Suppression of creatine kinase-catalyzed phosphotransfer results in increased phosphoryl transfer by adenylate kinase in intact skeletal muscle.

The kinetics of creatine kinase (CK) and adenylate kinase (AK) activities were monitored in intact diaphragm muscle by 18O phosphoryl oxygen exchange to assess whether these two phosphotransferases provide an interrelated function integral to high energy phosphoryl metabolism. This possibility was examined by quantitating the net rates of CK- and AK-catalyzed phosphoryl transfer in comparison to the total cellular ATP metabolic rate when CK activity in the intact diaphragm muscle was progressively inhibited by 2,4-dinitrofluorobenzene. In noncontracting muscle from untreated rats, net rates of CK- and AK-catalyzed phosphotransfer were equivalent to 88 and 7%, respectively, of the total ATP metabolic rate. These results were compared with reported 31P NMR analyses of total creatine phosphate flux to estimate that each creatine phosphate molecule produced undergoes about 50 unidirectional CK-catalyzed phosphotransfers in transit to an ATP consumption site in the intact muscles. Graded inhibition by 2,4-dinitrofluorobenzene of intracellular CK activity by up to 98% resulted in a progressive shift in phosphotransferase catalysis from the CK to the AK system; the sum of the net rates of phosphoryl transfer by combining the increasing AK and decreasing CK activities continued to approximate the total cellular ATP metabolic rate. These results indicate that in diaphragm muscle CK and AK operate as interrelated cellular high energy phosphoryl transfer systems through which the majority of newly generated ATP is processed prior to its utilization.

The kinetics of creatine kinase (CK) and adenylate kinase (AK) activities were monitored in intact diaphragm muscle by 18 O phosphoryl oxygen exchange to assess whether these two phosphotransferases provide an interrelated function integral to high energy phosphoryl metabolism. This possibility was examined by quantitating the net rates of CK-and AK-catalyzed phosphoryl transfer in comparison to the total cellular ATP metabolic rate when CK activity in the intact diaphragm muscle was progressively inhibited by 2,4-dinitrofluorobenzene. In noncontracting muscle from untreated rats, net rates of CK-and AK-catalyzed phosphotransfer were equivalent to 88 and 7%, respectively, of the total ATP metabolic rate. These results were compared with reported 31 P NMR analyses of total creatine phosphate flux to estimate that each creatine phosphate molecule produced undergoes about 50 unidirectional CK-catalyzed phosphotransfers in transit to an ATP consumption site in the intact muscles. Graded inhibition by 2,4-dinitrofluorobenzene of intracellular CK activity by up to 98% resulted in a progressive shift in phosphotransferase catalysis from the CK to the AK system; the sum of the net rates of phosphoryl transfer by combining the increasing AK and decreasing CK activities continued to approximate the total cellular ATP metabolic rate. These results indicate that in diaphragm muscle CK and AK operate as interrelated cellular high energy phosphoryl transfer systems through which the majority of newly generated ATP is processed prior to its utilization.
The view put forth by Bessman and co-workers (1,2) that the metabolic function of creatine kinase (CK) 1 is to transfer energy-rich phosphoryls from intracellular generation to utilization sites is supported by kinetic studies in vitro and in vivo by the distinctive localization of isozymic forms of CK and their functional coupling to ATP-consuming and ATP-generating processes (for reviews see Refs. [3][4][5]. In opposition to this view is the relatively intact physiological performance of muscle when CK activity is impaired following protracted ingestion of creatine analogs (6,7) or deletion of the gene encoding the major cytosolic isoform of muscle CK (i.e., M-CK) (8). It is also con-tradicted by reports that the total unidirectional rate of CKcatalyzed transfer of phosphoryls assessed by 31 P NMR technology in intact skeletal muscle is workload-independent or even reduced with the onset of muscle contraction (9,10).
An importance in intracellular energy transfer similar to that of CK has been suggested for adenylate kinase (AK)catalyzed phosphotransfer (4,11,12). From studies of AK kinetic behavior in intact diaphragm muscle, increases in net AK-catalyzed phosphotransfer were found to be directly proportional to the frequency of stimulated muscle contraction (12), and the stoichiometry between net AK-catalyzed phosphoryl transfer and anaerobic glycolytic ATP generation was nearly equivalent over a greater than 20-fold range of stimulated fluxes (13). The interpretation of these results was that AK-catalyzed phosphotransfer functionally couples ATPconsuming processes with anaerobic glycolytic ATP generation (13).
These studies of the dynamic behavior of phosphotransferase activities in intact diaphragm muscle also indicated that CKand AK-catalyzed phosphoryl transfer may be functionally interrelated; marked increases in the rate of AK-catalyzed phosphotransfer occurred when CK catalytic velocity was suppressed in muscle deprived of oxygen (13). The interpretation that this represented a compensatory increase in the activity of one phosphotransferase to offset the impaired activity of another translated into CK and AK representing closely interrelated phosphotransfer systems integral to cellular energy metabolism. If so, it would help to explain at least in part how muscle performance and metabolic integrity remain relatively intact when CK activity is markedly suppressed (6, 7) or deleted (8) as described above.
The investigations reported here were intended to determine if and to what extent AK-catalyzed phosphotransfer can replace phosphoryl transfer catalyzed by CK when the latter activity becomes impaired. This was assessed by determining the quantitative relationship between the sum of the net rates of CK-and AK-catalyzed phosphotransfer compared with the total cellular high energy (i.e., ATP) metabolic turnover rate in intact diaphragm muscle. These quantitative assessments were made by 18 O phosphoryl oxygen exchange analysis, which measures only net enzyme-catalyzed phosphoryl fluxes (12)(13)(14)(15) in contrast to 31 P NMR analysis using saturation transfer which, measures total unidirectional phosphoryl exchange rates (16). These phosphoryl transfer rates were measured when CK catalysis was progressively suppressed by treatment of intact diaphragm muscle with the CK inhibitor, 2,4dinitrofluorobenzene (DNFB).
The results show that in noncontracting rat diaphragm muscle net CK-catalyzed phosphoryl transfer is equivalent to nearly 90%, whereas AK-catalyzed phosphotransfer accounts for about 7% of the total ATP metabolic flux. Impairment of intracellular CK activity by DNFB results in a quantitative shift in phosphotransferase catalysis from the CK to the AK system. As a result, the sum of the net transfer of phosphoryls catalyzed by the progressively suppressed CK activity and incrementally greater AK activity accounts for the processing of almost all of the cellular ATP undergoing metabolic turnover in rat diaphragm muscle.

EXPERIMENTAL PROCEDURES
Preparation and Incubation of Rat Diaphragms-Diaphragms were obtained from anesthetized (50 mg⅐kg Ϫ1 pentobarbital, intraperitoneally) male Sprague-Dawley rats (150 -200 g) fed standard rat chow ad libitum. The procedures for incubating, stimulating, and preparing the tissues for analysis of 18 O content of metabolite phosphoryls has been described in detail (12,14). Briefly, after excising the diaphragms, the muscle was dissected free of the rib cage, washed, and preincubated in physiological buffer (12) continuously bubbled with O 2 /CO 2 (95%/5%). The whole diaphragms were preincubated for 40 min in the oxygenated medium without stimulation in the absence or the presence of DNFB (50 -200 M) before they were transferred to medium enriched in 25-40 atom % [ 18 O]water (Monsanto Co. and Isotec) not containing DNFB and incubated for an additional 1, 2, or 4 min before they were freezeclamped with aluminum blocks cooled with liquid nitrogen. The frozen tissues were ground to a fine powder in liquid nitrogen and extracted with 3 M perchloric acid at Ϫ10°C. Small samples of powdered tissue were extracted with buffer containing 20 mM Tris-HCl, pH 8.0, 10 mM EGTA, 2 mM EDTA, 2 mM dithiothreitol, and 0.004% leupeptin supplemented with 0.1% Triton X-100 for assay of CK and AK activities. The details of procedures for purification of nucleotides, CrP, and P i and isolation of specific metabolite phosphoryls for analysis of 18 O content was carried out as described previously (12)(13)(14)(15).
Analysis of the 18 O Content of Phosphoryls-Following enzymatic transfer of the ␤and ␥-phosphoryls of ADP and ATP and phosphoryls of CrP (via ␥-ATP) to glycerol with glycerol kinase to generate glycerol 3-phosphate, the latter was converted to the trimethylsilyl derivative as described previously (12,14). The 18 O enrichment of the phosphoryls of ␥-ATP, ␤-ATP, ␤-ADP, CrP, and P i was accomplished using gas chromatography-mass spectrometry (Hewlett-Packard 5970B) to monitor mass ions (m/z) of 357, 359, 361, and 363 corresponding to phosphoryl species of 18 O 0 to 18 O 3 , respectively. The trimethylsilyl derivative of orthophosphate yielded mass ions (m/z) 299, 301, 303, 305, and 307 corresponding to species of orthophosphate containing from 0 to 4 atoms of 18 O.
Analysis of 18 O Phosphoryl Labeling Data-The net rate of AKcatalyzed phosphotransfer in the intact muscle was calculated from the rate of appearance of 18 O-containing ␤-phosphoryls in ADP and ATP (12). These ␤-phosphoryls derive from the ␥-phosphoryls of ATP, which undergo more rapid replacement with 18 O-labeled phosphoryl species. The in situ net velocity of AK catalysis is, therefore, a function of both the rate of AK-catalyzed phosphotransfer and the kinetics of [␥- 18 O]ATP appearance. The net CK-catalyzed phosphoryl transfer rate was determined by the rate of appearance of CrP species with 18 Olabeled phosphoryls. The phosphoryl of CrP also derives from ␥-ATP. To assure that the precursor ␥-ATP was more rapidly replaced with 18 Olabeled species of phosphoryl than CrP, DNFB-treated diaphragms in which CK-catalyzed phosphoryl transfer was diminished (while the rate of ␥-ATP labeling with 18 O was relatively unaffected) were used to create this circumstance (see "Results"). A computer model (Stella, High Performance Systems, Lyme, NH) of AK catalysis was used to estimate the net velocities of AK-catalyzed phosphorylation of AMP as described previously (12) from the rates of formation of [␤- 18 O]ADP and [␤-18 O]ATP, which could also be used to determine the rate of the reverse reaction in which AK catalyzes the generation of AMP and ATP from 2 ADP. A similar computer model (14) was used to estimate the net rates of CK-catalyzed phosphorylation of creatine (Cr) from the rates of formation of [ 18 O]CrP in DNFB-treated diaphragms. These progressively inhibited rates of CK catalysis were used to determine the uninhibited rate by extrapolation (see Fig. 5).
Determination of Total Cellular ATP Metabolic Flux-Total cellular ATP metabolic turnover was estimated from the sum of the total number of 18 O atoms appearing in the major phosphoryl-containing cellular metabolites and orthophosphate as described previously (12,14).
Other Methods-The concentrations of AMP, ADP, ATP, CrP, and lactate were determined in neutralized acid extracts of diaphragm muscle by coupled enzymatic analysis using fluorometric detection (18). CK and AK activities in buffer extracts of a portion of the frozen powdered tissue (see above) were determined spectrophotomerically as described previously (19). Protein was determined by the BCA method (Pierce). Enzymes and other chemicals were obtained from Sigma and Boehringer Mannheim.
Statistics-All values shown represent means Ϯ S.E. unless otherwise indicated. Differences between means were assessed using Student's t test.

Estimation of Net CK Velocity in Intact Rat
Diaphragm Muscle-A problem in determining accurate net CK-catalyzed phosphoryl transfer rates in the intact rat diaphragm muscle cells is posed by the very similar rates of 18 O-labeled phosphoryl appearance in CrP and in the ␥-position of ATP (14). Because the 18 (12,14).
To deal with this potential problem, a strategy was devised to reduced the rate of CrP metabolism to less than that of ␥-ATP by decrementally suppressing the activity of CK in the intact muscle with a relatively selective inhibitor of this enzyme and then determining the uninhibited velocity by extrapolation. An agent frequently used to inhibit CK activity in numerous types of cells is DNFB (17).
The effectiveness of DNFB to impair CK activity after treatment of intact rat diaphragm muscle with the inhibitor followed by cell disruption and in vitro assay of CK activity is shown in Fig. 1. Because the inhibitory action of DNFB is time-dependent, an optimal duration of tissue exposure was pretested and determined to be 40 min for the range of concentrations used (50 -200 M). When assessed in extracts of diaphragm muscle treated with 50, 100, or 200 M DNFB, CK activity was found to be inhibited by 49, 65, and 84%, respectively. This corresponds to previous reports (17) of the effectiveness of DNFB to inhibit CK activity in different types of cells by this ex vivo assessment. In the inset of Fig. 1 it is shown that the tissue concentrations of CrP and ATP were relatively unaffected by 50 or 100 M DNFB but that at 200 M the concentration of CrP was decreased by 10% and that of ATP was decreased by 37%.
In intact rat diaphragm muscle, the activity of CK determined by the rate of appearance of 18 O-labeled phosphoryls in endogenous CrP exhibited a similar sensitivity to the inhibitory action of DNFB (Fig. 2) as that shown by the ex vivo assessment (see Fig. 1). This inhibition of CK-catalyzed 18 O phosphoryl labeling of CrP was equivalent to 35, 75, and 98% suppression of the intracellular CK activity with 50, 100, and 200 M DNFB, respectively. The relative rate of [␥-18 O]ATP appearance was not diminished but rather increased, at least during the first 2 min of monitoring the DNFB treated muscles (Fig. 2, inset). This probably resulted from a disproportionate amount of the newly generated ␥-ATP not transferring its 18 O-labeled ␥-phosphoryl to Cr because the CK catalyzing this reaction was inhibited.
Reduction by DNFB of CrP turnover rate to a level substantially lower than that of ␥-ATP permitted estimation of net velocities of CK in intact muscle using computer modeling of the 18 O phosphoryl labeling kinetics (12,14). These calculated net catalytic rates were used to construct the plot in Fig. 3 showing the relationship between the inhibition of 18 O labeling of CrP versus computer-derived estimates of net CK velocities in intact rat diaphragm muscle. This relationship is very nearly linear, and the intercept on the abscissa, representing the uninhibited CK velocity in the intact muscle cells, was equal to 146 Ϯ 8 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 .
Relationship between CK-and AK-catalyzed Net Phosphotransfer Rates in DNFB-treated Rat Diaphragm Muscle-It had been reported previously (13) that in rat diaphragm muscle in which CK activity was found to be diminished when the tissue was oxygen-deprived, there was a reciprocal increase in net AK-catalyzed phosphoryl transfer rate. To study whether this occurs in muscles when CK activity is directly inhibited by DNFB treatment, the rate of AK-catalyzed 18 O appearance in the ␤-phosphoryls of ADP and ATP was examined in rat diaphragm muscle treated with the CK inhibitor DNFB. That the in situ activity of AK-catalyzed phosphotransfer increases when CK activity is inhibited by DNFB is clearly demonstrated in Fig. 4. The rates of AK-catalyzed appearance of 18 O-labeled species of ␤-ADP (as well as ␤-ATP, not shown) are enhanced incrementally with concentrations of DNFB that progressively inhibit CK activity in the intact muscle (Fig. 2).
In Fig. 5 the net rates of CK-and AK-catalyzed phosphotransfer as a function of DNFB concentration are compared, and the relationship of these phosphotransfer rates to the total cellular ATP flux are also shown. The total ATP turnover rate determined by total 18 O appearance in cellular metabolites (12,14) was equal to 166 Ϯ 12 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 in the muscle not treated with DNFB. This basal ATP metabolic flux, which also corresponds to ATP turnover rates estimated from oxygen consumption data (14), was reduced by 11, 20, and 39% with 50, 100, and 200 mM DNFB, respectively (Fig. 5).
The net CK velocity in the control muscle was shown in the earlier experiment (see Fig. 3) to be equal to 146 Ϯ 8 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 . This is equivalent to about 88% of the total ATP metabolic flux. In this basal state (i.e., noncontracting and without DNFB) the net phosphoryl transfer rate catalyzed by AK (i.e., conversion of ADP) was determined to be 12 Ϯ 0.6 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 , which was equivalent to approximately 7% of the total ATP flux; net phosphoryl transfer catalyzed by AK plus CK could therefore account for 95% of the total ATP metabolic flux in the control muscle. Treatment with increasing concentrations of DNFB resulted in decremental lowering of the net CK-catalyzed phosphotransfer velocity so that at 200 M DNFB, this catalytic activity declined approximately 98%. The response of the AK-phosphotransfer system was virtually the reciprocal of that exhibited by the CK system (Fig. 5). Treatment with increasing concentrations of DNFB led to incrementally greater rates of net AK-catalyzed phosphotransfer so that at 200 M DNFB the net AK velocity increased from 12 Ϯ 0.6 to 96 Ϯ 3.1 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 . The percentage of the total ATP flux attributable to phosphoryl transfer catalyzed by AK increased from 7% without DNFB to the equivalent of 19, 45, and 94% with 50, 100, and 200 M DNFB, respectively. CK-catalyzed phosphotransfer, on the  Fig. 2) were calculated by computer modeling (Stella, High Performance Systems, Lyme, NH) as described under "Experimental Procedures." Each mean value (ϮS.E.) of inhibited CK velocity was determined from the data collected at three times (1, 2, and 4 min) in duplicate diaphragms from a total of 18 rats. The percentage of inhibition (mean Ϯ S.E.) of 18 O phosphoryl labeling of CrP was plotted against these determined CK velocities to estimate by extrapolation the uninhibited CK velocity, which was equal to 146 Ϯ 8 nmol⅐min Ϫ1 ⅐mg protein Ϫ1 . The numbers in parentheses indicate the DNFB concentration in M. Other details are the same as described in Fig. 2. other hand, decreased from the equivalent of 88% in the non-DNFB-treated muscle to 65, 32, and 3% of total ATP turnover rate in muscle treated with 50, 100, and 200 M DNFB, respectively. When this apparent shift from CK-to AK-catalyzed phosphotransfer occurs with DNFB suppression of CK activity, the total phosphotransfer capability of AK and CK catalysis combined could be calculated to account for 84, 77, and 97% of the ATP metabolic flux with 50, 100, or 200 M DNFB, respec-tively. This shift in phosphotransfer function from CK to AK catalysis occurs with a progressive decline to a maximum of 39% in total ATP metabolic flux. DISCUSSION The results of the experimentation reported here provide quantitative information about the kinetic behavior and functional interrelationship of CK-and AK-catalyzed phosphotransfer in the environment of the architecturally intact cell cytosol. This new information supports the concept that phosphoryl transfer catalyzed by these two enzymes is integral to the management of cellular energy metabolism.
Stoichiometry between Net CK-catalyzed Phosphotransfer and Total Cellular ATP Metabolic Flux-The analysis of CKand AK-catalyzed phosphotransfer in noncontracting, intact muscle cells by 18 O phosphoryl oxygen exchange shows that the sum of the net fluxes catalyzed by CK (88%) and AK (7%) can account for most of the ATP metabolized in the noncontracting rat diaphragm muscle. This could be interpreted to indicate that in this basal state most all of the newly generated molecules of ATP are processed by either the CK or AK phosphotransfer system.
The CK catalytic rates determined here by 18 O phosphoryl oxygen exchange analysis differ by more than an order of magnitude with those determined by 31 P NMR saturation transfer technology (16,20). This difference undoubtedly stems from the different kinetic parameters quantified by these procedures. 31 P NMR analysis assesses total unidirectional phosphoryl exchange rates catalyzed by CK (16), whereas the 18 O phosphoryl oxygen exchange measurements (12-14) assess net phosphoryl flux (i.e., appearance of newly generated molecules of [ 18 O]CrP). A difference of the magnitude found between these total and net flux measurements is predictable if phosphoryl transfer by CK catalysis was accomplished, as previously suggested (13), by the mechanism of vectorial ligand conduction (21).
According to this mechanism this difference would represent the average number of CK-catalyzed equilibration reactions involved in the transport of a newly synthesized molecule of CrP from its site of generation to an ATP utilization site, where ADP is re-esterified to ATP by CrP. Total unidirectional CKcatalyzed phosphoryl flux measured by 31 P NMR saturation transfer technology has been reported (20) to be equivalent to 7,000 nmol⅐mg protein Ϫ1 ⅐min Ϫ1 in noncontracting rat quadriceps muscle. By 18 O phosphoryl oxygen exchange analysis the net CK-catalyzed phosphoryl flux in noncontracting rat diaphragm was 146 nmol⅐mg protein Ϫ1 ⅐min Ϫ1 . Therefore, a first approximation of the average number of CK-catalyzed equilibration reactions involved in the transfer of a newly synthesized molecule of CrP from its generation to utilization sites would be 7,000 Ϭ 146 or 48. This is a provisional value because no 31 P NMR saturation transfer measurements have been made with rat diaphragm for a direct comparison with the measurements made here. Measurements by 18 O phosphoryl exchange that we have made in rat gastrocnemius muscle 2 indicate that net CK-catalyzed phosphoryl flux is lower than in diaphragm, so the number of total transfers in the gastrocnemius would be greater than the 48 estimated here.
The basic construct of the previous (13) and current working models of the operation of the AK-and CK-catalyzed phosphoryl transfer systems defines their importance in coupling ATPconsuming and ATP-generating processes through rapid equilibration of reactants along a series of CK-or AK-catalyzed reactions. The latter is viewed as a prerequisite for the operation of a vectorial ligand conduction system (21) and is consist-  Fig. 2 was determined, and the experimental details are the same as described in Fig. 2. Each value shown is the mean of duplicate diaphragms analyzed in triplicate. No value varied by more than 5% from the mean. This is a representative experiment of two very similar experiments from which almost identical results were obtained. Very similar effects of DNFB were observed on 18 O labeling of ␤-ATP (not shown).
FIG. 5. Reciprocal changes in net CK-and AK-catalyzed phosphotransfer rates in relation to total ATP metabolic rate in intact noncontracting rat diaphragm muscle treated with DNFB. The uninhibited and inhibited net rates of CK-catalyzed phosphoryl transfer were derived from the data presented in Fig. 3. The net rates of AK-catalyzed phosphoryl transfer from the direction of ADP conversion were estimated from the time course of 18 O-labeled phosphoryl appearance in ␤-ADP (Fig. 4) and ␤-ATP (not shown) using computer modeling of the rate data as described under "Experimental Procedures" and previously (13).The cellular ATP metabolic rate was determined by summing the total 18 O appearing in the major phosphoryl-containing cellular metabolites as described previously (13). Other experimental details were the same as described in the legend of Fig. 2. Each value of net phophoryl transfer rate represents a mean derived from data collected from six individual tissues (i.e., two diaphragms at each time point) analyzed in triplicate. All of the CK and AK net velocity values for DNFB-treated tissues are significantly different from the non-DNFB-control value (p Ͻ 0.005). The ATP turnover rates for the muscles treated with 100 and 200 M DNFB were significantly different from the control value (p Ͻ 0.05). ent with the theoretical principles governing the behavior of near equilibrium reactions comprising intracellular flux transfer chains (26 -29). By this mechanism each phosphorylbearing molecule entering at one end of the chain of reactions promotes propagation of sequential equilibrations resulting in simultaneous release of an equivalent molecule at the distal end. Thus, at steady state, the time for accomplishing this should be negligible. The delay in increased muscle O 2 consumption noted during the "rest to work" transition (30) could represent the interval required for this series of equilibrating reactions to achieve the new steady state.
Additionally, this mechanism would also help to resolve the problem created by the abundant and very catalytically active myofibrillar and cytosolic CK and AK enzyme protein molecules with the potential to bind and catalytically transform ATPase-generated ADP before it could freely diffuse from its site of generation to an ATP regenerating site (22)(23)(24)(25). Another feature of this proposed conduction system underscoring its potential thermodynamic efficiency is its capability to operate with minimal or no concentration gradient of reactants (31)(32)(33). This could explain why sought after changes in cellular adenine nucleotide concentrations are most often not observed even with marked increases in metabolic flux (34).

Relationship and Plasticity of AK and CK Catalysis in Situ-
The results showing that chemical inhibition of CK activity by DNFB in intact diaphragm muscle leads to reciprocal increases in the rate of AK-catalyzed phosphotransfer are consistent with the view that these two enzyme activities share a comparable functional importance. On the other hand, the expression of CK in these muscle cells may have been intended to provide for a specific function that AK may not take over in toto. How phosphotransfer function is shifted from the CK to the AK system is not presently understood, but it appears to involve the recruitment of the activity of pre-existing AK enzyme protein. There is no detectable increase in the total activity of AK assayable in extracts of muscle obtained after DNFB treatment (not shown). This ostensibly rapid and efficient switch from CK-to AKcatalyzed phosphotransfer is consistent with the coexistence of AK and CK in discrete subcellular locales (4,22,23). This may indicate that when ADP from ATP hydrolysis cannot be processed by CK and accumulates to a small but critical localized concentration favoring the K m requirement of AK for ADP, AK takes over the metabolic processing of this ADP.
The effect of acute suppression of CK catalytic capability by an agent such as DNFB may not be identical to the outcome resulting from chronic impairment of CK activity achieved, for example, by gene deletion (8) or protracted feeding of creatine analogs (6,7). Chronic CK impairment may involve other adaptive, phosphotransfer-related changes not yet defined. This was concluded from observations that protracted feeding of the creatine analog, cyclocreatine, markedly suppressed CKcatalyzed phosphorylation of Cr (77%) and also increased rat diaphragm muscle AK-catalyzed phosphotransfer (9-fold), but the sum of the net transfer of phosphoryls catalyzed by only these two enzyme systems accounted for no more than 75-80% of the total ATP metabolic flux. 2 Another characteristic of this shift from CK to AK catalysis is the plasticity with regard to the ATP-generating process to which the phosphotransfer system is functionally coupled. From earlier experimentation showing nearly equivalent stoichiometry between net phosphoryl transfer by AK and lactate generation (14), it appeared that phosphotransfer catalyzed by AK was linked to the production of ATP by anaerobic glycolysis. The present experiments indicate otherwise; AK-catalyzed phosphotransfer can also be linked to oxidatively produced ATP. This is apparent when CK-catalyzed phosphotransfer is impaired by DNFB; net AK-catalyzed phosphotransfer exceeds by severalfold net anaerobic glycolytic ATP generation, as indicated by the rate of lactate production (not shown). In other words, under circumstances where CK catalysis is impaired but oxidative phosphorylation remains operational (i.e., unlike the circumstance previously tested (14)), a large proportion of net phosphoryl transfer catalyzed by AK can be accounted for by ATP generated by oxidative phosphorylation.
These results expand the potential importance and versatility of the AK component in the previously proposed models of the operation of the CK and AK phosphoryl transfer systems. What the specific role(s) or advantages may be of one or the other system and the extent to which the networking of these two phosphotransferase systems may vary among tissues or species or even from one metabolic circumstance to another will require further exploration.