The molecular details of WPD-loop movement differ in the protein-tyrosine phosphatases YopH and PTP1B

Article abstract of DOI:10.1016/j.abb.2012.06.002

The movement of a conserved protein loop (the WPD-loop) is important in catalysis by protein tyrosine phosphatases (PTPs). Using kinetics, isotope effects, and X-ray crystallography, the different effects arising from mutation of the conserved tryptophan in the WPD-loop were compared in two PTPs, the human PTP1B, and the bacterial YopH from Yersinia. Mutation of the conserved tryptophan in the WPD-loop to phenylalanine has a negligible effect on k _cat in PTP1B and full loop movement is maintained. In contrast, the corresponding mutation in YopH reduces k_cat by two orders of magnitude and the WPD loop locks in an intermediate position, disabling general acid catalysis. During loop movement the indole moiety of the WPD-loop tryptophan moves in opposite directions in the two enzymes. Comparisons of mammalian and bacterial PTPs reveal differences in the residues forming the hydrophobic pocket surrounding the conserved tryptophan. Thus, although WPD-loop movement is a conserved feature in PTPs, differences exist in the molecular details, and in the tolerance to mutation, in PTP1B compared to YopH. Despite high structural similarity of the active sites in both WPD-loop open and closed conformations, differences are identified in the molecular details associated with loop movement in PTPs from different organisms.

Full text of DOI:10.1016/j.abb.2012.06.002

Archives of Biochemistry and Biophysics 525 (2012) 53–59
Contents lists available at SciVerse ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
The molecular details of WPD-loop movement differ in the protein-tyrosine
phosphatases YopH and PTP1B q
a,
Tiago A.S. Brandão a,b, Sean J. Johnson a, , Alvan C. Hengge
⇑
⇑
a Department of Chemistry and Biochemistry, Utah State University, Logan, UT 84322-0300, United States
b Departamento de Química, ICEX, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais 31270-901, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 May 2012
and in revised form 3 June 2012
Available online 12 June 2012
The movement of a conserved protein loop (the WPD-loop) is important in catalysis by protein tyrosine
phosphatases (PTPs). Using kinetics, isotope effects, and X-ray crystallography, the different effects aris-
ing from mutation of the conserved tryptophan in the WPD-loop were compared in two PTPs, the human
PTP1B, and the bacterial YopH from Yersinia. Mutation of the conserved tryptophan in the WPD-loop to
phenylalanine has a negligible effect on kcat in PTP1B and full loop movement is maintained. In contrast,
the corresponding mutation in YopH reduces kcat by two orders of magnitude and the WPD loop locks in
an intermediate position, disabling general acid catalysis. During loop movement the indole moiety of the
WPD-loop tryptophan moves in opposite directions in the two enzymes. Comparisons of mammalian and
bacterial PTPs reveal differences in the residues forming the hydrophobic pocket surrounding the con-
served tryptophan. Thus, although WPD-loop movement is a conserved feature in PTPs, differences exist
in the molecular details, and in the tolerance to mutation, in PTP1B compared to YopH. Despite high
structural similarity of the active sites in both WPD-loop open and closed conformations, differences
are identified in the molecular details associated with loop movement in PTPs from different organisms.
Ó 2012 Elsevier Inc. All rights reserved.
Keywords:
Phosphatase
Protein-tyrosine phosphatase
Introduction
eral activities, including negative regulation of insulin signaling by
dephosphorylation of the insulin receptor tyrosine kinase [9]. As a
Protein tyrosine phosphatases (PTPs)1 comprise a large family of
enzymes responsible for dephosphorylation of intracellular Tyr resi-
dues, functioning in concert with protein tyrosine kinases (PTKs) to
modulate signal transduction pathways [1–4]. Two of the best-stud-
ied PTPs are the Yersinia protein tyrosine phosphatase, YopH, and the
mammalian protein tyrosine phosphatase 1B (PTP1B). YopH is an
essential virulence factor in the bacteria Yersinia sp., a genus that in-
cludes three species causative of human illness, ranging from gastro-
intestinal disease to Bubonic Plague [5]. YopH is also one of the most
powerful phosphatases, catalyzing the hydrolysis of phosphate
monoester dianions with kcat values [6] of about 1300 sÀ1 for phys-
iological phosphopeptide substrates at its pH optimum [7]. Compar-
ison of this value with the rate constant for the uncatalyzed reaction
of ꢀ10À20 sÀ1 ranks YopH as one of the most efficient enzymes
known [8]. PTP1B is a biomedically important phosphatase with sev-
result, this enzyme has been considered a significant target for treat-
ment of type 2 diabetes and obesity [10,11]. PTP1B also down-regu-
lates cell growth by dephosphorylating the epidermal growth factor
receptor (EGFR) [12]. Overexpression has been observed in human
breast and ovarian cancer, where it is believed to suppress potential
tumors by antagonizing signaling of oncogenic factors [13,14]. De-
spite a similar active site and an identical mechanism, at its pH opti-
mum PTP1B exhibits lower kcat values of ꢀ40 sÀ1 [15].
PTPs catalyze the hydrolysis of tyrosine phosphate esters by the
two-step mechanism shown in Fig. 1 [2,3]. In the first step a nucle-
ophilic cysteine thiolate attacks the substrate, resulting in forma-
tion of
a phosphoenzyme intermediate with release of the
peptidyl tyrosine. The second step involves attack of water on
the phosphoenzyme intermediate and yields inorganic phosphate
and the free enzyme. The central binding site for the phosphoryl
group of the substrate is the P-loop, a signature motif region with
the sequence CX5R that includes the nucleophilic cysteine, a highly
conserved arginine residue, and backbone amide groups oriented
in a semicircular fashion. This arrangement provides hydrogen
bonding to the phosphoryl group of the substrate and transition
state (TS) stabilization.
q
This work was supported by National Institutes of Health Grant GM47297
(A.C.H.).
⇑
Corresponding authors. Fax: +1 435 797 3390.
E-mail addresses: sean.johnson@usu.edu (S.J. Johnson), alvan.hengge@usu.edu
(A.C. Hengge).
Substrate binding in PTPs favors the closure of a flexible loop
bearing the conserved sequence WPD. The aspartic acid in this loop
functions as a general acid in the first step and as a general base
1
Abbreviations used: PTPs, protein tyrosine phosphatases; PTKs, protein tyrosine
kinases; PTP1B, protein tyrosine phosphatase 1B; EGFR, epidermal growth factor
receptor; TS, transition state; KIEs, kinetic isotope effects.
0003-9861/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.abb.2012.06.002

54
T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
Fig. 1. The general mechanism of the PTP-catalyzed reaction. The WPD-loop assumes a catalytically active closed conformation with the general acid in position to protonate
the leaving group during formation of the phosphoenzyme intermediate. In the second step this intermediate is hydrolyzed. After the phosphate product is released the WPD-
loop open conformation becomes favored.
Fig. 2. Orientation of key residues at the active site of PTP1B (light blue) and YopH (green). (a) Superimposition of the ligand-free structures with the WPD-loop in the open
conformation. (b) Superimposition of the vanadate-bound structures with WPD-loop in the closed conformation, using the same orientation as in (a). The lower, italicized
residue positions refer to YopH. For the sake of clarity the only backbone carbonyl group is that of W179/W354 shown in (b). Hydrogen bonds in dotted blue lines are shown
only for PTP1B. The PDB IDs used to generate each structure were: PTP1B wildtype: 2CM2 (ligand-free form, open WPD-loop) [42] and 3I80 (VO4 bound, closed WPD-loop)
[28]; YopH wildtype: 1YPT (ligand-free form, open WPD-loop) [19] and 2I42 (VO4 bound, closed WPD-loop). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
catalyst in the second step (Fig. 1) [16,17]. The WPD-loop has two
distinct conformations, an ‘‘open’’ conformation in which the
WPD-loop has negligible interaction with the P-loop; and a
‘‘closed’’ conformation, where the WPD-loop is folded over the ac-
tive site, bringing the conserved Asp residue up to 8 Å closer to the
bound substrate [18–22]. The loop-open and closed forms of the
active sites of YopH and PTP1B superimpose well with one another
(Fig. 2). The backbone residues of the P-loop align closely in both
the ligand-free and ligand-bound states. Unlike the WPD loop,
there is no evidence for significant motion of the P-loop during
catalysis.
Mechanistic investigations have shown that the overall mecha-
nism and the transition state for the chemical steps are the same
for these two PTPs, as they are for all PTPs that have been exam-
ined [23]. Despite the structural and mechanistic similarities, the
catalytic rates of YopH and PTP1B vary by more than an order of
magnitude, and rates vary even more within other members of
the PTP family. This has led us to inquire whether there are differ-
ences in the molecular details involved in loop movement between
PTPs. Thus, we have compared the effect of the Trp mutation in
PTP1B and YopH. This highly conserved residue is not directly in-
volved in catalysis, but is involved in WPD loop motion. During
loop movement the indole side chain slides within a hydrophobic
crevice. Kinetic data and isotope effects on the reaction with the
substrate pNPP were obtained, and crystal structures for ligand-
free and bound states of Trp to Phe mutant of PTP1B are reported.
Analogous kinetic and structural data for the corresponding mu-
tant of YopH have previously been reported [17,24,25], which al-
lows for direct comparisons with data obtained with PTP1B.
Unexpectedly, the conservative Trp to Phe mutation has signifi-
cantly different effects in the two enzymes.
Materials and methods
The plasmid pEt-19b encoding the 37 kDa form of the human
wildtype PTP1B (amino acid residues 1–321) was provided by Dr.
Nicholas K. Tonks and has been previously reported [26]. Labeled
and natural abundance p-nitrophenyl phosphate (pNPP, dicyclo-
hexylammonium salt) were synthesized according to previous
methods [27]. Sodium orthovanadate (Na3VO4) was purchased
from Fisher ScientificÓ.
Site-directed mutagenesis
Mutations in PTP1B were made by using the QuikChangeÒ site-
directed mutagenesis kit from StrategeneÓ. The oligonucleotide
primers used to insert the desired mutations were as follows:
W179F, ccactataccacattccctgactttggagtccctg; and W179A, ccacta-
taccacagcgcctgactttggagtccctg. The underlined bases refer to the
changes from the naturally occurring nucleotides. All mutations
were verified by DNA sequencing.
Protein expression and purification
PTP1B mutants were expressed in Escherichia coli and purified
to homogeneity as previously reported for the wild-type protein
[28]. All Trp mutations were verified by TOF–MS. Protein

T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
55
concentrations were monitored by UV using an A1mg/mL280nm = 1.10
for both tryptophan PTP1B mutants.
and the respective initial concentrations of ligands present in the
protein and precipitant solutions, and then flash-cooled in liquid
nitrogen.
Kinetics
Reactions were carried out at 23 °C using the substrate pNPP in
the presence of 1 mM DTT in a three-component buffer mixture of
100 mM sodium acetate, 50 mM Bis–Tris, and 50 mM Tris, which
provides a constant ionic strength over a wide pH range. Reactions
were initiated by addition of enzyme (final reaction volume of
Data collection, structure determination and refinement
Diffraction data were collected on a home source (Rigaku Raxis
IV++). Data for the PTP1B W179F ligand-free structure was indexed
and processed using DENZO and SCALEPACK in the HKL2000 pro-
gram suite [31] and PTP1B W179F bound with vanadate was using
d⁄TREK in the program Crystal Clear [32]. Molecular replacement
was performed using Phaser [33] from the CCP4 program suite
[34,35]. The search model was the ligand-free wildtype PTP1B
structure (PDB ID 2CM2) with the active site water molecules re-
moved. Refinement was performed using the program Phenix
[36]. Coot [37] and MolProbity [38] were used for model building
and validation. Structure factors were deposited in the Protein Data
Bank. Table 1 summarizes the crystallographic data, statistics of
structure refinement and PDB accession numbers. All figures
depicting crystallography data were prepared using Pymol [39].
300 lL) and quenched by addition of 30 lL of 10 M NaOH. The
non-enzymatic hydrolysis of the substrate was corrected by mea-
suring the respective reaction in the absence of enzyme. The initial
rate measurements for the enzyme catalyzed hydrolysis of pNPP
were monitored by product (p-nitrophenol) formation, which
was measured from the absorbance at 400 nm using a molar
extinction coefficient of 18,300 MÀ1 cmÀ1. Kinetic parameters were
determined by a fit of the initial rate (v) versus [pNPP] data to the
Michaelis–Menten equation.
Inhibition constants (Ki) for PTP1B by tungstate and molybdate
for the W179F mutant of PTP1B were determined by measuring
initial rates of pNPP hydrolysis as described above in the presence
of a range of concentrations of inhibitor (0–10 Ki) and [pNPP] (0.5–
6 Km) at pH 5.5 and 23 °C. Inhibition constants were obtained from
a global fit, considering that Kmap = Km (1 + [I]/Ki) in the presence of
inhibitors. The Vmax values were unchanged in the presence of
these inhibitors, as expected for competitive behavior.
Table 1
Data collection and refinement statistics for structures of PTP1B W179F ligand-free
and vanadate bound. Values in parentheses correspond to those in the outer
resolution shell.
Kinetic isotope effect determinations
PTP1B W179F
ligand-free
PTP1B W179F
VO4
Kinetic isotope effects (KIEs) were measured using the internal
competition method, and thus are isotope effects on V/K [29]. In
the commonly used notation, a leading superscript of the heavier
isotope is used to indicate the isotope effect on the following ki-
netic quantity; for example 15(V/K) denotes (V/K)14/(V/K)15, the
nitrogen-15 KIE on V/K.
Data collection
Beamline
Wavelength (Å)
Resolution range (Å)
Outer shell (Å)
Home source
1.5418
38.23–2.05
2.12–2.05
Home source
1.5418
38.37–2.20
2.28–2.20
No. of reflections
Unique
Total
Average redundancy
Mean I/(I)
29,808
24,220
Natural abundance pNPP was used for measurements of 15(V/K).
The 18O KIEs 18(V/K)bridge and 18(V/K)nonbridge were measured by the
remote label method, using the nitrogen atom in p-nitrophenol as a
reporter for isotopic fractionation in the labeled oxygen positions
[23]. The isotopic isomers used are shown in Supporting informa-
tion. Isotope effect determinations were made at 25 °C in 50 mM
Bis–Tris buffer, pH 5.5, containing 1 mM DTT. The pNPP concentra-
tion was 18 mM and the reactions were started by addition of
174,588
5.9 (6.1)
15.1 (2.7)
98.7 (97.8)
10.3 (64.9)
P3121
133,515
5.5 (5.6)
7.9 (2.7)
98.7 (97.3)
10.8 (49.4)
P3121
Completeness (%)
Rsym (%)a
Space group
Number of protein molecules per
asymmetric unit
1
1
Unit cell dimensions
PTP1B wild-type or mutant to a final concentration of 0.15 lM.
a, b, c (Å)
88.3, 88.3, 104.5
90.0, 90.0, 120.0
88.6, 88.6,
104.3
90.0, 90.0,
120.0
After reactions reached 40–60% completion (about 1 h.) they were
stopped by titrating to pH 3–4 with HCl. Protocols for isolation of
p-nitrophenol, isotopic analysis and calculation of the isotope ef-
fects were the same as previously described [30], and are described
in Supporting information.
a
, b,
c
(°)
Refinement
Rwork/Rfree (%)b
20.4/24.5
19.6/25.4
Atoms in the structure
Protein
Waters
Ligands/ions
Average B factors (Å2)
Protein
Water
Ligands/ions
Rmsd bond (Å)/angle (°)
2372
239
20
2485
276
31
Crystallization
Crystals were grown using sitting drop vapor diffusion at 4 °C.
38.19
45.48
61.79
0.008/1.104
41.97
49.81
66.04
0.009/1.172
The crystallization drop was prepared by mixing 2
solution, 0.5 L sucrose 30% (w/v) and 3 L of precipitant solution
(0.1 M Hepes pH 7.5, 0.2 M magnesium acetate and 15–20% poly-
ethylene glycol 8000). The well solution was 500 L of precipitant
lL of protein
l
l
c
l
Protein geometry
Ramachandran outliers (%)
Ramachandran favored (%)
Rotamer outliers (%)
PDB ID
0.35
97.6
0.76
3QKP
0.34
98.0
0.37
3QKQ
solution. The protein solution used to obtain the ligand-free crystal
was 12 mg/mL PTP1B W179F in 10 mM Tris pH 7.5, 25 mM NaCl,
0.2 mM EDTA and 3 mM DTT; the vanadate bound crystals were
grown using 15 lL of 12 mg/mL PTP1B W179F and 0.5 lL of
a
Rsym = (
R
|(I-<I>)|)/(
R
I), where <I> is the average intensity of multiple
60 mM of Na3VO4 (1–4.7 eq. mol protein–vanadate ratio). Single
crystals were visible after three days. Cryoprotection was per-
formed by transferring crystals stepwise into stabilization solution
with increasing glycerol amounts to a final concentration of 15%
measurements.
b
Rwork = (R|Fobs–Fcalc|)/(R|Fobs|) and is calculated using all data; Rfree is the R-
factor based on 5% of the data excluded from refinement.
c
Ramachandran statistics were calculated using the MolProbity server [38].

56
T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
Table 3
Results
Inhibition constants by oxyanions for wildtype and tryptophan mutants of PTP1Ba
and YopHb. All measurements were made at 25 °C using pNPP as substrate.
Kinetic properties of the PTP1B W179F and W179A mutants
Reaction
Ki (lM) for tungstate
Ki (lM) for molybdate
Table 2 shows the kinetic data at several pH values, near the re-
ported pH optimum for PTP1B, for hydrolysis of pNPP catalyzed by
PTP1B wildtype and W179 mutants. For comparison, in the lower
section of Table 2 are presented the previously reported kinetic
data for wildtype YopH and its corresponding tryptophan mutants.
The native enzymes exhibit bell shaped pH-profiles for kcat with
pH optima close to 5.5 [15,25]. The acidic limb results from depro-
tonation of the cysteine residue to produce the active thiolate
nucleophile (Cys215 in PTP1B). On the basic limb, enzymatic activ-
ity is reduced at high pH due to loss of general acid catalysis by the
aspartic acid residue (Asp181 in PTP1B). Our interest is focused on
the retention or loss of general acid catalysis by mutation of the
tryptophan in the WPD loop of PTP1B to Phe and Ala. The loss of
general acid catalysis should reduce kcat by orders of magnitude
even with the activated substrate pNPP; moreover, kcat values
would not decrease above the pH optimum due to the change from
bell to a half-bell pH dependency. In contrast, if general acid catal-
ysis is retained, a much smaller effect on kcat should be observed at
the optimum pH, and a monotonic decrease in this value above the
optimum would be expected due to the incorrect protonation state
of the general acid.
The kcat value for the W179F mutant at pH 5.5 is decreased only
ꢀ2-fold in relation to the wildtype enzyme, and kcat decreases as
pH increases (Table 2). This differs from the effect of the corre-
sponding mutation in YopH. The kcat values for YopH W354F are
very similar from pH 5.5 to 7.5, ranging from 3 to 1 sÀ1, and are
about 200-fold lower at pH 5.5 than for the wildtype enzyme.
The activity of PTP1B is more strongly affected by the W179A
mutation, and mirrors the effect of the corresponding mutation
in YopH (Table 2). When compared to the kcat values at pH 5.5
for the native enzymes, the kcat values in PTP1B W179A and YopH
W354A are approximately 320- and 480-fold lower, respectively.
In each case, kcat values are also similar in the pH range of 5.5
and 7.5, which indicates that general acid catalysis is strongly im-
paired by this mutation.
PTP1B wildtype
PTP1B W179F
YopH wildtype
YopH W354F
0.99(7)
0.62(4)
5.8
0.17(2)
0.26(4)
32.0
a
Values in parenthesis are the standard deviations in the last decimal place.
Data from Keng et al. [25] measured at pH 5.5.
b
Kinetic isotope effects
The KIEs for the hydrolysis of pNPP by wildtype PTP1B have
been previously reported [28]. These and the KIE data for the
PTP1B mutants obtained in this work are shown in Table 4. Be-
cause the enzymatic substrate of PTPs is the dianionic form, the
18(V/K)nonbridge isotope effect was corrected for the isotope fraction-
ation for protonation of the nonbridge oxygen atoms, as pNPP is
present as a mixture of monoanion and dianion forms under the
conditions of the experiment. The KIE data in the leaving group
for the W179F mutant are similar to those for the wildtype PTP1B,
with a negligible 15(V/K) KIE and a primary bridge-18(V/K) KIE rang-
ing from 1.2% to 1.4%. For comparison, previously reported KIEs for
wildtype YopH [30] and for YopH mutant [27] corresponding to the
Trp to Phe mutation in PTP1B that were examined in this study are
shown in the lower section of Table 4.
The expected ranges of the isotope effects in pNPP and their
interpretation have been discussed in detail previously [23,40],
and are summarized briefly here. The KIE at the nitro nitrogen
atom, 15(V/K), arises from negative charge development on the
nitrophenolate leaving group. When general acid catalysis main-
tains the leaving group in a neutral state there is no isotope effect
(KIE = unity), while in general acid mutants of PTPs 15(V/K) reaches
its maximum value of about 1.003, reflecting a full negative charge.
The KIE at the bridge oxygen atom, 18(V/K)bridge, arises from P–O
bond fission and is also affected by O–H bond formation by the
general acid. Bond fission produces normal isotope effects, while
bond formation gives rise to inverse effects. The normal effect from
P–O fission is large, and in general acid mutants of PTPs the 18(V/
K)bridge effect is near its maximum of 1.03 reflecting a largely bro-
ken P–O bond in the transition state. In native enzymes where gen-
eral acid catalysis takes place, the observed 18(V/K)bridge is reduced
by protonation, as shown in Table 4.
Table 3 shows the effect of tungstate and molybdate on the
hydrolysis of pNPP catalyzed by PTP1B wildtype and the W179F
mutant. For comparison purposes the lower section of this table
displays reported inhibition data for tungstate for YopH wildtype
and W354F mutant. The activity of PTP1B wildtype and W179F
are inhibited by similar extents in both cases, with inhibition
constants in the micromolar range for both competitive inhibitors.
On the other hand, YopH W354F exhibits reduced affinity (about
5-fold) toward tungstate compared to the wildtype [25]. This
reduced affinity has been attributed to an improper structure
alignment for oxyanion binding, which is not observed by the
corresponding W179F mutation in PTP1B.
The 15(V/K) and 18(V/K)bridge KIEs are the reporters for how gen-
eral acid catalysis might be compromised by mutation. Loss of gen-
eral acid will result in an increase in both of these isotope effects
relative to their values in the native enzymes. The isotope effect
in the nonbridging oxygen atoms, 18(V/K)nonbridge, monitors the
hybridization state of the transferring phosphoryl group. A loose
Table 2
Top, kinetic constants at different pHs for pNPP hydrolysis catalyzed by the wildtype PTP1B and tryptophan 179 mutants at 23 °C.a Below, kinetic constants for the reaction of
b
pNPP by the wildtype and YopH variants with mutations to the corresponding residue.
Reaction
Km (mM) at pHs
kcat (sÀ1) at pHs
5.5
6.5
7.5
5.5
6.5
7.5
PTP1B wildtype
PTP1B W179F
PTP1B W179A
0.58(2)
0.93(5)
4.41(9)
1.48(2)
0.86(3)
3.94(3)
2.20(1)
1.04(5)
2.85(5)
24.4(2)
11.8(2)
0.0750(5)
17.1(1)
2.42(2)
0.0667(1)
3.48(1)
0.151(3)
0.0370(2)
YopH wildtype
YopH W354F
YopH W354A
2.00
4.82
3.85
601
2.96
1.26
95.5
1.42
1.23
11.0
1.10
1.07
a
Values in parenthesis are the standard deviations in the last decimal place.
YopH data were taken from Keng et al. [25] and were measured at 30 °C.
b

T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
57
Table 4
Phe and Trp rings are similar in their positions and orientations.
Other residues at the bottom of the Trp/Phe pocket (i.e. Phe191
and Phe 269) and at the WPD-loop region of PTP1B wildtype and
W179F mutant also exhibit similar conformational changes from
the open to the closed forms. As a whole, the structure of PTP1B
W179F with vanadate bound represents a transition state analog
of the second catalytic step, showing structural characteristics sim-
ilar to those of a vanadate complex of the wildtype enzyme [28].
Top, isotope effects for catalyzed reaction of pNPP by the wildtype and W179F mutant
of PTP1B. Below, isotope effects for the reaction of pNPP by the wildtype and YopH
Trp mutation corresponding to that in PTP1B.a
Reaction
15(V/K)
18(V/K)bridge
18(V/K)nonbridge
Reference
PTP1B wildtypeb
PTP1B W179Fb
YopH wildtypec
YopH W354Fc
1.0004(2)
1.0006(1)
1.0121(9)
1.0140(9)
1.0018(5)
1.0021(7)
[28]
This work
0.9999(3)
1.0013(2)
1.0152(6)
1.0240(10)
0.9998(13)
1.0015(8)
[30]
[27]
a
b
c
Values in parenthesis are the standard errors in the last decimal place.
pH 5.5 and 25 °C.
pH 5.5 and 30 °C.
Discussion
Fig. 2 shows an overlay of the open and closed conformations in
the active sites of the YopH and PTP1B wild-type enzymes. Upon
oxyanion or substrate binding to the P-loop, the conserved Arg res-
idue rotates to make two hydrogen bonds with the oxyanion. In
addition, a new hydrogen bond formed between the guanidinium
group of Arg and the carbonyl oxygen atom of the conserved Trp
assists in stabilizing the WPD loop in the closed, catalytically active
position. All PTPs share these essential aspects of binding and
catalysis. In the following discussion we present our findings as re-
vealed by kinetics, structural analysis, and isotope effects of the
distinctive effect of Trp mutations in the catalysis of these two
enzymes.
The Trp residue (W179 in PTP1B and W354 in YopH) is highly
conserved in the PTP family and forms one of the so-called hinge
residues of the flexible loop. In the superposition of the ligand-free,
loop-open structures of the two enzymes, the respective indole
rings and backbones are adjacent to one another (ꢀ2 Å apart)
and the residues adopt very similar positions in the respective
or metaphosphate-like transition state gives rise to slightly inverse
effects. This isotope effect becomes increasingly normal as the
transition state grows more associative in nature.
Crystallography
We have solved X-ray structures for PTP1B W179F in the pres-
ence and absence of the competitive inhibitor vanadate. Fig. 3
shows the active site of PTP1B W179F. In the crystal grown in
the presence of vanadate, two conformations are observed at
approximately equal occupancy: a WPD loop-open ligand-free
form, and a closed vanadate-bound form (refined to 48% and 52%
occupancy, respectively). A difference density map for vanadate
is shown in Fig. S3 of Supporting information. Conformational dif-
ferences between these two forms are confined to the side chain of
Arg221 and the WPD-loop. The open form observed at 48% occu-
pancy in the crystal grown in the presence of vanadate is superim-
posable with both the ligand-free structure of W179F, as well as
the wildtype PTP1B (PDB ID 2CM2) (see Supporting information
Fig. S2). In the closed form, the bound vanadate exhibits a trigonal
bipyramidal conformation with apical positions occupied by the
Cys215 sulfur and an oxygen atom from a water molecule. The
three equatorial oxygen atoms hydrogen bond to main chain NH
amides in the P-loop and to the guanidinium moiety of Arg221.
Compared to the open form, the side chain of Arg221 is rotated
about 100° around chi-3 and hydrogen bonds to the carbonyl
group of Phe179. The side chain of Phe179 is embedded into a
hydrophobic pocket in both the open and closed forms, with the
Phe methylene groups of these two forms ꢀ1.4 Å apart. Similarly,
the respective distance between the Trp methylene groups in the
open and closed forms of the wildtype enzyme is ꢀ1.2 Å, and the
loop-closed structures (see Fig.
structures).
2 for an overlap of these
A previous study revealed that the conservative W354F muta-
tion in YopH results in a decrease in kcat by two orders of magni-
tude, loss of the basic limb of the pH-rate profile, and kinetic
isotope effects indicative of the leaving group departing as an an-
ion [25,27]. These data indicate the loss of general acid catalysis
in the mutant. Subsequent structural analyses in the presence
and absence of bound ligands showed that the WPD loop is fixed
in a quasi-open position by the W354F mutation, leaving the Asp
356 side chain in a position unproductive for catalysis [24].
In contrast, the corresponding W179F mutant of PTP1B has
minor effects on kcat (about 2-fold decrease at pH 5.5) compared
to wild-type (Table 2). The KIE data are similar to those of the wild-
type enzyme (Table 4), showing that general acid catalysis remains
effective. The ability of PTP1B to bind the competitive inhibitors
tungstate and molybdate is also not affected by the W179F muta-
tion (Table 3). In contrast, the corresponding mutation in YopH re-
duces the binding affinity for tungstate by about 6-fold at pH 5.5
[25].
The crystal structure of the W179F mutant of PTP1B shows both
loop open and closed forms, in contrast to structures of W354F
YopH. This clearly demonstrates the ability of the WPD loop in
W179F PTP1B to attain both the open and fully closed conforma-
tions. In light of the unexpected difference in the effect of this
mutation in the two enzymes, we closely examined these struc-
tures looking for differences that might explain the retention of full
loop movement in PTP1B W179F, while the loop of the correspond-
ing YopH W354F mutant locks in an unproductive position. In both
wildtype enzymes the indole ring of Trp is embedded in a hydro-
phobic pocket and undergoes a repositioning upon loop closure.
Interestingly, a comparison of the loop open and closed crystal
structures reveals a difference in the movement of the Trp side
chains (Fig. 4). In the PTP1B wildtype and W179F mutant, the
Trp/Phe side chains slide in the same direction as WPD-loop move-
ment. The Phe side chain in the W179F mutant encounters no
Fig. 3. Orientation of key residues at the active-site of PTP1B W179F in unbound,
loop-open (green) and vanadate-bound, loop-closed (light blue) states. (For
interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)

58
T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
Fig. 4. Comparison between PTP1B and YopH structures. Hydrophobic pocket of the conserved tryptophan in the hinge of the WPD-loop, and orientations of the Trp in the
native and Phe in the mutant PTPs in open and closed WPD-loop forms. The arrows indicate the movement direction of the Trp or Phe upon WPD-loop closure. These pictures
were made from PTP1B and YopH structures superimposed considering the P-loop region, no rotation was applied between each representation. The hydrophobic pockets for
all representations are those for the closed or quasi-open (YopH W354F) WPD-loop conformations. The PDB IDs used to generate each structure were: (a) PTP1B wildtype:
2CM2 (ligand-free form, open WPD-loop) [42], 3I80 (VO4 bound, closed WPD-loop) [28]; (b) PTP1B W179F: 3QKP (ligand-free form, open WPD-loop), 3QKQ (VO4 bound,
closed WPD-loop); (c) YopH wildtype: 1YPT (ligand-free form, open WPD-loop) [19], 2I42 (VO4 bound, closed WPD-loop); (d) YopH W354F: 3F99 (ligand-free form, quasi-
open WPD-loop), 3F9A (WO4 bound, quasi-open WPD-loop) [24].
steric clashes upon full loop closure. In contrast, in the YopH
wildtype the Trp side chain slides in a direction opposite to WPD
loop movement. We have previously proposed that the locked
quasi-open conformation in the W354F mutant results from
steric clashes that would arise between the Phe side chain and
important residues (e.g. Pro355 and Thr358) in a fully closed-loop
conformation. In fact, the position of Phe354 in the quasi-open
loop of W354F YopH closely matches that of Trp354 in the
open state of the wildtype enzyme, and shows little difference
between the structures of the ligand-free and bound states of YopH
W354F [24].
This interaction, along with a hydrogen bond between the Pro355
carbonyl and Thr358 amide groups, is probably important for sta-
bility of the WPD-loop in the closed position.
The importance of an aromatic hydrophobic side chain in the
Trp179 position is shown by the significantly reduced kcat of the
W179A mutant, which is 325-fold slower than the native enzyme
at pH 5.5. YopH also shows a strong response to the corresponding
mutation; the W354A mutant is 476-fold slower than native YopH
under similar conditions [25]. The similar kcat values between pH
5.5 and 6.5 [25] and large isotope effects in the leaving group of
pNPP [27] indicate the loss of general acid catalysis in W354A
YopH. The W179A mutation in PTP1B causes similar effects on its
pH dependency (Table 2), consistent with an impaired general acid
catalysis. Catalysis was too slow to permit measurement of kinetic
isotope effects with this mutant.
The significantly different responses to the conservative Trp to
Phe mutation in YopH compared to PTP1B most likely can be
attributed to the opposite direction of motion of this residue’s side
chain during loop closure. In order to accommodate the new posi-
tion of the Trp/Phe side chain, other side chains at the pocket (i.e.
Phe191 and Phe 269) undergo movement in PTP1B upon WPD-loop
closure. This is not observed in YopH, and the relatively tight Trp
pocket in this enzyme is less suitable for the movement of the
Phe side chain. This implies that the Trp/Phe pocket in PTP1B
exhibits more plasticity than the corresponding pocket in YopH.
It is likely that the Trp repositioning upon WPD-loop closure
participates in control of dynamic events and catalysis in PTPs.
Although referred to as the WPD-loop, this flexible region in PTPs
consists of a larger segment of approximately 10–12 amino acids.
The majority of vertebrate PTPs present two highly conserved pro-
line residues in this loop [41]. On one side of the WPD-loop, the
Pro180 (residue numbering for PTP1B) is essential for maintenance
of the loop secondary structure, while at the other side the Pro185
forms the upper lid of the Trp pocket and it is involved in the Trp/
Phe repositioning. It is interesting to observe that the second
highly conserved Pro in this position of human PTPs is not a con-
sensus for bacterial PTPs (a sequence alignment of this region is
shown in Fig. S4 of Supporting information). The corresponding po-
sition of Pro185 in PTP1B is occupied by Val360 in YopH. In addi-
tion to the interplay between side chains of Pro185 and Trp179,
the NH indole moiety of the Trp side chain has polar interactions
with the main chain carbonyl group of Gly183 in PTP1B; the same
interaction is observed with the corresponding residue Thr358 in
YopH. The Gly183 is conserved in several vertebrate PTPs, [41]
and the Thr358 is found only in bacterial PTPs. Although this Thr
is not conserved among bacterial PTPs, it is surely an important
residue in YopH. The hydroxyl moiety of Thr358 hydrogen bonds
to the carbonyl moiety of Pro355 at the other side of the WPD-loop.
Conclusions
Although WPD-loop movement is a shared feature of catalysis
in the PTP family, the Trp to Phe mutation in the conserved
WPD-loop has distinctly different results in PTP1B compared with
YopH. The Trp to Phe mutation in the WPD loop of YopH locks the
loop in a quasi-open position, compromising general acid catalysis;
in contrast, PTP1B functions only slightly less efficiently with this
mutation and general acid catalysis is unaffected. Crystal struc-
tures reveal that in YopH the indole side chain of this residue
moves in a direction opposite to WPD loop movement, while in
PTP1B the indole and loop move in the same direction. The former
case results in steric clashes that cause the loop of the Phe mutant
to assume a static position that does not respond to oxyanion
binding.
These results show that some molecular details of loop move-
ment are different in YopH and PTP1B. Structural analysis shows
that this difference arises from the opposite direction of movement
of the indole side chain as the WPD-loop oscillates between the
open and closed positions in the two enzymes. Other differences
in neighboring residues in eukaryotic and bacterial sources that
form the hydrophobic pocket in which the indole moiety moves
also account for differences in the mechanical working of this flex-
ible loop in the PTP family. Presumably, such variations exist in
other members of the PTP family as well, and may contribute to
variations in catalytic rates. A more complete understanding of
the causes underlying the apparent dynamical differences in the

T.A.S. Brandão et al. / Archives of Biochemistry and Biophysics 525 (2012) 53–59
59
[16] Z.Y. Zhang, Current Topics in Cellular Regulation 35 (1997) 21–68.
human and the bacterial enzymes may come from solution NMR
studies that are underway.
It will be an interesting area of future study to address the ex-
tent to which loop movement correlates with catalytic rates in
PTPs and whether or not this might explain the variations in cata-
lytic efficiency within this enzyme family.
[17] Z.Y. Zhang, Y.A. Wang, J.E. Dixon, Proceedings of the National Academy of
Sciences of the United States of America 91 (1994) 1624–1627.
[18] D. Barford, A.J. Flint, N.K. Tonks, Faseb Journal 8 (1994) A1322.
[19] J.A. Stuckey, H.L. Schubert, E.B. Fauman, Z.Y. Zhang, J.E. Dixon, M.A. Saper,
Nature 370 (1994) 571–575.
[20] E.B. Fauman, C. Yuvaniyama, H.L. Schubert, J.A. Stuckey, M.A. Saper, Journal of
Biological Chemistry 271 (1996) 18780–18788.
[21] M.R. Groves, Z.J. Yao, P.P. Roller, T.R. Burke, D. Barford, Biochemistry 37 (1998)
17773–17783.
[22] H.L. Schubert, E.B. Fauman, J.A. Stuckey, J.E. Dixon, M.A. Saper, Protein Science
4 (1995) 1904–1913.
Acknowledgments
We thank Nick Tonks for providing the plasmid encoding native
PTP1B. ACH expresses thanks to John Denu, in whose laboratory
some of the initial experiments on this project were carried out
during a sabbatical. We are also grateful to CAPES (Brazil) for a Fel-
lowship to T.A.S. Brandão. This research was supported by a Grant
from the National Institutes of Health (GM47297).
[23] A.C. Hengge, Accounts of Chemical Research 35 (2002) 105–112.
[24] T.A.S. Brandao, H. Robinson, S.J. Johnson, A.C. Hengge, Journal of the American
Chemical Society 131 (2009) 778–786.
[25] Y.F. Keng, L. Wu, Z.Y. Zhang, European Journal of Biochemistry 259 (1999)
809–814.
[26] D. Barford, J.C. Keller, A.J. Flint, N.K. Tonks, Journal of Molecular Biology 239
(1994) 726–730.
[27] R.H. Hoff, A.C. Hengge, L. Wu, Y.F. Keng, Z.-Y. Zhang, Biochemistry 39 (2000)
46–54.
[28] T.A.S. Brandao, A.C. Hengge, S.J. Johnson, Journal of Biological Chemistry 285
(2010) 15874–15883.
Appendix A. Supplementary data
[29] W.W. Cleland, Enzyme mechanisms from isotope effects, in: A. Kohen, H.-H.
Limbach (Eds.), Isotope Effects in Chemistry and Biology, CRC Press, Boca
Raton, FL, 2006, pp. 915–930.
[30] A.C. Hengge, G.A. Sowa, L. Wu, Z.-Y. Zhang, Biochemistry 34 (1995) 13982–
13987.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.abb.2012.06.002.
References
[31] Z. Otwinowski, W. Minor, Macromolecular Crystallography, Part A 276 (1997)
307–326.
[32] Rigaku/MSC, CrystalClear: An Integrated Program for the Collection and
Processing of Area Detector Data, 2006.
[33] A.J. Mccoy, R.W. Grosse-Kunstleve, P.D. Adams, M.D. Winn, L.C. Storoni, R.J.
Read, Journal of Applied Crystallography 40 (2007) 658–674.
[34] S. Bailey, Acta Crystallographica Section D-Biological Crystallography 50
(1994) 760–763.
[35] E. Potterton, P. Briggs, M. Turkenburg, E. Dodson, Acta Crystallographica
Section D-Biological Crystallography 59 (2003) 1131–1137.
[36] P.D. Adams, R.W. Grosse-Kunstleve, L.W. Hung, T.R. Ioerger, A.J. McCoy, N.W.
Moriarty, R.J. Read, J.C. Sacchettini, N.K. Sauter, T.C. Terwilliger, Acta
Crystallographica Section D-Biological Crystallography 58 (2002) 1948–1954.
[37] P. Emsley, K. Cowtan, Acta Crystallographica Section D-Biological
Crystallography 60 (2004) 2126–2132.
[38] I.W. Davis, A. Leaver-Fay, V.B. Chen, J.N. Block, G.J. Kapral, X. Wang, L.W.
Murray, W.B. Arendall 3rd, J. Snoeyink, J.S. Richardson, D.C. Richardson,
Nucleic Acids Research 35 (2007) W375–W383.
[39] W.L. DeLano, The PyMOL Molecular Graphics System, DeLano Scientific LLC,
Palo Alto, CA, USA, 2007.
[40] A.C. Hengge, FEBS Letters 501 (2001) 99–102.
[41] J.N. Andersen, O.H. Mortensen, G.H. Peters, P.G. Drake, L.F. Iversen, O.H. Olsen,
P.G. Jansen, H.S. Andersen, N.K. Tonks, N.P.H. Moller, Molecular and Cellular
Biology 21 (2001) 7117–7136.
[42] P.J. Ala, L. Gonneville, M.C. Hillman, M. Becker-Pasha, M. Wei, B.G. Reid, R.
Klabe, E.W. Yue, B. Wayland, B. Douty, P. Polam, Z. Wasserman, M. Bower, A.P.
Combs, T.C. Burn, G.F. Hollis, R. Wynn, Journal of Biological Chemistry 281
(2006) 32784–32795.
[1] T. Hunter, Cell 80 (1995) 225–236.
[2] M.D. Jackson, J.M. Denu, Chemical Reviews 101 (2001) 2313–2340.
[3] Z.-Y. Zhang, Critical Reviews in Biochemistry and Molecular Biology 33 (1998)
1–52.
[4] A. Alonso, J. Sasin, N. Bottini, I. Friedberg, I. Friedberg, A. Osterman, A. Godzik, T.
Hunter, J. Dixon, T. Mustelin, Cell 117 (2004) 699–711.
[5] R.R. Brubaker, Clinical Microbiology Reviews 4 (1991) 309–324.
[6] Z.Y. Zhang, J.C. Clemens, H.L. Schubert, J.A. Stuckey, M.W.F. Fischer, D.M. Hume,
M.A. Saper, J.E. Dixon, Journal of Biological Chemistry 267 (1992) 23759–
23766.
[7] Z.Y. Zhang, A.M. Thiemesefler, D. Maclean, D.J. Mcnamara, E.M. Dobrusin, T.K.
Sawyer, J.E. Dixon, Proceedings of the National Academy of Sciences of the
United States of America 90 (1993) 4446–4450.
[8] C. Lad, N.H. Williams, R. Wolfenden, Proceedings of the National Academy of
Sciences of the United States of America 100 (2003) 5607–5610.
[9] T. Hunter, Cell 100 (2000) 113–127.
[10] S. Zhang, Z.Y. Zhang, Drug Discovery Today 12 (2007) 373–381.
[11] T.O. Johnson, J. Ermolieff, M.R. Jirousek, Nature Reviews Drug Discovery 1
(2002) 696–709.
[12] F.G. Haj, B. Markova, L.D. Klaman, F.D. Bohmer, B.G. Neel, Journal of Biological
Chemistry 278 (2003) 739–744.
[13] Z.Y. Zhang, Annual Review of Pharmacology and Toxicology 42 (2002) 209–
234.
[14] A. Ostman, C. Hellberg, F.D. Bohmer, Nature Reviews Cancer 6 (2006) 307–320.
[15] A.K. Pedersen, X.L. Guo, K.B. Moller, G.H. Peters, H.S. Andersen, J.S. Kastrup, S.B.
Mortensen, L.F. Iversen, Z.Y. Zhang, N.P.H. Moller, Biochemical Journal 378
(2004) 421–433.