Exploration of the six tryptophan residues of Escherichia coli cystathionine β-lyase as probes of enzyme conformational change

Article abstract of DOI:10.1016/j.abb.2013.07.006

Cystathionine β-lyase (CBL) catalyzes the hydrolysis of l-cystathionine (l-Cth), producing l-homocysteine (l-Hcys), pyruvate and ammonia, in the second step of the transsulfuration pathway of bacteria and plants. A series of 17 site-directed variants of Escherichia coli CBL (eCBL) was constructed to probe the contributions of the six tryptophan residues (W131, W188, W230, W276, W300 and W340) to the fluorescence spectrum of eCBL and to assess their mutability and utility as conformational probes. The effects of these Trp → Phe substitutions on k_cat and K_m ^l-Cth are less than 2-fold, with the exception of the 8-fold increase in K_m^l-Cth observed for eCBL-W340F. The midpoint of thermal denaturation, as monitored by circular dichroism spectroscopy, is reduced 4.7 C by the W188F substitution while the targeted replacement of the other five tryptophans alter T_m by less than 1.7 C. The fluorescence spectrum of eCBL is dominated by W230 and the contribution of W340, situated in the active site, is minor. The observed 5-fold increase in the 336 nm fluorescence emission of W188 between 0 and 2 M urea, suggests a conformational change at the domain interface. Residues W188 and W340, conserved in proteobacterial CBL enzymes, are situated at the core of the domain interface that forms the active-site cleft. The results of this study suggest that W188 is a useful probe of subtle conformational changes at the domain interface and active site.

Full text of DOI:10.1016/j.abb.2013.07.006

Archives of Biochemistry and Biophysics 538 (2013) 138–144
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
Exploration of the six tryptophan residues of Escherichia coli
cystathionine b-lyase as probes of enzyme conformational change
⇑
Allison F. Jaworski, Susan M. Aitken
Department of Biology, Carleton University, Ottawa K1S 5B6, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 May 2013
and in revised form 6 July 2013
Available online 19 August 2013
Cystathionine b-lyase (CBL) catalyzes the hydrolysis of L-cystathionine (L-Cth), producing L-homocysteine
(L-Hcys), pyruvate and ammonia, in the second step of the transsulfuration pathway of bacteria and
plants. A series of 17 site-directed variants of Escherichia coli CBL (eCBL) was constructed to probe the
contributions of the six tryptophan residues (W131, W188, W230, W276, W300 and W340) to the fluo-
rescence spectrum of eCBL and to assess their mutability and utility as conformational probes. The effects
of these Trp ? Phe substitutions on kcat and Km -Cth are less than 2-fold, with the exception of the 8-fold
Keywords:
L
Pyridoxal 50-phosphate
Tryptophan fluorescence
Site-directed mutagenesis
Amino acid metabolism
Cystathionine
L-Cth
increase in Km
observed for eCBL-W340F. The midpoint of thermal denaturation, as monitored by
circular dichroism spectroscopy, is reduced 4.7 °C by the W188F substitution while the targeted replace-
ment of the other five tryptophans alter Tm by less than 1.7 °C. The fluorescence spectrum of eCBL is dom-
inated by W230 and the contribution of W340, situated in the active site, is minor. The observed 5-fold
increase in the 336 nm fluorescence emission of W188 between 0 and 2 M urea, suggests a conforma-
tional change at the domain interface. Residues W188 and W340, conserved in proteobacterial CBL
enzymes, are situated at the core of the domain interface that forms the active-site cleft. The results of
this study suggest that W188 is a useful probe of subtle conformational changes at the domain interface
and active site.
Ó 2013 Elsevier Inc. All rights reserved.
Introduction
because they are not present in mammals, which derive L-Cys from
L-Hcys via the reverse transsulfuration pathway. There are structures
available for 20 members of the c-subfamily within the large and
Cystathionine b-lyase (CBL)1 follows cystathionine
c-synthase
(CGS) in the bacterial transsulfuration pathway. These two closely
related enzymes catalyze the pyridoxal 50-phosphate (PLP)-depen-
dent formation of L-cystathionine (L-Cth), via the condensation of
O-succinyl-L-homoserine (OSHS) and L-cysteine (L-Cys), and its sub-
sequent hydrolysis, producing pyruvate, ammonia and L-homocyste-
ine (L-Hcys), the immediate precursor of L-methionine [1–4]. The
enzymes of the bacterial transsulfuration pathway are attractive
targets for the development of novel anti-microbial compounds
catalytically diverse fold-type I classification of PLP-dependent en-
zymes. These represent seven distinct enzymes from diverse species:
CBL, CGS, cystathionine
verse transsulfuration pathway, methionine
cysteine -lyase, O-acetylhomoserine sulfhydrylase and O-
c
-lyase (CGL) – the second enzyme of the re-
c-lyase (MGL), homo-
c
succinylhomoserine sulfhydrylase [5]. Comparison of these enzymes
reveals conservation of overall structure as well as active-site fea-
tures and residues [2,3,6]. Recent studies probing the active-sites
of Escherichia coli CBL (eCBL) and CGS (eCGS) for determinants of
substrate and reaction specificity have reported that the roles of con-
served residues are dependent on the context of the specific enzyme
[5,7,8]. This illustrates the subtlety of the mechanisms underlying
specificity and demonstrates the need to explore structure–function
relationships, including the subtle differences in active-site architec-
ture and dynamics and the positioning of key residues, to enable the
design of selective inhibitors as well as engineering of the enzymes
within this family.
⇑
Corresponding author. Fax: +1 (613) 520 3539.
E-mail address: susan_aitken@carleton.ca (S.M. Aitken).
Abbreviations used: AAT, aspartate aminotransferase; CBL, cystathionine b-lyase;
1
eCBL, Escherichia coli CBL; CGL, cystathionine c-lyase; CGS, cystathionine c-synthase;
eCGS, E. coli CGS; L-Cth, L-cystathionine; DTNB, 5,50-dithiobis(2-nitrobenzoic acid);
FRET, fluorescence resonance energy transfer; L-Hcys, L-homocysteine; MGL, methi-
onine
c-lyase; Ni–NTA, Ni–nitrilo triacetic acid; L-OSHS, O-succinyl-L-homoserine;
PLP, pyridoxal 50-phosphate; sOASS,Salmonella typhimurium O-acetylserine sulfhydr-
ylase. eCBL Trp ? Phe variants (defined individually in Table 1) are identified as
follows: individual Trp ? Phe substitutions: W131F, W188F, W230F, W276F, W300F,
W340F; triple substitution variant (tX): W131F + W230F + W276F; pentuple substi-
tution variants (pX) are identified by the remaining tryptophan residue, e.g.
pXW131 = W188F + W230F + W276F + W300F + W340F; all tryptophan residues of
the hextuple variant (hX) are replaced: W131F + W188F + W230F + W276F +
W300F + W340F.
The eCBL monomer is comprised of three domains, which each
possess active-site residues. Residues 1–60 form the amino-termi-
nal domain that contributes to the active-site of the neighboring
monomer. The catalytic (residues 61–256) and carboxy-terminal
0003-9861/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.abb.2013.07.006

A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
139
Fig. 1. The location of the six tryptophan residues (yellow) of eCBL in the context of the (A) homotetramer and (B) catalytic dimer. (C) The active site, containing the PLP
cofactor (yellow) is situated at the interface between the catalytic (residues 61–256, white) and carboxy-terminal (residues 257–395, grey) domains. The amino-terminal
domain (residues 1–60, dark blue) contributes to the active site of the neighboring subunit. Tryptophan residues are highlighted to illustrate their relative positions: W131
(orange), W188 (teal), W230 (pink), W276 (purple), W300 (blue) and W340 (green). (For interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
L-Cth
L
(residues 257–395) domains form the two sides of the active-site
cleft (Fig. 1) [2]. The central location of W188 and W340 within
the active-site cleft provides potential probe(s) of changes in ac-
tive-site conformation, either during catalysis or resulting from
protein engineering studies. A series of 17 site-directed variants
of eCBL residues W131, W188, W230, W276, W300 and W340
was constructed and characterized to determine the efficacy of na-
tive tryptophan residues as conformational probes.
the Michaelis–Menten equation and kcat -Cth/Km
independently from Eq. 1.
was obtained
v
½Eꢁ
k
cat=KLꢀCth ꢂ ½L ꢀ Cthꢁ
m
¼
ð1Þ
1 þ ½L ꢀ Cthꢁ=KmLꢀCth
Fluorescence and far-UV circular dichroism spectra of eCBL
Fluorescence spectra of wild-type and variants of eCBL in
20 mM potassium phosphate, pH 7.5, were recorded with a Varian
Cary Eclipse spectrophotometer in a 1 cm path-length cell at 25 °C,
Materials and methods
with 5 nm slits. Similarly, far-UV CD spectra of 5.6 lM enzyme
Reagents
were recorded on a Jasco-J815 spectropolarimeter in a 1 mm
path-length cell by averaging five scans from 260 to 200 nm
(1 nm bandwidth) in 0.2 nm steps at a rate of 20 nm/min. The ellip-
ticity at 222 nm (1 nm bandwidth) was measured between 20 and
L
-Cth was purchased from Sigma. Ni-nitrilotriacetic acid (Ni-
NTA) resin and 5,50-dithiobis(2-nitrobenzoic acid) (DTNB) were ob-
tained from Qiagen and Pierce, respectively. Oligonucleotide prim-
ers were synthesized by Integrated DNA Technologies and site-
directed mutants were sequenced by BioBasic prior to expression
and purification to ensure only the desired mutation(s) were pres-
ent for each Trp ? Phe variant.
80 °C with instrument parameters set to
DT of 18 °C/hr with a 1 °C
step resolution.
Urea denaturation of eCBL
Wild-type and tryptophan replacement variants of eCBL were
incubated at 25 °C with 0–8 M urea for 26 h in 20 mM phosphate
Construction, expression and purification of site-directed variants
buffer, pH 7.5. Fluorescence spectra of 1
Trp ? Phe variants and 5.6 M wild-type eCBL, and far-UV CD
spectra of 5.6 M wild-type enzyme were recorded. To assess the
ability of eCBL to refold following urea denaturation, 30 M wild-
type eCBL was incubated in 1.4, 2.5, 3, 5 and 8 M urea. Enzyme
from each treatment was subsequently diluted in 20 mM phos-
lM wild-type eCBL and
Site-directed variants of eCBL, constructed via the overlap-
extension polymerase chain reaction method, were inserted into
the pTrc-99aAF plasmid [9]. Wild-type and site-directed variants
of eCBL were expressed in the E. coli ER1821 metC::cat strain, in
which the gene encoding eCBL is replaced by cat, encoding resis-
tance to chloramphenicol, to prevent contamination with the
wild-type E. coli enzyme [9]. The eCBL enzymes were expressed
and purified, via Ni-NTA affinity chromatography, as described by
Farsi et al. [9]
l
l
l
phate, pH 7.5, to a final enzyme concentration of 1
bated for further 26 h at 25 °C prior to recording of
fluorescence spectra and measurement of specific activity at
lM, and incu-
a
2.5 mM L-Cth.
Determination of steady-state kinetic parameters
Results
Enzyme activity was measured in a total volume of 100
25 °C on a Spectramax 340 microtiter plate spectrophotometer
(Molecular Devices). The assay buffer was comprised of 50 mM
l
L at
Residues W188 and W340 are situated at the core and W131 and
W300 are located at the periphery of the interface of the catalytic
and carboxy-terminal domains, which forms the active-site cleft
of eCBL (Fig. 1) [2]. These tryptophans provide potential probe(s)
of inter-domain and/or active-site conformational changes. A set
of 17 Trp ? Phe replacement variants, targeting the six tryptophan
residues of eCBL (W131, W188, W230, W276, W300 and W340),
was constructed to characterize the contribution of each to the
Tris, pH 8.5, with 20
using 2 mM DTNB (
sulfhydryl group of the
l
e
M PLP. The hydrolysis of
412 = 13,600 Mꢀ1sꢀ1), which reacts with the
-Hcys product [10]. A background reading
was recorded before initiating the reaction with the eCBL enzyme.
L-Cth was detected
L
Values of kcat -Cth and Km -Cth were obtained by fitting of the data to
L
L

140
A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
fluorescence spectrum of eCBL and the effect of their substitution
on activity and stability.
40
20
20
10
0
0
-5
-10
-15
-20
-25
-30
-35
A
B
-10
-15
-20
-25
-30
-35
Effect of Trp ? Phe substitution on the activity, stability and
fluorescence spectrum of eCBL
0
The 17 site-directed variants of eCBL are soluble with yields of
at least 7 mg/L. The yields of the Trp ? Phe replacement variants
targeting one, three or four, and five or six of the six tryptophan
residues of eCBL were 35–68, 8–22 and 7–10 mg/L, respectively.
This is similar to the 15–41 and 56 mg/L values reported for 15 sin-
gle-substitution active-site variants of eCBL and the wild-type en-
zyme, respectively [5,7,9]. Replacement of W340 with
phenylalanine precludes the formation of hydrogen bonds with
24
32
40
48
56
64
72
Temperature (oC)
-10
-20
-30
-40
-20
-40
-60
0
-2
C
-4
-5
-6
-10
-15
-20
-25
-30
-35
-8
-10
-12
-14
-16
-18
the 30 hydroxyl moiety of PLP and the
a-carboxylate group of the
24
32
40
48
56
64
72
substrate, bound in Schiff base linkage with the cofactor, and re-
o
Temperature ( C)
sults in a 8-fold increase in Km -Cth (Table 1) [2,7]. With the excep-
L
200
210
220
230
240
250
260
tion of those including the W340F substitution, all variants possess
Wavelength (nm)
near-native levels of the physiological
a
,b-elimination activity. The
L-Cth
kcat and Km
values of the W131F, W188F, W230F, W276F and
Fig. 2. Residue W188 possesses distinct properties compared to the other Trp
residues of eCBL. The circular dichroism (A) spectra and (B and C) thermal
denaturation profiles at 222 nm of 5.6 lM wild-type (solid line, open diamond) and
the single W188F (dashed line, open circle) and pentuple pXW188 (dotted line,
W300F variants are within 1.6-fold of the wild-type enzyme (Ta-
-
L
ble 1). The minor 3–30% and 2–45% changes in kcat and kcat/Km
Cth of W131F, W188F, W230F, W276F and W300F are largely addi-
tive and, for example, account for the 42% and 54% decreases in
these parameters, respectively, of the pXW340 variant, for which
all tryptophans except W340 have been replaced by phenylalanine.
closed circle) variants of eCBL.
respectively (Table 1, Figs. 2 and 3). The 4.4 °C difference between
pXW188 and hX is similar to the 4.7 °C decrease in Tm of W188F
compared to wild-type eCBL.
Loss of a-helical structure, as determined by circular dichroism
spectroscopy, was employed to assess the relative thermostability
of each site-directed variant, compared to the wild-type enzyme.
Substitution of W131, W230 or W300 results in increases in Tm
of 1.2–1.7 °C. In contrast, W340F has no effect and W276F de-
creases Tm by 1.0 °C. The 4.7 °C decrease in Tm observed for
W188F is unique among the single substitution variants and is evi-
dent in each of the multi-residue substitution variants containing
the W188F substitution (Table 1, Fig. 2). For example, the Tm value
of the tX (W131F, W230F, W276) variant is 5.6 °C higher than
tX + W188F, and the latter is decreased 2.7 °C compared to the
wild-type enzyme. The minor stabilizing effect of the other
Trp ? Phe substitutions is generally additive, as evidenced by the
pXW188 variant in which all Trp residues, with the exception of
W188, have been replaced with Phe, and hX, which contains all
six Trp ? Phe substitutions. The Tm values of the pXW188 and
hX proteins, which differ only in the presence of W188 in the for-
mer, are 6.1 and 1.7 °C higher than the wild-type eCBL enzyme,
The far-UV circular dichroism spectrum of eCBL is not affected
by substitution of any of the six tryptophan residues, as illustrated
when comparing the spectra of wild-type eCBL and the W188F,
pXW188 (Fig. 2A) and hX (Fig. 3B) variants. Comparison of the fluo-
rescence spectra (kex = 295 nm) of the single and pentuple
Trp ? Phe substitution variants, replacing one and five of the six
tryptophan residues of eCBL, respectively, demonstrates that while
W230 dominates the fluorescence spectrum of eCBL, the contribu-
tion of W340 is minor (Fig. 4).
Urea denaturation of eCBL
The fluorescence profile for urea denaturation of eCBL is com-
plex. The broad emission spectrum of eCBL is increased 5-fold in
intensity at 336 nm between 0 and 2 M urea, with subsequent
Table 1
Tm values and kinetic parameters of the
a
,b-elimination of L-Cth for wild-type site-directed variants of eCBL.a.
Enzyme
Trp remaining
kcat (sꢀ1
)
Km -Cth (mM)
kcat/KmE -Cth (Mꢀ1sꢀ1
)
Tm (°C)
L
L
eCBL
W131F
W188F
W230F
W276F
W300F
W340F
tX (W131F/W230F/W276F)
tX + W188F
tX + W300F
tX + W340F
pXW131
pXW188
pXW230
pXW276
pXW300
W131, W188, W230, W276, W300, W340
W188, W230, W276, W300, W340
W131, W230, W276, W300, W340
W131, W188, W276, W300, W340
W131, W188, W230, W300, W340
W131, W188, W230, W276, W340
W131, W188, W230, W276, W300
W188, W300, W340
W300, W340
W188, W340
W188, W300
W131
W188
W230
W276
W300
54.5 0.7
45.3 0.7
38.1 0.2
55.9 0.7
60.4 0.6
42.5 0.4
60.7 0.5
54.5 0.4
25.4 0.9
0.116 0.007
0.101 0.008
0.121 0.003
0.19 0.01
0.131 0.006
0.161 0.007
0.91 0.05
0.171 0.006
0.27 0.04
0.31 0.03
1.0 0.1
(4.7 0.3) ꢂ 105
(4.5 0.3) ꢂ 105
(3.15 0.06) ꢂ 105
(3.0 0.1) ꢂ 105
(4.6 0.2) ꢂ 105
(2.6 0.1) ꢂ 105
(6.6 0.1) ꢂ 104
(3.2 0.1) ꢂ 105
(9.4 0.1) ꢂ 104
(2.4 0.2) ꢂ 105
(6.6 0.7) ꢂ 104
(2.1 0.1) ꢂ 104
(4.3 0.2) ꢂ 104
(3.3 0.2) ꢂ 104
(1.8 0.1) ꢂ 104
(2.3 0.1) ꢂ 104
(2.13 0.09) ꢂ 105
(2.5 0.1) ꢂ 104
62.9 0.2
64.1 0.2
58.1 0.4
64.1 0.2
61.9 0.2
64.6 0.3
63.2 0.3
65.8 0.2
60.2 0.3
65.6 0.2
66.7 0.3
62.6 0.3
69.0 0.7
63.4 0.3
66.4 0.6
64.6 0.3
63.1 0.4
64.6 0.3
74
65
42
54
1
3
1
1
2.0 0.1
1.26 0.08
1.05 0.09
2.1 0.2
1.8 0.1
0.148 0.007
1.8 0.2
35.1 0.9
38
40
2
1
pXW340
hX (all W ? F)
W340
No Trp
31.7 0.3
46
2
a
Kinetic parameters reported are for hydrolysis of l-Cth. Reaction conditions: 2 mM DTNB, 0.01–6.4 mM
L
-Cth and 0.068–6.6 lM wild-type or eCBL variant, depending on
L-Cth
L-Cth
the activity of the enzyme, in assay buffer at 25 °C. The data were fit to the Michaelis–Menten equation to obtain kcat and Km
and Eq. 1 to obtain kcat and Km
.

A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
141
450
400
350
300
250
200
150
100
50
0
-5
-5
390
380
370
360
350
340
330
320
B
30
20
30
20
10
0
A
A
B
10
0
-10
-15
-20
-25
-10
-15
-20
-25
-30
-35
-10
-20
-30
-40
-10
-20
-30
0
1
2
3
4
5
6
7
8
[urea] (M)
200
210
220
230
240
250
260
Wavelength (nm)
0
1
2
3
4
5
6
7
8
24
32
40
48
56
64
72
[urea] (M)
Temperature (oC)
400
350
300
250
200
150
100
50
Fig. 3. Circular dichroism (A) thermal denaturation profiles at 222 nm and (B)
spectra of 5.6 M wild-type eCBL (d, solid line) and the hextuple Trp ? Phe (hX)
variant (s, dashed line).
400
3
D
C
350
300
250
l
3
4
200
150
5
100
2
50
1
600
1000 B
0
4
2
A
340
350
360
370
380
390
Wavelength (nm)
800
500
400
300
200
100
600
400
200
0
5
1
0
320
360
400
440
480
520
Wavelength (nm)
330
340
350
360
370
380
390 400
Wavelength (nm)
Fig. 5. The fluorescence (A) intensity at 336 nm and (B) kmax of the urea
denaturation profiles of 1 M wild-type eCBL (open diamond), W188F (open circle)
l
and pXW188 (closed circle) and the corresponding fluorescence spectra of (C) wild-
type eCBL and (D) pXW188 at (1) 0, (2) 1, (3) 2, (4) 4 and (5) 8 M urea.
0
320
360
400
440
480
520
Wavelength (nm)
by similar 380 nm peaks observed upon excitation of the
tX + W340F and pXW188 variants at 295 (Fig. 6A) and 330 nm
(Fig. 6B) [11]. The sharp decrease in fluorescence intensity between
3 and 3.5 M urea is likely due to aggregation, as centrifugation of
samples prior to collection of spectra results in a decrease in the
emission intensity in this range of urea concentration (Fig. 7).
The circular dichroism spectrum of eCBL is dominated by the 208
and 222 nm minima characteristic of a-helices (Fig. 2) and the in-
crease in 222 nm absorption between 5 and 6 M urea suggests that
secondary structure is maintained between 0 and 5 M urea (Fig. 7).
Fig. 4. Fluorescence spectra (kex = 295 nm) of 20 lM (A) wild-type eCBL (solid line),
W230F (dashed line) and W340F (dotted line) and (B) the pentuple Trp ? Phe
variants retaining W230 (pXW230, dashed line) and W340 (pXW340, dotted line).
decreases and red-shifts between 3 and 5 M and above 5 M urea
(Fig. 5A and B). The emission maximum is shifted to 345 and
360 nm at 4 and 8 M urea, respectively (Fig. 5B and C). Comparison
of the denaturation profiles of the 12 single and pentuple
Trp ? Phe variants identified W188 as the residue that is released
from quenching between 0 and 2 M urea as its replacement
(W188F) results in elimination of the hyperfluorescent species,
which is present in only one (pXW188) of the six pentuple
Trp ? Phe variants (Fig. 5A). Variants that lack W340 and possess
W188 (e.g. pXW188, tX + W340F and W340F) undergo a unique
shift in fluorescence maxima to 380 nm upon incubation with
0.3–1.4 M urea (Figs. 5B and D and 6A). The 380 nm species results
from fluorescence resonance energy transfer (FRET) between
W188 and the enolamine tautomer of the internal aldimine of
the PLP cofactor, in Schiff base linkage with K210, as demonstrated
The wild-type enzyme (30
2.5, 3, 5 and 8 M urea prior to dilution to a final enzyme concentra-
tion of 1 M in 20 mM phosphate buffer, pH 7.5, to assess the abil-
lM) was incubated for 26 h with 1.4,
l
ity of eCBL to refold. The presence of 2 M urea reduces eCBL activity
by 95%. Recovery of 93%, 95% and 82% activity was observed fol-
lowing dilution of enzyme incubated in 1.4, 2.5 and 3 M urea,
respectively. In contrast, dilution of enzyme incubated in 5 and
8 M urea results in the recovery of only 0.5% and 4% of activity,
respectively, thereby providing support for proposed aggregation
of the enzyme between 3 and 3.5 M urea and precluding thermo-
dynamic analysis.

142
A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
provides an effective system for studies probing reaction specific-
600
500
400
300
200
100
0
240
200
160
120
80
B
ity as structures of inhibitor complexes are available and the roles
of 15 active-site residues have been characterized [2,5,7,13,14].
This study aims to provide tools to probe protein conformational
changes in order to facilitate mechanistic and protein engineering
A
studies of eCBL and related enzymes of the c-subfamily.
40
Fluorescence and fluorescence quenching are effective meth-
ods to detect changes in protein conformation. In the case of pro-
teins containing multiple tryptophan residues, characterization of
each is necessary to access their efficacy as probes [15,16]. This is
illustrated by studies employing site-directed mutagenesis to
investigate the individual tryptophan residues of Salmonella
typhimurium O-acetylserine sulfhydrylase (sOASS) and human
cD-crystallin [17,18]. However, the conservation of tryptophan
in protein sequences is high and its substitution is commonly
associated with reduced stability [17]. The six tryptophan resi-
0
380 385 390 395 400 405 410 415 420
Wavelength (nm)
dues of eCBL are common to
In contrast, conservation at the corresponding positions is not ob-
served for the closely related enzymes of the -subfamily, which
c-proteobacterial CBL sequences.
340 360 380 400 420 440 460 480 500
Wavelength (nm)
c
Fig. 6. Effect of 1.2 M urea on the fluorescence spectrum of 1 lM wild-type eCBL
suggests the feasibility of their conservative replacement with
phenylalanine for the purpose of characterizing the contribution
of each to the fluorescence spectrum of eCBL and to assess their
ability to serve as conformational probes. The location of eCBL-
W188 and W340 at the core and on opposite sides of the inter-
face between the catalytic and carboxy-terminal domains, which
form the active-site cleft, is well-suited to this purpose. Similarly,
Campanini and coworkers characterized the tyrosine-substitution
variants of the two tryptophan residues (W50 and W161) of
sOASS, which are situated such that they provide sensitive probes
of conformational changes in the amino and carboxy-terminal do-
(solid line), the pentuple Trp ? Phe variant pXW188 (dashed line), which maintains
only W188, W340F (sparse-dotted line), and quadruple Trp ? Phe variant
tX + W340F (close-dotted line), which maintains only W188 and W300. Fluores-
cence spectra for excitation at (A) 295 nm and (B) 330 nm were recorded in the
absence of urea (gray lines) and in the presence of 1.2 M urea (black lines).
600
500
400
300
200
100
0
-5
-10
-15
-20
-25
mains of this enzyme [17].
L-Cth
With the exception of the 8-fold increase in Km
of W340F,
the effect of substituting the six tryptophan residues with phenyl-
alanine on the kinetic parameters of eCBL is minor [7]. The kcat and
Km -Cth of the W131F, W188F, W230F, W276F and W300F variants
L
are within 1.6-fold and the catalytic efficiency of eCBL-pXW340,
which possesses only W340, is reduced by 2.2-fold, compared to
the wild-type enzyme (Table 1). Similarly, the minor differences
between the CD spectra of the Trp ? Phe variants and the wild-
type enzyme likely reflect the absorption of tryptophan in the
far-UV region rather than changes in secondary structure (Figs. 2
and 3) [17,19]. The fluorescence spectra of the pentuple Trp ? Phe
replacement variants were compared with the corresponding dif-
ference spectra, produced by subtraction of the spectrum of the
single tryptophan replacement variant (i.e. the spectrum of the
remaining five tryptophan residues) from that of the wild-type en-
zyme. The spectrum of pXW340 is broadened and red shifted com-
pared to the corresponding difference spectrum, suggesting a
minor conformational perturbation that is localized to the inter-
face between the two major domains and likely indicates a shift
in the equilibrium between the open and closed conformations of
0
1
2
3
4
5
6
7
8
[Urea] (M)
Fig. 7. Urea denaturation profiles of 5.6 lM wild-type eCBL monitored via
fluorescence (kex = 295 nm, kem = 338 nm), prior to (black closed circle) and after
(grey closed circle) centrifugation at 13500 rpm for 15 min, and circular dichroism
spectroscopy at 222 nm (closed diamond).
Discussion
the active site. Loss of a-helical secondary structure, monitored by
The enzymes of the c-subfamily of fold-type I provide a useful
circular dichroism, resulting from thermal denaturation was em-
ployed to determine the relative effects of replacing the six trypto-
phan residues of eCBL with phenylalanine. With the exception of
the 4.7 °C decrease in Tm observed for W188F, the effect of the indi-
vidual Trp ? Phe substitutions on the thermostability of eCBL is
minor. Replacement of W131, W230 and W300 with phenylalanine
results in slight stabilization of the enzyme as the Tm of these vari-
ants is increased by 1.2–1.7 °C (Table 1). The observed additivity of
the Trp ? Phe substitutions on thermostability is demonstrated by
the 6.1 and 1.7 °C increases in Tm values, compared to the wild-
type enzyme, of the pXW188 and hX proteins, respectively, which
differ only in the presence of W188 in the former (Table 1, Figs. 2
and 3). In contrast with the common trend for correlation between
increased stability and decreased activity, resulting from the
model system for investigations of the structure–function relation-
ships underlying substrate and reaction specificity in PLP-depen-
dent enzymes as their strong structural similarity extends
beyond a common overall fold to the conservation of several key
active-site residues [4,6,12]. The role(s) of the conserved active-
site residues, shared by eCBL and eCGS, are influenced by the con-
text of the specific active site [5,7,8]. However, the underlying fea-
tures acting as determinants of specificity, including active-site
dynamics and conformational changes, are not readily apparent
from comparison of the available structures. Therefore, exploration
of the complex and subtle nature of the determinants of reaction
specificity among the enzymes of the
c-subfamily will be assisted
by the development of conformational probes. The eCBL enzyme

A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
143
corresponding reduction in enzyme flexibility, no change in activ-
ity is observed for variants with increased thermostability, com-
pared to the wild-type enzyme (Table 1) [18,20,21].
Thermodynamic parameters could not be calculated for eCBL be-
cause its denaturation is not fully reversible. Only 0.5% and 4%
activity is recovered following refolding from 5 and 8 M urea,
respectively.
transition dipoles are no longer appropriately oriented with re-
spect to one another to result in FRET [16]. Residues W188 and
W340 of eCBL are bridged by L310, which contributes to the main-
tenance of their relative orientations [2]. Substitution of W340 may
allow L310 to move, resulting in the reorientation of W188 with re-
spect to the cofactor and enable the observed FRET, which demon-
strates that maintenance of W340 is required for native active-site
architecture. The release from quenching of W188 between 0 and
2 M urea and the observed FRET to the cofactor in the context of
the W340F substitution in the presence of 0.3–1.4 M urea, demon-
strate the sensitivity of W188 to changes in active-site
architecture.
Substitution of W131, W230, W276 and W300 with phenylala-
nine, either individually or together (tX + W300F), does not affect
the kinetic parameters, secondary structure or stability of eCBL
(Table 1). This suggests that the quadruple Trp ? Phe substitution
variant tX + W300F can be employed as a model enzyme and raises
the possibility of introducing other tryptophan residues in the con-
text of the tX + W300F variant to directly probe other regions of
interest. For example, substitution of a residue at the dimer inter-
face (e.g. Y250 or Y343) with tryptophan would assist in identify-
ing the region within the complex eCBL denaturation profile that
corresponds to dissociation of the homotetramer to catalytic di-
mers since none of the six native tryptophan residues are located
in proximity to the dimer interface.
The fluorescence spectrum of eCBL is dominated by W230
(Fig. 4). The feasibility of substituting this residue to reveal the
fluorescence of residues better positioned to act as probes of ac-
tive-site conformation is demonstrated by the negligible effects
of W230F on the stability and kinetic parameters of eCBL. In con-
trast with W230, the contribution of W340 to the fluorescence
spectrum is minor. Quenching of tryptophan fluorescence in PLP
enzymes is common due to FRET between tryptophan and PLP or
by direct interaction with the cofactor via short-range electrostatic
interactions [15,22]. Quenching of W340 may reflect the latter pro-
cess, given the hydrogen bond formed between the side chain of
this residue and PLP-O30 and since FRET is not observed in the
pXW340 variant, which possesses only W340 (Fig. 4). The minor
contribution of W340 to the fluorescence spectrum of eCBL in com-
bination with the observed 8-fold increase in Km -Cth of W340F sug-
L
gests that maintaining this residue, required for native activity,
will not interfere with the use of other residue(s), such as W188,
as conformational probe(s).
The residue responsible for the 5-fold increase in the 336 nm
emission of eCBL between 0 and 2 M urea was identified as
W188 by comparing the denaturation profile of the wild-type en-
zyme to those of the 12 single and pentuple Trp ? Phe substitution
variants (Fig. 5A). Similarly, a hyperfluorescent species correspond-
ing to sOASS-W161 was reported, with maximum emission inten-
sity between 0.8 and 1 M guanidinium hydrochloride, in the
denaturation profile of the W50Y variant of sOASS [17]. The ob-
served hyperfluorescence in eCBL may indicate a stable intermedi-
ate in the denaturation pathway or be due to release of quenching
resulting from the reorientation of W188 caused by increased pre-
transition flexibility of the native state [17]. The latter seems likely
given the location of W188 at the interface of the two major struc-
tural domains that form the active-site cleft and the probability
that movement of the domains relative to one another is required
for catalysis. The sharp decrease in the emission intensity and red-
shift to 345 nm between 3 and 3.5 M urea (Fig. 5) in addition to the
loss of fluorescence intensity in this range following centrifugation
is indicative of aggregation (Fig. 7). This suggests that flexibility,
particularly at the domain interface, is increased between 0 and
2 M urea followed by a conformational change between 3 and
3.5 M urea that exposes a hydrophobic surface, which results in
the aggregation that precludes the refolding of eCBL.
The hyperfluorescence of W188 enables this residue, located in
the active-site cleft, to act as a sensitive probe of subtle active-site
conformational changes, as recently reported for chimeric variants
of eCBL [23]. Comparison of eCGS and eCBL reveals two structurally
distinct segments situated in proximity to the amino (region 1) and
carboxy (region 2) termini, which are located at the entrance of the
active-site. Analysis of 12 chimeric variants interchanging regions
1 and/or 2 of eCGS and eCBL demonstrated that exchange of region
2, in the context of eCBL, results in a ꢃ3-fold increase in fluores-
cence emission at 336 nm, which is likely due to release of W188
from quenching, thereby reflecting a subtle alteration in the
three-dimensional structure [23].
The enzymes of the c-subfamily of PLP-dependent enzymes
demonstrate strong conservation in both structure and active-site
residues. Recent studies have demonstrated that the roles of se-
lected conserved active-site residues (e.g. eCBL-S339, which corre-
sponds to eCGS-S326) depend on the context of the active site
[5,8]. This demonstrates the subtlety of the features that act as
determinants of specificity in this enzyme family and illustrates
the necessity of developing tools to probe the complex struc-
ture–function relationships that underlie specificity. This study
has identified the tX + W300F variant, in which residues W131,
W230, W276 and W300 are replaced by phenylalanine, as a suit-
able model enzyme for mechanistic and protein engineering stud-
ies. The tX + W300F variant maintains wild-type kinetic
parameters and stability while simplifying the fluorescence spec-
trum of eCBL, thereby enabling effective use of the sensitive probe
of conformational changes at the active site and domain interface
provided by the unique fluorescence properties of W188.
Acknowledgment
The authors are very grateful to Drs. Joanne Turnbull, Judith
Kornblatt, and Jack Kornblatt for their assistance with the circular
dichroism experiments, suggestions for further experiments and
interpretation of the eCBL denaturation pathway. This work was
supported by the Natural Sciences and Engineering Research Coun-
cil of Canada.
References
[1] J. Belfaiza, C. Parsot, A. Martel, C. Bouthier de la Tour, D. Margerita, G.N. Cohen,
I. Saint Girons, Proc. Natl. Acad. Sci. USA 83 (1986) 867–871.
[2] T. Clausen, R. Huber, R. Laber, H.D. Pohlenz, A. Messerschmidt, J. Mol. Biol. 262
(1996) 202–224.
[3] T. Clausen, R. Huber, L. Prade, M.C. Wahl, A. Messerschmidt, EMBO J. 17 (1998)
6827–6838.
[4] S.M. Aitken, P.H. Lodha, D.J.K. Morneau, Biochim. Biophys. Acta 1814 (2011)
1511–1517.
[5] P.H. Lodha, S.M. Aitken, Biochemistry 50 (2011) 9876–9885.
[6] A. Messerschmidt, M. Worbs, C. Steegborn, M.C. Wahl, R. Huber, B. Laber, T.
Clausen, Biol. Chem. 384 (2003) 373–386.
In wild-type eCBL, residue W188 is oriented such that it is
quenched and FRET is not observed. However, the combination of
the W340F substitution in the presence of 0.3–1.4 M urea, allows
reorientation of W188 and/or PLP to permit FRET. Similarly,
W177 of the b2 subunit of E. coli OASS was proposed to rotate rel-
[7] P.H. Lodha, A.F. Jaworski, S.M. Aitken, Protein Sci. 19 (2010) 383–391.
[8] A.F. Jaworski, P.H. Lodha, A.L. Manders, S.M. Aitken, Protein Sci. 21 (2012)
1662–1671.
ative to the cofactor, upon binding of the
a subunit, such that the

144
A.F. Jaworski, S.M. Aitken / Archives of Biochemistry and Biophysics 538 (2013) 138–144
[9] A. Farsi, P.H. Lodha, J.E. Skanes, H. Los, N. Kalidindi, S.M. Aitken, Biochem. Cell
[17] B. Campanini, S. Raboni, S. Vaccari, L. Zhang, P.F. Cook, T.L. Hazlett, A.
Mozzarelli, S. Bettati, J. Biol. Chem. 39 (2003) 37511–37519.
[18] J. Chen, S.L. Flaugh, P.R. Callis, J. King, Biochemistry 45 (2006) 11552–11563.
[19] F.X. Schmid, in: T.E. Creighton (Ed.), Protein Structure, vol. 174, IRL Press,
Oxford, 1997, pp. 261–297.
Biol. 87 (2009) 445–457.
[10] S.M. Aitken, J.F. Kirsch, Biochemistry 42 (2003) 571–578.
[11] T.D. Turbeville, J. Zhang, G.A. Hunter, G.C. Ferreira, Biochemistry 46 (2007)
5972–5981.
[12] S.M. Aitken, J.F. Kirsch, Arch. Biochem. Biophys. 433 (2005) 166–175.
[13] T. Clausen, R. Huber, A. Messerschmidt, H.D. Pohlenz, B. Laber, Biochemistry 36
(1997) 12633–12643.
[20] Z. Markovic-Housley, B. Stolz, R. Lanz, B. Erni, Protein Sci. 8 (1999) 1530–1535.
[21] K.F. Fulton, S.E. Jackson, A.M. Buckle, Biochemistry 42 (2003) 2364–2372.
[22] J.E. Churchich, Biochemistry 4 (1965) 1405–1410.
[14] E.J. Ejim, J.E. Blanchard, K.P. Koteva, R. Sumerfield, N.H. Elowe, J.D. Chechetto,
E.D. Brown, M.S. Junop, G.D. Wright, J. Med. Chem. 50 (2007) 755–764.
[15] M. Arrio-Dupont, Eur. J. Biochem. 91 (1978) 369–378.
[16] A.N. Lane, Eur. J. Biochem. 133 (1983) 531–538.
[23] A.L. Manders, A.F. Jaworski, M. Ahmed, S.M. Aitken, Biochim. Biophys. Acta
1834 (2013) 1044–1053.