Structural Preferences for the Binding of Chromium Nucleotides by

Transcription

1 Structural Preferences for the Binding of Chromium Nucleotides by Beef Heart Mitochondrial ATPase* Mary J. Bossard and Sheldon M. SchusterS (Received for publication, August 12, 1980, and in revised form, December 30, 1980) From the Department of Chemistry and School of Life Sciences, University of Nebraska-Lincola, Lincoln, Nebraska The mono- and bidentate forms of adenosine V-di- The effectiveness and suitability of chromium nucleotides phosphate, chromium (III) salt (CrADP) were separated as probes to examine enzyme kinetic mechanisms is well using Sephadex G-10 column chromatography. The iso- established (2-5). Cleland s laboratory has tested chromium meric punty of the two forms was monitored using nucleotides and chromium phosphates with several kinases. high voltage electrophoresis and column chromatog- The enzymes displayed sirnfiar specificity for the base, but raphy. The same techniques were employed to assess showed different specificities towards the various geometrical the purity of the mono-, bi-, and tridentate forms of and optical isomers of the chromium nucleotides (3, 6, 7). By adenosine 5 -triphosphate, chromium (III)salt (CrATP). characterizing those isomers of the chromium nucleotides that Distinct differences in the interaction of beef heart were the most effective enzyme inhibitors, Cornelius and mitochondrial ATPase with the various isomers of CIeland (6) and Dunaway-Mariano and Cleland (7, 8) were chromium nucleotides were seen inkineticstudies. able to deduce preferred substrate and product conformations Monodentate CrADP was a competitive inhibitor of the for several kinases. Although it has been established that the ATP bydrolysis activity of both purified ATPase and Mg-nucleotide complex is the active substrate form for FI submitochondrial particles. However, when ltpase ac- tivity was examined,noncompetitiveinhibitionwas observed. The bidentate isomer of CrADP did not affect ATPase activity. Enzymatic synthesis of the transition state analog of ATP synthesis and hydrolysis, Pi- CrADP occurred exclusively with the monodentate isomer of CrADP. It was also found that only the monoand tridentate forms of CrATP were potent inhibitors of ATP hydrolysis by beef heart mitochondrial ATPase. These results are discussed in terms of possible ATP synthesis and hydrolysis mechanisms. The naturally occurring substrate utilized by beef heart mitochondrial ATPase is ATP in the presence of magnesium ion. Formation of a MgATP complex is required prior to catalytic activity (1). Several possible substrate geometries exist since one, two, or three of the phosphate groups could potentially coordinate to the metal. The particular form of the MgATP complex used by Fl is, however, unknown. Magnesium exchanges oxygen ligands so rapidly that it is dbcult; to study the role of the coordinated metal ion in the catalytic mechanism of ATP synthesis or hydrolysis. This experimental difficulty, however, is not shared by Cr(Il1)- nucleotide complexes. Cr(1II) has an exchange rate with oxygen ligands so much slower than magnesium that it forms an essentially stable analog of the metal-nucleotide complex (2). Therefore, the Cr(l1I) nucleotides can conceivably be used in the study of the metal-copplexed nucleotide as it interacts with-the enzyme active center. * This work was supported by National Science Foundation Grant PCM The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $Supported by Research Career Development Award lk04ca from the National Cancer Institute, Department of Health. Education, and Welfare. The abbreviations used are: F1, beef heart mitochondrial ATPase; BES, 2-[bis(2-hydroxyeth~yll)aminoJethanesu~fonic acid; MKS, 2(Nmorphilino)ethanesulfonic acid; CrADP, adenosine 5 diphosphate, chron~iu~n (111) salt; CrATP, adenosine 5 triphosphatq chromium (111) salt; P,-CrADP, adenosine 5 -diphosphate, chromium (111) salt with P, replacing one of the water ligands activity, nothing is known concerning the conformation of metal-nucleotide complex bound t,o the enzyme. The preparation of chromium nucleotides involves heating a chromium salt with the desired ligand followed by purification and concentration on an ion-exchange resin. According to the published procedure, the Length of the heating period determines which isomer predominates. Unfortunately, this synthesis praduces a significant amount of contamination by the alternate isomer which is not completely separated during the published purification st.ep utilizing ion exchange chromatography (2, 3, 8). Currently, a procedure is available to separate the various isomers of CrADP and CrATP using a cycloheptylamylose gel (6, 8). However, the preparation of this gel is costly and tedious, Because of this problem, we have devised an alternate isolation procedure for these nucleotides that utilized commercially available Sephadex. This method of separating and monitoring the geometrical isomers of CrADP or CrATP presented is comparatively easy, rapid, and inexpensive. However, it does not resolve optical isomers. Kinetic evidence is presented which shows monodentate &ADP is a more potent inhibitor of the hydrolysis reaction of both membrane-bound ATPase and soluble F, from beef heart mitochondria. This specificity was unchanged whether ATP or ITP was used as substrate. In addition, synthesis of Pi-CrADP by FI occurs only with monodentate CrADP, A binding preference for the mono- and tridentate forms of CrATP is also demonstrated. MATERIALS AND METHODS2 Beef heart nrtahondrral ATPase (FIJ was lsolatea by the method of Spltsberg and Blair (18). Submrtocbondnal particles *ere pwpared occordmg to Xnnles and Pmefsky (19). Protein ConcentratlDnS &we measured by the Biuret procedure (20). Portions of this paper (including Materials and Methods, Figs. 1 to 8, and Tables 1-11) are presented in miniprint as prepared by the authors. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD Request Document No. 80M-1710, cite authors, and include a check or money order for $6.80 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

2 6618 Binding of Chromium Nucleotides to f The esay buffer used in the initial Velocity experiments contalnea a flnal COncentratlOn Of 100 rdl ~LSIKOH. 1W M sucrese. 5 rdl P!.$l2. 2 M phosphoenol pyruvate. 3 n)( NADH (4 M *en ITP was the rubstrrte). 1 nll KCIS and 16 U g h l (20 pg/ml uith ITP as substrate) each In lactate dehydrogenase and pyruvate kinase. The F1 was centrifuged at 4% to remove (NH412S0, and ATP present in the storage buffer. TheATPase was resuspended In a buffer of 100 M sucrose and 10 rdl TrrsICl ph 8.0 at v m tenperatwe. Substrate and inhibltovs were added as indicated rn the figurer. At the Concentrations Of chrmium nucleotides used there was no effect on the coupling system. Also an equivalent volm of 0.1 M N& used to store the chrmim wcleotides was added to the reaction buffer to ensure it had no effect on the Coupling system. Reactionr (1.0 m1 flnal volune) were imtiated by the addltlon of F1 (0.025 mg) or sutaito- chondrial partlcler (0.lY mg). Al klnetlcs.ere done at 3O0C and ph 7.0. The reactions were monitored by the disappearance of absorbance due to NMH at 340 MI. Under the initial velocity reaction conditlonr used. no slgmficant ISMR~ intermnverslons Of the Cr-nucleotide5 was Observed. Hexok7nase activlty was lnearured as per Dunway-Marlano and Cleland (?I. Coupltng enrpes. nvcleotldes and yeast herc$inare type 111 were purchased frm Sryna. Al Other chemicals Mstained frm cmmercial Iowces *ere Of the highest purlty available. Syntksw and Storage Of CrMP Isomers Wnodentate CrADP was ryntheslzed by the WthCd of Danenber and Cleland (3) WIth nsdifl- Cattons to enhance the purity of the Ismr. monodentate CrdP was prepared by heating a solution of CrCI) and ADP (20 nw flnal conc.) 1n a bolling water bath. Wen the tenperature reached 80 C the flask was removed and one minute later (tenperature dld not drop more than 3-4 degrees) rlpldly cooled in an ice bath. The 004r-M-X2-H column chrrmatoyraphr puriflcatton procedur; (3) was then followed. Hmever If it uas desired to have solutions of 20 n~ or greater. the flnal elution buffer used was 0.1 R Na-acetate at ph 5.2. A total starting volume of 50 ml CrADP adsorbed onto 15 ml of Dner 50-X2-H+ In a 20 cc sjrlnge easily produced nonodentate CrADP 1n 95% or better lsmrlc purlty. The total synthesis and elutlon procedure did not exceed tm hours. mbnodentate CrMP used for kinetic stvd-es was uruallr used within th&e-days-of ynthens before conslderable Conversion to bidentate CrMP occurred. Conversion to the bidentate form uas retarded by storage at ph At thls ph Bout 101. conversion to bidentate CrMP occurred ln a month. The bldentate form of CrMP appeared to be themdynmically most Itdble. Under canditions of mildly lhl ph and I h l ionic strength (0.1 I) NaAc). the monodentate form of CrMP converted to bidentate CrMP. Solutions of CrMP ph 5.2.ere alloed to age untll nearly 1WX conversron to the bldentate fom had occurred (3-4 reeks) and renalned stable for weeks. SYntheS11 of CrATP ISMRFS. All three chelate 1sm~s of CrATP ere prepared by the procedures of Dunway-Marian0 and Cleland (8). lrmr purity was monitored using a colwn (Mo X 2.2 ca) of Sephadex 6-10 and high voltage electrophoresis as indlcated in Figure 1. FRACTION NO. Flgure 1: Separation of mow- and bldentate lsornerr of CrAOP. A 2.2 x 55 cm column of Sephadex G-10 was used to determlne relatlve amounts of mono- and bldentate CrADP 110Mrs. The fwst peak correrpands to bldentate CrAOP and the second peak Is monodentate CrbDP. RESULTS The original isomeric assignment of the mono- and bidentate forms of CrADP were from Cleland s observation that CrADP produced by a 1-min heating of reactants yielded product which effectively inhibited yeast hexokinase (monodentate CrADP), whereas the product of a 10-min heating process (bidentate CrADP) did not. Although each process produced a mixture of the two forms, the synthesis and purification procedures were adequate for some experimental purposes. When CrATP was first synthesized, attempts to separate unreacted ATP from CrATP were done using Sephadex G-10. As reported earlier (2), the CrATP peak was unsymmetrical. This trailing of the peak was attributed to nonspecific adsorption of CrATP to the Sephadex since separation using Bio- Gel resulted in a symmetrical peak (2). We have found that the adsorption phenomena can be used to resolve geometrical isomers of CrADP and CrATP. The resolution and separation of mono- and bidentate isomers of Cr-nucleotides was achieved by using a long column containing Sephadex G-10 as described under Materials and Methods. Fig. 1 illustrates the absorption elution profie obtained from a mixture of the two chelate isomers of CrADP. Heating a solution ofcrc13 and ADP for 1 min at 80 Cyielded predominately a substance eluting from the column corresponding to the second peak in the elution profie. This compound inhibited yeast hexokinase (Fig. 2) at ph 5.8 (7) and is therefore designated monodentate CrADP. The relative area of the first peak in Fig. 1 increased as the heating time lengthened. The compound contained in this fist peak did not inhibit hexokinase (Fig. 2) under the prescribed assay conditions and is therefore designated bidentate CrADP. The fact that CrAMP elutes from the column with one symmetrical peak (data not shown) supports the notion that these two peaks represent chelation isomers of CrADP, and not polymerization products. That the bidentate form of CrADP eluted before the monodentate form (which might be expected to be less tightly coiled and thus have a larger Stokes radius) is reasonable. Due to the large degree of cross-linking of Sephadex G-10, the WP ADP ATP Ci P IMnO-CTMP bj-crmp Pi-CrAOP mno-cratp bi-cratp trl-cratp Eldentate CrADP distance leparte ** Mlgratlon in cm frm Origln Capound ph 2.5 (3000Y) rt 7.0 The bl- and tndentate ~smrl of CrATP cannot be drstlnguisbd under these conditions, however the tridentate fom usually erhibits a snail additional spot *ere PiCrMP migrates. Also bidentate CrMP and CrATP migvate about the saw distance. Hnrrer. bl- CrATP fom a concise spot hereas bi-crmp always stveaks. High voltaqe electrophoretic characterizatmn of the varfo~s nucleotides was done dl follols. Sanples of nucleotides 5 1 at a t r n e were spotted mu dried on Yhstmm 13 paper 47 K 57 m. The pwer.a; then wet with ihe electrophoresis medium and bloftcd prior to imrsion.ln the tank. This mediva contained 20 ml formic acld (901) and 80 ml glacial Metit acld per 2 liters of Mater with a resultant ph Of 2.5. Sbwles were subjected to 3,030 volts for 45 mn. The tewrature of the Varral cooled tank stwted at 15% and did not exceed 22OC. The initial Current was around 100 ma and did not exceed 2W ma. After the paper dned, cqounds =re located by thelr UV abrorptlon.

3 Binding of Chromium Nucleotides to FI 6619 effect of separation based on the Stokes radius is outweighed by adsorption and charge effects when the ionic strength is less than 0.02 (10). Using 50 mm acetic acid as a solvent permits the small number of carboxyl groups on the Sephadex beads to interact effectively with the positively charged chromium nucleotides. Thus, the more positively charged monodentate CrADP elutes after the bidentate form. The designation of the isomeric assignment corresponds to previously pubiished assignments (3,7) and has been verified using high voltage electrophoresis at ph 2.5. As would be expected under these conditions, the monodentate form of CrADP moved further towards the cathode (Table I). Sephadex chromatography was also used for batch separation of concentrated mixtures of the CrADP isomers. Fractions containing individual isomers were recombined and concentrated on a small column containing Dowex 50. Fractions containing appreciable amounts of both isomers were combined, concentrated, and chromatographed using Sephadex G-10. A major advantage of the Sephadex G-10 chromatography procedure was the ability to monitor the isomeric purity of the CrADP solutions obtained by different synthetic or storage procedures. Monodentate CrADP, prepared by a 1-min heating period, synthesized, and purified on Dowex 50 in 2 h, could be obtained in 95% or better (according to relative areas under the absorbance elution profiles) purity (see Fig. 1). After 3 days at ph 5.2, however, a significant portion had converted to bidentate CrADP (Fig. 1). Monodentate CrADP used in kinetic studies was either freshly synthesized or freshly separated by Sephadex chromatography. The conversion of monoto bidentate CrADP could be retarded by storage in 0.1 M sodium acetate, ph Bidentate CrADP was obtained in high isomeric punty by conversion of the monodentate isomer. The bidentate isomer was thermodynamically favored under mildly acidic conditions at low ionic strength. Fig. 3 illustrated this conversion. Monodentate CrADP was synthesized and eluted from a column containing Dowex 50 with 0.1 M MES/KOH, ph 5.6. This peak was collected and a small aliquot was chromatographed using Sephadex G-10 yielding the absorbance elution profiie seen in Fig. 3A. This shows primarily the monodentate form of CrADP. Fig. 3B is the absorbance profie 1 day, C is 6 days, and D is 14 days after synthesis and elution from Dowex 50. After 3 to 4 weeks, total conversion to bidentate CrADP was achieved. Isolated beef heart mitochondrial ATPase catalyzes the FRACTION NO I IATP IrnM)" YM Mono-CrADP-., llat P [mm] " hydrolysis of ATP to ADP and inorganic phosphate. The hydrolysis of ATP exhibits negative cooperativity in the absence of activating anions (11). The presence of the inhibitor CrADP amplified the cooperativity effect illustrated in Fig. 4A. The curvature of the lines made it difficult, if not impossible, to obtain a value for K,. Nearly equal concentrations of mono- and bidentate forms of CrADP were used as inhibitors. The slight inhibition by bidentate CrADP may be partly due to a small amount of contaminating monodenate isomer. Monodentate CrADP was clearly a more potent inhibitor (Fig 4A). In the presence of an activating anion (20 mm bicarbon- ate), the cooperativity effect was abolished (Fig. 4B), but again, the monodentate isomer effected greater competitive inhibition. The small amount of inhibition by bidentate CrADP relative to the control indicated this isomer does not bind F1 very well. As is shown in Fig. 5, submitochondrial particle ATPase was inhibited by monodentate CrADP and displayed negative cooperativity with ATP as a substrate. Bidentate CrADP was not an effective inhibitor in submitochondrial particles (data not shown). F1-ATPase also hydrolyzed ITP but displayed no cooperativity effect. Again, bidentate CrADP (Fig. 6) was much less effective as an inhibitor than the monodentate isomer when ITP was the substrate. The bidentate isomer was such a poor inhibitor (Fig. 6) no attempt was made to determine a Ki

4 ~~ 6620 Binding of Chromium Nucleotides to Fl /ATP (mw" Flgure 5: Monodentate CrADP lnhlbltlnn of submltochondrlal ATPase actlvlty. Negative coaperatlvlty was observed using monodentate CrADP as an Inhlbltor of ATP hydrolysis I" submltochandrlal partlcler. The reactlon condltlons were no CrAOP present y and CrADP concentritlons as follows: 130 vm &, 260 UM Q, and 515 UM (0). reacted equally with both the free enzyme and the enzymesubstrate complex. F1 produces a transition state analog of ATP synthesis, Pi- CrADP, from CrADP and inorganic phosphate. This same compound is also produced by hydrolysis of bidentate CrATP (9). As is illustrated in Fig. 8 by incorporation of free radioactive Pi, 32Pi-CrADP was produced only from monodentate CrADP. Although not shown, substitution of bovine serum albumin for F1, omission of F1, and substitution of MgADP for CrADP in the incubation mixture all gave essentially the same results as bidentate CrADP which failed to incorporate free "Pi. The three geometrical forms (mono-, bi-, and tridentate) of CrATP were resolved using a longer column containing Sephadex G-10 (2 meters) than that needed for the diphosphate compounds. Again, elution profiies were correlated to synthesis procedures and high voltage electrophoresis. The high voltage conditions employed clearly distinguish between the mono- and bidentate forms of CrATP (Table 1). Although the high voltage conditions reported did not clearly resolve the bi- and tridentate forms of CrATP, these were readily observed by monitoring the absorbance elution profile from Sephadex G-10 chromatography. This column, however, takes -18 h for complete sample elution and was used only to check isomer purity. The tridentate form eluted first, the bidentate form second, and monodentate CrATP eluted last (data not shown). These assignments agree with hexokinase inhibition behavior (3). It was observed that each preparation of monodentate CrATP and tridentate CrATP contained significant quantities of bidentate CrATP. Bidentate CrATP was prepared with essentially no contamination by the other isomers. Kinetic studies were done using all three forms, and linear competitive inhibition resulted. Inhibition due to contaminating bidenate CrATP in the mono- and tridentate CrATP preparations was accounted for using the. equation for two mutually exclusive competitive inhibitors (12), as described in Table 11. The Ki values for the three forms of CrATP as ATP hydrolysis inhibitors are shown in Table 11. Flgure 6: Inhlbltlon of Fl-ITPare actlvlty by CrADP. Manodentate CrADP 1s a much more potent lnhlbltor Of ITP hydrolyslr than bldentate CrADP. CrADP concentpatlons were as Indicated relative to the control where no inhibitor was present. I A (I) 30 I I liitpl.uil Figure 7: F,-ITPase activity Inhlb?tlon by mono-cradp. Mowo-CrADPi,M.6'! : A - Monodentate CrADP is a noncampet?t,ve lnhibltor of ITP hydrolysis. concentrations are as indicated. Mano-CrADP -8: Replot of Slope and Intercept. 80th the replot of slope and intercept versus lip COnCentPatlOn lndlcate a K, of 220 um. value. In Fig. 7A, micromolar concentrations of monodentate CrADP produced classical noncompetitive inhibition. A replot of slope and intercept versus inhibitor concentration (Fig. 7B) both showed a Ki of 220 PM, indicating monodentate CrADP FRACTION NO. Figure 8: Pi-CrADP fornatlan only by mono-cradp. When CrADP and 32P, were Incubated wth F1, P,-CrAOP was synthesized frm the monodentate ~somer (0) and not from the bldentate form ( I. The enzyne catalyzed formation Of PI-CrADP was accomplished by the procedure of Bossard et al. (9). However 50 mm 32P, (210 cpm/nmol) flnal COnCentrltlOn was used 3s the buffer and 0.5 ng F1 was used. Fractions of 2.2 ml were collected fmm the 2.5 cm x 1.0 rn column Containing Sephadex G-10. Allquots of 1.0 ml were counted I" water.

5 Binding of Chromium Nucleotides to Fl 6621 DISCUSSION Differences in the kinetic behavior of enzymes toward the various substrate geometrical isomers lead to two important consequences. First, it becomes imperative to work with pure solutions or be able to correct for contaminants. Secondly, the geometrical specificity can provide important clues to the natural substrate geometry. However, due to multiple synthesis products and isomer interconversion, it was essential for further research to be able to check rapidly the purity of the mono-, bi-, and tridentate forms of chromium nucleotides. Two independent techniques have been employed for this purpose. Using the high voltage electrophoresis system, as many as 10 samples can be simultaneously and quantitatively analyzed by the relative UV absorbance intensities of the spots in -2 h. At ph 2.5, monodentate CrADP has one more positive charge than the bidentate form, allowing the two to be clearly distinguishable. Any free ADP resulting from CrADP decomposition could also be concurrently detected. The absorbance profiie of the eluant from the Sephadex C- 10 chromatography did not distinguish ADP from bidentate- CrADP. However, this second method provided a more quantitative analysis of the relative percentages of the two geo- metrical forms of CrADP. The chromatography was also useful for batch separation of the two CrADP forms from concentrated solutions containing a mixture of isomers. It was observed that the synthesis method of heating CrC13 and ADP for 1 min (3) did not always produce the same relative percentages of mono- and bidentate CrADP. It was also noted that fractions of mono-cradp eluted from Dowex 50 converted to the bidentate form at different rates, depending upon the ph of the resultant solution. Further work is needed to determine the optimum conditions for stabilization of mono-cradp. Similar variations occurred in per cent yield of desired isomers obtained from CrATP synthesis attempts (8). Although we were able to purify bidentate CrATP, the monoand tridentate preparations contained varying amounts of contaminating bidentate CrATP after elution from Dowex. The 2-m Sephadex chromatography, however, permits quan- titation of the two components so that the necessary concentration corrections can be made. The marked preference of both isolated FI and submitochondrial particle-bound Fi for monodentate CrADP binding is significant. The tight binding of monodentate CrADP and synthesis of Pi-CrADP from only the monodentate isomer indicated that the natural substrate is probably monodentate MgADP. The ratio of monodentate MgADP to bidentate MgADP in solution has been estimated to be 1:27 (7). Thus, it is possible that when ATP synthesis occurs, the least abundant form of ADP is utilized and the remaining bidentate ADP will not cause substrate inhibition. It is also interesting to note that both mono- and tridentate CrATP bind tightly to F1, but the bidentate form (again the most abundant) does not, Again, extrapolating to the native system where the primary function of ATPase is to synthesize ATP, the probable product released is bidentate MgATP. As the relatively high K, of bi-cratp indicates, the bidentate MgATP released would not be likely to rebind to FI, preventing possible product inhibition. It has been previously proposed that mitochondrial ATPase contains two types of nucleotide binding sites (4, 13). A regulatory site is thought to bind specifically adenine nucleotides. The catalytic site, however, is nonspecific and hydrolyzes other nucleoside triphosphates as well. Both mono- CrADP and all three forms of CrATP were competitive inhibitors of ATP hydrolysis (Figs. 4 and 5) (Table 11). In Figs. 6 and 7, however, noncompetitive inhibition of ITP hydrolysis by mono-cradp was observed. This is consistent with the kinetic studies presented earlier (4, 11). It would appear that ITP cannot compete with CrADP for binding to the F, regulatory site, resulting in the observed noncompetitive inhibition. These data are difficult to reconcile with the proposed alternating site models proposed by Boyer and his colleagues (13-15). It must be reemphasized that the methods of nucleotide separation and identification discussed here do not resolve optical isomers: Optical resolution can be achieved (6, 8), but is not required for all types of mechanistic studies. Carrently, other groups are using chromium nucleotides prepared by the heating procedures and comparing the products by inhibition of yeast hexokinase (16, 17). We feel we have found a simpler method for routine analysis of these compounds. If the binding of chromium nucleotides to F1 is a reasonable model system to extrapolate to the naturally occurring Mgnucleotide system, an interesting correlation results. Monodentate MgATP might be the ATP hydrolysis substrate and bidentate MgATP could be the ATP synthesis product. Since the equilibrium favors bidentate MgATP formation free in solution, the process of ATP synthesis would not be inhibited by the MgATP produced in the mitochondria. This could explain how ATP synthesis can occur in the presence of a substantial mitochondrial phosphate potential. Acknowledgments-We wish to thank Drs. D. Dunaway-Mariano and W. W. Cleland for preprints of their work. REFERENCES 1. Pullman, M. E., Penefsky, H. S., Datta, A,, and Racker, E. (1960) J. Biol. Chem., 235, DePamphilis, M. L., and Cleland, W. W. (1973) Biochemistry, 12, Danenberg, K. D., and Cleland, W. W. (1975) Biochemistry, 14, Schuster, S. M., Ebel, R. E., and Lardy, H. A. (1975) Arch. Biochem. Bzophys. 171, yip, B. P., and Rudolph, F. B. (1977) J. Biol. Chem. 252,

6 6622 Binding of Chromium Nucleotides to Fl 6. Cornelius, R. D., and Cleland, W. W. (1978) Biochemistry 17, Dunaway-Mariano, D., and Cleland, W. W. (1980) Biochemistry 19, Dunaway-Mariano, D. and Cleland, W. W. (1980) Biochemistry 19, Bossard, M. J., Vik, T. A,, and Schuster, S. M. (1980) J. Biol. Chem. 255, Gel Filtration Theory and Practice (1980) Pharmacia Fine Chemicals, Uppsala, Sweden 11. Schuster, S. M., Ebel, R. E., and Lardy, H. A. (1975) J. Biol. Chem. 250, Segel, I. H., (1975) Enzyme Kinetics p. 474, John Wiley & Sons, New York 13. Hackney, D. D., and Boyer, P. D. (1978) J. Biol. Chem. 253, Choate, G. L., Hutton, R. L., and Boyer, P. D. (1979) J. Biol. Chem. 254, Rosen, G., Gresser, M., Vinkler, C., and Boyer, P. D. (1979) J. Biol. Chem. 254, Yee, D., and Eckstein, F. (1980) FEBS Lett. 112, Gupta, R. K., and Benovic, J. L. (1978) J. Biol. Chem. 253, Spitsberg, V. L., and Blair, J. E. (1977) Biochim. Biophys. Acta 460, Knowles, A. F., and Penefsky, H. S. (1972) J. Biol. Chem. 247, Layne, E. (1957) Methods Enzymol. 3,