8
4
C.A. Rodrígues-Poveda et al. / Molecular & Biochemical Parasitology 182 (2012) 83–87
of metalloproteases involved in catalysis are conserved in the T.
used in the characterization of E. coli, human and Plasmodium
peptide-deformylase activity. As shown in Fig. 1A, while the activity
increased with substrate concentration up to 35 mM, it drasti-
cally dropped thereafter, showing that TcPDF-1ꢀN37 is inhibited
by an excess of substrate. A similar substrate inhibition has been
described for bacterial [14] and human PDFs [12]. Consequently,
20 mM f-MAS was used in the enzymatic assay henceforth. Km and
Vmax values were obtained by hyperbolic regression fit of the data
to the Hill equation using Sigma Plot® software. The Km and Vmax
1
14
121
cruzi protein: motif 1
GVGLAAPQ
in Arabidopsis thaliana and
44
51
93
176
180
GIGLAATQ in Escherichia coli, motif 2
EGCLS
in A. thaliana
89
and EGCLS in E. coli and motif 3, HEXXH which is strictly con-
114
served in all PDFs studied to the date. The Gly of motif 1 (Gly
in
in A. thaliana,
Leu in E. coli) are crucial for catalysis and are substituted in H. sapi-
4
4
179
A. thaliana, Gly in E. coli) and Leu of motif 2 (Leu
92
1
08
173
108
115
ens for Cys
(
and Glu
in motif 1 ( CVGLSAPQ ) and motif 2
1
70
174
EGCES ) respectively. T. cruzi PDF-1 displays a similar substi-
8
3
83
90
−1
−1
,
tution in motif 1, Cys
(
CISFSAPK ) but maintains a hydrophobic
for f-MAS were 22.37 ± 0.137 mM and 200 ± 2.3 nmol min mg
Ile in motif 2 ( ENCIS138).
1
34
respectively. The calculated kcat value was 0.093 ± 0.065 s . While
we cannot fully discard that the His-tag may be detrimental to
enzyme activity, these values are of similar order of magnitude
to those described for other truncated recombinant eukaryotic
enzymes when f-MAS was used as substrate although the Vmax was
significantly lower in comparison to that of bacterial PDFs [4,5,12].
It is evident that the peptide f-MAS used in this study is not
the optimal substrate for TcPDF-1. It was designed to characterize
bacterial PDFs and is based on the fact that the complete bacterial
proteome is associated to formylation. In contrast, formylation in
trypanosomatids is limited to only 13 mitochondrial DNA-encoded
proteins [9]. Recent studies on human PDF have shown that the
enzyme is more efficient in deformylating substrates based on
the organelle-encoded proteome than other nonspecific substrates
including f-MAS [15]. We therefore designed a tripeptide based
on the N-terminal sequence of subunit 7 of the NADH dehydro-
genase complex (formyl-methionyl-leucyl-phenylalanine, f-MLP)
coded by mitochondrial DNA but its high hydrophobicity made it
inadequate for assay in aqueous media.
−1
All eukaryotic PDF orthologues have an extended N-terminal
domain which makes expression in bacterial systems difficult. This
extension plays a role in organellar targeting of the protein to
plastids and mitochondria in A. thaliana [1], to the apicoplast in api-
complexan parasites [4] and to the mitochondria in mammals [3].
In this study, we explored the properties of a truncated version of
PDF-1 from T. cruzi lacking 37 amino acids from the amino terminus
using the formylated tripeptide formyl-methionyl-alanyl-serine (f-
MAS).
Initial attempts to produce a recombinant full-length soluble
T. cruzi enzyme were unsuccessful. Considering previous reports
on PDF from H. sapiens [3], and Plasmodium falciparum [4], we
generated two deletion mutants that lacked the N-terminal 33
and 37 amino acids. The full-length PDF-1 gene was amplified
ꢀ
by PCR from T. cruzi CL Brener genomic DNA with primers 5 -
ꢀ
ꢀ
GCC ATA TGC TGA GCC GTC TGT CAC GGA-3 and 5 -GGA TCC CAT
TGG TTC AGA CGA GCA TCC-3 , which were specifically designed
ꢀ
according to the information obtained from the genome sequence
database (GenBank accession number TcCLB.506871.100). Addi-
The influence of the ionic strength on deformylase activity was
assessed by supplementation of the reaction buffer with increasing
concentrations of KCl. An increase in ionic strength had a significant
impact on TcPDF-1ꢀN37, giving rise to a stimulation of the activity
which reached a maximum at 0.5 M KCl (Fig. 1B).
ꢀ
ꢀ
tional primers, 5 -GCC ATA TGG CGG AGG CGC AGG TGA AGT 3 and
5
ꢀ
ꢀ
-GCC ATA TGG TGA AGT CCC GAG TGG CCT-3 , were designed to
generate the truncated proteins where 33 (ꢀN33) and 37 (ꢀN37)
amino acids were eliminated from the N-terminus respectively. The
primers contain NdeI and BamHI restriction sites (underlined) for
cloning into the pET-22b vector. The cloning procedure resulted
in the addition of a six-histidine tag to the C-terminus of PDF. Of
the two constructs tested, major amounts of soluble enzyme were
obtained with TcPDF-1ꢀN37, a PDF lacking 37 amino acids from
the amino terminus.
Next, we studied the effect of the nature and concentration of
the metallic cofactor on TcPDF-1 deformylation capacity. To this
end, several oxidation insensitive divalent cations were added to
the reaction buffer at different concentrations ranging from 10 to
100 M. The enzyme was preincubated with the metallic cofactor
2
+
2+
2+
for 10 min before initiating the reaction. While Co , Mg , Mn
The purification and biochemical characterization of bacterial
PDFs have largely remained a major challenge because the enzyme
did not affect significantly the activity, Ni2+ had a pronounced
inhibitory effect on peptide deformylation at all concentrations
tested. On the other hand, the addition of Cu induced a 50%
2+
2+
is extraordinarily labile due to oxidation of its catalytic Fe . It has
been shown that substitution of the ferrous ion for Co2 or Ni2+ dur-
ing the purification procedure of bacterial [10,11], P. falciparum [4]
and human PDFs [12] gives highly active and stable enzyme prepa-
rations. Here TcPDF-1ꢀN37 protein was overexpressed in E. coli
cells that were grown in medium enriched with Co2 (100 M).
Under these conditions, TcPDF-1ꢀN37 was purified to over 99%
purity by metal-affinity chromatography and exhibits an estimated
molecular mass of ∼28–30 kDa (Fig. S2).
+
increase in the measured activity at 50 M (Fig. 1C). To our knowl-
edge, our study is the first to report the activation of a PDF enzyme
by this metal ion. Both P. falciparum and E. coli PDFs differ from
TcPDF-1 and are highly active with both Co2+ and Ni , retaining
the same catalytic activity as the iron-native enzyme with the lat-
2+
+
2+
ter [4,10,11]. In contrast, human PDF is only active with Co [12]
while A. thaliana PDF is highly active when associated to Zn2+
.
The effect of the chelating agents EDTA and 1,10-phenantroline
on deformylase activity was studied at concentrations ranging from
1 to 20 mM, and 1 to 5 mM respectively. E. coli PDF was included as
a control assay. For this purpose, a segment of the E. coli def gene
Truncation of the first 37 N-terminal amino acids gave a pro-
tein that proved to be catalytically active and exhibited kinetic
parameters similar to other eukaryotic PDFs. A similar strategy was
employed by Schneiders and coworkers who obtained an active
version of T. brucei PDF-1 by eliminating the first 30 amino acids,
thus suggesting the presence of a structural core responsible for
deformylase activity [5].
ꢀ
ꢀ
was amplified by PCR using forward primer 5 -CGC ATA TGT CAG
ꢀ
TTT TGC AAG TGT TAC-3 and reverse primer 5 -GGA TCC TTC AGC
ꢀ
GGT GAC AGA TAA TCC AT-3 and cloned in pET22b. The result-
ing expression construct codes for a bacterial PDF (EcPDFꢀC18)
that lacks 18 C-terminal residues and contains a C-terminal His-
tag. The conditions for expression and purification were similar
to those used for TcPDF-1ꢀN37 (see supplementary information).
Purified EcPDFꢀC18 was active and more stable than the full-
length enzyme [16]. Before measuring the activity, the enzyme
was preincubated with the chelating agent for 10 min. Strikingly,
both EDTA and 1,10-phenantroline strongly activated deformy-
lase activity while 1,10-phenantroline was a potent inhibitor of
The PDF activity assay was based on the quantification of for-
mate using formate dehydrogenase [13]. The reaction mixture
+
contained 50 mM KH PO pH 7.5, 12 mM NAD , 1.2 U/ml FDH, and
2
4,
TcPDF-1ꢀN37 enzyme as indicated. The reaction was performed at
◦
3
7 C for 300 s and the resulting increase in absorbance of NADH at
45 nm was measured with a UV-VIS spectrophotometer.
The catalytic activity of TcPDF-1ꢀN37 was assessed with f-
3
MAS as substrate. f-MAS is a bacterial PDF specific substrate