:

Cloning, expression, purification, and

biological activity of recombinant native
and variant human alpha
1-antichymotrypsins.

J. Biol. Chem. 1990, 265:1199-1207.

Access the most updated version of this article at http://www.jbc.org/content/265/2/1199

Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites.
Alerts:
When this article is cited
When a correction for this article is posted
Click here to choose from all of JBC's e-mail alerts
This article cites 0 references, 0 of which can be accessed free at
http://www.jbc.org/content/265/2/1199.full.html#ref-list-1

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

H Rubin, Z M Wang, E B Nickbarg, S

McLarney, N Naidoo, O L Schoenberger, J L
Johnson and B S Cooperman

THE JOURNAL OP BIOLOGICAL. CHEMISTRY

0 1990 by The American Society for Biochemistry

Cloning, Expression,
Recombinant
Native

Vol. 265, No. 2, Issue of January 15, pp. 1199-1207,199O

Printed in U.S. A.

and Molecular Biology, Inc.

Purification,
and Variant

and Biological
Activity
of
Human arl-Antichymotrypsins*
(Received for publication,

Harvey
Oeyvind

RubinS,
Zhi mei Wang&
Elliott
B. Nickbargg,
Sean McLarneyS,
Nirinjini
L. Schoenbergerg,
Jeffrey
L. JohnsonQll,
and Barry
S. Cooperman
the $Department
of Medicine,
University
of Pennsylvania,
Philadelphia,
Pennsylvania 19104-6073

From
of Chemistry,

University

of Pennsylvania,

Philadelphia,

Pennsylvania

(4).

The precise biological role of ACT has not been determined.

Based on its rapid rate of association
with cathepsin G, it
may regulate the activity of this neutrophil
serine protease
* This work was supported by a grant (to H. R.) from H & Q Life
Sciences
(San Francisco,
CA). The costs of aublication
of this article
were defrayed
in part by the payment
of page charges.
This article
must therefore
be hereby marked
advertisement
in accordance
with
18 U.S.C. Section
1734 solely to indicate
this fact.
11Kodak Fellow.
The abbreviations
used are: ACT, antichymotrypsin;
PBS, phosphate-buffered
saline;
rACT,
recombinant
antichymotrypsin;
SDS,
sodium dodecyl
sulfate;
FPLC,
fast protein
liquid chromatography.

and

the SDepartment

19104-6323

(1). However, other targets are also possible. Chymotrypsinlike enzymes and their inhibitors
have been identified
in a
wide variety of normal and abnormal
biological
processes
including modulation
of cellular functions (5-9), DNA binding
(lo), inhibition
of certain parasite functions
(11-15) and
processing of vasoconstrictor
proteins (16). ACT appears to
be a component of the amyloid deposit in Alzheimers plaques
(17) and is present in various carcinomas (18,19) and in some
tissues of the reproductive
system (20, 21).
ACT forms SDS-stable complexes with its target enzymes
(22,23), which is a general property of serpin/serine
protease
interactions.
Little of a detailed nature is known about the
nature of these complexes. Although
high resolution
crystal
structures have been determined
for a form of the related
serpin, human cYl-protease inhibitor,
in which the Pl-Pl
peptide bond has been hydrolyzed
(24), as well as for complexes of serine proteases and some smaller peptide inhibitors
(25-29), no direct structural
studies of ACT alone or as a
complex with a serine protease have been reported.
In this paper we report the cloning, expression and mutagenesis of the human cul-antichymotrypsin
gene, and the
purification
and characterization
of both the recombinant
protein and of two variants of the recombinant
protein produced by mutation at its Pl site.
MATERIALS

cul-Antichymotrypsin
(ACT) is a serine protease inhibitor
(serpin) (1). In its native, circulating
form, it is a glycoprotein
of between 55,000 and 66,000 daltons, with the variation
attributed
to microheterogeneity
in glycosylation
(2). It is
synthesized
predominantly
in the liver and has also been
reported in mast cells, sinus histiocytes, endothelial
cells, and
in cells of the histio/monocytic
line (3). In response to inflammatory stimuli, plasma levels of ACT increase more than 4fold within several hours (3). A familial form of ACT deficiency has been described in which heterozygotes
have 50%
of normal circulating
levels (4). No homozygote
has been
reported, and such a genotype may be incompatible
with life

NaidooQ,

AND

Isopropyl-@-thiogalactopyranoside,

testinal alkaline phosphatase, T4 DNA

METHODS

EcoRI,

PstI,

HindIII,

calf

in-

polymerase, mung bean nu-

clease, Klenow
fragment,
pUC19,
and pKK233
were obtained
from
Promega
(Madison,
WI). Diaminobenzidine,
DNA-cellulose,
bovine
pancreatic
chymotrypsin
and trypsin,
and all chromophoric
protease
substrates
were obtained
from Sigma. Human
thrombin
was from
Sigma or Behring
Diagnostics.
Porcine
pancreatic
elastase
was from
Behring
Diagnostics.
Human
neutrophil
elastase was from EPC (Pacific, MO). DH5,
JM101,
and JM105
cells were obtained
from the
Cell Center
of the University
of Pennsylvania.
pINomp/Ncol/b,
a
secretion vector that allows fusion of a cloned protein to the omp
leader peptide,
was obtained
from Professor
John Collins
and Dr.
Gerhard
Gross
(Gesellschaft
fiir
Biotechnologische
Forschung,
Braunschweig,
Federal
Republic
of Germany).
pKC30
and Escherichiu coli N4830-1
were from Pharmacia
LKB
Biotechnology
Inc.
The pAR 3039 vector
originates
from Studier
and Moffat
(30) as

described.
Human
serum ACT was prepared
using a procedure
based on the
work of Tsuda et al. (31). Briefly,
this method
affords
pure ACT in
three steps, batchwise elution from DNA-cellulose,
G-150 chromatography, and NaCl gradient
elution from DNA-cellulose.
A full description of this procedure
will appear elsewhere.3
Plasmid
constructions
were carried
out following
Maniatis
et al.

(32).
The
numbering
of the reactive
site sequence
follows
Schechter
and Berger
(57).
3 L. Kilpatrick,
J. L. Johnson,
T. F. Clifford,
B. S. Cooperman,
S.
D. Douglas,
and H. Rubin, manuscript
in preparation.

1199

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

Human
al-antichymotrypsin
has been cloned,
sequenced
and expressed
in Escherichia coli and recombinant
protein
as well as point-specific
mutants
have
been purified
and characterized.
The corrected
genededuced amino acid sequence
has 45% overall
identity
with al-protease
inhibitor,
which
is higher
than the
42% previously
reported
(Chandra,
T., Stackhouse,
R.,
Kidd, V. J., Robson,
J. H., and Woo, S. L. C. (1983)
Biochemistry
22, 5055-5060).
Recombinant
antichymotrypsin
(rACT)
is similar
to natural
antichymotrypsin with respect
to the specificity
of its interactions
with proteases.
Its second-order
rate constant
for association
with bovine
chymotrypsin
is 6-8 X 10 M-
s-l, which
is identical
to that of the serum-derived
inhibitor.
Site-specific
mutagenesis
has been used to
produce
two variants
of rACT in which the Pl position
has been changed
from leucine
to either
methionine
(L358M-rACT)
or arginine
(L358R-rACT).
L358MrACT
has a specificity
of inhibitory
activity
toward
serine proteases
closely similar
to that of native rACT.
By contrast,
the specificity
of L358R-rACT
is quite
different
from that of native
rACT,
most notably
in
efficiently
inhibiting
trypsin
and human
thrombin
while
showing
a decreased
ability
to inhibit
chymotrypsin.

June 22,1989)

1200

Recombinant

Identification

and Sequencing

Native and Variant Human

of the Gene for Human

cul-Antichymotrypsins

ACT

A human liver cDNA

library
in the phage expression
vector Xgtll
(generously
provided
by Mitchell
Weiss, Department
of Human
Genetics,
University
of Pennsylvania)
was screened
according
to the
method
of Young
and Davis
(33) with polyclonal
antisera
raised
against Cl esterase inhibitor
(DAKO,
Santa Barbara,
CA), a related
human
serine protease
inhibitor.
Positives
were picked,
rescreened,
and plaque-purified.
DNA sequencing
was performed
with the chain
termination
method (34) using oligonucleotide
primers
obtained
from
the Nucleic
Acid Synthesis
Center of the Wistar
Institute
(Philadelphia, PA).
Expression

full length
Ea....

gene

EcoRr

EmRI
T

AAI-X,

CTC. TGC, GAG....

BamHl

Klenow

Mung Bean Nuclease

TC. TGC. GAG....

CAP

Systems
1 ligation

Site-directed

Mutagenesis

Site-directed
mutagenesis
was carried
out using the Bio-Rad
Ml3
in vitro mutagenesis
kit and the synthetic
DNA
primers
(5-CTAATGCAGACATGAGGGTGATT-3
for L358M
and 5-TGCAGAACGGAGGGT-3
for L358R).
The altered genes were excised from
double-stranded
Ml3 with EcoRI
and inserted
into pKK233
as described for the wild-type
construction,
yielding
recombinants
denoted
pKKACT-M
and pKKACT-R
for the methionine
and arginine
mutants, respectively.
Both mutations
were confirmed
by DNA sequencfull length gene

22

EWRI

ECORI

pKK 233-2

. . .CC ATG GCT GCA GCC AAG CTT

1yifi;K
I

CAP
CC ATG GCT GCA GCC AAG CT

T-

AA l-K

FIG. 1. Scheme
in pKKACT.

for

the

construction

CAP

Xbal

Xbal

Xbe I and EmRV

i
1

EmRV
Klenow
EcoRv

t ligation

mid

FIG. 2. Scheme
in PACTS.

for

the

construction

ing. The corresponding

proteins
L358R-rACT-1,
respectively.
Small

Scale

Growth

of the

were

denoted

Conditions

expression

L358M-rACT-1

plas-

and

and Extraction

Fresh overnight
cultures
of JM105
transformed
with pKKACT,
pKKACT-M,
or pKKACT-R
were diluted
to 1.5% in LB broth
containing
ampicillin
(sodium
salt, 0.1 mg/ml)
and grown
to an
ODW.,
of 0.3, induced
with 1.25 mM isopropyl-@thiogalactopyranoside and grown for an additional
5 h. The cells were pelleted
and
then disrupted
in a French
press. The ACT proteins
purified
from
these transformed
cells are denoted
rACT-1,
L358M-rACT-1,
and
L358R-rACT-1,
respectively.
Fresh
overnight
cultures
of N4830-1
were transformed
with
pACT2,
grown overnight
at 30 C, diluted to 1% in LB broth containing ampicillin
(sodium
salt, 0.1 mg/ml)
and grown to an ODsw.,
of
0.2, then shifted
to 42 C and grown
for an additional
5 h at 42 C.
The cells were pelleted
and disrupted
as above. The ACT protein
purified
from this transformed
cell is denoted
rACT-2.
Fresh overnight
cultures
of DH5 cells transformed
with pKTACT
were diluted to 1.5% in LB broth containing
ampicillin
(sodium
salt,
0.1 mg/ml),
grown
for 7 h, and harvested
by centrifugation.
The
washed
cell pellet was suspended
in 20% sucrose,
50 mM Tris, pH
7.5, 10 mM EDTA,
shaken
for 7 min at room temperature,
and
centrifuged
at 13,000 x g for 10 min. The pellet was rapidly
resuspended in double-distilled
water, and frozen,
thawed,
and sonicated
three times. The resulting
mixture
was centrifuged
at 13,000 X g for
10 min, and the supernatant
was saved.

CTC TGC CAC CCT AAC . . .

Western
f.kNlQ Bean Nucleaee
I
C CTC TGC CAC CCT AAC
Leu Cys His Pro Asn
I

ATG GCT GCA GCC AA0 CTC CTC TGC CAC CCT AAC.. .
Met Ala Ala Ala Lys Leu Leu Cye His Pro Asn...

mid

Hpa I

of the

expression

plas-

Blots

Commercial
antisera
(DAKO)
were absorbed
prior to use according
to the following
method.
An overnight
culture
of JMlOl
(200 ml) was
pelleted
for 5 min at 4 C, rinsed
with phosphate-buffered
saline,
resuspended
in 6 ml of cold phosphate-buffered
saline and then frozen
and thawed
three times in Dry Ice-ethanol.
The resulting
mixture
was sonicated
six times for 30 s each and pelleted in a microcentrifuge
at room temperature
for 5 min. The supernatant
was then diluted to
2% in phosphate-buffered
saline. 1.5 ml of 2% supernatant
was added
to a piece of nitrocellulose
paper cut in 2.5 x 2.5-cm squares.
This
mixture
was shaken
at room temperature
for 1 h. The nitrocellulose
was then rinsed twice in phosphate-buffered
saline. The Escherichia
coli extract-saturated
paper
was added to 20 ml of rabbit
antiantichymotrypsin
serum,
diluted
1:600 in Blotto
(2% dry milk in

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

The EcoRI insert from the recombinant

Xgtll was subcloned
into
pUC19.
pKK233
was digested
with HindIII,
the overhanging
ends
were filled in by treatment
with Klenow
fragment
in the presence
of
the four deoxynucleotide
triphosphates,
and dephosphorylated
with
alkaline
phosphatase.
The EcoRI-EcoRI
fragment
containing
the
entire ACT-coding
sequence
was removed
from the pUC19
vector,
isolated
by agarose gel electrophoresis,
treated
with mung bean nuclease to generate
a blunt end fragment
in the correct
reading frame,
and ligated to the modified
pKK233
vector
described
above. The
recombinant
was denoted pKKACT
and yielded recombinant
protein
denoted rACT-1.
The construction
is described
in Fig. 1.
A heat-inducible,
high expression
vector denoted pT7PL
was prepared by first placing the Shine-Dalgarno
sequence
and start codon
from the T7 vector pAR3039
upstream
to the coding sequence
of the
antichymotrypsin
gene and then placing
the gene with these heterologous regulatory
sequences
under the control
of the Pn promoter
in
pKC30.
The recombinant
was denoted
pACT2
and yielded recombinant protein
denoted
rACT-2.
The construction
is described
in Fig.
2.
pKT280
(Clontech)
was digested
with PstI, phenol/chloroformextracted
and ethanol-precipitated.
The 3 overhanging
ends were
removed
with T4 DNA polymerase
using 2 units of enzyme/pg
DNA
in the presence
of 3.3 mM dNTPs.
The vector was then treated
with
calf alkaline
phosphatase.
The EcoRI-EcoRI
fragment
containing
the
entire antichymotrypsin-coding
sequence
was isolated by agarose gel
electrophoresis
and the 5 overhanging
ends filled in with Klenow
fragment
as described
above.
The resulting
DNA
was blunt-end
ligated to the vector prepared
as described,
yielding
the recombinant
labeled pKTACT.

Recombinant

Native and Variant Human

CTG TCT CTG GGG GCC CAT AAT ACC ACC CTG ACA GAG ATT
Lsu Ser Leu Gly Ala His Asn Thr Thr Leu Thr Glu Ile

J c
CTC AAA GGC C;,
Leu Lys Gly ~eu

Chandra

CTG TCT
Leo Ser

CTG GGG GCC CAT AAT ACC ACC CTG ACA GAG ATT
Leu Gly Ala His Asn Thr Thr Leu Thr Glu Ile

CTC AAG GCC TCG AGT

Leu Lys Ala Ser Ser

al PI

CTC TCC CTG GGG ACC AAG GCT GAC ACT CAC GAT GAA ATC
Leu Ser Leu Gly Thr Lys Ala Asp Thr His Asp Glu Ile

CTG GAG GGC CTG AAT

Leu Glu Gly Leu Asn

This
work

Phe

Chandm
PI

al

w
E. coli pKKACT(2-45-32)
Crude lysate
Fast Q
DNA-cellulose
E. coli pT7PL
Crude lysate
Fast Q
DNA-cellulose

0.47d
0.41
0.20

NW
132
107

TCA CCT CAC GGA GAC TTA

Ser Pro His Gly Asp Leu

CTG AGG CAG AAA TTC ACT CAG AGC TTC

Leu Arg Gln Lys Phe Thr Gln Ser Phe

CAG CAC CTC

Gln His Leu

TTC AAC CTC ACG GAG ATT

Phe Asn Leu Thr Glu Ile

CCG GAG GCT CAG ATC CAT GAA GGC TTC CAG GAA CTC
Pro Glu Ala Gln Ile His Glu Gly Phe Gin Glu Leu

Activity

Thr

Chandra

CGC GCA CCC TCA ATC AGT

Arg Ala Pro Ser Ila
Ser

al PI

CTC
CGT
Leu Arg
two E. coli

Total b Yield
protein
%
mg
300
33
0.30

2500
880
122

100
86
42

100
81

in Crude

Ser

Glu

Ala

Glu

Ile

His

TCC
Ser

ACC CTC AAC CAG CCA

Thr Leu Asn Gln Pro

Purification
factor

7.9
430

5.8

1.0% Triton,
0.05 M Tris, 10.0 mM EDTA)
and swirled
for 1 h at
room temperature.
Proteins
were transferred
to nitrocellulose
paper
following
the procedure
of Tobin
et al. (35). The resulting
Western
blots were stained using the ABC Vectastain
kit (Vector
Laboratories)
and the color was developed
with diaminobenzidine.
and Antitrypsin

Glu

ACC CTC AAT CAG TCC

Thr Leu Asn Gln Ser

'The amount
ofantichymotrypsin
was determinedbytitration
of
bovine chymotrypsin
as described
in the methods
section.
*The amountoftotalprotein
wasdeterminedusingthemethodof
Bradford
(43).
From
3.1 g of cell paste (wet weight).
The
amount
of antichymotrypsin
in the crude lysate was estimated using Mono
Q chromatography
followed
by titration
as described under Materials
and Methods.
e From 19.1 g of cell paste (wet weight).
Not determined.

ACT

Thr

Lysates

ACT
activity
could not be directly
measured
in crude
bacterial
lysates because of a large background
inhibitory
activity
in the lysate
itself. The background
activity
was separated
from the antichymotrypsin
by anion-exchange
chromatography
using a Mono Q HR5/5
anion-exchange
FPLC
column
(Pharmacia
LKB Biotechnology
Inc.)
fitted into a LKB 2150 pump, 2152 gradient
controller,
and Waters
440 UV absorbance
detector
with an extended
wavelength
module.
Chromatography
was typically
conducted
on the extract
from 200 mg
of cells. The separation
involved
an isocratic
wash (5 min) with 50
mM Tris-Cl
buffer,
pH 7.5, containing
50 mM KCl, followed
by a
linear gradient
of KC1 (50-350
mM in 30 min) at a flow rate of 1.0
ml/min.
Protein
absorbance
was monitored
at both 214 and 280 nm.
Fractions
(1.0 ml) were collected
and assayed for ACT or antitrypsin
activity,
measured
as the inhibition
of the chymotrypsin-catalyzed
hydrolysis
of substrate
N-succinyl-A-A-P-F-p-nitroanilide
(36) or of
trypsin-catalyzed
hydrolysis
of substrate
N-Bz-P-F-R-p-nitroanilide.

A typical
chymotrypsin
assay contained
(in 1.0 ml) 100 mM Tris-Cl
buffer,
pH 8.3, 0.005%
(v/v)
Triton
X-100,
bovine
pancreatic
chymotrypsin
(18 pmol)
and column
eluate (0.005-0.5
ml). The assay
mixture
was preincubated
at room temperature
for 5 min, substrate
(0.01 ml of a 10 mM solution
in 90% dimethyl
sulfoxide)
was added,
and remaining
chymotrypsin
activity
was determined
by the rate of
A typical
change
in Ano ,,,, caused by the release of p-nitroanilide.
trypsin
assay contained
(in 1.0 ml) 100 mM Tris-Cl
buffer,
pH 8.3,
0.005%
(v/v)
Triton
X-100,
bovine
trypsin
(8.6 pmol) and sample
(0.005-0.5
ml). The assay mixture
was preincubated
at room temperature for 10 min, substrate
(0.02 ml of a 15 mM solution
in 90%
dimethyl
sulfoxide)
was added, and remaining
trypsin
activity
was
determined
as above. Measurements
of optical
absorbance
were conducted
at 25 C using a Hewlett-Packard
8452A spectrophotometer
fitted with a temperature
controlled
sample compartment.
The amount
of active rACT-1,
rACT-2,
or L358M-rACT-1
present
was determined
by titration
of a solution
of chymotrypsin
of known
concentration
and activity
with varying
amounts
of partially
purified
rACT
fractions.
The amount
of active
chymotrypsin
present
after
incubation
with the inhibitor-containing
solutions
was then determined using the chymotrypsin
activity
assay. The amount
of active
L358R-rACT-1
present
was determined
in a similar
manner
by titration of a solution
of trypsin
of known concentration,
using the trypsin
activity
assay.
Concentrations
of chymotrypsin
and trypsin
were
determined
using the active-site
titration
method
of Ardelt
and Laskowski
(37).
Purification

and Characterization

of Recombinant

Antichymotrypsins

Large-scale
Growth
of E. coli-E.
coli JM105
strains were grown to
a density
of 4-5 ODhho.,
in LB medium containing
ampicillin
(sodium
salt, 0.1 mg/ml)
and glucose (0.1% w/v) at 37 C in a 15-liter
carboy
fitted
with an oxygen
bubbler.
Cells were harvested
by passage
through
a Sharpless
continuous-flow
centrifuge.
3.5-5 g of cell paste
(wet weight)
were obtained/liter
of culture.
E. coli N4830-1
transformed
with pACT2
was grown in LB medium
containing
ampicillin
(sodium
salt, 0.1 mg/ml)
in a 15-liter
carboy
fitted with a heating
coil, sampling
tube, temperature
probe,
and an air bubbler.
The
medium
was inoculated
with an overnight
culture
(200 ml) that had
beengrownat
aconstanttemperatureof30
"C.Growthwascontinued
at 30 C for 1 h until the culture
had reached
an ODsoo.,,, of 0.17. The
temperature
of the medium
was shifted
to 42 C by pumping
steam
through
the heating coil for a period of two min and then maintained
at that temperature
by a heating
circulation
bath for 6 h until the
culture
had reached
an OD,,,
of 0.9. The cells were harvested
with
a Sharpless
centrifuge
as described
above. 1.2 g of cell paste (wet
weight)
were obtained/liter
of culture.
Extraction
and Column Chromatographies-Purifications
of rACT1, rACT-2,
L358M-rACT-1,
and L-358R-rACT-1
were all carried
out
in an essentially
identical
manner,
with the exception
that in the
latter case an antitrypsin
rather
than an antichymotrypsin
assay was

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

Antichymotrypsin

LYS

CAG CAC CTC

Gln His Leu

Leu

CTG C:C
Leu Arg

from

AAG

CAG AGC TTC

Gln Ser Phe

Asn

This
work

TABLE I
of recombinant
antichymotrypsin
expression
systems
step

1201

This
work

FIG. 3. DNA
and amino
acid
sequences.
Comparison
of gene-deduced
cul-antichymotrypsin
(our
sequences
data and those of Ref. 39) and of the alprotease
inhibitor
sequence
(41). Positions of base pair deletions
and insertions are indicated
by arrows.
Residues
that are identical
between
our corrected
al-antichymotrypsin
sequence
and the
al-protease
inhibitor
sequence
are indicated by asterisks.

Purification

al -Antichymotrypsins

1202

Recombinant

Native and Variant Human

(~1 -Antichymotrypsins
RESULTS

Cloning

-69

::

-46

-30

ABCD

used to detect and quantitate

recombinant
inhibitor.
All purification
steps were carried
out at 4 C. In a typical
preparation
of rACT-1,
cell paste (3.1 g) was dispersed
in 10 mM potassium
phosphate
buffer,
pH 6.9 (25 ml) and lysed by three passes through
a French
press at
10,000 psi and 4 C. Cell debris was removed
by centrifugation
at
30,000 x g for 30 min at 4 C. The supernatant
(25 ml) was loaded
onto a column
(4.9 cm* x 37 cm) of Sepharose
Fast Q (Pharmacia
LKB Biotechnology
Inc.) that had been equilibrated
to 50 mM TrisCl, pH 7.5, containing
50 mM KCI. Protein
eluted with a linear
gradient
of KC1 in 50 mM Tris-Cl,
pH 7.5 (50-500
mM in 2 liters).
Fractions
(15 ml) were monitored
for protein
by AZBOnmand assayed
for antichymotrypsin
activity
as described
above. rACT-1
eluted at
approximately
200 mM KCl. Fractions
50-58,
containing
rACT-1,
were combined
and dialyzed
against
two volumes
(2.5 liters each) of
10 mM potassium
phosphate
buffer,
pH 6.9, over 48 h. The dialyzed
solution
was then applied to a DNA-cellulose
column
(1.7 cm2 X 20
cm) that had been pre-equilibrated
with 10 mM potassium
phosphate,
pH 6.9, containing
10 mM KCl. After loading,
the column
was first
washed with the same buffer
(20 ml). The column
was eluted with a
linear gradient
of KC1 (lo-400
mM, 300 ml) in the same buffer.
Fractions
(8 ml) were assayed
for protein
and antichymotrypsin
activity
as above.
rACT-1
eluted
between
350 and 400 mM KC1
(fractions
49-53).
Fractions
containing
antichymotrypsin
activity
were analyzed
for purity
by SDS-PAGE,
performed
according
to
Laemmli
(38). Fractions
in the early portion
of the antichymotrypsin
peak showed a higher level of purity
than those in the later portion.
Each portion
was concentrated
by ultrafiltration
using Amicon
YM10 membranes
and dialyzed
overnight
against 50 mM Tris-Cl,
pH 7.5
(500 ml). In some cases recombinant
proteins
were further
purified
on a FPLC
MonoQ
anion-exchange
column,
using the conditions
described
above.
Automated
Edman
sequence
analysis
on rACT-2,
using the Milligen/Biosearch
6600 Series Prosequencer
of the Protein
Facility
of
the Dental
School
of the University
of Pennsylvania,
yielded
a
sequence
in full accord with that predicted
in Fig. 2, beginning
with
the tripeptide
Ala-Ser-Met.

Kinetics of Complex Formation

The rates of inhibition
by human serum ACT, rACT-1,
rACT-2,
L358M-rACT-1,
and L358R-rACT-1
of bovine chymotrypsin,
bovine
trypsin,
human
thrombin,
porcine
pancreatic
elastase,
and human
neutrophil
elastase
were investigated
at 25 C under second-order
conditions
in reaction
mixtures
containing
equimolar
concentrations
of enzyme and inhibitor,
or under pseudo-first
order conditions
with
an excess of inhibitor.
As described
above, enzyme
and inhibitor
were
incubated
in 100 mM Tris-Cl
buffer, pH 8.3, containing
0.005% (v/v)
Triton
X-100 for varying
periods
of time, substrate
was then added,
and the amount
of remaining
active enzyme
was determined.
Alternatively,
at timed intervals
aliquots
of the inhibitor
plus enzyme
solution
were diluted into an assay solution
containing
the appropriate substrate
and protease
activity
was determined.

of the Gene for Antichymotrypsin-

The DNA sequence and the derived amino acid sequence of

the insert from one of the positive Xgtll cDNA clones contained the entire coding region of mature human cul-antichymotrypsin. The insert also included an extension on the 5end encoding additional amino acids that appear in the precursor of the mature protein. The mature, serum-derived
protein has been reported to contain 398 amino acids (MI
45,031), starting from the tripeptide Asn-Ser-Pro (Fig. 2) at
the amino terminus (39). More recently, we3 have demonstrated the presence of a second form of mature protein that
includes two additional amino acids, His-Pro, at the NH,
terminus (Fig. 2). The reactive center, Pl-Pl, Leu-Ser, is
found at positions 358-359. The COOH-terminal sequence is
in agreement with Hill et al. (40) and the remainder of the
sequence is in agreement with Chandra et al. (39) except for
the 15 amino acids from position 77 to 91 and the six amino
acids from 98 to 103. These differences can be explained by
three insertions and three deletions of single bases within the
Chandra sequence. The sequence reported here displays a
high degree of similarity with al-protease inhibitor in this
region (Fig. 3) and raises the overall identity with al-protease
inhibitor by 17 residues, to 44.5%. The intron/exon structure
of these proteins has recently been reported (42) to be identical, with five exons separated by four introns. Comparison
of the antichymotrypsin and al-protease inhibitor sequences
exon by exon shows an interesting pattern. Exons II, IV, and
V have large percentages of identical amino acid residues: 51,
44, and 46%, respectively. However, exon III has only 33%
identity. The reactive center of the inhibitor is found in exon
IV. Finally, we note that, in comparison with the earlier work
of Chandra et al. (39), our sequence shows a proline at position
44 rather than a leucine, a leucine at 174 rather than a proline,
an alanine for valine at 336, and a leucine for serine at 338.
The latter two amino acid substitutions are also reported by
Hill et al. (40).
Expression of the ACT Gene-Four different E. coli vectors,
two secretion systems (pINomp/iVcoI/b and pKT280) and
two non-secretion systems (pKK233 and pT7PL), were evaluated for expression of recombinant human antichymotrypsin. More than 100 recombinants in a modified pINomp vector
were screened and yielded plasmids with the insert in the
wrong orientation in every instance (results not shown). We
presume that the correct orientation constructs lead to high
level production of a toxic gene product. A second secretion
plasmid, pKT280, in which expression is driven from the /3lactamase promoter and which encodes part of the signal
sequence of p-lactamase, was also examined. The construct
yielded full length ACT in low yield with an amino-terminal
extension coding for an additional eight amino acids, HisPro-Gln-Phe-Leu-Cys-His-Pro, the first four originating from
the vector and the following four from the precursor of antichymotrypsin.
The expression plasmid pKK233, utilizes the strong trp-lac
promotor and yielded a recombinant protein, migrating with
an apparent M, of 45,000, that could be detected on Western
blots of crude extracts, using affinity purified antibody raised
against al-antichymotrypsin.
The construction of the expressed protein contains an amino-terminal extension of 10
residues (Fig. 1).
The highest level of expression was obtained with pACT2,
which utilizes the PL system (Fig. 2).
Purification
of Recombinant
Antichymotrypsins-The
purifications of rACT-1 and rACT-2 are summarized in Table
I, from which it is clear that much higher levels of expression

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

FIG. 4. SDS-stacking
gel electrophoresis
of purified
recombinant
antichymotrypsin.
Lane A, 5 pg of purified
antichymotrypsin from fractions
49-50 of the DNA-cellulose
column;
lane B, 25 Kg
of combined
fractions
50-58 from the Fast Q column;
lane C, 50 Kg
of crude lysate; lane D, molecular
weight markers.
Protein
was visualized with a Coomassie
Blue stain.

and Sequencing

Recombinant

Native and Variant Human cul-Antichymotrypsins

r ACTI

L358R-r

ACT

L358R-rACT

I
-200

-El\

----

i % *

--

tryps1n

thrombln

lhrombin

1;;

-46

I;;y(
-69
-

ibIg/

-46K

!
-30
ABCDEF

-30

%IL

I
trypsin

-_

porcme

E F

rACT- I
poncreotlc

ABCDE

ye
&u-

-46K

K
K

-45

-30

elostose
I gKK
-69K

-92
-69

L3!38M-rACT
porcine poncreotlc

-200
-

-92
-69

-k.raDD-45

K
K
K
K

-30

--30K
ABCDEF

I
elostose

ABCDEFG

FIG. 5. Western
blots of crude
recombinant
antichymotrypsins
in the presence
or absence
of various
serine
proteases.
Upper panel, rACT-1
interaction
with thrombin:
lane A, rACT-1
alone; lanes B-F, rACT-1
incubated
with 0.1, 0.5, 1, 5, and 10 ~1 of thrombin,
respectively;
L358R-rACT-1
interaction
with thrombin:
lane
A, L358R-rACT-1
alone; lanes B-F, L358R-rACT-1
incubated
with 0.1,0.5,1,5,
and 10 ~1 of thrombin,
respectively;
L358R-rACT-1
interaction
with trypsin:
lane A, L358R-rACT-1
alone; lanes B-G, L358R-rACT-1
incubated
with
0.1, 0.5, 1, 2, 5, and 10 ,ul of trypsin,
respectively.
Center panel, L358M-rACT-1
interaction
with chymotrypsin:
lane A, L358M-rACT-1
alone; lanes B-F, L358M-rACT-1
incubated
with 0.1, 0.5, 1, 5, and 10 ~1 of chymotrypsin,
respectively;
L358M-rACT-1
interaction
with trypsin:
lane A, L358M-rACT-1
alone; lanes B-F, L358M-rACT-1
incubated
with 0.1, 0.5, 1, 5, and 10 ~1 of trypsin,
respectively.
Lower panel, rACT-1
interaction
with porcine
pancreatic
elastase:
lane A, rACT-1
alone; lanes B-F, rACT-1
incubated
with 0.1, 0.5, 1, 5, and 10 ~1 of porcine
pancreatic
elastase,
respectively.
L358M-rACT-1
interaction
with porcine
pancreatic
elastase:
lane A, L358MrACT-1
alone; lanes B-G, L358M-rACT-1
incubated
with 0.1, 0.5, 1, 2, 5, and 10 ~1 of porcine
pancreatic
elastase,
respectively.
The following
stock protease
solutions
were used in these experiments:
thrombin,
0.12 mg/ml; trypsin,
1 mg/ml;
chymotrypsin,
1 mg/ml;
porcine
pancreatic
elastase,
1 mg/ml.
Crude antichymotrypsins
were prepared
from extracts
of 0.5-1.0 ml of bacterial
culture.

are achieved with the second expression system. In both cases,

SDS-PAGE
analysis of protein
gradient-eluted
from the
DNA-cellulose
column (see Materials
and Methods)
showed
a single band at approximately
45,000 daltons (Fig. 4). Purification results similar to those for rACT-1 were obtained for
L358M-rACT-1
and L358R-rACT-1.

Formation of SDS-Stable Complexes between Recombinant

Antichymotrypsins and Serine Proteases-The
complex that
human serum ACT forms with chymotrypsin
is stable in SDSpolyacrylamide
gels and migrates at a higher molecular weight
than ACT itself (23). We used Western blots of Laemmli gels
(Fig. 5) to determine whether or not SDS-stable
complexes
are formed between rACT (and rACT variants) and proteases,
giving the results summarized in Table II4 As may be seen,
the complexes themselves appear to be substrates for the
In addition
complexes
were
G and between
L358M-rACT-1
N. Schechter,

to the results
presented
in Table
II, SDS-stable
shown to be formed
between
rACT-1
and cathepsin
L358R-rACT-1
and kallikrein,
but not between
and Clr.
personal
communication.

corresponding
uncomplexed
proteases, since they are readily
degraded as the [protease]/[inhibitor]
ratio is increased.
The formation
of serine protease-rACT
complexes was
examined using either crude extracts containing
rACTs or
highly purified rACTs. In the former case, such experiments
provided a useful screen for the expression of active protein.
As the ability of a rACT mutant to form an SDS-stable
complex with a protease correlates very well with its affording
a measurable rate of protease inhibition
(Table II), we conclude that the mutant rACTs inhibit proteases by the same
mechanism
as does human serum ACT. The one apparent
exception to the above noted correlation
is the interaction
between L358M-rACT-1
and porcine pancreatic
elastase.
Here we see evidence for complex formation (Fig. 5) but were
unable to measure a rate of protease inhibition.
However, as
we would not detect inhibition
from a rate constant that was
more than lo-fold less than that for rACT-1, this exception
may be more apparent than real.
It is interesting
to note that when identical Western blots
were reacted with the polyclonal
serum raised against Cl

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

ABCD

ABCDEFG

!,-200

ABCDEF

L358M-rACT
chymoirypsin

Recombinant

1204

Native and Variant Human al -Antichymotrypsins

60 .

Time (min)
FIG. 6. Kinetics of inhibition
of chymotrypsin
by rACT (0)
and by human serum al-antichymotrypsin
(0). Reaction conditions are as described in Table II, using enzyme and inhibitor
concentrations of 9 nM.
inhibitor
that was used to screen the original Xgtll library,
no bands were detected, although serum-derived
Cl inhibitor
could be easily detected (gels not shown). This suggests that
in the Xgtll expression system some fraction of recombinant
proteins are folded in a fashion that exposes regions detectable
by antisera raised against related proteins.
Inhibition
of Serine Protease Activity by Recombinant Antichymotrypsins-Both
human serum and recombinant
ACTS
were tested for their inhibitory
activities toward several serine
proteases, giving the results summarized
in Table II. The
various antichymotrypsin:serine
protease pairs fall into three
groups. The first group comprises those pairs in which full
(>90%) inhibition
was observed at a 1:l ratio of inhibitor
to
protease. Examples of such pairs are bovine chymotrypsin
with either human serum ACT, rACT-1, or rACT-2. Secondorder plots of rate data for inhibition
with human serum ACT
and rACT-1 are shown in Fig. 6, demonstrating
that both
second order rate constants are identical
(8 x lo5 M- s-ithe corresponding
value determined with rACT-2 was 6 x lo5
M-i s-l). We measured virtually
the same rate constant for

human serum ACT using the assay conditions of Beatty et al.

(44). Independent
measurements
using a third set of assay
conditions
(phosphate-buffered
saline) also yielded an association rate constant on the order of 4 x lo5 M-i s-i at 25 C.
These values are an order of magnitude higher than the value
of 6 X lo4 M-' s- previously determined
by Beatty et al. (44)
for reaction of human serum ACT with bovine chymotrypsin
at 25 C. The reasons for this apparent
disagreement
are
unclear.
In the second group of pairs, full inhibition
required a
stoichiometric
excess of inhibitor
over protease. Chymotrypsin and L358R-rACT-1
and pancreatic porcine elastase and
either human serum ACT, rACT-1, or rACT-2 fall into this
group, as demonstrated
by the results displayed
in Fig. 7.
Plots of the degree of final inactivation
as a function of the
ratio of L358R-rACT-1
to chymotrypsin
and of rACT-2 to
porcine pancreatic
elastase show that full inactivation
is
achieved at extrapolated
ratios of 2.2 and 2.5, respectively.
Very similar results were obtained when human serum ACT
was used in place of rACT-2. This second group of pairs is
considered further below.
The third group of antichymotrypsin:protease
pairs showed
no detectable inhibition
(<3%), allowing estimation
of only
upper limits of second order rate constants. The estimates
presented in Table II are based upon the simple second order
kinetic scheme described by Beatty et al. (44), the concentration of enzyme and inhibitor
used, the total length of time
allowed for the reaction (usually 45-240 min), and assuming
a threshold level for detection of inhibition
of a 10% loss of
protease activity.
Inhibition
of Protease Activity Requiring a Stoichiometric
Excess of Inhibitor-The
requirement
of an inhibitor
to enzyme ratio of >l for full inactivation
has been observed for
reactions of other serpin:protease
pairs (45,46) and has been
shown to be due to concurrent formation of both an essentially
irreversibly
formed inactive complex and an inactivated
inhibitor.
In the case of L358R-rACT-1
and chymotrypsin,
support for such a model is provided by the independence
of
the extent of final inhibition
on the initial inhibitor
concentration, at a constant ratio of inhibitor
to protease (Fig. 8).

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

TABLE II
Interaction of recombinant antichymotrypsins and serine proteases
Formation of SDS-stable complexes and inhibition rate constants (measured at 25 C, uH
. 8.3). +. SDS-stable
complex formed, -, SDS-stable complex not formed, ND, not determined.
rACT-1 or rACT-2
L358M rACT
L358R rACT
Enzyme
Substrate (final molarity)
Molarity
Molarit
Molarity
k,
k
k,
IZM
nhf
IIM
Bovine chymotrypsin
Sue-A-A-P-F-p-nitroanilide
9
6-8 x lo5 (+)
18
3 X lo5 (+)
72
1 x 10 (+)
[2.2 x lO]C
(0.2 mM)
18
360
N-p-Tos-G-P-R-p-nitroanilide
<5 x lo* (-)
172
<lo* (-)
5.4 x lo6 (+)
Trypsin
172
8.6
(0.2 mM)
Human thrombin
N-p-Tos-G-P-R-p-nitroanilide
425
<3 x lo* (-)
213
<lo3 (ND)
4.3 x lo3 (+)
850
(0.2 mM)
Porcine pancreatic
Sue-A-A-A-p-nitroanilide
(1.0
480d
8.0 x lo3 (+)
193
<lo3 (+)
770
~10 (ND)
elastase
96V
[2.0 x lO]d
mM)
207
Human neutrophile
Sue-A-A-P-V-p-nitroanilide
207
<lo3 (ND)
<lo3 (+)
800
<lo2 (ND)
elastase
(1.0 mM)
a Equimolar concentrations of enzyme and inhibitor were reacted, unless noted otherwise. The concentration of
wild-type recombinant inhibitor and of the L358M mutant inhibitor were both determined by titration with bovine
chymotrypsin. The concentration of the L358R mutant inhibitor was determined by titration with bovine trypsin.
bM-l s-l
Corrected for partitioning of the @.I). complex. See text.
d The association rate constant of rACT with porcine pancreatic elastase was determined under pseudo-first
order conditions with an enzyme concentration of 24 nM. The rate constant for reaction with human serum ACT
was 6.9 X lo3 M- s-l.
eDetermined under pseudo-first order conditions with an enzyme concentration of 48 nM.

Recombinant

Native and Variant Human Lul-Antichymotrypsins

1205
0

(4

0
80

fa)

(W

300

50

100

(Molar

200

Time (min)

Time (min)

[l]/[E],

150

Ratio)

[l]/[E],

(Molar

Ratio)

FIG. 7. Stoichiometry
of inhibition
of chymotrypsin
activity
by L358R-ACT-1
and of porcine
pancreatic
elastase
activity
by rACT-2.
A: upper panel, chymotrypsin
inhibition
by L358R-ACT-l.
The
concentration
of chymotrypsin
was held constant
at 360 nM. [I]/[E]
ra t io values were 0 (a); 0.5 (b); 1.0 (c); 2.0
(d). Lower panel, plot of final activities
uersus [I]/[E]
ratio. B: upper panel, porcine
pancreatic
elastase
inhibition
by rACT-2.
The concentration
of elastase was held constant
at 39 nM. [I]/[,??] ratio values were 0 (a); 1.0 (b); 2.0
(c); 3.0 (d); 4.0 (e); 5.0 (f). Lower panel, plot of final activities
uersus [I]/[E]
ratio. Very similar
results
were
obtained
when human serum ACT replaced
rACT-2.

(W

E+I

SCHEME 1. Serpin
(EI)b is the irreversibly
inhibitor.

o-

150

200

time (min)
FIG. 8. Second
order
kinetics
of inhibition
of chymotrypsin
by L358R-ACT-l.
Reaction
mixtures
contained
equimolar
initial
concentrations
of enzyme
and inhibitor.
Curve a, [E] = [I] = 72 nM;
curue b, [E] = [I] = 360 nM. Curve c is a control
measuring
the
activity
of chymotrypsin
(360 nM) in the absence
of inhibitor.

More direct evidence for such a model in the case of porcine

pancreatic
elastase and human serum ACT comes from the
observation of Morii and Travis (23) that elastase cleaves the
inhibitor
between the P5 and P4 positions, thereby inactivating it.

k_

E.1

as a suicide
formed

k,

um

(E.I),/
\ -E+Ii
k,
inhibitor.
E is protease,
I is serpin,
inactive
complex.
It is the inactive

As has been pointed out by Fish and Bjork (47), when the
model applies the serpin may be formally
considered as a
suicide inhibitor.
An analytical
solution
for the classical
scheme of suicide inhibition
shown below (Scheme 1) has
been presented by Waley (48). Applying
this solution to the
results in Fig. 8 permits evaluation
of an apparent second
order rate constant for reaction of L358R-ACT-1
and chymotrypsin, equal to kJ~&/[(k-~
+ kz)(k3 + k4)], of 1 X lo4 M-
s-l. Since full inhibition
is obtained at a 1:l ratio of serpin to
protease when 12s << k4, the term for L358R-rACT-1
interaction with chymotrypsin
that is most comparable to the simple
second-order
rate constant for rACT-1 interaction
with chymotrypsin
is klkz/(kwl
+ kp), which is equal to 2.2 X lo4 Mm1
s -1.

Measured under pseudo-first

order conditions
(Table II),
the observed second-order
rate constant for the loss of enzy-

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

200

100

1206

Recombinant

Native and Variant Human

matic activity on incubation of porcine pancreatic elastase

with rACT-2 is 8.0 x lo3 M- s- which, corrected for (E.I),
partitioning by multiplying by the term (k3 + 124)/124
(Scheme
l), yields an apparent second-order rate constant for @.I),
formation of 2.0 X lo4 M-' s-i. We obtained similar results
when human serum ACT was used in place of rACT-2. Laine
et al. (49) have reported that an I:E ratio of 5.5 was required
for full, irreversible inhibition by human serum ACT of porcine pancreatic elastase. However, in this earlier work, performed under conditions similar to our own, only a 60-min
incubation was used before inhibition was measured. As seen
in Fig. 7B, such an incubation time would underestimate
inhibition obtained at lower concentration of inhibitor and
lead to an overestimate of the amount of inhibitor required
for full inhibition of elastase activity.
DISCUSSION

enzyme, cathepsin G, and an 8000-fold reduction in the rate

of inhibition of human neutrophil elastase (53).
By contrast, even though human al-protease inhibitor is a
potent elastase inhibitor, has a methionine in its Pl position
and, as we have demonstrated above, a higher (44.5%) overall
identity with rACT than antithrombin, the L358M mutation
of rACT-I fails to convert rACT into an effective inhibitor of
elastase. In related work, mutation of methionine 358 in
human cYl-protease inhibitor to a variety of other hydrophobic
residues, including valine, alanine, leucine, and isoleucine, has
only modest effects on the second order rate constant for
inhibition of human neutrophil elastase (50). Results obtained
using small synthetic oligopeptide substrates and inhibitors
(29) also suggest that the Sl site of human neutrophil elastase
can accommodate a variety of hydrophobic amino acid side
chains. Therefore, antielastase activity, unlike antithrombin
activity, has no comparable dependence on the presence of a
particular residue in the Pl position.
These latter results suggest that, for at least some serpinserine protease complexes, specificity-determining interactions take place with serpin residues other than that in the
Pl position. A major question is the extent to which such
residues fall within the reactive site loop of protease inhibitors
(from position P6 to position P3 (29)) or in other portions
of the protein. X-ray structure determination of serine protease complexes with smaller protease inhibitors such as
bovine pancreatic trypsin inhibitor (25), turkey ovomucoid
third domain inhibitor (28), eglin c (26), and secretory leukocyte protease inhibitor (27) have shown many contacts
between proteases and positions within the reactive site loop
of the inhibitor, and results with site-specific mutants (52,54,
55) and with small synthetic oligopeptide substrates and
inhibitors (29, 58, 59) provide evidence that such contacts
make important

contributions

to binding.

On the other hand,

some of the x-ray results (29) also show evidence for a second
contact domain on inhibitors well outside the reactive site
loop. Furthermore, Mierzwa and Chan (56) have presented
evidence based on chemical modification experiments that
elastase interacts with al-protease inhibitor at residues that
are far from the Pl site. We are presently undertaking a series
of experiments combining studies of the properties of pointspecific mutants with chemical modification and x-ray crystallographic analysis to closely define the structural details of
antichymotrypsin-chymotrypsin interaction. This work forms
part of our overall goals to investigate the interaction of
antichymotrypsin with its natural targets, study the regulation of the gene in response to mediators of inflammation,
and evaluate the biological role and potential therapeutic
applications of antichymotrypsin.
REFERENCES

1. Travis, J., and Salvesen,G. S. (1983)Annu.

Reo.

Biochem.

52,

655-709

2.

Tokes,Z. A., Gendler,S.J., and Dermer, G. B. (1981)

Strut.

Cell. Biochem.
3. Berninger, R. W. (1985)

17,69-77
J. Med.

16,

J. Supramol.

101-127

4. Erikson, S.,Lindmark, B., and Lilja, H. (1986)Acta

Med.

Scandia

220.447-453

5. King, G. H., Goralinik, C. H., Kleinhenz, P. J., Marino, J. A.,

Sedor, J. R.. and Mahmoud, A. A. (1987) J. Clin. Znuest. 79,
1091-iO98
6. Braun,
N. J., and Schnebli,
Seyler 368, 155-161

H.

P. (1987)

Biol.

Chem.

Hoppe-

7. Camussi,G., Tetta, C., Bussolino,F., and Baglioni, C. (1988) J.

l&p. Med. 168, 1293-1307

8. Yavelow,
J., Collins, M., Birk, Y., Troll, W., and Kennedy, A. R.
(1985)Proc. N&l. Acad. Sci. U. S. A. 82, 5395-5399
9. Yavelow, J., Caggana, M., and Beck, K. A. (1987) Cancer Res.
47, 1598-1601

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

A major focus of current studies of serpins and their target

proteases is the determination of the nature and extent of
contact regions between them, as a means of understanding
the specificities of such interactions. The results described in
this paper make two important contributions to such understanding. They show first, that glycosylation of the serum
protein (which accounts for approximately 30% of its total
molecular weight) is not essential for the interactions of ACT
with serine proteases and second, that the importance of the
amino acid residue in the Pl position of a rACT in determining its ability to inhibit serine proteases varies with the serine
protease under consideration.
The important structural difference between human serum
ACT and either rACT-1 or rACT-2 is that the serum protein
is glycosylated whereas the rACTs are not. Despite this difference, the two kinds of ACT show similar specificities
toward both formation of SDS-stable complexes with serine
proteins and inhibition of protease activity (Table II). In a
closely related study (50), recombinant cYl-proteinase inhibitor produced in E. coli was found to have an association rate
constant for neutrophil elastase (7.2 X lo6 M-l 6-l) only
slightly below that found for the human serum protein (1.1 X
lo7 M- s-l). It may therefore be general that the pattern and
extent of glycosylation is not a dominant factor in serpin
interactions with their target proteases.
Much of the literature on serpins has focused on the contribution of the Pl side chain to the rate of formation and
stability of serpin:serine protease complexes. The results obtained in the present work, as well as the related work of
others, provide evidence that this focus is more justified for
some serpin:serine protease interactions than for others.
Thus, the properties of L358R-rACT-1 support the conclusion
that an arginine in the Pl position is a major determinant of
serpin specificity in both a positive and negative sense. This
single amino acid change turns rACT into a serpin capable of
inhibiting thrombin and trypsin but having a considerably
reduced (30-fold) rate of reaction with chymotrypsin. In fact,
even though rACT and antithrombin III have only limited
(28%) overall identity (51), the specificity of L358R-rACT-1
resembles that of antithrombin III (an inhibitor having an
arginine in its Pl position), and the rate constants for inhibition of thrombin by these two inhibitors are virtually identical, 4.3 x lo3 M- s-l for L358R-rACT-1
in the present work
and 2.5 x lo3 M- s-l for antithrombin III measured by others
under similar conditions (52). Similarly, human cyl-protease
inhibitor was shown to be converted from an antielastase to
an antithrombin by the naturally occurring Pittsburgh
M358R mutation, a mutation that also resulted in a 25-fold
reduction in the rate of inhibition of the chymotrypsin-like

cul-Antichymotrypsins

Recombinant

Native and Variant Human al -Antichymotrypsins

5651-5656
23. Morii, M., and Travis,

J. (1983) J. Biol. Chem. 258,12749-12752

24. Lobermann,
H., Tokuoka,
R., Deisenhofer,
J., and Huber,
R.
(1984) J. Mol.. Biol. 177; 531-556
25. Marauart.
M.. Walter.
J.. Deisenhoffer.
J.. Bode. W.. and Huber.
R. (1983) Acta Cry&ilogr.
Sect. B. ktkct.
C&&logr.
Cryst:
Chem. 39,480-490
26. Bode. W.. Panamokos.
E., and Musil. D. (1987) Eur. J. Biochem.

166,673-692
27. Fujinaga,
M., Sielecki,
A. R., Read, R. J., Ardelt,
W., Laskowski,
M., Jr.. and James,
M. N. (1987) J. Mol. Biol. 195,397-418
28. Grute, M. G., Fendrich,
G., Huber, R., and Bode, W. (i988) Embo
J. 7,345-351
29. Bode, W., Meyer, E., Jr., and Powers,
J. C. (1989) Biochemistry
28,1951-1963
30. Studier,
F. W., and Moffatt,
B. A. (1986) J. Mol. Biol. 189, 113130
31. Tsuda, M., Ohkubo,
T., Kamiguchi,
H., Suzuki,
K., Nakasaki,
H.,
Mitomi,
T., and Katsunuma,
T. (1982) J. Exp. Clin. Med. 7,
201-211
32. Maniatis,
T., Fritsch,
E. F., and Sambrook,
J. (1982) Molecular
Cloning: A Laboratory
Manual,
Cold Spring Harbor
Laboratory
Press, Cold Spring Harbor,
NY
33. Young,
R. A., and Davis,
R. W. (1983) Proc. Natl. Acad. Sci. U.
S. A. 80, 1194-1198

34.

Sanger,
F., Nicklen,
S., and Coulson,
A. R. (1977) Proc. N&l.
Ad.
Sci. U. S. A. 74,5463-5467
35. Tobin,
H., Staehelin,
T., and Gordon.
J. (1979) Proc. Natl. Ad.
Sci. U. S. A. 76, 4350-4354
36. DelMar,
E. G., Largman,
C., Brodrick,
J. W., and Geokas,
M. C.
(1979) And. Biochem.
99, 316-320
37. Ardelt,
W., and Laskowski,
M., Jr. (1985) Biochemistry
24,5313-

5320
38. Laemmli.
I-J. K. (1970) Nature
227.680-685
39. Chandra,
T., Stackhouse,
R., Kidd,.V.
J., Robson,
K. J. H., and
Woo, S. L. C. (1983) Biochemistry
22,5055-5061
40. Hill, R. E., Shaw, P. H., Boyd, P. A., Baumann,
H., and Hastie,
N. D. (1984) Nature
311,175-177
41. Rosenberg,
S., Barr, P. J., Najarian,
R. C., and Hallewell,
R. A.
(1984) Nature
312, 77-80
42. Bao, J., Sifers, R. N., Kidd, V. J., Ledley,
F. D., and Woo, S. L.
C. (1987) Biochemistry
26, 7755-7759
43. Bradford,
M. M. (1976) Anal. Biochem.
72,248-254
44. Beatty,
K., Bieth, J., and Travis,
J. (1980) J. Biol. Chem. 255,
3931-3934
45. Bjork,
I., Jackson,
C. M., Jornvall,
H., Lavine,
K. K., Nordling,
K., and Salsgiver,
W. J. (1982) J. Biol. Chem. 257, 2406-2411
46. Bjork, I., and Fish, W. W. (1982) J. Biol. Chem. 257,9487-9493
47. Fish, W. W., and Bjork,
I. (1979) Eur. J. Biochem.
101, 31-38
48. Walev.
S. G. (1985) Biochem.
J. 227, 843-849
49. Lain;,
A., Davril,
M., Rabaud,
M., -Vercaigne-Marko,
D., and
Hayem,
A. (1985) Eur. J. Biochem.
151,327-331
50. Jallat, S., Carvallo,
D., Tessier,
L. H., Roecklin,
D., Roitsch,
C.,
Dgushi,
F., Crystal,
R. G., and Courtney,
M. (1986) Protein
Eng. 1,29-35
51. Bock, S. C., Skriver,
K., Nielsen,
E., Thegersen,
H-C., Wiman,
B., Donaldson,
V. H., Eddy, R. L., Marrinan,
J., Radziejewska,
E., Huber,
R., Shows,
T. B., and Magnusson,
S. (1986) Biochemistry
25,4292-4301
52. Holmes,
W. E., Lijnen,
H. R., and Collen, D. (1987) Biochemistry

26,5133-5140
53. Travis,
J., Matheson,
N. R., George,
P. M., and Carrell,
R. W.
(1986) Biol. Chem. Hoppe-Seyler
367,853-859
54. Stephens,
A. W., Siddiqui,
A., and Hirs, C. H. W. (1988) J. Biol.
Chem. 263,15849-15852
55. Matheson,
N.. Bathurst,
I., and Travis.
J. (1989) Biochem.
Biophys. R&. Cimmun.
154, 271-277
56. Mierzwa.
S.. and Chan. S. K. (1987) Biochem.
J. 246. 37-42
57. Schechter,
I., and Berg&,
A. (i967)Biochem.
Biophys:
Res. Commun. 27, 157-162
58. Glover,
G. I., Schasteen,
C. S., Liu, W.-S.,
and Levine,
R. P.
(1988) Mol. Zmmunol.
25, 1261-1267
59. Schasteen,
C. S., McLafferty,
S. A., Clover,
G. I., Han, C. Y.,
Mayden,
J. C., Liu, W.-S.,
and Levine,
R. P. (1988)
Mol.
Zmmurwl.
25,1269-1275

Downloaded from http://www.jbc.org/ at TENNESSEE TECH UNIVERSITY on March 17, 2014

10. Tsuda,
M., Umezawa,
Y., Masuyama,
M., Yamaguchi,
K., and
Katsunuma,
T. (1987) Biochem.
Biophys.
Res. Commun.
144,
409-414
11. Braun-Breton,
C., Rosenberry,
T. L., and Pereira
da Silva, L.
(1988) Nature
332, 457-459
12. Banyal,
H. S., Misra,
G. C., Gupta, C. M., and Dutta, G. P. (1981)
J. Purasitol.
67, 623-627
13. Dluzewski,
A. R., Rangachari,
K., Wilson,
R. J., and Gratzer,
W.
B. (1986) Exn. Parusitol.
62, 416-422
14. McKerrow,
J.-H.,
Pino-Heiss;
S., Lindquist,
R., and Werb,
Z.
(1985) J. Biol. Chem. 260, 3703-3707
15. Asahi,
M., Lindquist,
R., Fukuyama,
K., Apodaca,
G., Epstein,
W. L., and McKerrow,
J. H. (1985) Biochem.
J. 232, 139-144
16. Yanagisawa,
M., Kurihara,
H., Kimura,
S., Tomobe,
Y., Kobayashi,
M., Mitsui,
Y., Yazaki,
Y., Gota, K., and Masaki,
T.
(1988) Nature
332, 411-415
17. Abraham,
C. R., Selkoe,
D. J., and Potter,
H. (1988) Cell 52,
487-501
18. Tahara,
E., Ito, H., Taniyama,
K., Yokozaki,
H., and Hata, J.
(1984) Human
Puthol.
15,957-964
19. Ordone,
N. G., and Manning,
J. T. (1984) Am. J. Gastro.
79,
959-963
20. Schill, W. B. (1976) Andrologia
8, 359-364
21. Casslen,
B., and Ohlsson,
K. (1981) Contraception
23, 425-434
22. Travis,
J., Bowen,
J., and Baugh,
R. (1978) Biochemistry
17,

1207