help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Berry, L. R.
Right arrow Articles by Chan, A. K. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Berry, L. R.
Right arrow Articles by Chan, A. K. C.
American Journal of Respiratory Cell and Molecular Biology. Vol. 28, pp. 150-158, 2003
© 2003 American Thoracic Society
DOI: 10.1165/rcmb.4849

Effect of Covalent Serpin–Heparinoid Complexes on Plasma Thrombin Generation on Fetal Distal Lung Epithelium

Leslie R. Berry, Petr Klement, Maureen Andrew* and Anthony K. C. Chan

Henderson Research Centre, Hamilton; Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada; University of Veterinary and Pharmaceutical Sciences, Brno, Czech Republic; and Hospital for Sick Children, Toronto, Ontario, Canada

Address correspondence to: Dr. Anthony K.C. Chan, Henderson Research Centre, 711 Concession St., Hamilton, ON, L8V 1C3 Canada. E-mail: achan{at}thrombosis.hhscr.org


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Extravascular coagulation within the lung airspace is a hallmark of respiratory distress syndrome (RDS) in premature infants. We previously showed that covalent antithrombin–heparin complex (ATH) is superior to noncovalent antithrombin (AT) + heparin (H) mixtures at inhibiting plasma thrombin generation on rat fetal distal lung epithelium (FDLE) in vitro. However, heparin cofactor II (HC) has been shown to selectively inhibit thrombin, which may be advantageous if other enzyme activities are present in the airspace. We compared the abilities of ATH, covalent HC–heparin complex (HCH), and covalent HC–dermatan sulfate (HCD) to inhibit thrombin generation on FDLE in plasmas from either adults or newborns. In the presence of ATH, peak free thrombin generation in adult plasma on the cell surface was reduced by 92% compared with controls (no anticoagulant). However, whereas HCH reduced peak free thrombin generation in adult plasma by 81%, HCD was only able to reduce activity by 33%. All covalent complexes caused a greater decrease in thrombin activity compared with that with the corresponding noncovalent serpin + heparinoid mixtures. Experiments in plasma from newborns resulted in peak free thrombin that was less than or equal to that in adult plasma when covalent conjugates were added. Relative peak free thrombin was proportional to rate of prothrombin consumption and amount of thrombin–inhibitor complexes formed. In vivo, experiments in newborn rats showed that a greater percentage of intratracheally instilled ATH and HCH could be recovered in lung lavage fluid compared withwith that for HCD. In summary, ATH, HCH, and HCD are inhibitors of thrombin generation on FDLE superior to the corresponding noncovalent mixtures, with ATH and HCH being more potent than HCD. Covalent conjugates of AT or HC with H may be preferred in treatment of extravascular coagulation.

Abbreviations: antithrombin, AT • AT–H complex, ATH • bronchopulmonary dysplasia, BPD • fetal distal lung epithelium, FDLE • heparin, H • heparin cofactor II, HC • HC–dermatan sulfate, HCD • HC–H complex, HCH • minimum essential medium, MEM • respiratory distress syndrome, RDS


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Extravascular fibrin deposition is a hallmark of respiratory distress syndrome (RDS) in premature infants (1, 2), which predisposes the lung to long-term bronchopulmonary displasia (BPD) (3). During recovery in patients with RDS, increased cell proliferation leads to remodeling of the alveolar surface (4–7) with areas of increased fibroblast growth (fibrosis) (8–11), which are supported, in part, by fibrin (12, 13). The development of an anticoagulant that would prevent fibrin formation through the regulation of thrombin, yet be safely retained within the lung, offers an attractive potential form of adjuvant therapy for the treatment of RDS and prevention of BPD.

We have developed a covalent antithrombin–heparin complex (ATH) that has a number of unique properties that are potentially useful in prevention of extravascular coagulation. Thrombin reacts directly with ATH at a rate which is 4 to 10 times faster than that of noncovalent antithrombin (AT) + heparin (H) mixtures (14, 15). Surprisingly, ATH was also found to have significant catalytic activity towards inhibition of thrombin by endogenous plasma AT (14). In vivo experiments showed that ATH had much greater antithrombotic efficacy compared with heparin, without a significant increase in hemorrhagic side effects (16). Further work revealed that the enhanced antithrombotic potency of ATH was likely due to neutralization of procoagulant activity of the clot by inhibition of fibrin-bound thrombin (17). Similar to reactions on fibrin surfaces, experiments in plasma on fetal distal lung epithelial (FDLE) cells confirmed that ATH inhibited thrombin generation to a greater extent than equivalent doses of AT + H (18). This latter result suggested the potential use of ATH in prevention or treatment of intrapulmonary coagulation. Intratracheal instillation of ATH resulted in high anticoagulant activity in the lavage fluid for over 48 h, with no measurable activity systemically (14).

In the airspace, however, other proteases exist that also react with serine proteinase inhibitors (serpins) (19, 20). Because AT has been shown to inhibit a wide range of coagulation (21–23) and other (24, 25) proteases, intra-alveolar ATH may be predisposed to rapid consumption by noncoagulant enzymes. Heparin cofactor II (HC) is a selective thrombin inhibitor (26) whose reaction can be catalyzed by either H (27, 28) or dermatan sulfate (D) (26). Further work has shown the D only accelerates thrombin inhibition by HC (26). Previously, we prepared and characterized covalent complexes of HC with H (HCH) and HC with D (HCD) (29). Both HCH and HCD rapidly inhibited thrombin but, as expected, were unable to inhibit factor Xa. However, whereas HCH readily catalyzed inhibition of thrombin by either AT or HC, HCD was only able to catalyze reaction of HC with thrombin (29). Thus, the property of HCH and HCD to selectively inhibit thrombin in the presence of other proteases may overcome depletion of these conjugates by noncoagulant enzymes found in the alveolus during a state of acute lung injury (30, 31). In order to further assess the relative anticoagulant potencies of ATH, HCH, and HCD on nonvascular surfaces, the capabilities of the three serpin–heparinoid conjugates to inhibit thrombin generation in different plasmas on FDLE were compared.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Blood Sampling and Plasma Preparation
Blood samples were taken from human subjects with the approval of the ethics board of the Hamilton Health Sciences Corporation (Ontario, Canada). Venous blood was obtained from umbilical cords taken after full-term, uneventful deliveries at St Joseph's Hospital (Hamilton, ON, Canada) or from healthy adult volunteers. A minimum of 20 individuals was used to obtain blood for either newborn (cord) or adult groups. Blood samples were collected into tubes or syringes which contained 3.2% buffered sodium citrate (volume of blood:volume of citrate = 9:1), immediately centrifuged (3,000 x g for 20 min) and platelet-poor plasma removed and stored at -80°C for future coagulation studies. Plasmas harvested were each analysed by measuring Activated Partial Thromboplastin Time and Prothombrin Time values using an ACL machine (Instrumentation Laboratory, Milan, Italy). Prothombrin Time values were expressed as International Normalized Ratios. Analysis of newborn plasma samples from the 20 individuals showed that they were all in the normal range reported previously for that age group (32). For thrombin generation experiments, pools of newborn plasma were constructed by mixture of plasma from at least 5 individuals, and pools of adult plasma were made of samples from 20 individuals.

Preparation of Covalent Conjugates
AT was from Bayer (Mississauga, ON, Canada), HC was from Affinity Biologicals (Hamilton, ON, Canada), H was from Sigma (Mississauga, ON, Canada), and D (commercial nomenclature is Chondroitin Sulfate B, which is synonomous with the conventional term Dermatan Sulfate) was also from Sigma. Covalent ATH was prepared as described previously (14). HC was covalently linked to H (HCH) or D (HCD) and products purified as reported before (29). In brief, AT was incubated with H, and HC was incubated with H or with D at 40°C for 14 d. ATH, HCH, or HCD formed was purified by sequential chromatography on butyl agarose (Sigma) or DEAE Sepharose fast flow (Amersham Pharmacia Biotech, Uppsala, Sweden). Final ATH, HCH, and HCD stocks were > 95% free of unreacted starting materials.

Cell Culture
The study received approval from the animal research ethics board of the Hamilton Health Sciences Corporation. Animal experiments conformed to guidelines from the Canadian Council on Animal Care. FDLE cells were cultured as described previously (18, 33). Lungs from fetal rats (Wistar; Charles River, ON, Canada) at 21 d gestational age (term = 22 d) were removed, washed in ice cold Hank's balanced salt solution, and cut into ~ 1-mm3 pieces with scissors. After washing in Hank's solution, lung pieces were mixed at 37°C in 0.125% trypsin (GIBCO-BRL, Burlington, ON, Canada) 0.002% deoxyribonuclease (Worthington Biochemical, Freehold, NJ), followed by filtration through a Nitex 100 mesh filter and washing of the cell suspension by mixing with 10% fetal bovine serum (GIBCO-BRL) in minimal essential medium (MEM; GIBCO-BRL) and pelleting at 120 x g for 3 min. The pellet was mixed at 37°C for 15 min with 0.1% collagenase (Worthington Biochemical) in MEM and the cells recovered by centrifugation at 300 x g for 3 min. Fibroblasts were separated from epithelium by differential adherence to T-75 tissue culture flasks (GIBCO-BRL). FDLE (> 95% purity) were cultured in 24-well Nunclon plastic plates using 10% fetal bovine serum in MEM under a humified 95% room air–5% CO2 atmosphere. Cells were grown to form confluent monolayers and experiments performed at 2–3 d after harvest.

Thrombin Generation
Thrombin generations were carried out in either adult or newborn plasmas using a procedure reported in detail before (18). Plasmas were defibrinated by incubation at 37°C for 10 min with 0.18 U Arvin (Viprinex; Knoll Pharma Inc., Markham, ON, Canada)/ml plasma, followed by winding out of the clot on a wire loop, and another 10 min incubation on ice with winding out of any further clot formed. Plates were heated on a metal block in a Dri-bath at 37°C during thrombin generation experiments. Culture media was removed from the wells and the cell monolayer washed twice with 1 ml of 0.036 M Na acetate 0.036 M Na diethylbarbiturate 0.145 M NaCl pH 7.4 buffer. One hundred microliters of buffer + 200 µl of defibrinated plasma (each prewarmed at 37°C) were added to the washed FDLE surface with mixing and maintained at 37°C for 3 min. Following pre-incubation of the cells with buffer + defibrinated plasma, 100 µl of 0.040 M CaCl2 in buffer was added with mixing as a clock was started. At various time points after recalcification, 25 µl subsamples were removed and immediately mixed with 0.005 M Na2EDTA on ice in order to stop further conversion of prothrombin to thrombin. Total thrombin activity in each time sample was determined by mixing 25-µl aliquots of EDTA-reaction mixture with 0.16 mM S-2238 substrate (DiaPharma, West Chester, OH) in buffer and incubating at 37°C for 10 min, followed by addition of 200 µl of 50% (vol/vol) acetic acid to terminate substrate hydrolysis. The absorbance at 405 nm of the neutralized chromogenic reaction sample was measured and compared with that of similar reactions with purified human thrombin (Enzyme Research Laboratories, South Bend, IN) in order to calculate nM thrombin concentration. Calculations were corrected for dilution to determine the amount of thrombin that would be present at each time point in undiluted plasma. Because thrombin retains activity against small substrates when bound to {alpha}-2-macroglobulin ({alpha}2M) (34), the concentration of thrombin–{alpha}2M complexes at each time point was determined by carrying out experiments similar to those for total thrombin, except that uncomplexed thrombin was first inhibited by addition of AT + heparin. Subtraction of the thrombin–{alpha}2M complex concentration from that of total thrombin gave the amount of free thrombin.

The effect of various anticoagulants was studied by addition of reagent to the buffer that was mixed with defibrinated plasma on the cell surface. ATH, HCH, HCD, equimolar mixture (noncovalent) of AT + H, equimolar mixture of HC + H, or equimolar mixture of HC + D were tested in either adult or newborn plasma. Agents were added so as to give final concentrations in the recalcified defibrinated plasma reaction mixture of 8.5 nM.

Prothrombin Consumption and Inhibitor Complex Formation
Reaction mixture-EDTA time samples were analyzed for either prothrombin or inhibitor complexes of thrombin-AT or thrombin-HC in order to determine the amount of prothrombin consumed or inhibitor complexes formed over time. Prothrombin was analyzed using a zymogen-specific prothrombin ELISA kit from Affinity Biologicals. Thrombin-AT and thrombin-HC were each determined using thrombin-AT and thrombin-HC ELISA kits, also from Affinity Biologicals.

Intratracheal Instillation of Serpin–Heparinoids
ATH, HCH, and HCD were introduced into the lungs of newborn rats in order to determine the retention of these complexes in the airspace. Pups were delivered by SPF Wistar rats and allowed to be nurtured by the dams. Five days after birth, the pups were anaesthetized by placement in a container with an O2 (1 liter/min) + isoflurane (1–3%) atmosphere. In turn, each pup was removed for treatment, anaesthesia being maintained by O2 (1 liter/min) + isoflurane (1–3%), which was delivered by face mask. A small incision was made into the trachea. Serpin–Heparinoid compound (10–200 µg protein), dissolved in 100 µl of 0.15 M NaCl, was put into the lungs of the pup by injection from a syringe through a 30-gauge needle inserted into the tracheal opening. Once the compound had been instilled, skin over the tracheal opening was sutured shut and the pup was allowed to recover normally in room air before being returned to the dam. At different time periods after intratracheal instillation, each pup was re-anaesthetized as described above, sutured trachea reopened and the lungs lavaged by syringe with 200 µl of saline. Lavage fluids were analyzed for either: human AT (AT ELISA kit from Affinity Biologicals), human HC (HC ELISA kit from Affinity Biologicals) or (in the case of ATH experiments) anti-factor Xa heparin activity (Stachrom Heparin kit; Diagnostica Stago, Asniéres, France).

Statistical Analysis
For each experimental group, five trials were carried out. An analysis of variance (ANOVA) was used to compare test results between groups. In some cases a Student's t test was performed to analyze differences between groups. A P value of < 0.05 was considered significant. All values are expressed as means ± SEM.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Inhibition of Thrombin Generation in Adult Plasma by Covalent Conjugates
Free thrombin generation in recalcified defibrinated plasma was determined as the difference between total thrombin activity and residual activity from thrombin–{alpha}2M complexes. Curves for free thrombin generation time courses in adult plasma, with or without serpin–heparinoid conjugates, are shown in Figure 1 . Compared with control experiments with no anticoagulant added, peak thrombin generation was strongly reduced (by 92%) and delayed (8 min compared with 4 min) in the presence of ATH. Reactions containing HCH resulted in significant reduction in free thrombin activity compared with that of controls (81% decrease in peak concentrations), albeit to a slightly lesser extent than with ATH. In contrast to results with ATH and HCH, peak free thrombin was decreased by a much smaller degree (33% lower than control) when equivalent molar concentrations of HCD were added. Assessment of the thrombin potential from area under the curve calculations (Table 1) confirmed the trend in relative potencies given by the peak free thrombin values.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 1. Thrombin generation in adult plasma on fetal distal lung epithelium in the presence of anticoagulants. Free thrombin generation in recalcified defibrinated adult plasma was measured on fetal distal lung epithelial monolayers in the presence of either buffer (control, circles), covalent antithrombin–heparin (ATH, squares), covalent heparin cofactor II–heparin (HCH, triangles), or covalent heparin cofactor II–dermatan sulfate (HCD, diamonds). Concentration of each covalent complex in the reaction mixture on the cells was 8.5 nM. Values are mean ± SEM, n >= 5.

 

View this table:
[in this window]
[in a new window]
  		TABLE 1 Area under the curve for thrombin generation in adult and newborn plasmas on fetal distal lung epithelium

 
Prothrombin Consumption in Adult Plasma Containing Serpin-Heparinoids
Prothrombin concentrations were determined from EDTA-reaction mixture time samples. Prothrombin consumption on recalcified defibrinated adult plasma on FDLE, in the absence or presence of different serpin–heparinoid complexes, is shown in Figure 2 . Although the majority of prothrombin was consumed in each group, conversion of prothrombin to thrombin occurred at a somewhat slower rate with ATH and HCH, compared with buffer control experiments. Alternatively, no significant difference in rate for disappearance of prothrombin was seen for reactions containing HCD compared with that of controls. The decreased rate of prothrombin consumption for experiments in the presence of ATH and HCH compared with that with HCD was consistent with the increased free thrombin observed for the HCD group (compare Figure 2 with Figure 1 and Table 1).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 2. Prothrombin consumption in adult plasma on fetal distal lung epithelium in the presence of anticoagulants. Disappearance of prothrombin in recalcified defibrinated adult plasma was measured on fetal distal lung epithelial monolayers in the presence of either buffer (control, circles), covalent antithrombin–heparin (ATH, squares), covalent heparin cofactor II–heparin (HCH, triangles), or covalent heparin cofactor II–dermatan sulfate (HCD, diamonds). Values are mean ± SEM, n >= 5.

 
Comparison of the Inhibition of Thrombin Generation in Adult Plasma by Covalent Serpin–Heparinoids to that of Noncovalent Serpin 1 Heparinoid Mixtures
Thrombin generation in adult plasma on FDLE was carried out with either covalent serpin–heparinoids or the corresponding equimolar noncovalent serpin + heparinoid mixtures. Results are given in Figure 3 . In each case, peak free thrombin was decreased for ATH, HCH, and HCD compared with that with AT + H, HC + H, and HC + D, respectively. Similarly, total area under the curve values for free thrombin was decreased with ATH, HCH, and HCD in comparison with their corresponding noncovalent serpin + heparinoid mixtures (P < 0.05, Table 1).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 3. Comparison of the effect of covalent versus noncovalent serpin–heparinoid complexes on thrombin generation in adult plasma. Free thrombin generation in recalcified defibrinated adult plasma was measured on fetal distal lung epithelium. (A) The effect of covalent antithrombin–heparin complex (ATH, open squares) was compared with noncovalent antithrombin (AT) + heparin (H) (filled squares). (B) The effect of covalent heparin cofactor II–H complex (HCH, open triangles) was compared with noncovalent heparin cofactor II (HC) + H (filled triangles). (C) The effect of covalent HC–dermatan sulfate complex (HCD, open diamonds) was compared with noncovalent HC + dermatan sulfate (D) (filled diamonds). Values are mean ± SEM, n >= 5.

 
Inhibition of Thrombin Generation in Newborn Plasma by Covalent Conjugates
The effects of serpin–heparinoid complexes on free thrombin generation on FDLE in plasma from newborns on FDLE were compared. Curves for free thrombin generation time courses in newborn plasma, with or without serpin–heparinoid conjugates, are shown in Figure 4 . Amount of thrombin generated in newborn plasma containing either buffer, ATH, HCH, or HCD (Figure 4) was less than or equal to the levels observed in parallel experiments in adult plasma (Figure 1). Similar to experiments in adult plasma, peak thrombin generation in newborn plasma was strongly decreased in the presence of ATH (reduced by 86%) or HCH (peak height decreased by 90%), compared with buffer control. However, HCD caused a reduction in peak free thrombin of 43% relative to the peak in buffer control reactions. The decrease in peak free thrombin by HCD in newborn plasma was somewhat greater than the corresponding experiments in adult plasma (compare Figures 4 and 1). Area under the curve data again confirmed the trend in relative potencies given by the peak free thrombin values (Table 1).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 4. Thrombin generation in newborn plasma on fetal distal lung epithelium in the presence of anticoagulants. Free thrombin generation in recalcified defibrinated newborn (cord) plasma was measured on fetal distal lung epithelial monolayers in the presence of either buffer (control, circles), covalent antithrombin–heparin (ATH, squares), covalent heparin cofactor II–heparin (HCH, diamonds), or covalent heparin cofactor II–dermatan sulfate (HCD, diamonds). Concentration of each covalent complex in the reaction mixture on the cells was 8.5 nM. Values are mean ± SEM, n >= 5.

 
Prothrombin Consumption in Newborn Plasma Containing Serpin–Heparinoids
Starting prothrombin concentrations prior to initiation of thrombin generation (addition of calcium) was significantly decreased in newborn plasma compared with adult (compare Figures 5 and 2) . During thrombin generation, the rate of prothrombin consumption in newborn plasma on FDLE was decreased in the presence of ATH and HCH (Figure 5). As in the case of adult plasma, the prothrombin consumption in newborn plasma containing HCD was more rapid than that of the other conjugates, approaching that of plasma without added anticoagulant. The decreased rate of prothrombin consumption for experiments in the presence of ATH and HCH compared with that with HCD was consistent with the increased free thrombin observed for the HCD group (compare Figure 5 with Figure 4 and Table 1).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 5. Prothrombin consumption in newborn plasma on fetal distal lung epithelium in the presence of anticoagulants. Disappearance of prothrombin in recalcified defibrinated newborn plasma was measured on fetal distal lung epithelial monolayers in the presence of either buffer (control, circles), covalent antithrombin–heparin (ATH, squares), covalent heparin cofactor II–heparin (HCH, triangles), or covalent heparin cofactor II–dermatan sulfate (HCD, diamonds). Values are mean ± SEM, n >= 5.

 
Comparison of the Inhibition of Thrombin Generation in Newborn Plasma by Covalent Serpin–Heparinoids to that of Noncovalent Serpin 1 Heparinoid Mixtures
Thrombin generation in newborn plasma on FDLE was carried out with either covalent serpin-heparinoids or the corresponding equimolar noncovalent serpin + heparinoid mixtures. In each case, peak free thrombin was decreased for ATH, HCH, and HCD compared with that with AT + H, HC + H, and HC + D, respectively (Figure 6) . However, the magnitude of differences in newborn plasma thrombin generation between covalent and noncovalent complexes was not as striking as in the case of adult plasma.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 6. Comparison of the effect of covalent versus noncovalent serpin–heparinoid complexes on thrombin generation in newborn plasma. Free thrombin generation in recalcified defibrinated newborn plasma was measured on fetal distal lung epithelium. (A) The effect of covalent antithrombin–heparin complex (ATH, open squares) was compared with noncovalent antithrombin (AT) + heparin (H) (filled squares). (B) The effect of covalent heparin cofactor II–H complex (HCH, open triangles) was compared with noncovalent heparin cofactor II (HC) + H (filled triangles). (C) The effect of covalent HC–dermatan sulfate complex (HCD, open diamonds) was compared with noncovalent HC + dermatan sulfate (D) (filled diamonds). Values are mean ± SEM, n >= 5.

 
Inhibitor-Complex Formation
Concentrations of thrombin-inhibitor complexes formed after 30 min of thrombin generation in adult and newborn plasmas are shown in Table 2. Total inhibitor complexes were greater in adult plasma than in newborn plasma, regardless of whether anticoagulant was added or not. Furthermore, thrombin–{alpha}2M complexes were a greater proportion of the total complexes in newborn plasma compared with that in adult plasma for all systems tested. As expected, thrombin–AT complexes were significantly increased in the presence of ATH or AT + H (P < 0.05, Table 2) and thrombin–HC complexes were significantly elevated in the presence of HCD or HC + D (P < 0.05, Table 2). When HCH or HC + H were added to adult plasma, a greater amount of thrombin–AT was generated, whereas the proportion of thrombin–HC was insignificantly different from reactions without added anticoagulant. A similar trend was observed when newborn plasma was supplemented with either HCH or HC + H; however, the increase in thrombin–AT complexes compared with control experiments was not as great as in the adult system.


View this table:
[in this window]
[in a new window]
  		TABLE 2 Inhibitor complexes

 
Recovery of Serpin–Heparinoid Complexes Instilled into the Lung
Significant differences in effectiveness of the different serpin–heparinoids were observed for inhibition of thrombin generation in plasma on the FDLE surface. These comparisons were made between equimolar concentrations of the serpin–heparinoids. However, it was not clear if similar amounts of serpin–heparinoids would be retained in the fluid phase over the alveolar epithelium in vivo. Therefore, to assess bioavailability of covalent conjugates in the lung airspace, ATH, HCH, and HCD were intratracheally instilled into the lungs of newborn rats, followed at various times by attempted recovery from the alveoli by saline lavage. Figure 7 shows the data for analyses of lavage of lungs from newborn rats that received serpin–heparinoids. It can be seen that there was a significantly greater recovery of ATH and HCH at 24 h after instillation, compared with that for HCD. In further experiments, analysis of the recovery of ATH activity at various times after instillation showed that significant anti-factor Xa levels in lavage fluid could be detected for up to 4 d (Figure 8) . Thus, although all compounds could be retained in lung fluid for at least 24 h, ATH and HCH exhibited the greatest long-term activity in fluid on the epithelial surface in vivo.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 7. Recovery of serpin-heparinoids in lung lavage fluid. Antithrombin–heparin (ATH), heparin cofactor II–heparin (HCH), or heparin cofactor II-dermatan sulfate (HCD) were instilled into the lungs of 5-d-old rats and the lungs lavaged with saline 24 h after instillation. Lavage fluids were analyzed for either human antithrombin or human heparin cofactor II, and the % recovery of each complex instilled was calculated. Values are mean ± SEM, n >= 2.

 


View larger version (35K):
[in this window]
[in a new window]
  		Figure 8. Effect of time on recovery of antithrombin–heparin (ATH) activity in lung lavage fluid. ATH was instilled into the lungs of 5-d-old rats and the lungs of each rat were lavaged (only once for each rat) with saline at various times after instillation. Lavage fluids were analyzed for anti-factor Xa activity and the activity plotted versus time. Values are mean ± SEM, n >= 3.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Neonatal and adult RDSs are characterized by leakage of plasma proteins of varying sizes into the airspace (35–37), which leads to interstitial and intra-alveolar thrombin generation with subsequent fibrin deposition. Strong and convincing evidence exists which clearly shows that coagulation leading to fibrin formation is a key effector of lung injury (38). Although absent in normal lung, the presence of fibrin in alveolar and interstitial compartments during evolving diffuse alveolar damage is a hallmark of RDS (1, 2, 39, 40). Furthermore, fibrin deposition in the vasculature and pulmonary artery during RDS indicates the existence of vasoconstrictor mechanisms in the occurrence of increased pulmonary vascular resistance in RDS (41). Intra-alveolar fibrin deposition can have marked short- and long-term detrimental effects. Fibrin has been shown to significantly impair surfactant function (42, 43), whereas fibrin degradation products have been linked to increased alveolar–capillary membrane permeability (39, 44, 45), thus further increasing plasma protein leakage into the airspace. Leakage of protein into the lungs has itself been associated with the systemic activation of: clotting, complement, and polymorphonuclear lymphocytes (35). Fibroblasts are recruited by and proliferate in regions containing fibrin (12, 13), which leads to the evolution of pulmonary fibrosis (1), one of the key features of BPD. Thus, inhibition of fibrin formation in plasma on the alveolar cell surface may be an important factor in the control of acute and long-term damage to lung function.

Recently, we have developed and partially characterized active covalent conjugates of ATH, HCH, and HCD (14, 15, 29). ATH has been shown to rapidly inhibit thrombin as well as potently catalyze the reaction of thrombin with exogenous AT (14). Previous experiments on FDLE indicated that ATH inhibited thrombin generation in adult plasma to a greater extent compared with equivalent concentrations of noncovalent AT + H (18). ATH could also be retained in the lung for at least 48 h without loss to the systemic circulation (14). These properties indicated that ATH may be effective at preventing intra-alveolar coagulation leading to increased lung injury during RDS and developing BPD. HCH and HCD also inhibit thrombin at a rapid rate in vitro (29). Additionally, unlike ATH (14, 15), HCH and HCD selectively neutralize thrombin; this would be advantageous for inhibition of thrombin generated in the presence of other proteases. However, inhibition of thrombin generation on FDLE by HCH and HCD has not been studied. Furthermore, because RDS occurs in both adults and especially newborns, developmental differences in plasma components may be important. Thus, we decided to compare the inhibition of thrombin generation by covalent and noncovalent serpin–heparinoid complexes on FDLE in either adult or newborn (cord) plasma.

ATH and HCH were both significantly more effective at inhibiting thrombin generation on FDLE in adult plasma than HCD (Figure 1 and Table 1). This result is consistent with the increased anticoagulant potency on a weight basis of H compared with D found previously in plasma systems (46, 47). Further analyses revealed that the lower free thrombin observed in adult plasma experiments with HCH and, particularly, ATH resulted from a reduced rate of conversion of prothrombin to thrombin (Figure 2), combined with an increase in thrombin–AT inhibitor complex formation (Table 2). Thus, although the vast majority of prothrombin was consumed over the full time period studied, covalent serpin–heparin complexes would prevent the appearance of free thrombin and likely inhibited the rapid conversion of prothrombin to free thrombin. The increased catalytic generation of thrombin-AT by ATH and HCH (via plasma AT) would have decreased the feedback activation of prothrombin by thrombin (48–50) and also would decrease free thrombin activity, as observed. Interestingly, the total concentration of inhibitor complexes formed with HCD was comparable to that with ATH, HCH, or buffer, mainly due to increased thrombin-HC (Table 2). The fact that the final number of total inhibitor complexes formed is similar regardless of treatment is consistent with the observation that all starting prothrombin (Figure 2) and generated free thrombin (Figure 1) were fully consumed by the end point. It has been previously shown that catalysis of the AT inhibition of thrombin by heparin (15) has a faster rate constant than catalysis of the HC inhibition of thrombin by dermatan sulfate (29). Thus, although HCD inhibited all free thrombin by facilitation of the thrombin + HC reaction, the slower catalytic rate of thrombin inhibition via the D moiety on HCD compared with the H component of ATH or HCH would result in greater expression of free thrombin activity and concomitant rapid prothrombin consumption. Moreover, it was clear that direct noncatalytic inhibition of thrombin by the conjugates was not the critical factor in reducing thrombin activity, because HCH was more potent in our studies than HCD, even though the rate constants for direct thrombin reaction with each are comparable (29). In all cases, the conjugates reduced thrombin generation in adult plasma to a greater degree than their noncovalent counterparts (Figure 3). This was not surprising, for two reasons. First, specific catalytic antithrombin activities of all covalent serpin–heparinoids are at least equal to, and in some cases significantly greater than, the noncovalent mixtures (14, 29). Second, small amount of thrombin formed initially (that causes feedback activation) would be more rapidly neutralized by the covalent complexes. This latter property is due to the fact that the reaction step of serpin binding by the heparinoid, which is the rate-determining step for AT inhibition of thrombin (51), was eliminated by covalent linkage of the inhibitor to the catalyst. Therefore, similar to comparisons of HCD with ATH/HCH, although final inhibitor complex formation was at least as great in the presence of noncovalent serpin + heparinoid mixtures (Table 2), covalent serpin–heparinoids would directly inhibit thrombin at a faster rate. Experiments carried out on FDLE using newborn plasma gave trends that were similar to those with adult plasma. Again, ATH and HCH caused greater reduction in free thrombin generation (Figure 4, Table 1) and prothrombin consumption rate (Figure 5) compared with HCD or buffer. The impact of ATH and HCH on prothrombin consumption was more marked in newborn plasma than adult plasma, likely due to the elevated AT:prothrombin molar ratio in newborn compared with adult plasma (32, 52). In all systems studied, compared with adult plasma, thrombin–{alpha}2M complexes composed a higher proportion of the total inhibitor complexes in newborn plasma (Table 2). Previous work has revealed that the elevated {alpha}2M concentrations in the newborn (32) make it a relatively more important thrombin inhibitor, both in the fluid phase (53) and on cell surfaces (54, 55). Although ATH and HCH were found to more potent at inhibiting thrombin generation on FDLE compared with HCD in vitro, it was possible that there may be differences in retention of ATH, HCH or HCD in the lung fluid in vivo. Further experiments showed that a significantly greater proportion of ATH and HCH instilled into newborn rat lung could be recovered in the lavage after 24 h compared with that of HCD (Figure 7). In fact, ATH activity could be detected over a time course up to 4 d after a single bolus pulmonary instillation (Figure 8). Thus, in addition to increased potency against thrombin generation, ATH and HCH may also have an added advantage over HCD with respect to retention within the lung fluid.

In conclusion, as reported previously for ATH (18), covalent complexes of HC with H or D inhibited free thrombin generation in adult plasma on FDLE more effectively than the corresponding equimolar noncovalent mixtures. A similar trend was found for thrombin generation in newborn plasma, with even more striking decreases in rate of prothrombin consumption. Regardless of plasma system, epithelial surface thrombin generation was more efficiently reduced by covalent heparin complexes with AT or HC, relative to HCD. Comparison of results involving noncovalent AT·H, HC·H, and HC·D complexes gave a similar outcome. Thus, H conjugates may be superior to those with D for prevention of coagulant-related problems during RDS. In order to confirm the relative effectiveness of serpin–heparinoids in ameliorating lung damage, studies in animal models of lung injury are required, along with future clinical testing.


   Footnotes
 
* This work was supported by Project 4 from the Canadian Institutes of Health Research group in Developmental Lung Biology. A.C. holds a Research Scholarship award from the Heart and Stroke Foundation of Canada. During the course of this study, Dr. Maureen Andrew passed away. She was instrumental in both the development of serpin–heparinoids and guidance in the overall direction of these studies. The authors dedicate this manuscript to her memory. Back

Received in original form February 22, 2002

Received in final form September 3, 2002


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bachofen, M., and E. Weibel. 1982. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin. Chest Med. 3:35–56.[Medline]
  2. Gitlin, D., and J. Craig. 1956. The nature of the hyaline membrane in asphyxia of the newborn. Pediatrics 17:64–71.[Abstract/Free Full Text]
  3. Singhal, K. K., and L. A. Parton. 1996. Plasminogen activator activity in preterm infants with respiratory distress syndrome: relationship to the development of bronchopulmonary dysplasia. Pediatr. Res. 39:229–235.[Medline]
  4. Buckley, S., K. C. Bui, M. Hussain, and D. Warburton. 1996. Dynamics of TGF-beta III peptide activity during rat alveolar epithelial cell proliferative recovery from acute hyperoxia. Am. J. Physiol. 271:L54–L60.[Abstract/Free Full Text]
  5. Kelleher, M. D., E. T. Naureckas, J. Solway, and M. B. Hershenson. 1995. In vivo hyperoxic exposure increases cultured lung fibroblasts proliferation and C-Ha-rds expression. Am. J. Respir. Cell Mol. Biol. 12:19–26.[Abstract]
  6. Fracica, P. J., S. P. Caminiti, C. A. Piantadosi, F. G. Duhlongsod, J. D. Crapo, and S. L. Young. 1994. Natural surfactant and hyperoxic lung injury in primates: II. Morphometric analyses. J. Appl. Physiol. 76:1002–1010.[Abstract/Free Full Text]
  7. Coflesky, J. T., K. B. Adler, J. Woodcock-Mitchell, J. Mitchell, and J. N. Evans. 1998. Proliferative changes in the pulmonary arterial wall during short term hyperoxic injury to the lung. Am. J. Pathol. 132:563–573.[Abstract]
  8. Finkelstein, J. N. 1990. Physiologic and toxicologic responses of alveolar type II cells. Toxicology 60:41–55.[CrossRef][Medline]
  9. Adamson, I. J., L. Young, and D. H. Bowden. 1988. Relationship of alveolar epithelial injury and repair to the induction of pulmonary fibrosis. Am. J. Pathol. 130:377–383.[Abstract]
  10. Adamson, I. J., C. Hedgecock, and D. H. Bowden. 1990. Epithelial cell fibroblast interactions in lung injury and repair. Am. J. Pathol. 137:251–262.
  11. Bowden, D. H., L. Young, and I. Y. Adamson. 1994. Fibroblast inhibition does not promote normal lung repair after hypoxia. Exp. Lung Res. 30:251–262.
  12. Damiano, V., P. Cherian, F. Frankel, J. Steeger, M. Sohn, D. Oppenheim, and G. Weinbaum. 1990. Intraluminal fibrosis induced unilaterally by lobar instillation of CdCl2 into the rat lung. Am. J. Pathol. 137:883–894.[Abstract]
  13. Fukuda, Y., M. Ishizaki, Y. Masuda, G. Kimura, O. Kawanami, and Y. Masugi. 1987. The role of intraalveolar fibrosis in the process of pulmonary structural remodelling in patients with diffuse alveolar damage. Am. J. Pathol. 126:171–182.[Abstract]
  14. Chan, A. K., L. Berry, H. O'Brodovich, P. Klement, M. Mitchell, B. Baranowski, P. Monagle, and M. Andrew. 1997. Covalent antithrombin-heparin complexes with high anticoagulant activity: intravenous, subcutaneous and intratracheal administration. J. Biol. Chem. 272:22111–22117.[Abstract/Free Full Text]
  15. Berry, L., A. Stafford, J. Fredenburgh, H. O'Brodovich, L. Mitchell, J. Weitz, M. Andrew, and A. K. Chan. 1998. Investigation of the anticoagulant mechanisms of a covalent antithrombin-heparin complex. J. Biol. Chem. 273:34730–34736.[Abstract/Free Full Text]
  16. Chan, A. K., L. Berry, P. Klement, J. Julian, M. Mitchell, J. Weitz, J. Hirsh, and M. Andrew. 1998. A novel antithrombin-heparin covalent complex: Antithrombotic and bleeding studies in rabbits. Blood Coagul. Fibrinolysis 9:587–595.[Medline]
  17. Becker, D. L., J. C. Fredenburgh, A. R. Stafford, and J. I. Weitz. 1999. Exosites 1 and 2 are essential for protection of fibrin-bound thrombin from heparin-catalyzed inhibition by antithrombin and heparin cofactor II. J. Biol. Chem. 274:6226–6233.[Abstract/Free Full Text]
  18. Chan, A. K., L. Berry, L. Mitchell, B. Baranowski, H. O'Brodovich, and M. Andrew. 1998. Effect of a novel covalent antithrombin-heparin complex on thrombin generation on fetal distal lung epithelium. Am. J. Physiol. 274:L914–L921.[Abstract/Free Full Text]
  19. Pardo, A., and M. Selman. 1999. Proteinase-antiproteinase imbalance in the pathogenesis of emphysema: the role of metalloproteinase in lung damage. Histol. Histopathol. 14:227–233.[Medline]
  20. Kawano, N., H. Osawa, T. Ito, Y. Nagashima, F. Hirahara, Y. Inayama, Y. Nakatani, S. Kimura, H. Kitajima, N. Koshikawa, K. Miyazaki, and H. Kitamura. 1997. Expression of gelatinase A, tissue inhibitor of metalloproteinases-2, matrilysin, and trypsin(ogen) in lung neoplasms: an immunohistochemical study. Hum. Pathol. 28:613–622.[CrossRef][Medline]
  21. Nilson, I. M. 1987. Coagulation and fibrinolysis. Scand. J. Gastroenterol. 137:11–18.
  22. Buller, H. R., and T. Ten Cate. 1995. Coagulation and platelet activation pathways: a review of the key components and the way in which these can be manipulated. Eur. Heart J. 16:8–10.
  23. Butenas, S., C. van't Veer, and K. G. Mann. 1999. "Normal" thrombin generation. Blood 94:2169–2178.[Abstract/Free Full Text]
  24. Boudier, C., and J. G. Bieth. 2001. The reaction of serpins with proteinases involves important enthalpy changes. Biochemistry 40:9962–9967.[CrossRef][Medline]
  25. Hahn, B. S., S. Y. Cho, M. Y. Ahn, and Y. S. Kim. 2001. Purification and characterization of a plasmin-like protease from Tenodera sinensis (Chinese mantis). Insect Biochem. Mol. Biol. 31:573–581.[CrossRef][Medline]
  26. Tollefsen, D., C. Petska, and J. Monafa. 1983. Activation of heparin cofactor II by dermatan sulfate. J. Biol. Chem. 256:6713–6716.
  27. Bourin, M., and U. Lindhal. 1993. Glycosaminoglycans and the regulation of blood coagulation. Biochem. J. 289:313–330.
  28. Ofosu, F. A., G. J. Modi, M. A. Blajchman, M. R. Buchanan, and E. A. Johnson. 1987. Increased sulphation improves the anticoagulant activities of heparan sulphate and dermatan sulphate. Biochem. J. 248:889–896.[Medline]
  29. Monagle, P., L. Berry, H. O'Brodovich, M. Andrew, and A. Chan. 1998. Covalent heparin cofactor II-heparin and heparin cofactor II-dermatan sulfate complexes. J. Biol. Chem. 273:33566–33571.[Abstract/Free Full Text]
  30. Uchiba, M., K. Okajima, K. Murakami, H. Okabe, and K. Takatsuki. 1995. Endotoxin-induced pulmonary vascular injury is mainly mediated by activated neutrophils in rats. Thromb. Res. 78:117–125.[CrossRef][Medline]
  31. Taniguchi, H., Y. Kondoh, N. Ikuta, and K. Takagi. 1995. Diagnosis of acute respiratory distress syndrome: analysis of bronchoalveolar lavage fluid. Nihon Kyobu Shikkan Gakkai Zasshi 33:251–256.
  32. Andrew, M., B. Paes, R. Milner, M. Johnston, L. Mitchell, D. M. Tollefsen, and P. Powers. 1987. Development of the human coagulation system in the full-term infant. Blood 70:165–172.[Abstract/Free Full Text]
  33. O'Brodovich, H., B. Rafii, and M. Post. 1990. Bioelectric properties of fetal alveolar epithelial monolayers. Am. J. Physiol. 258:L201–L206.[Abstract/Free Full Text]
  34. Berry, L., M. Andrew, M. Post, F. Ofosu, and H. O'Brodovich. 1991. A549 lung epithelial cell line synthesize anticoagulant molecules on the cell surface, matrix, and in conditioned media. Am. J. Respir. Cell Mol. Biol. 4:338–346.
  35. Brus, F., W. van Oeveren, A. Heikamp, A. Okken, and S. B. Oetomo. 1996. Leakage of protein into lungs of preterm ventilated rabbits is correlated with activation of clotting, complement, and polymorphonuclear leukocytes in plasma. Pediatr. Res. 39:958–965.[Medline]
  36. Holter, J., J. Weiland, E. Packt, J. Gadek, and W. Davis. 1986. Protein permeability in the adult respiratory distress syndrome. Loss of size selectivity of the alveolar epithelium. J. Clin. Invest. 78:1513–1522.
  37. Sprung, C. L., W. M. Long, E. H. Marcial, R. M. H. Schein, R. E. Parker, T. Schomer, and K. L. Brigham. 1987. Distribution of proteins in pulmonary edema. Am. Rev. Respir. Dis. 136:957–963.[Medline]
  38. Idell, S. 2001. Anticoagulants for acute respiratory distress syndrome: can they work? Am. J. Respir. Crit. Care Med. 164:517–520.[Free Full Text]
  39. Idell, S. 1996. Extravascular coagulation and fibrin deposition in acute lung injury. New Horiz. 153:163–167.
  40. Kuhn, C., J. Boldt, T. E. King, E. Crouch, T. Vartio, and J. A. McDonald. 1989. An immunohistochemical study of architectural remodeling and connective tissue synthesis in pulmonary fibrosis. Am. Rev. Respir. Dis. 140:1693–1703.[Medline]
  41. Zapol, W. M., and R. Jones. 1987. Vascular components of ARDS: clinical pulmonary hemodynamics and morphology. Am. Rev. Respir. Dis. 136:471–474.[Medline]
  42. Seeger, W., G. Stohr, and H. R. D. Wolf. 1985. Alteration of surfactant function due to protein leakage: special interaction with fibrin monomer. J. Appl. Physiol. 58:326–338.[Abstract/Free Full Text]
  43. O'Brodovich, H., J. Weitz, and F. Possmayer. 1990. Effect of fibrinogen degradation products and lung ground substance on surfactant function. Biol. Neonate 57:325–333.[Medline]
  44. Saldeen, T. 1982. Fibrin-derived peptides and pulmonary injury. Ann. N. Y. Acad. Sci. 384:319–331.[Medline]
  45. Stevens, T., J. G. N. Garcia, D. M. Shasby, J. Bhattacharya, and A. B. Malik. 2000. Mechanisms regulating endothelial cell barrier function. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 279:L419–L422.
  46. Ofosu, F. A., M. A. Blajchman, G. J. Modi, L. M. Smith, M. R. Buchanan, and J. Hirsh. 1985. The importance of thrombin inhibition for the expression of the anticoagulant activities of heparin, dermatan sulfate, low molecular weight heparin and pentosan polysulphate. Br. J. Haematol. 60:695–704.[Medline]
  47. Cohen, A. T., M. J. Phillips, R. A. Edmondson, J. A. Skinner, S. K. Das, D. J. Cooper, E. M. Thomas, E. Melissari, and V. V. Kakkar. 1994. A dose ranging study to evaluate dermatan sulphate in preventing deep vein thrombosis following total hip arthroplasty. Thromb. Haemost. 72:793–798.[Medline]
  48. Naito, K., and K. Fujikawa. 1991. Activation of human blood coagulation factor XI independent of factor XII: factor XI is activated by thrombin and factor XIa in the presence of negatively charged surfaces. J. Biol. Chem. 266:7353–7358.[Abstract/Free Full Text]
  49. Ofosu, F. A., P. Sie, G. J. Modi, F. Fernandez, M. R. Buchanan, M. A. Blajchman, B. Boneu, and J. Hirsh. 1987. The inhibition of thrombin-dependent positive-feedback reactions is critical to the expression of the anticoagulant effect of heparin. Biochem. J. 243:579–588.[Medline]
  50. Ofosu, F. A., L. Liu, and J. Feedman. 1996. Control mechanisms in thrombin generation. Semin. Thromb. Haemost. 22:303–308.[Medline]
  51. Pletcher, C. H., and G. L. Nelsestuen. 1982. The rate-determining step of the heparin-catalyzed antithrombin/thrombin reaction is independent of thrombin. J. Biol. Chem. 257:5342–5345.[Abstract/Free Full Text]
  52. Vieira, A., L. Berry, F. Ofosu, and M. Andrew. 1991. Heparin sensitivity and resistance in the neonate: An explanation. Thromb. Res. 63:85–98.[CrossRef][Medline]
  53. Schmidt, B., L. Mitchell, F. Ofosu, and M. Andrew. 1989. Alpha-2-macroglobulin is an important progressive inhibitor of thrombin in neonatal and infant plasma. Thromb. Haemost. 62:1074–1077.[Medline]
  54. Xu, L., M. Delorme, L. Berry, F. Ofosu, L. Mitchell, B. Paes, and M. Andrew. 1995. Alpha-2-macroglobulin remains as important as ATIII for thrombin regulation in cord plasma in the presence of endothelial cell surfaces. Pediatr. Res. 37:1–6.[Medline]
  55. Andrew, M., L. Berry, and H. O'Brodovich. 1994. Thrombin inhibition by fetal distal lung epithelium is different in fetal and adult plasma. Am. J. Respir. Cell Mol. Biol. 11:35–41.[Abstract]



This article has been cited by other articles:


Home page
 Am. J. Respir. Cell Mol. Bio.Home page
G. H. Caughey
Building A Better Heparin
Am. J. Respir. Cell Mol. Biol., February 1, 2003; 28(2): 129 - 132.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Berry, L. R.
Right arrow Articles by Chan, A. K. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Berry, L. R.
Right arrow Articles by Chan, A. K. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Crit. Care Med.
Copyright © 2003 American Thoracic Society.