./. Mol. Hiol. (1991) 220, 531-538
Relation Between Stability, Dynamics and Enzyme Activity in 3-Phosphoglycerate Kinases from Yeast and Thermus thermophilus Paul G. Varley and Roger H. Paint Department of Biochemistry and Genetics The University of Newcastle upon Tyne, NE2 4HH, (Received
11 January
1991; accepted 11 April
[J.K.
1991)
3-Phosphoglycerate kinases from yeast and the extreme thermophilic bacterium Thermus thermophilus HB8 have been used as models for investigating t’he relationship between stability, dynamics and activity. It was found that while at a given temperature the thermophilic protein is more stable, its conformational dynamics as measured by the ability of acrylamide to quench the fluorescence of a buried tryptophan as well as its specific activity, are both lower than for the mesophilic protein. As the temperature is increased. the thermodynamic stability of the thermophilic protein approaches that of the mesophilit: protein at its working temperature. Its conformational dynamics and specific activity however were both shown to increase, until at the physiologically operational temperature, they become similar to those of the mesophilic enzyme at its operational temperature. These results confirm the proposal that a direct relationship and balance holds between thermodynamic stability, dynamics and specific activity in globular proteins. They demonstrate also the constraining effect of increased stability upon conformational dynamics and enzyme activity. Keywords:
3-phosphoglycerate
kinase; stability; dynamics; thermophilic proteins.
1. Introduction Comparison of the properties of mesophilic proteins and of their more stable counterparts from thermophilic bacteria has provided a natural approach to the investigation of protein stability (Amelunxen & Murdock, 1978). Comparison of the sequences and structures of these proteins has shown that extra stability is obtained by increasing the number of non-covalent interactions that stabilize all proteins (Perutz & Raidt, 1977; Harris et al., 1980). Improvement of packing and elimination of cavities is also suggested to increase stability (Kellis et al., 1988). While it has been generally assumed that the conformational dynamics of a particular protein are conserved under the conditions under which it operates in oivo and even at lower temperatures (Daniel et al.. 1982), the additional interactions of thermophilic proteins are expected to make them more rigid than their mesophilic counterparts at a given t,emperature. This functionally adverse effect of t Author to whom all correspondence should he addressed.
fluorescence
lifetime;
stability on the conformational dynamics of t,he protein has been observed in hydrogen exchange experiments (Wiithrich & Wagner. 1979; Delpierre et al., 1983; Wrba et al., 1990), as well as flexibility indices computed from crystallographic data (Vihinen, 1989). The catalytic activity of enzymes is widely believed to be dependent upon the conformational dynamics of the protein (Huber, 1979: Wrba et al., 1990). The conservation of enzyme activity within a family of proteins operating under widely different physiological environments would therefore be expected to be accompanied by a conservation of conformational dynamics at their respective physiological operating conditions. The aim of t,he work presented here is to provide further firm experimental evidence for the relationship between st’ability, dynamics and activity in 3-phosphoglycerate kinase (PGKf: EC 2.7.2.3) an enzyme $ Abbreviations used: PGK. 3-phosphoglyceratr kinase; thermus, Thermus thermophilus HBX: SDS/PAGE. SDS/polyacrylamide gel electrophorrsis: GAPDH. 3-phosphat,e dehydrogenase; P(:A 3-phosl”hoglyceric acid.
utilizing a propostd hii~gt, bending mec*hanisnr (H>wks rt cd., 1979; Wat,son d al.. 198%). The proteitts studied here are the mesophilic WK from yeast, and the PGK from t,he extreme thermophile ‘I’hwmus thermophilus IIH8 (thermus). stabilit,iw of both having heen well characterized (Nojima rt nl.. 1!)77: Adams rf d.. 1985: Griko et cd.. 1989). Steatfy-state fluorescence and fluorescence decay kinetics provides particularly useful methods t,o study and compare the dynamics of each protein sinc*cathe fluorescence of tr?ptophan is highly sensit iw to its surrounding enwronment and has dway lifetimes of the order of nanosewnds, which is the time-wale of wzymatically important, fluctuations in protSeins (Lakowicz, 1983). The kinetic rate of ac*rvlamide quenching of fluorescence depends upon the st,ructural fluctuations of a protein that allow thr penetration of the quencher into its interior. ac*c:ording to the mobile defect model (Lumry & Rosenberg. 1975). Measurement of this rate t,herefort, provides information about the level of tlynamics around the fluorophorr on itn enzymatically important, time-walr (Eftink & Ghiron, 1976). Thw-mus PGK is well suit.ecl to such a. study sinw it cont,ains only orw tryptophan in a buried rnvironmtant bttween tht. two Rossrnan folding units in the (‘-t,erminal domain (Kowcn et nl.. 1987). PUK from veast. however, contains two tryptophans at posiiions 333 (equivalent to that. in thermus PC:K) and 308, which is exposed t)o solvent (M’at,son 4 01.. 1952). cwmplicating the int,erpretation of fluoreswnw data (Dry&w Hr Pain. 1989). Rewnt.l? howrvw. we havck shown that thr> fiuorwwnw of thcb c~sposed FV308 ca,n be selectively quenched with snc4nimitlr while leaving the hurirtl M’333 utlaffwtrtl (Varlry d cd.. 1991). thus allowing dircxc-t. (.ornpa.rison of t.hc flnorescrnw and henw tlynarnics of’ the two proteins. This, combinrtl with cwmparc tiw measurements of the specific: activities and the st.abilities of the two proteins. provitlrs a m(‘ans of examining directly t,he relat’ionships betwrrn dynamicas. st,ability and nctivity.
2. Materials
Thermus EY:K wax purified according to thr rnethoci described by Lit~tlwhild rt al. (I 987). with a yield of .1-Smg of’ PGK per kg of cells. ‘irast IXK. owrexpwssed ill yrast (Wilson rt ((1.. 1987). gab purified wcwrding to t hc mrt,hotl dewribed by Scopw (1969). tSoth proteins gavts single bands on SI)S/Pr\(:lC. l~nltw xtat)ed. all c,titatnirals were Analar grade or equivalent from 13I)H, I’oolr. IT.K.
JWK activit?- was rneasurrtl bv a motlific:atiorl of a method described by Scopes (197.i). The assay mixtuw (I ml) wntained 10 m.v-Tris. H(‘I (pH 7.2). I ~~YI-EI)TA. 5 mwmagneaium chloridr. I6 rn~l-l-S-l)hosphogl~(~~ri(, acid (WA) (Sigma. Poole. V.K .). 6 mw.ATI’ (Jrmostatted to *WI ‘C’. Spwific activities wercl c.atcaulatetl assuming that I unit of PGK will ~ww the oxidation of t P.M of N’ADH,irnin and using a value of .A1g,,j= 0.5 at 280 nrn for thermus (t,ittlc child et r/J.. 1987) and for yeast PC:K (Adarns st al.. 1985). Thtx purified prot)eins from ytlast and thermus hatl spwitic. acativitiss at ,.> *)r ’ ( ’ of II00 units trig-l and 2.50 units trig- I. rrspectively. which are in good agrrrrnrnt with t,htl \-slurs previously published. The protrins were shah II t,cj tw ovc’r !X50,,,activr in I.1 41-suc,c,ininiict(,,’I wwrytarnitlc
Fluoresc~wc~r~ tlxperiment s were pc~rfi~rrrrrti in a CO1 wTris. 0.1 M-h’a(‘l. OW~I wE:I)TA buffer atijustrd to pH 7.2 with HC’I and filtered (022 pm Miltiporr) hrforcl use. Fluorescence quenching was carried out by the addition of 5 wwrylamide (IGc~tran grade. BI)H) to protrin
Table 1 Thumo.stability
and Methods
and speci$c
nctizdy
Stability,
Dynamics
and lhzyme
533
Actioity
l’hr himolecular quenching rate constants. k, are listed with the Stern-\‘olmer constant for the qwn~*hinp of PGK by acr$amidr at a series of temperat,urrs. Each ralur of k, is calculated using tht= average of the 2 fluorescence lifetimes (t), weighted for the fractional intensities (d). and thr Stern-Volmer constant (FL,,). Values are quoted as averages with standard errors of thr mean. c~alculatrtl from 6 determinations. rn
solutions of absorbance WI to W% at 295 nm. The recorded fluorescence intensity was adjust’ed for dilution and acr,vamide absorbanw using a molar extinction roefficient for
acrylamide of 0.%X? ml mg-’ cm-’ at 295 nm (Eftink & Ghn-on. 1981). For double quenching experiments with the yeast enzyme, hoth qurnrhrr and protein solutions I.1 >I-sucscinimidv (Analar. treated with vontainrd achtivatrd charcoal). mea,surements \ver’r nlatlt~ using a Steady-stat.e Perkin-hlmrr MI’F4-I, sprctrofluorimetc,r with an es
Relation Between Stability, Dynamics and Enzyme Activity in 3-Phosphoglycerate Kinases from Yeast and Thermus thermophilus Paul G. Varley and Roger H. Paint Department of Biochemistry and Genetics The University of Newcastle upon Tyne, NE2 4HH, (Received
11 January
1991; accepted 11 April
[J.K.
1991)
3-Phosphoglycerate kinases from yeast and the extreme thermophilic bacterium Thermus thermophilus HB8 have been used as models for investigating t’he relationship between stability, dynamics and activity. It was found that while at a given temperature the thermophilic protein is more stable, its conformational dynamics as measured by the ability of acrylamide to quench the fluorescence of a buried tryptophan as well as its specific activity, are both lower than for the mesophilic protein. As the temperature is increased. the thermodynamic stability of the thermophilic protein approaches that of the mesophilit: protein at its working temperature. Its conformational dynamics and specific activity however were both shown to increase, until at the physiologically operational temperature, they become similar to those of the mesophilic enzyme at its operational temperature. These results confirm the proposal that a direct relationship and balance holds between thermodynamic stability, dynamics and specific activity in globular proteins. They demonstrate also the constraining effect of increased stability upon conformational dynamics and enzyme activity. Keywords:
3-phosphoglycerate
kinase; stability; dynamics; thermophilic proteins.
1. Introduction Comparison of the properties of mesophilic proteins and of their more stable counterparts from thermophilic bacteria has provided a natural approach to the investigation of protein stability (Amelunxen & Murdock, 1978). Comparison of the sequences and structures of these proteins has shown that extra stability is obtained by increasing the number of non-covalent interactions that stabilize all proteins (Perutz & Raidt, 1977; Harris et al., 1980). Improvement of packing and elimination of cavities is also suggested to increase stability (Kellis et al., 1988). While it has been generally assumed that the conformational dynamics of a particular protein are conserved under the conditions under which it operates in oivo and even at lower temperatures (Daniel et al.. 1982), the additional interactions of thermophilic proteins are expected to make them more rigid than their mesophilic counterparts at a given t,emperature. This functionally adverse effect of t Author to whom all correspondence should he addressed.
fluorescence
lifetime;
stability on the conformational dynamics of t,he protein has been observed in hydrogen exchange experiments (Wiithrich & Wagner. 1979; Delpierre et al., 1983; Wrba et al., 1990), as well as flexibility indices computed from crystallographic data (Vihinen, 1989). The catalytic activity of enzymes is widely believed to be dependent upon the conformational dynamics of the protein (Huber, 1979: Wrba et al., 1990). The conservation of enzyme activity within a family of proteins operating under widely different physiological environments would therefore be expected to be accompanied by a conservation of conformational dynamics at their respective physiological operating conditions. The aim of t,he work presented here is to provide further firm experimental evidence for the relationship between st’ability, dynamics and activity in 3-phosphoglycerate kinase (PGKf: EC 2.7.2.3) an enzyme $ Abbreviations used: PGK. 3-phosphoglyceratr kinase; thermus, Thermus thermophilus HBX: SDS/PAGE. SDS/polyacrylamide gel electrophorrsis: GAPDH. 3-phosphat,e dehydrogenase; P(:A 3-phosl”hoglyceric acid.
utilizing a propostd hii~gt, bending mec*hanisnr (H>wks rt cd., 1979; Wat,son d al.. 198%). The proteitts studied here are the mesophilic WK from yeast, and the PGK from t,he extreme thermophile ‘I’hwmus thermophilus IIH8 (thermus). stabilit,iw of both having heen well characterized (Nojima rt nl.. 1!)77: Adams rf d.. 1985: Griko et cd.. 1989). Steatfy-state fluorescence and fluorescence decay kinetics provides particularly useful methods t,o study and compare the dynamics of each protein sinc*cathe fluorescence of tr?ptophan is highly sensit iw to its surrounding enwronment and has dway lifetimes of the order of nanosewnds, which is the time-wale of wzymatically important, fluctuations in protSeins (Lakowicz, 1983). The kinetic rate of ac*rvlamide quenching of fluorescence depends upon the st,ructural fluctuations of a protein that allow thr penetration of the quencher into its interior. ac*c:ording to the mobile defect model (Lumry & Rosenberg. 1975). Measurement of this rate t,herefort, provides information about the level of tlynamics around the fluorophorr on itn enzymatically important, time-walr (Eftink & Ghiron, 1976). Thw-mus PGK is well suit.ecl to such a. study sinw it cont,ains only orw tryptophan in a buried rnvironmtant bttween tht. two Rossrnan folding units in the (‘-t,erminal domain (Kowcn et nl.. 1987). PUK from veast. however, contains two tryptophans at posiiions 333 (equivalent to that. in thermus PC:K) and 308, which is exposed t)o solvent (M’at,son 4 01.. 1952). cwmplicating the int,erpretation of fluoreswnw data (Dry&w Hr Pain. 1989). Rewnt.l? howrvw. we havck shown that thr> fiuorwwnw of thcb c~sposed FV308 ca,n be selectively quenched with snc4nimitlr while leaving the hurirtl M’333 utlaffwtrtl (Varlry d cd.. 1991). thus allowing dircxc-t. (.ornpa.rison of t.hc flnorescrnw and henw tlynarnics of’ the two proteins. This, combinrtl with cwmparc tiw measurements of the specific: activities and the st.abilities of the two proteins. provitlrs a m(‘ans of examining directly t,he relat’ionships betwrrn dynamicas. st,ability and nctivity.
2. Materials
Thermus EY:K wax purified according to thr rnethoci described by Lit~tlwhild rt al. (I 987). with a yield of .1-Smg of’ PGK per kg of cells. ‘irast IXK. owrexpwssed ill yrast (Wilson rt ((1.. 1987). gab purified wcwrding to t hc mrt,hotl dewribed by Scopw (1969). tSoth proteins gavts single bands on SI)S/Pr\(:lC. l~nltw xtat)ed. all c,titatnirals were Analar grade or equivalent from 13I)H, I’oolr. IT.K.
JWK activit?- was rneasurrtl bv a motlific:atiorl of a method described by Scopes (197.i). The assay mixtuw (I ml) wntained 10 m.v-Tris. H(‘I (pH 7.2). I ~~YI-EI)TA. 5 mwmagneaium chloridr. I6 rn~l-l-S-l)hosphogl~(~~ri(, acid (WA) (Sigma. Poole. V.K .). 6 mw.ATI’ (Jrmostatted to *WI ‘C’. Spwific activities wercl c.atcaulatetl assuming that I unit of PGK will ~ww the oxidation of t P.M of N’ADH,irnin and using a value of .A1g,,j= 0.5 at 280 nrn for thermus (t,ittlc child et r/J.. 1987) and for yeast PC:K (Adarns st al.. 1985). Thtx purified prot)eins from ytlast and thermus hatl spwitic. acativitiss at ,.> *)r ’ ( ’ of II00 units trig-l and 2.50 units trig- I. rrspectively. which are in good agrrrrnrnt with t,htl \-slurs previously published. The protrins were shah II t,cj tw ovc’r !X50,,,activr in I.1 41-suc,c,ininiict(,,’I wwrytarnitlc
Fluoresc~wc~r~ tlxperiment s were pc~rfi~rrrrrti in a CO1 wTris. 0.1 M-h’a(‘l. OW~I wE:I)TA buffer atijustrd to pH 7.2 with HC’I and filtered (022 pm Miltiporr) hrforcl use. Fluorescence quenching was carried out by the addition of 5 wwrylamide (IGc~tran grade. BI)H) to protrin
Table 1 Thumo.stability
and Methods
and speci$c
nctizdy
Stability,
Dynamics
and lhzyme
533
Actioity
l’hr himolecular quenching rate constants. k, are listed with the Stern-\‘olmer constant for the qwn~*hinp of PGK by acr$amidr at a series of temperat,urrs. Each ralur of k, is calculated using tht= average of the 2 fluorescence lifetimes (t), weighted for the fractional intensities (d). and thr Stern-Volmer constant (FL,,). Values are quoted as averages with standard errors of thr mean. c~alculatrtl from 6 determinations. rn
solutions of absorbance WI to W% at 295 nm. The recorded fluorescence intensity was adjust’ed for dilution and acr,vamide absorbanw using a molar extinction roefficient for
acrylamide of 0.%X? ml mg-’ cm-’ at 295 nm (Eftink & Ghn-on. 1981). For double quenching experiments with the yeast enzyme, hoth qurnrhrr and protein solutions I.1 >I-sucscinimidv (Analar. treated with vontainrd achtivatrd charcoal). mea,surements \ver’r nlatlt~ using a Steady-stat.e Perkin-hlmrr MI’F4-I, sprctrofluorimetc,r with an es
