows: Carefully insert a thin needle from a syringe between the edge of the gel and the glass tube. Then slide the needle around the inside edge of the tube to loosen the gel. 11. Soak the gel in the buffer solution. We tested the formation of the gel when various amounts of albumin were used. The results showed that a firm polymer gel can be formed when at least 0.6 ml of albumin solution (i.e.. 6% albumin) is used. Increasing the albumin content of the gel made its yellow color dark- er. and increased the strength of the polymer. It was also found that a firm gel cannot be formed with 0.03 ml or less of the glutaraldehyde solution. To minimize steric effects. we decided to use the minimum amount of albumin. i.e.. 6%. needed for the enzyme to gel. In making an enzyme gel which was sensitive to AMP and adenosine. 125 units of deaminase and 156 units of phosphatase were dissolved in the 10% albumin solution used in the above procedure. After the gel had formed. a thin slice (about 2 mm thick) was cut from the gel with a razor blade. and fitted together side by side with the 54 gas—permeable membrane of an ammonia electrode. The lowest detectable limit of the AMP sensor with a 2 mm enzyme membrane was 0.001 M. A thinner membrane should increase the sensitivity by reducing the diffusion con- trol [63. However. since we did not have a microtome. we were not able to get a thinner membrane. 'The gel was stored in Tris-HCl buffer at 4 C. Leaking of the enzymes cannot occur since they are covalently bound. This method may prove suitable for long-term storage or commercial applications of immobilized enzyme membranes. Method 3: Entrapment of Enzymes that are Copolymerized with Glutaraldehyde This method was developed and tested after it was discovered that the sensitivity of the adenosine electrode prepared by Method 1 decreases from day to day. Our goal was to develop a method as convenient and inexpensive as Method 1. but also to produce an electrode with greater stability. This method combines Methods 1 and 2. Enzymes are entrapped between a cellophane membrane and a gas— permeable membrane (see Method 1). .The enzyme membrane of the electrode thus created is then soaked for two hours in a continuously stirred 25% glutaraldehyde solution that was saturated with substrates (see Method 2). The enzyme 55 mixture becomes a soluble glutaraldehyde-bound enzyme-and- protein mixture with a light yellow color (Figure 2). An AMP electrode was prepared by this method using the enzymes adenosine deaminase and alkaline phosphatase. Unfortunately. it showed poor sensitivity and response. The pH profile of the sensor sensitivity was also studied (Figure 3). The AMP electrode prepared by this method clearly shows a pH profile different from those of the electrodes prepared by Methods 1 and 4 (see Chapter VI). Its sensitivity is about half those of the AMP electrodes prepared by Methods 1 and 4. Method 4: Copolymerization and Cross~ Linking of Enzymes by Glutaraldehyde This method modifies Methods 2 and 3. An enzyme mem- brane is constructed by using glutaraldehyde to copolymer- ize the enzyme as well as to bind the enzyme covalently to the gas~permeable membrane of an ammonia sensor. An AMP electrode constructed using the method developed shows very good sensitivity and a fast response rate and proves the success of this method. The resulting membrane is a polymerized enzyme membrane which is strongly bound to its support. the gas membrane. The membrane can last for months without denaturization. Interestingly. the pH profile of the AMP electrode is quite different from those constructed by using Method 1 56 1. A electrode prepared by Method 1. N A electrode prepared by Method 3. . A electrode prepared by Method 4. 3 4 A enzyme—and-protein polymer gel prepared by ' Methode 2. Figure 2. Enzyme Electrodes and Enzyme Polymer Gels. 57 .m oeoapma an awesomom unoppooam wee no maauoom mm .m assume .m2< 2 .0H .m 9550 «commences E IOH .< 9550 rummage. 3.2ngan Smoé Adudv £3me Humane; 23.0 A. .. 0V .momm Pm mwmcommwn H3. a 92 I 3 ca _ _ _ _ 0. MW 58 and Method 3 (see Figures 3. 12. and 13). This is expect- Ed: since these immobilized enzymes prepared by different methods may actually differ in their chemical stuctures and prOperties. The following are detailed directions for the pro- cedure developed: gu Prepare a 2.5% glutaraldehyde solution in a 0.02 M phosphate buffer at pH 6.8 and saturated with the substrates of the enzymes to be immobilized. Assemble the bottom cap with a gas-permeable membrane according to the manufacturer’s instructions. Stand the cap so that the membrane faces up. Apply 20 ul of the enzyme mixture onto the gas— permeable membrane. The solution applied should contain a total of about 5 mg of protein. Refrigerate the membrane at 4 C for 15 hours to allow the solvent to evaporate. Place a 25 ul portion of glutaraldehyde solution onto the membrane. (An amount greater than 30 ul will fail to form a firm enzyme membrane.) 59 Refrigerate the membrane at 4 C for an additional 1.5 hours. A gelatinous enzyme membrane with a yellow color will formed (Figure 2). Carefully rinse the membrane. first in a glycine solution. then in water. in order to neutralize the excess polymerizing agent. CHAPTER V ADENOSINE ELECTRODE A. INTRODUCTION An electrode developed as a sensor for the nucleo- tide 5’-adenosine monophosphate (5’-AMP) has been re- ported EZBJ. He now have developed a sensitive electrode for the nucleoside adenosine (adenosine riboside). which has a detection limit in the range of 0.7 to 5.0 uM and is 10 to 20 times more sensitive than similar electrodes reported by other workers [23. 24. 30. 45. 70]. This chapter covers the design and characteristics of the adenosine sensor. The adenosine electrode is constructed with a layer of a suspended mixture of adenosine deaminase and bovine serum albumin between a dialysis membrane and the gas- permeable membrane of an ammonia-sensing membrane elec- trode. Adenosine is deaminated according to the equation adenosine . . deaminase adenosine (adenine r1boszde) + H20 —% inosine (hypoxanthine riboside) + NH3. When the electrode is in contact with a sample solution containing adenosine. the ammonia concentration produced at the electrode surface by the above reaction is 60 bl proportional to the concentration of the adenosine and produces the potentiometric response of the electrode system. The resulting electrode has excellent sensitivity and selectivity For adenosine. This study shows that sensor response depends upon pH and temperature; and that the Optimal pH for the electrode is much higher than that For the enzyme reaction alone. B. EXPERIMENTAL Apparatus. The ammonia-gas-sensing electrode used was an Orion model 95-10. The electrode consists of an inner combina- tion pH electrode outfitted with a gas-sensitive membrane. The internal solution is 5 x 10'5 M NHuCl. Electrode po— tential measurements were made using a Markson ElectroMark Analyzer. The readings were recorded manually. Measure- ments were made in a thermostated water bath with a pre- 0 cision at t 0.2 C. A model 2400 Beckman spectrophotometer was employed For the spectrophotometric rate determination 0F enzyme activity. Reagents. All chemicals used were reagent grade. Freshly deionized. ammonia-free water was used to prepare all solutions. The enzyme used in this study was adenosine deaminase (Type III. From calf intestinal mucosa. activity 235 units per milligram of protein at pH 7.5 and 250C. about 5 milligrams per milliliter solution. Sigma Chemical Company). The enzyme was used without further puriti- cation. A 15% solution of albumin (From serum bovine. crystallized and lyophilized. containing 1-3% globulin. Sigma) was prepared in 0.05 M Tris-H01 buFFer pH 7.5. The 0.01 M stock solutions oF adenosine. 2’-AMP. 3’-AMP. 5’-AMP. and 2’ and 3’-cyclic AMP were prepared in 0.05 M Tris-HCl buFFer. Adenosine was Found to have a solubility limit of about 0.01 M in the buffer at room temperature. To test the ammonia response at the ammonia electrode. a stock solution of 0.1 M ammonium chloride was prepared with deionized water. Procedure. The enzyme electrode used in this study was prepared with 7 ul 0? deaminase (corresponding to 8 units at pH 7.5 and 250C) and 3 ul 0? 15% albumin solution. These were dispersed uniformly on the surface of the gas-permeable membrane with a sterilized glass rod. A cellophane mem- brane. carried by plastic tweezers. was Fitted careFully 63 /z,//i //// T: - ‘Reference _ @ electrode Outer body——————T;\ Internal electrolyte /flf7 zf///7 // j /////// Inner body (pH electrode) ‘/////_//// /‘ " //// _____.O-ring L . spacer .Bottom cap Cellophane ammonia—permeable Membrane membrane Deamimase and albumin solution Figure in [Schematic Diagram of Adenosine Electrode. 64 on the enzyme~wetted membrane. so as to avoid trapping any air bubbles. Thus. a "sandwich" of a thin layer of enzyme solution between two membranes was formed. The spacer containing the membranes was installed in the button cap of the electrode so that the cellophane membrane faced down. toward the sample solution. The enzyme electrode was then soaked in a continually stirred 0.05 M Tris-HCl buffer for at least one hour before use. Leaking some- times occurred. but it was easily detected by observing a yellow coloration of the buffer solution. The electrode was stored in 0.05 M Tris-H01 buffer. pH 7.0. and refrigerated at 40C. All the solutions pre— pared were tested for ammonia contamination with the ammonia sensor before use. No ammonia contamination was found in any of the solutions. RESULTS AND DISCUSSION Calibration Curve. A typical calibration curve for the adenosine elec— trode in 0.05 M Tris-HCl buffer. pH 9.0 and 37°C. is shown in Figure 5. in which 4E is the difference between the potential reponses in the buffer background and the sample solution. The electrode develops a linear response over 65 240 - IBO— AE,mV 60*- 8 6 4 2 -log [adenosine] , M 0.05 M Tris-H01 buffer, pH 9.0, 37 °c Figure 5. Calibration Curve for Adenosine Electrode. 66 the range 7 X 10"7 to 1 X 10.2 M substrate concentration with a slope of 55 mV per decade. Effect of pH on Electrode Sensitivity. The concentration of ammonia in an aqueous solution is pH dependent. as is the activity of a given amount of enzyme. In aqueous solutions. the pH has a positive ef- fect on the ratio of ammonia to ammonium with ammonia being the predominant form at pH 11 or above [713. Un- like many other enzymes. adenosine deaminase is active over a very wide pH range [373. Our determination of enzyme activity as a function of pH in a 0.05 M Tris-H01 buffer is shown in Figure 6. Although the pH of maximum activity is observed near the neutral point. the activi- ties at pH 9.0 and pH 6.0 are 63% and 78% that at pH 7.0. respectively. Since the enzyme activity remains high at pH values greater than 7. and the ammonia/ammonium ratio is greater at higher pH. the maximum sensor sensitivity is expected to occur at a pH appreciably greater than 7. As shown in Figure 7. the potential measured in both the buffer back- ground and the sample solution decreases with rising pH. However. the difference in voltage between buffer and sample rises rapidly with pH. and achieves a maximum near pH 9.0. The pH of maximum sensor sensitivity is far more basic than the optimum pH for the enzyme reaction itself. 67 6) U0 1 I 'Relotive Enzyme Activity (M I 0 6L 3 IO (0- o) 0.05 M Tris-HCl buffer, (A-L) 0.02 M phosphate buffer Figure 6. pH profile of Adenosine Deaminase. 68 l60 — :20 — :> 5‘80— Lo c 40—- . I . i e (D 6 3 l0 pH All responses at 25°C; Curve A, buffer background; Curve B, 10" M adenosine; Curve C, the differences of Curve A and Curve E. Figure 7. .Adenosine Electrode Response as Function of pH. 69 which is at pH 7. At pH 9.0. the sensor sensitivity is eight times greater than at pH 7.0. and four times greater than at pH 7.5. Further investigation of the response in various con- centrations at pH 9.0 and pH 7.5 (Figure 8) demonstrated that higher sensitivity and greater linearity are obtained at pH 9.0. As explained above. while the enzyme activity at pH 9.0 is 37% less than at pH 7.0. this is more than offset by the increased ratio of ammonia to ammonium in the more basic solution. Effect of Temperature on the Electrode Sensitivity. A temperature increase has a positive effect on the diffusion rate. the partial pressure of ammonia. and. within limits. the enzyme activity. Therefore. the elec- trode sensitivity is expected to vary with temperature. 4 The electrode response to 10- M of adenosine was measured at 25. so. 32. 35. and 37°C (Figure 9). The response is essentially constant between 250 and 3000. but increases dramatically above 30°C. The effect of temperatures greater than 37°C (body temperature in humans) was not studied because of the increasing risk of denaturing the enzyme. The electrode response to different concentra- tions of adenosine was measured at 250 and 37°C in solu- tions of pH 7.5 and pH 9.0; the results are shown in 7O F A h60" _ - . C D l20-— > _ E25“) UJ . Q '40- Or- ' I l I I 1 2 6 4 -log [adenosine] All curves in 0.05 M TrisaHCl buffer; Curve A, 8H 9.0, 37 C; Curve B, pH 960’ 25 C; Curve C, pH 7.5.37 C; Curve D, pH 7.5, 25 C Figure 8. Calibration Curves for Adenosine Electrode as Function of pH and Temperature. 71 e :A l60r- 130- :>h00- E. uJ 4o 1 I l l 25 so 35 4 40 Temperature (°C) All curves in 0.05 M Tris-H01 buffer, pH 9.0; Curve A, buffer background; Curve B, 10’ M adenosine; Curve 0, the difference of Curve A and Curve E. Figure 9. .Effect of Temperature on Response of Adenosine Electrode. 72 Figure B. The electrode exhibits higher sensitivity and a more linear response at 370C for both pH 7.5 and pH 9.0. Response Time. The response time. i.e.. the time required for a steady—state potential to be reached. varies with sub- strate concentration. In these experiments. the potential is considered as having reached a steady state when the changes are less than 1 mV in a 2 minute period. The response time is about 7 to 12 minutes for concentrations 4 M. and shortens to 6 to 10 minutes in the con- 4 below 10- centration range of 10' to 10"3 M. and to 2 to 4 minutes 2 in 10‘ M solutions at pH 9.0 and 37°C. The time required to reach the steady-state potential was found to be shorter at higher temperatures. Response times for 10‘” M adenosine at pH 9.0 were 7. 20. and 27 minutes at 37. 30. and 25°C. respectively. Higher temperatures thus have a positive effect on both electrode sensitivity and response rate. Selectivity Study. It was reported that the deaminase from the intestine is specific for adenosine and will not deaminate other adenine derivatives [37]. However. it is difficult to separate completely the deaminase from the potent phos- phatase present in the intestine C72]. Sigma Chemical 73 Company has indicated that the commercial deaminase used in this research has a trace ( i 0.01%) activity for AMP. To test the selectivity. we measured the response of the electrode in solutions containing adenine. 2’-AMP. 3’-AMP. 5’-AMP. and 2’ and 3’-cyclic AMP. respectively. up to 0.01 M. The response to these possible interferants in various concentrations is listed in Table 5. It is clear that the trace of phosphatase contamination did not Jeop- ardize the selectivity. and the electrode has an exclusive selectivity for adenosine. Sensitivity and Stability. Both the sensitivity and stability of the enzyme electrode have been studied over a period of 32 days. The electrode was either used for determinations or exposed to a 37°C buffer solution for 4 to 7 hours on each of the first 9 days. Between uses. and from the 10th day until the 32nd day. the electrode was stored in buffer at 4°C. As shown in Table 6. the sensitivity dropped one order of magnitude over the first 4-day period. but showed no further reduction during the next 5-day period. A sig- nificant decline in the response slope and sensitivity was noted at the 32nd day. In all tests. the electrode ex- hibited a linear response and a detection limit in the range of 2 to 10 UM. even at its 32nd day. 74 .ooem .o.e Ia .eoesoe Hozimeeeo m o o e: a one 03 o o e- m 0 one on o o o o o om 03 o o o o 0 on on oz¢-.m o:<-.m a .m oz¢z.m o:<-.m oeecoem oeeuocoom z .coeemeoeoucou tillilltlIttltllItillttttltltitttiltltlIllttiti ttttttttttttttttttttttttt mytpumnow .coeeoeou >5 : coeeoe >eo >5 .mooepuwsm wfifimocwu< sou wwwooum auw>wpuwfiwm .m mfinmk .m 75 Table 6. Long-term Behavior of the Adenosine Electrode. 8. Response slope __-_EEE _____ “V per deiiif __ 1 55 3 4o 4 43 o 42 7 4o 9 43 32 ‘ 27 Detection limit. concentration. M X 10 10 10 10 10 10 10 0 aTris-HCl buffer. pH 9.0. 37 c. CHAPTER VI AMP ELECTRODE A. INTRODUCTION An electrode selective for adenosine monOphosphate (AMP) and adenosine has been devised. The electrode uses the enzymes adenosine deaminase and alkaline phosphatase in conJunction with a membrane electrode sensitive to ammonia gas. This chapter covers the design and char- acteristics of the AMP sensor. The enzyme sensor is constructed with an immobilized enzyme layer of deaminase and alkaline phosphatase. AMP (either 3’- or 5’-AMP) is converted to adenosine. which in turn is deaminated. The reactions are: alkaline h s hata e AMP (adenosine monophosphate)-—9 O p s > adenosine (adenine riboside) + Pi (1). adenosine . . deaminase adenosine (adenine riboside) + H20 > inosine (hypoxanthine riboside) + NH3 (2). Reaction (1) is the phosphatase-catalyzed dephosphoryl— ation of AMP to adenosine. The adenosine liberated is then susceptible to deamination by adenosine deaminase (Reaction (2)). The ammonia produced in Reaction (2) gives rise to a steady-state electrode potential which is related to the AMP and adenosine concentrations in the 76 77 sample solution. We chose alkaline phosphatase. rather than acid phosphatase. because it is active in basic solu- tions. where the ratio of ammonia to ammonium ions is greater. As was reported in Chapter V. the sensitivity of the adenosine-selective electrode. which employs the soluble enzyme adenosine deaminase. drops by an order of magnitude within four days. This led to the construction of a second electrode which uses glutaraldehyde. a bifunctional agent (i.e.. an agent having two functional groups). Glutaraldehyde polymerizes the enzymes by cross-linking and covalently binding the enzymes to the gas-permeable membrane. In many cases insoluble enzymes are more re- Sistant to atmospheric oxydation. heat. and microbial attacks [46]. Both electrodes are critically examined below for their sensitivity and pH profile. (A third AMP electrode. which employs soluble glutaraldehyde-bound en— zymes. but has poor sensitivity and response. is discussed in Section B of Chapter IV.) 8. EXPERIMENTAL Apparatus. The ammonia-gas-sensing electrode used was an Orion model 95—10. as described in Chapter V. Electrode t 78 potential measurements were made using a Heath elec- trometer. model EU-200-30. The readings were recorded manually. Experiments were performed in a Neslab thermo- stated water bath with a precision of 0.10C. Reagents. All chemicals used were reagent grade. Freshly de- ionized. ammonia-free water was used to prepare all solu- tions. The enzymes used in this study were: adenosine deaminase (Type III. from calf intestinal mucosa. activity 250 units per milligram of protein at pH 7.5 and 250C. about 5 milligrams per milliliter solution. Sigma Chemical Company); alkaline phosphatase (Type I. from calf in- testinal mucosa. activity 1.1 units per milligram of solid. Sigma); and alkaline phosphatase (Type VII. from calf intestinal mucosa. ammonia sulfate suspension. ac— tivity approximately 1000 units per milligram of protein at pH 10.4 and 37°C. about 2 milligrams per milliliter solution. Sigma). The inert protein used was albumin (from serum bovine. crystallized and 1yophilized. con- taining 1-3% glubulin. Sigma). The enzymes were used without further purification. The 0.01 M stock solutions of adenosine. 2’-AMP. 3’-AMP. 5’-AMP. 5’-ADP. and 5’-ATP were prepared in 0.05 M Tris-HCl buffer. The glutar- aldehyde solution was prepared at 2.5% in 0.02 M phosphate 79 buffer at pH 6.8 and then saturated with adenosine and AMP. Procedure. The immobilized enzyme membranes used in this study were prepared in two different ways. Membrane I was pre— pared by dispersing 5 ul of deaminase (corresponding to o units and 11 units at 25°C and 37°C at pH 7). 5 ul of phosphatase (Type VII) (corresponding to 11 units at pH 10.4 and 370C). and about 0.5 mg of albumin evenly over the gas-permeable membrane of the ammonia electrode. A piece of dialysis paper was placed over the enzyme layer to trap the enzymes and albumin. and prevent them from diffusing into the solution. The enzyme electrode was soaked in a continually stirred 0.05 M Tris-HCl buffer for at least one hour before use. Leaking sometimes occurred. but it was easily detected by observing a yellow colora- tion of the buffer solution. Membrane II was prepared using a mixture of 13 mg phosphatase (Type I). 30 ul deaminase. and 10 ul water. A portion of this mixture (20 ul) was placed onto the gas— diffusion membrane. The membrane was refrigerated at 40C for 15 hours to allow the solvent to evaporate. A 25 ul portion of glutaraldehyde solution. the copolymerizing agent. was then added. (The presence of the substrates adenosine and AMP in the glutaraldehyde solution prevents 80 the active sites of the enzyme from being bound by the glutaraldehyde. See Chapter IV.) The membrane was next refrigerated at 40C for an addtional 1.5 hours. Finally. it was carefully rinsed. first in glycine solution. then in water. in order to neutralize the excess bifunctional agent. Membrane II is a polymerized enzyme membrane with a yellow color. Electrodes I and II were then constructed using Membranes I and II. respectively. All the tests were performed at 370C. since it was found that the electrode exhibits a higher sensitivity and faster response at this temperature (see Chapter V). RESULTS AND DISCUSSION Calibration Curves. The results of calibrating a typical Electrode I. and a typical Electrode II. at different adenosine and AMP concentrations at pH 9.3 and 3730 are shown in Figures 10 and 11. Electrode I develops a linear response over the ranges 9 X 1047 to 1 X 10"2 M adenosine concentration. and 1 X 10.6 to 3 X 10"3 M AMP concentration. Electrode II -6 develops a linear response over the ranges 9 X 10 to - -6 5 X 10 M adenosine concentration. and 9 X 10 to 5 x 1073 M AMP concentration. While the response of both 81 ISO 120 > E“ Lu < 60 O I l J l . 4- 2 - log Conc, M All responses in 0.05M Tris-HCl buffer, pH 9.0, 37°C. Curve A, adenosine; Curve B, AMP Figure 10. Calibration Curves for AMP Electrode I. 82 (D nfi” I l I 1 l 1 v;J 5 4- 3 2 I 403 Con—(QM All responses in 0.05M Tris-H01 buffer, pH 9.0, 37°C. Curve A, adenosine; Curve B, AMP Figure 11. Calibration Curve for AMP Electrode II. 83 electrodes becomes nonlinear at high concentrations. the sensitivity of Electrode I increases at high concentra- tions. whereas the sensitivity of Electrode II falls off. Effect of pH on Electrode Sensitivity. As described in Chapter V. the concentration of am- monia and the enzyme activity depend on the pH of the solution. Alkaline phosphatase is active at higher pH -- its maximum activity is at pH 10.4. Figure 13 shows the potential response of Electrode II for adenosine and AMP samples. less the buffer background response. as a func- tion of pH. The curves rise sharply and achieve their maxima near pH 9.3. These pH profiles for adenosine and AMP are sharper than those for Electrode I. which are shown in Figure 12. One explanation of these differences is that the pH profile of a specific enzyme varies with the characteristics of the immobilized enzymes and their carriers (see Chapter IV). ”Response Time. As was the case for the adenosine electrode. the response times for Electrodes I and II vary with the sub- strate concentration. The response time of Electrode I is in the same range as that of the adenosine electrode in its optimal conditions (pH 9.0 and 37°C). The time required for Electrode II to reach the steady-state 84 LL 1 I J 8.0 9.0 p H All responses at 37°C, 0.05M Tris-H01 buffer; Curve A, 10-4M adenosine; Curve B, 10'4M AMP. Figure 12. Response of AMP Electrode I as Function of pH. 85 60'— \ . A AE,mV 20— ‘ B ‘ l . l l 08.0 9.0 l 0 pH All responses at 37° C, (a -O) 0.05 M Tris-H01. buffer, (A -A ) 0.02 M phosphate buffer; Curve A, 10 M adeno- sine; Curve B, 10 4' M AMP Figure 13. Response of AMP Electrode II as Function of pH. 86 potential was less than half that. The response of Elec- trode II at pH 9.3. 37°C. and all concentrations reaches 85% of the steady-state response within two minutes. Membrane II. which was prepared by cross-linking and covalently binding the enzyme molecules. thus gives a faster response. One explanation for this difference is that Membrane II is constructed without a dialysis membrane. so one would expect less diffusion control for Electrode II. Selectivity Study. It was shown that the electrode with immobolized adenosine deaminase has an exclusive selectivity for adenosine (Chapter V). A mixture of adenosine deaminase and alkaline phosphatase should be sensitive to adenosine. 2’~AMP. 3’~AMP. and 5’-AMP [31]. In order to test the selectivity. we tested the response of Electrode II in solutions containing adenine. adenosine. 2’-AMP. 3’-AMP. 5’*AMP. 5’-ADP. and 5’-ATP. respectively. up to 0.01 M. The responses to these substrates and possible inter- -ferants in various concentrations are listed in Table 7. It is clear that the electrode exhibits the same response for 2’-AMP. 3’-AMP. AND 5’-AMP. but exhibits no response to the interferants tested. 87 .oonm .m.o Io .eoeese_auximeeen a- 0 one no“ one e was on o m on 003 me 3: use on o 0 am mm mm o no on o o o s m 0 we as ee¢1.m oo<1.m ezci. ozci.m o:¢n.m monsoon oeemocoon : .coeeneecoucou llllllllllllllllllllllllllllllllllllll ititltliitltlltlitliittt!IIllllist outpumaom Acoeeeaon >e 1 posses >eo >5 s.HH oooeeuoaw oz< cos noeooem see>eouuaom .n seems CHAPTER VII ADP ELECTRODE A. INTRODUCTION It is difficult to identify adenosine diphosphate (ADP) or to determine its concentration in a solution. The assay methods used in laboratories today are either very tedious or very expensive. The enzymatic assay for ADP requires several enzymes (see Chapter III) and the commonly used chemiluminescense method for determining adenosine triphosphate (ATP) is not adaptable to ADP. We have now developed an electrode selective for ADP. 'adenosine monophosphate (AMP). and adenosine. The elec- trode uses the enzymes adenosine deaminase. alkaline phos- phatase. and myokinase in conJunction with a membrane electrode sensitive to ammonia gas. This chapter covers the design and characteristics of the ADP sensor. The enzyme sensor is constructed with an immobilized enzyme layer of deaminase. alkaline phosphatase. and myokinase. Half of the ADP molecules which react with the myokinase are converted to AMP. AMP is converted to adenosine. which in turn is deaminated. The reactions are: 88 89 2ADP (adenosine diphosphate) myOklnase > ATP (adenosine triphosphate) + AMP (adenosine monophosphate) (1). alkaline AMP (adenosine monophosphate) phosphatase > adenosine (adenine riboside) + Pi (2). adenosine adenosine (adenine riboside) + H20 deaminase > inosine (hypoxanthine riboside) + NH3 (3). Myokinase catalyzes the transFer of labile phosphate From one molecule 0? ADP to another. as shown in Reaction (1). The AMP liberated is then susceptible to dephosphorylation by alkaline phosphatase (Reaction (2)). and then to de- : amination by adenosine deaminase (Reaction (3)). The ATP 5 liberated in Reaction (1) does not react Further. but diF~ Fuses through the membrane back into the sample solution. The ammonia produced in Reaction (3) gives rise to a steady-state electrode potential which is related to the ADP. AMP. and adenosine concentrations in the sample solution. The enzyme myokinase. which is also called "adenylate kinase“ and ”ADP phosphomutase." is highly speciFic For adenine nucleotides. It is not active to inosine di- phosphate (IDP). guanosine diphosphate (GDP). or uridine diphosphate (UDP). and has at most slight activity to cytidine diphosphate (GDP) [73. 74. 75. 7b]. Magnesium ions. or. less suitably. manganous ions are required For 90 the enzyme’s activity. The optimum concentration 0F mag- nesium ions is about half the initial concentration oF ADP. The enzymatic activity is inhibited by heavy metal ions. such as silver. zinc. and mercuric ions. Cysteine. EDTA. and some inert proteins (e.g.. serum albumin or ATP-Cr transphosphorylase) stabilze the enzyme at high dilution £773. Myokinase is relatively unstable when the ionic strength of the solution is low. but extraordinarily stable when the ionic strength is high [76]. We chose myokinase. rather than nucleoside diphos- phatase. because the latter is active to all the nucleo- side diphosphates. Using myokinase helps to insure that the electrode is specific to ADP. The ADP electrode is critically examined below For its sensitivity. pH profile. and selectivity. This study shows that the sensor produced has good sensitivity. Its response is dependent on the pH of the sample solution. and the optimal pH is the same as that For the adenosine 5811501“. B. EXPERIMENTAL Apparatus. The ammonia-gas-sensing electrode used was an Orion model 95—10. as described in Chapter V. Electrode 91 potential measurements were made using a Heath elec- trometer. model EU-200-30. The readings were recorded manually. Experiments were performed in a Neslab thermo- stated water bath with a precision of i:0.lch. Reagents. All chemicals used were reagent grade. Freshly de- ionized. ammonia-Free water was used to prepare all solu- tions. The enzymes used in this study were: adenosine deaminase (Type III. From cal? intestinal mucosa. activity 250 units per milligram oF protein at pH 7.5 and 25°C. about 5 milligrams per milliliter solution. Sigma Chemical Company); alkaline phosphatase (Type VII. From cal? in- testinal mucosa. ammonia sulfate suspension. activity ap~ proximately 1000 units per milligram 0? protein at pH 10.4 and 3700. about 2 milligrams per milliliter solution. Sigma); and myokinase (Grade III. From rabbit muscle. suspension in 3.2 M ammonium sulFate. activity 1925 units per milligram 0? protein. 2 milligrams protein per milli- liter. Sigma). The inert protein used was albumin (From serum bovine. crystallized and lyophilized. containing l-BZ glubulin. Sigma). The enzymes were used without Fur- ther purification. The 0.01 M stock solutions of 5’-ADP. and 5’-CDP were prepared in 0.05 M Tris-HCl buFFer. 92 Procedure. The immobilized—enzyme membrane used in this study was prepared by dispersing 5 ul 0? deaminase (correspond- ing to 6 units and 11 units at 25°C and 3760 at pH 7). 5 ul of phosphatase (Type VII) (corresponding to 11 units at pH 10.4 and 37°C). 3 ul of myokinase (corresponding to 12 units at pH 7.6 and 37°C). and about 0.5 mg of albumin evenly over the gas-permeable membrane of the ammonia electrode. (The presence of the inert protein albumin in the immobolized enzyme membrane stabilizes the enzymes. particularly the myokinase.) A piece 0? dialysis paper was placed over the enzyme layer to trap the enzymes and albumin. and prevent them trom diFFusing into the solu- tion. The enzyme electrode was soaked in a continuously stirred 0.05 M Tris-H01 buFFer For at least two hours before use. A stock solution a? 1 M magnesium chloride was used to make buFFers which have magnesium ion concentrations 0? 0.005 M. in order to produce the maximum activity of myokinase. The electrode was stored in 0.05 M Tris-H81 buFFer. pH 7.0. with 0.005 M magnesium ion concentration. 0 All tests in this study were performed at 37 C. 93 RESULTS AND DISCUSSION Calibration Curves. A typical calibration curve For an ADP electrode. showing its response to various ADP concentrations in 0.05 M TrierCl buFFer with 0.005 M magnesium ions. at pH 9.0 and 37°C. is given in Figure 14. The electrode develops a detectable response at concentrations above 1 X 10-5 M. and has a linear response over the range 1.5 x 10'5 M to e x 10‘3 M ADP concentration with a slope of about 55 mV per decade. The detection limit For the ADP electrode is about an order of magnitude higher than the limits For the adeno- sine and AMP sensors described in Chapter V and VI. This is as expected. since only half the ADP molecules con; verted by myokinase become AMP. and since the ADP mole- cules must undergo two separate enzyme reactions to become adenosine. During these steps. which are relatively time comsuming. many 0? the intermediate molecules can diFFuse through the dialysis membrane back to the solution and become unavailable For the next reaction step. EFFect of pH on Electrode Sensitivity. As described in Chapter V. the concentration oF am~ monia and the enzyme activity depend on the pH 0? the solution. The maximum activity 0? myokinase is observed 91+ IZO" 80— ' AE,mV 40—- . l . L 5 4 3 -log[ADP] 0.05 M Tris-H01 buffer with 0.005M‘Mgf’l', at pH 9.0 and 37°C Figure 1.4. Calibration Curve for ADP Electrode. 95 near pH 8.0 [763. The sensor sensitivity. measured as the difference in voltage between buffer and sample. rises rapidly with increasing pH. and achieves a maximum near pH 9.0 (Figure 15). This is also the pH for maximum re- sponse from the adenosine electrode. which uses only the enzyme adenosine deaminase. Selectivity Study. The enzyme myokinase is known to be highly specific to ADP. It is not active to other nucleoside diphos- phates. with the possible exception of CDP [76]. It is therefore only necessary to test the ADP electrode’s se- lectivity toward CDP to learn its selectivity toward all the nucleoside diphosphates. we checked the response of the sensor in solutions containing ADP and CDP. respec- tively. in concentrations up to 0.01 M. The responses to these substrates are listed in Table 8. It is clear that the sensor does not respond to CDP. 96 50" > .. E; 0‘ DJ 4 30" IO . ' " ‘ 1 8.0 9.0 Ffld All responses in 10'4M+'ADP§in 0.05 M Tris-H01 buffer with o.oo5M Mg ‘. 37 c Figure 15. Response of ADP Electrode as a Function of pH. 97 Table 8. Selectivity Studies for ADP Electrode.a mV (mV buffer — mV solution) Substrate 33353333323311‘ __________ 5‘33 _________________ 32': _________ 10 4 0 1O 50 O 10 105 O --».. , .fl“~M’-* a...— aTris—HCI buffer containing 0.005 M MgTI. pH 9.0. 37°C. CHAPTER VIII ATP ELECTRODE A. INTRODUCTION An electrode selective for adenosinse triphosphate (ATP). adenosine diphosphate (ADP). adenosine monOphos- phate (AMP). and adenosine has been devised. The elec- trode uses the enzymes adenosine deaminase. alkaline phosphatase. and potato apyrase in conJunction with a membrane electrode sensitive to ammonia gas. This chapter covers the design and characteristics of the ATP sensor. The enzyme sensor is constructed with an immobilized enzyme layer of deaminase. alkaline phosphatase; and po~ tato apyrase. ATP is converted stepwise to ADP. AMP. and then to adenosine. which in turn is deaminated. The re— actions are: ATP (adenosine triphosphate) —POtatO apyrase ) ADP (adenosine diphosphate) + Pi (1). ADP (adenosine diphosphate) —POtato apyrase > AMP(adenosine monophosphate) + Pi (2). alkaline AMP (adenosine monophosphate) Aphosphatase ) adenosine (adenine riboside) + Pi (3). adenosine adenosine (adenine riboside) + HZO deaminase_; inosine (hypoxanthine riboside) + NH3 (4). 98 99 Reactions (1) and (2) are the dephosphorylation of ATP to ADP. and of ADP to AMP. Both reactions are catalyzed by potato apyrase. The AMP liberated is then susceptible to dephosphorylation by alkaline phosphatase (Reaction (3)). and then to deamination by adenosine deaminase (Reac- tion (4)). The ammonia produced in Reaction 4 gives rise to a steady—state electrode potential which reflects the ATP. ADP. AMP. and adenosine concentrations in the sample solution. The enzyme potato apyrase is an adenosine diphos- phatase as well as a triphosphatase. The presence of calcium ions in the sample solution can double the enzyme activity [36]. ATP is split somewhat faster than ADP when the compounds are added in equimolar amounts to solutions of potato apyrase. when both are added to a solution of potato apyrase. the rate of phosphate production is very nearly the average of the rates of the ATP and ADP alone. Thus. they appear to compete for the enzyme. we chose potato apyrase rather than muscle adenosine triphosphatase because the potato apyrase is both a diphosphatase and a triphosphatase. Thus. the myokinase used in constructing the ADP electrode (Chapter VII) can be omitted. While apyrase can catalyze the dephosphorylation of inosine triphosphate (ITP) and inosine diphosphate (IDP). the end product. inosine. is not susceptible to the deaminase reaction. 100 The resulting electrode can detect ATP at 10 UN at 0 pH 9.0 and 37 C. Operating variables have been critically examined to define the conditions for optimum sensitivity. B. EXPERIMENTAL Apparatus. The ammonia-gas-sensing electrode.used was an Orion model 95—10. as described in Chapter V. Electrode poten- tial measurements were made using a Heath electrometer. model EU-200-30. The readings were recorded manually. Experiments were performed in a Neslab thermostated water 0 bath with a precision of f0.1 C. Reagents. All chemicals used were reagent grade. Freshly de~ ionized. ammonia-free water was used to prepare all solu- tions. The enzymes used in this study were: adenosine deaminase (Type III. from calf intestinal mucosa. activity 250 units per milligram of protein at pH 7.5 and 2500. about 5 milligrams per milliliter solution. Sigma Chemical Company); alkaline phosphatase (Type VII. from calf in~ testinal mucosa. ammonia sulfate suspension. activity ap— proximately 1000 units per milligram of protein at pH 10.4 0 and 37 C. about 2 milligrams per milliliter solution. 101 Sigma); and potato apyrase (Grade I. from potato. 1yo- philized powder. ATPase activity approximately 5.1 units per milligram of protein. and ADPase activity approximate- ly 1.4 units per milligram of protein. both at pH 6.5 and 300C). The enzymes were used without further purifica- tion. The 0.01 M stock solutions of adenosine. 5’-ITP. and 5’-ATP were prepared in 0.05 M Tris-HCl buffer. Procedure. The immobilized-enzyme membranes used in this study were prepared by placing a suspended enzyme mixture be- tween a dialysis membrane and the gas-permeable membrane. Electrode I was prepared from an enzyme mixture consisting of 21 ul of deaminase (corresponding to 25.2 units at pH 7 and 2500). 12 ul of phosphatase (corresponding to 25.2 units at pH 10.4 and 37°C). 10 mg of apyrase (correspond— ing to 50.1 units for ATP and 14 units for ADP). and 20 ul _of deionized water. which was added to dissolve the apyrase completely. A 10 ul portion of this mixture was then dispersed evenly over the gas—permeable membrane of the ammonia electrode. A piece of dialysis paper was placed over the enzyme layer to trap the enzymes. and prevent them from diffusing into the solution. The enzyme electrode was soaked in a continuously stirred 0.05 M Tris~HCl buffer for at least one hour before use. Leaking sometimes occurred. but was easily detected by observing a yellow coloration of the buffer solution. Electrode I had a low sensitivity. so Electrode II was createdu The enzyme mixture for it consists of 21u1 of deaminase. 12 ul of phosphatase. and 18 mg of apyrase. No water was added to the mixture. which was thick with undissolved apyrase. A micropipet in the form of a capil- lary tube was used to place 10 ul of the mixture onto the gas-permeable membrane. The remainder of the procedure for preparing Electrode II is the same as for Electrode I. A stock solution of 5% CaCl was used to make buffers with calcium ion concentrations of 0.5 mg/ml (0.05%). which is needed for maximal enzyme activity. The elec- trode was stored in 0.05 M Tris-H01 buffer at pH 6.5 (op- timal for apyrase) with 0.05% calcium ion. as cofacter. and refrigerated at 4CD. All tests in this study were performed at 37 CJC. RESULTS AND DISCUSSION Calibration Curves. The results of calibrating a typical Electrode I. and a typical Electrode II. at different ATP concentrations at pH 9.0 and 37°C are shown in Figures 16. Electrode I has a detection limit of 1 X 10- M. and develops a linear 103 ISO - I 03 [AT P] All responses in 0.05 M Tris-H01 with 0.05% Caf+’ pH 9.0, 37 C; Curve A, ATP electrode II; Curve B, ATP electrode I Figure 16. Calibration Curve for ATP Electrodes I and II. 104 4 to 5 X 10.3 M ATP con- response over the range 1 X 10- centration with a slope of about 46 mV per decade. Elec- trode II has a detection limit of 2 X 10'"5 M. and develops a linear response over the range 2 X 10-5 to 3 X 10-3 M ATP concentration with a slope of about 46 mV per decade. Electrode II has a lower detection limit because of the higher enzyme concentration in the mixture used. Further increases in the apyrase concentration in the enzyme mem~ brane produced no corresponding increase in the sensor sensitivity. The detection limit for Electrode II is about an order of magnitude higher than the limit for the adenosine sensor described in Chapter V. This is as expected. since the ATP molecules must undergo three separate enzyme re- actions to become adenosine. During these steps. which are relatively time comsuming. many of the intermediate molecules can diffuse through the dialysis membrane back to the solution. and are thus unavailable for the next enzyme reaction. Effect of pH on Electrode Sensitivity. As described in Chapter V. the concentration of am— monia and the enzyme activity depend on the pH of the solution. Like adenosine deaminase. potato apyrase is active over a very wide pH range [36]. The pH of maximum activity is observed near pH 6.5. The activities at 105 4O " >_ .. E 0.? <3 20 '- O l L l I I 7.5 8.5 9.5 pH 10;? M ATP in 0.0 37 C. 5 M Tris-H01 buffer With 0.03% 0a++: Figure 17. Response of ATP Electrode as a Function of pH. 106 pH 8.0. 9.0. and 9.5. are about half that at pH 6.5. The sensor sensitivity. measured as the difference in voltage between buffer and sample. rises rapidly with increasing pH. and achieves a maximum near pH 9.0 (Figure 17). which is also the pH of maximum sensitivity for the adenosine electrode. The situation is similar to that of the adenosine electrode (Chapter V). The potato apyrase used in the ATP electrode has about half the activity at pH 9.0 that it has at pH 6.5. However. this is more than offset by the increased ratio of ammonia to ammonium in the more basic solution. Selectivity Study. Potato apyrase is known to activate the dephos- phorylation of ITP. as well as ATP C36. 783; and alkaline phosphatase is active to any nucleoside monophosphate. Thus inosine may be produced at the membrane when the sen- sor is placed in a solution contains ITP. Since inosine is not susceptible to deamination by adenosine deaminase. it was expected that the ATP electrode would not be sen-’ sitive to ITP. To check this assumption. the reSponse of Electrode II to solutions containing up to 0.01 M ATP and ITP. was tested. The responses to these substrates are listed in Table 9. The sensor does not respond to ITP. even at the 0.01 M concentration. _. ...__..__ Table 9. Selectivity Studies for ATP Electrode II.a mV (mV buffer - mV solution) Substrate c°n:32£:iE:32;_T__-__-----f35-----~——~-----~~-5:5 ————————— 10 50 O 10 91 0 10 164 0 ++ Ca . pH 9.0. 37°C. a Tris-H01 buffer with 0.05% CHAPTER IX CONCLUSIONS AND FUTURE PROSPECTS The mayor accomplishments of this work were: 1) de- velopment of four enzyme immobilization methods. including one which may be practical for commercial production of large numbers of enzyme membranes; and 2) development of four enzyme electrodes. one each for adenosine. AMP. ADP. and ATP. by coupling up to four different enzyme reactions. It was also demonstrated that different im- mobilization procedures can produce immobilized enzymes with different chemical properties. e.g.. different pH profiles (Chapter IV. Section B). This suggests that further work be done to develop theories and methods for creating immobilized enzymes tailored to the needs of specific prOJects. It was also shown that stepwise enzyme reactions can be employed in an enzyme electrode with a relatively small increase in the detection limit and little loss of sensi- tivity. By carefully choosing the enzymes and operating conditions for the electrode. it is thus proven possible to couple several enzymes in order to expand the range of chemicals which can be detected by enzyme electrodes. The ability of the electrodes to measure the in— dividual concentrations of adenosine. AMP. ADP. and ATP 108 109 in a solution containing all four was not experimentally tested. Several effects may complicate the procedure. First. competition for the active sites of the im— mobilized enzymes will occur in solutions containing adenosine. AMP. ADP. and ATP. For example. adenosine produced from AMP by phosphatase on the AMP. ADP. or ATP electrode will compete with the adenosine originally in the solution for the deaminase on the electrode. If the concentrations of both adenosine and AMP are quite high. the production of ammonia from the enzymatic deamination of the adenosine in the solution might be reduced by the competition from adenosine produced at the membrane. To reduce the effect. the enzymes could be concentrated or purified. However. when the substrate concentrations are relatively low. this effect will be insignificant. Second. the rate at which a substrate diffuses through the membrane may be affected by the presence of other substrates. This would also change the response to the substrate. Again. this effect becomes insignificant at low substrate concentrations. Third. the concentration of dissolved species in the sample solution affects the partial pressure of dissolved ammonia gas. and thus affects the response of the ammonia electrode C79]. For example. the ratio of the partial pressure of the ammonia to the ammonia concentration in solution is 25% higher in a 1 M solution of NaCl than in 110 distilled water. Samples and standards should thus contain about the same level of dissolved species [79]. Sodium and potassium ions occur in significant concentra— tions in most biological samples. Since the enzyme elec- trodes are sensitive. the effect of these ions can be reduced by diluting the sample. Since dissolved salts increase the ammonia partial pressure. certain salts might prove useful for enhancing the response of enzyme elec— trodes employing ammonia-gas sensors. Obviously. the type and concentration of salt should be chosen so as not to inhibit the activity of the enzymes in the electrode. There are several other areas in which additional work may be done to improve the performance and con- venience of the enzyme electrode system. One is to design systems which allow several simultaneous determinations to be made conveniently. For example. a multiplexer could be used to automatically switch the potentiometer input be~ tween several electrodes. A more advanced system could monitor the rate at which the electrode potentials were changing. and indicate which electrodes had reached their steady-state potentials. Because of the large size of the ammonia electrodes now available commercially. the enzyme sensors developed in this project require 2 to 3 ml of solution for a de- 111 termination. Miniaturization of the ammonia electrode would allow smaller enzyme electrodes to be built. thus reducing the amount 0? sample solution required For a determination. Further work may be necessary For using the enzyme electrodes developed For measurements from tissue extracts and other biological samples. For example. heavy metal ions which are present in the sample can inhibit the en- zymes in the electrode. The addition a small quantity of EDTA to the buFFer solution is one technique For avoiding this. LIST OF REFERENCES 10. ll. 13. 14. LIST OF REFERENCES Rechnitz. Chem. and Eng. News. Jan. 27. 1975. '00 I I u A 2D "W J. Updike and G. P. Hicks. Nature. 214. 986 6 19 7). L. C. Clark. Jr.. in ”Enzyme Engineering." L. B. Ningard. Jr.. Ed.. Wiley. New York. 1972. pp. 377— 394. G. G. Guilbault. in "Enzyme Engineering." L. B. Ningard. Jr.. Ed.. Wiley. New York. 1972. pp. 361- 376. N. J. Blaedel. T. R. Kissel. and R. C. Boguslaski. Anal. Chem.. 5g. 2036 (1972). L. D. Bowers and P. N. Carr. Anal. Chem.. 1g. 545A (1976). G. G. Guilbault and J. G. Montalvo. J. Am. Chem. Soc.. 3g. 2533 (1970). T. Antalt. A. Graneli. and D. Jayner. Anal. Lett.. g. 969 (1973). E. H. Hansen and J. Ruzicka. Anal. Chim. Acta. 2g. 353 (1975). L. D. Nell and J. T. Maloy. Anal. Chem . AZ. 299 (19753. G. G. Guilbault and E. Hrabankova. Anal. Chem.. fig. 1779 (1970). ' G. G. Guilbault and E. Hrabankova. Anal. Chim. Acta. 56. 285 (1971). R. A. Llenado and G. A. Rechnitz. Anal. Chem.. £3. 1457 (1971). M. Mascini and A. Liberti. Anal. Chim. Acta. Qfi- 177 (1974). 112 15. 16. 17. 18. 19. 113 H. E. Booker and J. L. Haslam. Anal. Chem.. fi6. 1054 (1974). G. Nagy. L. H. Von Storp. G. G. Guilbault. Anal. Chim. Acta. 66. 443 (1973). G. G. Guilbault and F. R. Shu. Anal. Chim. Acta. 56. 333 (1971). G. J. Papariello. A. K. MukherJi. and C. M. Shearer. Anal. Chem.. fig. 790 (1973). L. F. Cullen. J. F. Rusling. A. SchleiFer. and G. J. Papariello. Anal. Chem.. fi6. 1955 (1974). G. G. Guilbault and F. R. Shu. Anal. Chem.. fifi. 2161 (1972). G. G. Guilbault. G. Nagy. and S. K. Kuan. Anal. Chim. Acta. 61. 195 (1973). N. Nilsson. A. Akerland. and K. Mosbach. Biochim. Biophys. Acta. 320. 529 (1973). D. S. Papastathopoulos and G. A. Rechnitz. Anal. Chem.. fig. 862 (1976). w. J. Blaedel and R. A. Jenkins. Anal. Chem.. 56. 1240 (1976). G. G. Guilbault and G. T. Lubrano. Anal. Chim. Acta. 60. 254 (1972). G. G. Guilbault and G. T. Lubrano. Anal. Chim. Acta. 64. 439 (1973). G. G. Guilbault and G. Nagy. Anal. Chim. Acta. _§. 69-80 (1975). G. G. Guilbault and G. T. Lubrano. Anal. Chim. Acta. 62. 189 (1974). G. G. Guilbault and G. T. Lubrano. Anal. Chim. Acta. 62. 183 (1974). M. NanJo and G. G. Guilbault. Anal. Chem.. fi6. 1769 (19743. H. M. Kalckar. J. Biol. Chem.. 1 7. 445 (1947). A. Kornberg and w. E. Prieer. Jr.. J. Biol. Chem.. 193. 481 (1951). 33. 34. 35. 36. 37. 38. 39. 40. 41. 44. 45. 4s. 47. 4e. 49. so. 51. 114 G. Schmidt. Z. Physiol. Chem.. 1 9. 243 (1928). H. M. Kalckar. J. Biol. Chem.. 148. 127 (1943). S. P. Colowick and H. M. Kalckar. J. Biol. Chem . 148. 117 (1943). H. M. Kalckar. J. Biol. Chem.. 153. 355 (1944). ~ -v H. M. Kalckar. J. Biol. Chem.. 161. 461 (1947). H. U. Bergmeyer. Ed.. "Methods of Enzymatic Analysis." Academic Press. New York. 1963. R. N. Estabrook and P. K. Maitra. Anal. Biochem.. 3. 368 (1962). N. D. McElroy. Proc. Nat. Acad. Sci. U. S.. 33. 342 (1947). J. N. Hastings. Ann. Rev. Biochem.. 32. 603 (1968). N. D. McElroy. J. N. Hastings. J. Coulombre. and V. SonnenFeld. Arch. Biochem. and Biophys.. fi6. 399 (1953). 4 B. L. Strehler and J. R. Totler. Arch. Biochem. and Biophys.. 49. 2B (1952). - L. M. Aledort. R. Need. and S. B. Troup. Anal. Biochem.. 61. 268 (1966). C. Tran-Minh and G. Brown. Anal. Chem.. fiZ. 1359 (1975). D. N. Gray. M. H. Keyes. and B. watson. Anal. Chem.. fig. 1067A (1977). R. Haynes and K. A. Nalsh. Biochem. Biophys. Res. COCflfflUfl.) 1.3—6"; 235 (1969). J. N. Nelson and E. G. Griffin. J. Amer. Chem. Soc.. 33. 1109 (1916). N. GrubhoFer and L. Schlecth. Heppe Seyler’s Z. Physiol. Chem.. 297. 108 (1954). N. GrubhoFer and L. Schlecth. Naturwissenschaften. 40. 508 (1953). H. H. Weetall. Anal. Chem.. 46. 602A (1974). 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 64. 65. 66. 67. 68. 69. 70. 115 O. Zaborsky. "Immobilized Enzymes." CRC Press. Cleveland. Ohio. 1973. G. J. H. Melrose. Rev. Pure Appl. Chem.. 2;. 83 (1971). K. J. Skinner. Chem. and Eng. News. Aug. 18. 1975. G. R. Stark. "Biochemical Aspects of Reactions on Solid Supports." Academic Press. New York. 1971. H. H. Neetall. Separ. Puritication Methods. 2. 199 (1973). A. D. McLaren and L. Packer. Advan. Enzymol. Relat. Areas Mol. Biol.. 33. 245 (1970) M. L. Green and G. CrutchField. Biochem. J.. 115. 183 (1969) Langmuir and V. J. SchalFer. J. Amer. Chem. Soc.. . 1351 (1938). L 29.. c. A. Zittle. Advan. Enzymol.. Relat. Areas. Mol. Biol.. 11. 319 (1953). R. A. Messing. J. Amer. Chem. Soc.. 2;. 2370 (1970). R. A. Messing. Enzymologia. 33. 370 (1970). N. GaFField. Y. Tomimatsu. A. C. Olson. and E. F. Jansen. Arch. Biochem. Biophys.. 157. 405 (1973). E. K. Bauman. L. H. Goodson. G. G. Guilbault. and D. N. Kramer. Anal. Chem.. 31. 1378 (1965). H. Ozawa. J. Biochem.. 62. 419 (1967). A. Bar-Eli and E. Katchalski. J. Biol. Chem.. 2 S. 1690 (1963). A. Bar—Eli and E. Katchalski. Nature. 188. 856 (1960). L. Goldstein. Y. Levin. and E. Katchalski. Biochemistry. 3. 1913 (1964). H. E. Hornby. M. D. Lilly. and E. M. Crook. Biochem. J.. 107. 669 (1968). D. S. Papastathopoulos and G. A. Rechnitz. Anal. Chim. Acta. Z2. 17 (1975). 71. 73. 74. 75. 76. 77. 78. 79. 116 J. N. Butler. "Ionic Equilibrium. a Mathematical Approach." Addison-wesley. Reading. Mass.. 1964. p. 132. G. NikiFeruk and S. P. Colowick. Methods in Enzymology. Volume II. 472 (1955). L. Mode and S. A. Kuby. J. Biol. Chem.. 226. 541 (1957). L. Node. J. Biol. Chem.. 232. 237 (1958). T. A. Mahowald. E. A. Noltmann. and S. A. Kuby. J. Biol. Chem.. 237. 1535 (1962). L. Noda and S. A. Kuby. Methods in Enzymology. Volume VI. 223 (1963). D. T. Plummer. "An Introduction to Practical Biochemistry.“ McGraw-Hill. London. 1971. p. 293. P. S. Krishnan. Methods in Enzymology. Volume II. 591 (1955). Orion Research. "Instruction Manual 09 Ammonia Electrode Model 95-10." Orion Research. Inc.. Cambridge. Mass.. 1970. p. 24. «m ICH‘ IN RUNI . IBRAR IES IIHIHHIHHIllkHINIIIHHIIHllHllHHHIHNIWIH!\I 31293100632847