| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Biological Sciences, University of Waikato, Hamilton, New Zealand;
School of Biological Sciences, University of Auckland, Auckland, New Zealand;
College of Marine and Earth Studies, University of Delaware, Lewis, Delaware, USA;
Centre for Extremophile Research, Department of Biology and Biochemistry, and
|| Department of Biology and Biochemistry, University of Bath, Bath, UK
1Correspondence: Department of Biological Sciences, University of Waikato, Private Bag 3105, Hamilton 3240, New Zealand. E-mail: r.daniel{at}waikato.ac.nz
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Gcat and Ginact but also by two new intrinsic parameters, Heq and Teq, which describe the enthalpy and midpoint, respectively, of a reversible equilibrium between active and inactive (but not denatured) forms of enzyme. Twenty-one enzymes from organisms with a wide range of growth temperatures were characterized using the Equilibrium Model. Statistical analysis indicates that Teq is a better predictor of growth temperature than enzyme stability (Ginact). As expected from the Equilibrium Model, Heq correlates with catalytic temperature tolerance of enzymes and thus can be declared the first intrinsic and quantitative measure of enzyme eurythermalism. Other findings shed light on the evolution of psychrophilic and thermophilic enzymes. The findings suggest that the description of the Equilibrium Model of the effect of temperature on enzyme activity applies to all enzymes regardless of their temperature origins and that its associated parameters, Heq and Teq, are intrinsic and necessary parameters for characterizing the thermal properties of enzymes and their temperature adaptation and evolution.—Lee, C. K., Daniel, R. M., Shepherd, C., Saul, D., Cary, S. C., Danson, M. J., Eisenthal, R., Peterson, M. E. Eurythermalism and the temperature dependence of enzyme activity
Key Words: enzyme temperature optimum • Equilibrium Model • growth temperature • protein stability • temperature adaptation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Gcat, the free energy of activation of the catalytic reaction, and secondly thermal stability, which can be described using Ginact, the free energy of activation of the thermal denaturation process.
Some functional studies of enzyme temperature-linked properties have defined a so-called enzyme "temperature optimum" that arises from this two-parameter "Classical Model" (1)
. However, this "temperature optimum" is not an intrinsic enzyme property, being derived from a complex mixture of both activity and thermal stability effects (1)
and dependent on assay duration. Consequently, it is of limited value for measuring enzyme temperature adaptation. Evidence of anomalously low enzyme activity at high temperatures (2
, 3)
also suggests that the Classical Model is insufficient to fully describe the effect of temperature on enzyme activity. In addition, there has not been any quantitative tool for characterizing an enzyme’s eurythermalism. Traditionally, eurythermalism has been used to describe an organism’s ability to withstand a wide range of temperatures. To date, there is limited information on how eurythermalism is achieved in eurythermal organisms, and an important possibility is that special characteristics of enzymes in these organisms play an important role.
Enzyme thermal stability clearly has a general correlation with growth temperature; enzymes from thermophiles are typically more thermally stable than those from mesophiles, and those from mesophiles are generally more stable than their psychrophilic counterparts. However, there are exceptions, such as RNase (4
, 5)
, mammalian adenylate kinase (6
, 7)
, and catalase (8
9
10
11
12)
, which are much more stable than might be expected from their environmental temperatures. Factors other than a high temperature environment may in some cases be responsible for high thermal stability. Resistance to proteolysis (13)
and organic solvents (14)
, for example, are known to be associated with high thermal stability. In addition, almost all thermal stability determinations are done in the absence of substrate, which often affects stability.
Two new, intrinsic, temperature-dependent parameters of enzymes, Teq and Heq, have recently been described (1)
and validated (15)
. These arise from the Equilibrium Model, which proposes that the active form of enzyme (Eact) is in reversible equilibrium with an inactive form (Einact) and that it is the inactive form that undergoes irreversible thermal inactivation to the thermally denatured state (X), as described in (1)
.
|
|
Gcat, Ginact, Heq, and Teq as follows:
|
|
|
|
|
|
Teq is the temperature representing the midpoint of transition between the active and inactive (but undenatured) forms of the enzyme. In other words, Teq is the temperature at which Keq = 1 and Geq = 0 [we have previously (1)
used the term Tm to designate this temperature but now prefer the term Teq (15)
since it is now clear that Teq is entirely separated from and unrelated to thermal denaturation]. Of more than 40 enzymes examined in our laboratories (15
16
17)
, including those presented here, all followed the temperature-dependent behavior predicted by the Equilibrium Model, and none followed that expected from the Classical Model. Furthermore, Teq is independent of Gcat and Ginact. Preliminary evidence suggests that the active-inactive transition described by the Equilibrium Model arises from a conformational transition (15)
at or near the active site, since the active-inactive equilibrium is established in a timescale at least two orders of magnitude faster than that of thermal denaturation (15)
; stabilizing and destabilizing agents have no effect on Teq (16
, 18)
; and different substrates can affect Teq without altering stability (16
, 18)
.
Not only does the Equilibrium Model provide a description of enzyme temperature-linked behavior, it does so under assay conditions, so that the derived constants such as Teq and Heq are potentially physiologically based. Since enzyme temperature adaptation is the functional basis of metabolic and organism adaptation studies, the Equilibrium Model is not only useful but necessary for conducting such studies.
In this work, the intrinsic temperature-dependent properties of 21 enzymes, including several isoactive sets of enzymes derived from a wide range of growth temperature sources, were determined. Parameters generated from model fitting were analyzed to identify any associations with each other or with growth temperature optima.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Enzyme assays and data collection
In general terms, these were carried out as described previously (17)
. All enzyme assays performed in this work were continuous photometric assays measured using a Thermospectronic Helios 
-spectrophotometer equipped with a Thermospectronic single cell peltier-effect cuvette holder. All reactions were started by the rapid addition and mixing of a few microliters of enzyme solution at zero degrees into 400-3000 µl of temperature-equilibrated reaction mixture to enable data collection to begin within the first few seconds; the temperature of the reaction mixture remains within ±1°C of desired temperature during data collection. Absorbance data were collected at one-second intervals for three minutes using Vision (version 1.25, Thermo Spectronic) on a Windows PC connected to the Helios, within a range of temperatures depending on the source of the enzyme (see Material and Methods in Supplemental Data for assay details). In all cases, precautions were taken so that no decrease in reaction rates was due to substrate depletion (see Material and Methods in Supplemental Data). Where possible, substrate concentrations were maintained at over ten times Km at all temperatures, otherwise appropriate adjustments were made to compensate for the effect of increases in Km. No signs of substrate/product inhibition were observed under the experimental conditions described. Blank rates were measured at all temperatures and used to correct reaction rates if they were significant (see Table 3 in Supplemental Data for assay conditions for individual enzymes).
Based on the variation between the individual triplicate rates from which the parameters are derived for all the enzymes we have assayed thus far, we find that the experimental errors in the determination of Gcat, Ginact, and Teq, are <0.5% and <6% in the determination of Heq (17)
.
Data analysis
The method for processing experimental data is as described previously in detail (17)
. Absorbance data from Vision were first converted in Excel (version 11 for Windows, Microsoft) to progress curves of product concentration (M). The data were subsequently imported into Scientist (version 2.01, MicroMath), where a set of common estimates of thermodynamic parameters (see Supplemental Data) was optimized by "Simplex" searching through the complete data set (product concentration vs. time vs. temperature). The optimized estimates of parameters were then used as initial values to "Least Square" fit the complete data set to the Equilibrium Model, which generated the final values of the parameters; Gcat, Ginact, Heq, and Teq. The generated parameters were then used in the "Zero-time Model" (17)
to draw a zero-time activity vs. temperature plot. This plot was used to visually estimate Topt and HWHM. Initial rates estimated from assay data were also plotted on the zero-time plot to verify the validity of the fittings.
A standalone MATLAB (Version 7.1.0.246 (R14) Service Pack 3, The MathWorks, Inc.) application, enabling the facile derivation of the Equilibrium Model parameters from a Microsoft Office Excel file of experimental progress curves (product concentration vs. time), is available on CD from the corresponding author. This application is suitable for computers running Microsoft Windows XP and is for noncommercial research purposes only.
Protein determination
Protein determinations were performed using the Bradford assay (19)
.
Statistical analysis
All parameters were imported into STATISTICA (20)
for further analysis. Correlation analysis was done for nine parameters, i.e., growth temperature, Gcat, Ginact, Heq, Teq, Topt, HWHM, the difference between Teq and growth temperature, and the difference between Teq and Topt. The nonparametric Goodman-Kruskal Gamma test was used, and Gamma statistics and the associated two-tailed P values were obtained. Gamma statistic is equivalent to Kendall Tau in terms of the underlying assumptions and interpretation, but takes tied ranks (e.g., growth temperature) into account. Two-tailed statistical power values were calculated with these conditions: n = 21, Rho0 = 0, Alpha = 0.05, and "Exact" algorithm. Best-subset regression analysis was done using growth temperature as the response (dependent) factor and Gcat, Ginact, Heq, Teq, and Topt as predictors. A maximum step of 100 and a probability of 0.05 were used for stepwise selection, and best subsets measures were calculated using R squared values. Nonparametric t test was carried out using Mann-Whitney U test, and a two-tailed P value was calculated. Scatter plots were drawn using MiniTab (R14 for Windows, Minitab).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
and Fig. 2
). It should be noted that given the "fast-start" of the enzyme assays (see Materials and Methods), the decline in activity at zero time at temperatures above Topt is evidence that equilibration of the active-inactive forms of the enzyme (Eact and Einact) takes place over timescales significantly shorter than we can observe, i.e., <1–3 s and that the equilibration is about two orders of magnitude or more faster than the rate of irreversible activity loss as shown by the activity vs. time plots for each temperature in the 3D graphs in Fig. 2
(although it should be noted that the very slow irreversible denaturation for the ASB HK47 psychrophilic alkaline phosphatase above Topt is not typical of the enzymes examined here). The parameters of all enzymes are listed in Table 1
. They are derived by fitting assay data to the Equilibrium Model and thus relate to active enzymes in the presence of substrate and cofactor. The exception is the growth temperature optimum of the source organism, which is cited from various sources of reference (see Supplemental Data). Growth temperatures of the source organisms range from 2°C to 75°C. The difference between the source growth temperature and Teq ranges from –6.1 to 52.6°C.
|
|
|
Topt is the graphical optimum temperature of the enzyme at time zero (see Fig. 1
). In most cases, Teq is within a few degrees of Topt, and for almost all enzymes Teq, (and Topt) are higher than the respective source growth temperature. In the case of Teq, the mean difference is 21.1±15.2°C. This is not surprising, since at Teq half of the enzyme activity is unavailable, while at 20°C below Teq most of the enzyme will be in the active form. HWHM is the upper temperature half width of the zero-time temperature/activity peak at 50% activity (the upper temperature half width has been used because the low temperature half of this peak is dominated by Gcat) and is included to show its relationship to Heq (see Fig. 1
); a large Heq leads to a sharp decline in activity with increasing temperature above Topt (and thus a small HWHM), since Heq is the enthalpic change associated with the reversible conversion of the active to the inactive form of enzyme. There are wide variations in both HWHM (from 4.5 to 44°C) and Heq (from 86 to 826 kJ·mol–1), and thus considerable variations in the sensitivity of enzyme activity to temperature above Teq (and Topt). Ginact describes the stability of the enzyme under reaction conditions; as expected, the average Ginact of thermophilic enzymes tends to be higher, and the average Ginact of psychrophilic/psychrotrophic enzymes lower, than that of mesophilic enzymes.
3D plots of one thermophilic, one mesophilic, and one psychrophilic enzyme are shown in Fig. 2A-C
. Although the plots are very similar in form, the temperature axes show that the three enzymes have very distinct responses to temperature. Note that for the ASB HK47 alkaline phosphatase (Fig. 2A
), Ginact has relatively little effect on activity over time at temperatures somewhat above Topt, and physiological activity is therefore very dependent on Teq, but for the Bacillus cereus DHFR (Fig. 2B
), activity over time above Topt depends largely on Ginact.
Correlation analysis
A correlation analysis of the parameters in Table 1
is shown in Table 2
. Two sets of significant correlations are evident. The strongest correlation set (in red) involves Heq, HWHM (negatively), and the difference between Teq and Topt. Heq is the enthalpic change associated with the reversible, temperature driven, interconversion of an enzyme between its active and inactive state. It can be considered as a measure of the sensitivity of an enzyme’s catalytic activity to temperature. Heq will therefore influence the broadness of zero-time activity vs. temperature plots; a small Heq will lead to a broader zero-time activity vs. temperature peak, thus giving rise to a strong negative correlation between Heq and HWHM. Heq can thus be considered to be a quantitative measure of an enzyme’s ability to function over a temperature range: a small Heq indicating that the enzyme will function at relatively high activity over a wide range of temperatures, i.e., behave in an eurythermal manner. Conversely, a large Heq indicates stenothermal behavior. The positive correlation between Heq and the difference between Teq and Topt can be explained on the basis that a low Heq will confer a broader time-zero graphical peak, and thus Topt is likely to be higher than Teq.
|
Compared to the correlation set above, other correlations involving Heq, HWHM, or the difference between Teq and Topt are much weaker. It may be that Heq, and hence an enzyme’s ability to function over a wide range of temperatures (e.g., in an environment with large temperature fluctuations), is more likely to be determined by the temperature range or temperature variability of the environment(s) than it is by other thermal properties, although this hypothesis has not been tested here.
The second strong correlation set (in green) is that between Teq and Topt, and Teq and growth temperature; and moderate or weak correlations between all three of growth temperature, Topt, and Ginact, and between Teq and Ginact. Since Topt can theoretically be calculated from Teq (15)
, the correlation between these two is not surprising, and the weak correlation of these two with protein stability (Ginact) simply confirms the well-established general tendency of thermophiles to have more stable proteins than mesophiles. Of the other correlations in this set, the correlation between Teq and optimal growth temperature is particularly significant. It indicates that in a broad sense, Teq is a better indication of an enzyme’s temperature origin than either its stability (Ginact) or the graphically determined Topt. This is supported by the result from best-subset regression analysis, which identifies Teq as the single best parameter for predicting growth temperature (data not shown).
The moderate negative correlation (in blue) between growth temperature and the difference between Teq and growth temperature is interesting. From Table 1
, it is evident that at high growth temperatures (growth temperature 
55°C), i.e., in organisms that are clearly thermophilic, there tend to be relatively small differences between Teq and growth temperature (7.3±6.6°C); however, for psychrophiles, psychrotrophs, and low-temperature mesophiles (growth temperature 
31°C), they are larger (27.8±15.1°). A nonparametric t test between the Teq and growth temperature differences of thermophilic and nonthermophilic enzymes reveals that the two groups are significantly different (P value=0.008) in this respect. A previous study using kcat as a measure of temperature optima also suggested that enzymes tend to have optimal activity at temperatures higher than their hosts’ living temperature, and the higher the growth temperature of the organism the narrower the gap becomes (21)
. There are several possible reasons for this. There may be a poor match between the environmental temperature of thermophiles and their laboratory growth temperature optima; thermophiles are often isolated from sources that are appreciably hotter than their determined growth temperatures. Alternatively, it might be argued that the origin of the current mesophilic microbial life is thermophilic, i.e., that microbial evolution has proceeded down-temperature from a (hyper)thermophilic last common ancestor (22
23
24)
, and that the evolution of Teq lags behind optimal growth temperature. Furthermore, depending on how crucial their roles may be in the cell, different enzymes might evolve at different speeds to adapt their Teq to the environment, thereby resulting in the variation in Teq among enzymes from one organism (e.g., for Bos taurus, Table 1
).
Conclusions based on weak correlations must be speculative. Although its P value is slightly over the adopted threshold of 0.05, the correlation between Gcat and Ginact may suggest that Gcat is higher for more stable proteins and thus support various studies (21
, 25
, 26)
that suggested that a change in activation energy (Gcat) might be important to the temperature-adaptation of psychrophilic enzymes. To investigate this idea, correlation analysis was performed with psychrophilic and psychrotrophic enzymes (growth temperature20°C, not including plant enzymes) excluded; the correlation between Ginact and Gcat was found to deteriorate significantly (P value>0.1), tending to confirm that a lower Gcat might indeed be one of the main adaptations for psychrophilic enzymes, while nonpsychrophilic enzymes take on other means of adaptation.
The scatter plots in Table 2
are consistent with the correlations described.
Growth temperature
In general, any correlations with growth temperature must be regarded as tentative since determining environmental growth temperature from laboratory-determined optimal growth temperatures of bacteria is problematic, partly because the temperature of most microbial environments varies with time, including that of Escherichia coli, for example (27
, 28)
, and partly because the laboratory environment on which these are usually based may not represent environmental conditions. In the absence of better measures, however, we have generally adopted here the published data for organisms involved in this study (see Supplemental Data).
Individual enzymes
The DHFR from the strict psychrophile Moritella profunda, isolated from deep Atlantic sediments (29
, 30)
, is a surprisingly stable enzyme with a concomitantly high Teq, while its Heq (104 kJ·mol–1) is significantly lower than the 25th percentile (115 kJ·mol–1), suggesting eurythermal behavior. We speculate that its low temperature activity has been achieved by evolving a lower Heq into an enzyme with an incidentally high Teq, thus conferring eurythermalism. Part of the lower half (below Topt) of the enzyme activity curve is influenced by Heq, but the effect of Gcat may dominate, as stated above.
The alkaline phosphatase from isolate ASB HK47 (Fig. 2A
) gives another illustration of the relative effects of the four parameters on enzyme activity over a range of temperatures. For many enzymes (e.g., B. cereus DHFR, Fig. 2B
), the denaturation rate is such that it will play the major role in controlling enzyme activity over timescales important to the organism (over minutes rather than seconds) at temperatures above Teq. However, for the ASB HK47 alkaline phosphatase, it is evident that the lower rate of denaturation above Teq leads to a situation where activity will be dominated by Teq and Heq over significant timescales.
To achieve temperature adaptation, evolution imposes a strong selection force for thermophilic enzymes to have both high thermal stability and a high Teq. High thermal stability has been thought to be achieved at the expense of flexibility to explain why thermophilic enzymes generally exhibit relatively little activity at lower temperatures (31)
. However, the rigidity required for stability is global, while the flexibility required for catalysis is likely to be local (32)
, as demonstrated by some exceptions (33
, 34)
. It has been suggested that a combination of local flexibility at the active site and high overall stability (35
, 36)
, is behind such phenomena. In other words, enzymes active at low temperatures do not necessarily have to be thermally unstable (37
, 38)
. Recent reports also suggested that the higher flexibility of psychrophilic enzymes is local and on the microsecond to millisecond timescale (39)
, while their global flexibility may not be necessarily higher (38)
. In addition, random and directed mutagenesis studies (37
, 40
, 41)
have shown that it is possible to engineer an enzyme so that it retains low-temperature activity while gaining thermostability.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Heq is a new quantitative measure to characterize eurythermalism at an enzymatic level. The new parameters (i.e., Teq and Heq), combined with established parameters (i.e., Gcat and Ginact), define fully how temperature influences enzyme activity, while the former provide new tools to describe and investigate this influence.
Much of the data presented here comes from bacterial sources, although results from correlation analysis remain similar when data from eukaryotic sources are excluded (data not shown). Although the 21 sets of data from discrete enzymes presented here are a relatively small data set, the main correlations are robust and generally consistent with findings or proposals based on other data; they also provide new targets and perspectives for molecular and structural studies.
The development and verification of the Equilibrium Model have provided additional and quantitative measures of the thermal behavior of enzymes (1
, 15)
that are essential for describing the effect of temperature on enzyme activity and useful parameters for measuring the temperature adaptation of enzymes. Additionally, the results here show the Equilibrium Model’s potential usefulness as a tool to investigate the relationship between enzyme thermal properties and the influence of temperature on the physiology and evolution of the host organism.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 12, 2006. Accepted for publication January 18, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. K. Lee, S. C. Cary, A. E. Murray, and R. M. Daniel Enzymic Approach to Eurythermalism of Alvinella pompejana and Its Episymbionts Appl. Envir. Microbiol., February 1, 2008; 74(3): 774 - 782. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
