Main

Thirty years ago Laurell and Eriksson (1) observed that the major serum α1-globulin, α1AT, was deficient in a group of patients with severe obstructive lung disease. α1AT is a major antielastase in the lower respiratory tract, and any imbalance of neutrophil elastase and α1AT in favor of proteolytic activity may cause lung injury (2,3). Variants of α1AT are classified according to the Pi system (4). Individuals with homozygous PiZZ α1AT deficiency, the variant most commonly associated with clinical disease, have serum concentrations that are about 15% of the normal level, 1.40 g/L. Moderately reduced α1AT levels, about 35% of the normal, are found in PiSZ individuals who have a slightly increased risk of emphysema (5).

Children with α1AT-deficiency PiZZ and PiSZ were identified in the Swedish neonatal α1AT screening of 200 000 newborns in 1972-1974 (6). In the prospective follow-up, lung function was tested at 16 and 18 y of age. No differences in the occurrence of lung disease symptoms and only marginal deviation of values for certain lung function variables were found (7,8).

We hypothesized that compensatory increases in protease inhibitors or a decreased neutrophil leukocyte activity might favorably alter the protease-protease inhibitor balance during adolescence in α1AT-deficient subjects. This study includes the analyses of the protease inhibitors α2M, Achy, and SLPI, and a member of the lipocalin family located in the neutrophils, NGAL, as well as HEAT in 18-y-old subjects with α1AT deficiency compared with age-matched control subjects.

METHODS

At the age of 18 y, EDTA-plasma was drawn from 46 PiZZ, 22 PiSZ, and 41 control subjects. The control subjects were selected from a high school in Malmö and were age- and race-matched. The proportion of girls and boys in the groups was equal. The subjects had both a clinical checkup at the pediatric clinic and lung function tests performed at a specialized clinical physiologic laboratory. All of them were clinically healthy when the blood was drawn. The samples were centrifuged within 30 min, and plasma was stored within 24 h at -20°C until analyzed.

α2M and Achy concentrations were determined by electroimmunoassay using antibodies available at the laboratory; coefficient of variation was 5% (9). SLPI was measured by an ELISA coefficient of variation of 4.2% (10). NGAL was determined by an ELISA procedure, with a coefficient of variation of 5.4% (11). The elastase ELISA measures the quantity of elastase complexed with α1AT, HEAT, and a coefficient of variation of 7.5% (12). The concentrations of the protease inhibitors are given as percentages, 100% corresponding to 2.5 g of α2M/L, 0.5 g of Achy/L, and 26 µg of SLPI/L.

ANOVA (Bonferroni/Dunns test) was performed on an Apple Macintosh computer with the Statview 4.5 statistical package. The respective data were normally distributed, and the material was age- and sex-matched, hence no covariants were analyzed.

RESULTS

The plasma α2M, Achy, SLPI, NGAL, and HEAT concentrations of the α1AT-deficient and control individuals are given in Table 1. Significantly higher α2M concentrations were found in PIZZ (p < 0.0001) and PiSZ (p < 0.0001) subjects. The PiZZ and SZ subjects had low levels of NGAL (p < 0.0001), whereas only PiZZ subjects had low levels of HEAT (p < 0.0005) compared with the control individuals. Higher Achy concentrations were found in both PiZZ (p < 0.04) and PiSZ (p < 0.05) individuals. The concentrations were not affected by a delay up to 24 h before freezing; concentrations being measured on an adequate number of samples before and after up to 24 h at 20°C.

Table 1 The concentrations of α2M, Achy, SLPI, NGAL, and HEAT in PiZZ, PiSZ, and control (PiMM) individuals
Full size table

DISCUSSION

The natural history of liver and lung disease in subjects with severe and moderate α1AT deficiency up to 18 y of age is now properly known (8,13). None of the PiZZ and PiSZ adolescents followed up in the prospective Swedish study had clinical signs of chronic obstructive lung disease, and the lung function parameters were at most marginally different from age-matched control subjects (8). There is, however, in adulthood a large variation in the severity of lung disease both in smoking and nonsmoking individuals (14,15). Cigarette smoking is the important detrimental environmental factor in the PiZZ-deficiency state, reducing both quality of life and longevity substantially (14,16). Cigarette smoking may negatively affect the protease-protease inhibitor balance by oxidative inactivation (in vitro studies) of α1AT and SLPI and an increase of the number of leukocytes and the amount of leukocyte elastase. The present study and an earlier one indicate that an increased protease inhibitory capacity may protect the lung of children and adolescents with α1AT deficiency (17).

The quantitatively dominating leukocyte protease are elastase, NP4, and cathepsin G. NP4 was purified at about the same time in several laboratories, including ours, and was designated different names. It is identical to proteinase 3 (18). All three enzymes are located in the azurophil granules and are released in the active form to about the same extent on the stimulation of the leukocytes in vitro and in vivo (19,20). The enzymes show potent nonspecific proteolytic activity with the capacity to degrade structured and soluble proteins in tissues and body fluids (2,21–23). Both elastase and NP4 have been implicated as important pathogenic factors in lung emphysema (2,24). The elastolytic activity of elastase is also augmented by cathepsin G (25). In this study we determined NGAL as an indirect measurement of the above mentioned proteolytic enzymes as well as elastase complexed with α1AT. Phagocytosis experiments have indicated parallel release of NGAL, NP4, and elastase (11). Thus the significantly decreased levels of NGAL found in PiZZ and PiSZ subjects may indicate a down regulation of elastase and NP4 synthesis. However, a decreased concentration of HEAT was only found in the PiZZ individuals. These observations must be confirmed and further studied. If correct, the decreased NGAL and HEAT levels may indirectly indicate a positive influence on the protease-antiprotease balance in the α1AT-deficiency state. These findings also raise the question whether the number of polymorphonuclear leukocytes, not analyzed in this study, are similar in each group. The polymorphonuclear count was at age 12 y, 2.8 ± 1.1 × 109/L in PiZZ (n = 62) and 2.8 ± 1.0 × 109/L in PiSZ (n = 31) children, no control subjects were tested (T. Sveger, unpublished results).

α1-AT is normally the major inhibitor of leukocyte proteases on the alveolar and interstitial levels with the capacity to block the activity of all three major proteolytic enzymes, whereas SLPI is the dominating inhibitor on the mucus membranes of the respiratory tract (26,27). It is synthesized and secreted by nonciliated cells of the respiratory epithelium (27). SLPI protects the ciliated epithelium from degradation and may also have a protective role in the interstitial lung tissue. Furthermore, SLPI, but not α1AT, has been shown to inhibit elastin-bound elastase (28). Normal SLPI levels were found in α1AT-deficient subjects, and it is possible that SLPI is of particular importance in lung inflammation when large amounts are found even in the peripheral blood and very likely in the interstitial fluids (29). Achy is regarded as the main inhibitor of cathepsin G (30). The elastolytic capacity of cathepsin G is low. It has, however, been reported to enhance the elastolytic capacity of neutrophil elastase in solubilizing human lung elastin 2-5 times in vitro. With this perspective our finding of a marginally increased Achy concentration in α1AT-deficient subjects may decrease the risk of elastin damage. α2M, a major plasma protease inhibitor, is synthesized in the liver and locally by macrophages and fibroblasts. The proteases cleave peptide bonds in the "bait region" of the molecule, and the conformationally changed protein is rapidly eliminated by an α2M receptor (31). A passive transfer of elastase from α1AT to α2M and from SLPI to α1AT may occur in the interstitial tissues increasing the amount of free SLPI and α1AT available for further elastase inhibition. Because α2M is a large molecule (725 kD) about 70% occurs intravascularly in contrast to α1AT (53 kD), where about 60% exists intercellularly. Children and adolescents normally have higher levels of α2M than do adults, being 170% in 18 y olds. In α1AT deficiency the importance of α2M even intercellularly increases, and in this respect the considerably higher concentration of α2M found in α1AT-deficient subjects during childhood and adolescence should be of importance (17). The protease inhibitory role of α2M is well established. However, α2M may also serve as a proteinase-activated sensor for situations requiring coordinated cellular responses (32). It has been shown that administration of complexed α2M to macrophages results in various biologic effects, one of which being neutrophil proteinase production by murine macrophages (33). Engagement of the receptor for the α2M-protease complex resulted in decreased secretion of NP by the murine macrophages even when basal secretion was low.

α1AT replacement therapy is used for the treatment of severe α1AT deficiency to slow down the progression of emphysema (34). Its clinical efficacy has so far not been proved. High doses of α1AT are given i.v. every 4 wk, giving plasma α1AT concentrations ranging from 10 to 0.8 g/L during a 3-wk period. In view of our present findings the augmentation therapy may have potential drawbacks. The unphysiologically high α1AT levels may upset the compensatory up-regulation of α2M and down-regulation of leukocyte activity. In addition, an increased α1AT-elastase complex concentration may serve as a chemoattractant and increase both the influx and secretion of neutrophils (35). These complexes may also activate the serpinenzyme complex receptor by which more α1AT PiZZ molecule accumulates within the hepatocytes, increasing the risk of liver damage (36).

In conclusion, 18-y-old individuals with α1AT deficiency have increased concentrations of α2M and Achy combined with decreased levels of lipocalin and elastase. In this way the protease-protease inhibitory balance may be positively affected in the α1AT-deficiency state.