Effects of the G376E and G157D mutations on the stability of yeast enolase – a model for human muscle enolase deficiency Songping Zhao*, Bonny S. F. Choy* and Mary J. Kornblatt Department of Chemistry and Biochemistry, Concordia University, Montreal, Canada Enolase (EC 4.2.1.11), an essential enzyme of glyco- lysis and gluconeogenesis, catalyses the intercon- version of 2-phosphoglyceric acid (PGA) and phosphoenolpyruvate. Enolases from most species are dimeric, with subunit molecular masses of 40 000– 50 000 Da. Mammals have three genes for enolase, coding for the a, b and c subunits; the subunits asso- ciate to form both homo- and heterodimers. The a gene is expressed in many tissues, c primarily in neurones and b in muscle. In 2001, the first human case of enolase deficiency was reported [1]. The affected individual showed reduced levels of enolase activity in the muscles. Western blot analysis showed the presence of normal levels of aa-enolase, but no detectable bb-enolase. This individual was heterozy- gous for the gene for b-enolase, and carried two mis- sense mutations, one inherited from each parent. His muscle cells synthesized two forms of b-enolase, each carrying a different mutation. These mutations chan- ged glycine at position 374 to glutamate (G374E) and glycine at position 156 to aspartate (G156D). In order to study the effects of each of these mutations on the structure and function of enolase, we have made the corresponding changes, G376E and G157D, in yeast (Saccharomyces cerevisiae) enolase. We chose to work with yeast enolase, not bb-enolase, as yeast enolase has been extensively studied, a number of crys- tal structures are available [2,3] and the recombinant Keywords muscle enolase; mutations; proteolysis; stability; subunit interactions Correspondence M. J. Kornblatt, Department of Chemistry and Biochemistry, Concordia University, 7141 Sherbrooke Street, W., Montreal, QC, Canada, H3G 1A7 Fax: +1 514 848 2868 Tel: +1 514 848 2424, ext 3384 E-mail: judithk@alcor.concordia.ca *These authors contributed equally to this work (Received 12 September 2007, revised 1 November 2007, accepted 5 November 2007) doi:10.1111/j.1742-4658.2007.06177.x The first known human enolase deficiency was reported in 2001 [Comi GP, Fortunato F, Lucchiari S, Bordoni A, Prelle A, Jann S, Keller A, Ciscato P, Galbiati S, Chiveri L et al. (2001) Ann Neurol 50, 202–207]. The subject had inherited two mutated genes for b-enolase. These mutations changed glycine 156 to aspartate and glycine 374 to glutamate. In order to study the effects of these changes on the structure and stability of enolase, we have introduced the corresponding changes (G157D and G376E) into yeast enolase. The two variants are correctly folded. They are less stable than wild-type enolase with respect to thermal denaturation, and both have increased K d values for subunit dissociation. At 37 °C, in the presence of salt, both are partially dissociated and are extensively cleaved by trypsin. Under the same conditions, wild-type enolase is fully dimeric and is only slightly cleaved by trypsin. However, wild-type enolase is also extensively cleaved if it is partially dissociated. The identification of the cleavage sites and spectral studies of enolase have revealed some of the structural differ- ences between the dimeric and monomeric forms of this enzyme. Abbreviations AUC, analytical ultracentrifugation; MES, 2-(N-morpholino)ethanesulfonic acid; PGA, 2-phosphoglyceric acid; PhAH, phosphonoacetohydroxamate; Q-TOF, quadrupole time-of-flight; s 20,w , sedimentation coefficient at 20 °C in pure water; TLCK, N-a-tosyl- L-lysine chloromethyl ketone. FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 97 yeast enolase can be overexpressed and purified in quantity [4,5]. The basic three-dimensional structure of the monomeric unit is the same in all enolases crystallized to date [6–10]. Yeast and bb-enolase share 79% sequence similarity. All residues that have been described as being involved in subunit interactions (salt bridges and hydrogen bonds) [6], or contributing to the active site, are conserved [2–5]. These two gly- cines are conserved and are in highly conserved regions of the protein: G376E is in a totally con- served sequence of 25 amino acids, whereas G157D is in a loop in which 11 of the 15 residues are con- served. In view of the structural similarities between yeast and mammalian enolases and the high degree of sequence conservation, we believe that the effects of these mutations on the structure and function of yeast enolase will be similar to their effects on human bb-enolase. Figure 1 shows the basic fold of yeast enolase, including the location of the active site and glycines 157 and 376. Comi et al. [1] reported that the levels of mRNA for b-enolase were normal, and suggested that the lack of b-enolase protein in the muscle could be the result of improper folding and assembly, which, in turn, would lead to increased proteolysis of the pro- tein. In this article, we focus on the structure and stability of these variants relative to wild-type yeast enolase. As the substitution of glutamate or aspartate for glycine is nonconservative, we introduced alanine at these two positions, with the aim of determining whether any of the observed effects were a result of changes in the size and charge of the amino acid at these positions. Results Preliminary characterization The G157D and G376E mutations were successfully introduced into the gene for yeast enolase; sequencing of the plasmids confirmed that the desired mutations were present and that no other changes had been introduced. The variant proteins were expressed in Escherichia coli and purified. As a result of the low activity of both variants under standard assay condi- tions, SDS-PAGE was used to monitor the purifica- tion. Typical yields of pure protein, from a 4 L culture, were 350 mg for wild-type enolase, 80 mg for G157D and 100 mg for G376E. All enolases were highly pure, as judged by SDS-PAGE (not shown). The specific activities of the variants, relative to wild- type enolase, were 0.1% (G157D) and 0.01% (G376E). MS confirmed that the desired mutations were present (data not shown). Secondary and tertiary structure CD was used to examine the structure of these pro- teins. In the peptide bond region, there were no signifi- cant differences between wild-type enolase and the variants, indicating that the variants were folded cor- rectly. However, there were significant differences in the aromatic region. The spectrum of the G157D vari- ant was very similar to that of the wild-type protein (Fig. 2A) in the region 280–300 nm; however, there were differences in intensity below 280 nm. The CD spectrum in the aromatic region of the G376E variant was markedly different from that of the wild-type eno- lase (Fig. 2A). Quaternary structure and K d value for dissociation Analytical ultracentrifugation (AUC) was used to examine the quaternary structure of the variants. Wild-type enolase and the G157D variant had the same sedimentation coefficient at 20 °C in pure water (s 20,w ) at both 10.6 and 1.06 lm, indicating that they were both dimeric at these concentrations (Table 1). How can we determine the K d values for the wild-type and G157D variant? Previous experiments have shown that the incubation of yeast enolase in NaClO 4 results in the dissociation of the protein into monomers, as indicated by changes in s 20,w (see below). There was no loss of CD signal in the peptide bond region (210– 230 nm), indicating that the protein was not being unfolded. However, there were large changes in the Fig. 1. Yeast enolase (1one.pdb). The product, phosphoenolpyru- vate, and glycines 157 and 376 are space-filled. The G376E and G157D mutations in yeast enolase S. Zhao et al. 98 FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS CD signal in the aromatic region (260–300 nm) (Fig. 2B), as well as small changes in the UV spectrum (data not shown) and a loss of activity. All of these processes appeared to be complete by 0.3 m NaClO 4 . The changes in the aromatic CD spectrum and in s 20,w were used to calculate the K d value as a function of NaClO 4 . For wild-type enolase, the K d value, extrapo- lated to 0 m NaClO 4 (Fig. 3), was (1.5 ± 0.3) · 10 )8 . The K d value for the G157D variant, determined in a similar experiment, was increased by a factor of 10 relative to the wild-type (Table 1). Based on the AUC data, the mutation at position 376 had a major effect on the quaternary structure (Table 1). The s 20,w value for the G376E variant was measured at four protein concentrations and the K d value for this variant was calculated; K d was increased by a factor of 10 3 (Table 1). Thermal denaturation Temperature stability was studied by monitoring the loss of the CD signal at 222 nm (Fig. 4). Both variants were less stable than the wild-type by 4–5 °C (Table 1). The wild-type and G157D variant were both stabilized Fig. 2. CD spectra of wild-type yeast enolase and its variants in the aromatic region. Spectra were normalized to 10 l M protein. All samples are in TME buffer. (A) Full line, wild-type; short broken lines, G376E; long broken lines, G157D. (B) Full line, wild-type; long broken lines, wild-type in 0.3 M NaClO 4 ; short broken lines, W56F. The W56F sample contained 0.5 m M PhAH in order to ensure that the protein was fully dimeric. Table 1. K d and T m values of wild-type enolase and variants. Enolase s 20,w a K d (M) T m (°C) e Wild-type 5.61 ± 0.023 (1.5 ± 0.3) · 10 )8b 55.4 G367E 4.51 ± 0.14 (1.4 ± 0.3) · 10 )5c 51.2 G157D 5.65 ± 0.02 (1.8 ± 0.4) · 10 )7d 49.9 a Average and standard deviations of two (G376E), three (G157D) or four (wild-type) determinations; the standard deviation for individ- ual deviations was < 0.01. b From perchlorate dissociation experi- ment using AUC and CD data (Fig. 4). c Determined from AUC data at four protein concentrations. d From perchlorate dissociation experiment using CD data. e Mid-point of curves shown in Fig. 4. Fig. 3. K d value for dissociation of wild-type enolase by NaClO 4 . Enolase was incubated in varying concentrations of NaClO 4 and then analysed by CD and AUC, as described in Experimental proce- dures. Open circles, K d based on the CD signal at 284 nm; filled cir- cles, K d based on s 20,w . S. Zhao et al. The G376E and G157D mutations in yeast enolase FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 99 by the presence of 50 lm phosphonoacetohydroxamate (PhAH), with T m increasing by 11.6 °C (wild-type) and 7.6 °C (G157D). PhAH, even at 0.5 mm, increased the T m value for the G376E variant by only 1 °C. Proteolytic digestion The susceptibility of the two variants to limited prote- olysis was examined under physiological conditions: 0.15 m KCl, 37 °C. Trypsin cleaved both variants; under the same conditions, but little cleavage of the wild-type occurred (Fig. 5). Further experiments were performed at 15 °C, without the addition of KCl. Under these conditions, at a trypsin ⁄ enolase ratio of 1 : 1000, cleavage of the G376E variant was complete within 60 min at 15 °C, producing three fragments. Under the same conditions, there was partial cleavage at one site by chymotrypsin, but no cleavage by pep- sin, endoproteinase Glu-c or elastase (data not shown). The cleaved samples were analysed by quadrupole time-of-flight (Q-TOF) MS and the cleavage sites were identified (Table 2). At 15 °C, there was no cleavage of the wild-type enolase or of G157D by either trypsin or chymotrypsin. Does enolase become susceptible to cleavage by trypsin when it is partially dissociated? Table 3 sum- marizes the results of several experiments performed to determine whether there is a correlation between disso- ciation and susceptibility to cleavage by trypsin. For all three forms of enolase, cleavage by trypsin occurs when there is measurable dissociation. Shifting the equilibrium towards monomers (37 °C and KCl for G157D, NaClO 4 for the wild-type) promotes cleavage. Shifting the equilibrium towards dimers (G376E plus 0.5 mm PhAH) provides substantial protection against proteolysis. MS confirmed that the fragments produced Fig. 4. Thermal denaturation of wild-type enolase and its variants, as monitored by the CD signal at 222 nm. Full line, G376E at 10.6 l M; long broken lines, wild-type at 10.6 lM; short broken lines, G157D at 5.3 l M; broken lines–dots, wild-type at 5.3 lM. Fig. 5. SDS-PAGE analysis of tryptic digests. Samples were incu- bated at 37 °C in the presence (+) or absence of trypsin for 30 min; proteolysis was stopped by the addition of TLCK All sam- ples contained 0.15 M KCl, except for lane 8, which contained 0.15 M NaClO 4 . Lane 1, molecular mass markers of 97, 66, 45, 31, 21 and 14 kDa. Lanes 2 and 3, G157D. Lanes 4 and 5, G376E. Lanes 6, 7 and 8, wild-type. Table 2. Masses of fragments formed by cleavage of the G376E variant of yeast enolase. Following the proteolysis of enolase, the fragments were analyzed by Q-TOF MS. Enzyme Mass of fragment (Da) Cleavage site Trypsin 5282, 11 515, 29 982 49–50, 329–330 Chymotrypsin 6098, 40 662 56–57 Table 3. Relationship between dissociation and cleavage by tryp- sin. Enolase (wild-type and mutants) was incubated in the stated conditions and then analysed by AUC or subjected to limited pro- teolysis by trypsin. Enzyme and conditions Dimeric (%) a Cleavage by trypsin? G376E, 15 °C, TME 70 Yes G376E, 15 °C, TME, 0.5 m M PhAH 96 Slight G157D, 15 °C, TME 100 No Wild-type, 15 °C, TME 100 No Wild-type, 15 °C, TME, 0.3 M NaClO 4 0 Yes G376E, 37 °C, TME, 0.15 M KCl 50 Yes G157D, 37 °C, TME, 0.15 M KCl 88–90 Yes Wild-type, 37 °C, TME, 0.15 M KCl 100 Very slight a Based on s 20,w . The G376E and G157D mutations in yeast enolase S. Zhao et al. 100 FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS during the trypsin digest of wild-type monomers were the same as those produced from the G376E variant. Is monomeric enolase cleaved because it is partially unfolded? Although the CD spectra of monomeric and dimeric wild-type enolase appear to be identical in the region 205–240 nm, differences become apparent when spectra are recorded at lower wavelengths (Fig. 6). These spectra were analysed using dichroweb [11,12], with the variable selection method (cdsstr) [13]. According to this analysis, the percentage of unordered structure in enolase increases from 17% in the dimeric protein to 21% in the monomeric form. Three pairs of monomeric and dimeric enolases, including enzyme from two separate purifications, were examined. All three showed a 4% increase in unordered structure on dissociation. Origin of the CD signal A CD difference spectrum for the wild-type enzyme (monomeric enolase – dimeric enolase) resembled the spectrum of tryptophan, with a major peak at 284 nm and shoulders at 274 and 291 nm. This spectrum sug- gests that, on dissociation, there is a major change in the environment of one or more tryptophans. The only tryptophan near the interface is residue 56. This resi- due was changed to phenylalanine. The aromatic CD spectra of the dimeric forms of the wild-type enolase and the W56F variant are shown in Fig. 2B. The major change in the CD spectrum of the wild-type enolase, which is seen on dissociation, is mimicked by a loss of W56. Discussion The two mutations (G157D and G376E) were success- fully introduced into yeast enolase. The resulting vari- ant proteins had the correct secondary structure, based on the CD spectra in the peptide bond region. Both variants could bind Mg 2+ and substrate, as evidenced by their enzymatic activity. They also bound PhAH, a tight binding inhibitor [14], as evidenced by the effects of this compound on thermal denaturation, proteolysis and subunit dissociation. Both mutations clearly desta- bilized the protein towards thermal denaturation and decreased subunit affinity. The K d value for dissocia- tion of the subunit was increased by approximately 10 3 for the G376E variant: at 1 mgÆmL )1 and 15 °C, the protein was partially dissociated. The K d value for the G157D variant was also increased, but by a smaller amount, such that the protein was dimeric under our standard conditions of 1 mgÆmL )1 and 15 °C. Physio- logical conditions of ionic strength and temperature promoted the dissociation of both variants. This is not surprising, as it has been reported that both salt [15] and increasing temperature [16,17] favour dissociation of the wild-type enolase. Conditions which promoted dissociation also promoted proteolysis by trypsin. The initial observation that we are trying to under- stand is the lack of any bb-enolase in the muscle of the patient [1]. We recognize that yeast enolase is not identical to bb-enolase and that the cytoplasm of mammalian cells does not contain trypsin or chymo- trypsin. However, the results with the G376E and G157D variants of yeast enolase show that these muta- tions destabilize the protein and result in partial disso- ciation. If, in muscle cells, the monomer is recognized as abnormal and is degraded, the proteolysis of the monomer would continually shift the dimer–monomer equilibrium towards monomer, until all the enolase had been degraded. As the yeast enolase and its vari- ants were expressed in E. coli at 37 °C, the significantly lower yield of the variants may also be the result of dissociation followed by proteolysis of the monomers. The effects of these mutations on temperature stabil- ity is not surprising. Brewer et al. [18,19] have pre- pared a number of variants of yeast enolase and, in many cases, the introduction of mutations has decreased the temperature stability. Although a Fig. 6. Peptide bond CD of the monomeric and dimeric forms of wild-type yeast enolase. Enolase (10.6 l M) was in the usual buffer, except that Tris was titrated with H 2 SO 4 , not HCl; the monomeric form of enolase was prepared by incubating the enzyme in buffer plus 0.3 M NaClO 4 overnight at 15 °C. The spectra were recorded at 15 °C using a 0.01 cm path length cuvette. Full line, dimeric eno- lase; broken line, monomeric enolase. S. Zhao et al. The G376E and G157D mutations in yeast enolase FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 101 number of variant forms of yeast enolase have been produced by various groups, in almost no cases have the effects on subunit interactions been examined. Replacing glycine by glutamate or aspartate is a very nonconservative change, although one that occurs in nature as it requires only a single base change. Are the effects that we have observed a result of a change in the size or charge of the amino acid? Variants with alanine at these positions were also studied. For tem- perature denaturation, any change at these positions was destabilizing. G376A and G376E had identical T m values. At position 157, alanine had a smaller effect than aspartate, but even alanine decreased the T m value by 4 °C. There was no correlation between the degree of dissociation and T m . Under the conditions used for thermal denaturation, the G376E variant was 30% monomeric, whereas the G376A variant was 100% dimeric; however, their T m values were identical and about 4 °C lower than that of the wild-type. A different picture emerged when dissociation was stud- ied, at least for G376E. In this case, alanine had little or no effect on the K d value; G376A, at 1 mgÆmL )1 , had the same s 20,w value as wild-type enolase and was not cleaved by trypsin. How do these changes in amino acids decrease sub- unit interactions? Glycine 376 is in a small loop: resi- dues 373–381. This loop includes glutamate 379, whose side chain is hydrogen bonded to the side chain of asparagine 410 in the other subunit, and glutamate 377 and threonine 378, both within 4.0 A ˚ of the other sub- unit (Fig. 7). Glycine 376 is close to residues arginine 14, serine 403 and glutamate 404, all of which are involved in subunit interactions. Introducing a large, negatively charged amino acid at this position would probably change the positions of some of these side chains, thereby weakening interactions between the subunits. Mutations at position 373 also increase the subunit dissociation constant [19]. Glycine 157 does not seem to be close to residues involved in subunit interactions. However, it is also in a loop and the w ⁄ / angles at position 157 are in a region of the Rama- chandran plot that is allowed only for glycine. Substi- tuting any amino acid at this position would result in a change in the conformation of this loop. Changes in the conformation of the backbone at this point and changes in the orientation of other side chains, as a result of the introduction of the large charged aspar- tate residue, would undoubtedly have subtle effects on other residues that are involved in subunit interactions. During the course of this study, it was observed that both variants showed reduced enzymatic activity rela- tive to the wild-type enolase. This is not surprising, considering the location of these changes. Glycine 157 is in one of the loops that moves on binding of sub- strate and divalent cation. This loop contains histidine 159, which is essential for catalysis. Nearby residues that contribute to the stabilization of one of the transi- tion states include 152, 155 and 168 [20]. Glycine 376 is close to residues 373 and 374, which are also impor- tant for the reaction [20]. How does dissociation into monomers promote pro- teolysis? Studies on these variants have revealed some interesting differences between the monomeric and dimeric forms of enolase. Trypsin cleaves at arginine 49. This residue is in a long loop (residues 36–60), most of which is on the surface of the protein. How- ever, this residue points into the protein, is surrounded by other amino acids and is not accessible to trypsin. On dissociation, there must be significant changes in the conformation of this loop, leading to the exposure of arginine 49. The chymotrypsin cleavage site, between residues 56 and 57, is also in this loop. Tryp- tophan 56 is surrounded by residues from both mono- mers, and the backbone amide at position 56 of one subunit is hydrogen bonded to the side chain of gluta- mate 188 of the other subunit. Therefore, it is not sur- prising that it is not accessible in the dimer. The identification of the 56–57 bond as a site that is hidden in the dimer, but accessible to chymotrypsin in the monomer, leads to the question of whether trypto- phan 56 contributes to the large changes in aromatic CD that are observed on dissociation. As shown in Fig. 2B, the aromatic CD spectrum of the fully dimeric W56F variant is very similar to that of monomeric wild-type enolase. In the wild-type dimer, tryptophan Fig. 7. The subunit interface of wild-type enolase (1ebh.pdb). Those atoms of subunit B that are within 4.0 A ˚ of subunit A are shown as a surface. Loop 373–381 of subunit A is shown as sticks with Corey–Pauling–Koltun colouring; G376 is space-filled. Resi- dues 14, 403 and 404 of subunit A are shown as yellow sticks. The G376E and G157D mutations in yeast enolase S. Zhao et al. 102 FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 56 is relatively immobile and has a strong, negative CD signal. On dissociation, tryptophan 56 becomes mobile, resulting in a loss of this negative signal. Yeast enolase can also be dissociated by hydrostatic pressure or by KCl plus EDTA. In both cases, dissociation is accompanied by changes in intrinsic fluorescence of the protein [15,21,22], which are probably caused by a change in the environment of this residue. Loop 36–60 includes a highly mobile region, residues 37–41, which folds over the active site on binding of substrate and divalent cation. We have observed differ- ences, at positions 56 and 49, between the dimeric and monomeric forms of enolase. The loss of activity observed in most studies [23–25] on dissociation of enolase may be the result of changes in other regions of this same loop. In a study of the pressure dissocia- tion [26], we demonstrated that pressure dissociation and inactivation of yeast enolase is a multistep process. The first step, dissociation of the dimer into mono- mers, is accompanied by small changes in the UV spec- trum of the protein, changes which were attributed to changes in the environment of several tyrosine resi- dues. This is followed by conformational changes in the monomer, which are reflected in further spectral changes (both absorbance and fluorescence) and a loss of activity. Based on our current data, we now propose that the transition between the initial active monomers formed by pressure and the subsequent inactive mono- mers is a result of changes in the conformation of loop 36–60, changes similar to those observed in the current experiments. The other bond cleaved by trypsin is between resi- dues 329 and 330. This bond is located in the last turn of a small a-helix and far from the subunit interface. We have no idea why it becomes susceptible to cleav- age. We do not know whether the small increase in disordered structure, observed in the CD spectrum, affects this part of the protein, or whether there is transient unfolding of the end of this helix. There are examples of helices in proteins that undergo transient unfolding, unfolding that is not apparent from the crystal structure [27]. However, in neither case is it obvious why this region of the enolase monomer would be affected. Comi et al. [1] suggested that the lack of bb-protein in the subject’s muscle was a result of improper folding and assembly of the dimer, leading to increased prote- olysis. Our results indicate that the two variants are correctly folded and form normal dimers. However, because of the increased values of K d for subunit disso- ciation, both variants are partially dissociated; it is the presence of the monomeric form of enolase that leads to the increased proteolysis. Experimental procedures Oligonucleotides were obtained from BioCorp Inc. (Mon- treal, Canada), restriction enzymes from MBI Fermentas (Burlington, Canada), CM-Sepharose and Q-Sepharose from Amersham (Piscataway, NJ, USA) and phosphoenolpyru- vate from Roche (Basel, Switzerland). PGA was prepared enzymatically by either of two methods: (a) phosphoenol- pyruvate was converted to PGA enzymatically, following the procedure of Shen and Westhead [28] with minor modi- fications [29]; or (b) PGA was synthesized enzymatically from ATP and glyceric acid [30]. PhAH was synthesized according to Anderson and Cleland [14]. The plasmid containing the d-glycerate-2-kinase gene was a gift from G. Reed (University of Wisconsin, Milwaukee, WI, USA). A plasmid containing the gene for yeast enolase 1 (ENO1) was a gift from T. Nowak (University of Notre Dame, Notre Dame, IN, USA). The enolase gene was removed from this plasmid and inserted into pET-3a. XL1- Blue E. coli was used for the storage of plasmids containing the enolase genes (mutant or wild-type); BL21(DE3) E. coli was used for the expression of the protein. Mutagenesis was performed using the QuickChange method (Stratagene, La Jolla, CA, USA). The primer sequences were as follows: 5¢-GG GGT GTT ATG GTT TCC CATCGA TCT GAA GAA A CT GAA GAC (G376E) and 5¢-CCA TTC TTG AAC GTT TTA AAC GGT GAT TCC CAC GCT GGT GG (G157D). Each sequence differs from that of the wild-type in two ways (the bases changed are indicated in italic type): (a) a glycine codon was changed to either a glutamate or aspartate; and (b) silent mutations were introduced that produced new restriction sites. These sites, BglII for G376E and AhaIII for G157D, were used for screening purposes following mutagenesis of the gene. Ala- nine was introduced at these positions using the same strat- egy. DNA sequencing was performed by BioS&T, Inc. (Lachine, Canada). The expression of enolase was performed as described previously [4]. The cell paste was either used immediately or stored at )20 °C. Cell paste from 4 L of cells was sus- pended in 60 mL of TME buffer [50 mm Tris ⁄ HCl, pH 7.4, 1mm Mg(OAc) 2 and 0.1 mm EDTA] containing 1 mm phosphonoacetic acid and about 3 mg each of DNase and RNase. The suspension was sonicated, on ice, using six 30 s bursts per 10 g of cell paste. The suspension was cooled on ice for 30–60 s between bursts. The pH was adjusted to 7.4 using 1 m Tris base and the sonicated cell suspension was centrifuged at 24 000 g for 30 min at 4 °C. The supernatant was decanted and recentrifuged at the same speed for another 30 min. All subsequent steps were performed on ice or in a cold room. Enolase was precipitated between 40 and 85% (NH 4 ) 2 SO 4 ; the precipitated protein was dia- lysed against TME buffer and applied to a column of Q-Sepharose Fast Flow resin equilibrated in the same buffer. Enolase binds very weakly to this resin under these S. Zhao et al. The G376E and G157D mutations in yeast enolase FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 103 conditions and is slowly eluted with TME buffer. Based on the specific activity, SDS-PAGE and the ratios of absor- bance at 280 and 260 nm, the enolase was often highly pure at this step. The protein was precipitated in 4.3 m (NH 4 ) 2 SO 4 and stored at 4 °C as the precipitate. If further purification was necessary, the enzyme was pre- cipitated as above, centrifuged to collect the protein and di- alysed against 20 mm 2-(N-morpholino)ethanesulfonic acid (MES), pH 6.0, containing 2 mm Mg(OAc) 2 and 0.2 mm EDTA. It was then applied to a CM-Sepharose Fast Flow column equilibrated in MES buffer. Enolase was eluted by a gradient of 0–0.25 m KCl in the same buffer. The purified enolase was precipitated and stored as described above. Purification of the G157 variants was identical to that of the wild-type enzyme, except that the initial (NH 4 ) 2 SO 4 cut was 60–95%. For the G376E mutant, the (NH 4 ) 2 SO 4 cut was 50–85% and the order of the chromatography steps was reversed. The enzyme was first applied to the CM- Sepharose column and eluted with the KCl gradient. Fol- lowing precipitation of the pooled fractions by 4.3 m (NH 4 ) 2 SO 4 and dialysis against TME buffer containing 0.1 m NaCl, the enzyme was applied to a small (about 2 mL bed volume) column of Q-Sepharose and eluted with the same buffer. The purified enzyme was precipitated and stored as described above. During purification of the wild-type enolase, the enzyme activity was monitored by following the conversion of phosphoenolpyruvate to PGA at 244 nm. The buffer con- tained 50 mm imidazole, pH 7.1, 250 mm KCl, 1 mm Mg(OAc) 2 and 0.1 mm EDTA. The specific activities of the purified enzymes were measured in the same buffer by fol- lowing the conversion of PGA to phosphoenolpyruvate at 240 nm. During purification of the mutant enolases, the chromatography steps were monitored by absorbance at 280 nm and by SDS-PAGE of column fractions. Protein concentrations were measured at 280 nm; e = 8.46 · 10 4 m )1 cm )1 [26]. Sedimentation velocity experiments were performed in a Beckman (Fullerton, CA, USA) XL-I analytical ultracen- trifuge at the Concordia University Centre for Structural and Functional Genomics. Samples were prepared in TME buffer, containing 0.3 m Na(OAc), unless stated otherwise. Samples were centrifuged at 12 800 g at 15 °C, unless stated otherwise, and monitored at either 280 or 230 nm, depending on the protein concentration. Data were analysed using dcdt+, version 1.15 or 2.02 (J. Philo, www.jphilo.mailway.com); the viscosity and density of this buffer were determined by sednterp, version 1.07 (D. B. Hayes, T. Laue and J. Philo, available at www.bbri.org/ RASMB/rasmb.html). In order to determine the sedimen- tation coefficients of dimeric and monomeric enolase, measurements were made over a range of protein concen- trations. In the presence of 0.14 m Na(OAc), enolase is fully dimeric; s 20,w , over a range of 2.1–0.056 mgÆmL )1 , was constant, with an average value of 5.49 ± 0.16. Simi- larly, in the presence of 0.3 m NaClO 4 and protein con- centrations ranging from 1.22 to 0.122 mgÆmL )1 , s 20,w was also constant, with an average value of 3.35 ± 0.13. For enolases that were partially dissociated, the concentration of dimeric enzyme was calculated from the total protein concentration and the s 20,w value [31]: s w ¼ðs M ½Mþ2s D ½DÞ=ð½Mþ2½DÞ ð1Þ where s W , s M and s D are the s 20,w values for the sample, monomer and dimer, respectively. The concentrations of dimeric and monomeric enzyme were then used to calculate K d . MS was performed on a Q-TOF instrument at the Con- cordia University Centre for Biological Applications of Mass Spectrometry. CD spectra were recorded on a Jasco (Easton, MD, USA) J-810 spectropolarimeter, with a ther- mostatically controlled sample compartment. When spectra were being recorded, samples were scanned from 320 to 250 nm (aromatic region) or 260 to 200 nm (peptide bond region) at 20 nmÆs )1 , with a 1 nm bandwidth and a 1 s response time. A minimum of four scans was averaged; baseline subtraction and smoothing were performed using jasco software. For temperature denaturation studies, the sample was monitored at 222 nm. The temperature was increased at a rate of 15 °C per hour. The CD signal was used to calculate the fraction unfolded: f U ¼ðy F À yÞ=ðy F À y U Þð2Þ where y F and y U are the CD signals at 222 nm for the ini- tial and final forms of the protein, respectively [32] and y is the signal of the sample. Samples for all CD experiments were in TME buffer, unless otherwise stated. The protein concentration was either 0.5 or 1.0 mgÆmL )1 ; in any given experiment, mutant and wild-type enolases were at the same concentration. The K d value for subunit dissociation of wild-type and G157D enolases was determined using NaClO 4 to dissociate the enzyme. Samples were incubated in TME buffer con- taining varying amounts of Na(OAc) and NaClO 4 , such that the total salt concentration was 0.3 m. After incubation at 15 °C for 24 h, the CD spectra in the aromatic region were recorded with the sample compartment at 15 °C. The spectrum of the enzyme in 0.3 m Na(OAc) was taken as that of the fully dimeric enzyme. As the spectral changes were complete by 0.3 m NaClO 4 , the spectrum of this sam- ple was assumed to be that of fully monomeric enolase. For each sample, the CD signal at 284 nm was used to calculate the fraction dissociated. These data were used to calculate K d : K d ¼ 4½enolaseðf M Þ 2 =ðf D Þð3Þ where f M and f D are the fractions of monomeric and dimeric enzyme, respectively. A plot of K d versus [NaClO 4 ] gives K d at 0 m NaClO 4 . The K d value for wild-type enolase was also determined, using the same experimental design, The G376E and G157D mutations in yeast enolase S. Zhao et al. 104 FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS but measuring the s 20,w value of each sample, in addition to recording the CD spectrum. The data were then analysed as described above. Samples of enolase (1 mgÆmL )1 ) were incubated with N-a-tosyl-l-lysine chloromethyl ketone (TLCK)-treated trypsin (Sigma, St Louis, MO, USA) at a trypsin ⁄ enolase ratio of 1 : 1000. At varying times, aliquots were removed, an excess of TLCK (Roche) was added, followed by SDS- PAGE sample buffer. Samples were then boiled for 2 min and analysed by SDS-PAGE, using a 12% separating gel. A similar protocol, without addition of an inhibitor, was used with other proteolytic enzymes. Figs 1 and 7 were cre- ated using pymol (http://pymol.sourceforge.net). Acknowledgements We thank P. Ulycznyj (Concordia Centre for Struc- tural and Functional Genomics) for running many of the analytical ultracentrifugation samples, A. Padovani for making the W56F variant and J. A. Kornblatt for encouragement and advice. Financial support was provided by the Natural Sciences and Engineering Research Council of Canada. References 1 Comi GP, Fortunato F, Lucchiari S, Bordoni A, Prelle A, Jann S, Keller A, Ciscato P, Galbiati S, Chiveri L et al. (2001) b-Enolase deficiency, a new metabolic myopathy of distal glycolysis. Ann Neurol 50, 202–207. 2 Wedekind JE, Poyner RR, Reed GH & Rayment I (1994) Chelation of serine 39 to Mg 2+ latches a gate at the active site of enolase: structure of the bis(Mg 2+ ) complex of yeast enolase and the intermediate analog phosphonoacetohydroxamate at 2.1-A ˚ resolution. Biochemistry 33, 9333–9342. 3 Lebioda L & Stec B (1991) Mechanism of enolase: the crystal structure of enolase-Mg 2+ -2-phosphoglycer- ate ⁄ phosphoenolpyruvate complex at 2.2-A ˚ resolution. Biochemistry 30, 2817–2822. 4 Poyner RR, Laughlin LT, Sowa GA & Reed GH (1996) Toward identification of acid ⁄ base catalysts in the active site of enolase: comparison of the properties of K345A, E168Q, and E211Q variants. Biochemistry 35, 1692–1699. 5 Vinarov DA & Nowak T (1999) Role of His159 in yeast enolase catalysis. Biochemistry 38, 12138–12149. 6 Stec B & Lebioda L (1990) Refined structure of yeast apo-enolase at 2.25-A ˚ resolution. J Mol Biol 211, 235–248. 7 da Silva Giotto MT, Hannaert V, Vertommen D, de AS Navarro MV, Rider MH, Michels PA, Garratt RC & Rigden DJ (2003) The crystal structure of Trypano- soma brucei enolase: visualisation of the inhibitory metal binding site III and potential as target for selec- tive, irreversible inhibition. J Mol Biol 331, 653–665. 8 Chai G, Brewer JM, Lovelace LL, Aoki T, Minor W & Lebioda L (2004) Expression, purification and the 1.8 A ˚ resolution crystal structure of human neuron specific enolase. J Mol Biol 341, 1015–1021. 9 Kuhnel K & Luisi BF (2001) Crystal structure of the Escherichia coli RNA degradosome component enolase. J Mol Biol 313, 583–592. 10 Duquerroy S, Camus C & Janin J (1995) X-ray struc- ture and catalytic mechanism of lobster enolase. Biochemistry 34, 12513–12523. 11 Lobley A, Whitmore L & Wallace BA (2002) DICHROWEB: an interactive website for the analysis of protein secondary structure from circular dichroism spectra. Bioinformatics 18, 211–212. 12 Whitmore L & Walllace BA (2004) DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res 32, W668–W673. 13 Manavalan P & Johnson WC (1987) Variable selection method improves the prediction of protein secondary structure from circular dichroism. Anal Biochem 167, 76–85. 14 Anderson VE & Cleland WW (1990) Phosphonate ana- logue substrates for enolase. Biochemistry 29, 10498– 10503. 15 Brewer JM & Weber G (1968) The reversible dissocia- tion of yeast enolase. Proc Natl Acad Sci USA 59, 216– 223. 16 Keresztes-Nagy S & Orman R (1971) Dissociation of yeast enolase into active monomers. Biochemistry 10, 2506–2508. 17 Holleman WH (1973) The use of absorption optics to measure dissociation of yeast enolase into enzymati- cally active monomers. Biochim Biophys Acta 327, 176–185. 18 Brewer JM, Robson RL, Glover CV, Holland MJ & Lebioda L (1993) Preparation and characterization of the E168Q site-directed mutant of yeast enolase 1. Proteins 17, 426–434. 19 Brewer JM, Glover CV, Holland MJ & Lebioda L (1997) Effect of site-directed mutagenesis of His373 of yeast enolase on some of its physical and enzymatic properties. Biochim Biophys Acta 1340, 88–96. 20 Liu HY, Zhang YK & Yang WT (2000) How is the active site of enolase organized to catalyze two different reaction steps? Journal of the American Chemical Society 122, 6560–6570. 21 Paladini AA & Weber G (1981) Pressure-induced reversible dissociation of yeast enolase. Biochemistry 20, 2587–2593. 22 Brewer JM, Bastiaens P & Lee J (1987) Investigation of conformational changes in yeast enolase using dynamic S. Zhao et al. The G376E and G157D mutations in yeast enolase FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS 105 fluorescence and steady-state quenching measurements. Biochem Biophys Res Commun 147, 329–334. 23 Kornblatt MJ, Al-Ghanim A & Kornblatt JA (1996) The effects of sodium perchlorate on rabbit muscle eno- lase – spectral characterization of the monomer. Eur J Biochem 236, 78–84. 24 Kornblatt MJ, Lange R & Balny C (1998) Can mono- mers of yeast enolase have enzymatic activity? Eur J Biochem 251, 775–780. 25 Brewer JM, Faini GJ, Wu CA, Goss LP, Carreira LA & Wojcik R (1995) Characterization of the subunit dis- sociation of yeast enolase. In Physical Aspects of Pro- tein Interactions (Catsimpoolas N, ed.), pp. 57–78. Elsevier, North-Holland, Amsterdam. 26 Kornblatt MJ, Lange R & Balny C (2004) Use of hydrostatic pressure to produce ‘native’ monomers of yeast enolase. Eur J Biochem 271, 3897–3904. 27 Horn JR, Kraybill B, Petro EJ, Coales SJ, Morrow JA, Hamuro Y & Kossiakoff AA (2006) The role of protein dynamics in increasing binding affinity for an engi- neered protein–protein interaction established by H ⁄ D exchange mass spectrometry. Biochemistry 45, 8488– 8498. 28 Shen TYS & Westhead EW (1973) Divalent cation and pH-dependent primary isotope effects in the enolase reaction. Biochemistry 12, 3333–3337. 29 Kornblatt MJ & Klugerman A (1989) Characterization of the enolase isozymes of rabbit brain: kinetic differ- ences between mammalian and yeast enolases. Biochem Cell Biol 67, 103–107. 30 Sims PA & Reed GH (2005) Method for the enzymatic synthesis of 2-phospho-D-glycerate from adenosine 5¢- triphosphate and D-glycerate via D-glycerate-2-kinase. J Mol Catal, B Enzym 32, 77–81. 31 Cole JL (1996) Characterization of human cytomegalo- virus protease dimerization by analytical centrifugation. Biochemistry 35, 15601–15610. 32 Pace CN & Scholtz JM (1997) Measuring the confor- mational stability of a protein. In Protein Structure (Creighton TE, ed.), pp 299–321. IRL Press, Oxford. The G376E and G157D mutations in yeast enolase S. Zhao et al. 106 FEBS Journal 275 (2008) 97–106 ª 2007 The Authors Journal compilation ª 2007 FEBS . Effects of the G376E and G157D mutations on the stability of yeast enolase – a model for human muscle enolase deficiency Songping Zhao*, Bonny S conservation, we believe that the effects of these mutations on the structure and function of yeast enolase will be similar to their effects on human bb -enolase.