Determination of separate inhibitor and substrate binding sites in the dehaloperoxidase-hemoglobin from Amphitrite ornata

Article abstract of DOI:10.1021/bi9018576

Dehaloperoxidase-hemoglobin (DHP A) is a dual function protein found in the terrebellid polychaete Amphitrite ornata. A. ornata is an annelid, which inhabits estuary mudflats with other polychaetes that secrete a range of toxic brominated phenols. DHPA is capable of binding and oxidatively dehalogenating some of these compounds. DHP A possesses the ability to bind halophenols in a distinct, internal distal binding pocket. Since its discovery, the distal binding pocket has been reported as the sole binding location for halophenols; however, data herein suggest a distinction between inhibitor (monohalogenated phenol) and substrate (trihalogenated phenol) binding locations. Backbone ¹³Cα, ¹³Cβ, carbonyl ¹³C, amide ¹H, and amide ¹⁵N resonance assignments have been made, and various halophenols were titrated into the protein. ¹H- ¹⁵N HSQC experiments were collected at stoichiometric intervals during each titration, and binding locations specific for mono- and trihalogenated phenols have been identified. Titration of monohalogenated phenol induced primary changes around the distal binding pocket, while introduction of trihalogenated phenols created alterations of the distal histidine and the local area surrounding W120, a structural region that corresponds to a possible dimer interface region recently observed in X-ray crystal structures of DHP A. 2010 American Chemical Society.

Full text of DOI:10.1021/bi9018576

Biochemistry 2010, 49, 1199–1206 1199
DOI: 10.1021/bi9018576
Determination of Separate Inhibitor and Substrate Binding Sites in the
†,‡
Dehaloperoxidase-Hemoglobin from Amphitrite ornata
§
,#
,^
,§
Michael F. Davis, Benjamin G. Bobay, and Stefan Franzen*
§
Departments of Chemistry, and Molecular and Structural Biochemistry, North Carolina State University, Raleigh, North Carolina
2
Cancer Center and Department of Biochemistry and Biophysics, University of North Carolina, Chapel Hill, NC 27599.
^
#
7606, and North Carolina Research Campus, Kannapolis, North Carolina 28081. Present address: Lineberger Comprehensive
Received October 29, 2009; Revised Manuscript Received January 11, 2010
ABSTRACT: Dehaloperoxidase-hemoglobin (DHP A) is a dual function protein found in the terrebellid
polychaete Amphitrite ornata. A. ornata is an annelid, which inhabits estuary mudflats with other polychaetes
that secrete a range of toxic brominated phenols. DHP A is capable of binding and oxidatively dehalogenating
some of these compounds. DHP A possesses the ability to bind halophenols in a distinct, internal distal
binding pocket. Since its discovery, the distal binding pocket has been reported as the sole binding location
for halophenols; however, data herein suggest a distinction between inhibitor (monohalogenated phenol) and
1
3
13
13
1
substrate (trihalogenated phenol) binding locations. Backbone CR, Cβ, carbonyl C, amide H, and amide
15
1
15
N resonance assignments have been made, and various halophenols were titrated into the protein. H- N
HSQC experiments were collected at stoichiometric intervals during each titration, and binding locations
specific for mono- and trihalogenated phenols have been identified. Titration of monohalogenated phenol
induced primary changes around the distal binding pocket, while introduction of trihalogenated phenols
created alterations of the distal histidine and the local area surrounding W120, a structural region that
corresponds to a possible dimer interface region recently observed in X-ray crystal structures of DHP A.
1
The dehaloperoxidase-hemoglobin (DHP A) is a globin with
2,4,6-trihalophenols, bind to an external active site (13, 15). In
addition to inhibition of peroxidase activity, a bound 4-halophe-
nol in the distal cavity appears to interfere with the inherent role
of globins in oxygen transport and NO scavenging (11, 12,
15, 16). The role of 2,4,6-trihalophenol substrate binding is even
more elusive since it is not directly observed in the X-ray crystal
structure. Thus, there is a need for solution studies of both
substrate and inhibitor binding to clarify the structural conse-
quences of each binding event with DHP A.
A. ornata coinhabits estuary mudflats with other filter feeding
marine polychaetes that secrete halophenols presumably for
territorial protection (17-19). One of the coinhabitants, Noto-
mastus lobatus, produces and excretes 4-bromophenol (4-BP),
2,4-dibromophenol (2,4-DBP), and 2,4,6-tribromophenol (2,4,6-
TBP) to the surrounding sediment with a stoichiometric ratio of
1.8:0.9:1.0 (20). Brominated phenols are extremely toxic, parti-
cularly for aquatic organisms. The ability of DHP A to bind and
oxidatively dehalogenate halophenols may therefore play an
important role in the survival of A. ornata. Despite the globin
fold, DHP A readily oxidizes various trihalogenated phenols, as
illustrated in Scheme 1, with a total yield similar to that of
horseradish peroxidase (HRP) (8-10, 21). The oxidation rate,
however, is ∼13 times less than HRP at pH 5 (8). Since HRP is a
secretory peroxidase, it is optimized to function at pH 5. On the
other hand, DHP A is optimized to function at pH 7.5, which is
near the pH of cytosol. This value is consistent with the role of
DHP A as a coelomic hemoglobin.
peroxidase activity found in the terebellid polychaete Amphitrite
ornata. DHP A is one of two known proteins in A. ornata which
contain the globin fold. Two isoforms of the dehaloperoxidase-
hemoglobin are found in A. ornata, DHP A and DHP B (1). DHP
B differs from DHP A by only five mutations (I9L, R32K, Y34N,
N81S, and S91G) and has recently been cloned and expressed
in Escherichia coli (2, 3). Even though DHP A and DHP B are
structurally homologous to typical myoglobins (Mbs) (4-6),
DHP A was first isolated from A. ornata and characterized as
a peroxidase (7). DHP A was later cloned into E. coli, expressed,
and characterized as a function of pH (8, 9). Site-directed
mutations of the distal and proximal histidines, H55 and H89,
respectively, have shown their essential nature in the catalytic
mechanism of DHP A (10).
It has been known for more than a decade that DHP A has
a unique ability to bind halogenated phenols in a distal poc-
ket above the heme plane (6, 11-14). We have recently demon-
strated this internal binding site to be an inhibitor binding
site favored by 4-halophenols, while the most active substrates,
†
This project was supported by Army Research Office Grant 52278-
LS.
‡
The DHPCN A magnetic coordinates have been deposited in the
Biological Magnetic Resonance Data Bank (accession number 16401).
To whom correspondence should be addressed. Phone: 919-515-
915. Fax: 919-515-8920. E-mail: Stefan_Franzen@ncsu.edu.
*
8
1
Abbreviations: 4-BP, 4-bromophenol; 2,4,6-TBP, 2,4,6-tribromo-
phenol; 2,4,6-TCP, 2,4,6-trichlorophenol; 2,4,6-TFP, 2,4,6-trifluoro-
phenol; 2,6-DCQ, 2,6-dichloroquinone; Amp, ampicillin; CcP,
cytochrome c peroxidase; DHP A, dehaloperoxidase-hemoglobin;
DHPCN A, metcyano dehaloperoxidase-hemoglobin; HSQC, hetero-
nuclear single-quantum coherence; HRP, horseradish peroxidase;
The internal binding pocket was initially thought to be the
active site of DHP A, which was a reasonable assumption in
the absence of detailed kinetic data and a structure of the exter-
nal binding site (6). While the reactivity of DHP A toward
IPTG, isopropyl β-D-thiogalactopyranoside.
r 2010 American Chemical Society
Published on Web 01/12/2010
pubs.acs.org/Biochemistry

1
200 Biochemistry, Vol. 49, No. 6, 2010
Davis et al.
order to determine the conformation of the protein in response to
substrate binding. The low water solubility of 2,4,6-trihaloge-
nated phenols and apparent high KD values make them poor
choices for structural studies such as X-ray crystallography
or NMR experiments. Of the active 2,4,6-trihalophenols only
2
(
,4,6-trifluorophenol (2,4,6-TFP) and 2,4,6-trichlorophenol
2,4,6-TCP) have solubility greater than micromolar concentra-
tions. 2,4,6-TCP is the closest analogue to the native substrate
,4,6-TBP and is the best laboratory substrate due to its simi-
lar size and turnover rates. The most soluble trihalophenol,
,4,6-TFP, has been used in cryogenic FT-IR, EPR, and
HYSCORE studies to show binding at cryogenic temperature
<260 K) but not at room temperature (12, 23). Above 260 K the
,4,6-TFP substrate leaves the distal binding pocket, and no
2
2
(
2
F
IGURE 1: X-ray structure of the internal binding pocket of DHP A.
The 1EWA X-ray structure illustrates binding of monohalogenated
-IP in the distal pocket (14). The closest residues to bound 4-IP are
19
other internal perturbations are observed (12). F NMR and
relaxation experiments provided initial evidence of an exterior
binding site for 2,4,6-TFP at ambient temperatures (13). How-
ever, the exact location of 2,4,6-TFP binding has not been
reported. Moreover, the use of 2,4,6-TFP as a substrate analogue
may be a poor choice due to the relative size difference between
fluorine and bromine or chlorine substituents.
4
F35, V59, and F21, respectively.
Scheme 1: DHP A Catalyzes the Oxidative Dehalogenation of
Trihalophenol to the Corresponding Dihalogenated Quinone
a
2
Product in the Presence of H O
2
1
3
13
In this report, assignments of the backbone CR, Cβ,
1
3
1
15
carbonyl C, amide H ,and amide N resonances have been
1
3
15
made using C/ N labeling and 3D NMR techniques. Para-
magnetic low-spin metcyano DHP A was used in order to retain
continuity with previous NMR investigations. Assignment of
9
2% of the backbone resonances allowed for observations of
a
The scheme represents the reaction at pH 6 where the substrate
is in the phenol form.
structural changes occurring throughout the protein upon intro-
duction of various substrates or inhibitors. Titrations of inhibitor
4
-BP and substrates 2,4,6-TCP/2,4,6-TFP were performed and
2
,4,6-trihalophenols is common in many peroxidases, the internal
1
15
monitored using H- N HSQC experiments. Chemical shift
deviations of backbone amide protons were observed for each
substrate/inhibitor. The resonances exhibiting the greatest degree
of change were then mapped onto current X-ray structures of the
protein, and the local regions experiencing the largest deviations
were analyzed.
inhibitor binding pocket separates DHP A from any known
globin or peroxidase. In fact, many of our early papers were
based on this assumption that the internal binding pocket was the
true substrate binding site. This led to a great deal of speculation
about whether a one-electron or two-electron oxidation takes
place in DHP A (8, 9). However, the realization that an external
site is the substrate binding site makes DHP A similar to other
peroxidases and strongly suggests a one-electron oxidation
mechanism. Inhibition of peroxidase activity, due to inhibitor
binding in the distal pocket, is attributable to two main factors.
The X-ray structure shown in Figure 1 illustrates bound
MATERIALS AND METHODS
Protein Expression and Purification. The highly expres-
sing His-DHP A 4R gene was inserted into the pET-16b vector
and transformed into Rosetta (DE3) E. coli cells. The Rosetta
cells were used for their slower growth cycle and additional
chloramphenicol (Cam) resistance, not to supply tRNAs for
poorly expressed E. coli codons. The cells were plated onto LB
agar plates containing 100 μg/mL Amp and 34 μg/mL Cam and
allowed to grow for ∼18 h. Single colonies were isolated, and
4-halophenol in the distal pocket (14). The close binding proxi-
mity of the inhibitor to the heme iron impedes access of H O to
the heme iron. H O access is essential for formation of com-
2
2
2
2
pound ES and subsequent peroxidase activity of the protein (22).
Consequently, the steric interference of 4-halophenols provides
one mechanism for inhibition. Second, the distal histidine is
forced into a solvent-exposed position by the inhibitor. This
conformational change results in DHP A inhibition because the
distal histidine is the acid-base catalyst required for formation of
the active compound ES.
2
mL starter growths in LB broth (100 μg/mL Amp, 34 μg/mL
Cam) were allowed to grow at 37 ꢀC overnight. One milliliter of
the starter growth was then added to 1 L of Spectra 9-CN
1
3
15
(
Spectra Stable Isotopes, Inc.) medium with >98% C and N
isotopes in the aforementioned Amp and Cam concentrations.
At present, one of the major questions in DHP A function
regards the structure of the true substrate binding site. Substrate
binding is not observed in any of the X-ray crystal structures.
While we can observe the binding of 4-iodo-, 4-bromo-, 4-chloro-,
and 4-fluorophenol in the distal pocket, attempts to infuse
The 1 L double-labeled growth was allowed to grow at 37 ꢀC
with shaking until an OD600 of 0.3 was reached. Isopropyl β-D-1-
thiogalactopyranoside (IPTG) was added to a final concentra-
tion of 0.5 mM, and the growth temperature was reduced to
25 ꢀC. The cells were then allowed to grow for 20 h before being
collected via centrifugation. Cell lysis was performed as described
previously (13) except that the cleared lysate was incubated with
∼100 mg of hemin chloride dissolved in 1 mL of 0.2 M NaOH for
8 h with stirring at 4 ꢀC. The protein was purified using affinity
and ion-exchange chromatography as described elsewhere (13).
2,4,6-trihalophenols into crystals have not resulted in observed
binding in a single X-ray crystal structure (15). We have treated
the crystals with saturated solutions of 2,4,6-trihalophenols
(
substrates) exactly as we have done for 4-halophenols (inhi-
bitors). Consequently, we turn to solution studies of DHP A in

Article
Biochemistry, Vol. 49, No. 6, 2010 1201
FIGURE 2: UV-vis enzymatic assay illustrating 2,4,6-TCP and 4-BP activity. The differences in activity between 2,4,6-TCP (left) and 4-BP (right)
are shown. The figures represent a 600 s assay going from red (0 s) to blue (600 s). The 2,4,6-TCP molecule is completely converted to product
2
,6-dichloro-1,4-benzoquinone (left) while the 4-BP molecule does not experience turnover (right).
1
15
1
15
F
(
IGURE 3: H- N HSQC of DHPCN A collected at 298 K. The H- N HSQC shows minimal overlap, and the crowded area of the spectrum
circled region) is shown in the inset for clarity. Backbone amide H resonances are labeled.
1
13
15
Final yields of C/ N-labeled DHP A were 24.5 mg/L using this
protocol with A406/A280 ratios greater than 4.
RESULTS
Figure 2 illustrates the differences in activity between the
monohalogenated 4-BP and trihalogenated 2,4,6-TCP. Upon
addition of H O , complete conversion of 2,4,6-TCP to 2,6-DCQ
Sample Preparation and NMR Spectroscopy. All NMR
samples were prepared using 90% H O/10% D O and 100 mM
2
2
2
2
potassium phosphate buffer, pH 7. The protein was concentrated
to 1.6 mM for all 3D NMR experiments, and KCN was added
between 5ꢀ and 10ꢀ excess. All NMR experiments were
collected on a Varian Inova 600 MHz spectrometer equipped
is observed over the course of the assay (red = 0 s; blue = 600 s).
The oxidation of 2,4,6-TCP can be seen by the decrease in
absorption at 313 nm and the concomitant appearance of the
2
,6-DCP product band at 273 nm. 4-BP, on the other hand, has
1
13 15
with a Varian cryogenic H/ C/ N triple resonance probe. For
1
15
been shown to be an efficient inhibitor of DHP A activity (15).
the 4-BP and 2,4,6-TFP titrations, a H- N HSQC experiment
was collected at 0:1, 1:1, 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 10:1, 12:1,
and 15:1 halophenol to protein ratios. Only ratios of 0:1, 1:1, 2:1,
Upon addition of H O the 4-BP absorption band at 280 nm is
2
2
not diminished, and during the 600s assay no detectable product
band is observed.
3:1, and 4:1 were feasible for the 2,4,6-TCP titration due to its
1
5
1
Using N- H HSQC, HNCO, HCACOCANH, HNCOCA,
poor water solubility. All NMR experiments were conducted
at 298 K, processed via NMRPipe (24), and analyzed using
Sparky (25).
13
13
HNCA, and HNCACB, ∼92% of the backbone CR, Cβ,
1
3
1
15
carbonyl C, amide H, and N resonances were assigned in
1
5
1
metcyano DHP A. The quality of the N- H HSQC spectrum
Figure 3) shows good peak dispersion with minimal peak
UV-Vis Enzymatic Assays. For all experiments DHP A
was oxidized via addition of excess K Fe(CN) . The excess
(
3
6
overlap, indicating a well-folded stable protein. The inset shows
ferricyanide was removed, and DHP A was buffer exchanged
into 100 mM potassium phosphate, pH 7, using a Sephadex G-25
column. All absorption data were collected on a Hewlett-Packard
1
the crowded region between 9.2 and 8.2 ppm ( H) for clarity. The
1
15
1
cross-peaks at 11.8 ( H), 133.0 ppm ( N) and 11.2 ( H), 118.5
15
ppm ( N) arise from the nonbackbone W120 NεH and H55
NεH side chain resonances, respectively. DHP A contains one
Trp (W120) and two His residues (H89, H55). The two His
residues are the conserved proximal (H89) and distal (H55)
histidines and are both susceptible to the paramagnetic effects
of the CN-ligated heme iron. The exchangeable NεH of H89
exhibits considerable hyperfine shifting and was assigned in our
8
5
453 multiwavelength spectrometer. Spectra were collected every
s over a 600 s time frame. The conditions used for the assays
were ∼2.5 μM DHP A, 120 μM H O , and 120 μM substrate
2
2
or inhibitor. Turnover of the halophenols was determined by
the disappearance of its respective absorption bands at pH 7
(
4-BP, 280 nm; 2,4,6-TCP, 313 nm) and appearance of product
absorption bands (2,6-dichloro-1,4-benzoquinone (2,6-DCQ),
72 nm).
1
1
2
previous work at 19.9 ppm ( H) (13). In the H NMR spectrum of

1
202 Biochemistry, Vol. 49, No. 6, 2010
Davis et al.
1
15
1
F
IGURE 4: H- N HSQC of DHPCN A titrated with 4-BP (left column) and 2,4,6-TCP (right column). The amide H resonances which exhibit
the greatest change in chemical shift upon addition of 4-BP are shown. Upon addition of 2,4,6-TCP these same resonances show no observable
deviations.
a H55 V mutant, the hyperfine-shifted exchangeable at 11.2 ppm
observed in this local area upon addition of substrate 2,4,6-TCP.
In addition to deviations in the distal pocket, 4-BP also induces
changes in the local area near the dimer interface. These include
perturbations of R122 and G1. Together, the internal binding
pocket and dimer interface comprise 80% of the ten largest
chemical shift changes in the presence of 4-BP.
(
(
H55 NεH) disappears while the other exchangeable at 11.8 ppm
W120 NεH) remains unchanged, confirming assignment of these
resonances.
Amide proton chemical shift deviations upon introduction of
inhibitor 4-BP and substrates 2,4,6-TCP and 2,4,6-TFP were
quantified (eq 1) via a weighted chemical shift formula to account
The largest chemical shift deviations induced by the substrate
2,4,6-TCP occur in regions distinct from the internal binding
pocket. The R122 amide proton and exchangeable side chain
H55 NεH proton experience the largest degree of change in the
presence of substrate 2,4,6-TCP. G1, S129, L76, Y34, E45, L100,
D116, and N126 respectively experience the next eight largest
chemical shift alterations when 2,4,6-TCP is titrated in. Some of
the greatest deviations in the presence of 2,4,6-TCP can be
observed in Figure 5. With the exception of the distal H55
NεH, the average weighted chemical shift deviation for 2,4,6-
TCP was 11% lower than that of 4-BP. This may be attributed
to the relatively low water solubility of the halophenols, as this
permitted only a 4ꢀ excess of 2,4,6-TCP versus a 15ꢀ excess of
4-BP. 2,4,6-TFP was the only trihalogenated substrate soluble
enough to reach a 15ꢀ excess concentration. While the smaller
2,4,6-TFP molecule is an active substrate, it is not as active as
others such as 2,4,6-TCP or 2,4,6-TBP (13, 28). With the addition
of 2,4,6-TFP the greatest changes in chemical shift were observed
in K99, Q4, H55 NεH, G1, M108, W120 NεH, L100, F60,
F35, and E45, respectively (see Supporting Information). At a
15
1
for the difference in backbone peak dispersion in the N and H
dimensions (8.29-fold difference) (26, 27).
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
15
2
ð N δ ðppmÞÞ
1
2
Δδ ðppmÞ ¼ ð H δ ðppmÞÞ þ
ð1Þ
8
:29
Residues F60, F35, K99, R122, and V59 respectively represent
the five largest deviations and V128, S42, G1, F21, and K58
respectively experience the next largest chemical shift alterations
in the presence of 4-BP. Three of the top five (and five of the top
ten) deviations occur in residues comprising the hydrophobic
internal distal binding pocket: F21, F35, K58, V59, and F60.
Current X-ray structures indicate that F35, V59, and F21 are the
closest residues to bound 4-iodophenol (4-IP) having respective
˚
C-C distances of 2.98, 2.98, and 3.41 A (14). The panels on the
left in Figure 4 show these deviations in the presence of inhibitor
4-BP, while the panels on the right show responses of these
same resonances upon addition of substrate 2,4,6-TCP. A large
disparity was seen in the effects these two molecules had on
the internal binding pocket, as no changes in chemical shift were

Article
Biochemistry, Vol. 49, No. 6, 2010 1203
1
15
1
F
IGURE 5: H- N HSQC spectrum of DHPCN A titrated with 2,4,6-TCP. The amide H resonancesexhibiting someofthe largest deviationsare
shown. These include the exchangeable distal histidine H55 NεH and amide protons of G1, R122, and D121.
1
5ꢀ excess of 2,4,6-TFP, the average chemical shift deviation was
the lowest of the halophenols studied (72% lower than that of
-BP). 2,4,6-TFP is believed to be the weakest binding trihalo-
4
genated phenol. In the monohalogenated phenol series, replace-
ment of the chlorine substituent with a fluorine atom results in
an increase in KD from 1.7 to 3.7 mM (15). It is unclear if this
decrease in binding affinity translates to the trihalogenated series.
This assumption justifies the significantly lowered chemical shift
deviations observed with 2,4,6-TFP.
DISCUSSION
The NMR backbone data presented here support the hypothe-
sis that substrate and inhibitor have distinct binding sites in
DHP A. The large effect of 4-bromophenol on binding poc-
ket residues F21, F35, K58, V59, and F60 is consistent with
internalization of the inhibitor in the distal pocket. The initial
proposal that the internal site was the substrate binding site was
reasonable on the grounds that such a well-defined site would be
assumed to have a function (6, 14). However, it is now clear that
the function of binding in this site is to inhibit catalysis by steric
interference at the heme iron (15). Steric interference both
decreases the binding affinity of H O at the heme iron and
2
2
removes the acid-base catalyst, H55, from the distal pocket.
In order to extract structural information from the NMR data,
the chemical shift perturbations caused by 4-BP, 2,4,6-TCP, and
2,4,6-TFP have been mapped onto the 2QFK X-ray structure of
the protein (Figure 6) (5). The five largest deviations for each
halophenol are shown in red, while the next five largest are shown
in blue. The heme and W120 side chain are displayed in black.
The heme and W120 sites serve as focal points for substrate
binding analysis, because both are capable of retaining a reactive
radical species necessary for the oxidation of 2,4,6-TCP or 2,4,6-
TFP (29, 30). The top panel in Figure 6 illustrates the effects of
F
IGURE 6: The residues experiencing the largest chemical shift devia-
tions have been mapped onto the 2QFK X-ray structure (5). The side
chains in red represent the five largest chemical shift changes, and the
residues in blue represent the next five largest deviations. The heme
and W120 residue (black) are provided as focal points. The panels
display the largest deviations caused by addition of 4-BP, 2,4,6-TCP,
and 2,4,6-TFP, respectively.
4-BP addition. With the exception of R122, the most prominent
changes (red) surround the internal binding pocket. The internal
binding pocket residues F21, F35, K58, V59, and F60 exhibit
large chemical shift deviations. However, it is not immediately
apparent why residues near the dimer interface (R122 and G1)
also exhibit significant perturbations. This could indicate either
an allosteric effect or nonspecific binding interactions at the
interface. Given the correlation between existing X-ray data and
1
15
current H- N HSQC data, there is little doubt regarding the

1
204 Biochemistry, Vol. 49, No. 6, 2010
Davis et al.
internalization of inhibitor 4-BP. However, since it has not been
possible to obtain an X-ray structure of the bound substrate, and
evidence points to an external site, H- N HSQC data are one
of the few current measurements that can be used to locate the
binding site of substrates 2,4,6-TCP and 2,4,6-TFP under ambi-
ent conditions.
second highest chemical shift change in 2,4,6-TCP, and third
largest in 2,4,6-TFP, occurs at the distal H55 NεH. Earlier FT-IR
experiments on the isoelectric, ferrous DHPCO A adduct in-
dicated the orientation of H55 was in the “closed” conformation
at pH 7, 290 K (12). If H55 is in the closed conformation, then
the observed deviations of H55 NεH are likely the result of a
perturbation in the Nε-H NC hydrogen bond, where NC
1
15
As can seen in Figure 4, in contrast to 4-BP, the internal pocket
resonances do not exhibit a noticeable difference when titrated
with the substrate 2,4,6-TCP. Instead, the local area surrounding
the dimer interface shows four out of the five largest chemical
shift perturbations (Figure 6). Of the five largest changes (R122,
H55 NεH, G1, S129, and L76, respectively) only the distal
histidine, H55 NεH, is not located near the interface. This result
is surprising given the distal histidine is separated from the dimer
3
3 3
refers to bound cyanide. The lack of internal pocket deviations
upon 2,4,6-TCP addition rules out binding of the molecule on the
internal side of H55 (i.e., the distal 4-BP binding pocket). Binding
of the molecule on the solvent-exposed, external side of H55
could be the cause for the deviation in the H55 NεH chemical
shift. This scenario would place the substrate near the heme edge
in an external fashion which is typical of other heme peroxidases,
such as HRP. Peroxidases like HRP commonly perform sub-
strate oxidation at the heme in an edge-on conformation (29).
In addition, recent resonance Raman data on metaquo DHP A
indicate an increase in 6-coordinate heme upon addition of 2,4,6-
TCP consistent with a “push” of H55 toward the protein interior,
adding credence to this mode of binding (15). There is also a long-
standing precedent for allosteric changes at dimer interfaces as a
result of axial ligand perturbations. Although cooperative bind-
ing and allosteric changes are typically related to tetrameric Hbs,
many well-known studies have illustrated similar changes in more
primitive, dimeric Hbs such as DHP A (31-33). Alterations in
H55 NεH may affect the hydrogen bond between the H55 side
chain and the axial ligand, ultimately resulting in allosteric
communication via the dimer interface.
˚
interface by a distance of nearly 25 A (5). Binding of substrate
2
2
,4,6-TFP has aspects of contributions seen in both 4-BP and
,4,6-TCP. As observed in Figure 6, alterations in chemical shift
occurred near the internal pocket (F35, H55 NεH, and F60)
and at a surface-exposed tryptophan near the interface (G1, Q4,
M108, and W120). Previous NMR studies showed that internal
heme substituents (most notably the 3-CH heme methyl) were
3
not affected by addition of 2,4,6-TFP but showed significant
changes in the presence of 4-BP. These data indicate that the
larger 4-BP has a much more pronounced effect in the distal
pocket than 2,4,6-TFP (13). Thus, 2,4,6-TFP has some access
to the distal pocket, but significantly less than 4-BP (13) and by
extension other 4-halophenols (15). On the basis of these data we
hypothesize that 2,4,6-TFP may act as both a substrate and a
weakly bound inhibitor, which would explain the decreased
activity of 2,4,6-TFP compared to 2,4,6-TCP (13, 28).
The disparity in perturbations caused by monohalogenated
inhibitor versus trihalogenated substrate indicates the impor-
tance of the 2- and 6-positions of the phenol in determining the
binding site. In this study, the monohalogenated inhibitor
primarily affects the internal distal pocket, while both trihalo-
genated substrates create perturbations externally near the heme
edge (H55) and dimer interface region near W120. The presence
of 2- and 6-position substituents on the phenol ring appears to be
a decisive factor in controlling substrate and inhibitor binding
locations. Steric and/or electrostatic effects of halogens at the
A second scenario involves direct binding of the substrate in
the region near the interface. This scenario is highly plausible
given the majority of residues with altered chemical shifts are
located near a potentially redox-active tryptophan (W120). Upon
inspection of the X-ray structures, this is the only residue in the
interface area capable of acting as an oxidizing equivalent (i.e., a
tryptophanyl radical). While the local shift deviations caused by
2,4,6-TCP appear evenly spread about the interface, the changes
induced by 2,4,6-TFP are more localized near W120 (Figure 6).
Many peroxidases, such as lignin peroxidase, manganese peroxi-
dase, and versatile peroxidase, are known to bind and oxidize
haloaromatics with high redox potentials near an exposed trypto-
phanyl radical (34-36). One of the most commonly studied
peroxidases, cytochrome c peroxidase (CcP), is also known to
generate a tryptophanyl radical (37). Upon addition of requisite
cosubstrate H O , X-band EPR experiments do indicate the
2
- and 6-positions likely restrict substrate access to the internal
distal binding pocket. While both substrates (2,4,6-TCP and
,4,6-TFP) have a significant effect on the distal H55, both differ
2
slightly in the perturbations induced near the dimer interface
region (Figure 6, red spheres). Binding of the smaller 2,4,6-TFP
substrate creates more localized deviations near W120, while
2
2
presence of a protein radical in DHP A that has tentatively been
assigned as a tyrosyl radical (22). In fact, electronic similarities
between the DHP A protein radical signature and the Trp191
radical species of CcP compound ES have led to naming this
reactive intermediate “DHP A compound ES” (22). Although the
EPR data reported for DHP A have been assigned to a tyrosyl
radical when initially formed, analysis of the freeze-quenched
radical signal at longer incubations times (> 12 s) led to the
suggestion that the EPR signature could be a mixture of tyrosyl
and tryptophanyl radicals (22). In structurally homologous
sperm whale myglobin, there are three separate trappable protein
radicals, including two tyrosyl (Y103 and Y151) and one
tryptophanyl (W14) centered radical (38, 39). In human hemo-
globin up to four trappable radicals, including two tyrosyl (Y24
and Y42), one cysteinyl (C93), and one histidyl-based radical
(H20), have been detected via the DMPO spin trap and sub-
sequent MS analysis (39, 40). Hence, there is precedent for
formation of multiple protein radicals in structurally homolo-
2,4,6-TCP binding appears to have a greater influence at the
dimer interface. It is possible that the substrate binding site
involves specific interactions with the halogens, and native 2,4,6-
TBP may bind even more tightly than 2,4,6-TCP or 2,4,6-TFP.
However, the low solubility of 2,4,6-TBP renders direct compar-
ison impossible. Indeed, similar size dependencies were recently
observed in various X-ray structures of bound monohalogenated
phenol in the distal pocket (15). It was found that smaller
monohalogated phenols (4-fluorophenol and 4-chlorophenol)
bind with lower affinities and do not penetrate the internal
pocket as efficiently when compared to larger monohalogenated
phenols (4-bromophenol and 4-iodophenol).
In general, the NMR data indicate that binding of substrate
may involve one of two possible scenarios. First, binding of the
substrate may occur primarily on the external edge of the distal
histidine, H55. This conclusion is reached on the basis that the

Article
Biochemistry, Vol. 49, No. 6, 2010 1205
gous globins, and there is no evidence that precludes the notion of
radical hopping in DHP A. If tryptophanyl radical formation at
W120 is a possibility, then a scenario involving 2,4,6-TCP or
2,4,6-TCP resulted in deviations largest at the distal H55 NεH
and dimer interface. No significant changes were observed in
internal binding pocket resonances upon addition of 2,4,6-TCP,
clearly illustrating that binding of substrate 2,4,6-TCP is sub-
stantially different and separate from that of inhibitor 4-BP.
Introduction of the smaller 2,4,6-TFP substrate showed a combi-
nation of the chemical shift changes observed for 4-BP and 2,4,6-
TCP. These deviations include alterations at the internal pocket
and surface tryptophan (W120) near the interface. Two scenarios
were presented that may tie together the activity of 2,4,6-TCP and
its observed binding perturbations. First, binding of the substrate
on the external side of H55 could create allosteric changes at
the dimer interface. This is in agreement with long-standing
precedents relating allosteric changes at dimer interfaces to axial
ligand perturbations. A second scenario involves direct binding
near W120 itself. In both cases, the substrate molecule would be
in favorable position for oxidation either along the heme edge or
at a potential tryptophanyl radical, respectively. The NMR
results have provided crucial insight into binding interactions
between the protein and a highly active substrate, 2,4,6-TCP.
Increased knowledge of the substrate binding site in DHP A will
provide an understanding into toxic haloaromatic binding events
in this bifunctional hemoglobin and possibly increase the oppor-
tunity to use DHP A for bioremediation processes beyond its
native function that involves oxidation of 2,4,6-TBP.
2,4,6-TFP binding and oxidation near W120 should not be ruled
out. Binding of the substrate at this site must however account for
the change in H55 NεH. Because the heme edge and W120 are
˚
separated by nearly 9 A, substrate binding near W120 must
produce transferable long-range binding effects to the axial CN
and likewise the H-bonded H55 NεH. Conducting enzymatic
assays of various W120 mutants will help to determine any role
W120 may have in substrate binding and overall protein activity.
The metcyano DHP adduct utilized in this study has been used
previously to investigate halophenol binding interactions near the
heme via 1D and 2D NMR experiments (13). Previous experi-
ments demonstrated that DHPCN A exhibits a wide dispersion
1
of chemical shifts (-12 to 27 ppm in the H NMR spectra), with
resonances of the heme and nearby residues being hyperfine-
shifted out from under the diamagnetic envelope. The hyperfine
shifts are due to a combination of contact, pseudocontact, and
1
dipolar shift contributions arising from the S = / paramagnetic
2
heme center (41). Several resonances of nonligated residues
experiencing large hyperfine shifts (i.e., proximal side F97 Cζ,
Cε, and Cδ ring protons) were found to lack broadening and/or
chemical shift deviations in the presence of substrate 2,4,6-TFP.
This is significant as variations in dipolar contributions to the
overall chemical shift are highly dependent on the orientation of
the magnetic susceptibility tensor, χ, which is directly correlated
to the Fe-CN tilt angle, β, from the heme normal (42-44).
Chemical shift deviations in hyperfine-shifted resonances of
nonligated residues, such as F97, would be expected if substrate
binding induced structural rearrangements at or near the residue
or if substrate binding altered the relative Fe-CN tilt angle, β.
The Fe-CN tilt angle does not appear to be altered in the
presence of active substrates like 2,4,6-TFP and suggests that
chemical shift deviations observed in the distal H55NεH are not
due to slight changes in the dipolar field. This rationale is
supported by the lack of chemical shift deviations in other
strongly paramagnetically influenced resonances, such as F97
CδHs, CεHs, and CζH and H89 CβHs (13).
SUPPORTING INFORMATION AVAILABLE
1
15
Full backbone chemical shift assignments and H- N HSQC
data upon titration of 2,4,6-TFP. This material is available free of
charge via the Internet at http://pubs.acs.org.
REFERENCES
1
. Han, K. P., Woodin, S. A., Lincoln, D. E., Fielman, K. T., and Ely, B.
2001) Amphitrite ornata, a marine worm, contains two dehaloperoxi-
dase genes. Mar. Biotechnol. 3, 287–292.
(
2. de Serrano, V., D’Antonio, J., Thompson, M. K., Franzen, S., and
Ghiladi, R. A. (2010) Crystal structure of dehaloperoxidase B at 1.58
˚
A and structural characterization of the A/B dimer from Amphitrite
ornata, J. Biol. Chem. (submitted for publication).
3. D’Antonio, J., D’Antonio, E. L., Bowden, E. F., Smirnova, T. S.,
Franzen, S. F., and Ghiladi, R. A. (2010) Spectroscopic and mecha-
nistic investigations of dehaloperoxidase B from Amphitrite ornata,
J. Biol. Chem. (submitted for publication).
Deviations in the Fe-CN tilt angle cannot, however, be ruled
out in the presence of inhibitor 4-BP. Previously reported NMR
data indicated that resonances with large dipolar shifts (e.g., F97
ring protons) experience broadening and chemical shift devia-
tions upon titration of 4-BP and not 2,4,6-TFP (13). Data from
4
. Chen, Z., De Serrano, V., Betts, L., and Franzen, S. (2009) Distal
histidine conformational flexibility in dehaloperoxidase from
Amphitrite ornata. Acta Crystallogr., Sect. D: Biol. Crystallogr. 65,
34–40.
1
15
the current H- N HSQC experiments indicate titration of 4-BP
primarily results in alterations of distal side amide protons. In a
previous report it was not fully understood why the ring protons
5
. de Serrano, V., Chen, Z. X., Davis, M. F., and Franzen, S. (2007)
X-ray crystal structural analysis of the binding site in the ferric and
oxyferrous forms of the recombinant heme dehaloperoxidase cloned
from Amphitrite ornata. Acta Crystallogr., Sect. D: Biol. Crystallogr.
of F97, a residue on the proximal side of the 3-CH heme methyl,
3
63, 1094–1101.
had such a sensitive response to the presence of 4-BP. One
possible explanation could lie in a slight alteration of the Fe-CN
tilt angle induced by steric interactions with inhibitor 4-BP in the
distal pocket. This explanation is logical since NMR (vide supra)
and X-ray data now indicate 4-BP bound in the distal pocket less
6. Lebioda, L., LaCount, M. W., Zhang, E., Chen, Y. P., Han, K.,
Whitton, M. M., Lincoln, D. E., and Woodin, S. A. (1999) An
enzymatic globin from a marine worm. Nature 401, 445.
7
. Weber, R. E., Mnum, C. P., Steinman, H., Bonaventura, C., Sullivan,
B., and Bonaventura, J. (1977) Hemoglobins of two terebellid poly-
chaetes: Enoplobranchus sanuigneus and Amphitire ornata. Comp.
Biochem. Physiol. 56A, 179–187.
˚
than 4 A from the heme iron (14, 15).
8
. Belyea, J., Gilvey, L. B., Davis, M. F., Godek, M., Sit, T. L., Lommel,
S. A., and Franzen, S. (2005) Enzyme function of the globin dehalo-
peroxidase from Amphitrite ornata is activated by substrate binding.
Biochemistry 44, 15637–15644.
CONCLUSION
The NMR study shows a distinct difference in the binding
interactions of a 4-halophenol inhibitor and 2,4,6-trihalogenated
phenol substrates. Binding of a 4-halophenol inhibitor, 4-BP,
produced the greatest deviations in backbone chemical shifts in
the distal binding pocket. Addition of trihalogenated substrate
9. Franzen, S., Gilvey, L. B., and Belyea, J. L. (2007) The pH dependence
of the activity of dehaloperoxidase from Amphitrite ornata. Biochim.
Biophys. Acta 1774, 121–130.
0. Franzen, S., Chaudhary, C., Belyea, J., Gilvey, L., Davis, M. F., Sit,
T. L., and Lommel, S. A. (2006) Proximal cavity, distal histidine
and substrate hydrogen-bonding mutations modulate the acti-
1

1
206 Biochemistry, Vol. 49, No. 6, 2010
Davis et al.
vity of Amphitrite ornata dehaloperoxidase. Biochemistry 45, 9085–
9
1. Nienhaus, K., Nickel, E., Davis, M. F., Franzen, S., and Nienhaus,
G. U. (2008) Determinants of substrate internalization in the distal
pocket of dehaloperoxidase hemoglobin of Amphitrite ornata. Bio-
chemistry 47, 12985–12994.
28. Chaudhary, C. (2003) Functional studies of the push-pull mechanism
in dehaloperoxidase from Amphitrite ornata, Master’s Thesis, North
Carolina State University.
094.
1
1
1
29. Ator, M. A., and Ortiz de Montellano, P. R. (1987) Protein control of
prosthetic heme reactivity: Reaction of substrates with the heme edge
of horseradish peroxidase. J. Biol. Chem. 262, 1542–1551.
30. Pogni, R., Baratto, M. C., Giansanti, S., Teutloff, C., Verdin, J.,
Valderrama, B., Lendzian, F., Lubitz, W., Vazquez-Duhalt, R., and
Basosi, R. (2005) Tryptophan-based radical in the catalytic mecha-
nism of versatile peroxidase from Bjerkandera adusta. Biochemistry
44, 4267–4274.
2. Nienhaus, K., Deng, P. C., Belyea, J., Franzen, S., and Nienhaus, G. U.
(
2006) Spectroscopic study of substrate binding to the carbonmonoxy
form of dehaloperoxidase from Amphitrite ornata. J. Phys. Chem. B
10, 13264–13276.
1
3. Davis, M. F., Gracz, H., Vendeix, F. A. P., de Serrano, V.,
Somasundaram, A., Decatur, S. M., and Franzen, S. (2009) Different
modes of binding of mono-, di- and trihalogenated phenols to the
hemoglobin dehaloperoxidase from Amphitrite ornata. Biochemistry
31. Royer, W. E., Hendrickson, W. A., and Chiancone, E. (1990)
Structural transitions upon ligand-binding in a cooperative dimeric
hemoglobin. Science 249, 518–521.
4
8, 2164–2172.
32. Royer, W. E., Pardanani, A., Gibson, Q. H., Peterson, E. S., and
Friedman, J. M. (1996) Ordered water molecules as key allosteric
mediators in a cooperative dimeric hemoglobin. Proc. Natl. Acad. Sci.
U.S.A. 93, 14526–14531.
33. Knapp, J. E., Bonham, M. A., Gibson, Q. H., Nichols, J. C., and
Royer, W. E. (2005) Residue F4 plays a key role in modulating oxygen
affinity and cooperativity in Scapharca dimeric hemoglobin. Biochem-
istry 44, 14419–14430.
1
1
4. Zhang, E., Chen, Y. P., Roach, M. P., Lincoln, D. E., Lovell, C. R.,
Woodin, S. A., Dawson, J. H., and Lebioda, L. (1996) Crystallization
and initial spectroscopic characterization of the heme-containing
dehaloperoxidase from the marine polychaete Amphitrite ornata. Acta
Crystallogr., Sect. D 52, 1191–1193.
5. Nicoletti, F. P., Thompson, M. K., Howes, B. D., Franzen, S., and
Smulevich, G. (2010) New Insights into the Role of Distal Histidine
Flexibility in Ligand Stabilization of Dehaloperoxidase-hemoglobin
from Amphitrite ornata. Biochemistry (DOI: 10.1021/bi9020567).
34. Doyle, W. A., Blodig, W., Veitch, N. C., Piontek, K., and Smith, A. T.
(1998) Two substrate interaction sites in lignin peroxidase revealed by
site-directed mutagenesis. Biochemistry 37, 15097–15105.
35. Perez-Boada, M., Ruiz-Duenas, F. J., Pogni, R., Basosi, R., Choi-
nowski, T., Martinez, M. J., Piontek, K., and Martinez, A. T. (2005)
Versatile peroxidase oxidation of high redox potential aromatic
compounds: Site-directed mutagenesis, spectroscopic and crystallo-
graphic investigation of three long-range electron transfer pathways.
J. Mol. Biol. 354, 385–402.
36. Ruiz-Duenas, F. J., Pogni, R., Morales, M., Giansanti, S., Mate,
M. J., Romero, A., Martinez, M. J., Basosi, R., and Martinez, A. T.
(2009) Protein radicals in fungal versatile peroxidase: Catalytic trypto-
phan radical in both compound I and compound II and studies on
W164Y, W164H, and W164S variants. J. Biol. Chem. 284, 7986–7994.
37. Sivaraja, M., Goodin, D. B., Smith, M., and Hoffman, B. M. (1989)
Identification by ENDOR of Trp191 as the free-radical site in
cytochrome-c peroxidase compound ES. Science 245, 738–740.
38. Gunther, M. R. (2004) Probing the free radicals formed in the
metmyoglobin-hydrogen peroxide reaction. Free Radical Biol. Med.
36, 1345–1354.
1
6. Shikama, K. (2006) Nature of the FeO
hemoglobin: A new molecular paradigm. Prog. Biophys. Mol. Biol. 91,
3–162.
2
bonding in myoglobin and
8
1
7. Fielman, K. T., Woodin, S. A., and Lincoln, D. E. (2001) Poly-
chaete indicator species as a source of natural halogenated organic
compounds in marine sediments. Environ. Toxicol. Chem. 20,
7
38–747.
8. Fielman, K. T., Woodin, S. A., Walla, M. D., and Lincoln, D. E.
1999) Widespread occurrence of natural halogenated organics
1
1
2
(
among temperate marine infauna. Mar. Ecol.: Prog. Ser. 181, 1–12.
9. Woodin, S. A., Lindsay, S. M., and Lincoln, D. E. (1997) Biogenic
bromophenols as negative recruitment cues. Mar. Ecol.: Prog. Ser.
1
57, 303–306.
0. Lincoln, D. E., Fielman, K. T., Marinelli, R. L., and Woodin, S. A.
2005) Bromophenol accumulation and sediment contamination by
(
the marine annelids Notomastus lobatus and Thelepus crispus. Bio-
chem. Systemat. Ecol. 33, 559–570.
1. Osborne, R. L., Taylor, L. O., Han, K. P., Ely, B., and Dawson, J. H.
2
2
2
2
(
2004) Amphitrite ornata dehaloperoxidase: Enhanced activity for the
39. Mason, R. P. (2004) Using anti-5,5-dimethyl-1-pyrroline N-oxide to
detect protein radicals in time and space with immuno-spin trapping.
Free Radical Biol. Med. 36, 1214–1223.
40. Bhattacharjee, S., Deterding, L. J., Jiang, J., Bonini, M. G., Tomer,
K. B., Ramirez, D. C., and Mason, R. P. (2007) Electron transfer
between a tyrosyl radical and a cysteine residue in hemoproteins: Spin
trapping analysis. J. Am. Chem. Soc. 129, 13493–13501.
catalytically active globin using MCPBA. Biochem. Biophys. Res.
Commun. 324, 1194–1198.
2. Feducia, J., Dumarieh, R., Gilvey, L. B. G., Smirnova, T., Franzen,
S., and Ghiladi, R. A. (2009) Characterization of dehaloperoxidase
compound ES and its reactivity with trihalophenols. Biochemistry 48,
9
95–1005.
3. Smirnova, T. I., Weber, R. T., Davis, M. F., and Franzen, S. (2008)
Substrate binding triggers a switch in the iron coordination in
dehaloperoxidase from Amphitrite ornata: HYSCORE experiments.
J. Am. Chem. Soc. 130, 2128–2129.
4. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and
Bax, A. (1995) NMRPipe: A multidimensional spectral processing
system based on UNIX pipes. J. Biomol. NMR 6, 277–293.
5. Goddard, T. D., and Kneller, D. G. SPARKY 3, University of
California, San Francisco.
41. La Mar, G. N., and de Ropp, J. S. (1993) in Biological Magnetic
Resonance, pp 1-73, Plenum Press, New York.
42. La Mar, G. N. (2007) Application of the paramagnetic dipole field for
solution NMR active site structure determination in low-spin, cya-
nide-inhibited ferric hemoproteins. IUBMB Life 59, 513–527.
43. La Mar, G. N., Chen, Z., Vyas, K., and McPherson, A. D. (1995) An
interpretive bases of the hyperfine shifts in cyanide-inhibited horse-
radish peroxidase based on the magnetic axes and ligand tilt. Influence
of substrate binding and extensions to other peroxidases. J. Am.
Chem. Soc. 117, 411–419.
44. Zhu, W., Li, Y., Wang, J., Ortiz de Montellano, P. R., and La Mar,
G. N. (2006) Solution NMR study of environmental effects on
substrate seating in human heme oxygenase: Influence of polypeptide
truncation, substrate modification and axial ligand. J. Inorg. Biochem.
100, 97–107.
2
2
6. Kordys, D. R., Bobay, B. G., Thompson, R. J., Venters, R. A., and
Cavanah, J. (2007) Peptide binding proclivities of calcium loaded
calbindin-D28k. FEBS Lett. 581, 4778–4782.
2
7. Palmer, S. M., Schaller, M. D., and Campbell, S. L. (2008) Vinculin
tail conformation and self-association is independent of pH and H906
protonation. Biochemistry 47, 12467–12475.