The effect of hyaluronan on elastic fiber injury in vitro and elastase-induced airspace enlargement in vivo.

This laboratory has previously described a method of preventing air-space enlargement in experimental pulmonary emphysema using aerosolized hyaluronan (HA). Although it was found that HA preferentially binds to elastic fibers (which undergo breakdown by elastases in emphysema), it remains to be shown that such attachment actually prevents damage to the fibers. In the current study, cell-free radiolabeled extracellular matrices, derived from rat pleural mesothelial cells, were used to test the ability of low molecular weight ( approximately 100 kDa) streptococcal HA to prevent elastolysis. Coating the matrices with HA significantly decreased elastolysis (P<0.05) induced by porcine pancreatic elastase (43%), human neutrophil elastase (53%), and human macrophage metalloelastase (80%). Concomitant in vivo studies examined the ability of an aerosol preparation of the streptococcal HA to prevent experimental emphysema induced by intratracheal administration of porcine pancreatic elastase. As seen with earlier studies involving bovine tracheal HA, a single aerosol exposure significantly decreased elastase-induced airspace enlargement, as measured by the mean linear intercept (107.5 vs 89.6 microm; P < 0. 05). Furthermore, repeated exposure to the HA aerosol for 1 month did not reveal any morphological changes in the lung. The results provide further evidence that aerosolized HA may be an effective means of preventing pulmonary emphysema and perhaps other lung diseases that involve elastic fiber injury.

The pulmonary matrix, glycosaminoglycans and pulmonary emphysema.

This paper reviews recent evidence of the effect of intratracheal hyaluronan (HA) to limit the induction of experimental emphysema in hamsters. Experimental emphysema was induced by both neutrophil and pancreatic elastase instilled intratracheally. Emphysema was quantified anatomically by measurement of alveolar mean linear intercept. Hyaluronidase, instilled intratracheally, enhanced the induction of experimental emphysema. Air-space size measured one week after intratracheal instillation of elastase showed that administration of 1 mg HA immediately following elastase administration resulted in a marked reduction in air-space enlargement (82 microM vs 122 microM, p < 0.01). Similarly, animals given either 1 or 2 mg HA 2 h before elastase or 2mg HA 1 h after elastase showed a significant decrease in air-space enlargement compared to controls (96 microM, 88 microM vs 120 microM and 66 microM vs 104 microM, respectively; p < 0.05. Experimental emphysema induced by neutrophil elastase was also limited by the administration of 1 or 4 mg of HA, administered 2 h prior to elastase (57 and 59 microM, respectively vs 64 for controls, p < 0.05). Characterization of administered HA showed a mean molecular weight of 104,800 Da, less than 5% protein and a uronic acid/hexosamine ratio of 1, which is characteristic of HA. Studies using fluorescein-labeled hyaluronan (HA) showed fluorescence associated with interstitial, pleural and vascular elastic fibers. The mechanism of attachment of the administered HA to elastin remains unknown. Fluorescein labeling of elastin was visible for at least 4 h post-instillation. These studies indicate a protective effect of hyaluronan against elastase degradation of pulmonary elastin in vivo by both pancreatic and neutrophil elastases. The anatomical studies further suggest a mechanism of protective coating of hyaluronan which may limit access to pulmonary elastin from neutrophils and alveolar macrophages. Results also suggest that a reduction in pulmonary hyaluronan content increases the susceptibility of elastin to degradation by elastases. These studies provide evidence for an antielastase effect of hyaluronan which is not dependent upon enzyme inhibition but on anatomical protection of pulmonary elastin by other mechanisms.

Aerosolized hyaluronic acid decreases alveolar injury induced by human neutrophil elastase.

This laboratory has previously shown that an intratracheally instilled solution of hyaluronic acid (HA) protects the lung from elastase-induced airspace enlargement. In those studies, fluorescein-labeled HA was found to bind preferentially to lung elastic fibers, suggesting a mechanism for the protective effect. The current investigation extends these findings by examining the capacity of an aerosol preparation of HA to similarly inhibit elastase-induced lung injury. Syrian hamsters were exposed to aerosolized bovine tracheal HA (0.1% solution in water) for either 25 or 50 min, then immediately instilled intratracheally with 80 units of human neutrophil elastase. One week later the lungs were examined for airspace enlargement, using the mean linear intercept method. Animals exposed to HA for 50 min showed a significant decrease in airspace enlargement compared to controls exposed to aerosolized water alone (68.2 microm vs 85.9 microm; P < 0.05). The 25-min exposure to the HA aerosol also reduced the mean linear intercept compared to controls (73.7 microm vs 85.9 microm), but this decrease was not statistically significant. With regard to possible inflammatory effects of HA, there was no difference in the percentage of lavaged neutrophils between HA-treated and control lungs at 24 hr (1.4% vs 1.8%, respectively). As with earlier experiments using intratracheally instilled HA, aerosolized fluorescein-labeled HA was found to bind to lung elastic fibers. These results suggest that aerosolized HA may prevent elastase-mediated injury in pulmonary emphysema.

Further investigation of the use of intratracheally administered hyaluronic acid to ameliorate elastase-induced emphysema.

Previously, this laboratory has shown that intratracheally administered hyaluronic acid (HA) significantly reduces air-space enlargement in a hamster model of emphysema induced with pancreatic elastase. Whereas HA was given immediately following elastase in those initial studies, the current investigation determined the effect of instilling HA up to 2 h before or after intratracheal administration of elastase to hamsters. Both 1 and 2 mg HA, given 2 h before pancreatic elastase, significantly decreased (p < .05) air-space enlargement compared to controls (as measured by the mean linear intercept). Instillment of 2 mg HA, 1 h after pancreatic elastase, had a similar effect (p < .05). In contrast, 1 mg HA, given 1 or 2 h after pancreatic elastase, did not significantly affect the mean linear intercept. Against human neutrophil elastase, HA exhibited the same protective effect. While neutrophil elastase induced less air-space enlargement than pancreatic elastase, both 1 and 4 mg of HA, given 2 h prior to the enzyme, still produced a significant reduction (p < .05) in the mean linear intercept. HA exerted this effect despite the fact that it initiates a transient influx of neutrophils into the lung. Since HA does not slow the clearance of intratracheally instilled [14C] albumin from the lung, its mechanism of action may not involve physical interference with the movement of elastase through the lung, but may instead depend on interaction with elastic fibers. Evidence for an association between these two matrix constituents was provided by studies using fluorescein-labeled HA. Overall, these results further suggest that HA may be useful in preventing lung injury by elastases.

Modulation of airspace enlargement in elastase-induced emphysema by intratracheal instillment of hyaluronidase and hyaluronic acid.

The study examined how lung hyaluronic acid content influences airspace enlargement in elastase-induced emphysema. To determine the effect of a decrease in hyaluronic acid, hamsters received a single intratracheal instillment of hyaluronidase 24 h prior to administration of pancreatic elastase by the same route. One week later, these animals showed significantly greater airspace enlargement than controls sequentially instilled with saline and elastase (128 vs. 100 microns; p < .05). Conversely, intratracheal administration of hyaluronic acid immediately after elastase instillment resulted in a marked decrease in airspace enlargement at 1 week compared to controls receiving elastase followed by saline (82 vs. 122 microns; p = .005). Since hyaluronic acid has no elastase inhibitory capacity, its effect may involve extracellular matrix interactions not directly related to elastic fiber breakdown. This concept is supported by the finding that animals treated with hyaluronidase and elastase showed no greater loss of lung elastin than that observed in the saline/elastase control group, despite demonstrating a marked increase in airspace enlargement. Further work is needed to determine how hyaluronic acid influences airspace enlargement and to evaluate the potential use of this substance as a treatment for emphysema.

Pulmonary air-space enlargement induced by intratracheal instillment of hyaluronidase and concomitant exposure to 60% oxygen.

Although emphysema is generally characterized by damage to pulmonary elastic fibers, the causes of such injury appear to be complex and are not entirely explained by a singular imbalance between elastases and their inhibitors. Other factors could compromise elastic fiber integrity. To test the validity of this argument, hamsters were instilled intratracheally with a nonelastolytic enzyme, hyaluronidase (which reduces lung hexuronic acid content by 21% after 24 h), then exposed to an otherwise nontoxic concentration of oxygen (60%) for 4 days. Additional groups were given (1) hyaluronidase and room air, (2) saline and 60% oxygen, and (3) saline and room air. Treatment with both hyaluronidase and 60% oxygen resulted in a significant increase in air-space enlargement at 4 days (67.1 vs. 57.9 microns for saline/room air controls; p < .05), which was accompanied by only minimal inflammatory changes, as determined by both light microscopy and lavage cytology. Animals receiving either hyaluronidase or 60% oxygen alone showed no significant increases in air-space size compared to those given saline and exposed to room air. While the mechanisms responsible for these results are unclear, the marked increase in radiolabeling of lung elastin cross-links (desmosine and isodesmosine) in animals receiving both hyaluronidase and 60% oxygen (429 vs. 168 cpm/g dry lung for saline/room air controls; p < .05), as well as a significant decrease in total lung desmosine and isodesmosine (32.5 vs. 37.7 micrograms/lung for saline/room air controls; p < .05), suggests that elastic fiber damage is a potential factor. Moreover, only those animals receiving both hyaluronidase and 60% oxygen showed a significant rise in cell-free elastase activity in lavage fluids compared to saline/room air controls (83.3 vs. 48.3 ng; p < .05). On the basis of these findings, it is concluded that while elastic fiber damage may be a common pathway in emphysema, the factors that initiate the disease may be more varied than previously suspected and not always related to the balance between elastases and their inhibitors.

Effects of amiodarone on elastin biosynthesis in primary hamster lung cell cultures.

Amiodarone is a Class III antiarrhythmic agent that has been implicated as a cause of human pulmonary fibrosis. Pulmonary fibrosis is associated with increased levels of connective tissue proteins such as collagen and elastin. The purpose of this investigation was to determine whether elastin synthesis would be altered by in vitro amiodarone administration. Primary hamster lung cell cultures were utilized. Cultures were treated with 2, 10, and 20 micrograms/ml amiodarone. Following treatment, elastin synthesis was monitored by a biochemical tracer assay based on the presence of the cross-linking amino acids: desmosine/isodesmosine. These cross-links are found only in elastin. Addition of [14C] lysine to cultures results in uptake of the radiolabel into the cross-links. Cross-links were isolated and identified using chromatography and electrophoresis. At all doses of amiodarone, elastin synthesis was seen to increase above control levels. Light and electron microscopy confirmed the presence of an extracellular matrix. The morphologic studies also revealed the presence of cytoplasmic inclusion bodies and vacuoles that are often associated with cationic, amphiphilic drugs such as amiodarone.

The effect of 60% oxygen on air-space enlargement and cross-linked elastin synthesis in hamsters with elastase-induced emphysema.

Hyperoxia is routinely administered to patients with severe emphysema. To gain insight into the possibly adverse effects of such treatment, hamsters were exposed to 60% oxygen for 5 days, beginning 48 h after induction of pulmonary emphysema by intratracheal instillment of pancreatic elastase. Control groups consisted of (1) animals instilled with elastase and exposed to room air, (2) animals instilled with saline and exposed to 60% oxygen, and (3) animals instilled with saline and exposed to room air. Cross-linked elastin content and synthesis in the lung were measured immediately following termination of hyperoxia, and the mean linear intercept was determined 4 wk later. Cytologic examination of bronchoalveolar lavage fluids was also performed. Statistical significance was determined by a two-way analysis of variance. Results indicate that exposure to 60% oxygen significantly affected (p less than 0.05) air-space size, causing a 51% increase among elastase-treated hamsters (124 versus 82 microns) but only a 4% increment among saline-treated animals (52 versus 50 microns). When compared to other groups, animals treated with both elastase and hyperoxia had a significantly greater (p less than 0.01) percentage of neutrophils (28%) in their lung lavage fluids immediately following exposure to 60% oxygen. Although total lung elastin content was not altered by hyperoxia at this time, labelling of elastin cross-links was significantly increased (p less than 0.05). These studies demonstrate that exposure to 60% oxygen enhances elastase-induced lung injury. They also raise the possibility that oxygen therapy may, under certain circumstances, accelerate the progression of human emphysema.

Neutrophil kinetics in O2-exposed rabbits.

Hyperoxic injury results in an influx of polymorphonuclear leukocytes (PMN) into the lung. To better understand the role of the PMN in this injury, kinetic studies were used to assess the survival of PMNs in the circulation. The rate of deposition of PMNs in the lungs of rabbits exposed to hyperoxia was also examined. The half-lives (T1/2) of [3H]thymidine-labeled PMNs in the circulation in rabbits exposed to air or to 95% O2 for less than or equal to 48 h varied between 3.9 and 4.5 h. After 72 h of hyperoxic exposure, T1/2 fell to 2.2 h, the marginal and circulating PMN pool increased and 3H deposition in the lung increased 10-fold. Autoradiographs confirmed that [3H]thymidine was initially nuclear- and cellular-associated but, with time, [3H]thymidine dispersed throughout the lung, suggesting PMN disintegration. These PMN events seem to occur in the later phases of O2 toxicity, and because PMNs are an additional source of oxyradicals, they may further amplify oxidant injury.

Orientation of elastic fibers in the human cervix.

We have determined the microanatomy of the cervix in relation to elastic fibers by serial dissection of human cervix in three tissue planes: cross, sagittal, and frontal sections. Analysis suggests that elastin is localized to specific regions of the uterine cervix and not dispersed throughout the connective tissue stroma. By Musto stain the majority of elastic fibers are noted to be oriented from the external os to the periphery and from there in a band upward toward the internal os where they become sparse in the area of the cervix with the greatest amount of smooth muscle just below the internal os. Elastic fibers were noted to be sparsely distributed in the cervical stroma.

Measurement of cross-linked elastin synthesis in bleomycin-induced pulmonary fibrosis using a highly sensitive assay for desmosine and isodesmosine.

Cross-linked elastin synthesis was measured in the intratracheal bleomycin model of interstitial pulmonary fibrosis by incorporation of 14C-lysine into the elastin-specific crosslinks, desmosine and isodesmosine. Detection of the labeled crosslinks was facilitated by development of a highly sensitive assay utilizing thin-layer electrophoresis. The results indicate that crosslinked elastin synthesis is significantly elevated from controls (p less than 0.05) at 1 to 3 weeks after exposure to bleomycin and returns to normal by 5 weeks. The increases in labeled elastin synthesis are not directly related to changes in either total lung protein synthesis or the pool size of the 14C-lysine. In comparison with collagen and glycosaminoglycan synthesis in this model of lung injury, maximal increases in cross-linked elastin formation occur later, but overlap with the elevated synthesis of these other connective tissue components. The marked increase from normal in cross-linked elastin synthesis in this model suggests that this tissue component is an important part of the fibrotic response of the pulmonary parenchyma and may play a role in the observed alterations in lung structure and function.

Amiodarone-induced pulmonary fibrosis in hamsters.

Amiodarone, a cardiac antiarrhythmic agent, has been associated with the development of interstitial pulmonary fibrosis in patients receiving prolonged therapy with the drug. To further assess the toxic effects of amiodarone on lung tissue, Syrian hamsters were given a single intratracheal insufflation of the agent and evaluated for histologic evidence of lung injury. Control animals received intratracheal insufflations of the vehicle in which amiodarone was dissolved. After an initial, transient alveolitis in both experimental and control animals, the amiodarone-treated lungs developed increased interstitial thickening due to fibrinous exudates, alveolar epithelial hyperplasia, inflammatory cell infiltrates, and marked deposition of collagen manifested on trichrome staining. Controls, in contrast, showed nearly complete resolution of the initial alveolitis. An unusual feature of the amiodarone-induced lung injury was reemergence of the alveolitis between 5 and 14 days, which included a marked influx of eosinophils into the lung. Although the precise mechanism of the lung injury is not known, the persistence of the acute inflammatory cells as well as the presence of eosinophils suggests a hypersensitivity-type reaction. Furthermore, the progression of lung injury to fibrosis after a single insult with the drug suggests that mere discontinuation of amiodarone therapy in humans may not reverse the disease process, but that corticosteroid therapy may also be required. Amiodarone appears to be a useful agent to induce diffuse fibrotic reactions in the lung that morphologically resemble idiopathic pulmonary fibrosis in humans.

Glycosaminoglycan synthesis in bleomycin-induced pulmonary fibrosis: biochemistry and autoradiography.

At 5, 15, and 45 days following induction of interstitial pulmonary fibrosis by intratracheal administration of bleomycin in hamsters, glycosaminoglycan synthesis was measured, using [35S]sulfate. Total labeled sulfate incorporation into lung glycosaminoglycans was maximally increased over that of saline-instilled controls at 5 days (P less than or equal to 0.05), declined markedly at 15 days, and returned to control values at 45 days. Separation of the various labeled glycosaminoglycans by chondroitinase digestion and chromatography revealed a transient rise from controls (P less than or equal to 0.05) in the proportion of labeled chondroitin 4-sulfate at 5 days, followed by an increase from controls (P less than or equal to 0.05) in proportionate labeling of dermatan sulfate at 15 and 45 days postbleomycin. Autoradiography, using [35S]sulfate, performed at 21 days postbleomycin, revealed an increase from controls in film grain formation in areas of interstitial reaction. Grain formation was greatly reduced by pretreatment of the slide sections with hyaluronidase and chondroitinase, demonstrating the specificity of the label for glycosaminoglycans. The results indicate that glycosaminoglycan synthesis is significantly altered from normal in this model of interstitial lung disease and that dermatan sulfate is preferentially synthesized during the fibrotic phase of the lung reaction.

Glycosaminoglycan synthesis in explants derived from bleomycin-treated fibrotic hamster lungs.

Glycosaminoglycan synthesis was studied in explant cultures of hamster lungs 15 and 45 days following intratracheal administration of Bleomycin. At both time points, a statistically significant increase in 35S-sulfate incorporation into glycosaminoglycans was seen in the Bleomycin-treated explants compared with that of the controls. Furthermore, the percentage of label associated with dermatan sulfate was significantly higher in the treated explants than in controls at both 15 and 45 days. Conversely, the percentage of labeled heparin and/or heparan sulfate was significantly lower for the treated explants compared to controls at these times. These results indicate that glycosaminoglycan synthesis is altered from normal in this model of interstitial lung disease. Comparison of these data with previous measurements of glycosaminoglycan synthesis in another model of interstitial lung disease, induced by N-nitroso-N-methylurethane, reveals marked similarity in the changes from normal in 35S-labeling.

Collision tumor of the frontal sinus: evidence of prior intrasinus instillation of thorotrast.

Carcinoma of the frontal sinus is relatively rare. We describe a case of two independent carcinomas, squamous cell and adenoid cystic, in the right frontal sinus region of a patient who received sinus irrigation with an unidentified, substance 30 years previously. Radioautography performed on sections of the tumor tissue revealed linear tracks of the type produced by thorium decay, strongly suggesting that the substance was thorotrast. In addition to being the first description of a "collision tumor" of the frontal sinus, the case would represent the first reported instance of frontal sinus carcinoma following intrasinus instillation of thorotrast.

Impaired angiotensin conversion and bradykinin clearance in experimental canine pulmonary emphysema.

Chronic hypoxic lung diseases are associated with abnormal blood pressure regulation. Because the lung is the principal site of angiotensin conversion and because hypoxia decreases converting enzyme activity, we examined whether angiotensin converting enzyme activity was impaired in lung disease. 12 dogs received a 6 wk course of aerosolized and intratracheal papain that produced moderate panlobular emphysema. These dogs and 24 control dogs were anesthetized and sampling catheters were placed under fluoroscopic control. Angiotensin conversion was measured by a blood pressure response bioassay. Pulmonary converting enzyme activity was also assessed by infusing bradykinin (BK) and using radioimmunoassay to measure the instantaneous clearance of BK and the concentration of BK in the pulmonary artery which first produced spillover of BK into left atrial blood. Angiotensin conversion was reduced in the emphysematous dogs to 81.1% (13.2 SD) from 92% (6 SD) in the control dogs (P < 0.01). Instantaneous clearance of BK in the emphysematous dogs was only slightly reduced (93%), despite reduction in their Pao(2) to 75 mm Hg, indicating that the greatest proportion of the perfused vascular bed was exposed to alveolar Po(2) of >90 mm Hg. However, the barrier to BK passage provided by the lung, and measured by the spillover level, was reduced (1/4) to (1/2) that observed in control animals. That the defect was promptly corrected by supplemental oxygen indicates that regional pulmonary vascular converting enzyme activity had been impaired by regional alveolar hypoxia, which permitted some peptide to pass through the lungs unmetabolized. Determination of peptide metabolism in the lungs may provide a useful measure of regional alveolar hypoxia and may lead to new ways of assessing lung injury.

Alpha 1-antitrypsin deficiency and increased susceptibility to elastase-induced experimental emphysema in a rat model.

Administration of 200 mg of D-galactosamine/kg intraperitoneally to rats produced a decrease in the serum concentrations of trypsin and elastase inhibitory capacities. Induction of emphysema by intravenous injection of pancreatic elastase resulted in significantly increased severity of the disease in the animals depleted of alpha 1-antitrypsin. The degree of severity of the disease determined by mean linear intercept suggested a correlation with trypsin and elastase inhibitory capacities at the time of elastase injection.