Properties of tryptophan indole-lyase from a piezophilic bacterium, Photobacterium profundum SS9

Article abstract of DOI:10.1016/j.abb.2010.11.002

Tryptophan indole-lyase (Trpase), PBPRA2532, from Photobacterium profundum SS9, a piezophilic marine bacterium, has been cloned, expressed in Escherichia coli, and purified. The P. profundum Trpase (PpTrpase) exhibits similar substrate specificity as the enzyme from E. coli (EcTrpase). PpTrpase has an optimum temperature for activity at about 30 °C, compared with 53 °C for EcTrpase, and loses activity rapidly (t_1/2 ～ 30 min) when incubated at 50 °C, while EcTrpase is stable up to 65 °C. PpTrpase retains complete activity when incubated more than 3 h at 0 °C, while EcTrpase has only about 20% remaining activity. Under hydrostatic pressure, PpTrpase remains fully active up to 100 MPa (986 atm), while EcTrpase exhibits only about 10% activity at 100 MPa. PpTrpase forms external aldimine and quinonoid intermediates in stopped-flow experiments with l-Trp, S-Et-l-Cys, S-benzyl-l-Cys, oxindolyl-l-Ala, l-Ala and l-Met, similar to EcTrpase. However, with l-Trp a gem-diamine is observed that decays to a quinonoid complex. An aminoacrylate is observed with l-Trp in the presence of benzimidazole, as was seen previously with EcTrpase [28] but not with S-Et-l-Cys. The results show that PpTrpase is adapted for optimal activity in the low temperature, high pressure marine environment.

Full text of DOI:10.1016/j.abb.2010.11.002

Archives of Biochemistry and Biophysics 506 (2011) 35–41
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
Properties of tryptophan indole-lyase from a piezophilic bacterium, Photobacterium
profundum SS9
b
a
b
c
Robert S. Phillips a,b, , Rashin Ghaffari , Peter Dinh , Santiago Lima , Douglas Bartlett
⇑
a Department of Chemistry, University of Georgia, Athens, GA 30602, USA
b Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA 30602, USA
c Scripps Institute of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 August 2010
and in revised form 8 November 2010
Available online 21 November 2010
Tryptophan indole-lyase (Trpase), PBPRA2532, from Photobacterium profundum SS9, a piezophilic marine
bacterium, has been cloned, expressed in Escherichia coli, and purified. The P. profundum Trpase
(PpTrpase) exhibits similar substrate specificity as the enzyme from E. coli (EcTrpase). PpTrpase has an
optimum temperature for activity at about 30 °C, compared with 53 °C for EcTrpase, and loses activity
rapidly (t1/2 ꢀ 30 min) when incubated at 50 °C, while EcTrpase is stable up to 65 °C. PpTrpase retains
complete activity when incubated more than 3 h at 0 °C, while EcTrpase has only about 20% remaining
activity. Under hydrostatic pressure, PpTrpase remains fully active up to 100 MPa (986 atm), while
EcTrpase exhibits only about 10% activity at 100 MPa. PpTrpase forms external aldimine and quinonoid
Keywords:
Pyridoxal-50-phosphate
Piezophilic
Hydrostatic pressure
Tryptophan
Reaction specificity
Reaction intermediate
intermediates in stopped-flow experiments with
and -Met, similar to EcTrpase. However, with -Trp a gem-diamine is observed that decays to a quinonoid
complex. An aminoacrylate is observed with -Trp in the presence of benzimidazole, as was seen previ-
ously with EcTrpase [28] but not with S-Et- -Cys. The results show that PpTrpase is adapted for optimal
activity in the low temperature, high pressure marine environment.
Ó 2010 Elsevier Inc. All rights reserved.
L-Trp, S-Et-L-Cys, S-benzyl-L-Cys, oxindolyl-L-Ala, L-Ala
L
L
L
L
Introduction
a wide variety of physiological processes, including biofilm forma-
tion [3,4] and plasmid replication [5]. Thus, the expression of Trpase
is stringently regulated in E. coli, with induction by high concentra-
Tryptophan indole-lyase (Trpase1, tryptophanase, EC 1.4.99.1) is
a pyridoxal-50-phosphate (PLP) dependent enzyme found in a wide
range of bacteria, especially in the enterobacteria, such as Escherichia
coli, Vibrio cholerae, and Proteus vulgaris [1]. Trpase catalyzes the
reversible hydrolytic carbon–carbon bond cleavage reaction
of L-Trp to give indole and ammonium pyruvate (Eq. (1)). Indole
tions of L-Trp [6], but repression in the presence of glucose [7].
Recently, the genome of a piezophilic and psychrophilic marine bac-
terium, Photobacterium profundum SS9, was sequenced, and it was
found to contain 3 different copies of tnaA, the gene coding for
Trpase [8], although E. coli only contains a single copy of tnaA. P.
profundum SS9 has an optimal growth temperature of 15 °C and
pressure of 28 MPa, and can grow at pressures up to 90 MPa [8].
A number of genes were found to be selectively expressed in
P. profundum SS9 grown at 28 MPa compared to growth at
0.1 MPa (0.986 atm), including one of the tnaAs, PBPRA2532 [8].
In this work, we have cloned, expressed, and purified the
PBPRA2532 Trpase from P. profundum (PpTrpase), and we have
compared its properties with the E. coli enzyme (EcTrpase).
ð1Þ
has long been considered as simply a waste product of bacteria
metabolism of proteins, but recent work has found that indole is
a chemical signal for bacteria, regulating gene expression [2] and
Experimental methods
Materials
⇑
Corresponding author. Address: Department of Chemistry, University of Geor-
gia, Athens, GA 30602, USA. Fax: +1 706 542 9454.
S-(o-Nitrophenyl)-
described [9]. Lactate dehydrogenase (LDH) from rabbit muscle,
-tryptophan, S-methyl- -cysteine, S-ethyl- -cysteine, S-benzyl-
cysteine, PLP, -methionine and NADH were purchased from
L-cysteine (SOPC) was prepared as previously
E-mail address: rsphillips@chem.uga.edu (R.S. Phillips).
1
Abbreviations used: PLP, pyridoxal-50-phosphate; Trpase, tryptophan indole-lyase
L
L
L
L-
(tryptophanase) [EC 4.1.99.1]; EcTrpase, Trpase from Escherichia coli; PpTrpase, Trpase
from Photobacterium profundum SS9 PBPRA2532; SOPC, S-(o-nitrophenyl-L-cysteine).
L
0003-9861/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.abb.2010.11.002

36
R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
United States Biochemical Corp. (USB). b-Cl-
L
-Alanine hydrochlo-
desired temperature, and the immediate decrease in absorbance
at 370 nm was measured. For the temperature stability measure-
ments, the enzymes were incubated at 0 °C on ice, or at 50 °C in
a water bath, and aliquots were removed at various times and
assayed with SOPC at 25 °C.
ride was prepared from -serine as described [10].
L
a
-[2H]-
L-Trp
was prepared as previously described [11].
Cloning and expression of P. profundum Trpase
Forward and reverse primers based on the DNA sequence of
P. profundum SS9 tnaA (PBPRA2532) were designed, and PCR was
performed with 1 ng of genomic DNA. The primers used were
50-CACCATGGAAAACTTTAAACACTTAC-30 for the forward primer
and 50-GATCGTGATTAAGCTTCTAGAATC-30 for the reverse primer.
The PCR product was cloned with pET-100/D-TOPO (Invitrogen).
Plasmids were isolated and sequenced to confirm the presence
and correct sequence of the P. profundum PBPRA2532 gene. The
plasmid pET100-PBPRA2532 was then used to transform E. coli
BL21(DE3) tn5:tnaA [12] for expression of PpTrpase without
contamination by EcTrpase.
Stopped-flow reactions
Stopped-flow experiments were carried out at ambient temper-
ature (ca. 22–23 °C) using an RSM-1000 instrument from OLIS, Inc.
(Bogart, Georgia, USA), equipped with a stopped-flow mixing
chamber, as described previously [17,18]. The stopped-flow mixer
has a 10 mm path length and a dead time of less than 2 ms. Absor-
bance spectra were collected over the wavelength range from 240
to 800 nm at a rate of 1000 scans sꢁ1. Enzymes for stopped-flow
measurements were passed through a PD-10 gel filtration column
equilibrated with 0.05 M potassium phosphate, pH 8.0, to remove
excess PLP, immediately prior to use.
An overnight culture of E. coli BL21(DE3) tn5:tnaA pET100-
PBPRA2532 was grown in 5 mL of LB medium containing 35
mL of kanamycin sulfate and 100 g/mL of ampicillin. The over-
night culture was used to inoculate 1 L of LB medium containing
35 g/mL of kanamycin sulfate and 100 g/mL of ampicillin, and
the culture was grown with shaking at 37 °C in a water bath until
the OD600 was 0.6–0.8. The temperature of the bath was then rap-
idly reduced to 20 °C by addition of ice, 0.1 mM IPTG was added to
the culture, and the cell suspension was shaken at 20 °C for 18 h.
The cells were then collected by centrifugation for 20 min at
10,000g, suspended in 30 mL 0.1 M potassium phosphate, pH 7.0,
lg/
l
Enzyme activity and absorption spectra under hydrostatic pressure
l
l
The effects of hydrostatic pressure on the rates and absorption
spectra were measured using a Cary 14 UV/Vis spectrophotometer
modified by OLIS, Inc. to contain a high pressure cell from ISS
(Champaign, Illinois, USA), equipped with a manual pressure pump
from High Pressure Equipment Co., using spectroscopic grade eth-
anol as the pressurizing fluid. The cell was maintained at 25 °C
with an external circulating water bath. The enzyme solutions
were contained in 1 mL quartz bottles with a 9 mm pathlength,
capped with Teflon tubing. A buffer blank at 1 bar was used to ob-
tain a baseline reading. The buffer, triethanolamine hydrochloride,
1 mM EDTA, 5 mM 2-mercaptoethanol, 50 lM PLP, and frozen at
ꢁ78 °C until used for enzyme purification.
Enzyme purification
pH 8.0, was chosen since the pKa is 7.88 and the
DV0 for ionization
is 4.5 0.3 mL/mol [19], so the pressure change will not signifi-
cantly change the pH. The pressure-dependent activity was mea-
sured at 25 °C in 0.05 M triethanolamine hydrochloride, pH 8.0,
EcTrpase was purified by hydrophobic chromatography on a col-
umn of CL-Sepharose 4B, as described previously [13], except that a
gradient of (NH4)2SO4 from 40% to 20% saturation was used for elu-
tion, rather than a stepwise elution. For PpTrpase, a column of Phe-
nyl-Sepharose CL-4B was used, since in contrast to EcTrpase, the
enzyme was not retained on a column of Sepharose CL-4B in 40%
saturated (NH4)2SO4. The crude extract and protamine treatment
was performed as described previously [13], and then the solution
was brought to 20% saturation by addition of solid ammonium sul-
fate. After loading, the column was washed with 0.1 M potassium
containing 100 mM
synthesis of tryptophan at 290 nm (
L
-serine and 250
lM indole, measuring the
e
= 1800 Mꢁ1 cmꢁ1) [20], in or-
der to avoid potential problems with the lactate dehydrogenase
coupling reaction under pressure.
Data analysis
The rapid-scanning stopped-flow data were analyzed by global
analysis of all spectra at all wavelengths using the Globalworks
program provided by OLIS [21]. The spectra were fitted to the min-
imum number of species and exponential processes to adequately
describe the data based on residuals and standard deviation, using
Eq. (2), where At is the absorbance at a wavelength at time t, Ai is
the absorbance at that wavelength for each phase, ki is the rate
constant for each phase, and A0 is the baseline absorbance at that
wavelength of the reaction mixture. The effects of hydrostatic
X
phosphate, pH 7.0, 1 mM EDTA, 5 mM 2-mercaptoethanol, 50 lM
PLP, containing 20% ammonium sulfate until the absorbance re-
turned to baseline, then it was eluted with a gradient of decreasing
ammonium sulfate. Peak fractions were pooled and frozen at
ꢁ78 °C. Protein was determined by the method of Bradford [14],
with purified wild-type EcTrpase as a standard. Enzyme activity
during purification was routinely measured with 0.6 mM
S-(o-nitrophenyl)-
phate, pH 8.0, 50
absorbance at 370 nm (
L-cysteine (SOPC) in 50 mM potassium phos-
l
M PLP, at 25 °C [15], following the decrease in
D
e
= ꢁ1.86 ꢂ 103 Mꢁ1 cmꢁ1) using a Cary
At ¼
Ai ꢃ expðꢁki ꢃ tÞ þ A0
ð2Þ
1E UV/vis spectrophotometer equipped with a 6 ꢂ 6 Peltier temper-
ature controlled cell compartment. Enzyme activity with
S-alkyl- -cysteines, and b-chloro- -Ala were performed in the same
buffer using the lactate dehydrogenase coupled assay [16], with
L-Trp,
pressure on the absorption spectra were analyzed by global analysis
using the Globalworks program, as previously described [17,18].
The absorption spectra were corrected for solvent compressibility
using a modified Tait equation [22]. The pressure dependent
changes at a single wavelength were fitted to a Boltzmann function,
L
L
0.1 mM NADH and 20
the absorbance decrease at 340 nm (
lg/mL lactate dehydrogenase, measuring
D
e
= ꢁ6.22 ꢂ 103 Mꢁ1 cmꢁ1).
in Eq. (3), where Ap is the observed absorbance at pressure P,
the pressure-dependent absorbance change, Keq is the pressure
independent value of the equilibrium constant, V is the reaction
DA is
Temperature dependence of Trpase reactions
D
The temperature dependence of the reactions of PpTrpase and
EcTrpase was performed with SOPC in 50 mM KPi, pH 8.0, 50 lM
PLP. For the temperature dependence of the activity, the enzyme
was added last to the reaction mixtures pre-equilibrated at the
volume change, and A is the absorbance background at infinite
1
pressure.
Ap ¼
D
AðKeq ꢃ expðꢁP ꢃ
D
VÞÞ=ð1 þ Keq ꢃ expðꢁP ꢃ
D
VÞÞ þ A
ð3Þ
1

R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
37
Results
Expression and purification of P. profundum Trpase
It was found that induction of the P. profundum tnaA
PBPRA2532 gene at 37 °C did not result in significant activity in cell
extracts, but large amounts of insoluble inclusion bodies were seen
in the pellets after cell disruption. However, induction with IPTG
followed by incubation overnight at 20 °C provided very high
expression of soluble, active Trpase. The expression was performed
in E. coli BL21(DE3)tn5:tnaA with a transposon in the genomic tnaA
to avoid contamination by the E. coli enzyme. In contrast to the
E. coli enzyme, PpTrpase did not adsorb to a column of Sepharose
CL-4B in 40% saturated ammonium sulfate [13], so the purification
was performed on a Phenyl-Sepharose CL-4B column, with adsorp-
tion in 20% ammonium sulfate saturated buffer and elution with a
linear gradient of buffer containing no ammonium sulfate. The
purified Trpase was >95% pure by PAGE and exhibited a specific
activity of 10 U/mg with SOPC at 25 °C, compared to the specific
activity of EcTrpase, about 34 U/mg [13].
Fig. 1. Effect of temperature on activity of Trpases from E. coli and P. profundum.
Open circles: EcTrpase; closed circles, PpTrpase.
Substrate specificity of P. profundum Trpase
PpTrpase has high activity for b-elimination of
L-Trp, SOPC, and
b-Cl- -Ala, but values of kcat and kcat/Km are 3- to 5-fold lower than
L
the E. coli enzyme (Table 1). It is particularly interesting that S-al-
kyl-cysteines with small substituents have at least 10-fold lower
values of kcat for PpTrpase (Table 1) than for EcTrpase [11,13,23] .
Effect of temperature on P. profundum Trpase activity
The optimal temperature for PpTrpase activity with SOPC is
30 °C, compared with an optimum greater than 50 °C with EcTr-
pase (Fig. 1). Incubation of PpTrpase at 50 °C result in rapid loss
of activity, while EcTrpase is considerably more stable on incuba-
tion (Fig. 2). In contrast, when EcTrpase is incubated at 0 °C, it loses
activity rapidly, as seen in Fig. 3 [24,25]. However, PpTrpase shows
little change in activity on incubation at 0 °C (Fig. 3).
Fig. 2. Effect of incubation at 50 °C on activity of Trpases from E. coli and P.
profundum. Open circles: EcTrpase; closed circles, PpTrpase.
Effect of pressure on P. profundum Trpase
The effect of increasing hydrostatic pressure on the specific
activity of PpTrpase and EcTrpase was determined (Fig. 4). The
activity of EcTrpase decreases at pressures above 50 MPa, and by
100 MPa is less than 10% of the activity at 1 bar. When the samples
incubated at 100 MPa were decompressed to 1 bar, activity was
immediately recovered to about 75% that of samples measured at
1 bar. In contrast, PpTrpase has greater than 90% activity remaining
Table 1
Substrate specificity of P. profundum and E. coli Trpase.a
Substrate
P. profundum
kcat (sꢁ1 kcat/Km (Mꢁ1 sꢁ1
E. coli
kcat (sꢁ1
6.8b
)
)
)
kcat/Km (Mꢁ1 sꢁ1
)
2.5 0.10 (7.6 0.4) ꢂ 103
13.9 4.2 (2.9 0.3) ꢂ 104
3.0 ꢂ 104
b
L
-Trp
b
SOPC
44b
5.0c
6.0d
5.1 ꢂ 105
c
S-Me-L-Cys 0.2 0.07 60 50
330
d
S-Et-
S-Bzl-
b-Cl- -Ala
L-Cys
0.4 0.05 280 130
1.1 0.16 (1.1 0.7) ꢂ 104
3.0 0.10 770 70
9.0 ꢂ 103
e
L-Cys
5.2 0.4e
(8.1 0.3) ꢂ 104
Fig. 3. Effect of incubation at 0 °C on activity of Trpases from E. coli and P.
profundum. Open circles: EcTrpase; closed circles, PpTrpase.
L
12.4 0.6 (2.4 0.1) ꢂ 103
a
The reactions were performed in 50 mM potassium phosphate, pH 8.0, with
50
l
M PLP, at 25 °C.
From Ref. [13].
From Ref. [11].
From Ref. [12].
From Ref. [23].
b
c
at 100 MPa (Fig. 4). It is interesting that both enzymes show an ini-
tial increase in activity with pressure, reaching a maximum of
about 140% at 30 MPa for EcTrpase and 60 MPa for PpTrpase.
d
e

38
R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
EcTrpase [28–30]. There is an initial rapid increase at about 340 nm
and decrease at 420 nm, together with a small increase at 505 nm,
followed by a decrease at 340 nm as the 420 and 505 nm peaks
increase (Fig. 4B), and finally a slow increase at 505 nm, with
1/
When the experiment was performed with
rate constants were 1/
1 = 95 10 sꢁ1, 1/ 2 = 12.2 1.7 sꢁ1 and
1/
3 = 0.92 0.20 sꢁ1 (data not shown). Thus, there is a small
s
1 = 230 22 sꢁ1, 1/
s
2 = 10.2 1.3 sꢁ1 and 1/
s
3 = 1.36 0.20 sꢁ1
-[2H]-
-Trp, the
.
a
L
s
s
s
kinetic isotope effect of 2.4 on the fast phase of the reaction. The
decay of the 340 peak and concomitant increase at 420 and
505 nm do not show an isotope effect. When 5 mM benzimidazole
is included in the reactions with tryptophan, after an initial
increase at 505 nm, the 420 and 505 nm peaks decay concomitant
with the formation of a strong peak at about 350 nm (Fig. 4C and
D), with 1/s s s3 = 0.69
1 = 260 22 sꢁ1, 1/ 2 = 20.0 2.3 sꢁ1 and 1/
0.06 sꢁ1. This is very similar to what was seen with EcTrpase
[28–30].
When EcTrpase is mixed with substrates and inhibitors such as
-Met, S-Et-L-Cys, S-benzyl-L-Cys, oxindoly-L-Ala or L-Ala in the
Fig. 4. Effect of hydrostatic pressure at 25 °C on activity of Trpases from E. coli
and P. profundum. Open circles: EcTrpase; closed circles, PpTrpase.
L
stopped-flow instrument, strongly absorbing quinonoid intermedi-
ates with kmax about 500 nm are formed [28,31]. We observed sim-
ilar results with PpTrpase. It is particularly interesting that S-Et-L-
Cys rapidly forms a quinonoid intermediate at 508 nm, as can be
seen in Fig. 7, with kobs of 18 sꢁ1, very similar to EcTrpase [28], even
though the kcat value with this substrate is much slower than that
for EcTrpase (Table 1). Furthermore, when the stopped-flow exper-
EcTrpase exhibits absorption peaks at about 340 and 420 nm,
corresponding to two different forms of the PLP cofactor [26,27].
Application of hydrostatic pressure decreases the 420 nm band
and increases the 340 nm band for EcTrpase [25], consistent with
a conformational change coupled with the change in spectrum.
PpTrpase exhibits a very similar absorption spectrum, and similar
effects of increasing pressure (Fig. 5). The absorption changes at
420 nm are fully reversible upon decompression, as can be seen
by comparing the open circles (compression) and the filled circles
(decompression) in the inset Fig. 5. Fitting the spectra in Fig. 5 pro-
iment with S-Et-L-Cys is performed with PpTrpase in the presence
of 5 mM benzimidazole, there is no observed decay of the quino-
noid intermediate to the 350 nm band of the aminoacrylate (data
not shown), while EcTrpase forms an aminoacrylate intermediate
vides
D
V = ꢁ39 7 mL/mol and Keq = 0.106 0.016, compared to
from S-Et-
rapid-scanning stopped-flow spectra for the reactions of
-Ala, S-benzyl- -Cys and oxindolyl- -Ala with PpTrpase are in-
L-Cys in the presence of benzimidazole [28–31]. The
D
V = ꢁ38 3 mL/mol and Keq = 0.65 0.03 for EcTrpase [25]. The
L-Met,
dashed lines are the calculated spectra from the global analysis.
The solid line in the inset is the calculated curve using Eq. (3)
and the parameters above.
L
L
L
cluded in the Supplementary data.
Discussion
Stopped-flow kinetics of P. profundum Trpase
There has been considerable interest in the properties of en-
zymes from organisms adapted to extreme environments in recent
Mixing of PpTrpase with 10 mM L-Trp in the stopped-flow spec-
trophotometer results in formation of intermediates absorbing at
about 340, 420 and 505 nm (Fig. 6A), similar to what is seen with
B
A
0.2
0.1
420 nm
1
340 nm
505 nm
6
0
D
6
C
0.2
340 nm
1
420 nm
505 nm
0.1
0
320 360 400 440 480 520 560
Wavelength (nm)
0.01
0.1
1
Time (seconds)
Fig. 6. Reaction of PpTrpase and
the reaction of 1.5 mg/mL PpTrpase with 10 mM
L
-Trp. (A) Rapid-scanning stopped-flow spectra for
-Trp in 0.02 M KPi, pH 8.0, 0.16 M
L
KCl. Scans are shown at 0 (Curve 1), 0.020 (Curve 2), 0.04 (Curve 3), 0.08 (Curve 4),
0.32 (Curve 5) and 0.64 (Curve 6) seconds after mixing. (B) Time courses for the
reaction in A at 340, 420 and 505 nm. (C) Rapid-scanning stopped-flow spectra for
Fig. 5. Effect of hydrostatic pressure at 25 °C on the absorption spectrum of
PpTrpase. Curve 1, 300 bar; Curve 2, 700 bar; Curve 3, 1000 bar; Curve 4, 1400 bar;
Curve 5, 1800 bar; Curve 6, 2000 bar. The dashed and dotted curves are the fitted
spectra of the two components at 0 and infinite pressure, respectively. Inset:
Absorbance at 420 nm as a function of pressure. Filled circles, compression; open
circles, decompression. The line is the calculated curve using Eq. (3) with the
parameters given in the text.
the reaction of 1.5 mg/mL PpTrpase with 10 mM L-Trp and 5 mM benzimidazole in
0.02 M KPi, pH 8.0, 0.16 M KCl. Scans are shown at 0 (Curve 1), 0.020 (Curve 2), 0.04
(Curve 3), 0.08 (Curve 4), 0.32 (Curve 5) and 0.64 (Curve 6) seconds after mixing.
The dashed line is the spectrum of the enzyme alone. (D) Time courses for the
reaction in C at 340, 420 and 505 nm.

R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
39
partially responsible for the increased cold stability of PpTrpase
shown in Fig. 3.
There are also dramatic differences in the effects of hydrostatic
pressure on the activity of EcTrpase and PpTrpase. PpTrpase re-
tained complete activity up to at least 100 MPa (Fig. 4), well above
the physiological pressures (28–45 MPa) that P. profundum is rou-
tinely exposed to in its marine environment. In contrast, EcTrpase
exhibits less than 10% activity at 100 MPa (Fig. 4). The loss of activ-
ity of EcTrpase under pressure is rapidly reversible, since the activ-
ity is immediately restored when the samples are decompressed.
This suggests that the pressure-dependent reversible loss of activ-
ity is the result of a transition state effect, due to a positive
D
Và,
rather than subunit or cofactor dissociation, since cold-inactivated
EcTrpase recovers activity relatively slowly by a second-order reas-
sociation of PLP with the apoenzyme [24,25]. It is interesting that
PpTrpase has somewhat lower activity than EcTrpase at 25 °C
and 0.1 MPa, but higher activity at pressures above 60 MPa. This
may be due to the structural modifications that are required for
maintenance of activity under low temperature and high pressure
making the enzyme more flexible. The effects of pressure on the
spectra of the resting enzymes from P. profundum and E. coli are
very similar (Fig. 5), changing the ratio of intensity of the 420
and 340 nm bands. This change in ratio is associated with a blue
shift in the 420 nm band (Fig. 5), suggesting a change in the polar-
ity of the cofactor environment that favors the 340 nm form. The
420 nm peak has been assigned to the ketoenamine tautomer of
the internal aldimine [26], while the 340 nm peak has been sug-
gested to be an enolimine tautomer [26] or a nucleophilic adduct
[27] of the internal aldimine. The dotted lines in Fig. 5 are the fitted
spectra from global analysis. Since both the high and low pressure
fitted spectra still contain both peaks, and the volume change is
negative, the pressure dependence of the spectra is likely due to
a conformational change which alters the environment of the PLP
and affects the equilibrium. The ratio of intensity of the 340 and
420 nm peaks is also pH-dependent [13,33,34], with an apparent
pKa of 7.5.
Fig. 7. Reaction of PpTrpase and S-Et-
spectra for the reaction of 1 mg/mL PpTrpase with 20 mM S-Et-
pH 8.0, 0.16 M KCl. Scans are shown at 0.001 (Curve 1), 0.047 (Curve 2), 0.095
(Curve 3), 0.207 (Curve 4), 0.393 (Curve 5) and 0.495 (Curve 6) seconds after mixing.
The dashed line is the spectrum of the enzyme alone. (B) Time courses for the
reaction in A at 340, 420 and 505 nm.
L
-Cys. (A) Rapid-scanning stopped-flow
-Cys in 0.02 M KPi,
L
years. Most of these studies have focused on enzymes from ther-
mophilic microorganisms, adapted to living at high temperatures
in terrestrial hot springs or around marine thermal vents. In the
case of Trpase, the enzymes from a symbiotic thermophile, Symbi-
obacterium thermophilum, have been cloned, expressed and puri-
fied [32]. In contrast, there has been much less work done with
psychrophilic and piezophilic enzymes. We were interested in
the Trpases from P. profundum SS9, since this organism is adapted
to both high pressure (pressure optimum 10–30 MPa, growth up to
90 MPa) and low temperature growth, with an optimum of 15 °C.
In particular, one of the three Trpases in the genome of P. profun-
dum SS9, PBPRA2532, is preferentially expressed when cells are
grown at 28 MPa [8]. Thus, it seemed likely to us that this enzyme
would show an adaptation to the high pressure and low tempera-
ture environment that would be reflected in its physical properties,
so we cloned, expressed and purified it and compared it with the
E. coli enzyme.
Catalytic properties
PpTrpase has similar, but slightly lower, activities for b-elimina-
tion of
L-Trp, b-Cl-L-Ala, S-benzyl-L-Cys and SOPC (Table 1) com-
pared to EcTrpase. However, the kcat values for S-Me and S-Et-
L
-
Cys are only 10% that of EcTrpase. Rapid-scanning stopped-flow
experiments showed that quinonoid intermediates are formed
from these substrates with rate constants similar to the E. coli en-
zyme (Fig. 5). Thus, the slow reaction of these substrates must be
due to much slower elimination of the small alkylthiols from the
quinonoid complex. This proposal is supported by the lack of for-
mation of an aminoacrylate intermediate absorbing at 350 nm
Physical properties
The well-characterized EcTrpase is relatively thermostable.
Indeed, a heat treatment for 15 min at 65 °C is one of the steps
in the classic purification procedure [16]. The temperature
optimum for EcTrpase is between 50 and 60 °C (Fig. 1), whereas
PpTrpase shows a sharp decrease in activity above 40 °C. This
decrease is probably due to denaturation, since the PpTrpase loses
activity rapidly when incubated at temperatures above 40 °C
(Fig. 2). Thus, PpTrpase is clearly adapted for the low temperatures
(4–15 °C) at which the organism grows. This is also seen strikingly
in the cold inactivation experiment, where EcTrpase rapidly loses
activity upon incubation at 0 °C (Fig. 3) [24,25], while PpTrpase
retains about 80% activity. This cold lability of EcTrpase has been
attributed to facile dissociation of the tetramer into non-catalytic
dimers, with concomitant loss of PLP [24,25]. The loss of activity
of EcTrpase in the cold is reversible, since activity is restored on
incubation at 37 °C for about 1 h [24,25]. Previously, residues
Val15, Ile16 and Val59 were identified as forming a hydrophobic
core which helps to stabilize the tetramer of EcTrpase [25]. In a
sequence alignment, we find that Val59 of EcTrpase is changed to
Ile59 in PpTrpase. This Ile residue increases the strength of the
hydrophobic interaction in the core and thus may be at least
when PpTrpase is mixed with S-Et-L-Cys and benzimidazole. Since
EcTrpase does form an aminoacrylate intermediate under the same
conditions [28], this observation implies that elimination of the
alkylthiol is much slower than the release of pyruvate, resulting
in very little accumulation of the aminoacrylate intermediate,
which binds benzimidazole as an uncompetitive inhibitor [28]. It
is possible that proton transfer to the leaving thiolate group is inef-
ficient for these substrates. The kcat value increases for alkyl Cyste-
ines with bulky groups like benzyl (Table 1), suggesting that the
binding of large groups can lower the activation energy for
elimination.
The reaction of PpTrpase with L-Trp shows the formation of
external aldimine and quinonoid intermediates (Scheme 1),
absorbing at 420 and 505 nm, similar to EcTrpase [28–30], in ra-
pid-scanning stopped-flow experiments (Fig. 6). The formation of
the quinonoid intermediate is biphasic, with the fast phase show-
ing a kinetic isotope effect of 2.4 when
a L-Trp is used. This
-[2H]-

40
R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
Scheme 1.
compares to an isotope effect of 3.6 on quinonoid formation with
EcTrpase [29,35]. However, an additional intermediate absorbing
at 340 nm is seen with PpTrpase (Fig. 6). This intermediate forms
rapidly at the same rate as a 505 nm quinonoid intermediate, but
then decays to give an equilibrating mixture of external aldimine
and quinonoid species (Fig. 6). There is no kinetic isotope effect
japonica [37]. An isopropylmalate dehydrogenase from Shewanella
benthica DB21MT-2 isolated at a depth of 11,000 m in the Mariana
Trench also showed evidence of piezophilicity [38]. The effects of
pressure on oligomerization of single-stranded DNA binding pro-
teins from various species of Shewanella were examined, and the
protein from a piezosensitive strain was found to undergo pres-
sure-dependent dissociation of the tetramer more readily than that
from piezotolerant or piezophilic strains [39]. Our data reported
herein indicate that the Trpase PBPRA2532 from P. profundum
SS9 should be included in the list of known piezophilic enzymes.
on the decay of the 340 nm peak when a L-Trp is used. The ra-
-[2H]-
pid formation of the 340 nm peak from the internal aldimine and
the decay to an external aldimine suggests that the 340 nm inter-
mediate is a gem-diamine, the conversion of which to the external
aldimine is rate-determining. Thus, the 340 nm intermediate is as-
signed to a gem-diamine complex of L-Trp and PLP (Scheme 1).
Conclusions
There seem to be two different forms of the enzyme, one of which
reacts rapidly to form external aldimine and quinonoid intermedi-
ates, and one which reacts slowly. These enzyme species must be
in slow equilibrium. The resting enzyme exists in two forms, which
have absorption maxima at 338 and 420 nm (Fig. 5), similar to
EcTrpase. It is possible that the interconversion of these two spe-
cies is slow for PpTrpase but relatively fast for EcTrpase. The reac-
Although EcTrpase and PpTrpase PBPRA2532 are highly homol-
ogous (>80% identity), they exhibit significant differences in phys-
ical and catalytic properties. EcTrpase is much more heat-stable,
while PpTrpase is more stable to cold and retains higher activity
under hydrostatic pressure. Both enzymes catalyze eliminations
of a range of b-substituted amino acids, as well as tryptophan,
but S-alkyl cysteines are much poorer substrates for PpTrpase.
Rapid-scanning stopped-flow experiments show that similar
intermediates are formed with both enzymes, but a gem-diamine
tion of PpTrpase with
shows
L-Trp in the presence of benzimidazole
a
transient aminoacrylate peak absorbing at 350 nm
(Fig. 6C and D) forming with a rate constant of 18 sꢁ1, similar to
the rate constant of 30 sꢁ1 for aminoacrylate formation by EcTr-
is formed with PpTrpase and L-Trp that is not observed with
pase [29]. Thus, the slower kcat for PpTrpase with L-Trp is not due
to slow elimination.
EcTrpase. The results show that PpTrpase PBPRA2532 is adapted
for optimal activity in the low temperature, high pressure marine
environment.
Piezophilic enzymes
References
In contrast to thermophilic enzymes, there have only been a few
piezophilic enzymes described to date. Although no evidence for
high pressure adaptation of a dihydrofolate reductase from Moritel-
la profunda was found, based on structure and activity comparisons
[36], a recent study of several dihydrofolate reductases found evi-
dence for piezophilicity in the enzymes from the piezophiles, P.
profundum and Moritella yayanosii, compared to the enzymes from
E. coli, and moderate piezophiles, Shewanella violacea and Moritella
[1] E.E. Snell, Adv. Enzymol. Relat. Areas Mol. Biol. 42 (1975) 287–333.
[2] A. Collet, S. Vilain, P. Cosette, G.A. Junter, T. Jouenne, R.S. Phillips, P. Di Martino,
Antonie Van Leeuwenhoek 91 (2007) 71–85.
[3] P. Di Martino, R. Fursy, L. Bret, B. Sundararaju, R.S. Phillips, Can. J. Microbiol. 49
(2003) 443–449.
[4] R.S. Mueller, S. Beyhan, S.G. Saini, F.H. Yildiz, D.H. Bartlett, J. Bacteriol. 191
(2009) 3504–3516.
[5] E.L. Chant, D.K. Summers, Mol. Microbiol. 63 (2007) 35–43.
[6] C. Yanofsky, V. Horn, P. Gollnick, J. Bacteriol. 173 (1991) 6009–6017.

R.S. Phillips et al. / Archives of Biochemistry and Biophysics 506 (2011) 35–41
41
[7] H. Isaacs Jr., D. Chao, C. Yanofsky, M.H. Saier Jr., Microbiology 140 (1994) 2125–
2134.
[25] A. Kogan, G.Y. Gdalevsky, R. Cohen-Luria, Y. Goldgur, R.S. Phillips, A.H. Parola,
O. Almog, BMC Struct. Biol. 9 (2009) 65.
[8] S. Campanaro, A. Vezzi, N. Vitulo, F.M. Lauro, M. D’Angelo, F. Simonato, A.
Cestaro, G. Malacrida, G. Bertoloni, G. Valle, D.H. Bartlett, BMC Genomics 6
(2005) 122.
[26] C.M. Metzler, R. Viswanath, D.E. Metzler, J. Biol. Chem. 266 (1991)
9374–9381.
[27] H. Ikushiro, H. Hayashi, Y. Kawata, H. Kagamiyama, Biochemistry 37 (1998)
3043–3052.
[28] R.S. Phillips, Biochemistry 30 (1991) 5927–5934.
[29] M.S. Sloan, R.S. Phillips, Biochemistry 35 (1996) 16165–16173.
[30] R.S. Phillips, B. Sundararju, N.G. Faleev, J. Am. Chem. Soc. 122 (2000) 1008–
1114.
[31] D.S. June, C.H. Suelter, J.L. Dye, Biochemistry 20 (1981) 2714–2719.
[32] S. Suzuki, T. Hirahara, S. Horinouchi, T. Beppu, Agric. Biol. Chem. 55 (1991)
3059–3066.
[33] Y. Morino, E.E. Snell, J. Biol. Chem. 242 (1967) 2800–2809.
[34] D.S. June, C.H. Suelter, J.L. Dye, Biochemistry 20 (1981) 2707–2713.
[35] R.S. Phillips, J. Am. Chem. Soc. 111 (1989) 727–730.
[36] S. Hay, R.M. Evans, C. Levy, E.J. Loveridge, X. Wang, D. Leys, R.K. Allemann, N.S.
Scrutton, Chembiochem 10 (2009) 2348–2353.
[9] R.K. Dua, E.W. Taylor, R.S. Phillips, J. Am. Chem. Soc. 115 (1993) 1264–1270.
[10] C.T. Walsh, A. Schonbrunn, R.H. Abeles, J. Biol. Chem. 246 (1971) 6855–6866.
[11] D.M. Kiick, R.S. Phillips, Biochemistry 27 (1988) 7339–7344.
[12] R.S. Phillips, N. Johnson, A.V. Kamath, Biochemistry 41 (2002) 4012–4019.
[13] R.S. Phillips, P.D. Gollnick, J. Biol. Chem. 254 (1989) 10627–10632.
[14] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[15] C.H. Suelter, J. Wang, E.E. Snell, FEBS Lett. 66 (1976) 230–232.
[16] Y. Morino, E.E. Snell, Methods Enzymol. 17A (1970) 439–446.
[17] R.S. Phillips, E.W. Miles, G. Holtermann, R.S. Goody, Biochemistry 44 (2005)
7921–7928.
[18] R.S. Phillips, G. Holtermann, Biochemistry 44 (2005) 14289–14297.
[19] Y. Kitamura, T. Itoh, J. Sol. Chem. 16 (1987) 715–725.
[20] E.W. Miles, R. Bauerle, S.A. Ahmed, Methods Enzymol. 142 (1987) 398–414.
[21] I.B.C. Matheson, Comput. Chem. 14 (1990) 49–57.
[22] Y.-H. Li, J. Geophys. Res. 72 (1967) 2665–2678.
[37] C. Murakami, E. Ohmae, S. Tate, K. Gekko, K. Nakasone, C. Kato, J. Biochem. 147
(2010) 591–599.
[23] R.S. Phillips, I. Richter, P. Gollnick, P. Brzovic, M.F. Dunn, J. Biol. Chem. 266
(1991) 18642–18648.
[38] R. Kasahara, T. Sato, H. Tamegai, C. Kato, Biosci. Biotechnol. Biochem. 73 (2009)
2541–2543.
[24] T. Erez, R.S. Phillips, G. Gdalevsky, A. Parola, Biochim. Biophys. Acta 1384
(1998) 365–372.
[39] L.N. Chilukuri, D.H. Bartlett, P.A. Fortes, Extremophiles
383.
6 (2002) 377–