Exploring the role of a glycine cluster in cold adaptation
of an alkaline phosphatase
Konstantinos Mavromatis
1,
*, Iason Tsigos
2,
*, Maria Tzanodaskalaki
2
, Michael Kokkinidis
1,3
and Vassilis Bouriotis
1,2
1
Department of Biology, Division of Applied Biology and Biotechnology, University of Crete, Greece;
2
Institute of Molecular Biology
and Biotechnology (IMBB), Enzyme Technology Division, and the
3
Institute of Molecular Biology and Biotechnology,
Crystallography Division, Heraklion, Crete, Greece
In an effort to explore the role of glycine clusters on the cold
adaptation of enzymes, we designed point mutations aiming
to alter the distribution of glycine residues close to the active
site of the psychrophilic alkaline phosphatase from the
Antarctic strain TAB5. The mutagenesis targets were
residues Gly261 and Gly262. The replacement of Gly262 by
Ala resulted in an inactive enzyme. Substitution of Gly261
by Ala resulted to an enzyme with lower stability and
increased energy of activation. The double mutant G261A/
Y269A designed on the basis of side-chain packing criteria
from a modelled structure of the enzyme resulted in restor-

Although the strategy of adaptation is unique to each
enzyme [15], it has been observed that amino-acid residues
involved in the catalytic mechanism are generally conserved
in psychrophilic and mesophilic enzymes [1]. This suggests
that generally the molecular basis of cold adaptation is
associated with sequence changes outside the active site.
However, recent work from our group indicated that the
psychrophilic character of an enzyme could also be altered
or masked by mutating active site residues [16]. Several
sequence patterns have been associated with psychrophilic
adaptations, such as decreased levels of Pro and Arg
residues, weakening of intramolecular interactions,
increased solvent interactions, decreased charged residues
interactions, and disulfide bonds [1,2,17]. Increased levels of
Gly residues or the establishment of Gly clusters have been
frequently suggested to be associated with psychrophilicity
[2]. This could be a result of increased local structural
flexibility due to the intrinsic flexibility of Gly residues [18].
However, recent studies of Gly clusters [19] appear to
contradict this assumption. It seems that the correlation
between the occurrence of Gly residues and the stability of
proteins is complex as several parameters from the whole
protein structure are involved and not just the intrinsic
flexibility of Gly residues [20].
We have recently reported the cloning, sequencing and
overexpression of the gene encoding alkaline phosphatase
from the Antarctic strain TAB5 [16]. Based on the crystal
structure (at 2.4 A
˚
)ofanEscherichia coli alkaline phospha-

and MINOTECH (Heraklion, Greece). All chemicals
were of analytical grade for biochemical use. PCR primers
were purchased from the Microchemistry Laboratory of
IMBB.
Enzymatic assay
Alkaline phosphatase activity was followed spectrophoto-
metrically utilizing p-nitrophenyl phosphate (pNPP) as
substrate. The release of product, p-nitrophenolate, was
monitored by measuring the absorbance at 405 nm using a
PerkinElmer photometer. Specific activity was determined
in a buffer containing 1
M
diethanolamine/HCl (pH 10),
10% glycerol, 10 m
M
MgCl
2
,1m
M
ZnCl
2
,and10m
M
pNPP at 20 °C. Enzyme units were calculated as previously
described [22].
Steady-state enzyme kinetics
Steady-state enzyme kinetics were performed in the tem-
perature range 5–25 °C. The program
HYPER
v1.01was

deviations do not exceed 10%.
Site-directed mutagenesis
Site directed mutagenesis was performed using standard
PCR methods [23]. For the construction of the mutations
the following primers were synthesized: Gly261 to Ala,
upper primer 5¢-d(CAAATAGATTGGGCTGGCCATG
CAAATAAT)-3¢, lower primer 5¢-d(TATTTGCATGGCC
AGCCCAATCTATTTGAG)-3¢; Gly262 to Ala, upper
primer 5¢-d(ATAGATTGGGGTGCCCATGCAAATAA
TGCA)-3¢, lower primer 5¢-d(ATTATTTGCATGGGCA
CCCCAATCTATTTG)-3¢; Tyr269 to Ala, upper primer
5¢-d(TAATGCATCCGCTTTAATTTCTGAAATTA
ATG)-3¢, lower primer 5¢-d(TCAGAAATTAAAGCGG
ATGCATTATTTGCATG)-3¢.
The upstream primer containing the NdeI restriction site
(underlined) was: 5¢-d(GCTAG
CATATGAAGCTTAAA
AAAATTG)-3¢ and the downstream primer containing the
EcoRI restriction site (underlined) was: 5¢-d(TT
GAATTC
GTTTATTGATTCCACTTTG)-3¢.
The PCR reaction mixtures were incubated on an
Eppendorf thermal cycler for 30 cycles of 94 °Cfor1min,
49 °C for 1 min, and 72 °C for 1 min. The amplified
product was isolated by agarose gel electrophoresis, gel
purified using QIAEX (Qiagen) and digested with NdeIand
EcoRI restriction enzymes. The resulting NdeI–EcoRI
fragment was inserted into the pRSETA vector previously
digested with these enzymes. The ligation mixture was used
to transform competent cells of E.colistrain XL1-MRF.

Ó FEBS 2002 Mutagenesis of a psychrophilic alkaline phosphatase (Eur. J. Biochem. 269) 2331
respectively. By introducing an Ala residue in the place of
Gly it is expected that the conformational flexibility of the
main chain can be constrained with a minimum perturba-
tion of the local structure, resulting to a more rigid protein
(Fig. 2). Moreover, Ala residues are common among
phosphatases at these positions (Fig. 1).
On the basis of the molecular model [21] Ala261 is
expected to introduce steric clashes with the side chain of
Tyr269 (Fig. 2B), which are not present in the structure of
the psychrophilic enzyme with the smaller Gly residue at
position 261 (Fig. 2A). Replacement of Tyr269 by Ala in
the double mutant G261A/Y269A is expected to remove
most of the spatial constraints of the side chain interactions
(Fig. 2C).
Temperature dependence of activity in wild-type
and mutant enzymes
The specific activity of all mutants was measured over the
entire range of temperature (5–25 °C) where wild-type
alkaline phosphatase is stable (Fig. 3A). Mutant G262A
had no significant activity at all temperatures tested making
it impossible to measure the specific activity or any kinetic
parameters of this mutant. We could only measure traces of
activity after prolonged incubation (24 h).
The mutant G261A is more active at elevated tempera-
tures (20–25 °C) compared to wild-type protein, while the
mutant G261A/Y269A is less active at any given tempera-
ture. However, compared to the mesophilic enzyme from
E.coli, these enzymes are approximately 10 times more
active.

thenativeone.
DISCUSSION
Recent studies have established that, adjustment of con-
formational flexibility is essential for the temperature
adaptation of enzymes [26]. Moreover, localized increases
in conformational flexibility constitute an essential element
in cold adaptation [9]. However, our incomplete under-
standing of the relation between enzyme properties and
conformational flexibility limits the exploitation of the full
potential of protein engineering in the redesign of psychro-
philic enzyme properties [15]. In particular, the effects of
local flexibility in psychrophilic enzyme properties have
been so far studied only for regions, which indirectly affect
the mobility of active site structures, but not for the active
sitesthemselves[9].
Fig. 2. Drawing of the three dimensional model of the wild type (A) and
mutant alkaline phosphatases G261A (B) and G261A/Y269A (C); only
residues that where studied are shown.
2332 K. Mavromatis et al. (Eur. J. Biochem. 269) Ó FEBS 2002
In a previous study [16], we explored the possibility of
modifying the psychrophilic properties of an enzyme by
introducing, via mutagenesis, predictable flexibility changes
to key active site residues of the psychrophilic alkaline
phosphatase from the Antarctic strain TAB5. This
approach was based on an approximate homology-based
three-dimensional model of the psychrophilic enzyme and
sequence comparisons with mesophilic sequences. The
mutagenesis targets were residues Trp260 and Ala219 of
the catalytic site and His135 of the Mg
2+

diethanolamine-
HCl pH 10, 10% glycerol, 10 m
M
MgCl
2
,1m
M
ZnCl
2
,and10m
M
pNPP. E
a
values were calculated from the slope of the Arrhenius plots in the
temperature range 5–25 °C for native and G261A/Y269A mutant, and 5–15 °C for the G261A mutant. Thermodynamic parameters DG
#
, DH
#
,
TDS
#
were calculated as described previously [27].
Enzyme
k
cat
(s
)1
)
E
a

)1
)
TD(DS
#
)
n-m
(kJÆmol
)1
)
Native 1212 42.8 52.48 40.45 )12.03
G261A 423 106.5 54.96 104.15 49.19 )2.48 )63.7 )61.22
G261A/Y269A 310 45.1 55.69 42.75 )12.94 )3.21 )2.3 0.91
Fig. 3. Kinetic studies of wild-type and mutant alkaline phosphatases.
(A) Temperature dependence of k
cat
of TAB5 (d), mutants G261A
(r), G261A/Y269A (j)andE.coli (·) alkaline phosphatases at
temperature range 5–25 °C. k
cat
values were determined in a buffer
containing 1
M
diethanolamine-HCl pH 10, 10% glycerol, 10 m
M
MgCl
2
,1m
M
ZnCl
2

functional importance of the Gly pair, located in the vicinity
of the active site of the cold adapted enzyme and to study its
potential role in the establishment of its psychrophilic
character.
This work uses, in accordance with more or less generally
established concepts, the energy of activation, E
a
,asthe
main criterion for the evaluation of the psychrophilic nature
of enzyme variants. In cold adapted enzymes, this param-
eter generally tends to be lower compared to their
mesophilic counterparts [27]. Furthermore, as a measure
of enzyme stability, thermal inactivation at 50 °Cisused.
We refer to stability in an activity sense and not in a
thermodynamic sense. We therefore assume that even low
enzymatic activity is associated with a mutant that retains to
a considerable extent the overall fold of the wild-type
protein and that loss of activity is associated either with
perturbation of the native structure or local disruption of
the metal binding or the active site.
The point mutation of Gly262 fi Ala results in an
enzyme that exhibits very low activity (less than 1 : 1000
of the native enzyme). This fact did not allow the study
of its kinetic parameters and its thermal inactivation
profile. However, this mutation demonstrates that at
position 262 the presence of Gly is essential, and a
mutation altering this residue results in a practically
inactive enzyme. This Gly may provide the necessary
flexibility required for catalysis. Several alkaline phos-
phatases have one Gly at the corresponding positions

at temperatures > 20 °C, indicating that the E
a
value in this
temperature range is considerably lowered. On the basis of
the model, the behavior of the G261A variant can be
interpreted in terms of constraints introduced by the Ala
side chain. The presence of the additional Gly at position
261 possibly offers increased flexibility to the adjacent
residue Trp260 that forms part of the active site and
therefore facilitates the catalysis at low temperatures.
Consequently, when the mutant G261A is driven to operate
in a cold environment, and the lack of Gly261 does not
allow the reaction to proceed as efficiently as in the case of
the native enzyme. At higher temperatures, the additional
energy provided by the environment is sufficient and the
mutant can proceed with the catalysis as efficiently as the
wild type (Fig. 3A). Investigation of the three-dimensional
homology-based model of the enzyme revealed that the
methyl group of Ala261 side-chain could produce steric
clashes with the aromatic ring of Tyr269, and these
unfavorable interactions could lead to a decrease of local
flexibility and an increased E
a
value.
The validity of the above interpretation was further
reinforced by the construction of the double mutant
G261A/Y269A. The additional substitution of Tyr269 fi
Ala was designed with the aim of reducing the spatial
constraints originating from the side-chain interactions
between Tyr269 and Ala261 (Fig. 2C). The main difference

engineered catalysts was analyzed in various temperatures.
However, it was found that although the HGGG loop was a
critical structural element for the catalytic efficiency of the
enzyme, the cold adaptation of the enzyme could not be
attributed to the presence of the Gly cluster in this element.
The present study supports the idea that the Gly cluster,
in combination with its structural environment, is an
essential feature of the psychrophilic character of TAB5
alkaline phosphatase. It seems that the volume of the side-
chains at positions 261 and 269 controls the psychrophilic
character as judged from the levels of the E
a
. In the G261A
mutant, this volume is increased (Fig. 2B) and the enzyme
proves to be as efficient as the native at elevated but not at
lower temperatures. The presence of Gly residues at both
positions 261 and 262 is necessary for the enhanced specific
activity of the enzyme in its natural environment; catalysts
harboring a Gly fi Ala mutation in any of these positions
exhibit a significantly decreased specific activity (Fig. 3A).
Consequently, the Gly cluster at this position plays a dual
role, contributing both to higher catalytic efficiency and
lower E
a
.
Moreover, the present work provides evidence that
mutations introduced to Gly cluster produced enzymes that
still exhibit psychrophilic properties while suitable compen-
sating mutations may even produce mutants with increased
stability. To our knowledge, the present study along with a

Arnold, F.H. (2000) Directed evolution study of temperature
adaptation in a psychrophilic enzyme. J. Mol. Biol. 297, 1015–
1026.
7. Wintrode, P.L., Miyazaki, K. & Arnold, F.H. (2001) Patterns of
adaptation in a laboratory evolved thermophilic enzyme. Biochim.
Biophys. Acta 1549, 1–8.
8. Wintrode, P.L., Miyazaki, K. & Arnold, F.H. (2000) Cold adap-
tation of a mesophilic subtilisin-like protease by laboratory evo-
lution. J. Biol. Chem. 275, 31635–31640.
9. Fields, P.A. & Somero, G.N. (1998) Hot spots in cold adaptation:
localized increases in conformational flexibility in lactate dehy-
drogenase A4 orthologs of Antarctic notothenioid fishes. Proc.
Natl Acad. Sci. USA 95, 11476–11481.
10. Lonhienne, T., Zoidakis, J., Vorgias, C.E., Feller, G., Gerday, C.
& Bouriotis, V. (2001) Modular structure, local flexibility and
cold-activity of a novel chitobiase from a psychrophilic Antarctic
bacterium. J. Mol. Biol. 310, 291–297.
11. Aghajari, N., Feller, G., Gerday, C. & Haser, R. (1998) Crystal
structures of the psychrophilic alpha-amylase from Alteromonas
haloplanctis in its native form and complexed with an inhibitor.
Protein Sci. 7, 564–572. [published erratum appears in Protein Sci.
(1998) 7, 1481]
12. Alvarez, M., Zeelen, J.P., Mainfroid, V., Rentier-Delrue, F.,
Martial, J.A., Wyns, L., Wierenga, R.K. & Maes, D. (1998)
Triose-phosphate isomerase (TIM) of the psychrophilic bacterium
Vibrio marinus. Kinetic and structural properties. J. Biol. Chem.
273, 2199–2206.
13. D’Amico, S., Gerday, C. & Feller, G. (2001) Structural deter-
minants of cold adaptation and stability in a large protein. J. Biol.
Chem. 276, 25791–25796.

tions. Eur. J. Biochem. 267, 1230–1238.
22. Garen, A. & Levinthal, C. (1960) A fine structure genetic and
chemical study of the enzyme alkaline phosphatase of E.coli.1.
Purification and characterization of alkaline phosphatase.
Biochim. Biophys. Acta 38, 470–483.
23. Horton, R.M., Cai, Z.L., Ho, S.N. & Pease, L.R. (1990) Gene
splicing by overlap extension: tailor-made genes using the poly-
merase chain reaction. Biotechniques 8, 528–535.
24. Kim, E.E. & Wyckoff, H.W. (1991) Reaction mechanism of
alkaline phosphatase based on crystal structures. Two-metal ion
catalysis. J. Mol. Biol. 218, 449–464.
25. Guex, N. & Peitsch, M.C. (1997) SWISS-MODEL and the Swiss-
PDB Viewer: An environment for comparative protein modeling.
Electrophoresis 18, 2714–2723.
26. Zavodszky, P., Kardos, J. & Svingor & Petsko, G.A. (1998)
Adjustment of conformational flexibility is a key event in the
thermal adaptation of proteins. Proc. Natl Acad. Sci. USA 95,
7406–7411.
27. Lonhienne, T., Gerday, C. & Feller, G. (2000) Psychrophilic
enzymes: revisiting the thermodynamic parameters of acti-
vation may explain local flexibility. Biochim. Biophys. Acta 1543,
1–10.
Ó FEBS 2002 Mutagenesis of a psychrophilic alkaline phosphatase (Eur. J. Biochem. 269) 2335