Aconoidasida: Plasmodiidae), Using Fluorogenic Peptide Substrates

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS http://www.rccb.uh.cu

ARTÍCULO ORIGINAL Kinetic characterization of recombinant PfAM1, a M1-aminopeptidase from Plasmodium falciparum (Aconoidasida: Plasmodiidae), using fluorogenic peptide substrates

Caracterización cinética de la PfAM1 recombinante, una aminopeptidasa M1 de Plasmodium falciparum (Aconoidasida: Plasmodiidae), con sustratos peptídicos fluorogénicos

Jorge O. González-Bacerio1, Adriana K. Carmona2, Marcos L. Gazarini3, María Á. Chávez1 and Maday Alonso del Rivero1*

1 Centro de Estudio de Proteínas, Facultad de Biología, Universidad de ABSTRACT La Habana, La Habana, Cuba. Plasmodium falciparum neutral M1-aminopeptidase (PfAM1) is a validated 2 Departamento de Biofísica, target against malaria, the main human parasitic disease in tropical regions. Universidade Federal de São Paulo, Recently, our group obtained a recombinant variant of this enzyme (rPfAM1) São Paulo, Brasil. in Escherichia coli with a yield of 24 mg of active protein per L of culture. As 3 Departamento de Biociências, Universidade Federal de São Paulo, a way to confirm the suitability of this recombinant aminopeptidase to be Santos, São Paulo, Brasil. used as a target for the identification of new potent and selective inhibitors of the endogenous enzyme, and adjusting the experimental parameters of the inhibition assays, here we determined the main kinetic characteristics of * Autor para correspondencia: rPfAM1 using L-aminoacyl-7-amido-4-methylcoumarin (AMC)-conjugated 2+ [email protected] fluorogenic substrates. Firstly, the Zn cation inhibited rPfAM1 activity at concentrations > 1μ mol/L, an effect qualitatively analogous to that observed for other metallo-aminopeptidases. In agreement with earlier reports, the near-neutral pH (7.2-7.4) is the optimum against the 4 substrates tested (Ala -, Leu-, Met- and Arg-AMC). The KM values at pH 7.2 (Met-AMC: 23 ± 5 μmol/ L; Arg-AMC: 37 ± 4 μmol/L; Ala-AMC: 106 ± 9 μmol/L; Leu-AMC: 60 ± 7 μmol/ L) are closer to those previously reported for the native enzyme than those described for another recombinant variant of PfAM1. On the other hand, the -1 - kcat values obtained here at pH 7.2 (Met-AMC: 99 ± 9 s ; Arg-AMC: 129 ± 6 s 1; Ala-AMC: 268 ± 20 s-1; Leu-AMC: 79 ± 6 s-1) are the highest reported to date for PfAM1, indicating that our enzyme preparation is particularly active. Recibido: 2015-04-11 Diminution of pH from 7.2 to 5.2, affected more the catalytic efficiency of Aceptado: 2015-09-01 rPfAM1 than affinity toward Leu-, Met- and Arg-AMC. The contrary effect was observed with Ala-AMC. The kinetic parameters at near-neutral pH varied as a function of the identity of the substrate amino-terminal residue, in a

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 41 JORGE O. GONZÁLEZ-BACERIO ET AL. similar way to that reported by another group using natural dipeptide substrates against a third reported recombinant form of PfAM1. Finally, the rPfAM1 here obtained was not sensitive to the inhibitors PMSF, E64 and pepstatin A, but was inhibited by amastatin, 1,10-phenanthroline and bestatin. This is an inhibition profile typical of neutral and basic metallo-aminopeptidases, also reported for native PfAM1. The obtainment of the kinetic parameters of rPfAM1 toward aminoacyl-AMC peptides will allow using these substrates to assess the inhibition of the endogenous enzyme on isolated live parasites.

Keywords: malaria, kinetic characterization, kinetic parameters,- M1 family aminopeptidase, fluorogenic peptide substrates, AMC, Plasmodium falciparum, PfAM1

RESUMEN La aminopeptidasa M1 neutra de Plasmodium falciparum (PfAM1) es un blanco validado contra la malaria, la principal enfermedad parasitaria humana en las regiones tropicales. Recientemente, nuestro grupo obtuvo una variante recombinante de esta enzima (PfAM1r) con un rendimiento de 24 mg de proteína activa por L de cultivo. Para confirmar la conveniencia de esta aminopeptidasa recombinante como blanco para la identificación de nue- vos inhibidores potentes y selectivos de la enzima endógena, y para ajustar los parámetros experimentales de los ensayos de inhibición, en este trabajo se determinaron las principales características cinéticas de la PfAM1r frente a sustratos fluorogénicos conjugados L-aminoacil-7-amida-4-metilcumarina (AMC). En primer lugar, el catión Zn2+ inhibió la actividad de la PfAM1r a concentraciones > 1 μmol/L, un efecto análogo cualitativamente al obser- vado para otras aminopeptidasas de tipo metalo. En correspondencia con informes anteriores, el pH cercano al neutro (7,2-7,4) es el óptimo con los 4 sustratos evaluados (Ala-, Leu-, Met- y Arg-AMC). Los valores de KM al pH 7,2 (Met-AMC: 23 ± 5 μmol/L; Arg-AMC: 37 ± 4 μmol/L; Ala-AMC: 106 ± 9 μmol/L; Leu-AMC: 60 ± 7 μmol/L) son más cercanos a los informados previamente para la enzima nativa que los obtenidos por otro grupo de autores con otra PfAM1 recombinante. Por otra parte, los valores de kcat obtenidos en este trabajo a pH 7,2 (Met-AMC: 99 ± 9 s-1; Arg-AMC: 129 ± 6 s-1; Ala-AMC: 268 ± 20 s-1; Leu-AMC: 79 ± 6 s-1) son los mayores informados hasta ahora para la PfAM1, lo que indica que la preparación enzimática obtenida es particularmente activa. La disminu- ción del pH de 7,2 a 5,2, afectó más la eficiencia catalítica de la PfAM1r que la afinidad frente a la Leu-, Met- y Arg-AMC. El efecto contrario se observó con la Ala-AMC. Los parámetros cinéticos a valores de pH cercanos al neutro variaron en función de la identidad del residuo amino terminal del sustrato, de manera similar a lo informado por otro grupo, los que utilizaron una tercera PfAM1 recombinante y sustratos dipéptidos naturales. Final- mente, la PfAM1r no resultó sensible a los inhibidores PMSF, E64 y pepstatina A, pero se inhibió por la amastatina, la 1,10-fenantrolina y la bestatina. Este es un perfil de inhibición típico de aminopeptidasas de tipo metalo neutras y básicas, informado también para la PfAM1 nativa. La obtención de los parámetros cinéticos de la PfAM1r frente a los péptidos aminoacil-AMC permitirá el empleo de estos sustratos para evaluar la inhibición de la enzima endógena en parásitos vivos aislados.

Palabras clave: malaria, caracterización cinética, parámetros cinéticos, aminopeptidasa de la familia M1, sustratos peptídicos fluorogénicos, AMC,Plasmodium falciparum, PfAM1

INTRODUCTION Metalloproteases constitute the most diverse cata- metallo-APs are a distinctive group which develops lytic class of proteases, are broadly spread across liv- crucial biological functions for different organisms, ing beings, and contain in their active site at least one including their essential involvement in the life cycles divalent metallic cation (generally Zn2+, Co2+ or Mn2+) of different parasites. Therefore, neutral metallo-APs acting as a ligand for catalytic nucleophilic water are recognized as targets in relevant human parasitic (Barret et al., 2003). Among these enzymes, metallo- diseases. For example, a leucyl AP belonging to the aminopeptidases (APs) hydrolyze peptide bonds from M17 family of metalloproteases was identified as a the amino termini of polypeptide chains. Neutral target in the kinetoplastid parasite Trypanosoma

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 42 JORGE O. GONZÁLEZ-BACERIO ET AL. cruzi, the causative agent of Chagas disease (Cadavid- system Escherichia coli (McGowan et al., 2009; Azim- Restrepo et al., 2011). The ortholog of this enzyme is zadeh et al., 2010; Ragheb et al., 2011). All these re- also present in the genus Leishmania, which causes combinant proteins are truncated forms of PfAM1, leishmaniasis (Morty and Morehead, 2002). Similarly, and only the variants produced in 2009 and 2011 the protozoan unicellular parasite Plasmodium falci- (termed here rPfAM1195-1085 and rPfAM1192-1085, re- parum contains two neutral metallo-APs, belonging to spectively) were kinetically characterized using fluoro- the M1 (PfAM1) and M17 families, which have genic substrates (McGowan et al., 2009; Ragheb et al., emerged as promising targets in malaria (Florent et 2011; Poreba et al., 2012). Since their kinetic proper- al., 1998; Gardiner et al., 2006; Harbut et al., 2011). ties are similar to those of nPfAM1 (Allary et al., Native PfAM1 (nPfAM1) (EC 3.4.11.2) is a monomeric 2002), these proteins have been used as targets for the identification of inhibitors (McGowan et al., 2009; enzyme of 1085 amino acids and 126 kDa (primary Harbut et al., 2011; Velmourougane et al., 2011; Skin- protein product) (Florent et al., 1998; Allary et al., ner-Adams et al., 2012; Kannan Sivaraman et al., 2002), structured in 4 domains (McGowan et al., 2013). 2009). It has a general fold very similar to that of bac- terial neutral metallo-APs (Addlagatta et al., 2006). By using a synthetic gene optimized for the E. coli PfAM1 is expressed in the erythrocytic stages of the system, recently we expressed high levels of another parasite life cycle as two active and soluble processed recombinant variant of PfAM1 (rPfAM1) in the cytosol forms of 96 and 68 kDa (Florent et al., 1998; Allary et of E. coli BL21 (González-Bacerio et al., 2014). This al., 2002; Azimzadeh et al., 2010), and has been local- protein has a predicted molecular mass of 106.13 kDa ized on the cytosol (Allary et al., 2002; McGowan et and two six-histidine tags (at the- N and C-termini), al., 2009), digestive vacuole and nucleus (Dalal and allowing its purification in a single step by immobi- Klemba, 2007; Ragheb et al., 2011) of P. falciparum. lized metal ion affinity chromatography (IMAC). Here This AP contains a single Zn2+ ion in the active site we report the kinetic characterization of rPfAM1 to- (McGowan et al., 2009), has a pH optimum of 7.0-7.4 ward four L-aminoacyl-7-amido-4-methylcoumarin (Allary et al., 2002; McGowan et al., 2009), and it is (AMC) fluorogenic peptide substrates. This study al- specific for peptides (natural or artificial) with basic lows (i) confirming the possibility of using rPfAM1 as a and hydrophobic residues at the amino terminus (P1 model of the endogenous enzyme for the identifica- position) (Allary et al., 2002; McGowan et al., 2009; tion of new PfAM1 inhibitors, (ii) to establish the ap- Dalal et al., 2012; Poreba et al., 2012; González- propriate experimental conditions for the inhibition Bacerio et al., 2014). assays, and (iii) the further use of these substrates to test the inhibition of nPfAM1 on isolated live para- This essential and non-redundant enzyme (Dalal and sites. Klemba, 2007) participates in vacuolar (Ragheb et al., 2011) and/or cytosolic (Kolakovich et al., 1997) cleav- age of dipeptides derived from the degradation of the MATERIALS AND METHODS host hemoglobin. This process is nutritionally neces- Obtainment of rPfAM1 sary for parasite development during their clinically- rPfAM1 was expressed in E. coli and purified in a sin- relevant asexual erythrocytic stages (ring, trophozo- gle IMAC step as was previously described (González- ite, schizont and merozoite) (Liu et al., 2006), and has Bacerio et al., 2014). been profusely targeted for antimalarials development (Yeh and Altman, 2006). In addition, it has been proposed Protein concentration assay that PfAM1 plays a role on ring-trophozoite transition Concentration of rPfAM1 was determined by the (Harbut et al., 2011), trophozoite development (Gavigan Bradford method, using bovine serum albumin as et al., 2001), erythrocyte reinvasion (Allary et al., 2002), standard protein, and corroborated from absorbance and has an unknown function in the nucleus (Dalal and at 280 nm using the extinction coefficient of Klemba, 2007; Ragheb et al., 2011). rPfAM1192-1085 previously reported: Since the purification of nPfAM1, to obtain enough 1.15 x 105 L×mol-1×cm-1 (Ragheb et al., 2011). amounts for kinetic studies and further searching for Aminopeptidase activity assay inhibitors, is a difficult task, the enzyme has been ex- Aminopeptidase enzymatic activity (EA) was deter- pressed by several groups using the heterologous mined by a continuous kinetic method, using Ala-, Leu-,

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 43 JORGE O. GONZÁLEZ-BACERIO ET AL.

Arg- or Met-AMC fluorogenic substrates (Amino Tech pH 7.2 and 5.2, in order to in vitro reproduce the elec- P & D, Brazil) solubilized in dimethyl sulfoxide trostatic microenvironment of the active site of (DMSO). The fluorescence due to the release of AMC nPfAM1 in the cytosol and the food vacuole (Kuhn et (excitation: 380 nm, emission: 460 nm, slits: 10 nm) al., 2007) of P. falciparum, respectively. Substrate was monitored during 10 min using a microplate spec- selection was based on the known specificity of this trofluorometer (Hitachi F7000, Japan). Kinetic assays enzyme (Allary et al., 2002; McGowan et al., 2009; were carried out with concentrations of rPfAM1 that Dalal et al., 2012; Poreba et al., 2012; González- were linearly related to the initial rates, at 37°C, in 96- Bacerio et al., 2014). well black plates (200 μL final volume). The activity Assays were performed with 1.7 x 10-10 mol/L rPfAM1 buffers were: 50 mmol/L Tris-HCl pH 7.2, for the as- (at pH 7.2) and 4.2 ´ 10-10 mol/L rPfAM1 (at pH 5.2), at says performed at this pH; and constant ionic strength 8 concentrations of each substrate prepared by serial universal buffer (25 mol/Lm acetic acid, MES and gly- dilutions and spanning the following ranges: Ala-AMC cine, 75 mmol/L Tris), for pH optimum determination (3.2-406.3 μmol/L), Leu-AMC (2.7-347.1 μmol/L), Arg- and the assays performed at pH 5.2. The final concen- AMC (2.4-302 μmol/L), Met-AMC (2.6-326.7 μmol/L). tration of DMSO was less than 2 % (v/v). Only the linear The experimental conditions were equal to those portions of progress curves, corresponding to a sub- above described. The kinetic parameters Michaelis- strate consumption lower than 5 %, were used to 2 Menten’s constant (KM) and catalytic constant (kcat) measure the reaction rates. Slopes with R < 0.98 were calculated by fitting the experimental data to were not considered. The proportionality constant the Michaelis-Menten’s rectangular hyperbola func- between fluorescence and AMC concentration was tion (Copeland, 2000) using GraFit software (version determined by fully converting 1, 2, 5 and 10 μmol/L 5.0.13; Erithacus Software Limited *http:// Leu-AMC to products, measuring the final fluores- www.erithacus.com/grafit+). Results are presented as cence and subtracting the background values corre- means ± SE. sponding to the non-hydrolyzed substrate controls. All assays were performed by triplicate. Determination of rPfAM1 inhibition profile Assessment of effect of Zn2+ concentration on The inhibition profile of the rPfAM1 activity was de- rPfAM1 enzymatic activity termined using the inhibitors (Sigma, USA) PMSF (2 2+ mmol/L), E64 (5 μmol/L), pepstatin A (50 μmol/L), The effect of the Zn concentration on the EA of which recognize serine-, cysteine- and aspartic prote- rPfAM1 was evaluated in a range of 0.01-100 μmol/L -10 ases, respectively, in addition to amastatin (10 μmol/ ZnCl2 (Sigma, USA), at pH 7.2 using 1.7 x 10 mol/L L, metallo-APs inhibitor), 1,10-phenanthroline (1 rPfAM1 and 34.7 μmol/L Leu-AMC. The remaining mmol/L, metallo-proteases inhibitor) and bestatin (20 experimental conditions were equal to those above μmol/L, inhibitor of neutral and basic metallo-APs; described. Residual activity was defined as the ratio Bachem, Sweden). The enzyme-inhibitor mixtures, between the reaction rates in the presence and in the -10 2+ containing 1.7 x 10 mol/L rPfAM1, were incubated absence of exogenous Zn . for 30 min at 37°C and pH 7.2 before the addition of Determination of pH optimum for rPfAM1 101.6 μmol/L Ala-AMC (~1 KM; table 1). The remaining The effect of pH was evaluated from pH 4.0 to 11.0, experimental conditions were equal to those above using universal buffer, 1.7 x 10-10 mol/L rPfAM1, and described. Control assays were prepared by pre- incubating the enzyme with the same volume of the 406.3 μmol/L Ala-AMC (~2 KM for nPfAM1; Allary et solvent used to dissolve each inhibitor. Residual activ- al., 2002), 347.1 μmol/L Leu-AMC (~6 KM for nPfAM1), ity was defined as the ratio between the reaction rate 302 μmol/L Arg-AMC (~3 KM for nPfAM1) or 326.7 in the presence of the inhibitor and the rate of the μmol/L Met-AMC (KM not determined for nPfAM1). The experimental conditions were equal to those control. above described. Relative activity was defined as the Statistical analyses ratio between the reaction rate at a given pH and the Normal distribution and homogeneity of variances for maximal rate measured. the data were verified using the Statistica software Determination of kinetic parameters of rPfAM1 (version 8.0; StatSoft Inc. *http://www.statsoft.com+) The kinetic parameters of rPfAM1 were determined at with default parameters. The same software was used for means comparison by the Dunnett test.

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 44 JORGE O. GONZÁLEZ-BACERIO ET AL.

RESULTADS We determined the main kinetic characteristics of rPfAM1, previously obtained by our group (González- Bacerio et al., 2014), by using aminoacyl-AMC fluorogenic substrates. This is essential to further correlate the results of inhibition of the recombinant enzyme with the inhibition of nPfAM1 in the context of intact live parasites (this last assay requires a fluorogenic substrate). Since some unknown amount of Zn2+ may be lost from the active site during purification of rPfAM1 by IMAC (González-Bacerio et al., 2014), we 2+ 2+ Figure 1. Effect of Zn concentration on aminopeptidase activity also assessed the effect of Zn concentration on the of rPfAM1. Assays were performed with 34.7 μmol/L Leu-AMC at EA of the purified enzyme. No augment of activity was pH 7.2. Activity in the absence of exogenous Zn2+ was considered registered in the range 10 nmol/L - 1 μmol/L ZnCl2 as 100%. Values are represented as means (n = 3). Asterisks (fig. 1), but inhibition at concentrations higher than 1 represent significant differences by the Dunnett test (p < 0.05) with μmol/L was obtained, indicating that addition of Zn2+ activity in the absence of Zn2+. to the enzymatic preparation is unnecessary for the Figura 1. Efecto de la concentración de Zn2+ sobre la actividad activity assays. As expected, the optimum pH for the aminopeptidasa de la PfAM1r. Los ensayos se realizaron con Leu- AP activity of rPfAM1 toward Ala-, Leu-, Arg- and Met- AMC 34,7 μmol/L a pH 7,2. La actividad en ausencia de Zn2+ exó- AMC is in the 7.2-7.4 range (fig. 2). geno se consideró como 100%. Los valores se representan como las medias (n = 3). Los asteriscos representan diferencias significa- Determination of the kinetic parameters of rPfAM1 tivas de la prueba de Dunnett (p < 0,05) con la actividad en ausen- 2+ showed that the highest kcat values at both pH were cia de Zn . obtained with Ala at the P1 position of the aminoacyl- AMC substrates (table 1). At pH 7.2, the kcat against Met-, Arg- and Leu-AMC were 2.7, 2.1 and 3.4 times lower, respectively, than the kcat for Ala-AMC. Howev- er, at pH 5.2 the values were 8, 4 and 16 times lower, respectively. In contrast, the lowest KM values at both pH were obtained with Met-AMC. Comparing with Met-AMC, the KM values were 1.6, 4.6 and 2.6 times higher, at pH 7.2, and 2, 48 and 1.3 times higher, at pH 5.2, for Arg-, Ala- and Leu-AMC, respectively (table 1).

The KM values increased only 1.5-1.9 times at pH 5.2 compared to those at pH 7.2 (KM for Leu-AMC even decreases slightly), with the exception of Ala-AMC, for Figure 2. Effect of pH on aminopeptidase activity of rPfAM1. As- which a 16-fold augment was observed. In contrast, says were performed with 406.3 μmol/L Ala-AMC (circles), 347.1 catalytic efficiencies were more affected due to the pH μmol/L Leu-AMC (squares), 302 μmol/L Arg-AMC (diamonds) and fall, with diminutions of 16-40 times in kcat, except for 326.7 μmol/L Met-AMC (triangles). Activity was normalized using Ala-AMC, with an 8-fold decrease. In accordance with the maximum activity level (corresponding to pH = 7.2-7.4) as these variations, the specificity constants (kcat/KM) di- 100%. Values are represented as means (n = 3). minished 31-137 times at acid pH. As is shown from Figura 2. Efecto del pH sobre la actividad aminopeptidasa de la the kcat/KM values, the substrate specificity of PfAM1 at PfAM1r. Los ensayos se realizaron con Ala-AMC 406,3 μmol/L pH 7.2 for the four P1 amino acids tested was: Met > (círculos), Leu-AMC 347,1 μmol/L (cuadrados), Arg-AMC 302 Arg > Ala > Leu, which suffered a slight variation at pH μmol/L (rombos) y Met-AMC 326,7 μmol/L (triángulos). La activi- 5.2: (Met = Arg) > Leu > Ala (table 1). On the other dad se normalizó utilizando el nivel máximo de actividad hand, rPfAM1 was not sensitive to PMSF, E64 and pep- (correspondiente a pH = 7,2-7,4) como 100%. Los valores se statin A, but it was inhibited by amastatin, 1,10- representan como las medias (n = 3). phenanthroline and bestatin (fig. 3).

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 45

JORGE O. GONZÁLEZ-BACERIO ET AL.

)

L×

×s

6 6

cat

0.0046

(x10

1.3 ± 0.1 ± 1.3

mol

: recombinant

0.04 ± 0.003 0.04

1085

)

AMC 195

cat

1.5 1085

79 ± 79 6

2 ±2 0.1

Leu

195

mol/L)

329.9

. (2002). rPfAM1

47 ± 47 4 ± 60 7

(

et al

1085

)

195 1

M Figure 3. Inhibition profile of the aminopeptidase activity of

L×

×s

6 6

1 -

ND rPfAM1. Assays were performed with 101.6 μmol/L Ala-AMC (~1

cat

a la PfAM1n y la PfAM1r y la PfAM1n la a

k 0.0023

(x10 2.5 ± 0.2 ± 2.5

mol KM) at pH 7.2. Tested inhibitors were: 2 μmol/L PMSF, 5 μmol/L 0.02 ± 0.01 0.02

E64, 50 μmol/L pepstatin A, 10 μmol/L amastatin, 1 mmol/L 1,10-

phenanthroline and 20 μmol/L bestatin. Residual activity is ex-

)

AMC 1). nPfAM1: PfAM1 nativa caracterizada por Allary et al. (2002).

characterized by Allary

- cat

2 pressed as a percentage of the activity of the enzyme incubated

M1 fAM Ala

32 ± 32 16 without inhibitors. Asterisks represent significant differences by the 268 ± 20 268

Dunnett test (p < 0.05) with activity in the absence of inhibitor.

Figura 3. Perfil de inhibición de la actividad aminopeptidasa de la

K 200

mol/L) PfAM1r. Los ensayos se realizaron con Ala-AMC 101,6 μmol/L (~1

888.9

μ 106 ± 9 106

( KM) a pH 7,2. Se evaluaron los siguientes inhibidores: PMSF 2 mmol/ 1 674 ± 824 674 1 L, E64 5 mol/L, pepstatina A 50 mol/L, amastatina 10 mol/L, 1,10

μ μ μ

)

-fenantrolina 1 mmol/L y bestatina 20 μmol/L. La actividad residual se

L×

×s 6 6

/K expresa como porcentaje de la actividad de la enzima incubada sin

- ND

cat inhibidores. Los valores se representan como las medias (n = 3). Los

k 0.0015

(x10

AMC, en comparación con los informados par informados los con comparación en AMC,

3.5 ± 0.2 ± 3.5

mol - 0.1 ± 0.01 ± 0.1 asteriscos representan diferencias significativas de la prueba de

Dunnett (p < 0,05) con la actividad en ausencia de inhibidor.

AMC

)

cat

ND 1.1

(s Arg

DISCUSSION

8 ±8 0.8 129 ± 6 129

The kinetic characteristics determined in this work for

M rPfAM1 are crucial in order to confirm that this en-

100

mol/L) 717.4

37 ± 37 4 zyme is similar to nPfAM1, in terms of their kinetic

70 ± 70 16 (

AMC fluorogenic substrates, compared to those reported for nPfAM1 and rPfAM1 and nPfAM1 for reported those to compared substrates, fluorogenic AMC behaviour toward aminoacyl-AMC conjugates. Such

)

condition is necessary to identify potential inhibitors

L×

×s

6 6 /K

1 of nPfAM1 using rPfAM1 as a target.

ND ND

cat . (2009). NotND: determined

k 2+

(x10 4.3 ± 0.4 ± 4.3

mol First, we detected that rPfAM1 is inhibited by Zn at

0.1 ± 0.007 ± 0.1

et al et

concentrations > 1 μmol/L (figure 1), which is con-

)

AMC -

cat sistent with our previous characterization of this en-

ND ND

(s 99 ± 99 9 4 ±4 0.2 zyme using chromogenic substrates (González-Bacerio Met 2+

et al., 2014). The inhibition of metallo-APs at high Zn

concentrations has been reported by various authors.

ND ND

mol/L) 23 ± 23 5

35 ± 35 5 For example, a neutral AP from the larvae of the silk- μ

( worm Bombyx mori is inhibited at more than 100

μmol/L Zn2+ (Hua et al., 1998). Moreover, the near-

5.2 7.2 7.4 7.5

: PfAM1: recombinante obtenida y caracterizada por McGowan et ND: (2009). al. no determinado neutral pH optimum here shown for the AP activity of

1085

Kinetic parameters of rPfAM1 toward aminoacyl toward rPfAM1 of parameters Kinetic

rPfAM1 (figure 2) was also previously observed when

195 1085

- using the chromogenic substrate Leu-aminoacyl-p-

195

rPfAM1 rPfAM1

nPfAM1

Enzyme

Table 1. Table aminoacil fluorogénicos sustratos a frente PfAM1r la de cinéticos Parámetros 1. Tabla PfAM1 obtained and characterized by McGowan Los valores se representan como las medias (n = 3) e incluyen losrPfAM1 errores estándares (± ES) en este estudio con la PfAM1r (rP Values are represented as means (n = 3) and include the standard errors (± SE) in this study using rPfAM1. nPfAM1: native PfA nitroanilide (pNA) (González-Bacerio et al., 2014), and is in agreement with previous reports for nPfAM1 and

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 46 JORGE O. GONZÁLEZ-BACERIO ET AL. the recombinant variant obtained in 2009: rPfAM1195-1085 negative effect on catalysis and a substantial loss of (Allary et al., 2002; McGowan et al., 2009). These re- affinity for P1 residues with small side chains, such as sults are compatible with the cytosolic localization alanine. Since the specificity constants remain at high previously proposed for the endogenous AP in P. falci- values at pH 5.2 (>104 L×mol-1×s-1; table 1), our kinetic parum (Kolakovich et al., 1997; Gavigan et al., 2001; results do not allow discarding a possible localization Allary et al., 2002; Azimzadeh et al., 2010). of nPfAM1 in the parasite acidic food vacuole, as was previously suggested (Dalal and Klemba, 2007; Harbut The KM values obtained in this work toward Arg-, Leu- and Ala-AMC substrates, at near-neutral pH, are clos- et al., 2011; Ragheb et al., 2011). In this sense, high er to those reported by Allary et al. (2002) for nPfAM1 KM values at acid pH toward peptides containing ala- (100, 60 and 200 μmol/L, respectively) than those nine at P1 position (table 1) would be compensated obtained with rPfAM1195-1085 (717, 330 and 889 μmol/L, by high concentrations of such peptide substrates in respectively; McGowan et al., 2009) (table 1). In this the P. falciparum digestive vacuole (alanine consti- sense, our recombinant variant of PfAM1 is similar to tutes 12.46 % amino acids in human hemoglobin; nPfAM1. On the other hand, neither Met-AMC nor McGowan et al., 2009), as was previously reasoned acidic conditions in the kinetic assays were tested in (Ragheb et al., 2011). those previous works, being this report the first one The obtainment of the highest kcat values with the Ala- about the kinetic parameters of PfAM1 at acid pH AMC substrate (table 1) is consistent with the alterna- toward aminoacyl-AMC substrates. Similarly, kcat values tive denomination of this enzyme as alanyl AP have not been reported for the native enzyme toward (Skinner-Adams et al., 2010). Variation of the rPfAM1 these substrates. Hence, here we performed a deeper kinetic parameters at near-neutral pH in function of kinetic characterization of the enzyme. the P1-residue identity (table 1) is similar to that observed with rPfAM1192-1085 toward aminoacyl-Ala natu- The kcat values here obtained at pH 7.2 are 53-134 times higher than those reported for rPfAM1195-1085 ral dipeptide substrates (Dalal et al., 2012). In the (McGowan et al., 2009) (table 1). The discrepancies in present work, the kinetic parameters varied as follows (in descending order): KM: Ala > Leu > Arg > Met; kcat: KM and kcat between the recombinant variant of PfAM1 obtained by our group (González-Bacerio et Ala > Arg > Met > Leu; and kcat/KM: Met > Arg > Ala > al., 2014) and the form produced by McGowan et al. Leu. In the reference study, the order is the same for (2009): rPfAM1195-1085, may be a consequence of as- KM, whereas slightly different for kcat: Ala > (Arg = Met sessing the kinetic parameters of the latter at pH 7.5, = Leu) and kcat/KM: Met > Arg > Leu > Ala (Dalal et al., which is different to the pH optimum reported for this 2012). These coincidences confirm the relevance of enzyme (pH 7.0; McGowan et al., 2009). The dispari- the P1 site for the PfAM1 substrate specificity (Allary et al., 2002; McGowan et al., 2009; Dalal et al., 2012; ties in kcat may be also caused by dissimilar percent- ages of active molecules in both enzyme preparations Poreba et al., 2012; González-Bacerio et al., 2014) and endorses using these aminoacyl-AMC substrates to (we assumed 100 % activity to calculate kcat). Since a assess the inhibition of the native and recombinant possible cause of overestimating the kcat is the under- estimation of the enzyme concentration, we assessed enzymes. The substrate specificity exhibited by the rPfAM1 concentration by two distinct methods rPfAM1 against P1 natural amino acids (table 1) also (Bradford method and absorbance at λ280 nm), which matches that of EtAPN1, the PfAM1 ortholog from the allowed the accurate quantification of the protein. avian apicomplexan parasite Eimeria tenella (Gras et al., 2014). Since the effect of pH upon the rPfAM1 kinetic parameters is common for substrates with basic (arginine) The inhibition profile of rPfAM1 (fig. 3) is similar to or non-ionizable (methionine, leucine) P1 side chains that of nPfAM1 and typical of neutral and basic metal- (table 1), we reject the pH-dependent variation of lo-APs (Allary et al., 2002). All results derived from the electrostatic interactions between the enzyme active kinetic characterization of rPfAM1 using aminoacyl- site and the P1 side chain as the cause of the catalytic AMC fluorogenic substrates, like those previously ob- efficiency impairment at pH 5.2. Our data could be tained with aminoacyl-pNA chromogenic substrates influenced by a conformational change, and/or an un- (González-Bacerio et al., 2014), indicate that our re- favourable protonation state of catalytic residues, in combinant protein has a kinetic behavior which is very the rPfAM1 active site at acid pH, resulting in a general similar to that of the native enzyme. Consequently,

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 47 JORGE O. GONZÁLEZ-BACERIO ET AL.

rPfAM1 can be used as a model of nPfAM1 to search S1, S1’ and S2’ subsites drives efficient catalysis of peptide for inhibitors. In addition, the kinetic parameters ob- bond hydrolysis by the M1-family aminopeptidase from Plasmo- tained here with the fluorogenic peptides, at near- dium falciparum. Mol. Biochem. Parasitol. 183: 70-77. neutral and acid pH (the latter for the first time), al- Florent, I., Z. Derhy, M. Allary, M. Monsigny et al. (1998) A Plasmo- low the use of these substrates for the evaluation of dium falciparum aminopeptidase gene belonging to the M1 family of zinc-metallopeptidases is expressed in erythrocytic nPfAM1 activity and its inhibition directly in parasites. stages. Mol. Biochem. Parasitol. 97: 149-160.

Gardiner, D.L., K.R. Trenholme, T.S. Skinner-Adams, C.M. Stack et ACKNOWLEDGMENTS al. (2006) Overexpression of leucyl aminopeptidase in Plasmo- dium falciparum parasites: Target for the antimalarial activity of This work was supported by grants F/4730-1 and F/4730-2F bestatin. J. Biol. Chem. 281(3): 1741-1745. from the International Foundation for Science (Sweden) to JGB, grant 169/12 from the Coordenação de Aperfeiçoa- Gavigan, C.S., J.P. Dalton and A. Bell (2001) The role of aminopeptidases in haemoglobin degradation in Plasmodium falci- mento de Pessoal de Nível Superior (CAPES, Brazil) - Minis- parum-infected erythrocytes. Mol. Biochem. Parasitol. 117: 37- terio de Educación Superior (MES, Cuba) to MLG and MAR, 48. and Red del Programa Iberoamericano de Ciencia y Tecnología para el Desarrollo (CYTED-PROMAL, 210RT0398) Gras, S., A. Byzia, F.B. Gilbert, S. McGowan et al. (2014) Ami- “Proteómica y Quimiogenómica de Inhibidores de Proteasas nopeptidase N1 (EtAPN1), an M1 metalloprotease of the de Origen Natural con Potencial Terapéutico en Malaria” to apicomplexan parasite Eimeria tenella, participates in parasite development. Eukaryotic Cell 13(7): 884-895. doi: 10.1128/ MÁC. JGB received a fellowship from CAPES. MLG was fund- EC.00062-14. http://ec.asm.org/ ed by FAPESP (2009/54598-9). AKC was funded by CNPq (482400/2013-7). González-Bacerio, J., J. Osuna, A. Ponce, R. Fando et al. (2014) High-level expression in Escherichia coli, purification and kinetic characterization of Plasmodium falciparum M1-aminopeptidase. CITED LITERATURE Protein Expr. Purif. 104: 103-114. Addlagatta, A., L. Gay and B.W. Matthews (2006) Structure of Harbut, M.B., G. Velmourougane, S. Dalal, G. Reissa et al. (2011) aminopeptidase N from Escherichia coli suggests a compart- Bestatin-based chemical biology strategy reveals distinct roles mentalized, gated active site. Proc. Natl. Acad. Sci. USA 103 for malaria M1- and M17-family aminopeptidases. Proc. Natl. (36): 13339-13344. Acad. Sci. USA 108(34): E526-E534. Allary, M., J. Schrével and I. Florent (2002) Properties, stage- Hua, G., K. Tsukamoto, R. Taguchi, M. Tomita et al. (1998) Char- dependent expression and localization of Plasmodium falcipa- acterization of aminopeptidase N from the brush border mem- rum M1 family zinc-aminopeptidase. Parasitology 125: 1-10. brane of the larvae midgut of silkworm, Bombyx mori as a zinc enzyme. Biochim. Biophys. Acta 1383(2): 301-310. Azimzadeh, O., C. Sow, M. Gèze, J. Nyalwidhe et al. (2010) Plas- modium falciparum PfA-M1 aminopeptidase is trafficked via the Kannan Sivaraman, K., A. Paiardini, M. Sieńczyk, C. Ruggeri et al. parasitophorous vacuole and marginally delivered to the food (2013) Synthesis and structure-activity relationships of phos- vacuole. Malaria J. 9: 189-204. doi:10.1186/1475-2875-9-189. phonic arginine mimetics as inhibitors of the M1 and M17 ami- http://www.malariajournal.com/content/9/1/189 nopeptidases from Plasmodium falciparum. J. Med. Chem. 56: 5213-5217. Barret, A.J., N.M. Rawlings and J.F. Woessner (2003) Introduction: metallopeptidases and their clans. In: Barret, A.J., N.M. Rawl- Kolakovich, K., I. Gluzman, K. Duffin and D. Goldberg (1997) Gen- ings and J.F. Woessner (Eds.), Handbook of proteolytic en- eration of hemoglobin peptides in the acidic digestive vacuole of zymes, 2nd Edition. Academic Press, London. pp. 334-542. Plasmodium falciparum implicates peptide transport in amino acid production. Mol. Biochem. Parasitol. 87: 123-135. Cadavid-Restrepo, G., T.S. Gastardelo, E. Faudry, H. de Almeida et al. (2011) The major leucyl aminopeptidase of Trypanosoma Kuhn, Y., P. Rohrbach and M. Lanzer (2007) Quantitative pH cruzi (LAPTc) assembles into a homohexamer and belongs to measurements in Plasmodium falciparum-infected erythrocytes the M17 family of metallopeptidases. BMC Biochem. 12: 46-59. using pHluorin. Cell Microbiol. 9(4): 1004-1013. doi:10.1186/1471-2091-12-46. http://www.biomedcentral.com/1471 Liu, J., E.S. Istvan, I.Y. Gluzman, J. Gross et al. (2006) Plasmodi- -2091/12/46 um falciparum ensures its amino acid supply with multiple acqui- Copeland, R.A. (2000) Kinetics of single-substrate enzyme reac- sition pathways and redundant proteolytic enzyme systems. tions. In: Copeland, R.A., Enzymes: A practical introduction to Proc. Natl. Acad. Sci. USA 103(23): 8840-8845. structure, mechanism, and data analysis. 2nd Edition. Wiley- McGowan, S., C.J. Porter, J. Lowther, C.M. Stack et al. (2009) VCH, Inc., New York. pp. 109-145. Structural basis for the inhibition of the essential Plasmodium Dalal, S. and M. Klemba (2007) Roles for two aminopeptidases in falciparum M1 neutral aminopeptidase. Proc. Natl. Acad. Sci. vacuolar hemoglobin catabolism in Plasmodium falciparum. J. USA 106(8): 2537-2542. Biol. Chem. 282(49): 35978-35987. Dalal, S., D.R.T. Ragheb and M. Klemba (2012) Engagement of the REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 . KINETICS OF RPFAM1 WITH FLUOROGENIC SUBSTRATES 48 JORGE O. GONZÁLEZ-BACERIO ET AL.

Morty, R.E. and J. Morehead (2002) Cloning and characterization bioavailable inhibitor of murine malaria. Antimicrob. Agents of a leucyl aminopeptidase from three pathogenic Leishmania Chemother. 56(6): 3244-3249. species. J. Biol. Chem. 277(29): 26057-26065. Skinner-Adams, T.S., C.M. Stack, K.R. Trenholme, C.L. Brown et Poreba, M., S. McGowan, T.S. Skinner-Adams, K.R. Trenholme et al. (2010) Plasmodium falciparum neutral aminopeptidases: New al. (2012) Fingerprinting the substrate specificity of M1 and M17 targets for anti-malarials. Trends Biochem. Sci. 35: 53-61. aminopeptidases of human malaria, Plasmodium falciparum. Velmourougane, G., M.B. Harbut, S. Dalal, S. McGowan et al. PLoS ONE 7(2): e31938 doi:10.1371/journal.pone.0031938. (2011) Synthesis of new (-)-bestatin-based inhibitor libraries http://www.plosone.org reveals a novel binding mode in the S1 pocket of the essential Ragheb, D., S. Dalal, K.M. Bompiani, W.K. Ray et al. (2011) Distri- malaria M1 metalloaminopeptidase. J. Med. Chem. 54: 1655- bution and biochemical properties of an M1-family aminopepti- 1666. dase in Plasmodium falciparum indicate a role in vacuolar hemo- Yeh, I. and R.B. Altman (2006) Drug targets for Plasmodium falci- globin catabolism. J. Biol. Chem. 286(31): 27255-27265. parum: A post-genomic review/survey. Min.-Rev. Med. Chem. 6: Skinner-Adams, T.S., C.L. Peatey, K. Anderson, K.R. Trenholme et 177-202. al. (2012) The aminopeptidase inhibitor CHR-2863 is an orally

  

Editor para correspondencia: Dra. Yamile González González

REVISTA CUBANA DE CIENCIAS BIOLÓGICAS o RNPS: 2362 • ISSN: 2307-695X • VOL. 4 • N. 2 • MAYO— SEPTIEMBRE • 2015 • pp. 40 - 48 .