Bronchial Mucus Hypersecretion in Acute Quadriplegia Macromolecular Yields and Glycoconjugate Composition1-3
K. RAMAKRISHNAN BHASKAR, ROBERT BROWN, DONNA DEFEUDIS O'SULLIVAN, STEPHEN MELIA, MARIE DUGGAN, LYNNE REID, and the Respiratory Therapists at the BrocktonlWest Roxbury Department of Veterans Affairs Medical Center
Introduction
Quadriplegia resulting from accidents causing severe spinal cord injury is associated with paralysis of muscles of respiration making it difficult for patients to clear mucus from their lungs (1). In such patients the clinical presentation of excess mucus is most commonly due to acute bronchitis or pneumonia, but irritation from endotracheal or tracheostomy tubes must be considered. We have noted that even in the absence of these provocative factors, about one of five acute quadriplegics develops within days of injury a syndrome characterized byunexplained production of excessive and tenacious mucus. In affected patients, mucus obtained by assisted cough and/or aspiration may exceed 1 L per day, in striking contrast to unaffected persons in whom mucus production is minimal and not a clinical problem. Accumulation of mucus causes airway obstruction, atelectasis, and respiratory failure. In many, the complications prove fatal, whereas in others mucus hypersecretion resolves spontaneously after weeks to a few months. The nature and amount of airway mucus production in quadriplegia has not been studied. Our concern was to establish whether the excessive amount of mucus produced in the airways of such patients is qualitativelyabnormal. An idiopathic condition known as bronchorrhea is also characterized by copious amounts of sputum (2); the macromolecular yields from such sputum are lower than from other hypersecretory diseases such as chronic bronchitis and cystic fibrosis (3). Is the macromolecular yield from sputum of quadriplegic patients similar to the latter or to bronchorrhea? Our recent examination of individual bronchial aspirates from healthy volunteers using density gradient methods has shown that epithelial glycoprotein of typical buoyant den640
SUMMARY In acute quadriplegia we have noted that about one in five patients develops unexplained production of markedly excessive and tenacious bronchial mucus. Spontaneous recovery from mucus hypersecretion usually occurs Within weeks to months. Mucus samples collected from 12 patients have been found to be abnormal. Macromolecular contents of single aspirates yielded as much as 500 mg. Analytical Ultracentrifuge analysis showed the mucus to contain considerable epithelial glycoprotein (GP) of typical buoyant density; Its amino acid and carbohydrate compositions were characteristic of the GP from hypersecretory bronchial mucus such as In chronic bronchitis and cystic fibrosis. In five patients studied after recovery from hypersecretion, there tended to be relatively less GP. The mucus samples contained a high density glycoconjugate (GC): this had sugars of GP but also reacted positively with a monoclonal antibody to keratan sulfate. Its amino acid composition was different from that of GP: threonine was lower and glycine was higher than in GP.In mucus from one patient who died, chondroitin sulfate ABC and hyaluronic acid were Identified as well. This suggests proteoglycans are involved In the pathophysiology of mucus hypersecretion. The sudden onset and spontaneous recovery of hypersecretion suggests that it is not due to gland hypertrophy. Wespeculate that in acute quadriplegia it is due to disturbed neuronal control of bronchial mucous gland secretion, perhaps related to initial dlsappeerance and later reappearAM REV RESPIR DIS 1991; 143:640-648 ance of peripheral sympathetic nervous system tone.
sity and composition is virtually absent in normal secretion; the glycoconjugate has higher buoyant density and features of both glycoprotein and proteoglycan (4). Is the mucus from the quadriplegic patients similar to such "normal" aspirates or to mucus from hypersecretory diseases such as chronic bronchitis and cysticfibrosis? Is the composition of mucus similar in all patients who develop excessive secretion (hypersecretion)? In patients in whom hypersecretionresolves, does the mucus change in composition as the amount decreases? We report here analysis of the mucus produced in the lungs of quadriplegic patients in the days and weeks after injury and also after recovery from hypersecretion. Isolation and characterization of lipids from the mucus is reported separately (5). Methods Patients Twelve male acute quadriplegic patients admitted to the Brockton/West Roxbury VA Medical Center were included in this study
(table 1). All had clinical evidence of mucus hypersecretion at the time first collections were obtained for analysis.
Collection of Samples Mucus samples were collected from patients during routine bronchial toilet by assisted cough, nasotracheal aspiration, bronchoscopy,or aspiration through their tracheostomies. From each of the five patients studied after recovery from mucus hypersecretion, one
(Received in original form January 26, 1990 and in revised form October 23, 1990) I From the Department of Pathology, Children's Hospital, and the Department of Medicine, Harvard Medical School, Boston, and the Brockton/West Roxbury Department of VeteransAffairs Medical Center, West Roxbury, Massachusetts. 2 Supported by the Spinal Cord Research Foundation, Department of Veterans Affairs Merit Review Funding, and by Grant HL-19170 from: the National Heart, Lung, and Blood Institute. J Correspondence and requests for reprints should be addressed to Dr. Robert Brown, Chief, Pulmonary Section, Department of Veterans Affairs Medical Center, 1400 VFW Parkway, West Roxbury, MA 02132.
641
MUCUS HYPERSECRETION IN ACUTE QUADRIPLEGIA
TABLE 1 CLINICAL INFORMATION OF QUADRIPLEGIC PATIENTS INCLUDED IN A STUDY OF MUCUS HYPERSECRETION AND CLINICAL DATA ON THE SPINAL CORD Date of
Patient No.
Injection
Tracheostomy
Level of Inj.
Clinical Course
Smoking History
1 2 3 4 5 6 7 8 9 10 11 12
7/14/85 4/17/86 7/05/83 8/04/84 3/14/85 6/14/84 2/28/87 10/23/83 10/23/83 3/18/84 7/17/84 12/02/83
7/14/85 4/19/86 7/15/83 8/15/84 4/19/85 7/03/84 3/24/87 1/07/86
C 4-5 C 2, complete C 7-8, complete C 4-5, complete C 6-7, complete C 4-5 C 4-5 C4 C 5-6, complete C 4, complete C6 C5, complete
Recovered, trach. closed Recovered Recovered, trach. closed, 11/83 Recovered Recovered Recovered, trach. closed Recovered Died, 8/2/86 Died, 6/29/84 Died, 1/7/85 Died Died, 8/12/84
5 pack-years Stopped smoking, 10 yr 1 ppd years? None 25 pack-years 1 ppd years? Pipe, cigars Heavy smoker 100 pack-years 40 pack-years 40 pack-years 40 pack-years
3/28/84 NAt 1/12/84
• Patient admitted to WRVA on 1/10/84 with tracheostomy plugged in; tracheostomy revised on 4/4/84. Not applicable; patient had endotracheal tube put in on 7/17/84.
t
sample was obtained. Patients 1, 3, and 6 had their tracheostomies closed; samples from them wereobtained through cough at the time of their visit to the hospital for routine checkups. Samples from Patients 4 and 5 were obtained through their tracheostomies, which were left in place for reasons unrelated to excessivemucus in the airways. Wedid not analyze any specimens from patients in whom respiratory infections were apparent after careful clinical and radiologic assessment. Specifically, specimens were used for study only when patients had no fever, no elevation in white blood cell count, nonpurulent gross appearance of sputum, and no chest radiographic abnormalities consistent with pneumonia. These specimens were used only for diagnostic purposes (e.g., culture and sensitivity).
Storage of Samples Phosphate-chloride buffer (pH, 6.8), containing in addition to EDTA and sodium azide the protease inhibitors phenyl methyl sulfonyl fluoride (PMSF) and N-ethyl maleimide, was added to the samples, which were then stored frozen until use. Determination of Macromolecular Yields Samples were dialyzed for 24 h in the cold (40 C) against severalchanges of distilled water and lyophilized. Macromolecular yields are reported as milligrams per sample and in selected cases as milligrams per milliliter of sputum. Examination in the Analytical Ultracentrifuge Representative mucus samples (at least two from each patient) were examined in a cesium bromide (CsBr) gradient in a Beckman model E analytical ultracentrifuge (Beckman Instruments, Schiller Park, IL) as described in our earlier publications (6, 7). Mucus samples were dispersed at a uniform concentration of about 4 mg/ml in phosphate chloride buffer containing CsBr (44OJo wt/wt) to give
a solution density corresponding to the buoyant density of bronchial epithelial glycoproteins (7). Equilibrium density gradient patterns wererecorded with Schlierenoptics. Mucous solutions were examined in either single-sectoror double-sector cells.In doublesector cells, one sector is filled with the mucous solution and the other with CsBr solution of weight fraction (and hence density) identical to that used for the mucous solution. In one case, an aliquot of the mucous solution was also examined in the presence of added DNA to establish its (DNA) level of detectability in the density gradient analysis.
Isolation of Glycoconjugates Glycoconjugates were separated from lipids and proteins by preparative density gradient ultracentrifugation (DGD) in a CsBr gradient followingprocedures described previously (7). The glycoconjugate fraction was subjected to further DGD in a gradient of cesium chloride (CsCI) to isolate typical epithelial glycoprotein free from any noncovalently linked protein and the high density glycoconjugates (figure 1).The glycoprotein preparations were monitored for purity in the analytical ultracentrifuge, and additional DGD in a CsCI gradient containing 4 M guanidinium hydrochloride was employed when necessary to obtain pure preparations for chemical analysis. Characterization of Glycoconjugates Carbohydrates were analyzed by gas liquid chromatography of trimethylsilyl derivatives following the procedures of Clamp and coworkers (8). Amino acid analysis was carried out on samples hydrolyzed under vacuum for 24 h at 1100 C according to the method of Moore and Stein (9). Oligosaccharides were cleaved by treating with 1 M NaBH 4 in 0.05 M NaOH at 45 0 C for 3 days (10). The mixture was neutralized with 4 M acetic acid, passed through a column of DowexAG 50WX8, 100 to 200 mesh to remove Na', and
evaporated to dryness with methanol (three times) to removeexcessmethylborate. The residue was dissolved in 200 IIIof 50% methanol/water, applied to Rediplate silica gel G plates (Fisher Scientific, Springfield, NJ), and developed in butanol:acetic acid:water 3:3:2. Glucose polymers obtained from Karo syrup, ranging in size from 3 to 15sugar residues were used as markers for size (11). The plates weresprayed with orcinol and heated at 1400 C to reveal the oligosaccharides. Glycosaminoglycans were identified by two-dimensional electrophoresisfollowingthe procedure of Schmid and coworkers (12). Glycosaminoglycans were liberated by digestion with papain and pronase for a total of 72 h, and this digest was subjected to electrophoresis on cellulose acetate squares 10by 10em, in the first dimension in 0.1M pyridineformate buffer (pH, 3.0) at 20 rnA for 70 min, and in the second dimension at right angles to the first, in 0.1 M Ba(Ach buffer (pH, 8.0)
MUCUS
+ buffer + Gs8r
IDGU
(38% w/wl
at 850009 for ""?2h
I Top holf
I Bot tom half
I
I
Lipids, proteins including serum type
Glycoconjugoles ~40
K. RAMAKRISHNAN BHASKAR, ROBERT BROWN, DONNA DEFEUDIS O'SULLIVAN, STEPHEN MELIA, MARIE DUGGAN, LYNNE REID, and the Respiratory Therapists at the BrocktonlWest Roxbury Department of Veterans Affairs Medical Center
Introduction
Quadriplegia resulting from accidents causing severe spinal cord injury is associated with paralysis of muscles of respiration making it difficult for patients to clear mucus from their lungs (1). In such patients the clinical presentation of excess mucus is most commonly due to acute bronchitis or pneumonia, but irritation from endotracheal or tracheostomy tubes must be considered. We have noted that even in the absence of these provocative factors, about one of five acute quadriplegics develops within days of injury a syndrome characterized byunexplained production of excessive and tenacious mucus. In affected patients, mucus obtained by assisted cough and/or aspiration may exceed 1 L per day, in striking contrast to unaffected persons in whom mucus production is minimal and not a clinical problem. Accumulation of mucus causes airway obstruction, atelectasis, and respiratory failure. In many, the complications prove fatal, whereas in others mucus hypersecretion resolves spontaneously after weeks to a few months. The nature and amount of airway mucus production in quadriplegia has not been studied. Our concern was to establish whether the excessive amount of mucus produced in the airways of such patients is qualitativelyabnormal. An idiopathic condition known as bronchorrhea is also characterized by copious amounts of sputum (2); the macromolecular yields from such sputum are lower than from other hypersecretory diseases such as chronic bronchitis and cystic fibrosis (3). Is the macromolecular yield from sputum of quadriplegic patients similar to the latter or to bronchorrhea? Our recent examination of individual bronchial aspirates from healthy volunteers using density gradient methods has shown that epithelial glycoprotein of typical buoyant den640
SUMMARY In acute quadriplegia we have noted that about one in five patients develops unexplained production of markedly excessive and tenacious bronchial mucus. Spontaneous recovery from mucus hypersecretion usually occurs Within weeks to months. Mucus samples collected from 12 patients have been found to be abnormal. Macromolecular contents of single aspirates yielded as much as 500 mg. Analytical Ultracentrifuge analysis showed the mucus to contain considerable epithelial glycoprotein (GP) of typical buoyant density; Its amino acid and carbohydrate compositions were characteristic of the GP from hypersecretory bronchial mucus such as In chronic bronchitis and cystic fibrosis. In five patients studied after recovery from hypersecretion, there tended to be relatively less GP. The mucus samples contained a high density glycoconjugate (GC): this had sugars of GP but also reacted positively with a monoclonal antibody to keratan sulfate. Its amino acid composition was different from that of GP: threonine was lower and glycine was higher than in GP.In mucus from one patient who died, chondroitin sulfate ABC and hyaluronic acid were Identified as well. This suggests proteoglycans are involved In the pathophysiology of mucus hypersecretion. The sudden onset and spontaneous recovery of hypersecretion suggests that it is not due to gland hypertrophy. Wespeculate that in acute quadriplegia it is due to disturbed neuronal control of bronchial mucous gland secretion, perhaps related to initial dlsappeerance and later reappearAM REV RESPIR DIS 1991; 143:640-648 ance of peripheral sympathetic nervous system tone.
sity and composition is virtually absent in normal secretion; the glycoconjugate has higher buoyant density and features of both glycoprotein and proteoglycan (4). Is the mucus from the quadriplegic patients similar to such "normal" aspirates or to mucus from hypersecretory diseases such as chronic bronchitis and cysticfibrosis? Is the composition of mucus similar in all patients who develop excessive secretion (hypersecretion)? In patients in whom hypersecretionresolves, does the mucus change in composition as the amount decreases? We report here analysis of the mucus produced in the lungs of quadriplegic patients in the days and weeks after injury and also after recovery from hypersecretion. Isolation and characterization of lipids from the mucus is reported separately (5). Methods Patients Twelve male acute quadriplegic patients admitted to the Brockton/West Roxbury VA Medical Center were included in this study
(table 1). All had clinical evidence of mucus hypersecretion at the time first collections were obtained for analysis.
Collection of Samples Mucus samples were collected from patients during routine bronchial toilet by assisted cough, nasotracheal aspiration, bronchoscopy,or aspiration through their tracheostomies. From each of the five patients studied after recovery from mucus hypersecretion, one
(Received in original form January 26, 1990 and in revised form October 23, 1990) I From the Department of Pathology, Children's Hospital, and the Department of Medicine, Harvard Medical School, Boston, and the Brockton/West Roxbury Department of VeteransAffairs Medical Center, West Roxbury, Massachusetts. 2 Supported by the Spinal Cord Research Foundation, Department of Veterans Affairs Merit Review Funding, and by Grant HL-19170 from: the National Heart, Lung, and Blood Institute. J Correspondence and requests for reprints should be addressed to Dr. Robert Brown, Chief, Pulmonary Section, Department of Veterans Affairs Medical Center, 1400 VFW Parkway, West Roxbury, MA 02132.
641
MUCUS HYPERSECRETION IN ACUTE QUADRIPLEGIA
TABLE 1 CLINICAL INFORMATION OF QUADRIPLEGIC PATIENTS INCLUDED IN A STUDY OF MUCUS HYPERSECRETION AND CLINICAL DATA ON THE SPINAL CORD Date of
Patient No.
Injection
Tracheostomy
Level of Inj.
Clinical Course
Smoking History
1 2 3 4 5 6 7 8 9 10 11 12
7/14/85 4/17/86 7/05/83 8/04/84 3/14/85 6/14/84 2/28/87 10/23/83 10/23/83 3/18/84 7/17/84 12/02/83
7/14/85 4/19/86 7/15/83 8/15/84 4/19/85 7/03/84 3/24/87 1/07/86
C 4-5 C 2, complete C 7-8, complete C 4-5, complete C 6-7, complete C 4-5 C 4-5 C4 C 5-6, complete C 4, complete C6 C5, complete
Recovered, trach. closed Recovered Recovered, trach. closed, 11/83 Recovered Recovered Recovered, trach. closed Recovered Died, 8/2/86 Died, 6/29/84 Died, 1/7/85 Died Died, 8/12/84
5 pack-years Stopped smoking, 10 yr 1 ppd years? None 25 pack-years 1 ppd years? Pipe, cigars Heavy smoker 100 pack-years 40 pack-years 40 pack-years 40 pack-years
3/28/84 NAt 1/12/84
• Patient admitted to WRVA on 1/10/84 with tracheostomy plugged in; tracheostomy revised on 4/4/84. Not applicable; patient had endotracheal tube put in on 7/17/84.
t
sample was obtained. Patients 1, 3, and 6 had their tracheostomies closed; samples from them wereobtained through cough at the time of their visit to the hospital for routine checkups. Samples from Patients 4 and 5 were obtained through their tracheostomies, which were left in place for reasons unrelated to excessivemucus in the airways. Wedid not analyze any specimens from patients in whom respiratory infections were apparent after careful clinical and radiologic assessment. Specifically, specimens were used for study only when patients had no fever, no elevation in white blood cell count, nonpurulent gross appearance of sputum, and no chest radiographic abnormalities consistent with pneumonia. These specimens were used only for diagnostic purposes (e.g., culture and sensitivity).
Storage of Samples Phosphate-chloride buffer (pH, 6.8), containing in addition to EDTA and sodium azide the protease inhibitors phenyl methyl sulfonyl fluoride (PMSF) and N-ethyl maleimide, was added to the samples, which were then stored frozen until use. Determination of Macromolecular Yields Samples were dialyzed for 24 h in the cold (40 C) against severalchanges of distilled water and lyophilized. Macromolecular yields are reported as milligrams per sample and in selected cases as milligrams per milliliter of sputum. Examination in the Analytical Ultracentrifuge Representative mucus samples (at least two from each patient) were examined in a cesium bromide (CsBr) gradient in a Beckman model E analytical ultracentrifuge (Beckman Instruments, Schiller Park, IL) as described in our earlier publications (6, 7). Mucus samples were dispersed at a uniform concentration of about 4 mg/ml in phosphate chloride buffer containing CsBr (44OJo wt/wt) to give
a solution density corresponding to the buoyant density of bronchial epithelial glycoproteins (7). Equilibrium density gradient patterns wererecorded with Schlierenoptics. Mucous solutions were examined in either single-sectoror double-sector cells.In doublesector cells, one sector is filled with the mucous solution and the other with CsBr solution of weight fraction (and hence density) identical to that used for the mucous solution. In one case, an aliquot of the mucous solution was also examined in the presence of added DNA to establish its (DNA) level of detectability in the density gradient analysis.
Isolation of Glycoconjugates Glycoconjugates were separated from lipids and proteins by preparative density gradient ultracentrifugation (DGD) in a CsBr gradient followingprocedures described previously (7). The glycoconjugate fraction was subjected to further DGD in a gradient of cesium chloride (CsCI) to isolate typical epithelial glycoprotein free from any noncovalently linked protein and the high density glycoconjugates (figure 1).The glycoprotein preparations were monitored for purity in the analytical ultracentrifuge, and additional DGD in a CsCI gradient containing 4 M guanidinium hydrochloride was employed when necessary to obtain pure preparations for chemical analysis. Characterization of Glycoconjugates Carbohydrates were analyzed by gas liquid chromatography of trimethylsilyl derivatives following the procedures of Clamp and coworkers (8). Amino acid analysis was carried out on samples hydrolyzed under vacuum for 24 h at 1100 C according to the method of Moore and Stein (9). Oligosaccharides were cleaved by treating with 1 M NaBH 4 in 0.05 M NaOH at 45 0 C for 3 days (10). The mixture was neutralized with 4 M acetic acid, passed through a column of DowexAG 50WX8, 100 to 200 mesh to remove Na', and
evaporated to dryness with methanol (three times) to removeexcessmethylborate. The residue was dissolved in 200 IIIof 50% methanol/water, applied to Rediplate silica gel G plates (Fisher Scientific, Springfield, NJ), and developed in butanol:acetic acid:water 3:3:2. Glucose polymers obtained from Karo syrup, ranging in size from 3 to 15sugar residues were used as markers for size (11). The plates weresprayed with orcinol and heated at 1400 C to reveal the oligosaccharides. Glycosaminoglycans were identified by two-dimensional electrophoresisfollowingthe procedure of Schmid and coworkers (12). Glycosaminoglycans were liberated by digestion with papain and pronase for a total of 72 h, and this digest was subjected to electrophoresis on cellulose acetate squares 10by 10em, in the first dimension in 0.1M pyridineformate buffer (pH, 3.0) at 20 rnA for 70 min, and in the second dimension at right angles to the first, in 0.1 M Ba(Ach buffer (pH, 8.0)
MUCUS
+ buffer + Gs8r
IDGU
(38% w/wl
at 850009 for ""?2h
I Top holf
I Bot tom half
I
I
Lipids, proteins including serum type
Glycoconjugoles ~40
