Tài liệu Báo cáo khoa học: Comparative studies of endonuclease I from cold-adapted Vibrio salmonicida and mesophilic Vibrio cholerae docx

Comparative studies of endonuclease I from cold-adapted
Vibrio salmonicida and mesophilic Vibrio cholerae
Bjørn Altermark
1
, Laila Niiranen
2
, Nils P. Willassen
1,2
, Arne O. Smala
˚
s
1
and Elin Moe
1
1 Norwegian Structural Biology Centre, Faculty of Science, University of Tromsø, Norway
2 Department of Molecular Biotechnology, Faculty of Medicine, University of Tromsø, Norway
The marine and estuarine environment harbors a vast
diversity of bacteria. Some of the most extensively
studied marine or estuarine bacteria belong to the
genus Vibrio, with Vibrio cholerae being the most
notorious species as it is the cause of cholera in
humans. V. cholerae is found in tropical and temper-
ate areas, and can be classified as a mesophilic bac-
terium with growth optimum around 37 °C. It prefers
estuarine waters, is halotolerant, and does not require
NaCl for growth [1,2]. The bacterium with one of the
lowest growth optimum temperatures found in the
genus Vibrio is the fish pathogen Vibrio salmonicida.
It has an optimal growth temperature of % 15 °C and
requires NaCl for growth [3]. It can therefore be
classified as a psychrophilic and mildly halophilic

bacterium.
A living cell can be considered as a chemical factory
which produces many substances. The speed of pro-
duction is limited by reaction rates. The reaction rates
are in turn limited by, among other things, environ-
mental factors such as pH, salinity, pressure and tem-
perature. Temperature is a very important factor for
growth and proliferation of the cells. At high tempera-
tures, at which thermophiles thrive, chemical reaction
rates are very high, and the main challenge for cells is
to adapt their enzymes, membranes and molecules to
cope with the heat. At low temperatures, the chemical
reaction rates are lower, and hence, in order to be
competitive and grow fast at low temperatures, evolu-
tionary pressure favors enzymes that are more efficient
than their high-temperature counterparts. This higher
efficiency at low temperatures is believed to be caused
Keywords
cold adaptation; endonuclease I;
psychrophilic enzymes; salt adaptation;
stability
Correspondence
E. Moe, Norwegian Structural Biology
Centre, Faculty of Science, University of
Tromsø, N-9037 Tromsø, Norway
Fax: +47 77644765
Tel: +47 77646473
E-mail:
(Received 14 September 2006, revised
2 November 2006, accepted 9 November

2006)
doi:10.1111/j.1742-4658.2006.05580.x
Endonuclease I is a periplasmic or extracellular enzyme present in many
different Proteobacteria. The endA gene encoding endonuclease I from the
psychrophilic and mildly halophilic bacterium Vibrio salmonicida and from
the mesophilic brackish water bacterium Vibrio cholerae have been cloned,
over-expressed in Escherichia coli, and purified. A comparison of the enzy-
matic properties shows large differences in NaCl requirements, optimum
pH, temperature stability and catalytic efficiency of the two proteins. The
V. salmonicida EndA shows typical cold-adapted features such as lower
unfolding temperature, lower temperature optimum for activity, and higher
specific activity than V. cholerae EndA. The thermodynamic activation
parameters confirm the psychrophilic nature of V. salmonicida EndA with
a much lower activation enthalpy. The optimal conditions for enzymatic
activity coincide well with the corresponding optimal requirements for
growth of the organisms, and the enzymes function predominantly as
DNases at physiological concentrations of NaCl. The periplasmic or extra-
cellular localization of the enzymes, which renders them constantly exposed
to the outer environment of the cell, may explain this fine-tuning of bio-
chemical properties.
Abbreviations
DSC, differential scanning calorimetry; VcEndA, recombinant Vibrio cholerae endonuclease I; VsEndA, recombinant Vibrio salmonicida
endonuclease I.
252 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
by a more flexible structure, and the increased flexi-
bility is thought to be the reason for the lower thermo-
stability of cold-adapted enzymes [4].
Endonuclease I is a periplasmic or extracellular
enzyme known to cleave both RNA and DNA at
unspecific internal (endo) sites. It also cleaves plasmids

and single-stranded DNA [5]. The enzyme cleaves at
the 3¢ side of the phosphodiester bond, leaving prod-
ucts with 5¢ phosphate ends. A histidine functions as a
general base, which activates a water molecule for an
in-line attack on the scissile phosphate in the DNA
substrate. The role of the active-site magnesium ion is
to stabilize the phosphoanion transition state and
make a proton available for the 3¢-oxygen leaving
group, via a bound water molecule. An arginine is
believed to stabilize the product via a hydrogen bond
to the phosphate, which also decelerates the reverse
reaction [6]. A chloride atom is found buried in the
structure of V. cholerae endonuclease I and probably
also in the Vibrio vulnificus endonuclease I structure
[7].
Orthologues of endonuclease I from many bacterial
species are described in the literature [5,8–13], but
there seems to be an uncertainty about the main func-
tion of this enzyme in vivo. It is well known for its
ability to reduce the level of transformation [14–16],
but appears to have no effect on conjugation [5]. The
enzyme is shown not to be involved in the patho-
genicity of V. cholerae [15], V. vulnificus [5] or Erwinia
chrysanthemi [17]. Most of the bacteria that harbor the
gene live in close contact with eukaryotic hosts, which
may provide nutritious DNA and RNA through their
mucus barriers. The mucus itself becomes less viscous
if the DNA is broken down, and this may facilitate the
movement of the bacterium through the mucus layer
[15]. The enzyme is reported to be constitutively

expressed in V. vulnificus [5] and Erwinia chrysanthemi
[18].
Here we report the cloning, expression and purifica-
tion of the endonuclease I enzymes from the psychro-
phile V. salmonicida (VsEndA) and the mesophile
V. cholerae (VcEndA). The two orthologous enzymes
have been biochemically and biophysically character-
ized to reveal possible differences related to environ-
mental adaptation.
Results
Sequence similarity and composition
VsEndA and VcEndA show 71% identity and 80%
similarity (Blosum62) at the amino acid level, when the
active enzymes are compared without their N-terminal
signal peptide. A sequence alignment of VcEndA and
VsEndA is shown in Fig. 1. The first two amino acids
at the N-terminus (Thr and Met) are encoded by the
expression vector.
An analysis of the amino-acid composition shows
that VsEndA contains 13 more lysines and two fewer
arginines than Vc EndA, resulting in a very high R ⁄ K
ratio for the mesophilic enzyme (1.6 versus 0.6,
respectively). In addition VsEndA contains two less
D + E than VcEndA. However all the cysteines
involved in disulfide bridge formation in VcEndA are
also found in the sequences of VsEndA (Fig. 1). The
theoretical pI was 9.61 for VsEndA and 8.62 for
VcEndA.
Expression and purification
From 7 L Escherichia coli culture, a total of 24 and

50 mg pure recombinant VsEndA and VcEndA pro-
teins, respectively, were obtained (Fig. 2). The final
NaCl concentration after cation-exchange chromato-
graphy was estimated to 0.8 m for VsEndA and 0.65 m
for VcEndA.
Fig. 1. Sequence alignment showing the amino acids of VsEndA and Vc EndA. Numbers indicate cysteines involved in disulfide bridges; stars
indicate Mg
2+
-co-ordinating residues, triangles indicate the catalytically important His80 and Arg99, and squares indicate Cl
)
-co-ordinating
residues. The sequence numbering is according to the structure of V. vulnificus endonuclease I in complex with a DNA octamer, PDB id.
1OUP [6].
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 253
The calculated molecular masses were 25.0 and
24.7 kDa for VsEndA and VcEndA, respectively,
which is in agreement with the results from the
SDS ⁄ PAGE analysis shown in Fig. 2.
Enzyme properties
To find the optimal buffer conditions for the biochemi-
cal characterization of the enzymes, we carried out an
analysis of the NaCl requirements and pH optimum
of VsEndA and VcEndA. The optimum NaCl concen-
trations for activity were found to be 425 mm for
VsEndA and 175 mm for VcEndA, respectively
(Fig. 3).
The optimum pH for activity of VsEndA and
VcEndA was % 8.5–9.0 and 7.5–8.0, respectively, when
measured in Tris ⁄ HCl and diethanolamine ⁄ HCl buffers

as shown in Fig. 4. The pH optimum was unaffected
by the NaCl concentration in the buffers. When tested
in glycine buffer at pH 9.0, the enzymes showed very
low activity compared with that in diethanolamine and
Tris buffers at the same pH, indicating that glycine
inhibits the enzymes. VcEndA activity decreases stee-
ply below pH 6.5 (measured in Bis-Tris buffer, data
not shown).
The optimum temperature for activity was deter-
mined using a modified Kunitz assay. The results
showed optimum activity at % 45 °C for VsEndA and
50 °C for VcEndA, as shown in Fig. 5.
Kinetic constants for VsEndA and VcEndA were
measured by incubating the enzymes in the presence of
substrate with different concentrations and at different
temperatures. The kinetic constants for the two
enzymes at 5, 15, 25, 30 and 37 °C are shown in
200
116.3
97.4
66.3
55.4
36.5
31.0
21.5
14.4
6.0
3.5
Fig. 2. SDS ⁄ PAGE. Lane 1, Mark12 MW ladder; lane 2, % 5 lg
VcEndA; lane 3, % 5 lg VsEndA. The relative molecular masses of

the standard are shown on the left.
0 200 400 600 800
0
25
50
75
100
VsEndA
VcEndA
[NaCl] (mM)
Activity (%)
Fig. 3. Optimum NaCl concentration for DNase activity. DNaseAlert
was used as substrate, and activity was measured in increasing
amounts of NaCl. Each replicate is plotted and the mean values are
drawn.
pH optimum VsEndA
A
6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Tris
DEA
pH
Activity (Rfu/s)
B

pH optimum VcEndA
6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
1.0
1.5
2.0
2.5
3.0
Tris
DEA
pH
Activity (Rfu/s)
Fig. 4. Optimum pH for activity. (A) VsEndA; (B) VcEndA. Buffers
used are 75 m
M Tris ⁄ HCl, pH 7–9, and 75 mM diethanolamine ⁄ HCl,
pH 8–10. DNaseAlert was used as a substrate in the assay. Each
replicate is plotted and the mean values are drawn.
Endonuclease I from V. salmonicida and V. cholerae B. Altermark et al.
254 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
Table 1. VsEndA possesses a higher k
cat
than VcEndA
at all temperatures, and the K
m
values of VcEndA are
slightly lower than for VsEndA at all temperatures.
The physiological efficiency is highest for VsEndA, but
the difference decreased with concomitant increase in
temperature.
As determined from Arrhenius plots, the energy
of activation (E

a
) is 35.7 kJÆmol
)1
for VsEndA and
76.3 kJÆmol
)1
for Vc EndA. The calculations of the
enthalpy (DH
#
) and entropy (DS
#
) of activation revealed
much lower values for VsEndA as shown in Table 2.
Also note that TDS
#
values for VsEndA were negative,
whereas those from VcEndA were positive (Table 2).
Temperature stability was analyzed by evaluating
thermal unfolding using differential scanning calorime-
try (DSC). The results revealed a T
m
of 44.8 °C for
VsEndA and 52.8 °C for VcEndA as shown in Fig. 6.
The calorimetric enthalpy (area under the transition) is
also much lower for VsEndA (328 kJÆmol
)1
) than for
VcEndA (480 kJÆmol
)1
).

The rate of irreversible unfolding was analyzed by
incubating both enzymes at 70 °C. Samples were
removed after 10 min and incubated for 1 h on ice
0 102030405060
0
25
50
75
100
VsEndA
VcEndA
Temperature (°C)
Activity (%)
Fig. 5. Optimum temperature for activity. The enzymes were
assayed using the modified Kunitz assay. Each replicate is plotted
and the mean values are drawn.
Table 1. Kinetic constants for VsEndA and VcEndA at 5, 15, 25, 30 and 37 °C.
T(°C) VsEndA VcEndA VsEndA ⁄ VcEndA
K
m
(nM) 5 246 ± 15 118 ± 13 2.1
15 202 ± 9.6 131 ± 10 1.5
25 169 ± 20 156 ± 17 1.1
30 208 ± 14 161 ± 12 1.3
37 181 ± 10 174 ± 10 1.0
k
cat
(s
)1
) 5 9.41 1.03 9.1

15 14.7 3.10 4.7
25 18.5 7.18 2.6
30 32.8 15.6 2.1
37 48.4 32.1 1.5
k
cat
⁄ K
m
(s
)1
ÆnM
)1
) 5 0.0383 0.00873 4.4
15 0.0728 0.0237 3.1
25 0.109 0.0461 2.4
30 0.158 0.0972 1.6
37 0.268 0.185 1.4
Table 2. Activation energy parameters were calculated (kJÆmol
)1
) for the psychrophilic VsEndA (p) and mesophilic VcEndA (m). The differ-
ences in values (p ) m) is also shown.
T
(°C) Enzyme DG
#
DH
#
TDS
#
D(DG
#

)
p-m
D(DH
#
)
p-m
TD(DS
#
)
p-m
5 p 62.8 33.4 ) 29.4 ) 5.1 ) 40.6 ) 35.4
m 67.9 74.0 6.1
15 p 64.0 33.3 ) 30.7 ) 3.7 ) 40.6 ) 36.8
m 67.8 73.9 6.1
25 p 65.8 33.2 ) 32.6 ) 2.3 ) 40.6 ) 38.2
m 68.1 73.8 5.7
30 p 65.5 33.2 ) 32.3 ) 1.9 ) 40.6 ) 38.7
m 67.4 73.7 6.4
37 p 66.1 33.1 ) 32.9 ) 1.1 ) 40.6 ) 39.5
m 67.1 73.7 6.6
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 255
before being assayed. Figure 7 shows that the rate
of irreversible unfolding for VsEndA is higher than
for Vc EndA, with a half-life of % 13 and 33 min,
respectively.
Substrate specificity analysis
An analysis of the substrate specificity for DNA of the
enzymes shows that they both cleave plasmid DNA,
dsDNA and ssDNA as shown in Fig. 8.

To test the RNA specificity of the enzymes, we used
the RNaseAlert assay and compared the results with
those obtained using the DNaseAlert assay. VsEndA
has over 900-fold higher preference for DNA than
RNA when measured in buffer with NaCl concentra-
tion optimal for DNase activity. The RNase activity is
inhibited in the presence of NaCl as shown in Fig. 9,
and at 425 mm NaCl the VsEndA is predominantly a
DNase. The VcEndA shows the same trend, with very
low RNase activity at NaCl concentration optimal for
DNase activity.
Discussion
The choices of enzyme orthologues, and their phylo-
genetic relationship, which has been investigated in
order to elucidate the cold-adapted properties of the
enzymes, have previously been criticized [19]. Here,
orthologue monomeric enzymes from species within
the same genus are studied to minimize other adapta-
tional strategies that may have affected these enzymes
differently. When the amino acid compositions of
VsEndA and VcEndA are compared, a remarkably
low R ⁄ K ratio in VsEndA is found. In addition, there
is a slight decrease in D + E. The difference in pI
reflects this substitution of charged residues, by being
one unit higher for VsEndA. VsEndA also binds much
more strongly to the SP Sepharose column because of
its higher positive charge compared with VcEndA. The
primary structure of VsEndA also contains an extra
lysine (Lys52a) which creates a gap in the alignment in
Fig. 1. The differences in charge between the two

enzymes may be involved in temperature adaptation;
however, two properties, which are not related to tem-
perature adaptation, clearly distinguish these enzymes.
The two enzymes respond notably differently to varia-
tions in both NaCl concentration and pH.
A notable increase in activity against the DNase-
Alert substrate was observed for the two enzymes
when NaCl was added to the assay buffer. The optimal
NaCl concentrations coincide with the salinities
encountered by the bacteria in their natural habitats.
Seawater at 3.5% salinity is composed of about
470 mm Na
+
ions and 540 mm Cl
–
ions [20]. The
30 40 50 60
0
10
20
30
40
50
60
VsEndA
VcEndA
Temperature (°C)
Cp (kJ mol
-1
K

-1
)
Fig. 6. DSC endotherms of VsEndA and VcEndA. Baseline subtrac-
ted data have been normalized for protein concentration.
0102030405060
1
10
100
VsEndA
VcEndA
Time (min)
Residual activity (%)
Fig. 7. Kinetic stability of VsEndA and VcEndA. Enzyme was incu-
bated at 70 °C. Samples were removed after 10 min and incubated
for 1 h on ice before being assayed using the DNaseAlert QC sys-
tem kit. Each replicate is plotted and the mean values are drawn.
AB
Fig. 8. Cleavage of plasmid, dsDNA and ssDNA. (A) 14 nM VcEndA
incubated at 23 °C for 5 min with plasmid (lane 2), dsDNA (lane 4)
and ssDNA (lane 6). Substrate incubated without enzyme is in
lanes 1, 3 and 5, respectively. (B) Substrate incubated with and
without 14 n
M VsEndA, as explained for VcEndA.
Endonuclease I from V. salmonicida and V. cholerae B. Altermark et al.
256 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
optimum NaCl concentration found for VsEndA cor-
responds quite well to this. V. cholerae resides in more
brackish water with lower salinity, and the lower
[NaCl] optimum of VcEndA reflects this. The optimal
salt concentrations were measured in a 75 mm Tris

buffer. The optima may be higher in a Tris buffer of
lower ionic strength, but this was not tested. Two
terrestrial orthologous endonucleases, one from the
plant pathogen Erwinia chrysanthemi and one from the
ruminal bacterium Fibrobacter succinogenes, are also
described in the literature [11,21]. The optimum
concentrations of NaCl for these enzymes are 0–
75 mm and 10 mm, respectively, with DNA as sub-
strate. It seems that the salt optima of the enzymes are
fine-tuned to match the salinity of their environment.
The outer membrane and cell wall of Gram-negative
bacteria do not restrict passage of ions, and the peri-
plasmic proteins are, like the extracellular proteins,
constantly exposed to the salinity of the surrounding
water. Knowledge on cold adaptation is in many
cases based on marine secreted enzymes. Detailed data
on salt adaptation of marine cold-adapted secreted
enzymes is lacking and may be a source of error in
the conclusions drawn [22]. For the endonuclease I
enzymes studied here, the effect of NaCl is very prom-
inent and underlines the need to dissect the different
adaptational strategies in future studies. The differ-
ences observed in the number of charged residues,
especially lysine, are probably related to adaptation to
both salinity and temperature. The K
m
of VsEndA is
higher than that of VcEndA; therefore, the more posit-
ive surface of VsEndA does not seem to significantly
increase the affinity for the negatively charged sub-

strate, and is apparently not a factor that aids VsEndA
in improving its catalytic efficiency. It is possible that
the K
m
is highly affected by the NaCl concentration in
the buffer, but this is not tested. Halophilic enzymes
have been reported to be more enriched in negatively
charged amino acids than their nonhalophilic counter-
parts [23,24]. This is the opposite to that found for the
enzymes studied here, in which the number of posi-
tively charged amino acids is increased. The chloride
atoms probably position themselves around the posit-
ive charges and make electrostatic interactions between
surface amino acids and between surface amino acids
and the substrate weaker. To counteract this, the
VsEndA may have developed a more positively
charged surface. It is possible that the surface charges
of the two enzymes are similar at their respective phy-
siological salt concentrations. The higher number of
lysines seen in VsEndA may result in increased flexibil-
ity, if the extra lysines repel other parts of the enzyme
and do not form stabilizing salt bridges or hydrogen
bonds. This may also lower the stability of the enzyme.
The Na
+
ions may affect the solvation of the phos-
phate groups in the DNA substrate, and it is possible
that the enzymes also have adapted strategies to
remove Na
+

around the phosphates of DNA before
catalysis can take place. It seems clear that the salt-
adapted and cold-adapted properties of VsEndA are
intertwined.
The differences in optimum pH for activity were
% 0.5–1 unit between the two enzymes as shown in
Fig. 4, with the optimum for VsEndA being shifted to
VsEndA
A
0 125 250 375 500
0.0
0.3
0.6
0.9
1.2
1.5
1.8
RNase
DNase
[NaCl] (mM)
Rfu/s
B
VcEndA
0 50 100 150 200
0.0
0.3
0.6
0.9
1.2
1.5

1.8
DNase
RNase
[NaCl] (mM)
Rfu/s
Fig. 9. DNase and RNase activity with increasing amounts of NaCl.
(A) VsEndA; (B) VcEndA. Enzyme was assayed using the DNase-
Alert and RNaseAlert QC system kits. Each replicate is plotted and
the mean values are drawn.
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 257
a higher pH. The pH optima for endonuclease activity
show a similar trend to that for growth of the corres-
ponding organisms. Activity in glycine buffer was very
low compared with that in Tris buffer at the same pH.
Citrate buffer has been shown to be inhibitory to Pro-
teus mirabilis endonuclease I [13]. Citrate and perhaps
also glycine may act as chelators that bind Mg
2+
and
thereby inhibit enzymatic activity, similarly to EDTA.
The carboxy group of the small amino-acid, glycine,
may also replace water molecules, which are bound
around the Mg
2+
ion of the enzyme and thereby inhi-
bit activity.
The kinetic analysis performed under optimal condi-
tions for each enzyme (Table 2) shows that VsEndA is
a better catalyst than VcEndA at all temperatures, and

the differences in catalytic efficiency (k
cat
⁄ K
m
) increase
with concomitant decrease in temperature. K
m
values
for VcEndA are lower than for VsEndA, indicating
that the former has slightly greater affinity for the sub-
strate. However, k
cat
is very different for the two
enzymes, especially at low temperatures, being 9 times
higher for VsEndA than for VcEndA. It is clear that
VsEndA adapts to lower temperatures by increasing
the k
cat
. The similar K
m
values of the two enzymes
may indicate that VsEndA is meant to function at high
substrate concentrations, at which the increase in k
cat
is more important for adaptation to low temperatures
[25]. The k
cat
values associated with both VsEndA
and VcEndA increase exponentially at temperatures
between 5 °C and 37 °C in accordance with the

Arrhenius equation:
k ¼ Ae
ÀE
a
=RT
ð1Þ
According to Eqn (1), there is an exponential decrease
in reaction rates (k) with decreasing temperature (T),
and the extent of this decrease depends on the activa-
tion energy, E
a
. The less steep slope for VsEndA when
the temperature is lowered in the temperature opti-
mum curve shown in Fig. 5 is a direct consequence of
the lower energy of activation.
Results from the thermodynamic calculations
(Table 2) reveal that there is a slight difference in the
free energy of activation between the two enzymes ori-
ginating from both the lower activation enthalpy
(DH
#
) and activation entropy (TDS
#
)ofVsEndA. The
TDS
#
values for VcEndA are positive, and, if we
assume that VcEndA is more rigid than VsEndA,
binding of substrate will not decrease the entropy of
activation to the same extent as in the psychrophilic

(and flexible) VsEndA. However, the method of calcu-
lation, especially for DS
#
, must be carefully interpreted
as stated by Cornish-Bowden [26]. Enthalpy calcula-
tions based on the experimentally determined values of
E
a
give more precise information, and it is clear that
VsEndA has adapted to low temperatures by lowering
the enthalpy of activation.
DSC measurements show that VsEndA is less ther-
mostable than VcEndA with an unfolding temperature
that is 8 °C lower. This is in agreement with results
from stability analysis of other cold-adapted enzymes,
which show reduced temperature stability compared
with their mesophilic homologues [27,28]. The results
support the theory of increased structural flexibility
leading to lower thermostability in cold-adapted
enzymes. The NaCl concentrations in which the ther-
mal scans were performed mimic the physiological con-
ditions that each of the enzymes face in their natural
environments. Thermal scans of VcEndA at [NaCl]
optimal for VsEndA (425 mm) revealed a higher T
m
,
and a thermal scan performed on VsEndA at [NaCl]
optimal for VcEndA (175 mm) revealed a lower T
m
than those found in optimal buffers (data not shown).

This highlights again that it is crucial to perform the
comparative analysis under physiological conditions
for each enzyme, as salt interferes with both the activ-
ity and stability of enzymes. Reversibility could be
detected by DSC, but the signal was very weak for
both proteins, probably because of aggregation and
destruction caused by the relatively long period at
elevated temperatures. As shown in Fig. 7, VsEndA
transforms into an irreversible unfolded state much
faster than VcEndA. However, a half-life of 13 min at
70 °C for VsEndA is substantially higher than that of
other cold-adapted enzymes [22]. Endonuclease I is
located in the periplasmic or extracellular space, and
the selective pressure to maintain stability must there-
fore be high. It would be a waste of energy to secrete
enzymes that denature quickly, so it is in the bacter-
ium’s interest for the secreted enzymes to be long lived.
However, it seems that, in order to achieve appropriate
activity at low temperatures, the enzyme must sacrifice
some of its stability. It has previously been suggested
that the lower thermal stability of cold-adapted
enzymes is simply a consequence of the lack of select-
ive pressure for stability [29]. A lack of selective pres-
sure for stability is not the case for this
periplasmic ⁄ extracellular protein, and our results indi-
cate that in order for it to be active at low tempera-
tures, its stability must be reduced.
The enzymes did not show any apparent difference
in ability to degrade plasmid DNA, dsDNA or
ssDNA. However, both VsEndA and VcEndA dis-

played decreasing activity against the RNaseAlert sub-
strate with concomitant increase in [NaCl], as shown
in Fig. 9. At physiological NaCl concentration, the two
enzymes have extremely low RNase activity and may be
Endonuclease I from V. salmonicida and V. cholerae B. Altermark et al.
258 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
considered solely as DNases. The highest RNase activity
is in buffer without added NaCl, but it is still % 3.5 times
(VsEndA) and 14 times (VcEndA) lower than the
DNase activity in the same buffer. The opposite effect
that NaCl addition seems to have on the RNase activity
of the enzymes may be linked to an increase in Na
+
around the phosphate groups and the 2¢-OH, which
reduces the negative charge, and hence the affinity of the
enzyme decreases with increasing NaCl concentration.
However, it seems clear that both VsEndA and VcEndA
are intended to function purely as DNases in vivo.
Conclusion
Endonuclease I from the psychrophilic bacterium
V. salmonicida is an enzyme that shows cold-adapted
features, such as lower thermal stability, lower tem-
perature optimum, and higher catalytic efficiency,
when compared with the corresponding enzyme from
the related mesophilic bacterium V. cholerae. The peri-
plasmic or extracellular localization of these enzymes
means that they are constantly exposed to the external
environment of the bacterium. Their differences in
enzymatic properties, such as pH optimum, salt opti-
mum and catalytic efficiency, seem to be fine-tuned to

match their respective environments. The salt-sensitive
and relatively low RNase activity of the enzymes indi-
cates that their physiological substrate is DNA. To our
knowledge, VsEndA is the first endonuclease described
that displays more than 90% activity against DNA in
0.5 m NaCl. This unique property in combination with
high activity at low temperatures and low RNase acti-
vity may be advantageous for future commercial
exploitation. Determination of the crystal structure of
VsEndA is in progress and will facilitate a detailed
explanation of the mechanisms behind the observed
cold-adapted properties, in addition to interesting dif-
ferences in pH and salt optima.
Experimental procedures
Bacterial strains and molecular biology materials
Genomic DNA from V. cholerae ATCC14035 and V. salm-
onicida LFI1238 was extracted using the Wizard Genomic
DNA Purification kit from Promega (Madison, WI, USA)
according to the manufacturer’s protocol for Gram-nega-
tive bacteria. The expression vector pBAD ⁄ gIII and chem-
ically competent E. coli TOP10 cells were purchased from
Invitrogen (Carlsbad, CA, USA). Oligonucleotide primers
(Table 3) were purchased from Invitrogen and Sigma-
Aldrich Co. (St Louis, MO, USA). Phusion DNA polym-
erase from Finnzymes (Espoo, Finland) and Vent and Taq
polymerase from Promega were used in the PCRs. Restric-
tion enzymes NcoI and SalI were purchased from New
England Biolabs (Ipswich, MA, USA), and T4 DNA ligase
was purchased from Sigma-Aldrich. DNaseAlert
TM

and
RNaseAlert
TM
QC System kit was purchased from Ambion
Inc. (Austin, TX, USA) and Integrated DNA Technologies
(Coralville, IA, USA).
Construction of the expression plasmids
The nucleotide sequences of VsEndA and VcEndA have the
GenBank accession nos. DQ263597 and DQ263605,
respectively.
To facilitate cloning of the VsEndA gene into the pBAD ⁄
gIII b vector, a restriction site for SalI was first removed by
point mutation using the overlap extension procedure [30].
PCR was conducted using primers 3 +4 and 1 +4 (Table 3),
with genomic DNA from V. salmonicida as a template. In a
0.2-mL PCR tube, a total of 50 lL reaction mix containing
37.5 lL water, 5 lL10· ThermoPol reaction buffer, 3 lL
25 mm MgCl
2
,1lL10mm dNTP, 1 lL each primer
(10 lm), 1 lL template, and 1 U Vent polymerase was sub-
jected to PCR using a DNA Engine (PTC-200) Peltier Ther-
mal Cycler from Bio-Rad (Hercules, CA, USA). Thermal
cycling conditions were 3 min at 94 °C followed by 30 cycles
of 30 s at 94 °C, 30 s at 50 °C and 90 s at 72 °C. The pro-
gram was ended by an extension step at 72 °C for 5 min, and
then cooled to 4 °C. This PCR yielded one 656-bp and a 254-
bp product when run on a 1% agarose gel. The 656-bp frag-
ment was used as a template in a second PCR conducted
under the same conditions as above, but with primers 3 +2.

This PCR yielded a product of 423 bp. Purified 254-bp and
423-bp fragments were then used as a template in a third
PCR using primers 3 +4. Thermal cycle conditions were the
same as above except for an annealing temperature of 55 °C
and use of 1 U Taq polymerase instead of Vent polymerase.
The two primers 3 +4 contain restriction sites for SalI and
NcoI, respectively, and the primers were created so that the
gene would be amplified without the native N-terminal peri-
plasmic signal. Instead, the periplasmic signal incorporated
into the pBAD ⁄ gIII b vector would be used to transport the
recombinant enzyme into the periplasmic space. The final
PCR product was analyzed on an agarose gel and purified
using the Qiaquick gel extraction kit from Qiagen (Hilden,
Table 3. List of PCR primers. Restriction sites are underlined.
No Sequence
1 GCTTTTAAAGTTGACTTCAAAG
2 CTTTGAAGTCAACTTTAAAAGC
3 CTA
CCATGGCACCTCCTTCTTCTTTCTCAA
4 GCT
GTCGACTTATTTAGTGCATGCTTTATAAACAA
5 CTA
CCATGGCCCCCATCTCTTTTAGTCAT
6 GCT
GTCGACTCAGTTCGGGCATTGCTCAC
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 259
Germany). The DNA fragment and the pBAD ⁄ gIII b vector
were then digested with SalI and NcoI. The insert and vector
were purified from a 1% agarose gel using the Qiaquick gel

extraction kit. Vector and insert were ligated overnight at
16 °C using T4 DNA ligase before transformation into
E. coli Top10 cells using the heat shock method. Positive col-
onies resistant to ampicillin were selected and used for
expression. The VcEndA was cloned using the same proce-
dure as for VsEndA, but no mutation was necessary. The clo-
ning primers for VcEndA are listed in Table 3. The plasmids
were thereafter sequenced using the PE Biosystems BigDye
Terminator Cycle Sequencing kit, ABI 377 Genetic Analyzer
and ABI Sequence Analysis software according to the proto-
col supplied by Applied Biosystems (Foster City, CA, USA).
Enzyme expression and purification
A Chemap CF 3000 fermentor (Chemap AG,
1
Volketswil,
Switzerland) was used for production of the recombinant
nucleases. First 7 L 2 · Luria–Bertani medium supplemen-
ted with 60 mL 20% glucose was inoculated with a 200-mL
overnight preculture and grown at 22 °C. The enzyme pro-
duction was induced by adding 50 mL 14.5% l-arabinose
when the glucose was depleted. The pH was held constant
at 7.4 by addition of 1 m NaOH or 2 m H
2
SO
4
. Oxygen
levels were automatically adjusted by increasing agitation
speed when the level went below 20% of maximum. The
cells were harvested 7 h after induction by centrifugation at
4225 g for 15 min at 4 °C. The cells were subjected to a

combined lysozyme ⁄ osmotic shock treatment [31] to separ-
ate the periplasmic fraction containing the recombinant
protein. Harvested cells were resuspended in 800 mL of a
fractionation buffer containing 20% sucrose, 1 mm EDTA
and 100 mm Tris ⁄ HCl, pH 7.4. Lysozyme (Sigma) was
added to a final concentration of 500 lgÆmL
)1
, and the cell
suspension was incubated for % 20 min at room tempera-
ture. After centrifugation at 8281 g for 20 min, the superna-
tant was collected as the periplasmic fraction and frozen at
)80 °C. The thawed periplasmic fraction was centrifuged at
13 180 g for 20 min before application on a SP Sepharose
FF column (2.6 ⁄ 10 cm; Amersham Pharmacia Biotech,
Uppsala,
2
Sweden) pre-equilibrated with 100 mL buffer A
(20 mm Tris ⁄ HCl, 5 mm MgCl
2
pH 8.3). The enzyme was
eluted using a linear gradient from 0 to 100% buffer B
(buffer A + 1 m NaCl). Fractions containing nuclease
activity were pooled and concentrated using Centriprep
Centrifugal Filter Units (molecular mass cut off, 10 kDa)
from Millipore at 3000 g at 4 °C.
Enzyme analysis
The enzyme purity was analyzed by applying 5 lg protein
to a 4–12% NuPAGE Novex Bis-Tris SDS ⁄ PAGE gel (In-
vitrogen). The gel was stained with Simply Blue Safe Stain
(Invitrogen) according to the manufacturer’s protocol. The

protein concentration was determined using Bio-Rad Pro-
tein Assay based on the method of Bradford [32] and
according to the microtiter plate protocol described by the
manufacturer using BSA as standard. N-Terminal signal
sequence cleavage sites were predicted using the SignalP
server [33]. Sequence alignment was performed using
BioEdit [34], and the alignment was visualized using the
ESPript server [35]. Theoretical isoelectric point, molecular
mass and sequence composition were calculated using the
protparam web-tool at ExPASy [36].
Enzyme assay
The DNaseAlert
TM
QC System kit was used in the deter-
mination of kinetic constants, pH optimum and optimum
NaCl concentration of the two enzymes. The DNase-
Alert
TM
substrate is a synthetic DNA oligonucleotide that
has a HEX
TM
reporter dye (hexachlorofluorescein) on one
end and a dark quencher on the other end. In all reactions,
except for the kinetic measurements, 200 nm substrate was
used. The reaction volumes were adjusted to 90 lL with
nuclease-free water. Reactions were started by pipetting
10 lL of the diluted enzyme solution into eight wells with a
multichannel pipette to a total reaction volume of 100 lL.
Non-binding 1.5-mL tubes from Eppendorf (Hamburg,
Germany) were used for enzyme dilution. New dilutions

were made before each measurement because of the sticky
nature of the enzymes. Black 96-well, low-protein-binding
trays from Corning (Corning, NY, USA) were used in com-
bination with a Spectramax Gemini fluorimeter from
Molecular Devices (Sunnyvale, CA, USA) to detect the
emitted fluorescence. The wavelengths for excitation ⁄ emis-
sion were 535 ⁄ 556 nm, respectively. The initial velocity was
calculated from a minimum of three linear readings on the
time versus fluorescence curve using the program softmax
pro (Molecular Devices). The fluorimeter was set to auto-
mix for 1 s before the first read. A minimum of two parallel
readings were determined under each condition at 23 °C.
[NaCl] optimum, pH optimum, temperature
optimum
The optimum concentration of NaCl was measured in
75 mm Tris buffer with various concentrations of NaCl (0–
750 mm). The pH optimum was measured in 75 mm dietha-
nolamine ⁄ HCl, pH 8–10, and 75 mm Tris ⁄ HCl, pH 7–9. In
addition, 175 and 425 mm NaCl were added to the solution
when VcEndA and VsEndA, respectively, were assayed.
A modified Kunitz DNase assay was used for measuring
optimum endonuclease activity of the two enzymes at dif-
ferent temperatures. In all reactions, 200 lg calf thymus
DNA (Sigma) dissolved in diluted TE buffer (1 mm
Tris ⁄ HCl, pH 8.0, 0.1 mm EDTA) was used as substrate.
Reactions were performed in assay buffers that were opti-
mal for each enzyme [VsEndA, 425 mm NaCl ⁄ 20 mm
Endonuclease I from V. salmonicida and V. cholerae B. Altermark et al.
260 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
Tris ⁄ HCl (pH 8.5) ⁄ 5mm MgCl

2
; VcEndA, 175 mm
NaCl ⁄ 20 mm Tris ⁄ HCl (pH 8.0) ⁄ 5mm MgCl
2
], and the
buffer pH was adjusted at the respective assay tempera-
tures. The total reaction volume was 1 mL. Reaction mix-
tures were preincubated for 5–10 min at the respective
assay temperatures before the addition of enzyme. The
same amount of enzyme (VsEndA 1.5 ng, VcEndA 4.2 ng)
was used at each temperature. Reactions were allowed to
proceed for 20 min and then stopped by adding 0.5 mL ice-
cold 12% perchloric acid. For blank reactions, enzyme was
added after the addition of perchloric acid. Quenched assay
solutions were incubated on ice for 20 min, centrifuged for
10 min at 16 000 g, and the A
260
was measured for the
supernatants in triplicate.
Enzyme kinetic measurements
Fixed amounts of enzyme were incubated at seven different
substrate concentrations ranging from 23 to 1470 nm at 5,
15, 25, 30 and 37 °C in a total reaction volume of 100 lL.
The amounts of VsEndA enzyme used were 0.69, 0.44,
0.21, 0.15 and 0.069 ng at 5, 15, 25, 30 and 37 °C, respect-
ively. For VcEndA the amounts used at these temperatures
were 4.2, 0.21, 0.56, 0.31 and 0.14 ng, respectively. Assay
buffer was optimal for each enzyme [VsEndA, 425 mm
NaCl ⁄ 75 mm diethanolamine (pH 8.5) ⁄ 5mm MgCl
2

;
VcEndA, 175 mm NaCl ⁄ 75 mm diethanolamine (pH 8.0) ⁄
5mm MgCl
2
]. The buffer pH was adjusted at the respective
assay temperatures. The initial velocities were recorded and
the program sigma plot (Systat
3
Software, Inc., San Jose,
CA, USA) was used for estimation of the V
max
and K
m
for
each enzyme by fitting the velocity data to the Michaelis–
Menten equation using nonlinear regression. All measure-
ments were performed in triplicate for each substrate
concentration. The k
cat
values were calculated using the for-
mula V
max
⁄ [enzyme]. The amount of fluorescence emitted
per nmol substrate was calculated from a standard curve
obtained by measuring the maximum fluorescence emitted
as a function of various substrate concentrations. By using
this linear standard curve (slope, 0.88; intercept, 21.8),
values of V
max
were converted from relative fluorescence

unitsÆs
)1
to nmolÆs
)1
. The calculated molecular masses
for VsEndA and VcEndA were 25005.41 gÆmol
)1
and
24731.72 gÆmol
)1
, respectively.
Thermodynamic activation parameters were calculated as
described by Lonhienne et al. [37]. Activation energy, E
a
,
was extracted from the slope of the linear regression curve
obtained from an Arrhenius plot of 1 ⁄ T versus lnk
cat
.
Stability measurements
DSC measurements were performed using the Nano-Differ-
ential Scanning Calorimeter III, model CSC6300 (Calori-
metry Sciences Corp., Lindon, UT, USA). The IUPAC
(International Union of Pure and Applied Chemistry)
recommendations for DSC measurements and analysis [38]
were used as a guideline. The scan rate was set to
1 °CÆmin
)1
, and the scans were performed from 25 to 85 °C
at a constant pressure of 304 kPa. All samples were dia-

lyzed overnight against 50 mm Hepes, pH 8.0, containing
5mm MgCl
2
and 175 mm NaCl or 425 mm NaCl at 4 °C.
The dialysates were used in the reference cell and for buffer
baseline determination. The thermograms obtained were
analyzed using the computer program cpcalc (Calorimetry
Sciences Corp.), and the T
m
(temperature corresponding to
the maximum of the peak) was extracted. The exact protein
concentrations (typically between 0.5 and 1 mgÆmL
)1
) were
measured before DSC analysis. Reversibility of unfolding
was checked by rapid cooling to 4 °C, waiting for 1 h, fol-
lowed by a second scan. The molecular masses used to con-
vert the DSC data to molar heat capacity are as described
above. Kinetic stability was determined by incubating equal
amounts of enzyme (dissolved in optimal buffer for activity
as described above) in a PCR machine heated to 70 °C.
Samples were removed after 10 min and incubated for 1 h
on ice before being assayed using the DNaseAlert QC Sys-
tem kit. Samples incubated for 1 h on ice only served as
the 100% activity reference.
Measurement of substrate specificity
Enzyme specificity towards dsDNA, ssDNA and plasmid
were analyzed using linearized pBAD ⁄ gIII plasmid, linea-
rized and denatured plasmid and intact plasmid. The
pBAD ⁄ gIII plasmid was linearized using SalI and dena-

tured by incubation at 98 °C in a PCR machine for 3 min,
and then kept on ice. Approximately 300 ng of the various
substrates was mixed with 30 ng enzyme in a total volume
of 20 lL containing 1 mm MgCl
2
and 75 mm diethanolam-
ine buffer with optimal [NaCl] and pH for each enzyme.
After 5 min of incubation at 23 °C, the reaction was
stopped by the addition of 5 lL 0.5 m EDTA. The samples
were analyzed on a 1% agarose gel for 1 h at 90 V and
visualized by ethidium bromide staining. The substrates
were also incubated without enzyme as a reference.
Activity towards RNA was measured using the RNase-
Alert QC System kit with the same instrumental set up as
for the DNaseAlert system mentioned above, except that
the wavelengths used for excitation ⁄ emission were
490 ⁄ 520 nm, respectively. Measurements were taken every
64 s for 20 min. The effect of [NaCl] on the RNase activity
was measured in 75 mm diethanolamine ⁄ HCl at pH 8.5 for
VsEndA and pH 8.0 for VcEndA with increasing concen-
trations of NaCl (0–425 mm for VsEndA, 0–175 mm for
VcEndA) including 5 mm MgCl
2
per 100 lL reaction mix-
ture. The maximum fluorescence obtained with 200 nm
RNaseAlert and DNaseAlert was measured by adding 5 lL
RNase A (0.01 UÆmL
)1
) to wells with RNaseAlert substrate
after the initial measurements and 2 lL undiluted VcEndA

to wells with DNaseAlert substrate. The initial velocities
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 261
(relative fluorescence unitsÆs
)1
) obtained using the RNase-
Alert substrate at various concentrations of NaCl were then
corrected with respect to the values obtained using an equal
amount of DNaseAlert substrate. Equal concentrations of
enzyme were used in the DNaseAlert and RNaseAlert
assays.
Acknowledgements
The present study was supported by the National Pro-
gram for Research in Functional Genomics in Norway
(FUGE), the Research Council of Norway, and by
Biotec Pharmacon ASA.
References
1 Borroto RJ (1997) Ecology of Vibrio cholerae serogroup
01 in aquatic environments. Rev Panam Salud Publica 1,
3–8.
2 Colwell RR (2004) Infectious disease and environment:
cholera as a paradigm for waterborne disease. Int
Microbiol 7, 285–289.
3 Colquhoun DJ, Alvheim K, Dommarsnes K, Syvertsen
C & Sorum H (2002) Relevance of incubation tempera-
ture for Vibrio salmonicida vaccine production. J Appl
Microbiol 92, 1087–1096.
4 Hochachka PW & Somero GN (1984) Biochemical
Adaptation. Princeton University Press, Princeton, NJ.
5 Wu SI, Lo SK, Shao CP, Tsai HW & Hor LI (2001)

Cloning and characterization of a periplasmic nuclease
of Vibrio vulnificus and its role in preventing uptake of
foreign DNA. Appl Environ Microbiol 67, 82–88.
6 Li CL, Hor LI, Chang ZF, Tsai LC, Yang WZ & Yuan
HS (2003) DNA binding and cleavage by the periplas-
mic nuclease Vvn: a novel structure with a known active
site. EMBO J 22 , 4014–4025.
7 Altermark B, Smalas AO, Willassen NP & Helland R
(2006) The structure of Vibrio cholerae extracellular
endonuclease I reveals the presence of a buried
chloride ion. Acta Crystallogr D Biol Crystallogr 62,
1387–1391.
8 Chang MC, Chang SY, Chen SL & Chuang SM (1992)
Cloning and expression in Escherichia coli of the gene
encoding an extracellular deoxyribonuclease (DNase)
from Aeromonas hydrophila. Gene 122, 175–180.
9 Focareta T & Manning PA (1987) Extracellular proteins
of Vibrio cholerae: molecular cloning, nucleotide
sequence and characterization of the deoxyribonuclease
(DNase) together with its periplasmic localization in
Escherichia coli K-12. Gene 53, 31–40.
10 Lehman IR, Roussos GG & Pratt EA (1962) The
deoxyribonucleases of Escherichia coli. II. Purification
and properties of a ribonucleic acid-inhibitable endonu-
clease. J Biol Chem 237, 819–828.
11 MacLellan SR & Forsberg CW (2001) Properties of the
major non-specific endonuclease from the strict anae-
robe Fibrobacter succinogenes and evidence for disulfide
bond formation in vivo. Microbiology 147, 315–323.
12 Nakasone K, Kato C & Horikoshi K (2002) Cloning

and sequencing of the endA gene encoding
deoxyribonuclease from a deep-sea piezophilic bacter-
ium, Shewanella violacea strain DSS12. Reports of Japan
Marine Science and Technology Centre
4
45, 63–68.
13 Salikhova ZZ, Sokolova RB, Ponomareva AZ &
Iusupova DV (2001) Endonuclease from Proteus
mirabilis.
Prikl Biokhim Mikrobiol 37, 43–47.
14 Hanahan D (1983) Studies on transformation of
Escherichia coli with plasmids. J Mol Biol 166, 557–580.
15 Focareta T & Manning PA (1991) Distinguishing
between the extracellular DNases of Vibrio cholerae and
development of a transformation system. Mol Microbiol
5, 2547–2555.
16 Kawagishi I, Okunishi I, Homma M & Imae Y (1994)
Removal of the periplasmic DNase before electropora-
tion enhances efficiency of transformation in the marine
bacterium Vibrio alginolyticus. Microbiology 140, 2355–
2361.
17 Moulard M, Condemine G & Robert-Baudouy J (1993)
Search for the function of the nuclease NucM of Erwi-
nia chrysanthemi. FEMS Microbiol Lett 112, 99–103.
18 Moulard M, Condemine G & Robert-Baudouy J (1993)
Characterization of the nucM gene coding for a nucle-
ase of the phytopathogenic bacteria Erwinia chry-
santhemi. Mol Microbiol 8, 685–695.
19 Sheridan PP, Panasik N, Coombs JM & Brenchley JE
(2000) Approaches for deciphering the structural basis

of low temperature enzyme activity. Biochim Biophys
Acta 1543, 417–433.
20 Turekian KK (1968) Oceans. Prentice Hall, Englewood
Cliffs, NJ.
21 Moulard M, Condemine G, Nasser W & Robert-Bau-
douy J (1995) Purification and characterization of the
nuclease NucM of Erwinia chrysanthemi. Biochim Bio-
phys Acta 1262, 133–138.
22 Siddiqui KS & Cavicchioli R (2006) Cold-adapted
enzymes. Annu Rev Biochem 75, 403–433.
23 Mevarech M, Frolow F & Gloss LM (2000) Halophilic
enzymes: proteins with a grain of salt. Biophys Chem
86, 155–164.
24 Oren A, Larimer F, Richardson P, Lapidus A &
Csonka LN (2005) How to be moderately halophilic
with broad salt tolerance: clues from the genome of
Chromohalobacter salexigens. Extremophiles 9, 275–279.
25 Feller G & Gerday C (1997) Psychrophilic enzymes:
molecular basis of cold adaptation. Cell Mol Life Sci
53, 830–841.
26 Cornish-Bowden A (2002) Enthalpy-entropy compensa-
tion: a phantom phenomenon. J Biosci 27, 121–126.
Endonuclease I from V. salmonicida and V. cholerae B. Altermark et al.
262 FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS
27 Bae E & Phillips GN Jr (2004) Structures and analysis
of highly homologous psychrophilic, mesophilic, and
thermophilic adenylate kinases. J Biol Chem 279,
28202–28208.
28 Georlette D, Damien B, Blaise V, Depiereux E, Uversky
VN, Gerday C & Feller G (2003) Structural and func-

tional adaptations to extreme temperatures in psychro-
philic, mesophilic, and thermophilic DNA ligases. J Biol
Chem 278, 37015–37023.
29 Wintrode PL, Miyazaki K & Arnold FH (2000) Cold
adaptation of a mesophilic subtilisin-like protease by
laboratory evolution. J Biol Chem 275, 31635–31640.
30 Ho SN, Hunt HD, Horton RM, Pullen JK & Pease LR
(1989) Site-directed mutagenesis by overlap extension
using the polymerase chain reaction. Gene 77, 51–59.
31 French C, KeshavarzMoore E & Ward JM (1996)
Development of a simple method for the recovery of
recombinant proteins from the Escherichia coli peri-
plasm. Enzyme Microb Technol 19, 332–338.
32 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein uti-
lizing the principle of protein-dye binding. Anal Biochem
72, 248–254.
33 Bendtsen JD, Nielsen H, von Heijne G & Brunak S
(2004) Improved prediction of signal peptides: SignalP
3.0. J Mol Biol 340, 783–795.
34 Hall TA (1999) BioEdit: A user-friendly biological
sequence alignment editor and analysis program
for windows 95 ⁄ 98 ⁄ NT. Nucleic Acids Symp Ser 41,
95–98.
35 Gouet P, Courcelle E, Stuart DI & Metoz F (1999)
ESPript: analysis of multiple sequence alignments in
PostScript. Bioinformatics 15, 305–308.
36 Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel
RD & Bairoch A (2003) ExPASy: the proteomics server
for in-depth protein knowledge and analysis. Nucleic

Acids Res 31, 3784–3788.
37 Lonhienne T, Gerday C & Feller G (2000) Psychrophilic
enzymes: revisiting the thermodynamic parameters of
activation may explain local flexibility. Biochim Biophys
Acta 1543, 1–10.
38 Hinz HJ & Schwarz FP (2001) Measurement and analy-
sis of results obtained on biological substances with
differential scanning calorimetry. Pure Appl Chem 73,
745–759.
B. Altermark et al. Endonuclease I from V. salmonicida and V. cholerae
FEBS Journal 274 (2007) 252–263 ª 2006 The Authors Journal compilation ª 2006 FEBS 263