Published online before print July 6, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Surgery, Division of Thoracic Surgery, Akita University School of Medicine, Akita City, Japan; and
ZLB Behring, Marburg, Germany
1Correspondence: Department of Surgery, Division of Thoracic Surgery, Akita University School of Medicine, 1-1-1 Hondo, Akita City 010-8543, Japan. E-mail: hsaito{at}doc.med.akita-u.ac.jp
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, ß, or low heparin-affinity isoforms of AT (1 IU/ml, 20 min), and deformability was evaluated using a filter assay system. Neutrophils were also stained with fluorescein isothiocyanate-phalloidin and subjected to a fluorescein-activated cell sorter scan to assess F-actin content. The results showed that pretreatment with any of the three AT isoforms similarly inhibited the decreased neutrophil deformability and increased F-actin content of stimulated cells. Notably, heparinase had no effect on deformability or F-actin content in the presence or absence of AT, which was somewhat unexpected, as heparin sulfate proteoglycans likely function as AT receptors. These findings suggested that AT inhibits the increase in neutrophil stiffness seen during inflammatory processes by inhibiting actin polymerization via a heparin-independent pathway.
Key Words: F-actin • fMLP • stiffness
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antithrombin III (AT) is a single-chain heparin-binding glycoprotein in plasma that acts as the major inhibitor of thrombin and interferes with several plasma proteases, including kallikrein and factors IXa, Xa, Xia, and XIIa, thereby playing a central role in regulating coagulation [8 ]. Apart from its role in hemostasis, it is also used as a therapeutic agent in patients with sepsis and disseminated intravascular coagulation (DIC) [9 ]. In addition, a number of recent studies have shown that AT also has anti-inflammatory properties, which are independent of its effects on coagulation. For instance, it is known that the binding of AT to heparin-like glycosaminoglycans on endothelial cells induces production of prostacyclin and mediates an anti-inflammatory effect [10 ]. However, one recent report showed that in vitro, AT may also exert a prostacyclin-independent, anti-inflammatory effect by binding directly to syndecan-4 on the surface of human neutrophils and thus, inhibiting their migration [11 ].
The and ß isoforms of AT avidly bind heparin, although the ß isoform, which lacks glycosylation at Asn135, binds heparins with greater avidity and glycosaminoglycans, with higher affinity than the fully glycosylated isoform. Both isoforms contribute to the anti-inflammatory effect by inhibiting nuclear factor-
B activation in human monocytes and endothelial cells. By contrast, a third isoform, low heparin-affinity AT (also known as latent AT) exerts no such effect [12
], which suggests that the interaction with its heparin sulfate proteoglycan (HSPG) binding site is important for the AT-induced anti-inflammatory effect. Conversely, LHA-AT was recently shown to exert an antiangiogenic effect by disrupting endothelial cell actin reorganization [13
].
In the present in vitro study, we evaluated the effect of AT on the deformability of human neutrophils and correlated deformability with the F-actin content of the cells. We found that the , ß, and latent isoforms efficiently inhibit the increased neutrophil deformability and decreased actin polymerization seen in stimulated cells and in a heparin-independent manner.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, ß, and latent isoforms) was obtained from ZLB Behring (Margurg, Germany) [8
]. To isolate the respective isoforms, 20,000 units AT (Kybernin P) were dissolved in 2.0 L water and pumped onto a heparin column (500 ml) that had been equilibrated with 15 mM Na2H(PO4)-2-hydrate, 60 mM NaH2(PO4)-2-hydrate, and 50 mM NaCl (pH 7.2). LHA-AT was eluted by washing the column with equilibration buffer containing 0.2 M NaCl, while the isoform was eluted with buffer containing 0.8 M NaCl, and the ß isoform was eluted with 2 M NaCl. All preparations were then dialyzed against 10 mM Na-citrate and 150 mM NaCl (pH 7.4). The purity of each fraction, according to size-exclusion chromatography, was for AT, 99.7%; for ß AT, 96.1%; and for latent AT, 94.6%. The AT heparin cofactor activity, which was tested by Berichrom AT assay (Dade Behring GmbH, Marburg, Germany), was for AT, 7.8 IU/mg; for ß AT, 6.77 IU/mg; and for latent AT, 0.18 IU/mg. Once isolated, all three isoforms were used at a concentration of 1 IU/ml; by definition, 1 IU AT equals the average amount found in 1 ml plasma and is equivalent to 150 µg/ml or 2.6 µM.
Neutrophil isolation
Human peripheral blood neutrophils were collected in the forearm venous blood of healthy volunteers and anticoagulated using 1.6 mg EDTA/ml blood. The neutrophils were then isolated using Histopaque density gradients (Sigma Chemical Co., St. Louis, MO) according to the manufacturer’s protocols, after which the isolated neutrophils were washed, resuspended in polymorphonuclear neutrophil (PMN) buffer [phosphate-buffered saline (PBS) containing 5 mM glucose, pH 7.4] at a concentration of 5 x 106 PMN/ml, and then kept at room temperature until used for experimentation. The purity of the isolated neutrophils was >95%.
Filter assay
The effect of AT on neutrophil deformability was examined using the microfilter technique [1
, 4
, 5
, 14
15
16
17
18
], in which the pressure needed to pass a bolus of neutrophils through a polycarbonate filter with a uniform pore diameter of 5 µm (Nucleopore, Pleasanton, Canada) was measured. Neutrophil deformability was evaluated after treating the cells for 2 min with 1 µM fMLP (Sigma Chemical Co.) [6
, 7
] or vehicle, and in some cases, the cells were also pretreated with AT (1 IU/ml) for 20 min at room temperature before being assayed. The cells (200 µl 5x106 PMN/ml) were then injected into the filter chamber through a port located upstream of the polycarbonate filter, immediately after which phosphate buffer was infused continuously at a constant rate of 2 ml/min using an infusion pump (KD Science, Holliston, MA). Also upstream of the filter chamber, pressure (mmHg) was measured continuously using a pressure transducer, which was connected to a strip-chart recorder (AD Instruments, Mountain View, CA; Fig. 1
). The pressure measured under conditions of no flow was taken as baseline. In some experiments, cells were pretreated with 0.5% bovine serum albumin (BSA) as nonfunctional protein against AT [11
].
 View larger version (21K): [in a new window] |
Figure 1. Neutrophil deformability was examined using the microfilter technique, which measured the pressure (mmHg) needed to pass a bolus of neutrophils through a polycarbonate filter with a uniform pore diameter of 5 µm. Injection of isolated human neutrophils (200 µl with 5x106 PMN/ml) through a port located upstream of the filter was immediately followed by continuous infusion of phosphate buffer at a constant flow rate of 2 ml/min using an infusion pump. Pressure was monitored continuously using a pressure transducer, which was connected to a strip-chart recorder.
|
To determine the distribution of F-actin within neutrophils, the cells were similarly stained with rhodamine-conjugated phalloidin (0.33 µM) for 60 min in the dark, after which randomly selected images of 100 neutrophils from each group were examined using a confocal laser-scanning microscopy system (LSM 410, Zeiss, Germany) incorporating an Axiovert 135 fluorescence microscope (Zeiss). After staining, 200 randomly selected neutrophils were observed, and neutrophil cytoskeletal rearrangement was expressed quantitatively as the percentage of neutrophils containing a pseudopod appearance.
Pretreatment with heparinase
To determine the effect of disrupting the binding of AT to HSPG, isolated neutrophils were preincubated with heparinase (Sigma Chemical Co.; 50 mU/ml) for 50 min at room temperature before treatment with AT [11
]. Neutrophil surface heparin sulfate proteoglycan pentasaccharide was cleaved by the treatment with heparinase, which is specific for N-sulfated disaccharides that contain IdoA,2S, and the susceptible linkages (GlcNSO3[a1-4]IdoA,2S) are located in the sulfated domains [19
]. After washing three times with PBS, neutrophil deformability and F-actin content and distribution were assessed as described above.
AT binding
To confirm that AT actually binds to the neutrophils and that pretreatment with heparinase inhibits that binding, AT (, ß, and LHA) was conjugated with fluorescein isothiocyanate (FITC) using an Alexa Fluor 488 protein labeling kit (Molecular Probes), after which the degree of conjugation was calculated according to the manufacturer’s protocols. Each group of neutrophils was then pretreated with the FITC-conjugated AT, and binding was measured by FACScan.
Statistics
Values are expressed as means ± SD. The significance of differences between groups was assessed by one-way ANOVA with Scheffe’s multiple comparison tests. Values of P< 0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß isoforms, although the latter is known to have a higher affinity for cell-surface proteoglycans and latent-AT, which is known to have a low affinity for heparin and displayed the same effect as the and ß isoforms. Furthermore, the peak pressures did not differ significantly in the presence or absence of heparinase, suggesting the affinity of AT for HSPG does not contribute to the observed reduction in neutrophil deformability.
 View larger version (16K): [in a new window] |
Figure 2. The effect of AT on neutrophil deformability. Isolated human neutrophils were pretreated for 20 min with AT [, ß, or latent (L)] at a final concentration of 1 IU/ml at room temperature. BSA (0.5%) was used as nonfunctional protein against AT. Neutrophil deformability was evaluated after stimulating the cells for 2 min with 1 µM fMLP or control vehicle using the microfilter technique, as described in the legend to Figure 1
. In some experiments, the cells were also pretreated for 50 min at room temperature with heparinase (50 mU/ml) prior to the addition of AT. Bars depict means ± SD; *, P < 0.05, versus vehicle pretreatment with no exposure to fMLP; +, P < 0.05, versus vehicle-pretreated neutrophils with exposure to fMLP; n = 5 in each group.
|
, ß, or latent) attenuated the effect significantly (Fig. 3
). Moreover, the F-actin content of the neutrophils correlated significantly with stiffness (Fig. 4
). Pretreating the cells with heparinase had no effect on F-actin levels in the presence or absence of AT, further confirming that to the extent that it is mediated by an increase in neutrophil deformability, the anti-inflammatory effect of AT is not associated with its ability to bind HSPG.
 View larger version (16K): [in a new window] |
Figure 3. Effect of AT [, ß, or latent (L)] on the F-actin content of neutrophils was determined by FACScan using FITC-phalloidin. F-actin content is reflected by the signal intensity expressed in arbitrary units. Bars depict means ± SD; *, P < 0.05, versus vehicle pretreatment with no exposure to fMLP; +, P < 0.05, versus vehicle-pretreated neutrophils with exposure to fMLP; n = 5 in each group.
|
 View larger version (11K): [in a new window] |
Figure 4. F-actin content of neutrophils was significantly correlated with cell stiffness (r=0.758; P<0.01).
|
 View larger version (31K): [in a new window] |
Figure 5. Effect of AT (, ß, or latent) on the intracellular distribution of F-actin. Shown are representative confocal images of rhodamine-phalloidin fluorescence, which is indicative of the F-actin distribution. (a and f) Images obtained in the absence of AT; (b and g) pretreated with 0.5% BSA. (c and h, d and i, and e and j) Cells pretreated with the , ß, and latent isoforms of AT (1 IU/ml for 20 min at room temperature), respectively. The cells shown in a–e were treated with vehicle, and those in f–j were stimulated with fMLP (1 µM, 2 min).
|
, ß, or latent) appeared to have little or no effect on the distribution of F-actin within untreated neutrophils or those treated with a control vehicle (Fig. 5c
5d
5e)
. On stimulation with fMLP, however, pretreatment with AT attenuated the redistribution of F-actin otherwise evoked. Only small changes in cell shape were seen, consisting primarily of small irregularities in the plasma membrane and occasional pseudopods (Fig. 5h
5i
5j)
. Again, heparinase had no effect on the distribution of F-actin in stimulated or unstimulated cells in the presence or absence of AT (Fig. 6a
6b
6c
6d
6e
6f
6g
6h
6i
6j
).
 View larger version (34K): [in a new window] |
Figure 6. Effect of heparinase on the distribution of F-actin in neutrophils in the presence and absence of AT. Isolated neutrophils were pretreated by heparinase (50 mU/ml) for 50 min at room temperature before addition of AT. (a and f ) Images of rhodamine-phalloidin fluorescence from cells in the absence of AT; (b and g) pretreated with 0.5% BSA. (c and h, d and i, and e and j) Cells pretreated with the , ß, and latent isoforms of AT (1 IU/ml for 20 min at room temperature), respectively. The cells shown in a–e were treated with vehicle, and those in f–jwere stimulated with fMLP (1 µM, 2 min). Note that heparinase had no effect on the distribution of F-actin within neutrophils in the presence or absence of AT.
|
|
View this table: [in a new window] |
Table 1. Stimulation with fMLP Rapidly Increased Neutrophils with Pseudopod Appearance
|
, ß, or latent) was conjugated with FITC, after which the strength of the cell labeling with FITC-AT was measured (Fig. 7
). For each of the isoforms tested, the intensity of the fluorescent signal from the neutrophils was significantly greater than what was obtained with unconjugated AT, indicating that the signal was dependent on the binding of FITC-AT. Pretreatment with heparinase had no effect on FITC-AT binding, suggesting that in this case, AT is binding to a receptor other than HSPG.
 View larger version (14K): [in a new window] |
Figure 7. Binding of AT to neutrophils was analyzed by FACScan using FITC-AT as a probe. Binding is reflected by signal intensity expressed in arbitrary units. Bars depict mean ± SD; *, P < 0.05, versus unconjugated AT pretreatment for each group of isoforms; n = 5 in each group.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß isoforms of AT, as well as latent AT, similarly inhibit neutrophil stiffening elicited by fMLP, perhaps by preventing actin polymerization. Moreover heparinase, which degrades HSPG, had no effect on deformability or F-actin content in the presence or absence of any of the AT isoforms. Thus, AT apparently exerts an anti-inflammatory effect by increasing neutrophil deformability in a manner that is independent of its HSPG binding properties. During an acute inflammatory response in the lungs, such as that which occurs during bacterial pneumonia or adult respiratory distress syndrome, large numbers of neutrophils are first sequestered in the microvasculature and then emigrate into the lungs. Earlier studies demonstrated that the initial sequestration of circulating neutrophils within the pulmonary capillary bed in response to inflammatory mediators does not require any of the commonly used adhesion molecules [24 25 26 ]. It is therefore generally accepted that changes in the mechanical properties of the cells are important in the process of sequestration [1 , 3 , 5 , 7 , 14 , 25 , 26 ]. In normal lungs, most neutrophils must become elongated to pass through the capillary bed into the venules, as many capillaries are narrower than spherical neutrophils. Inflammatory stimuli induce neutrophils to become stiffer and less able to change their shape [1 , 3 , 5 , 7 , 14 ], thereby trapping them within the pulmonary capillary beds. In that regard, it was reported recently that in a rat endotoxin model, infusion of AT reduced sequestration of neutrophils within the lungs and their F-actin content [27 ]. Our findings in human neutrophils are consistent with those earlier findings in rat, which suggests that one of the anti-inflammatory effects of AT is the inhibition of actin polymerization.
AT is a typical serpin with 432 residues, making up nine helices and three ß-sheets [28
]. The reactive (thrombin-binding) site is located at Arg393-Ser394 in the C-terminal loop structure, whereas the HSPG binding site is located in the N-terminal region [29
]. Many of the anti-inflammatory actions of AT appear to result from its binding to HSPG on cells and the subsequent down-modulation of various cellular reactions [10
, 11
, 30
, 31
]. Although the and ß isoforms are known to contribute to the anti-inflammatory effects of AT, latent AT, such as the Trp49-modified molecule, reportedly has no anti-inflammatory activity, reflecting the importance of heparin binding for those effects [12
, 27
]. Conversely, latent AT was shown recently to exert an antiangiogenic effect by disrupting actin reorganization in endothelial cells [13
, 32
]. Our findings show that in an analogous manner, latent AT also inhibits actin polymerization in neutrophils, thereby increasing cell deformability, and that this effect is heparin-independent.
Further confirming the heparin-independence of the effect of AT on neutrophil deformability was the finding that the ß isoform was as effective at reducing F-actin content and increasing deformability as the isoform. As the affinity of the ß isoform, which lacks glycosylation at Asn135, is known to bind heparin with greater avidity and glycosaminoglycans with higher affinity than the fully glycosylated isoform, this finding is consistent with the idea that the observed effect of AT on actin polymerization in neutrophils is independent of its heparin-binding properties.
Our observation that latent AT inhibited fMLP-induced F-actin formation to the same degree as the and ß isoforms differs from our earlier findings [27
], as well as from earlier reports by others showing that latent AT has no anti-inflammatory effects in vivo [10
, 20
, 30
]. We suggest that this inconsistency might be caused by differences in the way neutrophils and endothelial cells interact in living animal models; for example, latent AT does not promote release of prostacyclin from endothelial cells during inflammatory responses [20
, 30
]. Uchiba et al. [20] reported that the plasma concentration of 6-keto-prostaglandin F1a, a metabolite of prostacyclin, in rats receiving lipopolysaccharide (LPS) plus Trp49-modified AT was significantly lower than those of animals administered LPS plus native AT [20
]. Nevertheless, Trp49-modified AT showed a tendency to reduce the inflammation, although the effect did not reach statistical significance, suggesting some of the anti-inflammatory effects exerted by AT on the neutrophils are prostacyclin-dependent, and others are not [11
]. Indeed, our in vitro findings suggest that all AT isoforms exert an anti-inflammatory effect by reducing the F-actin content of neutrophils. In vivo, however, this direct effect might be somewhat diminished in importance under conditions where the interaction of neutrophils with endothelial cells makes the critical contribution to the anti-inflammatory event.
The available evidence indicates that syndecan-4 is required for the formation of focal adhesions and stress fibers, which is regulated by Rho family GTPases [33 , 34 ]. Syndecans are proteoglycans consisting of a core protein to which unbranched carbohydrate polymers (glycosaminoglycans) are covalently attached [35 ]. As syndecan-4 is a transmembrane proteoglycan that bears an extracellular HSPG, it is one of the important cell receptors of AT [36 37 38 39 ], and AT binding to syndecan-4 on the surface of human neutrophils inhibits their migration in vitro [11 ]. In the present study, all three AT isoforms tested, including latent AT, reduced actin polymerization in fMLP-stimulated neutrophils in a manner that was HSPG- and presumably, syndecan-4-independent, suggesting the presence of an alternative AT signaling pathway in neutrophils. We speculate the decreased deformability of neutrophils via actin polymerization (initial sequestration) might be in a heparin-independent manner; however, the next step of inflammation (transmigration) is in a heparin-dependent manner during the inflammatory process [11 ]. It seems plausible, therefore, that neutrophils express an as-yet unknown receptor for AT, via which it regulates actin filament organization. However, this question awaits further study.
Received October 15, 2004; revised December 20, 2004; accepted May 24, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B activation in human monocytes and vascular endothelial cells Blood 99,4015-4020 1-antitrypsin for structure and function of serpins Biochemistry 28,8951-8966[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
