- Email: [email protected]
The pathophysiology of hereditary angioedema
Clinical Immunology 114 (2005) 3 – 9 www.elsevier.com/locate/yclim
Short Analytical Review
The pathophysiology of hereditary angioedema Alvin E. Davis III * Department of Pediatrics, CBR Institute for Biomedical Research, Harvard Medical School, Boston, MA 02115, USA Received 13 April 2004; accepted 20 May 2004 Available online 2 July 2004
Abstract Hereditary angioedema (HAE), characterized by recurrent episodes of angioedema involving the skin, or the mucosa of the upper respiratory or the gastrointestinal tracts, results from heterozygosity for deficiency of the serine proteinase inhibitor (serpin), C1 inhibitor (C1INH). The primary biological role of C1INH is to regulate activation of the complement system, the contact system, and the intrinsic coagulation system. During attacks of angioedema, together with decreasing levels of C1INH, the complement and contact systems are activated: C2 and C4 levels fall and high molecular weight kininogen is cleaved. Although previous data suggested that symptoms in HAE might be mediated via complement system activation, a combination of recent clinical data, in vitro studies, and analysis of C1INH-deficient mice all indicate that the major mediator of angioedema is bradykinin: (1) a vascular permeability enhancing factor can be generated in vitro in C1INH-depleted, C2-deficient plasma, but not from C1INH-depleted, contact system-deficient plasma; this factor was identified by sequence analysis as bradykinin; (2) bradykinin can be detected in the plasma of HAE patients during attacks of angioedema; (3) in several members of one family, expression of a C1INH variant that inhibits contact system proteases but has defective inhibition of C1r and C1s does not result in HAE; (4) C1INH-deficient (C1INH/) mice have a defect in vascular permeability that is suppressed by treatment with specific plasma kallikrein inhibitors and by bradykinin type 2 receptor (Bk2R) antagonists, and is eliminated in C1INH/, Bk2R/ double-deficient mice. D 2004 Elsevier Inc. All rights reserved. Keywords: Hereditary angioedema; C1 inhibitor; Mucosa
Introduction Hereditary angioedema Hereditary angioedema (HAE) results from deficiency of the plasma protease inhibitor, C1 inhibitor (C1INH). Individuals with HAE are heterozygous for deficiency, which results in autosomal dominant inheritance. Complete deficiency has never been reported. Absence of expression from one allele, which results simply in decreased expression of C1INH in the plasma, is called type 1 HAE, while expression of a dysfunctional C1INH protein, together with decreased levels of normal protein, is termed type 2 HAE. C1INH function C1INH is a serpin. Most, but not all members of this group of proteins, are protease inhibitors; they share amino * Department of Pediatrics, CBR Institute for Biomedical Research, Harvard Medical School, 800 Huntington Avenue, Boston, MA 02115. Fax: +1-617-278-3490. E-mail address: [email protected]. 1521-6616/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.clim.2004.05.007
acid sequence homology and a similar distinctive threedimensional structure. Inactivation of proteases by serpins is initiated following protease recognition of a pseudosubstrate reactive center loop displayed over the surface of the molecule. This results in cleavage of the inhibitor at the reactive center peptide bond, covalent bond formation between the reactive center amino acid residue of the serpin and the active site serine of the protease, and distortion of the protease catalytic triad [1]. Complex formation results in inactivation of both the protease and the inhibitor. Serpins, therefore, are called suicide substrates [2]. Based both on amino acid sequence homology and on functional criteria, the mechanism of protease inactivation by C1INH clearly is the same as with other serpins [2,3]. Most serpins, in addition to a protease inhibitory domain, have an amino terminal domain that does not share homology with those of other serpins. The amino terminal domain of C1INH is the longest among the serpins (approximately 120 amino acids). In addition, this domain in C1INH is extremely heavily glycosylated, with three N-linked and seven O-linked carbohydrates [4]. The amino terminal domain is not involved in protease inhibitory function [5 –7].
4
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
Protease inhibitor activity
Regulation of contact system activation
Regulation of complement system activation
C1INH also regulates activation of the contact system and the intrinsic coagulation pathway via the inactivation of plasma kallikrein, and of coagulation factors XIa and XIIa. The original concept of contact system activation was based on the activation of factor XII, and subsequently of kallikrein, following the addition of artificial negatively charged substances to plasma. Zymogen factor XII binds to the negatively charged surface, which induces autoactivation to the active protease, factor XIIa. It then, by limited proteolysis, activates prekallikrein to active kallikrein. Factor XIIa also activates additional factor XII to XIIa, which results in enhanced contact system activation. Prekallikrein, which circulates in plasma complexed with high molecular weight kininogen (HK), following activation to kallikrein, cleaves HK at two sites to release bradykinin, which mediates some biological activities including vasodilation, increased vascular permeability, constriction of uterine and gastrointestinal smooth muscle, constriction of coronary and pulmonary vasculature, and bronchoconstriction [25]. The biological equivalent of the artificial negatively charged surface has not been identified. The primary site of contact system activation very likely is the endothelial cell surface [26 – 29], which provides a site for the binding of the prekallikrein –HK complex [30 –38]. Prekallikrein then is activated to kallikrein, which then cleaves the HK to release bradykinin. Two hypotheses have been proposed to explain contact system activation on the endothelial surface. According to one, cleavage of prekallikrein is mediated by endothelial cell-associated factor XIIa, similar to the mechanism of activation by negatively charged surfaces [35,39 – 43]. According to the second, prekallikrein is activated by an endothelial cell protease, prolylcarboxypeptidase [44 – 48]. The relative importance of these two potential activation pathways on the endothelial cell surface remains to be resolved. In either case, C1INH, via inactivation of kallikrein and/or factor XIIa, is the primary regulator of contact system activation [10 – 12,14,49 –51].
The primary biological activities of C1INH are to regulate activation of the complement [8,9] and contact systems [10 –14] (Table 1). In addition, C1INH is able to inactivate several other proteases. Inhibition of fibrinolysis, via complex formation with both tissue plasminogen activator and plasmin may occasionally be biologically important (Table 1) [15 –17]. C1INH controls activation of the classical complement pathway via inactivation of the proteases C1r and C1s following the activation of C1 by an immune complex, which thereby limits the consumption of C2 and C4 [8,9,18]. However, C1INH also prevents spontaneous activation of C1 through formation of a reversible complex with the zymogen forms of C1r and C1s [19,20]. The mechanism and the relative importance of this interaction in the regulation of C1 activation, however, are not clear. It appears highly likely that C1INH also regulates activation of the lectin pathway via inactivation of the mannan-binding lectin-associated proteases (MASPs) 1 and 2 [21,22]. The MASPs, although similar to C1r and C1s, differ in that MASP2, like C1s, is responsible for cleavage of C2 and C4, while MASP1 does not have activities similar to either C1r or C1s [23]. C1INH also may regulate activation of the alternative pathway [24] by binding to C3b, which thereby inhibits formation of the alternative pathway convertase by preventing factor B binding, similar to inhibition of alternative pathway activation by factor H. This function is not dependent on protease inhibitory activity. At physiological concentrations, C1INH suppressed alternative pathway activation, which suggests that this activity may be a normal biological function of C1INH [24].
Table 1 The inhibitory spectrum of C1 inhibitor Protease
Complement system C1r C1s MASP1 MASP2 Contact system Plasma kallikrein Coagulation factor XIa Coagulation factor XIIa Fibrinolytic system Tissue plasminogen activator Plasmin
Proportion of plasma inhibitory capacity provided by C1 inhibitor 100% 100% * * 42 – 84% 47% 90% § §
*Probably 100%, but has not been specifically determined. §C1INH plays little, if any, role in the inhibition of these proteases under normal circumstances, but small amounts of complexes with tPA and plasmin may be observed in some situations (endotoxin shock, exhaustive exercise).
The mediation of increased vascular permeability in HAE Complement and contact system activation in HAE The first indication that C1INH is required for normal regulation of vascular permeability was the discovery that hereditary angioedema (HAE) was associated with very low levels of C1INH [52]. During asymptomatic periods, plasma C1INH levels average approximately 30% of normal. The explanation for this decrease below 50%, which would be expected in a heterozygous disorder, is that the decreased inhibitor level results in increased activation of some or all of the proteases regulated by C1INH. These activated proteases consume C1INH thereby producing a plasma level
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
well below 50% of normal [53,54]. Studies of the in vivo catabolism of C1INH in patients with HAE are consistent with this explanation [55]. In addition, in patients with some mutations, it is possible that the abnormal mRNA or protein may produce a suppressive effect on either transcription or translation [56,57]. During attacks of angioedema, levels decrease further, and this decrease is associated with apparent spontaneous activation of both the complement and contact systems. Evidence for complement system activation consists of low levels of C2 and C4, the natural substrates of C1s, and the presence of circulating complexes of both C1r and C1s with C1INH [58]. Contact system activation is most readily indicated by the presence of high molecular weight kininogen cleavage [59 – 62]. It, therefore, seemed likely that angioedema was mediated by products of activation of either (or both) of these systems. However, because both systems are activated, it proved difficult to unequivocally determine which pathway was responsible for generating the mediator of symptoms. The C2 kinin A series of studies in the 1970s demonstrated that a kinin-like peptide could be generated in HAE plasma by incubation at 37jC. Furthermore, the data suggested that production of this kinin was dependent on complement system activation [63 – 65]. Later studies indicated that this activity may have resulted from a peptide released from C2 after sequential cleavages by C1s and plasmin [66,67]. However, the specific activity of this C2-derived peptide was relatively low. Cleavage of large quantities of C2 would have been required to generate sufficient amounts of the peptide. In addition, the peptide has never been detected in the plasma of patients with angioedema. Several early studies suggested that bradykinin might be the mediator [68 – 70]. Subsequently, a series of experiments designed to isolate and characterize the kinin-like peptide in C1INHdepleted plasma provided additional strong evidence that this peptide was bradykinin [71]. The activity (as demonstrated by smooth muscle-contracting activity and by the ability to increase vascular permeability in guinea pig skin) could be generated in C2-deficient plasma, but not in plasma that was deficient in any contact system protein; the only peptide isolated from these plasmas was bradykinin. In addition, the peptide generated during incubation of such plasma was isolated and sequenced, and shown to be bradykinin. Lastly, Nussberger et al. [72] showed that bradykinin levels were elevated in the plasma of HAE patients during attacks of angioedema. A dysfunctional C1INH with normal inhibition of contact system proteases and defective inhibition of complement proteases The description of a family, several members of whom had low C4 levels together with a dysfunctional C1INH
5
molecule [73], led to further support for the hypothesis that the mediator of angioedema is bradykinin. The index case presented with systemic lupus erythematosus, not with HAE. Subsequent studies demonstrated that this dysfunctional C1INH contained a Val rather than an Ala at amino acid 443, which is at the P2 position [74,75]. A recombinant C1INH protein with this substitution had diminished activity against C1r and C1s but normal activity against plasma kallikrein and factor XIIa [74,75]. Several other family members who also had low C4 levels had the same amino acid substitution, but no family member, had ever experienced an episode of angioedema. These observations, therefore, strongly suggested that insufficient regulation of complement system activation alone does not result in angioedema. C1 inhibitor-deficient mice Analysis of C1INH knockout mice has lent further support to the evidence that increased vascular permeability in HAE is mediated by bradykinin via contact system activation [76,77]. Homozygous (C1INH/)- and heterozygous (C1INH+/)-deficient mice are normal at birth, reproduce, grow and develop normally, and appear normal throughout their lives. They do not have obvious episodes of angioedema involving the skin. Blood from C1INH/ mice contains no C1INH protein, while that from C1INH+/ mice contains somewhat less that half normal levels, similar to humans with HAE. Most deficient mice have decreased C4 levels, indicating the presence of unregulated complement activation. No mice have been observed with episodes of angioedema of the skin. However, a small number of the deficient mice (eight mice from two different litters) have experienced intestinal angioedema with obstruction, and one mouse asphyxiated from laryngeal edema. These were isolated incidents, but they occurred only in C1INH-deficient mice (both C1INH/ and C1INH+/), and not in their wild-type littermates. The factors that may have triggered these episodes have not been identified and it has not been possible to reproduce the attacks. Although these episodes looked very much like attacks of gastrointestinal and laryngeal angioedema, because so few have occurred and the lack of biochemical data during the attacks make it difficult to unequivocally conclude that they were the result of C1INH deficiency. Although the mice did not show any symptoms of angioedema, both the C1INH/ and the C1INH+/ mice had increased vascular permeability, as demonstrated by analysis of the extravasation of intravenously injected Evans blue dye. This dye binds to serum albumin and therefore serves as a marker for the transport of fluid and protein across the endothelium. Proof that the vascular permeability defect was due to C1INH deficiency, rather than from some inadvertent associated defect, was provided by the demonstration that the increased vascular permeability was reversed by intravenous injection with human C1INH. In
6
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
addition, following application of mustard oil to the ears, the increase in dye extravasation in the deficient mice (both heterozygous and homozygous) was two- to fivefold greater than in the wild type. Mustard oil is an irritant that induces a local inflammatory response with increased vascular permeability. This effect is enhanced in the absence of C1INH. The fact that the C1INH+/ and the C1INH/ mice were identical in terms of their vascular permeability was a surprise. However, the C1INH/ mice not only are viable, but are normal except for the increased vascular permeability, to which they are well-compensated. There are some possible explanations for the relatively mild phenotype. It may be that below some critical C1INH level, activation of the proteolytic pathways regulated by C1INH is maximal and that a further decrease in the level does not change the degree of activation. However, at least with respect to the complement system in the deficient mice, this does not seem to be the case because C4, although low, is not depleted. Preliminary unpublished data suggest that the contact system also has not been maximally activated and that high molecular weight kininogen is not depleted. A likely possibility is that other inhibitors may play a greater role in the regulation of contact system activation than in the human. This might explain the fact that the phenotypes of both the C1INH/ and C1INH+/ mice are less severe than the human disease, and that the phenotypes of the two mice are identical. To test the hypothesis that increased vascular permeability in C1INH deficiency is mediated by bradykinin, C1INH/ mice were treated with a recombinant highly specific plasma kallikrein inhibitor, DX88 (Dyax, Cambridge, MA), with the C1INH Ala443!Val substitution mutant that is a relatively specific inhibitor of plasma kallikrein and factor XIIa, with the bradykinin type 2 receptor (Bk2R) antagonist Hoe140 (Icatibant) (Sigma, St. Louis, MO), and with the angiotensin converting enzyme inhibitor, captopril (Sigma) [76,77]. DX88, the C1INH variant, and Icatibant all completely reversed the increased vascular permeability in the C1INH-deficient mice, while captopril treatment, which inhibits bradykinin degradation by angiotensin converting enzyme (ACE), resulted in increased vascular permeability (Table 2). Mice deficient both in C1INH and the Bk2R were normal with no increase in vascular permeability in comparison with wild-type mice.
Table 2 Modulation of vascular permeability in C1INH/ mice Agent
Mode of action
Effect on Interpretation permeability
DX88
kallikrein inhibitor #
C1INH kallikrein inhibitor # Ala443!Val Icatibant Bk2R antagonist # Captopril
ACE inhibitor
zz
decreased bradykinin generation decreased bradykinin generation decreased signaling via Bk2R decreased bradykinin degradation
Fig. 1. Hypothesized mechanism for the initiation of an attack of angioedema. Trauma, infection, other febrile illness, and other unknown factors lead to activation of the complement, contact, and/or fibrinolytic systems. This leads to consumption of C1INH via complex formation with C1r, C1s, MASPs, factor XIIa, kallikrein, tissue plasminogen activator, and/ or plasmin. Depletion of C1INH results in complete deregulation of the complement and contact systems with decreasing levels of C4 and C2 and the generation of bradykinin. Bradykinin, via its interaction with the Bk2R on endothelial cells, mediates the increase in vascular permeability.
Vascular permeability in the C1INH-deficient mice, therefore, is almost certainly mediated via the Bk2R. Bradykinin is the only known ligand for this receptor. Conclusions The major inhibitory roles of C1INH in vivo are to regulate activation of the complement and the contact systems. Suppression of contact system activation via inactivation of plasma kallikrein and factor XIIa, which prevents excessive generation of bradykinin, helps to maintain the endothelial barrier and thereby modulate vascular permeability. This function is dramatically demonstrated by the recurrent episodes of angioedema that develop in patients with HAE. As outlined in this review, this conclusion is based on multiple observations using a variety of approaches including in vitro studies, analysis of HAE patients, and analysis of C1INH-deficient mice. The C1INH knockout mice, although not a perfect model for HAE, clearly share with C1INH-deficient humans an increase in vascular permeability mediated by bradykinin. The trigger or triggers that induce attacks of angioedema remain ill-defined. Although trauma (such as dental
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
manipulations, particularly) or intercurrent febrile illness frequently precede episodes of angioedema, many attacks develop in the absence of any identifiable predisposing factors. Furthermore, the mechanism through which an attack develops also is unclear. Presumably, in an individual with HAE, whose C1INH level always is low (averaging approximately 30% of normal), activation of any protease that is inactivated by C1INH leads to consumption of the inhibitor (because it is a suicide inhibitor) to the extent that activation of the complement and contact systems becomes completely unregulated (Fig. 1). Therefore, initial activation of either the complement system (C1r, C1s, and/or the MASPs), the contact system (factor XIa, factor XIIa, plasma kallikrein), or the fibrinolytic system (tPA, plasmin) might lead to such increased consumption. This may explain the apparent effectiveness of plasmin inhibitors in HAE. The decreased synthesis rate of C1INH in an individual who is heterozygous for deficiency is apparently insufficient to maintain homeostasis in the face of increased consumption. The role of complement system activation in HAE, therefore, is indirect and may serve to initiate and/or perpetuate an attack rather than mediate symptoms. If this is correct, it is possible that a therapeutic inhibitor that inactivates complement system, contact system, and fibrinolytic proteases might be more effective, or less likely to result in an early recurrence of symptoms, than an inhibitor that inactivates only contact system proteases because complement and fibrinolytic system activation would remain unchecked. References [1] J.A. Huntington, R.J. Read, R.W. Carrell, Structure of a serpin – protease complex shows inhibition by deformation, Nature (2000) 923 – 926. [2] P.A. Patston, P. Gettins, J. Beechem, M. Schapira, Mechanism of serpin action: evidence that C1 inhibitor functions as a suicide substrate, Biochemistry, (1991) 8876 – 8882. [3] I.G.A. Bos, C.E. Hack, J.P. Abrahams, Structural and functional aspects of C1-inhibitor, Immunobiology, (2002) 518 – 533. [4] S.C. Bock, K. Skriver, E. Nielsen, H.C. Thogersen, B. Wiman, V.H. Donaldson, R.L. Eddy, J. Marrinan, E. Radziejewska, R. Huber, T.B. Shows, S. Magnussen, Human C1 inhibitor: primary structure, cdna cloning, and chromosomal localization, Biochemistry, (1986) 4292 – 4301. [5] I.G.A. Bos, Y.T.P. Lubbers, D. Roem, J.P. Abrahams, C.E. Hack, E. Eldering, The functional integrity of the serpin domain of C1-inhibitor depends on the unique n-terminal domain, as revealed by a pathological mutant, J. Biol. Chem. (2003) 29463 – 29470. [6] M. Coutinho, K.S. Aulak , A.E. Davis III, Functional analysis of the serpin domain of C1 inhibitor, J. Immunol., (1994) 3648 – 3654. [7] A. Reboul, M. Prandini, M. Colomb, Proteolysis and deglycosylation of human C1 inhibitor: effect on functional properties, Biochem. J., (1987) 117 – 121. [8] R.B. Sim, A. Reboul, G.J. Arlaud, C.L. Villiers, M.G. Colomb, Interaction of 125I-labelled complement components C1r and C1s with protease inhibitors in plasma, FEBS Lett., (1979) 111 – 115. [9] R.J. Ziccardi, Activation of the early components of the classical complement pathway under physiological conditions, J. Immunol., (1981) 1768 – 1773.
7
[10] M. Schapira, C.F. Scott, R.W. Colman, Contribution of plasma protease inhibitors to the inactivation of kallikrein in plasma, J. Clin. Invest., (1982) 462 – 468. [11] F. van der Graaf, J.A. Koedam, B.N. Bouma, Inactivation of kallikrein in human plasma, J. Clin. Invest., (1983) 149 – 158. [12] P.C. Harpel, M.F. Lewin, A.P. Kaplan, Distribution of plasma kallikrein between c1 inactivator and a2-macroglobulin in plasma utilizing a new assay for a2-macroglobulin-kallikrein complexes, J. Biol. Chem., (1985) 4257 – 4263. [13] A.d. Agostini, H.R. Lijnen, R.A. Pixley, R.W. Colman, M. Schapira, Inactivation of factor XII active fragment in normal plasma. Predominant role of C1-inhibitor, J. Clin. Invest., (1984) 1542 – 1549. [14] R.A. Pixley, M. Schapira, R.W. Colman, The regulation of human factor XIIa by plasma proteinase inhibitors, J. Biol. Chem., (1985) 1723 – 1729. [15] N.A. Booth, E. Walker, R. Maughan, B. Bennett, Plasminogen activator in normal subjects after exercise and venous occlusion: T-PA circulates as complexes with C1-inhibitor and PAI-1, Blood, (1987) 1600 – 1604. [16] M. Cugno, C.E. Hack, J.P.d. Boer, A.J. Eerenberg, A. Agostoni, M. Cicardi, Generation of plasmin during acute attacks of hereditary angioedema, J. Lab. Clin. Med., (1993) 38 – 43. [17] L.G. Huisman, J.M. van Griensven, C. Kluft, On the role of C1inhibitor as inhibitor of tissue-type plasminogen activator in human plasma, Thromb. Haemostasis, (1995) 466 – 471. [18] O. Ratnoff, I. Lepow, Some properties of an esterase derived from preparations of the first component of complement, J. Exp. Med., (1957) 327 – 343. [19] E.J. Folkerd, B. Gardner, N.C. Hughes-Jones, The relationship between the binding ability and the rate of activation of the complement component C1, Immunology, (1980) 179 – 185. [20] R. Ziccardi, Demonstration of the interaction of native C1 with monomeric immunoglobulins and C1 inhibitor, J. Immunol., (1985) 2559 – 2563. [21] N.K.H. Wong, M. Kojima, J. Dobo, G. Ambrus, R.B. Sim, Activities of the mbl-associated serine proteases (MASPs) and their regulation by natural inhibitors, Mol. Immunol., (1999) 853 – 861. [22] S.V. Petersen, S. Thiel, L. Jensen, T. Vorup-Jensen, C. Koch, J.C. Jensenius, Control of the classical and the MBL pathway of complement activation, Mol. Immunol., (2000) 803 – 811. [23] K. Hajela, M. Kojima, G. Ambrus, K.H. Wong, B.E. Moffatt, J. Ferluga, S. Hajela, P. Gal, R.B. Sim, The biological functions of MBL-associated serine proteases (MASPS), Immunobiology, (2002) 467 – 475. [24] H. Jiang, E. Wagner, H. Zhang, M.M. Frank, Complement 1 inhibitor is a regulator of the alternative complement pathway, J. Exp. Med., (2001) 1609 – 1616. [25] A.P. Kaplan, K. Joseph, Y. Shibayama, S. Reddigari, B. Ghebrehiwet, M. Silverberg, The intrinsic coagulation/kinin-forming cascade: assembly in plasma and cell surfaces in inflammation, Adv. Immunol., (1997) 225 – 272. [26] E.J. Gustaafson, A.H. Schmaier, Y.T. Wachtfogel, N. Kaufman, U. Kucich, R.W. Colman, Human neutrophils contain and bind high molecular weight kininogen, J. Clin. Invest., (1989) 28 – 35. [27] E.J. Gustaafson, D. Schutsky, L. Knight, A.H. Schmaier, High molecular weight kininogen binds to unstimulated platelets, J. Clin. Invest., (1986) 310 – 318. [28] J.S. Greengard, J.H. Griffin, Receptors for high molecular weight kininogen on stimulated washed human platelets, Biochemistry, (1984) 6863 – 6869. [29] F.J. Meloni, E.J. Gustafson, A.H. Schmaier, High molecular weight kininogen binds to platelets by its heavy and light chains and when bound has altered susceptibility to kallikrein cleavage, Blood, (1992) 1233 – 1244. [30] A.H. Schmaier, A. Kuo, D. Lundberg, S. Murray, D.B. Cines, The expression of high molecular weight kininogen on human umbilical vein endothelial cells, J. Biol. Chem., (1988) 16327 – 16333.
8
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
[31] F. Van Iwaarden, P.G. deGroot, B.N. Bouma, The binding of high molecular weight kininogen to cultured human endothelial cells, J. Biol. Chem., (1988) 4698 – 4703. [32] R.W. Colman, A.H. Schmaier, Contact system: a vascular biology modulator with anticoagulant, profibrinolytic, antiadhesive, and proinflammatory attributes, Blood, (1997) 3819 – 3843. [33] A.A.K. Hasan, T. Zisman, A.H. Schmaier, Identification of cytokeratin 1 as a binding protein and presentation receptor for kininogens on endothelial cells, Proc. Natl. Acad. Sci. U. S. A., (1998) 3615 – 3620. [34] H. Herwald, J. Dedio, R. Kellner, M. Loos, W. Muller-Esterl, Isolation and characterization of the kininogen-binding protein p33 from endothelial cells, J. Biol. Chem., (1996) 13040 – 13047. [35] K. Joseph, B. Ghebrehiwet, E.I.B. Peerschke, K.B.M. Reid, A.P. Kaplan, Identification of the zinc-dependent endothelial cell binding protein for high molecular weight kininogen and factor XII: identity with the receptor that binds to the globular ‘‘heads’’ of C1q (gC1q-r), Proc. Natl. Acad. Sci. U. S. A., (1996) 8552 – 8557. [36] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Mapping binding domains of kininogens on endothelial cell cytokeratin 1, J. Biol. Chem., (1999) 7137 – 7145. [37] R.W. Colman, R.A. Pixley, S. Najamunnisa, W. Yan, J. Wang, A. Mazar, K.R. McCrae, Binding of high molecular weight kininogen to human endothelial cells is mediated via a site within domains 2 and 3 of the urokinase receptor, J. Clin. Invest., (1997) 1481 – 1487. [38] F. Mahdi, Z. shariat-Madar, R.F. Todd III, C.D. Figueroa, A.H. Schmaier, Expression and co-localization of cytokeratin 1 and urokinase plasminogen activator receptor on endothelial cells, Blood, (2001) 2342 – 2350. [39] S.R. Reddigari, Y. Shibayama, T. Brunnee, A.P. Kaplan, Human hageman factor (factor xii) and high molecular weight kininogen compete for the same binding site on human umbilical vein endothelial cells, J. Biol. Chem., (1993) 11982 – 11987. [40] K. Joseph, Y. Shibayama, B. Ghebrehiwet, A.P. Kaplan, Factor XII dependent contact activation on endothelial cells and binding proteins gc1qr and cytokeratin 1, Thromb. Haemostasis, (2001) 119 – 124. [41] K. Joseph, B.G. Tholanikunnel, A.P. Kaplan, Heat shock protein 90 catalyzes activation of the prekallikrein – kininogen complex in the absence of factor xii, Proc. Natl. Acad. Sci. U. S. A., (2002) 896 – 900. [42] K. Joseph, B.G. Tholanikunnel, A.P. Kaplan, Activation of the bradykinin-forming cascade on endothelial cells: a role for heat shock protein 90, Int. Immunopharmacol., (2002) 1851 – 1859. [43] A.P. Kaplan, K. Joseph, M. Silverberg, Pathways for bradykinin formation and inflammatory disease, J. Allergy Clin. Immunol., (2002) 195 – 209. [44] F. Mahdi, Z. Shariat-Madar, C.D. Figueroa, A.H. Schmaier, Factor XII interacts with the multiprotein assembly of urokinase plasminogen activator receptor, gC1qr, and cytokeratin 1 on endothelial cell membranes, Blood, (2002) 3585 – 3596. [45] Y. Zhao, Q. Qiu, F. Mahdi, Z. Shariat-Madar, R. Rojkjaer, A.H. Schmaier, Assembly and activation of hk – pk complex on endothelial cells results in bradykinin liberation and no formation, Am. J. Physiol.: Heart Circ. Physiol., (2001) H1821 – H1829. [46] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Identification and characterization of prolylcarboxypeptidase as an endothelial cell prekallikrein activator, J. Biol. Chem., (2002) 17962 – 17969. [47] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Assembly and activation of the plasma kallikrein/kinin system: a new interpretation, Int. Immunopharmacol., (2002) 1841 – 1849. [48] A.H. Schmaier, The plasma kallikrein – kinin system counterbalances the renin – angiotensin system, J. Clin. Invest., (2002) 1007 – 1009. [49] I. Gigli, J.W. Mason, R.W. Colman, K.F. Austen, Interaction of plasma kallikrein with the C1 inhibitor, J. Immunol., (1970) 574 – 581. [50] D.J. McConnell, Inhibitors of kallikrein in human plasma, J. Clin. Invest., (1972) 1611 – 1623. [51] A. de Agostini, H.R. Lijnen, R.A. Pixley, R.W. Colman, M. Schapira, Inactivation of factor-XII active fragment in normal plasma: predominant role of C1-inhibitor, J. Clin. Invest., (1984) 1542 – 1549.
[52] V.H. Donaldson, R.R. Evans, A biochemical abnormality in hereditary angioneurotic edema, Am. J. Med., (1963) 37 – 44. [53] P. Lachmann, F. Rosen, The catabolism of C1-inhibitor and the pathogenesis of hereditary angioedema, Acta Pathol., Microbiol. Immunol. Scand., (1984) 35 – 39. [54] P. Woo, P.J. Lachmann, R.A. Harrison, N. Amos, C. Cooper, F.S. Rosen, Simultaneous turnover of normal and dysfunctional c1 inhibitor as a probe of in vivo activation of C1 and contact activatable proteases, Clin. Exp. Immunol., (1985) 1 – 8. [55] M. Quastel, R. Harrison, M. Cicardi, C. Alper, F. Rosen, Behavior in vivo of normal and dysfunctional C1 inhibitor in normal subjects and patients with hereditary angioneurotic edema, J. Clin. Invest., (1983) 1041 – 1046. [56] S.C. Ernst, A. Circolo, A.E. Davis III, K. Gheesling-Mullis, M. Fliesler, R.C. Strunk, Impaired production of both normal and mutant C1 inhibitor proteins in type I hereditary angioedema with a duplication in exon 8, J. Immunol., (1996) 405 – 410. [57] J. Kramer, F. Rosen, H. Colten, K. Rajezy, R. Strunk, Transinhibition of C1 inhibitor synthesis in type I hereditary angioneurotic edema, J. Clin. Invest., (1993) 1258 – 1262. [58] A.B. Laurell, U. Martensson, A.G. Sjoholm, C1 subcomponent complexes in normal and pathological sera studied by crossed immunoelectrophoresis, Acta Pathol. Microbiol. Scand., (1976) 455 – 464. [59] R. Buhler, J.K. Hovinga, I. Aebi-Huber, M. Furlan, B. Lammle, Improved detection of proteolytically cleaved high molecular weight kininogen by immunoblotting using an antiserum against its reduced 47 kda light chain, Blood Coagulation Fibrinolysis, (1995) 223 – 232. [60] M. Berrettini, B. Lammle, T. White, M.J. Heeb, H.P. Schwarz, B. Zuraw, J. Curd, J.H. Griffin, Detection of in vitro and in vivo cleavage of high molecular weight kininogen in human plasma by immunoblotting with monoclonal antibodies, Blood, (1986) 455 – 462. [61] E.W. Nielsen, H.T. Johansen, K. Hogasen, W. Wuillemin, C.E. Hack, T.E. Mollnes, Activation of the complement, coagulation, fibrinolytic and kallikrein – kinin systems during attacks of hereditary angioedema, Scand. J. Immunol., (1996) 185 – 192. [62] M. Cugno, M. Cicardi, R. Coppola, A. Agostoni, Activation of factor XII and cleavage of high molecular weight kininogen during acute attacks in hereditary and acquired C1-inhibitor deficiencies, Immunopharmacology, (1996) 361 – 364. [63] V.H. Donaldson, Kinin formation in hereditary angioneurotic edema (HANE) plasma, Int. Arch. Allergy, (1973) 206 – 209. [64] V.H. Donaldson, E. Merler, F.S. Rosen, K.W. Kretschmer, I.H. Lepow, A polypeptide kinin in hereditary angioneurotic edema plasma: role of complement in its formation, J. Lab. Clin. Med., (1970) 986. [65] V.H. Donaldson, O.D. Ratnoff, W.D.d. Silva, F.S. Rosen, Permeability-increasing activity in hereditary angioneurotic edema plasma, J. Clin. Invest., (1969) 642 – 653. [66] C.J. Strang, K.S. Auerbach, F.S. Rosen, C1s-induced vascular permeability in C2-deficient guinea pigs, J. Immunol., (1986) 631 – 635. [67] C. Strang, S. Cholin, J. Spragg, A. Davis III, E. Schneeberger, V. Donaldson, F. Rosen, Angioedema induced by a peptide derived from complement component C2, J. Exp. Med., (1988) 1685 – 1698. [68] J.G. Curd, J.L.J. Prograis, C.G. Cochrane, Detection of active kallikrein in induced blister fluids of hereditary angioedema patients, J. Exp. Med., (1980) 742 – 747. [69] J.G. Curd, M. Yelvington, N. Burridge, N.P. Stimler, C. Gerard, J.L.J. Prograis, C.G. Cochrane, H.J. Muller-Eberhard, Generation of bradykinin during incubation of hereditary angioedema plasma, Mol. Immunol., (1982) 1365 (abstract). [70] T. Fields, B. Ghebrewihet, A. Kaplan, Kinin formation in hereditary angioedema plasma: evidence against kinin derivation from C2 and in support of spontaneous formation of bradykinin, J. Allergy Clin. Immunol., (1983) 54 – 60. [71] L.R. Shoemaker, S.J. Schurman, V.H. Donaldson , A.E. Davis III, Hereditary angioneurotic edema: characterization of plasma kinin
A.E. Davis III / Clinical Immunology 114 (2005) 3–9 and vascular permeability-enhancing activities, Clin. Exp. Immunol., (1994) 22 – 28. [72] J. Nussberger, M. Cugno, C. Amstutz, M. Cicardi, A. Pellacani, A. Agostoni, Plasma bradykinin in angio-oedema, Lancet, (1998) 1693 – 1697. [73] J.J. Wisnieski, T.C. Knauss, I. Yike, D.G. Dearborn, R.L. Narvy, G.B. Naff, Unique C1 inhibitor dysfunction in a kindred without angioedema. I. A mutant c1 inhibitor that inhibits C1s but not C1r, J. Immunol., (1994) 3199 – 3209. [74] R. Zahedi, J.J. Bissler, A.E. Davis III, C. Andreadis, J.J. Wisnieske, Unique C1 inhibitor dysfunction in a kindred without angioedema. II. Identification of an Ala443-Val substitution and functional analy-
9
sis of the recombinant mutant protein, J. Clin. Invest., (1995) 1299 – 1305. [75] R. Zahedi, J. Wisnieski, A.E. Davis III, Role of the P2 residue of complement 1 inhibitor (Ala443) in determination of target protease specificity, J. Immunol., (1997) 983 – 988. [76] E.D. Han, R.C. MacFarlane, A.N. Mulligan, J. Scafidi, A.E. Davis III, Increased vascular permeability in C1 inhibitor-deficient mice is mediated by the bradykinin type 2 receptor, J. Clin. Invest., (2002) 1057 – 1063. [77] E. Han Lee, E. Pappalardo, J. Scafidi, A.E. Davis III, Approaches toward reversal of increased vascular permeability in C1 inhibitor deficient mice, Immunol. Lett., (2003) 155 – 160.
Short Analytical Review
The pathophysiology of hereditary angioedema Alvin E. Davis III * Department of Pediatrics, CBR Institute for Biomedical Research, Harvard Medical School, Boston, MA 02115, USA Received 13 April 2004; accepted 20 May 2004 Available online 2 July 2004
Abstract Hereditary angioedema (HAE), characterized by recurrent episodes of angioedema involving the skin, or the mucosa of the upper respiratory or the gastrointestinal tracts, results from heterozygosity for deficiency of the serine proteinase inhibitor (serpin), C1 inhibitor (C1INH). The primary biological role of C1INH is to regulate activation of the complement system, the contact system, and the intrinsic coagulation system. During attacks of angioedema, together with decreasing levels of C1INH, the complement and contact systems are activated: C2 and C4 levels fall and high molecular weight kininogen is cleaved. Although previous data suggested that symptoms in HAE might be mediated via complement system activation, a combination of recent clinical data, in vitro studies, and analysis of C1INH-deficient mice all indicate that the major mediator of angioedema is bradykinin: (1) a vascular permeability enhancing factor can be generated in vitro in C1INH-depleted, C2-deficient plasma, but not from C1INH-depleted, contact system-deficient plasma; this factor was identified by sequence analysis as bradykinin; (2) bradykinin can be detected in the plasma of HAE patients during attacks of angioedema; (3) in several members of one family, expression of a C1INH variant that inhibits contact system proteases but has defective inhibition of C1r and C1s does not result in HAE; (4) C1INH-deficient (C1INH/) mice have a defect in vascular permeability that is suppressed by treatment with specific plasma kallikrein inhibitors and by bradykinin type 2 receptor (Bk2R) antagonists, and is eliminated in C1INH/, Bk2R/ double-deficient mice. D 2004 Elsevier Inc. All rights reserved. Keywords: Hereditary angioedema; C1 inhibitor; Mucosa
Introduction Hereditary angioedema Hereditary angioedema (HAE) results from deficiency of the plasma protease inhibitor, C1 inhibitor (C1INH). Individuals with HAE are heterozygous for deficiency, which results in autosomal dominant inheritance. Complete deficiency has never been reported. Absence of expression from one allele, which results simply in decreased expression of C1INH in the plasma, is called type 1 HAE, while expression of a dysfunctional C1INH protein, together with decreased levels of normal protein, is termed type 2 HAE. C1INH function C1INH is a serpin. Most, but not all members of this group of proteins, are protease inhibitors; they share amino * Department of Pediatrics, CBR Institute for Biomedical Research, Harvard Medical School, 800 Huntington Avenue, Boston, MA 02115. Fax: +1-617-278-3490. E-mail address: [email protected]. 1521-6616/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.clim.2004.05.007
acid sequence homology and a similar distinctive threedimensional structure. Inactivation of proteases by serpins is initiated following protease recognition of a pseudosubstrate reactive center loop displayed over the surface of the molecule. This results in cleavage of the inhibitor at the reactive center peptide bond, covalent bond formation between the reactive center amino acid residue of the serpin and the active site serine of the protease, and distortion of the protease catalytic triad [1]. Complex formation results in inactivation of both the protease and the inhibitor. Serpins, therefore, are called suicide substrates [2]. Based both on amino acid sequence homology and on functional criteria, the mechanism of protease inactivation by C1INH clearly is the same as with other serpins [2,3]. Most serpins, in addition to a protease inhibitory domain, have an amino terminal domain that does not share homology with those of other serpins. The amino terminal domain of C1INH is the longest among the serpins (approximately 120 amino acids). In addition, this domain in C1INH is extremely heavily glycosylated, with three N-linked and seven O-linked carbohydrates [4]. The amino terminal domain is not involved in protease inhibitory function [5 –7].
4
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
Protease inhibitor activity
Regulation of contact system activation
Regulation of complement system activation
C1INH also regulates activation of the contact system and the intrinsic coagulation pathway via the inactivation of plasma kallikrein, and of coagulation factors XIa and XIIa. The original concept of contact system activation was based on the activation of factor XII, and subsequently of kallikrein, following the addition of artificial negatively charged substances to plasma. Zymogen factor XII binds to the negatively charged surface, which induces autoactivation to the active protease, factor XIIa. It then, by limited proteolysis, activates prekallikrein to active kallikrein. Factor XIIa also activates additional factor XII to XIIa, which results in enhanced contact system activation. Prekallikrein, which circulates in plasma complexed with high molecular weight kininogen (HK), following activation to kallikrein, cleaves HK at two sites to release bradykinin, which mediates some biological activities including vasodilation, increased vascular permeability, constriction of uterine and gastrointestinal smooth muscle, constriction of coronary and pulmonary vasculature, and bronchoconstriction [25]. The biological equivalent of the artificial negatively charged surface has not been identified. The primary site of contact system activation very likely is the endothelial cell surface [26 – 29], which provides a site for the binding of the prekallikrein –HK complex [30 –38]. Prekallikrein then is activated to kallikrein, which then cleaves the HK to release bradykinin. Two hypotheses have been proposed to explain contact system activation on the endothelial surface. According to one, cleavage of prekallikrein is mediated by endothelial cell-associated factor XIIa, similar to the mechanism of activation by negatively charged surfaces [35,39 – 43]. According to the second, prekallikrein is activated by an endothelial cell protease, prolylcarboxypeptidase [44 – 48]. The relative importance of these two potential activation pathways on the endothelial cell surface remains to be resolved. In either case, C1INH, via inactivation of kallikrein and/or factor XIIa, is the primary regulator of contact system activation [10 – 12,14,49 –51].
The primary biological activities of C1INH are to regulate activation of the complement [8,9] and contact systems [10 –14] (Table 1). In addition, C1INH is able to inactivate several other proteases. Inhibition of fibrinolysis, via complex formation with both tissue plasminogen activator and plasmin may occasionally be biologically important (Table 1) [15 –17]. C1INH controls activation of the classical complement pathway via inactivation of the proteases C1r and C1s following the activation of C1 by an immune complex, which thereby limits the consumption of C2 and C4 [8,9,18]. However, C1INH also prevents spontaneous activation of C1 through formation of a reversible complex with the zymogen forms of C1r and C1s [19,20]. The mechanism and the relative importance of this interaction in the regulation of C1 activation, however, are not clear. It appears highly likely that C1INH also regulates activation of the lectin pathway via inactivation of the mannan-binding lectin-associated proteases (MASPs) 1 and 2 [21,22]. The MASPs, although similar to C1r and C1s, differ in that MASP2, like C1s, is responsible for cleavage of C2 and C4, while MASP1 does not have activities similar to either C1r or C1s [23]. C1INH also may regulate activation of the alternative pathway [24] by binding to C3b, which thereby inhibits formation of the alternative pathway convertase by preventing factor B binding, similar to inhibition of alternative pathway activation by factor H. This function is not dependent on protease inhibitory activity. At physiological concentrations, C1INH suppressed alternative pathway activation, which suggests that this activity may be a normal biological function of C1INH [24].
Table 1 The inhibitory spectrum of C1 inhibitor Protease
Complement system C1r C1s MASP1 MASP2 Contact system Plasma kallikrein Coagulation factor XIa Coagulation factor XIIa Fibrinolytic system Tissue plasminogen activator Plasmin
Proportion of plasma inhibitory capacity provided by C1 inhibitor 100% 100% * * 42 – 84% 47% 90% § §
*Probably 100%, but has not been specifically determined. §C1INH plays little, if any, role in the inhibition of these proteases under normal circumstances, but small amounts of complexes with tPA and plasmin may be observed in some situations (endotoxin shock, exhaustive exercise).
The mediation of increased vascular permeability in HAE Complement and contact system activation in HAE The first indication that C1INH is required for normal regulation of vascular permeability was the discovery that hereditary angioedema (HAE) was associated with very low levels of C1INH [52]. During asymptomatic periods, plasma C1INH levels average approximately 30% of normal. The explanation for this decrease below 50%, which would be expected in a heterozygous disorder, is that the decreased inhibitor level results in increased activation of some or all of the proteases regulated by C1INH. These activated proteases consume C1INH thereby producing a plasma level
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
well below 50% of normal [53,54]. Studies of the in vivo catabolism of C1INH in patients with HAE are consistent with this explanation [55]. In addition, in patients with some mutations, it is possible that the abnormal mRNA or protein may produce a suppressive effect on either transcription or translation [56,57]. During attacks of angioedema, levels decrease further, and this decrease is associated with apparent spontaneous activation of both the complement and contact systems. Evidence for complement system activation consists of low levels of C2 and C4, the natural substrates of C1s, and the presence of circulating complexes of both C1r and C1s with C1INH [58]. Contact system activation is most readily indicated by the presence of high molecular weight kininogen cleavage [59 – 62]. It, therefore, seemed likely that angioedema was mediated by products of activation of either (or both) of these systems. However, because both systems are activated, it proved difficult to unequivocally determine which pathway was responsible for generating the mediator of symptoms. The C2 kinin A series of studies in the 1970s demonstrated that a kinin-like peptide could be generated in HAE plasma by incubation at 37jC. Furthermore, the data suggested that production of this kinin was dependent on complement system activation [63 – 65]. Later studies indicated that this activity may have resulted from a peptide released from C2 after sequential cleavages by C1s and plasmin [66,67]. However, the specific activity of this C2-derived peptide was relatively low. Cleavage of large quantities of C2 would have been required to generate sufficient amounts of the peptide. In addition, the peptide has never been detected in the plasma of patients with angioedema. Several early studies suggested that bradykinin might be the mediator [68 – 70]. Subsequently, a series of experiments designed to isolate and characterize the kinin-like peptide in C1INHdepleted plasma provided additional strong evidence that this peptide was bradykinin [71]. The activity (as demonstrated by smooth muscle-contracting activity and by the ability to increase vascular permeability in guinea pig skin) could be generated in C2-deficient plasma, but not in plasma that was deficient in any contact system protein; the only peptide isolated from these plasmas was bradykinin. In addition, the peptide generated during incubation of such plasma was isolated and sequenced, and shown to be bradykinin. Lastly, Nussberger et al. [72] showed that bradykinin levels were elevated in the plasma of HAE patients during attacks of angioedema. A dysfunctional C1INH with normal inhibition of contact system proteases and defective inhibition of complement proteases The description of a family, several members of whom had low C4 levels together with a dysfunctional C1INH
5
molecule [73], led to further support for the hypothesis that the mediator of angioedema is bradykinin. The index case presented with systemic lupus erythematosus, not with HAE. Subsequent studies demonstrated that this dysfunctional C1INH contained a Val rather than an Ala at amino acid 443, which is at the P2 position [74,75]. A recombinant C1INH protein with this substitution had diminished activity against C1r and C1s but normal activity against plasma kallikrein and factor XIIa [74,75]. Several other family members who also had low C4 levels had the same amino acid substitution, but no family member, had ever experienced an episode of angioedema. These observations, therefore, strongly suggested that insufficient regulation of complement system activation alone does not result in angioedema. C1 inhibitor-deficient mice Analysis of C1INH knockout mice has lent further support to the evidence that increased vascular permeability in HAE is mediated by bradykinin via contact system activation [76,77]. Homozygous (C1INH/)- and heterozygous (C1INH+/)-deficient mice are normal at birth, reproduce, grow and develop normally, and appear normal throughout their lives. They do not have obvious episodes of angioedema involving the skin. Blood from C1INH/ mice contains no C1INH protein, while that from C1INH+/ mice contains somewhat less that half normal levels, similar to humans with HAE. Most deficient mice have decreased C4 levels, indicating the presence of unregulated complement activation. No mice have been observed with episodes of angioedema of the skin. However, a small number of the deficient mice (eight mice from two different litters) have experienced intestinal angioedema with obstruction, and one mouse asphyxiated from laryngeal edema. These were isolated incidents, but they occurred only in C1INH-deficient mice (both C1INH/ and C1INH+/), and not in their wild-type littermates. The factors that may have triggered these episodes have not been identified and it has not been possible to reproduce the attacks. Although these episodes looked very much like attacks of gastrointestinal and laryngeal angioedema, because so few have occurred and the lack of biochemical data during the attacks make it difficult to unequivocally conclude that they were the result of C1INH deficiency. Although the mice did not show any symptoms of angioedema, both the C1INH/ and the C1INH+/ mice had increased vascular permeability, as demonstrated by analysis of the extravasation of intravenously injected Evans blue dye. This dye binds to serum albumin and therefore serves as a marker for the transport of fluid and protein across the endothelium. Proof that the vascular permeability defect was due to C1INH deficiency, rather than from some inadvertent associated defect, was provided by the demonstration that the increased vascular permeability was reversed by intravenous injection with human C1INH. In
6
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
addition, following application of mustard oil to the ears, the increase in dye extravasation in the deficient mice (both heterozygous and homozygous) was two- to fivefold greater than in the wild type. Mustard oil is an irritant that induces a local inflammatory response with increased vascular permeability. This effect is enhanced in the absence of C1INH. The fact that the C1INH+/ and the C1INH/ mice were identical in terms of their vascular permeability was a surprise. However, the C1INH/ mice not only are viable, but are normal except for the increased vascular permeability, to which they are well-compensated. There are some possible explanations for the relatively mild phenotype. It may be that below some critical C1INH level, activation of the proteolytic pathways regulated by C1INH is maximal and that a further decrease in the level does not change the degree of activation. However, at least with respect to the complement system in the deficient mice, this does not seem to be the case because C4, although low, is not depleted. Preliminary unpublished data suggest that the contact system also has not been maximally activated and that high molecular weight kininogen is not depleted. A likely possibility is that other inhibitors may play a greater role in the regulation of contact system activation than in the human. This might explain the fact that the phenotypes of both the C1INH/ and C1INH+/ mice are less severe than the human disease, and that the phenotypes of the two mice are identical. To test the hypothesis that increased vascular permeability in C1INH deficiency is mediated by bradykinin, C1INH/ mice were treated with a recombinant highly specific plasma kallikrein inhibitor, DX88 (Dyax, Cambridge, MA), with the C1INH Ala443!Val substitution mutant that is a relatively specific inhibitor of plasma kallikrein and factor XIIa, with the bradykinin type 2 receptor (Bk2R) antagonist Hoe140 (Icatibant) (Sigma, St. Louis, MO), and with the angiotensin converting enzyme inhibitor, captopril (Sigma) [76,77]. DX88, the C1INH variant, and Icatibant all completely reversed the increased vascular permeability in the C1INH-deficient mice, while captopril treatment, which inhibits bradykinin degradation by angiotensin converting enzyme (ACE), resulted in increased vascular permeability (Table 2). Mice deficient both in C1INH and the Bk2R were normal with no increase in vascular permeability in comparison with wild-type mice.
Table 2 Modulation of vascular permeability in C1INH/ mice Agent
Mode of action
Effect on Interpretation permeability
DX88
kallikrein inhibitor #
C1INH kallikrein inhibitor # Ala443!Val Icatibant Bk2R antagonist # Captopril
ACE inhibitor
zz
decreased bradykinin generation decreased bradykinin generation decreased signaling via Bk2R decreased bradykinin degradation
Fig. 1. Hypothesized mechanism for the initiation of an attack of angioedema. Trauma, infection, other febrile illness, and other unknown factors lead to activation of the complement, contact, and/or fibrinolytic systems. This leads to consumption of C1INH via complex formation with C1r, C1s, MASPs, factor XIIa, kallikrein, tissue plasminogen activator, and/ or plasmin. Depletion of C1INH results in complete deregulation of the complement and contact systems with decreasing levels of C4 and C2 and the generation of bradykinin. Bradykinin, via its interaction with the Bk2R on endothelial cells, mediates the increase in vascular permeability.
Vascular permeability in the C1INH-deficient mice, therefore, is almost certainly mediated via the Bk2R. Bradykinin is the only known ligand for this receptor. Conclusions The major inhibitory roles of C1INH in vivo are to regulate activation of the complement and the contact systems. Suppression of contact system activation via inactivation of plasma kallikrein and factor XIIa, which prevents excessive generation of bradykinin, helps to maintain the endothelial barrier and thereby modulate vascular permeability. This function is dramatically demonstrated by the recurrent episodes of angioedema that develop in patients with HAE. As outlined in this review, this conclusion is based on multiple observations using a variety of approaches including in vitro studies, analysis of HAE patients, and analysis of C1INH-deficient mice. The C1INH knockout mice, although not a perfect model for HAE, clearly share with C1INH-deficient humans an increase in vascular permeability mediated by bradykinin. The trigger or triggers that induce attacks of angioedema remain ill-defined. Although trauma (such as dental
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
manipulations, particularly) or intercurrent febrile illness frequently precede episodes of angioedema, many attacks develop in the absence of any identifiable predisposing factors. Furthermore, the mechanism through which an attack develops also is unclear. Presumably, in an individual with HAE, whose C1INH level always is low (averaging approximately 30% of normal), activation of any protease that is inactivated by C1INH leads to consumption of the inhibitor (because it is a suicide inhibitor) to the extent that activation of the complement and contact systems becomes completely unregulated (Fig. 1). Therefore, initial activation of either the complement system (C1r, C1s, and/or the MASPs), the contact system (factor XIa, factor XIIa, plasma kallikrein), or the fibrinolytic system (tPA, plasmin) might lead to such increased consumption. This may explain the apparent effectiveness of plasmin inhibitors in HAE. The decreased synthesis rate of C1INH in an individual who is heterozygous for deficiency is apparently insufficient to maintain homeostasis in the face of increased consumption. The role of complement system activation in HAE, therefore, is indirect and may serve to initiate and/or perpetuate an attack rather than mediate symptoms. If this is correct, it is possible that a therapeutic inhibitor that inactivates complement system, contact system, and fibrinolytic proteases might be more effective, or less likely to result in an early recurrence of symptoms, than an inhibitor that inactivates only contact system proteases because complement and fibrinolytic system activation would remain unchecked. References [1] J.A. Huntington, R.J. Read, R.W. Carrell, Structure of a serpin – protease complex shows inhibition by deformation, Nature (2000) 923 – 926. [2] P.A. Patston, P. Gettins, J. Beechem, M. Schapira, Mechanism of serpin action: evidence that C1 inhibitor functions as a suicide substrate, Biochemistry, (1991) 8876 – 8882. [3] I.G.A. Bos, C.E. Hack, J.P. Abrahams, Structural and functional aspects of C1-inhibitor, Immunobiology, (2002) 518 – 533. [4] S.C. Bock, K. Skriver, E. Nielsen, H.C. Thogersen, B. Wiman, V.H. Donaldson, R.L. Eddy, J. Marrinan, E. Radziejewska, R. Huber, T.B. Shows, S. Magnussen, Human C1 inhibitor: primary structure, cdna cloning, and chromosomal localization, Biochemistry, (1986) 4292 – 4301. [5] I.G.A. Bos, Y.T.P. Lubbers, D. Roem, J.P. Abrahams, C.E. Hack, E. Eldering, The functional integrity of the serpin domain of C1-inhibitor depends on the unique n-terminal domain, as revealed by a pathological mutant, J. Biol. Chem. (2003) 29463 – 29470. [6] M. Coutinho, K.S. Aulak , A.E. Davis III, Functional analysis of the serpin domain of C1 inhibitor, J. Immunol., (1994) 3648 – 3654. [7] A. Reboul, M. Prandini, M. Colomb, Proteolysis and deglycosylation of human C1 inhibitor: effect on functional properties, Biochem. J., (1987) 117 – 121. [8] R.B. Sim, A. Reboul, G.J. Arlaud, C.L. Villiers, M.G. Colomb, Interaction of 125I-labelled complement components C1r and C1s with protease inhibitors in plasma, FEBS Lett., (1979) 111 – 115. [9] R.J. Ziccardi, Activation of the early components of the classical complement pathway under physiological conditions, J. Immunol., (1981) 1768 – 1773.
7
[10] M. Schapira, C.F. Scott, R.W. Colman, Contribution of plasma protease inhibitors to the inactivation of kallikrein in plasma, J. Clin. Invest., (1982) 462 – 468. [11] F. van der Graaf, J.A. Koedam, B.N. Bouma, Inactivation of kallikrein in human plasma, J. Clin. Invest., (1983) 149 – 158. [12] P.C. Harpel, M.F. Lewin, A.P. Kaplan, Distribution of plasma kallikrein between c1 inactivator and a2-macroglobulin in plasma utilizing a new assay for a2-macroglobulin-kallikrein complexes, J. Biol. Chem., (1985) 4257 – 4263. [13] A.d. Agostini, H.R. Lijnen, R.A. Pixley, R.W. Colman, M. Schapira, Inactivation of factor XII active fragment in normal plasma. Predominant role of C1-inhibitor, J. Clin. Invest., (1984) 1542 – 1549. [14] R.A. Pixley, M. Schapira, R.W. Colman, The regulation of human factor XIIa by plasma proteinase inhibitors, J. Biol. Chem., (1985) 1723 – 1729. [15] N.A. Booth, E. Walker, R. Maughan, B. Bennett, Plasminogen activator in normal subjects after exercise and venous occlusion: T-PA circulates as complexes with C1-inhibitor and PAI-1, Blood, (1987) 1600 – 1604. [16] M. Cugno, C.E. Hack, J.P.d. Boer, A.J. Eerenberg, A. Agostoni, M. Cicardi, Generation of plasmin during acute attacks of hereditary angioedema, J. Lab. Clin. Med., (1993) 38 – 43. [17] L.G. Huisman, J.M. van Griensven, C. Kluft, On the role of C1inhibitor as inhibitor of tissue-type plasminogen activator in human plasma, Thromb. Haemostasis, (1995) 466 – 471. [18] O. Ratnoff, I. Lepow, Some properties of an esterase derived from preparations of the first component of complement, J. Exp. Med., (1957) 327 – 343. [19] E.J. Folkerd, B. Gardner, N.C. Hughes-Jones, The relationship between the binding ability and the rate of activation of the complement component C1, Immunology, (1980) 179 – 185. [20] R. Ziccardi, Demonstration of the interaction of native C1 with monomeric immunoglobulins and C1 inhibitor, J. Immunol., (1985) 2559 – 2563. [21] N.K.H. Wong, M. Kojima, J. Dobo, G. Ambrus, R.B. Sim, Activities of the mbl-associated serine proteases (MASPs) and their regulation by natural inhibitors, Mol. Immunol., (1999) 853 – 861. [22] S.V. Petersen, S. Thiel, L. Jensen, T. Vorup-Jensen, C. Koch, J.C. Jensenius, Control of the classical and the MBL pathway of complement activation, Mol. Immunol., (2000) 803 – 811. [23] K. Hajela, M. Kojima, G. Ambrus, K.H. Wong, B.E. Moffatt, J. Ferluga, S. Hajela, P. Gal, R.B. Sim, The biological functions of MBL-associated serine proteases (MASPS), Immunobiology, (2002) 467 – 475. [24] H. Jiang, E. Wagner, H. Zhang, M.M. Frank, Complement 1 inhibitor is a regulator of the alternative complement pathway, J. Exp. Med., (2001) 1609 – 1616. [25] A.P. Kaplan, K. Joseph, Y. Shibayama, S. Reddigari, B. Ghebrehiwet, M. Silverberg, The intrinsic coagulation/kinin-forming cascade: assembly in plasma and cell surfaces in inflammation, Adv. Immunol., (1997) 225 – 272. [26] E.J. Gustaafson, A.H. Schmaier, Y.T. Wachtfogel, N. Kaufman, U. Kucich, R.W. Colman, Human neutrophils contain and bind high molecular weight kininogen, J. Clin. Invest., (1989) 28 – 35. [27] E.J. Gustaafson, D. Schutsky, L. Knight, A.H. Schmaier, High molecular weight kininogen binds to unstimulated platelets, J. Clin. Invest., (1986) 310 – 318. [28] J.S. Greengard, J.H. Griffin, Receptors for high molecular weight kininogen on stimulated washed human platelets, Biochemistry, (1984) 6863 – 6869. [29] F.J. Meloni, E.J. Gustafson, A.H. Schmaier, High molecular weight kininogen binds to platelets by its heavy and light chains and when bound has altered susceptibility to kallikrein cleavage, Blood, (1992) 1233 – 1244. [30] A.H. Schmaier, A. Kuo, D. Lundberg, S. Murray, D.B. Cines, The expression of high molecular weight kininogen on human umbilical vein endothelial cells, J. Biol. Chem., (1988) 16327 – 16333.
8
A.E. Davis III / Clinical Immunology 114 (2005) 3–9
[31] F. Van Iwaarden, P.G. deGroot, B.N. Bouma, The binding of high molecular weight kininogen to cultured human endothelial cells, J. Biol. Chem., (1988) 4698 – 4703. [32] R.W. Colman, A.H. Schmaier, Contact system: a vascular biology modulator with anticoagulant, profibrinolytic, antiadhesive, and proinflammatory attributes, Blood, (1997) 3819 – 3843. [33] A.A.K. Hasan, T. Zisman, A.H. Schmaier, Identification of cytokeratin 1 as a binding protein and presentation receptor for kininogens on endothelial cells, Proc. Natl. Acad. Sci. U. S. A., (1998) 3615 – 3620. [34] H. Herwald, J. Dedio, R. Kellner, M. Loos, W. Muller-Esterl, Isolation and characterization of the kininogen-binding protein p33 from endothelial cells, J. Biol. Chem., (1996) 13040 – 13047. [35] K. Joseph, B. Ghebrehiwet, E.I.B. Peerschke, K.B.M. Reid, A.P. Kaplan, Identification of the zinc-dependent endothelial cell binding protein for high molecular weight kininogen and factor XII: identity with the receptor that binds to the globular ‘‘heads’’ of C1q (gC1q-r), Proc. Natl. Acad. Sci. U. S. A., (1996) 8552 – 8557. [36] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Mapping binding domains of kininogens on endothelial cell cytokeratin 1, J. Biol. Chem., (1999) 7137 – 7145. [37] R.W. Colman, R.A. Pixley, S. Najamunnisa, W. Yan, J. Wang, A. Mazar, K.R. McCrae, Binding of high molecular weight kininogen to human endothelial cells is mediated via a site within domains 2 and 3 of the urokinase receptor, J. Clin. Invest., (1997) 1481 – 1487. [38] F. Mahdi, Z. shariat-Madar, R.F. Todd III, C.D. Figueroa, A.H. Schmaier, Expression and co-localization of cytokeratin 1 and urokinase plasminogen activator receptor on endothelial cells, Blood, (2001) 2342 – 2350. [39] S.R. Reddigari, Y. Shibayama, T. Brunnee, A.P. Kaplan, Human hageman factor (factor xii) and high molecular weight kininogen compete for the same binding site on human umbilical vein endothelial cells, J. Biol. Chem., (1993) 11982 – 11987. [40] K. Joseph, Y. Shibayama, B. Ghebrehiwet, A.P. Kaplan, Factor XII dependent contact activation on endothelial cells and binding proteins gc1qr and cytokeratin 1, Thromb. Haemostasis, (2001) 119 – 124. [41] K. Joseph, B.G. Tholanikunnel, A.P. Kaplan, Heat shock protein 90 catalyzes activation of the prekallikrein – kininogen complex in the absence of factor xii, Proc. Natl. Acad. Sci. U. S. A., (2002) 896 – 900. [42] K. Joseph, B.G. Tholanikunnel, A.P. Kaplan, Activation of the bradykinin-forming cascade on endothelial cells: a role for heat shock protein 90, Int. Immunopharmacol., (2002) 1851 – 1859. [43] A.P. Kaplan, K. Joseph, M. Silverberg, Pathways for bradykinin formation and inflammatory disease, J. Allergy Clin. Immunol., (2002) 195 – 209. [44] F. Mahdi, Z. Shariat-Madar, C.D. Figueroa, A.H. Schmaier, Factor XII interacts with the multiprotein assembly of urokinase plasminogen activator receptor, gC1qr, and cytokeratin 1 on endothelial cell membranes, Blood, (2002) 3585 – 3596. [45] Y. Zhao, Q. Qiu, F. Mahdi, Z. Shariat-Madar, R. Rojkjaer, A.H. Schmaier, Assembly and activation of hk – pk complex on endothelial cells results in bradykinin liberation and no formation, Am. J. Physiol.: Heart Circ. Physiol., (2001) H1821 – H1829. [46] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Identification and characterization of prolylcarboxypeptidase as an endothelial cell prekallikrein activator, J. Biol. Chem., (2002) 17962 – 17969. [47] Z. Shariat-Madar, F. Mahdi, A.H. Schmaier, Assembly and activation of the plasma kallikrein/kinin system: a new interpretation, Int. Immunopharmacol., (2002) 1841 – 1849. [48] A.H. Schmaier, The plasma kallikrein – kinin system counterbalances the renin – angiotensin system, J. Clin. Invest., (2002) 1007 – 1009. [49] I. Gigli, J.W. Mason, R.W. Colman, K.F. Austen, Interaction of plasma kallikrein with the C1 inhibitor, J. Immunol., (1970) 574 – 581. [50] D.J. McConnell, Inhibitors of kallikrein in human plasma, J. Clin. Invest., (1972) 1611 – 1623. [51] A. de Agostini, H.R. Lijnen, R.A. Pixley, R.W. Colman, M. Schapira, Inactivation of factor-XII active fragment in normal plasma: predominant role of C1-inhibitor, J. Clin. Invest., (1984) 1542 – 1549.
[52] V.H. Donaldson, R.R. Evans, A biochemical abnormality in hereditary angioneurotic edema, Am. J. Med., (1963) 37 – 44. [53] P. Lachmann, F. Rosen, The catabolism of C1-inhibitor and the pathogenesis of hereditary angioedema, Acta Pathol., Microbiol. Immunol. Scand., (1984) 35 – 39. [54] P. Woo, P.J. Lachmann, R.A. Harrison, N. Amos, C. Cooper, F.S. Rosen, Simultaneous turnover of normal and dysfunctional c1 inhibitor as a probe of in vivo activation of C1 and contact activatable proteases, Clin. Exp. Immunol., (1985) 1 – 8. [55] M. Quastel, R. Harrison, M. Cicardi, C. Alper, F. Rosen, Behavior in vivo of normal and dysfunctional C1 inhibitor in normal subjects and patients with hereditary angioneurotic edema, J. Clin. Invest., (1983) 1041 – 1046. [56] S.C. Ernst, A. Circolo, A.E. Davis III, K. Gheesling-Mullis, M. Fliesler, R.C. Strunk, Impaired production of both normal and mutant C1 inhibitor proteins in type I hereditary angioedema with a duplication in exon 8, J. Immunol., (1996) 405 – 410. [57] J. Kramer, F. Rosen, H. Colten, K. Rajezy, R. Strunk, Transinhibition of C1 inhibitor synthesis in type I hereditary angioneurotic edema, J. Clin. Invest., (1993) 1258 – 1262. [58] A.B. Laurell, U. Martensson, A.G. Sjoholm, C1 subcomponent complexes in normal and pathological sera studied by crossed immunoelectrophoresis, Acta Pathol. Microbiol. Scand., (1976) 455 – 464. [59] R. Buhler, J.K. Hovinga, I. Aebi-Huber, M. Furlan, B. Lammle, Improved detection of proteolytically cleaved high molecular weight kininogen by immunoblotting using an antiserum against its reduced 47 kda light chain, Blood Coagulation Fibrinolysis, (1995) 223 – 232. [60] M. Berrettini, B. Lammle, T. White, M.J. Heeb, H.P. Schwarz, B. Zuraw, J. Curd, J.H. Griffin, Detection of in vitro and in vivo cleavage of high molecular weight kininogen in human plasma by immunoblotting with monoclonal antibodies, Blood, (1986) 455 – 462. [61] E.W. Nielsen, H.T. Johansen, K. Hogasen, W. Wuillemin, C.E. Hack, T.E. Mollnes, Activation of the complement, coagulation, fibrinolytic and kallikrein – kinin systems during attacks of hereditary angioedema, Scand. J. Immunol., (1996) 185 – 192. [62] M. Cugno, M. Cicardi, R. Coppola, A. Agostoni, Activation of factor XII and cleavage of high molecular weight kininogen during acute attacks in hereditary and acquired C1-inhibitor deficiencies, Immunopharmacology, (1996) 361 – 364. [63] V.H. Donaldson, Kinin formation in hereditary angioneurotic edema (HANE) plasma, Int. Arch. Allergy, (1973) 206 – 209. [64] V.H. Donaldson, E. Merler, F.S. Rosen, K.W. Kretschmer, I.H. Lepow, A polypeptide kinin in hereditary angioneurotic edema plasma: role of complement in its formation, J. Lab. Clin. Med., (1970) 986. [65] V.H. Donaldson, O.D. Ratnoff, W.D.d. Silva, F.S. Rosen, Permeability-increasing activity in hereditary angioneurotic edema plasma, J. Clin. Invest., (1969) 642 – 653. [66] C.J. Strang, K.S. Auerbach, F.S. Rosen, C1s-induced vascular permeability in C2-deficient guinea pigs, J. Immunol., (1986) 631 – 635. [67] C. Strang, S. Cholin, J. Spragg, A. Davis III, E. Schneeberger, V. Donaldson, F. Rosen, Angioedema induced by a peptide derived from complement component C2, J. Exp. Med., (1988) 1685 – 1698. [68] J.G. Curd, J.L.J. Prograis, C.G. Cochrane, Detection of active kallikrein in induced blister fluids of hereditary angioedema patients, J. Exp. Med., (1980) 742 – 747. [69] J.G. Curd, M. Yelvington, N. Burridge, N.P. Stimler, C. Gerard, J.L.J. Prograis, C.G. Cochrane, H.J. Muller-Eberhard, Generation of bradykinin during incubation of hereditary angioedema plasma, Mol. Immunol., (1982) 1365 (abstract). [70] T. Fields, B. Ghebrewihet, A. Kaplan, Kinin formation in hereditary angioedema plasma: evidence against kinin derivation from C2 and in support of spontaneous formation of bradykinin, J. Allergy Clin. Immunol., (1983) 54 – 60. [71] L.R. Shoemaker, S.J. Schurman, V.H. Donaldson , A.E. Davis III, Hereditary angioneurotic edema: characterization of plasma kinin
A.E. Davis III / Clinical Immunology 114 (2005) 3–9 and vascular permeability-enhancing activities, Clin. Exp. Immunol., (1994) 22 – 28. [72] J. Nussberger, M. Cugno, C. Amstutz, M. Cicardi, A. Pellacani, A. Agostoni, Plasma bradykinin in angio-oedema, Lancet, (1998) 1693 – 1697. [73] J.J. Wisnieski, T.C. Knauss, I. Yike, D.G. Dearborn, R.L. Narvy, G.B. Naff, Unique C1 inhibitor dysfunction in a kindred without angioedema. I. A mutant c1 inhibitor that inhibits C1s but not C1r, J. Immunol., (1994) 3199 – 3209. [74] R. Zahedi, J.J. Bissler, A.E. Davis III, C. Andreadis, J.J. Wisnieske, Unique C1 inhibitor dysfunction in a kindred without angioedema. II. Identification of an Ala443-Val substitution and functional analy-
9
sis of the recombinant mutant protein, J. Clin. Invest., (1995) 1299 – 1305. [75] R. Zahedi, J. Wisnieski, A.E. Davis III, Role of the P2 residue of complement 1 inhibitor (Ala443) in determination of target protease specificity, J. Immunol., (1997) 983 – 988. [76] E.D. Han, R.C. MacFarlane, A.N. Mulligan, J. Scafidi, A.E. Davis III, Increased vascular permeability in C1 inhibitor-deficient mice is mediated by the bradykinin type 2 receptor, J. Clin. Invest., (2002) 1057 – 1063. [77] E. Han Lee, E. Pappalardo, J. Scafidi, A.E. Davis III, Approaches toward reversal of increased vascular permeability in C1 inhibitor deficient mice, Immunol. Lett., (2003) 155 – 160.