Use of hydrostatic pressure to produce ‘native’ monomers of yeast enolase M. Judith Kornblatt 1 , Reinhard Lange 2, * and Claude Balny 2, * 1 Department of Chemistry and Biochemistry, Concordia University, Montreal, Quebec, Canada; 2 INSERM Unite 128, IFR 122, Montpellier, France The effects of hydrostatic pressure on yeast enolase have been studied in the presence of 1 m M Mn 2+ . When com- pared w ith apo-enolase, and Mg-enolase, the Mn-enzyme differs from th e others in three ways. Exposure t o hydro- static pressure does not inactivate the enzyme. If the experiments are performed i n t he presence of 1 m M Mg 2+ , or with apo-enzyme, the enzyme is inactivated [Kornblatt, M.J., Lange R., Balny C. (1998) Eur. J. Biochem 251, 775– 780]. The UV spectra of the high pressure forms of the Mg 2+ - and apo-forms of enolase are identical and distinct from the spectrum of the form obtained in the presence of 1m M Mn 2+ ; this s uggests that M n 2+ remains bound to the high pr essure form of enolase. With Mn-enolase, the various spectral changes do not occur in the same pressure range, indicating that multiple processes are occurring. Pressure experiments were performed as a function of [Mn 2+ ]and [protein]. One of t he changes in th e UV spectra shows a dependence on protein concentration, indicating that eno- lase is dissociating into monomers. The small changes in the UV spectrum a nd the retention of activity lead to a model in which enolase, in t he presence of high concentrations of Mn 2+ , dissociates into native monomers; upon release of pressure, the enzyme isfully active. A lthough f urther spectral changes occur at higher pressures, there is no inactivation as long as Mn 2+ remains bound. We propose t hat the relatively small and polar nature of the subunit interface of yeast enolase, including the presence of several salt bridges, is responsible for the ability of hydrostatic pressure to disso- ciate this enzyme into monomers with a native-like structure. Keywords: dissociation; enolase; hydrostatic pressure; native monomers. Many enzymes normally exist as oligomeric proteins. In some cases, the e nzyme is a regulatory enzyme; allosteric kinetics require multiple subunits. In other cases, the active site is at the interface of the subunits, with two subunits each contributing residues. In many cases, however, it is not obvious what role is played by the oligomeric structure. Attempts to study the relationship between oligomeric state and the structure a nd function of the protein usually involve dissociating the protein into its subunits and then compar- ing the properties of the monomeric and oligomeric forms. Often, the r esulting monomers a re catalytically inactive. Because tertiary and quaternary structure are maintained b y similar forces, agents, such as temperature and chemical denaturants, that disrupt quater nary structure may also disrupt tertiary structure. Thus, when faced with inactive monomers of an active oligomeric protein, it is difficult to know if the conformation of the monomer has b een altered or if the quaternary structure is, i n some way, necessary for activity. Hydrostatic pressure is a useful tool for s tudying p rotein structure a nd function. If an equilibrium system, A Ð B, is subjected to pressure, the equilibrium will be displaced in the direction of t he system that occupies the smaller v olume. In the case of a solution of a protein, hydrostatic pressure may change the conformation, promote binding or dissociation of a ligand, denature the protein or dissociate an oligomeric protein [1–3]. Factors that contribute t o d ifferences in volume between an oligomer and its subunits include removal of p acking defects, hydration of buried surfaces, and disruption of salt bridges. As a general rule (although there are exceptions), the pressure required to dissociate an oligomeric protein is less than that required to denature monomeric proteins. It t herefore seems reasonable t o e xpect that pressure could d issociate oligomeric protein s while having relatively little effect on the secondary and tertiary structure of the resulting monomers. In spite of this expectation, most monomers produced by hydrostatic pressure have been inactive [1]. Yeast enolase (EC 4.2.1.11), which catalyzes the inter- conversion of 2-phosphoglycer ic a cid and pho sphoenol- pyruvate, is a homodimer. High resolution X-ray structures are available for the yeast enzyme [4–6], as well as en olase from lobster [ 7], Escherichia coli [8] and Trypanosoma brucei [9]. Each subunit has two domains; the larger domain is an a/b barrel, while the smaller is a mixture of b-sheet and a-helices. The dimer interface includes two helices in the large domain and two b-strands in the small domain. The active site is at the bottom of the barrel and is totally Correspondence to M. J. Kornblatt, Department of Chemistry and Biochemistry, Concordia University, 7141 Sherbrooke W, Montreal, Que. H4B 1R6 Canada. Fax: +1 514 848 2868, E-mail: judithk@vax2.concordia.ca Enzyme: enolase, EC 4 .2.1.11 *Current address: Universite ´ Montpellier 2, EA3763, Place E uge ` ne Bataillon, 34095 Montpellier cedex 5, France. (Received 7 June 2004, revised 2 August 2004, accepted 6 August 2004) Eur. J. Biochem. 271, 3897–3904 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04326.x contained within the monomer. The small domain of the same subunit contributes a loop which closes over the active site when substrate b inds. Enolase is a metalloenzyme; substrate can not bind unless a divalent cation, usually Mg 2+ or Mn 2+ ,isboundattheactivesite[10]. Hydrostatic pressure [11–13], a nd salts [14–18] have been used to dissociate enolases. Most of these procedures have produced inactive, but folded, monomers. As the active s ite is physically contained w ithin each s ubunit and as the interface contains elements of secondary structure which should be relatively unperturbed by mild treatments, we have been puzzled by our inability t o produce active monomers. Brewer, in his extensive studies on the dissoci- ation of yeast enolase [ 14], dissociated the enzyme by d iluting it into millimolar solutions of EDTA. These monomers, formed in the a bsence of a d ivalen t cation, were inactive. O ur studies using hydrostatic pressure indicated that the Mg 2+ was l ost during p ressure-induced dissociation [12]; these monomers were also inactive. We proposed that, in o rder to maintain the structure of the active site, divalent cation and, perhaps, substrate had to be bound to the enzyme. This article presents a study of the pressure-induced dissociation of yeast enolase in the presence of Mn 2+ .We demonstrate the ability of hydrostatic pressure to produce native monomers – monomers which have undergone no apparent conformational changes and maintain enzymatic activity upon return to low pressure. Materials and methods Buffer for all experiments contained 25 m M Mes, 25 m M Tris, pH 7.1. The pH of this buffer is relatively insensitive to pressure. Mg 2+ or Mn 2+ was present at the stated concentration; EDTA was present at 1/10th the concentra- tion of the Mg 2+ or Mn 2+ . EDTA binds divalent cations such as Co 2+ ,Ni 2+ ,Cu 2+ and Zn 2+ more strongly than it binds Mn 2+ and Mg 2+ . EDTA was added t o t he buffer in the hope of minimizing the free concentration of o ther divalent cations. Yeast enolase ( Sigma) was dialyzed against buffer prior to use. For the experiments that compared apo- enolase with enolase containing Mg 2+ or varying concen- trations o f Mn 2+ , the enzyme was dialyzed against buffer and t hen passed through a small chelex column ( Bio-Rad, Hercules, C A, USA) in order to remove divalent cations. The stated concentrations of EDTA, M g 2+ or Mn 2+ were then added. The concentration of yeast enolase was determined from its absorbance at 280 nm, using, e ¼ 0.9 mLÆmg )1 Æcm )1 [19] and a molecular mass of 94 000 Da. Protein concentrations are expressed as the concentration o f dimeric enzyme. Enzyme activity was measured by following the change in absorbance a t 240 nm, due to the production of phosphoenol pyruvate; the concentration of 2-phosphoglyceric acid (Sigma) i n the assay w as 1 m M . Pressure dissociation and pressure inacti- vation experiments were performed at 15 °C. UV spectra were recorded under pressure, using a Varian Cary 3 spectrophotometer (Varian, Australia) interfaced to a high pressure bomb. After correction for increased absorption due to compression, the 4th derivatives of the spectra were calculated, as described previously [20]. Data collection was optimized in order to maintain spectral quality. When the protein concentration was 9 l M (0.9 mg ÆmL )1 ), spectra were r ecorded in a square cuvette with path length of 1 cm. Samples were scanned from 260 to 305 nm, using a 1 nm band pass, 1 s signal averaging time and a 0.1 nm data interval. At 53 l M (5 mgÆmL )1 ), the same procedures were used except that the p ath length of the cuvette was 0.2 cm. At 2.2 l M (0.2 mg ÆmL )1 ), samples were scanned from 270 to 296 nm, using 5 s signal averaging time. Preliminary experiments were performed in order to determine the pressure range in which spectral changes were occurring and the length of time necessary for equilibrium to be reached. One s ample was used for each complete pressure curve. At low pressures (the r ange where no spectral changes are occurring) the pressure was held fo r 5–10 min, the spectrum was recorded and the pressure then increased to the next value. In the pressure range where spectral changes were occurring, pressure was held for 45 min prior to recording the spectrum. At high pressures (spectral changes are complete), the time at pressure was decreased to 10 min. At the end of the experiment, the pressure was lowered to 5–10 MPa and, after 10 min, a final spectrum was recorded. At 2.2 l M enolase, 22 min was required to record each spectrum; the time at pressure was reduced to 7 min at low a nd high pressur es and 37 min in the region where spectral changes were occurring. In all cases, in the range where c hanges were occurring, the total time at pressure (hold time plus recording time) was at least 53 min. The s pectra were analyzed using four parameters (Fig. 1): (a) the absorbance value at 296 nm, after correction for pressure; (b) D1 ¼ (maximum value of the 4th derivative in the region of 291 nm) – (minimum value of the 4th derivative in the region of 295–296 nm); (c) D2 ¼ (maxi- mum value of the 4th derivative at 287–288 nm) ) (mini- mum value at 283–284 nm); (iv) D3 ¼ (value of the 4th derivative at 276.6 nm) – (value at 279.6 nm). In order to compare different e xperiments and different spectral changes, the values of these parameters were normalized as follows. The low pressure and high pressure values for each parameter were determined from the spectra. Values for intermediate pressures were expressed as a fraction of the total change in that parameter that occurred between low and high pressures. The values of D2 were used to calculate K d for dissociation o f t he dimer into monomers, assuming that the low pressure value of D2is that of dimeric enzyme and the high pressure value of D2 is that of the monomer: K d ¼ 4[eno lase](fraction mono- meric) 2 /(fraction dimeric). As ¶(ln K d )/¶P ¼ –DV/RT,aplotoflnK d vs. press ure gives DV, the volume change for the process, and K d at 0.1 MPa. Pressure inactivation of enolase was measured by subjecting dilute solutions of enolase to pressure for varying times, returning to 0.1 MPa, and immediately assaying the sample for enzymatic activity. Samples used for equilibriu m inactivation experiments contained 0.04 mgÆmL )1 BSA. The albumin was a dded in order to minimize losses of activity due to absorption of enolase to the sides of the cuvette during the 45 min incubation under pressure. Results Exposure of yeast enolase to hydrostatic pressure in the range o f 0.1–240 MPa dissociates the enzyme reversibly into 3898 M. J. Kornblatt et al. (Eur. J. Biochem. 271) Ó FEBS 2004 monomers [11–13]; there is no spectral evidence for denaturation occurring in these samples. There are a number of small changes that occur in the UV spectrum of the protein. Figure 1 shows the zero order UV spectra of the p rotein at low and high pressures a nd the 4th derivatives of the same spectra. Changes in the UV spectra of Fig. 1A, which a re small, are magnified by calculating the 4th derivative; in addition, it is e asier to quantify t he changes in the U V s pectra if one u ses t he 4th derivatives. In our analysis, we use the following spectral characteristics: (a) changes in the zero order spectra at 296 nm; (b) changes in three regions of the 4th derivative o f the spectra, as indicated on Fig. 1B. The changes in these three regions – D1, D2, D3 – were c alculated as d escribed in Material and methods. These spectral changes are fully and rapidly (within 10 min) reversible upon return to 5–10 MPa. The changes in the UV spectra that occur during exposure to hydrostatic pressure indicate that changes are occurring in the environment of at least some of the aromatic residues. This is consistent with earlier work [11] showing changes in the intrinsic fluores- cence emission spectrum of enolase during pressure-induced dissociation. In an earlier study [12], we concluded that Mg 2+ disso- ciates from eno lase during pressure-induced dissociation of enolase into monomers; the resulting monomers were inactive. W e reasoned that the monomers produced by hydrostatic pressure might maintain their native structure and activity if t he divalent cation remained bound. We therefore compared the spectral changes occurring under pressure with Mg 2+ ,Mn 2+ or no divalen t cation present. The s pectra in Fig. 1 w ere recorded in the presence of Mn 2+ . F igure 2 shows the spectr a of the low pres sure (Fig. 2A) and high pressure (Fig. 2B) forms of apo-, Mg 2+ - and Mn 2+ -enolase. The low pressure form is fully dimeric; based on previous experiments, we assume that the high pressure forms are monomeric. T he presence or absence of divalent cation has a small effect on the spectrum of dimeric enzyme. Although the Mg 2+ and Mn 2+ forms have identical spectra at low pressure, they do not have the same high pressure spectra; this is most apparent for the parameter, D1. With the Mn 2+ enzyme, D1 decreases very slightly at high pressure; with the Mg 2+ form, there is a noticeable increase in D1 a t high pressure. Figure 2B shows that the spectra of the apo- and Mg 2+ forms of enolase at high pressure are identical and differ from that of the Mn 2+ enzyme. As the spectrum of Mn 2+ -enolase differs from the other two spectra, we conclude that Mn 2+ remains bound to the high pressure form of the enzyme. Fig. 2. Fourth derivative spectra of the apo-, Mg 2+ and Mn 2+ forms of yeast enolase a t high and low pressures. Enolase was passed through a small chelex colum n and dilut ed, with chelexed buffer, to 10 m M . Additions were then made such that the samples contained 1 m M EDTA (apo-form, solid line), 1 m M Mg 2+ and 0.1 m M EDTA (dotted line) or 1 m M Mn 2+ and 0.1 m M EDTA (dashed line). (A) Spe ctra were recorded at 10 MPa. (B) Spectra were recorded at 220 MPa. Fig. 1. UV spectra o f yeast enolase at high and low pressures. The spectrum of yeast enolase, 10.6 m M , i n Mes/Tris b uffer containing 1m M Mn 2+ and 0.1 m M EDTA was recorded at 10 (continuous line) and 220 (dashed line) MPa. (A) T he zero order spectra, which have been c orrected for the volume ch ange due to pressure. (B) The 4th derivative of the spectra shown in (A); the thr ee arrows indicate the parameters that were used to an alyze the ch anges that occur upon exposure to pressure. Ó FEBS 2004 Native monomers of yeast enolase produced by pressure (Eur. J. Biochem. 271) 3899 If Mn 2+ stays bound to enolase during the exposure to pressure, what happens to enzymatic activity? In order to answer this question, a dilute solution of enolase, 9 n M ,was exposed to pressure for short periods of time. Immediately upon returning to 0.1 MPa, the sample was removed and assayed for enzymatic activity. We were unable to demon- strate inactivation of enolase when the sample contained 1m M Mn 2+ . In the presence of 1 m M Mg 2+ ,1minat 240 MPa caused a 50% loss of activity; with 1 m M Mn 2+ in the sample, no loss of activity was observed at 240 MPa, even when the t ime at pressure w as increased t o 15 min. Either the high pressure form of Mn-enolase is active or its activity is completely recovered within 1.5 min at atmospheric pressure. Pressure exp eriments performed in the presence of 1 m M Mn 2+ differed from th ose in the presence of 1 m M Mg 2+ in another significant manner. When Mg 2+ was the cation, all the spectral changes (D1, D2, D3, absorbance at 296 nm), occurred in the same pressure range. This is also observed with apo-enzyme. When Mn 2+ was the cation, the various spectral changes did not all occur at once (Fig. 3), indicating the existence of multiple processes. In order to determine if any of the spectral changes monitor dissociation, the pressure experiments were performed at three protein concentrations – 2.2 l M ,9.4l M and 53 l M ;allthreewere performed at 1 m M Mn 2+ . Changes in D2showedaclear dependence on protein concentration (Fig. 4A). This indi- cates that exposure to hydrostatic pressure does dissociate the enzyme i nto monomers a nd that D2 monitors the dissociation. In addition, if D2isusedtocalculateK d for dissociation as a function of pressure, data from all three protein concentrations fall on the same line (Fig. 4B). From this data , w e ca lculate t hat K d at 10 MPa is 4.5 · 10 )9 and DV ¼ )120 mLÆmol )1 . Changes in the other parameters, D3 and absorbance at 2 96 (nm) (Fig. 3), begin at h igher pressures and show a much smaller dependence upon protein concentration (not shown). There is sufficient scatter in this data that we cannot say if both of these changes occu r at the same pressure. At a minimum, two steps – dissoci- ation into monomers, and conformational changes in the monomers – are occurring. Changes in D3 and absorbance at 296 nm were shifted to slightly higher pressures with the 53 l M sample (not shown). This suggests that these spectral changes only occur in the monomeric form of the protein. We have now established t hat yeast enolase, in the presence of 1 m M Mn 2+ , is dissociated by pressure and that the spectral parameter D2 is a measure of dissociation. The differences seen in the high pressure spe ctra of F ig. 2B indicate that the Mn 2+ remains bound t o the monomers. Upon release of pressure, the enzyme is fully active. As the Mn 2+ concentration is decreased, the high pressure spectra approach that of apo-enolase (Fig. 5) and inactivation occurs. Does this inactivation parallel dissoci- ation? An equilibrium pressure-inactivation experiment was performedwith6n M enolase a nd 25mM Mn 2+ .Usingthe data shown in Fig. 4B, we can calculate that, at 6 n M ,the Fig. 3. Effects of pressure on the spectral parameters. Asolutionof enolase, 2.2 l M , containing 1 m M Mn 2+ , was subjected to increasing pressure; spectra were recorded, a nalyzed, and normalized as described in Materials and methods. The parameters are D3(s), D2(d)and absorbance at 296 nm (.). Fig. 4. Changes in D2dependonproteinconcentration.(A) A solution of enolase, containing 1 m M Mn 2+ , was subjected to increasing pres- sure; spectra were recorded, analyzed, and n ormalized as de scribed in Materials and meth ods. Th e conc entrations o f en olase are 2.2 lM(s), 9.4 lM(d), 53 lM(.). (B) Th e data shown in (A) were used to calculate K d , as d escribed in Materials a nd methods. Enolase con- centrations were 2.2 l M (s), 9.4 l M (.)and53l M (d). 3900 M. J. Kornblatt et al. (Eur. J. Biochem. 271) Ó FEBS 2004 concentration used for the p ressure-inactivation e xperi- ments, the enzyme would be 95% dissociated by 90 MPa. However, inactivation does not begin until 120 MPa (8% inactivation), with 50% inactivation occu rring a t a bout 170 MPa. D issociation is not accompanied by loss of activity. The loss of activity that occurs at the h igher pressures is r eversible, with 90% o f the initial a ctivity recovered within 12 min at 10 MPa. Pressure dissociates enolase Depending on the identity and concentration of the divalent cation, two different forms of the monomer are produced. One, which we call ÔnativeÕ, has spectral properties almost identical to that of the dimeric enzyme and still has the cation bound. Upon return to 0.1 MPa, it i s fully active. The second has lost the divalent cation, has greater spectral differences and is inactive upon return to 0.1 MPa. We do not know if 1.5 min after return to 10 MPa, the enzyme is still monomeric. At 9 n M enzyme, 95% reassociation within 1.5 min would mean that t he rate constant for reassociation was 1 · 10 7 s )1 Æ M )1 . A lthough this i s fast, it is within the range of observed rate constants for protein–protein reactions [21]. What we do know is that the presence of bound Mn 2+ stabilizes the monomer such that either it is fully active or requires nothing more than reassociation to be active. We will use the term Ônative monomerÕ to refer to the form of the enzyme produced by dissociation under pressure that is fully active on return to 0.1 MPa. Our results can be summarized by the following model (Fig. 6). With Mg 2+ as the divalent cation, dissociation, loss of Mg 2+ , inactivation, and conformational changes in the monomer all occur in one step (step 1). When Mn 2+ is the cation, dissociation occurs (step 2) to p roduce mon omers which still have Mn 2+ bound and are fully active upon return to 0.1 MPa. As the pressure is raised still higher, conformational ch anges occur in the monomer. Depending on the concentration of M n 2+ and the K d of the monomer for Mn 2+ ,Mn 2+ and activity may be retained (step 3) or Mn 2+ may b e lost (step 4), yielding inactive m onomers. As both the empty Mn 2+ site and the free Mn 2+ would be hydrated, it is not surprising that dissociation of the Mn 2+ is promoted by hydrostatic pressure. Based on this model, we predicted that, at high Mg 2+ concentrations, the Mg 2+ form of the enzyme would behave as the Mn 2+ – i.e. the monomers formed initially wouldretainMg 2+ and activity. This prediction has been confirmed. Pressure-inactivation experiments were per- formed as a function of the Mg 2+ concentration. Exposure of 3 n M enolase to 220 MPa for 4 min results in almost complete inactivation of the enzyme w hen t he sample contains 0.45 m M Mg 2+ . If, however, the sample contains 5m M Mg 2+ , there was only a 13% loss of activity. Increasing the time at 220 MPa or increasing the pressure to 260 or 300 MPa did not result in any further loss of activity. Discussion A large number of oligomeric enzymes, including phospho- fructokinase [22], hexokinase [23], lactate dehydrogenase [24], and creatine kinase [25], have been examined by using pressure. I n these examples, and others, pressure both dissociated and inactivated the protein. In addition, recov- ery of activity was a slow process and was ofte n incomplete. We are aware of only two other studies reporting the production by hydrostatic pressure of native monomers. Two of the partial activities of carbamoyl-phosphate synthetase were largely unaffected when the dimeric enzyme was dissociated [26]; in this experiment, assays were begun within 15 s of returning to 0.1 MPa. Based on electropho- resis under pressure and activity staining of the gels, hydroxylamine oxidoreductase is dissociated but not inac- tivated by pressure [27]. We believe that the ability to produce native monomers of enolase depends on at least two factors: the properties of Fig. 5. Changes in spectral parameters as a function of [Mn 2+ ]. Apo- enolase, prepared by passing a sample of enolase through a small chelex column, wa s used to prepare samp les containing varying con - centrations of Mn 2+ . Protein concentration w as 2.2 l M .Spectrawere recorded for each sample after 1 0 min at 10 MPa and after 45 min at 2200 MPa. The fourth derivatives were calculated as described in Materials and methods. The ch anges in spect ral parameters D1(j) and D2(h), are expressed as the ratio of the high pressure to low pressure values. Fig. 6. Model for the effects of hydrostatic pressure on yeast enolase. Species in bold a re enzymatica lly active; monomer and monomer* indicate different conformations of the monomer. Ó FEBS 2004 Native monomers of yeast enolase produced by pressure (Eur. J. Biochem. 271) 3901 the enzyme and the experimental conditions and approach used. Tsai et al. [28] have examined the role of the hydrophobic effect in protein–protein interactions. Although subunit interfaces are more hydrophobic than the exposed surface of the protein, they are less hydrophobic than the interior. In addition, several polar amino acids, especially arginine, lysine, glutamine and glutamate are found more frequently at the interface than in the interior. The degree of hydrophobicity of the i nterface and the percentage surface area buried at the interface v ary from protein to protein; as a general rule, the greater the percentage buried, the greater the degree of hydrophobicity. Both Tsai et al. [28] and Janin et al. [29] suggest that in oligomers with large interfaces, the isolated monomers would be unstable, due to the exposure of the large hydrophobic surface to solvent. The subunit interface of yeast e nolase i s s mall by the criteria of Tsai et al. with only 13% of the surface b uried [4]. In addition, there are a large number of polar groups at the interface, many of which participate in subunit-subunit hydrogen bonds or electro- static interactions. E nolase monomers appear t o be relatively stable under p ressure. E ven the inactive apo- monomers, formed in the presence of low Mg 2+ or Mn 2+ and held a t 240 MPa for 45 min or more, rapidly and completely recover both a ctivity a nd spectral properties upon depressurization. There is no e vidence for irreversi- bility or Ôconformational driftÕ [30]. In the presence of bound divalent cation, the monomers remain native, even after 45 min at 220 MPa. Exposure of a system at equilibrium to increasing hydrostatic pressure will shift t hat equilibrium towards the system that occupies the smaller volume. The native structure of a protein – secondary, tertiary and quaternary structure – reflects a balance between opposing factors. Conformational entropy disfavors the native structure, while van der Waals interactions, electrostatic inter- actions, hydrogen bonds and hydrophobic interactions are favorable. Hydrostatic p ressure, by r educing the s ize of internal cavities, decreases flexibility of the protein core. At the same time, portions of the p rotein near the surface become more flexible since pressure promotes hydration of the protein [31]. Electrostatic inte ractions are disrupted by pressure, with a DVof)17 to )35 mLÆmol )1 [32]; this volume change is d ue to electrostriction of water around charged groups. Hydrogen bond disruption has a small, positive DV; these bonds are s trengthened a nd diversified by pressure [33–36]. The direction and magnitude of the volume change for disruption o f h ydrophobic bonds is still under debate [3,37]. Creating active monomers of an oligomeric enzyme may require selectively disrupting those interactions that maintain quaternary structure without perturbing those that maintain tertiary structure. According to the crystal structures o f yeast enolase, there are two glutamate and two arginine residues per subunit that form salt bridges with the two arginine and two glutamates on the other subunit. Subunit interactions in yeast enolase are not very strong; in the presence of 1m M Mn 2+ , K d is 4.5 · 10 )9 and DV ¼ )120 mLÆmol )1 (Fig. 5). Given the large negative volume change for disruption of salt bridges, t he pressure-induced dissoci- ation of yeast enolase may be driven primarily by disruption of these interactions. If the monomer of an enzyme maintains the same secondary and tertiary structure it had in the oligomer, how will dissociation b e detected? Standard techniques for monitoring changes in size, such as gel filtration, fluorescence polarization or dynamic light scattering, are not widely used. As a result, most pressure studies focus on conditions in which major spectral changes are occurring; smaller c hanges occurring in lower ranges of pressure are often not examined. Interpreting the se small changes is complicated by the fact that l ow pressure may affect the structure of an oligomeric protein without causing dissociation [38]; similarly, pressure can produce changes in spectra and activity of monomeric enzymes in the absence of denaturation [39]. W e were fortunate to find a spectral change that monitored dissociation of enolase, and to find conditions in which the monomer was stable. Although the observed changes in the UV spectrum of yeast enolase are small, they provide information on the changes that occur during e xposure to pressure. The first parameter to change D2, shows a clear d ependence upon protein concentration, indicating that D2 is monitoring dissociation of t he protein into monomers. In the 4th derivative of the U V spectrum, the region of D2, 282– 288 n m, contains contributions from both t yrosine a nd tryptophan residues [20]. Simulations of the enolase spectra, using standard spectra of tyrosine and tryptophan in various solvents, s how t hat t he c hanges i n D2thatare produced by pressure are due to changes in the environ- ment of tyrosine residues. In an earlier study [13], we proposed that the observed decrease in the polarity of the environment of the tyrosine residues was due to two residues which point into a cleft, between the subunits, that is filled w ith immobilized water. U pon dissociation, the water w ould no l onger be immobilized and its average polarity would decrease. As far as we can tell from the UV and fluorescence spectra (not shown), nothing else is happening to the p rotein at pressures b elow 150 MPa. The enzyme is b eing dissociated into monomers which maintain their native conformation. A comparison of t he results o f apo-enolase [ 12] with those of enolase in t he presence of 1 m M Mg 2+ [12], low (50 l M ) and high (1 m M )Mn 2+ , gives the following picture: (a) The presence of divalent cations stabilizes the dimeric s tructure of enolase, as has been demonstrated previously [14]. This implies that Me 2+ binds more tightly to the dimer than to the monomer. (b) The dimeric structure stabilizes the conformation of enolase; we do not observe changes in t he spectra until dissoci- ation occurs. (c) The presence of divalent cations also stabilizes the monomer of enolase. The stabilizing effect of the divalent c ation is not a unique property of Mn 2+ , but is also observed with Mg 2+ . We can now begin to explain the role of the dimeric structure of y east enolase. The dimeric structure stabilizes the structure of the monomer and favors the binding of divalent cations, which in turn stabilize the dimer. We find it difficult to believe that there are not other dimeric proteins that could b e d issociated by pressure into native monomers. Although the same forces are involved in maintaining tertiary a nd quaternary structure, in many cases they will not make the same relative 3902 M. J. Kornblatt et al. (Eur. J. Biochem. 271) Ó FEBS 2004 contributions to both levels of structure. We believe the key points for success are threefold: (a) finding conditions that stabilize the monomer without excessively stabilizing the dimer. This m ay require exploring a range of temperatures, pH, ion concentrations, etc; (b) examining with care the pressure range in which major spectral changes are not occurring and (c) using, under pressure, techniques s uch as fluorescence polarization [ 11] or dynamic light scattering that are direct measures of the size of a protein. Acknowledgements We thank C onc ordia Un iversity f or the sabbatical l eave during which time these experiments were performed, and J. A. Kornblatt for encouragement. Financial support was provided by the N atural Sciences and Engineering Research Council of Canada and INSERM. References 1. Gross, M. & Jaenicke, R. (1994) Proteins under pressure. Eur. J. Biochem. 221, 617–630. 2. Mozhaev, V.V., Heremans, K., Frank, J., Masson, P. & Balny, C. (1996) High pressure e ffects on protein structure and function. Proteins 24, 81–91. 3. Boonyaratanakornkit, B.B., Park, C.B. & Clark, D.S. (2002) Pressure e ffects on in tra- and intermole cular interactions within proteins. Biochim. Biophys. Acta 1595, 235–249. 4. Stec, B. & Lebioda, L., (1990) Refined structure of y east apo- enolase at 2.25 A ˚ resolution. J. Mol. Biol. 211, 235–248. 5. Zhang, E., Brewer, J.M., M inor, W., Carreira, L.A. & Lebio- da, L. ( 1997) Mec hanism of enolase: the crystal structure of asymmetric dimer enolase-2-phospho- D -glycerate/enolase-phos- phoenolpyruvate at 2.0 A ˚ resolution. Biochemistry 36, 12526– 12534. 6. Wedekind, J.E., P oyner, R.R., Reed, G.H. & Rayment, I. (1994) Chelation o f serine39 to Mg 2+ latches a gate at t he active site of enolase: structure of the bis (Mg 2+ ) complex of yeast e nolase a nd the i ntermediate analog p hosphono acetohydro xamat e at 2.1 A ˚ resolution. Biochemistry 33, 9333–9342. 7. Duquerroy, S., Camus, C. & Janin, J. (1995) X-ray structure and catalytic mechanism of lobster enolase. Biochemistry 34, 12513– 12523. 8. Kuhnel, K. & Luisi, B.F. (2001) Crystal structure of the Escher- ichia coli RNA degradoso me co mpon ent en olase. J. Mol. Biol. 313, 583–592. 9. da Silva Giotto, M.T., Hanna ert, V., d e Vertomm en, D.A.S., Navarro, M.V., Rider, M.H., Michels, P.A.M., Garratt, R.C. & Rigden, D.J. (2003) T he crystal structure of Trypanosoma brucei enolase: visualisation of t he inhibitory metal binding site III and potential as target for selective, irreversible inhibition. J. Mol. Biol. 331, 653–665. 10. Brewer, J.M. ( 1985) Specificity and mechanism of action of meta l ions in yeast enolase. FEBS Lett. 182, 8–14. 11. Paladini, A.A. & Weber, G., (1981) Pressure-induced reversible dissociation of enolase. Biochemistry 20, 2587–2593. 12. Kornblatt, M.J., Lange, R. & Balny, C., (1998) Can monomers of yeast enolase have enzymatic activity? Eur. J. Biochem. 251, 775–780. 13. Kornblatt, J.A., Kornblatt, M.J. & Hui Bon Hoa, G. (1995) Second derivative spectroscopy of enolase at high hydrostatic pressure: an approach to the study of macromolecular inter- actions. Biochemistry 34, 1218–1223. 14. Brewer, J.M., Faini, G.J., W u, C.A., Goss, L.P., Carreira, L.A. & Wojcik, R. (1978) Characterization of the sub unit dissociation of yeast enolase. I n Physical Aspe cts of P rotein Interactions (Catsimpoolas, N., ed.), pp. 57–78. Els evier/North-Holland, Amsterdam. 15. Trepannier, D ., Wong, C. & Kornblatt, M.J. (1990) The salt- induceddissociationandinactivationofamammalianenolase: evidence for the formation of active monomers. Arch. Biochem. Biophys. 283, 271–277. 16. Kornblatt, M.J., Al-Ghanim, A . & Kornblatt, J.A. (1996) The effects of sodium perchlorate on rabbit muscle enolase: spectral characterization of the monomer. Eur. J. Biochem. 236, 78–84. 17. Marangos, P.J., Parma, A.M. & Goodwin, F.K. (1978) Func- tional properties of ne uronal a nd glia l isoenzymes of b rain enolase. J. Neuroc hem . 31, 727–732. 18. Chorazyczewski, J., Danielewicz, R. & Wolny, M. (1987) Immobilization of pig muscle enolase. Studies on the activity of subunits. Acta Biochim. Polon. 34, 421–430. 19. Chin, C.C.Q., Brewer, J.M. & Wold, F., (1981) The a mino acid sequence of yeast enolase. J. Biol. Chem. 256, 1377–1384. 20. Lange, R., Frank, J., Saldana, J L. & Balny, C. (1996) Fourth derivative UV-spectroscopy of proteins under high p ressure. I. Factors affecting the fourth derivative spectrum of the aromatic amino acids. Eur. Biophys. J. 24, 277–283. 21. Fersht, A. (1999) Structure and Mechanism in Protein Science. W.H. Freeman, New York. 22. Deville-Bonne, D. & Else, A. (1991) Reversible high hydrostatic pressure inactivation of phosphofructokina se from Escherichia coli. Eur. J. Biochem. 200, 747–750. 23. Ruan, K. & Weber, G. (1988) Dissociation of yeast hexokinase by hydrostatic pressure. Biochemistry 27, 3295–3301. 24. Mu ¨ ller, K., Lu ¨ demann, H D. & Jaenicke, R. (1982) Thermo- dynamics and mechanism of high-pressure deactivation and dissociation of p orcine lactic dehydrogenase. Biophys. Chem. 16 , 1–7. 25. Zhou, J M., Zhu, L. & Balny, C. (2000) Inactivation o f creatine kinase by high pressure may precede dimer dissociation. Eur. J. Biochem. 267, 1247–1253. 26. Guy, H.I., Schmidt, B., Herve ´ , G. & Evans, D.R. (1998) Pressure- induced dissociation o f carbamoyl-pho sphate synthetase do mains. J. Biol. Chem. 273, 14172–14178. 27. Masson, P., Arciero, D.M., Hooper, A.B. & Balny, C. (1990) Electrophoresis at elevated hydr ostatic pressure of the multih eme hydroxylamine oxidoreductase. Electrophoresis 11, 128–133. 28. Tsai, C J., Lin, S.L., Wolfson, H.J. & Nussinov, R. (1997) Studies of p rotein–prote in interfaces: a statistical an alysis of the hydro- phobic effect. Protein Sci. 6, 53–64. 29. Janin, J., Miller, S. & Clothia, C., (1988) Surface, subunit inter- faces and in terior of oligomeric proteins. J. Mol. Biol. 204, 155–164. 30. Silva, J.L., Miles, E.W. & Weber, G. (1986) Pressure dissociation and conformational drift of th e beta dimer of tryptophan syn- thase. Biochemistry 25 , 5780–5786. 31. Cioni, P. & Strambini, G.B. (1994) Pre ssure effects on protein flexibility monomeric p roteins. J. Mol. Biol. 242, 291–301. 32. Heremans, L. & Heremans, K. (1989) Raman s pectrosc opic study of the changes in secondary structure of chymotrypsin: effect of pH and p ressure on the salt bridge. Biochim. Biophys. Acta 999, 192–197. 33. Kitchen, D.B., R eed, L.H. & Levy, R.M. (1992) Molecular dynamics simulation of solvatedproteinathighpressure.Bio- chemistry 31, 10083–10093. 34. Li, H., Yamada, H. & Akasaka, K. (1998) Effect of pressure on individual hyd rogen bonds in proteins. Basic pancreatic trypsin inhibitor. Biochemistry 37, 1167–1173. Ó FEBS 2004 Native monomers of yeast enolase produced by pressure (Eur. J. Biochem. 271) 3903 35. Kundrot, C.E. & Richards, F.M. (1988) Effect of hydrostatic pressure o n the solvent in crystals of hen egg-white lysozyme. J. Mol. Biol. 200, 401–410. 36. Paci, E. (2002) High pressure simulations of biomolecules. Bio- chim. Biophys. Acta 1595, 185–200. 37. Royer, C.A. (2002) Revisiting volume changes in pressure-induced protein unfolding. Biochim. Biophys. Acta 1595, 201–209. 38. Cioni, P. & Strambini, G.B. (1996) Pressure e ffects on the structure of oligomeric proteins prior to subunit dissociation. J. Mol. Biol. 263, 789–799. 39. Fujiwara, S., Kunugi, S., Oyama, H. & Oda, K. (2001) Effects of pressure on the act ivity and spectroscopic properties of carboxyl proteinases. Eur. J. Biochem. 268, 645–655. 3904 M. J. Kornblatt et al. (Eur. J. Biochem. 271) Ó FEBS 2004 . Use of hydrostatic pressure to produce ‘native’ monomers of yeast enolase M. Judith Kornblatt 1 , Reinhard Lange 2, *. the pressure- induced dissociation of yeast enolase in the presence of Mn 2+ .We demonstrate the ability of hydrostatic pressure to produce native monomers