© 1997 American Heart Association, Inc.
Articles |
Induction of the Acute-Phase Reaction Increases Heparin-Binding Proteins in Plasma
From the Departments of Pathology and Medicine, McMaster University, and the Hamilton Civic Hospitals Research Centre, Hamilton, Ontario, Canada.
Correspondence to Dr Edward Young, Division of Clinical Chemistry, Hamilton Civic Hospitals, Henderson Division, 711 Concession St, Hamilton, Ontario, Canada L8V 1C3.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract We have previously demonstrated that the nonspecific binding of unfractionated heparin (UFH) to plasma proteins has a marked modulating effect on its anticoagulant activity. Since some heparin-binding proteins are also acute-phase-reactant proteins, we explored the possibility that the induction of the acute-phase response can increase the plasma concentrations of heparin-binding proteins. The recovery of a fixed amount of UFH or low-molecular-weight heparin (LMWH) added in vitro to rat plasma samples obtained at various time intervals after the administration of intravenous endotoxin or subcutaneous turpentine was compared with that of saline-treated control animals. The anti–factor Xa activity was measured in the plasma samples before and after the addition of a chemically modified low-affinity heparin (LAH) to displace the proportion of the added heparin that is reversibly bound to plasma proteins. Our results show that at 6 hours post–endotoxin and at 24 hours post–turpentine treatment, virtually no anti–factor Xa activity could be measured in the plasma samples, while the expected levels were obtained for control plasma. After the addition of LAH to displace protein-bound UFH, essentially the same anti–factor Xa levels were measured in the plasma from all three treatment groups. These results indicate that induction of the acute-phase reaction can dramatically increase the levels of heparin-binding proteins in rat plasma. In addition, we compared the anti–factor Xa recovery of UFH with that of an LMWH from the plasma of endotoxin- and saline-treated rats and demonstrated that LMWH binds less to plasma proteins than UFH, even in plasma in which the levels of heparin-binding proteins are markedly elevated. The recovery of a fixed amount of UFH added in vitro to human plasma from septic patients was also reduced, but not to the same extent as seen in rat plasma. Removal of candidate heparin-binding and acute-phase proteins by immunodepletion indicated that vitronectin plays an important role in the nonspecific binding of UFH in patient plasma.
Key Words: heparin-binding proteins • acute-phase reaction • unfractionated heparin • low-molecular-weight heparin
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The clinical efficacy of UFH is dependent on obtaining an anticoagulant effect, which is achieved by adjusting the dosage to maintain the activated partial thromboplastin time response within a defined therapeutic range.1 2 The dose response to UFH is highly variable among patients with thromboembolic disease.3 Some patients require very high doses to exceed the lower limit of the therapeutic range; these patients have been characterized as being heparin resistant.4
The mechanisms for the marked variability in dose response and for heparin resistance have been clarified by the results of recent studies. The major anticoagulant effect of heparin is mediated by a unique pentasaccharide sequence with high-affinity binding to AT III, which is present in only one third of heparin molecules.5 6 7 8 Heparin binding to AT III produces a conformational change in the cofactor and so markedly accelerates its ability to inactivate the coagulation enzymes thrombin (factor IIa), factor Xa, and factor IXa.9 However, UFH also binds nonspecifically to a number of other plasma proteins, and there is evidence that these heparin-binding proteins compete with AT III for the binding to the anticoagulantly active heparin molecules, thereby impairing their anticoagulant effect.10 11 12 The proportion of anticoagulantly active UFH that binds nonspecifically to plasma proteins can be demonstrated by measuring the anti–factor Xa activity before and after the addition of an excess of LAH to plasma containing anticoagulantly active UFH. The chemically modified LAH, which is essentially devoid of anti–factor Xa activity, produces a substantial increase in anti–factor Xa activity when added to normal heparin-containing plasma, because it displaces anticoagulantly active UFH bound nonspecifically to plasma proteins, allowing the anticoagulant to interact with AT III.13 14 The levels of these heparin-binding proteins are increased in many patients with thromboembolic disease. Therefore, compared with the effects observed with normal plasma, when UFH was added to the plasma of patients with thromboembolism, the recovery of UFH, as anti–factor Xa activity, was reduced, and there was a greater and more variable increase in activity after the addition of LAH.14 Similarly, reversible nonspecific binding to plasma proteins has also been observed in plasma samples obtained from patients treated with UFH, and this phenomenon of reversible heparin binding was shown to be a major determinant of the anticoagulant response to fixed doses of UFH in these patients.15 Based on these observations, we have proposed that the variable dose response to UFH, of which one manifestation is heparin resistance, is contributed to in an important way by heparin-binding proteins, which are increased to a variable extent during illness. In contrast to UFH, LMWHs exhibit less binding to plasma proteins,16 a property that may explain their longer plasma half-life and less variable anticoagulant response to a fixed dosage.17 18 19
Since some of the heparin-binding proteins, such as fibronectin,20 vitronectin,21 and heparin cofactor II,22 are also acute-phase-reactant proteins,23 24 25 we hypothesized that increased levels of acute-phase-reactant proteins may contribute to the variability of anticoagulant response to UFH observed in sick patients. To explore the influence of acute-phase-reactant proteins on the phenomenon of heparin resistance, we induced an acute-phase response in rats and studied its effect on both plasma heparin recovery and reversible heparin binding. The acute-phase response was induced either by intravenous injection of endotoxin or the subcutaneous injection of turpentine. We also studied the effect of the acute-phase reaction on heparin recovery in human plasma by adding a fixed amount of UFH to the plasma of patients with sepsis. Our findings show that induction of the acute-phase reaction can lead to a marked reduction in the recovery of UFH (measured as anti– factor Xa activity) from the plasma of endotoxin- and turpentine-treated animals. A similar but less prominent reduction was obtained in the plasma from sick, septic patients. Further, we present evidence that plasma vitronectin is responsible for a large part of the nonspecific binding of UFH in patient plasma.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Induction of the Acute-Phase Reaction in Rats
We employed two well-characterized experimental models to induce acute inflammation in the rat.24 26 Male Sprague- Dawley rats (250 to 300 g, n=3 to 5 per time point) were treated with either an intravenous injection (200 µL, 4 mg/kg) of purified bacterial lipopolysaccharide (Escherichia coli, 055:B5, Sigma Chemical Company) via the tail vein or a subcutaneous injection (100 µL) of commercially available turpentine. The levels of stimuli were chosen to elicit a strong acute-phase response. The control animals received 200 µL of sterile saline intravenously.
Human Blood Samples
Blood samples were obtained from patients with a clinical
diagnosis of sepsis. The diagnosis of sepsis was based on clinical
criteria, including a suspected or established infection, shaking
chills, fever or hypothermia, tachycardia, and
leukocytosis. Blood samples were also obtained from healthy
volunteers.
Preparation of Blood Samples
At various times after injection, the animals were
anesthetized by isoflurane inhalation and killed by cardiac
puncture exsanguination. Blood was collected 9:1 into 3.8% sodium
citrate. Blood from septic patients and normal volunteers was collected
via clean venipuncture into evacuated tubes (Vacutainer,
Becton Dickinson) containing buffered sodium citrate. Both animal and
human studies were conducted in accordance with the institutional
guidelines of McMaster University and Hamilton Civic Hospitals. The
citrated blood was centrifuged at 2000g for 15
minutes at room temperature and the platelet-poor plasma stored
frozen at -70°C until assay.
Heparin Assay
The recovery of a fixed amount of UFH (porcine mucosal heparin,
178 USP U/mg, Sigma Chemical Company) added in vitro to the plasma
samples was determined as anti–factor Xa activity according to the
method of Teien and Lie.27 The Stachrom heparin kit from
Diagnostica Stago was adapted for use on a centrifugal
analyzer.28 This assay measures the ability of UFH
to potentiate the inhibition of factor Xa by AT III by using a
chromogenic substrate. This assay contains excess AT III in
the reaction mixture. In some experiments, LMWH (enoxaparin,
Rhône-Poulenc Rorer) was used in place of UFH. Pooled rat or
human plasma to which known concentrations of UFH or LMWH were added in
vitro were used as standards. Pooled plasma was also used to dilute the
heparinized samples into the range of the standard curves. All
anticoagulant assays were performed in duplicate.
Measurement of Reversible Heparin Binding
LAH was used to displace anticoagulantly active heparin, which
is bound nonspecifically to plasma proteins other than AT III
(reversible heparin binding). LAH was prepared from unfractionated pig
mucosal heparin by periodate oxidation followed by borohydride
reduction as previously described.13 29 Any remaining
traces of high-affinity material were removed by affinity
chromatography on an AT III–Sepharose column. The
chemically modified heparin retains its molecular weight distribution
and charge density, but its anticoagulant activity is drastically
reduced. The specific anti–factor Xa activity of the LAH was <1.0
U/mg.13
Acute-Phase-Response Proteins
Total plasma protein was determined by the biuret method (Abbott
Laboratories). Plasma protein electrophoresis was performed in agarose
gels by using the Paragon electrophoresis system (Beckman Instruments).
The stained albumin and fibrinogen bands were quantified by
densitometric scanning and expressed as a percentage of the total
protein value. C-reactive protein was measured by using a nephelometric
assay (Kallestad Inc).
Immunodepletion of Patient Plasma
Specific antibodies bound to protein G Sepharose (Gamma Bind G
Sepharose, Pharmacia Biotech) were used to immunodeplete candidate
heparin binding from pooled plasma obtained from septic patients. Based
on the binding capacity of protein G, excess amounts of each antibody
(see below) were mixed with separate aliquots of protein G Sepharose at
4°C for 18 hours followed by extensive washing with
phosphate-buffered saline to remove unbound antibody.
Anti–histidine-rich glycoprotein (rabbit antiserum) was
obtained from Diagnostica Stago. The following antibodies
were all supplied by Affinity Biologicals: anti-fibronectin (rabbit IgG
fraction), anti-vitronectin (affinity-purified sheep IgG
fraction), anti–platelet factor 4 (sheep IgG fraction), and normal
rabbit and sheep IgG fractions. For immunodepletion, 250 µL of pooled
patient plasma was mixed with 125 µL of antibody-bound protein G for
1 hour at 4°C. The plasma was recovered by
centrifugation at 1000g for 5 minutes. To
ensure complete removal of vitronectin or fibronectin from
their respective plasma, the immunodepletion procedure was repeated
three times with fresh aliquots of antibody-bound protein G. As
controls, pooled patient plasma was treated with either normal sheep or
rabbit IgG bound to protein G and taken through the entire
immunodepletion procedure as indicated above. Western blotting (7.5%
SDS– polyacrylamide gel electrophoresis transferred to
nitrocellulose) was used to confirm the completeness of
immunodepletion. Goat anti-rabbit–conjugated peroxidase (Bio-Rad) and
rabbit anti-sheep–conjugated peroxidase (KPL) were used as the
secondary antibodies followed by chemiluminescence detection (DuPont
NEN).
Statistical Analysis
Results are presented as mean±SD. The results were
analyzed using ANOVA followed by Scheffé's test for
pairwise comparisons or by Student's t test. Differences
are considered significant when P<.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Recovery of UFH Added In Vitro to the Plasma From Endotoxin- or Turpentine-Treated Rats
A fixed amount of UFH (0.35 anti–factor Xa U/mL) was added in vitro to the plasma samples according to the specific anticoagulant activity provided by the manufacturer. The heparin concentration was determined in each plasma sample in terms of anti-factor Xa activity. As shown in Fig 1A
, when 0.35
anti–factor Xa U/mL UFH was added to the plasma obtained from
endotoxin-treated animals, the amount of UFH recovered ranged from
0.22±0.03 to 0.34±0.06 anti–factor Xa U/mL during the first 3 hours.
This trend for a reduction in recovery of added UFH is not
statistically significant. However, when 0.35 anti–factor Xa U/mL UFH
was added to the plasma obtained from endotoxin-treated animals 6 hours
after receiving intravenous endotoxin, virtually no
anti–factor Xa activity could be measured, and this effect was still
present at 24 hours. These results indicate that the anticoagulant
activity of the added UFH was being neutralized by factors present
in the plasma samples. A similar but delayed effect was observed in the
rat plasma from turpentine-treated animals (Fig 1B). Thus, when 0.35
anti–factor Xa U/mL was added, the recovery of the added UFH was
between 0.32±0.01 and 0.42±0.03 anti–factor Xa U/mL during the first
6 hours, and complete neutralization of the anticoagulant activity of
the added UFH was evident 24 hours postinjection. In contrast, the
recovery of UFH added in vitro to the plasma from saline-treated
control animals did not change significantly (between 0.28±0.06 and
0.34±0.06 anti–factor Xa U/mL) over 24 hours (Fig 1C).
|
The plasma concentrations of two typical acute-phase-reactant proteins
(albumin and fibrinogen) were determined in the 24-hour samples
from the endotoxin- and turpentine-treated animals and compared with
the levels found in control animals. The 24-hour samples were chosen
because the recovery of UFH added in vitro to these samples was
undetectable. As shown in Table 1, there
is an approximate 30% decrease in albumin levels in the
endotoxin- and turpentine-treated groups, while the fibrinogen levels
increased by 50% and 440%, respectively, compared with the control,
saline group. These results indicate that both endotoxin and turpentine
produced an acute-phase response in these animals.
|
Reversible Heparin Binding in the Plasma From Endotoxin- or
Turpentine-Treated Rats
The proportion of UFH added to the plasma samples that was
reversibly bound and neutralized by nonspecific binding to plasma
proteins was determined as anti– factor Xa activity after the addition
of LAH (45 µg/mL). Since LAH possesses minimal anti–factor Xa
activity, the observed increases in anti–factor Xa activity were due
to the displacement of anticoagulantly active UFH from plasma protein
binding sites.
As shown in Fig 2C, the anti–factor Xa
activity measured in the plasma from saline-treated animals is between
1.26±0.09 and 1.59±0.06 U/mL after the addition of LAH. Thus, the
anti–factor Xa activity increased approximately fourfold after the
addition of LAH, since only 0.35 anti–factor Xa U/mL was added
initially in vitro to the plasma samples. A similar fourfold increase
(ranging from 1.41±0.09 to 1.56±0.09 anti–factor Xa U/mL) was also
observed during the first 6 hours in the plasma from the
turpentine-treated animals (Fig 2B). Even in the 24-hour samples in
which no anti–factor Xa activity could be measured in spite of the
addition of 0.35 U/mL UFH (see Fig 1B
), the anti–factor Xa activity
recovered at 24 hours post–turpentine treatment was 1.17±0.21 U/mL.
This level was lower than those seen during the other time points, but
the difference is not statistically significant. After the addition of
LAH to the plasma samples obtained from endotoxin-treated animals, the
anti–factor Xa levels were found to be between 1.18±0.05 and
1.47±0.11 U/mL for the first 3 hours (Fig 2A). This is approximately a
fivefold increase from the levels obtained before LAH addition (Fig 1A). In the 6- and 24-hour samples, in which no anti–factor Xa
activity could be demonstrated, the addition of 45 µg/mL LAH
increased the anti–factor Xa levels to 1.07±0.16 and 1.01±0.17 U/mL,
respectively. The 6-hour samples are significantly lower than the
3-hour samples, while the 24-hour samples are significantly lower than
the 1-hour and 3-hour samples. However, when 90 µg/mL LAH was
added to the 6- and 24-hour samples, higher anti–factor Xa levels were
obtained that were similar to the levels found for the earlier
time points (see Table 2). These results
show that the induction of the acute-phase reaction by either
turpentine or endotoxin administration produced a marked increase in
nonspecific binding of heparin to plasma proteins. The proportion of
UFH that was reversibly neutralized by binding to these plasma proteins
was displaced by the addition of LAH. The extent of nonspecific binding
was so great in the 6- and 24-hour samples from endotoxin-treated
animals that twice the amount of LAH was required to fully displace the
bound UFH.
|
|
Comparison Between the Recovery of UFH and LMWH From Rat
Plasma
The recovery of UFH and LMWH added in vitro was compared in the
24-hour plasma samples from endotoxin-treated animals and in 24-hour
samples from saline-treated controls. As shown in Table 2, when 0.35
U/mL UFH was added in vitro, no detectable anti–factor Xa activity
could be recovered from the plasma of the endotoxin group, while
0.28±0.06 U/mL was recovered from the plasma of the saline group. In
contrast, when 0.35 U/mL LMWH was added, 0.26±0.06 and 0.43±0.01 U/mL
was recovered from plasma of the endotoxin and saline groups,
respectively. The amount of either UFH or LMWH added in vitro was based
on the specific anticoagulant activities provided by the manufacturer.
UFH is standardized using citrated sheep plasma, while LMWH is
standardized using human plasma. Thus, the unitage assigned to each
preparation may not be directly applicable to rat plasma. This may
explain why the recovery of UFH was lower than expected, while that of
LMWH was higher than expected in the saline group. The measurement of
anti–factor Xa activity was repeated after the addition of either 45
µg/mL or 90 µg/mL LAH to displace the UFH and LMWH
that were bound nonspecifically to plasma proteins. In the case of UFH,
approximately the same level of anti–factor Xa activity was recovered
from the plasma of endotoxin-treated and saline-treated animals
(1.47±0.15 and 1.33±0.04 U/mL, respectively). However, 90
µg/mL LAH was necessary to fully displace the proportion of
UFH that was protein bound. Similarly, for LMWH, the anti–factor Xa
activity recovered is also approximately equal (0.80±0.08 and
0.75±0.06 U/mL) in both groups, but the amount recovered is about
twofold lower than that of UFH. Thus, 1.47±0.15 anti–factor Xa U/mL,
or 100% of the UFH added in vitro, was bound and neutralized by plasma
proteins compared with 0.54±0.08 U/mL, or 68% of the LMWH, in the
endotoxin group. In the saline group, the amount of anti– factor Xa
activity reversibly bound and neutralized by plasma proteins is
1.06±0.05 U/mL, or 80% of the added UFH, compared with 0.32±0.07
U/mL, or 43% of the LMWH. These results indicate that the anti–factor
Xa of LMWH is less affected by nonspecific binding to plasma proteins
in both plasma from saline-treated animals and plasma from
endotoxin-treated animals.
Recovery of UFH Added In Vitro to Plasma From Septic Patients and
Normal Volunteers
Plasma was obtained from five patients judged clinically to be
septic and from five healthy volunteers. A fixed amount of UFH (0.35
anti–factor Xa U/mL) was added in vitro to the plasma samples
according to the specific anticoagulant activity provided by the
manufacturer. The heparin concentration was determined in each plasma
sample in terms of anti–factor Xa activity before and after the
addition of 45 µg/mL LAH to displace UFH bound nonspecifically
to plasma proteins.
As shown in Table 3, when 0.35
anti–factor Xa U/mL UFH was added to plasma from normal volunteers,
the expected amount was recovered (0.35±0.03 U/mL). In contrast, when
the same amount of UFH was added to the plasma from septic patients,
only 0.26±0.04 U/mL UFH was recovered, which is significantly lower
than that recovered from normal plasma (P<.005). However,
after the addition of excess LAH (45 µg/mL) to displace UFH
bound nonspecifically to plasma proteins, the anti– factor Xa levels
found in septic plasma (0.75±0.09 U/mL) are not significantly
different from those measured in normal plasma (0.69±0.05 U/mL,
P>.2). Thus, the amount of UFH bound nonspecifically to
plasma proteins is 44% higher in plasma from septic patients than in
plasma from healthy individuals (0.49±0.09 versus 0.34±0.05 U/mL;
P<.02).
|
The plasma concentration of C-reactive protein was measured in
the plasma from septic patients as a marker for the acute-phase
response. As shown in Table 3, the levels of C-reactive protein
(150±38 mg/L) are elevated compared with normal (<5
mg/L). These values are indicative of acute inflammation and
moderate bacterial infection.
Recovery of UFH Added In Vitro to Patient Plasma Immunodepleted
of Candidate Heparin-Binding Proteins
To determine whether a single plasma protein or several
proteins acting in concert are involved in the increased nonspecific
binding of UFH, we immunodepleted either platelet factor 4,
histidine-rich glycoprotein, fibronectin, or
vitronectin from aliquots of pooled plasma from septic
patients. The candidate proteins are well-established heparin-binding
proteins and were chosen because of their relative abundance in plasma
or because of their potential heparin-binding capacity. Fibronectin and
vitronectin are also acute-phase-reactant proteins.
The recovery of a fixed amount of UFH (0.35 U/mL) added in vitro to
aliquots of immunodepleted patient plasma and to control plasma is
shown in Table 4. As controls, patient
plasma was taken through the entire immunodepletion procedure with
nonimmune rabbit or goat IgG fractions. After immunodepletion of either
platelet factor 4 or histidine-rich glycoprotein from
patient plasma, the recovery of UFH was unchanged from that of the
control plasma. In the fibronectin-depleted plasma, there was only a
slight (7%) increase in UFH recovery. In contrast, there was a 46%
increase in UFH recovery in the vitronectin-depleted plasma
com-pared with its control. These results indicate that
vitronectin accounts for a large part of the nonspecific
heparin binding in the pooled patient plasma.
|
As judged by Western blots, >90% of each heparin-binding protein was removed by the immunodepletion procedure (data not shown). However, there was unavoidable dilution of the starting plasma. The recovery of UFH from the pooled starting plasma was 0.27 anti– factor Xa U/mL, as expected. The effect of dilution can be seen in the controls, in which 0.33 U/mL was recovered after one treatment (platelet factor 4 and histidine-rich glycoprotein) and 0.41 U/mL after three additional treatments with the antibody-coated beads (fibronectin and vitronectin).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study, we examined the effect of the acute-phase response on the nonspecific binding of UFH and LMWH to plasma proteins. We provide evidence that the induction of the acute-phase response markedly increases the reversible heparin-neutralizing capacity of rat plasma and to a lesser extent of human plasma and propose that this phenomenon is due to increased levels of heparin-binding proteins.
We employed two established experimental models (endotoxin and
turpentine administration) to induce acute inflammation in the
rat.24 26 30 The decreased levels of albumin (a
negative acute-phase protein) and increased levels of fibrinogen (a
positive acute-phase protein) measured in the rat plasma 24 hours
posttreatment with either endotoxin or turpentine confirm that the
animals underwent an acute-phase response (Table 1). Using the same
models, we demonstrated previously that endotoxin injection and
turpentine injection also increase plasma concentrations of
interleukin-6, cysteine proteinase inhibitor (the major
acute-phase reactant in rats),26 31 and
vitronectin (a heparin-binding protein).24
Our results show that when a fixed amount of UFH is added to the plasma
obtained from endotoxin-treated animals, the recovery of the added UFH
measured as anti–factor Xa activity was markedly reduced at 6 hours
and 24 hours, while in animals that received turpentine, an acute-phase
stimulus that slowly increases over 24 hours, there was a marked
reduction in the recovery of anti–factor Xa activity in plasma samples
taken at 24 hours. In contrast, the recovery of UFH from the plasma of
control, saline-treated animals did not change significantly over 24
hours (Fig 1). However, when LAH was added to the same samples to
displace UFH bound nonspecifically to plasma proteins, essentially the
same amount of UFH could be recovered from the plasma samples from all
treatment groups (Fig 2). We attribute these findings to the increase
in the concentration of heparin-binding proteins in the plasma from
endotoxin- and turpentine-treated rats.
In contrast to UFH, we previously demonstrated that LMWHs display
less nonspecific binding to plasma proteins.14 16 This
property of LMWHs may contribute to the more predictable dose response
and greater bioavailability of these compounds.17 18 19 In
keeping with our previous findings, the present study also shows
that the LMWH (enoxaparin) is less affected by nonspecific binding to
plasma proteins than UFH. Thus, in the plasma from control animals,
43% of a fixed dose of LMWH was bound to plasma proteins compared with
80% of a fixed dose of UFH (see Table 2). These values are
approximately onefold to twofold higher than those obtained when
comparable concentrations of UFH or LMWH were added to normal human
plasma and the heparin recovery determined before and after the
addition of LAH.14 15 16 This variance in UFH and LMWH
recovery from rat and human plasma is likely due to species
differences. Reduced nonspecific binding of LMWH is also demonstrable
in the plasma from endotoxin-treated animals. Thus, in plasma in which
100% of a fixed-dose UFH was bound to plasma proteins, only 68% of a
similar dose of LMWH was bound to plasma proteins.
The reduced plasma recovery of UFH observed in the animal models
of acute inflammation is also apparent in plasma obtained from septic
human patients. However, the amount of UFH recovered from rat plasma
(Table 2) was less than that recovered from human plasma (Table 3).
These findings suggest that the plasma from septic patients contained
lower levels of acute-phase-reactant proteins, which are also
heparin-binding proteins. The amount of UFH recovered from the plasma
of septic patients is similar to that previously reported for
heparin-resistant patients with venous
thrombosis.15 Aside from possible species differences, the
rats received levels of stimuli chosen to elicit a strong acute-phase
response.24 26 It is unlikely that the septic patients
were stimulated to the same degree by their disease process, since all
were treated routinely for sepsis and no cases of septic shock were
reported. The increased levels of C-reactive protein measured in the
septic patient plasma are also consistent with only a moderate
acute-phase response. Thus, the amount of acute-phase-reactant proteins
that also bind nonspecifically to UFH would be expected to be less in
septic human plasma than in rat plasma, which may explain the
differences in UFH recovery.
Both endotoxin and turpentine administration are able to drastically alter the hepatic production and plasma concentration of acute-phase proteins.30 31 Some of these plasma proteins, such as fibronectin and vitronectin, are also heparin-binding proteins.23 24 In the case of bacterial endotoxin, the response occurs more rapidly and is larger than that caused by turpentine. It has been shown that endotoxin can directly alter endothelial cell morphology and permeability,32 as well as cause a disseminated intravascular coagulation.33 34 Thus, it is possible that heparin-binding proteins such as platelet factor 4,35 36 von Willebrand factor,37 and lactoferrin38 are released in response to endotoxin injury from activated platelets, endothelium, and neutrophils, respectively. The amount of UFH that could be recovered as anti– factor Xa activity was unmeasurable in the plasma of endotoxin-treated animals 6 hours after injection compared with 24 hours for the turpentine group. Since it is likely that the 6-hour response in endotoxin-treated animals is too rapid to be caused by significant de novo synthesis of heparin-binding proteins by the liver, our findings suggest that in addition to acute-phase proteins, heparin-binding proteins may be released from storage sites in endothelial and other vascular cells.
The results of our immunodepletion experiments demonstrate that
vitronectin is an important heparin-binding protein in
patient plasma. In contrast, platelet factor 4 and histidine-rich
glycoprotein appear not to contribute to the nonspecific
binding of UFH, while fibronectin has only a marginal effect. Our group
has recently shown that vitronectin is regulated as an
acute-phase-reactant protein in both humans and rats.24
This finding is consistent with our hypothesis that increased
levels of acute-phase-reactant proteins may contribute to an increase
in nonspecific binding of heparin. Although vitronectin is
a major heparin-binding protein in patient plasma, it cannot account
for all of the nonspecific binding of UFH. Assuming at least 0.49 U/mL
UFH is bound to plasma proteins in patient plasma (see Table 3),
approximately 50% of the nonspecific binding of UFH is attributable to
vitronectin. The remaining 60% is likely bound to other
heparin-binding proteins in plasma, whose identities remain to be
elucidated. Thus, the nonspecific binding of UFH in patient plasma is
the result of several heparin-binding proteins working in concert and
is not due to a single protein.
In summary, this study has demonstrated that it is possible to increase dramatically the concentration of heparin-binding proteins in rat plasma by using either endotoxin or turpentine to induce the acute-phase inflammatory response. This observation is also relevant in human plasma, since the plasma from septic patients displays a similar but less pronounced response. These findings may have clinical implications, since nonspecific binding to plasma proteins decreases the anticoagulant effect of heparin by limiting the amount of heparin available to bind to AT III. Since some of the heparin-binding proteins are also acute-phase reactants, the levels of which increase during illness, nonspecific binding to these proteins may contribute to the variability of the anticoagulant response in patients with thromboembolic disease and to the phenomenon of heparin resistance.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by the Heart and Stroke Foundation of Ontario. T.J. Podor is a Career Investigator of the Heart and Stroke Foundation of Ontario. The authors wish to thank P. Joshua for skilled technical assistance and J. Michelis for excellent secretarial assistance.
Received January 31, 1996; accepted November 5, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Hirsh J. Heparin. N Engl J Med. 1991;324:1565-1574.[Medline] [Order article via Infotrieve]
2. Hull RD, Raskob GE, Hirsh J, Jay RM, LeClerc JR, Geerts WH, Rosenbloom D, Sackett DL, Anderson C, Harrison L, Gent M. Continuous intravenous heparin compared with intermittent subcutaneous heparin in the initial treatment of proximal-vein thrombosis. N Engl J Med. 1972;287:324-327.
3.
Hirsh J, van Aken WG, Gallus AS, Dollery CT, Cade JF,
Yung WJ. Heparin kinetics in venous thrombosis and
pulmonary embolism. Circulation. 1976;53:691-695.
4.
Cruickshank MK, Levine MN, Hirsh J, Roberts R,
Siguenza M. A standard heparin nomogram for the management of
heparin therapy. Arch Intern Med. 1991;151:333-337.
5. Choay J, Petitou M, Lormeau JC, Sinay P, Casu B, Gatti G. Structure-activity relationship in heparin: a synthetic pentasaccharide with high affinity for antithrombin III and eliciting high anti–factor Xa activity. Biochem Biophys Res Commun. 1983;116:492-499.[Medline] [Order article via Infotrieve]
6.
Lindahl U, Backström G, Höök M,
Thunberg L, Fransson LA, Linder A. Structure of the
antithrombin-binding site of heparin. Proc Natl Acad Sci
U S A. 1979;76:3198-3202.
7. Höök M, Björk I, Hopwood J, Lindahl U. Anticoagulant activity of heparin: separation of high-activity and low-activity species by affinity chromatography on immobilized antithrombin. FEBS Lett. 1976;66:90-93.[Medline] [Order article via Infotrieve]
8. Lam LH, Silbert JE, Rosenberg RD. The separation of active and inactive forms of heparin. Biochem Biophys Res Commun. 1976;69:570-577.[Medline] [Order article via Infotrieve]
9. Rosenberg RD. The heparin-antithrombin system: a natural anticoagulant mechanism. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and Thrombosis: Basic Principles and Clinical Practice. Philadelphia, Pa: Lippincott; 1987:1373-1392.
10. Lane DA. Heparin binding and neutralizing proteins. In: Lane DA, Lindahl U, eds. Heparin: Chemical and Biological Properties, Clinical Applications. Boca Raton, Fla: CRC Press; 1989:363-391.
11. McKay EJ, Laurell CB. The interaction of heparin with plasma proteins: demonstration of different binding sites for antithrombin III complexes and antithrombin III. J Lab Clin Med. 1980;95:69-80.[Medline] [Order article via Infotrieve]
12. Zammit A, Pepper DS, Dawes J. Interaction of immobilised unfractionated and LMW heparins with proteins in whole human plasma. Thromb Haemost. 1993;70:951-958.[Medline] [Order article via Infotrieve]
13. Young E, Hirsh J. Contribution of red blood cells to the saturable mechanism of heparin clearance. Thromb Haemost. 1990;64:559-563.[Medline] [Order article via Infotrieve]
14. Young E, Cosmi B, Weitz J, Hirsh J. Comparison of the nonspecific binding of unfractionated heparin and low molecular weight heparin (enoxaparin) to plasma proteins. Thromb Haemost. 1993;70:625-630.[Medline] [Order article via Infotrieve]
15. Young E, Prins M, Levine MN, Hirsh J. Heparin binding to plasma proteins, an important mechanism for heparin resistance. Thromb Haemost. 1992;67:639-643.[Medline] [Order article via Infotrieve]
16. Young E, Wells P, Holloway S, Weitz J, Hirsh J. Ex-vivo and in-vitro evidence that low molecular weight heparins exhibit less binding to plasma proteins than unfractionated heparin. Thromb Haemost. 1994;71:300-304.[Medline] [Order article via Infotrieve]
17. Prandoni P, Lensing AWA, Buller HR, Carta M, Cogo A, Vigo M, Casara D, Ruol A, ten Cate JW. Comparison of subcutaneous low-molecular-weight heparin with intravenous standard heparin in proximal deep-vein thrombosis. Lancet. 1992;339:441-445.[Medline] [Order article via Infotrieve]
18. Hull RD, Raskob GE, Pineo GF, Green D, Trombridge AA, Elliott CG, Lerner RG, Hall J, Sparling T, Brettell HR, Norton J, Carter CJ, George R, Merli G, Ward J, Mayo W, Rosenbloom D, Brant R. Subcutaneous low-molecular weight heparin compared with continuous intravenous heparin in the treatment of proximal-vein thrombosis. N Engl J Med. 1992;326:975-982.[Abstract]
19.
Hirsh J, Levine MN. Low molecular weight
heparin. Blood. 1992;79:1-17.
20.
Mosesson MW, Amrani DL. The structure and
biological activities of plasma fibronectin. Blood. 1980;56:145-158.
21.
Preisnner KT, Muller-Berghaus G. Neutralization
and binding of heparin by S-protein/vitronectin in the
inhibition of factor Xa by antithrombin III. J Biol
Chem. 1987;262:12247-12253.
22.
Tollefsen DM, Majerus DW, Blank MK. Heparin
cofactor II: purification and properties of a heparin-dependent
inhibitor of thrombin in human plasma. J
Biol Chem. 1982;257:2162-2167.
23. Hagiwara T, Suzuki H, Kono I, Kashiwagi H, Akiyama Y, Onozaki K. Regulation of fibronectin synthesis by interleukin-1 and interleukin-6 in rat hepatocytes. Am J Pathol. 1990;136:39-47.[Abstract]
24. Seiffert D, Geisterfer M, Gauldie J, Young E, Podor TJ. IL-6 stimulates vitronectin gene expression in vivo. J Immunol. 1995;155:3180-3185.[Abstract]
25. Toulon P, Vitoux JF, Fiessinger JN, Sicard D, Aiach M. Heparin cofactor II: an acute phase reactant in patients with deep vein thrombosis. Blood Coagul Fibrinol. 1991;2:435-439.[Medline] [Order article via Infotrieve]
26. Podor TJ, Hirsh J, Gelehrter TD, Zeheb R, Torry D, Guigoz Y, Sierra F, Gauldie J. Type 1 plasminogen activator inhibitor is not an acute phase reactant in rats. J Immunol. 1993;150:225-235.[Abstract]
27. Teien AN, Lie M. Evaluation of an amidolytic heparin assay method: increased sensitivity by adding purified antithrombin III. Thromb Res. 1977;10:399-410.[Medline] [Order article via Infotrieve]
28. Young E. The monitoring of low molecular weight (LMW) heparins in the clinical chemistry laboratory. In: Abstracts of the Fourth Asian-Pacific Congress of Clinical Chemistry; 1988:160. Abstract.
29. Casu B, Diamantini G, Fedeli G, Montavani M, Oreste P, Pescador R, Porta R, Prino G, Torri G, Zoppetti G. Retention of antilipemic activity by periodate-oxidized nonanticoagulant heparins. Arzneimittelforschung. 1986;36:637-642.[Medline] [Order article via Infotrieve]
30. Schreiber G, Tsykin A, Aldred AR, Thomas T, Fung W-P, Dickson PW, Cole T, Birch H, De Jong FA, Millard J. The acute phase response in the rodent. Ann N Y Acad Sci. 1989;557:61-85.[Medline] [Order article via Infotrieve]
31. Castell JV, Andus T, Kunz D, Heinrich PC. Interleukin-6: the major regulator of acute-phase protein synthesis in man and rat. Ann N Y Acad Sci. 1989;557:87-101.[Medline] [Order article via Infotrieve]
32. Siefert PS, Haeffner-Cavaillon N, Appay MD, Kazatchkine MD. Bacterial lipopolysaccharides alter human endothelial cell morphology in vitro independent of cytokine secretion. J Lab Clin Med. 1991;118:563-569.[Medline] [Order article via Infotrieve]
33.
Mathison JC, Ulevitch RJ. The clearance, tissue
distribution and cellular localization of intravenously
injected lipopolysaccharide in rabbits. J
Immunol. 1979;123:2133-2143.
34. Mathison JC, Wolfson E, Ulevitch RJ. Participation of tumor necrosis factor in the mediation of gram negative bacterial lipopolysaccharide-induced injury in rabbits. J Clin Invest. 1988;81:1925-1937.
35. Rucinski B, Niewiaroski S, Strzyzewski M, Holt JC, Mayo KH. Human platelet factor 4 and its C-terminal peptides: heparin binding and clearance from the circulation. Thromb Haemost. 1990;63:493-498.[Medline] [Order article via Infotrieve]
36.
Maccarana M, Lindahl U. Mode of interaction
between platelet factor 4 and heparin.
Glycobiology. 1993;3:271-277.
37. Schorer AE, Moldow CF, Rick ME. Interleukin 1 or endotoxin increases the release of von Willebrand factor from human endothelial cells. Br J Haematol. 1987;67:193-197.[Medline] [Order article via Infotrieve]
38.
Wu H-F, Lundblad R, Church FC. Neutralization of
heparin activity by neutrophil lactoferrin. Blood. 1995;85:421-428.
This article has been cited by other articles:
 |
|
 |
|
  C.-E. Dempfle, J. Eichner, N. Suvajac, P. Ahmad-Nejad, M. Neumaier, and M. Borggrefe The Reduced Anticoagulant Effect of Fondaparinux at Low Antithrombin Levels Anesth. Analg., September 1, 2009; 109(3): 712 - 716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Breuhl, G. Moore, M. B. Brooks, and J. C. Scott-Moncrieff A Prospective Study of Unfractionated Heparin Therapy in Dogs With Primary Immune-Mediated Hemolytic Anemia J. Am. Anim. Hosp. Assoc., May 1, 2009; 45(3): 125 - 133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Niccoli and A P Banning Heparin dose during percutaneous coronary intervention: how low dare we go? Heart, October 1, 2002; 88(4): 331 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Podor, S. Campbell, P. Chindemi, D. M. Foulon, D. H. Farrell, P. D. Walton, J. I. Weitz, and C. B. Peterson Incorporation of Vitronectin into Fibrin Clots. EVIDENCE FOR A BINDING INTERACTION BETWEEN VITRONECTIN AND gamma A/gamma ' FIBRINOGEN J. Biol. Chem., February 22, 2002; 277(9): 7520 - 7528. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||