1. Some bronchiectatic patients persistently expectorate purulent secretions containing many polymorphonuclear leucocytes (PMN) and bacteria, suggesting ineffective microbial clearing from the lung. We have therefore studied several aspects of PMN recruitment to the lung by chemotaxis. 2. Circulating PMN from patients with bronchiectasis show a variable chemotactic response to the synthetic peptide N -formyl- l -methionyl- l -leucyl- l -phenylalanine. This is similar to the response of PMN from healthy controls and suggests that failure to clear microbial organisms is not due to defective chemotaxis as measured here. 3. Purulent sputum solphase demonstrates chemotactic activity against control PMN. This activity is greater ( P < 0.001) than that found with mucoid sputum solphase, which is consistent with the neutrophil content of the secretions in vivo. 4. Successful antibiotic therapy reduces the chemotactic activity of purulent sputum solphase from a median value of 183 (range 28–201) cells/high power field to 33.5 (range 0.7–64) cells/high power field ( P < 0.0025). This indicates that the chemotactic activity is largely associated with bacterial load.

1. Anti-elastase function in sputum sol-phase from patients with α 1 -proteinase inhibitor (α 1 PI) deficiency was compared with sol-phase from patients with cigarette smoke-induced bronchitis and emphysema. 2. Both α 1 PI (2 P < 0.01) and anti-leucoprotease (ALP) (2 P < 0.01) concentrations were lower in sol-phase from the α 1 PI-deficient group, although α 2 -macroglobulin (α 2 M) levels were similar. 3. There was no difference in α 1 PI function between the two groups, but the inhibitor was only ≃ 30% active. 4. The absolute neutrophil elastase (NE) inhibitory capacity was similar in both groups (median 185 μg of NE inhibited/ml of sputum, range 80–480, for the α 1 PI-deficient group; median 175, range 80–300, for the bronchitic group). A substantial proportion of NE inhibition in secretions could not be accounted for by the amount of α 1 PI, ALP and α 2 M present (median 74.8%, range 43.2–97.4, for α 1 PI-deficient sol-phase; median 50.0%, range 0–80.8, for bronchitic sol-phase). 5. Gel filtration of sol-phase demonstrated the presence of NE inhibition in the low molecular weight fractions which was markedly sensitive to changes in substrate concentration and ionic strength, in contrast to purified α 1 PI and ALP. 6. Sputum sol-phase from both groups failed to prevent hydrolysis of elastin–fluorescein or succinyltrialanyl- p -nitroanilide by NE completely during prolonged incubation in the presence of an excess of functional inhibitors. This was more apparent in secretions from subjects with α 1 PI deficiency and may explain why such patients have a more rapidly progressive form of emphysema.

1. The anti-elastase composition of bronchoalveolar lavage (BAL) fluid from (α 1 -proteinase inhibitor (α 1 PI) deficient and bronchitic patients was determined by immunological and functional assays, together with the cell profile of the BAL fluid. 2. α 1 PI, anti-leucoprotease and α 2 -macroglobuin were present in all the samples. BAL fluid α 1 PI concentrations were significantly lower in the group with serum α 1 PI deficiency. 3. Lavage fluid from α 1 PI deficient subjects inhibited less porcine pancreatic elastase than bronchitic BAL fluid ( P < 0.005). However, the α 1 PI was only about 50% active as an inhibitor in both groups. 4. There was no difference in the amount of neutrophil elastase (NE) inhibited per ml of lavage fluid or per mol of the measured inhibitors in the secretions, but both groups inhibited more enzyme than would be expected for these inhibitors (α 1 PI deficient: median 4.78 mol of NE/mol of known inhibitors, range 0.88–78.80; bronchitic: 1.14, 0.21–4.66), suggesting that an additional inhibitor is present. 5. The total leucocyte and neutrophil counts were elevated (2 P < 0.01) in the lavages of α 1 PI deficient patients, suggesting a greater potential elastase burden than subjects with normal α 1 PI.

1. Two-dimensional immunoelectrophoresis and conventional sodium dodecyl sulphate-polyacryl-amide gel electrophoresis was performed on various mixtures of purified α 1 -antitrypsin (α 1 AT) and leucocyte elastase (LE). 2. The results confirm that α 1 AT inhibits LE by the formation of enzyme-inhibitor complexes demonstrable by both techniques. 3. The complexes break down with time and are not affected by pH in the presence of excess α 1 AT. However, the breakdown is more rapid in the presence of excess enzyme only at pH values where LE remains active. The resultant products of the complex breakdown include inactivated LE and α 1 AT that has undergone limited proteolysis. 4. It is concluded that the presence or absence of α 1 AT-enzyme complexes as demonstrated by two-dimensional immunoelectrophoresis must be interpreted with caution when studying α 1 AT function in lung secretions. The absence of such complexes does not mean that previous interaction with enzyme has not occurred, thereby accounting for a reduction in α 1 AT inhibitory capacity.

1. The quantification of α 1 -antitrypsin (α 1 AT) by standard immunological techniques is altered by interaction of the protein with leucocyte elastase. 2. The results obtained for α 1 -antitrypsinleucocyte elastase mixtures in the presence of a functional excess of the inhibitor were relatively accurate for the first 6 h. However, continued incubation for more than 24 h led to a major over-estimation of the α 1 AT as the result of breakdown of the enzyme-inhibitor complex releasing a partially proteolysed form of the inhibitor. 3. In the presence of excess enzyme up to a twofold overestimation of α 1 AT occurred within 1 h and the degree of overestimation increased with time (up to threefold at 24 h). This was eventually associated with the presence of only a partially proteolysed form of α 1 AT (mol wt. ⋍ 50000). 4. Different results for each sample were obtained when different polyclonal antisera were used to quantify the α 1 AT. 5. Complete inactivation of α 1 AT by oxidation resulted in little change in the immunological quantification of the protein. However, further addition of H 2 O 2 led to a progressive underestimation of the α 1 AT. 6. The effect of physicochemical alteration on the immunological quantification of α 1 AT by different antisera should be borne in mind for all studies assessing this protein in lung secretions.

The pathogenesis of many acute and chronic lung diseases remains a mystery. However, recent years have seen a rapidly increasing interest in the role of proteolytic enzymes and their inhibitors in modifying the inflammatory, destructive and reparative changes that occur in the lung. Much of this interest owes its existence to two observations in the early 1960s: firstly, the recognition that subjects with an inherited deficiency of α 1 -antitrypsin (α 1 -AT; the main serum inhibitor of proteolytic enzymes) had a high incidence of pulmonary emphysema [1], and secondly the demonstration by Gross et al. [2] that a proteolytic enzyme (papain) was capable of producing lesions similar to emphysema in experimental animals. These observations ultimately led to the proteinase—anti-proteinase theory of emphysema, which predicts that a state of balance occurs in the healthy lung in which the proteolytic enzyme inhibitors functionally equal or exceed the enzymes. Destructive lung disease occurs when the enzymes functionally exceed the inhibitors such that they remain active within the lung, resulting in digestion of connective tissue. This general concept of a disturbed proteinase—anti-proteinase balance within the lung has been recently applied to many other lung diseases, and some will be mentioned later. However, it is in the study of chronic bronchitis and emphysema that the concept has become most well established.

1. Acute hypoxaemia has been reported to stimulate vasopressin release in animals. 2. Hypoxaemia induced by breathing 9·3% oxygen for 15–20 min failed to produce a rise in plasma arginine vasopressin concentration in six out of eight healthy human subjects. The two subjects who developed an increase in plasma arginine vasopressin concentration had a significant rise in serum cortisol. 3. Breathing 100% nitrogen until impairment of consciousness caused no rise in plasma arginine vasopressin concentration.

1. Oxygen breath tests were performed in seven patients with thyrotoxicosis. The inspired air was replaced by oxygen without their knowledge for 20 s and the fall in ventilation was recorded. 2. Ventilation fell significantly in all the patients within 40 s. The mean fall was 21·1% (se ± 1·7) below the control value. 3. β-Adrenoreceptor blockade with propranolol had little effect on this reduction in ventilation. The fall observed in five patients receiving propranolol was 19·6% (se ± 1·0). 4. After definitive treatment, which rendered the patients euthyroid, the observed reduction in ventilation after breathing oxygen was significantly less at 11·7% (se ± 1·2). 5. It is concluded that patients with thyrotoxicosis have an increased resting reflex hypoxic drive to respiration, when compared with the euthyroid state, and this is unaffected by β-adrenoreceptor blockade.

1. Oxygen breath tests were performed in nine patients with diffuse pulmonary infiltration whose resting arterial P o 2 ( P a,o 2 ) ranged from 8·9 kPa to 11·8 kPa. The inspired air was suddenly replaced with oxygen for 30 s and the percentage fall in ventilation over the last 10 s was measured. 2. P a,o 2 rose rapidly and exceeded 16 kPa within 20 s in each patient. 3. The ventilation fell significantly in seven of the nine patients. It is concluded that these seven patients had a demonstrable reflex hypoxic respiratory drive at rest. This tended to be greatest in patients with the lowest P a,o 2 . The percentage falls in ventilation observed were similar to those previously reported at comparable P a,o 2 levels in patients with chronic bronchitis. 4. The resting arterial P co 2 ( P a,co 2 ) ranged from 50 to 5·8 kPa. It did not change by more than 0·3 kPa during the oxygen breath tests in any patient.