Selectivity of pyruvate kinase for Na + and K + in water/dimethylsulfoxide mixtures Leticia Ramı ´ rez-Silva and Jesu ´ s Oria-Herna ´ ndez Departamento de Bioquı ´ mica, Facultad de Medicina, Universidad Nacional Auto ´ noma de Me ´ xico, Me ´ xico In aqueous media, muscle pyruvate kinase is highly selective for K + over Na + . We now studied the selec- tivity of pyruvate kinase in water/dimethylsulfoxide mix- tures by measuring the activation and inhibition constants of K + and Na + , i.e. their binding to the monovalent and divalent cation binding sites of pyruvate kinase, respect- ively [Melchoir J.B. (1965) Biochemistry 4, 1518–1525]. In 40% dimethylsulfoxide the K 0.5 app for K + and Na + were 190 and 64-fold lower than in water. K iapp for K + and Na + decreased 116 and 135-fold between 20 and 40% dimethylsulfoxide. The ratios of K iapp /K 0.5 app for K + and Na + were 34–3.5 and 3.3–0.2, respectively. There- fore, dimethylsulfoxide favored the partition of K + and Na + into the monovalent and divalent cation binding sites of the enzyme. The kinetics of the enzyme at sub- saturating concentrations of activators show that K + and Mg 2+ exhibit high selectivity for their respective cation binding sites, whereas when Na + substitutes K + ,Na + and Mg 2+ bind with high affinity to their incorrect sites. This is evident by the ratio of the affinities of Mg 2+ and K + for the monovalent cation binding site, which is close to 200. For Na + and Mg 2+ this ratio is approximately 20. Therefore, the data suggest that K + induces con- formational changes that prevent the binding of Mg 2+ to the monovalent cation binding site. Circular dichroism spectra of the enzyme and the magnitude of the transfer and apparent binding energies of K + and Na + indicate that structural arrangements of the enzyme induced by dimethylsulfoxide determine the affinities of pyruvate kinase for K + and Na + . Keywords: dimethylsulfoxide; magnesium ion; potassium ion; pyruvate kinase; sodium ion. Rabbit muscle pyruvate kinase catalyzes the transfer of the phosphoryl group of phosphoenolpyruvate to ADP. The reaction is largely favored toward the formation of pyruvate and ATP [1]. An important characteristic of pyruvate kinase is that it has an absolute requirement for K + [2]. The enzyme also catalyzes the reaction in the presence of NH 4 + , Rb + and Tl + , but the activity is 80–60% of that with K + ; with Na + and Li + , it is only 8% and 2%, respectively, and with Tris [3] and (CH 3 ) 4 N + it is 0.02% [4]. Although monovalent cations support markedly different catalytic activities, the attempts to correlate the catalytic activity with changes in structural properties such as immunoelectro- phoretic patterns [5], ultraviolet absorption [6], and circular dichroism of the enzyme [7] have failed. In contrast to dialkylglycine decarboxylase, where the substitution of K + for Na + causes distinct structural changes [8], the structure of pyruvate kinase cocrystallized with Mg 2+ -ATP, oxalate, Mg 2+ ,andeitherK + or Na + exhibits only subtle differences [9]. Likewise, the similarities in the coordination number, the polarizability, and stereochemistry of the Na + and K + ligand interactions do not account for the marked discrimination of pyruvate kinase for these cations [10–13]. An important contributing factor in the selectivity for Na + and K + may be the difference in the dehydration free energy required for stripping the water molecules from the two cations. It is 85 kJÆmol )1 more favorable for K + [10,14– 16]. In fact, when pyruvate kinase was entrapped in reverse micelles with low water content, it was found that Na + was an effective activator of the enzyme [17], and that its effectiveness was comparable to that of NH 4 + and Rb + . This suggested that partition of Na + into the activating site is hindered by energetic barriers [18], and that a low water environment favored the transfer of Na + into pyruvate kinase. Nevertheless, under all conditions, it was observed that at equivalent concentrations of K + and Na + ,the activity of pyruvate kinase was always higher with K + [17]. Pyruvate kinase also requires two Mg 2+ ions per active site for activity [19]. One binds directly to the protein in the absence of substrates (site I) [20] and the other to the nucleotide substrate site (site II) [19]. Other divalent cations may substitute them; ions smaller than Mn 2+ bind prefer- entially to site II, whilst the larger ions bind to site I [21]. However, regarding the characteristics of the sites for divalent and monovalent cations it has been shown that the sites are not entirely specific. In the absence of K + ,Mg 2+ Correspondence to L. Ramı ´ rez-Silva, Departamento de Bioquı ´ mica, Facultad de Medicina, Apartado Postal 70–159, Universidad Nacional Auto ´ noma de Me ´ xico, 05410 Me ´ xico D.F., Me ´ xico. Fax: + 525 6162419, Tel.: + 525 6232510, E-mail: lramirez@laguna.fmedic.unam.mx Abbreviations: V,limitvelocity;K 0.5 , activator constant; K 0.5 app , apparent activator constant; h,Hillcoeficient;k cat , catalytic constant; K iapp , apparent inhibition constant; K i , inhibition constant; K app , apparent Michaelis–Menten constant; K, Michaelis–Menten constant. Enzymes: lactate dehydrogenase (EC 1.1.1.27); pyruvate kinase (EC 2.7.1.40). (Received 18 February 2003, revised 31 March 2003, accepted 4 April 2003) Eur. J. Biochem. 270, 2377–2385 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03605.x canbindtothesiteforK + and vice versa, in the absence of Mg 2+ ,K + canbindtotheMg 2+ binding site [22]. This is of importance to the kinetics of the enzyme, as the occupancy of the two sites by either monovalent or divalent cations yields an inactive enzyme, or an enzyme with very low activity [23]. To ascertain the factors that control the selectivity of the two-cation binding sites, the kinetic and structural differ- ences of pyruvate kinase with K + and Na + were studied. A limitation in the study of the activation of pyruvate kinase by Na + is the low affinity that the enzyme exhibits for this monovalent cation [17]. Reports on the Na + /K-ATPase [24–27] show that the apparent affinities of the enzyme for both Na + and K + are higher in water/dimethylsulfoxide mixtures. The authors indicate that dimethylsulfoxide favors a conformation of the enzyme that exhibits higher apparent affinities for the monovalent cations. Therefore, it was thought that water/dimethylsulfoxide mixtures could be used to determine if energetic barriers or structural arrangements of the enzyme control the selectivity of pyruvate kinase for Na + and K + . This work shows that dimethylsulfoxide induced structural arrangements that favored the partition of both cations into the enzyme. It was also found that in water/dimethylsulfoxide mixtures, K + binds preferentially to the monovalent cation binding site, whereas the smaller cation Na + binds to the divalent cation (inhibitory) binding site with a higher affinity (20-fold) than K + . Materials and methods Materials Rabbit muscle pyruvate kinase and hog muscle lactate dehydrogenase were obtained as ammonium sulfate sus- pensions from Boehringer. The cyclohexylammonium salts of ADP and phosphoenolpyruvate were from Sigma. NADH sodium salt was converted to the cyclohexyl- ammonium salt by ion exchange following the protocol of the manufacturer (Sigma). Analytical and spectroscopy grades of dimethylsulfoxide were from Merck. Prior to the experiments, the suspensions of pyruvate kinase and lactate dehydrogenase were centrifuged and the pellets dissolved in 90 lLof50m M Tris/HCl pH 7.6. The solutions were passed twice through Sephadex G-25 insulin centrifuge columns [28]. As noted elsewhere [29], the concentration of contaminating NH 4 + ,Na + or K + in the assay mixtures was below the limits of detection (10 l M ). Assay of pyruvate kinase activity The formation of pyruvate was measured at 25 °Cina coupled system with lactate dehydrogenase and NADH [30]. In water or in the binary water/dimethylsulfoxide system, 1 mL of reaction mixture contained 1 m M phos- phoenolpyruvate, 3 m M ADP, 3 m M MgCl 2 ,0.24m M NADH, 25 m M Tris/HCl pH 7.6 and the concentrations of NaCl and KCl indicated in the Results and discussion section. To avoid ionic strength effects (CH 3 ) 4 NCl was added to give a final salt concentration of 100 m M .Inall experiments, the overall mixture contained 10, 15, 20, 40, 45, 50 and 60 lg of lactate dehydrogenase when mixtures 0, 5, 10, 20, 25, 30 and 40% dimethylsulfoxide (w/v), were, respectively, used; this was carried out because the activity of lactate dehydrogenase is inhibited by dimethylsulfoxide. The specific activity of pyruvate kinase in water/dimethyl- sulfoxide mixtures was not increased by the fivefold inclusion of lactate dehydrogenase. Activity was initiated by introducing pyruvate kinase. Fluorescence experiments Fluorescence emission spectra of pyruvate kinase in 100% water and in various water/dimethylsulfoxide mixtures were determined in an ISS PCI Photon Counting Spectro- fluorometer (ISS, Urbana, Il) thermoregulated at 25 °C. Excitation wavelength was set at 295 nm with excitation and emission slits of 4 nm. Emission was measured from 300 to 450 nm. The samples contained 30 lgÆmL )1 pyruvate kinase, 25 m M Tris/HCl pH 7.6 and either K + ,Na + or (CH 3 ) 4 N + with or without 1 m M phosphoenolpyruvate and 3m M Mg 2+ . The fluorescence spectra of blanks (no protein) were subtracted from those that contained the enzyme. From the difference, the spectral center of mass, or average emission wavelength was calculated with the software provided by ISS Inc. as indicated elsewhere [31]. Circular dichroism experiments CD measurements were carried out on a Jasco J-720 spectropolarimeter. A 5-mm quartz cell was used for near- ultraviolet (UV) CD experiments. The experiments were conducted at room temperature. The concentration of pyruvate kinase was 1 mgÆmL )1 . Spectral scans were run from 270 to 300 nm at intervals of 0.5 nm and a time constant of 5 s. The spectra of blanks were subtracted from those that contained the protein. CD is expressed as molar ellipticity. Protein concentrations were determined by measuring the absorbance at 280 nm using the absorption coefficients of 0.54 mLÆmg )1 Æcm )1 for pyruvate kinase [32] and 1.45 mLÆmg )1 Æcm )1 for lactate dehydrogenase [33]. Results and discussion Comparative effect of Na + and K + on the activity of pyruvate kinase TheeffectofNa + and K + concentration on the activity of pyruvate kinase was determined in mixtures with various dimethylsulfoxide concentrations; a saturating concentra- tion of Mg 2+ was maintained constant (Fig. 1). In the concentrations of Na + and K + thatcouldbeassayedin 100% water, K + induced a strong activation that exhibited saturation kinetics; Na + induced a relatively small enhance- ment of activity that was far from saturation. In water/ dimethylsulfoxide mixtures the titration curves were bipha- sic; the progressive increase in activity was followed by inhibition. The increase in activity is due to the progressive occupancy of the monovalent cation binding site by Na + or K + to an enzyme that has the site for divalent cation filled with Mg 2+ . According to Buchbinder and Reed [21], Na + and K + bind to the divalent cation binding site I. Thus, the inhibition shown in Fig. 1 would result from the 2378 L. Ramı ´ rez-Silva and J. Oria-Herna ´ ndez (Eur. J. Biochem. 270) Ó FEBS 2003 displacement of Mg 2+ by relatively high concentrations of the monovalent cations [3,17,22,23]. A notable feature of the data in Fig. 1 is that as dimethylsulfoxide concentration was increased, the curves shifted to the left. This is because the concentration of Na + and K + required for half-maximal activation and inhibition decreased. The kinetic constants at the various dimethyl- sulfoxide concentrations are in Table 1. Although increas- ing dimethylsulfoxide concentrations decreased the K 0.5 app and K iapp for both monovalent cations, it is relevant that the decrease of the K 0.5 app was more important for K + than for Na + ;theratioofK 0.5 app for K + inwatertothatin40% dimethylsulfoxide increased 190-fold, whereas with Na + the ratio increased 64-fold (Fig. 2). Compared to the twofold to fourfold decrease in K 0.5 for Na + and K + in the Na + /K- ATPase [24–27], this increase in affinities of pyruvate kinase for monovalent cations is the highest reported in the presence of dimethylsulfoxide and is similar to that previ- ously found in reverse micelles [17]. In order to explore how dimethylsulfoxide affects the partition of Na + and K + into the divalent metal cation binding site, the K iapp for Na + and K + at different cosolvent concentrations were compared. At variance to the difference in K 0.5 app for Na + and K + ,itwasfoundthat dimethylsulfoxide affected to similar extents their respective K iapp (Table 1). This was clearly evident in the ratios K iapp / K 0.5 app (Fig. 2, inset). For Na + , this ratio varied from 3.3 in 20% dimethylsulfoxide to 0.2 in 40% dimethylsulfoxide, whereas with K + , this decreased from 34 to 3.5 in the same range of dimethylsulfoxide concentrations. The decrease in the ratio indicates that the partition of Na + and K + into the divalent cation binding site is more sensitive to dimethylsulfoxide than the partition of Na + and K + into the monovalent cation binding site. However in 20–40% dimethylsulfoxide, K + binds preferentially to the mono- valent cation binding site, whereas Na + binds with similar affinity to both, monovalent and divalent cation binding sites (Fig. 2, inset). Competition between Na + or K + and Mg 2+ To gain further insight into how dimethylsulfoxide affects the competition between the monovalent cations and Mg 2+ in pyruvate kinase, we determined the kinetics of saturation for Na + and K + in presence of 1.35 and 6.46 m M Mg 2+ free (Fig. 3). With 6.46 m M Mg 2+ free ,theK app and V for Na + were approximately four and two times higher than with 1.35 m M Mg 2+ free ;theK app and V for K + were not affected by Mg 2+ concentration. It is noted that Mg 2+ did not affect the K iapp for Na + , albeit, the K iapp for K + was approxi- mately six times higher at the higher Mg 2+ concentration. The results indicate that in comparison to K + ,Na + binds very tightly to the divalent cation binding site. When Mg 2+ replaces Na + from the divalent cation binding site, the inhibitory effect of Na + is released and the maximal activity increased by twofold. It is also relevant that Mg 2+ binds more easily to the monovalent cation binding site in the Na + -pyruvate kinase complex than in the K + -pyruvate kinase complex. We carried out additional kinetic studies in order to further characterize the competition between Na + and Mg 2+ andthatofK + and Mg 2+ .Inmediawith40% dimethylsulfoxide, the kinetics of Na + activation at Mg 2+ free concentrations in the range of 1.35–9.38 m M were determined. For determination of the kinetics of Mg 2+ ,the concentrations of Na + and K + were varied between 1.4 and 11.8 m M and from 6.46 to 33.14 m M , respectively. The subsaturating regions of the curves were used to generate Lineweaver–Burk plots (Fig. 4A–C). To calculate K and K i ,theK app values derived from the data in Fig. 4A–C were replotted against the concentrations of the inhibitor (Fig. 4D–F). In all cases, the double reciprocal plots showed a competitive pattern. The linear fit of the replots indicates that inhibition is purely competitive. Fig. 1. Effect of Na + (A) and K + (B) on the activity of pyruvate kinase in 100% water and in various water/dimethylsulfoxide mixtures. One millilitre of reaction mixture contained 1 m M phosphoenolpyruvate, 3m M ADP, 3 m M MgCl 2 ,0.24m M NADH, 25 m M Tris/HCl pH 7.6 and 10–60 lg lactate dehydrogenase. The concentrations of Na + and K + were varied as indicated and (CH 3 ) 4 NCl was added in each case to give a final salt concentration of 100 m M in order to maintain constant ionic strength. The activities in the presence of the indicated concen- trations of the logarithm of NaCl and KCl in 100% water (d); 5% dimethylsulfoxide (h); 10% dimethylsulfoxide (.); 20% dimethyl- sulfoxide (n); 25% dimethylsulfoxide (r); 30% dimethylsulfoxide (s) and 40% dimethylsulfoxide (j) are shown. The reaction was started by the addition of pyruvate kinase. Different amounts of pyruvate kinase were used ranging from 70 lgÆmL )1 without monovalent cation added, to 0.2 lgÆmL )1 and 0.1 lgÆmL )1 when the mixtures contained Na + and K + , respectively. The basal cation-independent activities (0.025, 0.041, 0.15, 1.1, 3, 5.9 and 19.3 lmolÆmin )1 Æmg for 0%, 5%, 10%, 20%, 25%, 30% and 40% (w/v) dimethylsulfoxide, respectively) were subtracted. The temperature was 25 °C. The mean of three to six experiments is shown. The standard deviations are shown in Table 1. Ó FEBS 2003 Selectivity of pyruvate kinase for Na + and K + (Eur. J. Biochem. 270) 2379 The kinetic constants resulting from Fig. 4 are shown in Table 2. The K values for Na + ,K + and Mg 2+ were 0.017 m M , 0.019 m M and 0.013 m M , respectively. These results indicate that in 40% dimethylsulfoxide the enzyme binds with the same affinity Na + ,K + and Mg 2+ . Under conditions in which Na + competes with Mg 2+ ,theK i values for Na + and for Mg 2+ were 0.22 m M and 0.41 m M , respectively. In the competition of Mg 2+ with K + ,theK i for Mg 2+ was 4.38 m M . Therefore, in comparison to Na + and Mg 2+ , there is a higher selectivity between K + and Mg 2+ for their respective sites. This is clearly evident in the ratios K i Mg2+ /K Na+ , K iNa+ /K Mg2+ and K iK+ /K Mg2+ that are 24, 17 and 231, respectively. The overall results must be explained in terms of the intrinsic differences between Na + and K + . Because the V with K + is always higher than with Na + in all experimental conditions (Tables 1 and 2), it is possible that K + gives rise to a particular conformation in order to prevent Mg 2+ inhibition. Another alternative is that in comparison to K + , the similar ionic radii and bond distances of Na + and Mg 2+ [11,16], account for their ability to occupy with high affinity the divalent and monovalent cation binding sites, Table 1. Kinetics of the activation of pyruvate kinase by Na + and K + in 100% water and in various water/dimethylsulfoxide mixtures. The program MICROCAL - ORIGIN version 3.73 was used to calculate the apparent kinetic constants from the data of Fig. 1. The sigmoidal data were fitted to v ¼ VÆS h /K 0.5 h + S h . Data that showed inhibition were fitted to v ¼ VÆS/K + S +(S) 2 /K i and those that exhibited both sigmoidicity and inhibition were fitted to v ¼ VÆS h /K 0.5 h + S h +(S) 2 /K i . The fit of various curves to the Hill model was better than when adjusted to the Michaelis– Menten equation, which may be due to the decrease in affinities for phosphoenolpyruvate and ADP-Mg 2+ when K + is nearly or totally absent [30]. The mean and standard deviations from three to six experiments are shown. The kinetic constants estimated for Na + in 100% water and 40% dimethylsulfoxide exhibit large standard deviations. This is because in 100% water, the concentration of Na + required for half-maximal activation was higher than the value that could be experimentally assayed; thus the values shown derive from large extrapolation of the experimental data. In 40% dimethylsulfoxide, the K iapp for Na + is approximately sixfold smaller than the K app for Na + , making it difficult to obtain a good fit of the experimental points. Dimethylsulfoxide (%, w/v) Na + K + k cat (s )1 ) K app a (m M ) hK i app (m M ) k cat (s )1 ) K app a (m M ) hK i app (m M ) 0 362 ± 150 179 ± 86 1.39 ± 0.14 – 1592 ± 296 21 ± 6 1.22 ± 0.10 142 ± 69 5 339 ± 23 58 ± 6 1.44 ± 0.09 – 2287 ± 90 23 ± 3 – 88 ± 14 10 427 ± 25 46 ± 4 1.46 ± 0.09 – 1714 ± 75 8 ± 0.7 – 92 ± 11 20 604 ± 43 20 ± 2 1.53 ± 0.02 65 ± 12 972 ± 19 1.3 ± 0.07 – 44 ± 2.4 25 695 ± 43 12 ± 1 1.19 ± 0.07 33 ± 9 1019 ± 23 0.76 ± 0.06 – 15.9 ± 1.2 30 604 ± 43 7 ± 0.6 – 16.8 ± 1.9 1157 ± 28 0.33 ± 0.02 – 4.5 ± 0.3 40 569 ± 316 2.8 ± 1.8 – 0.48 ± 0.32 466 ± 59 0.11 ± 0.03 – 0.38 ± 0.08 a K app represents the K 0.5 app and K app for the data fitted to the Hill and to the Michaelis–Menten equations, respectively. Fig. 2. Ratios of K 0.5 app for pyruvate kinase activatedbyNa + (d)orK + (m)inwaterto K 0.5 app at the indicated dimethylsulfoxide (DMSO) concentrations. Theratioswerecal- culated from the data in Table 1. The inset illustrates the ratios of the apparent inhibition constants to K 0.5 app at the indicated water/ dimethylsulfoxide mixtures. The data point indicated by (j) represents the normalized K 0.5 app of pyruvate kinase with Na + or K + in 100% water. 2380 L. Ramı ´ rez-Silva and J. Oria-Herna ´ ndez (Eur. J. Biochem. 270) Ó FEBS 2003 respectively. In this respect, the example of fructose 1,6-bisphosphatase is illustrative. The small cation Li + , displaces Mg 2+ from its divalent cation binding site, but it is unable to occupy the K + binding site [34–36]. Solvation and energetic of binding of Na + and K + to pyruvate kinase It has been hypothesized [18] that the difference between Na + and K + in the activation and inhibition of pyruvate kinase could be related to their different solvation energy. Accordingly, we examined if the energy of transfer of Na + and K + between solvents correlates with their ability to affect the kinetics of pyruvate kinase. Due to solvent–solvent interactions, water/dimethylsulfoxide mix- tures are more structured than water or dimethylsulfoxide alone [37–40]. In such mixtures, small cations like Na + show positive enthalpies of transfer from water to mixtures with low dimethylsulfoxide concentrations. Maximum desolvation of Na + occurs with approximately 30% dimethylsulfoxide (v/v); at higher dimethylsulfoxide concentrations, Na + becomes more strongly solvated due to strong ion–dipole interactions [41]. These changes in ion solvation reflect on the free energy of transfer between solvents. Indeed, when ion transfer is between organic solvents, the free energy of solution (DDG s )is predominantly governed by the enthalpy of solution (DDH s ) [41]. However, when water is one of the solvents, entropy changes (TDS) contribute strongly to the free energy of transfer [41]. In the light of the latter data, we explored whether there is a relationship between the transfer energies of Na + and K + from water to water/dimethylsulfoxide and the affinity for Na + and K + of pyruvate kinase. As shown in Table 3, the transfer energies of monovalent cations become increasingly negative as dimethylsulfoxide concentration is increased; this reflects the solvation of cations in such media [42]. Table 3 also shows that the DG° t of both, Na + and K + from water to dimethylsulfoxide mixtures are negative, which indicates that the monovalent cations are more solvated in water/dimethylsulfoxide mixtures than in 100% water. However, the data also show that the DG° t is far more negative with Na + than with K + . The latter data suggested that the lower solvation of K + would favor its partition into pyruvate kinase. Thus, we compared the values of DG° t to the apparent binding energies of Na + and K + in pyruvate kinase. For the latter calculations, we assumed that the K 0.5 app is equal to the K d , and from the equilibrium constants we calculated the apparent binding energies. The effect of dimethylsulfoxide on the affinity of pyruvate kinase for both cations can be calculated from the difference of apparent binding energies in water/dimethylsulfoxide mixtures minus the values obtained in 100% water (DDG° b ¼ DG° b (water–dimethylsulfoxide) ) DG° b (100% water) ).For K + ,theDDG° b -values calculated in 20 and 40% dimethyl- sulfoxide were )6.86 and )13.02 kJÆmol )1 ;withNa + the Fig. 3. Effect of Mg 2+ on the activity of pyruvate kinase at various concentrations of Na + (A) and K + (B) in 40% dimethylsulfoxide. The experimental conditions were as in Fig 1, except that the experiments were performed in the presence of 40% dimethylsulfoxide and two concentrations of Mg 2+ free ,1.35m M (d)and6.46m M (s). The basal cation-independent activities were subtracted. The temperature was 25 °C. The kinetic constants were calculated as in Table 1. The average from two experiments is shown. Ó FEBS 2003 Selectivity of pyruvate kinase for Na + and K + (Eur. J. Biochem. 270) 2381 values were less negative )5.40 and )10.29 kJÆmol )1 .As shown in Table 3, the transfer energies are much less negative than the apparent binding energies (DG° b ) . This indicates that solvation energy is not the predominant factor in the control of the affinity for Na + and K + in pyruvate kinase. Fig. 4. Competition between Na + or K + and Mg 2+ . One millilitre of reaction mixture contained 1 m M phosphoenolpyruvate, 3 m M ADP, 0.24 m M NADH and 25 m M Tris/HCl pH 7.6. The temperature was 25 °C. The free concentrations of Mg 2+ were calculated using the Mg 2+ -ADP association constants reported for aqueous media with the computer program CHELATOR [45]. Na + varied from 0 to 10 m M at various Mg 2+ free fixed concentrations: 1.35 m M (m), 3.7 m M (h), 6.46 m M (d) and 9.38 m M (,). K + varied from 0 to 10 m M at various fixed Mg 2+ concentrations: 6.46 m M (m), 15.24 m M (h), 24.17 m M (d), and 33.14 m M (,). Conversely, Mg 2+ free varied from 0.014 to 12.3 m M at various fixed Na + concentrations: 1.4 m M (m), 2.8 m M (h), 5.75 m M (d) and 11.2 m M (,). The subsaturating regions of the curves were used to generate the Lineweaver–Burk plots (A–C). The K app values derived from these figures were plotted vs. the concentrations of the inhibitor (D–F). The kinetic constants are shown in Table 3. Table 2. Kinetic constants derived from competitive inhibition plots between monovalent cations (Na + and K + )andMg 2+ in 40% (w/v) dimethyl- sulfoxide. The apparent kinetic constants were calculated from data of Fig. 4, A–C, except that the activity without monovalent cation was not subtracted. The average from two experiments is shown. The standard error was approximately 5% in all cases. The K and K i were calculated from the replots fitted to K app ¼ K/K i [I] + K (Fig. 4, D–F). The linear correlation coefficients were above 0.995. Na + K + Mg 2+ Mg 2+ free (m M ) K app (m M ) V (lmolÆmin )1 Æmg) Mg 2+ free (m M ) K app (m M ) V (lmolÆmin )1 Æmg) Na + (m M ) K app (m M ) V (lmolÆmin )1 Æmg) 1.35 0.086 45 6.46 0.048 73 1.4 0.11 38 3.70 0.158 44 15.24 0.085 73 2.8 0.15 39 6.46 0.273 45 24.17 0.115 66 5.75 0.36 34 9.38 0.416 41 33.14 0.165 78 11.8 0.69 39 K (m M ) 0.017 K (m M ) 0.019 K (m M ) 0.013 K i (m M ) 0.41 K i (m M ) 4.38 K i (m M ) 0.22 2382 L. Ramı ´ rez-Silva and J. Oria-Herna ´ ndez (Eur. J. Biochem. 270) Ó FEBS 2003 Conformation of pyruvate kinase in water and 40% dimethylsulfoxide solutions In the light of the latter data it was considered that the different catalytic activities of pyruvate kinase with Na + and K + could be related to distinct structural features of the enzyme. Therefore, we determined the intrinsic fluorescence spectra of pyruvate kinase with and without cations and various concentrations of dimethylsulfoxide. Differences in the intrinsic fluorescence of pyruvate kinase incubated with and without ligands (K + ,Mg 2+ and phosphoenolpyruvate) have been reported previously, and it has been proposed that the distinct spectra correspond to the active and inactive conformations of pyruvate kinase [17,29,43]. The spectral center of mass of pyruvate kinase and pyruvate kinase in complex with Mg 2+ and phosphoenolpyruvate in the presence of Na + ,K + ,or(CH 3 ) 4 N + were recorded in mixtures that contained 0%, 5%, 10%, 20%, 30% and 40% (w/v) dimethylsulfoxide. In agreement with previous data [31], the recordings showed that in presence of 30 and 40% dimethylsulfoxide and in the absence of ligands, pyruvate kinase exhibited the active conformation. Moreover, in the presence of dimethylsulfoxide and ligands, pyruvate kinase continued to exhibit the active conformation; this was Table 3. Transfer energies for K + and Na + from 100% water to water/ dimethylsulfoxide solutions, and apparent binding energies for pyruvate kinase-K + and pyruvate kinase-Na + complexes in water and water/ dimethylsulfoxide mixtures. Transfer energies (DG° t )weretakenand transformedtokJÆmol )1 from the data of Kundu and Das (1979) [42]. The apparent binding energies (DG° b ) were calculated from the data of Table 1. Transfer energies were calculated in water/dimethylsulfoxide mixtures by volume (v/v), and the apparent binding energies in water/ dimethylsulfoxide by weight (w/v). Dimethylsulfoxide (%, w/v) Na + K + DG  t (kJÆmol )1 ) DG  b a (kJÆmol )1 ) DG  t (kJÆmol )1 ) DG  b a (kJÆmol )1 ) 0–)4.27 – )9.58 20 )1.72 )9.67 )1.30 )16.44 40 )3.00 )14.56 )2.01 )22.6 a Apparent binding energies were calculated assuming that the K 0.5 app is equal to K d . Fig. 5. Near-UV-CD spectra of pyruvate kinase in aqueous media (A) and 40% dimethylsulfoxide (w/v) (B) in the presence of (CH 3 ) 4 N + ,Na + and K + . The spectra were obtained in mixtures that contained pyruvate kinase at a concentration of 1 mgÆmL )1 in 25 m M Tris/HCl, pH 7.6, 1 m M phosphoenolpyruvate, 3 m M Mg 2+ and the monovalent cation indicated. Na + and K + were added at concentrations in which maximal activation of the enzyme was achieved. (A) 100 m M (CH 3 ) 4 N + (h), 100 m M Na + (n)and90 m M K + (s) were included in the reaction mixtures, (B) 100 m M (CH 3 ) 4 N + (h), 1.5 m M Na + (n)and0.2m M K + (s) were added. In all cases, salt concentration (100 m M ) was kept constant by including appropriate amounts of (CH 3 ) 4 N + . Ó FEBS 2003 Selectivity of pyruvate kinase for Na + and K + (Eur. J. Biochem. 270) 2383 independent of which monovalent cation was in the media. It is also relevant that in the latter condition, dimethylsulf- oxide did not modify the spectra (data not shown). Previous attempts to find if there is a correlation between the activating effect of various monovalent cations and structural alterations of the enzyme have been unsuccessful [7]. These studies have been performed in water and in presence of K + ,Li + and (CH 3 ) 4 N + , but not Na + . Here, we examined the effect of Na + and K + on the circular dichroism spectra of pyruvate kinase in water media and in 40% dimethylsulfoxide. The concentrations of Na + and K + that induced maximal activation were used (Fig. 5). In all cases, the enzyme showed transition bands at 282 and 289 nm where Tyr and Trp residues overlap [44]. The intensities of the spectra with the nonactivating cation (CH 3 ) 4 N + were very similar in water and 40% dimethyl- sulfoxide. In the presence of K + and Na + , however, the intensities of the spectra were, respectively, 25% and 17% lower in 40% dimethylsulfoxide than in water. It is also noteworthy, that in the presence of any cation, the CD bands at 282 nm and 289 nm were sharper in 40% dimethylsulfoxide than in aqueous media. This suggests that in 100% water, the CD spectra reflect the average of different conformations, whereas in dimethylsulfoxide there would seem to be an enrichment of enzymes with the same conformation. Conclusions The results of this study explain the high discrimination of pyruvate kinase for Na + and K + .K + and Mg 2+ exhibit high selectivity for their respective cation binding sites, whereas Na + and Mg 2+ are promiscuous with high affinity for their incorrect sites. This is clearly evident in the ratio of the affinities of Mg 2+ and K + for the monovalent cation binding site, which is approximately 200. For Na + and Mg 2+ this ratio is approximately 20. In the light of kinetic data, Na + and K + bind to the monovalent binding site with the same affinity. However the catalytic rates with K + are always higher; this suggests that K + induces particular conformational changes that are favorable for catalysis and that at the same time prevent inhibition by Mg 2+ . In regard to ionic selectivity, the physical characteristics of the ions (ionic radius, coordination number, geometry) and their solvation energy must be taken into account. The differences in ionic radii of K + and Na + undoubtedly contribute to the selectivity of monovalent and divalent cation binding sites in pyruvate kinase. As to the energetics of binding, we have explored if the magnitude of the transfer energies of Na + and K + from water to dimethylsulfoxide is related to the apparent binding energies of Na + and K + to pyruvate kinase (in 40% dimethylsulfoxide the DG° t and DG° b are )3.0 and )14.56 kJÆmol )1 for Na + and )2.01 and )22.6 kJÆmol )1 for K + ). We found that in presence of dimethylsulfoxide, the factor that determines the affinity and selectivity of pyruvate kinase for K + and Na + is not dehydration. Instead, intrinsic fluorescence and near ultraviolet CD studies show that dimethylsulfoxide induces structural arrangements in which the whole enzyme population acquires the high affinity active conformation. Acknowledgements The authors thank Rocı ´ oPatin ˜ o from Instituto de Quı ´ mica, Univer- sidad Nacional Auto ´ noma de Me ´ xico for technical assistance in the CD measurements and A. Go ´ mez-Puyou and M. Tuena de Go ´ mez-Puyou for valuable suggestions, discussion and revision of the manuscript. This work was partially supported by grants IN227202-3 from DGAPA-UNAM and 32033-N from Consejo Nacional de Ciencia y Tecnologı ´ a, Me ´ xico. References 1. McQuate, J.T. & Utter, M.F. (1959) Equilibrium and kinetic studies of the pyruvic kinase reaction. J. Biol. Chem. 234, 2151–2157. 2. Kachmar, J.F. & Boyer, P.D. (1953) Kinetic analysis of enzyme reactions. II. The potassium activation and calcium inhibition of pyruvic phosphoferase. J. Biol. Chem. 200, 669–683. 3. Kayne, F.J. (1971) Thallium (I) activation of pyruvate kinase. Arch. Biochem. Biophys. 143, 232–239. 4. Nowak, T. (1976) Conformational change required for pyruvate kinase activity as modulated by monovalent cations. J. Biol. Chem. 251, 73–78. 5. Sorger, G.J., Ford, R.E. & Evans, H. (1965) Effects of univalent cations on the immunoelectrophoretic behavior of pyruvic kinase. Proc.NatlAcad.Sci.54, 1614–1621. 6. Wilson, R.H., Evans, H.J. & Becker, R.R. (1967) The effect of univalent cation salts on the stability and on certain physical properties of pyruvate kinase. J. Biol. Chem. 242, 3825–3832. 7. Wildes, R.A., Evans, H.J. & Becker, R.R. (1971) The effect of univalent cations on the circular dichroism of pyruvate kinase. Biochim. Biophys. Acta 229, 850–854. 8.Toney,M.D.,Hohenester,E.,Cowan,S.W.&Jansonius,J.N. (1993) Dialkylglycine decarboxylase structure: Bifunctional active site and alkali metal sites. Science 261, 756–759. 9. Larsen, T.M., Benning, M.M., Rayment, I. & Reed, G.H. (1998) Structure of the bis (Mg 2+ )-ATP-oxalate complex of the rabbit muscle pyruvate kinase at 2.1 A ˚ resolution: ATP binding over a barrel. Biochemistry 37, 6247–6255. 10. Cotton, F.A. & Wilkinson, G. (1980) Inorganic chemistry of group Ia and IIa metals. In Advanced Inorganic Chemistry, pp. 14–63. Wiley-Interscience, New York. 11. Glusker, J.P. (1991) Structural aspects of metal liganding to functional groups in proteins. Advances Prot. Chem. 42, 1–76. 12. Sharpe, A.G. (1992) Inorganic Chemistry 3rd edn. Longman Scientific and Technical, Harlow, Essex. 13. Woehl, E.U. & Dunn, M.F. (1995) The roles of Na + and K + in pyridoxal phosphate enzyme catalysis. Chem. Rev. 144, 147–197. 14. Mullins, L.J. & Moore, R.D. (1960) Movement of thallium ions in muscle. J. Gen. Physiol. 43, 759–773. 15. Eisenman, G. (1962) Cation selective glass electrodes and their mode of operation. Biophys. J. S 2, 259–323. 16. Marcus, Y. (1994) A simple empirical model describing the thermodynamics of hydration of ions widely varying charges, sizes, and shapes. Biophys. Chem. 51, 111–127. 17. Ramı ´ rez-Silva, L., Oria, J., Go ´ mez-Puyou, A. & Tuena de Go ´ mez- Puyou, M. (1997) The contribution of water to the selectivity of pyruvate kinase for Na + and K + . Eur. J. Biochem. 250, 583–589. 18. Miller, C. (1993) Potassium selectivity in proteins: oxygen cage or in the F ace? Science 261, 1692–1693. 19. Gupta, R.K. & Oesterling, R.M. (1976) Dual divalent cation requirement for activation of pyruvate kinase; essential roles of both enzyme- and nucleotide-bound metal ions. Biochemistry 15, 2881–2887. 20. Reuben,J.&Cohn,M.(1970)Magneticresonancestudiesof manganese (II) binding sites of pyruvate kinase. Temperature 2384 L. Ramı ´ rez-Silva and J. Oria-Herna ´ ndez (Eur. J. Biochem. 270) Ó FEBS 2003 effects and frequency dependence of proton relaxation rates of water. J. Biol. Chem. 245, 6539–6546. 21. Buchbinder, J.L. & Reed, G.H. (1990) Electron paramagnetic resonance studies of the coordination schemes and site selectivities for divalent metal ions in complexes of pyruvate kinase. Biochemistry 29, 1799–1806. 22. Suelter, C.H., Singleton, R. Jr, Kayne, F.J., Arrington, S., Glass, J. & Mildvan, A.S. (1966) Studies on the interaction of substrate and monovalent and divalent cations with pyruvate kinase. Biochemistry 5, 131–138. 23. Melchoir, J.B. (1965) The role of metal ions in the pyruvic kinase reaction. Biochemistry 4, 1518–1525. 24. Albers, R.W. & Koval, G.J. (1972) Sodium-potassium-activated adenosine triphosphatase VII. Concurrent inhibition of Na + -K + - adenosine triphosphatase and activation K + -nitrophenylphos- phatase activities. J. Biol. Chem. 247, 3088–3092. 25. Albers, R.W. & Koval, G.J. (1973) Sodium-potassium-activated adenosine triphosphatase of Electrophorus electric organ. 8. Monovalent cation sites regulating phosphatase activity. J. Biol. Chem. 248, 777–784. 26. Swann, A.C. & Albers, R.W. (1975) Sodium-potassium-activated ATPase of mammalian brain regulation of phosphatase activity. Biochim. Biophys. Acta 382, 437–456. 27. Swann, A.C. (1983) Brain (Na + ,K + )-ATPase. Opposite effects of ethanol and dimethyl sulfoxide on temperature dependence of enzyme conformation and univalent cation binding. J. Biol. Chem. 258, 11780–11786. 28. Kasahara, M. & Penefsky, H.S. (1978) High affinity binding of monovalent Pi by beef heart mitochondrial adenosine triphos- phatase. J. Biol. Chem. 253, 4180–4187. 29. Ramı ´ rez-Silva, L., Tuena de Go ´ mez-Puyou, M. & Go ´ mez-Puyou, A. (1993) Water-induced transitions in the K + -requirements for the activity of pyruvate kinase entrapped in reverse micelles. Biochemistry 32, 5332–5338. 30. Bu ¨ chner, T. & Pleiderer, G. (1955) Methods in Enzymology (Colowick, S.P. & Kaplan, N.O., eds), Vol. 1, pp. 435–440. Academic Press, New York. 31. Ramı ´ rez-Silva, L., Ferreira, T.S., Nowak, T., Tuena de Go ´ mez- Puyou, M. & Go ´ mez-Puyou, A. (2001) Dimethyl sulfoxide pro- motes K + -independent activity of pyruvate kinase and the acquisition of the active catalytic conformation. Eur. J. Biochem. 268, 3267–3274. 32. Cottam, G.L., Hollenberg, P.F. & Coon, M.J. (1969) Subunit structure of rabbit muscle pyruvate kinase. J. Biol. Chem. 244, 1481–1486. 33. Jeanicke, R. & Knof, S. (1968) Molecular weight and quaternary structure of lactic dehydrogenase. 3. Comparative determination by sedimentation analysis, light scattering and osmosis. Eur. J. Biochem. 4, 157–163. 34. Zhang, R., Villeret, V., Lipscomb, W.N. & Fromm, H.J. (1996) Kinetics and mechanism of activation and inhibition of por- cine liver fructose-1,6-bisphosphatase by monovalent cations. Biochemistry 35, 3038–3043. 35. Nakashima, K. & Tuboi, S. (1976) Size-dependent allosteric effects of monovalent cations on rabbit liver fructose-1,6-bisphosphatase. J. Biol. Chem. 251, 4315–4321. 36. Villeret, V., Huang, S., Fromm, H.J. & Lipscomb, W.N. (1995) Crystallographic evidence for the action of potassium, thallium, and lithium ions on fructose-1,6-bisphosphatase. Proc. Natl Acad. Sci. 92, 8916–8920. 37. Cowie, J.M.G. & Toporowski, D.M. (1961) Association in the binary liquid system dimethyl sulfoxide-water. Can. J. Chem. 39, 2240–2243. 38. Fuchs, R., McCrary, G.E. & Bloomfield, J.J. (1961) Mechanisms of nucleophilic displacement in aqueous dimethyl sulfoxide solu- tions. J. Am. Chem. Soc. 83, 4281–4284. 39. Safford, G.J., Schaffer, P.C., Leung, P.S., Doebbler, G.F., Brady, G.W. & Lyden, E.F.X. (1969) Neutron inelastic scattering and X-ray studies of aqueous solutions of dimethylsulphoxide and dimethylsulphone. J. Chem. Phys. 50, 2140–2159. 40. Bell, G., Janssen, A., E.M. & Halling, P.J. (1997) Water activity fails to predict critical hydration level for enzyme activity in polar organic solvents: Interconversion of water concentrations and activities. Enzyme Microb. Technol. 20, 471–478. 41. Fuchs, R. & Hagan, P.C. (1973) Single ion entalphies of transfer from water to aqueous dimethyl sulfoxide solutions. J. Phys. Chem. 77, 1797–1800. 42. Kundu, K.K. & Das, A.K. (1979) Transfer free energies of some ions from water to dimethylsulfoxide-water and urea-water mix- tures. J. Solution Chem. 8, 259–265. 43. Suelter, C.H. (1967) Effects of temperature and activating cations on the fluorescence of pyruvate kinase. Biochemistry 6, 418–423. 44. Horowitz, J., Strickland, E.H. & Billups, C. (1970) Analysis of the vibrational structure in the near-ultraviolet circular dichroism and absorption spectra of tyrosine derivatives and ribonuclease-A at 77°K. J. Am. Chem. Soc. 92, 2119–2126. 45. Schoenmakers, T.J.M., Visser, G.J., Flik, G. & Theuvenet, P.R. (1992) CHELATOR: an improved method for computing metal ion concentrations in physiological solutions. Biotechniques 12, 870–879. Ó FEBS 2003 Selectivity of pyruvate kinase for Na + and K + (Eur. J. Biochem. 270) 2385 . active and inactive conformations of pyruvate kinase [17,29,43]. The spectral center of mass of pyruvate kinase and pyruvate kinase in complex with Mg 2+ and. fluorescence spectra of pyruvate kinase with and without cations and various concentrations of dimethylsulfoxide. Differences in the intrinsic fluorescence of pyruvate kinase