BIOCHEMICAL
Vol. 168, No. 2, 1990 April 30, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 878-885
STRUCTDRAL EVIDENCE FOR TWO ISOZYMIC FORMS AND THE CARBOBYDRATE ATTACHMENT SITE OF HUMAN GASTRIC CATHEPSIN E S.B.P.Athauda,
Osamu Matsuzaki, Takashi and Kenji Takahashil
Department of Biophysics The University of *Department
of
Kageyama
and Biochemistry, Faculty Tokyo, Hongo, Tokyo 113,
*
of Science, Japan
Primate Research Institute, Biochemistry, University, Inuyama, Aichi 484, Japan
Kyoto
Received March 20, 1990 SUMMARY: The amino acid sequences in the NH2-terminal region and some other parts of human gastric cathepsin E were investigated. The NH2-terminal sequencing revealed that the cathepsin E preparation which had been activated at pH 4.0 contained one major and one minorisozymes in an approximate molar ratio of 3 : sequence of the former was very similar to 1. The NH2-terminal but partly different from that predicted from cDNA sequencing by whereas the latter had an NH2-terminal sequence Azuma et al., identical with the predicted sequence. These results provide structural evidence for the presence of at least two isozymic forms in human gastric cathepsin E. In addition, the site of carbohydrate attachment was elucidated by isolation and analysis of a glycopeptide fraction from an enzymatic digest of cathepsin E. A single carbohydrate chain was deduced to be attached to the asparagine residue at position 34 in the major isozyme and to the corresponding asparagine residue in the minor isozyme. 01990 Academic Press,
Inc.
Human gastric protease proteinase pepsinogens this
cathepsin
present
about
size.
in
E (previously
D-like
(2).
85,000
The enzyme
to
human gastric
apparent
The enzyme
was shown should
to
All rights
0 1990 by Academic Press, of reproduction in any form
Inc. reserved.
two
and different
from
we first
a molecular
subunits
be addressed.
moving an acid
be a glycoprotein,
878
etc.)
and investigated
was shown to have of
as slow is
D. In 1980,
0006-291X/90$1.50 Copyright
(2)
mucosa
homogeneity
and composed
1To whom correspondence
called
proteinase
A and C and cathepsin
proteinase
properties of
(I),
cathepsin
of
purified several weight
an identical containing
2
Vol.
168, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
glucosamine
and 8 mannose residues
this
were
enzyme
further
studied
per
RESEARCH COMMUNICATIONS
subunit.
by Samloff
The properties
et
al.
of
(3) and by us
(4). Very
recently,
Azuma et
sequence
of
gastric
analysis
of
cell
line.
residue
cDNA clones
proenzyme
derived amino
determined
acid
sequence
on the
protein
signal
So far,
sequence
the
acid
from
the
adenocarcinoma includes
peptide, however,
and the
site
a 379and three
there
of human gastric
level
amino
E predicted a gastric
sites.
amino acid
reported
from
and a 17-residue
glycosylation on the
(5) have
procathepsin
The predicted
potential report
human
al.
is
cathepsin of
no E as
carbohydrate
attachment. In
the
sequences
present
of
the
presence
cathepsin
4.0. to
which
The major but
sequence
sequence.
These results
presence
addition, in
the
MATERIALS
of
we have
The NH2-terminal
that
hand,
the
with
provide
isozymes-
elucidated
forms
sequence
minor
isozyme
predicted
first
in human gastric
the
site
of
cathepsin
very
similar the
the
evidence cathepsin
carbohydrate
at pH
cDNA
had an NH2from
structural
of
revealed
enzyme
from
acid
parts
in the
predicted
that
the
amino
sequencing
to the mature
from
identical
isozymic
the
and some other
had an NH2-terminal
other
terminal
the
region
and one minor
different On the
investigated
had been activated
isozyme
partly
sequencing.
E.
of one major
E preparation
we
NH2- terminal
the
human gastric
study,
cDNA for E. In
attachment
enzyme. AND
METHODS
Materials: Human gastric cathepsin E was purified essentially as described previously (2,4) and the major component (SMP-I(4)) was peptides and Staphylococcus aureus V8 peptides were used. Tryptic prepared from the reduced and carboxymethylated (RCm-)protein and fractionated by high performance liquid chromatography on a TSKGel ODS-120T column ( 0.4 x 25 cm) in a manner as described The glycopeptide fraction was previously (6) (data not shown). prepared by digestion of the protein with thermolysin and B. subtilis aminopeptidase followed by chromatography on a Sephadex 879
BIOCHEMICAL
Vol. 166, No. 2, 1990
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
G-25 column (1.6 x 40cm) as described previously (2). Reagents for automated amino acid sequencing were obtained from Applied Biosystems. Reagents for manual Edman degradation were purchased from Wako Pure Chem. Ind.,Ltd. Other reagents used were of the highest grade available. Amino acid sequence analysis: The NH -terminal amino acid si by using a protein sequence of the protein was determine sequencer model 477A/12OA in the presence of pepstatin (about 2fold molar excess). The protein was analyzed before or after treatment at pH 4.0 and purification on pepstatin-Sepharose. In the latter case the enzyme was dialyzed against 0.05 M sodium acetate buffer (pH 4.0)-0.2 M NaCl at 4OC for 3 h and applied to a pepstatin-Sepharose column (1.0 x l.Ocm) equilibrated with the same buffer. After washing the column with same buffer (60 ml), the protein was eluted with 0.05 M Tris-HCl buffer (pH 8.0)-l M NaCl, and further purified by mono Q chromatography (data not After desalting through a Sephadex G-25 column, the shown). protein fraction was submitted to automated sequencing. Peptides were sequenced partly by the sequencer and partly by a modification (6,7) of the manual method (8) of Edman degradation. RESULTS
AND
DISCUSSION
E2-Terminal
amino acid
When the directly
by
amino acid that
the
human gastric the
with
the
pyroglutamyl
residue,
other
after
the
original
of
sequence
could
the of
mature amino
be deduced
an asparagine (Fig.1).
the
The
blocked.
This
is
residue
of
converted
followed
procathepsin
E
mature
at
enzyme revealed
and the
except
for
the
sequence
the
the
corresponding of
880
residues
the
NH2-
suggest before
that
pH 4.0-
converted
pH 4.0. presence
NH2-terminal residue
affinity
obtain
was
a
On the
by
These results
1.
pro-form
acids
into
Edman degradation.
we could
enzyme
in
indicating
was the
this
into
sequence
is
to
preparation
that
autocatalytically
which
resistant
(PTH)-
amount,
glutamine
be easily
as shown in Fig.
enzyme and
sequence
may
on pepstatin-Sepharose,
treatment
major
is
pH 4.0-treatment
sequences
degradation
protein
NH2-terminal
which
was analyzed
no phenylthiohydantoin
the
putative
E (5),
E preparation
in a significant
of
procathepsin
purification
almost
be detected
NH2-terminal
hand,
cathepsin
sequencer,
could
consistent
terminal
sequence
Edman of
one
39 residue at
position
procathepsin 22-37
was further
34, E
Vol.
168, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Protein 1
10
Major
:
IQFTExxxMDQMAKEPLINYLDM
Minor
:
7'"' IQFTES?SMDQSAKE--777777777777777 37
cDNA(5)
:
20
777777777777777 10 15
40
50
59
IQFTESCSMDQSAKEPLINYLDM
Protein 21 Major
:
V8-peptide
:
cDNA(5)
:
30
39
EYPGTISIGSPPQ?FTVIF--7777777777777777-?77
Y F G T I S I G S P P Q ? F(T)V 7777777777777777 60
--70
70
EYFGTISIGSPPQNFTVIF---
Fis.1. Comparison of the NH2-terminal amino acid sequences of human gastric cathepsin E with the corresponding sequence of the procathepsin E predicted from its cDNA sequence (5). The protein sequence was obtained with human gastric cathepsin E sample treated at pH 4.0 and then affinity-purified on pepstatinSepharose. The sequence of residues 22 to 37 in the major sequence was further confirmed by sequencing of a V8-peptide. Procathepsin E numbering is used for the sequence predicted from the cDNA sequence. deletion. each cycle of Edman degradation. (?), ambigu%s identificaty&.
confirmed
by isolation
peptide
as shown
similar
to
that
in Fig.
predicted
a few positions. the
al.
(5)
in
the
intermolecular
Further, which
suggested
lacks
presence used.
In of
approximately sequenced
this
ratio
at
this
1 : 3, and the
simultaneously.
This
the
cysteine
position
in
the
sequence 881
of
the
may be
the
procathepsin
also
revealed
major
15
major
of a serine
to
the
Azuma
formation.
9 instead
enzyme
NH2-terminal
deletion
Since
mature
sequence
at
may be involved
bond
degradation
very
differs
residue(s)
in
a V8-
is
42-44).
residue
position
Edman
of
is
disulfide
corresponding
sequence
but
bond formation. other
of
obtained
E residues
intermolecular
addition,
thus
difference
cysteine
a methionine
the
analysis
cDNA sequence,
(procathepsin
residue,
one minor
The molar
the
disulfide
the
we found
sequence.
that
this
for
occupies
The sequence
from
sequence
et
responsible
1.
sequence
The most notable
Ser-CyS-Ser
sequence
and partial
to
the
preparation sequence
residues
appeared
E
was
could
be identical
be
Vol.
BIOCHEMICAL
168, No. 2, 1990
with
that
results
predicted
provide
two
isozymes,
are
products
al.
(5)
a gastric
in
of
different
the
on the
Amino acid
In addition
sequence
to the
protein.
in the
to obtain
For example,
latter
seems to
residues
based
proteinases.
but
proline,
COOB-terminal
of
peptides
coincide
indeed
and 372-379 predicted
residues
of
two
peptide 80-88,
were
This
is
et al.
E.
about
can be assumed
with
The those
of
and
site
882
is
not
to
of
had
the
arginine
or
from
the
these
124-138,
with
the
respectively,
three 258-268,
respectively,
sequence
acid
aspartic
to be derived
3%. Further,
360-367,
the
The
peptide
residues
coincide
of
aspartic
among
which
and the
to
acid
respectively.
peptide),
VS-peptides,
amino
E
sequenced
sequences
procathepsin
confirmed
the
regions
tryptic
of
the
RCm-cathepsin
homology one
protein.
297-310,
pro-
1, some of
from
active
COOH-terminal
(5) of
other
the
hand,
COOH-terminal
sequence
sequences tryptic
(the
the
long.
were partially
sequence
so it
the
cathepsin
GCQAIVDTGTS---,
other
VGLAPAVP, the
This
and cancer
that
COOH-terminal
one of
the
On the
sequence, lysine
and
on
of rat
et
may be
made by Yonezawa
information
two V8-peptides
contain
normal
36 residues
peptides
and the
be TGSLSGIIGADQVSV---
which
genes
shown in Fig.
tryptic
interior
E,
regions
V8-peptide
and the
least
mucosa.
indicate
E (5) is
NH2-terminal
sequenced
sequences
results
at
corresponds
two
the
These
by Azuma
line
the
in
recently
V8-peptides also
that
present
of other
of
in the human gastric
prediction
sequences
presence
cell
proportion
the
1).
of human cathepsin
possibility
procathepsin
from
(9) based
were
isozyme
a different
the
the
adenocarcinoma
minor
In addition,
sequence
for
(Fig.
The cDNA obtained
genes.
an interesting
cells.
other
evidence
different
to the
expressed
cDNA sequence
and hence isozymogens,
from
suggests
the
structural
of
apparently
from
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
in the of
the
partial another
sequences in
of the
Vol.
168,
No.
predicted
sequence
therefore,
(5)
protein
differences
on this
of
the
NH2-terminal to
COMMUNICATIONS
far
the
sequence
necessary
as
examined,
sequence
of
except
for
Further
draw
the the
protein
a definite
conclusion
attachment
glycopeptide
80 % from
a thermolysin-aminopeptidase
chromatography The of
mol
acid
peptide)
Gly
0.78,
24h).
In
addition,
after
7.5h),
glucosamine
(1.80
amino
acid
Ile
after
mannose
chromatography
sequence
degradation
peptides,
: 2. Their
deduced
analysis revealed
major
as of
the
I and
II,
sequences
as
shown
were
below. I:
column
as
mol
(8
peptide an
per
0.95,
of
mol
llO°C,
HCl,
100°C,
found
The
by mannual
approximate a mixture
degradation,
by
peptide)
(2).
was
Pro
HCl,
was
fraction
as
Edman
N
peptide)
mol
E
described
Glu
(1
of
cathepsin
(6N
fraction
analyzed
Upon
of
a yield
glycopeptide
1.03,
previously
peptide
in
Ser
the
hydrolysis
described
this
that
of
hydrolysis
acid per
in
digest
of
1.00,
acid
mild
and
G-25
Asp
mol
isolated
composition
was
0.63
analysis
was
a Sephadex
amino
1.88,
gas-liquid
fraction
on
(2).
per
acid
so
regions.
A single
(mol
In
between
predicted
is
RESEARCH
E.
detected
the
carbohydrate
previously
3
procathepsin was
however,
BIOPHYSICAL
point.
The site
two
AND
and in
sequencing,
over
of
no difference
actual
by
BIOCHEMICAL
2, 1990
by amino Edman
a mixture
of
molar
ratio
of
and
could
be
each
residue
Ile-Gly-Ser-Pro-Pro-Gln-Asx 7777777
II:
Gly-Ser-Pro-Pro-Gln-Asx 777777
could
be
identified
terminal
residue.
residue.
However,
by
dansylation
residue
is
as No it
PTH-amino
could
followed thought
a PTH-amino
to
acid acid
be identified by
acid
be originally 883
except
was as
for
obtained
the
COOH-
from
this
dansyl-aspartic
hydrolysis. an asparagine
acid
Therefore, to
this which
the
BIOCHEMICAL
Vol. 168, No. 2, 1990
carbohydrate
chain
coincide
with
cathepsin
the
the
position
34
(procathepsin position
in
described
glycosylation obtained
The minor
since from
consistent
the
same
the
site
0-glycosylation
fact
the is
also
in
the
only I
thought
the
the
results
residue
at
sequence
asparagine
at
one
site
result
the
carbohydrate
of was
is also
potential
73) and two E
as
1)
fraction
This
at
of
corresponding
to be the
are
residue of
73 in
(Fig.
the
procathepsin
site
monkey pepsinogen
at
E residue
same aspargine
major
Edman degradation
VB-peptide
there
the
the
68) to
why the
isozymes.
that
(procathepsin
34 in
glycopeptide
of both
sites
Interestingly,
a single
to
E isozyme
in the
is also
glycopeptides
asparagine
asparagine
isozyme only
be the
the
the
Therefore,
explains
as
of
67 (or
(5).
to
This
well
mixture
with
glycosylation
Japanese
as
29)
residues
be identified
protein
in the
(or
cathepsin
73).
not
above.
position
major
E residue
mature
or
site
the
34 could
28
E sequence
glycosylation
RESEARCH COMMUNICATIONS
The sequences
sequence
procathepsin
establish
position
bound. residues
E isozyme
predicted
the
is
AND BIOPHYSICAL
N-
potential
sequence
(5).
corresponding attachment
in
(10).
ACKNOWLEDGMENTS
We are
most
Nagoya)
for
We also
thank
of
Medicine,
Hideshi
Inoue
the
grateful
to Dr.
generous
Dr.
supply
Takayuki
The University for
their
help
Masanori of
specimens
Takahashi, of
Ukai
Tokyo),
and useful
Dr.
of
(Ukai
human stomachs.
Masao Ichinose Dr.
Hospital,
Masao Tanji
(Faculty and Dr.
suggestions.
REFERENCES
1.
2. 3. 4.
Samloff, I.M. (1969) Gastroenterol. 57, 659-669. Kageyama, T., and Takahashi, K. (1980) J. Biochem. 87, 725735. Samloff, I. M., Taggart, R. T., Shiraishi, T., Branch, T., Reid, W. A., Heath, R., Lewis, R. W., Valler, M. J., and Kay, J. (1987) Gastroenterol. 93, 77-84. Matsuzaki, O., and Takahashi, K. (1988) Biomed. Res. 9, 515523. 884
Vol. 168, No. 2, 1990
5. 6. 7.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Azuma, T., Pals, G., Mohandas, T. K., Couvreur, J. M., Taggart, R. T. (1989) J. Biol. Chem. 264, 16748-16753. Takahashi, K. (1987) J. Biol. Chem. 262, 1468-1478. van Eerd, J. -P., and Takahashi, K. (1976) Biochemistry
and 15,
1171-1180. 8.
9.
Edman, P., and Henschen, A. (1975) in Protein Sequence Determination (Needleman, S. B., ed.) 2nd Ed, pp.232-279, Springer, New York. Yonezawa, S., Takahashi, T., Ichinose, M., Miki, K., Tanaka, J and Gasa, S. (1990) Biochem. Biophys. Res. Commun. 166, 10;2-1
10.
038.
Kageyama, T., and Takahashi, Commun. 74, 789-795.
K.
885
(1977)
Biochem.
Biophys.
Res.