Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Most Northern Europeans are homozygous for the M variant of the serine protease inhibitor (or serpin) ₁-antitrypsin. Approximately 4%, however, carry the Z allele, which in the homozygote results in profound plasma deficiency (1), liver disease (2, 3), and early onset panlobular emphysema (4). The Z mutation results from the substitution of a lysine for a glutamic acid at position 342 in ₁-antitrypsin (5, 6). This mutation lies at the head of strand 5 of the main -sheet A of the ₁-antitrypsin molecule (Figure 1) and the base of the mobile reactive center loop (7, 8). The Z mutation increases access to -sheet A and this allows the spontaneous insertion of the reactive loop of a neighboring molecule of ₁-antitrypsin (9). The successive interaction between the reactive loop of one molecule and the -sheet A of a second molecule results in the formation of chains or polymers of ₁-antitrypsin (8). This loop sheet linkage underlies the plasma deficiency of Z ₁-antitrypsin because polymers tangle within the endoplasmic reticulum of the hepatocyte and hence fail to be secreted (9, 12, 13). It is these tangles that give rise to the characteristic PAS-positive, diastase-resistant inclusion bodies that are associated with cirrhosis and neonatal hepatitis (3, 14). The lack of circulating ₁-antitrypsin predisposes the Z homozygote to early onset panlobular emphysema (4).

View larger version (25K): [in this window] [in a new window]   Figure 1.   Illustration of the structure of the native, polymeric, and reactive loop-cleaved conformations of 1-antitrypsin. The reactive center loop is at the apex of native antitrypsin but is inserted into  sheet A following cleavage (left). The severe Z deficiency variant of antitrypsin is at the head of strand 5 of  sheet A and the base of the reactive center loop. The mutation causes the reactive loop of one molecule to link with  sheet A of another to form chains of polymers (right). It is these polymers that tangle within the liver to form the inclusion bodies that are associated with liver disease.

We have demonstrated previously that complexes between members of the serpin superfamily and their target protease bind to the low-density lipoprotein receptor- related protein (LRP) (15). It is worthy of note that the binding of such complexes to LRP can be blocked by ₁-antitrypsin polymers, which suggests that complexes and polymers share similar epitopes. Others have reported that serpin-protease complexes are chemotactic for human neutrophils in vitro (16) and are proinflammatory in cell culture (17). These findings prompted us to assess the direct effects of ₁-antitrypsin polymers on neutrophil function. The importance of this investigation is underscored by our detection in a previous study of polymers of ₁- antitrypsin in the bronchoalveolar lavage fluid from patients with Z ₁-antitrypsin deficiency-related emphysema (18). We show here that polymers of ₁-antitrypsin display striking proinflammatory effects on human neutrophils in vitro. These findings suggest an additional and novel mechanism for the pathogenesis of emphysema associated with Z ₁-antitrypsin deficiency.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Preparation and Characterization of Conformations of ₁-Antitrypsin

Native ₁-antitrypsin was purified from human plasma by ammonium sulfate fractionation followed by glutathione and anion exchange chromatography (19). Native ₁-antitrypsin migrated as a single band on nondenaturing and sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis (PAGE), and showed a normal unfolding transition on transverse urea gradient PAGE (20). It was 70-80% active when assessed by titration against bovine -chymotrypsin (19). The native protein was then used to prepare the other ₁-antitrypsin conformations (Figure 1). Reactive loop-cleaved ₁-antitrypsin was obtained by incubation with Staphylococcus aureus V8 (Sigma Chemical Co., Poole, UK) protease at 37°C for 2 h. The protease was then removed by anion exchange chromatography using a mono-Q column. Polymers of ₁-antitrypsin were formed by heating native protein at 1 mg/ml for 3 h at 60°C (20). The mixture was reheated as necessary to ensure that the polymers were free from contamination with monomeric protein. Plasma purified M ₁-antitrypsin was used to prepare the polymeric conformation of ₁-antitrypsin. This is predicted to have the same loop sheet linkage as Z ₁-antitrypsin polymers, thus the same tertiary structure and accordingly behave in the same manner.

Conformations used in all assays, except the under agarose migration, were assayed for lipopolysaccharide (LPS) using the LAL turbimetric assay (Assoc. of Cape Cod, Liverpool, UK) with the concentration of LPS being determined using Softmax Pro software. The heat-stable cleaved conformation of ₁-antitrypsin was heated (60°C, 4 h) and filtered through a 100-kD centricon membrane to ensure complete removal of LPS. The non-heat-stable monomeric ₁-antitrypsin was decontaminated using END-X beads (Assoc of Cape Cod). All samples used in the subsequent experiments contained less than 10 ng/ml of LPS. Protein solutions were snap-frozen in phosphate-buffered saline (PBS) and stored at 80°C until required.

Preparation of Human Neutrophils

Human neutrophils were purified from the peripheral blood of healthy donors as previously detailed (21). In brief, 50 ml of anticoagulated blood was centrifuged (300 × g, 20 min) and sedimented with Dextran T500 (final concentration 6% wt/vol), and the cells from the resultant leukocyte-rich plasma centrifuged through a 51-42% (vol/vol) discontinuous plasma/Percoll gradient. Leukocytes were harvested from the 51-42% plasma/Percoll interface. The cells were sequentially washed with platelet-poor plasma, Dulbecco's PBS (without added calcium or magnesium cations; Sigma Chemical Co.) and Dulbecco's PBS (with calcium and magnesium cations). This method typically gave cells which were > 95% pure (< 0.1% mononuclear cell contamination) and 98% viable (as assessed by trypan blue exclusion).

As purified neutrophils are not required for the under agarose migration assays, a mixed leukocyte suspension was used. Following gravity sedimentation of whole blood, the buffy coat was aspirated, cells were then washed twice in calcium and magnesium free buffer, and resuspended in Hanks' balanced salt solution supplemented with human albumin (0.1% wt/vol) and glucose (0.2% wt/vol). Cell suspensions were standardized by granulocyte count.

Neutrophil Chemotaxis and Chemokinesis Assays

Cell migration was assessed using a modified Boyden chamber (22) (Receptor Technologies, Adderbury, UK). A suspension of neutrophils (225 µl, 3 × 10⁶ cells/ml) in PBS (with divalent cations) was placed in the upper wells with exactly 30 µl of test substance or PBS in the lower wells. Additional control test solutions included ovalbumin (2 mg/ml) and LPS (8 ng/ml), and maximally effective concentrations of C5a (100 nM), IL-8 (100 ng/ml), and fMLP (10 nM). The upper and lower wells were separated by a nitrocellulose filter containing 5-µm-diameter pores at a density of 4,000/cm² (Neuroprobe, Gaithersburg, MD). The chamber was then incubated in a 5% CO₂ incubator at 37°C for 90 min. Cell migration was assessed by manually counting the number of cells that had passed into the lower well using a standard hemocytometer. For each experiment a minimum of triplicate repeats were undertaken for each test solution.

Neutrophil migration was also independently assessed by means of the under agarose method, using gravity sedimented neutrophils as detailed above (23). In brief, 0.1% (wt/vol) agarose was heated and supplemented with 10% vol/vol heat-inactivated fetal calf serum. The resulting solution was then poured into a petri dish and allowed to solidify. Two-millimeter wells equally spaced were cut into the agarose in a line of three. The test substance was placed in the outer well with vehicle control (buffer) in the inner and the cells in the middle well and the dish incubated for 90 min at 37°C in a 5% CO₂ incubator. Chemotaxis was quantified by counting the number of cells entering a low-power microscope field tangential to the well containing the cell suspension.

Checkerboard analysis was used to distinguish chemotaxis from chemokinesis and was performed by setting up positive and negative test solution gradients as detailed in the legend for Table 1.

Neutrophil Shape Change

The effect of ₁-antitrypsin conformers on neutrophil polarization was assessed both visually and by fluorescence-activated cell sorter (FACS) analysis as detailed previously (24). Additional test solutions included fMLP (100 nM) as a positive control and five times the maximal concentration of LPS detected in the conformers of ₁-antitrypsin. The incubations were stopped at predetermined time points by the addition of an equal volume of ice-cold glutaldehyde (final concentration 1.6% vol/vol). Visual scoring was performed by counting the number of spherical and nonspherical (polarized) cells using a standard hemocytometer. The experiments were all performed in triplicate with a minimum of 300 cells counted per condition. Assessment of neutrophil shape change by flow cytometry was performed using Becton FACSort (24) and the data analyzed using FCS press software (FCS Press, Cambridge, UK).

Neutrophil Myeloperoxidase Release

The ability of ₁-antitrypsin polymers to cause neutrophil degranulation was examined by assaying the release of myeloperoxidase (MPO). The maximal extent of agonist-stimulated degranulation was determined by preincubating neutrophils (12.5 × 10⁶ cells/ml) with the priming agent TNF- (200 U/ml) for 30 min before stimulation with 100 nM fMLP. The reactions were quenched at times up to 90 min by the addition of 100 µl ice-cold PBS and placing the samples on ice. MPO release was measured by determining the change in absorbance at 460 nm induced by the oxidization of 0-dianisidine in the presence of hydrogen peroxide (25). The amount of MPO released was expressed as a percentage of the total cellular MPO released following the addition of 4 µl of 10% vol/ vol Triton X-100. All reactions were performed in quadruplicate.

Neutrophil Respiratory Burst Activity

The effect of ₁-antitrypsin polymers on the production of superoxide anions was determined by measuring the SOD-inhibitable reduction of cytochrome C (26). In brief, polymers or buffer were added to 90 µl of cells (12.5 × 10⁶ cells/ml) to achieve a final polymer concentration of 0.1 mg/ml and the neutrophils incubated at 37°C in the presence of cytochrome C (1.2 mg/ml) for 10 to 90 min. Superoxide dismutase was included in one set of each quadruplicate samples. Because agonist-induced superoxide anion generation is only observed in primed neutrophils, an additional series of experiments were undertaken to determine whether ₁-antitrypsin conformers could induce superoxide anion generation in TNF--pretreated cells (200 U/ml for 30 min), and similarly whether the conformers of ₁-antitrypsin could modulate the fMLP stimulated response. The reactions were stopped by centrifugation (14,000 × g, 4°C) and the optical density of the supernatants determined using a scanning spectrophotometer measuring the peak height at 550 nm. All determinations were undertaken in triplicate. The total superoxide anion output was calculated in nmol/10⁶ cells using the extinction coefficient of 21 × 10³ M¹cm¹.

Neutrophil Adhesion Assays

A 96-well plate was coated with heat-inactivated fetal calf serum for a minimum of two hours at 37°C, and then washed three times to remove excess serum. Purified neutrophils (8 × 10⁶ cells/ml in PBS) were incubated for 30 min at room temperature with 1 µM Calcein AM (Molecular Probes Europe BV, Leiden, The Netherlands) and, following washing with PBS and buffer exchange into Na HEPES (50 mM), were added (90 µl, 3 × 10⁶ cells/ml) to the precoated wells containing the test substance (27). After incubation at 37°C for 30, 60, or 90 min, the cells were fixed for 10 min with an equal volume of ice-cold glutaldehyde (final concentration 1.6% vol/vol) and the nonadherent cells gently aspirated. The plates were read in a fluorescent plate reader using 490 nm as the excitation wavelength and 530 nm as the emission wavelength.

Assessment of Neutrophil Apoptosis In Vitro

Neutrophils were re-suspended at 5 × 10⁶ cells/ml in Iscove's Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (vol/vol) autologous serum and 50 U/ml of penicillin and streptomycin. Native and polymeric ₁-antitrypsin were diluted in DMEM and 15 µl of ₁-antitrypsin conformer (final concentration 0.001-0.1 mg/ml) was added to 135 µl of cell suspension in a 96-well plate (Costar ultra low cluster; Corning Ltd., High Wycombe, UK). Control cells were incubated with the appropriate buffer diluted in DMEM. The cell suspensions were incubated at 37°C in a 5% CO₂ incubator for 6 and 20 h and cytospins prepared by centrifuging at 300 × g for 3 min. The cytospins were air dried, fixed in methanol, and stained with Diff Quik. Apoptosis was quantified by determining the proportion of the cells displaying the highly distinctive apoptotic morphology. Cell viability was > 95% at all time points examined. A minimum of 300 cells were assessed for each incubation with triplicate determinations performed for each condition.

Statistical Analysis

All data represent the mean of n separate experiments. Differences between groups were assessed with the Wilcoxon sign rank test for nonparametric data and Student's t test using Graph-pad Prism software. A P value of < 0.05 was considered to be significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Polymeric ₁-Antitrypsin Is Chemotactic for Human Neutrophils

The native, reactive loop-cleaved, protease complexed, and polymeric conformations of ₁-antitrypsin all caused significantly increased migration of human neutrophils compared with matched buffer or ovalbumin controls (P < 0.05, Figures 2A and 2B). However, the migration of neutrophils to polymeric ₁-antitrypsin was significantly greater than that observed to the native protein and other ₁-antitrypsin conformers (P < 0.05) and was comparable in magnitude to the chemotactic response induced by the major chemotatic peptide C5a. There was no evidence of processing of the conformations on Western blot analysis of the lower well samples (data not shown). The chemotactic response of ₁-antitrypsin and its conformers was not mediated by any residual contaminating LPS because adding this agent alone did not induce neutrophil migration (Figure 2A) or modify the effect of the individual ₁-antitrypsin conformers (data not shown). The effect of ₁-antitrypsin polymers on neutrophil migration was concentration dependent (EC₅₀ 0.0045 ± 0.002 mg/ml) and apparent over a range of physiologically relevant concentrations (Figure 2C) (18). No desensitization of the migratory response could be demonstrated at concentrations up to 10 mg/ml. In all further experiments ₁-antitrypsin polymers were used at 0.1 mg/ml, which represented the concentration of polymers required to induce a just maximal (~ EC₉₀) chemotactic response.

View larger version (18K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Figure 2.   Chemotactic properties of conformers of 1-antitrypsin for human neutrophils. The chemotactic properties of isolated human neutrophils were assessed in a modified Boyden chamber (A). Cells were placed in the upper wells with the various test substances in the lower wells. The horizontal axis indicates the conditions present in the lower well and the vertical axis the number of cells that have migrated into the lower chamber. The negative controls used were ovalbumin (2 mg/ml), LPS (8 ng/ml), and PBS, with fMLP (10 nM), IL-8 (100 ng/ ml), and C5a (100 nM) as positive controls. After a 90-min incubation, the number of cells which had migrated into the lower wells was assessed by manual counting. All the conformers of 1-antitrypsin were examined at 2 mg/ml. The data represent the mean (± SEM) of six independent experiments (*P < 0.05 compared with control, §P < 0.05 polymeric 1-antitrypsin compared with native 1-antitrypsin). Neutrophil migration was also assessed by the under agarose method (B). Cells were placed in the middle well with the test substance and buffer in the opposing outer wells. All test proteins were assayed at a concentration of 1 mg/ml. After 90-min incubation, cells entering a low power microscope field tangential to the well containing the cell suspension were counted. The data represent the mean (± SEM) of 11 independent experiments (*P < 0.001 compared with control, §P < 0.001 polymers of 1- antitrypsin or complexes compared with native 1-antitrypsin). Concentration response of the chemotactic effect of polymeric 1-antitrypsin was assessed using the Boyden chamber (C). The data are the mean (± SEM) of three independent experiments (*P < 0.05 compared with buffer control).

Because it is well recognized that a number of agents can induce random nondirectional neutrophil movement (chemokinesis) rather than true receptor-mediated directional migration (chemotaxis), further checkerboard analysis was undertaken (29). These results show that although ₁-antitrypsin polymers induced a degree of chemokinesis (Table 1, bold text), a major chemotactic response was still observed. In contrast, native ₁-antitrypsin induced a small neutrophil chemokinetic signal but no chemotaxis on checkerboard analysis (data not shown), and this most likely accounts for the small increase in the number of cells migrating in response to native ₁-antitrypsin (Figure 2A). This finding was in keeping with previous publications (16, 28).

The ability of ₁-antitrypsin polymers (and, to a smaller degree, complexes) to cause neutrophil chemotaxis was confirmed by assessing the migration of cells under agarose (Figure 2B).

Polymeric ₁-Antitrypsin Induces Neutrophil Shape Change

Neutrophils undergo a characteristic shape-change (polarization) when challenged with chemotactic agents under nongradient conditions. This can be assessed both visually and by FACS analysis (Figure 3). As a positive control, neutrophils were challenged with fMLP, which resulted in a significant shape change response that was maximal at early time points (83.7 ± 6.2%, 10 min) and thereafter gradually declined (63.9 ± 4.3%, 90 min), suggesting a degree of spontaneous de-priming or deactivation (30). ₁-Antitrypsin polymers caused a significant but delayed polarization response that was first evident at 60 min, with 31 ± 7.7% of the neutrophils being shape changed by 90 min (Figure 3A). These experiments were all performed in the absence of serum, and under such conditions LPS at a concentration of 8 ng/ml had no effect on neutrophil shape in incubations up to 90 min (Figure 3A). To address more definitively the concern that the effects of polymeric ₁-antitrypsin were due to contamination with low concentrations of LPS, these experiments were repeated with LPS-free transgenic ₁-antitrypsin (Protein Pharmaceuticals Ltd, Edinburgh, UK). Polymers derived from this alternative source gave near identical results to ₁-antitrypsin polymers purified in the laboratory (Figure 3B). The ability of ₁-antitrypsin polymers to induce neutrophil shape change was also assessed by FACS analysis (Figure 3C) which confirmed the results obtained by visual analysis.

View larger version (16K): [in this window] [in a new window]   View larger version (26K): [in this window] [in a new window]   Figure 3.   Effect of polymeric 1-antitrypsin on human neutrophil shape change. Neutrophils were incubated with conformers of 1-antitrypsin for 0, 10, 30, 60, and 90 min at 37°C. The incubations were stopped by fixing the cells with an equal volume of ice-cold glutaldehyde (1.6% vol/vol) and the number of cells that had changed shape were scored by visual examination using a hemocytometer (A). All experiments were performed in triplicate (fMLP, 100 nM [circles]; LPS, 8 ng/ml [squares]; buffer control [triangles]; and polymers of 1-antitrypsin, 0.1 mg/ml [diamonds]). The data represent the mean (± SEM) of six experiments performed in triplicate. The extent of shape change for native 1-antitrypsin, reactive loop-cleaved 1-antitrypsin, and complexes at 90 min was 10.4 ± 1.9, 11 ± 1.7, and 17.7 ± 2.6%, respectively. Identical experiments were undertaken with neutrophils incubated with LPS-free commercial 1-antitrypsin (B). (C) purified human neutrophils were incubated with buffer (left upper panel) or fMLP (100 nM) for 10 min (right upper panel) and buffer (left lower panel) or 1-antitrypsin polymers for 90 min (right lower panel) and analyzed by FACS analysis. A single histogram (representative of six independent observations) is shown; the x-axis represents the mean forward scatter and the y-axis cell numbers.

MPO Release by ₁-Antitrypsin Polymers

fMLP caused neutrophils to release myeloperoxidase and, in agreement with previous studies, this response was considerably increased following pretreatment of cells with TNF- (Figure 4). Incubation of neutrophils with ₁-antitrypsin polymers (0.1 mg/ml) alone also increased myeloperoxidase secretion which, in contrast to the shape change response, appeared to be maximal after 10 min. It is worthy of note that pretreatment of cells with polymers did not enhance the subsequent fMLP response, and the polymer effect itself was not affected by pretreatment of cells with TNF- (Figure 4). In comparison, native ₁-antitrypsin had no primary secretagogue activity and indeed inhibited the ability of fMLP to induce the release of MPO (P < 0.05). In subsequent experiments cleaved ₁-antitrypsin also had no effect on MPO release, whereas complexed ₁-antitrypsin caused a small degranulation response. The magnitude of the effect was smaller than that for polymers (2.1-fold increase over control compared with 3.8-fold increase for polymers at 60 min). These data demonstrate the ability of polymeric ₁-antitrypsin to induce neutrophil degranulation and hence to act as a direct secretagogue agonist.

View larger version (29K): [in this window] [in a new window]   Figure 4.   Myeloperoxidase generation by polymeric 1-antitrypsin. Neutrophils (12.5 × 106 cells/ml) were incubated with native or polymeric 1-antitrypsin (final concentration 0.1 mg/ml) or TNF- (200 U/ml) before stimulation with buffer or fMLP (100 nM). The left panel demonstrates the ability TNF- to enhance fMLP-stimulated MPO release in purified human neutrophils. The ability of native and polymeric 1-antitrypsin (0.1 mg/ml) to stimulate MPO release was assessed under control and TNF--primed conditions (middle panel). Finally, the effect of 1-antitrypsin conformers to stimulate MPO release after 60 min was examined, together with the impact of conformer preincubation on subsequent fMLP release (right panel). The total releasable myeloperoxidase was assessed following cell lysis with 10% (vol/vol) Triton X-100. The data represent the mean (±SEM) of seven experiments. (*P < 0.05 compared with control, §P < 0.05 native + fMLP versus fMLP alone.)

Effect of ₁-Antitrypsin Polymers on Neutrophil Adhesion

Native ₁-antitrypsin had no effect on the adhesion of Calcein AM-labeled neutrophils to FCS coated plates (Figure 5). In contrast, polymeric ₁-antitrypsin induced a small but significant concentration-dependent increase in neutrophil adhesion (P < 0.05 at 1 mg/ml). Polymers had no additional effect on the degree of neutrophil adhesion observed with 100 nM fMLP (data not shown).

View larger version (17K): [in this window] [in a new window]   Figure 5.   Effect of native and polymeric 1-antitrypsin on neutrophil adhesion. Calcein AM-labeled neutrophils were added to 96-well plates precoated with heat-inactivated fetal calf serum. After the addition of native or polymeric 1-antitrypsin, cellular adhesion was assessed by fluorescence with an excitation wavelength of 480 nm and an emission wavelength of 530 nm. The results are the mean (± SEM) of six independent experiments each performed in triplicate. (*P < 0.05 compared with control.)

Effect of ₁-Antitrypsin Polymers on Neutrophil Superoxide Anion Generation and the Rate of Constitutive Apoptosis

In contrast to the above effects of polymeric ₁-antitrypsin on neutrophil chemotaxis, degranulation, and adhesion, polymers (0.1 mg/ml) had no direct effect on superoxide anion production and, unlike TNF-, were unable to modulate the subsequent response to fMLP (Figure 6A).

View larger version (27K): [in this window] [in a new window]   Figure 6.   The effects of 1-antitrypsin conformers on neutrophil superoxide anion generation and constitutive apoptosis. Neutrophils (12.5 × 106 cells/ml) were incubated with polymeric 1-antitrypsin (final concentration 0.1 mg/ml) or were primed with TNF- (200 U/ml) for 30 min. Superoxide anion release was then assessed following the addition of buffer or 100 nM fMLP at 10, 60, and 90 min (A). The data represent the mean (± SEM) of four experiments. To assess the effects of conformers of 1-antitrypsin on constitutive apoptosis (B) neutrophils were incubated in DMEM containing 10% (wt/vol) autologous serum in the presence of conformers of 1-antitrypsin for 6 to 20 h. Aliquots of cells were removed and cytospins prepared for each concentration of native and polymeric 1-antitrypsin. A minimum of 300 cells were counted per condition and the data expressed as a percentage of apoptotic cells. The data are the mean (± SEM) of three independent experiments.

Neither native nor polymeric ₁-antitrypsin had any effect on the rate of neutrophil apoptosis when incubated over a range of concentrations (1-100 µg/ml) for 6 or 20 h (Figure 6B).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The lungs are under continuous assault from inhaled materials that can trigger the inflammatory cascade. At the vanguard of the defense against intrinsic or extraneous proteolytic attack is ₁-antitrypsin (31). The importance of this protease inhibitor is underscored by the florid and rapid onset emphysema seen in Z homozygotes who have greatly reduced circulating levels of ₁-antitrypsin compared with the normal M homozygote (32, 33). We have previously shown that this plasma deficiency results from a spontaneous conformational transition of Z antitrypsin and the formation of Z ₁-antitrypsin polymers within hepatocytes (9, 12, 34). We have also shown that these polymers can form spontaneously within the lungs of patients with Z ₁-antitrypsin-deficiency related emphysema. In this study two out of five Z homozygotes were shown to have polymeric ₁-antitrypsin as the predominant conformation. In the remaining three subjects a mixture of complexes and reactive loop-cleaved ₁-antitrypsin was demonstrated (18). The formation of polymers within the lungs is predicted to cause additional loss of ₁-antitrypsin and further exacerbate the local deficiency of protease inhibitor. We now present the first evidence that this conformational transition also converts ₁-antitrypsin into a potent proinflammatory agent for human neutrophils. It is noteworthy that individuals with Z ₁-antitrypsin deficiency have increased release of LTB₄ secreted from alveolar macrophages and high levels of LTB4 and IL-8 in induced sputum compared with control subjects (35, 36). It is therefore likely that several factors in addition to polymers of ₁-antitrypsin contribute to the increased inflammation seen in these individuals.

Polymeric ₁-antitrypsin stimulated an increase in the number of neutrophils that successfully migrated into the lower well of the Boyden chamber. This finding was independently confirmed using an under agarose method to assess neutrophil migration. The magnitude of this effect was comparable to that induced by the major chemotactic peptide C5a, and checkerboard analysis demonstrated that the majority of this effect was the result of chemotaxis rather than chemokinesis. The chemotactic effect of polymers was apparent over a range of concentrations, including those reflecting the likely concentration of polymeric ₁-antitrypsin present in the alveolar lining fluid in vivo. Indeed, the calculated EC₅₀ for this response was 0.0045 mg/ml, which compares to circulating levels of ₁-antitrypsin in a Z homozygote of ~ 0.3 mg/ml. In keeping with the chemotactic response, polymeric ₁-antitrypsin also induced neutrophil shape change and stimulated adherence of cells to fibrinogen-coated wells. The relatively long time required to induce the maximal shape change response may represent the level of receptor expression or indeed the efficacy of this response, but is well within the time period over which other chemotactic agents operate. The finding that ₁-antitrypsin polymers can stimulate both chemotaxis and adhesion is important as they are both represent essential components of neutrophil transmigration across endothelial surfaces both in vitro and in vivo.

Polymeric ₁-antitrypsin induced neutrophil degranulation over a much shorter time period (10 min) compared with that required for shape change implying that polymers have direct secretagogue activity. Importantly, ₁-antitrypsin polymers had no effect on either the rate of constitutive neutrophil apoptosis or basal or fMLP-stimulated respiratory burst activity. Although the striking effect of polymers on neutrophil chemotaxis in the absence of any effect on superoxide anion generation is unusual when compared with classic chemotactic agents, such a dichotomy is well recognized with agents such as TGF-₁, substance P (37), and soluble FAS-L (38). Likewise, a number of major secretogogue and priming agents (including fMLP and PAF) have no effect on the rate of neutrophil apoptosis in vivo. These data imply that ₁-antitrypsin polymers activate signaling pathways independent of those required to stimulate the generation of superoxide anions or affect cell survival. Taken together, the proinflammatory effects of polymeric ₁-antitrypsin allow us to advance a new paradigm to explain the early onset of emphysema in patients with Z ₁-antitrypsin deficiency (Figure 7). There is little doubt that the lungs of these patients are vulnerable to proteolytic attack due to the lack of circulating protease inhibitor. This is underscored by the rare individuals who are null ₁-antitrypsin homozygotes; these patients have no circulating plasma ₁-antitrypsin and are at significant risk of developing emphysema (39, 40). It is also now apparent that the ₁-antitrypsin that diffuses into the lungs of individuals with Z ₁-antitrypsin deficiency, or indeed that which is produced locally by bronchial epithelial cells (41) and macrophages (42), can also spontaneously polymerize in vivo (18). This polymerization inactivates ₁-antitrypsin and further reduces the protection of the alveoli. Our current work suggests that this process also converts ₁-antitrypsin from a protease inhibitor into a chemoattractant and secretagogue for human neutrophils. Support for this hypothesis comes from studies which have shown excessive number of neutrophils in the bronchoalveolar lavage fluid from patients with ₁-antitrypsin deficiency compared with matched control subjects with a normal ₁-antitrypsin phenotype (43).

View larger version (15K): [in this window] [in a new window]   Figure 7.   Proposed model for the pathogenesis of emphysema in patients with PiZ 1-antitrypsin deficiency. The plasma deficiency of 1-antitrypsin is exacerbated by polymerisation within the lungs. This inactivates the inhibitor, thereby further reducing the antiprotease screen. Moreover, this aberrant conformational acts as a proinflammatory stimulus that attracts and activates neutrophils, thereby increasing tissue damage.

Cigarette smoking is known to significantly increase the risk of emphysema in patients with ₁-antitrypsin deficiency (4, 33, 44). Smokers have more neutrophils in lung lavage fluid compared with nonsmokers and this is predicted to further enhance tissue damage at sites where there is a lack of protective ₁-antitrypsin (45). Furthermore, polymer formation is rapidly accelerated at the low pH values that may be induced when individuals smoke cigarettes (10). It is therefore plausible that smoking provides an additional stimulus for Z ₁-antitrypsin to polymerize within the lung and that this in turn exacerbates the influx of neutrophils and thereby increases tissue damage and emphysema.

Finally, the chemotactic effect of ₁-antitrypsin polymers shown here may explain in part some of the other inflammatory conditions that are associated with Z ₁-antitrypsin deficiency. In addition to emphysema and cirrhosis, ₁-antitrypsin deficiency is also associated with bronchiectasis (46), vasculitis (47), panniculitis (48), and asthma (49). Much of the tissue damage observed in these conditions is neutrophil derived, and it is possible that ₁-antitrypsin polymers are one of the factors that drive inflammation and disease progression.

In summary, we have demonstrated that polymers of ₁-antitrypsin are proinflammatory for human neutrophils. A detailed understanding of the factors that favor polymerization has enabled us to propose a new paradigm for the pathogenesis of emphysema in patients with Z ₁-antitrypsin deficiency.