|
J Biol Chem, Vol. 273, Issue 47, 31230-31236, November 20, 1998
Transport and Activation of the Vacuolar Aspartic Proteinase
Phytepsin in Barley (Hordeum vulgare L.)*
Stefanie
Glathe §,
Jukka
Kervinen¶ ,
Manfred
Nimtz**,
Grace H.
Li ,
Gregory J.
Tobin ,
Terry D.
Copeland¶,
David A.
Ashford§§,
Alexander
Wlodawer¶, and
Júlia
Costa §¶¶||
From the Instituto de Biologia Experimental e
Tecnológica/Instituto de Tecnologia Química e
Biológica, Apartado 12, 2780 Oeiras, Portugal, ¶ Advanced
Bioscience Laboratories-Basic Research Program and
 Laboratory of Cell and Molecular Structure-Science
Applications International Corp., NCI, National Institutes of Health,
Frederick Cancer Research and Development Center, Frederick, Maryland
21702, ** Gesellshaft für Biotechnologische Forshung, D-38124
Braunschweig, Germany, §§ Glycobiology Research and
Analytical Facility, Department of Biology, University of York, UK,
and ¶¶ Unidade de Ciências Exactas e Humanas,
Universidade Algarve, 8000 Faro Portugal
 |
ABSTRACT |
The primary translation product of barley
aspartic proteinase, phytepsin (EC 3.4.23.40), consists of a signal
sequence, a propart, and mature enzyme forms. Here, we describe
post-translational processing and activation of phytepsin during its
transport to the vacuole in roots, as detected by using metabolic
labeling and immunoprecipitation. After removal of the signal sequence, the glycosylated precursor of 53 kDa (P53) was produced and further processed to polypeptides of 31 and 15 kDa (P31 + P15) and,
subsequently, to polypeptides of 26 and 9 kDa (P26 + P9), 45 min and
24 h after synthesis, respectively. The processing occurred in a
late-Golgi compartment or post-Golgi compartment, because brefeldin A
inhibited the processing, and P53 acquired partial endoglycosidase H
resistance 30 min after synthesis, whereas P15 was completely
resistant. The N-glycosylation inhibitor tunicamycin had no
effect on transport, but the absence of glycans on P53 accelerated the
proteolytic processing. Phytepsin was also expressed in
baculovirus-infected insect cells. The recombinant prophytepsin
underwent autoproteolytic activation in vitro and showed
enzymatic properties similar to the enzyme purified from grains.
However, a comparison of the in vitro/in vivo
processing sites revealed slight differences, indicating that
additional proteases are needed for the completion of the maturation
in vivo.
 |
INTRODUCTION |
Aspartic proteinases
(APs)1 (EC 3.4.23) constitute
one of the four superfamilies of proteolytic enzymes. They are present
in a wide variety of organisms, such as viruses, fungi, plants, and animals. Common features of APs include an active site cleft that contains two catalytic aspartic acid residues (32 and 215 in pepsin), acidic pH optima for enzymatic activity, inhibition by pepstatin A, a
conserved overall fold, and a preferential cleavage specificity for
peptide bonds between amino acid residues with bulky hydrophobic side
chains. Both intracellular and extracellular forms of APs are present
in animal tissues (1, 2).
Aspartic proteinases are synthesized as inactive precursors (zymogens)
in which the N-terminal propeptide is bound to the active site cleft,
thus preventing undesirable protein degradation and enabling spatial
and temporal regulation of proteolytic activity. Pepsinogen, the
inactive precursor of stomach pepsin, needs only a drop in pH for the
autocatalytic cleavage of the propeptide to result in an active enzyme
(1). Procathepsin D, which is targeted to the lysosome largely via the
mannose-6-phosphate receptor (3), is activated after cleavage of its
N-terminal 44 amino acids, most likely by lysosomal cysteine
proteinases (4). Procathepsin D is also capable of
acid-dependent autoactivation in vitro to yield
a catalytically active (pseudo)cathepsin D (5, 6). However,
autocatalytic removal of the remaining 18-residue propeptide or the
processing intermediate corresponding to pseudocathepsin D has not been
observed in vivo (7).
Barley AP (Hordeum vulgare AP), recently renamed phytepsin
(EC 3.4.23.40) (8), was originally isolated from grains in which it
exists as two enzymatically active two-chain forms (9). Sequence
alignment of phytepsin with animal and microbial APs shows a high
degree of similarity, with the exception of an inserted domain of
approximately 100 residues that is plant-specific (10-12) and very
similar to that of saposins (13). The exact function of phytepsin is
still controversial. Because phytepsin is an intracellular enzyme
residing in leaf and root vacuoles (14) and in scutellar and aleuronal
vacuole-like protein bodies in grains (15) and because it is able to
cleave the C-terminal vacuolar targeting signal of barley prolectin
in vitro (14), phytepsin may represent a cathepsin D-like
enzyme from plant cells. Accordingly, phytepsin may participate in
protein processing and metabolic turnover (Ref. 16 and references
therein). It has recently also been observed that phytepsin may play a
role in the active autolysis in plant tissues undergoing
developmentally regulated programmed cell death (17).
Modification of phytepsin during its intracellular route to vacuoles
involves several steps; however, the enzymology, sequence, and
intracellular localization of these events is not known. The study
presented here describes the mode of expression, processing, and
activation of phytepsin during its transport. Furthermore, we
demonstrate the autoactivation of phytepsin in vitro by
using a recombinant enzyme expressed in insect cells.
 |
EXPERIMENTAL PROCEDURES |
Materials--
H. vulgare (cv. Sereia) grains were
purchased from the Estação Nacional de Melhoramento de
Plantas, Elvas, Portugal. A purified antiserum against phytepsin was
prepared as described previously (14).
L-[35S]methionine and
L-[35S]cysteine (Pro-mixTM in vivo
cell labeling mix) and autoradiography films (HyperfilmTM-MP) were
purchased from Amersham Pharmacia Biotech. Molecular weight standards
for gel electrophoresis were purchased from Bio-Rad. Protein A coupled
to Sepharose was purchased from Amersham Pharmacia Biotech.
Endoglycosidase H (Endo H) was from Oxford GlycoSciences (Oxford, UK).
ImmobilonTM-P polyvinylidene difluoride (PVDF) transfer membranes were
purchased from Millipore Corp. (Bedford, MA). The remaining reagents
were of analytical grade.
Plant Culture, Protein Extraction, and
Immunoprecipitation--
Barley grains were surface-sterilized with
1% (w/v) sodium hypochlorite followed by 60% (v/v) ethanol and
germinated on Petri dishes containing 15 ml of 0.8% (w/v) agar for 3 days at 28 °C in the dark. For the extraction of root proteins,
nondenaturing solubilization buffer (30 mM Tris, pH 7.5, 1 mM EDTA, 0.25 M sucrose, 5% (w/v)
polyvinylpolypyrrolidone, 0.15% (v/v) -mercaptoethanol) was added
to the roots frozen in liquid nitrogen. Five root tips were homogenized
with 0.2 ml of the buffer using a Teflon homogenizer (Sigma). The
extracts were cleared by centrifugation for 5 min at 10,000 rpm, and
the supernatants were used for immunoprecipitation or Western blot
analysis. Protein A-Sepharose (3 mg/sample) was resuspended in
immunoprecipitation buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 0.1% (v/v) Triton X-100, 0.05% (w/v) sodium
deoxycholate, 10% (v/v) glycerol, 1% (w/v) bovine serum albumin, 1 mM EDTA) and washed twice. Rabbit antibodies were directly
coupled to the Protein A-Sepharose by shaking for 20 min at room
temperature. The proteins in the lysate supernatants were
immunoprecipitated with aliquots of the protein
A-Sepharose-antibody-protein complex for 1 h on ice.
Immunoprecipitates were washed twice each with high salt (50 mM Hepes, pH 7.5, 0.5 M NaCl, 5 mM
EDTA, 0.2% (v/v) Triton X-100, 0.1% (w/v) SDS), medium salt (50 mM Hepes, pH 7.5, 0.15 M NaCl, 5 mM
EDTA, 0.2% (v/v) Triton X-100), and low salt (10 mM Tris,
pH 7.5, 0.1% (v/v) Triton X-100) buffers (18). The complex was
resuspended in 25 µl of SDS-polyacrylamide gel electrophoresis (PAGE)
sample buffer, boiled for 5 min, and analyzed by SDS-PAGE (19).
Western Blot Analysis--
The proteins were transferred from
the gel to a PVDF membrane in a Bio-Rad semi-dry transfer cell for
1 h at 15 V. The PVDF membrane was blocked in Tris-buffered saline
containing 0.05% (v/v) Tween 20 and 3% (w/v) dried lowfat milk for 45 min. The antiserum and the second antibody (alkaline
phosphatase-coupled goat-anti-rabbit IgG) were used at 1/500 and 1/8000
dilutions, respectively, in blocking solution containing 1% dried
lowfat milk. Detection was performed by the enhanced chemiluminescence (ECL) method (Amersham Pharmacia Biotech) or with 4-chloro-1-naphthol (20).
Metabolic Labeling and Pulse-Chase Experiments--
For each
time point, 5 roots from 3-day-germinated grains were cut ~1 cm from
the tip, and the cut ends were dipped in 20 µl of essential B5 plant
growth medium containing 4% (w/v) sucrose and 100 µCi of Pro-mixTM
(L-[35S]methionine/cysteine) for the
indicated pulse periods. The chase was performed in 20 µl of
essential B5 plant growth medium containing 4% (w/v) sucrose, 1 mM L-methionine, and 0.5 mM
L-cysteine for the indicated time periods. Protein extracts
were prepared from roots frozen in liquid nitrogen and were
immunoprecipitated with anti-phytepsin serum. The proteins (~40 µg)
were separated by SDS-PAGE in a gradient gel (10-20%). The gel was
dried, and 35S-labeled proteins were detected by fluorography.
Endo H Digestion of Phytepsin--
Metabolic labeling and
pulse-chase experiments were performed as described above, except that
for each time point, 10 root tips were incubated with 200 µCi of
Pro-mixTM. After immunoprecipitation, the protein was dissociated from
the antibodies with 40 µl of 0.1 M glycine-HCl buffer, pH
3.0, then incubated in 50 mM citrate buffer, pH 5.5, containing 1 mg/ml SDS and 0.2 M -mercaptoethanol at
95 °C for 10 min. Phenylmethylsulfonyl fluoride, E-64, and pepstatin
A were added to final concentrations of 1 mM, 10 µg/ml, and 10 µg/ml, respectively. Six milliunits of Endo H were added to a
final volume of 100 µl, and the mixture was incubated at 37 °C for
18 h. The protein was then precipitated with ethanol and analyzed
by SDS-PAGE in 12 or 10-20% polyacrylamide gradient gels.
Experiments with Inhibitors--
Root tips (1 cm) were dipped
for 1 h in 14 µl of essential B5 plant growth medium containing
4% (w/v) sucrose and inhibitors at the following concentrations: 7 µM brefeldin A1 in methanol and 10 µg/ml
tunicamycin A1 in methanol. The control samples were incubated with the same volume of methanol added. The roots were radiolabeled as described above.
Determination of Incorporation of Radiolabeled Amino Acids into
Protein--
Duplicate samples of 1 µl of metabolically labeled
extract were spotted on small pieces of Whatman 3MM filter paper. The
papers were dried, and one of the duplicates was washed 5 times with 5 ml of 5% (v/v) trichloroacetic acid and dried. The papers were placed
at the bottom of plastic vials, 3 ml of liquid scintillation mixture
(Beckman) was added, and the radioactivity was counted in the carbon
channel of a Beckman scintillation counter.
Expression of Phytepsin in Insect Cells--
Sequences encoding
the complete preprophytepsin (1863 base pairs (10)) were cloned into
the KpnI restriction site of the baculovirus transfer vector
pBlueBac4.5 (Invitrogen Corp., Carlsbad, CA). The construct was
cotransfected with Autographa californica multiple nuclear
polyhedrosis viral DNA into Spodoptera frugiperda (Sf9) cells, and recombinant baculoviruses were derived using standard methodologies (21). For protein production, Sf9
suspension cultures were infected with A. californica
multiple nuclear polyhedrosis virus-phytepsin in complete Grace's
medium (about 600 ml) supplemented with 7% fetal bovine serum. Four
days after infection, the conditioned culture supernatant was clarified
for 20 min at 6,000 × g and concentrated to
approximately 2% of the initial volume at 4 °C by ultrafiltration
(Amicon YM30 membrane, Amicon, Danvers, MA). The concentrated
supernatant was diluted 1.3-fold with cold 0.5 M sodium
acetate, pH 4.0, and phytepsin was purified according to Sarkkinen
et al. (9) by affinity chromatography on a pepstatin-agarose column, with the exception that no washing with pH 7.5 buffer was
carried out and the elution was performed with 0.1 M
Tris-HCl, pH 8.8, 0.2 mM dithiothreitol, 0.1 M
NaCl. Further purification was performed by ion exchange chromatography
on a Mono Q column (Amersham Pharmacia Biotech).
In Vitro Processing of Phytepsin--
Recombinant phytepsin
purified on the Mono Q column (2.2 mg/ml in 20 mM Tris-HCl,
pH 8.0, 0.1 M NaCl) was mixed with an equal volume of 0.2 M incubation buffer and kept at 37 °C for up to 90 min.
Incubation buffers were as follows: sodium lactate, pH 3.7, sodium
acetate, pH 4.5 and 5.5, and sodium phosphate, pH 6.5. Samples were
removed after various times and frozen at -70 °C. Processing
products were separated by SDS-PAGE (PhastSystem, Amersham Pharmacia
Biotech) and stained with Coomassie Brilliant Blue R-250 or
electroblotted onto a PVDF membrane for N-terminal sequencing. The
Coomassie Brilliant Blue-stained protein bands in the membrane were
excised and individually sequenced on an Applied Biosystems 477A
gas-phase sequencer, and the phenylthiohydantoin amino acids were
identified on-line with a Model 120 analyzer (Applied Biosystems, Inc.,
Foster City, CA).
Proteinase Assay--
Phytepsin samples were diluted with 20 mM sodium lactate, pH 3.7, and preincubated for 1 h at
37 °C, and the activity was measured according to Sarkkinen et
al. (9) using bovine hemoglobin as the substrate at pH 3.7. The
assay temperature was 37 °C. One unit of activity corresponded to
the enzymatic activity that liberated trichloroacetic acid-soluble
reaction products equivalent to 1 mg of bovine serum albumin in 1 h at 37 °C.
Matrix-assisted Laser Desorption Ionization Time-of-flight Mass
Spectroscopy (MALDI-TOF-MS)--
Glycoproteins were analyzed using a
matrix of 22.4 mg of 3,5-dimethoxy-4-hydroxycinnamic acid in 400 µl
of acetonitrile and 600 µl of 0.1% (v/v) trifluoroacetic acid in
H2O as the UV-absorbing material. The solubilized samples
were mixed with the same volume of matrix, and 1 µl of the mixture
was spotted onto the stainless steel tip and dried at room temperature.
The concentration of the analyte was ~5-25 pmol/µl. Measurements
were performed on a Bruker REFLEXTM MALDI/TOF mass
spectrometer using a N2 laser (337 nm) with a 3-ns pulse
width and 107-108 watt/cm2 irradiance at the surface (0.2 mm2 spot). Spectra were recorded at an acceleration voltage
of 28.5 kV in the linear mode, using the delayed extraction facility.
 |
RESULTS |
Phytepsin in Grains and Roots--
Affinity-purified phytepsin
preparation from barley grains typically contains two enzyme forms of
approximately 32 + 16 kDa and 29 + 11 kDa and occasionally some higher
molecular mass precursors. In earlier studies (9, 10) molecular weight
estimations for different chains were based on SDS-PAGE analyses (Fig.
1) and on calculations from the
cDNA-derived protein sequence when the N-terminal sequence was
known. However, details about the processing of the different
polypeptides at the C terminus were not known. To more accurately
describe the sizes of the polypeptides resulting from the processing
events, we used MALDI-TOF-MS for the analysis of reduced processed
products that corresponded to the polypeptides constituting the two
isoforms of phytepsin from barley grains. The sizes determined for two
or three independent phytepsin preparations were 9.2-9.5, 15.3-15.8,
26.4-26.7, 30.5-30.8, and 46-47 kDa. Accordingly, the polypeptides
were named P9, P15, P27, P31, and P47 (indicated on the left side of
Fig. 1). When we analyzed barley root extracts by Western blotting with
the same anti-phytepsin antibody, we detected polypeptides P9, P15, and
P31. We also detected a polypeptide of ~26 kDa in roots instead of 27 kDa as previously observed in grains (14), and additional polypeptides
of approximately 42, 46, and 53 kDa (P42, P46, and P53, respectively)
(Fig. 1, lane 2). The similarity among the molecular weights
of polypeptides below 31,000 indicates that the processing of phytepsin
follows a similar pathway in both grains and roots. P53 and P46
correspond to glycosylated prophytepsin and one-chain phytepsin,
respectively. P42 probably corresponds to an additional independent
isoenzyme characteristic of the root, as we discuss later in the text.
Thus, the roots constitute a good model for studying the expression and
processing of phytepsin during the intracellular transport to its final
cellular location, the vacuole.

View larger version (53K):
[in this window]
[in a new window]
|
Fig. 1.
Western blot analysis of phytepsin.
Lane 1, phytepsin purified from grains, 4 µg of protein;
lane 2, barley root extract, 48 µg of protein. Samples 1 and 2 were run in the same gel, but sample 1 was developed with
4-chloro-1-naphthol, and sample 2 was developed by the ECL method.
Polypeptides from purified phytepsin, whose molecular weights were
detected by MALDI-TOF-MS, are indicated on the left, and those from the
roots are indicated on the right.
|
|
Processing Rate of Phytepsin in Root Cells--
The expression and
processing of phytepsin were followed by pulse-chase labeling of the
roots with [35S]methionine/cysteine. After labeling, the
roots were homogenized under nondenaturing conditions, and the
processing products were immunoprecipitated with anti-phytepsin
antiserum and analyzed by SDS-PAGE (Fig.
2). The precursor P53 appeared during the
first 30 min of the pulse, and its half-life was estimated to be 3 h because it was no longer detected 6 h after the beginning of the chase (Fig. 2, A and B). The primary processing
products, P31 and P15, appeared 45 min after the pulse (Fig.
2A), and the further processed polypeptides, P26 and P9,
were only observed after 24 h of chase (Fig. 2B). The
processing products of phytepsin were still detected 3 days after the
chase (data not shown), indicating a slow turnover rate. The
pulse-chase analysis also showed the appearance of an additional
protein, P42, 30 min after the pulse. The intensity of this protein
remained constant for the entire 24-h chase period, which suggests that
it is not related to the proteolytic processing described above.

View larger version (76K):
[in this window]
[in a new window]
|
Fig. 2.
Processing of phytepsin in barley root
tips. Roots were labeled with 100 µCi of Pro-mixTM
(L-[35S]methionine/cysteine) for the
indicated pulse periods, and the phytepsin was immunoprecipitated with
anti-phytepsin antibody. Approximately 40 µg of total
protein/lane was separated by electrophoresis in a gradient
(10-20%) SDS-polyacrylamide gel. The gel was dried and fluorographed.
In B, the pulse time was 2 h.
|
|
Brefeldin A Causes Accumulation of the 53-kDa
Precursor--
To study whether the processing of P53 occurred before
prophytepsin reached the Golgi complex, we treated the roots with the fungal antibiotic brefeldin A, which is known to inhibit Golgi-mediated vesicular traffic by disrupting the Golgi apparatus (22). Incubation of
root cells with brefeldin A before metabolic labeling followed by
pulse-chase experiments showed the accumulation of P53, whereas for the
nontreated cells at the same chase times, P53 was processed to the
two-chain form P31 + P15 (Fig. 3).
Brefeldin A affected the processing for several hours after treatment,
and only partial processing of P53 was seen at the 2- and 5-h time
points. The experiment with brefeldin A clearly shows that processing
of P53, which leads to the formation of the two-chain form of
phytepsin, occurs only after the precursor has reached the Golgi
complex or has migrated beyond it. Brefeldin A did not have any effect on the P42 polypeptide, further corroborating our assumption that it is
independent from the described proteolytic pathway.

View larger version (54K):
[in this window]
[in a new window]
|
Fig. 3.
Inhibition of phytepsin processing in root
tips by brefeldin A. Roots were preincubated for 1 h with 7 µM brefeldin (BFA). This was followed by
radiolabeling with 100 µCi of Pro-mixTM
(L-[35S]methionine/cysteine) per sample and
incubation for the indicated pulse periods. Phytepsin was
immunoprecipitated with anti-phytepsin antibody. Approximately 40 µg
of total protein/lane was separated by electrophoresis in a
gradient (10-20%) SDS-polyacrylamide gel. The gel was dried and
fluorographed.
|
|
P53 Acquires Endo H Resistance--
Endo H removes
oligomannose but not complex-type N-linked glycans from
glycoproteins. Processing of oligomannose to complex-type glycans
occurs in the Golgi complex, and therefore, resistance to Endo H
indicates localization of a glycoprotein at or beyond the Golgi
complex. The prophytepsin sequence contains a single N-glycosylation site located in P15, and the attached
glycans in the mature P31 + P15 form are known to be of the plant
complex type (23). During the time course of a 30-min chase, the 53-kDa precursor was partially sensitive to Endo H (detected as a wider band),
with a shift of about 2 kDa (Fig. 4).
This result shows that the glycans linked to the P53 proform are of the
oligomannose type for about 30 min after the beginning of the chase,
which corresponds to the time taken for the enzyme to reach the Golgi complex, where the glycans are modified. In contrast, the P15 chain was
not sensitive to Endo H digestion at any time point (Fig.
4B), indicating that P15 contains only complex-type glycans. Therefore, we conclude that P15 is produced only when prophytepsin has
passed the Golgi complex and, most likely, in transit to or within the
vacuole.

View larger version (78K):
[in this window]
[in a new window]
|
Fig. 4.
Endo H sensitivity of phytepsin from root
tips after pulse-chase analysis. Roots were treated as described
in Fig. 2. After immunoprecipitation with anti-phytepsin antibody, half
of the sample was digested with Endo H, and the other half was
mock-digested. Approximately 40 µg protein/lane was
separated by electrophoresis in a gradient (A) 10-20% or
(B) 10% SDS-polyacrylamide gel. The gels were dried and
fluorographed.
|
|
N-glycans Delay the Proteolytic Processing of Phytepsin--
Plant
APs contain a conserved utilized glycosylation site in their
plant-specific insert (10). Since it has been suggested that the
plant-specific insert might be important for transport to the vacuole
(13), we investigated the importance of glycans for the intracellular
transport. We found that incubation of root cells with the
N-glycosylation inhibitor tunicamycin did not inhibit processing of P53 (Fig. 5), but that the
processing of P53 to P31 + P15 and P26 + P9 was actually accelerated
when glycosylation was inhibited. These results suggest that the glycan
moiety of phytepsin protects the enzyme from premature proteolytic
cleavage in the Golgi apparatus.

View larger version (56K):
[in this window]
[in a new window]
|
Fig. 5.
Processing of phytepsin in root tips in the
presence of tunicamycin A1. Roots were preincubated
for 1 h with 10 µg/ml tunicamycin A1 homologue
(TM), radiolabeled with 100 µCi of Pro-mixTM
(L-[35S]methionine/cysteine)/sample. The
phytepsin was immunoprecipitated with anti-phytepsin antibody.
ngP15, nonglycosylated P15. Approximately 40 µg of total
protein/lane was separated by electrophoresis in a gradient
(10-20%) SDS-polyacrylamide gel. The gel was dried and
fluorographed.
|
|
Expression and Purification of Phytepsin Produced in Sf9
Cells--
To enable a closer study on the processing pattern of
phytepsin, we developed a recombinant expression method for this
enzyme. Although several APs, including pepsinogen (24), procathepsin D
(5, 6) and phytepsin,2 have
been produced in bacterial expression systems, a general problem with
these methods has been the very low yield of correctly folded product.
Therefore, we chose a baculovirus-infected insect cell expression
method for this study. Sf9 cells were infected by a recombinant
baculovirus genome containing the complete coding region of
preprophytepsin. After incubation for 4 days, a prominent polypeptide
of 53 kDa was found in the medium, as analyzed by Western blotting
using anti-phytepsin antiserum (not shown). After pepstatin-agarose
column chromatography followed by ion exchange chromatography, a
typical yield of about 0.5 mg of purified protein was obtained from 1 liter of the cell medium. MALDI-TOF-MS analysis revealed two proteins
of 52,847 and 53,062 Da (Fig. 6), which migrated in SDS-PAGE as one broad band at 53 kDa (Fig.
7A, lane 1).
N-terminal sequencing gave one unambiguous sequence of EAEGLVRIAL (Fig.
8). This N-terminal sequence is similar
to that previously obtained from in vitro expression in the
presence of canine pancreatic microsomes (14), which indicates that
Sf9 cells are able to cleave the signal sequence from this plant
protein to produce a secreted recombinant prophytepsin (rP53). In an
isoelectric focusing gel, rP53 migrated to a pI of ~5.3, which was
identical to that observed for phytepsin purified from barley grains
(data not shown).

View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7.
Autoproteolytic processing of recombinant
prophytepsin in vitro. A, purified prophytepsin
from insect cell medium (lane 1) was mixed with an equal
volume of 0.2 M sodium lactate, pH 3.7, and incubated at
37 °C. Samples were removed after 7, 15, and 60 min (lanes
2-4, respectively); lane 5, incubation for 60 min with
50 µM pepstatin; lane 6, phytepsin purified
from barley grains (3 µg). B, incubation of prophytepsin
for 90 min at pH 3.7, 4.5, 5.5, and 6.5 (lanes 1-4,
respectively). All samples were analyzed by electrophoresis in 20%
SDS-polyacrylamide gels and stained with Coomassie Brilliant
Blue.
|
|

View larger version (17K):
[in this window]
[in a new window]
|
Fig. 8.
Proteolytic processing scheme of
phytepsin. , in vivo cleavage site; , in
vitro autocatalytic cleavage site. Bold Double
underlining denotes the N-terminal polypeptides from in
vivo processing (10) and overlining N-terminal polypeptides from
in vitro autoprocessing. Bold cross,
glycosylation site. Approximate locations of catalytic residues in the
active site are marked by DTG/DSG. The plant-specific domain
is shaded.
|
|
The molecular weight for rP53 predicted from the cDNA is 51,779. The higher molecular weight observed by MALDI-TOF-MS is due to the
presence of oligosaccharides, confirmed by positive staining observed
using the periodic acid Schiff method to detect the protein in the gel
and indicates that the single potential glycosylation site is occupied.
The observed main peak at 52,847 is consistent with the presence on
pro-phytepsin of an N-linked oligosaccharide with the
structure Man3GlcNAcFucGlcNAc characteristic of proteins expressed in
insect cells (predicted molecular weight of 52, 839 (25), assuming the
presence of one sodium atom). The additional signals at 53,062 and
53,273 are probably due to the addition of one or two matrix molecules
(sinapinic acid) with the concomitant loss of water (M + nx206) or to
the presence of larger oligosaccharide chains.
Autoproteolytic Processing of Recombinant Phytepsin in
Vitro--
To study the capability of prophytepsin for autoproteolytic
processing, we incubated rP53 in buffers over the pH range 3.7-6.5 at
37 °C and removed the samples at various time points. At pH 3.7, the
processed polypeptides of 36 and 17 kDa (rP36 and rP17, respectively)
were detected after 7 min of incubation, indicating rapid cleavage of
prophytepsin to a two-chain form (rP36 + rP17) (Fig. 7A,
lane 2). The two polypeptides were further processed stepwise to the final products of 28 and 11 kDa (rP28 and rP11, respectively). At pH 3.7, processing of rP30 to rP28 occurred slower
than at pH 4.5, whereas processing of rP17 to rP11 was much slower at
pH 4.5 than at pH 3.7. Processing was severely inhibited at pH values
above 4.5 (Fig. 7B). The addition of the AP inhibitor
pepstatin to the incubation buffer prevented all processing and showed
that the processing was of an autoproteolytic nature (Fig.
7A, lane 5). Polypeptides rP28 and rP11 were the final processing products over the pH range 3.7-4.5 and up to 90 min
of incubation; furthermore, in some extended incubations at pH 4.5, no
additional processing was observed after 5 h. The sizes of the
proteolytic processing products were very similar to those of the
polypeptides present in phytepsin purified from barley grains (Fig.
7A, lane 6). To obtain detailed information on
the cleavage sites, we separated the processing products of rP53 by
SDS-PAGE, electroblotted them onto a PVDF membrane, and subjected
several of them to N-terminal sequencing. rP36, rP30, and rP28 each
gave one sequence, whereas rP17 and rP11 both gave two or three almost
identical sequences, indicating slight heterogeneity in these
processing sites. The sequences and processing sites are illustrated in
Fig. 8. The molecular weights of the polypeptides resulting from the
autolysis of rP53 were confirmed by MALDI-TOF-MS.
Specific Activity of Recombinant Phytepsin--
To test the
proteolytic efficiency of the purified and autoproteolytically
processed recombinant phytepsin, we subjected prophytepsin to
autocatalytic activation at pH 4.0 for 1 h at 37 °C and assayed the proteolytic activity against hemoglobin at 37 °C. The observed mean value for purified recombinant phytepsin from three individual expressions was 628 units/mg (range 456-788 units/mg), whereas the
purified enzyme from grains gave a value of 882 units/mg. These values
are consistent with the previously reported value for affinity-purified
phytepsin from grains (range 534-668 units/mg at 30 °C) (9) and
indicate that expression of recombinant phytepsin in
baculovirus-infected insect cells yields a correctly folded and active enzyme.
 |
DISCUSSION |
Phytepsin is synthesized and translocated into the rough
endoplasmic reticulum as a preproenzyme of 54 kDa, according to the cDNA sequence, where it becomes N-glycosylated at its
single glycosylation site. The enzyme undergoes several proteolytic
cleavages to produce the mature two-chain forms present in barley
grains, roots, and other tissues:
1) The first processing step consists of the removal of the signal
sequence of 25 amino acid residues upon entering the endoplasmic reticulum, yielding a product of 51,779 Da. Even though this cleavage has not been studied in vivo in barley, the cleavage site of
the signal sequence occurs between Ser-25 and Glu-26, as has been detected in vitro using microsomal membranes (14) or as
shown here for the recombinant preprophytepsin from
baculovirus-infected insect cells (Fig. 8). Concomitantly, the
N-glycan Glc3Man9GlcNAc2 is transferred onto the single glycosylation site of the enzyme and is
further processed to the structure
Man 3,6(±Man 3,6)(Xyl 2)Man 4 GlcNAc 4(Fuc 3)GlcNAc in
barley (23), resulting in an increase in molecular weight to
~53,000. Alternatively, when the enzyme is expressed in Sf9
insect cells, the oligosaccharide is processed to the structure
Man 6(Man 3)Man 4GlcNAc 4 (Fuc 6)GlcNAc, which corresponds to the glycosylated precursor rP53 (Fig. 6).
2) The following processing steps include the removal of the N-terminal
propeptide sequence of 41 residues and the formation of a two-chain
form of the enzyme. The two polypeptides (P31 + P15) were detected
1 h after synthesis. Considering the N-terminal sequence of P15
(Fig. 8), a molecular weight of 14,046 is predicted from the cDNA
sequence, and the addition of the glycan at the single glycosylation
site accounts for the observed extra 1.2 kDa. Thus, no processing at
the C terminus of this subunit occurred. However, if a single
proteolytic cleavage had occurred on the 53-kDa polypeptide to produce
P15, the larger subunit should be 38 kDa, taking into account the N
terminus of P31 (Fig. 8). This difference in molecular weight can be
attributed to proteolytic cleavage of a 5-kDa peptide upstream of
Ala-378, in the middle of the coding sequence of the enzyme. This
processing step is inhibited by brefeldin A and P53 accumulates.
Brefeldin A is known to inhibit transport beyond the Golgi complex, so
the results suggest that the proteolytic processing occurs after
phytepsin has passed the Golgi, most probably in the vacuole. Further
corroborating these results, P53 is sensitive to Endo H for 30 min, the
time required for P53 to reach the Golgi complex and for the glycans to
be processed from oligomannose to complex type by Golgi glycosidases and glycosyltransferases, whereas P15 is always Endo H-resistant, indicating that it is produced only after the glycans have been processed, and thus, the phytepsin has passed the Golgi complex.
3) Finally, P31 and P15 are further processed, resulting in
polypeptides P27 and P9, respectively. P27 results from P31 after C-terminal processing, since the N termini of P31 and P27 are identical
(Fig. 8). P15 is probably processed only from the N-terminal side,
since removal of 44 residues (~ 5 kDa) including the
N-linked glycosylation site with attached glycans decreases
the size to P9. This processing step occurs only 24 h after
synthesis in vivo.
In addition to the polypeptides described in the processing pattern
described above, a polypeptide of approximately 42 kDa is clearly
visible in a Western blot from roots (Fig. 1) and in the pulse-chase
experiments (Figs. 2-5). However, since the intensity of P42 remains
constant during the pulse-chase experiments, we propose that P42 is not
part of the processing scheme of P53, but rather, that it probably
constitutes a different AP-like isoenzyme from the root. Further
supporting a hypothesis that plant tissues contain several AP-like
enzymes, Nakano et al. (26) recently showed that chloroplast
nucleoids from tobacco cells contain a DNA-binding protein of 41 kDa
with clear sequence homology to the AP family, including the conserved
active-site residues Asp-Thr-Gly/Asp-Ser-Gly. Chen and Foolad (27) also
reported the cloning and characterization of an AP-like proteinase of
~45 kDa (nucellin) from barley ovaries, which is abundantly expressed
after pollination. Furthermore, a putative AP is encoded by part of the
BARE-1 retroelement in the barley genome (28).
Interestingly, the aforementioned AP-like enzymes do not contain a
plant-specific domain that in phytepsin is cleaved and results in the
two-chain enzyme form (Fig. 8).
Plant APs contain a conserved occupied N-glycosylation site
within the plant-specific insert and we have investigated the importance of the glycosylation for the intracellular transport and
processing of the enzyme. The results obtained with tunicamycin suggest
that the glycans are not essential for the transport or processing of
phytepsin, since proteolytic processing occurred despite the presence
of this compound. In fact, the rate of processing was accelerated by
treatment with tunicamycin, suggesting that the glycans may have a role
in protecting the enzyme from premature proteolytic cleavage.
The recombinant prophytepsin (rP53) expressed in baculovirus-infected
insect cells was detected in the medium as a glycosylated proenzyme.
The enzyme contained glycans of the insect complex type, indicating
that rP53 had passed the Golgi complex before secretion outside the
cell. It is known that cathepsin D and many other animal proteins are
targeted largely to the lysosome via the mannose-6-phosphate receptor
pathway (3). Such a targeting mechanism to the plant vacuoles has not
been described. Apparently, phytepsin was secreted from the insect
cells because it did not contain the appropriate intracellular
targeting signal functional in insect cells. It is also possible that
phytepsin was secreted because the cells could not hold the large
amount of foreign protein they were producing.
When recombinant prophytepsin was incubated at pH 3.7-4.5, it
underwent autoprocessing, even though the cleavage sites were distinct
from those occurring in vivo (for a comparison, see Fig. 8).
Results from N-terminal sequencing and the decrease in molecular masses
revealed that 1) the 53-kDa glycosylated prophytepsin was autoprocessed
in vitro to rP36 and rP17, with cleavage occurring around
Val-365; 2) rP36 was further processed from both N- and C-terminal
sides to yield the final proteolytic product, rP28; and 3) the
glycosylated polypeptide rP17 was processed only from its N-terminal
side to rP11 (Fig. 8).
The recombinant phytepsin possesses features similar to those of
phytepsin purified from barley grains (9, 16). It hydrolyzes peptide
bonds that usually contain at least one hydrophobic residue in either
side of the bond to be cleaved (Fig. 8), and the optimal pH for the
hydrolytic activity is 3.7-4.5 (Fig. 7). Furthermore, the enzymatic
efficiency of the recombinant phytepsin on hemoglobin was almost as
high as that measured for phytepsin purified from grains. Additionally,
in an ongoing related study, we have crystallized rP53, and a
preliminary x-ray analysis extending to 2.4 Å resolution shows that
the overall fold of prophytepsin, excluding the plant-specific domain,
is similar to that of mammalian
APs.3
Although the sizes of the autocatalytic processing intermediates and
final products of recombinant phytepsin closely resemble the in
vivo P31 + P15 and P27 + P9 forms of barley grain phytepsin, the
cleavage products formed in vitro are not exactly the same. rP28 still retains seven residues belonging to the prosequence, and
rP11 contains 13-14 extra residues in the N-terminal side. In
addition, the very hydrophilic and, thus, unfavorable residues, Arg-Ser
in the N terminus of P31 and P27, Arg-377 in P15, and Gly-Glu in P9
(Fig. 8), are flanking the final maturation sites in phytepsin purified
from barley grains. Therefore, although in vitro processing
of rP53 results in an active enzyme, the completion of maturation
in vivo probably requires other
proteinase/exopeptidase(s).
Both autocatalytic and heterocatalytic processing and activation
mechanisms are known for APs. For example, the activation of mammalian
lysosomal procathepsin D, the closest counterpart to phytepsin,
involves cysteine proteases in the lysosomes (4). However, in
vitro studies show that procathepsin D undergoes a pH-dependent intramolecular proteolysis that removes 26 residues from the 44-residue propart, yielding an active one-chain
enzyme, pseudocathepsin D (5, 6). On the other hand, when a mutant of
procathepsin D that was unable to autoactivate itself to
pseudocathepsin D in vitro was expressed in mouse cells, it
nevertheless was transported to the lysosomes and was processed
normally to the mature two-chain enzyme (7). Based on several
processing studies on intracellular APs, it thus seems likely that
alternative activation mechanisms, including autocatalytic and
heterocatalytic steps, exist for intracellular APs, as we have shown
for phytepsin in the present study. These mechanisms depend on the pH
and processing endo- and exopeptidases present in the particular
intracellular compartments traversed by the AP along its route to its
final location in the cell.
 |
ACKNOWLEDGEMENTS |
We gratefully acknowledge the scientific
advice of Dr. Pedro Fevereiro and Lucinda Neves and Anne Arthur for her
editorial assistance.
 |
FOOTNOTES |
*
This work was supported by Grants PBICT/C/BIO/2025/95 and
PRAXIS PCNA/P/BIO/52/96, Portugal (to J. C.), by the National
Institutes of Health (NCI) Department of Health and Human Services,
under contract with Advanced Bioscience Laboratories (to J. K.,
T. D. C., and A. W.), and by the protocol Junta Nacional de
Investigacae Científica e Tecnológica/British Council
423/RU (to J. C. and D. A. A.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
§
Recipients of fellowships BPD/4154/94 and BPD/11839/97, Portugal, respectively.
Supported in part by the Academy of Finland.
||
To whom correspondence should be addressed: ITQB/IBET,
Apartado 12, P-2780 Oeiras, Portugal. Tel.: 351-1-4469462; Fax:
351-1-4411277; E-mail: jcosta{at}itqb.unl.pt.
The abbreviations used are:
AP, aspartic
proteinase; Endo H, endoglycosidase H; PAGE, polyacrylamide gel
electrophoresis; PVDF, polyvinylidene difluoride; MALDI-TOF MS, matrix-assisted laser desorption ionization time-of-flight mass
spectroscopy; r-, recombinant.
2
J. Kervinen, unpublished information.
3
J. Kervinen and A. Zdanov, unpublished information.
 |
REFERENCES |
-
Davies, D.
(1990)
Annu. Rev. Biophys. Biophys. Chem.
19,
189-215[CrossRef][Medline]
[Order article via Infotrieve]
-
Rawlings, N. D.,
and Barrett, A. L.
(1995)
Methods Enzymol.
248,
105-120[Medline]
[Order article via Infotrieve]
-
Metcalf, P.,
and Fusek, M.
(1993)
EMBO J.
12,
1293-1302[Medline]
[Order article via Infotrieve]
-
Samarel, A. M.,
Ferguson, A. G.,
Decker, R. S.,
and Lesch, M.
(1989)
Am. J. Physiol.
257,
C1069-C1079[Abstract/Free Full Text]
-
Conner, G. E.,
and Richo, G.
(1992)
Biochemistry
31,
1142-1147[CrossRef][Medline]
[Order article via Infotrieve]
-
Beyer, B. M.,
and Dunn, B. M.
(1996)
J. Biol. Chem.
271,
15590-15596[Abstract/Free Full Text]
-
Richo, G. R.,
and Conner, G. E.
(1994)
J. Biol. Chem.
269,
14806-14812[Abstract/Free Full Text]
-
Barrett, A. J.
(1997)
Eur. J. Biochem.
250,
1-6[Medline]
[Order article via Infotrieve]
-
Sarkkinen, P.,
Kalkkinen, N.,
Tilgmann, C.,
Siuro, J.,
Kervinen, J.,
and Mikola, L.
(1992)
Planta
186,
317-323
-
Runeberg-Roos, P.,
Törmakangas, K.,
and Östman, A.
(1991)
Eur. J. Biochem.
202,
1021-1027[Medline]
[Order article via Infotrieve]
-
Cordeiro, M. C.,
Xue, Z. T.,
Pietrzak, M.,
Pais, M. S.,
and Brodelius, P. E.
(1994)
Plant Mol. Biol.
24,
733-741[CrossRef][Medline]
[Order article via Infotrieve]
-
Asakura, T.,
Watanabe, H.,
Abe, K.,
and Arai, S.
(1995)
Eur. J. Biochem.
232,
77-83[Medline]
[Order article via Infotrieve]
-
Guruprasad, K.,
Törmakangas, K.,
Kervinen, J.,
and Blundell, T. L.
(1994)
FEBS Lett.
352,
131-136[CrossRef][Medline]
[Order article via Infotrieve]
-
Runeberg-Roos, P.,
Kervinen, J.,
Kovaleva, V.,
Raikhel, N. V.,
and Gal, S.
(1994)
Plant Physiol.
105,
321-329[Abstract]
-
Marttila, S.,
Jones, B. L.,
and Mikkonen, A.
(1995)
Physiol. Plant.
93,
317-327[CrossRef]
-
Kervinen, J.
(1998)
in
The Handbook of Proteolytic Enzymes (Barrett, A. J., Woessner, F. F., and Rawlings, N., eds), Academic Press, Inc. London, in press
-
Runeberg-Roos, P.,
and Saarma, M.
(1998)
Plant J.
15,
139-145[CrossRef][Medline]
[Order article via Infotrieve]
-
Pinkas-Kramarski, R.,
Shelly, M.,
Glathe, S.,
Ratzkin, B.,
and Yarden, Y.
(1996)
J. Biol. Chem.
271,
19029-19032[Abstract/Free Full Text]
-
Laemmli, U. K.
(1970)
Nature
227,
680-685[CrossRef][Medline]
[Order article via Infotrieve]
-
Costa, J.,
Grabenhorst, E.,
Nimtz, M.,
and Conradt, H. S.
(1997)
J. Biol. Chem.
272,
11613-11621[Abstract/Free Full Text]
-
O'Reilly, D. R.,
Miller, L. K.,
and Luckow, V. A.
(1994)
Baculovirus Expression Vectors: A Laboratory Manual, Oxford University Press, New York
-
Satiat-Jeunemaitre, B.,
Cole, L.,
Bourett, T.,
Howard, R.,
and Hawes, C.
(1996)
J. Micros. (Oxf.)
181,
162-177
-
Costa, J.,
Ashford, D. A.,
Nimtz, M.,
Bento, I.,
Frazão, C.,
Esteves, C. L.,
Faro, C. J.,
Kervinen, J.,
Pires, E.,
Veríssimo, P.,
Wlodawer, A.,
and Carrondo, M. A.
(1997)
Eur. J. Biochem.
243,
695-700[Medline]
[Order article via Infotrieve]
-
Lin, X.,
Wong, R. N. S.,
and Tang, J.
(1989)
J. Biol. Chem.
264,
4482-4489[Abstract/Free Full Text]
-
Grabenhorst, E.,
Hofer, B.,
Nimtz, M.,
Jäger, V.,
and Conradt, H. S.
(1993)
Eur. J. Biochem.
215,
189-197[Medline]
[Order article via Infotrieve]
-
Nakano, T.,
Murakami, S.,
Tsubasa, S.,
Yoshida, S.,
Yamada, Y.,
and Sato, F.
(1997)
Plant Cell
9,
1673-1682[Abstract]
-
Chen, F.,
and Foolad, M. R.
(1997)
Plant Mol. Biol.
35,
821-831[CrossRef][Medline]
[Order article via Infotrieve]
-
Manninen, I.,
and Schulman, A. H.
(1993)
Plant Mol. Biol.
22,
829-846[CrossRef][Medline]
[Order article via Infotrieve]
Copyright © 1998 by The American Society for Biochemistry and Molecular Biology, Inc.

CiteULike Complore Connotea Del.icio.us Digg Reddit Technorati What's this?
This article has been cited by other articles:

|
 |

|
 |
 
J. Chen, J. Ding, Y. Ouyang, H. Du, J. Yang, K. Cheng, J. Zhao, S. Qiu, X. Zhang, J. Yao, et al.
From the Cover: A triallelic system of S5 is a major regulator of the reproductive barrier and compatibility of indica-japonica hybrids in rice
PNAS,
August 12, 2008;
105(32):
11436 - 11441.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
O. Zschenker, D. Oezden, and D. Ameis
Lysosomal Acid Lipase as a Preproprotein
J. Biochem.,
July 1, 2004;
136(1):
65 - 72.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
K. Tormakangas, J. L. Hadlington, P. Pimpl, S. Hillmer, F. Brandizzi, T. H. Teeri, and J. Denecke
A Vacuolar Sorting Domain May Also Influence the Way in Which Proteins Leave the Endoplasmic Reticulum
PLANT CELL,
September 1, 2001;
13(9):
2021 - 2032.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
C. Faro, M. Ramalho-Santos, M. Vieira, A. Mendes, I. Simoes, R. Andrade, P. Verissimo, X.-l. Lin, J. Tang, and E. Pires
Cloning and Characterization of cDNA Encoding Cardosin A, an RGD-containing Plant Aspartic Proteinase
J. Biol. Chem.,
October 1, 1999;
274(40):
28724 - 28729.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
C. Frazao, I. Bento, J. Costa, C. M. Soares, P. Verissimo, C. Faro, E. Pires, J. Cooper, and M. A. Carrondo
Crystal Structure of Cardosin A, a Glycosylated and Arg-Gly-Asp-containing Aspartic Proteinase from the Flowers of Cynara cardunculus L.
J. Biol. Chem.,
September 24, 1999;
274(39):
27694 - 27701.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
P. C. White, M. C. Cordeiro, D. Arnold, P. E. Brodelius, and J. Kay
Processing, Activity, and Inhibition of Recombinant Cyprosin, an Aspartic Proteinase from Cardoon (Cynara cardunculus)
J. Biol. Chem.,
June 11, 1999;
274(24):
16685 - 16693.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
C. Egas, N. Lavoura, R. Resende, R. M. M. Brito, E. Pires, M. C. P. de Lima, and C. Faro
The Saposin-like Domain of the Plant Aspartic Proteinase Precursor Is a Potent Inducer of Vesicle Leakage
J. Biol. Chem.,
December 1, 2000;
275(49):
38190 - 38196.
[Abstract]
[Full Text]
[PDF]
|
 |
|
Copyright © 1998 by the American Society for Biochemistry and Molecular Biology.
|
Advertisement
Advertisement
|