Structural basis of protein stability at poly extreme crystal structure of amya at 1 6 a resolution 2

26
CHAPTER 2
CRYSTAL STRUCTURE DETERMINATION OF AMYA FROM
HALOTHERMOTHRIX ORENII

2.1 INTRODUCTION
Some organisms can live in extreme conditions like temperatures that are
close to the boiling point of water, pressures that are many hundred times that of
atmospheric pressure and salinity that is orders of magnitude above typical
physiological conditions. These extremophiles use several cellular and structural
adaptation mechanisms to be able to survive, actively grow and propagate in extreme
environments. It has been reported that macromolecules, especially proteins from
extremophiles and mesophiles, have the same overall fold and molecular mechanism
for their function (Vieille et al, 1995; Russell et al, 1997; Bauer et al, 1998).
However, certain structural features make these proteins stable and optimally active
under extreme conditions. Considerable effort has been made to date to understand
the molecular mechanism of adaptation of proteins from thermophilic and halophilic
organisms at the atomic level.


2.1.1 Thermophilic protein stability
With the exception of phylogenetic variations, what differentiates
thermophilic and mesophilic enzymes is only the temperature

ranges in which they are
stable and active. Otherwise, thermophilic

and mesophilic enzymes are highly similar.
The sequences of

homologous thermophilic and mesophilic proteins are typically

40 to 85% similar (Davies et al, 1993), their three-dimensional structures

are
superimposable (Auerbach et al, 1998; Tahirov et al, 1998) and they have the same

27
catalytic mechanism (Bauer et al, 1998). Thermophilic proteins are the most widely
studied due to their extreme stability and ease of purification. From these studies it
appears that thermal stability is not determined by any single factor but the
combination of several factors, each with a relatively small effect (Perutz and Raidt,
1975; Argos et al, 1979; Vogt et al, 1997; Jaenicke and Bohm, 1998; Shih and Kirsch,
1995). These factors include the following.

2.1.1.1 Amino acid composition
The amino acid composition of a protein has long been thought to be
correlated to its thermostability. The first statistical analysis

comparing amino acid
compositions in mesophilic and thermophilic

proteins indicated the trends toward
substitutions such as Glycine to Alanine

and Lysine to Arginine. A higher Alanine
content in thermophilic proteins

was supposed to reflect the fact that Alanine was the
best helix-forming

residue (Argos et al, 1979). As more experimental data
accumulate, in particular

complete genome sequences, it is becoming obvious that
"traffic

rules of thermophilic adaptation cannot be defined in terms of

significant
differences in the amino acid composition" (Bohm and Jaenicke, 1994).

The
comparison of residue contents in hyperthermophilic and mesophilic

proteins based
on the genome sequences of mesophilic and hyperthermophilic organisms shows only
minor trends.

2.1.1.2 Hydrophobic interactions
The hydrophobic effect is considered to be a major driving force of protein
folding. Hydrophobicity drives the protein

to a collapsed structure from which the
native structure is defined

by the contribution of all types of forces (Dill, 1990).

28
Thermophilic proteins normally have extensive hydrophobic interactions and reduced

water accessible

hydrophobic surface area compared to their mesophilic counterparts
(Wigley et al, 1987).

2.1.1.3 Disulfide bridges
Disulfide bridges are believed to stabilize proteins mostly through an entropic
effect by decreasing the entropy of the protein's

unfolded state (Matsumura et al,
1989). The entropic effect of the disulfide bridge

increases in proportion to the
logarithm of the number of residues

separating the two bridged cysteines.

2.1.1.4 Aromatic interactions
Aromatic-aromatic interactions (aromatic pairs) are defined by a distance of
less than 7.0 Å between the phenyl ring centroids. Aromatic amino acid interactions
are known to be one of the determinants of thermal stability in thermophilic proteins
(Kannan and Vishveshwara, 2000; Serrano et al, 1991). A pair of aromatic
interactions contributes between -0.6 and -1.3 kcal/mol to protein stability (Serrano et
al, 1991).

2.1.1.5 Ion-pair
Salt bridges are formed by spatially proximal pairs of oppositely charged
residues in native protein structures. Often salt-bridging residues are also close in the
protein sequence and fall in the same secondary structural element, building block,
autonomous folding unit, domain, or subunit, consistent with the hierarchical model

for protein folding. Salt bridges are rarely found across protein parts which are joined
by flexible hinges, a fact suggesting that salt bridges constrain flexibility and motion.

29
While conventional chemical intuition expects that salt bridges contribute favorably to
protein stability, recent computational and experimental evidence shows that salt
bridges can be stabilizing or destabilizing. Due to systemic protein flexibility,
reflected in small-scale side-chain and backbone atom motions, salt bridges and their
stabilities fluctuate in proteins. At the same time, genomewide amino acid sequence
composition, structural, and thermodynamic comparisons of thermophilic and
mesophilic proteins indicate that specific interactions, such as salt bridges, may
contribute significantly towards the thermophilic-mesophilic protein stability
differential. Ion pair networks are energetically more favorable

than an equivalent
number of isolated ion pairs because for each

new pair the burial cost is cut in half:
only one additional residue

must be desolvated and immobilized (Yip et al, 1995).

2.1.1.6 Metal binding
Metals have long been known to stabilize and activate enzymes. In proteins,
metal ions are coordinated, usually by lone pair electron donation from oxygen or
nitrogen atoms. Some thermophilic and hyperthermophilic enzymes have been
reported that contain metal atoms that are not present in their

mesophilic homologs.
Experiments have shown that metal binding can contribute 6 - 9 kcal/mol to stability.

2.1.1.7 Extrinsic parameters
While most pure hyperthermophilic enzymes are intrinsically very stable,
some intracellular hyperthermophilic proteins get

their high thermostability from
intracellular environmental factors

such as salts, high protein concentrations,
coenzymes, substrates,

activators, polyamines, or an extracellular environmental
factor

such as

pressure.

30

2.1.2 Halophilic protein stability
Halophiles (salt-lover) can be defined as microorganisms that require high salt
in a concentration range of 2 - 5 M to grow (Richard & Zaccai, 2000). In order to
overcome the extreme osmotic pressure of these hyper saline environments, halophilic
bacteria and eukaryotes accumulate mostly neutral organic compatible solutes and
exclude most of the inorganic salts. In contrast, halophilic archaea balance the
external high salt concentration by intracellular accumulation of inorganic ions to
concentrations that exceed that of the medium. Therefore, all the cellular components
of the halophilic archaea must adapt to function at the extremely high intracellular salt
concentration.

Halophilic proteins require a minimum of 2 M salt concentration to be
optimally active and stable. At high salt concentrations, proteins are in general
destabilized due to enhanced hydrophobic interactions. Halophilic proteins have,
therefore, evolved specific mechanisms that allow them to be both stable and soluble
in high salt concentration. The adoptive mechanism of halophilic proteins has not
been studied as extensively as thermophilic proteins due to the difficulty in purifying
and crystallizing them at very high ionic strengths. Halophilic enzymes are usually
very unstable in low salt concentrations. Since some of the important fractionation
methods in protein chemistry, such as, electrophoresis and ion exchange
chromatography, cannot be applied at high salt concentrations, the available
fractionation methods for halophilic bacterial proteins are rather limited.

31
In silico analyses of the genome sequence of halophilic organisms suggest that
proteins from these organisms have unique amino acid compositions. They have at
least twice the number of acidic residues than basic residues (Fukuchi et al, 2003;
Bieger et al, 2003). Structural insights gathered from the known halophilic crystal
structures suggest that the acidic surface and the associated negative electrostatic
surface potential, is one of the major stabilizing forces and is a highly conserved
feature (Dym et al, 1995; Bieger et al, 2003), Fig. 2.1.
Figure 2.1 The electrostatic surface potential of malate dehydrogenase from
Haloarcula marismortui, a typical halophilic protein (PDB id: 1D3A). The
electrostatic drawings were produced using the program GRASP.

Surface colors
represent the potential from -10 k
B
T
-1
(red) to +10 k

B
T
-1
(blue).
Since all soluble halophilic enzymes have a highly negative surface charge,
once folded properly, their flexibility may be achieved by repulsive forces between
closely placed charged residues. The instability caused by the high surface charge
density should be somehow balanced. Otherwise, the polypeptide chain will unfold. It
was long believed that one of the roles of high salt concentration was to shield this
high surface charge. Indeed, classical electrostatic calculations using Poisson–
Boltzmann equation (Elcock and McCammon, 1998) suggest that at pH 7.0 the

32
stability of halophilic proteins is decreased by 18.2 kcal/mol on lowering the salt
concentration from 5 to 0.1 M.
Using thermodynamic theories to analyze various biophysical measurements
(Bonnete et al, 1993) it was calculated that, in its native state at 4 M NaCl, halophilic
malate dehydrogenase (MDH) binds approximately 200 molecules of salt and almost
3000 molecules of water. These values are significantly higher than those measured
for non-halophilic proteins under the same condition and also higher than the number
of salt and water molecules bound in low salt solutions in which the halophilic
enzyme is unfolded. These findings are the basis for the ‘halophilic stabilization
model’ for solutions in NaCl, KCl and MgCl
2
(Zaccai, 1989). According to this model
the tertiary and quaternary structures of native halophilic proteins co-ordinate
hydrated salt ions on their surface at higher local concentrations than in the
surrounding solution by specific interactions with the surface carboxyl groups.
Through the binding of hydrated salt ions, water molecules would be associated with
the protein structure with different local salt concentrations depending on the hydrated

interactions of the particular salt. When the bulk salt concentration is reduced, salt
will diffuse from the ‘quasi-crystalline’ protein-associated layer into the solvent bulk,
destabilizing the protein surface and causing dissociation of the enzyme into its
subunits and unfolding of the polypeptide chain. According to this model, the
stabilization is enthalpy driven. The entropic penalty derived from the organization of
the hydrated salt is compensated by the enthalpy of the binding of the hydrated salt to
the surface carboxyl groups.
This explanation for the role of salt in halophilic protein stabilization is
challenged by two experimental results. First, the high resolution three-dimensional
structure of Haloarcula marismortui ferredoxin (HmFd) demonstrates very clearly

33
that although the protein was crystallized from 3.8 M sodium–potassium phosphate,
very few counter ions were found to be bound to the protein and when bound, they
interact with the main-chain carbonyl oxygen and not with side-chain carboxylates
(Frolow et al, 1996) Second, sub-millimolar concentrations of NADH can effectively
replace the requirement for molar quantities of salt in the stabilization of halophilic
malate dehydrogenase (MDH). Therefore, neutralization of surface charge by salt may
not be required for protein stability (Irimia et al, 2003).
In addition, some of the thermophilic protein determinants like metal ion
binding and salt bridge networks also play a role in stabilizing halophilic proteins
(Dym et al, 1995; Bieger et al, 2003).

2.1.3 Poly-extremophiles
Poly-extremophiles can be defined as the organisms that require more than
one extreme condition for its optimal growth and survival. For example, the organism
Thermosipho japonicus isolated from a deep-sea hydrothermal vent in the Okinawa
area, Japan requires high temperature and high pressure for its optimal growth (Takai
et al, 2000).
An interesting category among this involves the organisms that require both

high temperatures and high salt concentrations. Extremophilic microbes of this kind
are rare in nature and those isolated so far are difficult to handle in routine
laboratories. To date only two such organisms have been reported, namely,
Halothermothrix orenii and Thermohalobacter

barrensis (Mijts and Patel, 2002;
Cayol et al, 1994), both of which are members of the low G+C DNA-containing
gram-positive phylum. H. orenii is a true halophilic and thermophilic anaerobic
bacterium that was isolated from the Tunisian salt lake in the Sahara desert. It requires

34
2 M NaCl and 60 °C temperature for optimal growth but still shows significant
growth up to 4 M NaCl (Cayol et al, 1994). Halothermothrix orenii is a member of
the family Haloanaerobiaceae, order Haloanaerobiales (Cayol et al, 1994), whereas
Thermohalobacter berrensis, a member of the order Clostridiales, grows readily at 70
°C in the presence of 15% NaCl (Cayol et al, 2000). These organisms drastically
differ from other extremophiles as they handle both physical and chemical extremes at
the same time.
The study of the molecular adaptation of proteins at more than one extreme
i.e. poly-extreme condition is very important in the understanding of the biology of
these organisms. A few specific questions may be posed in this halo- and thermophilic
group of poly-extremophilies.
1) Are the structural adaptations the same as those found in halophilic and
thermophilic organisms or would it be a completely new set of adaptations?
2) In the former case, would these adaptations be a simple addition of structural
features or are there complex interactions of these features?
3) What specific differences exist in the poly-extreme adaptations?
4) What is the impact of the structural adaptations on the function and
mechanism of the proteins?
To address these questions we have undertaken a biophysical study of AmyA, a

secretory α-amylase (Mijts and Patel, 2002) from Halothermothrix orenii. AmyA is
an endo-acting α-amylase and randomly cleaves the α-1,4-glycosidic linkages present
in starch and its constituent polysaccharides amylose and amylopectin (Mijts and
Patel, 2002). AmyA is active in a broader salt concentration ranging between 0 and 5
M NaCl. However, it has optimal activity at 2 M NaCl concentration and temperature
above 65 °C, similar to the optimal conditions for H. orenii growth. The obligatory

35
requirement of salt at molar levels and temperature above 60 °C makes AmyA a true
halo-thermophilic enzyme.

2.1.4 The α-amylases:
Alpha-Amylases (alpha-1,4-glucan-4-glucanohydrolase, EC 3.2.1.1) are Endo-acting
enzymes that catalyze the hydrolysis of alpha-1,4-glycosidic bonds in starch and related poly
and oligosaccharides.
The α-amylase family comprises a group of enzymes with a
variety of different specificities that all act on one type of substrate, being glucose
residues linked through an α–1,1, 1,4, or 1,6 glycosidic bond. The members of this
family share a number of common characteristics but at least 21 different enzyme
specificities are found within the family. These differences in specificity are based not
only on subtle differences within the active site of the enzymes but also on the
differences within the overall architecture of the enzymes. The α-amylase family can
roughly be divided into two subgroups: the starch-hydrolyzing enzymes and the
starch-modifying or transglycosylating enzymes. The hydrolases and transferases that
constitute the α-amylase family are multidomain proteins, but each has a catalytic
domain in the form of a (β/α)
8
-barrel with the active site being at the C-terminal end of the
barrel beta-strands. Although the enzymes are believed to share the same catalytic acids and a
common mechanism of action, they have been assigned to three separate families - 13, 70 and

77 - in the classification scheme for glycoside hydrolases and transferases that is based on
amino acid sequence similarities




36


Figure 2.2. Different enzymes involved in the degradation of starch. The
open ring structure symbolizes the reducing end of a polyglucose molecule.

2.1.4.1 Domain Architecture of Amylase:
The enzymes are multidomain proteins, but share a common catalytic domain
in the form of a (β/α)
8
-barrel, i.e., a barrel of eight parallel β-strands surrounded by
eight helices, the so-called domain A. This structure has been demonstrated by X-ray
crystallography in several enzymes of the α-amylase family, although in one instance
only seven of the eight helices in the barrel fold are present. In addition, studies of
amino acid sequence similarities have led to the prediction that many other enzymes
belong to this family and have a similar catalytic domain. Usually, the loops that link
β-strands to the adjacent helices carry amino acid residues of the active site; some of
these loops may be long enough to be considered as domains in their own right. Thus,
in most cases where the structure has been determined by crystallography, a large
loop between the third β-strand and third helix is discussed as a separate domain,
domain B. This loop has an irregular structure that varies from enzyme to enzyme,

37
and, it has been argued, should not always be considered a separate domain but, in

some cases, should be thought of as part of a structural unit containing other loops.
Similarities in domain B amongst members of sub-groups of the α-amylase family
have, however, been found, e.g., various glucosidases resemble each other, while a
relationship can be demonstrated between enzymes such as neopullulanase and
cyclomaltodextrinase.










Figure 2 3. The domain architecture of α-amylases.

2.1.4.2 The catalytic mechanism of amylases:
Catalytic steps in glycoside bond cleavage in retaining enzymes. The proton
donor protonates the glycosidic oxygen and the catalytic nucleophile attacks at C1
leading to formation of the first transition state. The catalytic base promotes the attack
of the incoming molecule ROH (water in hydrolysis or another sugar molecule in
transglycosylation) on the formation of the covalent intermediate resulting in a second
transition state, leading to hydrolysis or transglycosylation product

38


Figure2.4. The catalytic mechanism of α-amylases.

The amylase enzymes are among the most industrially important enzymes,
having wide applications, such as in brewing, starch processing, textile, alcohol
production, and detergent industries. In the past decades, we have seen a shift from
the acid hydrolysis of starch to the use of starch-converting enzymes in the production
of maltodextrin, modified starches, or glucose and fructose syrups. The conditions
prevailing in the industrial applications in which enzymes are used are rather extreme,
especially with respect to temperature and pH. Therefore, there is a continuing
demand to improve the stability of the enzymes and thus meet the requirements set by
specific applications. Most of the well-characterized starch degrading enzymes are
from thermophilic and hyperthermophilic prokaryotes and much less research have
been devoted to enzymes from other extremophiles such as halophiles. The structural
information from the H. orenii kind of extremophiles would be very useful for us to
design highly stable and commercially useful proteins.
In this thesis, we report the structure of AmyA at both low and high salt
concentrations at
1.6 and 1.83 Å resolution, respectively. The analysis of AmyA structure
reveals a novel surface feature and its implications for stability under poly-extreme
conditions. We also report the biophysical characterization studies of AmyA under a

39
broad salinity range and this provides insight into the stability of AmyA over the
entire salinity range and at high temperatures.

2.2 MATERIALS AND METHODS
2.2.1 Protein purification
The AmyA gene was cloned in the pTrcHisB vector (Invitrogen) by
amplifying the gene by PCR using H.orenii’s genomics DNA and the protein was
over expressed with an N-terminal hexahistidine tag in Escherichia coli strain TOP10
cells (Invitrogen). One colony of TOP10 cells containing the pTrcHis-AmyA
construct was used to inoculate 10 ml of LB medium with 100 µg ml

-1
of ampicillin
(LB-Amp medium) and the cells were grown at 37 °C for 16 h. This 10 ml culture
was added to 1 L of fresh LB-Amp medium and the cells were grown at 37 °C to an
OD
600
of 0.6. The protein was induced for 4 h by adding isopropyl-β-D-
thiogalactopyranoside
(IPTG) to a final concentration of 1 mM. The cells were
harvested by centrifugation at 5000g for 10 minutes and resuspended in 20 ml
phosphate buffer [20 mM sodium phosphate (pH 7.8), 500 mM NaCl]. The cells were
lysed at 4 °C using a French press at 6.9 MPa. DNaseI was added to the lysate at a
concentration of 5 µg ml
-1
and the sample was incubated on ice for 30 minutes to
increase the precipitation of heat denatured protein. Any insoluble material was
removed by centrifugation at 10,000g for 15 min. The supernatant was incubated at
65 °C for 30 min to denature all other E. coli proteins and transferred to ice for 30
minutes to maximize protein precipitation. AmyA remains active at temperatures of
up to 70 °C. Therefore most of the E. coli host proteins were precipitated by the heat-
precipitation technique and removed by subsequent centrifugation at 10,000g for 20
minutes. The clear supernatant was loaded to a column containing 2 ml Ni-NTA

40
agarose affinity resin (Qiagen). After washing with phosphate buffer containing 10
mM imidazole, the recombinant AmyA protein was eluted from the column with 200
mM imidazole in the phosphate buffer. The protein was further purified by gel-
filtration chromatography using a Hiload 16/60 Superdex75 column (Amersham
Pharmacia Biotech) with buffer consisting of 20 mM HEPES (pH 7.5), 500 mM
NaCl. The His-tag was not removed before crystallization and the yield of the protein

was 2mg/litre of culture. The pure protein fractions were pooled together and
concentrated to a final concentration of 5 mg ml
-1
using Centriprep and Centricon
YM-10 devices (Millipore).

2.2.2 Crystallization and data collection

The AmyA protein sample (at 5 mg ml
-1
, assayed by the Bradford method)
was set up for crystallization using the Crystal Screen Cryo kit (Hampton Research)
with the vapour-diffusion hanging-drop method. 1 µl of protein sample was mixed
with 1 µl of reservoir solution and equilibrated against 700 µl of the reservoir solution
at 24 °C. Single crystals formed under the condition 70 mM sodium acetate (pH 4.6),
5.6% PEG 4000 and 30% glycerol. The condition was further optimized by mixing 4
µl of the protein sample with 1 µl of the reservoir solution. Crystals were obtained
with maximum dimensions of 0.1 x 0.1 x 0.6 mm after 1 d (Fig. 2.2). The mother
liquor with glycerol increased by 5% was used as a cryoprotectant. The crystals were
flash-cooled in liquid nitrogen. X-ray diffraction data were collected at ALS,
Berkeley at -173 °C. The crystal belongs to the orthorhombic P2
1
2
1
2
1
space group and
contains one molecule per asymmetric unit. All data were indexed, integrated and
scaled using the programs DENZO and SCALEPACK (Otwinowski and Minor,
1997). The data collection and crystallographic statistics are summarized in Table 2.1.

41













Figure 2.5 lAmyA crystal picture. Crystals of AmyA, with maximum
dimensions of 0.1 x 0.1 x 0.6 mm.
Table 2.1 Crystal parameters and data collection statistics of
AmyA at low salt. Values in parentheses are for the last resolution
shell.

1
R
merge
= Σ
hkl
Σ
i
|I
i
(h k l) – <I(h k l)>| / Σ
hkl
Σ
i
I
i
(h k l)]

Space group
Unit cell dimensions (Å)


Wavelength (Å)
Resolution of data (Å)
No. of measured reflections
No. of unique reflections
Redundancy
Completeness (%)
Mean I/σ(I)
1
R
merge

P2
1
2
1
2
1

a = 55.126
b = 61.658
c = 147.625
0.99
99-1.6 (1.65-1.6)
153,954
65,039
2.37

96 (87.8)
10.8 (5.2)
0.062 (0.179)

42
2.2.3 Structure determination and refinement
The structure of AmyA was solved by the molecular replacement method
using the program MOLREP (Vagin and Taplyakov, 1997). The initial phases were
calculated by using the TIM barrel domain of maltogenic amylase (Protein Data Bank
accession code 1SMA) and the model was adjusted by using O (Jones, et al, 1991).
Subsequently, the C domain was built into electron density by using a partially refined
model. The final model was built by using ARP/wARP (Perrakis et al, 1999). This
model was used for rigid body refinement against the 2.0 Å native data. Iteratively,
the 2|F
o
|-|F
c
| and

|F
o
|-|F
c
| maps were created to correct the misfit of the model. Water
molecules were picked from the |F
o
|-|F
c
| map at the 3.0 σ level and checked with the
2|F

o
|-|F
c
|

map at the 2.0 σ level. Positional and temperature factor parameter
refinement was repeatedly performed until the structure was refined to an R-factor of
0.201 and an R
free
of 0.221 using reflections with |F| > 0 σ(|F|). The calcium and
chloride ions were modeled by the use of both the 2|F
o
|-|F
c
| and |F
o
|-|F
c
| SIGMAA
weighted maps. All stages of crystallographic refinement made use of CNS (Brunger
et al, 1998). The geometry of the final model was checked with PROCHECK
(Laskowski et al, 1993) and all parameters were within acceptable ranges. The
refinement statistics are listed in Table 2.2.

2.2.4 Construction of N-terminal signal peptide cleaved AmyA
The primary amino acid sequence of AmyA was analyzed by using the
SignalP V 2.0 (Henrik et al, 1997) server and we have found that AmyA has a signal
peptide at its N-terminal in residues 1-25. The N-terminal signal peptide removed
AmyA gene was subcloned into the pGEX-6p-1 vector (Pharmacia) between the
EcoRI and BamH1 sites as a Glutathione S-Tranferase (GST) fusion protein. The

43
Table 2.2 Refinement statistics of AmyA at low salt. Values in
parentheses are for the last resolution shell.




1
R
work
=∑
hkl
||Fo(hkl)| – |Fc(hkl)|| / ∑
hkl
|Fo(hkl)|.
2
R
free
is equivalent to R
work

and is calculated for a randomly chosen 10% of reflections but omitted from
the refinement process.


Over expression construct was transformed into E. coli BL21 (DE3) cells and the cells
were grown in LB broth supplemented with 100 μg ml
-1
Ampicillin. Over expression

Refinement
Resolution range (Å) 10–1.6 (1.7–1.6)
|F|/σ(|F|)
>0
Protein atoms 4025
Water molecules 551
Calcium 2
Chloride 0
R work (%)
1
20.1(23.6)
R free (%)
2
22.1(26.6)
Reflections (working/test) 56,419/6,331
RMS deviations from ideal stereochemistry
Bond lengths (Å) 0.006
Bond angles (º) 1.27
B factors
Mean B factor (Protein) (Å
2
)
17.9
Mean B factor (Water) (Å
2
)
31.5
Mean B factor (Ions) (Å
2
)

13.7

44
of AmyA was induced with 0.1 mM IPTG for overnight at 30 ºC. Cells were
harvested by centrifugation, lyzed in 50 mM Tris (pH 8.0), 1 mM DTT and 500 mM
NaCl. The over expressed AmyA protein was purified using Sepharose 4B column
chromatography following which the GST tag was cleaved with the precision protease
purchased Amersham. The resulting protein preparation was purified by a heat step
consisting of incubation at 65 ºC for 30 minutes and the denatured impurities resulted
after the heat incubation were removed by centrifugation. The Amya protein was
further purified using Superdex-75 gel filtration column chromatography and dialyzed
against 50 mM Tris (pH 8.0) containing different NaCl concentration depending upon
the experimental need.

2.2.5 Enzymatic assays
Amylase activity assays were performed by using the EnzCheck Amylase
Activity Assay kit (Molecular Probes). α-amylase from Bacillus sp, which was
provided with the kit, served as a positive control enzyme. One unit is defined as the
amount of enzyme required to liberate 1 mg maltose from starch in 3 minutes under
assay conditions. The data presented are the means of four individual readings,
collected from four individual experiments.

2.3 RESULTS
2.3.1 Overview of AmyA structure
AmyA consists of a single polypeptide chain of 488 residues and comprises a
total of 18 α-helices and 20 β-strands. The overall fold of AmyA consists of three
distinct domains, A, B and C (Fig. 2.3). The central A domain forms the (β/α)
8
TIM
barrel structure through residues 27–131 and 198–436. The symmetry of the central

45
(β/α)
8
TIM barrel is interrupted by domain B that consists of 66 residues (132–197),
which is inserted between strand β5 and helix α5. Domain B consists of two α-helices,
two beta strands and a highly flexible and long loop protruding close to the active site.
On a structural level, domain B is the least conserved segment among α-amylases
from different origins and other carbohydrate-processing enzymes.





Figure 2.6 A ribbon diagram of the AmyA molecule. The three
domains are represented in different colors and the calcium ions are
shown as red spheres.

The B domain of AmyA is very similar to that of oligo-1,6-glucosidase and
isomaltulose synthase, Fig. 2.4 (Watanbe et al, 1997; Zhang et al, 2003). These two
enzymes are structurally very similar to AmyA. Even though all three enzymes
possess the glycosidic activity, the glucosyltransferase activity is present only in

46
Figure 2.7 Comparison of the domain B structure of AmyA (green)
with that of oligo-1,6-glucosidase (PDB entry 1UOK, blue) and
isomaltulose synthase (PDB entry 1M53, magenta).

oligo-1,6-glucosidase and isomaltulose synthase and is absent in AmyA. A loop in
domain B that contains residues 159-173, which reside in the vicinity of the catalytic

site and have very high temperature factors, shows multiple occupancies in the
electron density map and does not have sequence homology with α-amylases.
However, this loop has high sequence homology with the carbohydrate binding loop
of a neuraminidase (Burmeister et al, 1992), suggesting that this loop could be
involved in substrate binding.
Domain C is located exactly opposite to domain B on the other side of domain
A. The C-terminal domain, which consists of 79 residues (437–515), forms a
separated folding unit, exclusively made up of β-sheets. Eight of the ten strands fold

47
into a β-sandwich structure with the ‘Greek key’ topology. Domain C is also highly
conserved among α-amylase family enzymes and is reported to be involved in
substrate binding (Zhang et al, 2003). However, the clear mechanism is not yet
known. The overall fold and domain structure of AmyA are similar to those of the
members of family 13 glycoside hydrolases. Despite these similarities, a key and
significant attribute that has been identified from sequence alignment and structural
comparison of the AmyA structure with other homologous hydrolases is that while
most of the conserved residues are buried, the surface accessible residues are unique
(Fig. 2.5).
There are marked differences between the loops connecting the secondary
structure elements of the members of the α-amylase family

of enzymes, especially
those that surround the active site. These

loops might be responsible for the distinct
catalytic properties

among these enzymes. Comparison of the loop structures between

members of the family reveals that AmyA has some novel extended loops that bind
calcium ions. The calcium binding loop that contain residues 66-73 and the loop
containing residues 230-239 immediately after the catalytic residues Arg222 and
Asp224 are absent in other α-amylases.
Structural alignment results from the DALI server (Holm and Sander, 1993)
show that the tertiary structure of AmyA has the highest structural similarity to oligo-
1,6-glucosidase from Bacillus cereus (Watanbe et al, 1997). A total of 468 residues
(all atoms) could be superimposed with a root mean square deviation (RMSD) of 2.79
Å. RMS differences for the individual domains(all atoms) have been calculated by
using the program TM-align (Zhang et al, 2005) and have values of 2.52, 1.78 and
2.43 Å for the A, B and C domains, respectively.

48
Figure 2.8 The sequence alignment of AmyA. The amino acid sequence alignment
of AmyA with mesophilic α-amylase from Bacillus Cereus (1UOK), thermophilic α-
amylases from Thermus Sp (1SMA) and Thermoactinomyces vulgaris R- 47 (1JI2).
Conserved residues are in blue boxes; identical residues among all the four sequences
are indicated by white letters with red background and similar residues are indicated
by red letters. The secondary structure elements of AmyA are shown on top of the
AmyA sequence. Surface accessibility of residues is shown at the bottom of the
aligned sequences. The most accessible residues are shown in dark blue and buried
residues are color coded in white.

49

2.3.2 Catalytic site
Sequence alignment with various α-amylases shows that the well conserved
catalytic triad, Asp224, Glu260, Asp330 (AmyA numbering), is located at the center
of the barrel (Fig. 2.6). The overall conservation of other catalytic residues and water
molecules in conjunction with the previous structural and biochemical analyses

suggests that the enzymatic mechanism of AmyA is very similar to the double-
displacement catalytic mechanism of other known α-amylases (Strobl et al, 1998).







Figure 2.9 A stereo view of the active site. A stereo view of the
active site cleft at the center of the TIM barrel domain is shown. The
conserved catalytic residues and water molecules are shown as sticks
and spheres, respectively.

2.3.3 Calcium binding
The AmyA structure binds to two calcium ions. The calcium sites have been
identified from the difference Fourier map as strong peaks with appropriate
coordination atoms. The ions are hexa-coordinated and the geometry is described as a
distorted octahedron, which is often observed in the metal ion binding sites of metallo

50
proteins. The residues and the water molecules that make coordination bonds with the
two calcium ions are listed below with measured distances, Table 2.3.

Table 2.3 Coordinating atoms of calcium and distances.

Calcium 1

Calcium 2







Ca 1, the calcium ion, which is coordinated by residues within 44 to 52, is
present in most α-amylases, Figs. 2.7, 2.8. Also, a novel calcium site, Ca2, is also
present in AmyA. This calcium ion is not present in other amylase family enzymes. In
the previous studies it has been shown that heating will release calcium from the
protein. Even though no calcium salts were added during enzyme purification or
crystallization and AmyA was heated up to 65 °C for 30 minutes during purification,
the presence of calcium ions in the structure indicates that these calcium binding
pockets have very high affinity for calcium. However, the conserved calcium binding
No Interacting Atoms Distance
in Å
1 ASP 44 OD1 2.47
2 ASP 46 OD1 2.49
3 ASP 48 OD1 2.50
4 ILE 50 O 2.45

5 ASP 52 OD2 2.48
6 WAT 27 2.70
No Interacting Atoms Distance
in Å
1 ASP 65 OD1 2.59
2 ASP 67 O 2.50
3 THR 70 O 2.64
4 THR 70 OG1 2.58
5 ASP 73 OD1 2.59
6 WAT 69 2.70