Plant Physiology and Biochemistry 98 (2016) 162e170

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry

journal homepage: www.elsevier.com/locate/plaphy

Research article

Characterization of the legumains encoded by the genome of

Theobroma cacao L
Juliano Oliveira Santana a, Las Freire a, Aurizangela Oliveira de Sousa a,
Virgnia Lcia Fontes Soares a, Karina Peres Gramacho b, Carlos Priminho Pirovani a, *
a
Biotechnology and Genetics Center, State University of Santa Cruz, 45662-900 Ilh

eus, BA, Brazil
b
Cocoa Research Center, CEPLAC/CEPEC, P.O. Box 7, 45600-970 Itabuna, BA, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Legumains are cysteine proteases related to plant development, protein degradation, programmed cell
Received 20 July 2015 death, and defense against pathogens. In this study, we have identied and characterized three legu-
Received in revised form mains encoded by Theobroma cacao genome through in silico analyses, three-dimensional modeling,
29 October 2015
genetic expression pattern in different tissues and as a response to the inoculation of Moniliophthora
Accepted 16 November 2015
perniciosa fungus. The three proteins were named TcLEG3, TcLEG6, and TcLEG9. Histidine and cysteine
Available online 2 December 2015
residue which are part of the catalytic site were conserved among the proteins, and they remained
parallel in the loop region in the 3D modeling. Three-dimensional modeling showed that the propeptide,
Keywords:
Cysteine-proteinase
which is located in the terminal C region of legumains blocks the catalytic cleft. Comparing dendrogram
Cocoa data with the relative expression analysis, indicated that TcLEG3 is related to the seed legumain group,
Modeling TcLEG6 is related with the group of embryogenesis activities, and protein TcLEG9, with processes
qPCR regarding the vegetative group. Furthermore, the expression analyses proposes a signicant role for the
Gene expression three legumains during the development of Theobroma cacao and in its interaction with M. perniciosa.
Witches' broom 2015 Universidade Estadual de Santa Cruz, CNPJ: 40738999/0001-95. Published by Elsevier Masson
SAS. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

1. Introduction
Legumains are cysteine proteases in the C13 family (EC
3.4.22.34) and are found in plant tissues (Santos-Silva et al., 2012),
parasites (Klinkert et al., 1989), and mammals (Chan et al., 2009).
These proteins are also called Vacuolar Processing Enzymes (VPEs),
due to their ability to recognize and cleave asparagine (Asn) or
aspartic acid (Asp) residue in polypeptides (Hara-Nishimura et al.,
1995; Mntz and Shutov, 2002).
VPEs are synthesized as inactive prolegumains, and they are
transferred to the vacuoles or to the cell wall. Once they reach those
compartments under acid pH conditions, prolegumains undergo
activation through self-cleavage in the propeptides that are located
in the amino and carboxy-terminal. Then they become active
legumains in order to play their biological role in plant tissues (Dall
and Brandstetter, 2012; Li et al., 2003). In plants, those proteins
were initially isolated from mature castor beans (Ricinus communis)
and from soybean cotyledons (Glycine max) (Hara-Nishimura et al.,
* Corresponding author.
E-mail address: pirovani@uesc.br (C.P. Pirovani).
1991; Scott et al., 1992), and they have conserved the Cys and His
catalytic residues which are preceded by a block of four hydro-
phobic amino acids (Chen et al., 1998).
In plants, VPEs are involved in several physiological processes,
and they are classied in three groups, according to the tissues
where they are expressed the most. VPEs in from the vegetative
group are involved in programmed cell death (PCD) regulation
(Hatsugai et al., 2004), senescence (Donnison et al., 2007), and
pathogen response (Rojo et al., 2004). VPEs from seeds are related
to the processing and mobilizing of reserve proteins during seed
germination (Nakaune et al., 2005; Radchuk et al., 2011). VPEs in
the embryogenesis group seem to play a role in the formation
process of the external tegument of seeds (Nakaune et al., 2005).
In Arabidopsis thaliana, three legumain genes were found, and
they are referred to as aVPE, bVPE, and gVPE (Kinoshita et al., 1995a,
1995b). Later, on the same plant, a fourth protein was reported
through the observation of its expression in young developing
seeds, called dVPE (Gruis et al., 2002). Studies regarding the
expression of VPE genes showed that aVPE and gVPE are inserted in
the vegetative group, showing higher expression in roots, senes-
cent leaves, and in PCD tissues (Kinoshita et al., 1999). bVPE and
dVPE were cataloged in seed VPEs; the rst of which was mainly

http://dx.doi.org/10.1016/j.plaphy.2015.11.010
0981-9428/ 2015 Universidade Estadual de Santa Cruz, CNPJ: 40738999/0001-95. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170
expressed in dry seeds, embryonic axis, and cotyledon, and dVPE
was expressed in developing seeds (Hara-Nishimura et al., 2005;
Nakaune et al., 2005).
Cocoa tree (Theobroma cacao L.) is highly important economi-
cally as source of almonds, raw material for chocolate and sweets,
motivating several biotechnological studies which led to the com-
plete sequencing of its genome (Argout et al., 2011). Fungal diseases
are responsible for signicant losses in the production of almonds,
especially witches' broom, caused by the basidiomycete fungus
Moniliophthora perniciosa (Aime and Phillips-Mora, 2005). This
fungus is hemibiotrophic and its life cycle is divided into two
phases, a biotrophic and a necrotrophic phase, causing histological,
physiological, morphological alterations, and death of infected tis-
sues in plants (Ceita et al., 2007; Orchard et al., 1994; Scarpari et al.,
2005).
In this study, we have identied and characterized the three
VPEs (TcLEG3, TcLEG6 e TcLEG9) that are coded by the Theobroma
cacao genome through in silico analyses, three-dimensional
modeling, genetic expression patterns in different tissues, and
propose a signicant role of VPEs during plant development and in
response against the M. perniciosa pathogen, which causes witch's
broom in the cocoa tree.
2. Materials and methods
2.1. Primary in silico analysis
The primary amino acid sequences that were predicted for the
VPEs which are coded by the Theobroma cacao genome were ob-
tained from the CocoaGenDB database (http://cocoagendb.cirad.fr/
). The three identied VPEs were submitted to different bioinfor-
matics analyses: sequence alignment e ClustalW (http://www.ebi.
ac.uk/Tools/msa/clustalw2/); signal peptide prediction e SignalP
4.1 Sever (http://www.cbs.dtu.dk/services/SignalP/) (Bendtsen
et al., 2004); tracking of functional domains e Pfam (http://pfam.
xfam.org/search/sequence), and analysis on MEROPS (http://
merops.sanger.ac.uk/).
The phylogenetic dendrogram was built with the help of MEGA
6 software (Kumar et al., 2004), using Neighbor-Joining method and
the condence of the branching order was veried by making 1000
bootstrap replicates. The dendrogram was built with 31 sequences
of homologous proteins that were distributed in eleven plant spe-
cies, as follows: Nicotiana tabacum (tobacco), Solanum lycopersicum
(tomato), Oryza sativa (rice), Vitis vinifera (grape), R. communis
(castor oil plant), Triticum aestivum (wheat), Hordeum vulgare
(barley), Cucumis sativus (cucumber), Sesamum indicum (sesame),
A. thaliana, and Theobroma cacao (cocoa). The searches for homol-
ogous sequences were conducted in NCBI database (National Cen-
ter for Biotechnology Information - http://www.ncbi.nlm.nih.gov/
BLAST), and the sequences identied with their access numbers.
One of the cystatins (TcCys1) of cocoa, which is a competitive in-
hibitor of cysteine protease, described by Pirovani et al. (2010), was
used as an outlier.
2.2. Three-dimensional modeling
Homology modeling was used to obtain the active and inactive
three-dimensional structures from the three T. cacao VPEs. The
searches for template structures as resolved through x-ray crys-
tallography or nuclear magnetic resonance (NMR) were carried out
by the PDB database (http://www.rcsb.org/pdb/home/home.do).
Modeller 9.12 software (Eswar et al., 2007) was used in the
three-dimensional modeling of the structures. The scripts for pro-
cessing were generated by the Notepad software (Version 6.5.1),
with one .ali extension le (Dynamics AX Label Index File) and
163
ve .py extension les (Python). The .ali contains information
regarding the target protein sequence, whereas the scripts in the
.py format have modeling commands and three-dimensional
structures of template proteins selected in a .pdb format. The
generated protein structures were rened and validated by soft-
wares (Charmm, Anolea, and Procheck) which analyze possible
structural errors in specic regions and stereochemical parameters
(Johnson et al., 1994) (Figs. S1eS3).
2.3. Tissues and fruits analyzed
Plants of the CCN51 variety were used to analyze the genetic
expression in different tissues (root, tegument, seed, leaf and stalk).
They were grown under natural lighting, drip irrigation, and room
temperature. The seed cotyledons of the variety Parazinho were
germinated in a growth chamber at a 27 2  C temperature, with
photoperiod of 16 h of light and 8 h of darkness, with
36 mmol m2 s1 irradiance. Samples were collected at ten different
germination stages, as follows: zero hour (quiescent seed), 24 h,
48 h, 72 h, day 4, day 6, day 8, day 12, day 20, and day 30, after the
germination. Fruits of the Parazinho variety were collected in four
different development stages in order to analyze the expression of
VPEs in the seed development phase (Fig. S4). Fruits were measured
with the aid of a caliper ruler, and weighed through precision
scales. After collection tissues were immediately frozen in liquid
nitrogen, lyophilized, and stored at 80  C, until DNA extraction.
2.4. Germination conditions of infected tissues
Two hundred seeds of the CCN51 variety were planted and kept
in the greenhouse under natural light, spray irrigation, and room
temperature conditions. After 20 growth days, seedlings were
taken to a moist chamber, making up for a total 100 plants infected
by M. perniciosa and 100 control plants inoculated with 0.3% agar-
water. Inoculation was conducted through the spraying of 5.0  105
basidiospores/mL of M. perniciosa in a 0.3% agar-water suspension
(Surujdeo-Maharaj et al., 2003). The seedlings, after the inocula-
tion, were stored at 23  C for 48 h, under relative humidity >97%,
and controlled by an automated nebulization system. Afterwards,
seedlings returned to the greenhouse. Biological sample collections
(meristems) were conducted on different days, as follows: 0 h (right
after the inoculation), 24 h, 48 h, 72 h, 15 days and 45 days after the
inoculation. The collected samples were immersed in liquid nitro-
gen and lyophilized prior to extraction.
2.5. Criteria for primer design
The oligonucleotides that were used for the qPCR (Table S1)
were designed with the v.3.0 Primer Express software (Applied
Biosystems). In order to generate the design of primers for the
VPEs (TcLEG3, TcLEG6, and TcLEG9), the following criteria were
considered: CG percentage, size between 18 and 25 bp, melting
temperature Tm (58e60  C), and amplicon size (50e150 pb). Actin
(ACT), malate dehydrogenase (MDH), and glyceraldehyde 3-
phosphate dehydrogenase (GAPDH) genes were selected as
endogenous controls, since they presented homogeneous expres-
sion in the analyzed tissues (Pinheiro et al., 2011).
The relative expression levels of genes TcLeg3, TcLeg6, and
TcLeg9 were calculated according to the delta Ct (DCt) method, and
analyzed in two biological replicates (different tissues), ve bio-
logical replicates (developing seed), ten biological replicates
(infected tissues). All tissues were analyzed in experimental quin-
tuplicates. After Ct values were obtained, the average value of the Ct
of each target gene was normalized with the Ct value from the
endogenous control that was established by the NormFinder
164 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170
software (Claus et al., 2004), which calculated the best stability
value among the genes that were used as endogenous controls. The
relative expression values of the target genes in plant tissues were
expressed as powers of two (2DCt). In order to analyze the genetic
expression in T. cacao plant tissues, the 2DDCt (Livak and
Schmittgen, 2001) method was used (Table S2). The statistical an-
alyses of data were conducted by the BioEstat 5.0 software,
applying Dunnett's statistical test with a 95% condence interval.
2.6. RNA extraction, cDNA synthesis, and RT-qPCR
The tissues of T. cacao were sprayed, and 100 mg used to extract
RNA with the help of the Zymo Research kit, according to the
manufacturer's recommendations. The RNA of infected tissues
(meristem) and developing seeds were puried with the RNAqu-
eous reagent (Ambion), and then submitted to ultrasonication (3
pulses of 2 s each, 70% range, with a 10 s interval) due to the excess
mucilage. Afterwards, the RNA was treated with DNAse I (Thermo
Scientic). RNA integrity was analyzed in 1% agarose gel stained
with GelRed, and using the NanoDrop 2000 spectrophotometer
(Thermo Scientic). cDNA was synthesized with RvertAid H Minus
First Strand cDNA Synthesis Kit (Thermo Scientic) according to the
manufacturer's recommendations and quantied and diluted to
10 pmol/mL.
For qPCR, the Maxima Sybr Green/Rox kit (Thermo Scienti-
c), was used. Each reaction was performed in a 22 mL volume,
containing 10 mL cDNA, 11 mL Sybr Green, and 0.5 mL of each
oligonucleotide, which were diluted at 10 pmol/mL. The amplica-
tion was conducted in an Mx 3005P device (Agilent Technologies),
according to the following conditions: UDG incubation for 2 min at
50  C, polymerase activation for 10 min at 95  C, followed by 40
15 s cycles at 95  C and 1 min at 60  C. The dissociation curve that
was obtained in the reaction was used to evaluate the condence of
the amplied region, and to observe the presence of unspecic
amplication (Figs. S5eS7).
3. Results
3.1. In silico analysis of VPE sequences
In CocoaGenDB, the T. cacao genome database, three VPE
cysteine proteases were found in loci Tc03_g025550, Tc06_g006030,
and Tc09_g004580 (Fig. S8). Those three proteins were named
TcLEG3, TcLEG6, and TcLEG9. Pfam software revealed that the three
identied sequences in the blast correspond to peptidases from the
C13 class; that is, from the VPE family.
The alignment conducted among cocoa tree VPEs and a legu-
main (gVPE) from A. thaliana, as described by Kinoshita et al.
(1995b), found several blocks of conserved amino acids and
revealed a high degree of identity (61.4%, 57.8%, and 77.1% in TcLEG3,
TcLEG6, and TcLEG9, respectively). The signal peptide was observed
in the amino-terminal regions of the three analyzed legumains
(Fig. 1). In TcLEG3, the signal peptide cleavage sites were between
A27 and A28. In TcLEG6, between the S21 and the E22 residue. In
contrast, for TcLEG9, the cleavage site was between the A21 and the
G22 residue. TcLEG9 presented the motif LPS (L32, P33, and S34) in
the N-terminal pro-peptide. Two conserved Asp residues indicate
the N-terminal and C-terminal propeptide cleavage sites in two
active proteins - D51 and D422 for TcLEG3, and D46 and D420 for
TcLEG9. However, in TcLEG6, the cleavage residues were located in
N49 and D426 cleavage sites. The probable His catalytic sites (H175,
H173, and H170 in TcLEG3, TcLEG6, and TcLEG9, respectively) Cys
catalytic sites (C217, C215, and C212 in TcLEG3, TcLEG6, and TcLEG9,
respectively), remained conserved among the analyzed proteins,
and both residues are preceded by a block of four hydrophobic
amino acid residues.
In the dendrogram, the VPEs of Theobroma cacao have been
separated into three groups: a vegetative type (a/gVPE), a seed type
(bVPE) and an embryogenesis type (dVPE). The TcLEG3 in group
bVPE, TcLEG6 in group dVPE and TcLEG9 in group a/gVPE (Fig. 2).
3.2. Three-dimensional modeling
Three-D models of the three legumains, TcLeg3, TcLeg6, and
TcLeg9 in their active and inactive forms, were obtained with the
three dimensional modeling. The template structure of Homo sa-
piens (4n6o, 4awb, and 4fgu) (Dall and Brandstetter, 2013, Dall et al.,
2015) and Mus musculus (4noj) (Zhao et al., 2014) were the best
models with a degree of identity greater than 46%. Models 4n6o,
4fgu, and 4awb were selected for the construction of three-
dimensional model of TcLEG3 and structures 4n6o, 4fgu, and
4noj, were selected as models for TcLEG6 and TcLEG9. Small dif-
ferences were found in the supercial molecular areas and in the
number of atoms in the structures of the three active VPEs (without
the propeptides). In TcLEG3, the supercial area was 27.157.854 ,
with 2.620 atoms; in TcLEG6, it was 26.952.984 with 2.625 atoms;
and in TcLEG9, an area of 26.825.139 with 2.572 atoms. The
number of segments with conforming b-leaves, a-helices, and loops
were equally distributed among the structures, and they were 8, 9,
and 17, respectively (Fig. 3). The distances between the catalytic
residues of histidine (His) and cysteine (Cys) in the loop regions of
the three VPEs were similar - 5.5 for TcLEG3, 6.3 for TcLEG6 and
5.1 for TcLEG9.
The structure validation process was shown to have a Ram-
achandran plot with 100% of residues in energetically favorable
regions for all three generated VPEs. The three-dimensional struc-
tures of the three inactive VPEs (Fig. 3) showed that the C-terminal
pro-peptide is positioned in the upper portion of the catalytic
domain (His and Cys), blocking the catalytic cleft and preventing
the coupling of the active site substrates or inhibitors.
3.3. Genetic expression
3.3.1. Different tissues
Gene TcLeg3 was found to have an accumulation of transcripts
for all analyzed tissues (Fig. 4a); it was least expressed in the root,
which was used as calibrator (1.00X), and most expressed in the
seed (21.7X) and tegument (21.5X). Still on the same gene, statis-
tically signicant differences are observed in the expression of the
tegument (21.5X), seed (21.7X), leaf (9.9X), and stalk (17.8X), and
those values were higher than the expression in the root (1.0X)
(calibrator).
The TcLeg9 gene showed expression in all organs and tissues
analyzed (Fig. 4b). The leaf was the organ with the highest
expression (25.5X) as compared to the calibrator (root). Statistically
signicant differences were observed in the tegument (4.7X), leaf
(25.6X), and stalk (19.0X) as compared to the expression in the root
(1.00X). Unlike for the TcLeg3 gene, the TcLeg9 gene in the seed
presented less expression (0.7X), with no signicant differences as
compared to the calibrator. Gene TcLeg6 expression was not iden-
tied in any of organs or tissues (seed, tegument, root, stalk, and
leaf) analyzed.
3.3.2. Cotyledon expression during germination
In the cotyledons of germinating seeds, the accumulation of
transcripts of gene TcLeg3 decreased from the initial time (quies-
cent seed) to the thirtieth day (1.00e0.018X) (Fig. 5a), a reduction of
approximately 90-fold. The results point towards statistically sig-
nicant differences from the rst to the thirtieth germination day.
Still on the same tissue, TcLeg9 expression rose from time zero to
 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170 165

Fig. 1. a: Alignment among T. cacao VPEs and a VPE from A. thaliana (NP_195020). Conserved amino acids among VPEs are indicated by an asterisk. Signal peptides are shaded. Box I
indicates the vacuolar motif that was described by Jackson et al. (2007). The arrows indicate the cleavage site of propeptides in N and C-terminal regions. Boxes II and IV indicate the
four hydrophobic residues which precede the catalytic residues that are indicated in box III and V. b: Global identity among amino acid sequences, the three cocoa tree VPEs, and the
gVPE from A. thaliana.

the twentieth day, with an approximate 64-fold increase as
compared to the control (0 time - quiescent seed) (Fig. 5b), un-
dergoing a high reduction of its relative expression, from 64.10X to
20.34X in the thirtieth day. Results also pointed toward statistically
signicant differences from the second to the thirtieth germination
day for this gene, in comparison to the control. Gene TcLeg6 was not
expressed in any of the stages in the germinating cotyledon tissues.
3.3.3. Developing seeds
The TcLeg3 genes TcLeg6 and TcLeg9 showed transcripts in all
stages of seed development (Fig. 6). Gene TcLeg3 did not present
signicant expression variation among the analyzed stages, from
1.0X in the rst stage (calibrator) to 0.8X in second, 0.8X third, and
1.4X in last stage. In contrast, gene TcLeg6 expression increased
from the rst stage of seed development to the third stage
(1.00e8.9X), remaining practically unaltered in the third and fourth
stages (8.9e8.7X). In same tissue, the transcripts of gene TcLeg9 did
not show signicant variation for the rst three stages, from 1.0X in
rst stage to 1.2X in second, 0.7X in third, and nally a signicant
increase of 2.5X in the fourth stage.
166 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170

Fig. 2. Dendrogram of VPEs of ten plant species. Black triangle: cocoa legumains. VPEs were separated into three groups: bVPE (brown), a/gVPE (green), and dVPE (yellow). The
analysis was conducted through MEGA 6 software. Signal peptides were excluded from the sequences. Access numbers on NCBI are within parentheses. (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of this article.)

3.3.4. Infected tissues
The analysis of the relative expression in infected tissues by
M. perniciosa showed low accumulation of gene TcLeg3 transcripts
(Fig. 7a) for all days after inoculation. Signicant variation was only
obtained in the plant initial necrosis period (45 days) as compared
to its calibrator (non-inoculated plant e 45 days) shifting from
0.76X to 0.23X. Gene TcLeg9 (Fig. 7b) in the same tissue presented
signicant expression variation on rst day (1.00X to 2.93X) and on
the forty-fth day after inoculation (1.00X to 1.82X).
Gene TcLeg6 was not expressed in the analyzed days in plants
that were infected by M. perniciosa fungus basidiospores, and
neither in control plants.
4. Discussion
4.1. TcLEG3, TcLEG6, and TcLEG9 are bVPE, dVPE, and a/gVPE,
respectively
Cocoa VPEs are synthesized as prolegumains, with the presence
of signal peptides and amino and carboxy-terminal which depend
on cleavage processing (Fig. 1). They are probably transported to
their destination sites in inactive forms, which can be self-activated
under acidic conditions, as so happens with the castor oil plant
VPEs (R. communis) (Hiraiwa et al., 1997), A. thaliana (Kuroyanagi
et al., 2002), and humans (Chen et al., 2000).
The amino acid sequence of TcLEG9 has the LPS motif (Fig. 1)
located on the N-terminal propeptide, which according to Jackson
et al. (2007) is a probable signal for the transportation of that
enzyme to the lytic vacuole. His and Cys residues remained
conserved among the VPEs. Those residues, according to Hiraiwa
et al. (1997), are part of the catalytic site. The three VPEs also pre-
sent conserved residues of Asp, which indicate the cleavage site of
N and C-terminal propeptides during the activation of the protein.
However, TcLEG6 in the N-terminal regions has the Asn residue
replacing the Asp.
TcLEG3, TcLEG6, and TcLEG9 VPEs were separated in the
dendrogram (Fig. 2) into three distinct groups: bVPE, dVPE, and a/
gVPE, respectively. According to Christoff et al. (2014) the groups
are related to the specicity of biological roles of VPEs during a
plant life cycle. It is important to point out that TcLEG6 is inserted in
the dVPE group, and was the last group to be specied (Nakaune
et al., 2005). Those proteins are of the embryogenesis type and
their expression takes place before the synthesis of storage proteins
(bVPE). That protein is involved in the process in removing the
internal layers that are required for the formation of external
tegument of developing seeds (Yamada et al., 2005).
Homology modeling is a reliable in silico technique for the
prediction of protein structures (Chothia and Lesk, 1986; Marti-
Renom et al., 2000). The three active VPEs modeled revealed that
the catalytic residues are arranged on the surface of the loop region
along folds, which facilitates the interaction between two mole-
cules (Bode and Huber, 1992). The modeling of inactive structures
showed how the propeptide which is located in the C-terminal
region of the VPEs blocks the catalytic cleft (Fig. 3), being positioned
in the upper part of the catalytic domain (His e Cys), preventing its
interaction with the substrate or inhibitor, at the same time, giving
enzyme latency and conformational stability to the protein (Dall
and Brandstetter, 2013; Zhao et al., 2014).
 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170 167

Fig. 3. Three-dimensional structure of T. cacao VPEs. Active VPEs e a: TcLEG3, c: TcLEG6, e: TcLEG9. Inactive VPEs e b: TcLEG3, d: TcLEG6, f: TcLEG9. Blue (a-helices), orange (b-
leaves), green (Loops). Highlighted loops (red) are the (His and Cys) catalytic residues. The propeptide which is located in the terminal C region, in yellow. (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of this article.)

Fig. 4. Expression pattern of TcLeg3 (a) and TcLEG9 (b) genes in different tissues of T. cacao. Values were normalized through GAPDH endogenous enzyme. Error bars represent the
average of two replicates (n 2). The asterisks indicate the signicant values (p < 0.05) as compared to the calibrator (root) according to Dunnett's method.

4.2. Genetic expression
The expression of VPE genes was detected in different levels in
T. cacao tissues. Genes TcLeg3 and TcLeg9 were found to have
detectable expression levels in all analyzed tissues, as well as sig-
nicant alterations in cotyledons during germination. In contrast,
bgene TcLeg6 was only expressed in the stages of seed development.
4.2.1. TcLeg3 is related to specic roles in seeds
The relative expression of TcLeg3 in different plant tissues
(Fig. 4a) was shown to have great accumulation of transcripts,
especially in the tegument and in the seed. In germinating cotyle-
dons (Fig. 5a), the expression pattern decreased from the initial
time to the thirtieth germination day, a reduction of approximately
90 times, which corroborates with the data from Kinoshita et al.
(1999), in which the expression of gene bVPE in Arabidopsis
decreased after the seed germinated, and kept gradually reducing
with seedling growth.
In the tissue that was infected by the M. perniciosa fungus, gene
TcLeg3 presented accumulation of transcripts reduced on the forty-
fth day after the inoculation, when the plant necrosis process be-
gins (Fig. 7a) and the disease changes to the saprophytic phase (Ceita
et al., 2007; Frias et al., 1991), with a 53X reduction in the transcript,
which suggests that this gene is not related to the defense against the
M. perniciosa pathogen, senescence and tissue degradation.
The obtained expression prole suggests that VPE TcLeg3 is
168 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170

Fig. 5. Expression pattern of TcLeg3 (a) and TcLEG9 (b) genes in germinating cotyledons of T. cacao. Values were normalized through MDH endogenous enzyme. Error bars represent
the average of two replicates (n 2). The asterisks indicate the signicant values (p < 0.05) as compared to the calibrator (quiescent seed), according to Dunnett's method.

gradual, since no transcripts of TcLeg6 gene were detected in the

analysis of the mature seed (quiescent) and germinating cotyledon.
According to the dendrogram (Fig. 2) TcLEG6 is inserted in the
dVPE group. That type of VPE is expressed in developing seeds
before bVPE (TcLeg3), and it is inserted in the embryogenesis-type
group, being involved in the process in of removing the internal
layers that are required for the formation of the external seed
tegument (Christoff et al., 2014; Hara-Nishimura and Hatsugai,
2011; Nakaune et al., 2005).

4.2.3. TcLeg9 is a VPE related to development and defense against

Fig. 6. Expression pattern of TcLeg3, TcLeg6, and TcLeg9 genes in developing seeds of
T. cacao. Values were normalized through ACT endogenous protein. Error bars repre-
sent the average of ve replicates (n 5). The asterisks indicate the signicant values
(p < 0.05) as compared to the calibrator (rst development stage), according to
Dunnett's method.
related to specic roles in seeds. That is corroborated through the
dendrogram analysis (Fig. 2) which includes the TcLEG3 in the bVPE
group and in the developing seeds during the expression (Fig. 6), in
which no signicant differences occurred, and reinforces the idea
that such VPE must play an important role in the mobilization of
proteins during the initial germination step (Mntz and Shutov,
2002; Nakaune et al., 2005; Radchuk et al., 2011).
4.2.2. TcLeg6 is related to embryogenesis
TcLeg6 gene was only expressed in the developing seeds, when
the transcript was observed to continuously increase from the rst
to the third seed development stage. Afterwards, a slight reduction
of the expression took place. That transcript reduction must be
pathogens
The TcLeg9 gene was the most expressed in leaves and stalk, as
compared to the calibrator (root) (Fig. 4b). In the germinating
cotyledons (Fig. 5b), the expression pattern increased from the
initial time to its overexpression in the twentieth day, decreasing
right after the thirtieth day. The expression reduction on the thir-
tieth day may be related to the period marked by the senescence of
cotyledons.
In the tissue that was infected by M. perniciosa, the expression of
gene TcLeg9 was shown to be high on the rst day after the inoc-
ulation, and on the forty-fth day, when the plant necrosis process
started (Fig. 7b), which suggests that this gene may be related to the
defense against pathogens or PCD. The signicant expression in-
crease on the rst day may be related to the increased hydrogen
peroxide production in that period, which was not enough to
characterize a hypersensitivity response, but instead a biochemical
mechanism against M. perniciosa infection (Dias et al., 2011). In
response to the increased peroxide production, in fungus-resistant
cocoa genotypes, increased ascorbate peroxidase (APx) activity was
observed in infected tissues (Camillo et al., 2013). In the rst signs
of the fungus attack, the calcium oxalate druses are converted into
carbon dioxide, oxygen, and hydrogen peroxide, but they do not

Fig. 7. Expression pattern of TcLeg3 (a) and TcLEG9 (b) genes in tissue (meristem) after inoculation of M. pernicious. Values were normalized through MDH endogenous enzyme.
Error bars represent the average of ten replicates (n 10). The asterisks indicate the signicant values (p < 0.05) as compared to the calibrator (control), according to Dunnett's
method.
 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170
cause cell death (Dias et al., 2011). In contrast, the signicant
expression increase on the forty-fth day may have been induced
by the tissue senescence and degradation, because the initial
symptoms of witches' witch are observed in this period after
inoculation, and the lengthening of the axillary buds and swelling
of the stalk (Ceita et al., 2007).
The expression prole of TcLeg9 the VPE suggests that is related
to the type vegetative functions. This is also corroborated through
its location in the dendrogram (Fig. 2), where it is included in the a
and gVPE group, which suggests it is involved in the plant devel-
opment (Kinoshita et al., 1995a e 1995b). According to Mntz and
Shutov (2002), during the germination of seeds, reserve proteins
that are deposited in the storage vacuoles are mobilized and used
for plant growth.
Even though genes TcLeg3 and TcLeg9 are respectively inserted
in the seed and vegetative groups and have been expressed in
several tissues, the classication of VPEs continues to be performed
based on the functions that these enzymes play during the plant life
cycle (Gruis et al., 2002; Kinoshita et al., 1999; Yamada et al., 2005).
5. Conclusion
We have presented the identication and characterization of the
three VPEs that are coded by the T. cacao genome, which are named
TcLEG3, TcLEG6, and TcLEG9. Three-dimensional modeling revealed
that the propeptide that is located in the C-terminal regions of VPEs
blocks the enzyme catalytic cleft, which prevents interactions and
causes enzyme latency. VPEs are separated in three groups: seed
(TcLeg3), embryogenesis (TcLeg6), and vegetative (TcLeg9) accord-
ing to where they are located in the dendrogram and how genes are
expressed in the various tissues of plants. TcLeg3 may be related to
the mobilization of reserve proteins, whereas TcLeg9 may be
involved in the vegetative phase during the life cycle of plants and
processes regarding senescence and response against the
M. perniciosa pathogen. In turn, TcLeg6 only accumulates tran-
scripts in the developing seeds, and it may be related to the for-
mation of the seed external tegument.
Contributions from the authors
Juliano Santana performed the experiments and wrote the
manuscript. Las Freire and Aurizangela Oliveira contributed with
the data analysis and with the execution of some experiments.
Virgnia Lcia, advice and guidance. Karina Gramacho contributed
with the experiment design and plant inoculations. Carlos Pri-
minho, since the rst idea, has supervised the investigation efforts
and contributed with reagents and materials.
Conict of interest
The authors state that no conict of interest exists.
Authors contributions
Experimental design: Juliano Oliveira Santana, Virgnia Lcia
Fontes Soares and Carlos Priminho Pirovani.
Data collect: Juliano Oliveira Santana, Las Freire, Aurizangela
Oliveira de Sousa, and Karina Peres Gramacho.
Data analysis: Juliano Oliveira Santana, Las Freire, Aurizangela
Oliveira de Virgnia Lcia Fontes Soares, Carlos Priminho Pirovani.
laboratory infrastructure: Carlos Priminho Pirovani and Karina
Peres Gramacho writing of the manuscript: Juliano Oliveira Santana
and Carlos Priminho Pirovani.
169
Acknowledgments
The project was funded by Fundaa ~o de Amparo a Pesquisa do
Estado da Bahia - FAPESB (PNE 005/2011 FAPESB/CNPq). We would
like to thank MARS Brasil for the infrastructure they provided in
their microbiology laboratory and CEPLAC for making their em-
ployees available for conducting the experiments at CEPEC's labo-
ratory. Pirovani CP gratefully acknowledges the Conselho Nacional
de Desenvolvimento Cientco and Tecnolo
gico (CNPq), Brazil
(305309/2012-9) for the concession of a scientic productivity
fellowship.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.plaphy.2015.11.010.
References
Aime, M.C., Phillips-Mora, W., 2005. The causal agents of witches' broom and frosty
pod rot of cacao (chocolate, Theobroma cacao) form a new lineage of Mar-
asmiaceae. Mycologia 97 (5), 1012e1022. http://dx.doi.org/10.3852/mycologia.
97.5.1012.
Argout, X., Salse, J., Aury, J., et al., 2011. The genome of Theobroma cacao. Nat. Genet.
43 (2), 101e108. http://dx.doi.org/10.1038/ng.736.
Bendtsen, J.D., Nielsen, H., Heijne, G., Brunak, S., 2004. Improved prediction of signal
peptides: SignalP 3.0. J. Mol. Biol. 340 (4), 783e795. http://dx.doi.org/10.1016/j.
jmb.2004.05.028.
Bode, W., Huber, R., 1992. Natural protein proteinase inhibitors and their interaction
with proteinases. Eur. J. Biochem. 204, 433e451. http://dx.doi.org/10.1111/j.
1432-1033.1992.tb16654.x.
Camillo, L.R., Filadelfo, C.R., Monzani, P.S., Corre ^a, R.X., Gramacho, K.P., Micheli, F.,
Pirovani, C.P., 2013. Tc-cAPX, a cytosolic ascorbate peroxidase of Theobroma
cacao L. engaged in the interaction with Moniliophthora perniciosa, the causing
agent of witches' broom disease. Plant Physiol. Biochem. (Paris) 73, 254e265.
http://dx.doi.org/10.1016/j.plaphy.2013.10.009.
Ceita, G. de O., Mace ^do, J.N.A., Santos, T.B., Alemanno, L., Gesteira, A. da S.,
Micheli, F., Mariano, A.C., Gramacho, K.P., Silva, D. da C., Meinhard, L.,
Mazzafera, P., Pereira, G.A.G., de M. Cascardo, J.C., 2007. Involvement of calcium
oxalate degradation during programmed cell death in Theobroma cacao tissues
triggered by the hemibiotrophic fungus Moniliophthora perniciosa. Plant Sci.
173, 106e117. http://dx.doi.org/10.1016/j.plantsci.2007.04.006.
Claus, L.A., Jens, L.J., Torben, F., 2004. Normalization of real-time quantitative
reverse transcription-PCR data: a model-based variance estimation approach to
identify genes suited for normalization, applied to bladder and colon cancer
data sets. Cancer Res. 64 (1), 5245e5250. http://dx.doi.org/10.1158/0008-5472.
CAN-04-0496.
Chan, C.B., Abe, M., Hashimoto, N., Hao, C., Williams, I.R., Liu, X., Nakao, S.,
Yamamoto, A., Zheng, C., Henter, J.I., Meeths, M., Nordenskjold, M., Li, S.Y., Hara-
Nishimura, I., Asano, M., Y e, K., 2009. Mice lacking asparaginil endopeptidase
develop disorders resembling hemophagocytic syndrome. PNAS 106 (2),
468e473. http://dx.doi.org/10.1073/pnas.0809824105.
Chen, J.-M., Rawlings, N.D., Stevens, R.A.E., Barrett, A.J., 1998. Identication of the
active site of legumain links it to caspases, clostripain and gingipains in a new
clan of cysteine endopeptidases. FEBS Lett. 441, 361e365. http://dx.doi.org/10.
1016/S0014-5793(98)01574-9.
Chen, J.-M., Fortunato, M., Barrett, A.J., 2000. Activation of human prolegumain by
cleavage at a C-terminal asparagine residue. Biochem. J. 352 (2), 327e334,
3520327. http://dx.doi.org/10.1042/0264-6021.
Chothia, C., Lesk, A.M., 1986. The relation between the divergence of sequence and
structure in proteins. EMBO J. 5 (4), 823e826.
Christoff, A.P., Turchetto-zolet, A.C., Margis, Rogerio, 2014. Uncovering legumain
genes in rice. Plant Sci. 215, 100e109. http://dx.doi.org/10.1016/j.plantsci.2013.
11.005.
Dall, E., Brandstetter, H., 2012. Activation of legumain involves proteolytic
and conformational events, resulting in a context and substrate-dependent
activity prole. Acta Cryst. F68 (1), 24e31. http://dx.doi.org/10.1107/
S1744309111048020.
Dall, E., Brandstetter, H., 2013. Mechanistic and structural studies on legumain
explainits zymogenicity, distinct activation pathways, and regulation. PNAS 110
(27), 10940e10945. http://dx.doi.org/10.1073/pnas.1300686110.
Dall, E., Fegg, J.C., Briza, P., Brandstetter, H., 2015. Structure and mechanism of an
aspartimide-dependent peptide ligase in human legumain. Angew. Chem. Int.
Ed. Engl. 54 (10), 2917e2921. http://dx.doi.org/10.1002/anie.201409135.
Donnison, I.S., Gay, A.P., Thomas, H., Edwards, K.J., Edwards, D., James, C.L.,
Thomas, A.M., Ougham, H.J., 2007. Modication of nitrogen remobilization,
grain ll and leaf senescence in maize (Zea mays) by transposon insertional
mutagenesis in a protease gene. New Phytol. 173 (3), 481e494. http://dx.doi.
org/10.1111/j.1469-8137.2006.01928.x.
170 J.O. Santana et al. / Plant Physiology and Biochemistry 98 (2016) 162e170
Dias, C.V., Mendes, J.S., dos Santos, A.C., Pirovani, C.P., da Silva Gesteira, A.,
Micheli, F., Gramacho, K.P., Hammerstone, J., Mazzafera, P.,
de Mattos Cascardo, J.C., 2011. Hydrogen peroxide formation in cacao tissues
infected by the hemibiotrophic fungus Moniliophthora perniciosa. Plant Physiol.
Biochem. 49, 917e922. http://dx.doi.org/10.1016/j.plaphy.2011.05.004.
Eswar, N., Webb, B., Marti-Renom, M.A., Madhusudhan, M.S., Eramian, D.,
Shen, M.Y., Pieper, U., Sali, A., 2007. Comparative protein structure modeling
using Modeller. Curr. Protoc. Protein Sci. 50, 2.9.1-2.9.31 http://dx.doi.org/10.
1002/0471140864.ps0209s50.
Frias, G.A., Purdy, L.H., Schmidt, R.A., 1991. Infection biology of Crinipellis perniciosa
on vegetative ushes of cacao. Plant Dis. 7 (6), 552e556. http://dx.doi.org/10.
1094/PD-75-0552.
Gruis, D.F., Selinger, D.A., Curran, J.M., Jung, R., 2002. Redundant proteolytic
mechanisms process seed storage proteins in the absence of seed-type mem-
bers of the vacuolar processing enzyme family of cysteine proteases. Plant Cell
14, 2863e2882. http://dx.doi.org/10.1105/tpc.005009.
Hara-Nishimura, I., Kaori, I., Mikio, N., 1991. A unique vacuolar processing enzyme
responsible for conversion of several proprotein precursors into the mature
forms. FEBS Lett. 294, 89e94. http://dx.doi.org/10.1016/0014-5793(91)81349-D.
Hara-Nishimura, I., Tomoo, S., Nagako, H., Mikio, N., 1995. Vacuolar processing
enzymes responsible for maturation of seed proteins. J. Plant Physiol. 145,
632e640. http://dx.doi.org/10.1016/S0176-1617(11)81275-7.
Hara-Nishimura, I., Hatsugai, Nakaune S., Kuroyanagi, M., Nishimura, M., 2005.
Vacuolar processing enzyme: an executor of plant cell death. Curr. Opin. Plant
Biol. 8, 404e408. http://dx.doi.org/10.1016/j.pbi.2005.05.016.
Hara-Nishimura, I., Hatsugai, N., 2011. The role of vacuole in plant cell death. Cell
Death Differ. 18, 1298e1304. http://dx.doi.org/10.1038/cdd.2011.70.
Hatsugai, N., Kuroyanagi, M., Yamada, K., Meshi, T., Tsuda, S., Kondo, M.,
Nishimura, M., Hara-Nishimura, I., 2004. A plant vacuolar protease, VPE, me-
diates virus-induced hypersensitive cell death. Science 305, 855e858. http://dx.
doi.org/10.1126/science.1099859.
Hiraiwa, N., Nishimura, M., Hara-Nishimura, I., 1997. Expression and activation of
the vacuolar processing enzyme in Saccharomyces cerevisiae. Plant J. 12 (4),
819e829. http://dx.doi.org/10.1046/j.1365-313X.1997.12040819.x.
Jackson, M.A., Rae, A.L., Casu, R.E., Grof, C.P.L., Bonnett, G.D., Maclean, D.J., 2007.
A bioinformatic approach to the identication of a conserved domain in a
sugarcane legumain that directs GFP to the lytic vacuole. Funct. Plant Bio 34,
633e644. http://dx.doi.org/10.1071/FP07024.
Johnson, M.S., Srinivasan, N., Sowdhamini, R., Blundell, T.L., 1994. Knowledge-based
protein modeling. Crit. Rev. Biochem. Mol. Biol. 29 (1), 1e68. http://dx.doi.org/
10.3109/10409239409086797.
Kinoshita, T., Nishimura, M., Hara-Nishimura, I., 1995a. Homologues of a vacuolar
processing enzyme that are expressed in different organs in Arabidopsis thali-
ana. Plant Mol. Bio 29, 81e89. http://dx.doi.org/10.1007/BF00019120.
Kinoshita, T., Nishimura, M., Hara-Nishimura, I., 1995b. The sequence and expres-
sion of the gamma-VPE gene, one member of a family of three genes for
vacuolar processing enzymes in Arabidopsis thaliana. Plant Cell Physiol. 36 (8),
1555e1562. http://dx.doi.org/.
Kinoshita, T., Yamada, K., Hiraiwa, N., Kondo, M., Nishimura, M., Hara-Nishimura, I.,
1999. Vacuolar processing enzyme is up-regulated in the lytic vacuoles of
vegetative tissues during senescence and under various stressed conditions.
Plant J. 19 (1), 43e53. http://dx.doi.org/10.1046/j.1365-313X.1999.00497.x.
Klinkert, M.-Q., Felleiseng, R., Link, A., Beck, E., 1989. Primary structure of Sm31/
Sm32 diagnostic proteins of Schistosoma mansoni and their identication as
proteases. Mol. Biocherm. Parasitol. 33, 113e122. http://dx.doi.org/10.1016/
0166-6851(89)90025-X.
Kumar, S., Tamura, K., Nei, M., 2004. MEGA3: integrated software for molecular
evolutionary genetics analysis and sequence alignment. Brief. Bioinform 5 (2),
150e163. http://dx.doi.org/10.1093/bib/5.2.150.
Kuroyanagi, M., Nishimura, M., Hara-Nishimura, I., 2002. Activation of Arabidopsis
vacuolar processing enzyme by self-catalytic removal of an auto-inhibitory
domain of the C-terminal propeptide. Plant Cell Physiol. 43 (2), 143e151.
http://dx.doi.org/10.1093/pcp/pcf035.
Li, D.N., Matthews, S.P., Antoniou, A.N., Mazzeo, D., Watts, C., 2003. Multistep
Autoactivation of asparaginyl endopeptidase in Vitro and in Vivo. J. Bio Chem.
278 (40), 38980e38990. http://dx.doi.org/10.1074/jbc.M305930200.
Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using
real-time quantitative PCR and the 2- DDCT method. Methods 25, 402e408.
http://dx.doi.org/10.1006/meth.2001.1262.
Marti-Renom, M.A., Stuart, A.C., Fiser, A., S
anchez, R., Melo, F., Sali, A., 2000.
Comparative protein structure modeling of genes and genomes. Annu. Rev.
Biophys. Biomol. Struct. 29, 291e325. http://dx.doi.org/10.1146/annurev.
biophys.29.1.291.
Mntz, K., Shutov, A.D., 2002. Legumains and their functions in plants. Trends Plant
Sci. 7, 340e344. http://dx.doi.org/10.1016/S1360-1385(02)02298-7.
Nakaune, S., Yamada, K., Kondo, M., Kato, T., Tabata, S., Nishimura, M., Hara-
Nishimura, I., 2005. A vacuolar processing enzyme, dVPE, is involved in seed
coat formation at the early stage of seed development. Plant Cell 17, 876e887.
http://dx.doi.org/10.1105/tpc.104.026872.
Orchard, J., Collin, H.A., Hardwick, K., Isaac, S., 1994. Changes in morphology and
measurement of cytokinin levels during the development of witches' brooms
on cocoa. Plant Pathol. 43 (1), 65e72. http://dx.doi.org/10.1111/j.1365-3059.
1994.tb00554.x.
Pinheiro, T.T., Litholdo, C.G., Sereno, M.L., Leal, G.A., Albuquerque, P.S.B., Figueira, A.,
2011. Establishing references for gene expression analyses by RT-qPCR in The-
obroma cacao tissues. Genet. Mol. Res. 10 (4), 3291e3305. http://dx.doi.org/10.
4238/2011.
Pirovani, C.P., da Silva Santiago, A., dos Santos, L.S., Micheli, F., Margis, R., da Silva
Gesteira, A., Alvim, F.C., Pereira, G.A., de Mattos Cascardo, J.C., 2010. Theobroma
cacao cystatins impair Moniliophthora perniciosa mycelial growth and are
involved on postponing cell death symptoms. Planta 232 (1), 1485e1497. http://
dx.doi.org/10.1007/s00425-010-1272-0.
Radchuk, V., Weier, D., Radchuk, R., Weschke, W., Weber, H., 2011. Development of
maternal seed tissue in barley is mediated by regulated cell expansion and cell
disintegration and coordinated with endosperm growth. J. Exp. Bot. 62 (3),
1217e1227. http://dx.doi.org/10.1093/jxb/erq348.
Rojo, E., Martn, R., Carter, C., Zouhar, J., Pan, S., Plotnikova, J., Jin, H., Paneque, M.,

nchez-Serrano, J.J., Baker, B., Ausubel, F.M., Raikhel, N.V., 2004. VPE<gamma>
Sa
exhibits a caspase-like activity that contributes to defense against pathogens.
Curr. Biol. 14 (21), 1897e1906. http://dx.doi.org/10.1016/j.cub.2004.09.056.
Santos-Silva, L.K., Soares-Costa, A., Gerald, L.T.S., Meneghin, S.P., Henrique-Silva, F.,
2012. Recombinant expression and biochemical characterization of sugarcane
legumain. Plant Physiol. Biochem. 57, 181e192. http://dx.doi.org/10.1016/j.
plaphy.2012.05.020.
Scarpari, L.M., Meinhardt, L.W., Mazzafera, P., Pomella, A.W.V., Schiavinato, M.A.,
Cascardo, J.M.C., Pereira, G.A., 2005. Biochemical changes during the develop-
ment of witches' broom: the most important disease of cocoa in Brazil caused
by Crinipellis perniciosa. J. Exp. Bot. 56, 865e877. http://dx.doi.org/10.1093/jxb/
eri079.
Scott, M.P., Jung, R., Muntz, K., Nielsen, N.C., 1992. A protease responsible for post-
translational cleavage of a conserved Asn-Gly linkage in glycinin, the major seed
storage protein of soybean. Proc. Natl. Acad. Sci. 89, 658e662. http://dx.doi.org/
10.1073/pnas.89.2.658.
Surujdeo-Maharaj, S., Umaharan, P., Butler, D.R., Sreenivasan, T.N., 2003. An opti-
mized screening method for identifying levels of resistance to Crinipellis per-
niciosa in cocoa (Theobroma cacao). Plant Pathol. 52, 464e475. http://dx.doi.
org/10.1046/j.1365-3059.2003.00865.x.
Yamada, K., Shimadab, T., Nishimura, M., Hara-Nishimura, I., 2005. A VPE family
supporting various vacuolar functions in plants. Physiol. Plant 123, 369e375.
http://dx.doi.org/10.1111/j.1399-3054.2005.00464.x.
Zhao, L., Hua, T., Crowley, C., Ru, H., Ni, X., Shaw, N., Jiao, L., Ding, W., Qu, L., Hung, L.,
Huang, W., Liu, L., Ye, K., Ouyang, S., Cheng, G., Liu, Z., 2014. Structural analysis
of asparaginyl endopeptidase reveals the activation mechanism and a reversible
intermediate maturation stage. Cell Res. 24, 344e358. http://dx.doi.org/10.
1038/cr.2014.4.