(Received for publication, February 19, 1997)
From the Departments of Pharmacology,
Pathology and Laboratory Medicine, and § Medicine,
¶ Center for Thrombosis and Hemostasis, The University of North
Carolina School of Medicine, Chapel Hill, North Carolina 27599
Heparin cofactor II (HCII) is presumed to be a
physiological inhibitor of the serine proteinase thrombin. The reaction
between HCII and thrombin is quite unique, because it involves an
unusual HCII-reactive site loop sequence of
Leu444-Ser445, requires the presence of
glycosaminoglycans for optimal activity and involves a protein-protein
interaction besides the reactive site loop-active site interaction
characteristic of serine proteinase inhibitor-serine proteinase pairs.
Two mutations at a unique HCII residue, Arg200 Ala or
Glu, were generated by site-directed mutagenesis. The mutations did not
alter either HCII binding to heparin-Sepharose or HCII inhibition of
thrombin in the presence of heparin or dermatan sulfate, suggesting
that Arg200 is not part of the glycosaminoglycan binding
site of HCII. In the absence of glycosaminoglycan, there was a
significant increase in
-thrombin inhibition by the
Arg200 mutants as compared with wild type recombinant HCII
(wt-rHCII), whereas inhibition rates with chymotrypsin were identical.
Inhibition of
T-thrombin, which lacks anion-binding
exosite 1 ((ABE-1), the region of
-thrombin that interacts with the
acidic domain of HCII), was significantly reduced compared with
-thrombin, but the reduction was more dramatic for the
Arg200-rHCII mutants. Hirugen, which binds to ABE-1 of
-thrombin, also diminished inhibition of
-thrombin by the
Arg200-rHCII mutants to nearly wt-rHCII levels. Both
Arg200-rHCII mutants had significantly increased
ka values as compared with wt-rHCII, whereas the
kd rates were unchanged. Collectively, these
results suggest that the improved inhibitory activity of the
Arg200-rHCII mutants is mediated by enhanced interactions
between the acidic domain and ABE-1, resulting in an increased
HCII-thrombin association rate.
Serine proteinase inhibitors (serpins)1 are a superfamily of proteins whose primary function is to regulate the proteolytic activity of serine proteinases involved in such processes as coagulation, fibrinolysis, complement activation, inflammation, and tumor metastasis (Refs. 1 and 2 and reviewed in Ref. 3). Heparin cofactor II (HCII) belongs to a subfamily of serpins whose activity is greatly accelerated upon binding to glycosaminoglycans, such as heparin, heparan sulfate, and dermatan sulfate (4, 5). In vivo, glycosaminoglycan-containing proteoglycans found on cell surfaces and in extracellular matrix serve to accelerate this reaction (6-8). The physiological target of HCII is presumed to be thrombin, a pluripotent coagulation proteinase that participates in inflammation and wound healing processes based on its chemotactic, mitogenic, and cytokine-like action on vascular smooth muscle cells, monocytes, and fibroblasts (9-12). Thrombin activity generated during blood coagulation is regulated primarily by antithrombin, a heparin-binding serpin that inhibits most coagulation proteinases. However, extravascular thrombin activity associated with inflammation and wound healing processes is thought to be regulated by HCII, which exhibits remarkable specificity for thrombin (13, 14).
HCII possesses several characteristics for thrombin specificity that render it unique among heparin-binding serpins. HCII is the only heparin-binding serpin that binds dermatan sulfate to accelerate thrombin inhibition (15). The heparin and dermatan sulfate binding sites, which are distinct but overlapping, are localized primarily in the D-helix region (16-19). HCII is also unique because it has Leu444 at the P1 position, whereas most thrombin-inhibiting serpins (like antithrombin and protein C inhibitor) and typical thrombin substrates contain an Arg at the P1 site (20, 21). The P1 residue, which is located on an exposed loop that interacts with the active site of the proteinase, determines in large part the proteinase specificity of the serpin (for a review, see Ref. 3). The presence of a P1 Leu in HCII enables it to inhibit chymotrypsin, a nonphysiological target, more rapidly than thrombin in the absence of glycosaminoglycans (22). HCII with an Arg substituted for the P1 Leu no longer inhibits chymotrypsin (23). Interestingly, this mutant has an increased thrombin inhibition rate in the absence of glycosaminoglycans, but is also proteolytically inactivated by thrombin in the presence of heparin (23, 24).
Although HCII is ~30% identical in primary structure to antithrombin and other serpins (25), it has an unusual amino-terminal extension of approximately 80 residues (26, 27). The amino terminus contains a tandem repeat of two acidic stretches that are somewhat homologous to the carboxyl terminus of the leech thrombin inhibitor, hirudin (28). The acidic domains of both HCII and hirudin bind to anion-binding exosite-1 (ABE-1) of thrombin (29). The acidic domain of HCII is also thought to bind intramolecularly to the D-helix, the glycosaminoglycan binding site on HCII. Binding of glycosaminoglycans to the D-helix is thought to displace the acidic domain and promote its interaction with ABE-1 of thrombin (30, 31). The acidic domain interaction with ABE-1 appears to be the driving force for the rapid inhibition of thrombin by HCII in the presence of glycosaminoglycans and compensates for the unfavorable P1 Leu residue (19, 30, 32).
We are studying HCII to better understand the role of specific amino
acid residues in this unique thrombin inhibition reaction. A comparison
of serpin sequences shows that HCII is the only heparin-binding serpin
with a basic residue at Arg200 (2). Arg200 of
HCII is in strand 2 of sheet A, adjacent to the dermatan sulfate-binding region of the D-helix; thus it may be poised to play a
unique role in regulating HCII activity. In this study, we have found
that Arg200 promotes the interaction of the acidic domain
with HCII in the absence of glycosaminoglycans, but is not involved in
glycosaminoglycan binding. We propose that the function of
Arg200 is to keep HCII essentially "inactive" when it
is circulating in the blood stream unbound to glycosaminoglycans.
Human wild type recombinant HCII (wt-rHCII) (cDNA
kindly provided by Dr. Douglas M. Tollefsen, Washington University
School of Medicine, St. Louis, MO) was previously expressed in the
baculovirus expression system and characterized (33). To generate
R200A-rHCII and R200E-rHCII, site-directed mutagenesis was performed by
the method of Kunkel on full-length human HCII cDNA subcloned via flanking EcoRI sites into the pBluescript SK+ mutagenesis
and cloning vector (Stratagene) (34). A degenerate oligonucleotide was
used to introduce two point mutations in the HCII cDNA
(683CGG GCG or GAG), which caused substitutions of Ala
or Glu, respectively, at residue Arg200. The mutations were
identified by DNA sequencing (Sequenase® Version 2.0, U. S.
Biochemical Corp.), and positive clones were sequenced in full to
verify the absence of erroneously introduced mutations. Subcloning into
baculoviral transfer vector pVL1392 (PharMingen) and cotransfection
with linearized BaculoGoldTM (PharMingen) Autographica
californica nuclear polyhedrosis virus in Spodoptera frugiperda (Sf9, Invitrogen) insect cells was performed
as described previously (33). The infectious medium was collected 4 days post-transfection and was further amplified in fresh
Sf9 cells. Production of rHCII was verified by immunoblot
analysis of whole cell lysates from infected cells. Sf9
cells were maintained in spinner flasks in Grace's medium (JRH
Scientific) supplemented with 10% fetal bovine serum (HyClone), 0.3 g/liter L-glutamine (Life Technologies, Inc.), and 50 µg/ml gentamicin (Life Technologies, Inc.).
High-FiveTM insect
cells (Invitrogen) grown at 27 °C in serum free ExCell 401TM medium
with L-glutamine (JRH Scientific) were used to express
rHCII. Two T150 flasks of High-FiveTM cells were infected with viral
stock (150 µl/flask), and the cell supernatant was collected after 2 days and centrifuged at 350 × g for 5 min to remove
cell debris. The medium was diluted with an equal volume of HPN buffer,
pH 6.5 (20 mM HEPES (Boehringer Mannheim), 0.1% polyethylene glycol 8000 (Sigma), 0.05% NaN3) and
batch-adsorbed with 0.5 ml of heparin-Sepharose beads (Pharmacia
Biotech Inc.) for 1 h at 4 °C. rHCII was eluted from the
heparin-Sepharose with 0.5 M NaCl in HPN buffer, pH 7.4, after two washes in 75 mM NaCl in HPN, pH 6.5, buffer. The
heparin-Sepharose eluate was diluted in HPN buffer to a final
concentration of 50 mM NaCl, pH 7.8, and batch-adsorbed
with 0.5 ml of Q-Sepharose (Sigma) for 1 h at 4 °C. After two
washes in 50 mM NaCl, the protein was eluted with 0.5 M NaCl in HPN, pH 7.8, buffer, aliquoted, and stored at
70 °C.
A direct enzyme-linked immunosorbent assay using a mouse anti-HCII monoclonal antibody and a goat IgG-conjugated to alkaline phosphatase (Sigma) was used as described previously to measure rHCII concentrations (33). Human plasma HCII, purified as described previously (35), was used for the standard curve. Assays were performed in 96-well microtiter plates and color development was monitored at 405 nm on a Vmax microplate reader (Molecular Devices).
Heparin-Sepharose ChromatographyThe relative affinity of wt-rHCII and the R200-rHCII mutants for immobilized heparin was determined by fast protein liquid chromatography using a Pharmacia Biotech Inc. system and a 1-ml heparin-Sepharose column. The samples were dialyzed into 20 mM HEPES, 50 mM NaCl, 0.1% polyethylene glycol, pH 7.4, loaded onto the column with dialysis buffer, and eluted with a linear 1.0 ml/min salt gradient of 50 mM to 0.5 M NaCl. 1.0 µg of recombinant HCII was loaded, and 20 × 1-ml fractions were collected. The fractions were analyzed by thrombin inhibition assays in the presence of 10 µg/ml heparin, as described below.
Proteinase Inhibition AssaysHuman -thrombin was
isolated and prepared as described previously (36), and bovine
chymotrypsin was purchased from Sigma.
T-Thrombin was
prepared by limited proteolysis of plasma-derived
-thrombin with
L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin (Cooper Biomedical) [(37) as modified in (32)]. The activity
of
T-thrombin was then verified by chromogenic substrate cleavage and fibrin clotting assays. Chromogenic substrates were tosyl-Gly-Pro-Arg-
-nitroanilide (150 µM; Boehringer
Mannheim) for thrombin and
N-succinyl-Ala-Ala-Pro-Phe-
-nitroanilide (500 µM; Sigma) for chymotrypsin. Hirugen (residues 53-64)
was from Multiple Peptide Systems. A control peptide corresponding to
the reverse sequence of the HCII acidic domain (residues 47-61) was synthesized on a SynergyTM peptide synthesizer (Applied Biosystems). Proteinase inhibition assays for wt-rHCII and the R200-rHCII mutants were performed in 96-well enzyme-linked immunosorbent assay plates (previously coated with 2 mg/ml bovine serum albumin) at room temperature in HNPN, pH 7.4, buffer containing 2 mg/ml bovine serum
albumin. In the absence of glycosaminoglycan, 100 nM HCII and 1 nM thrombin were incubated together for 30-180 min
in the presence of 50 µg/ml polybrene. The thrombin-HCII association time was 90 min for assays performed in the presence of hirugen. For
the heparin (Diosynth, Oss, The Netherlands) and dermatan sulfate
(Calbiochem; nitrous acid-treated to remove contaminating heparin and
heparan sulfate) template curves, 5 nM HCII, and 0.5 nM thrombin were incubated together for 20 s. The
reactions were quenched by the addition of a chromogenic substrate
solution containing 3 mg/ml polybrene, and color development was
monitored at 405 nm on a Vmax microplate reader.
Second order inhibition rate constants (k2,
M
1 min
1) were measured in
triplicate on two to four different preparations of rHCII. The rates
were obtained under pseudo-first order reaction conditions as described
previously and were calculated using the equation
k2 = (
ln a)/t [I],
where a is the residual proteinase activity, t is
the time, and [I] is the HCII concentration (38).
Slow binding kinetic assays were
performed in 96-well bovine serum albumin-coated enzyme-linked
immunosorbent assay plates at room temperature in HNPN, pH 7.4, buffer
containing 2 mg/ml bovine serum albumin. A low-evaporation lid coated
with anti-fogging agent (Molecular Devices) was used to minimize
evaporation. Chromogenic substrate was S-2266
(D-Val-Leu-Arg--nitroanilide; Kabi Pharmacia). S-2266
was selected from several chromogenic substrates tested because its
high Km (262 µM; Ref. 39) permitted
adequate inhibition of thrombin by HCII under the experimental
conditions. Color development was monitored as described above. The
reaction was started by the addition of 0.5 nM thrombin to
wells containing varying concentrations of HCII (25-400
µM range) and 500 µM S-2266, and readings
were taken at 10-min intervals for 8 h. Data points were excluded
from the analysis when the level of substrate utilization exceeded
10%. Control assays indicated that the thrombin was stable during the
course of the experiment.
The competition between substrate and inhibitor for thrombin can be described by the following scheme (39).
![]() |
![]() |
![]() |
(Eq. 1) |
![]() |
(Eq. 2) |
![]() |
(Eq. 3) |
R200A-rHCII and R200E-rHCII were engineered, expressed, and purified as described under "Experimental Procedures." The protein yield from a typical preparation was essentially the same as for wt-rHCII (50-100 µg purified from two T150 flasks). Immunoblot analysis and SDS-PAGE demonstrated the mutants had the same electrophoretic mobility as wt-rHCII (data not show).
Heparin-Sepharose Affinity Chromatography of wt-rHCII and the Arg200 MutantsGradient salt elution from heparin-Sepharose was performed to determine whether the mutations altered the apparent affinity of HCII for immobilized heparin. Since the intermolecular interactions between heparin and HCII are thought to be primarily ionic, the concentration of NaCl required to elute the proteins is a measure of their relative affinity (41). There were no significant differences in the NaCl concentrations (mM) required to elute the proteins: 270 ± 40 for wt-rHCII (n = 3), 260 ± 20 for R200A-rHCII (n = 4), and 270 ± 40 for R200E-rHCII (n = 3).
Glycosaminoglycan-acceleratedThe
Arg200 mutants were assayed for their ability to inhibit
thrombin in the presence of glycosaminoglycans. Fig. 1
shows that both HCII mutants exhibited typical bell-shaped curves for
inhibition of -thrombin in the presence of increasing concentrations
of heparin or dermatan sulfate. While the optimal heparin concentration is about 2-fold higher for the Arg200 mutants (500 µg/ml)
as compared with wt-rHCII (200 µg/ml) (Fig. 1A), the
second order inhibition rate constants (k2,
M
1 min
1) are not significantly
different (see Table I). Likewise, dermatan sulfate-accelerated thrombin inhibition was identical for wt-rHCII and
the Arg200 mutants, both in terms of optimal dermatan
sulfate concentration and maximal inhibition rate (Fig. 1B
and Table I).
|
wt-rHCII and
the Arg200 mutants were assayed for their ability to
inhibit -thrombin,
T-thrombin, and chymotrypsin in
the absence of glycosaminoglycans. The second order rate constants of
inhibition (k2, M
1
min
1) are summarized in Table I. With
-thrombin, both
mutants have significantly higher inhibition rates than wt-rHCII
(p < 0.05). The k2 values for
R200A-rHCII and R200E-rHCII are 3.5- and 5.6-fold higher, respectively,
than wt-rHCII. In contrast, the k2 values for
wt-rHCII and the Arg200 mutants were identical with
chymotrypsin (p > 0.05), indicating that the mutations
did not alter the conformation of the reactive site loop.
To determine whether the increased inhibitory activity of HCII with
-thrombin is due to enhanced interactions between the acidic domain
and ABE-1, inhibition assays were performed with
T-thrombin.
T-Thrombin lacks portions of
ABE-1, which have been removed by limited proteolysis with trypsin. All
three HCII variants had lower inhibition rates with
T-thrombin as compared with
-thrombin, but the
greater decrease in activity was observed with the Arg200
mutants. Although the k2 values of the mutants
are still higher than wt-rHCII (p < 0.05), they are
both now only 2.3-fold higher than wt-rHCII (Table I).
To further examine the role of ABE-1 in the
enhanced activity of the Arg200 mutants, the rate of
-thrombin inhibition in the presence of hirugen was determined. By
binding directly to ABE-1, hirugen interferes with acidic domain-ABE-1
interactions and reduces the rate of
-thrombin inhibition by HCII
(33). Increasing amounts of hirugen resulted in a
dose-dependent blockage of the HCII-thrombin reaction, with
a maximal response at 100 µM for all three HCII proteins
(data not shown). Hirugen has a greater effect on the ability of the
Arg200 mutants to inhibit
-thrombin than it does on
wt-rHCII. At 100 µM hirugen, the
-thrombin inhibition
rates (k2, M
1
min
1) for R200A-rHCII and R200E-rHCII were reduced to
2.2 ± 0.8 × 104 and 2.8 ± 0.6 × 104, respectively, as compared with 1.0 ± 0.4 × 104 for wt-rHCII (Fig. 2). This effect is
specific since a control peptide that was highly negatively charged
could not block the HCII-thrombin reactions (Fig. 2).
Slow Binding Kinetics for the Inhibition of
To
determine whether the enhanced activity of the mutants is due to an
increased HCII-thrombin association rate (ka) or to
a decreased HCII-thrombin dissociation rate (kd), a
slow binding kinetics assay was developed for HCII inhibition of
thrombin. In slow binding kinetics, the inhibitor and substrate compete
for binding to the proteinase. Shown in Fig. 3 are
representative time courses of -thrombin inhibition at different
HCII concentrations. Analysis of the raw data yielded a good fit to
Equation 1, typically with correlation coefficients
(r2) of 0.999. The values obtained for
k
were used to estimate ka and
kd, as described under "Experimental
Procedures." A representative plot of k
versus
HCII concentration, shown in Fig. 4, demonstrates the
significantly increased ka of the Arg200
mutants as compared with wt-rHCII. The ka values
(Table II) for R200A-rHCII and R200E-rHCII are 5- and
9-fold higher, respectively, than wt-rHCII. Interestingly, the
calculated kd for all the HCII variants are not
significantly different (p > 0.05).
|
We initially hypothesized that Arg200 of HCII would be important for the glycosaminoglycan-mediated inhibitory activity of HCII with thrombin because no other heparin-binding serpin has a basic residue at the analogous position. Furthermore, Arg200 is adjacent to the D-helix region of HCII, previously established to be important for activity. We expected that changing this positively charged residue to an uncharged alanine or a negatively charged glutamic acid would decrease both heparin-Sepharose affinity and glycosaminoglycan-mediated thrombin inhibition. However, mutations at Arg200 did not affect either heparin binding or heparin- and dermatan sulfate-accelerated thrombin inhibition. Although slightly higher concentrations of heparin were required to optimally stimulate inhibition of thrombin by the mutants, the maximal inhibition rate was nearly identical to wt-rHCII. Mutation of critical D-helix residues in previous studies has led to both significantly decreased glycosaminoglycan-accelerated thrombin inhibition and reduced heparin-Sepharose affinity (16-18). The finding that R200A-rHCII and R200E-rHCII are fully active in the presence of both heparin and dermatan sulfate and that their heparin-Sepharose elution profiles are nearly identical indicates that Arg200 is not critical for glycosaminoglycan binding.
Interestingly, the thrombin inhibition rates of R200A-rHCII and
R200E-rHCII in the absence of glycosaminoglycans were significantly increased. The increased inhibitory activity of the mutants appears to
be caused by an enhanced interaction between the HCII acidic domain and
ABE-1 of thrombin, since inhibition rates with
T-thrombin are more significantly reduced for the
mutants than for wt-rHCII. Furthermore, the addition of hirugen to the
HCII-thrombin reaction reduces the activity of the mutants to almost
the same level as that of wt-rHCII. However, even with
T-thrombin or
-thrombin/hirugen, the mutants are
still more active than wt-rHCII, most likely because the acidic domain
binding site on ABE-1 is neither totally removed in
T-thrombin nor completely blocked by hirugen. We have
shown previously that dermatan sulfate accelerates
T-thrombin inhibition by HCII by 30-fold, indicating
that the acidic domain still interacts with the remaining portions of
ABE-1 in a productive manner (32). Finally, the identical inhibition
rates of wt-rHCII and the Arg200 mutants with chymotrypsin
(which does not require either the acidic domain of HCII or
glycosaminoglycans for inhibition) indicate that the reactive site loop
has not been altered to an "activated" conformation, further
implicating a role for the acidic domain in the increased activity of
the mutants.
To determine whether the enhanced interaction between the acidic domain of the Arg200 mutants and ABE-1 of thrombin resulted in an increased association rate (ka) or a decreased dissociation rate (kd), a slow binding kinetics assay, previously used to characterize other serpin-proteinase reactions, was adapted for HCII-thrombin. These studies revealed that the increased inhibitory activity of the mutants is due to an increased association rate (ka) with thrombin. This suggests that the acidic domain is more involved in the initial "handshake" between HCII and thrombin than in the stabilization of the bimolecular serpin-proteinase complex. These results imply that for wild type HCII, Arg200 helps to maintain acidic domain-HCII intramolecular interactions, thus attenuating thrombin inhibition in the absence of glycosaminoglycans.
Previous studies have led to the hypothesis that the D-helix is the
intramolecular binding site for the acidic domain in HCII. A study
instrumental in developing this hypothesis showed that amino-terminal
truncation mutants of HCII that are missing one (67-rHCII) or both
(
74-rHCII) acidic repeats have greatly increased heparin-Sepharose
affinity (30). A D-helix double mutant (R184Q/K185Q-rHCII) generated by
site-directed mutagenesis was shown to have significantly decreased
heparin-Sepharose affinity, enhanced ability to form SDS-PAGE-stable
complexes with thrombin in the absence of glycosaminoglycans, but
reduced ability to form complexes in the presence of glycosaminoglycans (19). This study suggested that the increased formation of
R184Q/K185Q-rHCII-thrombin complexes in the absence of
glycosaminoglycans was due to a disruption in acidic domain
interactions with R184 and K185, resulting in increased acidic domain
interaction with ABE-1.
Our results imply that in the absence of glycosaminoglycan, the presence of an Arg at residue 200 either promotes acidic domain interactions with the D-helix of HCII or binds the acidic domain directly. Changing Arg200 to an Ala or Glu would alter the "equilibrium" and favor the interaction of the acidic domain and ABE-1, resulting in increased thrombin inhibition. These results are also consistent with the hypothesis that the mechanism of acidic domain binding to ABE-1 is part of the HCII-thrombin association reaction. Finally, this interaction would further contribute to maintaining blood plasma HCII in a conformation that would be a poor thrombin inhibitor in the absence of glycosaminoglycans.
We acknowledge the late Stuart R. Stone (1951-1996), our colleague and friend, for his advice on developing the slow binding kinetic assay used in this study.