Temperature and phosphate effects on allosteric phenomena of phosphofructokinase from a hibernating ground squirrel (Spermophilus lateralis) Justin A. MacDonald 1 and Kenneth B. Storey 2 1 Department of Biochemistry & Molecular Biology, University of Calgary, AB, Canada 2 Institute of Biochemistry and Department of Biology, Carleton University Ottawa, ON, Canada Environments with widely differing seasonal tempera- tures present thermoregulatory challenges to small mammals who aim to maintain a constant body tem- perature of about 37 °C. Winter is particularly difficult because energy use in support of homeothermy increa- ses dramatically in cold weather at the same time as the food supply declines. For many small mammals, the only survival solution to this combination of low food availability and low environmental temperatures is hibernation [1–3]. The mammalian hibernator aban- dons homeothermy and allows body temperature to drop to that of its surroundings (although regulating body temperature at 0–5 °C if ambient temperature falls below 0 °C). The mechanisms that control the entry into hibernation are still not fully understood but it is known that an active suppression of basal metabolic rate occurs (often to only 1–5% of the nor- mal resting rate), preceding and causing the fall in body temperature. Hibernation is also facilitated by the accumulation, during late summer feeding, of huge reserves of lipids; for example, in ground squirrels, body mass often increases by 50% or more. These lipids are the main fuel for winter energy metabolism during torpor and measurements of respiratory quo- tients confirm this. Lipid oxidation is supplemented to some extent by gluconeogenesis from amino acids but carbohydrate reserves are largely spared to be used only by tissues and organs that can oxidize little else Keywords glycolysis; mammalian hibernation; metabolic rate depression; phosphofructokinase; temperature effects Correspondence K. B. Storey, Institute of Biochemistry and Department of Biology, Carleton University, 1125 Colonel By Drive, Ottawa, ON, K1S 5B6 Canada E-mail: kenneth_storey@carleton.ca (Received 7 July 2004, revised 31 August 2004, accepted 14 September 2004) doi:10.1111/j.1432-1033.2004.04388.x Temperature effects on the kinetic properties of phosphofructokinase (PFK) purified from skeletal muscle of the golden-mantled ground squirrel, Spermophilus lateralis, were examined at 37 °C and 5 °C, values character- istic of body temperatures in euthermia vs. hibernation. The enzyme showed reduced sensitivity to all activators at 5 °C, the K a values for AMP, ADP, NH 4 + and F2,6P 2 were 3–11-fold higher at 5 °C than at 37 °C. Inhibition by citrate was not affected whereas phosphoenolpyruvate, ATP and urea became more potent inhibitors at low temperature. While typically considered an activator of PFK activity, inorganic phosphate per- formed as an inhibitor at 5 °C. Decreasing temperature alone causes the actions of inorganic phosphate to change from activation to inhibition. We found that K m values for ATP remained constant while V max dropped sig- nificantly upon the addition of phosphate. Phosphate inhibition at 5 °C was noncompetitive with respect to ATP and the K i was 0.15 ± 0.01 mm (n ¼ 4). The results indicate that PFK is less likely to be activated in cold torpid muscle; PFK is less sensitive to changing adenylate levels at the low temperatures characteristic of torpor, and PFK is clearly much less sensi- tive to biosynthetic signals. All of these characteristics of hibernator PFK would serve to reduce glycolytic rate and help to preserve carbohydrate reserves during torpor. Abbreviations PFK, 6-phosphofructo-1-kinase; F6P, fructose 6-phosphate; F2,6P 2 , fructose 2,6-bisphosphate. 120 FEBS Journal 272 (2005) 120–128 ª 2004 FEBS [2–4]. Glycolytic rate drops to low levels in most organs. Mechanisms that contribute to the suppression of glycolytic flux are important for hibernation for several reasons. Glycolytic rate suppression (a) contributes to the overall metabolic rate depression by suppressing ATP output from carbohydrate catabolism; (b) limits carbohydrate use for anabolic purposes during torpor, and (c) facilitates lipid oxidation as the primary ATP- generating pathway in most organs as well as gluconeo- genesis in selected organs. 6-Phosphofructo-1-kinase (PFK) is an enzyme of central importance to the regu- lation of glycolysis. PFK gates the commitment of hex- ose phosphates (derived from glycogen or glucose) into the triose phosphate portion of glycolysis. A wide vari- ety of regulatory mechanisms modulate PFK activity including allosteric control by powerful activators and inhibitors [5], pH effects [6–8], post-translational modi- fication by reversible protein phosphorylation [9], and enzyme binding to subcellular macromolecules [10,11]. However, the specific mechanisms that achieve inhibi- tion of PFK activity during hibernation have yet to be fully realized. Skeletal muscle PFK does not appear to be regulated by protein phosphorylation during hiber- nation in ground squirrels [12]. As respiratory acidosis develops during hibernation, effects of low pH on PFK activity and subunit assembly have been sugges- ted as a means by which glycolytic flux in skeletal muscle can be reduced, as the enzyme is highly sensi- tive to pH change [10,13]. However, recent work has shown that low pH inhibition of PFK may not be physiologically relevant [14]. Temperature-dependent mechanisms of enzyme control may contribute to the regulation of PFK activity as the body temperature of hibernators can drop by 30 °C or more [15,16]. Several examples of temperature-dependent changes in the kin- etic properties of hibernator enzymes have been repor- ted [16]. The present study analyzes the kinetic and regulatory properties of PFK purified from skeletal muscle of the golden-mantled ground squirrel (Spermo- philus lateralis). Particular attention is paid to the effects of temperature on enzyme allostery and the influence of temperature in the regulation of PFK activity at both high (euthermic) and low (hibernating) temperatures. Results Selected kinetic properties of PFK purified from S. lat- eralis muscle were compared at assay temperatures and pH values that mimic the hibernating (5 °C, pH 7.5) vs. euthermic (37 °C, pH 7.2) conditions found in vivo in S. lateralis skeletal muscle. Table 1 shows the effect of temperature on activation coefficient (K a ) values for allosteric activators and the values of concentration of the inhibitor that reduces control activity by 50% (I 50 ). The enzyme showed reduced sensitivity to all activa- tors at 5 °C, the K a values for AMP, ADP, NH 4 + , and F2,6P 2 being 5.3-, 3.4-, 2.6- and 11.3-fold higher at 5 °C, respectively, compared with the corresponding 37 °C values. In contrast, inhibitor constants were dif- ferentially affected by temperature change. Inhibition by Mg citrate was not affected whereas the sensitivity to inhibition by phosphoenolpyruvate and urea both increased at low temperature; I 50 values dropped to 53 and 60%, respectively, of the corresponding values at 37 °C. Inorganic phosphate is typically an activator of PFK, and Table 1 shows that the ground squirrel enzyme responded to phosphate as expected at 37 °C with a K a value of 2.0 ± 0.2 mm. However, inorganic phosphate was unable to elicit an activation of PFK activity at 5 °C; in fact, the addition of inorganic phos- phate resulted in an inhibition of PFK activity. Fur- ther analysis of phosphate effects on PFK at 5 °C seemed warranted and Fig. 1 shows an anomalous interaction between pH and phosphate effects on the enzyme. At pH 7.0, phosphate acted as a strong acti- vator of PFK and raised maximal activity by 5.1-fold. The calculated K a for phosphate was 1.73 ± 0.13 mm (n ¼ 3). However, increasing the pH slightly to a value of 7.3 dramatically altered the effect of phosphate and activation was seen only at low concentrations with a maximal 1.7-fold activation at 2 mm phosphate. At Table 1. Effects of temperature on the kinetic properties of skel- etal muscle phosphofructokinase from the golden-mantled ground squirrel, Spermophilus lateralis. Values are means ± SEM; n ¼ 4–6. F6P concentrations were subsaturating (0.1 m M) for K a and I 50 determinations. For I 50 determinations, inhibitor stock solutions were prepared with added magnesium in a 1 : 1 molar ratio for Mg.ATP and 2 : 1 for Mg.citrate. The pH values of assay mixtures in imidazole–HCl buffer were adjusted at 23 °C to predetermined values so that when the mixtures were cooled or warmed to the desired assay temperatures, the pH at 5 °C was 7.5 and the pH at 37 °C was 7.2. NA, no activation. Conditions 37 °C5°C K a AMP (lM) 34.9 ± 2.0 184 ± 5.9 a K a ADP (lM) 35.5 ± 1.0 122 ± 6.4 a K a NH 4 + (mM) 1.4 ± 0.1 3.6 ± 0.08 a K a Fructose-2,6-bisphosphate (nM) 39.9 ± 2.9 449 ± 15 a K a Inorganic phosphate (mM) 2.0 ± 0.2 NA I 50 Phosphoenolpyruvate (mM) 1.2 ± 0.1 0.64 ± 0.04 a I 50 Urea (mM) 314 ± 30 189 ± 13 b I 50 Mg.citrate (lM) 51.7 ± 5.4 48.2 ± 2.0 a Significantly different from the corresponding value at 37 °C, P < 0.005; b P < 0.025 (Student’s t-test, two-tailed). J. A. MacDonald and K. B. Storey Temperature effects on hibernator PFK allostery FEBS Journal 272 (2005) 120–128 ª 2004 FEBS 121 concentrations higher than 10 mm, phosphate pro- duced inhibitory effects. A further increase in pH value to 7.5 removed all activating characteristics of phos- phate and it acted as an inhibitor with an I 50 value of 8.47 ± 0.21 mm. At pH 8.0, the inhibition was even stronger with a decrease in the I 50 to 3.96 ± 0.36 mm. Table 2 shows that PFK exhibited significantly dif- ferent affinity for its F6P substrate at both high and low temperatures. The enzyme showed sigmoidal F6P kinetics at both temperatures with Hill coefficients of  2 in the absence of added phosphate. However, the S 0.5 for F6P was significantly lower (by 39%) at 5 °C than at 37 °C (at 0 mm phosphate). PFK exhibits sig- moidal F6P kinetics at lower pH values but converts to hyperbolic kinetics with the addition of allosteric activators (F1,6P 2 , F2,6P 2 , AMP, inorganic phosphate, or NH 4 + ) or a rise in pH value to near 8 [17–19]. As would be predicted for an activator, the addition of 10 mm phosphate to S. lateralis PFK in 37 °C assays increased the maximal velocity by 1.5-fold, reduced the S 0.5 by 27% and reduced the Hill coefficient to 1.21; this low n H value indicates a hyperbolic relationship between velocity and [F6P] (Table 2). However, the effects of inorganic phosphate on F6P kinetics at 5 °C and pH 7.5 were different. Figure 2 shows that the F6P saturation curve was shifted strongly to the left with the addition of as little as 5 mm inorganic phos- phate and S 0.5 was reduced by 65% (Table 2). The addition of phosphate also changed the V–[F6P] rela- tionship from sigmoidal to hyperbolic; the Hill coeffi- cient dropping by 50% in the presence of 5 mm phosphate. However, unlike with the situation at 37 °C, increasing phosphate concentrations caused a strong reduction in enzyme maximal velocity; V max was reduced by 29% at 5 mm and by 64% at 20 mm phosphate (an effect that can also be seen in Fig. 1). Effects of temperature and phosphate concentration on the ATP kinetics of S. lateralis PFK are shown in Table 3. The enzyme showed hyperbolic substrate sat- uration kinetics with respect to Mg.ATP concentration at both 5 and 37 °C with a much lower K m for Mg.ATP at low temperature. The value at 5 °C was only 15% of the value at 37 °C; hence, affinity for both substrates, F6P and ATP, increased at low tem- perature. As is common for PFK, ATP had inhibitory effects at higher levels and inhibition was stronger at 37 °C with an I 50 for Mg.ATP of 1.29 mm compared with 2.13 at 5 °C. The addition of phosphate at 37 °C produced the typical effects of an activator on ATP kinetics; 10 mm phosphate lowered the K m for Mg.ATP by 58% and increased the I 50 by 2.2-fold. Phosphate effects on ATP kinetics at 5 °C were differ- ent. Phosphate had virtually no affect on the K m of Fig. 1. Effects of pH and inorganic phosphate concentration on the activity of S. lateralis phosphofructokinase at 5 °C. Activities are expressed relative to the PFK activity in the absence of phosphate at each pH value. PFK activity was measured as described in Mate- rials and methods at pH 7.0 (j), pH 7.3 (r), pH 7.6 (d), and pH 8.0 (m). Data are means ± SEM for n ¼ 3 separate determinations. Table 2. The effect of inorganic phosphate concentration on fructose 6-phosphate kinetics of S. lateralis muscle PFK at two assay tempera- tures. Maximal velocity values are milliunits of enzyme activity obtained for 0.2 lg (at 5 °C) or 0.03 lg (at 37 °C) of purified S. lateralis skel- etal muscle phosphofructokinase. The pH values of the assay mixtures were fixed at 23 °C using 50 m M imidazole–HCl buffer and were then allowed to vary with temperature so that the pH at 5 °C was 7.5 and the pH at 37 °C was 7.2. The concentration of Mg.ATP was held at 0.5 m M. All other assay conditions are detailed in the Materials and methods. Data are means ± SEM, n ¼ 4 separate determinations. Temperature Phosphate (m M) S 0.5 F6P (mM) Hill coefficient (n H ) Maximal velocity (mU) 5 °C 0 0.071 ± 0.009 2.30 ± 0.43 9.4 ± 0.3 5 0.025 ± 0.002 a 1.23 ± 0.13 b 6.7 ± 0.2 a 15 0.032 ± 0.004 a 1.12 ± 0.11 b 3.7 ± 0.1 a 20 0.022 ± 0.002 a 0.82 ± 0.15 b 3.4 ± 0.1 a 37 °C 0 0.121 ± 0.004 2.03 ± 0.21 6.21 ± 0.2 10 0.088 ± 0.007 b 1.21 ± 0.06 b 9.09 ± 0.4 a a Significantly different from the corresponding value at 0 mM phosphate using the Student’s t-test (two-tailed) or one-way analysis of variance followed by the Student–Newman–Keuls test (two-tailed), P < 0.005, b P < 0.05. Temperature effects on hibernator PFK allostery J. A. MacDonald and K. B. Storey 122 FEBS Journal 272 (2005) 120–128 ª 2004 FEBS Mg.ATP at 5 °C, but it did alleviate ATP inhibition. The I 50 for Mg.ATP increased by 2-fold in the pres- ence of 5 mm phosphate and by 5-fold with 20 mm phosphate. However, as was also seen in Table 2, the maximum velocity of PFK at 5 °C was reduced in the presence of phosphate by 49 and 79%, respectively, at 5 and 10 mm phosphate. This effect of phosphate on PFK velocity is shown as a Hanes–Wolf plot in Fig. 3. Phosphate inhibition was found to be noncompetitive with respect to Mg.ATP and the K i was determined to be 0.15 ± 0.01 mm (n ¼ 4). Temperature effects on the activities of PFK from a hibernating (ground squirrel) and nonhibernating (rab- bit) mammal were investigated. Arrhenius plots were constructed from V max values determined in assays run in buffer only (50 mm imidazole) or in buffer plus phosphate (50 mm imidazole plus 10 mm phosphate) under optimal Mg.ATP conditions (0.5 mm) (Fig. 4). The rabbit enzyme showed enhanced activity in the presence of phosphate over the entire temperature range tested (Fig. 4A). In contrast, the ground squirrel enzyme was activated by phosphate at higher tempera- tures, but at temperatures below 9 °C, phosphate reduced enzyme velocity as compared with assays with- out phosphate (Fig. 4B). The temperature relationships of rabbit PFK in both the presence and absence of phosphate exhibited sharp breaks in the Arrhenius plot at 15 °C (Fig. 4A). The activation energy (Ea) values for the 3–15 °C tempera- ture range of the plots were 103.2 ± 0.2 and 80.0 ± 1.2 kJ mol )1 in the absence and presence of phosphate. For the 15–45 °C temperature range, the Ea values were 25.7 ± 0.7 and 26.2 ± 1.1 kJ mol )1 , in the absence and presence of phosphate, respectively. The temperature relationship for ground squirrel PFK was linear from 3 °Cto28°C with an Ea value of 54.1 ± 0.8 kJ mol )1 (n ¼ 4) when assayed in imida- zole under optimal ATP concentrations (Fig. 4B). When assayed in the presence of phosphate under opti- mal Mg.ATP conditions, a very sharp break in the Arrhenius relationship was seen at 12 °C; the Ea value was 43.9 ± 0.2 kJ mol )1 over the range from 12 to 45 °C and rose sharply by 2.8-fold to 122.2 ± 6.1 kJ mol )1 (n ¼ 4) between 2 and 12 °C. Temperature effects on ground squirrel PFK were also assessed under inhibitory (but physiological) con- centrations of ATP (Fig. 4C). In this situation, phos- phate inhibition of ground squirrel PFK occurred at all temperatures below 27 °C. The Ea for the linear portion of the plot (3 °Cto29°C) was 61.9 ± 0.9 kJ mol )1 (n ¼ 4). Temperature had little effect on ground squirrel PFK activity under inhibitory Mg.ATP concentrations when assayed in the absence Fig. 2. The effect of inorganic phosphate on F6P kinetics of S. lat- eralis PFK at 5 °C. PFK activity was measured as described in Materials and methods with phosphate at: 0 m M (j), 5 mM (r), 15 m M (d), or 20 mM (m). Data shown are the result of one trial but are representative of n ¼ 4 determinations from separate pre- parations of enzyme. Table 3. The effect of inorganic phosphate concentration on Mg.ATP kinetics of S. lateralis muscle PFK at two different temperatures. Maxi- mal velocity values are milliunits of enzyme activity obtained for 0.2 lg (at 5 °C) or 0.03 lg (at 37 °C) of purified S. lateralis skeletal muscle phosphofructokinase. The pH values of the assay mixtures were fixed at 23 °C using 50 m M imidazole–HCl buffer and were then allowed to vary with temperature so that the pH at 5 °C was 7.5 and the pH at 37 °C was 7.2. The concentration of F6P was held at 0.5 m M. All other assay conditions are detailed in the Materials and methods. Data are means ± S.E.M., n ¼ 4 separate determinations. Temperature Phosphate (m M)K m Mg.ATP (lM) I 50 Mg.ATP (mM) Maximal Velocity (mU) 5 °C 0 19.6 ± 0.37 2.13 ± 0.22 9.25 ± 0.37 5 17.7 ± 0.23 a 4.70 ± 0.50 a 4.72 ± 0.13 a 10 18.9 ± 0.37 10.7 ± 0.64 a,b 1.92 ± 0.31 a,b 37 °C 0 127.0 ± 6.6 1.29 ± 0.19 7.23 ± 0.51 10 53.7 ± 3.2 a 2.79 ± 0.09 a 7.04 ± 0.18 a Significantly different from the corresponding value at 0 mM phosphate using one-way analysis of variance with the Student–Newman– Keuls test (two-tailed), P < 0.01; b significantly different from the value with 5 mM phosphate. J. A. MacDonald and K. B. Storey Temperature effects on hibernator PFK allostery FEBS Journal 272 (2005) 120–128 ª 2004 FEBS 123 of phosphate; a slight decline in activity was observed with increased temperature. Figure 5 presents a plot of log W vs. 1 ⁄ T, where W is the ratio of maximal velocity with 10 mm phosphate to maximal velocity in the absence of phosphate [17]. Under optimal Mg.ATP levels, the data was character- ized by a roughly linear relationship from 45 °Cto 9 °C with a slightly negative slope ()0.8). The tem- perature at which phosphate had no allosteric effect was 6.6 ± 0.04 °C(n ¼ 4). However, when the enzyme was assayed in the presence of inhibitory levels of Mg.ATP (5 mm), the relationship was characterized by a sharp negative slope ()5.3) that crossed through zero at 28.1 ± 0.1 °C(n ¼ 4). Discussion A reduction in glycolytic rate occurs during hiberna- tion with the putative major site of inhibitory control being phosphofructokinase. Major physiological mani- festations during hibernation that are predicted from the PFK control site include: (a) the suppression of shivering thermogenesis; (b) a shift to a lipid-based metabolism; (c) the conservation of muscle glycogen stores for use during arousal, and (d) the stability of plasma glucose levels [1,3,4]. In hibernators, the aci- dotic conditions found in muscle during hibernation can lead to an increase in histidine protonation and these effects may be translated into alterations in enzyme kinetics and structural properties which in turn inhibit glycolysis [13,18,19]. Considerable work has been completed in an effort to determine the mechan- ism by which regulation of PFK occurs during hibernation. PFK kinetic constants in S. lateralis skel- etal muscle showed no differences when euthermic and hibernating animals were compared [12]. This was in contrast to results obtained from the small hibernators, the meadow jumping mouse (Zapus hudsonius) [20], and the little brown bat (Myotis lucifugus) (K. B. Storey, unpublished data), which exhibited altered PFK kinetic properties during hibernation that were consistent with inactivation by post-translational modi- fication. Another approach focused on temperature and pH dependent shifts in PFK activity and subunit assembly [6,7,21,22]. The effects of acidosis and tem- perature were proposed to cause reversible inhibition of PFK via inactive dimer formation. Previous work in this laboratory [14] has questioned whether physiologi- cally relevant pH changes under simulated in vivo con- ditions of protein crowding affect PFK activity via subunit assembly at hibernating temperatures. The data presented here also conflict with the idea of tem- perature and pH induced tetramer–dimer regulation; in fact, the phosphate inhibition of PFK shown in Fig. 1 at low temperature decreased with decreasing pH. Solute interactions have also been proposed to act synergistically with the pH and temperature interactions detailed above for the regulation of hibernator PFK. Significant inactivation of PFK has been suggested to occur at physiological levels of urea and inactivation is proposed via unfolding of native macromolecules through increased solvent exposure of subunit inter- action sites [6,21]. However, the counteracting solute theory proposed by Somero and coworkers has not been shown to have major effects on the properties of PFK either from estivating [23,24] or hibernating species [14]. Fig. 3. (A) The effect of inorganic phosphate on MgATP kinetics of S. lateralis PFK at 5 °C. PFK activity was measured as described in Materials and methods with phosphate at: 0 m M (j), 5 mM (m), and 10 mM (d). Data shown are the result of one trial but are representative of n ¼ 4 determinations from separate preparations of enzyme. (B) Data replotted as a Hanes–Wolff plot showing the calculated K i for phos- phate inhibition. Inset: Secondary plot of PFK maximal velocity vs. phosphate concentration at 5 °C. Data are means ± SEM, n ¼ 4 separate determinations. Temperature effects on hibernator PFK allostery J. A. MacDonald and K. B. Storey 124 FEBS Journal 272 (2005) 120–128 ª 2004 FEBS Allosteric activators and inhibitors are generally considered to induce or bind to distinctly different enzyme conformations and thereby convey altered functionality to a regulatory enzyme. Allosteric modifi- ers act either by changing the strength of subsequent substrate binding by the enzyme or by changing the activation energy of the enzyme-catalyzed reaction [17]. The tissue-specific responses of PFK in S. lateralis during hibernation illustrated the importance of allo- steric activators in regulating PFK activity in heart and leg muscle [12]. In the case of leg muscle, the con- centration of the potent activator, F2,6P 2 , decreased from a level five times the K a value in euthermic ani- mals to one-half the K a value in hibernating animals, suggesting that F2,6P 2 levels may influence glycolytic rates by directly regulating PFK activity in these tis- sues. Temperature effects on inorganic phosphate allo- stery of PFK may also have a significant contribution to the regulation of glycolytic metabolism in the skel- etal muscle of this hibernating mammal. Fig. 5. Plot of log W vs. 1 ⁄ T summarizing the effect of 10 mM phosphate on V max for S. lateralis PFK under optimal (h) and inhibi- tory (j) Mg.ATP levels. W is the ratio of maximal PFK velocity with 10 m M phosphate to maximal PFK velocity in the absence of phos- phate. Data are means ± SEM, n ¼ 4 separate determinations. Fig. 4. Arrhenius plots of skeletal muscle PFK activity vs. tempera- ture under optimal or inhibitory ATP concentrations. (A) Rabbit PFK measured under optimal substrate conditions: 1.0 m M F6P, 0.5 mM Mg.ATP, 5 mM MgCl 2 ,50mM KCl, and 0.15 mM NADH with buffer (d)50m M imidazole, pH 7.5 at 5 °Cor(s)50mM imidazole +10 m M K 2 HPO 4 ⁄ KH 2 PO 4 , pH 7.5 at 5 °C. (B) ground squirrel PFK measured under the same conditions, and (C), ground squirrel PFK measured under conditions of inhibitory Mg.ATP: 5.0 m M F6P, 5.0 mM Mg.ATP, 5 mM MgCl 2 ,50mM KCl, and 0.15 mM NADH with buffer (d)50mM imidazole, pH 7.5 at 5 °Cor(s) 50 m M imidazole +10 mM K 2 HPO 4 ⁄ KH 2 PO 4 , pH 7.5 at 5 °C. Data are means ± SEM, n ¼ 4 separate determinations. J. A. MacDonald and K. B. Storey Temperature effects on hibernator PFK allostery FEBS Journal 272 (2005) 120–128 ª 2004 FEBS 125 Kinetic findings interpreted for mammalian muscle PFK have indicated the presence of not less than seven substrate, inhibitor and de-inhibitor sites on the enzyme [5]. ATP inhibition of PFK activity is over- come by small increases in ADP, AMP and inorganic phosphate, all of which increase in the cell whenever ATP use exceeds ATP production. This effect is parti- ally expressed in the low temperature F6P kinetics of hibernator PFK; increasing phosphate levels initially result in a hyperbolic shift in the F6P substrate curve but with an unusual decrease in V max . However, fur- ther phosphate addition does not have an effect on substrate affinity and continues to lower V max . A sim- ilar phenomenon was seen with respect to the allosteric influences of phosphate on PFK Mg.ATP kinetics. The effect of phosphate is pH dependent as lowering the pH changes inhibition to activation (Fig. 1). It would appear that phosphate inhibition is induced by low temperature as the addition of phosphate at 37 °C showed strong activation at a pH that showed inhibi- tion at 5 °C. Experimental pH values determined in hibernating skeletal muscle by in vivo 31 P NMR spectroscopy [25] suggest that PFK inhibition could occur under in vivo conditions. The temperature-induced inversion of allosteric phosphate effects is observed only at saturating Mg.ATP concentrations. Unlike temperature-induced changes in Mg.ADP allostery effects seen for F6P kinetics for Bacillus stearothermophilus PFK [17], the effects on hibernator PFK are due to a change in the activation energy of the enzyme-catalyzed reaction induced by the allosteric ligand and not by changes in the extent to which the binding of allosteric ligand modifies the affinity of enzyme for substrate. The Hanes–Wolf plot in Fig. 3(B) conclusively demon- strates that the K m value for Mg.ATP was not affected by increasing phosphate levels at low temperature as the data set was typical of noncompetitive inhibition. Braxton et al. [17] previously defined the effect of an allosteric ligand on V max via the use of Arrhenius plots that graph the ratio of maximal velocities when the allosteric ligand is saturating and when the allosteric ligand is absent. The plot of W vs. 1 ⁄ T for S. lateralis PFK under optimal conditions is relatively horizontal and only crosses below 0 at temperatures less than 7 °C. So, under optimal conditions, the allosteric effect of phosphate is consistent throughout the temperature range investigated and has little effect on V max . How- ever, under inhibitory concentrations of Mg.ATP, the plot of log W vs. 1 ⁄ T is a linear relationship with a sharply negative slope such that log W is equal to zero at 29 °C (Fig. 5). This result demonstrates that tem- perature has pronounced effects on phosphate allostery of PFK with activating effects becoming inhibitory at temperatures less than 29 °C. A comparison of the two relationships indicate that the decrease in maximal activity associated with phosphate is independent of Q 10 effects. Interestingly, PFK from a nonhibernating mammal (i.e. rabbit) lacks the temperature influences on phosphate allostery (Fig. 4A). It should be noted that the inhibitory levels of ATP used are actually within the range of physiological ATP concentrations present in ground squirrel skeletal muscle during hibernation. ATP and other adenylates have been quantified in S. lateralis after a week long hibernation bout; the total adenylate pool decreased in concentration without a corresponding decrease in energy charge. ATP levels dropped by 29% to a hiber- nating value of 2.9 mm [26] so that during hibernation, at least over the short-term, [ATP] appears to remain inhibitory with respect to temperature-induced phos- phate inhibition. Thus, the response of PFK under inhibitory ATP in the presence of added phosphate is perhaps the closest mimic to the in vivo state. Although a 0.3 unit pH difference exists between the 5 °C and 37 °C assay environments, we do not feel that this alone explains the different kinetics observed. It is well known that as PFK is subjected to higher pH values, the S 0.5 for F6P decreases and the enzyme looses allosteric properties [5]; however, our results show that the sigmoidal character of the F6P kinetics is retained at 5 °C at pH 7.5 and indicate that changes in kinetic parameters were most probably due to tem- perature alone. Sensitivity to adenylates (ATP inhibi- tion, AMP and ADP activation) was reduced when the enzyme was assayed at 5 °C, a temperature character- istic of hibernation, as was sensitivity to F2,6P 2 . PFK is generally thought of as being sensitive to two mes- sages: (a) overall cellular energy status, via adenylate and NH 4 + levels, and (b) biosynthetic demands for carbohydrates, via F2,6P 2 signaling that is the way that extracellular hormones also influence PFK (via reversible phosphorylation control over PFK-2, the enzyme that synthesizes F2,6P 2 ). Brooks and Storey [12] observed that the concentration of F2,6P 2 in hibernating golden-mantled ground squirrel leg muscle decreased to 20% of the euthermic value. The 11-fold increase in the K a value for F2,6P 2 at 5 °C when cou- pled with this decrease in tissue F2,6P 2 levels would effectively eliminate regulatory effects on PFK by this allosteric molecule during hibernation. The data also indicate that the enzyme is less likely to be activated by rising AMP and ADP in the torpid muscle (hence, muscle glycolysis is less sensitive to changing energetic state when torpid) and is clearly much less sensitive to biosynthetic signals (F2,6P 2 ) which would help to Temperature effects on hibernator PFK allostery J. A. MacDonald and K. B. Storey 126 FEBS Journal 272 (2005) 120–128 ª 2004 FEBS preserve carbohydrate reserves during torpor. This enzyme is particularly well suited to the metabolic conditions of hibernation as a decreased response to adenylates and to F2,6P 2 as well as the increased cit- rate levels that accompany a switch to fatty acid oxida- tion would all serve to suppress PFK activity and permit carbohydrate sparing during hibernation. In summary, we suggest that temperature-induced alterations in PFK activity via phosphate allostery could be a means by which suppression of PFK activ- ity, and hence glycolytic flux, occurs during metabolic depression in mammalian hibernators. Our results demonstrate that PFK is less likely to be activated in cold torpid muscle, PFK is less sensitive to changing adenylate levels at low temperatures characteristic of torpor, and PFK is clearly much less sensitive to bio- synthetic signals. All of these characteristics of hiber- nator PFK would serve to reduce glycolytic rate and help to preserve carbohydrate reserves during torpor. Materials and methods Animals and chemicals Adult golden-mantled ground squirrels, S. lateralis, were obtained from the Crooked Creek area of the White Moun- tains of California. Details of animal holding, feeding and hibernation were described in [15]. All possible measures were taken to minimise pain and discomfort during animal euthanasia in accordance with protocols approved by the Carleton University Animal Care and Use Committee. Squirrels were killed by decapitation and hind leg skeletal muscle was quickly excised and flash frozen in liquid nitro- gen. Tissues were transported to Carleton University on dry ice and were then stored at )80 °C until use. All bio- chemicals and coupling enzymes were obtained from Boeh- ringer Mannheim (Montreal, PQ) or Sigma Chemical Co. Purification and standard assay of phosphofructokinase PFK (EC 2.7.1.11) was purified from skeletal muscle of euthermic ground squirrels as described previously [14]. Rabbit skeletal muscle PFK was purified following a proce- dure modified from Ramadoss et al. [27]. Purified PFK was used immediately or stored for up to a week in 30% (v ⁄ v) glycerol at )20 °C. PFK activity was measured by a cou- pled enzyme assay [14] and the change in absorbance at 340 nm as a result of NADH consumption was monitored with a Dynatech MR5000 microplate reader with biolynx data capture software. Enzyme activities were analyzed with a microplate analysis program [28] and kinetic param- eters were determined using a simple computer program [29]. Assay temperature was manipulated by using the microplate thermal controller for 37 °C assays or by placing the entire microplate reader into a low temperature incuba- tor for 5 °C studies; in the latter case, thermistors placed in selected wells and attached to a YSI Model 42 SL telether- mometer were used to confirm assay temperature. All reac- tions were initiated with the addition of purified PFK. Standard assay conditions were: 20 mm imidazole-HCl buffer, 5 mm MgCl 2 ,50mm KCl, 0.2 mm F6P, 0.5 mm Mg.ATP, 10 mm 2-mercaptoethanol, 0.15 mm NADH, and 1 U each of aldolase, triosephosphate isomerase and gly- cerol-3-phosphate dehydrogenase. Ammonium sulfate was removed from the coupling enzymes by centrifugation through a small (5 mL) column of Sephadex G-25 equili- brated in 20 mm imidazole-HCl buffer pH 7.2 (at 37 °C) containing 5 mm MgCl 2 and 10 mm 2-mercaptoethanol [30]. Imidazole buffer pH was adjusted at 23 °C to produce pH values of 7.5 or 7.2 at 5 °Cor37°C, respectively; this was calculated assuming a +0.017 unit increase in pH per 1 °C decrease for imidazole buffer [14]. Due to the sensitiv- ity of PFK to minor pH variation, the pH of the inorganic phosphate solutions were also adjusted to be pH 7.5 at 5 °C or pH 7.2 at 37 °C. Protein concentration was determined by the Coomassie blue G-250 dye binding method using the Bio-Rad pre- pared reagent and bovine serum albumin as the standard [31]. Acknowledgements The authors thank Dr Craig Frank, Fordham Univer- sity for supplying the ground squirrel tissues and J. M. Storey for critical commentary on the manuscript. The work was supported by an N.S.E.R.C. Canada discov- ery grant (KBS) and postgraduate scholarship (JAM). JAM is currently the holder of a Protein Engineering Network of Centres of Excellence Chair in Protein Sciences. References 1 Wang LCH & Lee TF (1996) Torpor and hibernation in mammals: metabolic, physiological and biochemical adaptations. In Handbook of Physiology: Environmen- tal Physiology. (Fregley MJ & Blatteis CM, eds), Sec. 4, Vol. 1, pp. 507–532. Oxford University Press, New York, NY. 2 Hochachka PW & Somero GN (1984) Biochemical Adaptation. Princeton University Press, Princeton, NJ. 3 Carey HV, Andrews MT & Martin SL (2003) Mamma- lian hibernation: cellular and molecular responses to depressed metabolism and low temperature. Physiol Rev 83, 1153–1181. 4 Wang, LCH (1989) Ecological, physiological and bio- chemical aspects of torpor in mammals and birds. In J. A. MacDonald and K. B. Storey Temperature effects on hibernator PFK allostery FEBS Journal 272 (2005) 120–128 ª 2004 FEBS 127 Comparative and Environmental Physiology. 4: Animal Adaptation to Cold (Wang LCH, ed.), pp. 361–401. Springer-Verlag, Berlin, Germany. 5 Kemp RG & Foe LG (1983) Allosteric regulatory prop- erties of muscle phosphofructokinase. Mol Cell Biochem 57, 147–154. 6 Hand SC & Somero GN (1983) Phosphofructokinase of the hibernator Citellus beecheyi, temperature and pH regulation of activity via influences on the tetramer- dimer equilibrium. Physiol Zool 56, 280–388. 7 Bock PE & Frieden C (1976) Phosphofructokinase: mechanism of the pH-dependent inactivation and reacti- vation of the rabbit muscle enzyme. J Biol Chem 251, 5630–5636. 8 Bock PE & Frieden C (1976) Phosphofructokinase: role of ligands in pH-dependent structural changes of the rabbit muscle enzyme. J Biol Chem 251, 5637–5643. 9 Benoit MA, Debauche P & Devos PK (1994) Phospho- fructokinase from the posterior gills of the euryhaline crab, Eriocheir sinensis: evidence for its regulation by phosphorylation. J Comp Physiol B 164, 165–171. 10 Brooks SPJ & Storey KB (1991) A quantitative evalua- tion of the effect of enzyme complexes on the glycolytic rate in vivo: mathematical modeling of the glycolytic complex. J Theor Biol 149 , 361–375. 11 Brooks SPJ & Storey KB (1988) Re-evaluation of the ‘glycolytic complex’ in muscle: a multi-technique approach using trout white muscle. Arch Biochem Biophys 267, 13–22. 12 Brooks SPJ & Storey KB (1992) Mechanisms of glyco- lytic control during hibernation in the ground squirrel Spermophilus lateralis. J Comp Physiol B 162, 23–28. 13 Malan A, Rodeau JL & Daull F (1985) Intracellular pH in hibernation and respiratory acidosis in the European hampster. J Comp Physiol B 156, 251–258. 14 MacDonald JA & Storey KB (2001) Reassessment of the cold labile nature of phosphofructokinase from a hibernating ground squirrel. Mol Cell Biochem 225, 51–57. 15 MacDonald JA & Storey KB (2002) Purification and characterization of fructose bisphosphate aldolase from the ground squirrel, Spermophilus lateralis: Enzyme role in mammalian hibernation. Arch Biochem Biophys 408, 279–285. 16 Storey KB & Storey JM (2000) Gene expression and protein adaptations in mammalian hibernation. In Life in the Cold (Heldmaier G & Klingenspor M, eds), pp. 303–313. Springer-Verlag, Heidelberg, Germany. 17 Braxton BL, Tlapak-Simmons VL & Reinhart GD (1994) Temperature induced inversion of allosteric phenomena. J Biol Chem 269, 47–50. 18 Malan, A. (1982) Respiration and acid-base state in hibernation. In Hibernation and Torpor in Mammals and Birds (Lyman CP, Willis JS, Malan A & Wang LCH, eds), pp. 273–282. Academic Press, New York, NY. 19 Malan A (1986) pH as a central factor in hibernation. In Living in the Cold: Physiological and Biochemical Adaptations (Heller HC, Musacchia XJ & Wang LCH, eds), pp. 61–70. Elsevier, New York, NY. 20 Storey KB (1987) Regulation of liver metabolism by enzyme phosphorylation during mammaliam hiberna- tion. J Biol Chem 262, 1670–1673. 21 Hand SC & Somero GN (1982) Urea and methylamine effects on rabbit muscle phosphofructokinase. Catalytic stability and aggregation state as a function of pH and temperature. J Biol Chem 257, 734–741. 22 Somero GN, Lowery MS & Roberts SJ (1991) Com- partmentalization of animal enzymes: physiological and evolutionary significance. Am Zool 31, 493–503. 23 Grundy JE & Storey KB (1994) Urea and salt effects on enzymes from estivating and non-estivating amphibians. Mol Cell Biochem 131, 9–17. 24 Cowan KJ & Storey KB (2002) Urea and KCl have dif- ferential effects on enzyme activities in liver and muscle of estivating versus non-estivating species. Biochem Cell Biol 80, 745–755. 25 McArthur MD, Hanstock CC, Malan A, Wang LCH & Allen PS (1990) Skeletal muscle pH dynamics during arousal from hiberantion measured by 31 P NMR spec- troscopy. J Comp Physiol B 160, 339–347. 26 MacDonald JA & Storey KB (1999) Regulation of ground squirrel Na + K + -ATPase activity by reversible phosphorylation during hibernation. Biochem Biophys Res Commun 254 , 424–429. 27 Ramadoss CS, Luby LJ & Uyeda K (1976) Affinity chromatography of phosphofructokinase. Arch Biochem Biophys. 175, 487–494. 28 Brooks SPJ (1994) A program for analyzing the enzyme rate data obtained from a microplate reader. Biotech- niques 17, 1154–1161. 29 Brooks SPJ (1992) A simple computer program with statistical tests for the analysis of enzyme kinetics. Bio- techniques 13, 906–911. 30 Helmerhorst E & Stokes HH (1980) Microcentrifuge desalting: a rapid, quantitative method for desalting small amounts of protein. Anal Biochem 104, 130–135. 31 Bradford MM (1976) A rapid, sensitive method for the quantification of microgram quantities of protein util- izing the principle of protein-dye binding. Anal Biochem 72, 248–254. Temperature effects on hibernator PFK allostery J. A. MacDonald and K. B. Storey 128 FEBS Journal 272 (2005) 120–128 ª 2004 FEBS . phosphate and activation was seen only at low concentrations with a maximal 1.7-fold activation at 2 mm phosphate. At Table 1. Effects of temperature on the. an anomalous interaction between pH and phosphate effects on the enzyme. At pH 7.0, phosphate acted as a strong acti- vator of PFK and raised maximal activity