Metallopeptidase activities in hereditary angioedema: effect of androgen prophylaxis on plasma aminopeptidase P

Abstract

Background: Aminopeptidase P (APP) plays an important role in the catabolism of kinins in human plasma, mostly for des-Arg(9)-bradykinin. Impaired degradation of this active bradykinin metabolite was found to be associated with a decreased APP activity in hypertensive patients who experienced angioedema while being treated with angiotensin I-converting enzyme inhibitors. The pathophysiology of hereditary angioedema is presently attributed only to a quantitative/qualitative C1 inhibitor (CI-INH) defect with increased bradykinin release.

Objectives: In the context of androgen prophylaxis, increased CI-INH function cannot fully explain protection from angioedema attacks alone because of the limited reversion of the CI-INH defects. Therefore we hypothesized that androgen prophylaxis could enhance plasma APP activity.

Methods: Patients with hereditary angioedema were investigated for plasma metallopeptidase activities responsible for kinin catabolism (APP, angiotensin I-converting enzyme, and carboxypeptidase N) and for CI-INH function in treated and untreated patients.

Results: APP activity was asymmetrically distributed in untreated patients (n = 147): the mean value was significantly lower than the value in a reference healthy and unmedicated population (n = 116; P < or = .001). Prophylaxis with androgen induced a significant increase in APP activity (P < or = .001), whereas it did not affect the other metallopeptidase activities. In both patient groups, APP activity showed a significant inverse relationship to disease severity (P < or = .001).

Conclusion: In addition to the effect on circulating CI-INH levels, the increase in APP levels brought on by androgens could contribute to a more effective control of the kinin accumulation considered to be responsible for the symptoms of angioedema.

FIG 1

Distribution of APP activity in patients with HAE with no prophylaxis (solid columns, n = 147) and in healthy individuals used as the reference group (open columns, n = 116).

FIG 2

Plasma APP activity (A) and CI-INH serpin function (B) in patients treated with long-term androgen prophylaxis (200 mg of danazol or 4 mg of stanozolol per 1–4 days; solid symbols) or not treated (open symbols). Data are presented in box whisker plots showing the median (horizontal bars within boxes), the inter-quartile range (boxes), and the 5th to 95th percentiles (vertical bars). Values greater than and less than these levels were dotted separately. Data comparisons between groups of patients taking androgen or not are presented according to the Mann-Whitney or Student nonparametric tests.

FIG 3

Plasma APP activity and CI-INH serpin function of patients 1 and 2 with HAE undergoing short-term prophylaxis with danazol (400 and 600 mg/d, respectively, for 10 days). Differences between values observed in the context of danazol treatment (solid columns) or no treatment (open columns) are indicated as increasing factors.

FIG 4

CI-INH function or APP activity versus disease severity. Box whisker plots of CI-INH function (A and B) or APP activity (C and D) represent the median (horizontal bars within boxes) and the interquartile range (boxes) with the 5th to 95th percentiles (vertical bars). Data were expressed for each disease severity class (1–2, severe disease; 3–5, mild to asymptomatic disease) in patients with no treatment (n = 157; Fig 4, A and C) and those with androgen prophylaxis (n = 59; Fig 4, B and D). Data comparisons for severity of disease between groups follow the equality of means with the Student t test (Fig 4, B) or the Mann-Whitney rank sum test (Fig 4, A, C, and D).

FIG 5

Risk estimate for the association of the biologic parameters with disease severity. Odds ratios were calculated from plasma CI-INH function and APP, CPN, and ACE activity. A, Untreated patients (n = 98). B, Treated patients (n = 36). Dashed lines represent the 95% Cis, and vertical bars represent the median values.