Download Leishmania donovani - Oxford Academic

794–802
Nucleic Acids Research, 2002, Vol. 30, No. 3
© 2002 Oxford University Press
An insight into the active site of a type I DNA
topoisomerase from the kinetoplastid protozoan
Leishmania donovani
Aditi Das, Chhabinath Mandal1, Arindam Dasgupta, Tanushri Sengupta and
Hemanta K. Majumder*
Molecular Parasitology Laboratory and 1Drug Design, Development and Molecular Modelling, Indian Institute of
Chemical Biology, 4, Raja S.C. Mullick Road, Calcutta 700032, India
Received August 29, 2001; Revised and Accepted November 21, 2001
ABSTRACT
DNA topoisomerases are ubiquitous enzymes that
govern the topological interconversions of DNA
thereby playing a key role in many aspects of nucleic
acid metabolism. Recently determined crystal structures
of topoisomerase fragments, representing nearly all
the known subclasses, have been solved. The type IB
enzymes are structurally distinct from other known
topoisomerases but are similar to a class of enzymes
referred to as tyrosine recombinases. A putative
topoisomerase I open reading frame from the kinetoplastid Leishmania donovani was reported which
shared a substantial degree of homology with type IB
topoisomerases but having a variable C-terminus. Here
we present a molecular model of the above parasite
gene product, using the human topoisomerase I crystal
structure in complex with a 22 bp oligonucleotide as a
template. Our studies indicate that the overall structure
of the parasite protein is similar to the human enzyme;
however, major differences occur in the C-terminal
loop, which harbors a serine in place of the usual
catalytic tyrosine. Most other structural themes
common to type IB topoisomerases, including
secondary structural folds, hinged clamps that open and
close to bind DNA, nucleophilic attack on the scissile
DNA strand and formation of a ternary complex with the
topoisomerase I inhibitor camptothecin could be visualized in our homology model. The validity of serine
acting as the nucleophile in the case of the parasite
protein model was corroborated with our biochemical
mapping of the active site with topoisomerase I
enzyme purified from L.donovani promastigotes.
INTRODUCTION
DNA topoisomerases represent a family of DNA-processing
enzymes that essentially solve the topological problems arising
PDB accession no. 1JUW
in cells during DNA replication, transcription, recombination,
chromatin assembly and chromosome segregation (1–2). They
are broadly classified into type I and type II topoisomerases.
Type II enzymes are dimeric, cleaving both strands of a duplex
and passing another region of intact DNA through this transient double-stranded break (in an ATP-dependent manner)
resulting in changes in linking number in steps of two. In
contrast, type I enzymes are monomeric, require no additional
energy cofactor and they transiently nick a single strand of
DNA duplex, allowing for single-step changes in the linking
number of circular DNAs. Type I enzymes are further divided
into type IA (bacterial) and IB (eukaryotic and vaccinia virus)
enzymes. Type IA topoisomerases are able to relax only
negatively supercoiled DNA, require magnesium and a singlestranded stretch of DNA for function and become transiently
covalently attached to the 5′ end of the nicked single strand.
However, topoisomerases IB are able to relax both positively
and negatively supercoiled DNA with equal efficiency, do not
require a single-stranded region of DNA or metal ions for
function and form a transient covalent bond with the 3′ end of
the nicked DNA strand. With the structural characterizations of
fragments of Escherichia coli topoisomerase I (a type IA
enzyme) (3,4), yeast topoisomerase I (type IB enzyme) (5),
vaccinia topoisomerase (type IB enzyme) (6), and more
recently determined structures of human topoisomerase I in
complex with DNA (7–9) and of a catalytic fragment of
vaccinia topoisomerase (10), it is clear that the type IA and IB
topoisomerases share no structural similarity (11).
The fact that topoisomerases are proven drug targets (12,13)
has led to the study of these enzymes from an array of parasitic
organisms, so as to delineate common and distinguishing
features between the host and parasite enzymes. Work on DNA
topoisomerases from the kinetoplastid protozoan parasites (14)
has been a major focus of interest in this regard. Recently, the
DNA sequence of a topoisomerase I-like gene from the kinetoplastid Leishmania donovani has been reported (15). The
deduced amino acid sequence of this gene showed an extensive
degree of homology with the central core of other eukaryotic
type IB topoisomerases, including several conserved motifs.
Only the C-terminus of the parasite protein was markedly
*To whom correspondence should be addressed. Tel: +91 33 473 6793; Fax: +91 33 473 5197; Email: [email protected]
The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors
Nucleic Acids Research, 2002, Vol. 30, No. 3
dissimilar from the C-terminal domain of type IB topoisomerases
that normally contains the active site tyrosine within a conserved
SKXXY motif (15). Prediction analysis using NetPhos indicated
that within the C-terminal domain of this parasite protein,
serine residues and not tyrosine had maximum probability of
acting as potential phosphorylation sites for enzyme catalysis.
From our laboratory, we have previously reported ATPindependent, Mg-dependent type I topoisomerase activity from
cell extracts of L.donovani promastigotes (16), which relaxes
both positive and negative supercoils in topological steps of
unity. The enzyme is sensitive to topoisomerase I-specific
inhibitors like camptothecin (CPT) and is also a target for
potential chemotherapy of anti-leishmanial drugs (17–19). The
homogenously purified active enzyme is 67 kDa. The present
work reports the mapping of the active site residue of the
endogenously purified L.donovani topoisomerase I (LdTOP1)
enzyme. The work also includes molecular modeling studies to
gain an insight into the tertiary structure of the protein from the
deduced amino acid sequence of the LdTOP1-like gene, its
binary complex with the substrate DNA duplex and a ternary
complex involving the enzyme, DNA and CPT, a potent inhibitor
of the enzymatic process. The structure prediction and computational analysis was based on available crystal structures of
human topoisomerase I. These studies indicate a common
structural and functional theme for the human topoisomerase I
and the parasite topoisomerase I-like gene product. So this
work lends strong support to the existence of a topoisomerase
I protein in the kinetoplastid protozoon L.donovani, which
diverges from other eukaryotic type IB topoisomerases in the
possession of a variable active site.
MATERIALS AND METHODS
Enzyme, DNA and chemicals
Type I topoisomerase was purified from L.donovani strain
D1700 (MHOM/SD/68/SI) promastigotes as previously
described by Chakraborty et al. (16). Rat liver topoisomerase I
was purified according to a published protocol (20). Plasmid
pHOT1 DNA (containing the high-affinity topoisomerase I
cleavage site that is derived from the Tetrahymena ribosomal
gene repeat, a hexadecameric sequence) was purchased from
TopoGEN. The nick translation kit was purchased from Boehringer
Mannheim (Roche). DNase I and Sephadex G-25 were from
Amersham Pharmacia Biotech. Thin layer chromatography
(TLC) was performed on silica gel plates (Macherey-Nagel
Polygram SilG).
Biochemical analysis of covalent linkage through catalytic
amino acid residue
A 50 µl reaction mixture containing 50 mM PIPES, pH 6.0,
75 mM KCl, 10 mM MgCl2, 0.5 mM DTT, 0.5 mM EDTA,
30 µg/ml BSA, 25 ng of [α-32P]dCTP nick-translated pHOT1
DNA and 80 U purified LdTOP1 enzyme was incubated at
37°C for 30 min. The reaction was terminated by adding
NaOH to 0.2 M. DNase I treatment was done at a concentration
of 100 µg/ml followed by further addition of 30 U Bal31
nuclease. Reactions were stopped by adding 10 mM EDTA
and 0.5% SDS before processing the DNA–protein complexes
through two cycles of a Sephadex G-25 spin column and
acetone precipitation. Next, proteolysis was carried out with
795
50 µg/ml of proteinase K at 37°C for 30 min, the digested
products vacuum dried and re-suspended in 30 µl of constant
boiling with 6 M HCl. After removal of HCl, the dried hydrolysates were resuspended in 10 µl of H2O containing 200 nmol
each of phosphoserine, phoshphothreonine and phosphotyrosine. A few microliters of the samples were spotted on the
silica gel plate. Phosphoamino acid mix markers were also
spotted on the same plate in the same horizontal line with the
earlier spotted samples. TLC, ninhydrin staining and autoradiography of the resolved amino acids of the spots were
carried out following published procedures (21).
Molecular modeling and structural analysis
Predictions of three-dimensional structures of topoisomerase I,
its binary complex with DNA and ternary complex with DNA
and CPT were done by knowledge-based homology modeling
using InsightII 98.0 of MSI (Biosym Technologies), ABGEN
(22) and our in-house package of MODELYN and ANALYN
(23) in UNIX as well as in the MS Windows environment.
Energy minimization and molecular dynamics were performed
with the InsightII 98.0/Discover package using cff91 forcefield on a
Silicon GraphicsR OCTANE workstation. Energy minimizations
were done with a convergence criterion of 0.001 kcal/mol,
using a combination of steepest descent and conjugate gradient
methods of 100 steps each; these steps were repeated until
satisfactory conformational parameters were obtained. Molecular
dynamics simulations were carried out using a time step of 1 fs
over a period of 0.1 µs. Distance constraints were applied to
the hydrogen bonding of base pairs of the DNA duplex to keep
them intact while running minimization and dynamics on the
complexes. Secondary structures were predicted from
sequences using SOPM and analyzed from structures using
MODELYN and RIBBONS. The electrostatic potential
surface of the protein model was determined by MOLMOL
(24). TREEVIEW software package version 1.5 (25) was used
for the construction of a phylogenetic tree, which read
PHYLIP style treefiles produced by CLUSTAL W (26).
RESULTS
Active site mapping of the L.donovani topoisomerase I
enzyme
A 67 kDa protein with topoisomerase I activity had been
homogenously purified from the nuclear extracts of L.donovani
promastigotes (16). From its enzymatic characteristics, the
protein could be classified as an eukaryotic type I topoisomerase.
When the LdTOP1 was incubated with supercoiled pHOT1
DNA followed by termination of reaction with SDS, nicked
species of DNA were the main end products of the reaction,
which became further enriched in the presence of CPT (19).
Analysis of CPT-induced DNA breaks by an assay using endlabeled DNA (27) show that the Leishmania enzyme is
covalently linked to the 3′ end of the broken DNA (data not
shown). In order to investigate the chemical nature of the covalent
linkage formed between the substrate DNA and the LdTOP1,
biochemical mapping studies were carried out by labeling the
purified enzyme as described in Materials and Methods, positive
control being set up with a rat liver topoisomerase I. Acid
hydrolysis of the labeled oligonucleotide–protein complexes
yielded orthophosphate as the major labeled species, as shown
796
Nucleic Acids Research, 2002, Vol. 30, No. 3
Figure 1. Identification of the phosphoamino acid linkage in the hydrolysate of
topoisomerase I–DNA covalent complexes. Lane 1 shows the migration profile
of the standard cold phosphoamino acids. Lanes 2 and 3 represent the migration of
32P-labeled acid hydrolysates from rat liver and L.donovani samples, respectively.
The positions of the marker compounds (seen by ninhydrin staining) are indicated
by dotted outlines. The point of application of samples, i.e. the origin (ori), as
well as the solvent front run in the silica gel plate are marked.
by TLC analysis of the products (Fig. 1). In the case of the
LdTOP1, a significant fraction of the radioactivity was found
to co-chromatograph with chemically synthesized phosphoserine/phosphothreonine on thin-layer silica gel in the solvent
system mentioned in the Materials and Methods. Under similar
conditions, for the rat liver topoisomerase I, the radioactive
spot was found to co-migrate with standard phosphotyrosine.
Alkaline hydrolysis of the labeled protein led to the disappearance
of the radioactive signal of the LdTOP1–oligo complex on
TLC plates (data not shown), further confirming that the labeled
amino acid is either alkali-labile phosphoserine or phosphothreonine. The signal of the mammalian topoisomerase I–oligo
complex was found to be resistant to alkaline pH. Thus, the
covalent linkage formed between the L.donovani enzyme and
DNA was identified as Ser-P/Thr-P and not the usual Tyr-P as
observed for other eukaryotic topoisomerase I (28).
Three-dimensional model of putative Leishmania
topoisomerase I
Based on combined primary sequence and secondary structure
alignments against the known crystal structures of human
topoisomerase I (PDB accession nos 1A36, 1A35 and 1A31), a
three-dimensional model (PDB accession no. 1JUW) has been
presented here for a LdTOP1-like sequence. Like other eukaryotic
type I enzymes, the human topoisomerase I is a monomeric
protein and it contains 765 amino acids (29). Limited proteolysis
studies of human topoisomerase I (30) indicate that it is comprised
of four structural domains: a highly charged NH2-terminal domain,
a major positively charged core domain, followed by a small linker
domain and finally a COOH-terminal domain containing the active
site residue. In the crystal structures of engineered catalytically
active components of human topoisomerase I, the hydrophilic
N-terminal (which is known to be dispensable for activity) had
been removed and the protein was in a complex with a 22 bp
blunt-ended duplex DNA oligonucleotide. Using these structures
as a template, the approximate protein model of LdTOP1-like
protein (PDB accession no. 1JUW) was simulated (Fig. 2A) in
complex with an identical duplex DNA. The DNA sequence
had the preferred consensus sequence of ACTT∧ at positions –4
to –1 in the scissile strand adjacent to the cleavage site (∧), that
corresponded to the high affinity eukaryotic topoisomerase I
binding site from Tetrahymena rDNA (31). The most reliable
homology model generated for the parasite protein comprised
544 residues and showed an RMSD of 0.8 Å over Cα positions
for 445 equivalent residues of the human topoisomerase I
structure, 1A36, the superposition being depicted in Figure 3.
The Leishmania model appeared to be a bi-lobed protein that
clamps completely around its DNA substrate. One lobe
(comprising two domains resembling the core subdomains I
and II of human topoisomerase I) sits on top of the DNA
duplex (as oriented in Fig. 2A) just like the ‘cap’ structure
referred to in the human protein (32). It is composed of mixed
α and β secondary-structural elements and contains two unique
‘nose-cone’ helices that extend 25 Å from the body of the
molecule (Fig. 2B). The surfaces of these helices, which are
proximal to the DNA, are highly positively charged and are
positioned above the DNA major groove, but do not directly
interact with the DNA. The second lobe (also comprising two
domains; the former resembling the core subdomain III of
human topoisomerase I and the latter representing a short
variable C-terminal domain) sits below the DNA, is composed
of an all α-helical structure except for one three-stranded β-sheet
(Fig. 2B) and contains the catalytic residues implicated in the
strand cleavage and religation reactions. These two lobes are
covalently linked through a continuous helical chain on one
side of the DNA molecule labeled the ‘connector’, and touch
each other in a small contact region, similar to the ‘lip’ structure in
the human topoisomerase I (32). The coiled-coil linker domain
extends 50 Å from the body of the second lobe and exposes a
highly positively charged face towards the DNA similar to the
‘nose-cone’ helices; however, this domain also fails to make
direct interactions with the DNA.
Protein–DNA contacts in the Leishmania protein–DNA model
The model of the Leishmania protein in complex with topoisomerase I–suicide DNA substrate reveals how the parasite
enzyme can bind to the DNA and align it for strand cleavage
and covalent attachment. In fact, the Leishmania protein is
observed to contact the DNA (Fig. 4) mostly via protein to
DNA phosphate interactions just like the human topoisomerase I
(8,11). Only four protein–DNA contacts derive from two residues, Arg190 and Lys352, which enter the minor groove of the
DNA, adjacent to the cleavage site. One of these contacts is base
specific, i.e. the hydrogen bond between the side chain NH2 group
Nucleic Acids Research, 2002, Vol. 30, No. 3
797
Figure 3. Superposition of the modeled LdTOP1 (green) on human
topoisomerase I (red), both proteins being depicted in Cα trace.
topoisomerase I, interacting more frequently with the 5 bp
upstream of the cleavage site than it does with an 8 bp region
downstream of the cleavage site. The electrostatic potential
surface of the Leishmania protein model that plays an important
role in its circumferential binding around the negatively
charged DNA duplex has been illustrated in Figure 5.
Putative active site view of the Leishmania protein
homology model
Figure 2. Ribbon representation of the LdTOP1 model (PDB accession no. 1JUW).
(A) The homology model of LdTOP1 in complex with a 22 bp DNA duplex (31)
shown in sticks. The LdTOP1 model is colored domain-wise. Core subdomains I,
II and III are colored yellow, blue and red, respectively. The linker and C-terminal
domain are shown in orange and green, respectively. The part rendered in dark
red represents the missing part in the human topoisomerase I crystal structure.
(B) The homology model of LdTOP1 depicting its secondary structures: α-helices
rendered in sky blue, β-sheets in yellow and random coils in orange. The N- and
C-termini of the modeled protein are marked. The length of the nose-cone helices
and linker are depicted in angstroms. Also marked are the ‘connector’- and
‘cap’-like structures.
of Lys352 and the oxygen atom of the favored –1 thymidine base.
This lysine residue is conserved at this position in all known
sequences of type IB topoisomerases. This interaction explains
the observed preference for a thymidine at the –1 position
adjacent to the cleavage site, as thymidine is the only base with
a hydrogen bond acceptor freely available at this position in the
minor groove. The parasite enzyme even exhibits a distinct
asymmetry in its contact with the DNA (Fig. 4) like the human
The model of the LdTOP1-like gene provides a view of the
detailed architecture of the proposed active site of the enzyme
(Fig. 6). For eukaryotic type IB topoisomerases the DNA
strand cleavage reaction usually proceeds via a nucleophilic
attack on the scissile phosphorus atom by the catalytic residue.
Interestingly, sequence analysis revealed the absence of the
usual SKXXY motif within the C-terminal domain of the parasite
topoisomerase I-like open reading frame and a serine (Ser553)
aligned with the conserved Tyr723 catalytic residue of the
human enzyme. The Ser553 was found to be positioned
perfectly for nucleophilic attack and subsequently for covalent
attachment to the 3′ end of the broken strand because it is
within suitable distance of the O5′-P scissile bond. Other
residues near the phosphodiester bond that is cleaved are two
arginine and one histidine residue found to be conserved with
the human topoisomerase I enzyme in structurally equivalent
positions. For the catalytic mechanism of LdTOP1, it seems
reasonable to propose that this residue triad contributes to the
stabilization of the pentavalently co-ordinated transition state
intermediate through hydrogen-bonding; Arg314 and Arg410
lying within hydrogen-bonding distance of one non-bridging
oxygen, O1, and His453 lying near the other non-bridging
oxygen, O2. In the LdTOP1 model, there is no side-chain atom
within a 4 Å radius of the hydroxyl of Ser553 to act as a
general base for activating the nucleophilic oxygen. The
closest such amino acid is His453, which is unlikely to be
involved as a general base since the Nε2 atom of His453 is
4.5 Å away from the hydroxyl oxygen of Ser553. Analogously
the Nε2 atom of His632 is 4.3 Å away from what would be the
798
Nucleic Acids Research, 2002, Vol. 30, No. 3
Figure 4. Schematic representation of the protein–DNA contacts of modeled LdTOP1 with the 22 bp DNA duplex. Major interactions within distances of 3.5 Å
are depicted. The phosphate-deoxyribose backbone of the intact strand is shown in blue, and that of the scissile strand is shown in magenta to indicate regions
upstream or downstream of the cleavage site. The observed interactions are limited exclusively to protein–phosphate interactions (large gray dots). The –1 and +2 bases are
the only bases that are contacted by the protein, as indicated with bold black boxes. Both contacts occur in the minor groove, one between the Lys352 and O2 atom
of –1 thymidine and the other between the Arg190 and N3 atom of the guanidine base. Small red dots indicate interactions between Arg190 and the ribose oxygens
in positions +2 and +3 on the cleaved strand. The cleavage site, the catalytic triad of Arg314, Arg410 and His453 plus the serine nucleophile (Ser553) contacting
the scissile phosphate at the active site are marked.
hydroxyl oxygen of Tyr723 in the crystal structure of the
reconstituted human topoisomerase I non-covalent complex. A
possible explanation drawn from analogy with the human
structure is that the proton of the nucleophile might be
abstracted by a hypothetical water molecule as catalysis
proceeds. Thus, analysis of the position of participating
catalytic residues in the Leishmania protein model is compatible
with the human topoisomerase I crystal structure in many
aspects but deviating from the host enzyme in its possession of a
serine in place of the universal tyrosine. A view of the C-terminal
domains of the human topoisomerase I with its catalytic Tyr723
and the Leishmania protein with its putative catalytic Ser553 is
presented in Figure 7.
Proposed mechanism of topoisomerization by
Leishmania-modeled protein
The spatial arrangement of the DNA in the model of LdTOP1,
based on the human topoisomearse I co-crystal structure,
renders unlikely an enzyme-bridging mechanism for DNA
relaxation analogous to what has been proposed for Type IA
topoisomerase (Escherichia coli). Stewart et al. (8) proposed a
‘controlled rotation’ mechanism for the relaxation of DNA
superhelical tension by type IB topoisomerases from observations
concerning the nature of the human topoisomerase I–DNA
interactions. Many of these mechanistic features could be
visualized in the homology model of LdTOP1–DNA complex.
An alternative hypothesis is shown in Figure 8, where it is
proposed that once cleavage has occurred, the Leishmania
enzyme opens up at the putative hinge region located at the top
of the connector in subdomain III to allow rotation and perhaps
assumes a conformation approximating that proposed for the
DNA-free form of the enzyme. Opening and closing of the
protein clamp during DNA binding and release must involve
the breaking of the connection between the lips and the lifting
of the cap away from the base. In this model, there may be little
or no hindrance to DNA rotation, DNA binding being naturally
Nucleic Acids Research, 2002, Vol. 30, No. 3
Figure 5. Surface representations of the Leishmania protein model (MOLMOL)
revealing the prominent groove where DNA duplex can bind. (A) Region spanning
core subdomains I and III including the putative hinge of the protein. (B) Part of
subdomain III, linker and C-terminal domain. The surface is colored to represent
electrostatic potential (–15kbT/e to +15kbT/e) where kbT is the product of the
Boltzmann constant and temperature and e is the electronic charge, with
negatively and positively charged regions shown in red and blue, respectively.
directed by the surface and charge complementation. Either
model could accommodate multiple rotation events before
religation of the DNA.
Proposed mode of binding of CPT to the Leishmania
protein–DNA covalent complex
Topoisomerase I is the sole intracellular target for a well-studied
class of compounds, the CPTs, which have emerged as potent
anticancer drugs and which act by stabilizing the covalent
topoisomerase I–DNA complex. CPT inhibition of LdTOP1
has been reported (19). Recently, some modes of binding of
CPT to the covalent complex of human topoisomerase I and
DNA have been proposed (7,33,34). The model of Redinbo et al.
(7) places the CPT molecule into the DNA duplex in a position
799
Figure 6. Detailed architecture of the active site of putative LdTOP1. (A) Structural
view of the amino acid residues in the putative active site of the Leishmania
protein model. The three-dimensional relationship between the side chains of
residues Arg314, Arg410, His453, Lys352 and Ser553 and the +1 and –1 residue
belonging to the scissile strand of DNA is shown. The depiction is in solid
sticks (carbon, green; nitrogen, blue; oxygen, red; and phosphorus, pink).
Hydrogen bonds (dotted lines) and bond distances (in angstroms) are shown.
(B) Schematic diagram depicting a possible intermediate stage in the cleavage
phase of catalysis by the putative LdTOP1. The active site residues Arg314,
Arg410 and His453, from subdomain III as well as the putative catalytic
Ser553 of the C-terminal domain, are depicted surrounding the scissile
phosphate (+1), which is coordinated in a trigonal bipyramidal fashion.
that partially overlaps with the +1 guanine of the scissile
strand; this placement requires that the guanine base be flipped
out of the duplex and stacked on the planar CPT. Our results on
placing the CPT in an identical manner with the modeled
Leishmania protein–DNA covalent complex highlighted that
this alkaloid exerts an analogous mode of action on this
parasite protein. In this model (Fig. 9), the immutable lactone
800
Nucleic Acids Research, 2002, Vol. 30, No. 3
Figure 7. Close-up views of the active site pocket of human topoisomerase I
(A) and the putative LdTOP1 (B). The proteins are depicted in ribbon representation,
coloring being according to amino acid sequence. The positions of the nucleophilic
tyrosine (Phe723Tyr) and serine (Ser553) are marked, respectively, in the
active sites of human topoisomerase I (A) and the putative LdTOP1 model (B).
Figure 8. A possible mechanism of topoisomerization by the Leishmania protein.
The bilobed protein model (space-filling) is colored blue for the ‘cap’ lobe
(subdomains I and II) and yellow for the ‘catalytic’ lobe (subdomain III, linker
and the C-terminal domain) the view being down the axis of the bound DNA.
The DNA is represented as space-filling and colored according to usual atom
colors. (A) The modeled closed form of the Leishmania protein presumably
catalyzes cleavage of the scissile strand upon binding the DNA. (B) The parasite
protein modeled into an open configuration for DNA binding by allowing
upward movement at the putative hinge point.
and 20(S)-hydroxyl moieties of CPT are found to interact with
Arg190 and Asp353 and form a stacking interaction with
guanine (favored base in the scissile +1 position for CPT
action) such that the guanine N-3 is in close proximity to the C-7
of CPT. The amino acid residues whose mutation renders the
human topoisomerase I insensitive to CPT could also be
identified as important in the Leishmania protein, namely
Arg190 and Asp353, equivalent to Arg364 and Asp533 of
human topoisomerase I. Only the stacking interaction of CPT
with Asn722 of human topoisomerase I was missing in the
parasite enzyme. The important feature highlighted in the
model of the ternary complex is that the re-entry of the +1
guanine into the DNA duplex is hindered by the stacked CPT
and that the 5′-hydroxyl of the dG(+1) is prevented from
Figure 9. (A) Model of the inhibitor CPT bound to the covalent complex of
putative LdTOP1 and DNA. In this model, CPT (space-filling representation
with nitrogen atoms in blue, oxygen in red, carbon in green and the A, B and E
rings of the molecule indicated) stacks in the DNA duplex in place of the
guanine base at the +1 Gua position of the scissile strand, which has been
flipped out of the duplex. CPT makes a stacking interaction with this +1 base
bringing the C7 atom of the drug in close proximity to the N3 atom of +1 Gua.
Two of the CPT-resistant mutations reported in human, represented by Arg190
and Asp353 in the parasite protein model, were found to interact in an
analogous manner with the required lactone and 20(S)-hydroxyl moieties of
CPT. (B) A schematic representation of the key hydrogen bonds and ring-stacking
interactions made between the LdTOP1–DNA covalent complex model and
CPT in the proposed CPT-binding mode is shown.
attacking the enzyme–DNA phosphate bond (phosphoserine in
this case) thereby explaining the inhibition of topoI-mediated
DNA religation.
DISCUSSION
With both rat liver and L.donovani DNA topoisomerase I
enzymes, formation of the covalent complex with 32P-labeled
supercoiled DNA, followed by nucleolytic treatment of the
complex, yield a 32P-labeled protein. Analysis of the product
after hydrolysis and TLC identified Tyr-P as the labeled amino
acid in the case of rat liver topoisomerase I indicating that the
covalent bond that forms between DNA and the rat liver
enzyme is phosphotyrosine, like topoisomerase enzymes from
other species. However, similar analysis of the chemical nature of
Nucleic Acids Research, 2002, Vol. 30, No. 3
the labeled LdTOP1–DNA products indicated that the covalent
linkage is most likely a phosphoserine/phosphothreonine and
not a phosphotyrosine.
Recently, Brocoli et al. (15) reported a sequence from
L.donovani that codes for a putative topoisomerase I protein,
identical to other topoisomerase I sequences deposited in the
GenBank. The predicted protein analysis which uses MAXHOM
multiple sequence alignment aligns the query sequence of
putative LdTOP1 with different eukaryotic type IB DNA
topoisomerase sequences with an identity score of 37% and a
similarity score of 49% with the human topoisomerase I. This
homology allowed us to build a three-dimensional model of
the putative LdTOP1 protein based on the crystal structure of
human topoisomerase I (PDB accession no. 1A36). The model
predicts that the parasite protein adopts an overall conformation
similar to that of the human topoisomerase I, organized into
four subdomains. The superposition of the three-dimensional
model on the human topoisomerase I structure clearly showed
that most of the secondary structural elements are conserved in
both the proteins and that they are largely present in the same
relative positions. Other interesting features of the Leishmania
protein model are the presence of ‘connector’-, ‘cap’- and ‘lip’-like
structures similar to that of the human topoisomerase I. The
majority of the protein–DNA contacts in our model were found
to be equivalent to those in the human topoisomerase I–DNA
co-crystal structure, the protein contacting only the central 10 bp
of the DNA (–4 to +6). Topoisomerization by the human
enzyme could also be visualized in the open and closed
modeled conformations of the Leishmania protein, with the
hinge-bending movement being controlled by the electrostatic
charged surface of the protein.
The predicted active site pocket of the Leishmania protein
model is located in a structurally equivalent position to that of
the crystallized human topoisomerase I. The four catalytic
residues conserved in the human protein including Arg488,
Arg590, His632 and Lys532 cluster in the core subdomain III.
Mutagenesis studies have indicated that the above two arginine
and one histidine residues play an essential role in the nicking
closing reaction. The analogous residues in domain III of the
Leishmania protein model are Arg314, Arg410, His453 and
Lys352. However, the striking feature inferred from analysis
of the C-terminal domain of our model is that the reactive tyrosine found in the human topoisomerase I C-terminus was
missing in the structurally equivalent active site pocket of the
Leishmania protein. Instead, the C-terminal domain of our
model was found to harbor a serine residue (Ser553) within a
suitable distance of the 5′-OH of the scissile bond in the
modeled topoisomerase I–suicide DNA substrate complex.
The parasite protein ternary complex model with the CPT
inhibitor could account for the interaction with two conserved
residues as in human. Interestingly, the stacking interaction
with asparagine (Asn722 in human) was absent in the
Leishmania model. Ala552 and Ser553 of the parasite protein
were found to be occupying positions analogous with Asn722
and Tyr723, respectively, of human topoisomerase I. This
unique presence of a serine in our structural model at the
position occupied by tyrosine in the human enzyme points
towards serine acting as the reactive nucleophile for the
parasite enzyme. This observation is strongly supported by
identification of Ser-P/Thr-P as the protein–DNA linkage in the
801
Figure 10. Phylogenetic tree using the amino acid sequences of type I topoisomerases from L.donovani and other organisms. The phylogram is displayed
on TREEVIEW that reads PHYLIP format treefiles under the CLUSTAL W
program. The evolutionary scale bar is shown on the left, which indicates the
relative distance on the tree in arbitrary units.
covalent complex between LdTOP1 and DNA by biochemical
mapping studies.
Broadly, type I topoisomerases and site-specific recombinases
are believed to be derived from a common ancestral strand
transferase (35). Just as there are two families of site-specific
recombinases, one using a tyrosine nucleophile-based DNA
catalysis, e.g. tyrosine recombinases like λ integrase (36),
while the other performs a serine nucleophile-mediated strand
cleavage reaction, e.g. the resolvase-invertase group that
includes E.coli γδ resolvase (37); the kinetoplastid protozoan
parasite L.donovani possesses a topoisomerase I enzyme with
a reactive serine unlike the topoisomerase IB enzymes of other
eukaryotic members from yeast to human, that use a catalytic
tyrosine.
Since DNA topoisomerase I genes code for ubiquitous
proteins, their alignments and comparative analyses can form a
strong base for molecular phylogeny. A phylogenetic tree has
been constructed (Fig. 10) using the LdTOP1-like sequence
and representative topoisomerase I sequences from other
eukaryotes including yeast to human. The tree indicates that
the kinetoplastid protozoan parasite L.donovani diverged very
early from the main line of eukaryotic evolution. The absence of
the conserved motif surrounding the active site in the C-terminal
domain of LdTOP1 also points to this direction. A possible
hypothesis is that this conserved motif in topoisomerase I
proteins encompassing the active site tyrosine might be a later
acquisition in higher eukaryotes.
A future aspect of this work lies in exploiting the differences
between the active site domain of host and parasite enzymes by
integrating structural information with biochemical
802
Nucleic Acids Research, 2002, Vol. 30, No. 3
experimentation. Such information will provide crucial
insights for designing newer drugs with selective toxicity for
the parasitic enzyme. This drug screening system would,
thereby, form an exciting base for developing antiparasitic
chemotherapy.
ACKNOWLEDGEMENTS
We thank Professor S. Bhattacharyya, the Director of our
Institute, for his interest in this work. Purified rat liver
topoisomerase I was kindly provided by A. Roychowdhury.
S. Mandal is thanked for the assistance in TLC of samples.
This work is supported by a grant (BT/PRO493/MED/09/096/96)
from the Department of Biotechnology, Government of India.
REFERENCES
1. Champoux,J.J. (1990) Mechanistic aspects of type-I topoisomerases. In
Wang,J.C. and Cozzarelli,N.R. (eds), DNA Topology and Its Biological
Effects. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY,
pp. 217–242.
2. Wang,J.C. (1996) DNA topoisomerases. Annu. Rev. Biochem., 65, 635–692.
3. Lima,C.D., Wang,J.C. and Mondragon,A. (1994) Three-dimensional
structure of the 67 kD N-terminal fragment of E. coli DNA topoisomerase I.
Nature, 367, 138–146.
4. Yu,L., Zhu,C.X., Tse-Dinh,Y.C. and Fesik,S.W. (1995) Solution structure
of the C-terminal single-stranded DNA-binding domain of Escherichia
coli topoisomerase I. Biochemistry, 34, 7622–7628.
5. Lue,N., Sharma,A. and Mondragon,A. (1995) A 26 kDa yeast DNA
topoisomerase I fragment: crystallographic structure and mechanistic
implications. Structure, 3, 1315–1322.
6. Sharma,A., Hanai,R. and Mondragon,A. (1994) Crystal structure of the
N-terminal fragment of vaccinia virus DNA topoisomerase I. Structure, 2,
767–777.
7. Redinbo,M.R., Stewart,L., Kuhn,P., Champoux,J.J. and Hol,W.G.J.
(1998) Crystal structure of human topoisomerase I in covalent and
noncovalent complexes with DNA. Science, 279, 1504–1513.
8. Stewart,L., Redinbo,M.R., Qiu,X., Hol,W.G.J. and Champoux,J.J. (1998)
A model for the mechanism of human topoisomerase I. Science, 279,
1534–1541.
9. Redinbo,M.R., Stewart,L., Champoux,J.J. and Hol,W.G.J. (1999)
Structural flexibility in human topoisomerase I revealed in multiple
non-isomorphous crystal structures. J. Mol. Biol., 292, 685–696.
10. Cheng,C., Kussie,P., Paveltich,N. and Shuman,S. (1998) Conservation of
structure and mechanism between eukaryotic topoisomerase I and
site-specific recombinases. Cell, 92, 841–850.
11. Redinbo,M.R., Champoux,J.J. and Hol,W.G.J. (1999) Structural insights
into the function of type IB topoisomerases. Curr. Opin. Stuct. Biol., 9,
29–36.
12. Schneider,E., Hsiang,Y.H. and Liu,L.F. (1991) DNA topoisomerases as
anticancer drug targets. Adv. Pharmacol., 21, 149–183.
13. Wang,J.C. (1994) DNA topoisomerases as targets of therapeutics:
an overview. Adv. Pharmacol., 29A, 1–19.
14. Shapiro,T.A. (1993) Inhibition of topoisomerases in African
trypanosomes. Acta Trop., 54, 251–260.
15. Broccoli,S., Marquis J.-F., Papadopoulou,B., Olivier,M. and Drolet,M.
(1999) Characterization of a Leishmania donovani gene encoding a
protein that closely resembles a type IB topoisomerase. Nucleic Acids
Res., 27, 2745–2752.
16. Chakraborty,A.K., Gupta,A. and Majumder,H.K. (1993) A type 1 DNA
topoisomerase from the kinetoplast hemoflagellate Leishmania donovani.
Ind. J. Biochem. Biophys., 30, 257.
17. Ray,S., Majumder,H.K., Chakravarty,A. and Mukhopadhyay,S. (1996)
Amarogentin, a naturally occurring secoiridoid glycoside and a newly
recognised inhibitor of topoisomerase I from Leishmania donovani.
J. Nat. Prod., 59, 27–29.
18. Ray,S., Sadhukhan,P.K., Mandal,N.B., Mahato,S.B. and Majumder,H.K.
(1997) Dual inhibition of DNA topoisomerases of Leishmania donovani by
novel indolyl quinolines. Biochem. Biophys. Res. Commun., 230, 171–175.
19. Ray,S. Hazra,B., Mittra,B.M., Das,A. and Majumder,H.K. (1998)
Diospyrin, a bisnapthoquinone; a novel inhibitor of type I DNA
topoisomerase of Leishmania donovani. Mol. Pharmacol., 54, 994–999.
20. Champoux,J.J. and McConaughy,B.L. (1976) Purification and
characterization of the DNA untwisting enzyme from rat liver.
Biochemistry, 15, 4638–4642.
21. Tse,Y.-C., Kirkegaard,K. and Wang,J.C. (1980) Covalent bonds between
protein and DNA. J. Biol. Chem., 255, 5560–5565.
22. Mandal,C., Kingery,B.D., Anchin,J.M., Subramanium,S. and
Linthicum,D.S. (1996) ABGEN: a knowledge based automated approach
for antibody structure modeling. Nat. Biotechnol., 14, 323–328.
23. Mandal,C. (1998) MODELYN—A Molecular Modelling Program.
Version PC-1.0, Indian Copyright No. 9/98.
24. Koradi,R., Billeter,M. and Wuthrich,K. (1996) MOLMOL: a program for
display and analysis of macromolecular structures. J. Mol. Graph., 14,
51–55, 29–32.
25. Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-specific gap penalties and weight
matrix choice. Nucleic Acids Res., 22, 4673–4680.
26. Page,R.D.M. (1996) TREEVIEW: an application to display phylogenetic
trees on personal computers. Comput. Appl. Biosci., 12, 357–358.
27. Hsiang,Y.-H., Hertzberg,R., Hecht,S. and Liu,L.F. (1985) Camptothecin
induced protein-linked DNA breaks via mammalian DNA topoisomerase I.
J. Biol. Chem., 260, 14873–14878.
28. Eng,W.K., Pandit,S.D. and Sternglanz,R. (1989) Mapping of the active
site tyrosine of eukaryotic DNA topoisomerase I. J. Biol. Chem., 264,
13373–13376.
29. D’Arpa,P., Machlin,P.S., Ratrie,H.,III, Rothfield,N.F., Cleveland,D.W.
and Earnshaw,W.C. (1988) cDNA cloning of human DNA topoisomerase I:
catalytic activity of a 67.7 kDa C-terminal fragment. Proc. Natl Acad. Sci.
USA, 85, 2543.
30. Stewart,L., Ireton,G.C., Parker,L.H., Madden,K.R. and Champoux,J.J.
(1996) Topoisomerase I has a strong binding preference for a conserved
hexadecameric sequence in the promoter region of the rRNA gene from
Tetrahymena pyriformis. J. Biol. Chem., 271, 7593.
31. Andersen,A.H., Gocke,E., Bonven,B.J., Neilsen,O.F. and Westergaard,O.
(1985) Topoisomerase I has a strong binding preference for a conserved
hexadecameric sequence in the promoter region of the rRNA gene from
Tetrahymena pyriformis. Nucleic Acids Res., 13, 1543–1557.
32. Champoux,J.J. (2001) DNA topoisomerases: structure, function and
mechanism Annu. Rev. Biochem., 70, 369–413.
33. Fan,Y., Weinstein,J.N., Kohn,K.W., Shi,L.M. and Pommier Y. (1998)
Molecular modelling studies of the DNA topoisomerase I ternary
cleavable complex with camptothecin. J. Med. Chem., 41, 2216–2226.
34. Kerrigan,J.E. and Pilch,D.S. (2001) A structural model for the ternary
cleavable complex formed between human topoisomerase I, DNA and
camptothecin. Biochemistry, 40, 9792–9798.
35. Cheng,C., Kussie,P., Pavletic,N. and Shuman,S. (1998) Conservation of
structure and mechanism between eukaryotic topoisomerase I and
site-specific recombinases. Cell, 92, 841–850.
36. Kwon,H.J., Tirumalai,R., Landy,A. and Ellenberger,T. (1997) Flexibility
in DNA recombination: structure of the λ integrase catalytic core. Science,
276, 126–131.
37. Yang,W. and Steitz,T.A. (1995) Crystal structure of the site-specific
recombinase γδ resolvase complexed with a 34 bp cleavage site. Cell, 82,
193–207.