Quantitative analysis and functional evaluation of zinc ion in the D-hydantoinase from Pseudomonas putida YZ-26

Article abstract of DOI:10.1007/s10534-009-9267-7

D-Hydantoinase (HDT) is a metal-dependent enzyme that is widely used in industrial bioconversion to D-amino acids as valuable intermediates in the fields of food, pharmaceutical industry and agriculture. In this report, we prepared apo-HDT (metal-removed HDT) and Zn²⁺-HDT (Zn ²⁺-added HDT) in vitro from a recombinant HDT (re-HDT) expressed in E. coli. The Zn²⁺-HDT and re-HDT contain 2.17 and 0.95 mol Zn ²⁺ per mol subunit, respectively, and they have comparable enzymatic activities. In contrast, the apo-HDT only retains 0.04 mol Zn²⁺ per mol subunit with less than 10% activity, compared with the re-HDT. When the apo-HDT was reconstituted with ZnCl₂, the enzymatic activity recovery was about 75%. Moreover, the fluorescence intensity, circular dichroism spectra and thermo-stability of the apo-HDT and Zn²⁺-HDT are quite different from those of the re-HDT. These data suggest that the re-HDT may have two Zn²⁺-binding sites, one is an intrinsic or tight-binding site (zinc-α) essential for its activity and the other is a vacant or loose-binding site (zinc-β) possibly non-essential for the activity. Springer Science+Business Media.

Full text of DOI:10.1007/s10534-009-9267-7

Biometals (2010) 23:71–81
DOI 10.1007/s10534-009-9267-7
Quantitative analysis and functional evaluation of zinc ion
in the D-hydantoinase from Pseudomonas putida YZ-26
•
•
•
Xueyao Zhang Jingming Yuan Lixi Niu
Aihua Liang
Received: 23 November 2008 / Accepted: 9 September 2009 / Published online: 5 November 2009
Ó Springer Science+Business Media, LLC. 2009
Abstract D-Hydantoinase (HDT) is a metal-depen-
dent enzyme that is widely used in industrial
bioconversion to D-amino acids as valuable interme-
diates in the fields of food, pharmaceutical industry
and agriculture. In this report, we prepared apo-HDT
(metal-removed HDT) and Zn2?-HDT (Zn2?-added
HDT) in vitro from a recombinant HDT (re-HDT)
expressed in E. coli. The Zn2?-HDT and re-HDT
contain 2.17 and 0.95 mol Zn2? per mol subunit,
respectively, and they have comparable enzymatic
activities. In contrast, the apo-HDT only retains
0.04 mol Zn2? per mol subunit with less than 10%
activity, compared with the re-HDT. When the apo-
HDT was reconstituted with ZnCl2, the enzymatic
activity recovery was about 75%. Moreover, the
fluorescence intensity, circular dichroism spectra and
thermo-stability of the apo-HDT and Zn2?-HDT are
quite different from those of the re-HDT. These data
suggest that the re-HDT may have two Zn2?-binding
sites, one is an intrinsic or tight-binding site (zinc-a)
essential for its activity and the other is a vacant or
loose-binding site (zinc-b) possibly non-essential for
the activity.
Keywords D-Hydantoinase ꢀ Zn2? analysis ꢀ
Functional evaluation
Introduction
D-Hydantoinase (HDT, EC 3.5.2.2) catalyses the
substrate 50-monosubstituted hydantoin to enantio-
merical N-carbamoyl-amino acids which, in turn, can
be chemically or enzymatically converted into the
corresponding optically active amino acids (Alten-
buchner et al. 2001). Hydantoinase is ubiquitously
found in microorganisms, plants and animals (Pozo
et al. 2002). The enzyme is also well known as
dihydropyrimidinase in higher organisms and is
responsible for dihydropyrimidine hydrolysis as the
second step in the reductive catabolism of pyrimi-
dine. Based on its chiral specificity for the substrate
and product, HDT can be divided into three types:
L-, D- and DL-configuration. A number of bacterial
D-hydantoinases with different enantioselectivities
and substrate specificities have been used in indus-
trial bioconversion to optically active D-amino
acids. As unnatural chiral products, D-amino acids
are important intermediates in the synthesis of
various b-lactam semisynthetic antibiotics, such as
X. Zhang ꢀ J. Yuan (&) ꢀ L. Niu ꢀ A. Liang
Key Laboratory of Chemical Biology and Molecular
Engineering of National Ministry of Education, Institute
of Biotechnology, Shanxi University, 92 Wu-cheng Road,
030006 Taiyuan, People’s Republic of China
e-mail: jmyuan@sxu.edu.cn
X. Zhang
e-mail: zxyletter00@163.com
123

72
Biometals (2010) 23:71–81
ampicillin, amoxicillin, aspoxicillin and cefbuperaz-
one, as well as of other products, such as antiviral
agents, artificial sweeteners, pesticides, peptide hor-
mones and pyrethroids etc. (Altenbuchner et al.
2001).
stored at -80°C in this laboratory. Hydantoin,
4-(2-hydroxyerhyl) piperazine-1-ethanesulfonic acid
(HEPES, ultra pure grade) and Tris (Hydroxymethyl)
aminomethane (Tris, ultra pure grade) were obtained
from Sigma-Aldrich (St. Louis, MO, USA). Pyridine-
2, 6-dicarboxylic acid (DPA) and EDTA were
purchased from Alfa Aesar (Ward Hill, MA, USA).
The protein purification equipment and separation
materials were from Amersham Pharmacia Biotech.
All reagents and buffer solutions used here were
prepared using ultra-pure water with more than
18.2 MX m from a Milli-Q water purification system
(Millipore Corporation, France). All other chemicals
used were of the highest purity commercially avail-
able, unless otherwise indicated.
HDT, similar to other amidohydrolases, has been
characterized as a typical metal-dependent enzyme,
either in the fundamental research or in the industrial
application (May et al. 1998a, b). It is demonstrated
that divalent metal ions, especially Zn2?, play an
important role in the catalytic process of the enzyme
(May et al. 1998a, b). In the last decades, some reports
regarding the metal dependence of HDT have been
mainly focused on the function of Zn2? and its
quantitative determination (Gojkovic et al. 2003;
Cheon et al. 2003; Kikugawa et al. 1994; Brooks
et al. 1983). Thereinto, most HDTs contain two Zn2?
ions per subunit (Gojkovic et al. 2003; Cheon et al.
2003). The first Zn2? may bind to the carbonyl oxygen
of hydantoin and subsequently polarize the carbonyl
bond, while the second Zn2? only activates the bridge
water and does not bind to the substrate. In contrast, a
few HDTs only contain one catalytic Zn2? per subunit,
such as the HDTs from bovine liver and rat liver
(Kikugawa et al. 1994; Brooks et al. 1983). No matter
how many Zn2? per subunit HDT contains, they are all
essential for the catalytic reaction. However, the
potential structural and functional roles for Zn2? have
not been well established in hydantoinase family,
although the role of Zn2? has been well established in
other amidohydrolases. Our previous work indicated
that the D-hydantoinase from Pseudomonas putida
YZ-26 was dissociated into monomer subunit from its
dimer in the presence of 1 mM Zn2? (Shi et al. 2006).
To further explore the function of Zn2? in the HDT,
we prepared re-HDT, apo-HDT and Zn-HDT to study
the relationship between its zinc content and catalytic
activity. This report preliminarily demonstrates that
the re-HDT has two different Zn2?-binding sites, a
tight one linked to its activity is filled with Zn2? and a
loose one related to its structure is vacant.
Expression and purification of the recombinant
D-hydantoinase
Recombinant plasmid pE-p479 was introduced into
E. coli BL21 and expressed in 1 l Luria-Bertani (LB)
culture supplemented with ampicillin (100 lg/ml) at
30°C for 12–14 h without adding any inducers. The
culture was harvested by centrifugation at 6,000g for
10 min at 4°C. Then pellets were suspended in
chilled 50 mM Tris–HCl buffer, pH 8.0 (TB8) and
sonicated. The resulting supernatant was purified by a
two-step purification procedure to obtain the purified
D-hydantoinase, as described previously (Zhang et al.
2008). The protein concentration was performed by
the Bradford or Folin-phenol methods with bovine
serum albumin as the assay standard.
Enzymatic assay
D-Hydantoinase activity was measured by a colori-
metric method (Niu et al. 2007). An aliquot of the
enzyme was added into the reaction mixture in a total
volume of 1.5 ml, contained 100 mM hydantoin and
50 mM TB8. After being incubated at 37°C for
30 min, the reaction mixture was terminated by
adding 0.25 ml trichloro-acetic acid (10% w/v), then
0.25 ml dimethylaminobenzaldehyde (DMBA) solu-
tion (10% w/v in 6 M HCl) and finally diluted with
distilled water to 3 ml. After centrifugation, the
product, N-carbamoylglycine, in the supernatant was
measured at 430 nm wavelength and its amount was
calculated from a standard calibration plot. One
D-hydantoinase unit is equivalent to the formation of
Materials and methods
Strain, plasmid, reagents and equipments
The host cell, Escherichia coli BL21 (DE3) and a
recombinant plasmid pE-p479 used in this study were
123

Biometals (2010) 23:71–81
73
1 lmol of N-carbamoylglycine per min under the
above assay condition.
a colorimetric Zn2?-chelator, 4-(2-pyridylazo) resor-
cinol (PAR). Re-HDT (1 lM) and Zn2?-HDT (1 lM)
in 10 mM HEPES buffer, pH 7.5 were used as the
control and sample, respectively. Buffer alone was
used as a blank. During the titration with PAR (1–
35 lM), the data were collected at 500 nm (Shi et al.
2008), and then the optical absorbance was corrected
by subtracting the trace from buffer blank.
Apo-HDT and Zn2?-HDT preparation in vitro
Apo-HDT was prepared by mixing re-HDT and
100 mM DPA to remove the intrinsic Zn2? at 25°C
for 8 h. Then the reaction mixture was thoroughly
dialyzed against 10 mM TB8 at 4°C with buffer
changes for four times to remove trace metal ions.
Zn2?-HDT was prepared with the re-HDT by addi-
tion of ten fold (ca. 0.3 mM) ZnCl2. The reaction
mixture was dialyzed as described above. Finally, the
enzymatic activity, protein concentration and Zn2?
content of each sample were examined by DMBA,
Folin-phenol and ICP-AES methods, respectively.
The apparent binding constant of the vacant zinc-b
site in re-HDT was estimated by a competition assay
with a colorimetric Zn2?-chelator, 2-carboxy-2-
hydroxy-5-(sulfoformazyl) benzene (Zinco) (Armas
et al. 2006). A typical assay is performed as follows:
an amount of 20 lM Zinco and 10 lM ZnCl2 in
10 mM HEPES buffer was added into a cuvette, and
the spectrophotometer was blanked against this
solution. Then the re-HDT was added to the cuvette
step by step, and the absorbance value was acquired
at 620 nm after sufficiently mixing.
Determination of divalent metal ions
Divalent metal ions in re-HDT, apo-HDT and Zn2?
-
HDT were determined by ICP-AES (inductively
coupled plasma-atomic emission spectrometry, an
ICP-AtomScan 16 apparatus from Thermo Jarrell Ash
Co., USA) with a direct injection. All samples were
prepared to be SDS-PAGE pure. Metals were analyzed
Circular dichroism and fluorescence
emission spectra
Circular dichroism (CD) spectra were carried out
with a recording spectropolarimeter (Jasco 810) at
25°C under a nitrogen atmosphere. CD spectrum was
conducted at an enzyme concentration of 0.2 mg/ml
in 20 mM HEPES (pH 7.4) with a 1 mm path-length
cell. The spectrum was recorded from 190 to 250 nm
at a scan rate of 50 nm min-1. A blank was also
recorded in the absence of the enzyme using the same
buffer as a control. The CD spectrum of each sample
or control was measured at least three times.
atthe following wavelengths: Zn2?: 213.856 nm; Co2?
:
236.379 nm; Mn2?: 257.690 nm; Ni2?: 231.604 nm
and Fe2?: 259.640 nm. Data were calibrated against the
standard of Zn2?, Co2?, Mn2?, Fe2? and Ni2?, respec-
tively. Metal ion content and protein concentration were
simultaneously performed each time to evaluate the
reliability of metal: protein ratio. The last external
dialyzed buffer was used as a blank in ICP-AES, and the
equivalent amount of BSA was used as a control in the
protein concentration assay.
Fluorescence emission spectra were recorded at
25°C using a spectrofluorometer (HITACHI, F-2500)
with an excitation wavelength of 295 nm. The
spectrum was scanned from 290 to 400 nm at a scan
rate of 60 nm min-1. Samples with an 8 lM enzyme
concentration were measured in a 1 cm path-length
fluorescence cuvette.
Thermo-stability
The thermo-stability of re-HDT and Zn2?-HDT was
determined by pre-incubation in 50 mM TB8 at
various temperatures (from 35 to 60°C) for 60 min.
Samples were then chilled on ice for at least 5 min,
and then the residual activity was performed under
the standard assay condition.
Results
Preparation of re-HDT, apo-HDT and Zn2?-HDT
Apparent binding constant of zinc-b site
The re-HDT was purified by a two-step purification
procedure from E. coli as described previously
(Zhang et al. 2008). Zn2?-HDT and apo-HDT was
The apparent binding constant of the zinc-b site in
Zn2?-HDT was estimated by a competition test with
123

74
Biometals (2010) 23:71–81
prepared as described in ‘‘Materials and methods’’.
To avoid the inactivation of the enzyme during the
dialysis and ensure the reliability of the subsequent
metal analysis, three samples (re-HDT, apo-HDT and
Zn2?-HDT) must be treated simultaneously for each
test. All three species of HDTs show high purity and
exhibit the same molecular weight on SDS-PAGE
(Fig. 1). Moreover, the Zn2?-HDT activity reaches a
similar level as the re-HDT (9.5 U/mg), while the
apo-enzyme has little activity compared with re-
HDT.
Inactivation of re-HDT by metal chelators
The chelator such as EDTA or pyridine-2,6-dicar-
boxylate (DPA) is generally used as a reagent to
remove metal ions in a protein, accompanied by
losing the activity or changing the conformation.
However, when the re-HDT was incubated with
EDTA or DPA, its inactivation profile was quite
different as shown in Fig. 2. In the presence of
100 mM EDTA, the re-HDT is very stable, and its
activity retains more than 90% after 8 h incubation.
The effect of other chelators, such as 1,10-phenan-
throline or 8-HQSA on the re-HDT is roughly the
same as that of EDTA (data not shown). This is likely
due to that the intrinsic Zn2? in the re-HDT is too
tight to be removed by above chelators. However, the
inactivation of the re-HDT obviously occurred in the
presence of 100 mM DPA. The enzymatic activity
gradually decreased to less than 10% after 8 h
Fig. 2 Inactivation of re-HDT in the presence of chelator,
EDTA or DPA. (d) Control, (s) EDTA, () DPA. The
enzyme was incubated with 100 mM DPA or 100 mM EDTA
at 25°C. At time intervals, aliquots were removed to measure
the residual activity
incubation, indicating that the inactivation was
related to some metal ions. This result is similar to
that of the hamster dihydroorotase domain, a member
of the hydantoinase family, in which the metal
chelating ability of EDTA is weaker than that of DPA
(Huang et al. 1999).
Quantitative determination of divalent
metal ions in the re-HDT
Based on our experiences, a repeated dialysis proce-
dure was quite efficient to thoroughly remove free
trace metal ions. Three samples of pure HDTs
(re-HDT, apo-HDT and Zn2?-HDT) were dialyzed
simultaneously and extensively with 10 mM TB8
prepared by Milli-Q ultra-pure water and ultra-pure
reagents as described in ‘‘Materials and methods’’. In
order to confirm there is no metal ion contamination
in samples, equal amounts of BSA as a control and
the final external dialysate as a blank were measured
by same procedures. ICP-AES analysis shows in
Table 1 that the re-HDT contains 0.95 mol of Zn2?
per subunit, equivalent to a protein: Zn2? ratio of 1:1.
However, the apo-HDT treated with DPA only has
0.04 mol of Zn2? per subunit, indicating that the
intrinsic Zn2? (so called zinc-a) was thoroughly
removed by this dialysis procedure. On the contrary,
Fig. 1 SDS-PAGE analysis of re-HDT, apo-HDT and Zn2?
-
HDT. Lane M: Molecular mass marker (phosphorylase 97 kDa,
bovine serum albumin 66 kDa, ovalbumin, 45 kDa, carbonic
anhydrase, 30 kDa); Lane 1: Re-HDT; Lane 2: Apo-HDT;
Lane 3: Zn2?-HDT
123

Biometals (2010) 23:71–81
75
Reactivation of apo-HDT by divalent metal ion
Table 1 Divalent metal ion content in various types of D-
hydantoinasesa
To further explore the effect of Zn2? on the activity
of the apo-HDT it was titrated with ZnCl2. As shown
in Fig. 3, the activity of the apo-HDT is gradually
increased until the ratio of Zn2? to enzyme is 1:1.
The recovery of enzyme activity is about 75% of the
re-HDT. Beyond 1:1 ratio, the enzymatic activity no
Enzyme or control Metal ion (mol)/monomer enzyme(mol)
Mn2? Co2? Ni2?
Zn2? Fe2?
Re-HDT
Zn-HDT
Apo-HDTb
BSAc
0.09
0.03
0.06
0.03
0.01
0.02
0.95
2.17
0.08
0.02
\0.01 \0.01 \0.01
0.04 \0.01
longer increases, even in the presence of excess Zn2?
.
\0.01 \0.01 \0.01 \0.01 \0.01
External dialysatec \0.01 \0.01 \0.01 \0.01 \0.01
In order to investigate reactivation of the apo-HDT,
the above metal ions were added individually. As
shown in Fig. 3, the reactivation capability of the apo-
a
The above data were carried out in triplicate by ICP-AES
b
Apo-HDT is the re-HDT incubated with 100 mM DPA at
25°C for 8 h
HDT seems in the sequence of Co2? [ Mn2? [ Ni2?
,
implying that these three metal ions are different from
Zn2? not only on the activation level but also on the
binding characteristics.
c
Blanks are the equal amount of BSA and the final external
dialysate of different samples
the Zn2? content of Zn2?-HDT is up to 2.17 mol per
monomer enzyme with no change on its activity. The
result shows that Zn2? may occupy other binding site
(so called zinc-b), which may be nonessential for
enzymatic activity. Moreover, the quantity of other
metal ions, such as Co2?, Ni2?, Mn2? and Fe2? are
negligible in comparison with Zn2? as indicated in
Table 1.
Thermo-stability of Zn2?-HDT
The thermo-stability of Zn2?-HDT (Fig. 4) is higher
than that of the re-HDT under the identical condition. A
possible explanation is that the zinc-b site filled with
Zn2? results in the conformational change of the
protein to enhance its thermo-stability. In fact, it is not
Fig. 3 Reactivation of apo-HDT in the presence of different
metal ions. Apo-HDT was mixed with various molar ratios of
Zn2?, Ni2?, Co2? and Mn2?, respectively, at 4°C for 2 h, and
then 20 ll each was added into 1.5 ml reaction system for the
activity assay. The activity of re-HDT was normalized to 100%
123

76
Biometals (2010) 23:71–81
re-HDT can be calculated by resolving Eq. 1 for
Kapp Znꢁ
.
2þHDT
ð
Þ
Þ
Â
Ã
Znꢁ2þHDT ½Zincoꢂ
Kapp Znꢁ
2þHDT
ð
¼
ð1Þ
KappðZnꢁZincoÞ
½re-HDTꢂ½Zn-Zincoꢂ
It is shown in Fig. 5a that the optical absorption at
620 nm is gradually decreased with increased con-
centration of the re-HDT. The absorption change
(D620 nm) reflects the transfer of Zn2? from the
Zn2?-Zinco complex to the loose zinc-b site in the
re-HDT, therefore, the following Eqs. 2–7 would be
produced.
Â
Ã
½re-HDTꢂ ¼ ½re-HDTꢂaddꢁ Zn2þ-HDT
ð2Þ
ð3Þ
ð4Þ
ð5Þ
Â
Ã
Zn2þ-HDT ¼ D½Zn-Zincoꢂ ¼ A620=e620
½Zn-Zincoꢂoriginal¼ 10 lM
½Zincoꢂtotal¼ 20 lM
Fig. 4 Comparison of thermo-stability between re-HDT and
Zn2?-HDT. (d) Re-HDT, (s) Zn2?-HDT. The enzyme was
pre-incubated at different temperatures in a 50 mM Tris–HCl
buffer, pH 8.0. An aliquot of the pre-incubated enzyme was
removed and chilled on ice for 5 min. Then the residual
activity was assayed as described in ‘‘Materials and methods’’
½Zincoꢂ ¼ ½Zincoꢂtotalꢁ½Zn-ZincoꢂoriginalþD½Zn-Zincoꢂ
ð6Þ
½Zn-Zincoꢂ ¼ ½Zn-ZincoꢂoriginalꢁD½Zn-Zincoꢂ
ð7Þ
Finally, by the known binding constant of
Zn-Zinco, the apparent zinc-b binding constant of
Zn2?-HDT can be calculated at the level of 5 9
104 M-1. The result shows that Zn2? in Zn-Zinco can
be transferred into the vacant zinc-b site of the re-
HDT, but its binding constant is not very high
compared to the one for zinc-a site.
unusual that the addition of metal ions to an enzyme
solution results in changes of the protein conformation
and thermo-stability. For example, the thermo-stability
of the PhoACY from Pyrococcus horikoshii was
improved when it was incubated with Zn2?, Mn2? or
Ni2? and the presence of Mn2? or Co2? can increase
the thermo-stability of the Burkholderia pickettii
D-hydantoinase (Xu et al. 2003; Vieille and Zeikus
2001). Therefore, the enhancement of the Zn2?-HDT
thermo-stability may be dependent on the binding
feature of zinc-b site.
Similarly, the zinc-b binding constant of the Zn2?
-
HDT was measured with the colorimetric Zn2?-chela-
tor, 4-(2-pyridylazo) resorcinol (PAR). The titration
experiment (Fig. 5b) shows that the reagent, PAR only
interacts with zinc-b site of the Zn2?-HDT, but not with
the zinc-a site of the re-HDT at the same concentration.
It was reported that an excess PAR mainly became a
complex with Zn2? in a 2 to 1 stoichiometry
(Zn(PAR)2), which produces a distinctive absorption
at500 nm (e = 66,000 M-1 cm-1) and has an apparent
binding constant of 2 9 1012 M-1(Shaw et al. 1991).
Apparent binding constant of zinc-b site
The apparent binding constant between Zn2? and a
protein can be estimated by a competition test with a
Zn2? transferring reagent. The apparent Zn2?-bind-
ing constant of the loose zinc-b site in re-HDT was
estimated by the colorimetric Zn2?-chelator, Zinco.
Therefore, the apparent binding constant of the Zn2?
-
The reagent, Zinco can form 1:1 complex with Zn2?
,
HDT can be calculated by resolving the Eq. 8 for
Kapp Zn
, just as in the case of Zinco.
which has a distinct absorption at 620 nm (e =
23,500 M-1 cm-1) with a binding constant (Kapp) of
7.9 9 104 M-1 (Armas et al. 2006). Therefore,
the binding constant of the loose zinc-b site in the
2þꢁHDT
ð
Þ
Þ
Â
Ã
2
Znꢁ2þHDT ½PARꢂ
Kapp Znꢁ
2þHDT
ð
Â
Ã
¼
ð8Þ
Kapp ZnðPARÞ
½re-HDTꢂ ZnðPARÞ2
ð
Þ
2
123

Biometals (2010) 23:71–81
77
Â
Ã
½re-HDTꢂ ¼ ZnðPARÞ2 ¼ A500=e500
ð9Þ
ð10Þ
ð11Þ
Â
Ã
½PARꢂ ¼ ½PARꢂaddꢁ2 ZnðPARÞ2
Â
Ã
Zn2þ-HDT
¼ 1 lM
original
Â
Ã
Â
Ã
Â
Ã
Zn2þ-HDT ¼ Zn2þ-HDT
ꢁ ZnðPARÞ2
original
ð12Þ
Finally, according to the known binding constant
of Zn(PAR)2, the apparent binding constant of zinc-b
site in Zn2?-HDT should be around 5 9 1012 M-1
.
This experiment demonstrates that the zinc-b in the
Zn2?-HDT is very hard to be transferred into PAR. In
other words, as soon as the vacant zinc-b site is filled
with Zn2?, the interaction strength between Zn2? and
the protein is too tight to be removed.
Circular dichroism and fluorescence spectra
Circular dichroism (CD) is often used to evaluate
whether the secondary and tertiary structures of a
protein are changed under various conditions.
Figure 6a shows the far-UV CD spectra of the
re-HDT in the absence and presence of Zn2?. They
display the characteristic spectra of a mainly b-sheet
structure with a negative peak at 215 nm (Mishima
et al. 2006). The data is consistent with our previous
work that the three dimensional structure of the re-
HDT is rich in b-sheet (Zhang et al. 2008). The
negative peak is diminished by gradually adding
Zn2? ion, indicating that binding of Zn2? to zinc-b
site induces a modification of the re-HDT secondary
structure, resulting in the decrease of b-sheet content
in the enzyme.
Fig. 5 a Determination of the binding constant at the vacant
zinc-b site in re-HDT by Zinco. Zn-Zinco complex (20 lM
Zinco, 10 lM ZnCl2) was titrated with re-HDT (1–21 lM) in
10 mM HEPES, pH 7.5. The absorbent change at 620 nm is
used as an index for Zn2? transfer from Zn-Zinco to re-HDT. b
Determination of the binding constant of the zinc-b site in
Zn2?-HDT by PAR. (d) Zn2?-HDT, () Re-HDT. 1 lM
Zn2?-HDT or 1 lM re-HDT was titrated with PAR (1–35 lM)
in 10 mM HEPES, pH 7.5, respectively. The increase of the
absorbance at 500 nm was regarded as the formation of
Zn(PAR)2 complex
Fluorescence spectra reflect the state of intrinsic
fluorescent residues as a reliable indicator of protein
conformation changes. Figure 6b shows that the fluo-
rescence spectra of the re-HDT exhibit a maximum
emission at about 337 nm, which is a characteristic of
tryptophan residues buried inside a protein (Vlasova
and Ugarova 2007; Zhao et al. 2004). Although there is
no obvious shift (\5 nm) in maximum emission
wavelength, significant differences in fluorescent
intensity are observed among these three different
species of HDTs. Fluorescence intensity of the Zn2?
-HDT is increased by 27%, while that of the apo-HDT
is quenched by 28%, compared with the fluorescence
intensity of the re-HDT. A possible explanation is that
Zn2?-binding on the re-HDT affects the structure of the
Figure 5b shows that the optical absorption at
500 nm is gradually increased with the addition of
PAR due to the formation of Zn(PAR)2 complex. The
absorbance increment at 500 nm implies the transfer
of Zn2? at zinc-b site from the Zn2?-HDT to PAR.
The same equation can be obtained as follows:
123

78
Biometals (2010) 23:71–81
However, the environment polarity of tryptophan
residues may not be changed; therefore there is no
obvious shift of the maximum emission wavelength.
A
Homology modeling
SWISS-MODEL server was used to construct the
three-dimensional structure model of the re-HDT via
the following steps. Human dihydropyrimidinase
(PDB ID: 2VR2) was selected as the template by
Swissmodel and Modeller programs from templates
library. The primary sequence identity between the
template protein and the re-HDT from P. putida YZ-
26, is 52.1%, which is much higher than 30% limit
generally considered to be the threshold limit for an
accurate homology modeling (Marti-Renom et al.
2000). Then, the sequence of the re-HDT was aligned
against the template structure, and the project file was
submitted to SWISS-MODEL server. After the
calculation, the resulting homology structure was
evaluated employing the PyMOL program. It is
found that the homology structure of the re-HDT is
not only extremely similar to its template structure as
expected, but also quite similar to the dihydropyrim-
idinase from yeast Saccharomyces kluyveri
B
˚
(r.m.s.d. = 1.3 A), which has one tight zinc-a (five-
coordinate) and one loose zinc-b (five-coordinate).
The superposition (Fig. 7b) and the sequence align-
ment (Fig. 7a) demonstrate that the location of
Zn2?-binding residues in the dihydropyrimidinase is
conserved in the re-HDT. Overlap of Zn2? binding
residues (H62, H64, H255, H199, D358 and K167) of
the dihydropyrimidinase with the equivalent residues
of the re-HDT (H59, H61, H239, H183, D316 and
K150) yielded a r.m.s.d. value on the six Ca atoms of
˚
only 0.1 A. That suggests that the six equivalent
Fig. 6 a Far-UV circular dichroism spectra of re-HDT in the
presence of Zn2?. CD spectra of re-HDT (4 lM) in the presence of
Zn2? were scanned in 1 mm path-length cuvette at 25°C by a
Jasco spectropolarimeter J-810 under a nitrogen atmosphere. The
data were recorded from 190 to 250 nm at a scan rate of
50 nm min-1. The final concentrations of Zn2? are 0 mM (1:0),
0.01 mM (1:1.5), 0.05 mM (1:7.5) and 0.1 mM (1:15), respec-
tively. b Fluorescent spectra of re-HDT, apo-HDT and Zn2?-HDT.
The concentration of all samples was 8 lM in 10 mM Tris–HCl,
pH 8.0. The spectra were recorded at the excitation wavelength of
295 nm with 1 cm path-length cuvette and scanned from 290 to
400 nm at 60 nm min-1. 1, Zn2?-HDT; 2, Re-HDT; 3, Apo-HDT
histidine residues in the re-HDT may also participate
in Zn2? binding.
Discussion
Microbial D-hydantoinase is one of the important
enzyme sources in industrial bioconversion to
D-amino acids and their derivatives (Altenbuchner
et al. 2001). Although biochemists and chemists have
contributed to the work on mechanisms and applica-
tions of this enzyme for several decades, some new
features have being discovered from a variety of
enzyme molecule, which may slightly alter the micro-
environment surrounding tryptophan residues, result-
ing in the change of fluorescence quantum efficiency.
123

Biometals (2010) 23:71–81
79
Fig. 7 Structural comparison between yeast dihydropyrimi-
dinase and the re-DHT. a The sequence alignment of the
dihydropyrimidinase from yeast S. kluyveri (2FTY) and the
hydantoinase from P. putida YZ-26(II6). The Zn2?-binding
residues are indicated by green circular shapes. b Superposition
at the active sites of yeast dihydropyrimidinase (red) and
P. putida YZ-26 D-hydantoinase (green). The zinc atoms at
zinc-a site and at zinc-b sites for the dihydropyrimidinase are
shown as red and raspberry spheres, respectively. The putative
zinc atom at zinc-a and zinc-b sites for the D-hydantoinase are
show as green and lime green spheres, respectively. Water
molecules are shown as blue spheres (colour online)
organisms, especially from microorganisms. Gener-
ally speaking, D-hydantoinase is thought to be a
metal-dependent enzyme, which can be inhibited by
addition of metal chelating reagents, for example
DPA, to produce apo-enzyme. Similarly, it can also
be re-activated by the addition of divalent metal ions
to the apo-enzyme, such as Co2?, Mg2?, Mn2? and
Zn2? (May et al. 1998a, b). It is noteworthy that Zn2?
is the second most abundant transition metal in living
organisms after iron, responsible for maintaining pro-
tein structure, regulating protein function and partic-
ipating in catalysis. Approximately 9% structures in
Protein Data Bank (PDB) belong to Zn2?-containing
enzymes. Previous reports have substantiated that
divalent metal ions, especially Zn2?, play an impor-
tant role in the catalytic process of D-hydantoinase,
and in fact most HDTs are Zn2?-enzyme or Zn2?
-
dependent enzyme (Jahnke et al. 1993; Abendroth
et al. 2002). However, the data reported regarding
Zn2? content of D-hydantoinase were quite inconsis-
tent. For example, the hydantoinase from bovine liver
or rat liver contains only one Zn2? per subunit, while
the hydantoinase from Bacillus stearothermophilus
or Arthrobacter aurescens has two Zn2? ions per
123

80
Biometals (2010) 23:71–81
subunit (May et al. 1998a, b; Cheon et al. 2002). The
result herein from the quantitative analysis of Zn2?
content and the functional evaluation of the hydan-
toinase from Pseudomonas putida YZ-26 proves that
this enzyme differs from those as described above.
Based on our experimental data, one tight Zn2? (zinc-
a) per subunit is mostly responsible for the substrate
catalysis and the other loose Zn2? (zinc-b) per the
same subunit is only essential for maintaining the
protein conformation.
The result is consistent with CD spectrum that zinc-b
site can influence the secondary structure of the re-
HDT. These findings show that both zinc-a and zinc-
b sites are all required to maintain the structure of the
enzyme.
Interestingly, a phenomenon has been observed
that during the titration of the apo-HDT with ZnCl2,
the enzyme activity was proportionally increased
until Zn2? to subunit ratio reached 1:1, and after that,
the enzymatic activity was no longer enhanced
(Fig. 3). The result indicates that when both zinc-a
and zinc-b sites are vacant, the Zn2? preferentially
binds to zinc-a site and as soon as the zinc-a site is
bound to Zn2?, excess Zn2? ions gradually bind to
the zinc-b site. Obviously, the apparent binding
constant of zinc-a site should be much higher than
that of zinc-b site. Lohkamp et al. (2006) also found
that at least one Zn2? is loosely bound to the HDT
from S. kluyveri, based on metal exchange experi-
ments. The larger Zn2?-binding capacity of the re-
HDT may be from a longer time adaptation and
evolution, as well as to ensure the preferential supply
of Zn2? for zinc-a site in Zn2? deficient environment.
Lohkamp et al. (2006) suggested that one tight
zinc-a and one loose zinc-b sites might be a
general feature of HDT, which is consistent with
our conclusion. Moreover, the homology compari-
son of the D-hydantoinase is not only similar to the
gross structure of these two crystal structures as
described above but also identical in their Zn2?
binding sites. It shows that the Zn2? binding
characteristics of this enzyme should be similar to
that of these two known structures, which indicates
that the binding affinity of the five-coordinate zinc-a
site should be higher than that of the four-coordi-
nate zinc-b site.
It is shown from the Zn2? quantitative determina-
tion of the re-HDT and apo-HDT that the Zn2? at the
tight site is directly proportional to the enzymatic
activity, while apo-HDT has little activity, indicating
that the zinc-a site plays an important role in the
catalytic process. Subsequently, being confirmed by a
Zn2? titration experiment of the apo-HDT, the activity
recovery is related to Zn2? amount and reaches a
plateau when one Zn2? is saturated for each subunit.
This result demonstrated once again that only one Zn2?
in the re-HDT, so called the zinc-a, participates in the
enzymatic catalysis. Obviously, the apo-HDT activity
was not completely recovered to the level of the
re-HDT (approximately 75%) by Zn2? titration, sug-
gesting that the removal of Zn2? results in the
conformational change or partly enzymatic inactiva-
tion. In addition to the zinc-a site, the re-HDT has
another loose Zn2?-binding site (zinc-b), which is
different from the tight Zn2? binding site. ICP-AES
analysis shows if external Zn2? is added to the re-HDT,
the product, Zn2?-HDT has two Zn2? ions per subunit,
which is different from the re-HDT carrying one Zn2?
per subunit. Moreover, the identical activity of the re-
HDT and Zn2?-HDT indicates that Zn2? at zinc-b site
is not essential for the enzymatic activity.
In addition, the presence or absence of Zn2? for a
protein is often accompanied by the protein confor-
mational change (Erk et al. 2003). Golynskiy et al.
(2005) reported that the addition of Mn2? resulted in
not only a significant stabilization of MntR, but also
fluorescent quenching. This structural transformation
can be manifested by obvious changes in the emission
fluorescence intensity and by the special adsorption
of circular dichroism (CD) spectrum. In our study,
Zn2?-HDT quenches the fluorescence intensity by
28% and apo-HDT increases the intensity by 27%,
compared with that of re-HDT, indicating that there
exists the corresponding protein structural change
with respect to different Zn2? content in enzymes.
In conclusion, our present work predicts that the
D-hydantoinase from the strain P. putida YZ-26,
which is classified and identified by this laboratory, is
a
new-type metal-dependent or metalloenzyme
involving one mole Zn2? per one mole subunit, but
it has two Zn2? binding sites, a zinc-a and a zinc-b.
The zinc-a site is essential for both activity and
conformation, and the zinc-b site only contributes to
the protein conformation.
Acknowledgment We acknowledge that this project is
supported by the Natural Science Foundation of Shanxi
province in China (No. 031042).
123

Biometals (2010) 23:71–81
81
Marti-Renom MA, Stuart AC, Fiser A, Sanchez R, Melo F, Sali
A (2000) Comparative protein structure modeling of
genes and genomes. Annu Rev Biophys Biomol Struct
29:291–325
May O, Siemann M, Siemann MG, Syldatk C (1998a) The
hydantoin amidohydrolase from Arthrobacter aurescens
DSM 3745 is a zinc metalloenzyme. J Mol Catal B 5:367–
370
May O, Siemann M, Siemann MG, Syldatk C (1998b) Cata-
lytic and structural function of zinc for the hydantoinase
from Arthrobacter aurescens DSM 3745. J Mol Catal B
4:211–218
Mishima T, Ohkuri T, Monji A, Imoto T, Ueda T (2006)
Amyloid formation in denatured single-mutant lysozymes
where residual structures are modulated. Protein Sci
15:2448–2452
References
Abendroth J, Niefind K, May O, Siemann M, Syldatk C,
Schomburg D (2002) The structure of L-hydantoinase
from Arthobacter aurescens leads to an understanding of
dihydropyrimidinase substrate and enantio specificity.
Biochemistry 41:8589–8597
Altenbuchner J, Siemann-Herzberg M, Syldatk C (2001)
Hydantoinases and related enzymes as biocatalysts for the
synthesis of unnatural chiral amino acids. Curr Opin
Biotechnol 12:559–563
Armas A, Sonois V, Mothes E, Mazarguil H, Faller P (2006)
Zinc(II) binds to the neuroprotective peptide humanin. J
Inorg Biochem 100:1672–1678
Brooks KP, Jones EA, Kim BD, Sander EG (1983) Bovine
liver dihydropyrimidine amidohydrolase: purification,
properties, and characterization as a zinc metalloenzyme.
Arch Biochem Biophys 226:469–483
Cheon YH, Kim HS, Han KH, Abendroth J, Niefind K,
Schomburg D, Wang J, Kim Y (2002) Crystal structure of
D-hydantoinase from Bacillus stearothermophilus: insight
into the stereochemistry of enantioselectivity. Biochem-
istry 41:9410–9417
Cheon YH, Park HS, Lee SC, Lee DE, Kim HS (2003)
Structure-based mutational analysis of the active site
residues of D-hydantoinase. J Mol Catal B 26:217–222
Erk I, Huet JC, Duarte M, Duquerroy S, Rey F, Cohen J,
Lepault J (2003) A zinc ion controls assembly and sta-
bility of the major capsid protein of rotavirus. Virol J
77:3595–3601
Niu LX, Zhang XY, Shi YW, Yuan JM (2007) Subunit dis-
sociation and stability alteration of D-hydantoinase deleted
at the terminal amino acid residue. Biotechnol Lett 29:
303–308
´
´
Pozo C, Rodelas B, de la Escalera S, Gonzalez-Lopez J (2002)
D, L-Hydantoinase activity of an Ochrobactrum anthropi
strain. J Appl Microbiol 92:1028–1034
Shaw CF III, Savas MM, Petering DH (1991) Ligand substi-
tution and sulfhydryl reactivity of metallothionein.
Methods Enzymol 205:401–414
Shi YW, Niu LX, Feng X, Yuan JM (2006) Purification,
enzymatic properties of a recombinant D-hydantoinase and
its dissociation by zinc ion. World J Microbiol Biotechnol
22:675–680
Shi YW, Zhang L, Yuan JM, Xiao H, Yang XQ, Niu LX (2008)
Zinc binding site in PICK1 is dominantly located at the
CPC motif of its PDZ domain. J Neurochem 106:1027–
1034
Vieille C, Zeikus GJ (2001) Hyperthermophilic enzymes:
sources, uses, and molecular mechanisms for thermosta-
bility. Microbiol Mol Biol Rev 65:1–43
Vlasova TN, Ugarova NN (2007) Quenching of the fluorescence
of Tyr and Trp residues of firefly luciferase from Luciola
mingrelica by the substrates. Biochemistry (Moscow)
72:962–967
Xu Z, Liu Y, Yang Y, Jiang W, Arnold E, Ding J (2003)
Crystal structure of D-hydantoinase from Burkholderia
pickettii at a resolution of 2.7 Angstroms: insights into the
molecular basis of enzyme thermostability. J Bacteriol
185:4038–4049
Zhang XY, Niu LX, Shi YW, Yuan JM (2008) The flexibility
of the non-conservative region at the C terminus of D-
hydantoinase from Pseudomonas putida YZ-26 is extre-
mely limited. Appl Biochem Biotechnol 144:237–247
Zhao GH, Pu P, Hu XS, Zhao L (2004) Effect of Zn(II) on the
structure and biological activity of natural b-NGF. Acta
Biochim Biophys Sin 36:99–104
Gojkovic Z, Rislund L, Andersen B, Sandrini MPB, Cook PF,
ˇ
Schnackerz KD, Piskur
J (2003) Dihydropyrimidine
amidohydrolases and dihydroorotases share the same
origin and several enzymatic properties. Nucleic Acids
Res 31:1683–1692
Golynskiy MV, Davis TC, Helmann JD, Cohen SM (2005)
Metal-induced structural organization and stabilization of
the metalloregulatory protein MntR. Biochemistry 44:
3380–3389
Huang DTC, Thomas MAW, Christopherson RI (1999) Diva-
lent metal derivatives of the hamster dihydroorotase
domain. Biochemistry 38:9964–9970
Jahnke K, Podschun B, Schnackerz KD, Kautz J, Cook PF
(1993) Acid–base catalytic mechanism of dihydropyrim-
idinase from pH studies. Biochemistry 32:5160–5166
Kikugawa M, Kaneko M, Fujimoto-Sakata S, Maeda M, Ka-
wasaki K, Takagi T, Tamaki N (1994) Purification,
characterization and inhibition of dihydropyrimidinase
from rat liver. Eur J Biochem 219:393–399
ˇ
Lohkamp B, Andersen B, Piskur J, Dobritzsch D (2006) The
crystal structures of dihydropyrimidinases reaffirm the
close relationship between cyclic amidohydrolases and
explain their substrate specificity. J Biol Chem 281:
13762–13776
123