INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mucus is a critical component of the primary defense mechanism of the respiratory tract, trapping inhaled particulate and microbial material for removal via the mucociliary system. When this mechanism fails to clear sufficiently, mucus accumulates and must be coughed up as sputum; otherwise it is retained in the respiratory tract, encouraging colonization by microorganisms, which may lead to chronic lung inflammation and obstruction (2). In CF, airway mucus obstruction has long been considered the most insidious agent of morbidity and mortality. Therapies designed to thin the airway mucus and improve its clearability continue to be a major focus of attention (2, 3).

One of the primary aspects of the current treatment of CF lung disease is aimed at changing the physical properties of pulmonary secretions to improve their clearance from the airways. Treatment with rhDNase is based on the fact that the major factor involved in the elevated viscoelasticity of CF sputum is attributed to the presence of naked DNA released into the airway surface fluid (ASF) from bacteria, macrophages, and the other cellular debris (4). Enzymatic digestion of these DNA macromolecules effectively decreases mucus viscoelasticity and spinnability and enhances the clearability of airway secretions (5, 6). Other direct-acting mucolytic treatments such as N-acetylcysteine derivatives (7, 8), gelsolin (9), and hypertonic saline (10, 11) are effective in vitro in CF, but they may not necessarily show clinical efficacy. Indirect mucolysis such as with inhaled amiloride, which blocks the uptake of salt and water across the airway epithelium, is a strategy aimed at enhancing the degree of hydration and diluting the macromolecular component of the ASF (12). Combined mucokinetic therapies may aim to address more than one mechanism involved in the control of airway mucus secretion and clearance (8, 11, 13).

Dextran is a bacterial byproduct; the dextran macromolecule consists of glucan groups linked end to end. Its primary clinical uses are as a plasma volume expander and as an antithrombotic agent that has antiaggregation effects (14). Dextran has recently been shown to exhibit antiadhesive properties in airway epithelial cells (1), which may make it of value as an antimicrobial agent in preventing the Pseudomonas infection in CF. To investigate its possible direct effect on airway mucus clearability, we conducted an investigation of the effects of dextran on mucus rheology and on clearance and spinnability of CF sputum, using our laboratory model systems. Our main concern initially was to define any potentially deleterious effects of dextran prior to the initiation of a clinical trial. Subsequently, when the initial tests suggested that dextran might have direct, potentially beneficial effects on CF sputum, we conducted further studies to investigate these effects, and to determine whether noninfected mucus was also influenced by dextran.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Sputum samples were obtained by voluntary expectoration from 25 adolescent and young adult patients with CF. The patients were all infected with Pseudomonas aeruginosa. None of the patients was under current treatment with rhDNase. Approval to collect and use sputum for this in vitro analysis was obtained from the University of Calgary Research Ethics Board.

Dog mucus was obtained from healthy, anesthetized dogs from the endotracheal tube cuff. Dextran with a molecular weight of 4,000 was provided by Polydex Pharmaceuticals Ltd. (Toronto, ON, Canada).

Study Design

Protocol A (CF sputum, 0.4% dextran). Seven samples of sputum (0.40 to 0.60 g) were treated in 1.5 ml Eppendorf polypropylene centrifuge tubes as follows: (1) Baseline aliquot (no in vitro treatment). (2) Negative control aliquot, adding 10% vol. Ringer's solution and incubating for 30 min at 37° C. (3) Application of 10% vol. Ringer's containing 40 mg/ml Dextran 4,000 to give a final concentration of 0.4% (4 mg/ml) in the sputum; the sample was incubated for 30 min at 37° C.

Protocol B (CF sputum, 0.4% and 4% dextran). Eighteen samples of sputum (0.40 to 0.60 g) were treated in 1.5 ml Eppendorf polypropylene centrifuge tubes as follows. (1) Negative control aliquot, adding 10% vol. Ringer's solution and incubating for 30 min at 37° C. (2) Application of Ringer's containing 40 mg/ml Dextran 4000 to give a final concentration of 0.4% (4 mg/ml) in the sputum. (3) Application of Ringer's containing 400 mg/ml Dextran 4000 to give a final concentration of 4% (40 mg/ml) in the sputum.

Protocol C (canine tracheal mucus, 0.4% and 4% dextran). Seven samples of canine tracheal mucus (approximately 0.1 g) were treated in 1.5 ml Eppendorf polypropylene centrifuge tubes as follows. (1) Negative control aliquot, adding 10% vol. Ringer's incubated 30 min at 37° C. (2) Application of Ringer's containing 40 mg/ml Dextran 4000 to give a final concentration of 0.4% (4 mg/ml) in the sputum. (3) Application of Ringer's containing 400 mg/ml Dextran 4000 to give a final concentration of 4% (40 mg/ml) in the sputum, incubated 30 min at 37° C.

Frog Palate Mucociliary Transportability (FMT)

This was determined from CF sputum movement on the ciliated, mucus-depleted frog palate in a humidification chamber, relative to native frog control (relative velocity, FMT, expressed as a fraction). The movement of a 2 to 5 µl aliquot of CF sputum was timed; five measurements of sputum transport rate were taken to minimize measurement variability (15, 16). FMT was determined only for the samples in Protocols A and B.

Rheologic Measurements on CF Sputum

In this in vitro study, two techniques were used to measure the rheologic properties of sputum: spinnability of filancemeter and viscoelasticity by magnetic rheometry.

Spinnability is the thread-forming ability of mucus under the influence of low amplitude elastic deformation. The spinnability of CF sputum samples was measured using a Filancemeter (SEFAM, Nancy, France), in which a 20- to 30-µl mucus sample is stretched at a distraction velocity of 10 mm/s (17). An electric signal conducted through the mucus sample is interrupted at the point where the mucus thread is broken. The length of this thread is known as the mucus spinnability (measured in millimeters). Spinnability was determined only for sputum samples in Protocol B.

Viscoelasticity and Clearance Indices

The magnetic microrheometer technique was used to measure the viscosity and elasticity of the sputum samples. A 100-µm steel ball was positioned in a 5- to 10-µl sample of sputum, and the motion of this sphere under the influence of an electromagnet was used to determine the rheologic properties of the sputum. The image of the steel ball was projected via a microscope onto a pair of photocells whose output was amplified and transmitted to an oscilloscope. By plotting the displacement of the ball against the magnetic driving force, the viscoelastic properties of the mucus were ascertained (18).

The parameters of mucus viscoelasticity determined were the rigidity index or mechanical impedance, i.e., G*, reported here on a log scale, expressing the vector sum of "viscosity + elasticity" (18). Two derivative parametersmucociliary clearability index (MCI) and cough clearability index (CCI)were computed from in vitro relationships (19). These two indices predict mucus clearability by ciliary and cough mechanisms, respectively, based on the measured rheologic properties and observations of clearance from model studies. The respective formulas are as follows (20):

(1)

(2)

Statistical Analysis

Data from each protocol are presented as mean ± standard deviation (SD) of the mean. To analyze the significance of changes in spinnability, log G* at 1 rad/s, MCI, and CCI after administration of Ringer's control, 0.4% dextran and 4% dextran, the sputum from each patient served as its own control. Equality of means was tested by analysis of variance (ANOVA), post hoc analysis of changes from baseline was determined by the two-tailed, paired t test and regression. The paired t test was also used to determine the differences between spinnability and viscoelasticity after different treatments. Regression was used to determine the correlation between spinnability and visoelasticity. The StatView statistical package (Abacus Concepts, Palo Alto, CA) was used to carry out these analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In Protocol A, after administration of 0.4% Dextran 4000, FMT increased significantly (p = 0.046) compared with Ringer's control (Figure 1B). There was a modest dilution effect (reductions in mucus rigidity, log G*) associated with the Ringer's treatment, and a further reduction caused by the dextran (significant with respect to no treatment, p = 0.004) (Figure 1A). CCI (predicted from rheology), also increased significantly (p = 0.019). Mucociliary clearability on frog palate increased more than that predicted from rheology; this extra clearability could be an indication of a surface or antiadhesive effect.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Sputum samples were obtained from seven patients with CF not receiving rhDNase therapy. Aliquots of sputum were incubated at 37° C for 30 min with either Ringer's (10% added volume) or Dextran 4000 to achieve a final concentration of 0.4%. Rigidity modulus (viscoelasticity) was determined by magnetic rheometry, and mucociliary clearability was determined by observing the rate of sputum transport on excised frog palate. *Significantly different from baseline; significantly different from Ringer's.

After the initial tests suggested that Dextran 4000 might have direct, potentially beneficial effects on CF sputum, we conducted further experiments to investigate the concentration dependence of the effects (Protocol B) and whether noninfected mucus was also influenced by Dextran 4000 (Protocol C).

In Protocol B, compared with saline control, CF sputum spinnability decreased 34.2% by administration of 0.4% dextran (p = 0.0121) and 59.8% by administration of 4% dextran (p = 0.0016) (Figure 2B). Furthermore, the reduction in spinnability after 4% dextran was greater than that seen after 0.4% (p = 0.0046). FMT increased significantly with both 0.4% dextran (p = 0.0005) and 4% dextran (p = 0.0221), but there was no additional change between the two treatment groups (Figure 2C). At the same time, mucus viscoelasticity (log G* at 1 rad/s) was reduced by a factor of 2.43 (0.385 log units) by 0.4% dextran (p = 0.0404) and by a factor of 4.57 (0.660 log units) by 4% dextran (p = 0.0069) (Figure 2A). The additional reduction in mucus viscoelasticity observed at the higher dextran concentration was also significant (p = 0.0193). MCI and CCI (both predicted from rheology) improved significantly in both treatment groups (MCI: p = 0.0252, p = 0.003; CCI: p = 0.0227, p = 0.0423). There was a positive relationship between spinnability and viscoelasticity as well as negative correlations between spinnability and predicted mucociliary clearance and cough clearance.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Sputum samples were obtained from patients with CF not receiving rhDNase. Aliquots of sputum were incubated at 37° C for 30 min with Ringer's (10% added volume) or Dextran 4000 to achieve final concentrations of 0.4% (4 mg/ml) or 4% (40 mg/ml). Rigidity modulus (viscoelasticity) was determined by magnetic rheometry, and spinnability (cohesiveness) was determined by filancemeter (n = 8). Mucociliary clearability was determined by observing the rate of sputum transport on excised frog palate (n = 10). *Significantly different from Ringer's, significantly different from DEX 0.4%.

In Protocol C, the viscoelasticity of healthy dog mucus was decreased significantly by treatment with 0.4% dextran ( log G* = 0.269, p = 0.0048) and with 4% dextran ( log G* = 0.547, p = 0.0016) compared with saline control (Figure 3). Mucociliary clearability (predicted from rheology) only increased significantly for 4% dextran (p = 0.0108). Cough clearability (predicted from rheology) increased significantly in both treatment groups (p = 0.0385; p = 0.0459). The additional reduction in log G* between 0.4% and 4% dextran treatments did not achieve statistical significance (p = 0.09).

View larger version (17K): [in this window] [in a new window]   Figure 3.   Tracheal mucus was obtained from seven healthy anesthetized dogs from the endotracheal tube cuff. Aliquots of mucus were incubated at 37° C for 30 min with Ringer's (10% added volume) or Dextran 4000 to achieve final concentrations of 0.4% (4 mg/ml) or 4% (40 mg/ml). Rigidity modulus (viscoelasticity) was determined by magnetic rheometry. *Significantly different from Ringer's treatment.

Overall, whether for CF sputum or healthy dog mucus, Dextran 4000 treatment significantly reduced viscoelasticity and increased predicted mucociliary and cough clearability. These rheologic effects were significantly greater for 4% dextran treatment than for 0.4%. There was no significant correlation between FMT and mucus viscoelasticity, but there was a significant correlation between spinnability and mucus viscoelasticity (p = 0.0012), spinnability and mucociliary clearability (p = 0.0138), as well as cough clearability (p = 0.004) (both predicted from rheology).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our results demonstrate that Dextran 4000 (final concentrations, 4 mg/ml and 40 mg/ml) significantly reduces the viscoelastic modulus of CF sputum compared with saline control. In Protocols B and C, the absolute levels of control log G* with Ringer's treatment were quite similar, and the reductions in log G* achieved with the two levels of treatment were comparable, although the effect in CF sputum was marginally greater at each concentration. Because the dog mucus was noninfected mucus, we can conclude that the effects of dextran on mucus viscoelasticity and clearability are largely independent of the presence of high molecular weight DNA or other pathologic macromolecules.

The mechanisms for the change in viscoelasticity is not known. It could be due to a general osmotic effect, or it could be an antiaggregation effect (change in surface tension) associated with a decrease in the adhesive properties of the mucus, or it could arise because of the substitution of (relatively) low molecular weight saccharide moieties provided by dextran in place of the oligosaccharide moieties linked to the high molecular weight mucin peptides in the three-dimensional structure that makes up the mucous gel.

The decrease in viscoelasticity caused by dextran treatment is predicted to lead to a substantial increase in sputum cough clearability, based on model studies, and to a more modest increase in mucociliary clearability. However, direct measurements of sputum mucociliary clearability using the frog palate model indicate a substantial increase with dextran as well, beyond that predicted from the change in viscoelasticity. This discrepancy (extra clearability) could be an indication of a surface or antiadhesive effect of the dextran, but this remains to be proved. In the frog palate experiments, after administration of 0.4% dextran and 4% dextran, the CF sputa appeared to spread more easily, and some of the samples were quite liquid and difficult to pick up. It is possible that this phenomenon could be due to an osmolality or surface tension effect with dextran treatment. From the results, both 0.4% and 4% dextran treatment improved frog mucociliary clearance, compared with Ringer's control, but there was no apparent difference between the two treatment groups. This is probably a natural consequence of the relationship between mucociliary clearability and mucus viscoelasticity (15), indicating that the reduction in crosslink density with 4% dextran did not result in suboptimal clearability as a result of "overliquification" of the mucus.

The mean decrease in CF sputum viscoelasticity ( log G*) caused by 4 mg/ml Dextran 4000 was approximately 0.45 log units compared with no treatment, and approximately 0.25 log units with respect to vehicle control. This was intermediate between the effect of rhDNase on CF sputum at "high" concentration (50 µg/ml), where it caused an order-of-magnitude decrease in G*, and at "low" concentration (2.5 µg/ml or approximately 100 nM), where it caused virtually no change in G* with respect to saline control (11, 21). Most authorities would place the achievable in vivo rhDNase concentration in the range of 2 to 4 µg/ml. The concentration of 4 mg/ml dextran in the airway surface fluid is achievable with aerosol delivery of 80 mg/ml dextran solution, based on model calculations.

The interpretation of the rheologic data is complex because rhDNase, even at low concentration (2.5 µg/ml), does cause a significant decrease in another rheologic test of elasticity, namely, spinnability or cohesiveness, which measures finite extensibility (8). The present study also demonstrated that Dextran 4000 significantly decreases the CF respiratory mucus spinnability in vitro. In association with the spinnability decease, there was a significant decrease in mucus viscoelasticity and a significant increase in mucociliary clearability and cough clearability (as predicted from viscoelasticity). The mechanisms for the change of spinnability and its correlation with viscoelasticity, mucociliary clearability, and cough clearability is not known. These aspects of mucus rheology and clearability need further study (22).

The effects of dextran, a neutral polymer, on CF sputum are similar in many respects to those of salt (NaCl) treatment (10, 11) since both forms of treatment reduce sputum viscoelasticity in vitro. Although they may act through a generally related mechanism involving increased osmolality, it is likely that the specific mechanisms are different. In the case of NaCl, the reduction in mucous gel crosslink density is believed to involve shielding of excess fixed negative charges, reducing the size of the macromolecular coils and decreasing their intermolecular interactions, whereas in the case of dextran, the reduction in crosslink density most likely involves hydrogen bond substitution. In cystic fibrosis, it is possible that NaCl treatment could further elevate the periciliary ionic environment, thus exacerbating the problem of inadequate bacterial killing (23). Treatment with a nonionic mucotropic agent, on the other hand, should not alter the ASF ionic concentrations, and with the additional advantage of reducing adhesion (1), could result in improved bacterial clearance from the airway fluid milieu.

The dog mucus, which was collected from healthy, anesthetized dogs, was also reduced significantly in viscoelasticity with in vitro dextran treatment, whereas mucociliary and cough clearability, as predicted from the viscoelastic data, improved significantly. Thus, we can say that the effects of dextran are not specific to CF lung disease, and this form of treatment may have potential applicability to other lung diseases where mucus retention is a prominent feature. These results provide further support for clinical trials with dextran, which have been proposed because of the fact that Pseudomonas aeruginosa adherence to epithelial cells in inhibited by dextran. Dextran is an inexpensive and nontoxic agent that may be of benefit in patients with CF lung disease and perhaps in other respiratory diseases where mucus retention is a prominent feature.