Abstract
We investigated the physicochemical and transport properties of sputum samples collected in physiotherapy from a well-documented group of 27 cystic fibrosis (CF) patients with identified CF genotypes. Sputum samples were characterized ex vivo for their water content, surface properties (surface tension and contact angle), rheologic properties (viscosity and elastic modulus), and transport properties (mucociliary and cough transport). These data were analyzed in relation to the clinical status of the patients (FEV1, FVC, Shwachman score, Brasfield score, nutritional status), their genotype, and the degree of infection of their sputa (leukocyte and Pseudomonas aeruginosa counts). We observed negative and significant correlations between mucociliary transport and elastic modulus of the patients' sputum (r = − 0.63, p < 0.01), and between the cough transport and contact angle of the sputum (r = − 0.81, p < 0.0001), respectively. The P. aeruginosa count was also significantly correlated with the sputum water content (r = − 0.53, p < 0.02) as well as with the cough transport of the sputum (r = − 0.62, p < 0.01). In CF patients with a sputum leukocyte count > 2,000/mm3, the sputum water content (p < 0.02), FEV1 (p < 0.05) and FVC (p < 0.02) were significantly lower than those of CF patients with a leukocyte count ⩽ 2,000/mm3. CF patients with a homozygous Δ F 508 genotype had significantly greater values of sputum water content (p < 0.05), and cough-transport capacity (p < 0.05) than did heterozygous patients. No correlation was observed between the sputum properties and any of the clinical data. These results suggest that the control of infection should be emphasized in CF, since it can directly or indirectly modulate the degree of hydration, and therefore the physicochemical and transport properties, of airway secretions.
In cystic fibrosis (CF), alterations in cystic fibrosis transmembrane conductance regulator (CFTR) protein lead to an impermeability of the luminal surface to chloride (1-3). The resulting chloride-defective secretions, coupled with an excessive absorption of sodium (4), are responsible for the reduction in fluid secretion across the CF respiratory epithelium (5), contributing to substantial dehydration and alterations of osmolarity of the periciliary and gel layers. It is widely claimed that the dehydration and faulty transepithelial ion transport contribute to both abnormal rheologic (6, 7) and surface properties of CF respiratory mucus (8). A recent study by Smith and colleagues (9) reported hypertonic extracellular Na+ and Cl− concentrations in CF airway surface fluid. Abnormal acidification in the trans-Golgi endosomes and prelysosomes has also been suggested as a mechanism responsible for defective glycoprotein and lipid synthesis in CF, possibly in relation to alterations in the physical and rheologic properties of CF respiratory mucus (10). So far, most of studies devoted to characterization of the physicochemical properties of respiratory mucus in CF have used samples collected by expectoration from CF patients, without precise information about the degree of hydration, inflammation, and infection of the collected mucus samples. None of the studies has clearly established whether there are rheologic abnormalities in CF airway mucus that can be related to purulence and bacterial colonization, or the way in which these latter factors influence the mucociliary and cough-transport functions of the mucus. The possible influence of CF genotype on the biochemical properties of respiratory mucus has been elegantly demonstrated by Zhang and colleagues (11). The xenograft model that they used has the advantage of excluding the secondary effect of airway epithelial remodeling due to chronic bacterial infection, and therefore allows a better evaluation of the primary defect in CF mucus biochemistry. Although Zhang and colleagues observed a higher level of mucous sulfation in CF than in non-CF mucus, they reported a marked variability in mucous sulfation in genotypically different groups of CF patients. More interestingly, they reported that one homozygous ΔF 508 CF patient showed no increase in the level of mucous sulfation, suggesting that yet unidentified mutant alleles could provide some level of functional intracellular CFTR in CF patients, facilitating normal mucus glycoprotein processing. These recent findings suggest that physicochemical properties of mucus in CF may depend on the CFTR mutation. Other factors, such as the inflammatory response to chronic endobronchial infection, may also contribute to the dysregulation of ion and water secretion and therefore modify the rheologic profile of CF mucus and its transport functions. Whether these abnormalities relate to the clinical state of CF patients remains obscure.
The aim of the present study was to analyze the physicochemical properties, including the rheologic and physical properties and water content, and the ciliary and cough properties, of sputum samples from a group of CF patients with well-documented lung disease and identified genotype, in order to more clearly define whether these properties were dependent on the CF genotype and on the degree of infection and purulence of the patients' mucus.
Twenty seven CF patients (23 outpatients and four inpatients), consisting of 14 males and 13 females aged 5 to 29 yr, were included in the study. The disease was well documented, and recent clinical and paraclinical data were obtained for each patient. The patients were recruited from two CF centers, in Rennes and in Reims hospitals, respectively.
Clinical evaluation of the patients was based on several functional parameters determined during control visits. The pulmonary function of the patients was evaluated from their FVC and FEV1, each expressed as a percentage of predicted values. The clinical and radiologic severities of the CF patients were determined from their Shwachman (12) (SS) and the Brasfield (13) (BS) scores, respectively. The patients' nutritional status was estimated from the ratio of their actual weight versus the theoretical weight evaluated according to their measured height and their sex. Pancreatic sufficiency (PS) or insufficiency (PI) was noted.
Determination of each patient's genotype was performed by classical DNA extraction from blood samples. ΔF 508 deletions in exon 10 were noted, and several other exons, such as exons 4, 11, 19, 20, and 21, were screened for other mutations after amplification with the polymerase chain reaction (PCR). The CF group was screened by the Laboratory of Molecular Biology of the Hospital of Reims (P. Birembaut and C. Clavel) and by the Biochemical Laboratory of the Hospital of Rennes (J. Y. Le Gall).
Patients' treatments at the time of sputum collection were also reported, particularly, antibiotic treatment, antiinflammatory therapy, and mucolytic treatment, including N-acetylcysteine, ambroxol, or recombinant human deoxyribonuclease (rhDNase). All of these data are shown in Table 1.
| Subjects | Age(yr) | Sex | FVC(% pred ) | FEV1(% pred) | BS | SS(out 100) | LeukocyteCount | Treatment | Pathogens* | Mutations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 10 | M | 43 | 37 | 10 | 45 | > 2,000 | — | P. aer | ΔF508/F508 | ||||||||||
| 2 | 14 | M | 31 | 18 | 13 | 35 | > 2,000 | Ab | P. aer | ΔF508/ΔF508 | ||||||||||
| 3 | 18 | M | 86 | 75 | 6 | 90 | < 2,000 | — | S. aur | ΔF508/ΔF508 | ||||||||||
| H. inf | ||||||||||||||||||||
| 4 | 22 | F | 43 | 27 | 10 | 54 | > 2,000 | Ab | S. aur | ΔF508/ΔF508 | ||||||||||
| H. inf | ||||||||||||||||||||
| 5 | 25 | F | 60 | 31 | 11 | 35 | > 2,000 | — | P. aer | ΔF508/ΔF508 | ||||||||||
| S. α hem | ||||||||||||||||||||
| 6 | 11 | M | 107 | 101 | 5 | 90 | < 2,000 | — | S. aur | ΔF508/ΔF508 | ||||||||||
| S. α hem | ||||||||||||||||||||
| 7 | 20 | M | 36 | 26 | 14 | 35 | > 2,000 | Ab | P. aer | ΔF508//3272-26 A→ G | ||||||||||
| 8 | 22 | M | 50 | 48 | 12 | 53 | > 2,000 | Ab | P. aer | ΔF508/W846X | ||||||||||
| 9 | 19 | F | 81 | 55 | 8 | 85 | > 2,000 | Ab | P. aer | ΔF508/R553X | ||||||||||
| S. α hem | ||||||||||||||||||||
| 10 | 12 | M | 87 | 66 | 5 | 85 | > 2,000 | Ab + Mucol (N-acet) | P. aer S. aur | ΔF508/ΔF508 | ||||||||||
| S. pneu | ||||||||||||||||||||
| 11 | 16 | F | 95 | 44 | 7 | 60 | < 2,000 | Ab + Mucol (N-acet + amb) | P. aer | ΔF508/G551D | ||||||||||
| 12 | 17 | F | 36 | 25 | 14 | 35 | > 2,000 | Ab | P. aer | ΔF508/N1303K | ||||||||||
| S. aur | ||||||||||||||||||||
| 13 | 19 | F | 70 | 60 | 8 | 60 | > 2,000 | Ab | P. aer | ΔF508/621+1G→ T | ||||||||||
| 14 | 16 | M | 94 | 88 | 5 | 95 | > 2,000 | — | S. aur | N1303K/4005+1G→ A | ||||||||||
| 15 | 29 | F | 60 | 42 | 6 | 80 | > 2,000 | Ab + Mucol (amb) | P. aer S. α hem | ΔF508/ΔF508 | ||||||||||
| 16 | 13 | M | 74 | 52 | 3 | 80 | > 2,000 | — | P. aer | ΔF508/3905insT | ||||||||||
| S. aur | ||||||||||||||||||||
| H. inf | ||||||||||||||||||||
| 17 | 12 | F | 93 | 90 | 3 | 89 | > 2,000 | — | S. aur | ΔF508/? | ||||||||||
| S. pneu | ||||||||||||||||||||
| 18 | 24 | M | 69 | 46 | 9 | 87 | > 2,000 | Ab | P. aer | ΔF508/H1085R | ||||||||||
| 19 | 6 | F | 78 | 61 | 8 | 70 | > 2,000 | — | P. aer | ΔF508/G551D | ||||||||||
| S. aur | ||||||||||||||||||||
| 20 | 27 | M | 35 | 17 | 15 | 55 | > 2,000 | — | P. aer | ΔF508/R1162X | ||||||||||
| 21 | 18 | F | 103 | 111 | 6 | 95 | > 2,000 | — | S. aur | ΔF508/ΔF508 | ||||||||||
| H. inf | ||||||||||||||||||||
| 22 | 12 | F | 86 | 79 | 7 | 60 | < 2,000 | Mucol (N-acet) | P. aer | ΔF508/ΔF508 | ||||||||||
| 23 | 10 | M | 69 | 54 | 8 | 75 | > 2,000 | — | P. aer | ΔF508/? | ||||||||||
| S. α hem | ||||||||||||||||||||
| 24 | 6 | F | 71 | 59 | 7 | 75 | > 2,000 | Ab | S. aur | ΔF508/ΔF508 | ||||||||||
| H. inf | ||||||||||||||||||||
| 25 | 5 | F | 63 | 69 | 8 | 83 | < 2,000 | Ab | P. aer | ΔF508/ΔF508 | ||||||||||
| 26 | 10 | M | 74 | 57 | 7 | 80 | > 2,000 | Mucol (N-acet) | P. aer | ΔF508/ΔF508 | ||||||||||
| 27 | 6 | M | 91 | 103 | 4 | 90 | < 2,000 | Mucol (N-acet) | P. aer | ΔF508/ΔF508 |
Sputum was collected at physiotherapy. Expectorates were protected from salivary contamination through the use of dental cotton-wool swabs (14). Insofar as sputum samples were collected in different centers at varying distances from the Laboratory of Rheology (Rennes), a preliminary analysis was done in order to define optimal conditions for mailing sputum samples in such a way that their rheologic and surface properties were preserved. We verified that a 24-h-cooled mailing of sputum samples did not affect the rheologic, surface, or transport properties of aliquoted, paired sputum samples.
The analysis of sputum samples collected from our group of 27 CF patients included measurement of the samples' physical and transport properties, water content, viscosity, elastic modulus, surface tension, contact angle, and cough and relative mucociliary transport.
Cytobacteriologic analysis of sputum samples was done in the laboratories of bacteriology (Pr. C. Chippaux, Dr. O. Bajolet-Laudinat, Reims, France; Pr. J. L. Avril, Dr. P. Y. Donnio, Rennes, France) at the two CF centers involved in the study. The leukocyte counts in sputum samples were done in a Malassez cell in each hospital at the moment of sample collection, and were reported as ⩽ 2,000 leukocytes/ mm3 or > 2,000 leukocytes/mm3. A leukocyte count above 2,000 leukocytes/mm3 was representative of sputum purulence. The identification of Pseudomonas aeruginosa in sputum samples was done at the moment of collection. The concentrations of P. aeruginosa or other pathogens (Staphylococcus aureus, Haemophilus influenzae, Streptococcus pneumoniae, Streptococcus α hemolyticus) in sputum samples were also determined and those that exceeded 105 cfu/ml are reported in Table 1. The macroscopic aspect of each sputum sample, designated in the study by the purulence index, was evaluated visually for color and apparent viscosity, with values ranging from 1 for mucoid and fluid secretions to 5 for highly colored (green or yellow) and thick secretions, respectively. The evaluations were done by a single observer for all of the patients. Intermediate values for secretions were defined as 2 for opaque and fluid, 3 for white and moderately thick, and 4 for poorly colored and thick.
Water content. Water content was measured through double determination of the wet and dry weight of an aliquot of sputum after dehydration at 60° C for 12 h. The wet weight minus the dry weight divided by the wet weight represents the water content expressed as a percentage of water.
Viscoelasticity. Sputum viscoelastic properties were analyzed with a control stress rheometer (TA Instruments, Voisins le Bretorreux, France) equipped with a cone-plate (15). The angle between the cone and the plate was 0.017 rad and the required specimen volume was 100 μl. The measurements were done at 25° C, using the creep-test technique. A constant stress of 6.2 Pa was applied to the sample, and the resulting strain was recorded versus time. When a steady flow was achieved, the applied stress was suppressed and the recovery angle (γ) of the strain, representative of the sputum elasticity, was measured. The slope of the strain-versus-time curve was representative of the shear rate applied to the sputum sample. The shear stress/shear note ratio and the shear stress/shear strain ratio allowed us to calculate the viscosity and the elastic modulus, respectively, of the sputum sample.
Surface tension and contact angle. The surface tension (γlv) of the sputum samples was measured with the platinum ring method (16). In brief, a fixed platinum ring was put in contact with 200 μl of the sputum sample, which was then moved downward at a constant speed. A transducer connected to the ring measured the force required to separate the ring from the sample. The surface tension of the sputum sample was determined with reference to the known surface tension of pure water, which is 72.8 mN/m.
The contact angle (Θ) of a 20 μl drop of sputum on a reference glass slide was calculated with an image-analysis technique developed in our laboratory (17). At equilibrium in a 100% relative-humidity chamber, the tangent to the sputum drop at the sputum–air–glass interface was calculated.
Cough transport. Cough transport of sputum samples was measured in the cough machine developed by King and colleagues (18). A tank of 6 L volume was used as a reservoir for pressurized air, and was connected through a solenoid valve to a plastic tube representing the trachea. The floor of the tube was made of a glass slide on which a drop of sputum was deposited. Cough, as a pulse, was simulated by opening the solenoid valve and releasing the pressurized air through the model trachea with an airflow of 6 L/s. Cough transport, expressed in millimeters, represents the distance traveled by the sputum under the force of the airflow. For each sputum sample, the cough transport and the contact angle were measured in triplicate.
Mucociliary transport. In vitro measurements of sputum transport by ciliary activity were made using the frog-palate technique (19). Isolated palates from Rana esculenta frogs were placed into a Plexiglas chamber at a controlled temperature of 25° C and in 100% relative humidity. Twenty-four hours later, when endogenous mucus secretion had been exhausted, fresh mucus (1 μl) from a recently killed frog was placed on the depleted palate. The transport velocity of the reference and CF mucus specimens was measured by following the displacement of calibrated aluminum discs placed on each mucus drop. The relative mucociliary transport rate (tr), which corresponds to the ratio of the CF sputum transport velocity versus the frog mucus transport velocity, was measured in triplicate and expressed as arbitrary units.
Data for the CF patients were expressed as mean ± SD (range). Correlations between the different parameters were tested with a simple linear regression test. Further correlations between transport and physicochemical properties, and between the purulence index and the other parameters, respectively, were tested with a step-by-step regression analysis. The step-by-step regression analysis classifies the parameters according to the degree of correlation at each step, and interrupts the multiple correlation test when the multiple correlation coefficient is not significantly improved by the other steps.
Differences between subgroups in terms of genotype, the presence of P. aeruginosa in sputum samples, antibiotic treatment, and mucolytic treatment were compared through a one-way analysis of variance (ANOVA). A value of p < 0.05 was considered significant.
The individual CF patient characteristics are reported in Table 1. The mean age of the 27 CF patients was 15 ± 7 yr (range: 5 to 29 yr). The four inpatients were in exacerbation of CF, according to the criteria reported by Woods (20).
The mean SS of the study group was 69 ± 22 (range: 35 to 95). The mean nutritional status expressed as a percentage of the theoretical value was 90 ± 9% (range: 75 to 110%). All of the CF patients had pancreatic insufficiency. The mean BS was 8.1 ± 3.3 (range: 3 to 15). The mean FEV1, expressed as a percentage of the predicted value, was 57 ± 27% (range: 17 to 111%), and the mean FVC, expressed as a percentage of the predicted value, was 70 ± 20% (range: 31 to 107%).
None of the CF patients received antiinflammatory therapy. Thirteen and six of the 27 patients were receiving antibiotic or mucolytic agents, respectively, at the time of sputum collection, but none of the patients had ever received rhDNase. Patients treated with antibiotics had a significantly lower FEV1 (p < 0.05) than did those who did not receive antibiotics. Patients being treated with mucolytic agents at the time of sputum collection had a significantly lower purulence index (p < 0.05) than did those who did not receive mucolytic agents.
Twenty of the 27 patients were colonized with P. aeruginosa at the time of the study. Three of the 27 had never been colonized with P. aeruginosa, and of the 24 other patients, the mean age at colonization with P. aeruginosa was 10.6 ± 5.9 yr (range: 1 to 22 yr). Fourteen patients were homozygous for the ΔF 508 mutation. Twelve patients were heterozygous for ΔF 508 with an associated other mutation. Ten of these 12 patients had known mutations. Among these 10 known mutations, eight can be classified as severe mutations, including nonsense mutations (R553X [1/10], R1162X [1/10], W846X [1/ 10]), splicing mutations (621+1G→ T [1/10]), frameshift mutations (3905insT [1/10]), and missense mutations (G551D [2/ 10]), N1303K [1/10]). Only two are classified as mild mutations, consisting either of splicing mutations (3272-26A→ G [1/ 10]) or missense mutations (H1085R [1/10]). The single noncarrier of the ΔF 508 mutation had the mutations N1303K and 4005+1G→ T, and this patient was not classified as either a heterozygote or a homozygote.
No significant difference was observed within the CF group in clinical parameters including the SS and the BS, in the functional parameters FEV1 and FVC, or in nutritional status when the group was separated according to heterozygous or homozygous genotype for the ΔF 508 mutation. In both subgroups, patients in whom P. aeruginosa was identified in sputum samples at that time of sputum collection had altered values of the clinical parameters included in the study as compared with those whose sputum samples showed no P. aeruginosa at the time of sputum collection with an SS of 64.4 ± 19.7 versus 84.4 ± 14.9, respectively (p < 0.05), a BS of 8.9 ± 3.4 versus 6 ± 2.2, respectively (p < 0.05), an FEV1 of 49.1 ± 21.7% versus 78.7 ± 28.4%, respectively (p < 0.05), and an FVC of 63.9 ± 20.5% versus 85.3 ± 22%, respectively (p < 0.05).
The mean water content of CF sputum samples was 91.3 ± 3.1% (range: 84.5 to 97.1%). The mean contact angle was 58.8 ± 13.1 degrees (range: 31.2 to 90 degrees). The mean surface tension was 80.9 ± 23.6 mN/m (range: 53 to 167 mN/m). The mean viscosity was 38 ± 47.5 kPa · s (range: 0.8 to 183 kPa · s) and the mean elastic modulus was 2.2 ± 2.2 Pa (range: 0.2 to 10 Pa), respectively.
The mean cough transport and mean relative mucociliary transport of CF sputum samples were 23.5 ± 11.1 mm (range: 4 to 53 mm) and 0.66 ± 0.21 (range: 0.1 to 1), respectively. Simple regressions between the transport properties (cough transport and relative mucociliary transport) and the different physicochemical parameters of the sputum samples are shown in Table 2. In the step-by-step regression analysis, the relative mucociliary transport was significantly correlated with the elastic modulus (r = −0.63) (Figure 1A), and the cough transport was significantly correlated with the contact angle (r = −0.81) (Figure 1B); none of the other parameters improved the correlations. Transport properties of CF sputum samples did not correlate significantly with any of the clinical and paraclinical data.
| Water Content | Surface Tension | Contact Angle | Viscosity | Elastic Modulus | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cough transport | r = +0.72 | r = −0.63 | r = −0.81 | r = −0.42 | r = −0.48 | |||||
| p < 0.001 | p < 0.001 | p < 0.0001 | p < 0.05 | p < 0.05 | ||||||
| Relative mucociliary transport | r = +0.52 | r = −0.54 | r = −0.60 | r = −0.60 | r = −0.63 | |||||
| p < 0.01 | p < 0.01 | p < 0.01 | p < 0.01 | p < 0.01 |
Fig. 1. Relationships between relative mucociliary transport (tr) and elastic modulus (r = −0.63, p < 0.01) (A), and between cough transport and contact angle (r = −0.81, p < 0.0001) (B) of CF sputum samples.
[More] [Minimize]When patients were separated into the homozygous group or heterozygous group according to their genotype for the ΔF 508 mutation, significant differences were observed for sputum water content (92.2 ± 2.8% versus 89.7 ± 2.6%, respectively; p < 0.05), contact angle (50.2 ± 11.3 degrees versus 60.2 ± 12 degrees, respectively; p < 0.05) and cough transport (26.1 ± 10.1 mm versus 18 ± 7 mm, respectively; p < 0.05). The purulence index of the two groups was not significantly different.
The presence or absence of P. aeruginosa in sputum samples had no effect on the physical or transport properties of sputum samples, but the count of P. aeruginosa in the sputum samples was significantly and negatively correlated with the sputum water content (r = −0.53, p < 0.02; Figure 2A) and with cough transport (r = −0.62, p < 0.01; Figure 2B), as well as with FEV1 (r = −0.46, p < 0.05). When sputum samples were separated according to their leukocyte count, their water content was significantly lower in the group of patients with a leukocyte count > 2,000/mm3 than in the group with a leukocyte count ⩽ 2,000/mm3 (90.7 ± 3.1 versus 94.2 ± 1.8, respectively, p < 0.05). The group for which the number of leukocytes in the sputum was > 2,000/mm3 also had significantly lower values of FEV1 (52.2 ± 25.2 versus 80.4 ± 23.9, respectively, p < 0.05) and FVC (65.5 ± 21.7 versus 93 ± 8.7, respectively, p < 0.02), than did the group for which the leukocyte count was ⩽ 2,000/mm3.
Fig. 2. Relationships between P. aeruginosa count and water content (r = −0.53, p < 0.02) (A), and between P. aeruginosa count and cough transport (r = −0.62, p < 0.01) (B) of CF sputum samples.
[More] [Minimize]The macroscopic aspect of sputum samples, reported as the purulence index, was 2.8 ± 1.5 (range: 1 to 5). The purulence index was significantly correlated with the BS (r = +0.40, p < 0.05). The purulence index was also significantly correlated with the sputum viscosity (r = +0.53, p < 0.001), surface tension (r = +0.40, p < 0.05), water content (r = −0.67, p < 0.001), elastic modulus (r = +0.52, p < 0.01), contact angle (r = +0.55), and cough transport (r = −0.55, p < 0.01). In the step-by-step regression analysis the purulence index was significantly correlated with the sputum water content (r = −0.63); none of the other parameters improved the correlation. No correlation was observed between the visual purulence index and the P. aeruginosa count of sputum samples.
We report here a complete study of the physicochemical and transport properties of sputum samples in 27 CF patients with well-documented lung disease and identified CF genotypes. Our study confirmed results previously reported in the literature, achieved either with mucus simulants or with sputum samples collected from non-CF patients (8, 21-25); however, it represents the first complete study of sputum samples collected from genotyped CF patients with well-documented lung disease. The results of the study clearly demonstrate that the efficiency of mucociliary clearance is related to the rheologic properties (elastic modulus) of mucus, whereas the cough mechanism is highly dependent on the surface properties (contact angle) of respiratory mucus. In CF, altered surface properties of respiratory mucus, associated with marked changes in its phospholipid composition (12), an altered ionic composition (8) and altered fluid secretion may be responsible for the dramatic decrease in cough clearance. Interestingly, it appears from the present study that the water content of CF sputum samples is not systematically low, and that dehydration of respiratory mucus in CF is not a general feature of the disease, although a recent study done with a xenograft model demonstrated significantly higher basal fluid absorption in a CF differentiated respiratory epithelium than in a non-CF differentiated respiratory epithelium (26). It also appears from previous reports that an abnormal electrolyte composition of airway surface liquid (ASL) may not be systematically associated with fluid reabsorption by CF airway epithelium, and that an altered ionic composition of CF ASL could by itself induce altered viscoelastic properties of CF sputum (27, 28). The greater degree of purulence of sputum samples, including inflammatory and bacterial components, the lower the sputum water content. This suggests, apart from the primary CFTR defect in CF, that inflammatory or bacterial products may modify ionic transport, as previously reported by Stutts and colleagues (29), and may therefore limit fluid transport through the respiratory epithelium, in addition to the altered fluid transport that occurs across the CF respiratory epithelium without infection (26). At high doses, bacterial and leukocyte products could also be responsible for remodeling the epithelium in a way that reduces the number of cells or of well-differentiated cells (ciliated) to below that needed to ensure proper ionic and fluid transport through the respiratory epithelium (30). The macroscopic purulence index of sputum samples is highly correlated with their rheologic properties. Moreover, the purulence of sputum samples appears to be closely associated with their water content, surface properties, and cough clearance. Surprisingly, the presence of P. aeruginosa in the CF sputum samples from our infected CF patients in comparison with those from our noninfected CF patients had no effect on the rheologic, surface, or transport properties or on the water content of the sputum samples, although it contributes widely to the deterioration of clinical status according to a previous study by Kubesh and colleagues (31). This suggests that the purulence index reflects several factors, such as leukocyte transepithelial migration, cellular debris, plasma exudate, and the presence of other pathogens than P. aeruginosa. The absence of a significant correlation between the macroscopic aspect of sputum and the P. aeruginosa count may also reflect the possible persistence of inflammation even beyond episodes of acute infection. Interestingly, the water content, contact angle, and cough transport of sputum samples were better in our CF patients with a homozygous ΔF 508 mutation genotype than in CF patients with a heterozygous ΔF 508 mutation genotype. Our patient group that was heterozygous for the ΔF 508 mutation carried mutations on the second allele that are severe mutations (in eight of 10 patients), whereas two of the 10 patients in this group had mild mutations (32). These associated severe mutations could account for altered mucus hydration, surface properties, and cough transport, as a result of total absence of the CFTR protein, whereas ΔF 508 mutation is now described as a nonsystematically dramatic mutation in terms of protein expression and distribution. We were unable to correlate any of the mucus-transport properties examined in our study with clinical data, particularly with FEV1 and FVC, although a previous study by Puchelle and coworkers (33) reported that in CF patients with superinfection, including high values of P. aeruginosa and leukocyte counts, an increased sputum viscosity and altered mucociliary transport rate were associated with a low SS. Mucolytic treatment including N-acetylcysteine or ambroxol, and antibiotic treatment, significantly improved the visual purulence index of sputum and the respiratory function, respectively, in the CF group in Puchelle and coworkers' study, but these results should be followed with a study of a larger population.
Taken together, these results suggest that the control of infection should be emphasized early in the course of CF, since it can directly or indirectly modulate the degree of hydration and consequently the physicochemical as well as the transport properties of airway secretions in the disease.
The authors are grateful to Geneviève Jolly and Nelly Guillou, who helped them to collect sputum samples during physiotherapy, in the CF center in Rennes, and to Thierry Bienvenu, of the Laboratory of Genetic Biochemistry of the Hôpital Cochin in Paris, for classification of CFTR mutations.
| 1. | Quinton P. M.Cystic fibrosis: a disease in electrolyte transport. F.A.S.E.B. J.4199027092717 |
| 2. | Mac Pherson, M. A., and R. L. DormerMolecular and cellular biology of cystic fibrosis. Mol. Aspects Med.121991181 |
| 3. | Welsh M. J.Abnormal regulation of ion channel in cystic fibrosis epithelia. F.A.S.E.B. J.4199027182725 |
| 4. | Boucher R. C., Stutts M. J., Knowles M. R., Cantley L., Gatzy J. T.Na+ transport in cystic fibrosis respiratory epithelia: abnormal basal rate and response to adenylate cyclase activation. J. Clin. Invest.78198612451252 |
| 5. | Jiang C., Finkbeiner W. E., Widdicombe J. H., Mac P. B., Cray, Miller S. S.Altered fluid transport across airway epithelium in cystic fibrosis. Science2621993424427 |
| 6. | Verdugo P.Mucin exocytosis. Am. Rev. Respir. Dis.1441991S33S37 |
| 7. | Tacheuchi K., Suzumura E., Hirata K., Majima Y., Sakakura Y.Role of transepithelial ion transport as a determinant of mucus viscoelasticity in chronic inflammation of the maxillary sinus. Acta Otolaryngol.111199111331138 |
| 8. | Pillai R. S., Chandra T., Miller I. F., Lloyd-Still J., Yeates D. B.Work of adhesion of respiratory tract mucus. J. Appl. Physiol.72199216041610 |
| 9. | Smith J. J., Trevis S. M., Greenberg E. P., Welsh M. J.Cystic fibrosis airway epithelial cells fail to kill bacteria because of abnormal airway surface fluid. Cell851996229235 |
| 10. | Barasch J., Kiss B., Prince A., Saiman L., Gruenert D., Al-Awqati Q.Defective acidification of intracellular organelles in cystic fibrosis. Nature35219917073 |
| 11. | Zhang Z., Doranz B., Yankaskas J. R., Engelhardt J. F.Genotypic analysis of respiratory mucous sulfation defects in cystic fibrosis. J. Clin. Invest.96199529973004 |
| 12. | Shwachman H., Kulczycki L. L.Long-term study of one hundred and five patients with cystic fibrosis. Am. J. Dis. Child.961958629 |
| 13. | Brasfield D., Hilks G., Soong S., Tiller R. E.The chest roentgenogram in cystic fibrosis: a new scoring system. Pediatrics6319792329 |
| 14. | Puchelle E., Tournier J. M., Zahm J. M., Sadoul P.Rheology of sputum collected by a simple technique limiting salivary contamination. J. Lab. Clin. Med.1031984347353 |
| 15. | Puchelle E., Zahm J. M., Duvivier C., Didelon J., Jacquot J., Quemada D.Elastothixotropic properties of bronchial mucus and polymer analogs. Biorheology221985415423 |
| 16. | Puchelle E., Zahm J. M., Jacquot J., Plotkowski M. C., Duvivier C.A simple technique for measuring adhesion tension properties of human bronchial secretions. Eur. J. Respir. Dis.711987281282 |
| 17. | Vaquez-Girod S., Zahm J. M., Dionisius J. P., Pierrot D., Puchelle E.Automatic measurement of the wetting properties of biological fluids. Innov. Tech. Biol. Med.91988402412 |
| 18. | King M., Brock G., Lundell C.Clearance of mucus by simulated cough. J. Appl. Physiol.58198517761782 |
| 19. | Puchelle E., Tournier J. M., Petit A.The frog palate for studying mucus transport velocity and mucociliary frequency. Eur. J. Respir. Dis.741983293303 |
| 20. | Woods R. E.What is a “pulmonary exacerbation” in cystic fibrosis. J. Pediatr.1111987841842 |
| 21. | Girod S., Zahm J. M., Plotkowski C., Beck G., Puchelle E.Role of the physicochemical properties of mucus in the protection of the respiratory epithelium. Eur. Respir. J.51992477487 |
| 22. | King M., Zahm J. M., Pierrot D., Vaquez-Girod S., Puchelle E.The role of the mucus gel viscosity, spinnability and adhesive properties in clearance by simulated cough. Biorheology261989737745 |
| 23. | King, M. 1980. Rheological requirements for optimal clearance of secretions: ciliary transport versus cough. Eur. J. Respir. Dis. 61(Suppl. 110): 39–45. |
| 24. | Puchelle E., Zahm J. M.Influence of rheological properties of human bronchial secretions on the ciliary beat frequency. Biorheology211984265272 |
| 25. | Girod S., Galabert C., Lecuire A., Zahm J. M., Puchelle E.Phospholipid composition and surface-active properties of tracheobronchial secretions from patients with cystic fibrosis and chronic obstructive pulmonary diseases. Pediatr. Pulmon.1319922227 |
| 26. | Zhang Y., Yankaskas J., Wilson J., Engelhardt J. E.In vivo analysis of fluid transport in cystic fibrosis airway epithelia in bronchial xenografts. Am. J. Physiol. 270(Cell Phyisol.)391996C1326C1335 |
| 27. | Tomkiewicz R. P., App E. M., Zayas J. G., Ramirez O., Church N., Boucher R. C., Knowles M. R., King M.Amiloride inhalation therapy in cystic fibrosis. Am. Rev. Respir. Dis.148199310021007 |
| 28. | Quinton P. M.Viscosity versus composition in airway pathology. Am. J. Respir. Crit. Care Med.149199467 |
| 29. | Stutts M. J., Schwab J. H., Chen M. G., Knowles M. R., Boucher R. C.Effects of Pseudomonas aeruginosa on bronchial epithelial ion transport. Am. Rev. Respir. Dis.13419861721 |
| 30. | Zahm J. M., Bonnet N., Pierrot D., Dupuit F., Puchelle E.Airway cAMP-activated chloride efflux is preferentially expressed in ciliated cells. Pediatr. Pulmon.12(Suppl.)1995194A |
| 31. | Kubesch P., Dork T., Wulbrand U., Kalin N., Neumann T., Wulf B., Geerlings H., Weissbrodt H., Von der Hardt H., Tummler B.Genetic determinants of airways' colonisation with Pseudomonas aeruginosa in cystic fibrosis. Lancet3411993189193 |
| 32. | Zielenski J., Tsui L. C.Cystic fibrosis: genotypic and phenotypic variations. Annu. Rev. Genet.291995777807 |
| 33. | Puchelle E., Jacquot J., Beck G., Zahm J. M., Galabert C.Rheological and transport properties of airway secretions in cystic fibrosis-relationships with the degree of infection and severity of the disease. Eur. J. Clin. Invest.151985389394 |
