Bradykinin measurement by liquid chromatography tandem mass spectrometry in subjects with hereditary angioedema enhanced by cold activation

Abstract

Background: Bradykinin (BK), a 9-amino acid peptide, is a key mediator responsible for angioedema attacks in hereditary angioedema due to C1INH deficiency (HAE-C1INH). Current BK assays lack clinical utility due to the instability of BK (half-life < 30 seconds) and low pg/mL baseline levels.

Objective: We sought to develop a novel method to overcome the issues in current protease inhibitor-based methods for measuring endogenous BK.

Methods: Blood samples from subjects with HAE-C1INH and healthy volunteers were collected and subjected to cold activation for the contact system. Cold-induced BK was measured via LC-MS/MS. The protocol was developed and validated according to the US Food and Drug Administration Bioanalytical Method Validation guidelines. The procedure was optimized on specimen selection, collection, and process, as well as variable controls such as time windows, temperature, and storage conditions.

Results: EDTA whole blood samples without protease inhibitor were incubated at 4°C for 1 and 3 days. Total BK levels increased more than 100-fold in attack-free HAE-C1INH subjects compared with healthy volunteers (324.3 ± 54.7 ng/mL [n = 33] vs 2.3 ± 0.3 ng/mL [n = 43]; mean ± SEM, P < .001). The diagnostic sensitivity and specificity were 90.9% and 97.1%, respectively. BK levels highly correlated with plasma kallikrein activity in the same samples.

Conclusions: Whole blood under cold activation showed striking elevation of BK levels in subjects with HAE-C1INH, while minimally affecting healthy volunteers. The assay has been assessed for accuracy, precision and stability. It may serve as a reliable tool for HAE diagnosis and management.

Keywords: Hereditary angioedema; LC-MS/MS; angioedema; biomarker; bradykinin; cold activation; contact system; plasma kallikrein.

Fig 1

Cold activation preferably produced high levels of BK in whole blood samples from patients with HAE-C1INH. The validated LC-MS/MS method was used for BK measurements of EDTA whole blood samples from 6 HVs and 5 patients with HAE-C1INH incubated at RT or at 4°C for 24 hours, respectively. (A) In subjects with HAE-C1INH, the BK levels showed great differences between RT and 4°C incubation. (B) No significant differences were observed in HVs between RT and 4°C. Representative data from 3 independent experiments are shown.

Fig 2

Cold incubation consistently yielded total BK levels in EDTA whole blood, but not in citrate blood or plasma samples. BK levels in EDTA plasma showed far greater increases than expected, with plasma-to-whole blood ratios exceeding the theoretical proportion (approximately 1:0.55, reflecting the plasma fraction of whole blood) and exhibiting wider variabilities compared with EDTA whole blood after cold incubation. In addition, EDTA was favored over citrate, as the data showed that 24-hour cold incubation yielded high BK levels in EDTA, but low BK levels in citrate tubes. CV, Coefficient of variation.

Fig 3

Total BK delivers better diagnostic performance than individual peptides in HAE-C1INH. (A) Distribution profiles of individual BK metabolites (BK1-5, BK1-7, BK1-8, and BK1-9) varied among subjects. Most subjects exhibited peak total BK levels between cold activation for day 1 and day 3. (B) No repeated pattern of individual BK metabolite was found in HAE-C1INH samples. The plot of the total BK values was a better reflection of the cold activation of the contact system. (C) Receiver operating characteristic curve analysis (20 HAE-C1INH subjects and 20 HVs) showed that total BK had the highest area under the curve, indicating the best diagnostic value. AUC, Area under the curve.

Fig 4

Great blood sample stability after cold activation. Ethanol was used to stop cold activation by denaturing enzymatic proteins. BK and its metabolites remained stable in ethanol-treated blood samples for up to 4 weeks at 4°C (A) and up to 8 months at −80°C (B).

Fig 5

Cold-induced total BK levels and PKa activities in 43 HVs and 33 patients with HAE-C1INH in a validation cohort. (A) Both total BK concentrations and PKa were significantly increased in subjects with HAE-C1INH. Nearly all samples demonstrated changes in the same direction across control (B) and HAE (C) groups. Receiver operating characteristic analysis highlighted strong diagnostic performance for both BK and PKa assays (D). AUC, Area under the curve.