Thea“.

mcmom STATE UNIVERSITY BRARIES
l

\ llllllllllllllll

 

lllllllllllll l H

3 1293 01421 4930

 

 

 

 

l

 

 

This is to certify that the

dissertation entitled

presented by

Douglas Scott Burdette

has been accepted towards fulfillment
of the requirements for

Ph. D . degree in Biochemistrx

 

 

Datelgrch 19 . 1996

M5 U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

— — EMF—gs _

 

LIBRARY
Michigan State
University

 

 

 

PLACE ll RETURN BOX to remove thle checkout from your record.

TO AVOID F INES return on or betore date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-L_l::l

 

Sicc

 

 

THERMOANAEROBACTER ETHANOLICUS 39E
SECONDARY-ALCOHOL DEHYDROGENASE: MOLECULAR
BASIS FOR STABILITY AND CATALYSIS

By

Douglas S. Burdette

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry
1996

I.
Ala:

 

5-
d“ u .J'NL

 

ABSTRACT

THERMOANAEROBACTER ETHANOLICUS 3913 SECONDARY-
ALCOHOL DEHYDROGENASE: MOLECULAR BASIS FOR STABILITY
AND CATALYSIS
By

Douglas S. Burdette

The ath gene encoding Themtoanaerobacter ethanolicus 39E secondary-alcohol
dehydrogenase (2° ADH) was cloned, sequenced, and overexpressed in Escherichia coli
DHSa (~15% total prot.). The protein was purified to homogeneity by heating and
precipitation. The 1056 bp gene encoded a homotetrameric recombinant enzyme (37.7 kDa
subunits) that displayed a l70—fold greater catalytic efficiency toward NADP(H)—dependent
propan-2—ol than toward ethanol oxidation. The 2° ADH site directed mutants C378,
H59N, DlSON, D150E, and DISOC displayed 5 3% of wild type catalytic activity. These
data and the wild type enzyme inactivation by dithionitmbenzoate (DTNB) and
diethylpyrocarbonate chemical modification, supported sequence predictions based on
Hydrophobic Cluster Analysis (HCA) that Cys37, HisS9, and AsplSO residues are
catalytic Zn ligands. X-ray absorption spectrometry data supported the presence of a
protein bound Zn with a ZnSl(N-O)3.4 coordination sphere. Induction coupled plasma
emission spectrometry data of wild type and mutant enzyme Zn binding, implicated these
residues in 2° ADH Zn liganding. Thus, a catalytic Zn atom and its immediate environment
are important in 2° ADH activity. Analysis of the Gl98D mutant supported the HCA based
prediction that this 2° ADH binds NADP(H) in a common nucleotide binding motif, a
Rossmann fold. The substrate and cofactor binding sites in the 2° ADH are structurally
distinct because catalytic Zn ligand mutations did not change cofactor affinity. The
thermophilic 2° ADH) was optimally active at ~90°C and displayed a half-life of 1.2 days at
80°C. Analysis of temperature dependent unfolding of 2° ADH in guanidine hydrochloride

(GuHCl) indicated the enzyme was very rigid (50% unfolded at ~115°C). Thus, the
enzyme is nearly 100% folded at 90°C where it is maximally active. 2° ADH Activity
increased 2-fold at 37°C to 75°C in low concentrations of GuHCl (120190 mM); and,
activity remained in 1.6M GuHCl demonstrating its high resistance to chemical
denaturants. Catalytic rate enhancement was also seen with weakly chaotropic (i.e.,
KNO3) and neutral (i.e., KCl) inorganic salts, indicating that specific ionic and not
hydrophobic interactions are required for the catalytic rate enhancement. Linear Arrhenius
plots for the oxidation of propan-Z-ol by the native and recombinant 2° ADHs from 30°C to
90°C suggested that the thermal activity relationships are related to binding high kinetic
energy substrates and not to temperature dependent changes in enzyme unfolding.
Consequently it is suggested that the 2° ADH has evolved high rigidity for stability and

these molecular determinants are distinct from those which control catalysis.

Copyn'ght by
DOUGLAS scorr BURDBTI'E
1996

To Kate and to our families

ACKNOWLEDGEMENTS

I wish to thank my graduate advisor, Dr. Zeikus, and my committee members for
their guidance and patience. I would also like to thank Dr. Robert Phillips, Dr. Robert
Scott, and Dr. Bobby Ami whose collaboration on this project has provided greater depth
to the research. I am specifically indebted to Dr. Claire Vieille, Maris Laivenieks, Cindy
Petersen, and‘the rest of the Zeikus lab members for scientific guidance, technical
assistance, and friendship. I also thank Chris Jambor, the Hupes, the Santoros, the
Giegels, the Bordeners, the Castigliones, the Noons, and my family for giving me the
strength to endure. I am especially grateful for the consistent support of my grandparents,
specifically to Dora and Eugene Carbon whose kindness will live in my memory. Finally,
I thank Dr. Kate Noon who inspired me daily to persevere.

TABLE OF CONTENTS

page
LIST OF TABLES .............................................................................. ix
LIST OF FIGURES ............................................................................ xi
LIST OF ABBREVIATIONS ................................................................. xiii
CHAPTER I
Literature Review ............................................................................... 1
ALCOHOL DEHYDROGENASES
Alcohol dehydrogenase structure and function ..................................... 2
Alcohol dehydrogenase industrial applications ..................................... 5
THERMOPHILIC ENZYMES
Thermophilic enzyme sources and diversity ........................................ 8
Thermostability ......................................................................... 21
Thermophilicity ......................................................................... 23
Protein folding .......................................................................... 30
Protein unfolding ....................................................................... 32
Molecular mechanisms of thermostability
Intrinsic factors ..................................................................... 34
Multiple substitutions and protein stabilization ............................. 44
Substitutions and modification of the thermodynamics of unfolding 45
Prolines in loop regions ....................................................... 46
Salt bridges ..................................................................... 49
Hydrogen bonds ............................................................... 50
Hydrophobic interactions and core packing ................................ 51
Covalent destruction and irreversible denaturation ......................... 53
Extrinsic mechanisms .............................................................. 54
Glycosylation ................................................................... 56
Salts ............................................................................. 56
Other chemical effectors ....................................................... 58
Pressure effects ................................................................ 59
Molecular mechanisms of protein thermophilicity .................................. 59
Genetic engineering of therrnozymes
Modification of enzyme catalytic properties ..................................... 62
Thermostability and thermophilicity engineering ............................... 62
THESIS OBJECTIVES AND SIGNIFICANCE ............................................ 68
ACKNOWLEDGEMENTS .................................................................... 71
REFERENCES .................................................................................. 72
CHAPTER II

Cloning and Expression of the Gene Encoding the
Thennoanaerobacter ethanolicus 39E Secondary-Alcohol Dehydrogenase

and Enzyme Biochemical Characterization ................................................... 89
Abstract .................................................................................. 90
Introduction ............................................................................. 91
Materials and methods ................................................................. 92

 

Mt

St:

wwknw Dawn.

an:

MAW).

Results ................................................................................... 96

Discussion ............................................................................... 112

Acknowledgements ..................................................................... 116

References ............................................................................... 117
CHAPTER 111

Effect of Thermal and Chemical Denaturants on
Thennoanaerobacter ethanolicus 39E Secondary-Alcohol Dehydrogenase Stability 120

Abstract .................................................................................. 121
Introduction .............................................................................. 122
Materials and methods .................................................................. 124
Results ................................................................................... 128
Discussion ............................................................................... 144
Acknowledgements ..................................................................... 147
References ............................................................................... 148

CHAPTER IV

Mutagenic and Biophysical Analysis of T hermoanaerobacter ethanolicus

Secondary-Alcohol Dehydrogenase Activity ................................................. 151
Abstract .................................................................................. 152
Introduction ............................................................................. 154
Materials and methods ................................................................. 156
Results ................................................................................... 159
Discussion ............................................................................... 174
References ............................................................................... 185

CHAPTER V

CONCLUSIONS ................................................................................. 187

CI-IAP'I'ER VI

Directions for Future Research .................................................................. 192
Sequence based mutagenesis and kinetics ........................................... 193
Structural Analyses ..................................................................... 202
Biotechnological utility ................................................................. 204
References ............................................................................... 205

APPENDIX A

Sequence Comparison Based Predictions of Important Catalytic Domain and

Cofactor Binding Domain Amino Acids in 1° and 2° ADH Structures .................... 206

APPENDIX B

Computer programs used to calculate (hyper)thermophilic versus mes0philic
total amino acid composition and the theoretical Arrhenius data for thermophilic

and mesophilic enzymes ........................................................................ 214

APPENDD( C

Construction and Kinetic Characterization of Thennoanaerobacter ethwwlicus 39E

2° ADH Proline Deficient Mutants ........................................................... 239
Materials and methods ................................................................ 241
Results .................................................................................. 242
Discussion .............................................................................. 246
References .............................................................................. 248

 

 

 

Ci

In

To;

LIST OF TABLES

page

CHAPTER I
Table 1 - Thermozymes from thermophilic organisms ..................................... 10
Table 2 - Thermozymes from hyperthermophilic organisms .............................. 14
Table 3 - Amino acid composition comparison of enzymes from mesophilic,

thermophilic, and hyperthermophilic prokaryotes ............................ 20
Table 4 - Kinetic constants for type II xylose isomerases purified from

mesoPhiles and thermophiles .................................................... 28
Table 5 - Kinetic constants for glutamate dehydrogenases pmified from

mesophiles and thermophiles .................................................... 28
Table 6 - Representative site directed mutagenesis studies of enzyme

thermostability mechanisms ...................................................... 35
Table 7 - Comparison of xylose isomerase thermal parameters with the optimal

growth temperature of the respective microorganism ......................... 55
CHAPTER II
Table 1 - Comparison of primary structural similarity between horse liver and

bacterial alcohol dehydrogenases ................................................ 102

Table 2 - Effect of enzyme preincubation temperature on the activation energy
for propan-2-ol and ethanol oxidation by T. ethanolicus 39E 2° ADH ..... 111

 

CHAPTER III
Table 1 - Role of protein precipitation in T. ethanolicus 2° ADH

thermal inactivation ............................................................... 136
CHAPI'ER IV
Table l - Oligonucleotide primers for PCR amplification of

mutant ath gene DNA .......................................................... 157
Table 2 - Pmification of the recombinant wild type 2° ADH

expressed from plasmid pADHBKA4-kan ..................................... 160
Table 3 - Effect of site specific amino acid substitutions on

T. ethanolicus 2° ADH activity .................................................. 164
Table 4 - Effect of the Glyl98 to Asp mutation on

T. ethanolicus 2° ADH nicotinamide cofactor preference ..................... 175

CHAPTER VI
Table 1 - Comparison of conserved 2° ADH amino acids with the corresponding
1° ADH residues from HCA based sequence alignments ..................... 194

Table 2 - Comparison of conserved N ADP+ dependent 2° ADH amino acids
with the corresponding NAD+ dependent 1° ADH residues deduced
from HCA based sequence alignments ......................................... 198

Table 3 - Added thermophilic 2° ADH proline residues compared to the
mesophilic 2° ADH deduced from HCA based sequence alignments ........ 199

APPENDIX A
Table 1- Comparison of aligned amino acids involved in ADH catalytic
domain structure and function ................................................... 209

Table 2- Comparison of aligned amino acids involved in ADH nicotinamide
cofactor binding ................................................................... 211

LIST OF FIGURES

page

CHAPI'ER I
Figure l - Arrhenius dependence of reaction rate on the temperature and

activation energy of mesophilic and hyperthermophilic enzymes ............ 26
Figure 2 - Role of proline residues in protein structural stabilization .................... 48
Figure 3 - Flow chart of potential steps toward engineering enzyme

thermophilicity and thermostability .............................................. 65
CHAPTER II
Figure 1 - Restriction map of the T. ethanolicus 39E 2° ADH clone (pADHB25) ...... 99
Figure 2 - Nucleotide sequence and deduced amino acid sequence of

T. ethanolicus 39E ath and of the downstream open reading frame ....... 101
Figure 3 - HCA comparison of thermophilic and mesophilic 1° and 2° ADHs

centered on the putative nicotinamide binding motifs .......................... 104
Figure 4 - Arrhenius plots for the recombinant T. ethanolicus 39E 2° ADH

between 25°C and 90°C ............................................................ 110
CHAPTER 111
Figure 1 - Recombinant T. ethanolicus 39E 2° ADH thermophilicity ..................... 130
Figure 2 - Recombinant T. ethanolicus 39E 2° ADH thermostability ..................... 132
Figure 3 - Temperature dependence of 2° ADH precipitation .............................. 135
Figure 4 - Effect of GuHCl on T. ethanolicus melting temperature ....................... 138
Figure 5 - Effect of GuHCl on T. ethanolicus fluorescence ............................... 141
Figure 6 - GuHCl and temperature dependence of 2° ADH activity ...................... 143
CHAPTER IV
Figure 1 - SDS-PAGE of wild type recombinant 2° ADHs ................................ 163
Figure 2 - T. ethanolicus 2° ADH EXAFS analysis ........................................ 166
Figure 3 - PCR-based ath gene site directed mutagenesis scheme ..................... 168
Figtue 4 - SDS-PAGE of 2° ADH mutant proteins ......................................... 170
Figure 5 - Comparative 1° and 2° ADH active site models ................................. 181

CHAPTER VI
Figure 1 - Position of the non-conservative amino acid differences between

1° and 2° ADHs relative to the horse liver 1° ADH catalytic site ..............

APPENDD( A
Figure 1 - HCA comparison of the B. stearothennophilus 1° ADH, the
T. ethanolicus 2° ADI-I, and the C. beijerinckii 2° Adh amino acid

sequences to that of the horse liver 1° ADH ....................................
APPENDD( C
Figure 1 - Thermophilicity and thermostability profiles for T. ethanolicus 2° ADH
Pro residue mutants
Part A ...........................................................................
Part B ...........................................................................

196

208

1° ADH
2° ADH

DEPC
D'I'NB

EUI‘A
EPR
EXAFS

GAPDH
GuHCl
HCA
ICP
MAIDI
ORF
PEG

ABBREVIATIONS

primary-alcohol dehydrogenase

secondary-alcohol dehydrogenase

alcohol dehydrogenase

diethylpyrocarbonate

dithionitrobenzoate

dithiothreitol

reaction activation energy

(ethylenedinitrilo)tetraacetic acid trisodium salt

Electron paramagnetic resonance

X-ray absorption spectrometry

Fourier transformed infra-red spectroscopy
glyceraldehyde-3-phosphate dehydrogenase

Guanidine hydrochloride

Hydrophobic Cluster Analysis

induction coupled plasma emission spectrometry

matrix associated laser desorption ionization mass spectrometry
open reading frame

polyethylene glycol

Polymerase chain reaction

ribosome binding site

Enzyme melting temperature in the absence of denaturant
Enzyme melting temperature in the presence of guanidine

Chapter I

Literature Review

Adapted from an article published in Ann. Biotechnol. Rev.

2
ALCOHOL DEHYDROGENASES

Alcohol dehydrogenase structure and function

Alcohol dehydrogenases (ADHs) are integral to both prokaryotic and eukaryotic
metabolism. Categorized as 1° or 2° based on their displaying higher activity toward 1° or
2° alcohols, these enzymes typically act on a broad range of substrates. The 1° ADH
isolated from mammalian liver is proposed to oxidize alcohols and possibly steroids during
their detoxification [l]. ADHs are also involved in catabolic alcohol consumption by acetic
acid batceria [2]. More commonly however, catabolic ADHs mediate the terminal electron
transfer in solventogenic fermentations by yeasts [3—5] and bacteria [1]. Alcohol formation
by these organisms provides the oxidized nicotinamide cofactor necessary for glycolysis,
eliminating the need for an exogenous terminal electron acceptor. Catabolic 1° ADI-Is have
been identified in mesophilic (optimal growth between 25 °C and 50°C) [1,6], thermophilic
(optimal growth between 60°C and 80°C) [7-9], and hyperthermwhilic (optimal growth
above 80°C) [10,11] microorganisms. However, catabolic 2° ADHs have only been
reported in mesophilic [12,13] and thermophilic [8,14-16] bacteria. The mesophilic
bacterium, Closm'dium bezjierinckii, expresses a 2° ADH during propan-Z-ol production
whereas, the thermophilic Thennoanaerobacter sp. and Thennoanaerobacteriwn sp. use a
2° ADH to form ethanol - a 1° alcohol - from acetleoA [7,8]. These thermophilic bacteria
ferment sugars to ethanol and acetic acid (neither butanol, acetone, nor propan-2-ol have
been detected) and express a functional 1° ADH concurrently with the 2° ADH. The
Thermoanaerobacter ethanolicus 39E 1° ADH is proposed to function in electron transfer
between NAD(H) and NADP(H) and in ethanol consumption while the ethanol producing
2° ADH can function to reduce both acetleoA and ethanal [8] using the generalized

enzymatic reaction:

ii 9"
n—c—cu3 + NADPH+H+ —> n—é—crt, + NADP”
r‘r

ADHs are predominantly nicotinamide cofactor dependent Zn metalloproteins
although Fe [17] and ferredoxin linked enzymes have been reported [18]. Theorell and
Chance proposed an ordered 1° ADH kinetic mechanism where cofactor binding is
followed by substrate binding, hydride transfer, product release, then cofactor release [1];
however a semi-random mechanism for product and cofactor release has also been reported
[19]. Isomerization of enzyme bound to reduced cofactor is believed to be the slow step in
liver ADH catalysis [20,21]. Cofactor binding is proposed to induce a slow enzyme
conformational change after a rapid initial binding step, completing active site pocket
formation. Even the NAD+ molecular geometry required for horse liver ADH catalytic
activity has been determined [22]. Zn's catalytic role as a Lewis acid in the electrophilic
ADH mechanism is well established, with the metal mediating direct hydride transfer
between substrate and cofactor [23]. Kinetic and spectrophotometric measurements
provide direct evidence that the alcohol or carbonyl substrate binds to the catalytic Zn atom
[23-25]. This may explain both the broad ADH substrate specificities and why decades of
research into 1° ADH structure-function relationships have failed to identify residues
responsible for substrate binding and discrimination.

ADHs are either dimeric or tetrameric, composed of identical or nearly identical 35
kDa to 45 kDa subunits. Both 1° and 2° ADHs have been reported to contain catalytic Zn
atoms and Cys residues critical to catalytic activity [1.8.15]. The exact protein structure
has been determined for horse liver 1° ADH by x-ray crystallography [26] but no 3-
dimensional 2° ADH structure has been reported. X-ray structural data confirmed that 1°
ADHs bind a structural Zn atom with 4 Cys residues and a catalytic Zn with it His plus 2
Cys residues. The broad specificity of liver 1° ADH is attributed to a wide and deep
hydrophobic binding pocket seen in the x-ray structure [1]. Nicotinamide cofactor was

4
shown to associate with a Rossmann fold in the 1° ADH structures as reported for other

NAD(P)(H) binding proteins [27]. The identification of Rossmann fold consensus
sequences {Gly-Xaa-Gly-Xaa-Xaa-Gly-(Xaa)13-m[negatively charged amino acid for
NAD(H) dependent or neutral amino acid for NAD(P)(H)]} [28] provide sequence based
predictions of cofactor specificity linked to this particular structural fold.

Stereoselectivity is both a hallmark of enzymatic catalysis and an area of great
biotechnological interest. ADHs are highly enentioselective catalysts despite their
characteristically broad substrate specificities. In pyridine dinucleotide containing
enzymes, stereospecificity results from the cofactor attack angle according to Prelog [29].
Carbonyl reduction can result from hydride addition to the Re face, following Prelog's
Rule, or to the Si face, following Anti-Prelog's Rule. So based on the orientation of the
carbonyl relative to the cofactor and which cofactor hydrogen is transfered, there are 4
possible mechanisms for enantiospecific ADH reaction (El-E4) [30]. The E1 mechanism is
defined as the pro-R cofactor hydrogen (HR) on the substrate Si face, E2 involves cofactor
pro-S hydrogen (Hs) attack on the substrate Re face, HR transfer to the substrate Si face
results from an E3 mechanism, and reduction on the carbonyl Re face by the cofactor H3
describes mechanism E4. The horse liver and yeast 1° ADHs operate by the E3 pathway,
producing (S)-alcohols. E1 reactions catalyzed by L. kefir and Psuedomonas sp. (SBD6)
ADHs form (R)—alcohols [30,31].

Overall structural similarity among ADHs is presumed based on their similar gross
morphologies and functional characteristics despite their dissimilar peptide sequences [1].
The proposed structural similarity between ADHs and the difficulty in determining residues
involved in substrate specificity makes comparative analysis between enzymes with
overlapping but difl'erent specificities important. Cloning the genes encoding these
enzymes would provide adequate supplies of protein for structural studies and the means to
perform site directed mutagenesis experiments. Mesophilic (GenBank Acc. No. D90004,
L02104, M91440, X 17065, X59263), thermophilic (GenBank Acc. No. D90421), and

 

 

Alct

 

5
hyperthermophilic (GenBank Acc. No. 851211) 1° ADHs have been cloned and

characterized. The 2° ADH hour the obligately aerobic mesophile Alcaligenes eutrophus
has been clomd and characterized [32] and the meSOphilic C. berjen'nckii 2° ADH gene
sequence has been deposited in GenBank (Acc. No. M84723) but no characterization has
been published. Alignment of peptide sequences derived from the translated gene
sequences indicated the conservation of putative catalytic and critical core residues among
the A. eutrophus and liver enzymes [32]. However, an alignment between the A.
eutrophus 2° ADH, liver 1° ADH, and T. brockr‘i 2° ADH (The T. brockr'i peptide sequence
was determined by Edman degradation of purified protein fragments) led Peretz and
Burstein to conclude that insuficient similarity existed between the 3 sequences to make
comparative structural predictions [33]. These conflicting conclusions indicate that the
validity of 2° ADH structure-function hypotheses based on structural comparisons between

1° and 2° ADHs awaits assessment by genetic manipulation.

Alcohol dehydrogenase industrial applications
While industrial organic syntheses have typically been dominaed by chemical-

synthetic processes, the rising public environmental awareness (reflected by new
legislation) plus the recent developments in enzyme biotechnology suggest that enzymatic
and mixed chemo-enzymatic processes will progressively be substituted for polluting or
toxic chemical processes. Four enzyme groups - carbohydrases, lipases, proteases, and
oxidoreductases - have a high potential for synthesizing peptides (e.g., pharmaceuticals,
neuropeptides, and specific peptides for research purposes) [34]; flavors and fragrances
(e.g., benzaldehyde, naringin) [35]; non-peptide polymers (e.g., polyesters, polyphenols,
polyacrylates) [36]; chiral compounds [37]; and surfactants (e.g., monoglycerides, sugar
fatty acid esters, alkyl glucosides) [38a]. Typically known for their hydrolytic activity,
lipases and proteases can, in specific environments, be used as synthetic enzymes.

Enzymatic syntheses present numerous advantages over chemical syntheses: they are

6
usually highly specific (i.e., enantio-, regio-, and stereo-specific), environmentally

friendly, and their products are usually easily biodegradable.

Chiral alcohols are common in biological effectors (eg. the neurotransmitter
norepinephrine) and chiral alcohols can be used as stereospecific reaction centers in
bioactive compound manufacturing. With the top ten optically active drugs representing
sales of $10 billion (U .8. dollars) annually and with the recent FDA policy focused on
pharmaceutical enantiomeric pmity [37], strategies to produce chiral compounds as either
drugs or as synthetic intermediates have growing indusuial potential. ADHs,
cuantioselectively active on a wide range of substrates, have been the focus of substantial
research into enzymatic synthesis of industrially relevant chiral compounds. The horse
liver 1° ADH is optimized for the interconversion of aldehydes and 1° alcohols but it will
stereospecifically reduce ketones [1]. Mesophilic liver 1° ADH has already been used in the
analytical scale production of chiral cyclic alcohols, polyalcohols, alcohol containing
aldehydes (cg. L—glyceraldehyde) or ketones (cis-cyclohexanone-Z-ol), lactones, and
organosilicates [see 38,39]. Similarly diverse and numerous potentially valuable chiral
bioconversions have been performed in vitro and in vivo with yeast ADH [see 38,40,411.
Resting cells of the hyperthermophile S. sulfataricus were shown to stereospecifically
reduce ketones and to express a 1° ADH with activity toward ketones and 2° alcohols [42].
Ketones and 2° alcohols however, were poor substrates for this enzyme. The mesophilic,
(IO-specific ADH from L. kcfir has high activity toward alcohols near cyclic or aromatic
structures [31,43]. A range of chiral alcohols was also created using a mesophilic (R)-
specific 1° ADH from Pseudomonas sp. (ATCC 49794) [30].

The potential for enzymatic conversions in specialty chemical synthesis and racemic
resolution is widely recognized [38,44—51]. However, the expensive cofactor requirements
(NAD, NADP, FAD) of oxido-reductases such as ADHs may limit their applications
[37,52]. Several systems [51-54], have been demonstrated for cofactor recycling, but

current biotransformations are typically performed in whole cells, where the cellular

7
machinery retains and regenerates the cofactor [37,40,41,52]. The specific cases of chiral

diol synthesis using a host of microbial cells have been compiled and reviewed [55]. Many
important specialty chemicals and their precursors are sparingly soluble in water, requiring
a solvent stable biocatalyst. Horse liver ADH was shown to be catalytically active in
numerous organic solvents with specific activities similar to that in aqueous solution
[56,57]. The activity if horse liver 1° ADH was also examined in water-oil microemulsions
to enhance conversion rates for substrates with low water solubility [58]. The
hyperthermophile S. solfataricus 1° ADH [10], and its resting cells have been used in
chiral biotransformations [42]. This enzyme has high stability in the presence of solvents
[59], but its low catalytic efficiency toward 2° alcohols and ketones might limit its
applications.

The importance of biocatalyst longevity and stabilty to their large scale utilization
makes necessary the development of strategies to stabilize mesophilic enzymes or the
isolation of highly stable analogs to them. Inuinsically stable and active at high
temperatures, thermophilic enzymes offer major biotechnological advantages over
mesophilic enzymes: (i) once expressed in mesophiles, thermozymes may be purified by
heat treatment [60—62]; (ii) their thermostability is associated with a higher resistance to
chemical denaturants (such as a solvent or guanidine-HQ); (iii) performing enzymatic
reactions at high temperatures can allow higher reaction rates, higher subsuate
concenuations, and lower viscosity; and (iv) there is a higher product yield during certain
reactions due to chemical equilibrium shifts with high temperatme. Isolating and
characterizing thermophilic enzymes therefore, provides natural examples of structurally
stable enzymes with high active temperature ranges that are potentially robust industrial
biocatalysts. Comparisons of these thermophilic enzymes to their mesophilic analogs will
also provide insights into what molecular interactions confer high folded protein stability.

The extensive ADH structure and function research literature have been well reviewed [1].

 

 

 

 

AI-.
mu“

8
The emerging ideas regarding protein thermostability, thermophilicity, and folding

molecular mechanisms will be reviewed here.

THERMOPHILIC ENZYMES

Thermophilic enzyme sources and diversity

Originally, thermophilicity was a property associated exclusively with spore
forming bacteria. Thermophilic enzymes were believed to be unstable, with high protein
ttn'nover rates explaining why thermophiles did not grow faster than mesophiles [66].
However, Thermus aquaticus, a non-sporulating thermophile, was shown to express
inherently thermostable enzymes [63], overturning these hypotheses. Subsequently, most
thermophiles and hyperthermophiles [64—69] have been shown to possess inherently stable
enzymes that function at temperatures above the organism's optimal growth temperature
[65,70]. Enzyme thermostability is the protein’s capacity to resist irreversible thermal
inactivation, and is commonly reported as the enzyme’s half-life at a given temperature
[71]. Enzyme thermophilicity is defined as the temperature at which the enzyme is
optimally active [71]. An enzyme is thermophilic if it is optimally active at temperatures
above 60°C [71]. Although thermophilic enzymes typically originate from thermophiles
and hyperthermophiles, some mesophiles produce enzymes active and stable above 60°C
(e g. pancreatic ribonuclease A). Mesophilic enzymes are optimally active between 20°C
and 60°C [71]. Mesophilic enzymes, typically from mesophiles, include most eukaryotic
enzymes and those from mesophilic bacteria and archaea.

All known thermophiles (optimal growth from 60°C to 80°C) are microbial
including bacteria, arehea, and some blue-green algae which grow at temperatures up to
60°C. The predominance of prokaryotic life forms at thermophilic temperatures is
consistent with the appearance of prokaryotes while the Earth was much warmer, with
eukaryotic life forms evolving much later. Thermophiles have been isolated from hot

 

catchments inch;
sprigs, soils, shal‘
(ii) microbially se‘.
and (iii) industrial
sludge sysrems, o
diverse as their .
htmmphic‘ Ch:
among Others.
Tables 1
rhcrmophmc and
mmmbiliq' y
Itflecg the lack
Specific 111cm 0d
Cases, it inCluds
Potential blotec
thermophiles \t
Enzymes haVc
mmstablc “
growth tcmPCT
Sfiingenuy for
alcoho1 dehyd
[GAPDHs], e
function of m
thumophllic ‘
mbstmtc Dr C
D°3Di

usually haVe ]

9
environments including: (i) natural volcanic environments (continental solfataras, hot

springs, soils, shallow marine and deep-sea hot sediments, submarine hydrothermal vents);
(ii) microbially self-heated environments (e. g., manure, coal refuse piles, compost piles);
and (iii) industrial environments (e.g., food industry effluents, hot-water lines, sewage
sludge systems, oil drilling injection water systems). Thermophiles, as physiologically
diverse as their mesophilic counterparts, include species that are aerobic and anaerobic,
heteromophic, chemoorganotrophic, chemolithotrophic, automophic, and phototrophic
among others. .

Tables 1 and 2 list enzymes which have been characterized and/or cloned from
thermophilic and hyperthermophilic organisms, respectively. Thermophilicity and
thermostability properties are included where available. The heterogeneity of this data
reflects the lack of consensus on the way to guage these properties (we have proposed
specific methods for standardizing these data [71]). This list is not exhaustive, in most
cases, it includes only examples of each enzyme type and it focuses on enzymes with
potential biotechnological applications. An extensive list of enzymes purified from
thermophiles was published by Coolbear er al. [70] and detailed descriptions of individual
enzymes have also been compiled [64,70,72]. Thermophilic enzymes are inherently
thermostable with optimal activities at temperatures near the original organism’s optimal
growth temperature. The proximity to organism optimal growth temperature holds more
stringently for enzymes within a single su'uctural family (i.e., a—amylases, proteases,
alcohol dehydrogenases [ADHs], glyceraldehyde-3-phosphate dehydrogenases
[GAPDHs], etc.) than across a range of proteins, suggesting that stability is partly a
function of the match structural fold. Note that the Optimal activity temperatures for some
thermOphilic enzymes, oxidoreductases in particular, have not been determined because of
substrate or coenzyme (e.g., NAD, N ADP) instability.

Despite being synthesized in a mesophilic host, recombinant thermophilic enzymes
usually have kinetic and thermal stability characteristics identical to those of the native

 

 

 

rr\r=\elr\ Ii... hi- ..itl .Illllttl lit-istll: ilhlmufldddlll
idflllll A. “mu—N. ._.+ tr newcomeui can
. . - . REG
. - .. . 3 0 5w. NIQNU we ArU 8v .3.\\93 §~kt§h§¢§~b¥< gzkogsozé MN
31 I. bore 3 v o o mmm<hUDQWMOQ~NO
.. fl: 7 Emcee—505 hemom—EQOCCDE Engage. Signage
.0 Z GUM VHHHNWMWU . .0MCAN—hmu DEF-mm Emu-CUMLO ”EN—m

 

henna-SHED U=uutnhDELOCu Eavé HUEHNDEEO-FH o‘ O‘Conph

 

 

6.3,: 6.8V 5. §cm3§ec§es .

8 as .m .o .m sea 2&8 sex 336.? §§§§3§a usage

a as .m .o .m a. 3:855 sausage. d 6sa$a§a<a

% as .m 6 .m a. 336.8 §3§§§ d 63,5033...
6.83

a use .m .o .m ”use. 58 *2 095.8 6.3553acficesssmcaaq assuage
SE a as we

a .3 as .m .o .m 5 3368253 6.2 suaasEsecesa .m saga
E06 28 n +v

a m 6.825 ”NA.“ 6.8 assess seem 3535.

cm 0.3%; 3am an 35...: aaaecfi 8:82

a m cause 2 836.2. sumacEcsces. .m aaaeaaaxoae8<.z¢

mamas—SE»:

a m .o 6.93 323.2% a 88: gsasacfia

S .55 .m .U naiueetoeaoaoau .m 82e— §>32%e§.m

8.? as .m 6 6.3389. 6.8 33.8 35...: ESE? <75

2.“: as .m 6 65355. 9.3.35.8 8-83.3 secesficc a. Essence.

magma—age...

R m .o g. .a. cafiwcesfie unease

.e
2. Quebec 838:8 §to~§o§a§§o§ 88262 85.6
6&8 agencies...

.axo .m .U §~§§ goagogtofi .28? 583m

3 vegans. oaen 658$; 23.836 senescence;

28.. 6.3588 6.3 Eaasaescea. 33% 3.20

8M 23 8303523
an? 6.83%... «53.2 6.8:.62 Sausage: geesecéaé
mmmfioaemmeexe
38888 35:88:55 36:28:55 Age 539% 282—3
.62 com 30:00 gem gem :32."qu gem

 

2:35”... 3:53.505 5?... BEES—Eon... .n 2.2:.

Swami 2. Cf;
». 3 n. _ Uta i \1 §U>Um 0:33 .Ra. §.~\U~UEQKV~§§X»EVRN §=u~3=38~kfl=<
N... :53 .m .0 .w. U%_E Cw W: 35950 006$» QENQQSSM K 0vdca~==~a0~>ne<
Done 2:“
. «e w- Uonxxt 3:033. 66.. .L\Uocm-hn §~§§§3he§vktgvi~ gxfithéxb 0vé§~3=3nrv~kwt<

 

11

we
be
we
99 com
on new
we
2:

2. com
8 .2:

no“ .2:

9:08

QB 8m
9: 08

a
ma
3 3m
.3
3

mm
mm

3

md

.55 .m .0

£98 .m .U .m
.HQXD .m .0 .m

.55 .m .0
.axo .m .o d:
.hQKO am .0 .m
.HQXO a” .0 .m

m
.HQKO am .0 .m

m

§e§u§ .u
9858 3:308 aimsszssgm
9%“me «£508 608V .3. §§§o§§§s
5% 258 *8 09:83: Oévgaaasfié .u
0% cu afibeugaau
mange mm 9.2. §~8§~ g
935:. 2 08,503 .a. 338m
00828 w
BoonSS <3 3.5.»:
was} mm 3392 §§€§u§ s
<mm+
odmeoomtsa om hwmanaofi afibcefiauu
008 333‘83
§§fi§~8§8u
oon§_ $5008 §=8§§ U
008?? on
5% 3.85 33008 $39953“ .u
E mama—DOS §=§E§§a§su
60938
=33:vo
£25m i 98% 608.8
Sam 26% $2. max—893 “Em uuzuméauefisfiu H
00358 R 330% 358%» g
$392-3 $83..”
9253 3305? 3.338 a
2 0&2: 608V Em §gut§=§§ s
6&9 ”53.30398”
conga a. up? §Es§§3§$
Gone Bios
0°82 3.98 008 in §E§E§u§
0%? WESOS Goaegaésss» .5
Song 85
9.32 385. 93.5.2.2 §§§§9§~ §§o

§§§2¢
anagémé
Savages—0.8

9.8530
aéu8=6

asuooaaa
3.8355

885813
033%3 $.88

oasifi. {Scam
$5833; 25
§§=aé._ 2am

sausaa 1} 8am
945...»$ 25

gang—$8.53
@5396
$0836
93:2
gaggl—

own—553833
08335.5

0353.35
0853335

game—>5

N_—
———
0-—

.LQXO .

I‘Sun

AN—rvflu 2E W ATV
UochEE ON

hu-Hdu ”slung: *8

WEE

CS 3002.
Odin
vowels

~ 6:“ .Rn :QSCQMN
< a Vg “ .§h. §EKN~$N

”(VS m .Kh: SERV‘rN

h§\\-\Q§N~\\A~K~\V\h 5‘

9.6.5.29:
Omacmugé

022:»qu

OWQCQ-~==QAVOZ

 

12

0Q
mm“

as 000
«Q 000
mag

NS
-~
ofi
03.x:

b:
.m:

v:
m:

0:

N:
g:
a:
mo—

m0 .098 .m .0

m
m
m

m

m
.098 .m .0

0.00% 2

SUEZ

Ea .0EESBw

.+~w—2\+N:E+v
055m 038m

0owQEE 3
Ba“ 0380

0.0%}... 0350

0% we
0ontfia mm
oootfi

0cm» 3 a 2
000.00 0.600 gowA

0.228

0.3.: .8 055...
306 283
0.0%.... a

5 0.822
306 28 0 +0

0.238 cm
306 28 0 +0

0.0%.? 8
0.83

.00.. 268 $8

0.00.5.8
0.5.00.8

00%
002.

0.00.00.00.05
0330.2.
$03.8
$30.8

8.0::
3.3%
DOE
0.2
0030.8
3%
0.3.00.8
0.0%
0.800

053095052003. .m

53.000030000005030 .0
Cost §03m§§00§05
~§§03§02

33%....050880 .m
Eaimcumuu .h
ggoéfiuaogwi .0

fluuboguua .0
§0ub0§008 H

in §E§E§E0§
6.80 338%.. 0583955
E.» 300:8 3&0...”

2.9.2 .0... 3.....5
«0 3:80

05.3%.505200: .m
32 .5. 0.580
<32 .0. 35.2:
«$3 .5. 352:

§~3§~E050§0~u .m

8038M 083x

008088m 829.800 085.

Sgafioaoofimoam

00805:? 038085

805:? 00005

00.2030
0020833408002m
mmm<03
nz< $030028—
.mmm<>q

38.83“

30283“

9.8.53;
0082000003
38035.
85—352
0353:...
8.83:5
Oman—musk—
820.80
08:08.0

8083:0002

 

.tbntkyficcu :02. E: SE £8.22 :C 02...:5222Q £5 k0 hymum 0:”
4 WNNJ t3.v.§.¢«yk~w~\.ub~\h §~\\.fiv§\»\30 \QVUCCOK a

 

F. 1:: .5: 4m .L a: v: §>$~§E§S K 3.22:9: 961*]
am. R 0533. Qt 3003" an: uiuhuabu K 0255.94 0201*
.000 new
hN _ .NC 575 .m .U ~7va +9 0005 No. azhzlhvoomw ufifihkzlkwihgvs uh u§g~ 95th
IN—.FN- CLXU .V. NU 5:

P. . hm unabguw usurvhkwgvbhhu. uh GREECEN Oh°~sfiufl

13

@8298 9:55:88”: ”.98 "8803623 22%29méanonfibobc "mm—.20

” 2389655523508 ”E ”Egg 055808 “82 522—8885 um 8:3?6 9585 as»? "8 68538 um 60:20 no
dam—Boga 83 8: ms— :88». no Rufiaaa $5 mo .0168 05. 9

Anna gubcguuuu .55»ch bbauom a

 

 

 

55: .55 .w .o E 2 £3 855 s afiefias x

ad 3 Oman? c.5502 mm: 95:38.5 afiefi as?
~UOU ad

5% .55 .m .0 «cm: +v 955 Ne 2:308 SumEsasE§ g 035:3 as?

«2.3— .55 .m .0 E. 053808 gogcguua K E 803K

I ll-lnlld'H‘ ha‘lfiéh q
%u&%<&UDQHZC~=K:
. . «95.x: TEEEXNES 530:“ 35.55%
. b 52.537.5be 5.. «2.5% a .
a Z guy. {vamp—”flu“. 02:05:”.— Utrmhcl Sui—3M5 ”SSW
. . . . . .

nEa~=¢ML° 0~=IEOEL0£Pm0QH£ Eng EOEHNQELOCE. .N mine-3B

 

14

w:
>3

9;

a:

3:

m:

m:
3.3:
N:
.33:

2

3;

w»:
.5262

m2
we 08

V9
mmfiNmfi

S
9

 

.oZ mom

.55 .m .U
.55 .m .0

m0

9.0% 2
UoooHEN

93:3. 2:
00358 3.2

camsoooSNA
woman

9822
008% NE

DonSBm.m
UoOmEN

Gone :2 H8 03%

000%

UooaA
DonoA

gives
33.5%
98%

330%?
35.502
odmnaoafi

Dona

008A
33de
608A
”.mmeuoow
wwnfig
Dom?

608: .383 §Em§©
$33 .m

398:3 .:
Gone
Eugx fingegs:

Aoomwvwefiiuxaafixszgax

~85. K

Come §§ afieigwfi
Coco: .59: “.3885
60% gurus swag»:
3985 H

§ufi8§8 .m.

2.82% .m

608: vmm
£22.: .5

.38th .m
608: §uot§§uu8§£
6&8 asks: 398$st

Comm
not §utu~§8 azasebam

820»ng

@503on NE ofiuowofimm
3688 5:696

emanowofimm

Ao>uo§bnv~b aaaomeumm
$920

$956 $920

a<zv new

$925

$920
GEQKEQZV
ommcowegaou 038820
Egziév
ommcoweumsoc BERG
EQ<ZE<ZV
855ngon 088820
GEBQEQZV
eacuwegnoc 038820
9.888526 eagouafiom
C038 .5583 5283
Executom
88032625
Eocene 03%?
6E§¢E<E
amazemegaou 3:82
362M336
8303638 6:82
mHmfiwUDGHMOQ—NO

gum

 

2:3ng 95.395.555.32. 59: nthNoEuuch. .N 035—.

 

 

»: usmhgjk?
3 _ «2x: rat: :.5 :5. 3:... (be. 5A .2§§5 .\< omucumEchun 5&3} ~HNIU
3:39): :5. 2 :.~\v..c «.cotcunohémmt
an _ Sc: 033: $59 -néIELonh .60an T335». QKAQEQS»: Ovdcomeni‘cou E2 fume
mm. m .U “.33:ka .\< unacomemrncoh 03?:
. n _ .Cr._ .Exu .m .U Uonxxfieom mfilfibmnmA gtfi K omucomnvaukcoh 95.93'4
Uonxéur: Cw. r 6.0 fichfiA grg uh «Ntmoanvbhl NE Ohflcom‘pg:

3V—

 

 

15

ouQNB

0343
3 vow

of
on“
onfimmfi
52

$2

m2

v2
m2
N2
Eficfl

a:

:98 .m .0
.35 .m .U

m
.Exo .m

Dong—ca

Dong
Geog—m

Gone .3 é 035m
0082“ u

.5 DoowEn fl
00838 3

.5 Geog mm
avg 23

0358 8
Gamma EC
Deco 8 038m
9% 05

00858 2.2
AVOAENM
28 8: 05%:
Baud 038a abs“
23 05 Dog

3 356an >682
Geog 333“an
.3292 using...
Ba 0083
has 028a

$8“ 53x0 =8
0.6%:

Ban 2505 *8—

OomQEEOm
000258 cm

0

a
fi

0

can?

9E

92

3E .98
008A
nomeuoo?
fiasco?
nsmgboc?

cameo...»

-NvOm .-~v0m
2 SS 5
33:3,?

.3. 2m.5°8A
=3. 2 9S6
Agave:

Steven?
mgamg
DORA

3g .m

fleas s
guns g
333 a.
3.3% .k
95.29: .5
$3 .m
”333% .<

6.8 :fiméafimogué

Egg .3

E3 .3
0.3 533 3.333.»:
3.2% .3
SE2 .5

gtg K

a Saba—on <75
Agufiboq :63
fl unsung—om <29
gfigofiosaé‘v
3.353 .5.
mmm<mmhmz<~rr

5x8033—
$53835
Exouohoufiag
95032025
niouoboaagg
gag—ocflo
.6835 ”an;

$32
mn<z 588.5 8E

98303.
buvmv 850:3. gamma

35881.:
0383628 gamma
ccovcomuubnvmv
8203628 $25.56
game—“Eon 0332
3.83638 283;

85938; NE sagas?

 

_€ .uaxo .v. .U uS§.~>kN.\. .\< Omucmu~ 0~EOUK~WC£Q8£Q-N

3.5m TV Ecsuaczv -NVE
£.¥.._~¥ .2 n; + .-~..Ow .2 N . owonucuhigo.‘
«1.: AUOCOV O_£.Zm $87.. \WCIQOCOQ AUONQV ~\N\§ \Q E5105

RV hm»
me — hibhthNK-g 38°U°§Mu§lhflvQ -N OWQLUENOAN <ZQ

 

 

16

EA
3 3»
m2

mi
52

0562
v2.3;
~S
NS
EL
0:
OS

o2
mg

5363
n3
8

#2
m3

.Exo .m .0

. 00338 8
Domain .5 Dong—em
UowOQEE E.

00338 cm
Dong—N

Augean—«v

DOSES 3

8835205
00858 34:
002 <58 8

so“.

28 ~55 289
008:5:— n

305 289
Dongs

92

can «.06 289
05238 N“
008.38 cm
00838 cm

€58 2:5
0%
Doc“ <5.

0.69 3 gm
3% 968 $3
003 “Ema”

.Exo .m .0

23 3 03583
5.5g 2 3 +
6.8V 2%.... $87

2330.?

qfimbone
wanna

Done

2.3393

29.5.2
oh:

396 283
339.5

UoMNTo:

366 289
08.5.3
3:308
c.9503

363.508
03.5.8“
3.3%

608“
n.9eoo8_

anOnm
flqum 2 N
\QEDOS

$.55 .5
§§~§8 .m
E. swag»:

6.3 .a. $35:
SE .5

39.35%8 .w
Coma: §§§8 gfifié

.3985 K
vmm
.383 .m
gates .5
£55 E

0.68

Saw—SEE 3889535
.369: .k

.383 .m
”:83 .m

3..th .:

azuubfig §§8§3Q

emacaoamfiifiéxm
9.83833
aspifiséiiém

ans? $.25
08533 {Tecum—

ggoflao
x0388 oé<
0853:5335
guano—>3
gag—E
gages—mggé
35¢

35.3275

own—>38

Aaazoogfiv 0333.8
A§=o§§ 82550
mmmafiomA—w:

38: agaoamsfi.~

gamma—Em
.325de

E 3.53.8 <75

do.
Ni

—I—

Ox-
: 00v.

. Ecru $9.4m. no.2:

ongmg

35.2 35:53
onaxznd
Dunc—>5: CN

3: 3303
99¢.th
009 ~
O.h:n~\uv:

nah‘wxxx. .V
vaccxeaxxah

§h§~§\. .nN

.Qu. QankENSN
605 my K..~\.U.U b§UUQ§U§.N

 

859632vo :5 .1" mu 5
DEE AV 2810:8050:

§Umn83~o -5

OfiflnmfiogNOI
Olefivmmogzo

17

3208628 22%29338385 "man—<9 §§§-_§é
-mz ”Hm—23.010 nauounogfiofigoficoifiofiézwz "gamma 69.8898 359.5888 ".55 63838 ”m 628—0 "0

 

§ 98:58 2 9.218 Bangs 03588.32?
:2 55 .m .o 935$ SECOR 5388.5 38089.. as?
awfi u: v: .889: .m 8805:? 8520
5: 99:5 9.8.3 33.3: 9.505% 8.8920
53 0&2 o. 2&8 .2333 .2 > “$888699 5.5
mm 983 :z<
.mmm<mmzom_ .3355
22 Egg asuogofiea
83.: 9.35. 0.5.5: .& $395.3.“ E§§Q¢
9: 052388 c.2305: Gone .a .5885 9.855%...
3 8... 0.32% vane 33.3 .m 3.2.28 088%
98$? 8
3: 50.83.. can: names.“ .m gazing assaofiom
on”
3 8.. 9.822 9.2: .3822. 388333 0835
UES 5222
a: 5% 263 $§ Emma? 53 .2. 8%63205 guaa
as 3 03923
SEE 2 fl
.23 6.8 as 8
§ 5% 035... $8“ 3:502 gamut .< 823.3%: E2315
NM: 33%..” §fi< 23.055802
Sivan
:2 0.3..va Doc: §§é .m sagas
$8 .8353 .
om: oowafid qua? .& amass»: afissa
38m uoneaficm 0.5.38 38.353.“ 86.886

 

 

 

 

protein, allowing
thermophilic on:
war: cxtcnsivcly
fluorosccncc cmi:
(GuHCI‘fdcpcndz
constants, the 53::
calm: [141]. P
lht 83m: Michael
P. woesei cnzyrm
”WWW dept!
flammability w
dcpendcncc on pe
Wm adaptarim

Asmom 54

“Wins, mt CV'l(

 

 

18
protein, allowing the use of standard molecular biology techniques and hosts for

thermophilic enzyme studies. T. maritima and Pyrococcus woesei recombinant GAPDHs
were extensively studied. The T. maritima recombinant GAPDH had the same
fluorescence emission and circular dichroism (CD) spectra, the same Guanidine-HCI
(GuHCl)-dependent denaturation and heat-inactivation behaviors. the same Michaelis
constants, the same allosteric inhibitor effect, and same specific activity as the native
enzyme [141]. P. woesei recombinant GAPDH had the same heat—inactivation behavior,
the same Michaelis constants and Vmax, and the salt effect on stabilization as the native

P. woesei enzyme [143]. Thus, proper thermophilic protein folding was not strongly
temperature dependent, and all the information necessary for protein thermophilicity and
thermostability was encoded in the primary structure of the enzyme. This thermal property
dependence on peptide sequence suggests a direct genetic mutational mechanism for
enzyme adaptation to the thermal environment of its original source.

As more sequence information is gathered on thermophilic and thermostable
proteins, most evidence indicates that the catalytic machinery from thermophiles, like
thermophilic organism genetics, is homologous to that from mesophiles. Thermophilic
enzymes contain the same amino acids, the same catalytic consensus regions
[135.152.160.190], and, in many cases, enzyme structural analysis reveals that they share
the same 3-dimensional backbone as their mesophilic counterparts [151,160]. No distinct
structural features unique to thermophilic enzymes can account for their activity at elevated
temperatures and their resistance to heat denaturation. However. attempts have been made
to define amino acid “traffic rules” for high temperature molecular adaptation. Hypotheses
trying to explain the increased stability of thermostable proteins include: (i) or—helix
stabilization by substituting helix-stabilizing residues (such as alanines) for helix-
destabilizing residues (most often glycines) [192]; (ii) short-loop and [iv—turn rigidification
by introducing prolines [193]; (iii) increased average protein hydrophobicity [85.143]; (iv)

increased number of salt-bridges [194]; and (v) a decrease in cysteine and deamidable

 

residues. Howev
and sequenced. en
rules.

No trend 1
from the thoroegt
sparring an 82°C
Hrsoptilie, ther'm
families are comp.
that features acco;
3301316. Enzymcg
mixmmtes abet.
Table 3, uith few t
M Q'SItinc or d:
mmphflic Count
W5. Thc A
their temptramrc ft
Amfiln Content a;

to}.
C

h

as
frences, Thcsec

lithium, 1'
le Cmi

19
residues. However, as more genes from thermophiles and hyperthermophiles are cloned

and sequenced, examples accumulate of thermophilic enzyme sequences conuadicting these
rules.

No trend in thermOphilic GAPDH total amino acid compositions could be identified
from the thorough comparison of GAPDH sequences from twenty-six different organisms
spanning an 82°C growth temperature range [195]. In Table 3 the amino acid contents of
mesophilic, thermophilic, and hyperthermophilic enzymes from nine different enzyme
families are compared. No prominent general uend is observed and it is highly probable
that features accounting for protein thermostabilization differ from one enzyme class to
another. Enzymes from hyperthermophiles, in particular extracellular enzymes stable at
temperatures above 100°C, show a high resistance to amino acid covalent modification. In
Table 3, with few exceptions, enzymes from hyperthermophiles show a slight decrease in
their cysteine or deamidable residues contents compared to their mesophilic and
thermophilic counterparts. This uend is particularly noticeable among type II xylose
isomerases. The Asn-oGln contents of four type II xylose isomerases was compared to
their temperature for maximal activity [190], showing an inverse relationship between
Asn+Gln content and temperature for maximal activity. Although it is interesting. the
authors acknowledge that more information is needed on proteins from thermophiles and
hyperthermophiles to confirm the statistical significance of this uend. Contrary to the early
belief that thermophilic enzymes obtained their thermostability from a different amino acid
composition. enzyme thermostability appears to be enhanced by numerous subtle sequence
differences. These differences protect the highly labile residues and bonds (by modifying
their immediate environment). they limit destabilizing interactions, and they restrain protein
flexibility. These protein alignments and amino acid composition comparisons clearly
show that no simple, general protein thermostabilization traffic rule can be defined.
Therefore, elucidating the molecular mechanisms of protein thermostability and

thermOphilicity will require careful study of protein molecular structural interactions and

 

20

Table 3. Amino acid composition comparison of enzymes from mesophilic, thermophilic, and hyperthermophilic prokaryotes

 

 

Protein Class Gly Pro Ala Val Leu Ile Phe Trp Tyr His Cys Met Ser Thr Lys Arg Asp Glu Asn Gln Asn H— Polar Char
+ phob god
a]
a-amylases
meso (13) 8.5 4.5 8.8 5.7 8.0 4.4 3.8 2.9 4.1 2.5 1.4 2.0 7 6 6.8 4 1 5.5 5.8 3.7 5.2 4.5 9.7 51 20 29
therm (4) 8.9 4.8 8 1 6 4 6 9 3.8 4.4 4.0 5.9 2.3 0.75 1.9 5 6 8.3 5 1 4.1 7.4 3.6 4.0 3.3 7.3 53 19 28
hyper (1) 6.8 4.3 4.0 9.2 9.1 6.3 5.7 2.6 6.3 2.2 0.31 1.8 5 1 1.8 7 4 5.4 5.4 10 3.7 2.3 6.0 54 11 34
neutral proteases
meso (11) 7.7 3.7 8.4 6.6 7.4 4.2 3.2 0.90 5.2 2.3 0.57 1.2 8.6 6.7 7.3 4.4 5.9 4.7 6.8 4.3 11 47 19 33
them (2) 12 3.3 9.4 7.7 7.0 4.3 3.0 1.3 6.9 2.0 0.37 1.9 6.2 6.3 3.7 4.8 6.0 4.2 5.4 4.3 9.7 55 17 28
1° ADHs
meso (5) 8.2 7.4 11 6.4 7.4 4.9 2 1.4 2.6 2.1 1.6 2.8 7.2 5.3 5.0 8.4 5.0 4.0 3.8 2.5 6.3 52 19 29
therm (1) 10 4 7 9.9 12 8 5 6 4 1 9 0 88 3 5 2.3 2 0 1 7 3 4 3 5 7 5 3 8 4.7 7.0 2.9 3.2 6.1 58 13 29
hyper (1) 11 3.5 8.4 7.5 7.9 8.9 3 0.63 3.0 2.1 2.3 3.0 4.8 5.1 5.5 4.9 5.7 4.9 5.3 2.7 8.0 54 17 29
2° ADHs
meso (2) 11 3.5 8.4 7.5 7.9 8.9 3.2 0.63 3.0 2.1 2.3 3.0 4.8 5.1 5.5 4.9 5.7 4.9 5.3 2.7 8.0 54 17 29
therm (1) 12 6.2 9.9 10 6.5 7.4 4.0 0.14 0.17 2.8 0.11 4.3 0.26 3.7 6.8 3.1 5.7 6.0 2.8 0.11 2.9 60 14 26
Glu DHs
meso (4) 11 3.6 9.5 8.8 7.0 4.9 4.0 1.6 3.1 1.5 1.2 3.6 4.9 5.0 6.1 4.5 4.6 7.8 4.3 3.5 7.8 53 16 31
hyper (3) 8.1 4.6 9.4 8.4 5.5 7.5 2.3 2.4 4.8 0.87 0.32 3.4 3.4 5.7 8.1 4.7 6.0 8.3 3.6 2.5 6.1 53 14 33
DHs
meso (10) 6.8 4.6 10 6.8 8.8 7.3 3.8 1.1 3.1 2.4 0.94 2.4 6.1 5.4 6.3 4.9 5.1 6.0 4.5 3.6 8.1 52 17 30
therm (5) 9.4 3.6 12 10 8.9 6.6 3.7 0.95 3.3 2.2 0.38 1.3 4.1 4.1 3.5 6.8 5.6 6.8 3.2 2.6 5.8 59 12 28
hyper (1) 9.6 3.5 8.6 9.5 10 7.5 4.0 n.d. 2.8 2.0 n.d. 2.0 4 1 4 9 6 5 4 3 n.d. n d n d n d 10 56 13 32
MDHs
meso (1) 11 4.5 12 10 10 6.4 0.29 0.0 0.13 0.96 0.96 0.13 5.8 5.4 7.0 0.26 3.8 6 4 3.5 4.2 7.7 58 14 28
therm (1) 8.2 4.9 15 8.6 9.2 5.2 3.0 0.12 0.21 0.12 0.30 3.4 0.28 4.3 5.5 5.8 5.2 7 3 3.4 3.7 7 1 57 12 31
hyper (l) 9.7 4.4 7.7 6.5 8.0 11 0.26 0.29 0.24 0.18 0.59 0.26 4.7 5.0 9.7 0.24 5.6 8 8 5.3 0.88 6.2 52 ' 15 33
D—xylose isomerases
meso (8) 8.7 5.8 11 5.5 8.8 3.4 4.1 2.0 2.5 3.2 1.5 , 2.1 4.4 5.1 4.4 9.6 5.6 5.3 3.1 3.9 7.0 52 16 32
therm (3) 6.5 3.0 9.0 4.8 7.4 5.1 8.0 1.1 4.8 2.6 0.53 2.7 3.8 5.3 8.5 3.9 7.7 7.5 5.1 2.4 6.5 50 15 35
hyper (1) 7.6 3.6 7.9 5.0 9.0 5.6 7.9 0.11 4.3 2.5 0.68 0.18 2.7 5.6 8.6 4.7 7.2 9.0 3.4 0.18 3.6 52 13 35
Gln synthetases
mo (12) 7.0 5.3 8.1 6.2 8.3 5.6 5.0 1.1 3.8 2.7 1.2 3.0 5.7 5.3 6.9 4.3 6.0 7.2 4.4 2.7 7.1 50 18 32
therm (l) 7.2 6.8 6.8 6.6 8.7 5.5 4.7 0.13 5.9 2.1 0.21 2.3 5.5 5.9 6.6 5.1 6.8 7.0 2.7 2.1 4.8 54 16 30
hm (2) 7.3 5.9 7.1 6.3 8.2 7.2 5.5 1.4 5.1 2.4 0.34 2.6 4.4 4.0 7.3 4.7 5.8 10 3.5 0.91 4.4 54 14 &

 

n.d. - not determined

meso, enzymes from organisms with optimal growth temperatures below 60°C; thcmt, enzymes from organisms with optimal growth temperatures from 60-80°C; hyper, enzymes from organisms
with optimal growth temperatures above 80°C.

Numbers in parentheses indicate the number of proteins represented in the category. ,

GenBank database accession numbers or specific references for the proteins used in the comparison: Primary alcohol dehydrogenase (1° ADH): meso - D90004, 1302104, M91440, X17065.
X59263; therm ~ D9042]; hyper - $51211. Secondary alcohol dehydrogenase (2° ADH): meso -103362, M84723; therm - [196]. Lactate dehydrogenase (LDH): meso - D13405, L29327.
M22305, M72545, M82881, M95919, X01067, X55118, X55119, Z22737; therm - D00535, M19394, M19396. M28336, X04519; hyper - ref #320. Glutamate dehydrogenase (Glu DH): meso -
K02499, M24021, M65250, M76403; hyper - L12408. L19995, M97860. Malate dehydrogenase (MDH): meso - M95049; therm - 102598; hyper - X51714, Glutamine synthetase (Gln synth):
meso - D00513, D10020, 1.05609, L08256, M14536, M16626, M18966. M22811, M26107. M57275. X04880, X53509; therm - X53263; hyper - X60160. X60161. Neutral protease: mo -
D10773, K01985. K02497, M36694, M36723, M62845. M64809. M83910, X61286, X73315, X75070; therm - M11446. M63575. (it-amylase; meso - 101542, K00563, L19299, M18244.
M24516, M25263, 004956, X07796. X12725. X12726. X52755, X52756, X55799; them] - M11450, M34957, M57457, X02769; hyper - 122346. D-xylose isomerase (XI): meso - 1.12967,
M15050, M36269, M73789, M84564, X00772. X02795, X61059; therm - 105650, W699, M91248; hyper - L38994.

 

 

energetics. Them

organisms have bt

Thermostabi Ii t]

Protein the
lnacrisation is due
protein regions, or
btiicved to bc a 1‘3;
£59m°smdfic. n
pamcuhf emimnn:
[Emitting (detenni:
mgt) in a saxnplc a

Tommc and Kliban

nafiVC enzm

(aCfiVe) ‘

Representaums of [it

My linet'tlr. fining

William“ time }

 

Inlfcsj

 

 

21
energetics. Thermostable enzymes isolated from thermophilic and hyperthermophilic

organisms have been termed thermozymes [71].

Thermostability

Protein thermostability is the protein’s resistance to irreversible thermal inactivation.
Inactivation is due either to protein chemical modification which destroys specific critical
protein regions, or to protein denaturation (i.e., irreversible unfolding). Denaturation is
believed to be a rapid cooperative process. once a threshold molecular energy is reached.
Enzyme-specific, this threshold is related to the characteristic protein stabilizing energy in a
particular environment. Thermostability is typically measured as the fraction of activity
remaining (determined by standard enzyme assay in the enzyme’s activity temperature
range) in a sample after heating it at a constant temperature for a specific period of time.
Tomazic and Klibanov proposed a general model for enzyme thermoinactivation [197]:

native enzyme 4: non-native —-> scrambled structures [scheme 1]

(active) (inactive) (often precipitate)

Representations of the natural log of enzyme residual activity plotted versus time are
usually linear. fitting a pseudo-first order equation {Equation (1). with k=rate constant and

t=incubation time} .

ln(residua1 activity) = do [1]

Equation (1) contains no protein concentration term, suggesting that the rate-determining
step is intramolecular with respect to protein. The only way to ensure that a pseudo-first
order rate law reflects a truly intramolecular process however, is to measure thermal

inactivation at various initial protein concentrations, verifying that 1: remains constant.

Temazic ant
thermal inac'
Prote
protein mole.
thermophilic
recombinant
and in differ:
native enzm
Ihcrmosta‘oih
Whfional 1
Im'crsihle n
dcnamranon,
Min chum
infO‘TDation 11
(9pm? 304
m8 [203].

22
Tomazic and Klibanov’s model is consistent with an intramolecular rate~determining step in

thermal inactivation if the natine to non-native protein step rate determining.

Protein stability is due to numerous ionic and nonionic interactions within the
protein molecule and between the protein and the environment [198]. While some
thermophilic enzymes have been shown to be stabilized by glycosylation [199.200], most
recombinant thermozymes - expressed in mesophilic hosts, in the absence of glycosylation,
and in different cellular environments - remain as thermophilic and thermostable as the
native enzymes [141,143]. This observation demonstrates that thermophilicity and
thermostability are encoded in the peptide sequence; they are neither a consequence of post-
translational modifications. nor of non-covalent interactions with cellular components.
Irreversible inactivation of mesophilic proteins typically results from irreversible
denaturation, due to the disruption of numerous non-covalent interactions rather than to
protein chemical decomposition [201]. Thus, the original protein conformational
information remains intact in many denatured proteins. The protein stabilizing energy
(typically 30-65 kJ mol'1 [202]) is lO-fold smaller than the magnitude of the opposing
forces [203]. Since each additional hydrogen bond or salt bridge can contribute
approximately 2-20 kJ mol'1 to the stabilizing energy, and since the possibilities to add
hydrogen bonds or salt bridges in a protein are vast, it is likely that the protein covalent
modification rates, rather than unfolding, determine the theoretical upper limit of protein
thermostability. Ahem and Klibanov [204] reported that peptide depolymerization becomes
significant at temperatures above 100°C. but oxidative decomposition of cysteine.
asparagine. and glutamine residues has been shown to limit the thermostability of some
enzymes to temperatures below 100°C. The identification of enzymes active at
temperatures above 120°C suggests that the non-covalent protein structural interactions are
strong enough to allow stability to approach the theoretical maximum temperatures defined

by peptide bond destruction.

 

Thermophilic
While th
from 65 to 70°C
Thermophilic am
with variations ir
conditions of hi;
directly involved
denamration are .
umsr be more res
amines similar 1
atom positions lm
Attire Emptraturt
Differential scanni
Monably Constar
absent: 0f Signific
mm 3" tllJiCa

Mop-la] ”Gilliam

23
Thermophilicity

While the system kinetic energy increase (~RAT) is the same from 25 to 30°C or
from 65 to 70°C, the total molecular kinetic energy is greater at high temperatures.
Thermophilic and mesophilic enzymes similarly resist the structural changes associated
with variations in their molecular energy, but thermophilic enzymes must do so under
conditions of higher total kinetic energy. Therefore, while protein denaturation is not
directly involved in enzyme thermophilicity. structural interactions which oppose
denaturation are very important for thermophilicity. In other words, thermophilic enzymes
must be more resistant to denaturation, while having dynamic structural states allowing
activities similar to those of their mesophilic counterparts. Because of sub-A variations in
atom positions known to be tolerated in a functional active site, it is believed that, within its
active temperatrn'e range, an enzyme maintains its average su'ucture within strict limits.
Difi'erential scanning calorimetry measurements indicate that protein heat capacities remain
reasonably constant within their catalytically active temperature ranges, indicating an
absence of significant structural changes. Arrhenius plots for thermophilic and mesophilic
enzymes are typically linear. suggesting that mesophilic and thermophilic enzyme
functional architectures are similarly, tightly controlled tluoughout their respective
temperature ranges (significant su'uctural changes that alter the functional architecture are
expected to cause non-Arrhenius behavior). Biphasic Arrhenius plots reported for some
mesophilic and thermophilic enzymes [82,145,205] represent an important exception to the
typical Arrhenius-like behavior, however. discontinuities are not a specific trait of
thermophilic enzymes. In the cases of yeast and Thennoproteus tenax GAPDHs, it has
been proposed that the discontinuity extent indicates the degree of enzyme thermostability
and thermophilicity [145]. It is critical however. that such interpretations of Arrhenius data
are verified by careful biophysical and kinetic experimentation. More likely, the similar
Arrhenius plots obtained for mesophilic and thermophilic enzymes suggest that all proteins
respond similarly to temperature.

 

In their res;
names are more
personal communic
mesophilic counter
temperatures is obs
temperate rigidiz}
1027\in at these ten:
mural change or
more Vmax (kw,
throughout their ac:
MOM? Pmdicts
“mm. a set 11'.
Nuanandp a fun
M15 that a mom

Emily mt than a:

minimum AUhCI
mmnophilic analog I

magi-'8 COnthing

- Recent “'0'

T. ’lfhnosuzfun-gem

Change)

(—
. . _

 

 

24
In their respective temperature ranges, psychrophilic, mesophilic, and thermophilic

enzymes are more rigid at low temperatures than at higher temperatures (G. Petsko,
personal communication). Thermophilic enzymes are significantly more rigid than their
mesophilic counterparts at room temperature [198], however, increased rigidity at low
temperatures is observed in all enzyme categories. It has been proposed that the mesophilic
temperature rigidity of thermophilic enzymes [198,203] was responsible for their low
activity at these temperatures [140]. This implies that a catalytically significant enzyme
structural change occurs between low and high temperatures. However, the increasing
enzyme Vmax (kw) values with temperature seen for thermophilic and mesophilic enzymes
throughout their active ranges are typically consistent with the Arrhenius relationship. This
relationship predicts that the percent of maximal enzyme activity measured at some
temperature. a set number of degrees below the optimal activity temperature, is
predominantly a function of the reaction activation energy (Ea) (Fig. 1). Thus. the theory
predicts that a more thermophilic enzyme would have only a slightly broader temperature-
activity curve than an enzyme with a lower optimal temperatme and the same Ea.
Therefore, the Arrhenius equation can explain poor low temperature activity for
thermophilic analogs to mesophilic enzymes while assuming that only the system thermal
energy is controlling the reaction rate (there is no catalytically significant enzyme structural
change). Recent work in our laboratory demonstrated that E. coli. B. stearothermophilus,
T. thermosumm'genes. and T. neapolitana xylose isomerase Arrhenius plots were linear
[206]. Thus, the percent of maximal activity at any temperature below the optimal
temperature for an enzyme depended principally on the activation energy of the reaction.
These data clearly showed low thermophilic enzyme activity at mesophilic temperatures.
Also. the excellent agreement between the observed temperature-activity dependence and
Arrhenius theory suggested that throughout the active temperature range for each enzyme
no catalytically significant structural changes occurred.

Figure l. Arrhenius dependence of reaction rate on the temperature and

activation energy for mesophilic and hyperthermophilic enzymes. Tmax is
definedas the temperature formaximalenzyme activity andEais thereaction activation.

energy.

Relative activity

 

(7

 

Relative actlvrty

26

 

 

 

   

 

1.2
1 O I Ea = 80 Idlmol, Tmax = 90°C
' 'i I: Ea=80kllmol.Tmax=37°C 0
‘ e Ba:20kJ/mol,Tmax=90°C .30
0.8- o Ea:20kJ/mol,Tmax=37°C '0
o.° I
0.00
0.: f
0.66 0 ° q:
o
r .. O I
n
0.4- I
.D
In
'1:
0.2- '3
0.0 ::;:::::::IIIII-IIIIIIIIIIIIII'.."5.- .
.150 -100 -50 0

Change in temperature from maximal (°C)

 

 

The Arritcr
been attributed to a
rigid inefficient ca
Although a discor:
discontinuities can

other phenomena r.

 

 

til‘fifanphilic enzy:

27
The Arrhenius plot discontinuities reported for some thermophilic enzymes have

been attributed to a catalytically significant enzyme structural change from an excessively
rigid, inefficient catalyst to a less rigid sturcture optimized for enzymatic activity [140].
Although a discontinuous plot is not absolutely consistent with theory, Arrhenius
discontinuities can result from temperature dependent changes in the reaction slow step or
other phenomena not linked to enzyme structure [207,208]. However. whether
thermophilic enzyme Arrhenius plot discontinuities indicate a temperature dependent change
in enzyme structtne and whether such a change can account for the low mesophilic
temperatm'e activity of thermophilic enzymes, remains to be determined.

As kinetic data accumulate on thermophilic enzymes. it becomes evident that,
despite their activity at high temperatures, thermophilic enzymes catalyze reactions with
Vmax and Km values similar to those of their mesophilic counterparts at their respective
optimal temperatures. Thermophilic enzyme-catalyzed reactions were initially expected to
have high catalytic efficiencies based on extrapolations of mesophilic enzyme rates and the
assumption of similar substrate affinities for the mes0philic and themrophilic enzymes.
Instead, thermophilic and mesophilic enzyme Vmax values at their respective optimal
temperatures are typically similar. as is typified by a series of xylose isomerases (Table 4).
Substrate affinities for thermophilic and mesophilic enzymes at their respective optimal
temperatures are also usually similar (Table 4.5) [215]. These similar Km and Vmax data
yield similar optimal temperature catalytic efficiencies among analogous thermophilic and
mesophilic enzymes (Table 4). While higher reaction rates with increasing temperature can
be attributed to increased rates of collision and a larger fraction of the substrate population

possessing kinetic energies above the reaction activation energy, the temperature
dependence of Km is less clear [208]. Even for the simple Michaelis-Menten case

(equation 1)

..... 21.11. L.
Eases K
ghcfibxhfiocce K

 

 

 

 

 

 

u . .- evil :3 mo.

hm. 2.3mm Cmvfi. OK; We

hN~ CK: Cc. m 564 he Hutuktsgxadgus K

33m Cd? 95? .0 .3 cc SEQRENSQéfi .9

ON. Cm OCR. O MNtNCN \\00 .rQ
:.:=.—ZE\: =:=\: g‘gmog do. N

— .32 HOW ~ A Ev—bcov— flag! a .22....) EDMOTAK 50.x Eva ”ha-Egg Egg
ICB:SQOELOS. awn-I tfl-cfln-OIOE 50.: unfit-Luann IQQILUEOI- 00°~Hl hi UQNH he |~=I~I=09 U~sfia-ufi .‘ 0~A~I.~.

.592 €255 8%: =5 .338 a WEE as EB

.+£z Us 592 E 3325 a guage £936 883. 222 2:. a

‘2 Son 3%: $03,638 amassi 83H. a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we o2 M: o 2. no N... o _

m2 2 8d «:3 8-0 m8 3. 33H .m.
E E ”85 83. w: E N .a 388.5%“
5 mm 86 So 3 E 3 Eggs .m.
SN 2 mod ES 2 .1. 3m 33:3 g
SN 82 28.9 $3. 2 3. E 93%;» .2
:~ Sm 2.3 23 3o 3 on §S§Eb .m
as 2 ~35 35 2 Ed 3 a8 .m

g 632%.
52 fix €28“ Macaw E92 592 +£z sesaaoxo.~ assuage :5.an
925 SM ‘

 

28

8.23.52. .23 8.22805 E9: 35...:— uouauouobumsou 82:33» .8.— 3528 3.25— .m ~35.

 

 

 

 

I cocoa». ”ammo—3? 2 82m“? 8 29250 Bu mow—5.3x a
9: $2 $3 .2 8 . Us
:2 Wu» onmw do ca 3:33.: &
hfi odmu 82 .2 no Euuubguua H
5 3.2 $3 .2 no gummiassus s
SN adv 8% .3. co 32QSE§°§2 .m
m2 on com o nu.~u :8 .m

Egg; 3. 3 Aug 3 8
.oz mom EvSaoM “SM §E> 081x .8“ EM gig. Ewing

 

 

 

 

 

 

 

cue—£35.05 EB 3.2182: E9..— uoEEn 833E8— 82? fl 2:. .5.— 353289 2.05% .v 933—.

 

 

r , on
the I‘m VaIUC. C

for the temperatur:

Symbols A1, A4,;

activation energies

 

respectively. “1111
Mined m'thout a
With Ifirnper'cmme. 1
“311333111128 for be
dchydmgcnase (Ta
“31033 r0 mesophi
cmPOlattion of an
IElationship yields
Sltbsmc affinities
Sim"Jar substrate af
”1}
my 0r cmDIOy a

33m dcscfibed bv
WYUC late enhan

29

k1 k2
E+S<—>ES—>E+P (1)
511

the Km value, composed of 3 fundamental rate constants k], k], and k2, yields equation 2

for the temperature dependence of Km.

Km = (A-1A2/A12)e[(2Eal - Ea-1 - Bad/R11 (2)

Symbols A1, A4, A2, Eal, Ea.1, and Bag represent the preexponential factors (A) and
activation energies (Ea) for the reactions corresponding to rate constants k1, k_1, and k2,
respectively. While the specific effect of temperature on enzyme Km values cannot be
predicted without a great deal of mechanistic information, these values are expected to vary
with temperature. Reported Km values were indeed significantly higher at higher
temperatures for both T. maritima xylose isomerase [190] and P. furiosis glutamate
dehydrogenase (Table 5). If thermophilic enzymes are exact structural and functional
analogs to mesophilic enzymes excepting their greater resistance to denaturation, the
extrapolation of an enzyme's Km values to different temperatures using the Arrhenius
relationship yields results inconsistent with the assumption of generally similar enzyme
substrate affinities at mesophilic and thermophilic temperatures. The observations of
similar substrate affinities for thermophilic enzymes at high temperature and mesophilic
enzymes at low temperatures therefore, argue that thermophilic enzymes must expend more
energy or employ a different mechanistic strategy to so efficiently bind substrate. The circe
effect described by Jencks contends that excess energy from substrate binding is used for
catalytic rate enhancement [216]. Increased partitioning of this total energy to maintain
substrate binding affinity versus catalytic rate enhancement in thermophilic enzymes or their
structural limitations reducing the total efficiency of substrate energy utilization therefore,

may account for the activity and affinity differences between Arrhenius theory prediction

 

and observation.

    
  
   
  
   
  
   
  
 
 
  

increasing therrnor

stability and redue

Protein folding

Thennoph;
mmnes fold prep:
bl Similar mechan;
Wm: stability a;
“lime Chamctcris'
and involvcs the m;
We is believed .
mm is Stabiliu
lie-tum Parts of the
Staggering number c
rapidly in“) ”lei! acz;
Stiles 0f wen‘dcfine.

30
and observation. Reports of enzyme mutants that have consistently lower kw values with

increasing thermostability supports a direct structure-function link between folded protein
stability and reduced efficiency of substrate energy utilization [217,218].

Protein folding

Thermophilic and mesophilic proteins are typically similar and thermophilic
enzymes fold properly at mesophilic temperatmes arguing that both classes of proteins fold
by similar mechanisms. The link between protein folded structure, enzyme activity, and
enzyme stability argues that understanding protein folding may yield insights into these
enzyme characteristics. Protein folding begins as the peptide is synthesized at the ribosome
and involves the rapid condensation of particular regions, or nuclei, into native-like states.
Folding is believed to be driven primarily by hydmphobic forces, and the native protein
structure is stabilized by a variety of hydmphobic, covalent, and coulombic interactions
between parts of the protein and between the protein and the solvent [219]. Despite the
staggering number of theoretical structures they may occupy, the fact that proteins fold
rapidly into their active conformation indicates that they reach their final structure through a
series of well-defined intermediate states. Rapid, cooperative protein folding suggests that
partially folded intermediates are not stable and that the native state includes very few
structural conformers of similar energy. This conclusion is supported by the conserved
structures seen in x-ray crystallography and NMR Extending this conclusion to protein
folding energetics suggests that the free-energy well containing the protein native state is
swep-walled and deep, comprising most of the free energy range that describes the actual
folding pathway.

Freire et al. [220] proposed that at least two factors are specifically related to the
unfavorable energetics of exposing complementary surfaces to the solvent in the partially
folded/unfolded states: (i) the driving energy in protein folding and (ii) the protein

stabilizing forces. The authors define complementary surfaces as surfaces which are not

 

I
solventexposed i

unfolded states. 7|
partly foided’un f0j

aqueous solution cl

states. Their mod;

SOIVCHL and that t"
PTOtein characteris
Mildlm protein ii
r’3gvdless of their I
”ho“ also nored :
Wm of m: SOCor
aIldhave a highly J

eXistencc of mes: SI

 

 

3 l
solvent-exposed in the native state, but which become solvent-exposed in the partly

unfolded states. These result from regions of the folded core remaining condensed in
partly folded/unfolded proteins that expose surfaces uniquely present in these intermediate
states. Their model predicts that protein folding intermediates are highly unstable in
aqueous solution due to the exposure of hydrophobic complementary surfaces to the
solvent, and that these states are poorly populated. This CORE model relies on universal
protein characteristics, not on structural motifs (at—helices and B—sheets) specific to
individual protein folds, and is consistent with the observation that most proteins,
regardless of their secondary structural composition, fold in a two—state process. These
authors also noted reports that protein condensed phases which retain a “...significant
percent of the secondary structure content of the native state, exhibit considerable flexibility
and have a highly disrupted tertiary structure” are stable under certain conditions. The
existence of these stable condensed cores indicates that secondary structural element
interactions cannot, by themselves, control the protein folding pathway and folded protein
stability. Haynie and Freire proposed a thermodynamic model to determine the conditions
maximizing the stability of folded protein intermediate states [221,222]. Their model
predicts that, in the presence of a denaturant, intermediate-state stability is independent of
the free-energy variation between the intermediate and the native state, and that, in the
absence of denaturant, the intermediate state AH is derived from the thermodynamic
parameters of the unfolded state alone. This model implies that, thermodynamically,
protein folding is defacto a one-way process. The successive steps depend only on their
thermodynamic properties relative to the previous state and not to the subsequent one.
Furthermore, contrary to the current belief, they predict that the two-state protein folding
process relies on a small entropy contribution to the intermediate state stability.
Examining protein folding as a function of the total energies of the folded and
unfolded states further complicates this scenario [223]. At 25°C and high dilution it has
been established that non-polar solute transfer into water was opposed mainly by entropy

 

and not enthalp}
driven. However
transfer to water
tanperanne-inde;
highest between t

 

transfer is propogt
loosely appmxirn
Ma DOD-polar rl
t“literature. Abo
significam There
“My and enthaj

 

apmemamly ch
anta] ”511]“
0fpamnomng mu.

“mm in mCSO'

32
and not enthalpy. Under these conditions, protein folding can be considered to be entropy

driven. However, because of the large, positive heat-capacity change of non-polar-solute
transfer to water [223,224], the AH and A8 for this process cannot be considered
temperatme-independent. The AG of non-polar-solute transfer into water is predicted to be
highest between 130°C and 160°C, where TAS~O [223,224]. At these temperatures the
transfer is proposed to be completely enthalpy driven. Considering that the cell cytoplasm
loosely approximates a dilute water solution, and that a protein loosely approximates a
small, non-polar molecule, protein folding can be considered entropy driven only at room
temperanne. Above this temperature, both entropy and enthalpy contributions become
significant. Therefore, at temperatmes where many proteins are folded (37-100°C),
entropy and enthalpy contributions to the folding free-energy are expected to be
approximately equal [224,225]. These predicted energy contributions are consistent with
experimental results [225], and may help explain why, despite the temperature dependence
of partitioning entropy and enthalpy, thermophilic proteins are correctly folded when
expressed in mesophilic hosts. Protein folding is therefore, a robust process resistant to
entropy and enthalpy variations. Unfortunately, folding intermediates do not necessarily
contain local, condensed conformations identical to their counterparts in the native enzyme,
sothefreeenergiesofstrucnualintermediatescannotpredictthepardalfieeenergyofthe
corresponding region in the native protein [226].

Protein unfolding

Protein unfolding is fundamental to protein stability with the Gibbs energy of
unfolding a direct measure of the folded protein stabilizing energy. Protein destabilization
by thermal energy and chemical denaturing agents has been well documented [227].
Extensive irreversible denaturation (i.e., loss of active architecture that is not recovered by
the removal of the denaturing force) is far more common than extensive reversible

denaturation (where active structure is regained upon removal of the denatming force).

 

Since the system :
reversible unfold:

irreversible protti.’

 

image of prorein ft
folding intermedia
in protein thermal
folding. For syster'
pathway man be t*
un‘olding must era

Mess at equilibnt

 

processes does n0t .
completely unfoldet
for both folding and
Molding experimc:
folded enzyme stabi

33
Since the system is usually chemically or thermally altered to initiate protein unfolding,

reversible unfolding is not necessarily a mirror image of protein folding. The existence of
irreversible protein denaturation also indicates that irreversible unfolding is not a mirror
image of protein folding. Thus, unfolding intermediates do not necessarily represent
folding intermediates. For example, the molecular aggregation and precipitation common
in protein thermal denaturation did not occur for these protein molecules during proper
folding. For systems at equillibrium, the flux through each specific fundamental reaction
pathway must be the same in both directions. Thus, the mechanism of reversible protein
unfolding must exactly mirror the folding mechanism under the same conditions. Not a
process at equilibrium however, this microreversibility principle describing equilibrium
processes does not apply to irreversible protein denaturation. The AG between the
completely unfolded and native states in the same environment, however, must be identical
for both folding and unfolding. Thus, studies of protein folding energetics using data from
unfolding experiments can yield direct insights into the molecular determinants controlling
folded enzyme stability. But, extrapolations from these data to the protein folding
mechanisms may be misleading due to the potential for folding versus unfolding
asymmetry.

Protein denaturation is directly related to thermophilicity and thermostability.
Enzyme activity usually increases with temperatme until it falls precipitously above the
temperature of maximal activity. This rapid loss of activity is consistent with the loss of an
enzyme’s active structure by denaturation. Many enzymes, however, have long half-lives
at temperatures above their highest active temperatme. For these enzymes, the recovered
activity upon cooling into their active temperatme range has been attribuwd to reversible,
incomplete unfolding [228,229]. Denaturation is usually rapid and cooperative but the
observation that denatured proteins in solution retain some condensed structure indicates
that denatured protein is not necessarily complete unfolding. Mechanistically, denaturation

is the loss of native 3-dimensional structme sustained by solvent molecules invading the

 

unfoidint protein.

protein solvent in

Molecular met
Intrinsic Fact0

The reteu‘.
mesophilic hosts
pepu'de. These ir.
entropy of unfold
Others. The char.
fumtionally Sinai
COutpatrisons to cl
diminution NC
mlymes and me:

hllth'bd to be dll

34
unfolding protein. This hydration reaction (in aqueous solutions) must originate at the

protein solvent interface, its surface.

Molecular mechanisms of Thermostability
Intrinsic Factors

The retention of thermophilic properties by enzymes recombinantly expressed in
mesophilic hosts argues that these pr0tein characteristics result in part from the nature of the
peptide. These intrinsic factors include sequence specific amino acid replacement, altered
entropy of unfolding, hydrophobic core packing, and loop region architecture, among
others. The characterization of thermophilic and thermostable enzymes, structurally and
functionally similar to well characterized meSOphilic enzymes, allows the use of
comparisons to determine the factors involved in stabilizing enzyme architecture against
denaturation. N 0 universal mechanism explains the differences between thermophilic
enzymes and mesophilic enzymes. Thermostability and thermophilicity properties are
believed to be due to subtle changes throughout the amino acid sequences of thermophilic
enzymes. An extensive comparative amino acid analysis by Argos et 01. led to the
conclusion that thermal stability was related to (i) increased internal hydrophobic amino
acids and decreased external hydrophobic amino acids, (ii) the replacement of Gly, Ser,
Ser, Lys, and Asp by Ala, Ala, Thr, Arg, and Glu, respectively, and (iii) to helix
stabilization by more exclusive use of amino acids commonly found in helices [230].
Not all amino acid substitutions alter the function or stability of a protein. Specific
interactions and residues, rather than all‘amino acids, contribute significantly to protein
structural stability [231]. While single substitutions can increase the stability of an enzyme
over 10°C [232], the thermostability intrinsic to thermophilic enzymes most often results
from multiple amino acid substitutions [233.235]. The effects of single and multiple amino

acid substitutions on enzyme thermostability are shown in Table 6. The examples were

 

 

 

 

 

 

  

 

 

 

d. . . l I <‘.\ link-fie
v O
- . ,. e A 3.5% Easesguegsgi guests.
:CEWKEEA... «your. E .25: 5.: 663;. <
.5932. 35 5:45.. 232:3: “your. .e dear. EE MEG-m >
53.00 35:: 55.5.9037. U.£C—.QC._€.A£ Don.“ .fi 55¢. 55 VMQM ~
05:0: 25:: OED-9' 30; Uohn 2. 53¢ 5.: in 4 NBMNN
DON.“ nun .DSI cm:- NVASM U~h¢h~ gwfiggugt-d
nl na=vm279cou — ugnvhl 3% N van-EKU

 

lEI-p-II-BE

 

Hu-nnaioRSELU-uu United-u... kc Iain-93‘!

I‘lmv-Ibhw-s as Bubbhhhv

.IUI . be! §i ‘I Ii“..hnil§..' BI

 

t‘ .\ a .r.

35

 

 

 

 

 

 

 

 

 

 

 

 

«.2 3: ca
.368 098% e832 8:82: 3. «madam—MB»?
B ee§8§5 8:20 menses a E a.» 8»
«a mg 88
«8:82: Wm OH 23% 83—823
035m .2. ban—88:55 .5 some 5- «nus—82M 33838: cox—.55—
eswp.» a 96: :8 means—E. vie—=2 C as- 6.: «mm..— ade 0:29:55
Ge .5... 0e ages in 956 £288:
moo imam toga:
2.55 3» 3% >89: 38
BaaeEosaaasm $83.2 we «a E... eases:
853852855 «gnaw a 3m 3m 5.: 3338: fig
confine—55 cons—8 €35 0:33:55
85.35 05 gas» 53352255 2 Gov «F v.3 03.2» flan—082
we- a
3.. smegmgeafiizézafie
2:- Gangxiezaze
2:- Gmmagxxxgéi
ca. Gnomeuxxxauéx: :8
Cd K: Egan gap—vans 85—A—
. . s ., _. _ 8.....«828383
35:88:55 88265 3:: mg
geese—5.553295 2 389 53333»:
«See». co 5328623365 2.. "6L 8.2. m8? 23 excavate
333% 35333.»:
g Passage 39m <
asgefiusea 3832 came?» as 89m >
.33 as: 238% usage caneéase 3.9m _
use»: an»: 059 33 vegan... as 89m A 6E
09% a .58» as 99m 08m genie
28.8380 .8 £2: A 582:

 

 

 

 

«En—:28... 3.358535 2539 me 8:53 3839...:— 382... a...“ 3:85.689: 6 «3:.

 

 

I I (PM.
x: C. m .c E

Cd Cd E

---- «.3» ME.

~06 0; EV.

£2.33: 006 cd 5:.
800 Baku-D ho EOE—5133‘: 0>38t9¢2C0 “N. — Mia LHNN. \QVMM\
ha huauun—Ofin—OnuA—a UCG Ont-:0) CmBSU 0‘3n A L3:- _IDu-v Anvov 3 3| .. Ofléb MAI: ~
UAR- EG—Uh haugauv Ia \n-u-aCQ-‘u-h-EOE A - .3c<<. -=.—.< :a h=la 0:: :0 .5 is V

 

 

 

 

36

 

 

 

 

 

1mg: milk—23:3

3 ms§ea§u

3 ms:—

2 banana

3 5%

2. Ann.

o.» mxgzmini
.oéeaos 9m 22 5a
538§853§§u§5 2 =32 o§_
538553 38% 2a.: 3 "Gov 5.2 .5: 23:3... .935

841 3- fl:

8.? 9:- «5.

8a. 3- E

Fm- 2n .3.

own- 3 9:.

SN- 3. an.

Ra- q». 96.

8a- an- a:

$4- .3. <>m

m2- 2- m>m

m2- 2. >2.

“2. 2. <2.

8.? «a- at.

3.9 8- . Sm

an? 2- >6

3.... 2- a

one. «.9 E

8.... 8. ma

2.9 v6. 5..

8.9 2. 2...

ed 2. E

E. E. as an.

:3 2 .5

$828. 85 3 E.
28 BEE mo «Susana. 9358 mm; fin HF Sea
5 afiaofiogusssgfififiu 9-3... .85 GL . a E.
93:2 xufiaaaafisg: .03. a? 82 223.3036

 

 

 

 

2 ~..N.~.-..--N\
:KCs \ 0~=~HNAV§€ :=:=-~
_ In: 1-1-11:14 111-1.--- 1...:. :4 J LC<<I firs ~

— TAMMNV
< . - . . - .Cfiu‘ tufinhhu-§\
V?§h\l’-V \ Chunks.” I
awn-CA,- .v-...——:Z:u . v FC. N
.3:- .9:- sAA- uvb vuuovt I. AI-rgnzud. .V V-n .I AL... :n. G
.- huv 7. - ,. - . .

    

 

37

 

 

 

 

 

 

 

 

 

 

 

 

5... $2384
we. <3".
3. <93
2- $2.—
ana «.2- 5:3
3.558.. 9:9... ".585 326 he- <8.— SE
3- "Cove-2 $3 858: E.
of 2:- <
31 2:- >
bass“ 31 E.- c
835s ”is; 28 38.3.3 a 1n.- 3- a
2v 3- m
«5&8 £2 flags” 3- S- 23.
253825 aaEEaEB-m .88: 3mg 3 m.— 5a
EEq oESRE.
monta- <m2 a 5 _ RE
«.3 we- <8§222§>2§m
n33..- <Nm_z\<_m_>\§_m\<§n
«2:33 fiazzfiga
a... a 3 3222:;
2,52 2a 2.3 2 Sega—m
535398..» 832.. 2a a mud-.35 $2385
33.:- 5.24
«23 Susanaaéfi-a : 8.33 52> SE
n83... u 6.» e2 52m 2.38::-
§ 2 «.95
2- :- $8
2 8 .54.
. 29 9o 3. 083
26?: saga-852° 5%? 3 S- S. 92
8262 @526 3%.2 2&8 a». :- 3. 0886:..—
co 8%? Egabafim foe .85 60V . .3.me
I i .03. :2. $35:
‘ 9308.36 filmm-
So 30538353§3 v.3- "Gus-2» <§E $.85:

 

 

13.33: {c 1-3.2: (323'... I h.- o!.«.«-.-. 1 r?
mad mam-FR
A..._....:r. Uriah-b... p. — .C L.~.3< \V.7..H\
r=uuhul h: rut-7:95 uvaul I... 1.3.: .2-Xu. _ A - VF. - I A. ch :n-huf‘ VéflL- 5.-~A.H.H¢kl\- s35

 

 

38

 

 

 

.52: 2.

L‘

ad- NS-

One

 

 

 

 

 

 

 

 

 

 

 

usage E. fizz 2963253 3. S. mac—o
3- 2- <85
€3,853. 3 ma 3 ma RE
3.: mica 28 8.22: €52 - cine: E.
SE 2 M3 4
23.2 033a 28.3.. 8%? 3 3 I- <80
8262 “.536 828.. 2&5: a». 3 ma 3 ma 8E
2o 858 330553 a. 553m Gov 5.2 858». a.
w-MNW J32 a;
3- 29E
3 Anna
3. 5E
3:3qu «Fa—ca 88 33:05 AN 31 >mnE
2- {WE
28 as...“ 8. ma;
22.. bfififieuécaaaos : 2- :33
2- 53
"Snagsa £53 2. >93 5a
1:- u 6.9 5% <84 25982 a. .
Lnflw- vam an .23.... 9.2 2 a _
mm. o.: an €55§5$E<899§~§
2:- hfi an €53 0%;ng
Z- c.: a u «33 Una-ERUPEQBE
05% no. to c u 533 033585802
29.. 53950508... 88.. 22.. 2a 2- 3- : u sag-323
:53 592.. 23235382.. 2. 3. an .EERQHRU
8255383 Ho 853 be th- 65
6b 5.2 8585."
g
238 9 as»: 0395 a .o 8u§< a
8... 2n:-
aafimoaoaa «3 g 53
525.. 9.8.2 as assuage an. n O.» Eh m8... 258: E.

 

 

 

 

 

ICIIII‘EI

 

ab>::..x=x..: 7.2. .2532: ”xv-Wax- \NNQNEHNQQQ
12.71332. 53 vanity—.49 tow—:5 by: :3:- NmuK-u- - ->
Rao- 1.. RCC_-~..:= 97:3:25 hc «DEF—Wu A- Nth-V: “N n ~> NWM..~!\
v.r-.N- I «.9ov .2.F< 2~NC~2 Oat-«WIFSV .7.»-

 

 

 

 

39

 

 

 

 

 

 

 

 

 

”6. 055.3 + «AM 4 84
3444448” 53844 3 45:24:44 «4 «£44844 «6. <mn4n<<m3
28 8 048042: 38.44244»: 4495 4N4- 054544 + <mn4~4
a .40 94444.43 3 448444448» on :8 8:82: o.~4- <mn4n4
9.44443884. .4588 .4488 49545 44 m4.- AZEOD 888.4 + <84 find
24- u 604 5.2 <84 ciga-
=o4§=E E245.— 9 43.42 50% Nu mi 4.44.4 4m .
45448442 28 4804 .3 804435544 tn 44.4 QM 4,444 Emu
356.5» 38.4888 :8 3884 Gov 5.2 24.58.44 v.4.
8.2- 49>
n?- 5>
ohm- 48>
3.0. m3”:
84- m8?
$8. m8<
£35” 5224 «2. 82
Evanssgsusasifia 8d- 842
30305844? 8 54434433 488445 R. 4 - 949‘
4:348 44 :54. mfinaagu 29: m4 mm- 4- ma¢<
28 44432.4 05 :4 44:on .8424 a «5.95 3.4.- mm}. 44nd
:4- u 004 6.2 m4v< 258». E-
3 SN:
swan-8524353.. +46 4.83 00: 38.2 8.388 05488. E
L444. 484 4: 4 .484
mu. 4- Ann 4E4 4 4 44.5484
wed. 4mn4§4z>4§
8.4V. 4444;42334
«Wm. 182%: 4 4>
54.44.80.408 05 8.4 5588 $6- 44 4 338442
4:44.844 28 482 58.5488 3 8.44 4mm 4En434
Ed. 444 4>\m84
3.4442388 98445::— wmd. 48435434
33424-6 38440 4.8ng 84. 45>
804 m4 Sousa:— 344442: .40 484444 44 2.41 ~44 44> menu
an- n Gov EH< .4842 2.3895 v.4.

 

 

 

 

 

. .. - i . - a filial. n. t. the“. NXNN
0:.EANCU by... )2: 4.4-2.4.4.7 4.3.9 A::»:&...-: c.3- - nhP‘CL. «8.?sz bvdwpuztfih ~§u7v=
nurvAv- rib-IE A: Av‘ut— h: unavm-m-u-d< AV.’ fl ALUCV hvf,ph.<€ ghfl~c< b!\\n\\\ngkl\(\thk:lv\b. h‘

 

 

.4847:

 

 

 

 

 

 

 

 

 

39. 4.“ a:«anggmflmfiugmoamaafizae~34

3388 E4525? 2: an n42«8.448444453434848844834?444E434

3834835.. 98440253 3: 9m 44344438442343.4834“:8.44244

an. 45 mahamoamamommkoagfisaafiu

.844» 4852:0255 .8442. my. 4.4 mcbamwonmaaamfiaaafi

.5 4.. £32 4.5 25.4. =4 8.4m 2a a 2.. in 4§4§m§=m§~§

34.23 3 mSNmEEmENE
£25852. 885.4 5%: 3 44.4 fizmaafi was
to 0852. 88549588414 .44.. «.4 .4484 084.4482»
4.4445 N234 Gov 8.4.44 6.46340 .3488 .m

434- 4W8

2. on:

2- £8.

3. 448.4.

3. one.

.3 >34.

n4 2%..

9m .59

4.4. 484.
50850584435” 8&5” 5224 S "484. 68
24.39828 05.352883: 3 4.48.4 43.2%.“..8:
4.» u 604 8.2 4494. afigsafi .m
becamfipspa 4mm...
828%?2244 8825 4§¢zvoaseq§§
232.5%... .23 84.38.. 9.3434 «4 "6.4 8.2 m8? afiggud

0% an .4344

35355.3” 3:85.528 34. .4844
84.384.554.834 4.4443» S. .48» 44.3
aims 9a 4.434425554454544 2:. .484. 4%.4240899533:
§.§§spa4o§§< an ugcnaqa m$< ”sagas

 

1 lil I- I‘lfil‘

NI)» - .4 IVAEQAI ~ IS; 53.0 K 31‘

A...-.a-.X:2: 1.7.:- P —
file :5 .uov I.....p.-hrv--rv trU\.-.a-:-r — T: V A— N. n

 

 

I Ace; EFFQ \ NI}; :5» :23». TIAofiytav» -m $7.225- \\.u...» LN

 

41

 

an :8 5.2-

 

 

 

 

 

 

 

sgaéuaafifi 445 8.4 445 3.- $5
4.452.444.4425? 444.4485 4.384444 2 9 4.~ 43>
2 9 4.“ .4445
4442844 84.48 2.844444% 44.24
24547. 544544444 944444.. 54460 44 2 444.2 4: 8.42M .448 .m
4.64 «.6 24395444
4&4 «4 4.434442
34 44.4 22434
44.44 S 9434
2 3 9454
3 .44. S454
48¢» 4498445 a6 QN- 25.”—
2:22.38 42. 242844 044444.482: 4.44. 4.4- 34544444
9304444444855 344544 444.484.0544 4.4 «.44 $44444
448343 «4485444480 844044448 a6. WV. 24%
84.82.558.442 84442 444 2422.4 44.4 4.4- 4444444444452 TEomoSuaEm
044444448244 444 84.344443844440444. 44. 4d- ed. A4444 40 u «44.4
44.44 «.4 BaoToBusno
”8852444 44.43435“ 44.2- 344. 554.3446 T350469 "$4
a; 2242 82.448443444424444 44.44 to 4.323444
4% E- EngomoguE
445N425” an 3 04444 T 44945 u 84
933444 844 «4 44844444444409 344044448 n6 4mm mug..— T 0.4.4095 u 2
8884485 84434433243 4.4. «.4 4844 "34
u<m T E u 94
Erin—088 Nd- 0N. g T. MOM» u NM
3 5 4+4 3944 s 452248.44 S- 3. 44.444.44.42 T .42 n 444
on .4544 «48.440 945944442 44 3.4444 4464444 834
Gov 444.4.< 4: 8442M .443 .444
§
4.2 4.4.84.4 4648.4 4.484 as: 48 44
344444984. 9 44.44484 482 8.4a 4.5
4.4244 2.3 82 £243 5 :4 54 4a 2- 444.8
2 43334868144358
4.44483 .4442 2. 83.444434401443448 454
8 54.. 8.45.0 38444.48» 48 44 «4 u 604 5.4.44 83.48.80I44368 4: 8.454 448 .m

 

 

    
 

4 . 1. 1-..--.1111 1..--.. a: .-.--.ll-:..l..3 v v :C C r :C \«F.N.~»C-
—vl ~--. AVV-yv E.-.< hhi rU..v.-\2~§ .\\~\-\ 3..“

 

42

 

 

4444444434 44499444 244488454 0444-48244

4.4:. on "944%

44%-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4485445 44.43 84443444244 84444.34 4.4. uGov 52 04494404440 444444444443
48443444444444 8 mm- m- thmgm
8434444984. 448543.488 4.4.3-44 8- 4m. 0538
04.449448455344344. 844444444. 245 4:4 444. can
.44.: 44 gm
£5 44443 0444 Q .44.: na- UNE. fig.
94342 5224405844445... 444448444544 4.4-9+ 4.5- .2.444 44444445444
42. 4.44. 4444448444445 8.44344. 44 van-4422.445 Sa 3333444343 .44
Jag-.8555 .8 6442.44 n4 4- m4 4- Bag-0.4.3
44. 838.4 249244 45448.5 44444442244 84- 4:. 0326444.;
844444444...- 424 .4424. 4444443 0444-4444449 3 444- m 848488
$- 4? 84420.42
4544044442 9 84444444442 w v Unag :08
.4405 9 339:8 4c: 8044 444448844 4.4-n7 4.4-Q- 2.3 4434444444»
84.48444 5448444449384. 0444444444444 00403444454443 348-4344434443 .44
.4 42444449845542.5445 Zn? 5. .9. 44%???»
24. 44443.43 4.5824444344488444 2. S. 44%?
44.4- o. 4- E4.» 4448.
84444444444 44.43-44.4- 4454; 4.4. 8.4244
guage 9.44844 2848244224414 Gave-4h 434934443444
4].. $5
4.4 v.4 4.45
243244 2 hm .425
84448844445484.4448 445.44.84.84 2 4444 44.4 4444 45.
Gov 444.2 44.4 8.42% .448 .m
4.539: 8444.44.84 Tmm awT 4,454
=25 440 44.44. 244.95.44.48 84.54 44.44 4.4 984
4404 444 440444.484 444444.44 .444 44404448444440“ ad 4444 9m 4.444 Emu-8:
Gov 444.: 44-4 844244 .448 .M

 

 

- - - xanA-qx
.Iti-‘I-I ’Iulll uni-II:~l-l.‘ .IFI‘j-IF‘FIF-IlI-m F‘N. - x A.” “pkmg 94:0.5‘A5.‘
.B) 01-14—514 :5 lunavmu-J-z: LC Vat-Pawn A ~ APVDCN-v San-nun: AMUCVQflV new! . gazsfivn ENE-(K‘s): P-Efiufi
A .hv Q— cc hr-{Mh-usnsfitifi -\

 

43

04.543449444444884.
04.944444944444984.

B§E§§§§448€§44§34o4£§§§5§u3§

4.4444384438555894434534445834ogp3395n8544494456244.4424:
ago-44444 444 .6405 84.4 444 094540 n 44045
2444448444442 4844844444544 444 0949444 244 u 44.4.3

5844.5. 54oaaa§4u§u5<o

4.345233884492242344332305838445
atgfigsuganggofifiowqfiofluuapzo
ndu4§93284844§4884£4544438§a§44§oe94449444044444.4444

45443449444444.4444gé§u8flac§n§30§§§£om§flaug§<a

tam—4:88 so: u 6.44

 

 

 

 

 

 

444- 44844
n4 $4444? Gas
444444458585 4422488844 a 2843 89.438344
84.48.4443. 244888 42.3 4484.344 «4 4444443 §u44s§85§4
444 "44.454 S444 44.4443 5448448539444»:
4.44 an 44%
44.44 44 4444404
44484-44 444344.454 4444.4 8.6484 442 44 44.4 444 ‘ 4444404
44.4 8 44804
84484444 44444444438 44.4 84 44434 4844
455.44 48443443....- 44-4 s4 42 44 40.344.844- 6-8424? 8.54844
4444 $444 844444 4.4244835 .<
4.4 4.4 4.2-$42438
444- 44.44 5.8444448
8444944443844 44. 4.44- an?
444844444 444 ~444442444444 444.4344 «4 48§444é .4444 .24 0394
88.4444 44444-4444 84484644 44. .2 04454
4.4. «.4. 44384 4444-43 ,
344444 4.4585445448444855 44: 4.44 44454 3.2484
.2, 2.428883%... 444 48.4444 44 0.4685444... 40.2448 82444 4.424.443.4444
4.44 Q48 3.38.443-

 

 

 

 

 

 

 

. .‘
"w“.
3u¢m

REC

COT-’3

mum
cons:
mum

an (1.]

44
chosen from site-directed mutagenesis experiments in which the mutational effects were

thoroughly studied by biophysical structural analyses (e. g., crystallography, hydrogen
exchange measurements, Fourier transformed infra-red spectroscopy [FI'IR], calorimetry).
These examples are also representative of the known mutational effects. Typically, several
mechanisms are used to stabilize a single protein. Finally, because of the structural
similarities between mesophilic and thermophilic proteins, reviewing selected literature on
mesophilic protein thermostabilization may aid in understand the the general mechanisms
controlling thermophilic-protein folded stability.

Multiple substitutions and protein stabilization - A great deal of protein
thermostability research has focused on enhancing mesophilic enzyme thermostability
through mutagenesis. Because mutataional effects on protein stability are proposed to be
local structural phenomena, they can be independent (multiple amino acid substitutions
have the same effect as the sum of the individual substitutions) or cooperative (the multiple
mutant stability is either greater than or less than the sum of the single mutations) when
constructive or destructive interferences occurs between the spheres of individual
mutational effects [234]. When Matthews and collaborators introduced multiple alanines in
an a-helix of T4 lysozyme, the mutation effects were additive (Table 6). However, when
they constructed multiple mutations affecting a single internal cavity (where the side-chains
of the target residues interacted with each other), the effects of the multiple mutations were
cooperative and less predictable (Table 6). Further work on chicken egg-white lysozyme
has corroborated this range of mutational effects [240,241]. Therefore, from an
engineering standpoint, a continuum of thermostabilities should be attainable by protein
engineering, bounded only by the thermodynamic limits of the protein’s total molecular
energy. However, the dificulty in predicting the stabilizing effects of multiple
substitutions prevents the use of anrino acid thermostabilizing "traffic rules" to identify
specific amino acids responsible for one enzymes thermostability relative to another.

 

res
the
its
da

H

(is.

con
[hi-l

citfi

45
Substitutions and modification of the thermodynamics of unfolding -

Since entropy has traditionally been considered the main factor driving protein stability, it
has been proposed that reducing the entropy gained by protein unfolding stabilizes the
native structure [248]. Reducing the entropy gain upon protein unfolding by adding a
constraint not removed upon denaturation may be achieved by disulfide bond formation,
replacing Gly residues with any other amino acid, and replacing any amino acid with Pro
residues. While disulfide bond engineering has been shown to affect mesophilic enzyme
thermostability [242,246,261-263], the chemical instability of Cys residues at thermophilic
temperatures makes this strategy ineffective for high temperature applications [267], and it
does not appear as an evolutionary strategy for thermophilic enzymes. Moreover, enthalpy
appears to be a significant factor (often dominating entropy) in protein stabilization by
disulfide bonds even when these bonds remain intact in both the folded and unfolded
protein forms [242]. Lacking a B—carbon, glycine residues in solution have more
conformational flexibility and entropy than any other amino acid. Matthews et al. predicted
that replacement of glycine by any other amino acid containing a B—carbon would stabilize
either thermophilic or mesophilic protein folded structure by ~4 kJ mole‘l relative to the
unfolded state [248]. Unfortunately, results of their subsequent exhaustive mutagenic
study of T4 lysozyme were not completely consistent with the theory. While mutations of
poorly mobile amino acids (as determined by low average crystallographic B values) with
reduced solvent accessibility were far more likely to have a significant effect on protein
thermostability, stability mutants primarily affecting the fiee energy of the unfolded state
were rare [268]. Therefore, they concluded that folded protein local packing efficiency,
rather than an entropy difi'erence between the folded and unfolded states, was responsible
for the observed thermostabilization. Furthermore, the lack of significant differences
between the thermophilic and mesophilic Gly residue percent compositions (Table 3)
indicates that this strategy is also not employed to stabilize thermophilic proteins.
Matthews and colleagues, using similar arguments, predicted that the extra geometric

 

cor-scrim of pro.

mum. the fold;

 

configurations pr
been demonstrate
surroundin g the-at
of a protein stmc:

tcins[2 33,2.-

 

demonstrating th;
bt’nafum. These
Which can cnhan
:- PMto contain
COLIHICI'pans Am

{IT‘Opic COmPOner
[233,270] in): ex

Smalltrencrgw. gal
Seen.

azippcr, mounts &"
numerous coulonx

m This PMinez

 

46
constraint of proline residues present in both the folded and unfolded protein would

stabilize the folded protein by ~4 kJ mole'l [248]. The specific minimum—energy
configurations prolines can occupy have been carefully calculated [269]. It has further
been demonstrated that prolines play an important role in directing the local conformations
surrounding them [269], and that proline insertions are only well tolerated in specific parts
of a protein structure [269]. Proline substitutions have been shown to stabilize mesophilic
proteins [233,243,24825 8] and to further stabilize thermophilic proteins [254],
demonstrating that the practical limits of enzyme stability are higher than those engineered
by nature. These results suggest that careful proline substitution is a general technique
which can enhance enzyme thermostability, and, in some cases, thermophilic proteins do
appear to contain higher Pro residue percent compositions than their mesophilic
counterparts. Also, as seen in disulfide stabilization of folded proteins, both enthalpic and
entropic components were important to proline substitution thermostabilization effects
[233,270]. The extent of proline steric constraints relative to other amino acids predicts a
smaller energy gain from entropy due to proline introduction into a protein than is often
seen.

Prolines and loop regions - It has been proposed that analogous to the stop on
a zipper, prolines are used in constrained loop regions to prevent the sequential dissociation
of numerous coulombic stabilizing interactions between the two adjacent core elements
[71]. This proline-zipper model predicts a crucial role for loop regions in protein
thermostability (Fig. 2), contradicting the traditional thinking that, typically structurally
insignificant, loops act as flexible links maintaining a continuous peptide between two
properly positioned core elements. This traditional view is supported by the observations
that loop regions can withstand the accumulation of more neutral substitutions than can core
elements, and that there is usually a greater sequence variability in loop regions than in core
elements within an enzyme family [271]. However, the proline-zipper hypothesis
described here for thermostabilization by loops follows from the mechanistic understanding

47

Figure 2. Role of proline residues in protein structural stabilization.

 

 

*—

48

 

J\ * i it}
<0.25nm

 

 

 

   
  
  
   
    
   
  
  
  

that thc rapid cor
interface and the
cnrrging data St;
protein thermos;
cmsraincd loop
(11. [259] constru-
mtsophiliC/thcnr;
that targeted a [34 _
diChmism meaSuy
mOdtling, Kimur-
mated by the lys;
core elements, ’1‘;

htmgen bond.

49
that the rapid cooperative protein unfolding process originates at the protein-solvent

interface and that loops are usually found at the surface of protiens. Furthermore,
emerging data suggests that manipulating turn structures may be a common method for
protein thermostabilization in nature [259,272]. Studies focusing on B-tums suggest that
constrained loop regions play a major role in protein resistance to denaturation. Kimura et
al. [259] constructed E. coli RNaseH site-directed mutants, using the results of a previous
mesophilic/thermophilic enzyme sequence comparison [25 8].. Mutations K95G and K95N
that targeted a B—turn clearly stabilized the protein, as shown by activity assays and circular
dichroism measurements. Based on the enzyme x-ray structure and on molecular
modeling, Kimura et al. proposed that the K95G mutation eliminated some structural strain
created by the lysine residue, and allowed a better interaction between the two neighboring
core elements. The K95N mutation allowed the formation of an asparagine intra-residue
hydrogen bond. This bond created a structural constraint analogous to the constraint
introduced by prolines [259]. Additional prolines were observed in short loops and in
longer, constrained loops (those whos flexibility is limited by factors such as salt bridges
or multiple Pro residues) in two thermostable and one hyperthermostable xylose isomerase.
These prolines were absent in the mesophilic enzymes (see Table 3). These unpublished
data predict that loop stabilizing mutations are mOre effective when introduced into
structrnally constrained loops thus, the Pro residue positions in a peptide, as well as the
number, will control their effect on folded protein stability. Proline introduction in turn
regions appears to indicate a general importance for turn regions in protein stability and to
represent a naturally occm'ring thermostability control mechanism that can be applied to
protein engineering.

Salt bridges - Activities of ions immobilized on the same molecule are extremely
high, even compared to those in bulk solution at molar concentrations [273]. This fact
suggests that intramolecular salt bridges may be very stable, even at the surface of a protein

in a highly polar solvent environment. Comparing the thermophilic B. stearothermophilus

 

and mesophilic )'
extra salt bridges
tbcmnsnbihty [.l
phosphoglyceratrh
factors in these crh
Acn'noplanes mi; '
effects on S. solfal
connibutcd sic. '1"
number of surfacel
improtcm theme-I
Winona] Charge-2|
Ihcrmombihmg . I
in B, “moth-em,
Mabflity of the P I
irreversible damn
that “Om 0f the ad.
Hl'drogen
folded Prom“ SUUC
Slbsauenny, the n
mommy Studjcd
missed the i“11301
éfw httwch
WW” StrOnge
cPinanog’aphicany
Minn 3 ~ .
an Slop:fLS m M
Hfiondin y dependc
g “”11ch i

 

50
and mesophilic yeast phosphoglycerate kinase crystal structures led to the conclusion that

extra salt bridges present in loops in the thermophilic protein contribute to its
thermostability [274]. Crystal structure analysis of ribonuclease H [275] and 3-
phosphoglycerate kinase [276] allowed the identification of salt-bridges as important
factors in these enzymes’s thermostabilities. An intra-peptide salt bridge also stabilizes the
Actinoplanes missoufiensis xylose isomerase [265]. A careful study of chaotropic salt
effects on S. solfataricus carboxypeptidase similarly concluded that surface salt bridges
contributed significantly to the enzymes resistance to thermal denaturation [175]. A
number of surface residue to arginine substitutions also implicated electrostatic interactions
in protein thermostability [265]. It is important to note that the presence of suitable
additional charged residues in a peptide 1° structtne is not sufficient to idenu'fy this
thermostabilizing mechanism. Tomazic and Klibanov proposed that additional salt bridges
in B. stearothermophilus a-amylase reduced reversible unfolding, thus reducing the
probability of the partially unfolded enzyme forming scrambled structures and preventing
irreversible denaturation [138]. However, subsequent crystallographic exidence indicated
that none of the added Lys residues were involved in salt bridges [277].

Hydrogen bonds - The importance of hydrogen bonds (H-bonds) in maintaining
folded protein structure was first recognized by Mirski and Pauling in 1936 [201].
Subsequently, the role of these coulombic interactions in protein stability has been
thoroughly studied and reviewed [278]. A recent publication by Cleland and Kreevoy
addressed the importance of low-barrier H-bonds in catalysis [279]. Low-barrier H-bonds
are formd between residues carrying functional groups with similar pKa’s, and are
significantly stronger than ordinary double-well H-bonds. They can be detected
crystallographically by either their characteristic short bond distances (< 2.5 A) or by field
position shifts in NMR [280]. It has long been known that the pKa’s of ionizable groups
are strongly dependent on environmental factors; in proteins, the environment smrounding

H-bonding partners is often quite different from their environment in solution. These

 

   
  
  
  
  
  
   
   
    
   
   

ohscnations, cor.-
hcrrnophilic pro:-
thc'r mesophilic .
t‘mophilicin' n
hermaphdic pro:-
dfilcrtnccs in stn;
Stabilimtion (T832:-
lht geometric cor.
Substantiadng or ,
Smrtturts.

Hl’dI'Opht
believed to prom,
B- Wdovelox (I-a.
Wing ltsiducs U
bydpophollic inltra
Mommy, M
which Conmbutcd t
m mginmmd resi-

 

5 1
observations, combined with recent x-ray crystallography showing that the core of

thermophilic proteins contains fewer and smaller cavities (i.e., higher atomic density) than
their mesophilic counterparts [124], suggest that both enzyme thermostability and
thermophilicity may be enhanced by the formation of stronger H-bonds. Thus,
thermophilic proteins can substantially increase their stabilization energy with only small
differences in structure or amino acid content. The excess enthalpic component of protein
stabilization created by proline substitution in constrained loops may also be explained by
the geometric constraint strengthening the H—bonds between adjacent core elements.
Substantiating or refuting this hypothesis will require careful analysis of exact protein
structures.

Hydrophobic interactions and core packing - Hydrophobic interactions are
believed to provide the energy needed to fold proteins in aqueous solutions. Thermophilic
B. caldovelox (rt-amylase [85] and E84 GDH [135] contain more hydrophobic-interaction-
forming residues than their mesophilic counterparts. It has been proposed, therefore, that
hydrophobic interactions play a significant role in enzyme thermophilicity and .
thermostability. Matsumura et al. [281] created multiple substitutions of Ile3, a residue
which contributed to the major hydrophobic core in T4 lysozyme. The hydrophobicity of
the engineered residue (measured as the free energy of transfer from water to ethanol)
appeared clearly related to protein stability (measured as the difference between the free
energy of unfolding of the mutant and the wild-type protein). Water/ethanol transfer free
energies for the substituted residues could account for approximately 80% of the difference
in overall protein stability, indicating that hydrophobic interactions were the main
stabilizing factor at this position. Other amino acid substitutions increasing the
hydrophobic content of the protein core have increased protein thermostability
[247,25 1,253,255,282]. Furthermore, the stability cost of burying a polar hydroxyl group
in the core of T4 lysozyme was significant [251]. It is generally believed that hydrophobic

residues at the surface of proteins are unfavorable for protein stability. However, the

 

prom-din g hydro
the protein suria.
snhiity. The B. .
introducing bulk;
[356].

Extensive
Sillificant cat-ids
m Padang is St. I
substitutions [284
C01? ClllimnanI
based mutagenesis
thermosmbiiny by
dL‘PlflCing buried i

Satuence analysis

 

was amibunblc to

twitch higher P<
have also shown a 3
tfi‘iu'cncy in therm.
mmhmbm‘l' [28
m “We in 1
”links Since, in n

“mm chS[ bein

52
protruding hydrophobic surfaces of some surface residues create a hydrophobic pocket in

the protein surface, where the presence of a hydrophobic residue will favor protein
stability. The B. stearothennophilus neutral protease was significantly stabilized either by
introducing bulky hydrophobic residues or by Arg or Lys residues in such a surface pocket
[25 6].

Extensive x—ray crystallographic studies have shown (i) that folded proteins contain
significant cavities (some filled by water molecules, some not) [28 3], (ii) that protein local
core packing is smprisingly able to compensate for deformations due to amino acid
substitutions [284,285], and (iii) that aromatic-aromatic interactions in the hydrophobic
core environment may stabilize the folded structure [234,252]. Other crystallography-
based mutagenesis experiments have demonstrated the potential to increase protein
thermostability by filling cavities in the folded structure [236,240,247257286], and by
displacing buried H20 molecules with either amino acid sidechains [255] or benzene [25 3].
Sequence analysis of T. maritima GAPDH was used to argue that its high thermostability
was attributable to local packing interactions specifically involving the T. maritima
enzyme’s higher percentage of large hydrophobic groups [142]. Crystallographic studies
have also shown a link between increased core hychophobicity and better packing
efficiency in thermostable proteins, supporting the importance of protein hydrophobicity in
thermostability [287,288]. Some thermophilic enzymes are significantly stabilized by an
overall increase in hydrophobicity but this trait is not universal among thermostable
enzymes since, in many enzyme families, no significant differences in hydrophobic residue
content exist between thermostable and thermolabile proteins (Table 3). This observation
indicates that increased hydrophobicity is not the only stabilizing tool used in naturally
occurring proteins. While better core packing is often linked to increased hydrophobicity,
in some cases it can affect stability by more subtle means such as the reduction of bond
strain [285,289,290]. Not necessarily stable in solution, secondary structural elements are

still important in maintaining active enzyme architecture [224]. Based on the ability of high

 

 

densities of short

homing peptid
. . l
Stahihmtion in 0|

Denser internal p
sud-2m g force (I
l

Covalent
denamration by tel
orthe oxidation [ I
”Will lbctrnosta‘l
through calalflic II
indium)- lhrough [
aimgation. F0”

h"pcrmcm‘OPhilic -

hdieatgd that dean

 

53
densities of short peptides to form or—helices in crystals, it has been proposed that
increasing peptide concentration increases the stability of helices [224], and that helix
stabilization in core domains was implicated in protein thermostability [232,257,291].
Denser internal packing, therefore, may enhance protein thermostability either as a general
stabilizing force or as a factor altering the stability of secondary structures.

Covalent destruction and irreversible denaturation - Enzyme irreversible
denaturation by temperature-induced covalent modifications (i.e., Asn or Gln deamidation,
or the oxidation [134,204,267] of Cys to cysteic acid [284,293,294]) directly affects
protein thermostability. These covalent alterations possibly affect thermophilicity directly
through catalytic residue decomposition (e.g., Cys residue in ADHs or GAPDHs) or
indirectly through their role in irreversible denaturation due to protein degradation or
aggregation. For example, peptide cleavage at the C-terminal end of arginine residues in a
hyperthermophilic GAPDH limited its thermostability [151]. Klibanov and collaborators
indicated that deamidation was probably involved in the formation of scrambled structures
after significant protein unfolding occurred [295]. They also showed that deamidation was
the major mechanism controlling therrnoinactivation in Bacillus or-amylases [197] and hen
egg-white lysozyme [138]. Sicard er al. roported a similar conclusion for Streptomyces
violaceoniger and S. olivochromogenes glucose isomerase thermoinactivation [292]. The
rates of amino acid covalent degradation become significant only at elevated temperatures
(usually above 90°C), indicating that only under these conditions will such processes be a
significant cause of enzyme inactivation [296]. Tomazic and Klibanov showed that
Bacillus amyloliquefaciens and B. stearothennophilus or-amylases are inactivated by
unimolecular conformational scrambling, whereas the more thermostable Bacillus
licheniformis enzyme was inactivated by Asn/Gln deamidation [297].

An inverse relationship was seen between the percent of Asn plus Gln residues in D-xylose
isomerases and their respective temperature for maximal activity (ranging fiom 55°C to
95°C) [190]. The general trend of fewer Cys, Asn, and Gln residues in the more

 

hemphilic prt‘

 

induced decomp-
temperatures. In
isomerases [206)
50‘} unfolding) f I
The mesophilic E

rtearothenmphif.

and 85°C, respect

and 85°C, tespec:

 

 

54
thermophilic proteins provides indirect support for the hypothesis that the thermally

induced decomposition of these residues is undesirable for protein function at very high
temperatures. In a comparative study of mes0philic and thermophilic type II D—xylose
isomerases [206], we extrapolated the enzyme melting temperatures (i.e., temperature for
50% unfolding) from melting experiments performed in the presence of GuHCl (Table 7).
The mesophilic Ecoli, thermophilic T. thennosulfurigenes, and thermophilic B.
stearothermophihts xylose isomerases were 50% unfolded at temperatures (53°C, 80°C,
and 85°C, respectively) close to their respective maximal activity temperatures (55°C, 80°C
and 85°C, respectively) and their precipitation initiation temperatures (52°C, 80°C, and
85°C). The T. neapolitana enzyme's melting temperature (120°C), however, was 25°C
higher than its temperature for maximal activity (95°C) and for initiation of protein
precipitation (95 °C). Thus, the mechanism of T. neapolitana xylose isomerase inactivation
appears to differ from that of the other three xylose isomerases, and does not involve a high
percentage of initial unfolding. Further experimentation is required for verification, but we
suspect that T. neapolitana xylose isomerase inactivation occurs through deamidation (or
another covalent destruction mechanism) at temperatures where the enzyme is barely
unfolded [206]. These data argue that the engineering of protein hyperthermostability will

require either the removal or the modification of any thermolabile residues in the source

protein.

Extrinsic mechanisms

Extrinsic factors, those due to the influence of non-protein effectors on folded
protein stability (glycosylation, immobilization, stabilization by salts, metals, thermamines,
pressure effects, etc.), have also been implicated in enzyme thermostability and
thermophilicity. It is well known that in vitro enzyme activity is strongly affected by the
nature and concentration of effector molecules present in solution as well as by

environmental factors. Allosteric effectors, substrates, and ions (such as Ca2+ and PO43')

 

tltittltlttllt .‘lllalttvne ‘iCL-o‘ahha‘Uo- QUA\—>I HA8 Fights-Luzscc oh O‘RHQrF

 

 

‘ "“5“...-

.~ -m. "

 

55

E5 :08 Ba 58,—. a vogue 83 .058 098% a
wedge Eng 05 Bob pa chat 858038 538» gauge 83390 a

 

 

 

vogue .0: "Q2

8 as 8m 3 8 Egg .5
omega:

mm 3 92 mm we nagmezthEeua .m

8 8 a. 8 8 ”vignettes g
omega—85

mm mm e. V mm 5 a8 .m
amazemen—
mwe 00V 602 a is 9 mm Beacon—22

. a 89 03%: has a, e
egos—emu 083% Sago

 

53533me 3392.8.— 2: «e 2532.83

5E...» 3:53 2: 5t» fleece—Ema .322: 320.5% one—mu he eaten—ecu S 035—.

56
are specifically bound by proteins. It is not surprising, therefore, that numerous enzyme

thermostabilities are altered by such effectors [298]. Substrate molecules have long been
known to stabilize enzymes, presumably by interaction in specific binding sites [250,266],
and these effects are not reviewed here. Chemical cross-linking or immobilization,
glycosylation, inorganic salts, and high pressme also stabilize enzymes. This discussion
of these topics is confined to the current mechanistic theories and provides representative
examples supporting or opposing these theories.

Glycosylation - An increasing number of bacterial extracellular enzymes have
been shown to be glycosylated (for example, C. thermocellum cellulosome components
and T. saccharolyn’cum endoxylanase [298a-300]). Most of these glycosylated enzymes
retain their catalytic and stability properties when expressed in mesophilic hosts (not known
to extensively glycosylate recombinant proteins). Also, a major source of thermal
destabilization for the A. missouriensis glucose isomerase has been shown to be the
glycosylation of a Lys residue by glucose - a glycoside as well as the glucose isomerase
substrate [264]. While some glycosylated proteins have been shown to be more stable than
their non-glycosylawd forms [199,301], glycosylation is not a thermostabilization method
commonly found in nature.

salts - Inorganic salts stabilize proteins by either a specific salt effect, where a
metal ion interacts with the protein in a conformational manner or a general salt effect,
which mainly affects water activity. Ca2+ binding by ct-Amylases is an example of the
former mechanism. The tat-amylase catalytic site is located in a cleft between two domains
(an [oz/[313 barrel and a large loop). Coordinated by ligands belonging to these two
domains, Ca2+ is essential for the a—amylase's catalytic activity and thermostability [302].
Xylose isomerases bind two metal ions (either C02+, Mg2+, or W”). One cation is
directly involved in catalysis (the catalytic metal) and the second stabilizes the folded
protein (the structural metal) [303,304]. The two metal-binding sites can have different
specificities, and replacing one cation with another often significantly alters enzyme

 

acuity, subsm
ijsoz'tme was ]
protein. relativt
geeifrc salt eff
size, coordinati

A Study
K3PO4, N33PC
reduce the enzy
(1837:35ng cffa
mutation is t
0““ Wipitatt
Convenient Way
mosiabinry
Whilt the five 6
varied from en;
in the PTCSCDCC .
$13me by 0.1
1110:: efficiently
W“: effw

gamed by the i

[>1 H pOIaSSiLm

57
activity, substrate specificity, and thermostability [303,305]. Ca2+ stabilization of

lysozyme was proposed to be due to a decrease in the entropy of the Ca2+-bound unfolded
protein, relative to the entropy of the unfolded protein in the absence of Ca2+ [298]. This
specific salt effect was attributed to the characteristics of the specific binding sites (i.e.,
size, coordination number, and the nature of the liganding residues) [65].

A study of GAPDH thermostabilization by salt indicated that the relative effects of
K3PO4, Na3PO4, K2SO3, NaZSO3, KCl, and NaCl were consistent with their abilities to
reduce the enzyme solubility in aqueous solvent. Their action was attributed to their
decreasing effect on water activity, a general salt effect [76]. Ammonium sulfate protein
precipitation is a common application of the general salt effect for enzyme precipitation.
Once precipitated, enzymes are typically also stabilized. This procedure is a standard,
convenient way to store enzymes. Thauer and colleagues studied the effect of salts on the
thermostability and activity of five M. kandleri methanogenic enzymes [153-155,164,183].
While the five enzymes are activated and stabilized by salts, the extent of the salt effect
varied from enzyme to enzyme. CHO—H4MPT formyltransferase was optimally stabilized
in the presence of 1.5 M K2HP04 while the Ego-dependent CI-12=H4MPT reductase was
stabilized by 0.1 M KzHPO4. K+ and NH4+ cations typically improve enzyme stabilities
more efficiently than other cations. Of all the anions, 8042' and HPO42' had the strongest
activating effect [164]. Enzyme stabilizing salt requirements however, are not always
satisfied by the intracellular salt concentration. M. kandleri's intracellular salt concentration
(>1 M potassium plus 1.2 M 2,3-diphosphoglycerate) [164] seems to favor CH2=H4MPT
reductase activity (optimal at 2.0-2.5 M salt), but this intracellular salt concentration is far
from being the optimal enzyme stabilizing concentration (optimal between 0.1 and 1.5 M
salt) [155]. These studies illustrate the variety of situations encountered: (i) similar
enzymes from different organisms do not have the same stabilization and activation

requirements, (ii) enzymes from the same organism are not equally affected by an

 

    
 
   

tiiirnmntrti‘i
man‘s int:
lithinnihtsis
httnttn charge.
(13231306). ins:-
protcin desrabil:

the interface btt‘

 

This hrpothesis I
sat concentratin
M) I‘Oquired to 5
Mar uncharget
mtSOphiiic em
"“33" Wrens:
“lth the less pt
than the lime”
{hm PmdiCt:
“Kit Partially

Other
dminutes bi
Mammy!“l
”Elam to r
”gamism ;
know} [0 stab
mmmues (

Wins [3071
mitinp HOw

58
environmental (i.e., cellular) factor, and (iii) thermophilicity and thermostability are not

necessarily controlled or favored by the same factors.

The hypothesis that the general salt effect was due to direct electrostatic interactions
between charged amino acids and salt counterions was not corroborated by experimental
data [306]. Instead, a smeared charge repulsion mechanism was proposed: Salt-induced
protein destabilization was due to nonspecific charge repulsion and to ionization changes at
the interface between complementary peptide smfaces in the partially unfolded protein.
This hypothesis agrees with the well-documented protein stabilizing effect of intermediate
salt concentrations, in particular with the relatively high salt concentrations (1.0 M to 2.0
M) required to stabilize some thermophilic enzymes [306]. It also agrees with the reduced
polar uncharged residue content in some thermophilic enzymes, when compared to their
mesophilic counterparts [85,230]. At high temperatures, the system‘s higher kinetic
energy increases the potential for exposing buried surfaces to the solvent. Salt interactions
with the less polar residues in the partially unfolded thermophilic enzyme are less favorable
than the interactions with the more polar mesophilic enzyme residues. Furthermore, this
theory predicts that greater solvent ionic character will stabilize folded proteins relative to
their partially unfolded forms.

Other chemical effectors - Breitung er al. (1992) [164] found that the
differences between M. kandleri, M. thermoautotrophicum, Archaeoglobus fidgidus, and
Methanosarcina barkeri CHO—H4MPT formyltransferase activation by salts were directly
correlated to the intracellular 2,3-diphosphoglycerate concentration in the different
organisms. Polyalcohols such as glycerol, sorbitol, mannitol, sucrose, and starch and are
known to stabilize proteins [306]. Hyperthermophilic bacterial cytoplasms contain
thernmmines (polyamines) that have been shown to thermostabilize nucleic acids and
proteins [307]. Organic polymers such as polyethylene glycol (PEG) also stabilize folded

proteins. However, they can also induce precipitation and even inactivation.

 

Pressu
reported [30M
crysulhne solid
Unfolding a pro
Stable than the i
compact , foide
pressure has on
thermophiles h;
from 200 to 401
from organism
Nations are pf

at high pressurt

MOleClllar m

59
Pressure effects - The ability of high pressure to stabilize proteins has been

reported [308,309]. The observation that folded proteins have densities similar to
crystalline solids is the theoretical basis for pressure as a general stabilizing force.
Unfolding a protein would increase its volume, making a partially unfolded protein less
stable than the folded protein at increased pressures. Thus, at higher pressures, the
compact , folded protein form is more stable. Hei and Clark [308] have studied the effect
pressure has on enzymes from barophilic organisms. They showed that some barophilic
thermophiles have enzymes which are also barophilic (i.e., optimal activity at pressures
from 200 to 400 atrn) [308]. Hei and Clark also characterized barOphilic enzymes isolated
from organisms that were not, themselves, barophilic [308]. Since numerous chemical
reactions are performed at high temperatures and pressures, demonstrating enzyme stability

at high pressmes is potentially important for enzyme biotechnological applications [309].

Molecular mechanisms of protein thermophilicity ‘

The molecular basis for enzyme thermophilicity is not well understood, and the
literature on this subject is limited. Linear Arrhenius p10ts for numerous enzymes indicate
that if there are structural changes in the catalyst throughout the active temperature range,
they are either not catalytically significantor they are offsetting. The discontinuous
Arrhenius curves observed for catalysis by some enzymes [82,145,190] indicate that, in
these specific cases, temperature dependent, functionally significant structural changes may
occur. These discontinuous Arrhenius ewes were believed to be characteristic of
thermophilic enzymes, but, as mentioned previously, not all thermophilic enzyme
Arrhenius plots are discontinuous and some mesophilic enzyme Arrhenius plots are also
discontinuous [92]. Explanations for Arrhenius discontinuities have been proposed that do
not involve altering the catalyst’s structure (e.g., change in the slow step of the reaction)
[207 ,208]. Generally, enzyme activity in its active temperature range is well described by
the Arrhenius equation, and can be derived from the temperatme for maximal activity,

 

magnitude of
equation, hovt
Thus, an enzy
the theoretical
protein unfold
independend)

Most :
thermophiiieit
thtrtrrostabiiit
Whig site if.
inmascd its t]
d” abStnce or
and militant en
Minty of the
Minty 00nd;
its Own tempt
enlime therm
thermophilicit
iMimi)" of 2 t

60
magnitude of maximal activity, and the activation energy for the reaction. The Arrhenius

equation, however, cannot predict the optimal temperature for an enzymatic reaction.
Thus, an enzyme’s active temperature range appears to be determined at the lower end by
the theoretical limitation on activity determined by the Arrhenius relationship. Partial
protein unfolding or covalent chemical modifications appear to determine the upper limit,
independently from the Arrhenius equation.

Most studies on enzyme thermostabilization do not comment on the enzyme
thermophilicity. Only a few reports [250,310] describe mutations that altered both enzyme
thermostability and thermophilicity. Kuroki et al. [250] demonstrated that adding a Ca2+
binding site in human lysozyme by site directed mutagenesis both stabilized the protein and
increased its thermophilicity from 70°C (native enzyme) to 80°C (mutant holoenzyme). In
the absence of Ca2+, the mutant enzyme thermophilicity was reduced to 65°C. The native
and mutant enzyme catalytic rates were superimposable up to the temperature for maximal
activity of the wild-type enzyme (~70°C). Above that temperature the mutant enzyme
activity continued to increase, with the same apparent Arrhenius behavior, until it reached
its own temperature for maximal activity. The mutant behavior suggests that the wild-type
enzyme thermophilicity was limited by its thermostability, and that the higher
thermophilicity of the mutant was due to an increase in its thermostability. In one study,
insertion of 2 to 4 amino acids into Caldicellulosiruptor saccharolyticus xylanase generated
stability-only mutants and stability-plus-optimal temperature-altering mutants [310].
Insertion of a single Pro-Arg sequence reduced the enzyme's specific activity more than 2-
fold, the thermostability 4-fold, and the optimal temperature by 20°C. These mutations
support the hypothesis that, in this case, thermophilicity was also limiwd by folded protein
stability. Other C. saccharolyticus xylanase mutants, displayed reduced activity and
reduced stability (similar to the Pro-Arg mutant) while retaining native-like thermophilicity
(70°C), suggested that the molecular bases for protein thermostability and thermophilicity
can be structurally distinct. The Tomazic and Klibanov thermostability model (scheme (1),

61
and [197]) can potentially explain the observed difference between the two classes of C.

saccharotyticum xylanase mutants. The mutations which only reduce enzyme
thermostability would increase the rate of scrambled structure formation from the enzyme
non-native state, while the mutations that reduce both thermophilicity and thermostability
would result fi'om the shift of native to nonnative enzyme at lower temperatures. This
analysis predicts that a class of thermophilicity-reducing mutants that do not alter
thermostability could be generated if the normative state (resulting from reversible
unfolding) were stabilized with respect to the scrambled structure (the enzymes would only
reversibly inactivate).

While the relationship between enzyme kinetics and protein thermophilicity]
thermostability remains unclear, it has been proposed that the poor activity of thermozymes
at low temperatures is the result of excessive rigidity. This rigidity is believed to be
necessary to maintain active enzyme architecture at high temperatures [198,203,215]. Data
from crystallography, deuterium exchange, proteolytic susceptibility, and other
experimental approaches have demonstrated that folded thermozymes are indeed more rigid
at low temperatures than their mesophilic counterparts [198,203,215]. Tchemajenko er al.
[206], however, reported linear Arrhenius plots for the four xylose isomerases they
studied. Since these plots were perfectly fit by the Arrhenius equation, increased low-
temperature protein rigidity was not required to explain the poor low-temperature activity of
thermophilic enzymes [206]. The temperatme dependence of thermophilic enzyme activity
might, instead, only be determined by temperature-dependent substrate kinetic energy
variations. Therefore, determining the importance of protein flexibility to thermophilic
enzyme activity will require comparisons of corresponding thermophilic and mesophilic
enzyme flexibilities and activities throughout their active temperature ranges.

62
Genetic engineering of thermozymes

Modification of enzyme catalytic properties

The accumulating protein structural data justifies the use of enzyme engineering as a
rational research approach to make enzyme catalytic properties fit industrial processes.
Knowing the L-lactate dehydrogenase structure and catalytic mechanism, Holbrook and
colleagues attempted to systematically modify the B. stearothennophilus L-lactate
dehydrogenas into a malate dehydrogenase by a point mutation proposed to better
accommodate oxaloacetate in the substrate pocket [311], then to design a non-specific a-
hydroxy acid dehydrogenase from the lactate dehydrogenase [312]. This non-specific a-
hydroxy acid dehydrogenase mutant enzyme remained thermostable, and could be used for
various chemical syntheses. A similar approach was used by Meng et al. [313] to switch
the substrate preference of T. thermosulfun’genes xylose isomerase from xylose to
glucose. Arnold and colleagues used PCR-mdiated random mutagenesis to enhance
subtilisin B activity in polar organic solvents [313a,313b]. Using several screening steps,
they eventually selected a multiple mutant that was 256 times more efficient than the wild-
type enzyme in 60% dimethylformamide. A similar PCR based technique, was
successfully employed to enhance the Z. mobilis 1° ADH thermostability [217]. These
studies provide proof of principle for the general application of directed enzyme design

strategies using mutagenesis.

Thermostability and thermophilicity engineering

Research has shown that while thermophilicity and thermostability can be altered
independently, they are often structtn'ally related. Numerous molecular interactions might
be responsible for these properties, however, only a few are of genetic engineering value:
(i) protein sequences contain redundant information for proper folding, and protein
structures can often accommodate amino acid substitutions without significantly altering the

catalytic efficiency; (ii) a small number of amino acid substitutions throughout the protein

 

can sigru‘frem
protein are in
ones) is critic
stabil‘zation i
sufaee 100p 5
analysis has it
tremophilic ;
311mg protei
hehces by cap
same engineer
successfully sr
methods (eg. H
dinnhona] Str
identified (e. g.,
inc“mined 1.
MIME Pepi

applications (>9

mm Start

63
can significantly alter its thermophilicity and thermostability; and (iii) certain regions in a

protein are more labile than others—stabilizing these regions (rather than the more stable
ones) is critical to improve protein thermal properties. Four general strategies for thermal
stabilization have been identified: more efficient protein core packing, tit-helix stabilization,
surface loop stabilization, and prevention of chemical degradation. Recent crystallographic
analysis has indicated that more-rigid core structure (lower Brvalues) is a hallmark of
thermophilic proteins [124,198]. However, systematic engineering of core rigidity without
altering protein function is beyond the capabilities of current technology. Stabilizing cr-
helices by capping or by introducing alanines has had spotty success, and suffers from the
same engineering problems as do the core packing strategies. Loop engineering has
successfully stabilized proteins [233,243,248254258,259,272]. Sequence analysis
methods (eg. HCA) now reliably predict surface loop regions, even in the absence of 3-
dimensional structural information. Specific stabilizing amino acid substitutions have been
identified (e.g., substitution of surface lysines with arginines and introduction of prolines
in constrained loops, and salt bridges), making this approach a good engineering tool.
Preventing peptide chemical degradation is also important for very high temperature
applications (>90°C). Trends toward reduced Cys, Asn, and Gln contents in
hyperthermophilic enzymes have been identified in nature [303], and deamidation has been
shown to be a main factor responsible for irreversible enzyme inactivation [197]. The
appropriate strategy for engineering protein thermostability or thermophilicity depends on
the project goal, on the available structural information, and on the thermal mechanisms
limiting enzyme stability or activity (Fig. 3). Selecting an initial enzyme with thermal
properties as close to the target as possible clearly increases the chances of successful
engineering. The probability of engineering success is further enhanced by having the
initial enzyme's high-resolution 3-dimensional structtne as well as a protein in the same
class that has thermal characteristics close to the desired protein. Comparing
thermophilicity and thermostability properties indicates whether the limiting step in protein

Figure 3. Flow chart of potential steps toward engineering enzyme

thermophilicity and thermostability.

 

65

Design goal

Thermophilicity Thamostability

Reversible protein inactivtion Irreversible protein inactinipn
(Thermostability > Thermophilicity) (Then-nostabrlrty = 'I'hmnovhlllcrty)

N0 precipitation Precipitation

0?me
Precipitate Precipitate

active inactive
Prevent structural clung ' Thor-mo bilicity
General:
1) Random mutagenesis with selection
2) sequence comparison directed mutagenesis
3) substitute Pro residues into loops
4) replace Lys with Arg in loops
5) chemically aosslink V '
6) immobilize 1) Immobilize

 
 
 
   

Complete Partial

7) add stabilizing chemicals to the system 2) Chemically modify
a) protein specific a) succinylate Lys residues
i) salts (N (:1, KCl. etc.) b) intramolecular croaslink
ii) substrates/effectors 3) add polyalcohols to system
b) reduce solvent entropy
i) glycerol

ii) Hofmeister series ions

1) add divalent cation binding site

2) reduce bond strain by mutagenesis

3) till internal cavities by mutagenesis

4) stabilize helices
i) 11le to cap helix C-terminus
ii) Glu/ASP to cap helix N-tcrminus
iii) Ala to stabilize helix

v 0 C IV I

1) addGlytoloops

 

WW
1) reduce themunber ofAsn. Gln. and Cys residues

66
inactivation is the initial partial unfolding or ineversible structural changes (such as

precipitation or chemical degradation). If, in the temperature range of interest, the protein
thermostability half-fife is longer than the activity half-life, the reversibly partially unfolded
protein is not immdiately irreversibly inactivated If, however, these two half-lives are
similar, irreversible inactivation follows the initial unfolding almost instantly. The protein
concentration in the target application should be estimated based on the protein function
(receptor, enzyme, antibody, etc.), to provide a concentration range for these
characterizations. Determination of the thermostability half-life at different initial protein
concentrations within this range will indicate if inactivation is intramolecular (concentration
independent) under these specific conditions.

Having increased enzyme thermostability allows applications of repeated thermal
cycling (e.g., PCR), as well as permitting pre-application high-temperature processing
steps (e.g., processing enzymes in food or feed additives). To increase enzyme
thermostability, engineering efforts should focus on preventing irreversible protein
inactivation. Strategies to prevent protein aggregation (e.g., chemical modification of
lysine residues with succinic acid [197], addition of polyalcohols to the solution [306], or
immobilization on an inert matrix) might significantly thermostabilize the protein. This
approach would need to be undertaken if (i) enzyme precipitation coincides with
inactivation and the precipitated protein does not retain complete activity, or (ii) if
aggregation is undesirable for the target application. If these measures are insufficient or
inappropriate for the design goal, then prevention of inactivating conformational changes
through peptide modifications such as intramolecular cross-linking or through system
additives (cg. salt or glycerol) that stabilize the folded protein structure should be
examined. Salts have been shown to stabilize proteins through specific binding, charge
shielding, and modification of the system entropy. The addition of organic chemicals such
as glycerol and polyethyleneglycol has also been shown to stabilize some proteins, but they
often reduce activity and can destabilize some proteins. The specificity of intramolecular

67
cross-linking is difficult to control and often reduces protein activity but it covalently tethers

protein regions analogously to disulfide linkages.

Either random or site directed mutagenesis of the initial protein may be used to
thermostabilize proteins. Random mutagenesis has been successfully used to increase
enzyme thermostability and resistance to solvent-induced denaturation. Hageman and
colleagues [314] developed a system in which a gene encoding a mesophilic enzyme is
introduced into a thermophilic host, and variant enzymes with increased thermostability are
selected during growth at increasing temperatures. Heat-stable kanamycin
nucleotidyltransferase mutants were obtained using a B. stearothennophilus strain that
expressed a mesophilic kanamycin nucleotidyltransferase to select kanamycin resistant
variants at increasing temperatures [239]. Although attractive at first glance, this approach
is limited by the absence of genetic tools (i.e., shuttle vectors, transformation methods)
available for most thermophiles and hyperthermophiles, as well as by the limited enzymatic
activities directly selectable in B. stearothennophilus cultures. The cunent lack of
thermophilic cloning hosts also makes classic genetic complementation by thermophilicity
mutants a significant challenge. Practical use of random mutagenesis requires a powerful
mutant selection procedure or specific structural information (to limit the size of the target
thus, reducing the total number of possible mutants to be tested).

Site-directed mutagenesis usually requires protein structural information. Strategies
that include substitutions with residues present in structurally similar proteins with thermal
properties similar to the desired mutant require sequence information from both proteins. If
an enzyme targeted for hyperthermophilic applications (>80°C) is irreversibly inactivated,
reduction of the number of noncatalytic Gln, Asn, and Cys residues can be used to stabilize
the protein structme, using the single peptide sequence. Loop stabilization by introducing
Pro or Arg residues also requires only knowing the protein sequence. Exact structural
information from crystallography, NMR, or homology modeling allows the engineering of
multi—residue motifs such as metal binding sites or disulfide bridges (for moderate

68
temperature stabilization) to stabilize the enzyme structure, and allows specifically tailored

design schemes such as cavity filling mutations, helix stabilization by alanine insertions or
capping (eg. His at the c-terminus), and alleviation of specific bond strains. Deuterium
exchange, measured by NMR, can precisely identify the more labile protein regions to
stabilize [315]. Muheim et al. [316] created a dihistidine metal-chelating site on a surface
B—sheet of cytochrome c, and cross-linked it with the metal complex Run(2,2'-bipyridine).
This cross-linking increased the meltingltemperature of the mutant enzyme by more than
23°C.

Increasing enzyme thermophilicity requires preventing thermally induced unfolding.
Since enzyme thermophilicity is often limited by thermostability, engineering
thermophilicity usually requires enzyme thermostabilization. Also note that, for an enzyme
that precipitates upon hearing, if precipitation is not the first inactivation step, preventing

aggregation is unlikely to enhance thermophilicity.
THESIS OBJECTIVES AND SIGNIFICANCE

Because archaeal and bacterial thermophiles were the first organisms to have
evolved the study of these organisms and their enzymes is expected to provide insights into
the origins of biocatalysis on Earth [317]. With their remarkable thermostability and
activity, thermophilic enzymes provide a model for studying protein thermostability and
biocatalysis at high temperatures. Structurally similar to their mesophilic counterparts,
thermophilic enzymes are powerful tools for developing enzyme structural or functional
models through comparative analysis. Their high-temperature activity and high stability
make thermophilic enzymes excellent candidates for industrial enzymatic applications.
Therefore, the fundamental scientific examination of thermophilic life may also yield novel

catalysts of tremendous biotechnological value.

69
Numerous highly similar thermophilic 2° ADHs have been isolated from

thermoanaerobic bacteria [7,8,15,16] but despite more that a decade of research since the
first enzymes were purified in 1981 [14,15], no report of the cloning of a gene encoding a
thermophilic 2° ADH has been published. There is also no record of either mesophilic or
thermophilic 2° ADH characterization by mutagenesis. The only thermophilic 2° ADH
peptide sequence avaliable was determined by Edman degradation [33]. Alignment of the
liver 1° ADH and the T. brockii 2° ADH peptides by Peretz and Burstein [33] suggested
that the structural Zn binding loop was absent in the 2° ADH, but they concluded that
insufficient similarity existed between the A. eutrophus, horse liver and
Thennoanaerobacten'wn brocla'i enzymes for further comparative analysis. Earlier, Lamed
and Zeikus reported that T. brockii 2° ADH activity was both cysteine and Zn dependent
[l5] and Bryant and Ljungdahl measured Zn specifically bound to the T. ethanolicus
(ATCC 31550) enzyme [14]. All thermophilic 2° ADHs identified are tetrameric with
reported subunit molecular masses near 40 kDa. This coincidence of enzyme structural
properties and the similarities seen in kinetic comparisons of these NADP(H) linked 2°
ADHs led N agata er al. to conclude that all thermophilic 2° ADHs were extremely
structurally similar [16]. The greater than 75% sequence identity between the mesophilic
C. beijerinckii and thermophilic T. brockii peptides further argues that all catabolic 2°
ADHs may share significant 3-dimensional structural identity. These thermophilic and
mesophilic 2° ADH peptide sequences therefore, make an excellent system for
determination of protein thermostability structure-function relationships through
comparative analysis. However, reciprocal mutagenesis experiments to verify the sequence
based predictions require expressed clones of the genes encoding both proteins.

The T. brockii 2° ADH has been demonstrated to reduce aldehydes and ketones on
aliphatic molecules containing up to 10 carbons [318]. This enzyme requires ketones on
the second or third carbon, displaying higher activity toward unbranched noncyclic
molecules. The enantiospecificity was lower and (R)— specific in the reduction of very

 

Cb:

DOLE:

70
short ketones but greater than 90% and (S)- specific for reduction of ketones with aliphatic

chains greater than 5 carbons in length. The T. brockii 2° ADH, like the horse liver and
yeast 1° ADHs, produces (8)-alcohols using the E3 pathway [30]. For the T. brockii 2°
ADH, lower reaction temperatures also yielded higher enantiomeric excess [319].
Enantiomeric excess at equillibrium is a function of the Gibbs energy ratio between the (R)-
and (S)- specific reactions and this relationship indicated that the variation of T. brockii 2°
ADH reaction stereospecificity with temperatme resulted mainly from entropy
considerations. An ordered kinetic mechanism with cofactor binding followed by substrate
binding, hydride transfer, product release, then cofactor dissociation, similar to the liver 1°
ADH mechanism, was identified for T. brockii 2° ADH catalysis [320]. The T. brockii
enzyme is both thermostable and thermophilic [15]. Its optimal temperature for activity
was reported to be above 70°C with significant inactivation rates occrn'ring at temperatures
above 90°C. A discontinuity in the Arrhenius plot for propan-2—ol and butan-2-ol oxidation
by this enzyme was reported in the same study.

The T. brockii 2°ADH has already been used in the analytical scale production of a
chiral constituent of civet (a perfume additive) [321] and of an insect pheromone [322]. Its
potential value for other syntheses has also been examined [323]. Thermophilic 2° ADHs
have high temperature optima (80 to 95 °C), high thermostability, and reduced sensitivity to
oxygen (they are only reversibly inactivated, unlike 1°ADI-ls) [8]. The extent of T. brockii
2°ADH stability in the presence of organic solvent is still controversial. In the original
1981 study [323] the enzyme remained 80% active after 15 min at 52°C in the presence of
40% propanol-Z, whereas it was reviewed by Cowan [59] to rapidly inactivate at
temperatures above 45°C and in the presence of 10% organic solvent. T. brockr'i enzyme
catalysis was determined in vitro in supercritical 002 indicating that it is active in
nonaqueous conditions under high pressure [324]. The T. brockz'i enzyme was also active
toward both highly water soluble ketones and substrates with low water-solubility in a

range of water-oil emulsions containing either cationic or neutral surfactants [325]. A

 

tried

Q

flint

7 1
range of potential chiral transformations has already been demonstrated for ADI-Is. 2°

ADHs, optimized to interconvert ketones and 2° alcohols, are better suited to the production
of chiral alcohols that 1° ADHs, optimized to interconvert aldehydes and 1° alcohols.
Therefore, continued research into the molecular mechanisms responsible for 2° ADH
thermostability, thermophilicity, and substrate specificity will advance our understanding of
these general enzyme properties specifically using an enzyme with potential as an industrial
chiral catalyst.

The research objectives presented in this thesis advances knowledge on
thermophilic 2° ADH 2° ADH structure-function relationships in 5 ways. First, cloning and
characterizing the gene encoding the T. ethanolicus 39E 2° ADH will make mutagenic
analysis of thermOphilic 2° ADH structure-function relationships possible. Second, stable
overexpression of the properly active recombinant thermophilic 2° ADH in E. coli will
provide sufficient quantities of easily pmified protein for structural analysis. Third,
characterization of enzyme thermostability and thermophilicity will allow sequence
comparisons between this thermophilic enzyme and the similar mesophilic enzyme from C.
beijerinckii for identification of 2° ADH thermal property molecular determinants. Fourth,
site directed mutagenesis and biophysical analysis will test the predictions of 2° ADH
catalytic architecture based on comparison to those documented for 1° ADHs. Finally, the
biochemical mechanism of thermoinactivation and the forces stabilizing active the T.

ethanolicus 2° ADH structure will be elucidawd.
ACKNOWLEDGMENTS

I gratefully acknowledge Dr. Claire Vieille for her part in writing and editing the original
manuscript. I also am grateful to Dr. Claire Vieille, Dr. Vladimir Tchemajenko, Dr. Mariet
Van dc Werf, and Maris Laivenieks for their helpful discussions. I acknowledge Mr. Chris

Jambor for critical review of parts of the original manuscript.

10.

11.

12.

13.
14.

15.
16.
17.

18.
19.
20.

72
REFERENCES

Branden, C.-I., Jomvall, H., Eklund, H. and B. Fururgren (1975) in The
enzymes, 3rd ed., vol. XI part A" (P. D. Boyer, ed.) pp. 103-190. Academic
Press, New York.

Gottschalk G. (1986) Bacterial metabolism, 2nd edition. Springer-Verlag, New
York.

Schimpfessel, L. (1967) Arch. Int. Physiol. Biochim. 75, 368-369
Schimpfessel, L. (1968) Biochim. Biophys. Acta 151, 386—?.
Lutstorf, U. and Megnet, R. (1968) Arch. Biochem. Biophys. 126, 933-944

Sund, H. and Theorell, H. (1963) in The enzymes, 2051 ed., vol. 7, (Boyer, P D.,
Lardy, H. and Myrback, K., ed.), pp. 25-83 Academic Press, New York.

Bryant, F. 0., Weigel, J. and Ljungdahl, L. G. (1988) Appl. Environ. Microbiol.
54, 460-465

Burdettc, D. S. and Zeikus, J. G. (1994) Biochem. J. 302, 163-170
Kolb, E. and Harris, J. I. (1971) Biochem J. 124, 76P-77P

Rella, R., Raia, C. A., Pensa, M., Pisani, F. M., Gambacorta, A., De Rosa, M.,
and Rossi, M. (1987) Eur. J. Biochem. 107, 475-479

Ma, K., Robb, F. T., and Adams, M. w. w. (1994)App1. Environ. Microbiol.
60, 562-568

Ismaiel, A. A., Zhu, C.-X., Colby, G. D. and Chen, J .-S. (1993) J. Bacteriol.
175, 5097—5105

Steinbfichel, A. and Schlegel, H. G. (1984) Eur. J. Biochem. 141, 555-564

Bryant, F. 0. and Ljungdahl, L. G. (1981) Biochem. Biophys. Res. Corn. 100,
793-799

Lamed R. J. and Zeikus, J. G. (1981) Biochem. J. 195, 183-190
Nagata, Y., Maeda, K. and Scopes, R K. (1992) Bioseparation 2, 353-362

Neale, A. D., Scopes, R. K., Kelly, J. M., and Wettenhall, R. E. H. (1986) Eur.
J. Biochem. 154, 119-124

Bleicher, K. and Winter, J. (1991) Eur. J. Biochem. 200, 43-51
Ainsley, G. R. Jr. and Cleland, W. W. (1972) J. Biol. Chem. 247, 946-951
Brooks, R. L. and J. D. Shore (1971) Biochemistry 10, 3855-3858

 

41.

<2.

 

21.
22.

23.
24.

25.
26.

27.
28.
29.
31.

30.

32.

33.
34.
35.
36.
37.
38.

38a.

39.

40.
41.

42.

73
Bush, K., Shiner, V. J ., and H. R. Mahler (1973) Biochemistry 12, 4802-4805

Beijer, N. A., Buck, H. M., Sluyterman, L. A. AE., and Meijer, E. M. (1990)
Biochimica Biophysica Acta 1039, 227-233

Dunn, M. F. and Hutchison, J. S. (1973) Biochemistry 12:4882-4892

Jacobs, J. W., McFarland, Wainer, I., Jeanmaier, D., Ham, C., Hamm, K.,
Wnuk, M., and Lam, M. (1974) Biochemistry 13, 60-64

McFarland, Chu, Y.-H., and J. W. Jacobs (1974) Biochemistry 13, 65-69

Eklund, H., Nordstrbm, B., Zeppezauer, E., Soderlund, G., Ohlsson, I., Boiwe,
T. and Branden, C.-I. (1974) FEBS Lett. 44, 200-204

Rossmann, M. G. Moras, D., and K. W. Olsen (1974) Nature 250, 194-199
Wierenga, R. K. and H01, G. J. (1983) Nature 302, 842-844
Prelog, V. (1964) Pure Appl. Chem 9, 119-130

Bradshaw, C. W., Hummel, W. and Wong, C.-H. (1992) J. Organic Chem. 57,
1533-1536

Bradshaw, C. W., Shen, J.-G. and Wong, C.-H. (1992) J. Organic Chem. 57,
1526-1532

Jendrossek, D., Steinbuechel, A. and Schlegel, H. G. (1988) J. Bacteriol. 170,
5248-5256

Peretz, M. and Y. Burstein (1989) Biochemistry 28, 6549-6555
Kullmann, W. (1987) CRC Press, Inc., Boca Raton, FL

Cheetham, P. S. J. (1993) Trends Biotechnol. 11, 478-488

Dordick, J. S. (1992) Trends Biotechnol. 10, 287-293

Faber, K. and Franssen, M.C.R. (1993) Trends Biotechnol. 11, 461-470
Jones, J. B. (1986) Tetrahedron 42, 3351-3403

Sarney, D. B. and Vulfson, E. N. (1995) Trends Biotechnol. 13, 164-172

Zong, M.—H., Fukui, T., Kawamoto, T. and Tanaka, A. (1991) Appl. Microbiol.
Biotechnol. 36, 40-43

Fuganti, C. and Grasselli, P. (1985) Tetrahedron Let. 26, 101-104

217%? K., Inuoe, K., Ushio, K., Oka, s. and Ohno, A. (1987) Chem. let. 4,

Trincone, A., Lama, L., Lanzotti, V., Nicolaus, B., De Rosa, M., Rossi, M. and
Gambacorta, A. (1990) Biotech. Bioeng. 35, 559-564

LII

irl
. (I

61

63.

43.
44.
45.
46.
47.
48.

49.

50.

51.

52.
53.

54.
55.

56.
57.
58.

59.
60.

61.

62.

63.
64.

74

Hummel W. (1990)App1. Microbiol. Biotechnol. 34, 15-19

Whitesides, G. M. and Wong, C.-H. (1983) Aldrichimica Acta 16, 27-34
Rosazza, J. P. N. (1995) Amer. Soc. Microbiol. News 61, 241-245
Meyer, H.-P. (1991) Biol. Ferm. Eng. 8, 602-606

Wong, C.-H. (1989) Science 244, 1145-1152

Irwin, J. B., Lok, K. P., Huang K. W. C. and Jones, J. B. (1978) J. Chem. Soc.
Perkin I 12, 1636-1641

Bradshaw, C. W., Hummel, W. and Wong, C.-H. (1992) J. Org. Chem. 57,
1532-1536

Bradshaw, C. W., Fu, H., Shen, G.-J. and Wong, C.-H. (1992) J. Org. Chem.
57, 1526-1532

Hummel, W. and Kukla, M. R. (1989) Eur. J. Biochem. 184, 1-13

May, S. W. and Padgette, S. R. (1983) Bio/Technology 1, 677-686

Persson, M., MAnsson, M.-O., Biilow, L. and Mosbach, K (1991) Biotechnology
9, 280-284

Klibanov, M. (1983) Proc. Biochem. 18, 13-23

Eeg-Rghlgcgbk, Fauve, A., Renard, M. F. and Veschambre, H. ( 1992) Biocatalysis

Zaks, A. and Klibanov, A. M. (1988) J. Biol. Chem. 263, 8017-8021
Klibanov, A. M. (1989) Trends Biol. Sci. 14, 141-144

231911213, J. R, Lee, K. M. and Biellmann, J. F. (1987) Eur. J. Biochem. 163,

Cowan, DA. (1992) Trends Biotechnol. 10, 315-323

Lee, Y.-E., Lowe, S. E. and Zeikus, J. G. (1993) Appl. Environ. Microbiol. 59,
3134—3137

Lehmacher, A. and Hensel, R. (1994) Mol. Gen. Genet. 242, 163-168

Lee, C., Bhatnagar, L., Saha, B. C., Lee, Y.-E., Takagi, M., Imanaka, T.,
Eggasarian, M. and Zeikus, J. G. (1990) Appl. Environ. Microbiol. 56, 2638-
Zeikus, J. G. and Brock, T. D. (1971) Biochim. Biophys. Acta 228, 736-745

Adams, M. W. W. (1993) Ann. Rev. Microbiol. 47, 627-658

 

 

so.

65.

66.
67.
68.

69.
70.

71.

72.

73.

74.

75.

76.
77.
78.
79.

80.

81.

82.
83.

84.

75

Coolbear, T., Whittaker, J .M. and Daniel, RM. (1992) Biochem. J. 287, 367-
374

Daniel, R. M. (1992) Origins Life Evol. Biosphere 22, 33-42
Lowe, S. E., Jain, M. K., and Zeikus, J. G. (1993) Microbial. Rev. 57, 451-509

Stetter, K. 0., Fiala, G., Huber, G., Huber, R., and Segerer, A. ( 1990) FEMS
Microbiol. Rev. 75, 117-124

Wiegel, J. and Ljungdahl, L. G. (1984) CRC Crit. Rev. Biotechnol. 3, 39-108

Coolbear, T., Daniel, R. M. and Morgan, H. W. (1992) Adv. Biochem. Eng.-
Biatechnal. 45, 57-98

Vieille, C., Burdctte, D. S. and Zeikus, J. G. (1996) in Biotech. Ann. Rev., vol.
2, (M. R. E1 Geweley, ed.) pp. 1-83. Elsevier, Amsterdam, Neth.

Berquist, P. L. and Morgan, H. W. (1992) in Molecular biology and
Biotechnology of extremaphiles, (Herbert, R. A. and Sharp, R. J., eds.), pp. 44-
75. Chapman and Hall, NY

Winter, J ., Lerp, C., Zabel, H.-P., Wildenauer, F. X., Kanig, W. H., and
Schindler, F. (1984) System. Appl. Microbial. 5, 457-466

Schmitz, R.A., Richter, M., Linder, D. and Thauer, R.K. (1992) Eur. J. Biochem.
207, 559-565

O'Brien, W. E., Brewer, J. M. and Ljungdahl, L. G. (1973) J. Biol. Chem. 248,
403-408

Hatchikian, E. C., and Zeikus, J. G. (1983) J. Bacterial. 153, 1211-1220
Benachenhau-Lahfa, N., Labedan, B. and Farterre, P. (1994) Gene 140, 17-24
Lacher, K. and Schafer, G. (1993) Arch. Biochem. Biophys. 302, 391-397

35mg £16, Schmid, R., and Schafer, G. (1993) Arch. Biochem. Biophys. 307,

Lawyer, F. C., Staffel, S., Saiki, R. K., Myamba, K., Drummand, R. and
Gelfand, D. H. (1989) J. Biol. Chem. 264, 6427-6437

ggajses, G. J., Littlechild, J. A., Watson, H. C., and Hall, L. (1991) Gene 109,

Fabry, S., and Hensel, R. (1987) Eur. J. Biochem. 165, 147-155

Sakai, K., Narihara, M., Kasama, Y., Wakayama, M., and Mariguchi, M. (1994)
Appl. Environ. Microbial. 60, 2911-2915

Hartag, A. T. and Daniel, R. M. (1992) Int. J. Biochem. 24, 1657-1660

85.

86.

87.

88.

89.

90.
91.
92.

93.

94.

95.

96.
97.
98.
99.

100.

101.
102.

103.

104.

76

Bealin-Kelly, R, Kelly, C. T. and Fogarty, W. M. (1991) Appl. Microbial.
Biotechnol. 36, 332-336

Brosnan, M. R, Kelly, C. T., and Fagarty, W. M. (1992) Eur. J. Biochem. 203,
225-231

Gray, G. L., Mainzer, S. E., Rey, M. W., Lamsa, M. H., Kindle, K. L.,
Carmana, C. and Requadt, C. (1986) J. Bacterial. 166, 635-643

Fukusumi, S., Kamizana, A., Harinauchi, S., and Beppu, T. (1988) Eur. J.
Biochem. 174, 15-21

Harinauchi, S., Fukusumi, S., Ohshirna, T. and Beppu, T. (1988) Eur. J.
Biochem. 176, 243-253 ‘

Hyun, H. H. and Zeikus, J. G. (1985) Appl. Environ. Microbiol. 49, 1162-1167
Melasniemi, H. (1987) Biochem. J. 246, 193-197

Mathupala, S. P., and Zeikus, J. G. (1993) Appl. Microbial. Biotechnol. 39, 487-
493

Plant, A. R., Clemens, R. M., Morgan, H. W. and Daniel, R. M. (1987)
Biochem. J. 246, 537-541

Saha, B. C., Lamed, R., Lee, C.-Y., Mathupala, S. P. and Zeikus, J. G. (1990)
Appl. Environ. Microbial. 56, 881-886

Madi, E., Antranikian, G., Ohmiya, K., and Gottschalk, G. (1987) Appl. Environ.
Microbiol. 53, 1661-1667

Guy, G. R. and Daniel, R. M. (1982) Biochem. J. 203, 787-790
Schafer, G. and Meyering-Vos, M. (1992) Ann. N.Y. Acad. Sci. 671, 293-309
Saha, B. C. and Zeikus, I. G. (1990) Appl. Environ. Microbial. 56, 2941-2943

Wind, R. D., Liebl, W., Buitelaar, R. M., Penninga, D., Spreinat, A., Dijkhuizen,
L., and Bahl, H. ( 1995) Appl. Environ. Microbial. 61, 1257-1265

gaggaritis, A., and Merchant, R. F. J. (1986) CRC Crit. Rev. Biotechnol. 4, 327-

Schafield, L. R. and Daniel, R. M. (1993) Int. J. Biochem. 25, 609-617

Liithi, E., Jasmat, N. B. and Bergquist, P. L. (1990) Appl. Environ. Microbial.
56, 2677-2683

Lee, Y.-E., Lowe, S.E., Henrissat, B., and Zeikus, LG. 1993. J. Bacterial.
175:5890-5898.

Nakao, M., Nakayama, T., Harada, M., Kakuda, A., Ikemato, H., Kobayashi, S.
and Shibano, Y. (1994) Appl. Microbial. Biotechnol. 41, 337-343

l‘.
H

‘14
(‘fi'

105.
106.

107.

108.

109.
110.

111.

112.

113.

114.

115.

116.

117.

118.
119.
120.

121.
122.
123.

124.

77
Saha, B. C., and Zeikus, J. G. (1991) Appl. Microbial. Biotechnol. 35, 568-571

Shaa, W., Obi, S. K. C., Puls, J. and Wiegel, J. (1995) Appl. Environ.
Microbial. 61, 1077-1081

Lee, S.-G., Lee, D.-C., Hang, S.-P., Sung, M.-H., and Kim, H.-S. (1995) Appl.
Microbial. Biotechnol. 43, 270-276

Gibbs, M. D., Saul, D. J., Liithi, E. and Bergquist, P. L. (1992) Appl. Environ.
Microbiol. 58, 3864-3867

Kuriki, T., Okada, S. and Imanaka, T. (1988) J. Bacteriol. 170, 1554-1559

Freeman, 8. -A., Peek, K., Prescott, M., and Daniel, R. (1993) Biochem. J. 295,
463-469

Peek, K., Daniel, R. M., Monk, G, Parker, L., and Coolbear, T. (1992) Eur. J.
Biochem. 207, 1035-1044

Peek, K., Veitch, D. P., Prescott, M., Daniel, R. M., MacIver, B., and Bergquist,
P. L. (1993) Appl. Environ. Microbial. 59, 1168-1175

Kuriki, T. Park, J. -H., Okada, S. and Imanaka, T. (1988)App1. Environ.
Microbial. 54, 2881-2883

Shen, G.-J., Srivastava, K. C., Saha, B. C. and Zeikus, J. G. (1990) Appl.
Microbial. Biotechnol. 33, 340-344

Aubert, J. -P., Béguin, P. and Millet, J. (1993)Genetics and Molecular Biology of
Anaerobic Bacteria 1993, (Sebald, M., ed.). Springer-Verlag, NY

Nashihara, N., Nakamura, N., Nagayama, H. and Horikashi, K. (1988) FEMS
Microbiol. Lett. 49, 385-388

Plant, A. R., Morgan, H. W. and Daniel, R. M. (1986) Enz. Microbial. Technol.
8, 668-678

Richter, O.-M. H. and Schiifer, G. (1992) Eur. J. Biochem. 208, 343-349
Richter, O.-M. H. and Schafer, G. (1992.) Eur. J. Biochem. 209, 351-355

3230, W., DeBlais, S., and Wiegel, J. (1995) Appl. Environ. Microbiol. 61, 937-

Lee, Y.-E., and Zeikus, J. G. (1993) J. Gen. Microbial. 139, 1235-1243
Liithi, E., and Bergquist, P. L. (1990) FEMS Microbial. Lett. 67, 291-294

Barnes,E. M. Akagi,J. M. and Himes, R. H. (1971) Biochim. Biophys. Acta
227,199- 220

Russell, R. J. M., Haugh, D. W., Dansan, M. J. and Taylor, G. L. (1994)
Structure2, 1157- 1167

125.
126.
127.
128.

129.
130.

131.

132.

133.
134.
135.

136.

137.

138.

139.

140.

141.

142.

143.

1 44.

78
Kenealy, W. R. and Zeikus, J. G. (1982) FEMS Microbial. Lett. 14, 7-10

Shing, Y. W., Akagi, J. M. and Himes, R. H. (1975) J. Bacteriol. 122, 177-184
Lee, C. and Zeikus, J. G. (1991) Biochem. J. 273, 565-571

Lee, Y.-E., Ramesh, M. V. and Zeikus, J. G. (1993) J. Gen. Microbiol. 139,
1227-1234

Lehmacher, A. and Bisswanger, H. (1990) J. Gen. Microbiol. 136, 679-686

Dekker, K., Sugiura, A., Yamagata, H., Sakaguchi, K. and Udaka, S. (1992)
Appl. Microbial. Biotechnol. 36, 727-732

Dekker, K., Yamagata, H., Sakaguchi, K. and Udaka, S. (1991) J. Bacterial.
173, 3078-3083

Chan, M. K., Mukund, S., Kletzin, A., Adams, M. W. W. and Rees, D. C.
(1995) Science 267, 1463-1469

Mukund, S. and Adams, M. W. W. (1991) J. Biol. Chem. 266, 14208-14216
Ahem, T. and Klibanov, A. M. (1985) Science 228, 1280-1284

DiRuggiera, J., Robb, F. T., Jagus, R., Klump, H. H., Borges, K. M., Kessel,
M., Mai, X. and Adams, M. W. W. (1993) J. Biol. Chem. 268, 17767-17774

Britton, K. L., Baker, P. J., Barges, K. M. M., Engel, P. C., Pasqua, A., Rice,
D. W., Robb, F. T., Scandurra, R., Stillman, T. J. and Yip, K. S. P. (1995) Eur.
J. Biochem. 229, 688-695

Cansalvi, V., Chiaraluce, R., Paliti, L., Vaccaro, R., De Rosa, M., and
Scandurra, R. (1991) Eur. J. Biochem. 202, 1189-1196

Robb, F. T., Park, J .-B. and Adams, M. W. W. (1992) Biochim. Biophys. Acta
1120, 267-272

Cansalvi, V., Chiaraluce, R., Paliti, L., Gambacorta, A., De Rosa, M., and
Scandurra, R. (1991) Eur. J. Biochem. 196, 459-467

Wrba, A., Schweiger, A., Schultes, V. and Jaenicke, R. (1990) Biochemistry 29,
7584-7592

Igrggchy, A., Glockshuber, R. and Jaenicke, R. (1993) Eur. J. Biochem. 214,

gehultes, V., Deutzrnann, R. and Jaenicke, R. (1990) Eur. J. Biochem. 192, 25-

Zwickl, P., Fabry, S., Bogedain, C., Haas, A. and Hensel, R. (1990) J. Bacterial.
172, 4329-4338

Hensel, R., Jakob, 1., Scheer, H. and Lottspeich, F. (1992) Biochem. Soc. Symp.
58, 127-133

145.

146.
147.
148.
149.

150.

151.

152.

153.

154.

155.

156.

157.
158.

159.
160.

161.

162.

163.

79

Hensel, R., Laumann, S., Lang, J ., Heumann, H. and Lattspeich, F. (1987) Eur.
J. Biochem. 170, 325-333

Shah, N. N. and Clark, D. S. (1990) Appl. Environ. Microbial. 56, 858-863
Bryant, F. O. and Adams, M. W. W. (1989) J. Biol. Chem. 264, 5070-5079
Pihl, T. D., and Maier, R. J. (1991) J. Bacterial. 173, 1839-1844

Juszczak, A., Aana, S. and Adams, M. W. W. (1991) J. Biol. Chem. 266,
13834-13841

Wrba, A., Jaenicke, R., Huber, R. and Stetter, K. O. (1990) Eur. J. Biochem.
188. 195-201

Ostendorp, R., Liebl, W., Schurig, H. and Jaenicke, R. (1993) Eur. J. Biochem.
216, 709-715 .

Hanka, E., Fabry, S., Niermann, T., Palm, P. and Hensel, R. (1990) Eur. J.
Biochem. 188, 623-632

Breitung, J., Bdrner, G., Scholz, S., Linder, D., Stetter, K. O. and Thauer, R. K.
(1992) Eur. J. Biochem. 210, 971-981

Ma, K., Zirngibl, C., Linder, D., Stetter, K. O. and Thauer, R. K. (1991) Arch.
Microbial. 156, 43-48

Ma, K., Linder, D., Stetter, K. O. and Thauer, R. K. (1991) Arch. Microbial.
155, 593-600

Kunaw, J., Schwarer, B., Stetter, K. O. and Thauer, R. K. (1993) Arch.
Microbiol. 160, 199-205

Kunaw, J., Linder, D. and Thauer, R. K. ( 1995) Arch. Microbial. 163, 21-28

glamey, J. M. and Adams, M. W. W. (1993) Biochim. Biophys. Acta 1161, 19-
7

Blamey, J. M. and Adams, M. W. W. (1994) Biochemistry 33, 1000-1007
Blake, P. R., Park, J.-B., Bryant, F. 0., Aana, S., Magnusan, J. K, Ecclestan,
E., Howard, J. B., Summers, M. F. and Adams, M. W. W. (1991) Biochemistry
30, 10885-10895

Perler, F. B., Comb, D. 6., Jack, W. E., Moran, L. S., Qiang, B., Kucera, R.
B., Benner, J., Slatka, B. E., Nwankwa, D. 0., Hempstead, S. K., Carlaw, C.
K. S. and Jannasch, H. (1992) Proc. Natl. Acad. Sci. USA89, 5577-5581

Uemori, T., Ishina, Y., Tah, H., Asada, K and Kata, I. (1993) Nucleic Acids
Res. 21, 259-265

Hooper, A. B., Hansen, J. and Bell, R. ( 1967) J. Biol Chem. 242, 288-296

164.

165.

166.

167.

168.

169.

170.

171.
172.

173.

174.

175.

176.

177.

178.

179.
180.

181.

182.

80

Klein, A. R., Breitung, J., Linder, D., Stetter, K. O. and Thauer, R. K. (1993)
Arch. Microbial. 159, 213-219

Koch, R., Zablawski, P., Spreinat, A. and Antranikian, G. (1990) FEMS
Microbial. Lett. 71, 21-26

Laderman, K A., Davis, B. R., Krutzsch, H. C., Lewis, M. S., Grika, Y. V.,
Privalov, P. L., and Anfmsen, C. B. (1993) J. Biol. Chem. 268, 24394-24401

Laderman, K.A., Asada, K., Uemori, T., Mukai, H., Taguchi, Y., Kata, I. and
Anfinsen, CB. (1993) J. Biol. Chem. 268, 24402-24407

Koch, R., Spreinat, A., Lemke, K. and Antranildan, G. (1991) Arch. Microbial.
155, 572-578

Chung, Y. C., Kobayashi, T., Kanai, H., Alciba, T. and Kuda, T. (1995) Appl.
Environ. Microbial. 61, 1502-1506

Schumann, J ., Wrba, A., Jaenicke, R., and Stetter, K. O. (1991) FEBS Lett. 282,
122-126

Brown, S.H., and Kelly, RM. (1993) Appl. Environ. Microbial. 59, 2614-2621

Schuliger, J. W., Brown, S. H., Barass, J. A. and Kelly, R. M. (1993) Mol.
Marine Biol. Biotechnol. 2, 76-87

Phipps, B. M., Hoffman, A., Stetter, K. O. and Baumeister, W. (1991) EMBO J.
10, 1711-1722

Heyne, R. I. R., de Vrij, W., Crielaard, W. and Kanings, W. N. (1991) J.
Bacterial. 173, 791-800

Villa, A., Zecca, L., Fusi, P., Calumba, S., Tedeschi, G. and Tartara, P. (1993)
Biochem. J. 295, 827-831

Colombo, S., D'Auria, S., Fusi, P., Zecca, L., Raia, C. A. and Tortora, P. (1992)
Eur. J. Biochem. 206, 349-357

Brannenmeier, K., Kern, A., Liebl, W., and Staudenbauer, W. L. (1995) Appl.
Environ. Microbial. 61, 1399-1407

Simpson, H. D., Haufler, U. R. and Daniel, R. M. (1991) Biochem. J. 277, 413-
417

Ruttcrsmith, L. D., and Daniel, R. M. (1991) Biochem. J. 277, 887-890

Rugtersmith, L. D., and Daniel, R. M. ( 1993) Biochim. Biophys. Acta 1156, 167-
17

Castantina, H. R., Brown, S. H., and Kelly, R. M. (1990) J. Bacterial. 172,
3654-3660

Bowen, D., Littlechild, J. A., Fathergill, J. B., Watson, H. C. and Hall, L. (1988)
Biochem. J. 254, 509-517

183.

184.

185.

186.

187.

188.

189.

190.

191.

192.
193.

194.

195.
196.
197.
198.

199.
200.
201 .
202.

81

Breitung, J., Schmitz, R. A., Stetter, K. O. and Thauer, R. K. (1991) Arch.
Microbial. 156, 517-524

Eggen, R., Geerling, A., Watts, J. and deVos, W. M. (1990) FEMS Microbial.
Lett. 71,17-20

Marikawa, M., Izawa, Y., Rashid, N., Haaki, T. and Imanaka, T. (1994) Appl.
Environ. Microbial. 60, 4559-4566

Ruttersmith, L. D., Daniel, R. M. and Simpson, H. D. (1992) Ann. N.Y. Acad.
Sci. 672, 137-141

Slesarev, A. 1., Lake, J. A., Stetter, K. 0., Gellert, M. and Kazyavkin, S. A.
(1994) J. Biol. Chem. 269, 3295-3303

Robb, F. T., Masuchi, Y., Park, J .-B. and Adams, M. W. W. (1992) in
Biocatalysis at extreme temperatures, (Adams, M. W. W. and Kelly, R.M., eds.),
pp. 74-85. American Chemical Society, Washington, DC

Tibani, O., Cammarano, P. and Sanangelantani, A. M. (1993) J. Bacterial. 175,
2961-2969

Vieille, C., Hess, J. M., Kelly, R. M. and Zeikus, JG. (1995) Appl. Environ.
Microbial. 61, 1867-1875

grtsrvgré,6 S. H., Sj¢halm, C. and Kelly, R. M. (1993) Biotechnol. Bioeng. 41,
7 -

Menéndez-Arias, L. and Argos, P. (1989) J. Mal. Biol. 206, 397-406

Watanabe, K., Chishira, K, Kitamura, K. and Suzuki, Y. (1991) J. Biol. Chem.
266, 24287-24294

ngell, C.R., Przybyla, A. and Ljungdahl, L. G. (1990) Biochemistry 29, 5687-
5 4

Bahm, G. and Jaenicke, R. (1994) Int. J. Peptide Protein Res. 43, 97-106
Burdette, D. S., Vieille, C. and Zeikus, J. G. (1996) Biochem. J. 313 (in press)
Tomazic, S. J. and Klibanov, A. M. (1988) J. Biol. Chem. 263, 3086-3091

Fantana, A. (1990) in Life under extreme conditions: Biochemical adaptation, (di
Prisca, G., ed.), pp. 89-113. Springer-Verlag, Heidelberg

Meldgaard, M. and Svendsen, I. (1994) Microbial. 140, 159-166
Olsen, 0. and Thamsen, K K (1991) J. Gen. Microbial. 137, 579-585
Mirsky, A. E. and Pauling, L. (1936) Proc. Natl. Acad. Sci. USA 22, 439-447

Pfeil, W. (1986) in Thermodynamic data for biochemistry and biotechnology,
(Hinz, H.-J., ed.), pp. 349-376. Springer-Verlag, Heidelberg

1)

A.)

203.
204.
205.

206.

207.
208.
209.

210.

211.
212.
213.
214.

215.

216.

217.
218.

219.
220.

221.
222.
223.
224.

82

Jaenicke, R. (1991) Eur. J. Biochem. 202, 715-728
Ahem, T. J. and Klibanov, A. M. (1988) Meth. Biochem. Anal. 33, 91-127

Steigerwald, V. J., Beckler, G. S., and Reeve, J. N. (1990) J. Bacterial. 172,
4715-4718 ‘

Tchemajenka, V., Vieille, C., Burdette, D. S. and Zeikus, J. G. (1996) manuscript
in preparation

Londesborough, J. (1980) Eur. J. Biochem. 105, 211-215
Segel, I. H. (1975) Enzyme Kinetics. John Wiley & Sans, NY

Smith, C. A., Rangarajan, M. and Hartley, B. S. (1991) Biochem. J. 277, 255-
261

Sakamata, N ., Katre, A. M. and Savageau, M. A. (1975) J. Bacterial. 124, 775-
783

Caultan, J. W. and Kapoor, M. (1973) Can. J. Microbiol. 19, 439-450
Hooper, A. B., Hansen, J. and Bell, R. (1967) J. Biol. Chem. 242, 288-296
LéJahn, H. B., Suzuki, 1. and Wright, J. A. (1968) J. Biol. Chem. 243, 118-128

Hudson, R. C., Ruttersmith, L. D. and Daniel, R. M. (1993) Biochim. Biophys.
Acta 1202, 244-250

Glaser, P., Presecan, B., Delepierre, M., Surewicz, W. K., Mantsch, H. H.,
Bar-Lu, O. and Gilles, A.-M. (1992) Biochemistry 31, 3038-3043

Jencks, W. P. (1975) in Advances in Enzymalagy and related areas of molecular
biology vol. 43, (Meister, A., ed), pp. 219-410. John Wiley & Sans, NY

Rellas, P. and Scopes, R. K. (1994) Prat. Express. Purif. 5, 270-277

Nagara, T., Makira, Y., Yamamata, K., Urabe, I. and Okada , H. (1989) FEBS
lett. 253, 113-116

Kauzmann, W. (1959) Adv. Protein Chem. 14, 1-47

FrleiireésE” Haynie, D. T. and Xie, D. (1993) Proteins: Struct. Funct. Genet. 17,
1 -

Haynie, D. T. and Freire, E. (1993) Proteins: Struct. Funct. Genet. 16, 115-140
Haynie, D. T. and Freire, E. (1994) Biapalymers 34, 261-271

Privalov, P. L. and Gill, S. J. (1988) Adv. Protein Chem. 39, 191-234

Dill, K.A. ( 1990) Biochemistry 29, 7133-7155

225.

226.
227.

228.

229.

230.

231.
232.
233.

234.

235.

236.

237.

238.

239.
240.
241 .

242.

243.

244.

83

Dill, K. A., Alonso, D. O. V. and Hutchinson, K. (1989) Biochemistry 28, 5439-
5449

Creighton, T. E. (1990) Biochem. J. 270, 1-16

Creighton, TE. (1993) Proteins: structures and molecular properties 2nd edition.
W. H. Freeman and Company, NY

Tsubai, M., Ohta, S. and Nakanishi, M. (1978) in Biochemistry of thermaphily
(Friedman, M., ed.), pp. 251-266. Academic Press, NY

Wiitrich, K., Rader, H. and Wagner, G. (1980) in Protein folding, (Jaenicke, R.,
ed.), pp. 549-564. Elsevier/North Holland Biomedical Press, Amsterdam

Argos, P., Rossmann, M. G., Grau, U. M., Zuber, H., Frank, G. and Tratschin,
J. D. (1979) Biochemistry 18, 5698-5703

Matthews, B.W. (1993) Ann. Rev. Biochem. 62, 139-160
Zhang, X.-J., Baase, W. A. and Matthews, B. W. (1992) Protein Sci. 1, 761-776

Heming, T., Yutani, K., Inaka, K., Kuraki, R., Matsushima, M. and Kikuchi, M.
( 1992) Biochemistry 31, 7077-7085

Hurley, J. H., Baase, W. A. and Matthews, B. W. (1992) J. Mal. Biol. 224,
1143-1159

Watanabe, K., Masuda, T., Ohashi, H., Mihara, H. and Suzuki, Y. (1994) Eur. J.
Biochem. 226, 277-283

Erikssan, A. B., Baase, W. A., Zhang, X.-J., Heinz, D. W., Blaber, M.,
Baldwin, E. P. and Matthews, B. W. (1992) Science255, 178-183

Biro, J., Fabry, S., Dietrnaier, W., Bogedain, C. and Hensel, R. (1990) FEBS
Lett. 275, 130-134

Kaizumi, J.-I., Zhang, M., Imanaka, T. and Aiba, S. (1990) Appl. Environ.
Microbial. 56, 3612-3614

Matsumura, M., Yasumura, S. and Aiba, S. (1986) Nature 323, 356-3358
Shih, P. and J. F. Kirsch (1995) Prat. Sci. 4, 2050-2062
Shih, P. and J. F. Kirsch (1995) Prat. Sci. 4, 2063-2072

Kuraki, R., Inaka, K, Taniyama, Y., Kidakara, S.-I., Matsushima, M., Kikuchi,
M. and Yutani, K (1992) Biochemistry 31, 8323-8328

Heming, T., Yutani, K, Taniyama, Y. and Kikuchi, M. ( 1991) Biochemistry 30,
9882-9891

Bell, J. A., Becktel, W. J., Sauer, U., Baase, W. A. and Matthews, B. W. (1992)
Biochemistry 31, 3590-3596

 

 

 

,.p

f-_)

"J

245.
246.
247.

248.

249.

250.

251.

252.

253.

254.

255.

25 6.

257.

258.

259.

260.

261.
262.
263.

84
Pjura, P. and Matthews, B. W. (1993) Protein Sci. 2, 2226-2232

Matsumura, M., Signor, G. and Matthews, B. W. (1989) Nature 342, 291-293

Erikssan, A. B., Baase, W. A., and Matthews, B. W. (1993) J. Mol. Biol. 229,
747-769

Matthews, B. W., Nicholson, H. and Becktel, W. (1987) Proc. Natl. Acad. Sci.
USA 84. 6663-6667

Pjura, P., McIntosh, L. P., Wozniak, J. A. and Matthews, B. W. (1993) Proteins:
Struct. Funct. Genet. 15, 401-412

Kuraki, R., Taniyama, Y., Seka, C., Nakamura, H., Kikuchi, M. and Ikehara, M.
(1989) Proc. Natl. Acad. Sci. USA 86, 6903-6907

Blaber, M., Lindstram, J. D., Gassner, N., Xu, J ., Heinz, D. W. and Matthews,
B. W. (1993) Biochemistry 32, 11363-11373

Anderson, D. E., Hurley, J. H., Nicholson, H., Baase, W. A. and Matthews, B.
W. (1993) Protein Sci. 2, 1285-1290

Erikssan, A. E., Baase, W. A., Wozniak, J. A., and Matthews, B. W. (1992)
Nature 355, 371-373

Hardy, F., Vriend, G., Veltrnan, O. R., van der Vinne, B., Venema, G. and
Eijsink, V. G. H. (1993) FEBS Lett. 317, 89-92

Vriend, G., Berendsen, H. J. C., van der Zee, J. R., van den Burg, B., Venema,
G. and Eijsink, V. G. H. (1991) Protein Eng. 4, 941-945

Van den Burg, B., Dijkstra, B. W., Vriend, G., van der Vinne, B., Venema, G.
and Eijsink, V. G. H. (1994) Eur. J. Biochem. 220, 981-985

Ishikawa, K., Nakamura, H., Marikawa, K. and Kanaya, S. (1993) Biochemistry
32, 6171-6178

Kimura, S., Nakamura, H., Hashimato, T., Oabatake, M. and Kanaya, S. (1992)
J. Biol. Chem. 267, 21535-21542

éfizgmra, S., Kanaya, S. and Nakamura, H. (1992) J. Biol. Chem. 267, 22014-
17

Fabian, H., Schultz, C., Backmann, J., Hahn, U., Saenger, W., Mantsch, H. H.
and Naumann, D. (1994) Biochemistry 33, 10725-10730

Mitchinsan, C. and Wells, J. A. (1989) Biochemistry 28, 4807-4815
Wells, J. A. and Powers, D. B. (1986) J. Biol. Chem. 261, 6564-6570

Takagi, H., Takahashi, T., Mamase, H., Inauye, M., Maeda, Y., Matsuzawa, H.
and Ohta, T. (1990) J. Biol. Chem. 265, 6874-6878

 

7'11

“I
718
*1

279

264.

265.

266.

267.
268.

269.
270.

271.
272.
273.
274.
275.
276.

277.
278.
279.
280.

281.

282.

85

Quax, W. J ., Mrabet, N. T., Luiten, R. G. M., Schuurhuizen, P. W., Stanssens,
P. and Lasters, I. (1991) Bio/Technology 9, 738-742

Mrabet, N. T., Van den Braeck, A., Van den brande, I., Stanssens, P., Larache,
Y., Lambeir, A.-M., Matthijssens, G., Jenkins, J ., Chiadmi, M., van Tilbeurgh,
H., Rey, F., Janin, J., Quax, W. J., Lasters, 1., De Maeyer, M. and Wadak, S. J.
(1992) Biochemistry 31, 2239-2253

Meng, M., Bagdasarian, M. and Zeikus, J. G. (1993) Bio/Technology 11, 1157-
1161

Valkin, D. B. and Klibanov, A. M. (1987) J. Biol. Chem. 262, 2945-2950

Alber, T., Daa-pin, S., Nye, J. A., Muchmare, D. C. and Matthews, B. W.
(1987) Biochemistry 26, 3754-3758

MacArthur, M. W. and Thornton, J. M. (1991) J. Mal. Biol. 218, 3397-3412

Dixon, M. M., Nicholson, H, Shewchuk, L., Baase, W. A. and Matthews, B. W.
(1992) J. Mal. Biol. 227, 917-933

Perutz, M. F., Kendrew, J. C. and Watson, H. C. (1965) J. Mol. Biol. 13, 669-
678

El Hawrani, A. S., Moretan, K. M., Sessions, R. B., Clarke, A. R. and
Halbroak, J. J. (1994) Trends Biotechnol. 12, 207-211

Creighton, TE (1993) Proteins: structures and molecular properties 2nd ed., W.H.
Freeman and Company, NY

Hensel, R., Fabry, S., Biro, J., Bogedain, C., Jakob, I. and Siebers, B. (1994)
Biocatalysis 11, 151-164

Ishikawa, K., Okumura, M., Katayanagi, K., Kimura, S., Kanaya, S.,
Nakamura, H. and Marikawa, K. (1993) J. Mal. Biol. 230, 529-542

Davies, G. J., Gamblin, S. J., Littlechild, J. A. and Watson, H. C. (1993)
Proteins: Struct. Funct. Genet. 15, 283-289

Machius, M., Weingand, G. and Huber, R. (1995) J. Mol. Biol. 246, 545-559
Baker, E. N., and Hubbard, R. E. (1984) Prag. Biophys. Mol. Biol. 44, 97-179
Cleland, W. W. and Kreevoy, M. M. (1994) Science 264, 1887-1890

Hibberr, F. and Emsley, J. (1990) in Advances in Physical Organic Chemistry,
(Bethell, D., ed.), pp. 255-379. Academic Press Ltd., London

flagsumura, M., Becktel, W. J. and Matthews, B. W. ( 1988) Nature 334, 406-

Yutani, K., Ogasahara, K, Tsujita, T. and Sugina, Y. (1987) Proc. Natl. Acad.
Sci. USA 84, 4441-4444

283.
284.

285.
286.

287.
288.
289.

290.
291.
292.

293.

294.
295.
296.
297.
298.
298a.

299.

300.

86
Rashin, A. A., Iafin, M. and Hanig, B. (1986) Biochemistry 25, 3619-3625

McRee, D. E., Redford, S. M., Getzaff, E. D., Lepack, J. R., Hallewell, R. A.
and Tainer, J. A. (1990) J. Biol. Chem. 265, 14234-14241

Wilson, K P., Malcolm, B. A. and Matthews, B. W. (1992) J. Biol. Chem. 267 ,
10842-10849

Yamada, H., Kanaya, B., Inaka, K., Ueno, Y., Ikehara, M. and Kikuchi, M.
(1994) Biol. Pharm. Bull. 17, 192-196

Kanaya, S. and Itaya, M. (1992.) J. Biol. Chem. 267, 10184-10192
Sandberg, W.S. and Terwilliger, T. C. (1989) Science 245, 54-57

Baldwin, E. P., Hajiscyedjavadi, 0., Base, W. A., and Matthews, B. W. (1993)
Science 262, 1715-1718

Kwan, K.-S., Kim, J., Shin, H. S. and Yu, M.-H. (1994) J. Biol. Chem. 269,
9627-9631

Heinz, D. W., Baase, W. A. and Matthews, B. W. (1992) Proc. Natl. Acad. Sci.
USA 89, 3751-3755

Sicard, P. J., Leleu, J.-B., Duflat, P., Drocaurt, D., Martin, F., Tiraby, G.,
Petsko, G. and Glasfeld, A. (1990) Ann. N.Y. Acad. Sci. 613, 371-375

Hallewell, R. A., Imlay, K. C., Lee, R, Fang, N. M., Gallegos, C., Getzaff, E.
D., Tainer, J. A., Cabelli, D. E., Tekamp-Olsan, P., Mullenbach, G. T. and
Couscns, L. S. (1991) Biochem. Biophys. Res. Comm. 181, 474-480

Lepack, J. R., Frey, H. E. and Hallewell, R. A. (1990) J. Biol. Chem. 265,
21612-21618

Zale, S. E. and Klibanov, A. M. (1986) Biochemistry 25, 5432-5444

Valkin, D. B. and Middaugh, C. R. (1992) in Stability of protein pharmaceuticals,
Part A: Chemical and physical pathways of protein degradation, (Ahem, TJ. and
Manning, M.C., eds.), pp. 215-247. Plenum Press, NY

Tomazic, S. J. and Klibanov, A. M. (1988) J. Biol. Chem. 263, 3092-3096

Kuraki, R., Kawakita, S., Nakamura, H. and Yutani, K. (1992) Proc. Natl. Acad.
Sci. USA 89, 6803-6807

Gerwig, G. J., Kamerling, J. P., Vliegenthart, J. F. G., Marag, E., Lamed, R.
and Bayer, E. A. (1991) Eur. J. Biochem. 196, 115-122

17,25,7215" Lowe, S. E. and Zeikus, J. G. (1993) Appl. Environ. Microbial. 59,

Ng, T. K. and Zeikus, J. G. (1981) Biochem. J. 199, 341-35

301.

302.

303.

304.

305.
306.

308.
309.

310.

311.

312.

313.

313a.
313b.
306.

307.

314.

315.

316.

87

Chen, B.-l and King, J. (1991) in Protein Refalding, (Georgian, G. and De
Bemardez-Clark, B., eds.), pp. 119-132. American Chemical Society,
Washington, DC

Bael, E., Brady, L., Brzazawksi, A. M., Derewenda, Z., Dodson, G. G., Jensen,
V. J ., Petersen, S. B., Swift, H., Thim, L. and Waldike, H. F. (1990)
Biochemistry 29, 6244-6249

Marg, G. A. and Clark, D. S. (1990) Enzyme Microbial. Technol. 12, 367-373

Whitlow, M., Howard, A. J ., Finzel, B. C., Paulas, T. L., Winborne, E. and
Gillialand, G. L. (1991) Proteins Struct. Funct. Genet. 9, 153-173

Kasumi, T., Hayashai, K. and Tsumura, N. (1982) Agric. Biol. Chem. 46, 21-30

Stigter, D., Alonso, D. O. V., and Dill, K. A. (1991) Proc. Natl. Acad. Sci. USA
88, 4176-4180

Hei, D. J .and Clark, D. S. (1994) Appl. Environ. Microbial. 60, 932-939

Miller, J. R, Nelson, C. M., Ludlaw, J. M., Shah, N. N. and Clark, D. S. (1989)
Biotechnol. Bioeng. 34, 1015-1021

Ltithi, E., Reif, K., Jasmat, N. B. and Bergquist, P. L. (1992) Appl. Microbial.
Biotechnol. 36, 503-506

Wilks, H. M., Hart, K. W., Feeney, R., Dunn, C. R., Muirhead, H., Chia, W.
N., Barstaw, D. A., Atkinson, T., Clarke, A. R. and Halbroak, J. J. (1988)
Science 242, 1541-1544

Wilks, H. M., Halsall, D. J., Atkinson, T., Chia, W. N., Clarke, A. R. and
Halbroak, J. J. (1990) Biochemistry 29, 8587-8591

Meng, M., Lee, C., Bagdasarian, M. and Zeikus, J. G. (1991) Proc. Natl. Acad.
Sci. USA 88, 4015-4019

Arnold, EH. (1993) FASEB J. 7, 744-749
Chen, K. and Arnold, F. H. (1991) Bio/Technology 9, 1073-1077

De Cardt, S., Hendrickx, M., Maesmans, G. and Tabback, P. (1994) Biotech.
Bioeng. 43, 107-114

Oshima, T. (1986) in Thermophiles: General, molecular, and applied microbiology
(Brock, T. D., ed.) pp. 137-158. John Wiley & Sans, NY

Liza, £10, McKenzie, T. and Hageman, R. (1986) Proc. Natl. Acad. Sci. USA 83,
57 5

Bai, Y., Milne, J. S., Mayne, L. and Englander, S. W. (1994) Proteins: Struct.
Funct. Genet. 20, 4-14

Muheim, A., Todd, R. J., Casimira, D. R., Gray, H. B. and Arnold, EH. (1993)
J. Am. Chem. Soc. 115, 5312-5313

317.
318.

319.

320.

321.

322.

323.

324.
325.

88

Woese, C. R. (1994) Microbial. Rev. 58, 1-9

Keinan, E., Hafeli, E. K., Seth, K. K. and Lamed, R. L. (1986) J. Am. Chem.
Soc. 108, 162-169

Pham, V. T., Phillips, R. S. and Ljungdahl, L. G. (1989) J. Amer. Chem Soc.
111, 1935-1936

Periera, D. A., Gerson, F. P. and Oestreicher, E. G. (1994) J. Biotechnol. 34, 43-
50

Keinan, E., Seth, K. K. and Lamed, R. (1986) J. Am. Chem. Soc. 108, 3474-
3480

Keinan, E., Seth, K. K., Lamed, R., Ghirlanda, R. and Singh, S. P. (1990)
Biocatalysis 4, 1-15

Lamed, R. J., Keinan, E. and Zeikus, J. G. (1981) Enz. Microbial. Technol. 3,
144-148

Aaltanen, O. and Rantakyla, M. (1991) Chemtech 1, 240-248
Lee, K. M. and Biellmann, J.-F. (1987) Nuav. J. Chim. 11, 1-4

Chapter II
Cloning and Expression of the Gene Encoding the
Thermoanaerobacter ethanolicus 39E Secondary-Alcohol

Dehydrogenase and Enzyme Biochemical Characterization

Accepted for publication in The Biochemical Jamnal

89

90
ABSTRACT

The ath gene encoding Thermoanaerobacter ethanolicus 39E secondary-alcohol
dehydrogenase (2° ADH) was cloned, sequenced, and expressed in Escherichia coli. The
1056 bp gene encodes a hamatetrameric recombinant enzyme consisting of 37.7 kDa
subunits. The purified recombinant enzyme is optimally active above 90°C with a half-life
of approximately 1.7 h at 90°C. An NADP(H)-dependent enzyme, the recombinant 2°
ADH has 170-fold greater catalytic efficiency in propan-2-ol versus ethanol oxidation. The
enzyme was inactivated by chemical modification using dithionitrobenzoate (DTNB) and
diethylpyrocarbonate, indicating that Cys and His residues are involved in catalysis. Zinc
was the only metal enhancing 2° ADH reactivation after DTNB modification, implicating
the involvement of a bound zinc in catalysis. Arrhenius plots for the oxidation of propan-2-
01 by the native and recombinant 2° ADHs were linear from 25°C to 90°C when the
enzymes were incubated at 55°C prior to assay. Discontinuities in the Arrhenius plots for
propan-2-ol and ethanol oxidations were observed, however, when the enzymes were
preincubated at 0°C or 25 °C. The observed Arrhenius discontinuity, therefore, resulwd
from a temperature dependent, catalytically significant 2° ADH structural change.
Hydrophobic Cluster Analysis comparisons of both mesophilic versus thermophilic 2°
ADH and 1° versus 2° ADH amino acid sequences were performed. These comparisons
predicted that specific proline residues may contribute to 2° ADH thermostability and
thermophilicity, and that the catalytic Zn ligands are different in 1° ADI-Is (two Cys and a
His) and 2° ADHs (Cys, His, and Asp).

91
INTRODUCTION

Alcohol dehydrogenases (ADHs) (EC 1.1.1.1 [NADH] or EC 1.1.1.2 [NADPHD are well
studied as a structurally conserved class of enzyme [1]. The X-ray structure of the horse
liver primary-alcohol dehydrogenase (1° ADH) is known, and the properties of this enzyme
have been extensively detailed [1,2]. ADHs are typically dimeric or tetrameric pyridine
dinucleotide-dependent metalloenzymes with a zinc atom involved in catalysis. ADHs have
been classified as 1° or 2° ADHs based on their relative activities toward 1° and 2° alcohols.
It generally has been assumed that 1° and 2° ADHs are structm‘ally similar, and that their
substrate differences are due to relatively small changes in their active site architecture.
Tetrameric 2° ADI-Is have been repormd from a number of microorganisms [3-7]. The
Thermoanaerobacter ethanolicus 39E 2° ADH is a bifunctional ADH/acetleoA reductive
thioesterase [7]. It has been proposed to function physiologically by oxidizing
nicotinamide cofactor during ethanol formation, indirectly preventing glycolytic inhibition
at the glyceraldehyde dehydrogenase step [8].

2° ADHs are attractive subjects for research into chiral chemical production because
of their broad specificities and their highly enantiospecific conversion of prachiral ketones
to alcohols [9-12]. Two issues facing the commercial scale application of 2° ADHs in
chiral syntheses are the difficulty of regenerating and retaining expensive nicotinamide
cofactor and the lack of inexpensive, highly stable enzymes. Cofactor regeneration and
retention have been overcome by numerous strategies [13,14]. While thermophilic
enzymes are generally much more stable than their mesophilic counterparts, the organisms
that produce thermophilic 2° ADHs grow slowly and to low cell densities, making an
alternative expression system for these enzymes crucial for both their commercial
application and detailed protein structure-function studies.

The discovery of thermophilic bacteria has provided the opportunity to isolate
therrrrostable enzymes directly. The thermophilic anaerobes T. ethanolicus and

92
Thennoanaerobacter brocla'i [15] express extremely enantiospecific 2° ADI-ls that are stable

above 70°C. These two 2° ADHs have been proposed to be extremely structrn'ally similar
based on similar molecular weight and kinetic characteristics [16]. While the Clostridiwn
beijen‘nckii gene (ath) encoding a mesophilic NADP(H)-dependent 2° ADH has been
cloned and sequenced (Genbank, Acc. no. M84723), and while the amino acid sequence of
the T. brockii 2° ADH has been determined by Edman degradation [17], no cloned
thermophilic 2° ADH is available for detailed biochemical studies.

The work reported here describes the cloning, sequencing, and expression of the
gene encoding the thermophilic T. ethanolicus 2° ADH. The kinetic and thermal properties
of the recombinant enzyme are determined. The biochemical basis for enzyme
discontinuous Arrhenius plots is investigawd Finally, the protein sequence information is
used to examine the structural similarities between thermophilic and mesophilic 1° ADHs
and 2° ADHs. These comparisons are used to predict the 2° ADH catalytic Zn liganding
residues and the potential involvement of specific prolines in 2° ADH thermal stability.

MATERIALS AND METHODS

Chemicals and reagents

All chemicals were of at least reagent/molecular biology grade. Gases were provided by
AGA specialty gases (Cleveland, OH), and made anaerobic by passage through heated
copper filings. Oligonucleatide syntheses and amino acid sequence analyses were
performed by the Macromolecular Structure Facility (Department of Biochemistry,
Michigan State University). The kanamycin resistance GenBlock (EcoRI) DNA cartridge
used in expression vector construction was purchased from Pharmacia (U PPsala, Sweden)
[1 8] .

93
Media and strains

T. ethanolicus 39E (ATCC #33223) was grown in TYE medium as previously described
[19]. All batch cultures were grown anaerobically under an N2 headspace. Escherichia
coli DH50t containing the recombinant (1th gene was grown in rich complex medium (20
g 1'1 tryptone, 10 g 1'1 yeast extract, 5 g 1'1 NaCl) at 37°C in the presence of 25 pg ml-l
kanamycin and 100 ug ml'l ampicillin.

DNA manipulations and library construction

Plasmid DNA purification, restriction analysis, PCR, and colony and DNA hybridizations
were performed using conventional techniques [20,21]. T. ethanolicus chromosomal
DNA was purified as previously described [22]. Partially digested 2—5 kb Sau3A I
fragments were isolated by size fractionation from a 10-40% sucrose gradient [20] and
ligated into pUC18 BamH I/BAP (Pharmacia, Uppsala, Sweden). The ligation mixture was
transformed into E. coli DHSa by electroporation [21]. Degenerate primers (1) (5'-
ATGAA(R)GG(N)T1‘(H)GC(N) ATG(Y)T) and (2) (5'-
G(W)(N)GTCATCAT(R)TC(N)G(K)(D)ATCAT) were used to synthesize the
homologous probe for colony hybridization [23]. The DNA fragment containing the ath
gene was sequenced using the method of Sanger et al. [24].

Enzyme purification

The native 2° ADH was prnified by established techniques [7]. The recombinant enzyme
was pmified from E. coli DHSa aerobically. The pelleted cells from batch cultures were
resuspended (0.5 g wet wt. ml°1) in 50 mM Tris:HCl [pH 8.0] (buffer A) containing 5 mM
DTT, and 10 mM ZnClz, and lysed by passage through a French pressure cell. The
clarified lysate was incubated at 65°C for 25 min and then centrifuged for 30 min at
15000g. The supernatant was applied to a DEAE-sephacryl column (2.5 cm x 15 cm) that
was equilibrated with buffer A and eluted using a 250 ml NaCl gradient (0-300 mM).

94
Active fractions were diluted 4—fold in buffer A and applied to a Q-sepharose column (2.5

cm x 10 cm) equilibrated with buffer A. Purified enzyme was eluted using a 250 ml NaCl
gradient (0-300 mM).

Molecular mass determination

Recombinant holoenzyme molecular mass was determined by comparison with protein
standards (Sigma; St. Louis, MO) using gel filtration chromatography (0.5 ml min'l) with
a Pharmacia S300 column (110 cm x 1.2 cm) equilibrated with buffer A containing 200
mM NaCl. Subunit molecular mass values were determined by matrix associated laser
desorption ionization mass spectrometry (MALDI) at the Michigan State University Mass
Spectrometry Facility.

Kinetics and thermal stability
The standard 2° ADH activity assay was defined as NADP" reduction coupled to prapan-2-
ol oxidation at 60°C as previously described [7]. The enzyme was incubated at 55°C for 15
min prior to activity determination unless otherwise indicated. Assays to determine Kmllpp
and Vmaxapp were conducted at 60°C with substrate concentrations between 20meapp and
02me,“. Kinetic parameters were calculated from nonlinear best fits of the data to the
Michaelis-Menten equation using Kinzyme software [26]. Protein concentrations were
measured using the bicinchoninic acid procedure (Pierce; Rockford, IL) [27].

2° ADH thermostability was measured as the residual activity after timed incubation
at the desired temperatures. Thermal inactivation was stopped by incubation for 30 min at

25°C, and samples were prepared for activity assays by preincubation at 55°C for 15 min.
Incubations were performed in 100 [.11 PCR tubes (cat. #72.733.050, Sarstedt; Newton,

NC) using 200 pg ml'l protein in 100 111 buffer A. Activity was determined using the
unfractionated samples. The temperature effect on enzyme activity was studied using the

substrates propan-2-ol or ethanol at 10meapp concentrations.

95
To study the effect of enzyme preincubation temperature on enzyme reaction rates,

the 2° ADH was incubated at 0°C, 25°C, or 55 °C for 15 min prior to activity determinations
in the temperature range 30—90°C. TriszHCl buffer pH values were adjusted at 25 °C to be
pH 8.0 at the temperature they were used (thermal correction factor = -0.031 ApH °C'1).
The statistical significance of the differences between the Arrhenius plot slapes above and

below the discontinuity temperatures was determined by covariance analysis [25].

Chemical modification

Cysteine residues were reversibly modified using dithionitrobenzoate (DTNB) at 25 °C and
60°C [28]. DTNB-inactivated 2° ADH was reactivated using DTT in the presence of 0.01
mM to 1.0 mM metal salts or 0.5 mM to 3.0 mM EDTA. Histidine residues were
chemically modified with diethylpyrocarbonate (DEPC) by incubation with 20 mM or 40
mM DEPC in 50 mM phosphate buffer (pH 6.0) at 25°C for 1.0 h and the reaction was
quenched by addition of 0.5 x volume of 0.5 M imidazole (pH 6.5) [29].

Protein sequence comparisons

The peptide sequences of the Bacillus stearorhermophilus (acc. #D90421), Sulfolobus
solfataricus (acc. #851211), and Zymomonas mobilis (acc. #M32100) 1° ADHs and of the
Alcaligenes eutrophus (acc. #J03362) and Clostridium beijerinckii (acc. #M84723) 2°
ADHs were obtained from GenBank. The horse liver 1° ADH [1] and the T. brocla'i 2°
ADH [17] peptide sequences were obtained from the literature. Access to GenBank,
standard sequence alignments, and percentages of amino acid similarity/identity were
performed using the Program Manual for the Wisconsin Package, Version 8, Sept. 1994
(Genetics Computer Group; Madison, WI). Protein sequence alignments were performed
using Hydrophobic cluster analysis (HCA) [30]. HCA plots of individual protein
sequences were generated using HCA-Plot V2 computer software (Doriane; Le Chesnay,

France).

96

Nucleotide sequence accession number

The GenBank accession number for the sequence published in this article is U49975.

RESULTS

Cloning and sequencing of the T. ethanolicus ath gene

The T. ethanolicus 2° ADH was cloned from a T. ethanolicus chromosomal DNA library by
homologous hybridization. The N-terminal sequence of the native T. ethanolicus 2° ADH
(MKGFAML) was identical to those of C. beijerinckii and T. brockii 2° ADHs. This
sequence was used to generate primer (1). Alignment of the C. beijerinckii and T. brockii
2° ADH peptide sequences indicated another conserved region (residues 147-153) that
would reverse translate into a low-degeneracy Oligonucleatide (primer [2]). The PCR
product obtained using primers (1) and (2), and using T. ethanolicus chromosomal DNA as
the template was 470 bp, as expected from the position of primers (1) and (2) in the C.
Beijerinclcii and T. brockii 2° ADHs. This PCR product was used as a homologous probe
to screen the T. ethanolicus genomic library. The positive clone showing the highest 2°
ADH activity at 60°C was selected for frn'ther studies. Plasmid pADHBZS-C contained a
1.6 kb Sau3A I insert and was shown by subsequent sequencing and peptide analyses to
carry the complete ath gene. The 1.6 kb insert was subcloned into the
pBluescriptIlKS(+) Xba I site to construct the expression plasmid pADHB25. The
physical map of pADHB25 is shown in Figure 1. Plasmid pADI-IB25 was stabilized by
insertion of a kanamycin resistance cartridge into the vector EcoR I site, allowing dual
selection on kanamycin and ampicillin. This final construct (pADHB25-kan) was used for
all subsequent work.

97
The nucleotide sequence of the pADI-IB25 insert is shown in Figure 2. A unique

open reading frame (ORF) was identified that encoded a polypeptide which was highly
homologous to C. beijerinckii 2° ADH, and which started with the N-terminal sequence of
the native T. ethanolicus 2° ADH. Two consecutive ATG codons were identified as
potential translation initiation codons. The N-terminus of the native 2° ADH starts with a
single Met, suggesting that the first ATG codon is not translated or is post-translationally
removed. A potential ribosome binding site (RBS) (positions 223-228) is located 10 bp
upstream of the start codon. A potential promoter was identified approximately 70—100 bp
upstream of the RBS. The "-35" and "-10" regions are highly similar to the E. coli
consensus promoter sequences [31], and are separated by 16 bp. The "-10" region is
duplicated with the second copy overlapping the first. Because of its improper distance
from the "-35" region, this second copy may only provide an A+T rich sequence to aid in
strand separation. N a transcriptional stop site was identified downstream of ath. In this
region, instead, a truncated ORF, preceded by an RBS, was identified. It translated into a
45 amino acid peptide fragment 46% identical and 68% similar to the product of a similar
truncated ORF located downstream of C. beijerinckii ath gene. Little similarity was

found with other sequences in GenBank, so the function of this ORF remains unknown.

Sequence comparison of 1° and 2° ADHs

Standard alignments of 1° and 2° ADH amino acid sequences indicated a high level of
similarity among the three 2° ADI-Is from obligate anaerobes (Table 1). The T. ethanolicus
2° ADH differed from the T. brocla'i enzyme by only three residues, and was 75% identical
to the enzyme from the mesophile C. beijerinckii. The similarity was lower for
comparisons with the 2° ADH from the obligate aerobe A. cutrophus. The 1° ADHs
showed less sequence conservation (25 to 54% identity and 49 to 71% similarity) than the
2° ADHs, and showed only 20 to 27% identity (and 48 to 51% similarity) to the 2° ADHs.

Based on these standard alignments, the conservation of important core domain residues,

98

Figure 1. Restriction map of the T. ethanolicus 39E 2° ADH clone
(pADHB25). The flanking restriction sites are from the plasmid polylinker.

 

0 kb 0.5 kb 1.0 kb 1.5 kb

 

 

 

 

 

 

 

 

EcoR I} ‘Hind 111
Pst I EcoR V
KpnI / ‘ Pstl
Ss I
\ Del 14ch P \Ach .
Xbal EcoRV

 

100

Figure 2. Nucleotide sequence and deduced amino acid sequence of T.
ethanolicus 39E ath and of the downstream open reading frame. The
putative promoter -35 and -10 regions and RBSs are underlined. The ath stop cation is
indicated by three asterisks.

 

of T.

WM

101

I I 90
TGAACAATAGACAACCCCTTTCTGTGATCTTGTTTTTTGCAAATGCTATTTTATCACAAGAGATTTCTCTAGTTCTTTTTTACTTAAAAA
I I I I I I I I 180
AACCCTACGAAATTTTAAACTATGTCGAATAAATTATIGAIAATTTTTAACTATGTGCIAIIATATTATTGCAAAAAATTTAACAATCAT
_ -1
I I I 3? I I o I I 270
CGCGTAAGCTAGTTTTCACATTAATGACTTACCCAGTATTTTAGGAGGTGTTTAATGATGAAAGGTTTTGCAATGCTCAGTATCGGTAAA
RBS M K G F A M L S I G K
I I I I I I I I 360

GTTGGCTGGATTGAGAAGGAAAAGCCTGCTCCTGGCCCATTTGATGCTATTGTAAGACCTCTAGCTGTGGCCCCTTGCACTTCGGACATT
V G W I B K E K P A P G P F D A I V R P L A V A P C T S D I

450
CATACCGTTTTTGAAGGAGCCATTGGCGAAAGAATAACATGATACTCGGTCACCGAAGCTGTAGGTGAAGTAGTTGAAGTAGGTAGTGAG

H T V P E G A I G E R H N M I L G H E A V G E V V E V G S E

I I I I I I I I 540
GTAAAAGATTTTAAACCTGGTGATCGCGTTGTTGTGCCAGCTATTACCCCTGATTGGTGGACCTCTGAAGTACAAAGAGGATATCACCAG

V K D P K P G D R V V V P A I T P D N W T S E V O R G Y R Q

I I I I I I I I 630
CACTCCGGTGGAATGCTGGCAGGCTGGAAATTTTCGAATGTAAAAGATGGTGTTTTTGGTGAATTTTTTCATGTGAATGATGCTGATATG

H S G G H L A G N K F S N V K D G V F G E F F H V N D A D M

I I I I I I I . I 720
AATTTAGCACATCTGCCTAAAGAAATTCCATTGGAAGCTGCAGTTATGATTCCCGATATGATGACCACTGGTTTTCACGGAGCTGAACTG

N L A H L P K E I P L E A A V M I P D M M T T G P H G A B L.

I I I I I I I I 810
GCAGATATAGAATTAGGTGCGACGGTAGCAGTTTTGGGTATTGGCCCAGTAGGTCTTATGGCAGTCGCTGGTGCCAAATTGCGTGGAGCC

A D I E L G A T V A V L G I G P V G L M A V A G A K L R G A

I I I I I I I I 900
GGAAGAATTATTGCOGTAGGCAGTAGACCAGTTTGTGTAGATGCTGCAAAATACTATGGAGCTACTGATATTGTAAACTATAAAGATGGT
G R I I A V G S R P V C V D _A A K Y Y G A T D I V N Y K D G

I I I I I I I I 990
CCTATCGAAAGTCAGATTATGAATCTAACTGAAGGCAAAGGTGTCGATGCTGCCATCATCGCTGGAGGAAATGATGACATTATGGCTACA

P I E S Q I H N L T E G K G V D A A I I A G G N A D I M A T

I I I I I I I I 1080
GCAGTTAAGATTGTTAAACCTGGTGGCACCATCGCTAATGTAAATTATTTTGGCGAAGGAGAGGTTTTGCCTGTTCCTCGTCTTGAATGG

A V K I V R P G G T I A N V N Y F G E G B V L P V P R L E I

I I I I I I I I 1170
TGGCTCATAAAACTATAAAAGGCGGGCTATGTTCCGGTGGACGTCTAAGAATGGAAAGACTGATTGACCTTGTTTTTTAT

G C G H A R K T I K G G L C P G G R L R M B R L I D L V F Y

I I I I I I I I 1260
AAGCCTGTCGATCCTTCTAAGCTCGTCACTCACGTTTTCCAGGGATTTGACAATATTGAAAAAGCCTTTATGTTGATGAAAGACAAACCA

K P V D P S K L V T H V P Q G F D N I B R A P M L M R D K P

I I I I I I I I 1350
AAAGACCTTATCAAACCTGTTGTAATATTAGCAIAAAAATGGGGACTTAGTCCATTTTTATGCTAATAAGGCTAAATACACTGGTTTTTT
K D L I K P V V I L A ***

I I I I I I I I 14‘0
TATATGACACATCGGCCAGTAAACTCTTGGTAAAAAAATAACAAAAAATAGTTATTTTCTTAACATTTTTACGCCATTAACACTTGATAA

I I I I I I I I 1530
CATCATOGAAGAAGTAAATAAACAACTATTAAATAAAAGAAGAAGGAGGATTATCATGTTCAAAATTTTAGAAAAAAGAGAATTGGCACC
M F K I L E K R E L A P
t r r r t I r t 1620
TTCCATCAAGTTGTTTGTAATAGAGGCACCACTAGTAGCCAAAAAAGCAAGGCCAGGCCAATTCGTTATGCTAAGGATAAAAGAAGGAGG
S I K L P V I B A P L V A K K A R P G N F V M L R I K B G G

I

AGAAAGAATT 1630
E R I

—fl#" “um" I I

 

102

as. emeoofiefi u...

.55 Seen unz— e5 3235 .N £9: .N goreaboe. .m

fie. .m ”QEQQEbSESa .m Home. .m Sacha» .< S.» .< "agreemed "N3 .0 "383.563 H "wan esoaeeefie H ”5»

H G .8 commeofiomea $5235 9% e8 novice; .23 888E new UUG 05 .5 moan—3:8 203 «outage v5 83:82 agate

 

 

 

8e 35 2m son 33 see ”.2 came in one in $.49 3a 8.3 new oi
2: one on... 3:33 3.5 93 ewe new 933 “.3 83:3 oosN
2: 8.3 one 8.3 «.3 see in 8.5 eon 6.5 SN 3... .e...
8“ one in Goo ea 83 SN Goo SN 8:. .m...
ESQ.
8e Goo «.8 one «on 58 eon .3 .<
8“ see one see 4.: E .o
8“ $59 go one 6.
2: so .5...
g
be: 82 .N 3. .m one .m 3» .< E .o . one .5 so .5 fig.

 

 

"me—€338 05 Set 805.58 55 85.68 we Swain—mm .5 3:83 am

 

8353633 3:83 3:803 93 3.»: 88.. e833 bean—:8 3583» SEE we canton—=00 A o—efl.

103

Figure 3. HCA comparison of thermophilic and mesophilic 1° and 2° ADHs
centered on the putative nicotinamide binding motifs. The proposed catalytic Zn
liganding (O), structural Zn liganding Cl), and Rossmann fold binding motif 0) residues
are indicated. Symbols:*: proline”; glycine;l: serine; n: threonine.

 

104

    

UJD" I 8W5.

A
o N
KAf t I
l D K
N Ev
A 0'85
O N.
D
K LK

I. ethanolicus 39E 2° Adh

 

Q. beijerinckii 2° Adh

 

 

_B. stearothermophilus 1° Adh

horse liver 1° Adh

   

105
amino acids lining the active site pocket, and structurally conserved glycines identified for

the horse liver enzyme [1] was higher in all the peptides.

HCA comparisons allow the alignment of potentially similar 2° or 3° structural
regions between enzymes with dissimilar amino acid sequences. The Rossmann fold
consensus sequences {Gly-Xaa-Gly-Xaa-Xaa-Gly-(Xaa)13.20[negatively charged amino
acid for NAD(H) dependent or neutral amino acid for NAD(P)(H)]} [32] identified in each
of the ADH peptides were used to direct HCA alignments of 1° and 2° ADHs. Figure 3
shows a representative HCA multiple comparison between a thermophilic (T. ethanolicus )
and a mesophilic (C. beijerinckit) 2° ADH with a thermophilic (B. stearothermophilus)
and a mesophilic (horse liver) 1° ADH based on the identified Rossmann fold consensus
sequences. The consensus motifs of the N ADP(H) linked T. ethanolicus and C.
beiierinckii 2° ADHs were identified as Gly174 -Xaa-Gly175-Xaa-Xaa-Gly179, Gly193, and
those of the NAD(H) dependent B. stearothermophilus and horse liver 1° ADH were
identified as Gly172 - Xaa - Gly174-Xaa-Xaa-Gly177, Asp195 and Gly199-Xaa-Gly201-Xaa-
Xaa-Gly“, Asp223, respectively. This alignment also predicts 71 to 94% overall
similarity between 1° and 2° ADI-Is for the critical hydrophobic core residues (71 to 94%),
active site pocket residues (71 to 100%), and structurally critical glycines (100% in all
cases) identified in the horse liver enzyme.

The HCA comparison allowed the identification of corresponding cluster regions
and catalytically important residues in all fatn' dehydrogenases based on the corresponding
residues identified for the horse liver enzyme [1,2]. The horse liver 1° ADH contains two
Zn atoms per subunit, one structural and one catalytic [1]. The thermophilic B.

Stearothermophilus 1° ADH contained a region involving cysteine residues 97, 100, 103,
and 111 that is analogous to the structural Zn binding loop in the horse liver ADH
(involving Cys92, Cys95, Cys93, and Cy8105). However, no analogous structural Zn
binding loop regions were identified in the T. ethanolicus and C. beijerinckii 2° ADHs (Fig.
3). The catalytic Zn ligands in the horse liver enzyme have been established (Cys45, His53,

106
and Cysl74). The N-terminal cysteine and histidine ligands to the catalytic Zn atom appear

to be conserved in both the 1° and 2° ADH sequences (Fig. 3). However, the second

cysteine ligand in the horse liver 1° ADH (Cys174), conserved in the B. stearothermophilus

1 ° ADH(Cys143), is substituted with an aspartate residue (Asp150) in both the T.
ethanolicus and C. beijerinckii proteins. The A. eutrophus and T. brockii 2° ADHs also
conserved corresponding aspartate residues and appeared to lack 1° ADH-like structural Zn

binding loops (data not shown).

Sequence comparison of 2° ADH enzymes from a mesophile and

th ermophiles

The structln'al constraints introduced by proline residues have been proposed as a
Incchanism involved in protein thermostabilization [33]. Among the twelve
nonconservative sequence substitutions between the mesophilic C. beijerinckii 2° ADH and
the thermophilic T. ethanolicus 2° ADH, nine correspond to the introduction of prolines
(22, 24, 149, 177, 222, 275, 313, 316, and 347) in the T. ethanolicus protein. All but
P103 13 are also present in the thermophilic T. brockii 2° ADH. The mesophilic 2° ADH
subunit contained a total of 13 prolines (3.7%) while the T. ethanolicus and T. brockii
subunits contained 22 prolines (6.2%) and 21 prolines (6.0%), respectively. Prolines 20,
22. and 24 are in a 9 residue stretch of hydrophilic amino acids near the putative catalytic
2“ 1igand Cys37. Pro”9 interrupts a hydrophobic cluster and is next to another putative Zn
ligand, Asp150. The nicotinamide cofactor binding motif includes Pro177 which also
interrupts a hydrophobic cluster region. Prolines 222, 275, 313, 316, and 347 are located

In Shon (two to four residue) hydrophilic stretches that form putative turn regions.

PI“‘if‘ication and characterization of the recombinant 2° ADH
The recombinant r. ethanolicus 2° ADH was highly expressed in E. coli in the absence of

1ndfiction and no significant increase was seen upon induction with 5.0 mM isapropylthio-

107
b-D—galactoside. The enzyme expression level was similar in E. coli and in the native

organism (1% to 5% of total protein). The recombinant enzyme was purified 36-fold to
homogeneity (as determined by the presence of a single band on SDS-PAGE). The subunit
molecular masses were calculated from the average of 3 (native enzyme). and 5
(recombinant enzyme) determinations using MALDI, yielding masses for the native and
recombinant enzyme subunits of 37707 Da and 37854 Da, respectively. These values are
within the generally accepted error of the technique (~1%), and are in agreement with the
theoretical molecular mass for the native enzyme based on the gene sequence (37 644 Da).
N —terminal amino acid analysis of the recombinant enzyme indicated that in 72 i 5% of the
recombinant protein the two N-terminal ATG codons had been translated, while 28 i 2%
of the protein contained a single N-terminal Met residue, like the native enzyme. The
increased mass of 147 Da determined for the recombinant enzyme is consistent with the
mass of one Met residue (149.2 Da), although this difference is small compared to the
measurement error associated with MALDI. The recombinant holoprotein molecular mass
was determined to be 160 kDa by gel filtration chromatography demonstrating that the
recombinant T. ethanolicus 2° ADH, as the native enzyme, is a homotetramer.
The 2° ADH was completely inactivawd at both 25°C and 60°C by cysteine-specific
DTNB modification. Inactivation was partially reversed at 60°C by the addition of DTT
allowing the recovery of 34% initial activity (initial activity = 54 1 3.0 units mgl). The
addition of CdSO4 , FeClz, MnClz, CaClz, MgClz, NaCl (0.01 mM to 1.0 mM), or EDTA
(0:5 mM to 3.0 mM) did not affect enzyme reactivation by DTT, and the addition of CoC12
01’ Nicr; reduced the recovered activity to only 10% or 15% of the initial activity,
respectively. Zinc was the only metal to enhance the reactivation (up to 48% activity
rt"unwary in the presence of 100 M ZnClz). DEPC modification of histidine residues also

completely inactivated the enzyme.

108
Characterization of thermal and kinetic properties of the recombinant T.

ethanolicus 2° ADH
The activity of the recombinant 2° ADH toward ADH substrates was characterized. The
Vmaxapp for propan-2-ol (68 units mg‘l) was 3.6-fold higher than for ethanol (19 units
mg‘l), and the Kmapp toward the 1° alcohol (53 mM) was almost 50-fold higher than for
the 2° alcohol (1.1 mM). The catalytic efficiency of the recombinant enzyme was
determined to be approximately 170-fold greater toward the 2° alcohol (0.062 ml min'1 mg“
1) than toward the 1° alcohol (0.00036 ml min'1 mg'l). The Kmapp for NADP+ was 0.011
mM, and no NAD(H) dependent activity was detected. The temperature dependence of
native and recombinant enzyme activities were determined to be similar. Thus, only the data
obtained with the recombinant enzyme are reported here. T. ethanolicus 2° ADH activity
was detected below 25 °C and increased to beyond 90°C (Fig. 4A). The 2° ADH half-life at
90°C was 1.7 h. The Arrhenius plot for the oxidation of propan-2ool was linear from 30°C
to 90°C when the enzyme was incubated at 55°C prior to assay. Under the same
conditions, however, a distinct discontinuity was seen in the Arrhenius plot for ethanol
oxidation (Fig. 4B). Discontinuities were also observed at ~55°C and ~46°C for propan-2-
01 and ethanol oxidations, respectively, when the enzyme was preincubated at 0°C or 25°C
(Table 2). The slopes of the best fit regression lines above and below the discontinuity
were significantly different beyond the 95% confidence level except for propan-2-ol
oxidation by enzyme preincubated at 55°C, where the regression line slopes were similar at
the 95% confidence level. The activation energies for propan-2-ol and ethanol oxidations
were similar at assay temperatures above the discontinuities but were 15 to 20 kJ mol'1
higher for ethanol oxidation than for propan-Z—ol oxidation at temperatures below the
discontinuity temperatures (Table 2). Furthermore, the differences between activation
energies above and below the discontinuity temperatures (D) decreased with increasing
preincubation temperatures. and the differences for ethanol oxidation were at least 3-fold

higher than for propan-2-ol oxidation.

109

Figure 4. Arrhenius plots for the recombinant T. ethanolicus 39E 2° ADH
between 25°C and 90°C. (A) Temperature-activity data for propan-2-ol oxidation with
the enzyme preincubated at 55°C. Linear regression best fits to the data was determined to
be y = 13.647 - 3.3694x (R2 = 0.993). (B) Temperature-activity data for ethanol oxidation
with the enzyme preincubated at 55°C. Points above and below the discontinuity are
indicated by. and., respectively. Linear regression best fits to the data were determined
to be y = 9.7929 - 2.5036x (R2 = 0.946) for points above and y = 15.730 - 4.3899x (R2 =
0.969) for points below the discontinuity.

 

110

[(Bw/suun) Amuse ouioadslm

n 0. v: C. v: 0.
MMNN—iv—t

 

3.3

3.1

2.9

 

 

 

 

2.7

 

A
3.3

3.1

2.9

 

 

 

Lir Ifijfififii' I - a
Q '0 9. V? q W. Q
in V V M M N N

[win/SIN“) KJIAIIOB 01.1109dSJUI

3

10 x 1 [Temperature (K)

111

.293 858:8 e3 2.. a anemones 382.com a; 53:25:. as 23 osaosoo .oz Haze
Arm—55586 05 26% 3.88 533.03 - 33658:. 05 323 3.85 ecu—268v u an

.83 05 S 8:: comic." E “no; 05 .«o abuse

85580 .3 ~32 853:8 $3 05 a vogue 83 “ca 3205 2335 a .«e consumes ofiu

 

 

 

 

S on _N no nn
cm 3. mm 3 mm
R we 2 me o 3:an
fl S on 96 mm
o em on em mm
o 8 S R c l «receded
325.88% 33558:.
Q. 323 26% Gee Gov Between
88% 5%
9.38 we 3.85 eeuguo< 335.585— :cufleoemoun l

 

43. em mom gouge g .3 Scene
.0550 e5 ~e§q€9~ e8 3.85 52:68 05 no gages 5393.35 go we Scum .N 3%.”.

112
DISCUSSION

This first report of the cloning and expression of a thermophilic 2° ADH also
provides evidence on the molecular bases for enzyme activity, thermophilicity, and
discontinuous Arrhenius plots. Protein sequence comparisons predict that 1° and 2° ADHs
have some structure-function similarities related to catalytic Zn and nicotinamide cofactor
binding. However, differences in the hydrophobic clusters present in the overall 1° and 2°
ADH sequences predict significant differences in the overall structures of these enzymes.
The recombinant T. ethanolicus protein, with kinetic properties similar to those of the
native enzyme [6], is a thermophilic and thermostable NADP(H) dependent enzyme that
exhibits significantly greater catalytic efficiency toward 2° alcohols than 1° alcohols due to
both lower Kmapp and higher Vmaxapp values. Chemical modification experiments suggest
that enzyme activity requires at least Cys and His residues and a tightly bound Zn atom.
Finally, the magnitude of the bend in the Arrhenius plots varied inversely with enzyme
preincubation temperature, indicating the existence of a catalytically significant, temperature
dependent structural change in the enzyme.

The ath gene encoding the thermophilic T. ethanolicus 2° ADH was cloned by
hybridiution and expressed in E. coli. Comparisons of the native and recombinant enzyme
N-terminal sequences and molecular masses showed that the recombinant enzyme differed
from the native protein only by the addition of an N-terminal methionine. While the first
ATG is the main translation start codon in E. coli, it is unknown if it is the sole initiation
codon (and the first methionine is processed later), or if the second ATG is also used as an
initiation codon. The presence of a truncated ORF (preceded by an RES) 180 nucleotides
downstream of ath and the absence of any potential transcriptional stop signal suggested

that ath may be the first gene of an operon. Although confirmation of this hypothesis
will require additional experimentation, the similarity between, the truncated ORF reported

1 13
here and the one downstream of C. beijerinckii ath lends further support to this

hypothesis.

HCA alignments of the 1° and 2° ADHs directed by their pyridine dinucleotide
binding motifs indicated some structural similarity between these classes of enzymes. The
negatively charged Asp(223r 195» 215) residues are consistent with the NAD(H) dependence
of the horse liver, B. stearothennophilus , and A. eutrophus enzymes, respectively, while
the presence of an uncharged residue (Gly193) at the analogous position in the T.
ethanolicus, T. brockii and C. beijerinckii 2° ADHs is consistent with the NADP(H)
dependence of these enzymes [3 ,6,7]. Furthermore, residues identified in the enzyme
hydrophobic core, residues lining the active site cavity, and structurally important glycines
previously identified in the horse liver 1° ADH [l], were better conserved between proteins
in this alignment than the overall peptide sequences. Similarity of putative structurally
critical regions has been reported for other ADH comparisons [34]. The correlation
between enzyme structure-function properties and the predictions of this alignment support
its use in further ADH structure-function studies. However, the apparent lack of a
structural Zn binding loop in the 2° ADHs, the reports of only tetrameric 2° ADHs but both
tetrameric and dimeric 1° ADHs, and the significantly greater similarity between the 2°
ADHs than between the 1° and 2° ADHs predicted by HCA comparison also suggests that,
while functionally similar, 1° and 2° ADHs may be less str'uctm'ally similar than was it
previously believed.

The thermophilic T. ethanolicus 39E 2° ADH incubated at 90°C retained detectable
activity for more than one how while the mesophilic C. beijerinckii 2° ADH was
completely inactivated within 10 min at 70°C [6]. Still, these two enzymes shared more
than 85% sequence similarity. Nine of the nonconservative substitutions between the
subunits of these proteins correspond to prolines in the T. ethanolicus enzyme. The
similarly thermophilic T. brockii 2° ADH contains 8 of these additional prolines. This
difference in proline content of the thermophilic versus mesophilic 2° ADHs is consistent

1 14
with the hypothesis of Matthews er al. regarding the role of prolines in protein stabilization

[33] and with the observations by numerous investigators of protein stabilization due to
proline insertion [35 ,36], constrained loop regions [37], and proline substitution into loops
[38—40]. The additional prolines in the T. ethanolicus 2° ADH were either in short putative
loop regions or in longer putative loop regions containing multiple prolines, suggesting that
the specific placement as well as the number of prolines may be critical to their stabilizing
effect. The difference in proline content between the T. ethanolicus and C. beijerinckii
enzymes and their overall high sequence similarity makes these 2° ADHs an excellent
system for testing the effect of proline insertion on protein structtne stabilization.
Comparative sequence analysis predicted the involvement of specific Cys, His, and
Asp residues in 2° ADH catalysis. Chemical modification of the recombinant and native T.
ethanoll’cus 2° ADHs with DTNB established the importance of cysteine residues in
catalysis. The observation that Zn was the only metal enhancing the enzyme reactivation
after DTNB modification suggested that the enzyme is Zn dependent. This finding is in
agreement with the previous report that, once inactivated by the sulfhydryl modifying
reagent p-chloromercuribenzoate, the T. brockl'i 2° ADH recovered activity only in the
presence of ZnClz [3]. Here, the recovery of 34% initial activity upon reversal of DTNB
modification in the absence of added metal and the inability of EDTA to reduce the rate of
reactivation, even at 60°C, suggested that the catalytic metal remained tightly bound to the
protein. Histidine specific modification of the T. ethanolicus enzyme by DEPC was also
accompanied by complete enzyme inactivation, implicating a histidine residue in catalysis.
The apparent lack of a structural Zn binding region in the 2° ADH subunits argues that the
DTNB linked inactivation and Zn dependent reactivation are not due to the loss and
recovery of a cysteine-liganded structural Zn. Therefore, the catalytically important
cysteine and histidine residues may act as ligands to the catalytic (Zn as described for other
ADHs [1]. Determination of the importance of Asp150 to 2° ADH activity awaits
mutagenic, crystallographic, or physical biochemical analysis.

1 15
The temperature dependence of catalytic activities for the native and recombinant

enzymes were similar. The Arrhenius plot for propan-2-ol oxidation by T. ethanolicus 2°
ADH preincubated at 55°C was linear, unlike that previously reported for the T. brockii 2°
ADH [3]. However, statistically significant discontinuities in the Arrhenius plots for both
ethanol and propan-2-ol oxidation by the T. ethanolicus enzyme were seen when the
enzyme was preincubated at lower temperatures. A change in the rate determining step of
the overall reaction and other explanations that do not invoke alterations in catalyst structure
predict Arrhenius plot discontinuities [41,42]. A shift in the reaction slow step not related
to an alteration in enzyme structure would be independent of the initial temperature of the
enzyme, and is inconsistent with the inverse relationship observed here between enzyme
preincubation temperature and the difference in activation energies above and below the
discontinuity (D). At assay temperatures above the discontinuity, the reaction activation
energies were similar for enzymes preincubated at 0°C, 25°C and 55°C, unlike those at
lower assay temperatures. This observation is consistent with the enzyme attaining its
optimally active conformation rapidly enough at higher assay temperatures not to affect the
measured rate. The discontinuity temperatures, being below the lowest reported
temperatures for T. ethanolicus growth, are physiologically irrelevant, but they underscore
the importance of treating thermophilic enzymes differently from mesophilic enzymes when
conducting kinetic analyses. Furthermore, the differences between the low temperatme
activation energies for ethanol and propan-2-ol argue that substrate/product-protein
interactions are important to the rate determining step in catalysis.

The linear Arrhenius plot for propan-2-ol oxidation by T. ethanolicus 2° ADH
preincubated at 55°C indicates that the low mesophilic temperature activity of this enzyme
may be explained by the substrate energy alone. Arrhenius theory predicts increasing
Kmapp and Vmaxapp with increasing temperature [41], and this effect has been confirmed
for the Thermotoga neopolitana D-xylose isomerase [43]. The thermophilic 2° ADH has

lower Vmaxapp values than would be predicted from extension of the mesophilic enzyme

1 16
activity at 25°C. In fact, at 60°C the thermophilic 2° ADH maintains Kmapp and Vmaxapp

values toward substrates similar to those reported for the mesophilic enzyme at 25°C.
These similar kinetic values suggest that the thermophilic enzyme channels more total
reaction energy into substrate binding and less into turnover [44]. Therefore, an alternative
explanation for the comparatively poor T. ethanolicus 2° ADH low temperature activity
which argues that substrate affinity at high temperatures is maintained by sacrificing high

turnover number must also be considered.

ACKNOWLEDGMENTS

I gratefully acknowledge Dr. Claire Vieille for her aid in performing the molecular biology
and in writing this manuscript. 1 also gratefully acknowledge Maris Laivenieks for his
assistance in protein purification and molecular biology, as well as Drs. Vladimir
Tchemajenko and David Giegel for their helpful discussions. The MSU Mass
Spectrometry Facility is supported, in part, by a grant (DRR-00480) from the
Biotechnology Resource Technology Program, National Center for Research Resources,
National Institutes of Health. This research was supported by a grant from the Cooperative

State Research Service, US. Department of Agriculture, under the agreement 90-34189-
5014.

10.

ll.
12.

13.
14.

15.

l6.

17.
18.
19.

117
REFERENCES

Branden, C.-I., Jornvall, H., Eklund, H. and Furugren, B. (1975) in The enzymes,
vol. XI part A (Boyer, P. D., ed.), pp. 103-190, Academic Press, NY

Eklund, H., Nordstrom, B., Zeppezauer, B., Soderlund, G., Ohlsson, I., Boiwe, T.
and Branden, C.-I. (1974) FEBS Lett. 44, 200-204

Lamed, R. J. and Zeikus, J. G. (1981) Biochem. J. 195, 183-190
Steinbl'ichel, A. and Schlegel, H. G. (1984) Eur. J. Biochem. 141, 555-564

Bryant, F. 0., Wiegel, J. and Ljungdahl, L. (1988) Appl. Env. Microbiol. 54, 460-
465

Ismaiel, A. A., Zhu, C.-X., Colby,G. D. and Chen, J.-S. (1993) J. Bacteriol. 175,
5097-5105

Burdette, D. S. and Zeikus, J. G. (1994) Biochem. J. 302, 163-170
Lovitt, R. W., Shen, G.-S. and Zeikus, J. G. (1988) J. Bacteriol. 170, 2809-2815

Keinan, B., Hafeli, E. K., Seth, K. K. and Lamed, R. L. (1986) J. Am. Chem. Soc.
108, 162-169

Keinan, B., Hafeli, E. K., Seth, K. K. and Lamed, R. L. (1986) Ann. N. Y. Acad.

’ Sci. 501, 130-149

Hummel, W. (1990) Appl. Microbiol. Biotech. 34, 15-19

Keinan, B., Seth, K. K., Lamed, R. L., Ghirlando, R. and Singh, S. P. (1990)
Biocatalysis 4, 1-15

Hummel, W. and Kula, M.-R. (1989) Eur. J. Biochem. 184, 1-13

Persson, M., Mansson, M.-O., Biilow, L. and Mosbach, K. (1991) Biotechnology
9, 280-284

Lee, Y.-E., Jain, M. K., Lee, C., Lowe, S. B., and Zeikus, J. G. (1993) Int. J.
Syst. Bacteriol. 43, 41-51

Nagata, N., Maeda, K. and Scopes, R. K. (1992) Bioseparation
2. 353-362

Peretz, M. and Burstein, Y. (1989) Biochemistry 28, 6549-6555
Oka, A., Sugisaki, H. and Takanami., M. (1981) J. Mol. Biol. 147, 217-226

4Zfil4ctsls, J. G., Hegge, P. W. and Anderson, M. A. (1979) Arch. Microbiol. 122,

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.
32.
33.

34.

35.

36.

37.

l 1.8

Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning: A
laboratory manual 2nd edition. (Nolan, C., ed.), Cold spring Harbor Press, NY

Ausubel, F. M., Brent, R., Kingston, R. B., Moore, D. D., Seidman, J. G., Smith,
J. A. and Struhl, K. (1993) Current Protocols in Molecular Biology (Janssen, K.,
ed.), Current Protocols, NY

Doi, R. H. ( 1983) in Recombinant DNA Techniques: An Introduction (Rodriguez, R.
L, and Tait, R. C., eds.), pp. 162-163, Benjamin-Cummings, Menlo Park, CA

Lee, C. C., Wu, X., Gibbs, R. A., Cook, R. G., Muzny, D. M. and Caskey, C. T.
(1988) Science 239, 1288 - 1291

Sanger, F., Nicklen, S. and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74,
5463-5467

Remington, R. D. and Schork, M. A. (1985) Statistics with applications to the
biological and health sciences, pp. 296-309, Prentice-Hall, NJ

Brooks, S. (1992) Biotechniques 13, 906-911

Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H.,
Provenzano, M. D., Fugimoto, E. K., Goele, N. M., Olson, B. J. and Klenk, D. C.
(1985) Anal. Biochem. 150. 76-85

Habeeb, A. F. S. A. (1972) in Methods in enzymology vol. 25: Enzyme structure,
part B (Hirs, C. H. W. and Timasheff, S. N ., ed.), pp. 457-459, Academic press,
NY

Miles, E. W. (1977) in Methods in enzymology vol. 47 (Hirs, C. H. W. and
Timasheff, S. N., ed.), pp. 431-442, Academic press, NY

Lemesle-Verloot, L., Henrissat, B., Gaboriaud, C., Bissery, V., Morgat, A. and
Mornon, J. P. (1990) Biochimie 72, 555-574

Hawley, D. K. and McClure, W. R. (1983) Nucl. Acids Res. 11, 2237-2255
Wierenga, R. K. and H01, G. J. (1983) Nature 302, 842-844

Matthews, B. W., Nicholson, H. and Becktel, W. ( 1987) Proc. Natl. Acad. Sci.
USA 84, 6663-6667

Jendrossek, D., Steinbuechel, A. and Schlegel, H. G. (1988) J. Bacteriol. 170,
5248-5256

Heming, T., Katsuhide, Y., Taniyama, Y. and Kikuchi, M. (1991) Biochemistry
30, 9882-9891

Heming, T., Katsuhide, Y., Inaka, K., Matsushima, M. and Kikuchi, M. (1992)
Biochemistry 31, 7077-7085

Eijsink, V. G. H., Vriend, G., van de Burg, B., van de Zee, J. R., Veltman, O. R.,
Stulp, B. K. and Venema, G. (1993) Protein Engineer. 5, 157-163

38.

39.

40.

41.

42.
43.

119

Watanabe, K., Masuda, T., Ohashi, H., Mihara, H. and Suzuki, Y. (1994) Eur. J.
Biochem. 226, 277-283

Watanabe, K., Chishiro, K., Kitamura, K. and Suzuki, Y. (1991) J. Biol. Chem.
266, 24287-24294

Hardy, F., Vriend, G., Veltman, O. R., van de Vinne, B., Venema, G. and Eijsink,
V. G. H. (1993) FEBS Lett. 317, 89-92

Segel, I. H. (1975) in Enzyme kinetics: behavior and analysis of rapid equilibrium
and steady-state enzyme systems, pp. 929-934, John Wiley & Sons, NY

Londesborough, J. (1980) Eur. J. Biochem. 105, 211-215

Vieille, C., Hess, J. M., Kelly, R. M., and Zeikus, J. G. (1995) Appl. and Environ.
Microbiol. 61, 1867-1875

Jencks, W. P. (1975) in Advances in enzymology and related areas of molecular
biology, vol. 43, (Meister, A., ed.), pp. 219-411, John Wiley & sons, NY

Chapter 111
Effect of thermal and chemical denaturants on
Thermoanaerobacter ethanolicus 39E secondary-alcohol

dehydrogenase stability

Prepared for submission to Biotechnology and Applied Biochemistry

120

121
ABSTRACT

Thermoanaerobacter ethanolicus secondary-alcohol dehydrogenase (2° ADH) was
optimally active near 90°C with calculated thermostability half-lives of 1.2 days, 1.7 h, 19
min, 9.0 min, and 1.3 min at 80°C, 90°C, 92°C, 95°C, and 99°C, respectively. Enzyme
activity loss upon heating (90-100°C) was accompanied by precipitation, but the soluble
enzyme remaining after partial sample inactivation retained complete activity,
mechanistically linking enzyme inactivation and precipitation. Enzyme thermoinactivation
was modeled by a pseudo-first order rate equation suggesting that the rate determining step
was unimolecular with respect to protein, thus, also implying that thermoinactivation
preceded aggregation. Structural unfolding of the 2° ADH occurred with a Tm (50%
unfolding) at ~115°C. This Tm was much higher than the temperature for maximal activity,
indicating that the 2° ADH was rigid and tightly folded throughout its active temperature
range. Thermodynamic calculations indicated that the active folded structure of the 2°
ADH, like other proteins, is stabilized by a relatively small Gibbs energy (AGismb, = 110
kl mol'l). 2° ADH was also active in > 1.0 M GuHCl, demonstrating its high resistance to
chemical denaturation. Enzyme activities were increased 2-fold in low guanidine
hydrochloride (GuHCl) concentrations (150-230 mM). The additive denaturing effects of
GuHCl and temperature on T. ethanolicus 2° ADH TmlGu] predict that at higher
temperatures lower GuHCl concentrations should be required to reduce enzyme rigidity,
thus lower GuHCl concentrations should result in optimal enzyme flexibility and activity.
The increasing Gui-1C1 concentration that provided maximal 2° ADH catalytic rate
enhancement at higher temperatures was also seen with weakly chaouopic (KN 03) or
neutral (KCl) inorganic salts, suggesting that this 2° ADH catalytic rate enhancement results
from specific ionic and not hydrophobic interactions.

122
INTRODUCTION

As industrial catalysts, enzymes must have robust activities that are stable under
process chemical and thermal conditions. The saccharidases used in starch processing and
the proteases used in detergents typify current industrial enzyme standards. The potential
biotechnological application of enzymes as chiral chemical catalysts has long been
recognized [1-10]. The chiral synthesis of high value bioactive compounds such as
pharmaceuticals is increasingly important due to the recent FDA ruling mandating either
. enantiomeric purity or exhaustive toxicity testing to demonstrate the lack of any detrimental
effects due to enantiomeric impurities [11]. Traditional chemical synthetic methods can
often only ensure chiral purity using chiral feedstock chemicals because racemate resolution
is often too difficult and expensive. Thus, the economic biocatalytic synthesis of either
chiral feedstock chemicals or of the final chiral products can significantly advance current
technologies.

Alcohol dehydrogenases (ADHs) are almost exclusively nicotinamide cofactor
dependent, enantioselective enzymes with broad specificity [12]. While NAD(P)(H) is
costly, the demonstrated effectiveness of numerous NAD(P)(H) retention and redox
recycling strategies has removed this economic barrier to the scale-up of even moderate
value adding processes [9,10,13]. Analytical scale chiral alcohol formation by 1° ADHs
has been demonstrated for a diverse set of substrates [14.5.6]. These 1° ADHs are far less
active on 2° alcohols or ketones than on 1° alcohols or aldehydes. The emerging class of 2°
ADHs have greater catalytic efficiency for 2° alcohol oxidation and ketone reduction than
for 1° alcohol oxidation and aldehyde reduction and they display broad substrate
specificities [15-17]. The potential for 2° ADH catalyzed chiral chemical syntheses has
been demonstrated using the Thermoanaerobacten'wn brockii and Thermoanaerobacter
ethanolicus enzymes [18-22].

123
The kinetic similarity of the 2° ADHs isolated from a number of similar

thermoanaerobes led Nagata et al. to conclude that the enzymes were also structurally
similar [15]. A comparison of the T. brockl'l' 2° ADH peptide sequence, determined by
Edman degradation, to that of liver 1° ADH predicted some potential structure function
similarities [21], however, without a 2° ADH clone these predictions could not be tested.
Recently, the ath gene encoding the T. ethanolicus 2° ADH was cloned, characterized,
and overexpressed, demonstrating that the T. erhanolicus and T. brockii 2° ADHs differed
by only 3 amino acids [23]. Therefore, research using the T. ethanolicus 2° ADH may
provide knowledge representative of other 2° ADHs.

Thermozymes [24], with their greater resistance to both thermal and chemical
denaturation than mesophilic enzymes, may alleviate some of the limiting process
conditions for applications of enzymes and enhance enzyme lifespans. Irreversible
mesophilic and thermophilic protein thermoinactivation has been shown to result from the
loss of folded protein structure and from covalent peptide or amino acid destruction [see
24], suggesting that the higher stability of thermozymes is due to interactions
mechanistically similar but stronger than those stabilizing mesophilic enzymes [see 24].
Thermophilic enzymes recombinantly expressed in mesophilic hosts retain their thermal
characteristics, indicating that the peptide sequences encode these properties. Comparative
analyses of thermophilic and mesophilic enzymes have typically concluded, from the lack
of exotic amino acids or 1° structural differences, that the added stability of one enzyme
compared to another results from numerous subtle differences in the folded protein
structure [24,25]. Enzyme thermostabilizing mechanism theories including increased core
hydrophobicity and atomic density, additional salt bridges, 2° structural element
stabilization, and constrained surface loop architecture have been reported [see 24].

Detailed studies on thermal and chemical denaturation of a thermophilic 2° ADH
have not been reported The data presented here measure T. ethanolicus 2° ADH
thermophilicity, thermostability, and stability to denaturation by GuHCl. The relationship

124
between T. ethanolicus 2° ADH rigidity and thermostability is determined through kinetic

and biophysical characterization of enzyme activity, precipitation, and unfolding. Also, the
effect of the chemical denaturant guanidine hydrochloride (GuHCl) on the activity of this

robust biocatalyst is examined.

MATERIAL AND METHODS

Chemicals and reagents

All chemicals were of at least reagent/molecular biology grade. Gases were provided by
AGA specialty gases (Cleveland, OH) and made anaerobic by passage though heated
copper filings. Spectrophotometrically pure GuHCl was obtained as an 8.0 M aqueous
solution (Cat. #24115: Pierce; Rockford, IL). Tris buffer pH values were measured at

25°C and formulated for pH 8.0 at the temperature they were used (thermal correction
factor = -0.031 ApH °C'1).

Media and strains

Escherichia coli (DHSa) containing the 2° ADH recombinant plasmid [23] was grown in
rich, complex medium (20 g 1'1 tryptone, 10 g l-1 yeast extract, 5 g 1'1 NaCl) at 37°C with
25 tlg ml'1 kanamycin and 100 ug ml-1 ampicillin.

Enzyme purification
The recombinant 2° ADH was purified from E. coli (DHSa) aerobically. The pelleted cells

from batch cultures were resuspended (0.5 g wet wt. ml'l) in buffer A [50 mM Tris:HC1
(pH 8.0), 5 mM DTT, and 10 11M ZnClz} and lysed by passage though a French pressure

cell. The clarified lysate was incubated at 65°C for 25 min then centrifuged for 30 min at
15000g. The clarified supernatant was applied to a DEAE-sephacryl column (2.5 cm x 15

125
cm) equilibrated with buffer B {50 mM TriszHCI (pH 8.0)}. Active fractions were diluted

4-fold in buffer B and applied to a Q-sepharose column (2.5 cm x 10 cm) equilibrated with
buffer B. Purified recombinant enzyme was eluted using a 250 ml NaCl gradient (0-300
mM).

Enzyme activity

The standard assay for 2° ADH activity was defined as the reduction of NADP+ using
propan-2-ol at 60°C as previously described [17]. Kinetic parameters were calculated from
nonlinear best fits of the data to the Michaelis-Menten equation using Kinzyme software
[26] on an IBM PC. Protein concentrations were measured in all cases by the
bicinchoninic acid (BCA) procedure (Pierce; Rockford, IL). The effect of GuHCl on 2°
ADH activity was determined at 37°C, 50°C, 60°C, 70°C, and 75°C in the absence or
presence of 100 mM, 200 mM, 400 mM, 800 mM, or 1.6 M GuHCl. ‘

Ionic activities
Guanidinium ionic activities were determined using the Debye-Hiickel limiting law [see
27]. The mean ionic activity (at) is defined as
3:1: = 711111:
where rm; is the mean electrolyte molality and 71 is the mean activity coefficient. The mean

activity coefficient is determined as a function of temperature (T), ionic charge (2),

dielectric constant of water (a = 78.54), and ionic strength (I) using the equation:
log y: = {-1.824x106/[(er)(3/2)] } (lz.,.z.l)[(I)(1/2)]

where

126

I = (1/2)Z(mi2i2).

The GuHCl concentration dependent enzyme activity experiments were performed in 45
mM to 50 mM Tris (pH 8.0) at 37°C, 50°C, 60°C, 70°C, and 75°C. Tris buffer was
assumed to be 50% dissociation (pKa = 8.1). Because of GuHCl's pKa (> 10) it was

assumed to be completely ionized. All species ionic charges were 11.

Thermostability

Purified recombinant 2° ADH thermostability was evaluated by timed incubation at the
desired temperatures, followed by incubation for 30 min at 25 °C. Samples were then
equilibrated at 55°C for at least 15 min prior to activity determination. Activity was
determined using the unfractionated samples. Incubations were performed in 100 til PCR
tubes (cat. #72.733.050, Sarstedt; Newton, NC) using 0.2 mg ml'1 protein in 100 pl of
buffer B. Protein precipitation due to heating for 2 min at 99°C was quantified in similar
experiments by fractionation of the soluble and aggregated protein. First, the 70 pl of
sample remaining after thermostability activity determination was centrifuged for 5 min at
12000g. The supernatant was then removed from the pelleted protein aggregate, the pellet
was resuspended in 100 [ll of buffer B, centrifuged (12000g, 5 min), the wash buffer
removed, and the repelleted protein resuspended in 70 ul of buffer B. Enzyme activities
and protein concentrations were determined for the initial sample, supernatant, pellet wash,
and the resuspended pellet. Specific acivities and tom] protein values from the pellet wash
and resuspended pellet fractions were combined and reported as the precipitate specific

activity and total protein, respectively.

127
Heat induced enzyme precipitation

Heat-induced enzyme precipitation was monitored from 25 °C to 100°C by light scattering
(it = 5 80 nm), using protein solutions (0.2 mg ml'l) in buffer B. Absorbance
measurements were conducted in 0.3 ml quartz cuvettes (pathlength = 1.0 cm), using a
Gilford Response spectrophotometer (Coming; Oberlin, OH) equipped with a Peltier
cuvette heating system. The apparent OD530 increase is due to light scattering from protein

precipitation.

Heat-induced enzyme unfolding

2° ADH unfolding experiments were conducted in buffer B in the absence of GuHCl and in
the presence of 2.0—6.0 M GuHCl. Unfolding was monitored from 25°C to 100°C by
absorbance at 280 nm [28-30] in 0.3 ml quartz cuvettes (pathlength = 1.0 cm), using a
Gilford Response spectrophotometer equipped with a Peltier cuvette heating system.
Melting temperature (Tm) is defined as the temperature of the absorbance transition
midpoint during an increasing thermal gradient (1.0°C min'l). The 95% confidence interval
error boundaries for the predicted 2° ADH Tm were determined by standard statistical
treatment of the data [31].

Fluorescence spectrometry

Fluorimetric unfolding measurements were conducted with protein solutions (0.2 units
O.D.230) in buffer C [10 mM Tris:HCl (pH 8.0)] at 25°C. Fluorescence spectra
“excitation = 295 nm, Aemission = 315-450 nm) were recorded in 2.25 ml quartz cuvettes
(pathlengths: excitation = 0.5 cm, emission = 1.0 cm), using an SLM spectrofluorimeter
(Champain-Urbana, Il) equipped with a thermal, cell holder. Decreasing flomescence
intensity indicates exposure of the aromatic amino acid residues to the polar solvent

environment.

128
RESULTS

Thermophilicity and thermostability

T. ethanolicus 2° ADH activity increased with temperature to at least 90°C with no activity
observed at 100°C (Fig. 1). Accrnately quantitating enzyme activity at temperatures above
90°C was difficult due to NADP(H) instability and equipment constraints, however, no
product formation was detected during reactions at 100°C. Recombinant 2° ADH
thermostability was determined at 65°C, 80°C, 90°C, 92°C, 95°C, and 99°C (Fig. 2). The
enzyme was stable at 65°C, and, it had calculated half-lives of 1.2 days, 1.7 h, 19 min, 9.0
min, and 1.3 min at 80°C, 90°C, 92°C, 95°C, and 99°C, respectively. The rate of
thermoinactivation was described by equation 1 for all temperature measured (k=rate

constant and Ftime (min)).
ln(residual activity) = 1n(irlitial activity) - kt (1)

This pseudo first order relationship indicated that the slow step in T. ethanolicus 2° ADH
thermoinactivation did not depend on enzyme concentration. Arrhenus analysis of the
inactivation rate constants (6.70x10‘6 SCC’l, 1.34le seC'l, 6.66x10'4 seC'l, 1.60x10‘3
seC'l, and 1.87x10'2 sec1 at 80°C, 90°C, 92°C, 95°C, and 99°C, respectively) yielded a
transition state Gibbs energy barrier of 110 kJ mol'1 resulting from the difference between
AH? = 440 kJ mol'1 and A81 = 330 kJ mol'1 (temperature = 363 K).

Thermostability and unfolding
Enzyme inactivation was accompanied by precipitation and the effect of protein aggregation
on enzyme activity was investigated. Spectrophotometric determination of heat-induced

protein precipitation using a 10°C per minute thermal gradient indicated that the enzyme

129

Figure l
Recombinant T. ethanolicus 39E 2° ADH thermophilicity. 2° ADH propan-2-ol oxidation
activity was measured at temperatures spanning 30°C to 100°C.

130

 

 

 

100

80-
60"
40-
20-

A539:— uEEEE a=>39< 958.5

 

120

100

80

60

40

20

Temperature (°C)

131

Figure 2

Recombinant T. ethanolicus 2° ADH thermostability. 2° ADH residual activity was
determined after timed (1 min to 100 min) incubations at 65°C (0), 80°C (0), 90°C (A),
92°C (0), 95°C m, and 99°C a3).

132

 

 

 

 

A539:— 9532::
3:235 #:90an us—

 

100

80

60

4O

20

(min)

Time

133
began to precipitate at temperatures above 85°C (Fig. 3). Precipitation was a rapid,

cooperative process that was essentially complete at 100°C with the transition midpoint at
~95°C. The potential correlation between enzyme precipitation and loss of catalytic activity
was then investigated by direct measurement (Table 1). After heat treatment for 2 min at
99°C (approximately the enzyme half-life at this temperature), the specific activity of the
unfractionated heat treated sample was half that of the unheated enzyme. Of the 11 i 1.1
[lg of protein originally added, 10 i 1.0 pg were recovered after heat treatment.
Subsequent fractionation of the remaining sample (70 pl of the original 100 111) by
centrifugation indicated that the supernatant contained half of the initial enzyme. This
soluble enzyme had similar specific activity to the unheated sample. The resuspended
pellet, which did not redissolve, also contained half of the protein but displayed no
detectible enzyme activity. Once precipitate had formed, enzyme continued to inactivate
and aggregate slowly upon standing at room temperature (data not shown).

The role of protein unfolding in 2° ADH thermoinactivation was investigated by
measuring the effect of GuHCl on temperature dependent enzyme unfolding.
Spectrophotomeuically determined transition midpoints for protein unfolding (using a
1.0°C per minute thermal gradient) with 2.0 M, 2.25 M, 2.5 M, 2.75 M, and 3.0 M added
GuHCl yielded melting temperature (TmlGul) values of 68°C, 59°C, 58°C, 53°C, and 48°C,
respectively. Tm is defined as the midpoint of the unfolding transition and is interpreted as
the point where the protein is 50% unfolded. The inset in figme 4 of a representative 2°
ADH melting curve used to obtain the Tm[Gu] values indicates that unfolding was a rapid,
c00perative process. At GuHCl concentrations below 2.0 M enzyme precipitated prior to
completing the unfolding transition so no midpoint temperature could be identified. Also,
at concentrations above 3.0 M the absorbance change registered upon 2° ADH denaturation
became to small for reliable Tm[Gu] determination. Enzyme Tm[Gu] values decreased
linearly with increasing GuHCl concentrations (Fig. 4). The GuHCl dependence of 2°

134

Figure 3
Temperature dependence of 2° ADH precipitation. 2° ADH precipitation during an
incresaing thermal gradient (1.0°C min-1) was determined by sample absorbance at 580 nm.

OD. (580 nm)

135

 

 

 

 

Temperature (°C)

 

100

136

6088833 803
888.888 5208 E .388 088:» 0388 880382. 05 :ounamom “gage e8 888695 3 hem:
.ASegoeSh 8 BE 0388 20388 05 we mum—:8 :8 3588 293 3

cc 83.9 eoeoee a 8: was as 9 so... menace: .3 B .8838 one as? seen eosaoa Bob

.8on RE 05 e5 :33 05 8 805 .«o 88 05 8a Seen—E Egon 05 you concav—
moea> 2F .:eeaw_£b:oo .3 30:25: on: 8.95 five .23 3:33 83 88880.:— eoeo=om 2F 4:288:
:8; Bee Summer—:50 .3 ages—09$ ageE 0528 05 88a 3883 «:3 aux—Re 3538 33860.8 25.

.58 N to: come a 3385 2o.» m 88.. a moans soon 38:.

 

 

need a M: £3 a Ed 3 a. 2 1385
Se v 283 v serene.“ see
one a m Sod a w; some a cw assuage
nae?
2 H mm 8.3 a ”8 82. a : :8:
one sec 9.83 Ame
been ocean area Boa £2883

 

$383.85 38.—2: Ina. ou guagefie H :m 53:338.:— 538: he 23— n 935—.

 

137

Figure 4

Effect of GuHCl on T. ethanolicus 2° ADH melting temperature. The Tmmu] values from
absorbance were measured at 280 nm during an increasing thermal gradient (1.0°C min'l)
with 2.0 M, 2.25 M, 2.5 M, 2.75 M, and 3.0 M added GuHCl. The inset shows percent
protein unfolding versus temperature for 2° ADH in 2.5 M GuHCl.

138

 

 

 

 

 

 

 

wwwwaw
33.-=2.—

T2

 

 

120

100-
80q
60-
40'

Gov 9.59—2.83 ant—o:

20"

GuHCl concentration (M)

l 39
ADH denaturation determined by fluorescence intensity at 25 °C indicated a melting

transition midpoint at 3.7 M (Fig. 5). Extrapolating to zero GuHCl from both the
absorbance and fluorimetric Tm[Gu] data predicts a 2° ADH Tm of 115 i 5.8°C (at the 95%

confidence level).

Chemical stability and 2 °ADH activity

The generally poor low temperature activity of thermozymes has been attributed to their
high rigidity at these temperatures [32]. Therefore, reducing enzyme rigidity by adding a
chemical denaturant would be expected to enhance thermozyme low temperature activity.
The chemical denatm'ant GuHCl and thermal energy were also used to study the effect of
enzyme rigidity on catalysis. Figure 6 shows the effect of GuHCl concentrations up to 1.6
M on enzyme activity at temperatures between 37°C and 75°C. Maximal enzyme specific
activity was seen in the presence of 150 mM to 230 mM GuHCl at all temperatures tested
and it was approximately doubled that in the absence of GuHCl. The GuHCl concentration
corresponding to maximal enzyme activity increased with increasing temperature (147 mM,
184 mM, 206 mM, 237 mM, and 231 mM at 37°C, 50°C, 60°C, 70°C, and 75°C,
respectively). The mean ionic activity coefficients (7:) calculated from the Debye-Hiickel
limiting law for these maximal activity conditions averaged to 0.808 1!: 0.0051. The
calculated GuHCl activities also increased with increasing temperature (120 mM, 149 mM,
165 mM, 183 mM, and 187 mM at 37°C, 50°C, 60°C, 70°C, and 75°C, respectively).
Similar but less dramatic low concentration 2° ADH rate enhancements were seen for added

KCl and KN03 but not for K2804 (data not shown).

140

 

Figure 5
Effect of GuHCl on T. ethanolicus 2° ADH fluorescence. The flourescenoe measurements
(lacim=295 nm, Missim=340 nm) at 25 °C were conducted using protein in GuHCl

concentrations from 0.0 to 6.0 M.

 

Relative Florescence

141

 

 

 

 

GuHCl concentration (M)

 

142

Figure 6

GuHCl and temperature dependence of 2° ADH activity. 2° ADH catalyzed propan-2-ol
oxidation rates were determined at 37°C (A), 50°C (0), 60°C (0), 70°C a) and 75°C (0) in
GuHCl concentrations from 0.0 to 1.6 M.

143

 

 

 

100

q
0
8

A589:— wEEEB .3333. 95025

 

 

 

GuHCl concentration (M)

144
DISCUSSION

These data provide quantitative evidence that the T. ethanolicus 2° ADH is a highly
thermophilic and thermostable enzyme with catalytic activity that is also robust to high
levels of chemical denaturants. The enzyme was rapidly thermoinactivated only at
temperatures above 90°C. Enzyme thermoinactivation was modeled by a pseudo-first order
rate equation suggesting that the rate determining step was unimolecular with respect to
protein. The large thermoinactivation transition state enthalpy and entropy changes are
consistent with a significant protein structural change during this rate limiting step.
Enzyme activity loss upon heating was accompanied by precipitation, but the soluble
enzyme remaining after partial sample inactivation retained complete activity,
mechanistically linking enzyme inactivation and precipitation. Finally, the increased
GuHCl concentration necessary for enzyme rate enhancement at higher temperatures was
also seen with certain inorganic salts arguing that this 2° ADH catalytic property results
from specific ionic and not general hydrophobic interactions.

The highly thermophilic T. ethanolicus 2° ADH has a projected half-life of 2 months
at 70°C which compares to the complete inactivation of the structurally similar mesophilic
Closm'diwn beijen'nckii enzyme [23,33] is also more stable than the T. brocla'i 2° ADH
[33a] that differs by only 3 amino acids. Tomazic and Klibanov proposed that enzyme
molecules reversibly enter structural states that can lead to essentially irreversible protein
aggregation and inactivation [34]. With respect to protein, this process involves an initial
intramolecular structural change followed by an intermolecular aggregation step.
Thermoinactivation of the T. ethanolicus enzyme follows a pseudo-first order rate law,
suggesting that the rate is independent of the protein concentration as reported for a-
amylase [34] and D-xylose isomerase [35]. Thus, the process appears to be unimolecular
with respect to 2° ADH protein. The soluble enzyme recovered after incomplete enzyme
thermoinactivation retained specific activity comparable to the unheated enzyme, implying

145
complete inactivation of part of the enzyme population and not partial inactivation of the

entire enzyme population. These data suggest that catalytically compromised 2° ADH
intermediate structures did not persist, consistent with a rapid second step in irreversible
thermoinactivation (precipitation). The absence of detectible enzyme activity in the
precipitated protein demonstrated that loss of catalytic activity either accompanied or
preceded protein precipitation. Therefore, the observed unimolecular T. ethanolicus 2°
ADH thermoinactivation rate limiting step preceding precipitation of structurally scrambled
protein molecules fits the two step Tomazic and Klibanov model [34] for protein
thermoinactivation with a slow native to non-native enzyme structural transition followed
by rapid aggregation of thermoinactivated enzyme. Identifying the exact molecular
mechanism of irreversible 2° ADH thermoinactivation will require precise measurement of
inactivated protein molecular mass or detection of covalent destruction products like
ammonia to determine whether this rate limiting step involves covalent protein modification
(eg., deamidation) or partial loss of native enzyme structure.

The inactivation slow step transition state thermodynamics were evaluated to
determine the stabilizing energy for the 2° ADH. The transition state enthalpy, almost 20-
fold that reported for enzymatic catalysis [23], is lOO-fold greater than the expected
molecular kinetic energy (~RT; where R = 8.341 J mol'1 K4, and T is the system
temperature in Kelvin). Calculating the astmbmty from absolute rate theory requires the
assumption that the inactivated, partly unfolded protein is an intermediate following the
slow step transition state which is itself in rapid equilibrium with the native protein
structure. Assuming this mechanism, the 2° ADH is stabilized by a relatively small energy
difference (110 kJ mol'l) between large enthalpic (440 kJ mol'l) and entropic (330 1:] mol'
1 at 393 K) contributions as described for other proteins [36]. The AI-Ii therefore, appears
to be an effective barrier to 2° ADH thermoinactivation opposing a large destabilizing

transition state entropy change.

146
Unfolding of T. ethanolicus 2° ADH was measured to examine the nature of this

slow, inactivating 2° ADH structural transition. The Tmmu] data from absorbance
measurements predicted that a GuHCl concentration of 3.75 M would yield an enzyme
Tm[Gu] value of 25 °C. This predicted value was corroborated by Trp fluorescence
measurements. 2° ADH melting measured by both absorbance and fluorescence indicated
an essentially linear Tm[Gu] dependence on GuHCl concentration. Extrapolatin g this linear
dependence of enzyme Tm[Gu] on GuHCl concentration to zero GuHCl predicts a Tm value
of 115 3‘- 5.8°C. This temperature is 20°C higher than the midpoint temperature for 2° ADH
precipitation (~95°C) arguing that the precipitating protein is not significantly unfolded.
Consequently, the high thermal stability of T. ethanolicus 2° ADH is associated with its
high resistance (i.e., rigidity) to chemically and thermally induced unfolding.

Biophysical measurements have indicated that enzymes are more rigid at low
temperatmes than at high temperatures within their active temperature ranges [25.36.37].
However, the hypothesis that poor low temperature enzyme activity results from excessive
structural rigidity [32] has been challenged based on the general Arrhenius theory
adherence of enzyme catalytic rate temperature dependence [24,38] and specifically based
on the temperature activity behavior of the T. ethanolicus 2° ADH [25]. Unlike other
enzymes, the 2° ADH was totally folded at 90°C where it was optimally active. Here we
attempted to alter 2° ADH rigidity by the addition of GuHCl. The additive denaturing
effects of GuHCl and temperature on T. ethanolicus 2° ADH Tm[Gu] predict that at higher
temperatures lower GuHCl concentrations should be required to reduce enzyme rigidity,
thus lower (3qu concentrations should result in optimal enzyme flexibility and activity.
While added GuHCl did significantly enhance enzyme activity, the effect was equally
pronounced at low and at high temperatures. The slightly increased denaturant
concentration corresponding to maximal 2° ADH activity at higher temperatures is
inconsistant with the lower denaturant concentration predicted to yield optimal enzyme

rigidity at higher temperatures. Similar 2° ADH activity enhancement was seen with low

147
concentrations of inorganic salts which are not strong protein denaturants, supporting the

conclusion that ionic interactions and not optimal rigidity were responsible for the observed
increase in the catalytic rate.

Industrial biocatalysts must have activities robust to a broad range reaction
conditions. Both chemical and thermal stability are therefore important properties of a
potential industrial chiral catalyst. The T. ethanolicus 2° ADH is active from 30°C to over
90°C with active half-lives greater than 1 hr to temperatures exceeding 90°C. GuHCl
concentrations exceeding 1.0 M were necessary to eliminate enzyme activity and this
thermophilic 2° ADH was also highly active in ethanol concentrations greater than 2.0 M
(data not shown). Therefore, the extremely high thermal and chemical stabilty of T.
ethanolicus 2° ADH activity, demonstrates that this is an extremely robust enzyme catalyst.
As a chiral chemical catalyst, the effect of temperature on T. ethanolicus 2° ADH chirality
has been shown to be reaction specific [39]. The effect of chemical denaturants on 2° ADH
product chirality however, remains to be determined.

ACKNOWLEDGMENTS

I gratefully acknowledge Dr. Vladimir Tchemajencko for conducting the structural melting
experiments. I also gratefully acknowledge Maris Laivenieks for his assistance in protein

purification. This research was supported by a grant from the Cooperative State Research
Service, US Department of Agriculture, under the agreement 90-34189-5014.

10

11

12

13

14

15

148
REFERENCES

Whitesides, G. M. and Wong, C.-H. [1983] Enzymes as catalysts in organic
synthesis. Aldrichimica Acta 16, 27-34

Rosazza, J. P. N. [1995] Biocatalysis, microbiology, and chemistry: The power of
positive linking. Amer. Soc. Microbial. News 61, 241-245

Meyer, H.-P. [1991] Microbiology in the contemporary organic chemical
industry. Biol. Ferm. Eng. 8, 602-606

Wong, C.-H. [1989] Enzymatic catalysts in organic synthesis. Science 244,
1 145-1 152

Jones, J. B. [1986] Enzymes in organic synthesis. Tetrahedron 42, 3351-3403

Irwin, J. B., Lok, K. P., Huang K. W. C. and Jones, J. B. [1978] Enzymes in
organic synthesis. Influence of substrate structure on rates of horse liver alcohol
dehydrogenase-catalyzed oxidoreductions. J. Chem. Soc. Perkin I 12, 1636-1641

Bradshaw, C. W., Hummel, W. and Wong, C.-H. [1992] Lactobacillus kefir
alcohol dehydrogenase: a useful catalyst for synthesis. J. Org. Chem. 57, 1532-
1536

Bradshaw, C. W., Fu, H., Shen, G.-J. and Wong, C.-H. [1992] A Pseudomonas
sp. alcohol dehydrogenase with broad substrate specificity and unusual
stereospecificity for organic synthesis. J. Org. Chem. 57, 1526-1532

Hummel, W. and Kukla, M. R. [1989] Dehydrogenases for the synthesis of
chiral compounds. Eur. J. Biochem. 184, l-13

May, SW. and Padgette, SR. [1983] Oxidoreductase enzymes in biotechnology:
current status and future potential. Bio/Technology 1, 677-686

Faber, K. and Franssen, M. C. R. [1993] Prospects for the increased application of
biocatalysts in organic transformations. Trends Biotechnol. 11, 461-470

Brandén, C.-I., Jornvall, H., Eklund, H. and Fururgren, B. [1975] Alcohol

dehydrogenases, in The enzymes, 331 ed., vol. XI part A (P. D. Boyer, ed.)
pp. 105-106. Academic Press, New York

Klibanov, M. [1983] Biotechnological potential of the enzyme hydrogenase.
Proc. Biochem. 18, 13-23

Zong, M.-H., Fukui, T., Kawamoto, T. and Tanaka, A. [1991] Bioconversion of
organosilicon compounds by horse liver alcohol dehydrogenase: the role of the
silicon atom in enzymatic reactions. Appl. Microbiol. Biotechnol. 36, 40-43

Nagata, Y., Maeda, K. and Scopes, R. K. [1992] NADP-linked alcohol
dehydrogenases from extreme thermophiles: Simple affinity purification schemes,
and gomparative properties of the enzymes from different strains. Bioseparation 2,
353- 62

16

17

18

19

20

21

22

23

24

25

26

27

28

149

Keinan, B., Hafeli, E. K., Seth, K. K. and Lamed, R. L. [1986] Thermostable
enzymes in organic synthesis. 2. Asymmetric reduction of ketones with alcohol
dehydrogenase from Thermoanaerobium brockii. J. Am. Chem. Soc. 108,
162-169

Burdette, D. S. and Zeikus, J.G. [1994] Purification of acetaldehyde
dehydrogenase and alcohol dehydrogenases from Thermoanaerobacter ethanolicus
39B and characterization of the secondary-alcohol dehydrogenase (2° ADH) as a
bifunctional alcohol dehydrogenase-acetyl-CoA reductive thioesterase. Biochem. J.
302, 163-170

Zheng, C., Pham, V. T. and Phillips, R. S. [1992] Asymmetric reduction of
ketoesters with alcohol dehydrogenase from Thermoanaerobacter ethanolicus.
Bioorg. Med. Chem. Let. 2, 619-622

Keinan, B., Seth, K.K. and Lamed, R. [1986] Organic synthesis with

enzymes. 3. TBADH-catalyzed reduction of chloro ketones. Total synthesis of (+)-
(S ,S)-(cis-6-methyltetrahydropyran-2-yl)acetic acid: a civet constituent. J. Am.
Chem. Soc. 108, 3474-3480

Keinan, B., Seth, K.K., Lamed, R., Ghirlando, R. and Singh, SP. [1990]
thermostable enzyme in organic synthesis, 4. TBADH-catalyzed preparation of
bifunctional chirons. Total synthesis of S-(+)-Z=tetradec-5-en-13-olide.
Biocatalysis 4, 1-15

Peretz, M. and Burstein, Y. (1989) Amino acid sequence of alcohol
dehydrogenase from the thermophilic bacterium Thermoanaerobium brockii.
Biochemistry 28, 6549-6555

Lamed, R.J., Keinan, E. and Zeikus, J.G. [1981] Potential applications of
an alcohol-aldehyde/ketone oxidoreductase from thermophilic bacteria. Enz.
Microbiol. T echnol. 3, 144-148

Burdette, D.S., Vieille, C. and Zeikus, J .G. [1996] Cloning, expression, and
biochemical characterization of the thermophilic secondary-alcohol dehydrogenase
from Thermoanaerobacter ethanolicus 39E. Biochem. J ., in the press

Vieille, C., Burdette, D. S. and Zeikus, J. G. [1996] Thermozymes, in
Biotechnology Annual Reviews, vol. 2, (M. R. El Geweley, ed.) pp. 1-83.
Elsevier Press, Amsterdam, Neth.

Bbhm, G. and Jaenicke, R. [1994] Relevance of sequence statistics for the
properties of extremephilic proteins. Int. J. Peptide Protein Res. 43, 97-106

Brooks, S. [1992] A simple computer program with statistical tests for the analysis
of enzyme kinetics. Biotechniques 13, 906 - 911

Chang, R. [1977] Physical chemistry with applications to biological systems.
MacMillan Pub. Co., NY

Hermans, J. and Scheraga, H. A. [1961] Structural studies of ribonuclease V.
Reversible changes of configuration. J. Am. Chem. Soc. 83, 3283-3292

29

30

31

32

33

33a

34

35

36

37

38

39

150

Tiktopulo, E. I. and Privalov, P. L. [1974] Heat denaturation of ribonuctease.
Biophys. Chem. 1, 349-357

Klump, H., Di Ruggiero, J., Kessl, M., Park, J.-B., Adams, M. W. W. and
Robb, F. T. [1992] Glutamate dehydrogenase from the hyperthermophile
Pyrococcus furiasis J. Biol. Chem. 267, 22681-22685

Remington, R. D. and Schork, M. A. [1985] Statistics with applications to the
biological and health sciences, Prentice-Hall, Engelwood Cliffs, NJ

Wrba, A., Schweiger, A., Schultes, V., Zévodszky, P. and Jaenicke, R. [1990]
Extremely thermostable D—glyceraldehyde-3-phosphate dehydrogenase from the
eubacterium Thermatoga matin'ma. Biochemistry 29, 7584-7592

Ismaiel, A. A., Zhu, C.-X., Colby, G. D. and Chen, J .-S. [1993] Purification and
characterization of a primary-secondaryalcohol dehydrogenase from two strains of
Clastridium beijerinckii. J. Bacteriol. 175, 5097-5105

Lamed, R. J. and Zeikus, J. G. [1981] Novel NADP-linked alcohol
aldehyde/ketone oxidoreductase in thermophilic ethanologenic bacteria. Biochem J.
195, 183-190

Tomazic, SJ, and Klibanov, AM. [1988] Mechanisms of irreversible thermal
inactivation of Bacillus a-amylases. J. Biol. Chem. 263, 3086-3091

Meng, M., Bagdasarian, M. and Zeikus, J. G. [1993] The role of active-site
aromatic and polar residues in catalysis and substrate discrimination by xylose
isomerase. Biotechnology 11, 1157-1161

J aenicke, R. [1991] Protein stability and molecular adaptation to extreme
conditions. Eur. J. Biochem. 202, 715-728

Fontana, A. [1990] How natrne engineers protein (thermo) stability, in Life under
extreme conditions: Biochemical adaptation (G. di Prisco, ed.) pp. 89-113.
Springer-Verlag, Heidelberg.

Tchemajenko, V., Vieille, C., Burdette, D.S., and Zeikus, J .G. [1996]
Physicobiochemical studies of xylose isomerases from thermophilic and mesophilic
microorganisms. Manuscript in preparation

Phillips, R. S., Zheng, C., Pham, V. T., Andrade, F. A. C. and Andrade, M. A.
C. (1994) Effects of temperature on the ctereochemistry of enzyme reactions.
Biocatalysis 10, 77-86

Chapter IV

Mutagenic and Biophysical Analysis of Thermoanaerobacter

ethanolicus Secondary-Alcohol Dehydrogenase Activity

Prepared for submission to the Journal of Biological Chemistry

151

152
ABSTRACT

The Thermoanaerobacter ethanolicus 398 ath gene encoding the secondary-
alcohol dehydrobenase (2° Adh) was overexpressed in Escherichia cali (DH5a) at ~15% of
total protein. Recombinant enzyme was purified in high yield (67%) by simple heat
treatment at 85°C and ammonium sulfate precipitation. Purified site directed 2° Adh
mutants Cys37 to Ser, His59 to Asn, Asp150 to Asn, Asp150 to Glu, and Asp150 to Cys
were analyzed to test the peptide sequence comparison based predictions of amino acids
responsible for putative catalytic Zn binding. X-ray absorption spectrometry confirmed the
presence of a protein bound Zn atom (spectral transition at ~9662 eV) with a ZnSl(N-O)3.4
coordination sphere. Induction coupled plasma emmission spectrometry measured 0.48 i
0.021 Zn atoms per wild type 2° Adh subunit. The Cys37 to Ser, HisS9 to Asn, and
Asp150 to Asn mutant enzymes bound only 0.11, 0.13, and 0.33 Zn per subunit,
respectively, suggesting that these residues were involved in Zn liganding. The Asp150 to
Glu and Asp150 to Cys mutants retained 0.47 and 1.2 Zn atoms per subunit, suggesting
that an electron rich sidechain moiety at this position preserves the bound Zn. All five
mutant enzymes had 5 3% of wild type catalytic activity, indicating that the T. ethanolicus
2° Adh requires a catalytic Zn atom. Also, the Hi559 and Asp150 mutations altered 2° Adh
affinity for propan-2-ol over a 140-fold range while the overall change in affinity for
ethanol spanned a range of only 7-fold, supporting the importance of the metal in 2° ADH
substrate binding. The lack of significant changes in cofactor affinity due to these catalytic
Zn ligand mutations suggested that 2° ADH substrate and cofactor binding are structurally
distinct. Altering Gly198 to Asp reduced the enzyme specific activity 2.7-fold, increased
the Kmapp for NADP+ 225-fold, and decreased the Kmapp for NAD+ 3-fold, supporting
the prediction that this 2° Adh binds cofactor in a common nicotinamide cofactor binding
motif, a Rossmann fold. Therefore, these mutant enzyme data indicate that, unlike the liver
1° ADH, the NADP(H) linked T. ethanolicus 2° Adh binds its catalytic Zn atom using a

153
novel Cys-His-Asp motif, it does not bind a structural Zn atom, and it uses a Rossmann

fold to bind nicotinarrride cofactor with G1y198 significantly responsible for its NADP(H)
specificity.

154
INTRODUCTION

Alcohol dehydrogenases (ADHs), central to prokaryotic and eukaryotic
metabolism, are also potential biocatalysts for chiral chemical production [Keinan et al.,
1986a; Keinan et al., 1986b; Hummel, 1990; Keinan et al., 1990, Bradshaw et al.; 1992a,
Bradshaw et al.; 1992b]. The functionally and presumably structurally similar ADHs are
classified as primary or secondary based on their higher catalytic efficiencies toward
primary or secondary alcohols. These typically homodimeric or homotetrameric enzymes
are almost exclusively Zn containing metalloenzymes [Branden et al., 1975] that use
NAD(H)(EC1.1.1.1), NADP(I-l) (EC 1.1.1.2), or both (EC 1.1.1.71) as cofactor. Fe
linked [Scopes, 1983] and ferredoxin Ego-dependent [Bleicher and Winter, 1991] 1°
ADHs have also been reported The catalytic Zn's role in liver 1° ADH activity has been
proposed to involve binding directly to the substrate oxygen atom, facilitating transfer of a
hydride between the adjacent substrate carbon atom and the nicotinamide cofactor [Branden
et al., 1975]. While 1° ADHs from diverse sources have been extensively characterized
[Branden et al., 1975], little is known about the molecular basis for 1° ADH substrate
specificity. Far less is is known about 2° ADH structme—function, with relatively few
reported studies on enzyme purification and characterization [Lamed and Zeikus, 1981;
Steinbiichel et al., 1984; Bryant et al., 1988; Ismaiel et al., 1993; Burdette and Zeikus,
1994] and a complete lack of exact structrn'al or site directed mutagenic information.
Because of the 2° ADI-Is' central metabolic roles [Levitt et al., 1984; Burdette and Zeikus
1994] and their potential biotechnological value [Lamed et al., 1981], structure-function
analysis of catalysis and substrate specificity would contribute significantly to the basic
understanding of enzyme function and to the design of biocatalyst specificity.

The exact 3-dimensional structure of the 1° ADH from horse liver has been
determined [Eklund et al., 1974], indicating that the catalytic Zn atom was bound by two

Cys and a His residue and that the nicotinamide cofactor was bound in a Rossman fold

155
[Branden et al., 1975; Rossmann and Argos, 1976]. Research to date has failed to identify

ADH active site amino acid residues which are specifically responsible for substrate
binding. Instead, substrate appears to bind the catalytic metal directly [Dunn and
Hutchinson, 1973; Jacobs et al., 1974; McFarland et al., 1974]. The presumed similarity
between 1° and 2° ADH structures based on shared catalytic properties provided the
rationale for sequence based comparisons between these enzymes [Jendrossek et al., 1988;
Peretz and Burstein, 1989; Burdette et al., 1996]. Peptide sequence alignments of
thermophilic and mesophilic 1° and 2° ADHs were used to hypothesize that 2° ADHs
similarly bind nicotinamide cofactor in a Rossman fold but that 2° ADHs lack a structural
Zn binding loop and they use a unique Cys-His-Asp motif for catalytic metal liganding
[Burdette et al., 1996]. Chemical modification experiments have implicated Zn, Cys, and
His in Thermaanaerobium brackii and Thermoanaerobacter ethanolicus 39E 2° ADHs
catalysis [Lamed and Zeikus, 1981; Burdette et al., 1996] but validation of the predicted Zn
liganding motif has awaited assessment by biophysical analysis and site directed
mutagenesis. Furthermore, the strong similarities among 2° ADH peptide sequences,
kinetic parameters, subunit compositions, and molecular masses [Nagata et al., 1992;
Burdette et al., 1996] suggest that the results of T. ethanolicus 2° ADH catalytic structme-
function analysis will be generally applicable to other 2° ADHs.

The research reported here describes the construction of a recombinant
overexpression and purification system providing high yields of homogeneous enzyme.
The results of site directed mutagenesis to test sequence comparison based predictions of
amino acid residues important for T. ethanolicus 39E 2° ADH catalysis are also described.
Predictions of catalytic Zn binding ligands and the amino acid responsible for NADP(H)
cofactor specificity are assessed from site directed mutation effects on enzyme Kmapp,
Vmaxapp, and Zn binding. Finally, the mutant enzyme kinetic data on 2° ADH mutants and
exact structural information on 1° ADH are used to create a hypothetical working model for

156
the 2° ADH active site and to distinguish key structure-function differences between 2° and

1° ADHs.

EXPERIMENTAL PROCEDURES

Chemicals and reagents - All chemicals were of at least reagent/molecular biology
grade. Oligonucleotide synthesis and amino acid sequence analysis were performed by the
Macromolecular Structure Facility (Department of Biochemistry, Michigan State
University). The kanamycin resistance GenBlock (EcoRI) DNA cartridge used in
expression vector construction was purchased from Pharmacia (Uppsala, Sweden) [Oka et
al., 1981]. DNA for sequencing was isolated using the Wizard Miniprep kit (Promega;
Madison, WI).

Media and strains - Escherichia coli (DI-15a) containing the 2° ADH recombinant
plasmids were grown in rich complex medium (20 g l'1 tryptone, 10 g 1'1 yeast extract, 5 g
1-1 NaCl) at 37°C in the presence of 25 ug ml:1 kanamycin and 100 ug ml-l ampicillin.

Mutagenesis - All DNA manipulations were performed using established protocols
[Sambrook et al., 1989; Ausubel et al., 1993]. Point mutations were introduced into the
ath gene by PCR [Ausubel et al., 1993] using the T. ethanolicus 39E 2° ADH gene clonal
plasmid pADI-IB25-kan [Burdette et al., 1996] as template DNA. An oligonucleotide
primer (KA4 N-end) was synthesized to bind the noncoding strand and included a Kpnl
restriction enzyme site, the native ath gene ribosome assembly site, and the initiation
codon for the ath gene (Table I). An oligonucleotide primer (KA4 C—end) was
synthesized to bind the coding strand, it included the compliment of the ath termination
codon and an Apal restriction enzyme site. Complimentary 3045 base oligonucleotide
primers that contained the mutated bases were used in conjunction with two KA4 end
primers to amplify the N-terminal and C-terrninal segments of the ath gene. PCR

157

Table I

Oligonucleatide primers for PCR amplification of mutant ath gene DNA

List of PCR primers used in the construction of 2° Adh overexpression plasmids and
genes encoding mutant enzymes. The primers for the 5' and 3' end of the ath gene are listed
with the mutated bases underlined and the restriction sites in italics. Altered bases in the mutation

encoding primers are underlined.

 

Mutation Primer sequence
10.4 '
5'-end 5.CGGGGTACCCCGTA’I’I'ITAGGAGGTG’ITI‘AATGATGAAAGG3'
3'-end 5CAGTCCGGGCCCITATGC’I‘AATA'ITACAACAGG'I'I'I‘G3
lMetl . '
5'-end 5'CGGGGTACCCCGTA'ITI'I‘AGGAGGTGTI’I‘AITAATGAAAGG3
3'-end 5CAGTCCGGGCCC'I‘I‘ATGCI‘AA’I‘A’ITACAACAGG'I'I'l‘G3
Cys37 to Ser ' .
coding 5'GCI‘GTGGCCCCT'I‘QCAC'ITCGGACATI‘CATACC3
noncoding 5GGTATGAATG’I‘CCGAAGTGQAAGGGGC3
HisS9 to Asn ' '
coding S'CAroArAcrcoorAAcoAAocrorM.
noncoding 5C'ITCACCI'ACAGC'I'I‘CGTIACCGAG3
Asp150 to Asn ' '
coding 5.GGAAGCI‘GCACTI‘ATGATTCCCAATATGATGACCACI‘GG3'
noncoding 5GCTCCGTGAAAACCAGTGGTCATCA’I‘A'IIGGGAA’I‘CATA3
Asp150 to Cys . '
coding 5'GGAAGCTGCACITATGATI‘CCCIQTATGATGACCACI‘GG3 '
noncoding 5GCTCCGTGAAAACCAGTGGTCATCATACAGGGAATCATAAG3
Asp150 to Glu . .
coding 5.GGAAGCI‘GCACI‘I‘ATGA’ITCCCGAAATGATGACCACI‘GG?
noncoding 5GC'ICCGTGAAAACCAGTGGTCATCA’III‘CGGGAATCATA3
Gly 198 to Asp

coding

noncoding 5'CACAAACI‘GGTCI‘AC'I‘GICI‘ACGGCAATAA’ITC3'

5'GAA'ITATI‘GCCGTAGACAGTAGACCAG1T1‘GTG3'

 

158
syntheses of partial and complete mutated genes were performed using the Taqplus,

exonuclease containing polymerase (Strategene; La Jolla, CA). All clones were expressed
in pBluescriptII KS(+) with a kanamycin resistance cartridge introduced into the polylinker
EcoRI site. Mutations were verified by DNA sequencing using the method of Sanger et al.
( 1977).

Enzyme purification - The recombinant enzyme was purified from E. coli (DHSa)
aerobically. The pelleted cells from batch cultures were resuspended (0.5 g wet wt. ml'l)
in buffer A (50 mM TriszHCl [pH 8.0], 5 mM DTT, and 10 LLM ZnClz) containing 3 [lg
ml'1 lysozyme. The resuspended cells were incubated on ice for 30 min, frozen in N2
(liq), thawed and centrifuged for 30 min at 15 000g. The clarified lysate was incubated at
85°C for 15 min, cooled on ice for 30 min, and then centrifuged for 30 min at 15 000g.
We have recently shown that T. ethanolicus 2° Adh displayed a half-er of 1.2 days at 80°C
(see chapter 3). Ammonium sulfate {50% (wt/vol)} was added to the supernatant and
stirred at 4°C for 30 min. The 2° ADH was recovered from the supernatant after
centrifugation for 30 min at 15 000g and precipitated with stirring at 4°C for 30 min in 70%
(wt/vol) ammonium sulfate. After centrifugation for 30 min at 15 000g, the purified 2°
ADH was resuspended in buffer A and stored at 4°C.

Zn binding and molecular mass determination - EXAFS determination of 2° ADH
Zn coordination was performed by Dr. Bob Scott (Univ. of Georgia; Athens, GA).
Enzyme samples were concentrated to 1-10 mg ml'1 in 50 mM Tris:HCl (pH 8.0) and
stored in N2 (liq) prior to analysis. 2° ADH samples were dialyzed against 50 mM
NH4HCO3 (pH 8.0) containing 100 g 1'1 Chelex resin (BioRad; Hercules, CA) for 12 hr to
remove unbound or advantitiously bound metals. The metal cleared samples were
concentrated to between 1.2 and 12 mg ml'1 using centricon 30 ultrafiltration units
(Amicon; Beverly, MA) for induction coupled plasma emission spectrometry (ICP)
analysis (Chemical Analysis Laboratory at the University of Georgia, Athens). Subunit

159
molecular mass values were determined by comparison with standards (BioRad; Hercules,

CA) using SDS-PAGE (12% polyacrylamide).

Enzyme kinetics - The standard 2° ADH activity assay was defined as NADP“
reduction coupled to propan-2-ol oxidation at 60°C as previously described [Burdette and
Zeikus, 1994]. The enzyme was incubated at 55°C for 15 min prior to activity
determination unless otherwise indicated. Tris buffer pH was adjusted at 25°C to be pH
8.0 at 60°C (thermal correction factor = -0.031 ApH °C‘1). Assays to determine Km.pp and
Vmaxm, were conducted at 60°C with substrate concentrations between 20me.pp and
0.2mew. Kinetic parameters were calculated from nonlinear best fits of the data to the
Michaelis-Menten equation using Kinzyme software [Brooks, 1992] on an IBM PC.
Protein concentrations were measured using the bicinchoninic acid (BCA) procedure

(Pierce; Rockford, IL).

RESULTS

Enzyme overexpression - T. ethanolicus 2° ADH was purified from E. cali (DHSa)

harboring the pADHBKA4-kan clonal construct (Table II) yielding 67% recovery of
recombinant enzyme expressed at ~15% of total soluble protein without induction (Fig. I).
This construct contained the native ribosome assembly site and translational initiation
codons but removed the native transcriptional promoter regions, placing the gene under
transcriptional control of the pBluescriptKS (+) T3 promoter. The recombinant protein
population included 56 i 11% enzyme with a duplicated N-terminal methionine residue
(MMKGFA...) while 44 i 11% of the protein possessed a single methionine residue at the
N-terminus. Alteration of the "ATGATG" translational initiation site to "TTAATG" in the
pADHB lMl-kan expression plasmid eliminated the synthesis of 2° ADH with two N-
terminal methionine residues but did not alter the level of enzyme expression compared to

160

 

no we .3. an e: S. 5888 comma.
2

 

o2 d g 3a SN mm 8888.: 8o:
02 wé hm EN an 8988 :00
3% Sex. was 2.23 18 m8 .35. m8

.338 8088888 mam

2o; Seduced emooem cream memos Boer need new
do w 8 588888 08.38 888085 2255
oeoh Q SQEE mean m 88H 8:2. voweuwueeo 85 8 3.838.— 83 8208 REES 80383880 .3 @0qu 88
.88 on 88 03 o. 3382 .88 2 8C 0.. mm 8 38388 83 8.88 :8 8F datum—£88 .3 .8588 e5 .832.
.95 N2 5 assoc .o a. e 8e 8 3385 so; Eocene 18 w: m e5 888256 28 n .3 me 688s :8

cm 8 18 SB 83 m 08 £8 Becca—mag 33‘ em mom §ome=§8 .5 2383888 .88 0.8888 5285.85

=3-3¢~th<& 38.83 Eekeemmmxfiu $39 N 8% 3%: 282.5882 85 \e =§8R§m
a 23mm.

161
that seen for pADHBKA4-kan (Fi g. I) or the enzyme specific activities toward ethanol and

propan-2-ol (Table III). The Kmapp values toward propan-Z-ol, ethanol, and NADP+ were
1.1 i 0.22 mM, 53 i 9.0 mM, and 17 i 2.6 uM, for protein expressed from
pADI-IBKA4-kan and 0.87 i 0.32 mM, 37 i 4.2 mM, and 8.5 i 1.4 M for enzyme
expressed from pADHB lMl-kan.

Catalytic Zn liganding - Chemical modification experiments have implicated Zn,
Cys, and His in T. ethanolicus 2° ADH activity. Peptide alignments identified Cys37,
Hi859, and Asp150 as potential catalytic Zn ligands [Burdette et al., 1996]. EXAFS
analysis of the native and wild type recombinant enzyme suggested the presence of a single
type of specifically bound Zn atom in the enzyme (Fig. II). The transition at 9660-9665 eV
in Fig. 11a is characteristic of Zn. The Fourier transformed data (Fig. 11b) is best fit by a
coordination sphere of ZnSl(N,O)3.4. The sulfur signal is consistent with that of a
cysteine residue. The optimal data fit also indicated a single imidazole nitrogen ligand,
leaving 2-3 potential oxygen ligands which could be from an amino acid such as Asp,
water, or a hydroxide ion.

Mutant ath genes encoding amino acid substitutions at residues 37, 59, and 150
were constructed by PCR (Fig. III) to test the involvement of these amino acids in catalytic
Zn binding. Mutant proteins were expressed at levels similar to the wild type enzyme and
were purified to homogeniety as seen on SDS-PAGE (Fig IV). Sequencing the mutated
section of the ath gene confirmed the identity of each mutant. ICP analysis indicated that
the enzyme expressed from the wild type gene (pADHBKA4-kan) and the single N-
terminal Met mutant enzyme bound 0.47 and 0.50 Zn ions per subunit, respectively. The
Cys37 to Ser and His59 to Asn mutant enzymes contained 0.11 and 0.13 Zn per subunit,
respectively. Altering Asp150 to Glu preserved binding of 0.47 Zn per subunit while the
Asp150 to Asn mutant bound 0.33 Zn per subunit. Changing the putative Asp150 ligand
to Cys increased binding to 1.2 Zn per subunit. Finally,the Gly198 to Asp mutant enzyme
also bound 0.55 Zn per subunit. Mutant and wild type enzymes treated for Zn analysis by

162

Figure 1. SDS-PAGE of wild type recombinant 2° Adhs. Lanes: 1, clone 1M1
cell extract: 2, clone 1M1 dialized ammunium sulfate precipitate; 3, clone KA4 cell extract;
4, clone KA4 post heat treatment; 5, clone KA4 dialized ammunium sulfate precipitate.
The Mr values for the molecular mass markers flanking the sample lanes are indicated on
the left. Proteins were stained with Coomassee brilliant blue R-250.

l 2
97kDa--=::=
45 fig-
31 ‘13:!
22 s "

163

-. Il'mlll “

.\~\ 3“

164

5:0 3:88 28 own 8: 28 8K. 8 808: 80.8 88888 88? 088883.?

 

 

 

 

 

 

noon 3.. 8 n02 e: as «be: 2 mm 820
«-00: 88d 86 fiend mm Sad W006 ed end 2%:
03.: Red 8.0 woo: cm 3.0.0 088 8.0 8.0 288
no: :8. 2.0 woe: 9: 82. some 8 and :88
3.. 0886 ed 0.36 c: 86 0.00..“ v; QN Dom:—
~-0o.v Rood wed «.08? c2 8 N36: whos m6 $8 3.8
3 20.0 8 room mm a some 2 me E
888 2: 38 .ME 3 ~88 a: 28 7a“: b 18: a: :8 $8: D
38850 38850 38850
2580 ease: 35> crease ease: are; E ea58>
0835
£512 _ofiem Fetmtfiapa
§> m: 88.8: 83 38850 083000 .28 8h 8: cmm 8.0383 8088:0080
3:88 8:0 m8? 888.880 803 8:88 mummy 08 .3 88:88 3888 :8 > 2: 808% 0888088

2:. .880 :0 8 088388 05 a: _0.N-:mqe.8 .83 888.88: 0.83 88:.» 8880— 3942 .8888 05 0:
+32 88: 888, 88> 8:8: “8.3380 3023 88:8 DwEO 05 «a 00:0 05 8 8000.0 0908.8 tan—<2 .83
8:888“. 083 8: > 8880. .0888 0:: 3.8588 8§~ 8 0:83:80 08 83 8088:0080 888:...

88.. 88 50.8 o. .8588 meme as: .80 as ea 88 a a: 2:8 8528 B 8888

083 8.83 on08> 8: 8.08M .980 ~.28 an.» n 858$ Do 8 8 88888.: LAX—<2 2 080 00:03.83:
8 00.888 08 .3 88588 0:03 808 888 08.38“ mum—8:0 083:0 88:8 8.8 80888:: 0:08M

.3880 See... N g :0 30282.35. B00 08:8 08008.. 8&0 80%.8
E 03mm.

165

Figure 2. T. ethanolicus 2° ADH EXAFS analysis. (A) Spectrograph of
normalized signal intensity versus energy for native (—) and rewmbinant (- - -)2° Adhs.
(B) Plot of Fourier transformed data versus distance for native (—) and recombinant

(- - -)2° Adhs.

uni—L

166

 

|_"——'r .SAdH. recombinant B—KG' 'sdbtraotad "———'_T| TADOB
‘ "-73. 8AdH.aslsolated.ADOA(08194) _B(05195 ‘

 

J

 

Normalized Intensity

 

 

 

 

] J J I "'
9640 ' 9660 9680 9700 9720
Energy (eV)

 

 

 

 

167

Figure 3. PCR-based ath gene site directed mutagenesis scheme. The partial
gene PCR products from reactions using overlapping coding- and noncoding-strand
oligonucleotides encoding the mutation with one complimentaty to the 5' or 3' end were
combined. anealed, extended, and amplified in a second PCR reaction which yielded the
complete mutant (1th gene.

 

168

 

 

 

 

 

 

 

 

 

 

 

Kpn I
\—‘ mut. _‘
IIWéEiPEEIéIQEIEII IIEVEQEYZ'PEEHEIEIEI‘EII
vmut
Apa I
Amplify (26 cycles) 1
K pn I
mutant N—end mutant C—end
Apa I
Aneal (50°C)

K pn I

A I

+ P0
KP" I Apa I
lExtend (3 cycles)
K pn I
Complete mutant
7?)?!
Amplify (26 cycles)
K pn I
Complete mutant

 

Apa I

169

Figure 4. SDS-PAGE of 2° ADH mutant proteins. Lanes: 1, cell extract of E. coli
DHer containing pBluescriptII-KS (+) plus the kanr cartridge; 2, C378 mutant cell extract;
3, purified C378 mutant; 4, purified D150C mutant; 5, purified DISOE mutant; 6, purified
Gl98D mutant; 7, purified HS9N mutant. The Mr values for the molecular mass markers
flanking the sample lanes are indicated on the left. Proteins were stained with Coomassee
brilliant blue R-250.

 

as
E
l

.5
UI

to
N

y.

.
' n
C
.u 4 '.
.._0 ~
e
,
n - ‘
. .
v‘ .
.
. .

I

.9:
r r r r

170

 

l7 1
dialysis in buffer containing metal chelating resin were active and indicated no increase in

activity upon Zn addition (data not shown).

Point mutations altering Cys37, His59, and Asp150 significantly reduced enzyme
activity (Table III). The Cys37 to Ser mutation replaced the putative Zn ligand residue with
one less than 0.4 A shorter, but the Ser hydroxyl's ionization constant is 7 orders of
magnitude smaller than that of the Cys sulfhydryl. This mutant retained less than 1% of the
wild type enzyme activity toward both ethanol and propan-Z-ol. Alteration of HisS9 to Asn
eliminated the ionizable nitrogen atom and reduced enzyme activity to less than 0.5% of
wild type. The Asp150 to Cys mutant which presumably mirrored the Cys-His—Cys motif
for 1° ADH catalytic Zn binding, had the highest specific activity of the catalytic Zn
mutants, retaining approximately 3% of the wild type enzyme activity. Replacement of the
Asp carboxyl moiety with a sulfhydryl group preserved the strong negative ionic character
but shortened the amino acid side chain by approximately 0.6 A Mutating Asp150 to Glu
lengthened the side chain approximately 1.3 A and reduced the specific activity to 0.2% of
wild type. Both the Asp to Cys and Asp to Glu mutations retained the electronic
characteristics for metal binding but mutation of Asp150 to Asn significantly reduced the
polarity of this residue, maintaining the wild type residue geometry. The Asp150 to Asn
mutant retained less than 3% of the wild type activity.

Cofactor and substrate specificity - All of these catalytic Zn liganding residue
mutants had similarly high affinities for NADP" (Kmapp values less than 40 uM) and, like
the wild type enzyme, demonstrated very low activities using NAD‘”. The Asp150 to Asn
mutant Kmapp value for NADP+ was 37 “M with an unusually large standard deviation of
50%, making the difference between it and the wild type Kmapp (16 uM) insignificant.
The Asp150 to Cys mutant however, had an NADP+ Kmapp value near 1.0 M which is
16-fold lower than that of the wild type enzyme.

Weringa and Hoi, 1983, identified a specific amino acid position responsible for
confering NAD(H) versus NADP(H) cofactor specificity in Rossman fold containing

172
proteins. Glyl98 was predicted to occupy this position in the T. ethanolicus 2° ADH

sequence [Burdette et al., 1996] and is consistent with the enzyme's ability to bind
NADP(H) since its small uncharged sidechain would accomodate the additional cofactor
ribose phosphate, unlike the Glu or Asp residues present in N AD(H) linked enzymes that
repel the negatively charged phosphate moiety. Based on Weringa and Hoi's hypothesis,
mutation of Glyl98 to Asp would exclude N ADP(H) from the cofactor binding site but
allow NAD(H) binding. The wild type enzyme had a l40-fold lower Kmapp for NADP+
than for NAD+ and a catalytic efficiency for propan-2-ol oxidation 2400 fold greater using
NADP+ than using NAD+ (Table N). Mutation of Glyl98 to Asp increased the Kmapp for
NADP+ 225-fold and reduced the Vmaxapp using NADP" 4—fold. The mutant enzyme had
a 3-fold lower Kmapp toward NAD+ and a 5.5-fold higher propan-2-ol oxidation rate using
NAD+ compared to wild type, giving the mutant a 6.7-fold lower catalytic efficiency for
NADP“ linked propan—2-ol oxidation than for the NAD+ dependent reaction.

Mutational effects on enzyme affinities for 1° versus 2° alcohol substrates and the
catalytic efficiencies for their oxidation indicate a role for these putative catalytic Zn ligands
in substrate specificity as well as catalytic rate enhancement. Mutation of the Cy837 to Ser
increased the Kmapp 6-fold and 20-fold for propan-2-ol and ethanol, respectively (Table
III). Determining the Cy837 to Ser mutant enzyme Michaelis-Menten constant for ethanol
was complicated by the inability to measure enzyme initial rates using substrate
concentrations greater than 700 mM and less than 300 mM. Therefore, this Kmapp is
presented for completeness but it is an approximate value and is not included in subsequent
data analysis. Alteration of HisS9 to Asn caused a 6-fold increase in Kmapp toward
propan-2-ol and no significant change in the Kmapp toward ethanol. The Asp150 to Cys
mutant Kmapp for propan-2-ol was similar to the wild type value and that for ethanol was
increased only 2-fold. Mutating Asp150 to Glu caused a 3-fold increase in Kmapp for
ethanol but a 29-fold increase in Kmapp toward propan-Z-ol. Finally, the Asp150 to Asn
mutant enzyme had a Kmapp value toward propan-2-ol S-fold lower than the wild type and

173

 

25 e3 2 3.. 3 .. 93%

 

 

 

 

 

83 Ga v6 086 Nb
+92 :«32 :5 we: a 25 was a
e... IE and? IE tie - the
35850
queoD +Q<z tun—<2

 

 

ace 838 58 ca get accuse... 858 2.. e

coup— 2: 3 Bacon 83 35850 36¢on .880 as 5 8.3358 05 3 e3: ma? 3. announ—

eeeué e 2.9338 2e .23 889588 838 was deamén 9 awed—Ad

mean 388 89a 88 88 3:5 9 am 958 30528 .3 vogue 803 82g Edna—Er 98

.95. 5.5 7:8 «me u Scene 0. 8 e 8832 1592 9 e2. 85.5% e eases
05 .3 35808 803 8.5 325 gm dew—88 08.95 88:8 Bu guage cue—5—

8:8»er Snake.» memes—€83 53:. m 3% .8 .8383: Rev. S «3wa as \e Eebw
Z osah

174
a Kmapp value toward ethanol 4-fold greater than the wild type enzyme. The mutant

enzyme catalytic efficiencies were lower than those for the wild type enzyme in all cases but
the ratio of these values for propan-Z-ol oxidation compared to ethanol oxidation varied
from 4—fold lower than (the Asp150 to Glu mutant) to lOJold higher than (the Asp150 to
Asn mutant) that calculated for the wild type enzyme.

DISCUSSION

The research described here provides direct evidence for the molecular basis for
specific amino acid residues in T. ethanolicus 39E 2° ADH catalysis. The presence of a
single specifically bound Zn ion per enzyme subunit and the identities of at least two of the
three potential liganding protein atoms were confirmed by EXAFS measurements.
Induction coupled plasma emission spectrometry (ICP) analysis of wild type and mutant
enzymes implicated Cys37, HisS9, and Asp150 in catalytic Zn binding. Kinetic analysis of
mutant enzymes implicated Cy337, HisS9, and Asp150 in both catalytic rate enhancement
and substrate specificity. Transformation of the highly NADP(H) specific T. ethanolicus 2°
ADH into a NAD(H) preferring enzyme with the Glyl98 to Asp mutation further validates
the 1° to 2° ADH sequence comparisons, provides experimental evidence that this 2° ADH
binds nicotinamide cofactor in a Rossmann fold, and makes the T. ethanolicus enzyme a
more suitable as an industrial biocatalyst. Furthermore, construction of the overexpression
system provides sufficient yield of purified enzyme for structural analysis and potential
industrial scale biocatalyst production.

The original T. ethanolicus 39E 2° ADH clone was expressed in E. coli DHSa
under the control of its native transcriptional promoter at ~2% of total protein [Burdette et

al., 1996]. Removal of the native transcriptional promoter region, placing the (1th gene

175
under control of the pBluescriptII KS (+) T3 promoter (the pADI-IBKA4—kan construct)

increased recombinant protein expression to ~15% of total protein. This indicates that the
native T. ethanolicus promorer is not as strong as the T3 promoter in E. coli. Thermophilic
ADH purification using only 85°C treatment and ammonium sulfate precipitation reduced
enzyme losses to 33% while also eliminating time consuming and costly chromatography
steps. Protein expressed from the pADHBKA4—kan plasmid, like the original clone
[Burdette et al., 1996], consisted of a mixed recombinant enzyme population composed of
protein with one or two N-terminal Met residues. However, recombinant protein with only
a single N-terminal Met, like the mature enzyme isolated from T. ethanolicus 39E, was
detected in the population expressed from pADHBlMl-kan. Thus, altering the
"ATGATG" translational initiation site in the ath gene to "'ITAATG" resulted in the
production of a homogeneous, native-like, recombinant enzyme population. The
homogeneous and mixed recombinant enzyme populations were kinetically
indistinguishable as previously reported for enzyme isolated from T. ethanolicus and the
mixed recombinant protein [Burdette et al., 1996]. This provides direct evidence that the
additional Met did not significantly alter recombinant enzyme expression or function.

The prediction that Glyl98 was in part responsible for T. ethanolicus 2° ADH
N ADP(H) specificity was based on the hypothesis that the cofactor was bound in a
Rossmann fold [Burdette et al., 1996]. Identification of a Rossmann fold consensus
sequence in the 2° ADH peptide supported this hypothesis. However, the dramatic
decrease in enzyme affinity for NADP" resulting fiom mutation of Glyl98 to Asp is the
first experimental evidence for the prediction that T. ethanolicus 39E 2° ADH binds
nicotinamide cofactor in a Rossmann fold. The strong primary structural similarity
between 2° ADI-Is argues that this is a common 2° ADH feature. Mutation of Glyl98 to
Asp only increased enzyme affinity for N AD+ 3-fold, but this is still 48-fold lower affinity
than the wild type enzyme for NADP". Therefore, the negatively charged Asp residue
appears to repel the added NADP+ phosphate moiety, consistent with theory [Weringa and

176
Hoi,1983], but does not significantly decrease the T. ethanolicus 2° ADH's high Kmapp

toward NAD“: While the Glyl98 to Asp mutation increases the catalytic efficiency ratio
for N AD+ reduction to NADP+ reduction 16000-fold, the mutant enzyme Vmaxapp for
catalysis using either NAD+ or NADP+ is lower than the wild type enzyme's Vmaxapp for
NADP+ dependent propan-2-ol oxidation, suggesting that mutant enzyme-cofactor
complex is suboptimal for turnover. The Glyl98 to Asp mutant's Kmapp values for both
ethanol and propan-2-ol were similar to those of the wild type enzyme, suggesting that this
difference is not important to alcohol substrate affinity.

With few exceptions, ADHs require an enzyme bound Zn ion for catalysis. Two
Cys residues and a single His residue act as the catalytic Zn ligands in 1° ADHs [Branden er
al., 1975]. Reactivation of T. ethanolicus 2° ADH after sulfhydryl modification was
enhanced by Zn suggesting that enzyme activity may be Zn dependent [Btn'dette et al.,
1996] however, confirmation of this observation requires more rigorous structural
analysis. The wild type recombinant enzymes (expressed from plasmids pADI-IBKA4-kan
and PADHBlMl-kan) were shown by ICP analysis to bind one 211 for every two subunits
suggesting that the 2° ADH possesses a single competent active site per dimer.
Determining the accuracy of this prediction however, requires exact enzyme 3-dimensional
structural information. The similar Zn content observed in the Glyl98 to Asp mutant (0.55
Zn per subunit) suggests that this mutation does not interfere wth Zn binding and that the
Zn bindng site structure is retained. This analysis supports the sequence based prediction
that the 2° ADH subunits do not contain a Cys liganded structural Zn atom analogous to that
in the 1° ADHs [Peretz and Burstein, 1989; Burdette er al., 1996]. Tire EXAFS results for .
this enzyme identify a specifically bound Zn ion with sulfur, nitrogen, and oxygen ligands.
The coordination sphere {Zn81(N ,O)3.4} measured for the 2° ADH would be expected to
have a much larger number of sulfur ligands (5) if both a catalytic and a structural Zn (with
coordination sphere of 21184) were present. Thus, the EXAFS data further support the
predicted lack of a 2° ADH structural Zn. Although calculations hour the EXAFS spectra

177
indicated 1 or 2 possible sulfur atoms coordinated to the enzyme bound Zn, the best fit to

the data is for a single sulfur ligand. The EXAFS data also indicated the presence of a
single imidazole nitrogen ligand to the Zn atom. This result is consistent with peptide
sequence alignments and chemical modification experiments that have implicated Cys and
His residues in 2° ADH catalytic 2n binding [Burdette et al., 1996]. Finally, the EXAFS
analysis is is consistent with the peptide alignment-predicted Cys-His—Asp catalytic Zn
liganding motif among 2° ADHs.

Site directed mutations of the predicted Cys and His residues to eliminate them as
potential Zn ligands reduced enzyme activity to less than 1% of wild type. Also, the molar
ratio of bound Zn per subunit for these mutants was 4-fold lower than for wild type
enzyme. Therefore, both Cys37 and HisS9 are important to catalysis and to Zn binding.
The protein samples prepared for ICP measmements displayed enzyme activity and 10 M
added Zn did not enhance catalytic rates in the standard assays. Zn contaminated assay
components could have provided sufficient metal to reconstitute the mutant enzyme catalytic
sites, explaining the observed activity. ICP measurements indicating that ~10% of the
protein subunits bound metal after dialysis however, suggest that the low mutant catalytic
activities (<1% of wild type enzyme activity using either substrate) displayed by these
mutants did not result simply from the loss of Zn. The altered mutant enzyme affinities for
substrates also support this conclusion. Mutation of Asp150 to Asn also reduced enzyme
activity toward both ethanol and propan-2-ol to less than 1% of wild type indicating the
importance of Asp150 in 2° ADH catalysis. Zn binding by the Asp150 to Asn mutant was
one third less than wild type Zn binding. Furthermore, the mutant catalytic activity was
significantly lower that expected by the mere loss of Zn in 33% of the protein population,
arguing that this resudue, too, is directly involved in Zn binding and in catalysis.
Replacing Asp150 with Glu alters only the carbonyl position, yet this eliminated all but
0.3% of the wild type enzyme activity. This mutant, like the wild type enzyme, bound

0.47 Zn ions per subunit so its low activity is also not due to the lack of bound Zn. If

178
Asp150 provides the third 2° ADH catalytic Zn ligand, then substitution of Cys for Asp150

should have replicated the 1° ADH Cys-His-Cys Zn liganding structure, although the Cys
sidechain is slightly shorter that the Asp sidechain. This Asp150 to Cys mutant retained
only about 3% of the wild type enzyme activity. Fmthermore, this mutant enzyme bound
1.2 Zn atoms per subunit so reduced catalytic activity clearly did not result from loss of Zn.
The enhancement of Zn binding by this mutant may be due to stronger metal binding or
fi'orn the creation of a second competent Zn binding site per dimer. The effects of these
three mutations at residue 150 argue that both the size and ionic character of the Asp
sidechain are critical to enzyme function. Thus, changing the Zn binding site
charactertistics affected enzyme catalysis, providing direct evidence that Cy837, HisS9, and
Asp150 are Zn ligands and that, like 1° ADH activity, proper Zn binding is essential for 2°
ADH catalysis. The effects of mutations at amino acid residues 37, 59, and 150 provide
strong support for the sequence-comparison based predictions, but final confirmation of
this novel Zn binding motif awaits crystallographic determination of the 2° ADH 3-
dimensional structure.

Because the horse liver 1° ADH is so well characterized, it is an excellent
comparative reference for predicting the structures of other ADHs. Reports of 3-
dimensional structmal similarity between 1° and 2° ADI-Is based on peptide sequence
comparisons have suggested that although similarities exist, these enzyme groups may lack
complete structural similarity [Peretz and Burstein, 1989; Burdette et al., 1996]. The
similarities and differences identified here between the horse liver 1° ADH and the T.
ethanolicus 2° ADH are summarized in Fig 5a,b. Keinan et al., 1986a, proposed an
asymmetric 2 lobe T. brockii 2° ADH active site model based on its accomodation of
substrates varying in length but its requirement for the site of dehydrogenation on the first
to the third carbon atoms. The data presented here demonstrate that the T. ethanolicus 2°
ADH is a Zn dependent, Rossmann fold containing enzyme that appears to use a novel
Cys-His-Asp metal binding motif. Furthermore, kinetic analyses of site directed mutants

179
confirm the sequence-comparison based predictions of specific residues affecting cofactor

selectivity and catalysis. Thus, we propose a hypothetical working model for 2° ADH
structure-function studies (Fig. 5b). In addition, the previous discovery [Burdette and
Zeikus, 1994] of a physiological acetleoA linked reductive tlrioesterase activity for this T.
ethanolicus enzyme suggested that the previously hypothesized long lobe [Keinan et al.;
1986a] may be open ended, allowing the coenzymeA portion of the substrate to remain
outside the active site. The active site channel in the horse liver ADH has a short segment
(Fig. 5a), one wall of which consists of the bound nicotinamide cofactor, and a long
segment open at the far end, with the catalytic Zn atom at their junction [Eklund et al.,
1974]. This 1° ADH architecture is consistent with the 2° ADH active site requirements for
a short lobe to accomodate the C1 to C3 substrate segment, but, it also allows very long
moieties on the other end of the ketone.

The kinetic mechanisms proposed for both horse liver [Branden et al., 1975] and T.
brockii [Pereira et al., 1994] ADH catalysis require bound cofactor prior to solvent
substrate binding. The similar T. brockr'i and liver ADH kinetic mechanisms suggest that
cofactor binding is needed for solvent binding by both 1° and 2° ADHs. Analysis of the
horse liver 3-dimensional structure [Eklund et al., 1974] suggests that the cofactor may
function to complete the active site wall. Although exact crystallographic evidence is
needed to confirm this hypothesis, the similar kinetic mechanisms and predicted structural
characteristics suggests that, like the liver 1° ADH active site, nicotinamide cofactor may
complete 2° ADH catalytic pocket formation. Also, the HisS9 and Asp150 mutations
altered 2° ADH affinity for propan-2-ol over a 140-fold range while the overall change in
affinity for ethanol spanned a range of only 7-fold. The greater effects of these mutations
on propan-2-01 than on ethanol affinity suggest that the catalytic Zn's position affects
substrate specificity. The greater secondary alcohol sensitivity to zinc ligand mutations
suggests that substrate binds the 2° ADH catalytic Zn atom, like in the liver 1° ADH
[Branden et al., 197 5], and that 2°-alcohol binding is more susceptible to steric hinderance

180

Figure 5. Comparison of 1° and 2° ADH active site models. (A) The liver 1°
ADH active site model, and (B) The hypothesized 2° ADH active site model incorporating

the mutagenic and biophysical information.

181

 

 

 

 

 

182

 

 

 

183
in the catalytic pocket. The effect of mutations that alter the active site pocket depth on

enzyme specificity must wait for determination of which residues line the bottom of the
active site.

2° ADH have been employed in chiral compound synthesis and racemate resolution
[Keinan et al., 1986a; Keinan et al., 1986b; Hummel, 1990; Keinan et al., 1990; Bradshaw
et al.; 1992a, Bradshaw et al.; 1992b, Zheng et al., 1992]. Recombinant overexpression
of this enzyme at ~15% of total protein followed by purification using only heat and
ammonium sulfate precipitation made possible the economical production of large enzyme
quantities, vital for structural analyses and for its potential utility as a commercial
biocatalyst. Converting the strongly NADP(H) linked T. ethanolicus enzyme to one
preferring NAD(H) through the Glyl98 to Asp mutation created the first reported highly
stable, thermophilic 2° ADH with higher catalytic efficiency using the more thermally stable
NAD(H) cofactor without altering the substrate specificity. Thus, the Glyl98 to Asp
mutant enzyme retains the desirable 2° ADH catalytic properties but reduces the cost of
potential industrial processes by switching enzyme preference to a cofactor form (N AD(H))
that degrades more slowly. The evidence presented here of 2° ADH active site structural
similarity to 1° ADH active site structure has provided a revised thermophilic 2° ADH active
site model based on comparison to the horse liver enzyme. This mutagenic evidence for 2°
ADH- specific molecular structme-function relationships indicates that comparative analysis
between these two enzyme classes may yield insights into the molecular determinants of

both 1° and 2° ADH substrate specificities.

184
ACKNOWLEDGEMENTS

I gratefully acknowledge Dr. Robert Scott for conducting the EXAFS analysis and to Dr.
Robert Phillips for providing enzyme and productive discussions. I also gratefully
acknowledge Maris Laivenieks for his assistance in protein purification. Dr. Claire Vieille,
Cindy Peterson, and Lloyd Hough provided helpful discussions and assistance regarding
2° ADH molecular biology. This research was supported by a grant from the Cooperative
State Research Service, U.S. Department of Agriculture, under the agreement 90-34189-
5014.

185
REFERENCES

Ausubel, F. M., Brent, R., Kingston, R. B., Moore, D. D., Seidman, J. G., Smith,
J. A. and Struhl, K. (1993) Current Protocols in Molecular Biology (Janssen, K.,
ed.), Current Protocols, NY

Bleicher, K. and J. Winter (1991) Eur. J. Biochem. 200, 43-51

Bradshaw, C. W., Fu, H., Shen, G.-J. and Wong, C.-H. (1992) J. Organic Chem. 57,
1526-1532

Bradshaw, C. W., Hummel, W. and Wong, C.-H. (1992) J. Organic Chem. 57,
1533-1536

Branden, C.-I., mevall, H., Eklund, H. and Furugren, B. (1975) in The enzymes,
vol. XI part A (Boyer, P. D., ed.), pp. 103-190, Academic Press, NY

Brooks, 8. (1992) Biotechniques 13, 906-911
Bryant, F. 0., Wiegel, J. and Ljungdahl, L. (1988) Appl. Env. Microbiol. 54, 460-
465

Burdette, D. S. and Zeikus, J. G. (1994) Biochem. J. 302, 163-170
Burdette, D. S., Vieille, C. and J. G. Zeikus (1996) Biochem. J. 313, in press
Dunn, M. F. and Hutchison, J. S. (1973) Biochemistry 12:4882-4892

Eklund, H., Nordstrom, B., Zeppezauer, B., deerlund, G., Ohlsson, 1., Boiwe, T.
and Branden, C.-I. (1974) FEBS Lett. 44, 200-204

Hummel, w. (1990) Appl. Microbiol. Biotech. 34, 15-19

Ismaiel, A. A., Zhu, C.-X., Colby,G. D. and Chen, J.-S. (1993) J. Bacteriol. 175,
5097-5105

Jacobs, J. W., McFarland, Wainer, 1., Jeanmaier, D., Ham, C., Hamm, K., Wnuk, M.,
and Lam, M. (1974) Biochemistry 13, 60-64

Jendrossek, D., Steinbuechel, A. and Schlegel, H. G. (1988) J. Bacteriol. 170, 5248-
5256

a Keinan, E., Hafeli, E. K., Seth, K. K. and Lamed, R. L. (1986) J. Am. Chem. Soc.
108, 162-169

b Keinan, B., Hafeli, E. K., Seth, K. K. and Lamed, R. L. (1986) Ann. N. Y. Acad.
Sci. 501, 130-149

Keinan, 13., Seth, K. K., Lamed, R. L., Ghirlando, R. and Singh, S. P. (1990)
Biocatalysis 4, 1-15

Lamed, R. J ., Keinan, E. and Zeikus, J. G. (1981) Enz. Microbiol. Technol. 3, 144-148

186
Lamed, R. J. and Zeikus, J. G. (1981) Biochem. J. 195, 183-190

Lovitt, R. W., Longin, R. and Zeikus, J. G. (1984) Appl. Environ. Microbiol. 48,
171-177

McFarland, Chu, Y.-H., and J. W. Jacobs (1974) Biochemistry 13, 65-69

Nagata, N., Maeda, K. and Scopes, R. K. (1992) Bioseparation
2, 353-362

Oka, A., Sugisaki, H. and Takanami., M. (1981) J. Mol. Biol. 147, 217-226
Pereira, D. A., Gerson, F. P., and E. G. Oestreicher (1994) J. Biotech. 34, 43-50
Peretz, M. and Burstein, Y. (1989) Biochemistry 28, 6549-6555

Rossmann, M. G. and Argos, P. (1976) J. Mol. Biol. 105, 75-95

Sambrook, J ., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning: A
laboratory manual 2114 edition. (Nolan, C., ed.), Cold spring Harbor Press, NY

Sanger, F., Nicklen, S. and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74,
5463-5467

Scopes, R. K. (1983) FEBS Lett. 156, 303-306
Steinbiichel, A. and Schlegel, H. G. (1984) Eur. J. Biochem. 141, 555-564
Wierenga, R. K. and H01, G. J. (1983) Nature 302, 842-844

Zheng6,1(92, 6,2112”), V. T., and Phillips, R. S. (1992) Bioorgarric Medicinal Chem. 2,

Chapter V

Conclusions

187

188
CONCLUSIONS

The Thermoanaerobacter ethanolicus 2° ADH is a thermophilic enzyme
physiologically analogous to catabolic 1° ADHs. Both 1° and 2° ADI-ls have been used in
analytical scale, industrially valuable chiral syntheses. The thermophilic ADHs have
demonstrated generally higher stability than their mesophilic counterparts. Decades of
research have yielded detailed information about liver 1° ADH 3-dimensional structure,
kinetic mechanism, functional architecture, and biotechnological potential; yet, these studies
have failed to identify specific amino acid residues responsible for substrate descrimination.
Unanswered, this fundamental ADH structure-function question also prevents rational
protein engineering approaches to optimizing their design for chiral catalysis. 2° ADHs are
structurally and functionally similar to 1° ADHs, thus, they allow comparative analysis to
identify the molecular determinants of 1° versus 2° alcohol substrate specificity. Isolating
both mesophilic and thermophilic 1° and 2° ADHs similarly allows comparative analysis to
identify the molecular strategies used in ADHs to thermostabilize the folded proteins.
Finally, overexpressing the thermophilic 2° ADH cloned ath gene, the only thermophilic
2° ADH clone reported, allows site directed mutagenesis to test thermal feature or substrate
specificity hypotheses fiom comparative sequence analysis.

Biochemical, biophysical, and molecular biological approaches were employed here
to characterize the mechanisms responsible for T. ethanolicus 2° ADH thermal stability and
catalysis. First, the ath gene encoding the T. ethanolicus 39E 2° ADH was isolated,
sequenced, and the corresponding 2° ADH peptide compared to other 1° And 2° ADH
peptide sequences. Second, the recombinant protein was overexpressed, purified, and
kinetically characterized. Third, site directed mutagenesis and biophysical analysis were
used to validate the predictions of 2° ADH catalytic architecture based on peptide sequence
comparison to 1° ADHs. Finally, enzyme thermostability and thermophilicity were
characterized and their molecular biochemistry investigated.

189
The similar catalytic and thermal properties of the recombinant 2° ADH to those of

the catabolic enzyme purified from T. ethanolicus indicate that the isolated (1th gene
encodes the catabolic T. ethanolicus 2° ADH. Calculations using the translated T.
ethanolicus ath sequence coincided with the amino acid composition and molecular mass
experimentally determined for the native enzyme, arguing that the recombinant gene was
properly expressed in E. coli. The recombinant enzyme’s catalytic and thermal properties
indicated that the enzyme 1° structtnal information was sufficient to properly fold the
protein at both thermophilic and mesophilic temperatures. The deduced T. ethanolicus
peptide sequence was extremely similar to that determined for the T. brockii enzyme by
Edman degradation of the native enzyme. The T. ethanolicus 2° ADH amino acid sequence
was also quite similar to the peptide deduced fiom the gene encoding the mesophilic C.
Beijerinckit' 2° ADH, arguing that these enzymes also share overall structural similarity.
Finally, overexpression (~15% of total protein) of properly folded 2° ADH followed by
purification using only heat treatment and (NH4)2SO4 precipitation made production of
sufficiently large enzyme quantities for structure-function or biotechnological utility
analysis practical.

1° ADHs are either dimers or tetramers of identical or nearly identical 35 kDa to 45
kDa subunits. The characterized 2° ADHs are homotetramers of 36 kDa to 42 kDa
peptides. Both classes of enzymes typically require a nicotinamide cofactor for catalysis
and act through an ordered kinetic mechanism. Chemical modification experiments
demonstrated the importance of Zn, cysteine residues, and histidine residues in T.
ethanolicus 2° ADH catalytic activity, similar to that of 1° ADHs. EXAFS analysis
identified a bound Zn atom with a Coordination sphere of ZnS 1(N0)3.4 and ICP
measmements indicated a single Zn atom per wild type enzyme dimer. Mutations of
Cys37, HisS9, and Asp150 altered both enzyme catalytic activity and Zn binding. The
mutagenesis results and the Zn coordination sphere atomic composition together support

the comparative sequence prediction that this 2° ADH uses Cys37, HisS9, and Asp150 to

190
bind the catalytic Zn, a novel metal ligand structure compared to the Cys-His-Cys motif

seen in 1° ADHs. The low level of bound Zn and the prediction of only a single Cys
residue in the Zn coordination sphere also support the conclusion that the T. ethanolicus 2°
ADH subunits, unlike 1° ADH subunits, do not bind a structural Zn atom liganded by 4
Cys residues. Converting the NADP(H) specific T. ethanolicus 2° ADH into a NAD(H)
preferring enzyme by the Glyl98 to Asp point mutation as predicted from comparison of
the 2° ADH peptide to that of the horse liver 1° ADH. The dramatic reduction in mutant
enzyme affinity for NADP+ argues that both the sequence alignment and the assumption of
a Rossmann fold for 2° ADH nicotinamide cofactor binding are correct. Furthermore, the
experimental confirmation of both specific predicted structure-function similarities (eg., a
catalytic Zn atom liganded by Cys and His residues ) and differences (eg., the absence of a
2° ADH structural Zn and the Cys-His-Asp- for 2° ADH catalytic Zn binding versus the 1°
ADH’s Cys-His-Cys motif) between the enzyme classes, supports the assumption of
overall 1° versus 2° ADH structural similarity and validates the continued use of 1° ADH
structural data to direct future 2° ADH experiments.

T. ethanolicus 2° ADH catalyzed propan-2-ol oxidation was optimal near 90°C. The
half-life of 2° ADH activity was more than an hour at 90°C. Arrhenius analysis of the 2°
ADH thermoinactivation rate constants indicated a large activation energy (430 kJ mol'l).
The 2° ADH thermoinactivation rate is accelerated therefore, by factors influencing the large
positive arrhenius constant (1.8x1060). Precipitation, quantitated by spectrophotometric
absorbance, occurred at >85°C. Enzyme unfolding in the absence of chemical denaturant
was predicted - using both protein absorbance Tm[(3u] and fluorescence measurements in
the presence of GuHCl - to occur at higher temperatures (~115°C) than precipitation.
Thermally precipitated protein did not redissolve in buffer nor did it display catalytic
activity, indicating that thermoinactivation must precede or accompany precipitation. The
pseudo-first order thermoinactivation rate determining step argued that this step was

unimolecular with respect to protein and that it preceded multimolecular protein

191
precipitation. Scheme 1 summarizes the proposed thermophilic 2° ADH thermoinactivation

process.
. slow . . .
nanve Q non-nauve _) Precrprtate (1)
(active) step (inactive ?) (inactive)

Propan-2-ol oxidation by enzyme preincubated at 55°C showed no statistically
significant Arrhenius discontinuity. This temperature dependence of enzyme activity,
consistent with Arrhenius theory, is completely described by the change in average
substrate kinetic energy. Therefore, the discontinuous 2° ADH plots appear to result from a
slow enzyme structural change flour a highly active form to a less active form and that
thermozyme Arrhenius plots discontinuities are not necessarily evidence of the role of
reduced low temperature flexibility in the poor low temperature activity of thermozymes.
The observed 2° ADH catalytic rate enhancement at low chemical denaturant concentrations
required slightly greater GuHCl concentrations at higher temperatures contrary to the
prediction of a catalytically significant effect of GuHCl on enzyme flexibility. The similar
low effector concentration, enzyme rate enhancement was seen with added KCl, which is
not a strong denaturant, indicated that the rate enhancement resulted from ionic interactions.
These data argue that the observation of low temperature enzymatic rate enhancement by
added chemical denaturants also may not indicate a role for optimal flexibility in the
temperature dependence of thermozyme activity. The possible correlation between activity
and flexibility therefore, requires direct measurements of both enzyme properties.

192

Chapter 6

Directions for Future Research

193
SEQUENCE BASED MUTAGENESIS AND KINETICS

The HCA alignment between thermophilic and mesophilic 1° and 2° ADHs (see
chapter 2) predicted significant structural similarity despite dissimilar peptide sequences.
These predicted catalytic structural similarities were supported by the results of mutagenic
and biophysical analyses (see chapter 4). The predictive power of comparative sequence
analysis was proposed as a tool to identify the molecular determinants of ADH substrate
specificity. Furthermore, developing mutagenesis and active recombinant enzyme
expression systems provide the means to test these structure function predictions. An
extensive critical core residue comparison between the horse liver and B.
stearothermophilus 1° ADHs and the T. ethanolicus and C. beijerinckii 2° ADHs indicated
strong amino acid characteristic conservation among the peptides (see appendix A; tablel).
The conspicuous sequence differences between the 1° and 2° ADHs are listed in Table 1.
These residues are mostly clustered around the horse liver enzyme active site based on the
crystal structure (Fig. 1). Determining the effects of 1° versus 2° ADH comparison based
T. ethanolicus ath gene mutations on the 2° ADH substrate specificity and the effects
complimentary mutations have on 1° ADH specificity will advance our understanding of the
molecular biochemistry of ADH molecular biochemistry. This information may provide
site direcwd mutagenesis based ADH specificity optimization for industrially relevant
biotransformations.

The higher stability of NAD+ versus NADP+ makes it preferable for indusrtial
applications. A single Gly to Asp mutation shifted the T. ethanolicus 2° ADH catalytic
efficiency ratio from 2400-fold in favor of NADP+ linked activity to 6.7-fold in favor of
the N AD+ dependent reaction. This mutation's effectiveness was strong evidence
supporting the predicted 2° ADH Rossmann fold nicotinamide cofactor binding motif. The
mutation reduced the maximal NADP+ linked enzyme velocity 3-fold and increased the
NAD+ dependent reaction 5-fold creating a faster catalyst using NAD+ but a generally

194

Table 1. Comparison of conserved 2° ADH amino acids with the
corresponding 1° ADH residues deduced from HCA based
sequence alignments

 

 

 

 

P’ ADI-ls 7° ADI-ls
Horse liver B. stearotlierm. C. beijerinckii T. ethanolicus
AromanE-nonammafic
Va113 Va15 Trp15 Trp14
Val83 Leu77 Phe7 5 Phe75
Leu350 Val310 Phe328 Phe328
Phe93 Phe89 Thr86 lle86
conformational
Met40 Ile32 Pro31 Pro31
lle45 Val47 Pro36 Pro36
Charge
Thr59 Pr053 Asp51 Glu51
Leul71 Prol45 Met147 Met147
Phel76 Va1151 Met152 Met152
Size
Alall Ala3 Leu13 Va112
Ala12 Ala4 Gly14 Gly13
Leu14 Val6 lle16 Ile15
Va136 Va128 Ala27 Ala27
Va158 Va151 Gly50 Gly50
V3163 lle57 Met55 Met55
lle90 Ile85 Val83 Va183
Ala183 Ala155 Gly158 Gly158
Va1186 Va1158 Leu161 Leu161
Va1189 Ala161 Ile164 lle164
Leu342 Ala300 Va1314 Va1314
Ile346 Ile307 Val320 Val320
Glu353 Glu314 Asp328 Asp328
Gly35 8 Va13 19 Ala333 Ala333
Arg368 Arg331 Lys346 Lys346

 

195

Figure 1. Position of the non-conservative amino acid differences between
1° and 2° ADI-Is relative to the horse liver 1° ADH catalytic site. The liver
ADH structural coordinates were obtained from the Cambridge protein structure database
(filename "pdbadh6.ent"). Residues identified from the catalytic domain (red) and in the
cofactor binding domain (green) are shown in relation to the catalytic Zn atom (light blue),
the His ligand to the catalytic Zn (dark blue), and the two catalytic Zn cys residues
(yellow).

 

I
.
~t
I t
, t
.i
_\.‘\J
._l ,
v
r
t b

-

~—

..
t
J
t.» 3. , V
cl} .5“: A
_'.l

odd

 

 

197
lower turnover enzyme. Also mutant enzyme affinity for NAD+ increased only 3-fold

while decreasing the 2° ADH affinity for NADP+ 225-fold. This Glyl98 to Asp mutant
protein therefore, represents an incomplete transformation of nicotinamide cofactor
preferences. The reduction potentials for NAD+ and NADP+ are identical, NAD+ and
NADP+ structures differ only by a single phosphate group, the Rossmann fold structures
that bind the cofactor forrrrs differ only slightly, and NAD+ linked 2° ADH with high
turnover and high cofactor affinity exist in nature. These observations and the partial
success already achieved argue that this N ADP+ linked 2° ADH may be successfully
converted by further amino acid replacement into an improved functional NAD“ dependent
enzyme with high cofactor affinity. Sequence based comparative analysis of NAD+ and
NADP+ linked ADHs (Appendix A; table 2) indicates nine further mutations that may
complete this conversion (T able 2). These amino acid positions were identified based on
their being consistently different between the NAD(H) linked and the NADP(H) dependent
enzymes. The residue positions in the liver 3-dimensional structure (Fig. 1) indicate that
they are also clustered near the catalyitc Zn atom. Assessing the porential contributions of
these mutations on NAD(H) versus NADP(H) specificity requires additional kinetic
analysis of the mutated enzymes.

The greater than 85% similarity between the T. ethanolicus and C. beijert'nckt‘i 2°
ADH peptide sequences plus the vast difference in their reported thermostabilities makes
the reciprocal site directed mutation of these proteins based on comparative sequence
analysis a practicable and rational approach to understanding enzyme thermostability and
thermOphilicity. Of the 89 amino acid differences between these peptide sequences, only
12 were classified as nonconservative. Of these 12, nine were substituted Pro residues in
the thermophilic T. ethanolicus enzyme (Table 3). Substituting Pro residues into protein
loop regions have been shown to thermostabilize engineered proteins and to be important to
natural protein thermostability (see chapter 1). A Proline-zipper hypothesis has been
proposed to explain the enthalpic and entropic contributions of Pro residues to folded

198

83:

 

 

 

 

emcee 82m Gene acne 8988c. . 2
£5. ”E: 8e: 93. 229285
83o 8nd 824 use»; use
8&2 8&2 835 Read 8883

3&5 330 Race. 2.8:.

39:. ENE. fined 32d

82:. 82:. 3:3 «fie:
IIES Ea: Ewe meefi «8 use?
SeagéhF Eggnog U .Euefieeae .m b>m 88m Sufism 8
monument 38 can.“ moewma 3mm “Sex—AMP

 

855:2? 3:253 e82. <0: :8...— 3933 3:28..
:3. a 285%.. +3.: ”5888.98 2.. 5? 88
2:5: :3 en ace—.53.. tun—<2 eaten-.8 he commune—Sew .u 03.;

199

 

 

 

 

was: named cow. 28:2. 383%
8835 833 bad
2%: «Nae: 8&8 escapes .eaee sen:
REF. 2.3.5 £28 :a .«e EB e: - .32: 38880 a
38 53880“
as“: eN eggs on: 3. a8
83:. 9.3: See £2982 Sauteed:
new? See: cone £2852 382.2522
2:9 29:
2me mac: cone peace: eeeeaseeeea
«New ac: Aeaaficfiaavemao ace
~92 Nae: 8:2 cease: 2.829554
eaaee aged
finerebom .D 3.58:5» H
r in? em I cores.“ gene: :5 Sentence .5me $38.9

 

 

$555.3 3:253 e83 <0: See e355: :3 en ”em—Enema:—
e... 2 8.2.58 8:28.. 8:8.— ::< on 3.28.52: 82:. .n 2.2:.

200
thermozyme [1] stability. Importantly, in this thermophilic-mesophilic 2° ADH system,

each Pro residue's contribution to enzyme thermal properties can be assessed by both the
destabilizing effects due to thermophilic enzyme Pro removal and the stabilizing effects due
to mesophilic 2° ADH Pro addition. This type of extensive reciprocal study using a set of
multidomain, multisubunit enzymes has not been reported. Furthermore, the entensive
sequence similarity between the mesophilic and thermophilic 2° ADHs reduces the possible
number of thermal property determinants, allowing for potentially complete characterization
of the structural determinants of these properties. The PCR based mutagenesis system
developed here would also allow rapid construction of multiple mutations to assess the
independence or cooperativity of these mutational effects. Therefore this system appears to
be an excellent model to test the proline zipper hypothesis through biophysical
measurement of protein thermal unfolding and thermoinactivation (eg., temperature or
chemical denaturant dependent circular dichroism. differential scanning calorimetry,
temperature or chemical denaturant dependent absorbance spectrophotometry or
fluorimetry) (see chapter 3). i

The thermophilic-mesophilic enzyme reciprocal mutagenesis experiments assess the
contribution of each Pro to corresponding mesophilic peptide re$idue mutation to the
overall folded protein stability. The specific thermophilicity and thermostability
contributions of the added Pro residues as opposed to the replacement of the other amino
acid must be quantitated by constructing a series of mesophilic and thermophilic protein
mutants with a specific neutral (eg., Ser or Ala) residue at all Pro substitution positions and
measuring their thermal properties (see chapter 3). The effect of each mutation on the
enzyme thermal properties by itself, the effect due to interaction of the substituted residue
with the rest of the protein, and the contribution of the individual mutation to the total
difference in the thermal properties between the two 2° ADI-1s should be measured to
identify potential mechanisms to account for the mutational effect. The difference between

the completely neutral residue substituted thermophilic enzyme (thermophilic baseline

201
mutant) and the completely neutral residue substituted mesophilic enzyme (mesophilic

baseline mutant) thermal properties reflects the contribution of other peptide factors in
determining these thermal stability and activity properties. Next, subtracting the
thermophilicity and thermostability values of each baseline mutant from the conesponding
completely Pro substituted enzyme will quantitate the total stabilizing effect of these '
prolines on both 2° ADH folds. The difference between the mesophilic and thermophilic
total stabilizing effects measures the interaction of these Pro mutations with the rest of the
folded enzyme structure. The mutation specific and protein interactive effects of each Pro
could be examined in this way. A large difference would indicate that the proline effects
are strongly dependent on overall protein structure whereas a small difference would
indicate independent thermal property contributions by these residues. The latter result
would argue that Pro loop mutations are a predictable general thermostabilizing strategy.
The neutral residue used in this study should mimic either general amino acid or Pro
characteristics (aside from its side chain structural constraint). I propose that either Ala or
Ser be used because they are small, reducing the possibility of disrupted local packing,
they are uncharged but are often found at the protein surface, reducing the possibility of
altering local 3-dimensional structure or salt bridging, and unlike Gly, both have residue
geometric constraints typical of general amino acids. Furthermore, Ala's more Pro-er
aliphatic sidechain makes it a closer analog and, since the substitutions are proposed in
surface loops, the hypothesized tat-helix stabilizing Ala characteristic should not be relevant.
Loop stabilization has also been proposed to enhance enzyme thermostability (see
chapter 1). Determining the thermostability and thermophilicity effects of complimentary
mutations to swap loop regions between thethermophilic and mesophilic 2° ADHs could
specifically test this hypothesis. By identifying putative core regions the HCA
representation also proposes the putative loop positions (those stretches of hydrophilic
amino acids between the hydrophobic clusters). Like the Pro study, this work would
evaluate both the role of this strategy in native enzyme stability and the protein engineering

202
potential for loop stabilization. Furthermore, the relative effects of Pro insertion into 100ps

and loop stabilization strategies that do not involve prolines (i.e., added salt bridges) could
be examined using this system. Substituting stabilized loops which do not contain Pro into
both mesophilic and thermophilic 2° ADH loops that were shown to effect enzyme thermal
properties due to the presence or absence of Pro residues would directly compare the

relative efficacy Pro versus non-proline loop stabilization strategies.

STRUCTURAL ANALYSES

The research described in chapters 1 and 2 indicates that the poor low temperature
activity of the thermophilic 2° ADH can be explained by the Arrhenius theory. Therefore
the temperature dependence of enzyme activity can be described by changes in the average
substrate kinetic energy alone. The potential influence of protein flexibility temperature
dependence on the temperature dependence of enzyme activity however, has not been
explicitly determined. Thermal property altering mutants of both the T. ethanolicus and the
C. beijerincla'i enzymes (eg., the Pro residue or loop substituted mutants described in the
previous section) would provide an excellent system to examine this relationship using
electron paramagnetic resonance spectrometry (EPR). The flexibility of immobilized liver
1° ADH in solvents has been determined from the spin label environment dependent EPR
relaxation correlation times using a spin labelled compound coupled to Cys residues [1].
Enzyme flexibility has also been reported as the extent of deuterium exchange between the
D20 solvent and the protein (see chapter 1). Correlating the temperature dependence of
flexibility and activity for such a system of mutants would test the hypothesis that enzymes
require optimal flexibility for optimal activity. Furthermore, direct analysis of the
relationship between protein flexibility and activity would examine whether the reported
similar mesophilic and thermophilic protein flexibilities at their respective maximal activity
temperatm'es are necessary for maximal activity or simply an indication that the folded

structures are destabilizing.

203
2° ADH catalytic Zn binding was pr0posed to involve Cys, His, and Asp based on

HCA ADH sequence comparisons, and chemical modification experiments (see chapter 2).
The loss of catalytic activity and the predicted changes in protein Zn content for the mutant
enzymes supported the hypothesis that Cys37, HisS9, and Asp150 provided the catalytic
Zn ligands (see chapter 4). Also consistent with this hypothesis, EXAFS data identified a
bound Zn atom with a coordination sphere containing 1 Cys sulfur, 1 imidazole nitrogen,
and 3-4 other nitrogen or oxygen ligands. Similar EXAFS analysis of the Zn ligand mutant
enzymes will provide relative signal intensities for the Zn coordinated atoms. Comparing
the expected ligand specific EXAFS signals for each mutant to those observed will
determine if the mutated residue directly altered the metal coordination sphere. These data
would provide strong evidence to support or to refute the direct role of Cys37, HisS9, and
Asp150 in catalytic Zn binding. This work is currently underway in our laboratory.

Constructing a recombinant T. ethanolicus 2° ADH overexpression system provided
sufficient quantities of purified enzyme for crystallographic analysis. The exact 3-
dimensional structure of this thermophilic 2° ADH is currently being determined using x-
ray crystallography by our collaborator Dr. Bobby Ami. These data, the first 2° ADH exact
3-dimensional structure, would provide final confirmation of the overall 1° versus 2° ADH
structural similarities. Comparing the liver 1° ADH and T. ethanolicus 2° ADH exact
structures would also determine the validity of predictions based on the peptide sequence
alignments and mutagenic analysis of catalytic Zn and cofactor binding. An exact 2° ADH
structure would allow computer modeling to analyze the potential effects of proposed
mutations and allow the identification of subtle mutations that may be structurally or
functionally significant (ie., those based on the residue's position in the folded protein
which do not require significantly altered amino acid properties which cannot be identified
from the peptide sequence alone). Specifically, armed with both a 1° and a 2° ADH crystal
structme residues influencing enzyme catalytic site geometry and altering substrate

selectivity may be identified, allowing detailed analysis of the ADH substrate binding and

204
reaction mechanisms. Furthermore, the regions of subunit interaction would be identified,

determining if this tetrameric enzyme is formed by the association of two 1° ADH-like
dimers as currently believed. The number of su'ucturally competent active sites in the
holoprotein would also be determined. Crystallization of the 2° ADH with acetleoA will
also determine whether this physiological substrate is completely contained in the protein
during catalysis or whether, as we propose, the long coenzyme A portion of the molecule is
outside the protien during thioester reduction. If our hypothesis is correct, then the 2° ADH
may catalyze the reduction of a wide range of very large ketone substrates but if the
coenzyme is eopletely retained in the protein active site then the 2° ADH substrate range
would be expected to have a more limited size range. The crystallographic [3 values, the
uncertainties in the atomic positions, for protein regions could be compared in the wild type
enzyme x-ray structure and in thermal property altering mutant protein x-ray structures.
Significantly altered regional [3 values would point to specific regions potentially
responsible for folded enzyme structural integrity. Temperature dependent crystal
structural analysis could directly measure differences in regional flexibilities within the
protein (Petsco, personal communication). Comparing the measured flexibilities (residue
Bt-values) and the corresponding enzyme activities could directly address the role of protein
flexibility in enzyme activity.

BIOTECHNOLOGICAL UTILITY
Changing the cofactor specificity of the T. ethanolicus 2° ADH (eg., conversion
from an NADP" to an NAD+ linked enzyme), aside from yielding basic scientific protein
structure function information, would construct a more valuable biocatalyst due to more
desirable cofactor properties (eg., the greater thermal stability of NAD"). Om' collaborator,
Dr. Robert Phillips, is examining the utility of this thermophilic 2° ADH in transforming
ketones into chiral alcohols. Expanding the range of identified chiral reactions this 2° ADH

performs and the effect of reaction conditions (eg., solvents and temperatm'e) on both

 

 

_— 2.---..2.-- --- .....-. ..-_..

205
activity and enantiospecificity also increases the likelyhood that it will be incorporated into a

biotechnological production process. For very large substrates, solubility and the need for
aggitation may be avoided by reversibly binding the substrate to a solid matrix, exposing it
to enzyme in the presence of cofactor (similar to a chromatographic process), and releasing
the product. This type of system would interface with the current diversimer organic
synthesis technology used in drug development as the immobilized particulate enzyme
could be separated from large or small scale batch reactants by centrifugation and it could
be retained in a column for flow reacter systems. Furthermore, research identifying the
exact molecular determinants of T. ethwwlicus 2° ADH thermal and catalytic properties
would allow tailoring the properties of this stable chiral oxidoreductase. The ability to
modify this enzyme to perfom specific reactions under specific conditions identified by
industry would make this a strong candidate for chiral industrial processes.

REFERENCES

1 Vieille, c., Burdette, D. s. and Zeikus,.l. G. (1996) in Biotech. Ann. Rev., vol.
2, (M. R. E1 Geweley, ed.) pp. 1-83. Elsevier, Amsterdam, Neth.

2 Guinn, R. M., Skerker, P. S., Kavanaugh, P. and D. S. Clark (1991) Biotechnol.
Bioeng. 37, 303-308

 

APPENDIX A

206
Appendix A

Sequence Comparison Based Predictions of Important Catalytic
Domain and Cofactor Binding Domain Amino Acids in 1° and 2°
ADH Structures

Because of the relatively small number of exact 3-dimensional protein structures
determined and the typically well conserved structural folds within each enzyme family,
peptide sequence based comparisons are used to predict structurally analogous regions of
these proteins. ADHs characteristically share significant 3~dimensional structural
architecture but very little amino acid sequence similarity [l]. The horse liver 1° ADH x-ray
structure has been determined and its amino acid structure-function characteristics
extensively investigated [1]. HCA comparison of the thermophilic T. ethanolicus 39E 2°
ADH, the thermophilic B. stearothermophilus 1° ADH, and the mesophilic C. beijerinckii
and A. eutrophus 2° ADH peptides sequences to that of liver l° ADH (Fig. I) predicted a
Rossmann fold for nicotinamide cofactor binding and specific amino acids associated with
cofactor specificity and catalytic Zn liganding. The complete sequence alignments also
predict corresponding residues to those that are part of identified liver ADH structural
elements. Similarity among these structurally important residues, while not practically
testable by point mutations, provides evidence in support of the folded-architectural
similarity between the compared peptides. Tables 1 and 2 list the T. ethanolicus 2° ADH,
C. biejerinckii 2° ADH, and B. stearothermophilus 1° ADH residues predicted to
correspond to important residues identified in the horse liver enzyme. These tables also
indicate the percent amino acid similarity and identity between these four peptides for
specific sections of the horse liver ADH catalytic and cofactor binding structural domains.

 

 

 

Figure 1. HCA comparison of the B. stearothermophilus 1° ADH, the T.
ethanolicus 2° ADH, and the C. beiien’nckil 2° ADH amino Acid sequences
to that of the horse liver 1° ADH. The proposed catalytic Zn liganding (0). structure
Zn liganding CD, and Rossmann fold bindingm) residues are indicated. Symbols :‘k‘,
Pr01ine;°.slycinc;'i. scrinc;'n. threonine

I. W393 2° Adh

e

  

horse liver 1° Adh

 

209

Table 1. Comparison of aligned amino acids involved in ADH catalytic
domain structure and function

 

 

 

_Proposed role in Aligned peptide residues _
liver ADH Horse liver B. stearotherm. C. beijerinckii T. ethanolicus
structure
Active domain Alall A133 Leu13 Va112
core residues Alal2 A134 Glyl4 Gly13
Va113 V315 TrplS Trp14
Leu14 V316 Ilel6 Ile15
Phe21 Leul3 n.d. n.d.
Va126 N.D. V3121 Ala21
V3136 V3128 Ala27 A1327
Ile38 V3130 V3129 V3129
Met40 Ile32 Pr031 Pr031
Thr43 Cys45 V3134 Trp34
Ile45 Val47 Pro36 Pro36
Thr59 ProS 3 Asp51 Glu51
Val63 Ile57 Met55 Met55
lle64 Ile58 Ile5 6 Ile56
A1365 ProS 9 Leu57 Leu57
Ala69 Gly63 Ala61 Ala61
A1370 Val64 Val62 Val62
Val73 Ile67 ' V3166 V3166
V3180 V3174 V3172 V3172
V3183 Leu77 Phe75 Phe75
V3189 V3183 Ile82 V3182
Ile90 lle85 Val83 Val83
Pro91 Pr086 Pr084 Pr084
Vall69 Alal43 Alal45 Alal45
Leul7l Prol45 Met147 Met147
Ilel72 Ile146 Ile148 Ile148
Glyl73 Phe147 Thr149 Prol49
Phel76 VallSl Met152 Met152
Ala183 Ala155 Gly158 Gly158
V31186 V31158 Leul6l Leu161
A13187 Thr159 Alal62 Ala162
Va1189 Alal6l Ile164 Ile164
Amino acid
similarity (%)
Horse liver 100(100) 38(84) 25(80) 25(80)
B. stearotherm. 100(100) 2202) 2505)
C. beijen’nckii 100( 100) 78(94)
T. ethanolicus 100(100)
Active site Ser48 Thr40 Ser39 Ser39
pocket residues Leu57 ProSO Leu49 Ile49
Va158 V3151 Gly50 Gly50
Phe93 Phe89 Thr86 Ile86
, Phel 10 LeulOS Leu93 Ser93
Leu116 Tyr114 Phe99 T
Thrl78 Thr152 Ser154 Thr154

210

Amino acid

similarity (%)

Horse liver 100(100) 43(86) 28(78) 28(78)

B. stearotherm. 100(100) 14(86) 2801)

C. beijerinckii 100(100) 7 8(100)

T. ethanolicus 100(100)

Catalytic Zn Cys46 Cys38 Cys37 Cys37

ligands His67 His62 HisS9 HisS9
Cysl74 Cysl48 Asp150 Asp150

Amino acid

similarity (%)

Horse liver 100(100) 100(100) 67(67) 67(67)

B. stearotherm. 100(100) 67(67) 67(67)

C. beijerinckr’i 100(100) 100(100)

T. ethanolicus 100(100)

Structural Zn Cys97 Cys92 Cys85 n.d.

liganding CyleO Cys95 n.d. n.d.

residues Cys103 Cys98 n.d. n.d.
Cyslll Cyle6 n.d. n.d.

Amino acid

similarity (%)

Horse liver 100(100) 100(100) 25(25) mm

B. stearotherm. 100(100) 25 (25) ------

C. beijerinckii 100(100) ------

T. ethanolicus ------

Structural Gly Gly66 Gly60 Gly60 Gly60

residues Gly77 Gly71 Gly69 Gly69

Gly86 Gly8O Gly78 Gly78

Amino acid

similarity (%)

Horse liver 100(100) 100(100) 100(100) 100(100)

B. stearotherm. 100(100) 100(100) 100(100)

C. beijerinckii 100(100) 100(100)

T. ethanolicus 100(100)

Overall amino

acid similarity

(%) ,

Horse lrver 100( 100) 5 1 (88) 32(76) 3004)

B. stearotherm. 100(100) 29(25) 30(69)

C. beijerinckii 100(100) 79(94)

T. ethanolicus 100(100)

n.d.: no analogous residue was identified.

211

Table 2. Comaprison of aligned amino acids involved in ADH nicotinamide
cofactor binding

 

 

 

Proposed role in Aligned peptide residues

liver ADH

ADP- ribose 4

bindin Horse liver B. stearotherm. C.7§eijerinckii T. Zhanaicus

Core residues Aral 83 Thr153 A13159 Ala159
V31186 Tyr154 Leul61 Leul61
A13187 Ala156 Alal62 Ala162
Va1189 Va1159 Ile164 Ile164
Cys195 V31168 Vall70 Vall70
V31197 Ilel70 Vall72 Vall72
Leu200 Ile203 Ilel75 Ilel75
Val203 Leu206 V31178 V31178
A13237 A13209 Ala212 Ala212
Phe264 Val23l aArg23g A13238
Va1268 V31235 Ala242 Ala242
V31288 A13256 Ser263 Ala263
V31290 V31258 Ile265 Va1265
V31292 V3126O Tyr267 Tyr267
Ala317 Ala295 Gly239 Gly293
Alal96 Alal69 Vall7l Ala17l
Phel98 Tyrl7l Ile173 Leu173
Ile220 Vall92 Ilel95 Ile195
Va1222 Vall94 V31197 Val97
lle250 I-Ii5217 Ile263 Ile263
Leu254 A13221 V31227 I1e227
V31262 Gly229 V31236 V31236
Ala278 Ser250 Met255 Ile255
Cy5281 Arg252 Gly259 Gly259
CysZ 82 Arg253 Gly260 Gly260

Amino acid

similarity (%)

Horse liver 100( 100) 36(68) 36(68) 4406)

B. stearotherm. 100(100) 20(60) 32(68)

C. beijerinckii 100(100) 76(88)

T.’ ethanolicus 100(100)

Adenine pocket Phel98 Tyrl7l Ile173 Leu173

residues V31222 Vall94 V31197 V31197
Ile224 Leul96 Thrl99 Thrl99
Pro243 Pr0215 Tyr218 Tyr218
Ile250 His217 lle223 Ile223
lle269 V31236 Gly243 Gly243
Thr274 Thr237 Gly244 Gly244
Arg27l A13238 Ser246 A13246

Adenosine Asp223 Aspl95 Glyl98 Glyl98

ribose binding Glyl99 Gly172 Glyl74 Glyl74

residues Ile269 V31236 Gly243 Gly243

212

Asn225 Glyl97 Arg200 Arg200
Ly8228 LysZOO CysZO3 Cy3203
Pyrophosphate Arg47 His39 Thr38 Thr38
interacting Ile269 V31236 Gly243 Gly243
Nicotinamide Gly293 Gly26l Hi3268 Phe268
ribose biding
bAmino acid
similarity (%)
Horse liver 100(100) 5008) 21(28) 21(28)
B. stearotherm. 100(100) 14(28) 21(36)
C. beijerinckii 100(100) 78(86)
T. ethanolicus 100(100)
Overall amino
acid similarity
(%)
Horse liver 100(100) 4102) 30(52) 35(57)
B. stearotherm. 100(100) 18(48) 28(56)
C. beijerinckii 100(100) 77(87)
T. ethanolicus 100(100)

 

3Possibly a sequencing error.
bIle269 is only counted once in the calculation.

213
REFERENCES

Branden, C.-I., Jornvall, H., Eklund, H., and B. Fururgren (1975) Alcohol

dehydrogenases. In The enzymes, 3rd ed., vol. XI part A" p. 105-106, (P. D.
Boyer, ed.). Academic Press, New York.

APPENDIX B

214
Appendix B

Computer programs used to calculate (hyper)thermophilic
versus mesophilic total amino acid composition and the
theoretical Arrhenius data for thermophilic and mesophilic

enzymes

Computer program to compile amino acid composition data from peptide sequences
and to predict percent enzyme activity values at temperatures below the maximal
temperature for mesophilic or thermophilic enzyme activity were written using Turbo
Pascal 3.0 for IBM (Borland International; Scotts Valley, CA). Programs were compiled
and executed using an IBM PC. Program "TOTALAA" calculates the percent amino acid
composition of a peptide sequence input as a suing of letters in an ascii format text file, and
it calculates the average amino acid composition of a group of peptide sequence files input
in succession (used to calculate the values reported in chapter 1; Table 3). The output is
written to the default line printer and to the monitor. Program "Arrheniussimulation"
calculates the expected precent catalytic activities within a temperature range below the
temperature for optimal activity (Topt) specified by the low temperature (Told) in Kelvin
(see chapter 1; Fig. l). The calculation can be performed on enzymes with any reaction
activation energy (Ea) in kJ/mol. The output data is written to an ascii format text file

named "ARRSIMDA ".

215
program TOTALAA;

VAR
COMPARE: ARRAY [1..20] OF RECORD
LABELC: STRING[25];
NGC: REAL;
NPC: REAL;
NAC: REAL;
NV C: REAL;
NLC: REAL;
NIC: REAL;
NFC: REAL;
NWC: REAL;
NYC: REAL;
NHC: REAL;
NCC: REAL;
NMC: REAL;
NSC: REAL;
NTC: REAL;
NKC: REAL;
NRC: REAL;
NDC: REAL;
NEC: REAL;
NN C: REAL;
NQC: REAL;
END;
FILEIN: TEXT;
NAMEIN,CATEGORY: STRING[25];

216
AA,PAUSE,YNT: STRING[l];

i,J,NG,NP,NA,NV,NL,NI,NF,NW,NY,NH,NC,NM,NS ,NT,NK,NR,ND,NE,NN,NQ
,TOT: INTEGER;

N GT,NPT ,NAT,NVT,NLT,NIT,NFT,NWT,N YT,NHT,N CT ,NMT,NST,NT’I’,NKT,N
RT,NDT,NET,NNT,N QT: REAL;

SQNGT,S QN PT ,8 QNAT,SQNVT,SQNLT,SQNIT,SQNFT ,SQNWT ,SQNYT,SQNHT,
SQNCT,SQNMT,SQNST,SQNTT: REAL;
SQNKT,SQNRT,SQNDT,SQNET,SQNNT,SQNQT: REAL;
MGT,MPTMAT,MVT,MLT,MIT,NH71‘,MWT,MYT,MHT,MC1‘,MMT,MST,MTI‘,MKT
,MRT,MDT,MET,MNT,MQT: REAL;
GTSD,PTSD,ATSD,VTSD,LTSD,ITSD,FI‘SD,WTSD,YTSD,HTSD,CTSD,MTSD,STS
D,T'I‘SD,KTSD,RTSD,DTSD,ETSD,NTSD,QTSD: REAL;

TOTR: REAL;

YN,X: BOOLEAN;

PROCEDURE YESNO;
VAR
CK: BOOLEAN;
BEGIN
REPEAT
IF YNT = 'Y' THEN
BEGIN
YN := TRUE;
CK := TRUE;
END
ELSE IF YNT = 'y' THEN
BEGIN

217
YN := TRUE;
CK := TRUE;
END
ELSE IF YNT = 'N' THEN
BEGIN
YN := FALSE;
CK := TRUE;
END
ELSE IF YNT = 'n' THEN
BEGIN
YN := FALSE;
CK := TRUE;
END
ELSE
BEGIN
CK := FALSE;
WRITELN;
WRITELNCPLEASE ENTER EITHER Y OR N');
READ(YN'I);
WRITELN;

FUNCTION RD(A:REAL): CHAR;
VAR
S,SO: STRING[18];

218
i,M,P: INTEGER;

Y,YY: CHAR;

BEGIN
STR(A,S);
FOR i:= 1 TO 18 DO
BEGIN
IF S[i] = '.' THEN
BEGIN
M := i + 3;
P := i;
END;
IF S[i] = 'E' THEN
BEGIN
IF S[i+3] = '1' THEN
BEGIN
Y := S[P+1];
S[P] := Y;
S[P+1] :='.';
END;
IF S[i+3] = '3’ THEN
BEGIN
Y := S[P-l];
S[P-l] := ‘.';
S[P] := Y;
END;
IF S[i+3] = '4' THEN

 

2 l9
BEGIN

FORi:=1T018 DO
BEGIN
SO[i] := S[i];
END;
Y := SO[3];
S[3] := '.';
S[4] := '0';
S[S] := SO[3];
S[6] := SO[S];
S[7] == SO[6];
S[8] := SO[7];
M := 8;
END;
END;
END;
FOR i := 1 TO M DO
BEGDI
WRITE(S[i]);
END;
RD := ' ';
END;

FUNCTION PRD(A:REAL): CHAR;
VAR

S: STRING[18];

i.M.P: INTEGER;

 

 

ll"

220
Y: CHAR;

BEGIN
STR(A,S);
FOR i:= 1 TO 18 DO
BEGIN
IF S[i] = '.' THEN
BEGIN
M := i + 3;
P := i;
END;
IF S[i] = 'E' THEN
BEGIN
IF S[i+3] = '1' THEN
BEGIN
Y := S[P+l];
S[P] := Y;
S[P+1] := '.';
END;
END;
END;
FOR i := 1 TO M DO
BEGIN
WRITE(LST,S[i]);
END;
PRD := ' ';
END;

221

BEGIN

WRITELNCINPUT THE LABEL FOR THIS SET OF FILES (25 CHAR OR LESS)');
READ(CA'IEGORY);
WRITELN;
REPEAT
WRITELNCENTER INPUT FILE N AME');

READLNCNAMEIN);
ASSIGN(FILEIN,NAMEIN);

RESET(FILEIN);
CLRSCR;

 

i???
$29.2»;

ll
9

érégiéai
gee;

II
S?

222

II II
99

2%???

C?

9

NO := 0;

TOT := O;

WHILE NOT EOF(FILEIN) DO
BEGIN

READ(F1LEIN.AA);
IF UPCASE(AA)='A' THEN
BEGIN
NA := NA + 1;
END
ELSE IF UPCASE(AA) = 'G' THEN
BEGIN
NG := NG + 1;
END
ELSE IF UPCASE(AA) = 'P' THEN
BEGIN
NP := NP + 1;
END
ELSE IF UPCASE(AA) = 'V' THEN
BEGIN
NV := NV + 1;
END
ELSE IF UPCASE(AA) = 'L' THEN

 

 

223
BEGIN

NL := NL + 1;
END
ELSE IF UPCASE(AA) = '1' THEN
BEGIN

N1 := N1 + 1;
END
ELSE IF UPCASE(AA) = 'F' THEN
BEGIN

NF := NF + 1;
END
ELSE IF UPCASE(AA) = 'W' THEN
BEGIN

NW := NW + 1;
END
ELSE IF UPCASE(AA) = 'Y' THEN
BEGIN

NY := NY + 1;
END
ELSE IF UPCASE(AA) = 'H' THEN
BEGIN

NH := NH + 1;
END
ELSE IF UPCASE(AA) = 'C' THEN
BEGIN

NC := NC + 1;
END

224
ELSE IF UPCASE(AA) = ’M' THEN

BEGIN

NM := NM + 1;
END
ELSE IF UPCASE(AA) = '8' THEN
BEGIN

NS := NS + 1;
END
ELSE IF UPCASE(AA) = '1" THEN
BEGIN

NT := NT + 1;
END
ELSE IF UPCASE(AA) = 'K' THEN
BEGIN

NK := NK + 1;
END
ELSE IF UPCASE(AA) = 'R' THEN
BEGIN

NR := NR + 1;
END
ELSE IF UPCASE(AA) = 'D' THEN
BEGIN

ND := ND + 1;
END
ELSE IF UPCASE(AA) = 'E' THEN
BEGIN

NE := NE + l;

 

225
END

ELSE IF UPCASE(AA) = 'N' THEN
BEGIN
NN := NN + 1;
END
ELSE IF UPCASE(AA) = 'Q' THEN
BEGIN
NQ := NQ + 1;

END;
TOT := TOT + l;
TOTR := TOT;
WRITE(AA);

END;

WRITELN;

CIDSE(FH—EIN);

WITH COMPARE[i] DO
BEGIN
LABELC := NAMEIN;
NGC:: N GIT OTR;
NPC:: NP/TOT'R;
NAC:: NA/I‘OTR;
NV C:= NV/l'OTR;
NLC:= NIJTOTR;
NIC:= NI/I'OTR;
NFC:= NF/I‘OTR;
NW C:= NW/FOT‘R;
NYC:= NY [I OTR;

 

226
NHC:= NH/IOTR;

NCC:= NC/I'OTR;
NMC:= NMfI'OTR;
NSC:= NS/FOTR;
NTC:: NT/TOTR;
NKC:= NK/TOTR;
NRC:= NR/TOTR;
NDC:: ND/TOTR;
NEC:= NE/I‘OTR;
NNC:= NN/I'OTR;
NQC:: NQ/I‘OTR;
END;
WRITELNCPRESS ANY KEY TO CONTINUE);
READ(PAUSE);
CLRSCR;
X := FALSE;
IF X = TRUE THEN
BEGIN
WRITELNCGLY = ',NG);
WRITELN('PRO = '.NP);
WRITELNCALA = ',NA);
WRITELNCVAL = ',NV);
WRITELNCLEU = ',NL);
WRITELNCILE = ',NI);
WRITELNCPHE = ‘,NF);
WRITELNCTRP = ',NW);
WRITELNCTYR = ',NY);

 

227
WRITELNCHIS = '.NH);

WRITELNCCYS = ',NC);
WRITELNCMET = ',NM);
WRITELNCSER = ',NS);
WRITELNCTHR = ',NT);
WRITELNCLYS = ',NK);
WRITELNCARG = ',NR);
WRITELNCASP = ',ND);
WRI'I‘ELN('GLU = ',NE);
WRITELNCASN = ',NN);
WRITELNCGLN = ',NQ);
WRITELN;

WRITELNCTOT = ',TOT);
READ(PAUSE);

END;

CLRSCR;

GOTOXY(1,1);
WRITE(NAMEIN);

GOTOXY (1,2);

WRI'I‘E('AMINO ACID');
GOTOXY (15,2);
WRITE(NUMBER');

GOTOXY (30,2);

WRITE ('PERCENT COMPOSITION);
GOTOXY ( 1,3);
WRITELN('GLY',' ',NG,‘
WRITELNCPR ',' ',NP,’

'.RD(NGfl‘OTR));
',RD(NPrrOTR));

 

228

 

WRITELN('ALA',‘ ',NA,' '.RD(NAfrOTR));
WRITELN('VAL‘,‘ ',NV,‘ ',RD(NV/TOTR));
WRITELN('LEU‘,‘ ',NL,‘ ',RD(NLfl‘OTR));
WRITELN('ILE',‘ ',NI,‘ .RD(NI/TOTR));
WRITELN('PI—IE',‘ '.NF.' '.RD(NF/TOTR));
WRITELN('TRP',‘ '..'NW '.RD(NW/1‘ CTR»;
WRITELN('TYR',' '.NY.' '.RD(NY/TOTR));
WRITELN('HIS',‘ '.NH.' '.RD(NH/l'OTR));
WRITELN('CYS',‘ ',NC,‘ ',RD(NC/TOTR));
WRITELN('MET',‘ ,,'NM '.RD(NM/rOTR));
WRITELN('SER',‘ ',NS,‘ ',RD(NS/I‘OTR));
WRITELN('THR',‘ ',NT,’ ',RD(NT/TOTR));
WRITELN('LYS',‘ '..'NK '.RD(NK/I‘OTR));
WRITELN('ARG',’ ',NR,’ '.RD(NR/I‘OTR));
WRITELN(‘ASP',’ '.'.ND '.RD(ND/r0TR));
WRITELN('GLU',’ ',NE,‘ ',RD(NE/I‘OTR));
WRITELN(’ASN',‘ '.'.NN '.RD(NN/I‘OTR));
WRITELN('GLN',‘ '.NQ.' '.RD(NQfl‘0TR));
WRITELN('H-PHOBIC',’

',RD((NG+NA+NL+NI+NV+NP+NF+NW+NY)/TOTR));

WRITELN('POLAR',‘ ',RD((NS+NT+NH+NM+NC)/TOTR));
WRITELN('CHARGED',‘ '.
RD((NK+NR+NQ+NN+NE+ND)/TOTR));

X := FALSE;

IF X = TRUE THEN

BEGIN

WRITELN(LST,NAMEIN);

229

WRITELN(LST,‘AMINO ACID',‘ ','NUMBER',‘ 3% COMP');
WRITELN(LST,‘GLY',' ',NG,’ ',PRD(NG/TOTR»;
WRITELN(LST,‘PRO',' ',NP,‘ ',PRD(NP/TOTR»;
WRITELN(LST,'ALA',' '.NA,' ',PRD(NA/TOTR));
WRITELN(LST,'VAL',' ',',NV ',PRD(NV/I‘OTR));
WRITELN(LST,'LEU',' ‘,NL,‘ ',PRD(NL/TOTR»;
WRITELN(LST,'ILE',' ',NI,' ',PRD(NI/I‘OTR»;
WRITELN(LST,‘PHE',' ',NF,' ',PRD(NF/I‘OTR»;
WRITELN(LST,'TRP',' '..'NW ',PRD(NWrrOTR»;
WRITELN(LST,'TYR',' ",NY, ',PRD(NY/I‘OTR»;
WRITELN(LST,'HIS',' ',,'NH ',PRD(NH/I‘OTR»;
WRITELN(LST,‘CYS',' ',NC,’ ',PRD(NC/I‘OTR»;
WRITELN(LSTHMET ',,'NM ',PRD(NM/TOTR»;
WRITELN(LST,'SER',' ',NS,‘ ',PRD(NS/TOTR»;
WRITELN(LST,"I'I-IR',' ',NT,‘ ',PRD(NT/I‘OTR»;
WRITELN(IST,’I.YS',' '..'NK ',PRD(NKfl‘OTR»;
WRITELN(LST,‘ARG',' ',NR,' ',PRD(NR/TOTR»;
WRITELN(LST,’ASP',' ',ND,‘ ',PRD(ND/rorR»;
WRITELN(LST,‘GLU',' ',NE,‘ ',PRD(NE/TOTRD;
WRITELN(LST,‘ASN',' ',,'NN ',PRD(NN/I‘OTR»;
WRITELN(LST,'GLN',' ',NQ,‘ ',PRD(NQ/TOTR»;
WRITELN(LST,‘H-PHOBIC’,'

‘,PRD((NG+NA+NL+NI+NV+NP+NF+NW+NY)/TOTR));
WRITELN(LST,’POLAR','

',PRD((N S +NT+NH+NM+NC)/I‘OTR));
WRITELN(LST,'CHARGE ','

',PRD((NK+NR+NQ+NN+NE+ND)ITOTR));

 

230
END;

READ(PAUSE);
CLRSCR;
WRITELN('ADD ANOTHER SEQUENCE? (Y/N)');
READ(YNT);
WRITELN;
YESNO;
J := i;
i := i + l;
UNTIL YN = FALSE;
WRITELN('NUMBER OF SEQUENCES = ',J);
READ(PAUSE);

 

231
NRT := 0;

NDT := 0;
NET := O;
NNT := 0;
NQT := 0;
SQNAT := 0;
SQNGT := 0;
SQNPT := 0;
' SQNVT ;= 0;
SQNLT := O;
SQNIT := 0;
SQNFI‘ := 0;
SQNWT := 0;
SQNYT := 0;
SQNHT := 0;
SQNCI‘ := 0;
SQNMT := 0;
SQNST := 0;
SQN'I'I‘ := 0;
SQNKT := 0;
SQNRT := 0;
SQNDT := 0,
SQNET := O;
SQNNT := 0;
SQNQT := 0;
FORi := 1 TO I DO
BEGIN

 

232
WITH COMPARE[i] DO

BEGIN
NGT := NGT + NGC:
SQNGT := SQNGT + SQR(NGC);
NPT := NPT + NPC:
SQNPT := SQNPT + SQR(NFC);
NAT := NAT 4» NAC;
SQNAT := SQNAT + SQR(N AC);
NLT := NLT + NLC;
SQNLT := SQNLT + SQR(NLC);
NIT := NIT + NIC;
SQNIT := SQNIT + SQR(NIC);
NVT := NVT + NV C;
SQNVT := SQNVT + SQR(NVC);
NYT := NYT + NYC;
SQNYT := SQNYT + SQR(NY C);
NFI‘ := NFI‘ + NFC;
SQNPT := SQNFI‘ + SQR(NFC);
NWT := NWT + NWC:
SQNWT := SQNWT + SQR(NWC);
NHT ;= NHT + NHC;
SQNHT := SQNHT + SQR(NHC);
NCT := NCI‘ + NCC:
SQNCT := SQNCI‘ + SQR(NCC);
NMT := NMT + NMC;
SQNMT := SQNMT + SQR(NMC);
NST := NST + NSC;

 

233
SQNST := SQNST + SQR(NSC);

NTI‘ := N'I'I‘ + NTC;
SQN'I'I‘ := SQNTT + SQR(NT C);
NRT := NRT + NRC;
SQNRT := SQNRT + SQR(NRC);
NKT ;= NKT + NKC;
SQNKT := SQNKT + SQR(NKC);
NNT := NNT + NNC;
SQNNT := SQNNT + SQR(NNC);
NQT := NQT + NQC;
SQNQT := SQNQT + SQR(NQC);
NET := NET + NEC;
SQNET := SQNET + SQR(NEC);
NDT := NDT + NDC;
SQNDT := SQNDT + SQR(NDC);
END;

END;

MGT := NGT/J;

MPT := NPT/J;

MAT := NAT/J;

MLT := NLT/J;

MIT := NIT/J;

MVT := NVT/J;

MYT := NYT/J;

MFT := NFI‘IJ;

MWT := NWT/J;

MHT := NHT/J;

 

I!

234
MCT := NCI’IJ;

MMT := NMT/J;

MST := NST/J;

MTI‘ := N'IT/J;

NIRT := NRT/J;

MKT := NKT/J;

MNT := NNT/J;

MQT := NQT/J;

MET := NET/J;

MDT := NDT/J;

GTSD := SQRT((J*SQNGT - SQR(NG'I'))/(J*(J-1)));
PT SD := SQRT((J*SQNPT - SQR(NPD)/(J*(J-l)));
ATSD := SQRT((J*SQNAT - SQR(N AT»/(J *(J -1)));
LTSD := SQRT((J*SQNLT - SQR(NLD)/(J*(J-l)));
ITSD 3= SQRT((J*SQN1T ' SQR(NH'))/(J*(J-1)));
VTSD := SQRT((J*SQNVT - SQR(NVT))/(J*(J-l)));
YTSD := SQRT((J*SQNYT - SQR(NYD)/(J*(J-l)));
FI‘SD := SQRT((J*SQNFT - SQR(NFI'))/(J*(J-l)));
WTSD := SQRT((J*SQNWT - SQR(NW)/(J*(J-l)));
HTSD := SQRT((J*SQNHT - SQR(NH'I'))/(J*(J-1)));
CT SD := SQRT((J*SQNCI' - SQR(NCI')N(J*(J-l)));
MT SD := SQRT((J*SQNMT - SQR(NMT))/(J*(J-l)));
STSD := SQRT((J*SQNST - SQR(NST))/(J*(J-l)));
TI'SD := SQRT((J*SQN'IT - SQR(Nm)/(J*(J-1)));
RTSD := SQRT((J*SQNRT - SQR(NRDWPU-l)»;
KTSD := SQRT((J*SQNKT - SQR(NKT))/(J *(J -1)));
NT SD := SQRT((J*SQNNT - SQR(NNT))/(J*(J-l)));

235

QTSD == SQRT((J*SQNQT - SQR(NQD)/(J*(J-1)));
ETSD := SQRT((J*SQNET - SQR(NED)/(J*(J-1)));
DTSD := SQRT((J*SQNDT - SQR(NDD)/(J*(J-1)));
CLRSCR;

GOTOXY(60,2);

WRITE(CATEGORY);

GOTOXY(1,2);

WRITE(‘AMINO ACID');

GOTOXY(22,2);

WRITE(‘MEAN');

GOTOXY(42,2);

WRITE ('STD DEV');

GOTOXY(1,3);

WRITELN('GLY',‘ '.RD(MGT). '.RD(GTSD»;
WRITELN('PRO',’ .RD(MPT).' '.RD(PI‘SD»;
WRITELN('ALA',‘ RD'(MA.'T) '.RD(ATSD»;
WRITELN('VAL',‘ ,RD(MVT) ' '.RD(VTSD»;
WRITELN('LEU',‘ '.RD(MLT)‘ '.RD(LTSD»;
WRITELN('ILE',‘ ',RD(MIT),' '.RD(ITSD»;
WRITELN('PHE',‘ ,RD(MFI‘),' '.RD(FTSD»;
WRITELN('TRP',‘ ','.RD(MWT) '.RD(WTSD»;
WRITELN('TYR',' ‘.RD(MYT).' '.RD(YTSD»;
WRITELN('HIS',‘ ',RD(MHT),' '.RD(HTSD»;
WRITELN('CYS',‘ ',RD(MCI'),' '.RD(CTSD»;
WRITELN('MET',‘ ',RD(MMT),' '.RD(MTSD));
WRITELN('SER',‘ ',RD(MST).' '.RD(STSD));
WRITELN('I'HR'.’ '.RD(M'IT).' '.RD(TTSD»;

236
WRITELN('LYS',’ '.RD(MKT),' '.RD(KTSD»;

WRITELN('ARG',’ ',RD(MRT).' '.RD(RTSD»;

WRITELN('ASP',’ ',RD(MDT),' '.RD(DTSD»;

WRITELN('GLU',‘ ',RD(MET),' ',RD(ETSD));

WRITELN('ASN',‘ ',RD(MNT),' ',RD(NTSD»;

WRITELN('GLN',‘ ’,RD(MQT),' ',RD(QTSD));

WRITELN('H-PHOBIC',’
',RD(MGT+MAT+MLT+MIT+MVT+MPT+MFI‘+MWT+MYT));

WRITELN('POLAR',‘ ',RD(MST+MTI‘+MH'I‘+MMT+MCI'));

WRITELN('CHARGED',’ ',RD(MT+MRT+MQT+MNT+MET+MDT));

GOTOXY (60,4);

WRITE(FILES USED');

FORi := 1T0 I DO

BEGIN

WITH COMPARE[i] DO

BEGIN

GOTOXY(60,(i + 4));

WRITE(LABELC);

END;

 

END;
READ(PAUSE);
CLRSCR;
WRITELN('ANOTHER SET OF FILES (Y/N)?');
READ(YNT);
WRITELN;
YESNO;
UNTIL YN = FALSE;

237
END.

program Arrheniussimulation;

VAR
DATOUT: TEXT;
X,Y,Z: STRING[20];
TOPT,EA,DT,T,FK,TOLD: REAL;
TF: BOOLEAN;

BEGIN
ASSIGN(DATOUT,'ARRSIM.DAT');
WRITELN('INPUT T0pt (K)');
READ(I’OPT);
WRITELN;
WRITELN('INPUT Ea (kl/MOLD;
READ(EA);
WRITELN;
TOLD := 200,
TE := FALSE;
RESET(DATOUT);
APPEND(DATOUT);
WRITELN(DATOUT,'T'0pt = ',TOPT,‘ '.'Ea = '.EA);
WRITELN(DATOUT);
REPEAT
DT := TOPT-TOLD;
FK := EXP(((EA*1000)/8.314)*((1/TOPT) - (l/TOLD)»;

 

238
STR(TOLD,X);

STR(DT,Y);
STR(FK.Z);
WRITELN(DATOUTX,Y,Z);
T := TOLD + 2;
IF TOLD >= (TOPT) THEN
BEGIN
TF := TRUE; END;
TOLD := T;
UNTIL TF=TRUE;
QOSEmAmUD;
END.

 

APPENDIX C

 

239
Appendix C

Construction and Kinetic Characterization of
Thermoanaerobacter ethanolicus 39E 2° ADH Proline Deficient

Mutants

INTRODUCTION
Expression of the genes encoding thermOphilic and thermostable enzymes in mes0philic
hosts typically yields recombinant enzymes with native-like thermal properties. Therefore,
the molecular determinants of these characteristics appear to be in the amino acid sequences
themselves. The general amino acid compositional similarity between mesophilic and
thermophilic proteins ftu'ther suggests that the positions of specific amino acids and not
their number are critical to enzyme thermal stability. Finally, the significant energetic
contributions of individual amino acids to the marginal stabilizing energy of folded protein
su-uctures indicates that. similar to proper protein folding, enzyme thermostability is due
predominantly to energetic contributions from a subset of peptide amino acids. However,
the identities and positions of these critical amino acid residues cannot yet be predicted even
with exact 3-dimensional structure information.

Proline residues have been proposed to constrain protein surface loops and this is
proposed to stabilize the interactions between the associated hydrophobic core elements [1].
This proline-zipper hypothesis agrues that by preventing the disassociation of the surface
most inter-core-element interactions, a constrained loop thermostabilize the folded protein.
Loop engineering has been shown to thermostabilize protein folded structures (see chapter

1; prolines and loop regions). Furthermore, introducing Pro residues into tum regions has

240
thermostabilized both mesophilic and therm0philic proteins (see chapter 1; prolines and

loop regions). The identification of thermophilic enzymes containing additional Pro
residues compared to their mesophilic counterparts has suggesterd that this is also a protein
thermostabilization strategy used in nature (see chapter 1; prolines and loop regions). The
availability of functionally analogous thermophilic and mesophilic proteins with highly
similar peptide sequences would allow comparative analysis to identify potentially critical
Pro residues.

The recently cloned thermophilic Thermoanaerobacter ethanolicus 2° ADH amino
acid sequence differs from that deduced for the mesophilic Closm'dium beijerincla’i 2° ADH
fiom its cloned gene at 89 positions (see chapter 2). Of these, approximately 12 are
nonconservative and 9 of these are Pro residues in putative loop regions of the more
thermophilic enzyme. The overall sequence similarity among these 2 peptides is >85%
(See chapter 2). This is significantly higher than the similarity seen between the T.
ethanolicus 2° ADH and the horse liver 1° ADH (54%), yet mutagenic structure-function
analysis suggested that they were also architecturally analogous (see chapter 4). Because
the mesophilic C. beijerincla’i 2° ADH shares greater peptide sequence identity to the
thermophilic T. ethanolicus enzyme than the horse liver 1° ADH, the mesophilic C.
beijerinckii 2° ADH is also predicted to share greater folded structural similarity with the
thermophilic T. ethanolicus 2° ADI-I. Therefore, this is an excellent model system for
mutagenic analysis to test the hypothesis that proline residues, constraining surface loops,
can prevent hydrophobic core element separation and so thermostabilize proteins.

The data presented here examine the effect of mutating 3 individual T. ethanolicus
2° ADH added Pro residues to the corresponding C. beijerinckii enzyme residues predicted
based on sequence alignments. The thermOphilicities and thermostabilities of the wild type

and proline mutant enzymes are reported

 

 

 

241
MATERIALS AND METHODS

Chemicals and reagents

All chemicals were of at least reagent/molecular biology grade. Oligonucleotide synthesis
and amino acid sequence analysis were performed by the Macromolecular Structure Facility
(Department of Biochemistry, Michigan State University). The kanamycin resistance
GenBlock (EcoRI) DNA cartridge used in expression vector construction was purchased
from Pharmacia (Uppsala, Sweden). DNA for sequencing was isolated using the Wizard
Miniprep kit (Promega; Madison, WI).

Media and strains

Escherichia coli (DHSOt) containing the 2° ADH recombinant plasmids were grown in rich
complex medium (20 g l'1 tryptone, 10 g l'1 yeast extract, 5 g 1'1 NaCl) at 37°C in the
presence of 25 ug ml-1 kanamycin and 100 ug ml:1 ampicillin.

Muta genesis

All DNA manipulations were performed using established protocols [2,3]. Point. mutations
were introduced into the (1th gene by PCR [2] using the T. ethanolicus 39E 2° ADH gene
clonal plasmid pADHB25-kan (see chapter 2) as template DNA. An oligonucleotide primer
(KA4 N-end) was synthesized to bind the noncoding strand and included a Kpnl restriction
enzyme site, the native ath gene ribosome assembly site, and the initiation codon for the
ath gene. An oligonucleotide primer (KA4 C-end) was synthesized to bind the coding
strand, it included the compliment of the ath termination codon and an Apa! restriction
enzyme site. Complimentary 30-45 base oligonucleotide primers that contained the mutated
bases were used in conjunction with 2 KA4 end primers to amplify the N-terminal and C-
terminal segments of the ath gene. PCR syntheses of partial and complete mutated genes
were performed using the Taqplus, exonuclease containing polymerase (Strategene; La
Jolla, CA). All clones were expressed in pBluescriptII KS(+) with a kanamycin resistance
cartridge introduced into the polylinker EcoRI site. Mutations were verified by DNA
sequencing using the method of Sanger et al. [4].

242
Enzyme kinetics

The standard 2° ADH activity assay was defined as NADP“ reduction coupled to propan-2-
ol oxidation at 60°C as previously described [5]. The enzyme was incubated at 55°C for 15
min prior to activity determination unless otherwise indicawd. Tris buffer pH was adjusted
at 25°C to be pH 8.0 at assay temperature (thermal correction factor = -0.031 ApH °C'1).
Protein concentrations were measured using the bicinchoninic acid (BCA) procedure
(Pierce; Rockford, IL).

Thermostability

Recombinant 2° ADH thermostabilities were evaluated in E. coli cell extracts by timed
incubation at the desired temperatures, followed by incubation for 30 min at 25°C.
Incubations were performed in 100 p1 PCR tubes (cat. #72.733.050, Sarstedt; Newton,
NC) using 0.2 mg ml'1 protein in 100 ul of 50 mM TriszHCl (pH 8.0). Activity was
determined using the unfractionated samples.

RESULTS
The percent activities of the single and duplicated N—terminal Met containing wild type T.
ethanolicus 2° ADHs and the H022 to Ala, Prol49 to Thr, and H0222 to His mutants were
compared at temperatures from 30°C to 90°C (Fig. 1A). The catalytic activities calculated
for thecellextracts at90°Crangedfrom 6.0unitspermg total protein forthePro149 toThr
mutant to over 60 units per mg protein for the wild type enzymes. However, the percent
activities increased similarly for all enzymes from 30°C to 90°C. The enzyme residual
activities after timed incubations at 90°C were determined (Fig. 1B), indicating that
thermoinactivation of both the wild type and mutant enzymes was not pseudo-first order in
E. coli cell extracts. Also, the proline deficient mutant enzymes demonstrated greater
thermostability than either of the wild type 2° ADHs.

 

243

Figure l. Thermophilicity and thermostability profiles for T. ethanolicus 2°
ADH Pro residue mutants. (A) The effect of temperature on propan-2-ol oxidation by
recombinant wild type (0). Single N-terminal Met (0), Pro24 to Ser cl), Prol49 to Thr (A),
and Pr0222 to His fl), enzyme containing E. coli DHSa cell extracts. (B) The efi'ect of
timed incubations at 90°C on wild type (0), Single N-terminal Met (0), Pr024 to Ser Cb, ’
Prol49 to Thr (A), and Pm222 to His C), enzyme containing E. coli DHSa cell extract
residual percent activities.

 

244

 

 

 

 

 

1.4

1.2d

08
6
.4

024

u
0.
1

3.2.2. 0:358 .2552: ._e .8322..—

0.0

100

20

Temperature (°C)

245

 

 

 

 

 

 

 

o
2
1
0
ID 0 IO
1
AID 0
l0
8
AID ..
0
AID I6
AID .
lo
I D O 4
AI D O ..
AID O
o
I D .0 I2
I D00
I n! ..
ID.
F. . u 0
0 an. 9.. 3

2e: 53.8 .32825

Time (min)

246
DISCUSSION

These preliminary data suggest that the 3 single Pro mutations tested do not significantly
affect T. ethanolicus 2° ADH thermophilicity and thermostability. The wild type enzyme
temperature activity profile is similar to that reported for pmified enzyme (chapter 3). The
temperature dependence of mutant and wild type enzyme activities were similarly except for
the Prol49 to Thr mutant which may be inactivating at temperatures between 70°C and 90°
C. The decreased slope for average percent activity versus temperature above 70°C
compared to that below 70°C for this mutant is not consistent with wild type enzyme
behavior and suggests loss of optimal active enzyme structure in at least part of the enzyme
population. Furthermore, because the maximum Prol49 to Thr activity is less than that
expected from extrapolating the lower temperature activity data to 90°C, dividing the lower
temperature specific activities by the maximal activity (that at 90°C) to obtain the percent of
maximum catalytic rate would artificially raise the percent activity values at lower
temperatures. Recalculating the data using the temperature at 70°C as 100% activity
indicates that all ofthe mutants and the wild type enzyme had similar temperature activity
profiles below 70°C. Therefore replacing Pro 24 and Pr0222 with the corresponding C.
beijerinckii 2° ADH amino acids did not appear to significantly alter enzyme thermophilicity
but replacing Prol49 may have reduced enzyme thermophilicity. However, activity
determinations must be repeated using pmified enzyme to confirm these findings.

The Pro residue deficient mutants all displayed similar thermostabilities which were
greater than either wild type enzyme. The relationship between the logarithm of residual
percent activity and incubation time however, was nonlinear for all 5 enzymes. Precipitate
was seen in the thermoinactivated samples, consistent with data reported for the purified
wild type enzyme. The failure of even the wild type enzyme data collected using cell
extracts to fit a pseudo-first order rate equation suggests that the thermoinactivation slow
step under these conditions differs from that for purified enzyme. The relatively rapid loss
of 2° ADH activity in all 5 extracts compared to the purified enzyme further suggests that

 

247
purification stabilizes the 2° ADH against precipitation. While none of the Pro deficient

mutants demonstrated a substantial loss of thermostability, the inconsistency of the cell
extract data with that reported for the purified wild type enzyme makes analysis difficult.
The entire set of nine added T. ethanolicus 2° ADH Pro residue mutations must be
examined individually and together before drawing any conclusions but the preliminary
evidence presented here using these 3 Pro mutants indicates that T. ethanolicus 2° ADH
thermostability must be measured using purified enzyme.

248
REFERENCES

Vieille, C., Burdette, D. S. and Zeikus, J. G. [1996] Thermozymes, in
Biotechnology Annual Reviews, vol. 2, (M. R. El Geweley, ed.) pp. 1-83.
Elsevier Press, Amsterdam, Neth.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G.,
Smith, J. A. and Struhl, K. (1993) Current Protocols in Molecular Biology
(Janssen, K., ed.), Current Protocols, NY

Sambrook, J ., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning: A
laboratory manual 2051 edition. (Nolan, C., ed.), Cold spring Harbor Press, NY

Sanger, F., Nicklen, S. and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA
74, 5463-5467

Burdette, D. S. and Zeikus, J. G. (1994) Biochem. J. 302, 163-170

 

MICHIGAN STATE UNIV. LIBRQRIES
lllWWIIWI”11111111“llNIHIHIHIIHWIlHI
31293014214930