Hereditary angioedema (HAE) is an inherited disorder characterized by recurrent episodes of severe soft tissue swelling involving skin and mucous membranes. In most cases, edema is caused by excessive formation of the vasoactive nanopeptide bradykinin (BK) due to dysregulation of the plasma kallikrein-kinin system (KKS). The KKS consists of the zymogens prekallikrein (PK) and factor XII (FXII) and the cofactor/substrate high-molecular-weight kininogen (HK). PK and FXII reciprocally convert each other to the proteases plasma kallikrein (PKa) and FXIIa. PKa cleaves HK after Lys362 and Arg371 to release bradykinin. In most cases of HAE, increased KKS activation is due to low plasma activity of C1-Inhibitor (C1-INH), the main regulator of PKa and FXIIa. However, ~10% of HAE patients have normal C1-INH activity (HAEnC1). A relationship between fibrinolysis and kinin formation was first described in 1971. The fibrinolytic protease plasmin activates FXII and cleaves HK, although it is unclear that this results in significant bradykinin generation. Recently mutations in the PLG gene encoding plasminogen (Lys311Glu) and the KNG1 gene encoding HK and the related protein low-molecular-weight kininogen (LK, Met361Lys) were identified in patients with HAEnC1. Here we describe how the fibrinolytic system contributes to angioedema in patients with these mutations. Human Glu-plasminogen (Plg), HK and LK were expressed in HEK293 cells. Lys311 in plasminogen was changed to glutamic acid, and Met361 in HK and LK were changed to lysine, by site directed mutagenesis. Wild type (Plg-Lys311) and variant (Plg-Glu311) plasminogen were converted to plasmin (Plm-Lys311 and Plm-Glu311) with urokinase. The capacity of plasmin to cleave HK and LK were studied with SDS-PAGE, western blots and kinin ELISA. Cleavage of wild type HK-Met361 and LK-Met361 and variant HK-Lys361 and LK-Lys361 by PKa and plasmin were assessed in a similar manner. The types of kinin released during reactions were determined by mass spectroscopy. Normal plasma was supplemented with Plg-Lys311 or Plg-Glu311, and tPA was added to generate plasmin. Bradykinin release was substantially greater with Plm-Glu311 than Plm-Lys311. This held when reactions were run in plasma lacking FXII or PK, indicating activation of the KKS was not responsible for kinin production. Using plasma-derived proteins, PKa released bradykinin 100-fold faster from HK than LK, and released bradykinin from HK ~50-fold faster than did plasmin. PKa and plasmin released bradykinin at comparable relatively slow rates from LK. Plm-Glu311 released bradykinin from HK 10-fold faster, and from LK 3-fold faster, than in reactions with Plm-Lys311. In experiments in which kininogen deficient plasma was supplemented with either HK or LK to a physiologic concentration, the majority of bradykinin released by Plm-Glu311 was from LK. PKa-catalyzed release of bradykinin from wild type HK-Met361 and variant HK-Lys361 comparably. The same held for LK-Met361 and LK-Lys361. However, plasmin released kinin from HK-Lys361 and LK-Lys361 3.3-fold and 13-fold faster than from HK-Met361 and LK-Met361. While the kinin released from HK-Met361 and LK-Met361 was bradykinin, plasmin released the decapeptide Lys-bradykinin (kallidin) from HK-Lys361 and LK-Lys36. Our studies indicate that angioedema in some patients with HAEnC1 may be due to fibrinolytic activity, and not to KKS activity. Both HK and LK may be substrates in such patients. As the plasma LK concentration is ~4-fold higher than HK, it may be the main pathologic kinin source. The plasmin Lys311Glu substitution converts the protease into a more efficient kininogenase. Fibrinolytic activity is normally brisk in the mouth, perhaps explaining why patients with the Lys311Glu mutation primarily suffer from oral-lingual angioedema. The findings raise the possibility that oral-lingual edema after tPA infusion or ACE inhibitor administration may also be due to plasmin-mediated cleavage of LK and HK. The Met361Lys substitution in HK and LK introduces a novel protease cleavage site immediately upstream from a normal PKa cleavage site. The novel site is preferred over the normal site by plasmin, but not PKa. Several treatments for HAE directed at neutralizing PKa or FXIIa activity have been developed. Our results suggest that such drugs may be less effective in patients with HAEnC1 in which disease mechanisms operate independently of the KKS.
Read full abstract