Biophysical Chemistry 86 2000 109?118
Structure?function studies of the non-heme iron active
site of isopenicillin N synthase: some implications for
Rachel Kreisberg-Zakarin1, Ilya Borovok2, Michaela Yanko, Felix
Frolow, Yair Aharonowitz, Gerald Cohen?
Department of Molecular Microbiology and Biotechnology, The George S. Wise Faculty of Life Sciences, Tel A?i? Uni?ersity,
Ramat A?i? 69978, Israel
Received 10 December 1999; accepted 7 January 2000
Isopenicillin N synthase IPNS is a non-heme ferrous iron-dependent oxygenase that catalyzes the ring closure of
?- L-?-aminoadipoyl -L-cysteinyl-D-valineACVto form isopenicillin N. Spectroscopic studies and the crystal
structure of IPNS show that the iron atom in the active species is coordinated to two histidine and one aspartic acid
residues, and to ACV, dioxygen and H O. We previously showed by site-directed mutagenesis that residues His212,
Asp214 and His268 in the IPNS of Streptomyces jumonjinensis are essential for activity and correspond to the iron
ligands identified by crystallography. To evaluate the importance of the nature of the protein ligands for activity,
His214 and His268 were exchanged with asparagine, aspartic acid and glutamine, and Asp214 replaced with glutamic
acid, histidine and cysteine, each of which has the potential to bind iron. Only the Asp214Glu mutant retained
activity, ?1% that of the wild type. To determine the importance of the spatial arrangement of the protein ligands
for activity, His212 and His268 were separately exchanged with Asp214; both mutant enzymes were completely
defective. These findings establish that IPNS activity depends critically on the presence of two histidine and one
carboxylate ligands in a unique spatial arrangement within the active site. Molecular modeling studies of the active
site employing the S. jumonjinensis IPNS crystal structure support this view. Measurements of iron binding by the
wild type and the Asp214Glu, Asp214His and Asp214Cys-modified proteins suggest that Asp214 may have a role in
catalysis as well as in iron coordination. ? 2000 Elsevier Science B.V. All rights reserved.
Keywords: Isopenicillin N synthase; Non-heme iron dioxygenase; Active-site protein ligands; Mutagenesis; Streptomyces
?Dedicated to Heini Eisenberg.
?Corresponding author. Tel.: ?972-3-6409649; fax ?972-3-6409407.
E-mail address: firstname.lastname@example.org G. Cohen .
1,2Contributed equally to this study.
0301-4622?00?$ - see front matter ? 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 1 - 4 6 2 2 0 0 0 0 1 2 3 - X
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118 110
Isopenicillin N synthase IPNS is a non-heme
ferrous iron-dependent dioxygenase that mediates
a key step in the biosynthesis of penicillin and
cephalosporin antibiotics in bacteria and fungi 1 .
IPNS catalyzes the oxidative ring closure of ?- L-
form isopenicillinN in a reaction in which
molecular oxygen is completely reduced to water
? ? 2 . New insights into the mechanism of this re-
markable reaction have come from recent studies
on the structure of the iron coordination center
and its role in catalysis. The crystal structure of
the Aspergillus nidulans IPNS and its substrate
complex 3,4 and a wealth of spectroscopic stud-
ies of IPNS reviewed in 5
picture in which the iron atom in the catalytically
active species is attached to the tripeptide subs-
trate, to three protein ligands ? two histidines
and one aspartic acid ? and to ACV, dioxygen
and a water molecule. A similar 2-His-1-carboxy-
late triad is found in the crystal structures of
several other ferrous iron dioxygenases that cat-
alyze very different reactions 6 . In each case the
protein ligands are arranged in an array on one
face of an octahedron, anchoring the iron atom in
the active site while making accessible the re-
maining sites for binding of additional protein
residues and?or exogenous ligands such as the
substrate and cofactor.
In previous studies, we used site-directed muta-
genesis to determine which of the conserved his-
tidines and aspartic acids present throughout the
microbial IPNSs are necessary for function 7 .
We showed that residues His212, Asp214 and
His268 in the IPNS of Streptomyces jumonjinensis
are essential for activity and correspond to the
active site iron?protein ligands identified from
the A. nidulans IPNS crystal structure. A search
of the protein sequence data banks showed that
two conserved histidines and an aspartic acid or
glutamic acid are common to a large family of
non-heme ferrous iron oxygenases. A subset of
these ? including IPNS, 1-aminocyclopropane-
1-carboxylic acid ACC oxidase which converts
ACC to ethylene, and various other dioxygen-
activating enzymes that require 2-oxoglutarate as
are consistent with a
a cofactor and which are involved in the produc-
tion of secondary metabolites in bacteria, fungi
and plants and in the modification of collagen ?
all contain these three residues in a characteristic
sequence motif denoted His-X-Asp- 53-57 X-His
? ? 7 . Site-directed mutagenesis of the conserved
histidine and aspartic acid residues of the IPNS
of Cephalosporium acremonium 8,9 , of several
plant ACC oxidases, apple fruit 10 , kiwi fruit
11 and tomato 12 , a petunia flavanone 3?-hy-
droxylase 13 and a human prolyl 4-hydroxylase
14 has verified that these three residues are
essential for catalysis in each of these enzymes.
Chemical modification of histidine residues in
ACC oxidase and some other members of this
family support these conclusions 12,15,16 .
While there is compelling evidence in the above
dioxygen-activating enzymes that the histidine?
carboxylate triad functions in coordinating the
iron atom, facilitating binding of substrates, co-
factors and Oreviewed in 6 , it is not clear to
what extent the nature and arrangement of the
three endogenous ligands contribute to their cat-
alytic activity. The study described here is an
attempt to address this issue in IPNS using site-
directed mutagenesis to modify the active site
ligands in two ways: i by creating compositional
mutants in which each of the three protein lig-
ands was replaced with other amino acid residues
which have the potential to bind iron but differ in
their chemical and physical properties from the
native ligand; and ii by creating spatial mutants
in which the arrangement of the three protein
ligands in the active site was cyclically permuted
without changing their composition. We show that
the enzymatic activity of IPNS is critically depen-
dent on its possessing in its iron active site two
histidine and one aspartic acid residues in a
unique spatial arrangement.
2. Experimental procedures
Plasmid pOL18 containing the wild type S.
jumonjinensis IPNS gene under the control of the
phage T7 promoter was used for the construction
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118111
and expression in E. coli of mutant IPNS genes
as previously described 7 .
2.2. DNA manipulations and site-directed
Standard DNA manipulations procedures, the
preparation of competent cells and transforma-
tion of E. coli were performed as described 17 .
Site-directed mutagenesis was carried out using
the Amersham Sculptor kit Amersham, UK as
previously described 7 . Synthetic oligodeoxynu-
cleotides were obtained from Biotechnology Gen-
eral Rehovot, Israel and designed so as to incor-
porate or destroy a new restriction site in the
mutant region of the IPNS gene Table 1 . The
sequence of modified IPNS genes was verified
with the ABI Prism Automatic Sequencer Perkin
Elmer Biosystems, Foster City, CA, USA . IPNS
single mutants constructed in this study were:
H212D, H212N, H212Q; D214E, D214H, D214C;
H268D, H268N, H268Q. The IPNS double mu-
tant H212D, D214H was constructed using the
DNA encoding D214H as the template; the dou-
ble mutant D214H, H268D was constructed by
combining the individual single mutant DNAs.
2.3. Expression of recombinant mutant proteins,
purification and enzyme acti?ity assay
Plasmid pOL18 derivatives with mutant IPNS
genes were transformed into E. coli BL21?DE3
plysS for expression of recombinant proteins.
Briefly, cultures of transformants grown at 37?C
in Luria Broth were induced with 0.4 mM
isopropyl ?-thiogalactopyranoside for 2 h, the cells
harvested and disrupted by sonication and the
insoluble protein denatured in urea and rena-
tured as previously described 18 . The solubilized
IPNS proteins were greater than 95% pure as
trophoresis. The activity of recombinant proteins
was measured by following the conversion of ACV
to isopenicillin N by reverse phase HPLC 19 .
2.4. Iron binding
Quantitation of iron binding was performed
employing a colorimetric assay 20 with modifi-
cations 21 . Wild type and mutant proteins were
incubated with two equivalents of ferrous iron
from an anaerobic solution of ferrous ammonium
sulfate prepared in distilled water. Free iron was
separated from the bound iron by chromatogra-
phy on a P4 Biogel column. Protein concentra-
tions were determined using the extinction coef-
ficient at 280 nm, 35 mM
2.5. Molecular modeling
The crystal structure of the S. jumonjinensis
apo-IPNS and the IPNS-Fe II -?- L-?-aminoa-
Oligonucleotides used to perform site-directed mutagenesis of the active site histidine and aspartic acid ligands of IPNS
MutationOligonucleotide sequenceRestriction site
5?-CGAGGACCATCTG cat gtcTCGATGATC-3?
5?-CCGGCCCCGAAC aac cgggtgAAGTTC-3?
5?-CCGGCCCCGAAc gat cgGGTGAAGTTC-3?
5?-CCGGCCCCGAAC cag cgggtgAAGTTC-3?
Loss of AvaII
Gain of DrdI
Gain of PvuII
Loss of AatII
Loss of AatII
Loss of AatII
Loss of DrdIII
Gain of PvuI
Loss of DrdIII
aThe nucleotide changes that create the altered amino acid codons are shown underlined; the gain or loss of a restriction site
that enables the wild type and mutant gene to be differentiated is shown in lowercase.
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118112
hibitor complex unpublished results were used
to model the structures of the D214E, D214C,
D214H and H212D, D214H and D214H, H268D
mutant proteins using the programs O
Quanta 96, Molecular Simulations Inc, Burling-
ton, MA, USA and SETOR 23 .
3. Results and discussion
3.1. Acti?ity of compositional IPNS mutant proteins
Modified IPNS proteins were designed to re-
place the His212 and His268 ligands with as-
paragine, aspartic acid or glutamine, and the
Asp214 ligand with glutamic acid, histidine or
cysteine, each of which has the potential to bind
ferrous iron. Wild type IPNS and the nine single
compositional mutant proteins ? H212N,H212D,
H212Q, D214E, D214H, D214C, H268N, H268D,
H268Q and two double spatial mutant proteins
referred to below were highly overproduced em-
ploying the E. coli?T7 expression system and
were produced mainly in an insoluble form. The
recombinant wild type and mutant proteins were
solubilized as previously described 7 and their
mobility in SDS-polyacrylamide gel electrophore-
sis found to be indistinguishable data not shown .
With the sole exception of the D214E mutant
protein, no measurable activity was found in any
of the single mutant proteins less than 0.2% the
activity of wild type . Substitution of Asp214 by
glutamic acid resulted in an enzyme with ?1%
of the wild type activity. Table 2 shows the speci-
fic activity and apparent K
the wild type protein and the D214E mutant
protein. The apparent K
and kvalues for
of the D214E protein
for ACV was approximately threefold higher than
that of the wild type, while the apparent k
?35-fold lower. Despite the potential difficulties
in interpretation of these parameters 12 , partic-
ularly in the case for mutant enzymes with low
activity, the apparent k
type and mutant enzymes presumably reflect their
relative catalytic efficiency.
Table 3 lists the relative specific activities of
the S. jumonjinensis IPNS compositional mutant
proteins made in this study ? shown in bold. For
comparison, the table includes data from site-
directed mutagenesis studies of other related
dioxygenases, IPNS of C. acremonium, several
ACC oxidases from different plant sources, fla-
vanone 3?-hydroxylase of Petunia hybrida and
human prolyl 4-hydroxylase. We discuss first the
effects on activity of modifying the aspartic acid
ligand and then those of modifying the two histi-
?K values of the wild
3.2. Aspartic acid acti?e site mutants
Inspection of Table 3 shows that substitution of
glutamic acid for aspartic acid in each of the
above dioxygenases resulted in enzymes with par-
tial activities that varied from ?0.2% for tomato
ACC oxidase to ?1% for S. jumonjinensis IPNS
and kiwi fruit ACC oxidase, and ?15% for hu-
man prolyl 4-hydroxylase. Replacement of the
aspartic acid by asparagine in the petunia fla-
vanone 3?-hydroxylase was reported to give an
enzyme with ?0.4% residual activity, however,
no detectable activity was found for this change
for tomato ACC oxidase or human prolyl 4-hy-
droxylase. Although it was considered probable
that replacement of aspartic acid with glutamic
acid in the active site would not dramatically alter
Kinetic parameters of the S. jumonjinensis wild type and the Asp214Glu mutant IPNS proteins
EnzymeRelative specific activityK mMk mink
?K mM min
m cat catm
aResults reported are the averages based on three measurements.
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118 113
Relative specific activities of active site mutants of IPNS and related oxidases
Activity of purified modified enzymes expressed as percent of wild type; nd denotes no detectable activity ?0.2% ; modified
Streptomyces jumonjinensis IPNS proteins studied in this work indicated by bold letters.
Other data from: Streptomyces jumonjinensis IPNS 7 ; Cephalosporium acremonium IPNS 8,9 ; petunia flavanone 3?-hydroxyl-
ase, F3H 13 ; tomato fruit 12 ; apple fruit 10 ; and kiwi fruit 11 ACC oxidases, ACCO; human prolyl 4-hydroxylase, P4H 14 .
the iron II coordination chemistry, the effects of
replacing the aspartic acid with other residues
such as histidine which creates three histidines in
the active site or cysteine, both of which contain
a functional group ? the nitrogen atoms of the
imidazole ring and the sulfur atom of the
sulfhydryl group, respectively ? with the poten-
tial to bind iron, and which can occur as ligands
in the active site of IPNS and other ferrous iron
proteins, were not obvious. Molecular modeling
studies described below reveal that the positions
of these atoms in the modified enzymes are not
favorable for iron binding. The fact that only
glutamic acid is able to substitute for aspartic
acid in the above group of ferrous iron dioxyge-
nases with partial retention of activity indicates
that both the charge and side chain length of the
aspartic acid ligand are important for normal
3.3. Histidine acti?e site mutants
Examination of Table 3 shows the effect on
activity of individually replacing the two histidine
ligands in different dioxygenases with other iron
binding amino acids. Thus, substitution of glu-
tamine for His212 in S. jumonjinensis IPNS and
for the corresponding histidine in tomato ACC
oxidase produced in both cases inactive enzymes,
whereas the same modification in the petunia
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118 114
flavanone 3?-hydroxylase and the kiwi ACC oxi-
dase resulted in enzymes with ?0.15% and ?1%
activity, respectively. Also, substitution of that
histidine with glutamic acid in the tomato ACC
oxidase resulted in an enzyme with ?0.2% activ-
ity while in the human prolyl 4-hydroxylase it
produced an inactive enzyme. In contrast, re-
placement of His212 in S. jumonjinensis IPNS
with asparagine and aspartic acid, and the corre-
sponding histidine in the ACC oxidases of tomato
and apple with aspartic acid and asparagine, re-
spectively, created in each case inactive enzymes.
Similarly, in those enzymes where the second
histidine ligand, corresponding to His268 in S.
jumonjinensis, was changed to an asparagine, as-
partic acid or to a glutamine, the modified en-
zymes in all cases were completely inactive.
3.4. Acti?ity of spatial IPNS mutant proteins
Two IPNS double mutant proteins were con-
structed ? H212D, D214H and D214H, H268D
? in which the arrangement of the two his-
tidines and the aspartic acid in the wild type
active site were changed to create a spatially
different iron coordination center. Mutants of
this type have not previously been described. Un-
expectedly, both spatial mutants had no de-
tectable activity less than 0.2% of wild type ,
indicating that the activity of the native S. jumon-
jinensis IPNS depends on a unique spatial ar-
rangement of the three protein ligands in the
ferrous iron active site. The crystal structure of
the S. jumonjinensis IPNS see below shows this
arrangement to be identical to that reported for
the A. nidulans IPNS 3 . In the latter case, the
structure of the IPNS-Fe II -ACV substrate com-
plex4 reveals that the carboxylate ligand
Asp216 is positioned trans to the site where the
dioxygen analog nitric oxide binds, and a similar
arrangement of iron ligands occurs in the active
site of 2,3-dihydroxybiphenyl dioxygenase 24 . In
the deacetoxycephalosporin synthase-Fe II
oxoglutarate complex, the corresponding carboxy-
late ligand Asp185 binds trans to the 2-carbonyl
group of the cofactor, which is the equivalent
position to the proposed binding site of dioxygen
in IPNS 25 . This point is discussed further below.
3.5. Molecularmodelingof compositionaland spatial
IPNS mutant proteins
The refined coordinates of the S. jumonjinensis
apo-IPNS and the IPNS-Fe II -ACG
substrate inhibitor complex the crystal structures
will be described elsewhere were used to model
the effects on iron coordination of replacing the
aspartic acid at position 214 with glutamic acid,
cysteine and histidine, and of exchanging that
aspartic acid with either His212 or His268. Com-
parison of the structures of the apo-IPNS and the
IPNS inhibitor complex shows that the two struc-
tures are very similar, indicating that the IPNS
structure does not change significantly upon bind-
ing of iron and ACG. It was assumed, therefore,
in the molecular modeling analysis that the
geometry of the protein main chain in the native
protein is unchanged in the mutant proteins.
Molecular modeling of the D214E protein showed
that the longer side chain of the glutamic acid,
compared to aspartic acid, causes the iron atom
to be displaced. The most favorable arrangement
of the protein ligands for coordination of the iron
atom in the active site was found to be that in
which the iron atom is shifted by ?1.4 A from its
native position ? in the direction away from the
D214 C atom Fig. 1a . This movement is ac-
companied by rotations for His212 and His268 of
15? and 56? about ? , and 0? and 160? about ? ,
respectively. A similar analysis of the D214H and
D214C mutant proteins showed that the histidine
N?imidazole and the cysteine sulfhydryl atoms,
respectively, are positioned too far from the iron
atom to effectively contribute to its binding data
not shown . Molecular modeling of the H212D,
D214H and the D214H, H268D spatial double
mutant proteins, in which His212 and His268 are
placed in trans to the putative binding site for
dioxygen, showed that the relative positions and
geometry of the aspartic acid carboxylate oxygen
and the histidine imidazole N?liganding atoms
are significantly changed compared to the wild
type and are unfavorable for metal binding. In
part this reflects the quite different distances
separating the histidine ?6.2 A and the aspar-
tic acid ?4.8 A C atoms from the iron atom in
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118115
Fig. 1. a Superposition of the iron?protein ligands in the
active site of native S. jumonjinensis IPNS orange with the
corresponding ligands of the mutant Asp214Glu IPNS green .
Ž . bNon-symmetrical disposition of the C? atoms of the
His212, Asp214 and His268 iron ligands in the active site of
native S. jumonjinensis IPNS.
the wild type protein Fig. 1b . According to this
analysis, two of the three endogenous iron ligands
in the spatial mutants should be disrupted.
3.6. Iron binding
To evaluate whether the IPNS active site mu-
tant proteins are affected in their ability to bind
iron, the amount of iron bound by the wild type
and selected mutant proteins was determined.
The data reported in Table 4 have been corrected
for extraneous iron binding to IPNS, determined
as the amount of iron bound to a mutant IPNS
protein in which all three active site protein lig-
ands were exchanged with alanines. This protein
is entirely lacking in activity as are each of the
single alanine-substituted proteins 7 . In the con-
ditions of our experiments, wild type IPNS bound
?0.75 iron atoms?protein molecule. The weakly
active D214E enzyme bound less iron, ?0.31
iron atoms?protein molecule. We presume, on
the basis of the molecular modeling analysis, that
the iron present in its active site is displaced and
bound in a way that precludes it from effectively
promoting catalysis as compared with the wild
type. In contrast the histidine mutant, D214H,
and the D214C cysteine mutant, both of which
are totally devoid of activity, bound almost the
same amount of iron as the wild type enzyme.
Thus, the presence in each of these mutant IPNSs
of two of the native protein ligands, His212 and
His268, is sufficient for effective metal binding.
Taken together, these observations suggest that
the aspartic acid ligand may have two functions,
Iron binding by wild type and active site mutants of IPNS
Enzyme Iron bound?enzyme
aIron bound is expressed as equivalents of ferrous iron per
mole enzyme and was calculated after subtraction of non-
specific binding; the latter was determined as the amount of
iron bound, in the same experimental conditions, by a mutant
enzyme in which all three active site protein ligands had been
replaced with alanines and was found to be 0.49 eq.?mol
enzyme. Results reported are the averages based on two to
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118116
to bind iron and to participate in some other step
in the catalytic cycle. Measurement of iron bind-
ing in the D214H, H268D spatial double mutant
showed that the amount of iron bound was almost
the same as the level of extraneously bound iron.
This result is consistent with the molecular mod-
eling studies described above which predict that
in this mutant two of the protein ligands are
abolished. Because of the uncertainty in the esti-
mates of the iron bound at the active site, due to
the existence of additional iron binding site s ,
the interpretation of the results given in Table 4
should be viewed with some caution. In this re-
spect it is noteworthy that the crystal structure of
the A. nidulans IPNS complexed with manganese
reveals two binding sites for manganese, one at
the active site and one near the surface attached
to glutamine and histidine residues 3 .
The experiments reported here were designed
to help better define the role of the endogenous
IPNS ferrous iron ligands in catalysis. The crystal
structures of the A. nidulans 3,4 and the S.
jumonjinenesis IPNS unpublished results , deace-
toxycephalosporin C synthase 25,26 , and several
other non-heme ferrous iron oxygenases, all of
which possess the 2-His-1-carboxylate triad, in-
cluding tyrosine hydroxylase 27 , 2,3-dihydroxybi-
phenyl 1,2 dioxygenase 24 , soybean lipoxygenase
28 and iron superoxide dismutase 29 , indicate
that a primary role for the protein ligands is to
anchor the iron atom in the active site while
making available additional sites for binding of
substrate, cofactors and dioxygen 6 . The results
described here for IPNS compositional mutants,
in which the active site protein ligands were re-
placed with residues possessing a functional group
potentially able to bind iron, and for IPNS spatial
mutants in which the protein ligands were cycli-
cally rearranged in the active site, are consistent
with their having a role in iron binding. Further-
more, they raise the possibility that the protein
ligands, in particular the aspartic acid ligand, may
also play an additional role in the catalytic cycle.
This is suggested by the findings that replacement
of the S. jumonjinensis Asp214 ligand with either
histidine or cysteine resulted in completely defec-
tive enzymes but did not appreciably affect their
ability to bind iron; on the other hand substitu-
tion of that aspartic acid with glutamic acid cre-
ated an enzyme with a low, but significant, activity
while substantially decreasing iron binding. These
results demonstrate that the charge of the car-
boxylate ligand, the length of its side chain and its
spatial position in the IPNS active site are all
critical for function.
Elsewhere, it has been proposed that binding of
substrate to IPNS and related oxygenases primes
the iron center for binding of O by modulating
the redox potential of the iron center, and evi-
dence in support of this has come from crystallo-
graphic and spectroscopic studies of several of the
above cited enzymes see 6 . In IPNS the charged
aspartate carboxylate group is positioned trans to
the presumed O binding site and this may facili-
tate formation of an Fe?O
thought to be a necessary step in the catalytic
cycle 30 . Thus, not only the nature but also the
precise spatial organization of the functional
groups in the metal coordination center may be
crucial for this to occur. It will be of interest
therefore to analyze the ability of the two classes
of IPNS mutant enzymes discussed in this work
for their ability to bind ACV and the substrate
analog ACG. One approach will be to investigate
the metal catalyzed oxidative modification of the
complexes as recently described for ACC oxidase
12 ; another will be to employ spectroscopic
means to measure the interaction of modified
IPNSs with tripeptides.
adduct which is
We wish to thank Jim Remington for providing
data for determining the crystal structure of the
S. jumonjinensis IPNS, and for support from the
Israel Science Foundation, Grant No. 488.97 and
the Constantiner Institute for Molecular Genetics
at Tel Aviv University.
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118117
? ? 1 R. Kreisberg-Zakarin,
Aharonowitz, G. Cohen, Recent advances in the struc-
ture and function of isopenicillin N synthase, Antonie
van Leeuwenhoek 75 1999 33?39.
J.B. Baldwin, M. Bradley, Isopenicillin N synthase:
mechanistic studies, Chem. Rev. 90 1990 1079?1088.
P.L. Roach, I.J. Clifton, V. Fulop et al., Crystal structure
of isopenicillin N synthase is the first from a new struc-
tural family of enzymes, Nature 375 1995 700?704.
P.L. Roach, I.J. Clifton, C.M. Hensgens et al., Structure
of isopenicillin N synthase complexed with substrate
and the mechanism of penicillin formation, Nature 387
L. Que, R.Y.N. Ho, Dioxygen activation by enzymes
with mononuclear non-heme iron active sites, Chem.
Rev. 96 1996 2607?2624.
E.L. Hegg, L. Que Jr, The 2-His-1-carboxylate facial
triad: an emerging structural motif in mononuclear
non-heme iron II enzymes, Eur. J. Biochem. 250 1997
I. Borovok, O. Landman, R. Kreisberg-Zakarin, Y.
Aharonowitz, G. Cohen, Ferrous active site of isopeni-
cillin N synthase: genetic and sequence analysis of the
endogenous ligands, Biochemistry 35 1996 1981?1987.
D.S. Tan, T.S. Sim, Functional analysis of conserved
isopenicillin N synthase by site-directed mutagenesis, J.
Biol. Chem. 271 1996 889?894.
P. Loke, J. Sim, T.S. Sim, Functional analysis of a
conserved aspartate D218 in Cephalosporium acre-
monium isopenicillin N synthase, FEMS Microbiol. Lett.
157 1997 137?140.
J.F. Shaw, Y.S. Chou, R.C. Chang, S.F. Yang, Character-
ization of the ferrous ion binding sites of apple 1-
rected mutagenesis, Biochem. Biophys. Res. Commun.
225 1996 697?700.
V.J. Lay, A.G. Prescott, P.G. Thomas, P. John, Heterol-
ogous expression and site-directed mutagenesis of the
1-aminocyclopropane-1-carboxylate oxidase from kiwi
fruit, Eur. J. Biochem. 242 1996 228?234.
Z. Zhang, J.N. Barlow, J.E. Baldwin, C.J. Schofield,
Metal-catalyzed oxidation and mutagenesis studies on
theiron II bindingsite
R. Lukacin, L. Britsch, Identification of strictly con-
served histidine and arginine residues as part of the
active site in Petunia hybrida flavanone 3?-hydroxylase,
Eur. J. Biochem. 249 1997 748?757.
J. Myllyharju, K.I. Kivirikko, Characterization of the
I. Borovok, M. Yanko, Y.
? ? 2
? ? 3
? ? 4
? ? 5
? ? 6
? ? 7
? ? 8
? ? 9
iron- and 2-oxoglutarate-binding sites of human prolyl
4-hydroxylase, EMBO J. 16 1997 1173?1180.
A. Lamberg, T. Pihlajaniemi, K.I. Kivirikko, Site-di-
rected mutagenesis of the alpha subunit of human prolyl
4-hydroxylase. Identification of three histidine residues
critical for catalytic activity, J. Biol. Chem. 270 1995
C.C Lawrence, W.J. Sobey, R.A. Field, J.E. Baldwin,
C.J. Schofield, Purification and initial characterization of
proline 4-hydroxylase from Streptomyces griseo?iridus
P8648: a 2-oxoacid, ferrous-dependent dioxygenase in-
volved in etamycin biosynthesis, Biochem. J. 313 1996
J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular
Cloning: a Laboratory Manual, 2nd ed., Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, 1989.
O. Landman, D. Shiffman, Y. Av-Gay, Y. Aharonowitz,
G. Cohen, High level expression in Escherichia coli of
isopenicillin N synthase genes from Fla?obacterium and
Streptomyces, and recovery of active enzyme from inclu-
sion bodies, FEMS Microbiol. Lett. 68 1991 239?244.
S.E. Jensen, D.W.S. Westlake, S. Wolfe, High perfor-
mance liquid chromatographic assay of cyclization activ-
ity in cell-free systems from Streptomyces cla?uligerus, J.
Antibiot. Tokyo 35 1982 1026?1032.
W.W. Fish, Rapid colorimetric micromethod for the
quantitation of complexed iron in biological samples,
Meth. Enzymol. 158 1988 357?364.
K. Lauber, Determination of serum iron; a comparison
of two methods: teepol?dithionite?bathophenanthro-
line versus guanidine?ascorbic acid?ferrozine, J. Clin.
Chem. Clin. Biochem. 18 1980 147?148.
T.A. Jones, J.Y. Zou, S.W. Cowan, M. Kjeldgaard, Im-
proved methods for binding protein models in electron
density maps and the location of errors in these models,
Acta Crystallogr. A. 47 1991 110?119.
S.V. Evans, SETOR: hardware-lighted three-dimen-
sional solid model representations of macromolecules, J.
Mol. Graph 11 1993 134?138.
T. Senda, K. Sugiyama, H. Narita et al., Three-dimen-
sional structures of free form and two substrate com-
plexes of an extradiol ring-cleavage type dioxygenase,
the BphC enzyme from Pseudomonas sp. strain KKS102,
J. Mol. Biol. 255 1996 735?752.
K. Valegard, A.C. van Scheltinga, M.D. Lloyd et al.,
Structure of a cephalosporin synthase, Nature 394 1998
M.D. Lloyd, H.J. Lee, K. Harlos et al., Studies on the
active site of deacetoxycephalosporin C synthase, J. Mol.
Biol. 287 1999 943?960.
K.E. Goodwill, C. Sabatier, C. Marks, R. Raag, P.F.
Fitzpatrick, R.C. Stevens, Crystal structure of tyrosine
hydroxylase at 2.3 A and its implications for inherited
neurodegenerative diseases, Nat. Struct. Biol. 4 1997
() Download full-text
R. Kreisberg-Zakarinet al.?Biophysical Chemistry 86 2000 109?118118
28W. Minor, J. Steczko, B. Stec et al., Crystal structure of
soybean lipoxygenase L-1 at 1.4 A resolution, Biochem-
istry 35 1996 10687?10701.
M.S. Lah, M.M. Dixon, K.A. Pattridge, W.C. Stallings,
J.A. Fee, M.L. Ludwig, Structure?function in
cherichia coli iron superoxide dismutase: comparisons
with the manganese enzyme from Thermus thermophilus,
Biochemistry 34 1995 1646?1660.
P.L. Roach, I.J. Clifton, C.M. Hensgens et al., Anaer-
obic crystallization of an isopenicillin N synthase Fe II
substrate complex demonstrated by X-ray studies, Eur.
J. Biochem. 242 1996 736?740.