YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
Protein Expression and Purication xxx (2014) xxxxxx
1

Contents lists available at ScienceDirect

Protein Expression and Purication

journal homepage: www.elsevier.com/locate/yprep
5
6

High-level expression in Escherichia coli, purication and kinetic

characterization of Plasmodium falciparum M1-aminopeptidase

3
4
7
8
9
10
11
12
13
14
1 9
6
2
17
18
19
20
21
22
23
24
25
26
27
28

Q1

Jorge Gonzlez-Bacerio a,, Joel Osuna b, Amaia Ponce a, Rafael Fando c, Katherine Figarella d,
Yanira Mndez a, Jean-Louis Charli b, Mara de los . Chvez a
a

Centro de Estudio de Protenas, 25 # 455 entre J e I, Facultad de Biologa, Universidad de La Habana, Cuba
Instituto de Biotecnologa, Universidad Nacional Autnoma de Mxico (UNAM), Ave. Universidad 2001, Cuernavaca, Mor., Mexico
Centro Nacional de Investigaciones Cientcas, Ave. 25 y 158, 12100 La Habana, Cuba
d
Fundacin Instituto de Estudios Avanzados, Carretera Nacional de Hoyo de la Puerta, Valle de Sartenejas, Baruta 1080, Estado Miranda, Venezuela
b
c

a r t i c l e

i n f o

Article history:
Received 8 July 2014
and in revised form 3 August 2014
Available online xxxx
Keywords:
Expression in Escherichia coli
IMAC
M1-family aminopeptidase
Malaria
PfA-M1
Plasmodium falciparum

a b s t r a c t
Plasmodium falciparum neutral metallo-aminopeptidase (PfAM1), a member of the M1 family of metallo
proteases, is a promising target for malaria, a devastating human parasitic disease. We report the highlevel expression of PfAM1 in Escherichia coli BL21. An optimized gene, with a codon adaptation index and
an average G/C content higher than the native gene, was synthesized and cloned in the pTrcHis2B vector.
Optimal expression was achieved by induction with 1 mM IPTG at 37 C for 18 h. This allowed obtaining
100 mg of recombinant PfAM1 (rPfAM1) per L of culture medium; 19% of the E. coli soluble protein mass
was from rPFAM1. rPfAM1, fused to an amino-terminal 6His tag, was puried in a single step by immobilized metal ion afnity chromatography. The protein showed only limited signs of proteolytic degradation, and this step increased purity 27-fold. The kinetic characteristics of rPfAM1, such as a neutral
optimal pH, a preference for substrates with basic or hydrophobic amino acids at the P1 position, an inhibition prole typical of metallo-aminopeptidases, and inhibition from Zn2+ excess, were similar to those
of the native PfAM1. We have thus optimized an expression system that should be useful for identifying
new PfAM1 inhibitors.
2014 Published by Elsevier Inc.

30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

46

Corresponding author. Fax: +53 7 8321321.

E-mail addresses: jogoba@fbio.uh.cu (J. Gonzlez-Bacerio), joel@ibt.unam.mx
(J. Osuna), aponce@estudiantes.fbio.uh.cu (A. Ponce), rafael.fando@cnic.edu.cu
(R. Fando), kgarella@idea.gob.ve (K. Figarella), yanira@quimica.uh.cu (Y. Mndez),
charli@ibt.unam.mx (J.-L. Charli), mchavez@infomed.sld.cu (Mara de los .
Chvez).

Introduction

47

Aminopeptidases (APs)1 are proteolytic enzymes that hydrolyze

peptide bonds at the amino terminus of polypeptide chains. MetalloAPs use one or two active site metal ions for catalysis [1]. A neutral

48

1
Abbreviations used: AP(s), aminopeptidase(s); PfAM1, M1-aminopeptidase from P.
falciparum; p96, PfAM1 96-kDa form; p68, PfAM1 68-kDa form; ePepN, M1aminopeptidase from E. coli; rPfAM1, recombinant PfAM1; IMAC, immobilized metal
ion afnity chromatography; DNA, deoxyribonucleic acid; G, guanine; C, cytosine;
PCR, polymerase chain reaction; LB, Luria-Bertani; OD600(280)nm, optical density at 600
or 280 nm; IPTG, isopropyl-b-D-1-thiogalactopyranoside; SDSPAGE, sodium dodecylsulfate polyacrylamide gel electrophoresis; EA, enzymatic activity; Leu, leucine; pNA,
p-nitroanilide; C1, culture of E. coli BL21 non-transformed and induced; C2, culture of
E. coli BL21 transformed with the pTrcHis2B vector and induced; C3, culture of E. coli
BL21 transformed with the pTrcHis2B-rPfAM1 construct and non-induced; BSA,
bovine serum albumin; LCMS/MS, liquid chromatography coupled to tandem mass
spectrometry; UNAM, Autonomous National University of Mexico; His, histidine;
DMSO, dimethyl sulfoxide; U, unit of EA; MES, 2-(N-morpholino)ethanesulfonic acid;
Arg, arginine; Lys, lysine; Ala, alanine; Val, valine; Ile, isoleucine; Gly, glycine; Pro,
proline; bp, base pairs; CAI, Codon Adaptation Index; spEA, specic EA; AMC, 7amido-4-methylcoumarin; Glu, glutamic acid; bNA, b-naphtylamide; dNTPs, deoxynucleoside-triphosphates; PMSF, phenyl-methylsulfonyl-uoride.

http://dx.doi.org/10.1016/j.pep.2014.08.002
1046-5928/ 2014 Published by Elsevier Inc.

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

49
50

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
2

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

103

metallo-AP, belonging to the M1 family of metallo-proteases [2], is a

molecular target for malaria [3], a human parasitic disease caused by
several species of protozoan unicellular parasites belonging to the
genus Plasmodium, mainly Plasmodium falciparum [4]. This enzyme,
named PfAM1, has biological functions that are crucial for
P. falciparum intra-erythrocytic stages inside the human host. PfAM1
is involved in the degradation of erythrocytic hemoglobin [1,511],
trophozoite maturation [7,12,13], and reinvasion [7,9,11,1315]
and appears to be essential for the parasite during the erythrocytic
stage [1,8,10,1620]. Furthermore, inhibitors of PfAM1 block the
in vitro growth of the parasite at micromolar concentrations
[7,10,16,1822], and one of them reduces infection in a murine
malaria model [19]. The identication of potent and selective PfAM1
inhibitors is therefore a major target for malaria treatment.
PfAM1 is a monomeric enzyme of 1085 amino acid residues and
126 kDa (primary protein product), with an amino-terminal extension of 194 residues that is not conserved among the members of
the M1 family [20]. In P. falciparum erythrocytic stages, two active
and soluble 96- and 68-kDa PfAM1 processed forms (p96 and p68)
have been detected, both lacking the amino-terminal extension
[2,9,11]. The enzyme has 4 domains [20], a general fold very similar to that of bacterial neutral metallo-APs [2325], and 35%
identity to the full-length sequence of Escherichia coli M1-AP
(ePepN) [20]. PfAM1 localizes in the food vacuole and nucleus of
P. falciparum [10,26]. This AP is Zn2+-dependent, has an optimal
pH of 7.07.4 [9,20], and has a substrate specicity typical of
M1-APs: preference for basic and hydrophobic (including aromatic
and branched) residues at the P1 position (i.e., the amino acid
amino-terminal to the hydrolyzed peptide bond) [9].
Because of the difculty with purifying quantities of native
enzyme sufcient for kinetic studies and the search for inhibitors
[26], recombinant PfAM1 (rPfAM1) expression is a priority. McGowan et al. [20] were the rst to report recombinant expression of
PfAM1. In E. coli BL21, they expressed a synthetic gene that was initially optimized for the Pichia pastoris system. This gene encodes a
truncated PfAM1 form of approximately 100 kDa that lacks the
rst 194 residues. In this construct, the rst residue of rPfAM1 correlates with the start of the ePepN orthologue, and a six-histidine
tag is inserted at the carboxyl terminus for purication purposes.
The protein is active and soluble, with molecular and kinetic characteristics similar to those of the native enzyme [20]. On the other
hand, Azimzadeh et al. [11] obtained recombinant p68 (residues
191802) in E. coli BL21(DE3). This protein, expressed in an insoluble form, was puried by immobilized metal ion afnity chromatography (IMAC) in 6 M urea. A third recombinant variant of PfAM1
lacking the rst 191 residues and containing an histidine tag at its
amino terminus was amplied from parasite genomic DNA and
was expressed in a soluble and active form in the E. coli BL21(DE3)
Rosetta 2 optimized strain [26]. However, comprehensive data on
heterologous rPfAM1 expression are lacking. This work describes,
in detail, a strategy that allowed high-level expression of rPfAM1.
The puried rPfAM1 had kinetic characteristics similar to those
of the native enzyme.

104

Materials and methods

105

Codon optimization, synthesis of the rpfam1 gene and cloning in the

pTrcHis2B vector

51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102

106
107
108
109
110
111
112
113

The services of the company GeneArt AG (Germany) were contracted. The rpfam1 gene (protein sequence available in UniProtKB/
TrEMBL: Q8IEK1) was optimized for its expression in bacteria,
using the software GeneOptimizer. The optimization consisted
of changing the codons of the native gene to codons with the
highest usage in bacteria, increasing the G/C content and avoiding
the formation of sequences that can affect transcription and

translation. Synthesis was performed by PCR assembly of synthetic

oligonucleotides and PCR products. The sequences of the native
and synthetic rPfAM1 genes were analyzed and compared using
the web site GeneScript (http://www.genscript.com/cgi-bin/tools/
rare_codon_analysis). Cloning was performed in the bacterial
expression vector pTrcHis2B (Invitrogen, USA) under the control
of the strong, inducible and hybrid trc promoter [27], between
Pst I and Kpn I restriction sites. This vector contains the lacIq gene
to regulate the expression of the heterologous gene. The resultant
construct (pTrcHis2B-rPfAM1) was conrmed by DNA sequencing.
E. coli DH5a (supE44 DlacU169 (/80 lacZ DM15) hsdR17 recA1
endA1 gyrA96 thi-1 relA-1) competent cells were transformed with
the pTrcHis2B-rPfAM1 construct. Four colonies were arbitrarily
selected, and plasmid DNA was amplied and puried from liquid
cultures using a mini-prep kit (Roche, Switzerland). The quality
of the amplied plasmids was conrmed by agarose gel
electrophoresis.

114

Expression of rPfAM1 in E. coli

131

rPfAM1 was expressed in E. coli BL21 (hsdS gal (kcIts857 ind1

Sam7 nin5 lacUV5-T7 gene 1)). To check pTrcHis2B-rPfAM1 functionality, a small-scale proof of concept experiment was performed. For this, 1-mL aliquots of LuriaBertani (LB) broth (10 g/
L tryptone, 5 g/L yeast extract and 10 g/L NaCl) supplemented with
200 lg/mL ampicillin were inoculated with colonies of transformed cells and incubated overnight at 37 C. Then, 50-lL aliquots
were inoculated in 5-mL aliquots of ampicillin-supplemented LB
broth. The 5-mL cultures were incubated at 37 C until an OD600
nm of 0.40.8 was reached, at which point expression was induced
with 1 mM isopropyl-b-D-1-thiogalactopyranoside (IPTG, Sigma,
USA) over 4 h. The following negative controls were used: (i) the
non-transformed strain, induced in LB broth; (ii) the strain transformed with the pTrcHis2B vector, induced in ampicillinsupplemented LB broth; and (iii) the strain transformed with the
construct pTrcHis2B-rPfAM1, in ampicillin-supplemented LB
broth. Expression of rPfAM1 was evaluated by SDSPAGE and AP
enzymatic activity (EA), with Leu-p-nitroanilide (Leu-pNA) as a
substrate. To keep the positive clones, aliquots from 1-mL cultures
were mixed with glycerol and stored at 70 C.
To select the induction method, 5 mL of medium was inoculated
with an rPfAM1-producer bacterial clone stored in glycerol and
incubated overnight at 37 C. Thereafter, 100 mL of medium were
inoculated with 1 mL of clone-containing medium, and induction
was performed for 18 h at 37 C. Cultures used as negative controls
included (i) the non-transformed strain, induced (C1); (ii) the
strain transformed with the pTrcHis2B vector, induced (C2); and
(iii) the strain transformed with pTrcHis2B-rPfAM1, non-induced
(C3). Induction with IPTG was performed in ampicillinsupplemented LB broth with 0.1 mM IPTG at the start of the culture
or with 1 mM IPTG during the exponential phase (OD600 nm = 0.4
0.8). Negative control C1 was prepared with LB broth, whereas C2
and C3 controls were prepared with ampicillin-supplemented LB
broth. For auto-induction with 0.2% lactose, MDG non-inducing
medium (25 mM Na2HPO4, 25 mM KH2PO4, 50 mM NH4Cl, 5 mM
Na2SO4, 2 mM MgSO4, 100 lM FeCl3, 0.5% glucose, 0.25% aspartate;
[28]) supplemented with 200 lg/mL ampicillin was used for preparation of the 5-mL culture. The ZYMB-5052 auto-induction medium (1% N-Z-amine, 0.5% yeast extract, 1% NaCl, 25 mM Na2HPO4,
25 mM KH2PO4, 50 mM NH4Cl, 5 mM Na2SO4, 2 mM MgSO4,
10 lM FeCl3, 1 lM ZnSO4, 0.5% glycerol, 0.05% glucose, 0.2% lactose; [28]) supplemented with 200 lg/mL ampicillin was used
for the 100-mL culture. Negative control C1 was prepared with
ZYMB-5052 broth. Control C2 was prepared with ampicillinsupplemented ZYMB-5052 medium, and control C3 was prepared
with ampicillin-supplemented ZYMB-5260 broth (ZYMB-5052

132

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130

133
134
135
136
137
138
139
140
141

142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

179

without lactose and with 0.26% glucose). The nal OD600 nm was
measured, and expression was evaluated as previously mentioned.

180

rPfAM1 purication

181

210

Bacterial biomass was collected by centrifugation at 8000g for

10 min at 4 C. Cells were resuspended in cold 50 mM sodium
phosphate buffer, pH 8.0, 300 mM NaCl, to obtain cellular suspensions at a concentration of 10 units of OD600 nm. These suspensions
were subjected to 2 ultrasonic pulses (Branson Sonicator, USA) of
15 s at 40% output, with a 30-s pause on ice between pulses. The
extracts were centrifuged at 12,000g for 20 min at 4 C, and the
pellets and supernatants were separated and stored at 20 C.
Purication of rPfAM1 was performed by IMAC using a gravityow column with 1 mL of HisPur Cobalt Resin (Thermo Scientic/Pierce Biotechnology, USA). Protein extracts (10 mL) were
equilibrated at the binding conditions with a dilution of 1:2 (v/v)
with cold 50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl,
and 20 mM imidazole (ReagentPlus, 99%, SigmaAldrich, USA).
The matrix was equilibrated with 5-resin-bed volumes of binding
buffer (cold 50 mM sodium phosphate buffer, pH 8.0, 300 mM
NaCl, and 10 mM imidazole). The prepared protein extract was
added to the resin. The ow-through was collected, and the resin
was washed with the binding buffer until the OD280 nm of the
ow-through fraction approached baseline. The His-tagged protein
was eluted with 5-resin-bed volumes of elution buffer (cold 50 mM
sodium phosphate buffer, pH 8.0, 300 mM NaCl, and 50 mM imidazole). During the initial study, various elution buffer imidazole
concentrations were assayed: 20, 30, 40, 50, 100 and 250 mM.
Fractions of 1-resin-bed volume were collected (1 mL). The chromatographic runs were monitored by measuring the OD280 nm of
the fractions and were evaluated by SDSPAGE and determination
of AP EA. SDSPAGE was used as a criterion of purity. Eluates from
several runs were pooled and dialtered with a 10-kDa cut-off
membrane (Sigma, USA). Puried rPfAM1 was kept at 20 C.

211
212

solubilized in DMSO. The absorbance at 405 nm, due to the liberation of pNA, was monitored at 15-s intervals for 3 min. Kinetic
assays were conducted with volumes of protein extracts or concentrations of puried rPfAM1 that were linearly related to the initial
rates, at 37 C, in 96-well plates (200 lL nal volume) using a
microplate spectrophotometer (Multiscan FC, Thermo Scientic,
USA). The assays conditions were 50 mM TrisHCl buffer, pH
7.27.4. The nal DMSO concentration was not higher than 2%.
The used molar extinction coefcient at 405 nm for pNA was
9.87 mL lmol1 cm1 [31]. The EA unit (U) is the amount of
enzyme that hydrolyzes 1 lmol of substrate per minute under
the assays conditions. The assays were performed in triplicate,
and the results are presented as the mean standard deviation.
The inhibition of puried rPfAM1 by 1 mM 1,10-phenanthroline
(Sigma, USA), solubilized in DMSO, was tested pre-incubating the
enzyme-inhibitor mixture for 5 min at 25 C before the addition
of substrate. The control was prepared by pre-incubating the
enzyme with the same volume of DMSO. Residual activity was
dened as the ratio between the EA in presence of the inhibitor
and the activity of the control. The effect of the Zn2+ concentration
on the EA of rPfAM1 was evaluated with 0, 10 or 100 mM ZnCl2
(Sigma, USA) and 1.5 mM Leu-pNA substrate. To test the effect of
pH, the reaction mixtures were adjusted to different pH values with
100 mM MES (pH 5.2, 5.5 and 6.5) or 50 mM TrisHCl (pH 7.2, 7.5,
8.0, 8.5 and 9.0). To test substrate specicity, aminoacyl-pNA
chromogenic substrates (Arg-, Lys-, Leu-, Ala-, Val-, Ile-, Gly- and
Pro-pNA) (Bachem, Sweden) were tested at different concentrations
(range: 15.6500 lM) at pH 7.27.4 (50 mM TrisHCl buffer), and
the kinetic parameters KM and kcat were calculated from LineweaverBurk plots. The initial enzyme concentration was calculated
assuming a sequence-based predicted molecular weight for rPfAM1
of 106.13 kDa, determined using the Sequence Manipulation Suite v
2.0 software, (http://www.bioinformatics.org/sms2/).

239

Protein analysis

Results and discussion

272

Construction of the pTrcHis2B-rPfAM1 plasmid optimized for rPfAM1

expression in E. coli

273

236

AP EA assays

237

AP EA was determined by a continuous kinetic method, using

300 lM Leu-pNA chromogenic substrate (Bachem, Sweden)

The experimental design for rPfAM1 expression was based on

the work of McGowan et al. [20], with an important difference:
we designed and synthesized a gene optimized for the expression
in E. coli of a truncated amino-terminal form of the protein (residues 1951,085 of the native protein). This gene contains
2708 bp, 2 stop codons, is anked by Pst I and Kpn I restriction sites,
and encodes a protein with a predicted molecular mass of
106.13 kDa with a 6His tag at its carboxyl terminal end (Fig. 1).
In general, high-level expression correlates positively with the
codon adaptation index (CAI) of the gene in the selected expression
system. A CAI = 1.0 represents an ideal situation for gene expression in heterologous systems, when all codons match with those
of the highest usage frequency in the host. A CAI > 0.8 is sufcient
to achieve the successful expression of a gene [32]. For the
designed rPfAM1 gene, CAI = 0.96, a value notably higher than
0.65 for the native gene (Fig. 2A). Non-optimal codons were distributed relatively homogeneously along the sequence (Fig. 2B).
The optimal average G/C content for efcient gene expression
ranges from 3070%. Furthermore, low G/C content stretches can
negatively affect the efciency of gene transcription and translation (http://www.genscript.com/cgi-bin/tools/rare_codon_analysis). The average G/C content for the synthetic gene was 39%
(vs. 26% for the native gene). Regions with low G/C content were
homogeneously distributed along the optimized sequence
(Fig. 2C). The construction of plasmid pTrcHis2B-rPfAM1
(7098 bp) is shown in Fig. 3.

275

235

Protein samples were applied to 8% polyacrilamide gels in the

presence of SDS [29] under reducing conditions. Runs were performed at 30 mA, and staining was performed with Coomassie Blue
R-250. To evaluate expression, we loaded a constant amount of
supernatant (20 lg, measured by the Bradford method [30]) or pellet protein. In contrast, to evaluate the IMAC results, protein quantities applied to acrylamide gels were normalized by volume.
Commercial molecular weight markers were used (Dual Color,
BioRad, USA; or Bench Mark, Invitrogen, USA). Gel images were
captured with an image analyzer (Molecular Imager, Chemi Doc
XRS, BioRad, USA) using Quantity One 4.6.7 software (BioRad, USA).
Densitometric analysis of gels was performed using ImageJ 1.38d
software (USA).
The protein concentration was determined by the Bradford
method [30], with bovine serum albumin (BSA) as standard protein. Assays were performed in triplicate, and the results are presented as the mean standard deviation. For the identication of
rPfAM1 by the sequencing of tryptic peptides using liquid chromatography coupled to tandem mass spectrometry (LCMS/MS) from
gel-derived protein bands, the services of the Proteomic Unit from
Instituto de Biotecnologa, Universidad Nacional Autnoma de
Mxico (UNAM), were contracted. Immunodetection of rPfAM1
by Western blot was performed with an anti-His antibody (Thermo
Scientic/Pierce Biotechnology, USA).

178

182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209

213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234

238

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271

274

276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
4

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

Fig. 1. Schematic representation of the rpfam1 synthetic gene.

Fig. 2. Optimization of the rpfam1 gene for expression in E. coli. Data for the native pfam1 gene (without the region encoding the amino-terminal extension) and the synthetic
rpfam1 gene. Distribution of codons according to quality ranges (A). A value of 100 is set for the codon with the highest usage frequency for a given amino acid in E. coli [32].
Distribution of the quality of codons (B) and of the G/C content (C) along the sequence of the native and synthetic genes. Plots in (C) show the G/C content in a 40-bp window
centered at the indicated nucleotide position. For sequence analysis and graphic visualization, we used the web site GeneScript (http://www.genscript.com/cgi-bin/tools/
rare_codon_analysis).
301

Small-scale expression of rPfAM1

302

Four tested clones of E. coli BL21 cells, transformed with

pTrcHis2B-rPfAM1, were positive for recombinant protein expression. The results obtained with one of these are shown in Fig. 4. An
increase in the intensity of a protein band, with the size expected
for rPfAM1 (100 kDa), was observed in the lane corresponding to
the rupture supernatant of the culture of E. coli transformed with
the genetic construct pTrcHis2B-rPfAM1 and induced with IPTG.
This enrichment was in comparison with the rupture supernatants
corresponding to the culture of the non-transformed strain, to the
culture of the strain transformed with the pTrcHis2B empty vector,
and to the culture of the strain transformed with the pTrcHis2BrPfAM1 construct and not induced. The same result was obtained
with insoluble fractions (Fig. 4A). As is shown in Fig. 4B, a 10-fold

303
304
305
306
307
308
309
310
311
312
313
314

increase in specic EA (spEA) was observed in the extract obtained

from the E. coli BL21/pTrcHis2B-rPfAM1 culture (induced)
compared with basal spEA (most likely due to E. coli endogenous
M1-AP: ePepN). This EA was inhibited by 80 lM bestatin, a generic
inhibitor of neutral metallo-APs.
The denitive conrmation of recombinant enzyme expression
was performed by tryptic digestion of the 100-kDa enriched protein band and sequencing of resultant peptides by LCMS/MS. All
of the 20 sequences (51% of sequence coverage) matched the
native PfAM1 sequence deposited in the UniProt database
(Q8IEK1) (Fig. 4C). The identied peptide closest to the amino terminus began at residue 229 of native PfAM1. This nding was in
agreement with the design of the rPfAM1 construction, which lacks
the rst 194 amino acids of the native enzyme. Despite the high
sequence similarity between PfAM1 and ePepN [2], none of the

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

315
316
317
318
319
320
321
322
323
324
325
326
327
328
329

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

Fig. 3. Construction of plasmid pTrcHis2B-rPfAM1. (A) Cloning of the rpfam1 gene in the pTrcHis2B vector. The lacIq gene and the pBR322 replication origin were omitted in
the pTrcHis2B-rPfAM1 scheme for simplicity. (B) Four clones of the genetic construct pTrcHis2B-rPfAM1 were checked by electrophoresis on 1% agarose gel.

Fig. 4. Preliminary expression of rPfAM1 in E. coli BL21. Expression at a small scale (5 mL), in LB broth supplemented with ampicillin at 37 C; induction for 4 h with 1 mM
IPTG added at the exponential phase of growth. Negative controls of expression: non-transformed strain and induced, strain transformed with pTrcHis2B vector and induced,
and strain transformed with pTrcHis2B-rPfAM1 construct and not induced. Evaluation of expression by SDSPAGE (8% acrylamide, non-reducing conditions, Coomassie
staining) (A) and determination of AP spEA (Leu-pNA substrate) in cell-free soluble protein extracts (B). Lanes contained a constant amount of supernatant (20 lg) or of the
pellet protein. MWM: molecular weights markers. Data in (B) are the mean standard deviations (n = 3). (+): presence. (): absence. (C) Sequence of tryptic peptides (by LC
MS/MS) derived from the 100-kDa protein band (enclosed by an oval). Identied peptides that match with the PfAM1 sequence (51% of sequence coverage) are underlined.

332

20 identied peptides matched the ePepN sequence (Fig. A.1). This

result indicates that ePepN does not contribute appreciably to the
100-kDa protein band (Fig. 4A).

333

Selection of the induction method for rPfAM1 expression

334

To select the best induction method, we compared conventional

induction with 1 mM IPTG added at late exponential phase in LB

330
331

335

broth [33], induction with 0.1 mM IPTG added at the beginning

of the culture in LB broth (to decrease the typical toxicity at higher
IPTG concentrations and augment the induction time), and
autoinduction with 0.2% lactose in ZYMB-5052 culture medium
[28]. In each case, the intensity of the rPfAM1 band in the soluble
fractions was increased compared with the controls (Fig. 5AC).
Bestatin-sensitive AP spEA was enhanced by one order of magnitude over the negative controls (Fig. 5D). The three induction

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

336
337
338
339
340
341
342
343

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
6

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

Fig. 5. Comparative analysis of three induction methods for rPfAM1 expression in E. coli BL21. Evaluation of expression by SDSPAGE (8% acrylamide, non-reducing
conditions, Coomassie staining) after 18 h-induction with 0.1 mM IPTG added at the beginning of the culture in ampicillin-supplemented LB broth (A), 1 mM IPTG added at
late exponential phase of bacterial growth in ampicillin-supplemented LB broth (B), or auto-induction with 0.2% of lactose in ampicillin-supplemented ZYMB-5052 culture
media (C). Negative controls of expression are as presented in Fig. 4. Lanes contain a constant amount of supernatant (20 lg) or of pellet protein. rPfAM1 bands are enclosed
by ovals. (D) Evaluation of expression by determination of AP spEA (Leu-pNA substrate) in cell-free soluble protein extracts. The means standard deviations (n = 3). (+):
Presence. (): absence.

Table 1
Effect of three induction methods on rPfAM1 expression in E. coli BL21.
Induction method

0.1 mM IPTG
1 mM IPTG
Auto-induction
(0.2% lactose)

Final
OD600

3.6
3.9
3.7

nm

Approximate level of
expression
(% of total proteins)a
Insoluble fraction

Soluble fraction

2.0 0.3
4.1 0.2
3.6 0.9

9.4 2.2
19 0.8
11 1.2

Approximate volumetric
yield (mg of soluble rPfAM1/
L of culture)a

spEA
(104 U/mg of protein)

Expression level:
fold increment
over basal spEA

Total EA (U)

61 14
102 4
86 9

192 12
220 10
145 7

11 1b
10 2b
4 0.2c

1.25 0.08
1.17 0.05
1.15 0.06

Means standard deviations (n = 3). OD600 nm: optical density at 600 nm. EA: enzymatic activity. spEA: specic EA. U: unit of EA, dened as the amount of enzyme hydrolyzing
1 lmol of substrate per min under the assays conditions.
a
Determined by gel densitometry with ImageJ 1.38d software.
b
The mean activities registered in negative controls of expression (non-transformed strain, strain transformed with pTrcHis2B vector and strain transformed with genetic
construct pTrcHis2B-rPfAM1 but without induction) were taken as the basal activity.
c
In this case, we considered as basal activity the activity measured in the non-induced control.

344
345
346
347
348
349
350

methods resulted in similar nal growth density, spEA and total EA

(Table 1). The highest levels of the rPfAM1 band, as determined by
densitometric analysis, were obtained by induction with 1 mM
IPTG, which produced an amount of rPfAM1 equivalent to 4% of
the insoluble proteins and 19% of the proteins from the soluble
extract. Consequently, this method was the best for volumetric
yield, with approximately 100 mg of recombinant protein per L

of culture. This yield was considered adequate for the development

of purication and characterization of the recombinant enzyme.
Auto-induction with lactose was successfully used in our lab to
produce recombinant falcipain 2 in E. coli BL21(DE3) [34], but in
the present study, it was not better than conventional induction
(Fig. 5, Table 1). This could be due to a characteristic of the used
strain or the protein. The highest amount of soluble recombinant

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

351
352
353
354
355
356
357

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx
358
359
360
361
362

protein, a critical parameter to the success of the purication process, was obtained with 1 mM IPTG induction. Therefore, this
method was selected for routine production of rPfAM1.
Selection of induction time and number of ultrasonic cycles for
extraction of recombinant protein

372

A growth curve, representative of the E. coli BL21/pTrcHis2BrPfAM1 strain, is shown in Fig. A.2.A. The stationary phase was
reached 12 h after induction. The specic AP activity was
progressively augmented, with the highest value detected at 18 h
post-induction, which was selected for the production of the
recombinant enzyme. On the other hand, two ultrasonic cycles
were sufcient for the maximal recovery of total proteins and AP
specic activity, and spEA was notably reduced with four cycles
(Fig. A.2.B). Two cycles of ultrasonic treatment were selected for
rPfAM1 extraction from E. coli BL21/pTrcHis2B-rPfAM1 cells.

373

Modication of the genetic construct pTrcHis2B-rPfAM1

374

With the optimized parameters for rPfAM1 expression and

extraction, we attempted to purify the recombinant protein by
IMAC. Unexpectedly, the rst attempts to adsorb rPfAM1 on an
IMAC commercial resin, previously validated with a 6xHis-tagged
recombinant Green Fluorescent Protein (data not shown), were
unsuccessful. rPfAM1 adsorption was also not achieved by previous treatment of the protein extract with 8 M urea. Immunodetection of rPFAM1 in extracts with two distinct commercial anti-His
antibodies was negative. These ndings suggest that a conformational effect impeded the correct exposure of the histidine tag at
the rPfAM1 carboxyl terminus and hampered its interaction with
the IMAC resin.
As rPfAM1 co-exists with ePepN in the protein extract and both
enzymes share similar molecular weights and 35% sequence identity [20,24], the utilization of another purication method was
rejected because of the likely difculty of resolving two APs with
similar molecular characteristics. For this reason, we modied
the genetic construct pTrcHis2B-rPfAM1 to fuse a second histidine
tag at the amino terminus of rPfAM1 and facilitate purication by
IMAC (Fig. A.3). The modied construct was named pTrcHis2BrPfAM1-H6N.

363
364
365
366
367
368
369
370
371

375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417

Determination of the AP activity inhibition prole in the E. coli BL21

soluble protein extract enriched in rPfAM1
We determined the sensitivity of the AP activity in the E. coli
BL21 soluble protein extract enriched in the recombinant enzyme
toward inhibitors to obtain evidence about the mechanistic class
of the recombinant enzyme, prior to protein purication. The AP
activity in the extract was totally inhibited by bestatin (Fig. A.4).
In contrast, serine and cysteine protease inhibitors (PMSF, aprotinin, leupeptin and E64) were inactive, but pepstatin A (an aspartic
proteases inhibitor) inhibited the activity by 40%.
The metallo-AP inhibitors bestatin, nitrobestatin, amastatin and
epiamastatin inhibit, in a dose-dependent way, AP activity toward
the Leu-7-amido-4-methylcoumarin (AMC) uorogenic substrate
in P. falciparum trophozoite extracts [21]. Similarly, the puried
native enzyme is sensitive to several known metallo-AP inhibitors
and insensitive to protease inhibitors belonging to other mechanistic classes [9]. The inhibition prole for AP activity in the rPfAM1enriched bacterial extract (Fig. A.4) is in agreement with these
reports. The 40% inhibition caused by 15 lM pepstatin A may be
due to an interaction between the rPfAM1 S1 subsite and the isovaleric acid hydrophobic side chain at the amino terminus of the
pepstatin A hexapeptide; this suggestion requires experimental
conrmation.

Purication of rPfAM1

418

Purication of rPfAM1 in a single step from an E. coli BL21 protein extract enriched in the recombinant enzyme was performed
by IMAC. The chromatographic prole reveals that rPfAM1 elution
occurred at imidazole concentrations as low as 20 and 30 mM
(Fig. 6A). SDSPAGE indicated that these chromatographic fractions contained a 100-kDa protein that was highly concentrated
and accompanied by other proteins. The main additional bands
had molecular masses approximately 50 and 200 kDa. Then,
50 mM imidazole was selected for elution, as this concentration
value was sufcient to achieve total desorption of the recombinant
protein. The low afnity of rPfAM1 to the matrix in a native IMAC
could be explained by an incomplete exposure of the 6His tag in
the protein surface, and/or the use of a cobalt resin, which releases
histidine-tagged proteins with gentler conditions regarding nickel
resins (www.thermosher.com).
In Fig. 6B, a representative IMAC of the rPfAM1-enriched protein extract is shown. A 280-nm absorbance peak, associated with
the AP EA, was obtained. In contrast, if chromatography was
applied to a negative control extract (E. coli BL21/pTrcHis2B culture), this absorbance peak was not detected. The evaluation of
chromatographic runs by SDSPAGE and Coomassie staining indicated that if the negative control extract was loaded onto the column, elution with 50 mM imidazole did not lead to detection of
any band; on the contrary, if the IMAC column was loaded with
the extract from E. coli BL21/pTrcHis2B-rPfAM1-H6N, we observed
one highly concentrated 100 kDa protein in the eluates in addition to minor bands, including the 200- and 50-kDa bands.
Fractions 1 and 2 from the eluates of various chromatographic runs
loaded with the rPfAM1-enriched extract were pooled and dialtered with a 10-kDa cut-off membrane. Dialtration was performed to concentrate the recombinant protein and to remove
imidazole, which interferes with the EA assays for metallo-APs.
We sequenced tryptic peptides obtained from the 100- and
50-kDa protein bands by LCMS/MS (Fig. 7). We corroborated
that both bands corresponded to rPfAM1, with 24 sequenced peptides in the rst band (55% of sequence coverage) and 8 peptides
for the second band (23% of sequence coverage). The position of
the identied peptides in the PfAM1 sequence from the 50-kDa
band indicates that this protein is an amino-terminal degradation
product of rPfAM1. The identication of this amino-terminal degradation product is in agreement with the report of McGowan
et al. [20] about the purication of another recombinant variant
of PfAM1. These authors obtained an amino-terminal degradation
product with a molecular mass of 55 kDa, the presence of which
did not hamper the successful determination of the PfAM1
three-dimensional structure by X-ray crystallography with a
2.1- resolution. The selective adsorption of an amino-terminal
degradation product to the IMAC column (Fig. 7) is consistent with
the presence of a histidine tag at the carboxyl terminus of rPfAM1
that is properly exposed in the cleaved molecule for its interaction
with the Co2+ resin.
To test whether other protein bands coeluting with the
100-kDa band were additional rPfAM1 degradation products, a
Western blot with a commercial anti-6His tag antibody was performed on the eluates from IMACs loaded with extracts derived
from the E. coli BL21/pTrcHis2B and E. coli BL21/pTrcHis2BrPfAM1-H6N cultures (Fig. 8). The anti-His antibody did not interact with host proteins but recognized the 100-kDa band rPfAM1
and additional bands in the dialtered fraction, including that of
200-kDa. This indicates that the minority bands which elute
together with rPfAM1 on IMAC are degradation products of the
recombinant protein. Furthermore, as the antibody recognized several components in the rPfAM1-enriched extract, degradation
occurred before protein extract application to the column. The

419

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
8

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

Fig. 6. rPfAM1 purication by IMAC. (A) Determination of the imidazole concentration that allows elution of rPfAM1 adsorbed to a Co2+ matrix. (B) Representative IMAC for
rPfAM1 purication. A comparison with the extract that does not contain rPFAM1 (derived from an E. coli BL21/pTrcHis2B culture) is shown. The chromatographic runs were
evaluated by SDSPAGE (8% of acrylamide, reducing conditions, Coomassie staining). Protein samples applied on gels were normalized by volume. MWM: molecular weight
markers. Ext: rPfAM1-enriched protein extract. FT: ow-through sample of the column. EA: enzymatic activity. El: eluate.

Fig. 7. Identication of the 100-kDa (triangle) and 50-kDa (arrow) bands obtained as a result of rPfAM1 purication by IMAC. The identication was performed by the
sequencing of tryptic peptides derived from the gel bands (8% of acrylamide, reducing conditions, Coomassie staining) by LCMS/MS. The amount of protein applied in the gel
was normalized by volume. Ext: rPfAM1-enriched protein extract. FT: ow-through sample of the column. IMAC: pool of eluates from various IMAC runs. Dialt: dialtered
rPfAM1 with a 10-kDa cut-off membrane. M: molecular weight markers.

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

Fig. 8. Western-blot analysis of the expression and purication of rPfAM1 with an anti-histidine antibody. SDSPAGE (8% of acrylamide, reducing conditions); the gel was
stained with Coomassie (left); proteins were transferred to a nitrocellulose membrane and revealed with an anti-His antibody (right). The amount of protein applied to the gel
was normalized by volume. Ext: protein extract. FT: ow-through sample. El: eluate. M: molecular weight markers. Diaf: dialtered rPfAM1 with a 10-kDa cut-off membrane.

Table 2
Summary of the rPfAM1 purication process.
Purication step

EA (U)

Yield (%)

spEA (U/mg)

Purication
factor (fold)

Extract
IMAC
Dialtration

72.5 3.7
4.57 0.34
17.2 1.16

100
6.3
24

0.56 0.03
0.73 0.05
15.3 1.03

1
1.3
27.3

EA: AP enzymatic activity towards 1 mM Leu-pNA substrate. U: unit of EA, dened

as the amount of enzyme hydrolyzing 1 lmol of substrate per min under the assays
conditions. spEA: specic EA. Means standard deviations (n = 3).
483
484
485
486

recognition of the 200-kDa band suggests that it is an rPfAM1

dimer. Densitometric analysis of the gel indicated that the
100 kDa band, corresponding to intact rPfAM1, represented more
than 90% of total proteins in the dialtered preparation.

A summary of the purication process is presented on Table 2.

The low EA in the IMAC fraction is possibly due to the presence of
imidazole masking part of the potentially existent activity. rPfAM1
was puried in a single step from the extract, with a 24% yield and
a 27-fold purication factor.

487

Partial kinetic characterization of rPfAM1

492

The Zn2+ divalent cation is present in the active site of metalloAPs belonging to the M1 family [35], where it acts as an enzymatic
cofactor [36]. The presence of Zn2+ in the active site of rPfAM1 is
suggested by the 80% inhibition caused by 1 mM 1,10-phenanthroline, a metal-ion chelator. Fig. 9A shows that the AP activity of the
puried recombinant enzyme was diminished by 50% with 10 lM
ZnCl2 and by more than 80% when the concentration was raised to
100 lM. A qualitatively similar result was obtained with ZnSO4

493

Fig. 9. Partial kinetic characterization of rPfAM1. (A) Effect of the Zn2+ concentration (ZnCl2) on the rPfAM1 enzymatic activity. Substrate: 1.5 mM Leu-pNA. Buffer: 50 mM
TrisHCl, pH 7.27.4. Means (n = 3). (B) Effect of pH on the rPfAM1 EA. Substrate: 1.5 mM Leu-pNA. Buffers: 100 mM MES (pH 5.2, 5.5 and 6.5; triangles) and 50 mM TrisHCl
(pH 7.2, 7.5, 8.0, 8.5 and 9.0; squares). Means (n = 3). (C) Effect of substrates on rPfAM1 EA; substrate concentration range: 15.625500 lM. Buffer: 50 mM TrisHCl, pH 7.2
7.4. The kinetic parameters for each substrate were calculated from LineweaverBurk plots. The substrates with the highest specicity constants (kcat/KM) are highlighted in
bold. kcat: catalytic constant or turnover number. ND: Not determined.

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

488
489
490
491

494
495
496
497
498
499
500

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
10
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

(data not shown), indicating that the inhibitory effect is not caused
by the anion. This suggests that the Zn2+ associated to the active
sites is not lost during purication and indicates the uselessness
of supplementing the enzymatic preparation with this cation. A
similar result was reported for the neutral AP from Bombyx mori,
which is activated at very low Zn2+ concentrations but inhibited
with ion concentrations higher than 100 lM [37].
The maximal rPfAM1 activity toward the Leu-pNA substrate was
registered in the pH range 7.27.4, with an abrupt fall between pH
8.0 and 9.0 (Fig. 9B). The reduction of the activity in the acid zone
was more gradual, as indicated by the detection of 40% activity at
pH 5.25.5 compared with the maximal value. For native PfAM1,
AP activity toward Leu-AMC is optimum at pH 7.4, but at least
40% of activity remains between pH 5.8 and pH 8.6 [9]. As some
authors afrm that PfAM1 performs its biological function in the
lumen of the parasite acidic digestive vacuole [26], where pH values
range from 5.0 to 5.5 [38], attention has been focused on enzyme
activity in acidic conditions. Likewise, mammalian M1-family basic
AP acts in acidic vesicles and shows a high activity at pH 5.0 [39].
However, another group reported that native and recombinant
PfAM1 have an optimum pH of 7.0 toward Arg-AMC, with an activity <20% at pH 6.0 and a total loss of activity at pH 5.0 [20]. The discrepancy in remaining activity at pH < 6.0 might be due to
dissimilar enzyme concentrations or buffers in each pH range, as
enzyme activity can vary according to the buffer identity at a xed
pH and enzyme concentration [40]. Unfortunately, this information
is not always available. In correspondence with this idea, we
observed that the activity of rPfAM1 toward Leu-pNA at pH 5.2
reached 40% of maximal activity (obtained at pH 7.27.4) when
it was measured in a MES buffer, but it was only 10% of the maximal activity when an acetate buffer was used (data not shown).
This is the rst report of the kinetic parameters of PfAM1 determined with aminoacyl-pNA-conjugated substrates. The higher values of the specicity constant were obtained with the substrates
Lys-, Arg-, Leu- and Ala-pNA, in that order (Fig. 9C). It was not possible to determine the kinetic parameters with the isoleucine-,
valine-, glycine- and proline-based substrates because activity
was not detected, even at the maximal substrate concentration
used of 500 lM. The preference order of rPfAM1 for substrates is
in agreement in general terms with what has been previously
reported. With a single dose of aminoacyl uorogenic substrates
and neutral pH, the order of preference of native PfAM1 for the ve
best P1 residues was lysine > alanine > arginine > leucine > phenylalanine [9]. Two similar studies with the same rPfAM1 form that
determined the specicity constants toward uorogenic substrates
demonstrated a preference order for P1 amino acids of (methionine = leucine) > (alanine = arginine) > tryptophan > phenylalanine; the specicity constant toward lysine in the P1 position of
the substrate was not reported [20,41].
The substrate specicity of another variant of PfAM1 determined through the specicity constant toward the dipeptides XAla revealed the following order of preference for the seven best
P1 residues: lysine > methionine > arginine > (leucine = phenylalanine) > tyrosine > alanine [42]. The preference for lysine in the P1
position is in agreement with the results obtained in the present
work (Fig. 9C) as well as with what has been reported for native
PfAM1 toward aminoacyl-AMC conjugates [9]. The position of
methionine in the order of preference agrees with the data generated with uorogenic substrates [20,41]. The assignation of arginine and leucine to the third and fourth positions in the
hierarchy of PfAM1 specicity [42] is also consistent with our data
(Fig. 9C) and previous data. However, the relatively poor hydrolysis
of the dipeptide Ala-Ala does not match the specicity preference
toward the Ala-containing uorogenic conjugates, most likely due

to the structural differences between both groups of substrates

[9,20,41]. In summary, the highest specicity constants are
obtained for X-Ala peptides that have P1 sites that are occupied
by amino acids with linear (basic or uncharged: lysine, methionine,
arginine) or voluminous hydrophobic (e.g., leucine, phenylalanine)
side chains [42]. Neither histidine [41,42], the basic side chain of
which is too short to interact with a key Glu572 at the bottom of
the S1 pocket [2,22,43], nor proline [20,41,42], which does not
have a free amino group, allows efcient hydrolysis by PfAM1.
These reports are in agreement with the results obtained in the
present work (Fig. 9C). On the other hand, as acid side chains probably do not establish favorable interactions with the bottom of the
S1 subsite, aspartate and glutamate are not tolerated by native and
recombinant PfAM1 at the P1 position of uorogenic substrates
[9,20,41] or X-Ala peptides [42].
The presence of small amounts of rPfAM1 degradation products
in our puried protein preparation (Fig. 8) does not apparently
modify the main kinetic properties of the recombinant enzyme
compared with those of the native enzyme. This is in agreement
with the results obtained by Ragheb et al. [26], who proved that
the kinetic parameters of the non-processed form of native PfAM1,
the complex between an enzyme-processed form and its resulting
fragment, and a PfAM1 recombinant form, are very similar toward
the uorogenic substrate Arg-bNA. This indicates that the presence
of non-catalytic fragments of the enzyme does not interfere with
catalysis performed by the intact enzyme or by the fragment that
contains the active site. On the other hand, the low concentration
of degradation products in the nal puried preparation (<10%;
Fig. 8) indicates that it is not necessary to use protease inhibitors
during preparation of the E. coli soluble protein extract enriched
in rPfAM1.
The similarities in kinetic characteristics (substrate specicity,
activity variation in function of pH, requirement for metallic cofactors, and inhibition prole) between native PfAM1 and recombinant variants are notable. These coincidences indicate that the
rst 194 amino acids and the last 283 residues of PfAM1 (which
are alternatively absent in different rPfAM1 variants [11,20,26])
do not dictate the primary catalytic properties of the enzyme. In
this sense, the recombinant variant obtained in the present work
(lacking the rst 194 amino acids of native PfAM1) is not the
exception, as the studied kinetic properties are similar to those
that have been previously reported. This suggests the possibility
of using rPfAM1 as a model of native PfAM1 to search for and
develop inhibitors.

565

Conclusions

609

An rPfAM1 gene was designed and synthesized. The gene was

optimized for expression in E. coli, and its CAI and average G/C content were higher than that of the native gene. Transformation of
E. coli BL21 with pTrcHis2B-rPfAM1 and induction of expression
at 37 C for 18 h with 1 mM IPTG added at the late exponential
phase led to the production of a soluble and active form of rPfAM1.
The expression level was 19% of the soluble proteins, and the volumetric yield was 100 mg/L of culture. A histidine tag fused to the
amino terminus of rPfAM1 allowed a 27-fold purication in a single step by IMAC. Puried rPfAM1 showed a low level of proteolytic
degradation (<10% degradation products). rPfAM1 had kinetic
characteristics, such as a neutral optimal pH, substrate specicity
for basic and hydrophobic amino acids at P1 position, inhibition
prole typical of metallo-APs, and inhibition by Zn2+ excess, that
were similar to those of the native PfAM1. Thus, the expression
and purication strategy that we developed allows producing

610

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608

611
612
613
614
615
616
617
618
619
620
621
622
623
624
625

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

627

enzyme of sufcient quality and quantity to be useful to screen for

PfAM1 inhibitors.

628

Acknowledgments

629

651

This work was supported by Grant F/4730-1 from the International Foundation for Science (IFS, Sweden) (2010-2012), fellowship 811.06.03 from Secretara de Relaciones Exteriores del
Gobierno de Mxico (2010), and fellowship from Universidad de
las Naciones Unidas, Programa de Biotecnologa para Latinoamrica y el Caribe (UNU-BIOLAC) (2012) to J.G.B.; Grant J000.0453 from
Consejo Nacional de Ciencia y Tecnologa (CONACYT, Mexico)
Ministerio de Ciencia, Tecnologa y Medio Ambiente (CITMA, Cuba)
(20092011) to J.L.C.; Grant 176351 from CONACYT to J.O.; and
Red del Programa Iberoamericano de Ciencia y Tecnologa para el
Desarrollo (CYTED-PROMAL, 210RT0398) Protemica y Quimiogenmica de Inhibidores de Proteasas de Origen Natural con
Potencial Teraputico en Malaria to MC. The authors thank
Dr. Isel Pascual, Centro de Estudio de Protenas, Facultad de Biologa, Universidad de La Habana, Cuba, for helpful advice during
the conception of this work and critical reading of the manuscript;
Dr. Emir Salas, Universidad Nacional de General San Martn,
Argentina, for useful suggestions for the design of rpfam1 gene
and expression experiments; Fidelia Romero, Instituto de Biotecnologa, UNAM, for assistance in the Molecular Biology techniques; and Lic. Annamil lvarez, Fundacin Instituto de Estudios
Avanzados, Venezuela, for technical assistance during the Western
blot assay.

652

Appendix A. Supplementary data

653
654

Supplementary data associated with this article can be found, in

the online version, at http://dx.doi.org/10.1016/j.pep.2014.08.002.

655

References

626

630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650

656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694

[1] T.S. Skinner-Adams, C.M. Stack, K.R. Trenholme, C.L. Brown, J. Grembecka, J.
Lowther, A. Mucha, M. Drag, P. Kafarski, S. McGowan, J.C. Whisstock, D.L.
Gardiner, J.P. Dalton, Plasmodium falciparum neutral aminopeptidases: new
targets for anti-malarials, Trends. Biochem. Sci. 35 (2010) 5361.
[2] I. Florent, Z. Derhy, M. Allary, M. Monsigny, R. Mayer, J. Schrvel, A Plasmodium
falciparum aminopeptidase gene belonging to the M1 family of zincmetallopeptidases is expressed in erythrocytic stages, Mol. Biochem.
Parasitol. 97 (1998) 149160.
[3] M.B. Harbut, G. Velmourougane, S. Dalal, G. Reissa, J.C. Whisstock, O. Onder, D.
Brisson, Sh. McGowan, M. Klemba, D.C. Greenbaum, Bestatin-based chemical
biology strategy reveals distinct roles for malaria M1- and M17-family
aminopeptidases, Proc. Natl. Acad. Sci. U.S.A. 108 (2011) E526E534.
[4] A. Pain, U. Bhme, A.E. Berry, K. Mungall, R.D. Finn, A.P. Jackson, T. Mourier, J.
Mistry, E.M. Pasini, M.A. Aslett, S. Balasubrammaniam, K. Borgwardt, K. Brooks,
C. Carret, T.J. Carver, I. Cherevach, T. Chillingworth, T.G. Clark, M.R. Galinski, N.
Hall, D. Harper, D. Harris, H. Hauser, A. Ivens, C.S. Janssen, T. Keane, N. Larke, S.
Lapp, M. Marti, S. Moule, I.M. Meyer, D. Ormond, N. Peters, M. Sanders, S.
Sanders, T.J. Sargeant, M. Simmonds, F. Smith, R. Squares, S. Thurston, A.R.
Tivey, D. Walker, B. White, E. Zuiderwijk, C. Churcher, M.A. Quail, A.F. Cowman,
C.M.R. Turner, M.A. Rajandream, C.H.M. Kocken, A.W. Thomas, C.I. Newbold,
B.G. Barrell, M. Berriman, The genome of the simian and human malaria
parasite Plasmodium knowlesi, Nature 455 (2008) 799803.
[5] G.P. Curley, S.M. ODonovan, J. McNally, M. Mullally, H. OHara, A. Troy, S.A.
OCallaghan, J.P. Dalton, Aminopeptidases from Plasmodium falciparum,
Plasmodium chabaudi chabaudi and Plasmodium berghei, J. Eukaryot.
Microbiol. 41 (1994) 119123.
[6] K. Kolakovich, I. Gluzman, K. Dufn, D. Goldberg, Generation of hemoglobin
peptides in the acidic digestive vacuole of Plasmodium falciparum implicates
peptide transport in amino acid production, Mol. Biochem. Parasitol. 87 (1997)
123135.
[7] C. Gavigan, J. Dalton, A. Bell, The role of aminopeptidases in haemoglobin
degradation in Plasmodium falciparum-infected erythrocytes, Mol. Biochem.
Parasitol. 117 (2001) 3748.
[8] C. Gavigan, S.G. Machado, J. Dalton, A. Bell, Analysis of antimalarial synergy
between bestatin and endoprotease inhibitors using statistical responsesurface modelling, Antimicrob. Agents Chemother. 45 (2001) 113175113181.
[9] M. Allary, J. Schrvel, I. Florent, Properties, stage-dependent expression and
localization of Plasmodium falciparum M1 family zinc-aminopeptidase,
Parasitology 125 (2002) 110.

11

[10] S. Dalal, M. Klemba, Roles for two aminopeptidases in vacuolar hemoglobin

catabolism in Plasmodium falciparum, J. Biol. Chem. 282 (2007) 3597835987.
[11] O. Azimzadeh, C. Sow, M. Gze, J. Nyalwidhe, I. Florent, Plasmodium falciparum
PfA-M1 aminopeptidase is trafcked via the parasitophorous vacuole and
marginally delivered to the food vacuole, Malaria J. 9 (2010) 189204.
[12] D.L. Vander Jagt, B.R. Baack, L.A. Hunsaker, Purication and characterization of
an aminopeptidase from Plasmodium falciparum, Mol. Biochem. Parasitol. 10
(1984) 4554.
[13] D.L. Vander Jagt, W.S. Caughey, N.M. Campos, L.A. Hunsaker, M.A. Zanner,
Parasite proteases and antimalarial activities of the protease inhibitors, Prog.
Clin. Biol. Res. 313 (1989) 105118.
[14] P. Olaya, M. Wasserman, Effect of calpain inhibitors on the invasion of human
erythrocytes by the parasite Plasmodium falciparum, Biochim. Biophys. Acta
1096 (1991) 217221.
[15] A. Kitjaroentham, T. Suthiphongchai, P. Wilarairat, Effect of metalloprotease
inhibitors on invasion of red blood cell by Plasmodium falciparum, Acta Trop.
97 (2006) 59.
[16] M. Flipo, I. Florent, P. Grellier, C. Sergheraerta, R. Deprez-Poulain, Design,
synthesis and antimalarial activity of novel, quinoline-based, zinc metalloaminopeptidase inhibitors, Bioorg. Med. Chem. Lett. 13 (2003) 26592662.
[17] M. Flipo, T. Beghyn, V. Leroux, I. Florent, B.P. Deprez, R. Deprez-Poulain, Novel
selective inhibitors of the zinc plasmodial aminopeptidase PfA-M1 as potential
antimalarial agents, J. Med. Chem. 50 (2007) 13221334.
[18] M. Flipo, T. Beghyn, J. Charton, V.A. Leroux, B.P. Deprez, R.F. Deprez-Poulain,
A library of novel hydroxamic acids targeting the metallo-protease family:
design, parallel synthesis and screening, Bioorg. Med. Chem. 15 (2007)
6376.
[19] T.S. Skinner-Adams, J. Lowther, F. Teuscher, C.M. Stack, J. Grembecka, A.
Mucha, P. Kafarski, K.R. Trenholme, J.P. Dalton, D.L. Gardiner, Identication of
phosphinate dipeptide analog inhibitors directed against the Plasmodium
falciparum M17 leucine aminopeptidase as lead antimalarial compounds, J.
Med. Chem. 50 (2007) 60246031.
[20] S. McGowan, C.J. Porter, J. Lowther, C.M. Stack, S.J. Golding, T.S. Skinner-Adams,
K.R. Trenholme, F. Teuscher, S.M. Donnelly, J. Grembecka, A. Mucha, P. Kafarski,
R. DeGori, A.M. Buckle, D.L. Gardiner, J.C. Whisstock, J.P. Dalton, Structural
basis for the inhibition of the essential Plasmodium falciparum M1 neutral
aminopeptidase, Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 25372542.
[21] M. Nankya-Kitaka, G. Curley, C. Gavigan, A. Bell, J. Dalton, Plasmodium chabaudi
chabaudi and P. falciparum: inhibition of aminopeptidase and parasite growth
by bestatin and nitrobestatin, Parasitol. Res. 84 (1998) 552558.
[22] G. Velmourougane, M.B. Harbut, S. Dalal, S. McGowan, C.A. Oellig, N.
Meinhardt, J.C. Whisstock, M. Klemba, D.C. Greenbaum, Synthesis of new (-)bestatin-based inhibitor libraries reveals a novel binding mode in the S1
pocket of the essential malaria M1 metalloaminopeptidase, J. Med. Chem. 54
(2011) 16551666.
[23] A. Addlagatta, L. Gay, B.W. Matthews, Structure of aminopeptidase N from
Escherichia coli suggests a compartmentalized, gated active site, Proc. Natl.
Acad. Sci. U.S.A. 103 (2006) 1333913344.
[24] K. Ito, Y. Nakajima, Y. Onohara, M. Takeo, K. Nakashima, F. Matsubara, T. Ito, T.
Yoshimoto, Crystal structure of aminopeptidase N (proteobacteria alanyl
aminopeptidase) from Escherichia coli and conformational change of
methionine 260 involved in substrate recognition, J. Biol. Chem. 281 (2006)
3366433676.
[25] B. Nocek, R. Mulligan, M. Bargassa, F. Collart, A. Joachimiak, Crystal structure of
aminopeptidase N from human pathogen Neisseria meningitides, Proteins 273
(2007) 273279.
[26] D. Ragheb, S. Dalal, K.M. Bompiani, W.K. Ray, M. Klemba, Distribution and
biochemical properties of an M1-family aminopeptidase in Plasmodium
falciparum indicate a role in vacuolar hemoglobin catabolism, J. Biol. Chem.
286 (2011) 2725527265.
[27] A. Egon, J. Brosius, M. Ptashne, Vectors bearing a hybrid trp-lac promoter useful
for regulated expression of cloned genes in Escherichia coli, Gene 25 (1983)
167178.
[28] F.W. Studier, Protein production by auto-induction in high-density shaking
cultures, Protein Expr. Purif. 41 (2005) 207234.
[29] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head
of bacteriophage T4, Nature 227 (1970) 680685.
[30] M.M. Bradford, A rapid and sensitive method for the quantication of
microgram quantities of protein utilizing the principle of protein binding,
Anal. Biochem. 72 (1976) 248254.
[31] J.W. London, L.M. Shaw, D. Fetterolf, D. Garnkel, Determination of the
mechanism and kinetic constants for hog kidney c-glutamyltransferase,
Biochem. J. 157 (1976) 609617.
[32] P.M. Sharp, W.H. Li, The Codon Adaptation Index a measure of directional
synonymous codon usage bias, and its potential applications, Nucleic Acids
Res. 15 (1987) 12811295.
[33] A. Sivashanmugam, V. Murray, C. Cui, Y. Zhang, J. Wang, Q. Li, Practical
protocols for production of very high yields of recombinant proteins using
Escherichia coli, Protein Sci. 18 (2009) 936948.
[34] E. Salas, A. Cabrera, S.A. Trejo, M.A. Chvez, High-level expression of Falcipain2 in Escherichia coli by codon optimization and auto-induction, Protein Expr.
Purif. 83 (2012) 5969.
[35] N.D. Rawlings, A.J. Barret, Evolutionary families of metallopeptidases, Methods
Enzymol. 248 (1995) 183228.
[36] A. Mucha, M. Drag, J.P. Dalton, P. Kafarski, Metallo-aminopeptidase inhibitors,
Biochimie 92 (2010) 15091529.

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002

695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780

YPREP 4545

No. of Pages 12, Model 5G

14 August 2014
12
781
782
783
784
785
786
787
788
789
790
791
792
793

J. Gonzlez-Bacerio et al. / Protein Expression and Purication xxx (2014) xxxxxx

[37] G. Hua, K. Tsukamoto, R. Taguchi, M. Tomita, S. Miyajima, H. Ikezawa,

Characterization of aminopeptidase N from the brush border membrane of the
larvae midgut of silkworm, Bombyx mori as a zinc enzyme, Biochim. Biophys.
Acta 1383 (1998) 301310.
[38] Y. Kuhn, P. Rohrbach, M. Lanzer, Quantitative pH measurements in Plasmodium
falciparum-infected erythrocytes using pHluorin, Cell. Microbiol. 9 (2007)
10041013.
[39] S.R. Hwang, A. ONeill, S. Bark, T. Foulon, V. Hook, Secretory vesicle
aminopeptidase B related to neuropeptide processing: molecular
identication and subcellular localization to enkephalin- and NPY-containing
chromafn granules, J. Neurochem. 100 (2007) 13401350.
[40] W.T. Frankenberger, J.B. Johanson, Effect of pH on enzyme stability in soils, Soil
Biol. Biochem. 14 (1982) 433437.

[41] M. Poreba, S. McGowan, T.S. Skinner-Adams, K.R. Trenholme, D.L. Gardiner, J.C.
Whisstock, J. To, G.S. Salvesen, J.P. Dalton, M. Drag, Fingerprinting the
substrate specicity of M1 and M17 aminopeptidases of human malaria
Plasmodium falciparum, PLoS One 7 (2012) e31938, http://dx.doi.org/10.1371/
journal.pone.0031938.
[42] S. Dalal, D.R.T. Ragheb, M. Klemba, Engagement of the S1, S10 and S20 subsites
drives efcient catalysis of peptide bond hydrolysis by the M1-family
aminopeptidase from Plasmodium falciparum, Mol. Biochem. Parasitol. 183
(2012) 7077.
[43] P.M. Jones, M.W. Robinson, J.P. Dalton, A.M. George, The Plasmodium
falciparum malaria M1 alanyl aminopeptidase (PfA-M1): insights of catalytic
mechanism and function from MD simulations, PLoS One 6 (2011) e28589,
http://dx.doi.org/10.1371/journal.pone.0028589.

794
795
796
797
798
799
800
801
802
803
804
805
806
807

Please cite this article in press as: J. Gonzlez-Bacerio et al., High-level expression in Escherichia coli, purication and kinetic characterization of Plasmodium
falciparum M1-aminopeptidase, Protein Expr. Purif. (2014), http://dx.doi.org/10.1016/j.pep.2014.08.002