Probing the active site of rat porphobilinogen synthase using newly developed inhibitors

Article abstract of DOI:10.1016/j.bioorg.2008.11.001

The structurally related tetrapyrrolic pigments are a group of natural products that participate in many of the fundamental biosynthetic and catabolic processes of living organisms. Porphobilinogen synthase catalyzes a rate-limiting step for the biosyntheses of tetrapyrrolic natural products. In the present study, a variety of new substrate analogs and reaction intermediate analogs were synthesized, which were used as probes for studying the active site of rat porphobilinogen synthase. The compounds 1, 3, 6, 9, 14, 16, and 28 were found to be competitive inhibitors of rat porphobilinogen synthase with inhibition constants ranging from 0.96 to 73.04 mM. Compounds 7, 10, 12, 13, 15, 17, 18, and 26 were found to be irreversible enzyme inhibitors. For irreversible inhibitors, loose-binding inhibitors were found to give stronger inactivation. The amino group and carboxyl group of the analogs were found to be important for their binding to the enzyme. This study increased our understanding of the active site of porphobilinogen synthase.

Full text of DOI:10.1016/j.bioorg.2008.11.001

Bioorganic Chemistry 37 (2009) 33–40
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Probing the active site of rat porphobilinogen synthase using newly
developed inhibitors
Nan Li a,b, Xiusheng Chu a, Xiaojun Liu a, Ding Li a,
*
a Department of Biology and Chemistry, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong SAR, PR China
b Department of Chemical and Environmental Engineering, Wuyi University, Jiangmen, Guangdong Province 529020, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 October 2008
Available online 17 December 2008
The structurally related tetrapyrrolic pigments are a group of natural products that participate in many of
the fundamental biosynthetic and catabolic processes of living organisms. Porphobilinogen synthase cat-
alyzes a rate-limiting step for the biosyntheses of tetrapyrrolic natural products. In the present study, a
variety of new substrate analogs and reaction intermediate analogs were synthesized, which were used
as probes for studying the active site of rat porphobilinogen synthase. The compounds 1, 3, 6, 9, 14, 16,
and 28 were found to be competitive inhibitors of rat porphobilinogen synthase with inhibition constants
ranging from 0.96 to 73.04 mM. Compounds 7, 10, 12,13, 15, 17, 18, and 26 were found to be irreversible
enzyme inhibitors. For irreversible inhibitors, loose-binding inhibitors were found to give stronger inac-
tivation. The amino group and carboxyl group of the analogs were found to be important for their binding
to the enzyme. This study increased our understanding of the active site of porphobilinogen synthase.
Ó 2008 Elsevier Inc. All rights reserved.
Keywords:
Porphobilinogen synthase
PBG synthase
5-Aminolaevulinic acid dehydratase
ALA dehydratase
5-Aminolaevulinic acid
Porphobilinogen
Tetrapyrrole
1. Introduction
propionyl-containing half of PBG derives from P-side ALA. This
has also led to the terminology of the ‘‘A” and ‘‘P” binding sites
The structurally related tetrapyrrolic pigments are a group of
natural products that include the haems, the chlorophylls, the cor-
rinoids (e.g. coenzyme B12), the cyclic tetrapyrroles factor F430 and
the linear tetrapyrroles (bilins) [1,2]. Regulation of tetrapyrrole
biosynthesis has been found to be crucial to plant and bacteria
metabolism and gene expression [3–8]. These compounds partici-
pate in many of the fundamental biosynthetic and catabolic pro-
cesses of living organisms. They are all intensely colored and
almost every living organism has an absolute requirement for
one or more of them. It is for this reason that they are called the
‘‘pigments of life.”
Porphobilinogen synthase (PBG synthase; PBGS; EC 4.2.1.24) is
also named 5-aminolaevulinic acid dehydratase (ALA1 dehydra-
tase), which catalyses an asymmetric condensation of two mole-
cules of 5-aminolaevulinic acid (ALA) to give the monopyrrole
porphobilinogen (PBG) as shown in Fig. 1. The substrate that be-
comes the acetyl-containing half of PBG is called A-side ALA; the
in the enzyme which bind the A-side and P-side ALAs, respectively.
PBG synthase is highly conserved throughout the archaea,
eubacteria, and eukarya. This enzyme has been purified from a
variety of sources including human erythrocytes [9], bacteria such
as Escherichia coli [10], and plants such as spinach [11]. There are
some differences among these PBG synthases in terms of their me-
tal requirements, kinetic parameters, pH optima, inactivation by
inhibitors and susceptibility to oxidation. It is often claimed that
almost all PBG synthases require a divalent cation for activity with
animal enzymes using zinc and plant enzymes using magnesium.
Representatives of both zinc and magnesium classes exist in some
bacteria [12]. The human PBG synthase can adopt different nonad-
ditive quaternary assemblies (morpheein forms), which are a high
activity octamer, a low activity hexamer, and two structurally dis-
tinct dimer conformations [13,14].
The X-ray structures of PBG synthases from several species have
been determined [15–21]. In these structures, the enzyme is a
homo-octamer, and the active site of each subunit is located in a
pronounced cavity formed by loops at the C-terminal ends of the
b-strands. All eight active sites are oriented towards the outer sur-
face of the octamer and appear to be independent. The catalysis
has been shown to proceed by formation of a Schiff-base link at
the P-site at first between the 4-keto group of substrate ALA and
an invariant lysine residue, equivalent to Lys263 in yeast PBG syn-
thase. The structures of several PBG synthases complexed with the
inhibitors 4,7-dioxosebacic acid, 5-fluorolaevulinic, and several
* Corresponding author. Fax: +86 852 2788 7406.
E-mail address: bhdingli@cityu.edu.hk (D. Li).
1
Abbreviations used: ALA, 5-aminolaevulinic acid; DHP, 1,2-dihydropyrane; DMF,
dimethyl formamide; DMSO, dimethyl sulfuroxide; DTT, dithiothreitol; IPTG, isopro-
pyl-b-D-thiogalactopyranoside; PAGE, polyacrylamide gel electrophoresis; PBG, por-
phobilinogen; PBGS, porphobilinogen synthase; PCC, pyridinium chlorochromate;
PCR, polymerase chain reaction; PPTS, pyridinium p-toluenesulfonate; PTSA, p-
toluenesulfonic acid; SDS, sodium dodecylsulfate; THF, tetrahydrofuran; THP, tetra-
hydropyran; UV/vis, ultraviolet–visible spectroscopy.
0045-2068/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2008.11.001

34
N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
enzymes exhibit different susceptibilities to various inhibitors
COOH
COOH
HOOC
[44,45]. In the present research, we synthesized a variety of new
substrate analogs of PBG synthase, which were incubated with
the enzyme as probes. The study further increased our understand-
ing of this important enzyme.
HOOC
PBGS
O
+
+ 2 H2O
O
N
H
H2N
NH2
A-side
ALA
NH2
2. Methods and materials
P-side
ALA
PBG
+
2.1. Materials
Fig. 1. Reactions catalyzed by PBG synthase.
A Hi-Trap chelating metal-affinity column was purchased from
Amersham Pharmacia Biotech. Taq DNA polymerase, HB101 com-
petent cells, E. coli strain BL21(DE3) competent cells, agarose, Plas-
mid Mini kit, and synthesized oligonucleotides came from
Invitrogen Life Technologies. Restriction enzymes came from MBI
Fermentas. All other reagents were of research grade or better
and were obtained from commercial sources.
bisubstrate analogs have been solved at high resolution [22–24].
These inhibitors form two Schiff bases at the active site involving
Lys210 as well as Lys263 in yeast PBG synthase. The structural evi-
dence that both invariant lysines form Schiff bases with some
inhibitors suggests that catalysis involves a double Schiff-base
mechanism.
Two possible mechanisms have been proposed for the forma-
tion of PBG from ALA. They differ mainly in the order of C–C (aldol
condensation) and C–N bond formation (Schiff base) (Fig. 1). Nandi
and Shemin have proposed a mechanism that is analogous to the
reaction catalyzed by the class I aldolases suggesting C–C bond for-
mation happens before C–N bond formation [25]. Several years la-
ter, Jordan and Seehra proposed another mechanism suggesting a
C–N bond formation takes place at first [26]. After either C–C bond
or C–N bond forms at first in the enzymatic reaction, a long chain
dicarboxylate intermediate is generated. In order to distinguish be-
tween the two major mechanistic proposals, diacids containing se-
ven carbon atoms (suggesting that C–C bond formation occurs
first) or diacids containing ten carbon atoms (suggesting C–N bond
formation occurs first) were tested as inhibitors for E. coli PBG syn-
thase [27]. The diacids analogs containing ten carbon atoms have
been found to be either competitive or irreversible inhibitors for
E. coli PBG synthase, which supports Jordan mechanism. However,
the later studies on these compounds indicate that some PBG syn-
thases from human and other sources are not inhibited by these
compounds [22,28]. Therefore, the order of bond-making and
bond-breaking reactions in the PBGS-catalyzed reaction remains
uncertain and may not be conserved for all PBG synthases from dif-
ferent species.
It has been reported that PBG synthase activity is correlated
with some diseases such as diabetes, hypothyroidism [29,30]. In
humans, hereditary deficiencies in PBG synthase give rise to the
rare disease porphyria [31–33], and the clinical comment of this
acute porphyria is similar to lead poisoning. Human PBG synthase
is a metalloenzyme that requires ZnII for maximal catalytic activity
and it is an important molecular target for the widespread environ-
mental toxic metal Pb2+ [34,35]. Al3+, Ga3+, and In3+ inhibit bovine
liver PBG Synthase by competing with ZnII, whereas Tl3+ and In3+
inhibit bovine PBG synthase by directly oxidizing essential sulfhy-
dryl group [36]. PBG synthase also require thiol groups for catalytic
activity, and the purified enzyme is extremely sensitive to oxygen
and requires thiol-reducing agents to display maximal activity
[35]. In fact, mammalian PBG Synthases have several reactive cys-
teinyl residues that react with a variety of sulfhydryl reagents [37–
39], and it has been proposed that one role for ZnII is to prevent
disulfide formation between essential sulfhydryl groups in PBG
Synthase [40].
2.2. Cloning of the functional rat PBG synthase
A rat liver Quick-Clone cDNA library was purchased from Clon-
tech (Palo Alto, CA). The gene of rat PBG synthase was amplified by
PCR using primers that were designed to add six continuous histi-
dine codons to the 50 primer. The sequence of the forward primer
was 50-cg cgc gga tcc aggagga atttaaa atg aga gga tcg cat cac cat
cac cac cac cag tcc gtt ctg cac ag 30, containing a BamHI site (gga
tcc), a ribosome binding site (aggagga), codons for the amino acid
sequence MRGSHHHHHH (start codon and hexahistag), and codons
for amino acids 4–8 of rat liver PBG synthase. The sequence of the
reverse primer was 50-ctg cag gtc gac tta ctc ttc ctt cag cca ctt caa
cag-30, containing a SalI site (gtc gac), a stop anticodon (tta), and
anticodons for the last eight amino acids of rat liver PBG synthase.
The PCR product was gel purified, double digested, and ligated into
a pLM1 [46] expression vector resulting in the pLM1::PBGS plas-
mid. pLM1 vector has a T7 promoter-driven system with ampicillin
resistance and can amplify in E. coli HB101. The constructed
pLM1::PBGS plasmid was transformed into HB101 competent cells
according to an electroporation transformation procedure (Bio-
Rad) for screening purposes. The identified positive colony was
grown in LB medium containing ampicillin (50 mg/L), and the plas-
mid pLM1::PBGS was isolated and transformed to E. coli strain
BL21(DE3) competent cells for expression purposes. DNA sequenc-
ing of the cloned rat liver PBG synthase gene was performed, and
the inserted gene sequence was identified to be the same as that
previously deposited in NCBI without any mutation.
PBG synthase gene was also cloned into another vector pET28a+
using a similar method. The sequence of the forward primer was
50-cg cgc gct agc cac cac cag tcc gtt ctg cac ag-30, containing a NheI
site (gct agc), and codons for amino acids 2–8 of rat liver PBG syn-
thase. The sequence of the reverse primer was 50-ctg cag aag ctt tta
ctc ttc ctt cag cca ctt caa cag-30, containing a HindIII site (aag ctt), a
stop anticodon (tta), and anticodons for the last eight amino acids
of rat liver PBG synthase. The constructed pET28a+::PBGS plasmid
was sequenced, and the inserted gene sequence was identified to
be the same as that previously deposited in NCBI without any
mutation.
2.3. Expression and purification of soluble rat PBG synthase
A number of compounds have been synthesized as substrate or
product analogs in attempts to unravel PBG synthase mechanism
or to inhibit the enzyme in a specific manner [25,41–43]. Some
substrate analogs have been found to be competitive enzyme
inhibitors. It has been found that PBG synthase is not inhibited
by its direct product PBG. It has also been revealed that different
Established methods [47] were used to prepare the enzymes to
apparent homogeneity as analyzed by SDS–PAGE. The proteins
were stored at ꢀ80 °C in 50 mM Tris buffer, pH 7.5, 5% glycerol,
and 5 mM b-mercaptoethanol. The stability of the purified en-
zymes was tested by its activity and the His-tagged proteins were
proved to be highly stable. The enzymes can be stored at 4 °C for

N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
35
1 week without significant change of activity. The proteins were
normally stored in a ꢀ80 °C freezer and were stable for at least
one year tested on the basis of their activities.
Rat PBG synthase from recombinant plasmid pET28a+::PBGS
was incubated with thrombin to remove the His-tag. The enzyme
(150 mg) and the thrombin (5 mg) were incubated in Tris–Cl buffer
(50 mM, pH 8.0) at 25 °C for different time period, and the result
showed that 1-h incubation is enough for an obvious cleavage of
the protein.
zyme, and an inhibitor, which were incubated at 37 °C for 10 min.
Then the reaction was started by adding the substrate. A control
experiment was performed under identical conditions without
the inhibitor. The IC50 value was determined by Dixon plot [49].
3. Results and discussion
3.1. Gene cloning, expression in E. coli, and purification of rat PBG
synthase
2.4. Activity assay
Rat PBG synthase gene was cloned with two different vectors,
which gave two recombinant plasmid pLM1::PBGS and pE-
T28a+::PBGS. The expression in E. coli BL21 (DE3) was carried out
at different temperatures and IPTG concentrations. It was found
that both proteins expressed better at an IPTG concentration of
0.5 mM and that it resulted in a more soluble protein at room tem-
perature than 37 °C. Nickle metal-affinity resin columns were used
for single-step purifications of the two His-tagged rat PBG
synthases. The protein fractions were dialyzed against a 20 mM
potassium phosphate buffer, pH 7.5, 5% glycerol, and 5 mM b-
mercaptoethanol, as soon as possible after the purification. The
addition of 5% glycerol and 5 mM b-mercaptoethanol was essential
to maintain the long-term stability of the protein. The enzymes can
be stored at 4 °C for 1 week without significant change of activity.
The proteins were normally stored in a ꢀ80 °C freezer and were
stable for at least one year tested on the basis of its activity. The
purity of the enzymes was examined by SDS–PAGE and single
bands corresponding to about 37 kDa proteins were observed with
>95% purity (Fig. 2). The molecular mass is in agreement with the
data calculated from the amino acid sequences. The overall yields
of the two proteins in the purification procedure both were above
90%. For the plasmid pLM1::PBGS, 12 mg purified protein was ob-
tained from 0.5 L culture. For the plasmid pET28a+::PBGS, 2 mg
purified protein was obtained from 0.5 L culture. In order to test
the effect of the six histidines added in the N-terminus of the pro-
tein, thrombin was used to remove the His-tag from rat PBG syn-
thase from the recombinant plasmid pET28a+::PBGS. The purity
of the resulting enzymes was examined using SDS–PAGE and sin-
gle bands corresponding to about 37 kDa proteins were observed
with >90% purity (Fig. 3).
The activity of PBG synthase was determined by a calorimetric
assay based on the reaction between PBG and modified Ehrlich’s
reagent [48]. Enzyme (6–8
37 °C in Tris–Cl buffer (100 mM, pH 8.5, containing 10
and 10 mM 2-mercaptoethanol). The reaction was started by the
l
g) was preincubated for 10 min at
lM ZnCl2
addition of ALA (10 mM final concentration) to give a final volume
of 500
l
l and preceded for 5 min at 37 °C. The incubation was ter-
l mercuric chloride (0.1 M) in 20%
minated by the addition of 500
l
trichloroacetic acid at 0 °C to precipitate protein and thiol. The
solution was stayed on ice for 10 min and then centrifuged at
10,000 rpm for 5 min at room temperature. The aliquot (500 ll)
of supernatant was transferred and reacted with an equal volume
of modified Ehrlich’s reagent. After 10 min, PBG was detected spec-
trophotometrically by the absorbance of pyrrole at 555 nm with
the use of
60,200 Mꢀ1 cmꢀ1. One unit of enzyme activity is defined as the
amount of enzyme that produce 1 mol of PBG in 1 h at 37 °C.
a
molar absorption coefficient for PBG of
l
Determination of the KM and the Vmax was performed using the
same assay buffer with varying substrate concentrations ranging
from 0.1 to 5 mM.
2.5. Enzyme inactivation study
PBG synthase (40–50
priate amount of inhibitor, and 10
(100 mM, pH 8.5) at 37 °C. At various time intervals, 10
l
g, 15
l
l) was incubated with an appro-
M ZnCl2 in 45 l Tris buffer
l aliquot
l
l
l
of the solution was taken out individually to assay the enzyme
activity in a buffer containing 2-mercaptoethanol in order to react
with the remaining inhibitor. A control experiment was performed
under identical conditions without inhibitor. After 60 min incuba-
tion at room temperature to ensure complete inactivation of the
enzyme, the inactivated enzyme was dialyzed against 100 mM Tris
buffer (pH 8.5) over 2 days at 4 °C with seven buffer changes. The
residual enzyme activity was determined before and after dialysis
to test for the reversibility of the inactivation. A control experiment
was performed under identical conditions without inhibitor.
45kDa
35kDa
2.6. Determination of kinetic constant Ki of competitive inhibitors
The type of inhibition of the reversible inhibitors was deter-
mined by a double-reciprocal plot of 1/V versus 1/[S]. The Ki values
of the competitive inhibitors of PBG synthase were determined by
Dixon plot [49]. The reaction cocktail contained the buffer, the en-
zyme, and an inhibitor, which were incubated at 37 °C for 10 min.
Then the reaction was started by adding the substrate. If the inhi-
bition type of the inhibitor was not competitive, the IC50 values
were determined instead of the Ki values to compare the potency
of the inhibitors.
1
2
3
Fig. 2. SDS–PAGE of purified rat PBG synthase from different plasmids. Lane 1,
protein marker: 116.0, b-galactosidase from E. coli; 66.2, bovine serum albumin
from bovine plasma; 45.0, ovalbumin from chicken egg white; 35.0, lactate
dehydrogenase from porcine muscle; 25.0, REase Bsp98I from E. coli; 18.4, b-
lactoglobulin from bovine milk; 14.4, lysozyme from chicken egg white; lane 2,
purified rat PBG synthase from recombinant plasmid pLM1::PBGS; lane 3, purified
rat PBG synthase from recombinant plasmid pET28a+::PBGS.
2.7. Determination of IC50 for inhibitors
In determination of the IC50 values for the inhibitors, the con-
centration of the substrate was about two fold of the KM value
for the enzyme. The reaction cocktail contained the buffer, the en-

36
N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
Table 1
Kinetic constants of rat PBG synthase.
KM (mM)
Vmax (lmol/h/mg)
PBGS from pLM1:: PBGS
PBGS from pET28a+::PBGS
PBGS from pET28a+::PBGS, No His-tag
3.22 0.67
0.65 0.06
2.10 0.49
5.5 0.6
16.7 0.5
17.4 1.2
Cl buffer (containing 10 lM ZnCl2) in the range of pH 7.0–9.5. All
enzymes showed higher activity under basic conditions. The opti-
mum pH value for PBG synthase from pLM1::PBGS is 8.5. For PBG
synthase from pET28a+::PBGS, the enzymes with and without His-
tag showed same pH-activity profile with highest activity in the
range of pH 8.5–9.5.
The activities of three rat PBG synthases were tested at different
ZnCl2 concentrations in Tris buffer (pH 8.5). For two enzymes from
pET28a+::PBGS, the optimal concentration of Zn2+ ion was found to
be 10
l
M. For the enzyme from pLM1::PBGS, the optimal concen-
M. All three enzymes were
tration of Zn2+ ion was found to be 30
l
1
2
3
4
5
found to have considerable activities when no Zn2+ ion was used,
since there was a little Zn2+ ion in the LB culture for protein expres-
sion. All three enzymes were completely inactivated when 1 mM
EDTA was added to the assay buffer, indicating that the Zn2+ ion
was essential to the activity of rat PBG synthase.
Fig. 3. SDS–PAGE of rat PBG synthase from recombinant plasmid pET28a+::PBGS
cleaved by thrombin. Lane 1, protein marker; lane 2, PBG synthase from
recombinant plasmid pET28a+::PBGS; lane 3, PBG synthase cleaved by thrombin
at 25 °C for 1 h; lane 4, PBG synthase cleaved by thrombin at 25 °C for 2 h; lane 5,
PBG synthase cleaved by thrombin at 25 °C for 3 h.
For the kinetic analyses of rat PBG synthases, rates were mea-
sured at five or six substrate concentrations. The results of kinetic
studies obtained through nonlinear curve fitting using SigmaPlot
8.0 program are summarized in Table 1. The results indicate the re-
moval of His-tag does not have significant effect on the enzymatic
activity.
3.2. Characterization of rat PBG synthase
Rat PBG synthases from pLM1::PBGS, pET28a+::PBGS, and rat
PBG synthase without the His-tag were assayed in 100 mM Tris–
Fig. 4. The sequence of PBG synthase from various sources. Rn, Rattus norvegicus; Hs, Homo sapiens; Sc, Saccharomyces cerevisiae; Ec, Escherichia coli. Residues conserved in all
sequences are marked with *. Residues not conserved in all sequences but conserved in some sequences are marked with : or . based on the degree of conservation. The
sequence alignment was made by using ClustalX [53,54]. The important residues for catalysis and binding are highlighted.

N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
37
O
O
O
O
Br2
HCl
HCl
OH
O
OBn
Br
OH
CH3OH
H2O
H2O
1
2
O
O
O
O
O
O
O
KOH
Br2
CH3OH
Br
H2O
H2N
Br
OH
OEt
OEt
3
4
NaN3
DMF
O
O
O
O
1. H2, Pd/C
2. HCl, H2O
OEt
OH
6
5
N3
CH3OH
THF
O
O
O
DHP
NaN3
DMF
1. H2, Pd/C
H2N
Br
OH
OH
OTHP
PPTS
PTSA 2. HCl, H2O
7
8
9
Br2
CH3OH
0 oC
O
O
O
Br2
NaN3
DMF 2. HCl, H2O
1. H2, Pd/C
H2N
Br
Br
CH3OH
O
OH
PBr3
Et2O
O
11
10
O
O
Br2
CH3OH
Br
Br
13
12
KCN
DMSO
O
O
O
Br2
NaN3
1. H2, Pd/C
Br
H2N
CN
CN
CN
CH3OH
DMF 2. HCl, H2O
14
15
16
O
O
O
O
PBr3
Br2
Br
Br
OH
Br
Et2O
CH3OH
17
18
Bn-Br
KOH
O
O
NaN3
Br2
1. H2, Pd/C
OBn
Br
OBn
H2N
OH
CH3OH
DMF 2. HCl, H2O
19
20
21
Fig. 5. Syntheses of PBG synthase substrate analogs.
3.3. Sequence alignment of PBG synthases from various sources
in the methanol, and the resulting a-bromoketo ester was substi-
tuted by sodium azide. Then, the azido group was hydrogenated
using palladium on charcoal catalyst, which was then deprotected
through hydrolysis reaction to give compound 6.
The sequence alignment of PBG synthases from various sources
were carried out as shown in Fig. 4. It was found that PBG syn-
thases from mammalian sources, yeast, and bacteria have good se-
quence homology. Rat PBG synthase shares 88% sequence
homology with human PBG synthase. The important residues for
catalysis and binding are highly conserved. This indicates rat PBG
synthase is a good model enzyme for subsequent studies of syn-
thetic substrate analogs.
4-Hydroxy-2-butanone was used as the starting material for
syntheses of a variety of substrate analogs as shown in Fig. 5. It
was brominated in methanol at 0 °C to give compound 7, and the
hydroxyl group was stable in the reaction at low temperature.
The 4-hydroxyl group was protected by dihydropyran (DHP), and
the resulting compound 8 was reacted with sodium azide. After
deprotection, the azido group was hydrogenated to give compound
9. The hydroxyl group of 4-hydroxy-2-butanone was substituted
by bromine to give compound 12, and then the bromine was
substituted by cyano group to afford compound 14. Compounds
10, 11, 13, 15, and 16 were synthesized using same chemical reac-
tions as mentioned above. 5-Hydroxy-2-pentanone was used as
starting material for syntheses of several other substrate analogs,
including compounds 17, 18, and 21, using same chemical reac-
tions as mentioned above.
3.4. Synthesis of substrate analogs and reaction intermediate analogs
Some new substrate analogs of PBG synthase were synthesized
as molecular probes for the study of rat PBG synthase as shown in
Fig. 5. Compounds 1 and 3 were synthesized from corresponding
ester through hydrolysis reactions. 5-Bromolaevulinic acid (2)
was synthesized from benzyl laevulinate through bromination
and hydrolysis reaction. Ethyl 4-acetylbutyrate was brominated

38
N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
O
O
O
PPh3
benzene
Ph3+P
O
O
Na2CO3
Br
O
O
Ph3P
O
H2O/CH2Cl2
22
23
O
O
O
O
PCC
CH2Cl2
O
OH
O
O
O
24
O
O
H3PO4
H2O
OH
O
O
HO
O
O
26
25
27
O
O
O
H2, Pd/C
O
O
O
O
1. NaOH/H2O
2. HCl/H2O
OH
HO
28
O
Fig. 6. Syntheses of diacids intermediate analogs for PBG synthase catalyzed reaction.
Two new intermediate analogs (dicarboxylic acids) for PBG syn-
supported by the comparison of the structures and kinetic con-
stants of compounds 1 and 6. Comparing with compound 16, com-
pound 14 lacks the amino group and has a much higher Ki value.
This further indicates that the amino group plays an important role
for the binding.
thase catalyzed reaction were also synthesized as shown in Fig. 6.
The key intermediate compound 25 was synthesized through Wit-
tig reaction of aldehyde (24) and triphenyl phosphonium ylide
(23). Compound 25 is identified as an E-a,b-unsaturated com-
pound, which was hydrolyzed in the solution of phosphoric acid
to give compound 26. The compound 25 was also hydrogenated
using palladium on charcoal catalyst to give dimethyl 4-oxo-5-
decenedioate (27), which was then hydrolyzed with NaOH and
acidified with HCl to give product 28.
Compounds 9 and 16 were found to be weak competitive inhib-
itors of rat PBG synthase with similar inhibition constants. Com-
pounds 11 and 21 were not competitive inhibitors but their IC50
values are at the same level as those of compounds 9 and 16. These
four compounds all have the amino groups and carbonyl groups,
but no carboxyl groups. All of them were not substrates, but weak
inhibitors of the enzyme, which indicates the carboxyl group is
important for the binding. The X-ray structure of yeast PBG syn-
thase complexed with substrate showed the bound ALA was linked
to Lys263 in the P-site and the carboxyl group was hydrogen
bonded to two highly conserved residues, Ser290 and Tyr329
[18]. Therefore, if the carboxyl group is replaced by other func-
tional groups, the analogs are not able to be fixed in the appropri-
ate position through hydrogen bonding, and can not take part in
subsequent catalytic reaction.
3.5. Inhibition study of rat PBG synthase with newly synthesized
reversible inhibitors
All above synthetic substrate analogs were tested for their
interactions with rat PBG synthase as inhibitors. Some substrate
analogs were found to be reversible enzyme inhibitors, and their
inhibition constants (Ki or IC50) are summarized in Table 2. If the
inhibition type of an inhibitor was not competitive, the IC50 value
was determined instead of the Ki value to compare the potency of
the inhibitors.
The compounds1, 3, 6, 9, 14, 16, and 28 were found to be com-
petitive inhibitors of rat PBG synthase with inhibition constants
ranging from 0.96 to 73.04 mM, indicating that their binding affin-
ity with this enzyme was quite different. The compounds 1 and 6
are very similar to the substrate and their Ki values are close to
the KM, indicating that the two compounds are well adopted into
the active site. Compound 1 has been found to be a good compet-
itive inhibitor of PBG synthases from Pea, E. coli., and yeast [44].
The Ki value of compound 1 is a little lower than that of compound
6, indicating the chain length of the molecule is more important
than the amino group in the terminal for the binding.
3.6. Inhibition study of rat PBG synthase with newly synthesized
irreversible inhibitors
In the absence of thiols, PBG synthases are susceptible to aerial
oxidation. However, many known inhibitors of PBG synthase were
found to react with thiols such as dithiothreitol (DTT) and b-
mercaptoethanol. Therefore, the inactivation experiments were
carried out in well degassed thiol-free buffers, and little loss of
activity was observed under these conditions. Some new substrate
analogs were synthesized in the present study and were found to
be irreversible enzyme inhibitors.
Compound 28 has two carboxyl groups and was expected to
bind to both the A-site and P-site of the enzyme whereas the com-
pound 1 was found to bind to the P-site of E. coli. enzyme [16]. The
Ki value of compound 28 is three times more than that of com-
pound 1, indicating that the additional carboxyl group and longer
carbon chain can not fit well into the A-site. The compound 28
has been found to be a good irreversible inhibitor to E. coli enzyme,
but was a weak inhibitor to human enzyme [28].
Compound 3 has one more methylene group than compound 1
and was assumed to bind to the P-site like compound 1, but the
larger Ki value indicates the longer carbon chain is unfavorable
for the binding. The Comparison of compounds 3 and 6 indicates
that the amino group is important for binding. This result is also
Table 2
Inhibition constants of the reversible inhibitors.
Compounds
Ki (mM)
IC50 (mM)
Laevulinic acid (1)
6-Amino-5-oxohexanoic acid (6)
4-Oxo-sebacic acid (28)
0.96
1.07
2.93
3.05
18.48
26.52
73.04
—
—
—
—
—
—
—
—
30.00
32.50
5-Oxo-hexanoic acid (3)
5-Amino-4-oxopentanenitrile (16)
1-Amino-4-hydroxy-2-butanone (9)
4-Oxo-pentanenitrile (14)
1-Amino-4-methoxy-2-butanone (11)
1-Amino-5-hydroxy-2-pentanone (21)
—

N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
39
100
a 100
80
b
90
80
70
60
50
40
30
20
10
0
60
40
20
0
0
2
4
6
8
10
0
2
4
6
8
10
Time (min)
Time (min)
100
100
90
80
70
60
50
40
30
20
10
0
c
d
90
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
0
5
10
15
20
Time (min)
Time (min)
Fig. 7. (a) Inactivation of rat PBG synthase by 1 mM (j) and 5 mM (N) 5-bromo-laevulinic acid (2); (b) Inactivation of rat PBG synthase by 0.1 mM of compounds 10 (ꢀ), 7 (j),
and 15 (N); (c) Inactivation of rat PBG synthase by 0.1 mM compounds 17 (j), 12 (N), 18 (ꢀ), and 13 (h); (d) Inactivation of rat PBG synthase by 0.1 mM (ꢀ) and 0.5 mM (N) 4-
oxo-decenedioic acid (26).
The inactivation experiments were carried out by incubating a
relatively concentrated solution of thiol-free enzyme with the ana-
logs. At various time intervals, aliquots were taken and added to
the assay mixture containing ALA at a final concentration of
10 mM. The time-dependent loss of activity was noted during the
incubations (Fig. 7). It should be noted that some compounds have
been found to be irreversible inhibitors of PBG synthases from var-
ious sources, but none of their apparent dissociation constants (Ki)
and the inactivation rate constants (kinact) have been determined
[50], because the reaction involves two ALAs and the inhibitions
do not fit into Michael–Menton equation. Our attempt to deter-
mine the Ki and kinact values of our synthetic analogs was not suc-
cessful because of the same reason.
therefore both of them are stronger enzyme inhibitors than com-
pounds 12 and 17. This may indicate that more than one nucleo-
philic residue could be labeled by the same or different enzyme
inhibitors leading to enzyme inactivation.
Compound 26 was found to be a moderate irreversible inhibitor
of rat PBG synthase, and its time-dependent inactivation of the en-
zyme is shown in Fig. 7d. In comparison, compound 28 is the cor-
responding saturated derivative of compound 26 and was found to
be a moderate competitive inhibitor. This indicates that the b posi-
tion of compound 26 could be attacked by a nucleophilic residue of
rat PBG synthase through conjugate addition reaction leading to
enzyme inactivation.
All these compounds could inactivate rat PBG synthase by
reacting with one or more nucleophilic residues. There are four
cysteines in the active site of PBG synthase (Fig. 4), which can sub-
stitute the bromine atom of the substrate analog or react with the
Almost all our irreversible inhibitors contain bromide except
compound 26, which is
a,b-unsaturated dicarboxylic acid.
5-Bromolaevulinic acid (2) was found to be a weak irreversible
enzyme inhibitor, as shown in Fig. 7a. This is consistent with
the inactivation properties of 5-chlorolaevulinic acid for bovine
and E. coli. PBG synthases [43,51,52]. Comparing with 5-chloro-
laevulinic acid, compound 2 has a bromine atom, a more sensi-
tive leaving group at C-5 position. Therefore, compound 2 is
assumed to inactivate the enzyme in the same mechanism as
that of 5-chlorolaevulinic acid. For bovine PBG synthase, 5-chlo-
rolaevulinic acid modified the protein at Cys223, to which the
ZnA was bound in the active site. It is possible that compound
2 is attacked by the same nucleophilic cysteine.
Compounds 7, 10, and 15 showed considerable inactivation of
PBG synthase at low concentration (0.1 mM) (Fig. 7b). Comparing
with compound 2, these compounds lack the carboxyl group that
is very important to bind the compound to the active site. Interest-
ingly, the tight binding analog 2 led to weak inactivation of the en-
zyme, while loose-binding analog 7, 10, and 15 gave strong
inactivation.
a,b-unsaturated analog as good nucleophilic reagents. If an analog
is bound to the active site tightly, the compound may only be able
to react with one of these cysteines. But if an analog is bound to the
active site loosely, it may react with more cysteines leading to
stronger inactivation. This study further increased our understand-
ing of the active site of PBG synthase.
Acknowledgment
The work described in this paper was financially supported by
the City University of Hong Kong.
References
[1] P.M. Shoolingin-Jordan, K.-M. Cheung, Compr. Nat. Prod. Chem. 4 (1999) 61–
107.
[2] P.M. Shoolingin-Jordan, Biochem. Soc. Trans. 26 (1998) 326–336.
[3] J.E. Cornah, M.J. Terry, A.G. Smith, Trends Plant Sci. 8 (2003) 224–230.
[4] D.V. Vavilin, W.F.J. Vermaas, Physiol. Plantarum 115 (2002) 9–24.
[5] J.A. Brusslan, M.P. Peterson, Photosynth. Res. 71 (2002) 185–194.
[6] J. Papenbrock, B. Grimm, Planta 213 (2001) 667–681.
[7] M. Moulin, A.G. Smith, Biochem. Soc. Trans. 33 (2005) 737–742.
[8] B. Grimm, Curr. Opin. Plant Biol. 1 (1998) 245–250.
Compounds 12, 13, 17, and 18 were all found to be irreversible
inhibitors of rat PBG synthase as shown in Fig. 7c, which are simple
substrate analogs with bromine atom. Compounds 13 and 18 have
two bromine atoms to react with more nucleophilic residues,

40
N. Li et al. / Bioorganic Chemistry 37 (2009) 33–40
[9] P.M. Anderson, R.J. Desnick, J. Biol. Chem. 254 (1979) 6924–6930.
[31] E.K. Jaffe, L. Stith, Am. J. Hum. Genet. 80 (2007) 329–337.
[32] R. Akagi, C. Nishitani, H. Harigae, Y. Horie, L. Garbaczewski, A. Hassoun, R.
Mercelis, L. Verstraeten, S. Sassa, Blood 96 (2000) 3618–3623.
[33] N. Ishida, H. Fujita, T. Noguchi, M. Doss, A. Kappas, S. Sassa, Biochem. Biophys.
Res. Commun. 172 (1990) 237–242.
[10] P. Spencer, P.M. Jordan, Biochem. J. 290 (Pt 1) (1993) 279–287.
[11] W. Liedgens, C. Lutz, H.A. Schneider, Eur. J. Biochem. 135 (1983) 75–79.
[12] Q.F. Boese, A.J. Spano, J.M. Li, M.P. Timko, J. Biol. Chem. 266 (1991) 17060–
17066.
[13] T. Selwood, L. Tang, S.H. Lawrence, Y. Anokhina, E.K. Jaffe, Biochemistry 47
(2008) 3245–3257.
[14] S.H. Lawrence, U.D. Ramirez, L. Tang, F. Fazliyez, L. Kundrat, G.D. Markham, E.K.
Jaffe, Chem. Biol. 15 (2008) 586–596.
[15] P.T. Erskine, N. Senior, S. Awan, R. Lambert, G. Lewis, I.J. Tickle, M. Sarwar, P.
Spencer, P. Thomas, M.J. Warren, P.M. Shoolingin-Jordan, S.P. Wood, J.B.
Cooper, Nat. Struct. Biol. 4 (1997) 1025–1031.
[34] R. Sommer, D. Beyersmann, J. Inorg. Biochem. 20 (1984) 131–145.
[35] I. Tsukamoto, T. Yoshinaga, S. Sano, Biochim. Biophys. Acta 570 (1979) 167–
178.
[36] J.B. Rocha, S.M. Tuerlinckx, M.R. Schetinger, V. Folmer, Toxicol. Appl.
Pharmacol. 200 (2004) 169–176.
[37] R.C. Bolzan, V. Folmer, M. Farina, G. Zeni, C.W. Nogueira, J.B. Rocha, T.
Emanuelli, Pharmacol. Toxicol. 90 (2002) 214–219.
[16] P.T. Erskine, E. Norton, J.B. Cooper, R. Lambert, A. Coker, G. Lewis, P. Spencer, M.
Sarwar, S.P. Wood, M.J. Warren, P.M. Shoolingin-Jordan, Biochemistry 38
(1999) 4266–4276.
[17] P.T. Erskine, L. Coates, D. Butler, J.H. Youell, A.A. Brindley, S.P. Wood, M.J.
Warren, P.M. Shoolingin-Jordan, J.B. Cooper, Biochem. J. 373 (2003) 733–738.
[18] P.T. Erskine, R. Newbold, A.A. Brindley, S.P. Wood, P.M. Shoolingin-Jordan, M.J.
Warren, J.B. Cooper, J. Mol. Biol. 312 (2001) 133–141.
[19] P.T. Erskine, L. Coates, R. Newbold, A.A. Brindley, F. Stauffer, S.P. Wood, M.J. Warren,
J.B. Cooper, P.M. Shoolingin-Jordan, R. Neier, FEBS Lett. 503 (2001) 196–200.
[20] N. Frankenberg, P.T. Erskine, J.B. Cooper, P.M. Shoolingin-Jordan, D. Jahn, D.W.
Heinz, J. Mol. Biol. 289 (1999) 591–602.
[21] L. Coates, G. Beaven, P.T. Erskine, S.I. Beale, Y.J. Avissar, R. Gill, F. Mohammed,
S.P. Wood, P. Shoolingin-Jordan, J.B. Cooper, J. Mol. Biol. 342 (2004) 563–570.
[22] J. Kervinen, E.K. Jaffe, F. Stauffer, R. Neier, A. Wlodawer, A. Zdanov,
Biochemistry 40 (2001) 8227–8236.
[38] M. Farina, R. Brandao, F.S. de Lara, L.B. Pagliosa, F.A. Soares, D.O. Souza, J.B.
Rocha, Toxicology 184 (2003) 179–187.
[39] C.W. Nogueira, V.C. Borges, G. Zeni, J.B. Rocha, Toxicology 191 (2003) 169–
178.
[40] T. Emanuelli, J.B. Rocha, M.E. Pereira, P.C. Nascimento, D.O. Souza, F.A. Beber,
Pharmacol. Toxicol. 83 (1998) 95–103.
[41] P. Spencer, P.M. Jordan, Biochem. J. 300 (Pt 2) (1994) 373–381.
[42] P.S. Ebert, R.A. Hess, B.C. Frykholm, D.P. Tschudy, Biochem. Biophys. Res.
Commun. 88 (1979) 1382–1390.
[43] J.S. Seehra, P.M. Jordan, Eur. J. Biochem. 113 (1981) 435–446.
[44] N.M. Senior, K. Brocklehurst, J.B. Cooper, S.P. Wood, P. Erskine, P.M.
Shoolingin-Jordan, P.G. Thomas, M.J. Warren, Biochem. J. 320 (Pt 2) (1996)
401–412.
[45] K.M. Cheung, P. Spencer, M.P. Timko, P.M. Shoolingin-Jordan, Biochemistry 36
(1997) 1148–1156.
[23] F. Frere, W.D. Schubert, F. Stauffer, N. Frankenberg, R. Neier, D. Jahn, D.W.
Heinz, J. Mol. Biol. 320 (2002) 237–247.
[46] M. Sodeoka, C.J. Larson, L. Chen, K.P. LeClair, M. Verdine, L. Gregory, Bioorg.
Med. Chem. Lett. 3 (1993) 1089–1094.
[24] F. Frere, M. Nentwich, S. Gacond, D.W. Heinz, R. Neier, N. Frankenberg-Dinkel,
Biochemistry 45 (2006) 8243–8253.
[25] D.L. Nandi, D. Shemin, J. Biol. Chem. 243 (1968) 1236–1242.
[26] P.M. Jordan, J.S. Seehra, FEBS Lett. 114 (1980) 283–286.
[27] C. Jarret, F. Stauffer, M.E. Henz, M. Marty, R.M. Luond, J. Bobalova, P.
Schurmann, R. Neier, Chem. Biol. 7 (2000) 185–196.
[28] E.K. Jaffe, J. Kervinen, J. Martins, F. Stauffer, R. Neier, A. Wlodawer, A. Zdanov, J.
Biol. Chem. 277 (2002) 19792–19799.
[29] B. Fernandez-Cuartero, J.L. Rebollar, A. Batlle, R. Enriquez de Salamanca, Int. J.
Biochem. Cell. Biol. 31 (1999) 479–488.
[47] X. Chu, D. Li, Protein Expr. Purif. 27 (2003) 165–170.
[48] D. Mauzerall, S. Granick, J. Biol. Chem. 219 (1956) 435–446.
[49] B.T. Burlingham, T.S. Widlanski, J. Chem. Educ. 80 (2003) 214–218.
[50] F. Stauffer, E. Zizzari, C. Engeloch-Jarret, J.P. Faurite, J. Bobalova, R. Neier,
Chembiochem 2 (2001) 343–354.
[51] E.K. Jaffe, W.R. Abrams, H.X. Kaempfen, K.A. Harris Jr., Biochemistry 31 (1992)
2113–2123.
[52] E.K. Jaffe, M. Volin, C.B. Myers, W.R. Abrams, Biochemistry 33 (1994) 11554–
11562.
[53] J.D. Thompson, T.J. Gibson, F. Plewniak, F. Jeanmougin, D.G. Higgins, Nucleic
Acids Res. 25 (1997) 4876–4882.
[54] R. Chenna, H. Sugawara, T. Koike, R. Lopez, T.J. Gibson, D.G. Higgins, J.D.
Thompson, Nucleic Acids Res. 31 (2003) 3497–3500.
[30] J.B. Souza, J.B. Rocha, C.W. Nogueira, V.C. Borges, R.R. Kaizer, V.M. Morsch, V.L.
Dressler, A.F. Martins, E.M. Flores, M.R. Schetinger, Clin. Biochem. 40 (2007)
321–325.