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

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

´
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

,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

), 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):

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

-
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

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

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

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-

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+
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.

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

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

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

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.

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

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