Mechanistic and structural analyses of the roles of active site residues in yeast polyamine oxidase Fms1: Characterization of the N195A and D94N enzymes

Article abstract of DOI:10.1021/bi3011434

Flavoprotein Fms1 from Saccharomyces cerevisiae catalyzes the oxidation of spermine in the biosynthetic pathway for pantothenic acid. The same reaction is catalyzed by the mammalian polyamine and spermine oxidases. The active site of Fms1 contains three amino acid residues positioned to interact with the polyamine substrate, His67, Asn195, and Asp94. These three residues form a hydrogen-bonding triad with Asn195 being the central residue. Previous studies of the effects of mutating His67 are consistent with that residue being important both for interacting with the substrate and for maintaining the hydrogen bonds in the triad [Adachi, M. S., Taylor, A. B., Hart, P. J., and Fitzpatrick, P. F. (2012) Biochemistry 51, 4888-4897]. The N195A and D94N enzymes have now been characterized to evaluate their roles in catalysis. Both mutations primarily affect the reductive half-reaction. With N ¹-acetylspermine as the substrate, the rate constant for flavin reduction decreases ～450-fold for both mutations; the effects with spermine as the substrate are smaller, 20-40-fold. The k_cat/K_amine- and k_cat-pH profiles with N¹-acetylspermine are only slightly changed from the profiles for the wild-type enzyme, consistent with the pK_a values arising from the amine substrate or product and not from active site residues. The structure of the N195A enzyme was determined at a resolution of 2.0 ?. The structure shows a molecule of tetraethylene glycol in the active site and establishes that the mutation has no effect on the protein structure. Overall, the results are consistent with the role of Asn195 and Asp94 being to properly position the polyamine substrate for oxidation.

Full text of DOI:10.1021/bi3011434

Article
pubs.acs.org/biochemistry
Mechanistic and Structural Analyses of the Roles of Active Site
Residues in Yeast Polyamine Oxidase Fms1: Characterization of the
N195A and D94N Enzymes
Mariya S. Adachi, Alexander B. Taylor, P. John Hart, and Paul F. Fitzpatrick*,†
†
†
†,‡
†
Department of Biochemistry, University of Texas Health Science Center at San Antonio, San Antonio, Texas 78229, United States
‡
Department of Veterans Affairs, Audie Murphy Division, Geriatric Research, Education, and Clinical Center, South Texas Veterans
Health Care System, San Antonio, Texas 78229, United States
ABSTRACT: Flavoprotein Fms1 from Saccharomyces cerevisiae catalyzes the oxidation
of spermine in the biosynthetic pathway for pantothenic acid. The same reaction is
catalyzed by the mammalian polyamine and spermine oxidases. The active site of Fms1
contains three amino acid residues positioned to interact with the polyamine substrate,
His67, Asn195, and Asp94. These three residues form a hydrogen-bonding triad with
Asn195 being the central residue. Previous studies of the effects of mutating His67 are
consistent with that residue being important both for interacting with the substrate and
for maintaining the hydrogen bonds in the triad [Adachi, M. S., Taylor, A. B., Hart, P.
J., and Fitzpatrick, P. F. (2012) Biochemistry 51, 4888−4897]. The N195A and D94N
enzymes have now been characterized to evaluate their roles in catalysis. Both mutations primarily affect the reductive half-
1
reaction. With N -acetylspermine as the substrate, the rate constant for flavin reduction decreases ∼450-fold for both mutations;
1
the effects with spermine as the substrate are smaller, 20−40-fold. The k /K
− and k −pH profiles with N -acetylspermine
cat
amine
cat
are only slightly changed from the profiles for the wild-type enzyme, consistent with the pK values arising from the amine
a
substrate or product and not from active site residues. The structure of the N195A enzyme was determined at a resolution of 2.0
Å. The structure shows a molecule of tetraethylene glycol in the active site and establishes that the mutation has no effect on the
protein structure. Overall, the results are consistent with the role of Asn195 and Asp94 being to properly position the polyamine
substrate for oxidation.
8
9
he polyamines spermidine and spermine are polybasic1
molecules that are ubiquitous in living organisms.
tures of maize PAO and Fms1 are available, but not that of
any mammalian SMO or PAO. These structures establish that
the maize enzyme and Fms1 belong to the monoamine oxidase
T
Because polyamines are required by growing cells, the enzymes
of polyamine metabolism have been targets for the develop-
10
structural family. The sequences of the mammalian SMOs
and PAOs, while only ∼20% identical to those of maize PAO
and Fms1, indicate that they also have the same fold as
monoamine oxidase. Because the maize enzyme catalyzes a
reaction different from that of Fms1 and the mammalian
enzymes, Fms1 is at present the best model for understanding
the structural basis for catalysis and specificity of the
mammalian polyamine-oxidizing enzymes.
2,3
ment of anticancer drugs. Catabolism of polyamines in
4
mammals occurs by one of two pathways. Both spermine and
spermidine can be acetylated on a terminal nitrogen by
1
1
spermidine/spermine-N -acetyltransferase. The N -acetylsper-
1
mine or N -acetylspermidine can then be oxidized by the
peroxisomal flavoprotein polyamine oxidase (PAO, also known
as N-acetylpolyamine oxidase) to produce spermidine or
putrescine, respectively, and N-acetyl-3-aminopropanaldehyde
Scheme 1). Alternatively, spermine and spermidine can be
oxidized directly by the separate flavoprotein spermine oxidase
SMO) to spermidine or putrescine and 3-aminopropanalde-
The structure of Fms1 shows that the active site contains
three polar residues capable of interacting with a polyamine
substrate, His67, Asn195, and Asp94. On the basis of a
structure of Fms1 in the presence of spermine, these three
residues have been proposed to form hydrogen bonds with the
,
9 a
(
(
hyde. Plants and yeast also contain flavoproteins capable of
catalyzing polyamine oxidation. Saccharomyces cerevisiae protein
nitrogens of the amine and with one another (Figure 1).
1
His67, which is conserved in the mammalian enzymes, would
form a hydrogen bond with N4 of the substrate, the site of
oxidation, and with Asn195. Asn195 would form a hydrogen
bond with N12 of spermine and with Asp94, which in turn also
interacts with N12 of the substrate. Mechanistic and structural
Fms1 oxidizes spermine and N -acetylspermine like the
5
,6
mammalian enzymes. The 3-aminopropanal produced by
Fms1 is further metabolized to β-alanine, a precursor of
5
,6
pantothenic acid. The plant enzymes oxidize spermine and
spermidine and their acetylated derivatives on the endo side of a
nitrogen, producing 1,3-diaminopropane and the aldehyde form
1
1
7
of N -acetyl-spermidine from N -acetylspermine.
Received: August 23, 2012
Revised: October 2, 2012
Published: October 3, 2012
The structural bases for the different substrate specificities of
the polyamine-oxidizing flavoproteins are not known. Struc-
©
2012 American Chemical Society
8690
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
Scheme 1
Acros Organics (Geel, Belgium). The nickel-nitrilotriacetic acid
agarose and the Sephacryl S-100 HR resin were purchased from
Invitrogen (Carlsbad, CA) and Sigma-Aldrich, respectively.
Site-Directed Mutagenesis. The mutations of residues
Asn195 and Asp94 were introduced into the gene encoding
6
wild-type Fms1 using the QuikChange protocol (Stratagene).
The oligonucleotides used as forward primers, with the
m u t a t i o n s u n d e r l i n e d , w e r e 5 ′ - C A C C A A G -
GAAGGGCCGCCTTTGCTTTG-3′ and 5′-GTTTTTGA-
TAACGATAATTTTATTTATATCGACG-3′ for the N195A
and D94N mutations, respectively. The entire sequences of the
mutant genes were determined at the Nucleic Acid Core
Facility at the University of Texas Health Science Center at San
Antonio.
Protein Expression and Purification. Wild-type and
mutant Fms1 were expressed and purified following the
13
protocol for the wild-type enzyme. The levels of protein
expression and the yields of the purified mutant proteins were
similar to those for wild-type Fms1. For crystallization, the
N195A enzyme was purified further using a Sephacryl S-100
11
column as described previously. The purified protein was
dialyzed against three changes of 25 mM HEPES, 25 mM
NaCl, and 2% glycerol (pH 7.5) before crystallization.
Enzyme Assays. The enzymatic activities of both N195A
and D94N Fms1 were determined in buffer containing 10%
glycerol by following oxygen consumption with a Yellow
Springs Instrument model 5300 oxygen monitor at 25 °C. The
steady-state kinetic parameters k , K
, and k /K
were
cat
amine
cat
amine
determined at an oxygen concentration of 1.2 mM and varying
1
concentrations of spermine (0.05−3 mM) at pH 9.35 or N -
acetylspermine (0.05−2 mM) at pH 9.0. The parameters KO2
and k /KO2 were determined by varying the concentration of
cat
oxygen (0.065−1.2 mM) at 20 mM spermine at pH 9.35 or 20
1
mM N -acetylspermine at pH 9.0. The reaction mixture (1 mL)
was first equilibrated with the substrates at the desired
Figure 1. Active site residues in S. cerevisiae Fms1 showing the
proposed interactions with spermine. The structure is based on
subunit B of PDB entry 1XPQ.
concentrations, and the appropriate O /N2 mixture was
2
bubbled into the cell of the oxygen electrode for 10 min
prior to starting the reaction by adding enzyme. The buffers
were 200 mM bis-Tris at pH 6.8, 200 mM Tris-HCl from pH
7.0 to 8.75, 200 mM CHES from pH 9.0 to 9.75, and 200 mM
CAPS from pH 10.1 to 10.5. Because of the hydroscopic nature
11
studies of the effects of mutating His67 of Fms1 and His64 of
1
2
mouse PAO suggest that this histidine residue plays a role
both in properly positioning the substrate for hydride transfer
to the flavin and in maintaining the position of Asn195. In this
study, we describe the effects of replacing Asn195 of Fms1 with
alanine and Asp94 with asparagine.
1
of N -acetylspermine, its concentration was determined
enzymatically.
Rapid-reaction analyses were conducted using an Applied
Photophysics SX-20MV stopped-flow spectrophotometer in the
absorbance mode. The instrument was thermostated at 25 °C,
and enzyme and substrate solutions were made anaerobic as
MATERIALS AND METHODS
■
1
3
Materials. The pET28-based pJWL94 vector encoding
described previously. Enzyme (∼20 μM) was mixed
6
histidine-tagged Fms1 was kindly provided by R. Sternglanz
anaerobically with varying concentrations of spermine (0.03−
1
(
Stony Brook University, Stony Brook, NY). Escherichia coli
2 mM) or N -acetylspermine (0.03−1 mM) in 200 mM CHES
+
BL21(DE3) Codon RIL competent cells and the QuikChange
site-directed mutagenesis kit were from Stratagene (Santa
Clara, CA). The QIAprep Spin Miniprep kit was from Qiagen
(
pH 9.0), and the reaction was followed at 458 nm.
Data Analysis. Kinetic data were analyzed using Kaleida-
Graph (Adelbeck Software, Reading, PA). The steady-state
(
Valencia, CA). Oligonucleotides used for site-directed muta-
kinetic parameters k , k /K , and K were determined by
cat
cat
M
M
genesis and for sequencing of the mutant genes were
synthesized by the Nucleic Acid Core Facility at the University
of Texas Health Science Center at San Antonio. Kanamycin,
chloramphenicol, isopropyl β-D-thiogalactopyranoside, lyso-
zyme, Luria-Bertani agar and broth, phenylmethanesulfonyl
fluoride, and HEPES were from Fisher (Pittsburgh, PA). N -
Acetylspermine trihydrochloride and glucose oxidase were from
Sigma-Aldrich (Milwaukee, WI). Spermine was purchased from
fitting the data to the Michaelis−Menten equation. The k /
amine
Fms1 were fit to eq 1
cat
K
values at different pH values for both N195A and D94N
⎛
⎞
⎟
⎟
1
log Y = log⎜
C
⎜
H
K2
H ⎠
1
+
+
⎝
K1
(1)
8
691
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
where K and K are the dissociation constants for the ionizable
Table 1. Data Collection and Refinement Statistics
1
2
groups and C is the pH-independent k /K
value. The kcat
cat
amine
Data Collection
values for the mutant enzymes as a function of pH were fit to
eq 2
space group
cell dimensions
a, b, c (Å)
P1
⎛
K1
⎞
YL + YH
78.9, 81.9, 104.8
78.1, 79.5, 78.9
0.97920
⎜
⎜
⎝
H ⎟
log Y = log
α, β, γ (deg)
wavelength (Å)
resolution (Å)
Rsym
K1
⎟
1
+ H
⎠
(2)
100.0−2.0
0.109 (0.408)
where K is the dissociation constant for the ionizable group, Y
a
1
L
1
4
is the activity at low pH, and Y is the activity at high pH.
a
H
I/σI
10.5 (3.2)
Stopped-flow traces were fit to eq 3, which describes a biphasic
exponential process
a
completeness (%)
redundancy
89.0 (75.3)
a
3.4 (3.2)
−
λ t
−λ2t
Refinement
1
A = A + A e
+ A2e
(3)
∞
1
resolution (Å)
no. of reflections
Rwork/Rfree
79.1−2.0
149579
0.203/0.234
4
where λ and λ are the first-order rate constants for each phase,
A and A are the absorbances of each species at time t, and A
1
2
1
2
∞
is the absorbance at infinite time. Kinetic parameters for the
reductive half-reaction were determined using eq 4
no. of monomers per unit cell
no. of atoms
protein
15909
260
k S
Kd + S
2
kobs
=
ligand
(4)
solvent
B factor (Å2)
723
where kobs is the observed first-order rate constant for the
reduction of the enzyme-bound flavin at a specific concen-
protein
34.0
25.4
36.3
ligand
tration of the amine, k is the rate constant for reduction at
2
solvent
saturating concentrations of the amine, and K is the apparent
d
rmsd
dissociation constant for binding of the amine to the enzyme.
An alternative fit, in which a non-zero y-intercept was included,
did not improve the quality of the fits to the data and yielded
values for the y-intercepts not significantly different from zero.
Crystallization, Structure Determination, and Refine-
ment. Crystals of Fms1 N195A were grown in the UTHSCSA
X-ray Crystallography Core Laboratory using the Morpheus
crystallization screen kit (Molecular Dimensions Ltd., New-
market, U.K.) and a Phoenix crystallization robot (Art Robbins
Instruments, Sunnyvale, CA). The crystals grew within 1 week
by the sitting drop vapor diffusion method with the protein
solution mixed in a 1:1 ratio with buffer containing 30 mM
diethylene glycol, 30 mM triethylene glycol, 30 mM tetra-
ethylene glycol, 30 mM pentaethylene glycol, 10% ethylene
glycol, 20% polyethylene glycol 8000, and 0.1 M 2-(N-
morpholino)ethanesulfonic acid (MES)-imidazole (pH 6.5).
Data were collected from a crystal flash-cooled with liquid
nitrogen on beamline 24-ID-C at the Advanced Photon Source
bond lengths (Å)
bond angles (deg)
0.004
0.876
a
Values in parentheses are for the highest-resolution shell.
in the kcat/Kamine value of ∼1 order of magnitude with both
amine substrates, with a slightly larger effect on N -
acetylspermine oxidation. The effect of this mutation on the
cat value with spermine mirrors the effect on the kcat/K value,
while the kcat value with N -acetylspermine decreases only ∼2-
fold. Mutating Asn195 to alanine has no significant effect on
the kcat/K value for oxygen with either substrate. These results
M
are consistent with this mutation primarily affecting the
1
k
M
1
reductive half-reaction.
The D94N mutation has a much larger effect on the kcat
amine and kcat values for spermine than those for N -
/
1
K
acetylspermine (80- vs 4-fold and 20- vs 1.5-fold, respectively).
1
With either N -acetylspermine or spermine as the polyamine
(
Argonne National Laboratory, Argonne, IL). The data were
substrate, the k /KO2 value decreases ∼4-fold for the D94N
cat
15
integrated and scaled using HKL-2000. Initial phases were
obtained by the molecular replacement method implemented in
PHASER using Fms1 coordinates in PDB entry 4ECH as
enzyme. Thus, this mutation affects both the reductive and
oxidative half-reactions, with a larger effect on the former.
pH Dependence of the kcat and k /K
gain further insight into the basis for the changes in the k /KM
1
6
11
Values. To
cat amine
the search model. Coordinates were refined against the data
cat
1
7
using PHENIX, including simulated annealing, alternating
and kcat values, k /K
− and k −pH profiles were
cat
amine
cat
18
1
with manual rebuilding using COOT. Data collection and
refinement statistics are listed in Table 1. The coordinates have
been deposited in the PDB as entry 4GDP.
determined with N -acetylspermine as the amine substrate for
the mutant enzymes. For both enzymes, the k /K −pH
cat
amine
profiles are bell-shaped, with maxima of ∼9.0, similar to the
profile for the wild-type enzyme (Figure 2A). These pH profiles
show the importance of two ionizable groups in the free
enzyme or substrate, one of which must be protonated and one
unprotonated for full activity. The data were fit to eq 1 to
RESULTS
■
Steady-State Kinetic Parameters. To analyze the role of
Asn195 and Asp94 of Fms1 in binding and catalysis, Asn195
was mutated to alanine and Asp94 was mutated to asparagine.
The steady-state kinetic parameters of the mutant enzymes with
extract the pK values (Table 3). For the N195A enzyme, the
a
two pK values are sufficiently separated to yield discrete pKa
a
1
spermine and N -acetylspermine as substrates are given in
values that are unchanged from the values for the wild-type
Table 2. The analyses were conducted at the respective pH
enzyme. For the D94N enzyme, the two pK values are too
a
1
optima for the wild-type enzyme, 9.0 for N -acetylspermine and
close together to resolve, so that only the average pK value of
a
13
9
.35 for spermine. The N195A mutation results in a decrease
the two ionizable groups can be determined; this value is not
8
692
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
a
Table 2. Steady-State Kinetic Parameters for Wild-Type Fms1 and the N195A and D94N Enzymes
d
substrate
kinetic parameter
Fms1
N195A
D94N
1
b
kcat (s−1
c
N -acetylspermine
)
15.1 ± 0.4
1400 ± 200
10.9 ± 1.8
358 ± 20
8.1 ± 0.2
78 ± 13
104 ± 14
405 ± 72
9.7 ± 0.2
320 ± 50
30 ± 5
(mM s−1
−
1
)
c
k /K
cat
amine
c
Kamine (μM)
k /K (mM− s−1
1
e
)
100 ± 14
cat
O
2
e
KO2 (μM)
43.6 ± 2.3
20 ± 3
97 ± 14
f
kcat (s−1
k /K
c
spermine
)
39.0 ± 1.5
330 ± 60
118 ± 25
428 ± 77
4.9 ± 0.2
38.5 ± 8.3
127 ± 27
327 ± 67
2.0 ± 0.3
4.1 ± 1.0
320 ± 88
104 ± 36
(mM s−1
−
1
)
c
cat
amine
c
Kamine (μM)
k /K (mM− s−1
1
g
)
cat
O
2
g
KO2 (μM)
91 ± 16
15 ± 3
192 ± 60
a
b
c
d
e
At 25 °C. Determined at pH 9.0. Determined by varying the concentration of the amine at 1.2 mM oxygen. From ref 13. Determined by
1
f
g
varying the concentration of oxygen at 20 mM N -acetylspermine. Determined at pH 9.35. Determined by varying the concentration of oxygen at
0 mM spermine.
2
Table 3. pK Values for Wild-Type Fms1 and the D94N and
a
a
N195A Enzymes
enzyme
eq
kinetic parameter
pK1
pK2
b
wild-type Fms1
1
2
1
2
1
2
k /K
8.3 ± 0.1
7.7 ± 0.1
8.1 ± 0.1
9.0 ± 0.1
8.7 ± 0.1
8.4 ± 0.1
9.6 ± 0.1
−
9.4 ± 0.2
−
8.7 ± 0.1
−
cat
amine
amine
amine
kcat
N195A
D94N
k /K
cat
kcat
k /K
cat
kcat
a
1
b
Conditions: N -acetylspermine as a substrate, 25 °C. From ref 13.
significantly different from the average of the pK values for the
a
wild-type enzyme (9.0 ± 0.2).
The effects of pH on the k values for both mutant enzymes
cat
1
with N -acetylspermine as the amine substrate are shown in
Figure 2B. As is the case for the wild-type enzyme, the kcat
values for the mutant enzymes increase with pH to limiting
values at high pH, consistent with the requirement for an
unprotonated group for the fastest turnover of both the wild-
type and mutant enzymes. To obtain the pK values (Table 3),
a
the data were fit to eq 2. For both mutant enzymes, this pKa
value is higher than that seen for the wild-type enzyme. As a
result, while the kcat value for the wild-type enzyme is constant
at pH ≥9.0, the k values for the mutant enzymes do not reach
cat
a maximum until pH ≥10. The data in Figure 2B can be used to
obtain the pH-independent k values for the mutant enzymes
cat
−1
using eq 2. This yields kcat values of 11.5 ± 0.6 s for the
N195A enzyme and 12.7 ± 0.6 s− for the D195N enzyme;
1
−1
both are very close to the wild-type value of 15 s .
Rapid-Reaction Kinetics. Rapid-reaction methods were
used to determine directly the effects of the mutations on the
rate constant for flavin reduction. The N195A and D94N
1
enzymes were mixed with N -acetylspermine or spermine
anaerobically in a stopped-flow spectrophotometer at pH 9.0
and 25 °C, and the changes in the visible absorbance spectrum
of the FAD were followed. As was the case with the wild-type
Figure 2. pH profiles for Fms1 mutant enzymes. (A) Effect of pH on
1
the k /K
value with N -acetylspermine as the substrate for wild-
cat
amine
type (◆), D94N (■), and N195A (●) Fms1. The lines are from fits
1
13
of the data to eq 1. (B) Effect of pH on the k value with N -
acetylspermine for wild-type (◆), D94N (■), and N195A (●) Fms1.
The lines are from fits to eq 2. For the mutant enzymes, the
enzyme, the decrease in absorbance was biphasic with most of
cat
the absorbance change occurring in the more rapid first phase.
Only the rate constant for the fast phase exhibited a
dependence on the concentration of the amine substrate,
consistent with the mechanism of Scheme 2. The limiting rate
1
concentration of N -acetylspermine was varied at an oxygen
concentration of 1.2 mM at 25 °C. The data for the wild-type enzyme
are from ref 13.
constant for flavin reduction (k in Scheme 2) and the apparent
2
dissociation constant (K ) for binding of the amine to the
d
enzyme were determined by fitting the rate constant for the
8
693
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
Scheme 2
first phase as a function of the amine concentration to eq 4. For
both mutant enzymes, the rate constant for the slow phase was
independent of the amine concentration and much slower than
turnover, so that it is not along the catalytic pathway. Both
mutations result in significant decreases in the k values with
2
1
respect to the wild-type enzyme (Table 4). With N -
acetylspermine as the amine substrate, the value of k decreases
2
∼
600- and ∼350-fold for the N195A and D94N enzymes,
Figure 3. Refined N195A Fms1 structure (tan) superimposed onto a
SIGMA-A weighted electron density map with coefficients 2mF − F
respectively, compared to the value for the wild-type enzyme.
o
c
The apparent K values for the mutant enzymes are both ∼8-
fold lower than that for wild-type Fms1. The effect is weaker
and contoured at 1σ.
d
with spermine as a substrate. The value of k decreases ∼7- and
2
∼
20-fold while the apparent K value increases 4- and 12-fold
for the N195A and D94N enzymes, respectively.
d
Crystal Structures of N195A Fms1. Crystallization
screens were performed with both mutant proteins. For
D94N Fms1, the best crystal diffracted to 3.7 Å, which was
sufficient to establish that the mutation did not cause major
changes in the overall structure of the protein (results not
shown). In contrast, the structure of N195A Fms1 could be
determined at a resolution of 2.0 Å (Figure 3). The protein
crystallized as a tetramer, similar to the previous structures of
Figure 4. Comparison of active sites of wild-type (cyan) and N195A
tan) Fms1. TEG stands for tetraethylene glycol and Spm for
spermine. The wild-type structure is from subunit B of PDB entry
XPQ.
(
9
the wild-type enzyme in the presence of spermine and of the
1
11
H67Q mutant enzyme. The structure of N195A Fms1 lacks
the first six residues and residues 347 and 348, a mobile loop
that is also not in seen the wild-type enzyme. The overall
structure is similar to that of the wild-type enzyme, with an
rmsd of 0.437 Å for 482 α-carbons versus PDB entry 1YY5.
Most of the differences are in surface loops, especially residues
present in the crystallization solution. Tetraethylene glycol is a
reasonable analogue for the substrate spermine, although it is
one atom shorter. The nearest carbon of the tetraethylene
glycol is 3.2 Å from imidazole N3 atom of His67, consistent
with the proposed role of this residue in binding a secondary
nitrogen of the substrate. A terminal oxygen of the ligand is 3.2
Å from Cys488, suggesting that a role of this residue is to bind
N1 of the polyamine substrate. The other terminal oxygen is
too far from Asp94 for a hydrogen bond (4.1 Å), but the
additional atom of spermine could place the substrate N14
appropriately for an ionic interaction with this residue.
1
31−136 and 419−427; the latter is disordered in two of the
subunits of the mutant protein. The loop containing residues
4
56−459 missing in the structure of the wild-type enzyme is
observed in the mutant protein structure.
The structure of the active site is not disrupted by replacing
Asn195 with alanine, in that all of the active site residues,
including His67 and Asp94, occupy equivalent positions in the
mutant enzyme (Figure 4). The short alanine side chain at
position 195 of the mutant occupies the same position as the
methylene of the native aspartate. The isoalloxazine ring of the
FAD was best modeled with a slight bend of ∼7°. This bend
was not reported for the wild-type structures, nor was it
detected in our previous description of the structure of H67Q
DISCUSSION
■
13
The kinetic mechanism for Fms1 is shown in Scheme 3. The
effects of the N195A and D94N mutations on the values of the
Scheme 3
1
1
Fms1, all of which were determined at lower resolution than
9
the structure presented here (2.3−2.4 Å vs 2.0 Å).
Additional electron density is observed in the active site of
N195A Fms1 in the same position that spermine has been
reported to bind in the wild-type enzyme. This density fits well
9
with a molecule of tetraethylene glycol (Figure 3), which was
a
Table 4. Rapid-Reaction Kinetic Parameters for Wild-Type Fms1 and Mutant Enzymes
b
b
substrate
kinetic parameter
Fms1
N195A
D94N
H67Q
1
N -acetylspermine
Kd (μM)
k2 (s−1)
Kd (μM)
k2 (s−1)
484 ± 83
5490 ± 90
23 ± 8
60 ± 10
9.0 ± 0.2
84 ± 20
16.4 ± 0.6
73 ± 17
23 ± 1
58 ± 3
3300 ± 500
2.1 ± 0.5
15.6 ± 0.7
282 ± 38
6.10 ± 0.25
spermine
126 ± 3
a
b
Determined at pH 9.0 and 25 °C. From ref 11.
8
694
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
individual rate constants in the mechanism can be determined
from the steady-state and rapid-reaction analyses described
is lined primarily with backbone atoms. One exception is
Cys488; on the basis of the position of the tetraethylene glycol
in the active site of the mutant enzyme, the thiol of this residue
could interact with the uncharged N1 atom of the substrate.
here. The apparent K value for polyamine binding and the rate
d
constant for flavin reduction, k , were measured in stopped-flow
2
analyses of the reductive half-reaction in the absence of oxygen.
The second-order rate constant for the reaction of the reduced
flavin−product complex with oxygen is equal to the k /K
The k /K
−pH profiles of wild-type Fms1 support a
cat
amine
requirement that N1 and N4 be neutral for flavin reduction,
while N9 and N12 must be positively charged. The residues
in the Fms1 active site most likely to have pK values between 7
and 10 are His67 and Asp94; the finding that mutating these
residues does not affect the k /K
with this requirement. The higher k /K value for N -
acetylspermine is consistent with this requirement for a neutral
N1 and with the lack of negatively charged amino acid residues
that could interact with a positively charged N1. Interactions of
13
cat
O
2
13
value determined in steady-state kinetic analyses. With both
mutant enzymes, the effects of the mutation are greater on the
reductive half-reaction than on the oxidative half-reaction. This
a
−pH profiles is consistent
cat
amine
11
1
11
resembles the effects of mutating His67. Mechanistic studies
of amine-oxidizing flavoproteins, including the polyamine-
oxidizing enzymes, are most consistent with reaction involving
direct transfer of a hydride from the uncharged carbon−
cat
M
1
1
9−21
the neutral acetyl group at N1 of N -acetylspermine with the
hydrogen bond of the substrate to the flavin cofactor,
22
enzyme are thus likely to be mainly hydrophobic interactions
that would not be relevant for interactions with spermine. The
difference in interactions between the substrate tunnel and the
substrate N1 atom and its substituents provides a reasonable
explanation for the slower dissociation of the acetylated
substrate and product from the enzyme.
although other mechanisms have been proposed. In such a
mechanism, there is no need for an active site base because
catalysis requires that the substrate bind with the reacting
23−25
nitrogen in the unprotonated form.
Instead, the role of
active site residues is to properly position the substrate for
oxidation. These results are consistent with such a role for both
Asn195 and Asp94.
The D94N and N195A mutations significantly alter both the
1
K and k values with spermine. The D94N mutation has the
Both N -acetylspermine and spermine were used as
d
2
greater effect, increasing the K value for spermine by 1 order of
substrates in these analyses. The relative k /Kamine values for
d
cat
magnitude and decreasing the rate constant for flavin reduction
by 20-fold. The N195A mutation results in more modest
although still significant changes in both parameters. The
changes in the k /K value for spermine reflect these
changes in k and K . The effects of both mutations on all three
the two substrates suggest that wild-type Fms1 has only a slight
13
preference for the acetylated polyamine (Table 2), placing it
intermediate in specificity between the mammalian PAOs and
2
3,26,27
1
SMOs.
However, the rate constant for oxidation of N -
cat
amine
acetylspermine by Fms1 is ∼200-fold greater than the k value
2
d
2
kinetic parameters are significantly lower than the previously
reported effects of the H67Q mutation (K = 3.3 mM, and k =
for oxidation of spermine (Table 4). This difference in intrinsic
rate constants is not reflected in the k /K
values because
d
2
cat
amine
−1
11
1
2
.1 s ), consistent with the interaction with N4, the nitrogen
N -acetylspermine has a strong forward commitment to
catalysis with wild-type Fms1 because of the rate constant for
dissociation of this substrate from the oxidized enzyme being
in the bond being oxidized, being much more critical for
reactivity with this slow substrate. As noted above, the lack of
an acetyl group on N1 of spermine likely precludes interactions
11
smaller than the rate constant for flavin reduction. As a result,
1
1
the k /K
value for N -acetylspermine with wild-type Fms1
that stabilize the binding of N -acetylspermine. The magnitudes
cat
amine
reflects the second-order rate constant for substrate binding
rather than the rate constant for the chemical step. The kcat
value for this substrate similarly reflects the rate constant for
release of the oxidized amine from the oxidized enzyme rather
than for chemistry. Thus, changes in the steady-state kinetic
of the changes in the kinetic parameters for the D94N and
N195A enzymes are consistent with the loss of the proposed
interactions with the nonreacting nitrogen.
Because of the significant forward commitment to catalysis of
1
wild-type Fms1 with N -acetylspermine as the substrate,
1
parameters for N -acetylspermine for the mutant enzymes do
quantitative interpretation of the effects of the mutation with
this substrate is more difficult. Significant forward commit-
not fully reflect the effects of the mutations on individual rate
constants. In contrast, the more slowly oxidized substrate
spermine appears to have very little commitment to catalysis,
and product release is only approximately twice as fast as flavin
reduction for the wild-type enzyme; therefore, changes in
steady-state kinetic parameters with spermine more accurately
reflect changes in intrinsic kinetic parameters.
ments alter the K
experiments just as they alter K
compounded in this case by the very high k
value obtained from rapid-reaction
d
28−30
values.
This problem is
M
1
value with N -
2
acetylspermine as the substrate for Fms1, which is too fast to
measure at 25 °C directly and was only estimated by
1
1
extrapolation from measurements at 5−15 °C. Thus, the K
d
The structure of N195A Fms1 establishes clearly that the
only significant change in the mutant protein is the loss of the
amide moiety of Asn195. Thus, the effects of this mutation can
be attributed solely to local effects, primarily the loss of
hydrogen bonds between this residue and the substrate and
with both His67 and Asp94. The active site of Fms1 is a U-
shaped tunnel open to the solvent at both ends and passing by
value for the wild-type enzyme in Table 4 with this substrate is
not a reliable measure of the actual equilibrium constant for
binding of N -acetylspermine to wild-type Fms1. Still,
1
comparison of the k and K values for the three mutant
2
d
proteins suggests that mutagenesis of Asp94 or Asn195 has a
larger effect with this faster substrate than with spermine. This
1
may reflect more optimized interactions with N -acetylsper-
9
the flavin isoalloxazine ring. The active site of maize PAO is
mine that are more sensitive to any slight perturbation. In
8
similar, so that this shape is likely a common feature of PAOs
addition, both mutations result in larger effects on binding and
1
and SMOs. Within the Fms1 active site, His67, Asn195, and
Asp94 are appropriately placed to interact with N4 and N12 of
the polyamine substrate, as shown in Figure 1. The substrate
N1 atom extends up the tunnel toward the solvent. The tunnel
catalysis with N -acetylspermine than does mutagenesis of
−
1
11
His67 (K = 23 μM, and k = 58 s ). This suggests that the
d
2
1
interactions of the N -acetyl moiety with the protein
compensate for the weaker interaction with N4 in the H67Q
8
695
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
enzyme, but any such effect is attenuated for Asn195 and
Asp94, residues that interact with the more distant N12.
The D94N mutation results in a 4-fold decrease in the
second-order rate constant for the reaction of oxygen with the
complex of the reduced enzyme and oxidized amine, while the
N195A mutation does not appear to have any effect. The rate
constant for the reaction of reduced flavoproteins with oxygen
is very sensitive to the presence of positive charges in the
vicinity of the isoalloxazine ring; these can be supplied by the
Advanced Photon Source on the Northeastern Collaborative
Access Team beamlines, which are supported by grants from
the National Center for Research Resources (5P41RR015301-
10) and the National Institute of General Medical Sciences (8
P41 GM103403-10) from the National Institutes of Health.
Use of the Advanced Photon Source, an Office of Science User
Facility operated for the U.S. Department of Energy (DOE),
Office of Science, by Argonne National Laboratory, was
supported by the U.S. DOE under Contract DE-AC02-
06CH11357.
31−34
protein or the substrate.
For all three mutations of Fms1
in Table 4, there is a correlation between the effect of the
Notes
mutation on the K value for spermine and the effect on the
d
The authors declare no competing financial interest.
k /K value, with both parameters being affected the most by
cat
O
2
mutation of His67 (k /KO2 = 28 mM− s ) and the least by
1
−1 11
ACKNOWLEDGMENTS
cat
■
This material is based upon work supported in part by the
Department of Veterans Affairs, Veterans Health Adminis-
tration, Office of Research Development, Biomedical Labo-
ratory Research and Development. Support for the X-ray
Crystallography Core Laboratory by the University of Texas
Health Science Center at San Antonio Executive Research
Committee and the Cancer Therapy Research Center is
gratefully acknowledged.
mutation of N195A. If the positive charge on the oxidized
amine contributes to the catalysis of the reaction of the reduced
enzyme with oxygen, a disruption of the binding of the product
would likely disrupt the overall electrostatics of the active site.
Given the similar structures of the polyamine reactant and
product (Scheme 1), mutations that alter the binding of the
substrate would be expected also to alter the binding of the
oxidized amine product.
The k /K −pH profile for amine substrates for Fms1 has
cat
M
ABBREVIATIONS
PAO, polyamine oxidase; SMO, spermine oxidase; PDB,
Protein Data Bank.
been attributed to the need for the form of the substrate in
which only N8 and N12 are charged, because the pH optimum
matches the pH at which the doubly charged form of the
■
13
polyamine substrate is maximal. The lack of a change in the
general shape or pH optimum of the k /K
ADDITIONAL NOTE
In PDB entry 1XPQ, Asn195 is shown with the amide
−pH profile in
cat
amine
■
a
the mutant proteins is consistent with this pH profile reflecting
the protonation state of the substrate rather than any amino
nitrogen interacting with His67 and the oxygen interacting with
a carboxylate oxygen of Asp65. In response to a suggestion by a
reviewer, we are showing the reverse orientation for Asn195,
because a hydrogen bond between the amide nitrogen of
Asn195 and a carboxylate oxygen of Asp94 is more chemically
reasonable.
acid residue in the protein. The k /K
become narrower in the mutant proteins. Significant external
−pH profile does
cat
amine
forward commitments to catalysis will perturb pK values
a
35
outward, resulting in a broader pH profile. The narrower
profiles with the mutant proteins can be attributed to a decrease
in the commitment as the value of k decreases. The k −pH
2
cat
1
profile for wild-type Fms1 with N -acetylspermine reflects the
REFERENCES
■
pH dependence of k . The similar shapes of the k profiles for
5
cat
(1) Wallace, H. M., Fraser, A. V., and Hughes, A. (2003) A
perspective of polyamine metabolism. Biochem. J. 376, 1−14.
(2) Casero, R. A., and Woster, P. M. (2009) Recent Advances in the
Development of Polyamine Analogues as Antitumor Agents. J. Med.
Chem. 52, 4551−4573.
3) Casero, R. A., Jr., and Marton, L. J. (2007) Targeting polyamine
metabolism and function in cancer and other hyperproliferative
diseases. Nat. Rev. Drug Discovery 6, 373−390.
4) Pegg, A. E. (2009) Mammalian polyamine metabolism and
the mutant proteins rule out Asp94 as the source of the pKa;
His67 was similarly eliminated as a potential source by
1
1
mutagenesis. One possibility is that this pK reflects the
a
protonation state of the amine product. The change in the pKa
in the two mutant proteins can be attributed to a change in the
(
rate-determining step so that k becomes closer to the rate
2
constant for product release.
(
These results establish that both Asn195 and Asp94 are
important for properly positioning the polyamine substrate for
oxidation. The quantitative effects of mutating these residues
depend on the polyamine substrate used, in that the acetyl
function. IUBMB Life 61, 880−894.
5) White, W. H., Gunyuzlu, P. L., and Toyn, J. H. (2001)
(
Saccharomyces cerevisiae Is Capable of de Novo Pantothenic Acid
Biosynthesis Involving a Novel Pathway of β-Alanine Production from
Spermine. J. Biol. Chem. 276, 10794−10800.
(6) Landry, J., and Sternglanz, R. (2003) Yeast Fms1 is a FAD-
utilizing polyamine oxidase. Biochem. Biophys. Res. Commun. 303, 771−
1
moiety in N -acetylspermine provides interactions not available
to spermine. Moreover, both residues appear to be less critical
than His67, a residue conserved among PAOs and SMOs.
7
76.
(
7) Federico, R., Ercolini, L., Laurenzi, M., and Angelini, R. (1996)
AUTHOR INFORMATION
Corresponding Author
Phone: (210) 567-8264. Fax: (210) 567-8778. E-mail:
fitzpatrick@biochem.uthscsa.edu.
■
*
Oxidation of acetylpolyamines by maize polyamine oxidase.
Phytochemistry 43, 339−341.
(8) Binda, C., Coda, A., Angelini, R., Federico, R., Ascenzi, P., and
Mattevi, A. (1999) A 30 Å long U-shaped catalytic tunnel in the crystal
structure of polyamine oxidase. Structure 7, 265−276.
Funding
(
9) Huang, Q., Liu, Q., and Hao, Q. (2005) Crystal structures of
This work was supported in part by National Institutes of
Health Grant R01 GM058698 (to P.F.F.) and The Welch
Foundation Grants AQ-1399 (to P.J.H.) and AQ1245 (to
P.F.F.). This work is based upon research conducted at the
Fms1 and its complex with spermine reveal substrate specificity. J. Mol.
Biol. 348, 951−959.
(10) Binda, C., Mattevi, A., and Edmondson, D. E. (2002) Structure-
function relationships in flavoenzyme dependent amine oxidations. A
8
696
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697

Biochemistry
Article
comparison of polyamine oxidase and monoamine oxidase. J. Biol.
Chem. 277, 23973−23976.
enzyme concentration on the kinetics. J. Biol. Chem. 257, 12916−
12923.
(
11) Adachi, M. S., Taylor, A. B., Hart, P. J., and Fitzpatrick, P. F.
(30) Porter, D. J. T., and Bright, H. J. (1980) Interpretation of the
pH dependence of flavin reduction in the L-amino acid oxidase
reaction. J. Biol. Chem. 255, 2969−2975.
(
2012) Mechanistic and Structural Analyses of the Role of His67 in
the Yeast Polyamine Oxidase Fms1. Biochemistry 51, 4888−4897.
12) Tormos, J. R., Henderson Pozzi, M., and Fitzpatrick, P. F.
2012) Mechanistic studies of the role of a conserved histidine in a
mammalian polyamine oxidase. Arch. Biochem. Biophys. 528, 45−49.
13) Adachi, M. S., Torres, J. M., and Fitzpatrick, P. F. (2010)
(
(31) Roth, J. P., and Klinman, J. P. (2003) Catalysis of electron
(
transfer during the activation of O
oxidase. Proc. Natl. Acad. Sci. U.S.A. 100, 62−67.
32) Gadda, G. (2012) Oxygen Activation in Flavoprotein Oxidases:
The Importance of Being Positive. Biochemistry 51, 2662−2669.
33) Zhao, G., Bruckner, R. C., and Jorns, M. S. (2008) Identification
2
by the flavoprotein glucose
(
(
Mechanistic Studies of the Yeast Polyamine Oxidase Fms1: Kinetic
Mechanism, Substrate Specificity, and pH Dependence. Biochemistry
(
of the Oxygen Activation Site in Monomeric Sarcosine Oxidase: Role
4
9, 10440−10448.
14) Cleland, W. W. (1979) Statistical analysis of enzyme kinetic
data. Methods Enzymol. 63, 103−138.
15) Otwinowski, Z., and Minor, W. (1997) Processing of X-ray
diffraction data collected in oscillation mode. Methods Enzymol. 276,
07−326.
16) McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M.
of Lys265 in Catalysis. Biochemistry 47, 9124−9135.
(
(34) Henderson Pozzi, M., and Fitzpatrick, P. F. (2010) A lysine
conserved in the monoamine oxidase family is involved in oxidation of
the reduced flavin in mouse polyamine oxidase. Arch. Biochem. Biophys.
(
4
(
98, 83−88.
3
(
35) Cleland, W. W. (1986) Enzyme kinetics as a tool for
determination of enzyme mechanisms. In Investigation of Rates and
Mechanism (Bernasconi, C. F., Ed.) 4th ed., Vol. 6, Part I, pp 791−870,
John Wiley & Sons, New York.
D., Storoni, L. C., and Read, R. J. (2007) Phaser crystallographic
software. J. Appl. Crystallogr. 40, 658−674.
(
́
17) Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I.
W., Echols, N., Headd, J. J., Hung, L.-W., Kapral, G. J., Grosse-
Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R.
J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P.
H. (2010) PHENIX: A comprehensive Python-based system for
macromolecular structure solution. Acta Crystallogr. D66, 213−221.
(
18) Emsley, P., and Cowtan, K. (2004) Coot: Model-building tools
for molecular graphics. Acta Crystallogr. D60, 2126−2132.
19) Fitzpatrick, P. F. (2010) Oxidation of amines by flavoproteins.
Arch. Biochem. Biophys. 493, 13−25.
20) Henderson Pozzi, M., Gawandi, V., and Fitzpatrick, P. F. (2009)
(
(
Mechanistic Studies of para-Substituted N,N′-Dibenzyl-1,4-diamino-
butanes as Substrates for a Mammalian Polyamine Oxidase.
Biochemistry 48, 12305−12313.
(
21) Ralph, E. C., Hirschi, J. S., Anderson, M. A., Cleland, W. W.,
Singleton, D. A., and Fitzpatrick, P. F. (2007) Insights into the
mechanism of flavoprotein-catalyzed amine oxidation from nitrogen
isotope effects on the reaction of N-methyltryptophan oxidase.
Biochemistry 46, 7655−7664.
(
22) MacMillar, S., Edmondson, D. E., and Matsson, O. (2011)
Nitrogen Kinetic Isotope Effects for the Monoamine Oxidase B-
2
Catalyzed Oxidation of Benzylamine and (1,1- H )Benzylamine:
2
Nitrogen Rehybridization and CH Bond Cleavage Are Not Concerted.
J. Am. Chem. Soc. 133, 12319−12321.
(
23) Henderson Pozzi, M., Gawandi, V., and Fitzpatrick, P. F. (2009)
pH Dependence of a Mammalian Polyamine Oxidase: Insights into
Substrate Specificity and the Role of Lysine 315. Biochemistry 48,
1
(
(
508−1516.
24) Dunn, R. V., Marshall, K. R., Munro, A. W., and Scrutton, N. S.
2008) The pH dependence of kinetic isotope effects in monoamine
oxidase A indicates stabilization of the neutral amine in the enzyme-
substrate complex. FEBS J. 275, 3850−3858.
(
25) Kurtz, K. A., Rishavy, M. A., Cleland, W. W., and Fitzpatrick, P.
F. (2000) Nitrogen isotope effects as probes of the mechanism of D-
amino acid oxidase. J. Am. Chem. Soc. 122, 12896−12897.
(
26) Adachi, M. S., Juarez, P. R., and Fitzpatrick, P. F. (2010)
Mechanistic Studies of Human Spermine Oxidase: Kinetic Mechanism
and pH Effects. Biochemistry 49, 386−392.
(
27) Wu, T., Yankovskaya, V., and McIntire, W. S. (2003) Cloning,
sequencing, and heterologous expression of the murine peroxisomal
flavoprotein, N1-acetylated polyamine oxidase. J. Biol. Chem. 278,
2
(
0514−20525.
28) Strickland, S., Palmer, G., and Massey, V. (1975) Determination
of dissociation constants and specific rate constants of enzyme-
substrate (or protein-ligand) interactions from rapid reaction kinetic
data. J. Biol. Chem. 250, 4048−4052.
(
29) Fitzpatrick, P. F., and Massey, V. (1982) The kinetic mechanism
of D-amino acid oxidase with D-α-aminobutyrate as substrate: Effect of
8
697
dx.doi.org/10.1021/bi3011434 | Biochemistry 2012, 51, 8690−8697