This Article Abstract Full Text (PDF) Online Data Supplement Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Idell, S. Articles by Miller, E. Search for Related Content PubMed PubMed Citation Articles by Idell, S. Articles by Miller, E.

Single-Chain Urokinase Alone or Complexed to Its Receptor in Tetracycline-induced Pleuritis in Rabbits

Department of Specialty Care Services, The University of Texas Health Center at Tyler, Tyler, Texas; Attenuon, La Jolla, California; Department of Pathology and Laboratory Medicine, The University of Pennsylvania, Philadelphia, Pennsylvania; Louisiana State University Health Sciences Center, Shreveport, Louisiana; and Department of Surgery, North Shore-LIJ Research Institute, Manhasset, New York

Correspondence and requests for reprints should be addressed to Steven Idell, M.D., Ph.D., Chairman, Department of Specialty Care Services, The University of Texas Health Center at Tyler, 11937 U.S. HWY 271, Tyler, TX 75708. E-mail: steven.idell{at}uthct.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Intrapleural loculation can increase morbidity in hemothoraces or parapneumonic effusions. Intrapleural fibrin precedes visceral–parietal pleural adhesions. We speculated that single-chain urokinase plasminogen activator alone or bound to its receptor could prevent these adhesions by their relative resistance to local inhibition by plasminogen activator inhibitors. We found that recombinant human single-chain urokinase-bound rabbit pleural mesothelial cells or lung fibroblasts with kinetics similar to that reported for human cells (kD of approximately 5 nM). The receptor-bound fibrinolysin maintained in vitro fibrinolytic activity in the presence of pleural fluids from rabbits with tetracycline-induced pleural injury over 24 hours. In rabbits given intrapleural single-chain urokinase 24 and 48 hours after intrapleural tetracycline (n = 10 animals), adhesions were prevented, whereas the receptor-complexed form (n = 12) attenuated adhesions versus vehicle/tetracycline-treated rabbits (n = 22, p 0.005 in both cases). There were more adhesions in the complex than the single-chain urokinase group (p = 0.02). Residual antigenic but not functional evidence of the interventional agents remained in pleural fluids at 72 hours after tetracycline. No local or systemic bleeding occurred because of either interventional agent. The data demonstrate that single-chain urokinase inhibits, whereas lysin–receptor complexes attenuate, adhesion formation in tetracycline-induced pleural injury in rabbits.

Key Words: urokinase • fibrinolysis • pleural scarring

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Intrapleural loculation can occur early in the course of complicated parapneumonic effusions or frank empyemas and can adversely affect clinical outcome (1, 2). It has long been appreciated that an initial fibrinopurulent phase occurs as part of the inflammatory response. During this phase, intrapleural fibrin forms and can bridge the visceral and parietal pleural surfaces (1–3). If the inflammatory process persists, the fibrinous exudate undergoes organization, with fibroblast invasion and collagen deposition (2). Clinically, formation of a "pleural peel" and extensive intrapleural organization can encase the lung and impair its function. A similar process of intrapleural fibrin deposition and organization can occasionally occur in association with hemothoraceses, supporting the concept that intrapleural fibrin formation is integral to subsequent loculation and scarring in the pleural compartment (4).

Since the 1940s, it has been appreciated that intrapleural fibrin deposition could be targeted for therapeutic benefit. Tillett and Sherry originally used relatively crude preparations of streptokinase to degrade pleural loculations (5). This fibrinolytic strategy has been refined and remains a viable clinical option for treatment of loculated parapneumonic effusions (1, 2, 6–8). However, the use of intrapleural fibrinolysins can be logistically difficult and expensive in that multiple treatments are often required and hospital stay is often extended, providing the rationale to investigate whether alternative agents could be effective and safe for intrapleural application.

Based on the properties of single-chain urokinase plasminogen activator (scuPA) when bound to urokinase plasminogen activator receptor (uPAR) and our current understanding of the derangements of fibrinolysis in pleural injury, we reasoned that single-chain urokinase plasminogen activator (scuPA or prourokinase) could be particularly useful as an interventional agent for prevention of pleural loculation. When bound to its receptor, uPAR (either on the cell surface or in soluble form), scuPA expresses enhanced and sustained fibrinolytic activity that is also relatively resistant to plasminogen activator inhibitors (PAIs) (9–11). These properties suggest that use of this agent could be of advantage in organizing pleuritis, where the local concentrations of PAI are very high (12). We have previously shown that pleural mesothelial cells and fibroblasts express uPAR and that uPAR is upregulated in these cells by cytokines expressed in pleural injury (13, 14). Thus, we inferred that scuPA would likely bind uPAR at loci of injury where fibrin would likely be proximated. In this study, we evaluated the ability of novel fibrinolytic strategies using either intrapleural recombinant human scuPA or scuPA precomplexed to soluble recombinant uPAR (suPAR) to prevent pleural adhesion formation associated with tetracycline (TCN)-induced pleural injury in rabbits. Although both agents were protective, we found that protection due to scuPA against formation of intrapleural loculations in this model was virtually complete.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Induction of TCN-induced Pleural Injury
Female New Zealand white rabbits weighing 4 kg were used in this study. Briefly, freshly prepared intrapleural TCN with lidocaine (1 mg/ml) was administered under sterile conditions into the right pleural space with the animal in the left lateral decubitus position using an 18 gauge x 2-inch, 2.25-mm ball stainless steel feeding needle inserted through a cutaneous subscapular incision made with a #10 scalpel, as previously described (15). Anesthesia was accomplished using intramuscular Ketamine and Xylazine (15), and all animals were then carefully monitored throughout the perioperative and postoperative periods to assure stability and the absence of overt pain or distress. The animals were killed by administration of intravenous Euthasol after they were anesthetized using Ketamine and Xylazine at 72 hours after intrapleural TCN, at which time they underwent direct inspection of the operative site, screening thoracic and abdominal autopsy, quantification of collection of pleural fluid, and direct assessment of intrapleural adhesion formation. All animal protocols were approved by the Animal Research Committee of The University of Texas Health Center at Tyler and by a veterinarian (W. H.) who periodically monitored the induction of anesthesia, procedures, and perioperative animal welfare.

Preparation of Recombinant Human scuPA and scuPA–scuPAR Complexes
suPAR was expressed in Drosophila S2 cells (DES expression system; Invitrogen, Carlsbad, CA) and purified from S2 culture supernatants as previously described (16). Wild-type human scuPA (amino acids 1–411 of the mature human sequence) was also cloned and expressed using the DES system. suPAR-S2 cells were expanded to a cell density of 20–30 x 10⁶ cell/ml of culture media followed by the induction of expression using 0.5 mM of CuSO₄ (final concentration). Culture supernatants were typically collected 5–7 days after induction. Supernatants were clarified by centrifugation. The pH of the supernatants was adjusted to 8.8 using 1 M of BICINE, pH 9.0, and the media was sterile filtered through a 0.22-µm filter before purification. A column was prepared using SP-Sepharose (5 x 200 cm) and equilibrated with 20 mM of BICINE, pH 8.8, at 4°C, and the filtered supernatant was applied to this column at a flow rate of 2–5 ml/min. Typically, 2–5 L of supernatant were applied to this column at one time. The column was then washed with 20 mM of BICINE buffer, pH 8.8 (5 ml/min flow rate), until protein could no longer be detected in the flow through. A 0- to 0.5-M NaCl gradient was used to elute bound scuPA, which typically eluted between 0.1 and 0.2 M of NaCl with a more than 90% purity. Reverse-phase high-performance liquid chromatography using a semipreparative C₈ column was used to enhance the scuPA to a purity of more than 97%. scuPA prepared in this manner was active in complex with suPAR as well as when activated by plasmin, and the specific activity was estimated to be 150,000 U/mg. S2-derived scuPA also bound to uPAR and suPAR with a kD similar to those reported for scuPA derived from natural and other recombinant sources (data not shown).

Intrapleural Administration of Interventional Agents
At 24 and 48 hours after administration of intrapleural TCN, the animals were sedated with intravenous Domitor 1.5 mg/kg, after which they were placed in the prone position and then subjected to thoracentesis using a 21-gauge 1.25-inch needle on a 3-cc nonluer lock syringe containing a total volume of 1.1 ml of either scuPA, scuPA–suPAR, or the phosphate-buffered saline (PBS) vehicle, pH 7.4. Entry into the pleural space was confirmed by initial aspiration of a small amount of pleural fluid, after which the contents of the syringe were injected into the pleural space. In pilot studies, we determined that this procedure was more practical than imaging of the pleural effusion by intraoperative ultrasonography, followed by direct aspiration. The animals received a dose of 0.5 mg/ml of either interventional agent or a total dose of 2 mg of interventional agent at each of the two intrapleural administrations at 24 and 48 hours after TCN. The duration of sedation approximated 20–30 minutes, and the animals appeared to tolerate the repeated episodes of sedation without distress or apparent change in behavior.

Collection and Processing of Pleural Fluid and Blood Samples
Pleural fluids were immediately collected using a 60-ml plastic syringe after the right and left hemithoraces were inspected. The total pleural fluid volume was measured, and a sample was collected in 0.9% citrate and placed on ice. Samples of the citrated fluids were used to determine the total red and white blood cell counts using a Coulter counter (Beckman Coulter Inc., Hialeah, FL), and slides were prepared for differential white cell counts. Pleural fluid total protein and lactate dehydrogenase determinations were done, as previously described (15). Blood from an ear vein was also collected and placed in citrate immediately before administration of intrapleural TCN and again immediately before sacrifice by administration of intravenous Euthasol. Aliquots were used to determine the red and white cell counts, as well as differential white blood cell analyses.

Incidence of Extrapleural Administration of TCN
Thirteen of 57 rabbits in this study were not evaluable because of extrapleural administration of the TCN. The incidence of misplaced sclerosant was comparable in the three treatment groups. Extrapleural delivery of TCN was confirmed by autopsy analysis. These animals had apparent mediastinal or intrahepatic delivery of the sclerosant. Seven of 29 animals in the TCN–PBS group had extrapleural TCN administration. Six of 28 animals in the interventional groups, 4 in the scuPA group, and 2 in the scuPA–suPAR group had similar extrapleural administration of autopsy-confirmed extrapleural TCN. The proportion of nonevaluable animals did not differ between any of the groups, indicating the absence of statistic bias attributable to the error rate associated with the procedure used to induce pleural injury (p = 0.64).

Postmortem Intrapleural Adhesion Assessment
Pleural adhesions were quantitated by direct inspection of the right hemithorax after removal of all pleural fluid that could be harvested. In cases where florid adhesion formation filled the hemithorax and trapped the lung, the adhesions were designated as too numerous to count. Where adhesion formation was more limited, as was typically the case in the animals that received the interventional agents, the number of discrete adhesions between the visceral and parietal pleural surfaces was counted.

Binding of Human Recombinant scuPA to Cultured Rabbit Pleural Mesothelial Cells and Rabbit Lung Fibroblasts
Rabbit pleural mesothelial cells and lung fibroblasts were isolated from rabbit pleural tissues and were grown to confluence in RPMI medium, as previously described (13). Competition binding studies were performed using these cells and ¹²⁵I scuPA. Briefly, mesothelial cells and fibroblasts were plated in 24-well plates (25,000 cells/well) and cultured overnight at 37°C in 5% CO₂. Cells were briefly exposed to acidic conditions (0.5 M of glycine, 100 nM of NaCl, pH 3.0, for 3 minutes at room temperature) to remove endogenously bound rabbit uPA and then washed three times with PBS buffer. ¹²⁵I scuPA (1 nM) was added to each well either alone or in the presence of increasing concentrations of cold scuPA. After 2 hours of incubation at room temperature, the cells were washed using PBS (three times) followed by dissolution in 1 N NaOH. Each well was then aspirated, and the contents were placed into individual 12 x 75-mm tubes for analysis using a counter.

Lysis of Radiolabeled Clots by scuPA or scuPA–suPAR in the Presence of Rabbit Pleural Fluids
The ability of the interventional agents to lyse fibrin clots in the presence of pleural fluids was assessed by radioassay, as previously reported (17). Human fibrinogen was radiolabeled with ¹²⁵I-labeled fibrinogen, which was added in separate experiments to three randomly selected pleural fluids from rabbits challenged with intrapleural TCN (one collected at 24 hours after TCN and two others at 72 hours after TCN) (final concentration of 60,000 counts per minute per 150 µl). Plasminogen-depleted human fibrinogen was added to a final concentration of 2 mg/ml, CaCl₂ to a final concentration of 10 nM, and human thrombin to a final concentration of 0.2 U, after which the mixtures were immediately placed in polyethylene tubes in aliquots of 150 µL, and the total radioactivity was counted using a counter. Radiolabeled fibrin clots were then allowed to form as the mixtures were incubated at room temperature for 1 hour, with scuPA–suPAR or two-chain uPA (tcuPA)–suPAR complexes by adding equal volumes of 1-µM concentrations of either scuPA or tcuPA with 1-µM suPAR. The complexes were allowed to form by incubation at room temperature for 5 minutes and were then immediately added to a mixture of 150 µL of clot and 300 µL of PBS. Thirty microliters of scuPA, tcuPA, or either scuPA–suPAR or tcuPA–suPAR complexes or control supplemental suPAR alone (each 500 nM) were then added to each tube (final concentration 50 nM). The radioactivity of 10 µL of clot lysate was then measured at 1, 2, 4, 6, 7, 8, 10, and 24 hours using a counter.

Pleural Fluid Coagulation and Fibrinolytic Analyses
Pleural fluid procoagulant activity was determined by its ability to shorten the recalcification time of normal pooled human plasma, as previously reported (18). Antigenic determination of pleural fluid uPA was performed by enzyme-linked immunosorbent assay (American Diagnostics #894, Greenwich, CT), which recognizes human free uPA, uPA complexed to PAI, or uPA in complex with uPAR equivalently. Antigenic determination of soluble uPAR was likewise determined by enzyme-linked immunosorbent assay (American Diagnostics, #893). Pleural fluid and plasma D dimer was determined by a rapid latex agglutination slide test using mouse monoclonal antibodies (Diagnostica Stago, Asnieres-sur-Seine, France), as previously reported (18). Pleural fluid plasminogen activator activity was assessed using two independent methods. The plasminogen activator activity of pleural fluid samples (four TCN–PBS, one TCN–scuPA, and five TCN scuPA–suPAR-treated animals) was first assayed based on amidolytic activity using the Spectrolyse plasminogen activator/PAI activity assay kit (American Diagnostica #452). We also performed fibrin enzymography to evaluate the presence of functional uPA in pleural fluids and reverse fibrin enzymography to evaluate the presence of functional PAI activity as previously described (18).

Statistics
Nonparametric analyses of variables, including pleural fluid variables and adhesions, were chosen, as these variables are not known to be normally distributed. Intragroup variation was first determined by Kruskal Wallis one-way analysis of variance. Specific comparisons of variables between groups were performed using the Mann Whitney U-Wilcoxon rank sum W test. Fisher's exact probability was used to compare the variables for selected two-group comparisons.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Cleavage of Fibrin Clots by Interventional Agents in the Presence of Rabbit Pleural Fluids
Of the agents tested, only scuPA–suPAR complexes demonstrated ability to lyse fibrin clots at early (6 hours) as well as at late (20 hours) intervals in the presence of pleural fluids from TCN-treated rabbits (Figure 1) . As expected, scuPA or tcuPA in the presence of suPAR increased the release of radiolabeled fibrin compared with suPAR alone used as a control. Cleavage of radiolabeled fibrin by tcuPA complexed to suPAR was not detectable early (6 hours) but was detectable at 20 hours after its addition to radiolabeled clot in the presence of rabbit pleural fluid.

View larger version (32K): [in this window] [in a new window]   Figure 1. Radiolabeled clot lysis of selected fibrinolysins in the presence of pleural fluids from rabbits with TCN-induced pleuritis and pleural scarring. Net lysis refers to the radiolabeled fibrin released into the reaction mixture in the presence of pleural fluids from rabbits with TCN-induced pleural injury. cpm = counts per minute; scuPA/suPAR = scuPA complexed with solubilized uPA receptor. The ordinate of A is applicable to the data presented in both A and B. The data in A (dark bars) represent clot lysis achieved over 6 hours. The data in B (gray bars) represent clot lysis achieved by the selected agents over 20 hours.

Detection of Interventional Agents in Pleural Fluids
We confirmed the intrapleural delivery of the interventional agents by direct aspiration of pleural fluid immediately before injection of the agents or control vehicle. Our ability to aspirate fluids at 24 and 48 hours after intrapleural TCN confirms our previously reported findings in this model: that pleural fluids begin to form by 24 hours and then enlarge with extensive formation of adhesions by 72 hours after TCN (15, 18). We assayed 21 of 22 pleural fluids from the TCN–PBS-treated control animals for the presence of antigenic human uPA and soluble uPAR and could not detect either protein. We also tested the pleural fluids of three naïve animals that received intrapleural TCN but no subsequent intrapleural injections and likewise found no detectable levels of uPA or uPAR. All 10 of the scuPA-treated animals had detectable levels of this agent (median, 11 ng/ml; range, 3–31 ng/ml). Eleven of 12 of the scuPA–suPAR-treated animals likewise contained detectable amounts of uPA, likely reflecting the reported ability of the assay to detect human uPA bound to uPAR (median, 17 ng/ml; range, 2–75 ng/ml). Soluble uPAR was detectable in 11 of 12 pleural fluid samples from the suPA–suPAR-treated animals (median, 21 ng/ml; range, 0–24 ng/ml). The data therefore confirm that the recombinant human interventional agents could be detected in the appropriate interventional groups at least 24 hours after the last intrapleural dose was given and 48 hours after TCN.

We could not detect PA activity by amidolytic assay in the TCN–PBS pleural fluids: the TCN–scuPA or TCN–scuPA–suPAR fluids using this technique. By reverse fibrin enzymography, we found that these pleural fluids contained a zone of PAI activity that could be neutralized by an antibody to PAI-1, confirming the presence of this inhibitor within these fluids (data not shown), as previously reported in this model (18). By fibrin enzymography, we found that most of the fibrinolytic zones related to uPA or tissue plasminogen activator (tPA) bound to PAI, as previously reported (18). Plasminogen activator activity, which comigrated with the human uPA standard, was detected in only two of nine randomly selected TCN–PBS pleural fluid samples, in one of nine TCN–scuPA samples, and in three of nine TCN–scuPA–suPAR samples (data not shown). D-dimer concentrations were significantly increased in pleural fluids of scuPA or scuPA–suPAR-treated rabbits versus those of rabbits treated with intrapleural vehicle (Figure 2) , suggesting that local fibrinolytic activity was at some point increased in the pleural fluids of rabbits treated with either interventional agent.

View larger version (25K): [in this window] [in a new window]   Figure 2. D-dimer concentrations in pleural fluids of scuPA, scuPA–suPAR, and PBS vehicle-treated rabbits with TCN-induced pleural injury. Box plots illustrate the findings, and the vertical lines within the boxes illustrate median values. The boxes represent the inner quartile range, 50% of the data around the median value. Outlying values are indicated as individual data points. The concentrations are illustrated in µg/ml on the ordinate.

View larger version (35K): [in this window] [in a new window]   Figure 3. Numbers of adhesions in PBS, scuPA, and scuPA-suPAR–treated rabbits.

In the TCN–scuPA–suPAR group, 5 of 12 animals had no adhesions in the right pleural space; 6 of 12 animals in this group had one to five discrete adhesions. One animal in this group had extensive adhesions, resembling the response typically seen in the TCN–PBS-treated animals. The TCN–scuPA–suPAR group had significantly fewer than the control TCN–PBS group (p = 0.004) but more adhesions than the TCN–scuPA group (p = 0.02). When both interventional groups combined were compared with the control group, the TCN–PBS controls were found to have more adhesions (p < 0.0001).

In the PBS–TCN group, the right lung was in all cases restricted from full inflation by the intrapleural adhesions and was dusky (Figure E1). The contralateral (left) lung in all cases was pink and fully expanded. In animals treated with the intrapleural interventional agents, the right lung was often of a purplish hue, even in the absence of extensive adhesions (data not shown), likely reflecting the presence of a chemical pneumonitis induced by TCN and atelectasis, as previously reported (20).

Pleural Effusion Volumes
Right pleural effusions ipsilateral to the previously administered intrapleural TCN were identified in all animals at 72 hours after induction of pleural injury. These effusions generally appeared to be sanguinous and were carefully suctioned from the right hemithorax so as to permit assessment of the extent of visceral–parietal adhesion formation, as noted previously here. Rarely, small serous left pleural fluid collections occurred in association with the large right pleural effusions. For all groups combined, the median volume of right pleural fluid at 72 hours after TCN was 34 ml (range, 10–53 ml), comparable to our previously reported observations (15), and there were no significant differences versus the interventional groups (see Table E1 in the online data supplement).

Pleural Fluid Cell Counts and Biochemical Parameters
The pleural fluid total protein and lactate dehydrogenase concentrations were measured as general indices of the local inflammatory response (Table E1). The pleural fluid total protein concentration was nearly identical in all groups: the TCN–PBS median was 41.5 mg/ml versus a TCN–scuPA of 42.6 mg/ml and a TCN–scuPA–suPAR of 43.3 mg/ml; there was not any difference in total protein concentrations between the TCN–PBS and naïve controls (p = 0.23). There were likewise no significant differences in the concentrations of pleural fluid lactate dehydrogenase between the groups (p = 0.37) nor between the TCN–PBS group and naïve controls (p = 0.53) (see Table E1).

The median pleural fluid total white blood cell count was 2.64 x 10⁶ cells/ml in the TCN–PBS group, 6.07 x 10⁶ cells/ml in the TCN–scuPA group, and 4.42 x 10⁶ cells/ml in the TCN–scuPA–suPAR group, values comparable to those that we previously reported in the model (20). The total WBC counts were significantly lower in the TCN–PBS group (p = 0.04), but there was no difference between the other interventional groups (see Table E1 in the online data supplement). There were also no significant differences between the TCN–PBS group and the naïve controls (p = 0.15). The total red cell counts in the three groups did not differ significantly (p = 0.73), attesting to the absence of intrapleural bleeding caused by local administration of the fibrinolytic agents (Figure E2A). There were also no significant differences in the pleural fluid red blood cell counts in the TCN–PBS group and the naïve controls (p = 0.71), suggesting that the additional thoracenteses to which all experimental groups were subjected did not cause appreciable local bleeding. On differential WBC analyses, the percentage of neutrophils (polymorphonuclear cells) observed in the TCN–PBS group (median, 12%; range, 4–46%) was significantly lower than that of the other groups (p = 0.01) (see Table E1 in the online data supplement). The total numbers of neutrophils were likewise greater in the pleural fluids of either interventional group compared with the PBS controls (see Table E2 in the online data supplement).

Pleural Fluid Recalcification Times
We have previously shown that the pleural fluids in this model contain several procoagulants, including tissue factor (15). We therefore chose to analyze the aggregate procoagulant activity of the pleural fluids by measurement of the recalcification times, as we previously reported in this model (18). There were no significant differences in pleural fluid recalcification times (p = 0.15) (see Table E1 in the online data supplement).

Peripheral Blood Cell Counts before and after Pleural Injury
When all groups were compared, the peripheral red blood cell counts measured before induction of pleural injury were significantly lower in the TCN–scuPA–suPAR group (median, 4.13 x 10⁹; range, 3.87–4.27 x 10⁹ cells) versus that of the TCN–PBS (median, 4.47 x 10⁹, 3.55–4.91 x 10⁹ cells) or TCN–scuPA-treated animals (median, 4.42 x 10⁹, 3.89–4.85 x 10⁹ cells) (see Figure E2 in the online data supplement). After injury, the peripheral red blood cell count rose at 72 hours when compared with baseline levels (p < 0.0001, all groups were considered). The data indicate that neither of the intrapleural fibrinolysins caused detectable systemic blood loss or acute anemia.

There was no significant difference between the peripheral white blood cell counts (range of medians 5.04–6.87 x 10⁶ cells/ml of the groups either before or after injury, p = 0.06 and p = 0.52, respectively). Although the percentage of polymorphonuclear cells on differential white blood cell counts in the TCN–PBS group (median, 23; range, 6–54%) was significantly lower before injury versus the TCN–scuPA (30, 13–43%) and TCN–scuPA–suPAR groups (44, 20–61%), there was no difference in the percentage of polymorphonuclear cells after injury in any of the groups (p = 0.69).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
It appears that the pathogenesis of pleural loculation recapitulates that of wound healing, in which a progression of extravascular fibrin deposition and remodeling of transitional fibrin leads to fibrotic repair and scar formation (21). It is widely accepted that a phase of fibrinous adhesion formation initiates intrapleural loculation and fibrosis (1, 2). Experimental evidence supports this clinical impression. Morphologic evidence from TCN-induced pleural injury in rabbits directly links transitional extravascular fibrin deposition to the pathogenesis of intrapleural loculation after acute pleural injury. In this model, the initial injury is characterized by intrapleural fibrin, which persists through the first 72 hours of evolving intrapleural loculation (15, 18). At 72 hours after induction of pleural injury, fibrinous intrapleural loculation is characteristically florid, bridging the visceral and parietal pleural surfaces (15). The visceral–parietal strands demonstrate collagen within the strands by 72 hours after administration of intrapleural TCN (20). These observations strongly implicate intrapleural transitional fibrin in the development of intrapleural adhesions after acute injury induced by TCN.

In previous studies, we found that administration of intrapleural heparin or the no longer commercially available low molecular weight uPA Abbokinase (Abbott Laboratories, North Chicago, IL) could attenuate intrapleural adhesion formation induced by intrapleural administration of TCN in rabbits. Multiple doses of heparin or intrapleural low molecular weight uPA every 12 hours over 72 hours were required to achieve partial protection against formation of intrapleural adhesions in these animals (18). This study provides proof of principle that disruption of intrapleural fibrin could influence the subsequent development of intrapleural adhesion formation. Because protection provided by low molecular weight uPA was incomplete, we reasoned that alternative approaches should be explored.

A long chronology of clinical experience further supports the concept that intrapleural fibrin is essential to pleural organization and fibrosis. More than 50 years ago, Tillett and Sherry demonstrated that relatively crude preparations of fibrinolytic proteases could be used to clear intrapleural loculations associated with parapneumonic effusions and hemothoraces (5). The use of fibrinolytic agents is now considered to be a reasonable therapeutic option for selected patients with advanced loculations based on recommendations of a recent consensus conference (8).

On the other hand, there is ongoing debate about the place of currently available fibrinolytic interventions for intrapleural loculation. In a recent systemic review of all available randomized controlled trials, the pooled data showed that small benefits of fibrinolytic intervention could be anticipated in terms of reduction of hospital stay, time to defervescence, radiographic improvement, and ultimate need for surgical intervention (22). In another recent retrospective analysis, streptokinase was found to increase the volume of pleural fluid drained from patients with organized empyemas, but there was no significant impact on hospital stay, defervescence, need for surgical intervention, or mortality (23). These observations provide a clinically based rationale for the investigation of new interventional fibrinolysins for intrapleural use.

The injured pleural space is an inhibitor-rich environment, which limits expression of endogenous plasminogen activator activity. Pleural fluid fibrinolytic activity is generally depressed in exudative pleural effusions, and in this respect recapitulates the response observed in alveolar lining fluids in acute respiratory distress syndrome (24). For example, there is almost always no detectable fibrinolytic activity in pleural fluids from patients with parapneumonic effusions and empyema, and procoagulant activity is concurrently augmented (12). These conditions favor the deposition and maintenance of intrapleural fibrin in the setting of high-grade local inflammation. Although the pleural fluids contain both tPA and uPA, most of the immunoreactive plasminogen activator is bound and irreversibly inhibited by PAI, mainly PAI-1 (12). Fibrinolytic activity is likewise undetectable in pleural effusions that form after TCN-induced pleuritis in rabbits attributable to the identical expression of PAI-1 and antiplasmins (18). These circumstances likely limit the fibrinolytic activity of the currently used fibrinolysins.

Based on these observations, we reasoned that scuPA or scuPA–suPAR complexes might be particularly effective agents for interventional intrapleural use. First, scuPA bound to its receptor is relatively resistant to PAI-1 (9), a property that could be of advantage in the setting of pleural inflammation. scuPA appears to resist formation of irreversible covalent bonds with PAI-1, which thereby preserves its PA activity (11). Second, when associated with its receptor, uPAR, scuPA remains as a single-chain molecule that expresses enzymatic activity comparable to that of tcuPA and exhibits less susceptibility to inhibition by PAI-1 (10). Finally, fibrin clots contain PAI-1 derived from plasma and platelets, and scuPA complexed to suPAR has been shown to be a potent fibrinolysin. scuPA complexed to suPAR was found to lyse fibrin clots more efficiently than equimolar amounts of scuPA alone, tcuPA, or tcuPA bound to suPAR (17, 25).

We found that scuPA alone appeared to be a more effective intrapleural fibrinolysin than scuPA–suPAR complexes. The findings are consistent with the clot lysis data, in which scuPA complexed to its receptor was superior to scuPA alone, indicating that receptor-bound scuPA demonstrated relatively more fibrinolytic activity in the presence of rabbit pleural fluids containing increased quantities of PAI-1. The apparent superiority of scuPA–suPAR complexes versus scuPA in the in vitro system reflects the paucity of uPAR to which scuPA alone could bind, whereas intrapleurally administered scuPA could bind the abundant endogenous receptor. It is likely that scuPA bound to endogenous uPAR expressed at loci of pleural injury induced by TCN. Supporting this inference, we previously showed that primary cultures of lung fibroblasts and pleural mesothelial cells harvested from rabbits with pleural injury induced by TCN expressed uPAR (13). We also found that this receptor is upregulated by stimuli that have been implicated in the pathogenesis of pleural injury and fibrosis, including asbestos and the cytokines tumor necrosis factor-, and transforming growth factor-ß (13, 26–28). In addition, the recombinant human scuPA that we used bound uPAR at the surface of these rabbit cells with the same affinity as to human cells (19). The presence of detectable levels of human uPA related antigen in the fluids of rabbits treated with scuPA alone or in complex with suPAR 24 hours after intrapleural administration further suggests that the agents remained localized in the pleural space. Although we could not detect sufficient amounts of the proteins to determine the fates of either intrapleural scuPA or the scuPA–suPAR complexes, the prevention of adhesion formation suggests the agents were functional in vivo. Our inability to demonstrate consistently incremental uPA activity in the fluids by fibrin enzymography further suggests the likelihood that the scuPA acted, as expected, by receptor binding along the pleural surface rather than in fluid phase. Improved protection afforded by scuPA alone suggests the possibility that binding to endogenous receptor may be preferable to administration of preformed scuPA–suPAR complexes, in this setting where endogenous uPAR are increased at sites of pleural injury and could localize scuPA to preferred sites of action. Alternatively, the apparent superiority of the scuPA alone could reflect the twofold molar relative excess of scuPA alone versus that in complex, as both agents in this trial were administered at the same intrapleural dosage, and the suPAR moiety accounted for half of the administered complex.

Interestingly, the total number of neutrophils was increased in the pleural fluids of the scuPA and scuPA + suPAR interventional groups versus the vehicle controls (see Table E2). The numbers of neutrophils in peripheral blood of the scuPA and scuPA + suPAR groups were increased before but not after administration of the intrapleural fibrinolysins. The observations indicate that the primary alteration of neutrophil trafficking was the local increase in the injured pleural compartment in the presence of scuPA or scuPA + suPAR. Although the pleural neutrophilia could reflect a deleterious augmentation of the inflammatory response by these agents, the alternative possibility—that the increment of neutrophils was protective—is likewise tenable. If increased release of neutrophil proteases accompanied the increased neutrophil load, the proteases could have helped degrade forming adhesions. Alternatively, such proteases might accelerate local proteolysis of the interventional agents. Apart from potential complexity of their effects on intrapleural neutrophil traffic and local inflammation, it remains clear that the interventional agents exerted a salutary effect regarding the endpoint of intrapleural adhesion formation.

In summary, we found that scuPA and scuPA precomplexed to suPAR protected against development of intrapleural loculations in rabbits with TCN-induced pleural injury. Of the two regimens we used, intrapleural scuPA alone provided nearly complete protection from adhesion formation. The treatments were well-tolerated, and there was no increased incidence of local or systemic bleeding. We anticipated that this would be the case based on previous reports showing that intrapleural administration of fibrinolysins does not generally cause systemic activation of fibrinolysis or bleeding (29, 30). Although this application of scuPA or scuPA complexed to suPAR is, to our knowledge, novel, scuPA has previously been used in clinical practice as treatment for cerebral occlusions associated with cerebrovascular accidents (31). Future clinical trials may be warranted if confirmatory results derive from further preclinical trials of scuPA or scuPA–suPAR in other relevant models of pleural loculation.

	FOOTNOTES

 
Supported by National Institutes of Health RO-145,018 (S.I.) and the Temple Endowed Chair in Pulmonary Fibrosis (S.I.).

This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Received in original form April 12, 2002; accepted in final form May 29, 2002

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 

1. Sahn S. The pleura: state of the art. Am Rev Respir Dis 1988;138:184–234.[Medline]
2. Light RW. Parapneumonic effusions: pleural diseases, 4th ed. Philadelphia: Lippincott Williams and Wilkins; 2001. p. 151–181.
3. Sahn SA. Pleural disease: pulmonary medicine and critical care MKSAP, 2nd ed. Philadelphia: American College of Physicians and American Society of Internal Medicine; 1998. p. 262–283.
4. Light RW. Hemothorax: pleural diseases, 4th ed. Philadelphia: Lippincott, Williams, and Wilkins; 2001. p. 320–326.
5. Tillett WS, Sherry S. The effect in patients of streptococcal fibrinolysin (streptokinase) and streptococcal desoxyribonuclease on fibrinous, purulent and sanguinous pleural exudations. J Clin Invest 1949;28:173–190.
6. Sahn SA. Use of fibrinolytic agents in the management of complicated parapneumonic effusions and empyemas. Thorax 1998;53:S65–S72.[Abstract]
7. Light RW, Rodriguez RM. Management of parapneumonic effusions. Clin Chest Med 1998;19:373–382.[CrossRef][Medline]
8. Colice GL, Curtis A, Deslauriers J, Heffner J, Light R, Littenberg B, Sahn S, Weinstein RA, Yusen RD. Medical and surgical treatment of parapneumonic effusions: ACP consensus statement. Chest 2000;18:1158–1171.
9. Higazi AA, Mazar A, Wang J, Reilly R, Henkin J, Kniss D, Cines D. Single-chain urokinase-type plasminogen activator bound to its receptor is relatively resistant to plasminogen activator inhibitor type 1. Blood 1996;87:3545–3549.[Abstract/Free Full Text]
10. Higazi A. Enhancement of the enzymatic activity of single-chain urokinase plasminogen activator by soluble urokinase receptor. J Biol Chem 1995;270:17375–17380.[Abstract/Free Full Text]
11. Manchanda N. Interaction of single-chain urokinase and plasminogen activator inhibitor type 1. J Biol Chem 1995;270:20032–20035.[Abstract/Free Full Text]
12. Idell S, Girard W, Koenig KB, McLarty J, Fair DS. Abnormalities of pathways of fibrin turnover in the human pleural space. Am Rev Respir Dis 1991;144:187–194.[Medline]
13. Shetty S, Idell S. A urokinase receptor mRNA binding protein from rabbit lung fibroblasts and mesothelial cells. Am J Physiol 1998;274:L871–L882.[Abstract/Free Full Text]
14. Shetty S, Kumar A, Johnson AR, Pueblitz S, Holiday D, Raghu G, Idell S. Differential expression of the urokinase receptor in fibroblasts from normal and fibrotic human lungs. Am J Respir Cell Mol Biol 1996;15:78–87.[Abstract]
15. Idell S, Pendurthi U, Pueblitz S, Koenig K, Williams T, Rao LV. Tissue factor pathway inhibitor in tetracycline-induced pleuritis in rabbits. Thromb Haemost 1998;79:649–655.[Medline]
16. Tarui T, Mazar AP, Cines DB, Takada Y. Urokinase-type plasminogen activator receptor (CD87) is a ligand for integrins and mediates cell-cell interactions. J Biol Chem 2001;276:3983–3990.[Abstract/Free Full Text]
17. Higazi AA, Bdeir K, Hiss E, Arad S, Kuo A, Barghouti I, Cines DB. Lysis of plasma clots by urokinase soluble urokinase receptor complexes. Blood 1998;92:2075–2083.[Abstract/Free Full Text]
18. Strange C, Baumann MH, Sahn SA, Idell S. Effects of intrapleural heparin or urokinase on the extent of tetracycline-induced pleural disease. Am J Respir Crit Care Med 1995;151:508–515.[Abstract]
19. Mazar AP, Henkin J, Goldfarb RH. The urokinase plasminogen activator system in cancer: implications for tumor angiogenesis and metastasis. Angiogenesis 1999;3:15–32.[CrossRef][Medline]
20. Miller EJ, Kajikawa O, Pueblitz S, Koenig KK, Idell S. Chemokine involvement in tetracycline-induced pleuritis. Eur Respir J 2000;14:1387–1393.
21. Dvorak HF. Tumors: wounds that do not heal: similarities between tumor stroma generation and wound healing. N Engl J Med 1986;315:1650–1659.[Medline]
22. Cameron R. Intrapleural fibrinolytic therapy vs. conservative management in the treatment of parapneumonic effusions and empyema. Cochrane Database Syst Rev 2000;3:CD002312.
23. de Souza A, Offner PJ, Moore EE, Biffl WL, Haenel JB, Franciose RJ, Burch JM. Optimal management of complicated empyema. Am J Surg 2000;180:507–511.[CrossRef][Medline]
24. Idell S. Extravascular coagulation and fibrin deposition in acute lung injury. New Horiz 1994;2:566–574.[Medline]
25. Bdeir K, Muciano J-C, Tomaszewski J, Koniares L, Martinez J, Cines DB. Urokinase contributes to fibrinolysis in the pulmonary microcirculation. Blood 2000;96:1820–1826.[Abstract/Free Full Text]
26. Perkins RS, Broaddus CV, Shetty S, Hamlton S, Idell S. Asbestos upregulates expression of the urokinase plasminogen activator receptor on mesothelial cells. Am J Respir Cell Mol Biol 1999;21:637–646.[Abstract/Free Full Text]
27. Antony VB, Hott JW, Kunkel SL, Godbey SW, Burdick MD, Strieter RM. Pleural mesothelial cell expression of C-C (monocyte chemotactic peptide) and C-X-C (interleukin 8) chemokines. Am J Respir Cell Mol Biol 1995;12:581–588.[Abstract]
28. Idell S. Coagulation, fibrinolysis and fibrin deposition in lung injury and repair. In: Phan SH, Thrall RS, editors. Pulmonary fibrosis. New York: Marcel Dekker; 1995. p. 743–776.
29. Davies CWH, Lok S, Davies RJO. The systemic fibrinolytic activity of intrapleural streptokinase. Am J Respir Crit Care Med 1998;157:328–330.[Abstract/Free Full Text]
30. Sahn SA. Use of fibrinolytic agents in the management of complicated parapneumonic effusions and empyemas. Thorax 1998;53:S65–S72.
31. Brott T, Bogousslavsky J. Treatment of acute ischemic stroke. N Engl J Med 2000;343:710–722.[Free Full Text]

This article has been cited by other articles:

				S. Idell, T. Allen, S. Chen, K. Koenig, A. Mazar, and A. Azghani Intrapleural activation, processing, efficacy, and duration of protection of single-chain urokinase in evolving tetracycline-induced pleural injury in rabbits Am J Physiol Lung Cell Mol Physiol, January 1, 2007; 292(1): L25 - L32. [Abstract] [Full Text] [PDF]

				J. E. Heffner Multicenter Trials of Treatment for Empyema -- After All These Years N. Engl. J. Med., March 3, 2005; 352(9): 926 - 928. [Full Text] [PDF]

				Y. C. G. Lee Ongoing Search for Effective Intrapleural Therapy for Empyema: Is Streptokinase the Answer? Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 1 - 2. [Full Text] [PDF]

				L. Erkan, S. Fyndyk, O. Uzun, A. G. Atycy, and R. W. Light A New Radiologic Appearance of Pulmonary Thromboembolism: Multiloculated Pleural Effusions Chest, July 1, 2004; 126(1): 298 - 302. [Abstract] [Full Text] [PDF]

				A. H. Diacon, J. Theron, M. M. Schuurmans, B. W. Van de Wal, and C. T. Bolliger Intrapleural Streptokinase for Empyema and Complicated Parapneumonic Effusions Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 49 - 53. [Abstract] [Full Text] [PDF]

				S. Idell The Matrix Unloaded: Aerosolized Heparin or Urokinase for Pulmonary Fibrosis Am. J. Respir. Crit. Care Med., December 1, 2003; 168(11): 1268 - 1269. [Full Text] [PDF]

				A Jaffe and G Cohen Thoracic empyema Arch. Dis. Child., October 1, 2003; 88(10): 839 - 841. [Full Text] [PDF]

				M. J. Tobin Chronic Obstructive Pulmonary Disease, Pollution, Pulmonary Vascular Disease, Transplantation, Pleural Disease, and Lung Cancer in AJRCCM 2002 Am. J. Respir. Crit. Care Med., February 1, 2003; 167(3): 356 - 370. [Full Text] [PDF]

				V. B. Antony Fibrinolysis in the Pleural Space: Breaking the Bonds That Bind Am. J. Respir. Crit. Care Med., October 1, 2002; 166(7): 909 - 910. [Full Text]

This Article Abstract Full Text (PDF) Online Data Supplement Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Idell, S. Articles by Miller, E. Search for Related Content PubMed PubMed Citation Articles by Idell, S. Articles by Miller, E.