INTRODUCTION

TOP ABSTRACT INTRODUCTION CASE HISTORY DISCUSSION REFERENCES

Cilia are minute hairlike structures present on the surface of respiratory mucosa which sweep the respiratory tract clean by continuous beating at 10 to 18 Hz. The ultrastructural structure of cilia is conserved among species and at different levels of the airways (1). When there are ciliary ultrastructural abnormalities, ciliary dysfunction could result in development of bronchiectasis. This is exemplified by Kartagener's syndrome (sinusitis, bronchiectasis, and dextrocardia) which is the most obvious clinical expression of a spectrum of diseases caused by abnormal ciliary actions (2). Classically, patients with abnormal cilia suffer from early-onset recurrent respiratory infections leading to the development of cylindrical bronchiectasis in early childhood and then severe saccular bronchiectasis in adulthood (2).

A number of functional and structural ciliary defects have been identified in bronchiectasis and many apparently healthy individuals (3). Dyskinetic cilia are usually found in patients with primary ciliary dyskinesia although immotile cilia are sometimes encountered in this condition (2). Some patients with otherwise classic primary ciliary dyskinesia have normal ciliary beating and ultrastructure but abnormal orientation of the central microtubules (7). We have identified a previously undescribed ciliary ultrastructural defect in four patients who suffered from severe idiopathic bronchiectasis. This newly described defect could be the cause of bronchiectasis in these patients who had otherwise normal ciliary function and ultrastructure.

	CASE HISTORY

TOP ABSTRACT INTRODUCTION CASE HISTORY DISCUSSION REFERENCES

None of the patients (WYM, WWK, CWK, and KCH) had ever smoked and all suffered from early-onset bronchiectasis. The clinical and demographic details are shown in Table 1. All patients underwent extensive history evaluation, physical examination, saccharine test (10), and investigations which showed high-resolution computed tomography (HRCT)-proven idiopathic bronchiectasis (Figure 1) between July 1998 and September 1999 (11). There was no previous history of pertussis, measles, tuberculosis, or other significant conditions known to cause bronchiectasis. All four patients had regular sputum production, recurrent exacerbations, and occasional hemoptysis since their 20s. Some relevant information on the family members of these patients is shown in Table 2. A detailed inquiry revealed that none of the subjects had exposure to any occupational or environmental noxious agents. Patients 1, 2, and 3 have always been housewives since completion of secondary education. Patient 4 has worked as a clerk in a bank since he left college. Examination of the chest revealed bilateral lower zone crackles but no other abnormalities. Detailed general and systematic examination of the cardiovascular and abdominal systems were normal. Patients 1 and 2 did not suffer from nasal symptoms or other features of rhinosinusitis although Patients 3 and 4 had experienced sporadic nasal obstruction and rhinorrhea up to twice yearly for 3 and 14 yr respectively.

View larger version (117K): [in this window] [in a new window]   Figure 1.   High-resolution computed tomogram showing severe bronchiectatic changes in the lower lobes, and milder changes in the right middle lobe and lingula of Patient 1.

Lung function indices were measured using a SensorMedics 2200 Lung Function package (SensorMedics, Yorba Linda, CA) and standard clinical protocols by a trained technician (Table 1). Routine biochemistry including liver and renal function tests and hematology indices were normal. In addition, the following blood tests, which were negative or normal, were also performed: IgG for Pseudomonas pseudomallei, IgG subclasses, -1-antitrypsin level, viral titers, Aspergillus precipitins, complements 3 and 4, auto-antibodies (rheumatoid factor, antineutrophil cytoplasmic antibody [ANCA], antinuclear factor, and IgG against Ro, La, Jo₁ and smooth muscle), and serum IgG, IgA, IgE, and IgM levels.

Nasal mucociliary function was assessed by saccharine test as described previously (10). Briefly, a saccharine tablet (1 × 1 × 1 mm³) was inserted under direct vision with a pair of Tilley nasal dressing forceps to the medial aspect of the inferior turbinate of one nasal cavity. The time from saccharine placement until the subject, who sat forward quietly with the head bent forward, reported the first sensation of a sweet taste was measured to the nearest minute. The result was expressed as "nasal mucociliary clearance" (NMCC) time, which is normally between 20 and 60 min (10, 12). Patients 1 and 3 did not detect the taste after 20 and 30 min respectively and could not persevere with the saccharine test. Both of these patients could taste saccharine when placed on their tongue after the test was abandoned. Patients 2 and 4 had abnormally prolonged NMCC time (Table 1). Some of the patients' relatives consented to the saccharine test and their NMCC times are shown in Table 2.

Ciliary assessment using phase contrast and transmission electron microscopy (TEM) was performed when the patients were at steady state which was defined as "no significant alteration of 24 h sputum volume, FEV₁ and FVC, and in the absence of deterioration in respiratory symptoms for three consecutive weeks" (13). Nasal epithelium was obtained without anesthetics by using a cytology brush from the inferior turbinate of patients and then resuspended in 1.5 ml of medium 199 (Flow Laboratory, NY). The ciliated epithelium was examined using a Leica DM LB phase contrast microscope (with long working distance lenses) which had a warm stage maintained at 37° C (Leica, Wetzlar, Germany). The beating cilia, examined at ×400, were positioned to interrupt a light source and this frequency of interruption was conveyed to a MPV-COMBI photomultiplier (Leica) and a custom-made digital converter which translated this into ciliary beat frequency (Hz) as described previously (14). Light microscopy examination revealed normal beating cilia in three patients and slow beating cilia on Patient 1 (Table 1). No ciliary dyskinesia or immotility was observed.

The ciliary suspension was then fixed in 2.5% cacodylate-buffered glutaraldehyde (pH 7.2) and postfixed in 1% osmium tetroxide. This was followed by standard serial dehydration through alcohols and embedding in araldite. An ultrathin section (70 to 90 nm) through the central portion of each specimen was examined using TEM and a systematic evaluation of the ciliary cross section was performed including assessment on the central complex (central microtubules and central sheath), peripheral junctions (the radial spokes and nexin links), dynein arms, and peripheral doublets (Figure 2). The orientation of the ciliary central microtubules was assessed for each patient as described previously (7). Briefly, random sections of cilia were taken from each randomly chosen ciliated epithelial cell for further TEM examination. An image of these cilia was electronically captured and processed using an image analysis system (Improvision, London, UK). For each cilium, a line was drawn electronically through the central pairs of microtubules. The angle made by each of these lines to the horizontal axis was then measured. The standard deviation (SD) of these angles for the cilia from each of the epithelial cells was calculated. A mean SD was obtained from all the epithelial cells for each patient, which represented an index of ciliary central microtubular orientation (7).

View larger version (26K): [in this window] [in a new window]   Figure 2.   A schematic cross-sectional sketch of a normal ciliary shaft showing the "9 + 2" arrangement of microtubules, small amount of matrix, and ciliary membrane.

Altogether between 125 and 263 cilia from 6 to 8 epithelial cells were examined for the four patients (7). Between 2.3 and 12.4% of the cilia showed striking "cystlike" structures within the ciliary shafts (Figure 3). The ciliary ultrastructure and central microtubular orientation were otherwise normal in all four cases (Table 1). Longitudinal sections of these structures were obtained which suggest that these were of short dimension and predominantly situated at the bases of the cilia (Figure 4). In Patient 4, these cystic structures were also found within an ultrastructurally otherwise normal compound cilium (Figure 5).

View larger version (117K): [in this window] [in a new window]   View larger version (112K): [in this window] [in a new window]   Figure 3.   Transmission electron micrographs of cilia (cross section) obtained from (A) Patient 2 showing the presence of cystlike structures in the ciliary shaft with normal ciliary "9 + 2" microtubular arrangement. The central microtubules of 5 neighboring cilia had the same axes of orientation, as shown by the solid lines drawn through the central microtubules; (B) Patient 3 showing the lack of contents within these cystic structures.

View larger version (147K): [in this window] [in a new window]   View larger version (133K): [in this window] [in a new window]   Figure 4.   Transmission electron micrographs of cilia (longitudinal section) obtained from (A) Patient 2 showing the cystic structures encircling a "9 + 2" arrangement; and (B) Patient 3 showing the predominantly basal localization of these cystic structures (arrow points at plasma membrane).

View larger version (116K): [in this window] [in a new window]   Figure 5.   Transmission electron micrograph of the cross section of a compound cilium obtained from Patient 4, which also contains the "cystlike" structures.

	DISCUSSION

TOP ABSTRACT INTRODUCTION CASE HISTORY DISCUSSION REFERENCES

We have described the clinical and investigation profiles on four patients with severe and otherwise idiopathic bronchiectasis who had striking but previously undescribed ciliary ultrastructural abnormalities. These "cystlike" structures, some of which appeared to contain amorphous materials, although constituting between 2.3 to 12.4% of the cilia sampled, could severely hinder the overall mucociliary clearance in analogy to the disruption of a conveyor belt. Ciliary beat frequency was only slightly abnormal in Patient 1, and ciliary ultrastructure and central microtubular orientation were otherwise normal in all four patients. Light microscopy examination of cilia was normal despite the cross-sectional size of these defects. This might have been due to the predominantly basal location and the short dimension of these cystic structures (Figure 4). The presence of many normal cilia adjacent to these striking abnormalities strongly suggests that the latter are not artefacts. Although NMCC time was only obtained from two of the patients, they were abnormally prolonged. In the absence of other ciliary abnormalities, it is highly probable that this impairment in mucociliary clearance is caused by the presence of these "cystlike" structures in the ciliary shafts (Table 1). Only one of the four patients had a positive family history of bronchiectasis (Table 2), and most of these patients are female and appear to have developed symptoms in their twenties.

Ciliary shafts are usually very tightly packed structures which contain the membrane, a thin rim of matrix, and the "9 + 2" arrangement of microtubules. There are no known cellular organelles that resemble these cystlike abnormalities. Ciliary ultrastructure defects occur in bronchiectasis but have not been studied systematically with clinical correlation. Our preliminary study showed that microtubular defects occur in up to 34.1%, matrix abnormalities in 10.6%, and absence of one or both microtubules of the central pair in 8.3% of patients with bronchiectasis. However, the cilia of none of these patients displayed these cystic structures. Despite the extensive investigations into the origins of bronchiectasis, no other causes of bronchiectasis had been identified. Although this does not exclude "idiopathic" bronchiectasis as the underlying cause of the disease in these patients, repeated examination of the cilia obtained from all four patients showed that 2.3 to 12.4% of cilia displayed these defects. It is highly possible that these basally situated cystic structures could affect the beating not only of the affected but also of neighboring cilia owing to steric hindrance. It might be argued that this defect is still relatively infrequent and might not be the cause of, or a major contributing factor to, the development of bronchiectasis in these patients. It is possible that many of these defects are undetected as they are relatively short and could be missed on random sectioning. NMCC was impaired in two of the tested patients, which is important evidence for the presence of overall defective functioning of respiratory cilia (Table 1). Our reported defect could cause a significant disruption of mucociliary clearance, analogous to the overwhelming consequences of having numerous minor defects on a conveyor belt which could severely disrupt its functioning.

It can be argued that only nasal cilia were examined, which might not represent the tracheobronchial epithelium. However, ciliary ultrastructure is generally agreed to be conserved among all living species (1), and there is currently no evidence to suggest that nasal and bronchial epithelia have a different ultrastructure. Many researchers also used nasal cilia in the investigation of patients with suspected ciliary abnormalities (14, 15). The defect we found in our patients appears to be associated with early-onset severe bronchiectasis but not with other ciliary ultrastructural abnormalities, central microtubular disorientation, infertility, or other features of primary ciliary dyskinesia.

Primary ciliary dyskinesia is rare and has an estimated incidence of 1/15,000 to 1/35,000 (16). Kartagener's syndrome is most frequently associated with the absence of dynein arms (site of ATPase activity), and this was previously suggested to be an abnormality originating in the synthesis, degradation, or assembly of dynein subunits in ciliated structures (17). An impressive number of ciliary ultrastructural abnormalities has also been identified in primary ciliary dyskinesia and idiopathic bronchiectasis (4), in addition to the absence of dynein arms. These include: agenesis of cilia (5), absence of central microtubules or radial spokes (6), random ciliary orientation (7), bleb formation in the outer ciliary membrane (8), and transposition of ciliary microtubules (9). As ciliary ultrastructural assessment requires electron microscopy expertise and equipment, it is often not performed outside highly specialized centers which undoubtedly leads to underrecognition of these defects in bronchiectatic patients.

The exact origin of these cystlike structures remains obscure and there are no known organelles that resemble them. Large cilia have been described previously in the form of compound or fusion cilia. Compound cilia can further be subclassified as "bulging" and "adhesive" types. The former type comprises a small number of usually irregularly arranged microtubules embedded in a large amount of amorphous matrix within the ciliary shaft, whereas the latter type is characterized by an orderly arrangement of "9 + 2" structures within otherwise normal but enlarged ciliary shafts (18). While fusion, whether complete or partial, of neighboring cilia could lead to the formation of large cilia, most fusion cilia are not much larger than a single "9 + 2" size (19). Although membrane defects such as finlike extension from the ciliary membrane or defects of the shaft membrane have been described (19), our cystic structures do not resemble these. While defects of ciliary ultrastructures have been described, none of these defects has been evaluated for its impact on overall mucociliary clearance or ciliary motility in terms of beat frequency and character. Our study is therefore the first in associating an impaired NMCC with a newly described and ultrastructurally striking defect.

Current literature suggests that most patients with ciliary ultrastructural defects appear to have primary defects (3, 20, 21) despite in vitro data showing ciliary damage during bacterial and viral infections, or inflammation. In vitro studies have shown that ciliary dyskinesia and slowing can occur with experimental infection of respiratory mucosa by Pseudomonas aeruginosa (22) and Haemophilus influenzae (23). In addition, severe ultrastructural damage to respiratory mucosa also occurs in such experimental infections although ciliary ultrastructure has not been reported to develop any defects from these short-term experiments. Common cold or influenza could lead to sloughing of mucosa and damage of the respiratory mucosa but does not cause ultrastructural defects (24). In addition, chronic infective and inflammatory conditions such as cystic fibrosis (21) and asthma (3) are not associated with abnormal ciliary ultrastructure. The association with early- onset and otherwise idiopathic severe bronchiectasis suggests that this striking "cystlike ciliary shaft" could be a primary defect. Further studies should be performed to evaluate this defect and its impact on ciliary beating.