NZ727063B2 - Compositions and methods of inhibiting MASP-1, MASP-2 and/or MASP-3 for treatment of paroxysmal nocturnal hemoglobinuria - Google Patents

Abstract

Disclosed is a pharmaceutical composition comprising a MASP-2 inhibitory monoclonal antibody that specifically binds to human MASP-2 and a MASP-3 inhibitory monoclonal antibody that specifically binds to the serine protease domain of MASP-3 and inhibits factor D maturation. The composition inhibits both the lectin and alternative pathway of complement. both the lectin and alternative pathway of complement.

Description

Divisional application out of NZ 629675 NEW ZEALAND PATENTS ACT 1953 COMPLETE SPECIFICATION Compositions and methods of inhibiting MASP-1, MASP-2 and/or MASP-3 for treatment of paroxysmal nocturnal hemoglobinuria We, Omeros ation, of 201 Elliott Avenue West, Seattle, 98119, Washington, United States of America; and University of Leicester, of University Road, Leicestershire, LE1 7RH, United Kingdom hereby declare the invention, for which we pray that a patent may be granted to us and the method by which it is to be med, to be ularly described in and by the following statement: - 1 – (followed by page 1a) COMPOSITIONS AND METHODS OF INHIBITING MASP-1 AND/OR MASP-2 AND/OR MASP-3 FOR THE TREATMENT OF PAROXYSMAL NOCTURNAL HEMOGLOBINURIA CROSS-REFERENCE TO RELATED APPLICATION This application claims the benefit of Application No. 61/621,461 filed April 6, 2012. This ation was divided from New Zealand Patent ation No. 629675.

STATEMENT REGARDING SEQUENCE LISTING The sequence listing associated with this application is provided in text format in lieu of a paper copy and is hereby incorporated by reference into the specification. The name of the text file ning the sequence listing is MP_1_0146_PCT_Sequence_20130327_ST25.txt. The text file is 85 KB; was created on April 1, 2013; and is being submitted via EFS-Web with the filing of the specification.

OUND The complement system provides an early acting mechanism to initiate, amplify and orchestrate the immune response to microbial infection and other acute insults (M.K. Liszewski and J.P. Atkinson, 1993, in Fundamental Immunology, Third Edition, edited by W.E. Paul, Raven Press, Ltd., New York), in humans and other vertebrates.

While complement activation es a valuable first-line defense against potential pathogens, the activities of complement that promote a protective immune response can also ent a potential threat to the host (K.R. Kalli, et al., Springer Semin.

Immunopathol. 15 :417-431, 1994; B.P. Morgan, Eur. J. al Investig. 24 :219-228, 1994). For example, C3 and C5 proteolytic products recruit and activate neutrophils.

While indispensable for host defense, activated neutrophils are riminate in their release of destructive enzymes and may cause organ damage. In addition, complement activation may cause the deposition of lytic complement components on nearby host cells as well as on microbial s, resulting in host cell lysis.

The ment system has also been implicated in the pathogenesis of numerous acute and chronic disease , including: myocardial infarction, stroke, ARDS, reperfusion injury, septic shock, capillary e following thermal burns, -1a- (followed by page 2) postcardiopulmonary bypass inflammation, transplant rejection, rheumatoid arthritis, multiple sclerosis, myasthenia gravis, and mer's disease. In almost all of these conditions, complement is not the cause but is one of several factors involved in enesis. heless, complement activation may be a major pathological mechanism and represents an effective point for al control in many of these disease states. The growing recognition of the importance of complement-mediated tissue injury in a variety of disease states underscores the need for effective complement inhibitory drugs. To date, Eculizumab (Solaris®), an antibody against C5, is the only complement- targeting drug that has been approved for human use. Yet, C5 is one of several effector molecules located tream” in the complement system, and blockade of C5 does not inhibit activation of the complement system. Therefore, an inhibitor of the initiation steps of complement activation would have significant advantages over a “downstream” complement inhibitor.

Currently, it is widely accepted that the complement system can be activated through three distinct ys: the classical pathway, the lectin pathway, and the alternative pathway. The classical pathway is usually triggered by a x ed of host antibodies bound to a foreign particle (z'.e., an antigen) and thus requires prior exposure to an antigen for the generation of a specific dy response. Since activation of the classical pathway depends on a prior adaptive immune response by the host, the classical pathway is part of the acquired immune . In contrast, both the lectin and alternative pathways are independent of adaptive immunity and are part of the innate immune system.

The activation of the ment system results in the sequential activation of serine se zymogens. The first step in activation of the classical pathway is the binding of a specific recognition molecule, Clq, to antigen-bound IgG and IgM molecules. Clqis associated with the Clr and Cls serine protease proenzymes as a complex called Cl. Upon binding of Clq to an immune complex, autoproteolytic cleavage of the Arg-Ile site of Clr is followed by Clr-mediated cleavage and tion of Cls, which thereby acquires the ability to cleave C4 and C2. C4 is cleaved into two fragments, ated C4a and C4b, and, similarly, C2 is cleaved into C2a and C2b. C4b fragments are able to form covalent bonds with adjacent hydroxyl or amino groups and generate the C3 convertase (C4b2a) through noncovalent interaction with the C2a nt of activated C2. C3 convertase (C4b2a) activates C3 by proteolytic cleavage into C3a and C3b subcomponents g to generation of the C5 convertase (C4b2a3b), which, by cleaving C5 leads to the formation of the membrane attack complex (C5b combined with C6, C7, C8 and C-9, also referred to as “MAC”) that can disrupt cellular membranes ing in cell lysis. The activated forms of C3 and C4 (C3b and C4b) are covalently deposited on the foreign target surfaces, which are recognized by complement receptors on multiple phagocytes.

Independently, the first step in activation of the complement system through the lectin y is also the binding of specific recognition molecules, which is followed by the activation of associated serine protease ymes. However, rather than the binding of immune complexes by Clq, the recognition molecules in the lectin pathway comprise a group of carbohydrate-binding proteins (mannan-binding lectin (MBL), H-ficolin, M-ficolin, L-ficolin and C-type lectin CL-l 1), collectively referred to as lectins. See J. Lu et al., Biochim. Biophys. Acta 1572:387-400, ; Holmskov et al., Annu. Rev. Immunol. 21:547-578 (2003); Teh et al., Immunology [01:225-232 (2000)).

See also J. Luet et al., Biochim Biophys Acta 1572:387-400 (2002); Holmskov et al, Annu Rev Immunol 21:547-578 (2003); Teh et al., Immunology 101:225-232 (2000); Hansen et al, J. l 185(10):6096-6104 .

Ikeda et al. first demonstrated that, like Clq, MBL could activate the complement system upon binding to yeast mannan-coated erythrocytes in a C4-dependent manner (Ikeda et al., J. Biol. Chem. 262:7451-7454, (1987)). MBL, a member of the collectin protein family, is a calcium-dependent lectin that binds carbohydrates with 3-and 4- hydroxy groups oriented in the equatorial plane of the pyranose ring. Prominent s for MBL are thus ose and yl-D-glucosamine, while carbohydrates not fitting this steric requirement have ctable affinity for MBL (Weis et al., Nature 360:127-134, (1992)). The interaction between MBL and monovalent sugars is ely weak, with dissociation constants typically in the single-digit millimolar range.

MBL achieves tight, specific binding to glycan ligands by avidity, i.e., by interacting simultaneously with multiple monosaccharide residues located in close proximity to each other (Lee et al., Archiv. Biochem. Biophys. 299:129-136, (1992)). MBL izes the carbohydrate ns that commonly decorate microorganisms such as bacteria, yeast, parasites and n viruses. In contrast, MBL does not recognize ctose and sialic acid, the penultimate and ultimate sugars that usually decorate "mature" complex glycoconjugates present on mammalian plasma and cell surface glycoproteins. This WO 80834 binding specificity is thought to promote recognition of “foreign” es and help protect from “self-activation.” However, MBL does bind with high affinity to rs of high-mannose "precursor" glycans on N-linked glycoproteins and glycolipids sequestered in the endoplasmic reticulum and Golgi of mammalian cells (Maynard et al., J. Biol.

Chem. 25 73788-3794, (1982)). In addition, it has been shown that MBL can bind the polynucleotides, DNA and RNA, which may be d on necrotic and apoptotic cells (Palaniyar et al., Ann. NY. Acad. Sell, 1010:467-470 (2003); Nakamura et al., J. Leuk.

Biol. 86:737-748 (2009)). Therefore, damaged cells are potential targets for lectin pathway tion via MBL binding.

The ficolins possess a different type of lectin domain than MBL, called the fibrinogen-like . ns bind sugar residues in a CaII-independent manner. In humans, three kinds of ficolins (L-ficolin, in and H-ficolin) have been fied.

The two serum ficolins, L-ficolin and H-ficolin, have in common a city for N—acetyl-D-glucosamine; however, H-ficolin also binds N—acetyl-D-galactosamine. The difference in sugar specificity of L-ficolin, H-ficolin, CL-ll, and MBL means that the different lectins may be mentary and target different, though overlapping, glycoconjugates. This concept is supported by the recent report that, of the known lectins in the lectin pathway, only L-ficolin binds specifically to lipoteichoic acid, a cell wall glycoconjugate found on all Gram-positive bacteria (Lynch et al., J. Immunol. 1 72:1198-1202, (2004)). In addition to acetylated sugar moieties, the ficolins can also bind acetylated amino acids and polypeptides (Thomsen et al., M01. Immunol. 48(4):369- 81 (2011)). The collectins (i.e., MBL) and the ficolins bear no significant similarity in amino acid sequence. However, the two groups of proteins have similar domain organizations and, like Clq, assemble into oligomeric structures, which maximize the ility of multisite binding.

The serum concentrations of MBL are highly variable in healthy populations and this is genetically controlled by polymorphisms/mutations in both the promoter and coding regions of the MBL gene. As an acute phase protein, the expression of MBL is further upregulated during inflammation. L-ficolin is present in serum at concentrations r to those of MBL. Therefore, the L-ficolin branch of the lectin pathway is potentially comparable to the MBL arm in th. MBL and ficolins can also function as opsonins, which allow ytes to target MBL- and ficolin-decorated surfaces (see Jack et al., J Leukoc Biol., 77(3):328-36 (2004), Matsushita and Fujita, Immunobz'ology, 2013/035488 205(4-5):490-7 (2002), Aoyagi et al., J Immunol, 174(1):418-25(2005). This opsonization es the interaction of these proteins with phagocyte receptors (Kuhlman et al., J. Exp. Med. [69:1733, ; Matsushita et al., J. Biol. Chem. 271:2448-54, (1996)), the indentity of which has not been established.

Human MBL forms a specific and high-affinity interaction through its collagen-like domain with unique Clr/Cls-like serine proteases, termed MBL-associated serine ses (MASPs). To date, three MASPs have been described. First, a single enzyme "MASP" was identified and characterized as the enzyme responsible for the initiation of the complement cascade (i.e., cleaVing C2 and C4) (Matsushita et al., J Exp Med 176(6):1497-1502 (1992); Ji et al., J. Immunol. [50:571-578, (1993)). It was uently determined that the MASP actiVity was, in fact, a mixture of two proteases: MASP-l and MASP-2 (Thiel et al., Nature 386:506-510, (1997)). However, it was demonstrated that the MBL-MASP-2 x alone is sufficient for complement activation (Vorup-Jensen et al., J. l. [65:2093-2100, (2000)). Furthermore, only MASP-2 cleaved C2 and C4 at high rates (Ambrus et al., J. Immunol. [70:1374-1382, (2003)). Therefore, MASP-2 is the protease responsible for activating C4 and C2 to generate the C3 convertase, C4b2a. This is a significant difference from the Cl complex of the classical pathway, where the nated action of two specific serine proteases (Clr and Cls) leads to the activation of the complement system. In addition, a third novel protease, MASP-3, has been isolated (Dahl, M.R., et a1., ty 15:127-35, 2001). MASP-l and MASP-3 are alternatively spliced products of the same gene.

MASPs share identical domain zations with those of Clr and Cls, the enzymatic components of the Cl complex (Sim et a1., Biochem. Soc. Trans. 28:54.5, (2000)). These domains include an N-terminal Clr/Cls/sea urchin VEGF/bone morphogenic protein (CUB) domain, an epidermal growth factor-like domain, a second CUB domain, a tandem of complement control protein domains, and a serine protease domain. As in the Cl proteases, activation of MASP-2 occurs through ge of an Arg-Ile bond adjacent to the serine se domain, which splits the enzyme into disulfide-linked A and B chains, the latter consisting of the serine protease .

MBL can also associate with an alternatively spliced form of MASP-2, known as MBL-associated protein of 19 kDa (MAp19) or small MBL-associated protein , which lacks the catalytic ty of MASP-2. (Stover, J. Immunol. [62:3481-90, (1999); Takahashiet a1., Int. Immunol.11:859-863, (1999)). MAp19 comprises the first two domains of MASP-2, followed by an extra sequence of four unique amino acids. The function of Mapl9 is unclear (Degn et al., J Immunol. Methods, 2011). The MASP-l and MASP-2 genes are located on human chromosomes 3 and 1, tively eble et al., Immunobz’ology 205:455-466, (2002)).

Several lines of evidence suggest that there are different MBL-MASP complexes and a large fraction of the MASPs in serum is not complexed with MBL (Thiel,et al., J.

Immunol. [65:878-887, (2000)). Both H— and L-ficolin bind to all MASPs and activate the lectin complement pathway, as does MBL (Dahl et al., Immunity 15:127-35, (2001); Matsushita et al., J. Immunol.168:3502-3506, ). Both the lectin and classical pathways form a common C3 convertase (C4b2a) and the two pathways converge at this step.

The lectin pathway is widely thought to have a major role in host defense against infection in the naive host. Strong evidence for the involvement of MBL in host defense comes from analysis of patients with decreased serum levels of fianctional MBL (Kilpatrick, Biochim. Biophys. Acta 1572:401-413, ). Such patients display susceptibility to recurrent bacterial and fiangal infections. These symptoms are usually evident early in life, during an apparent window of vulnerability as maternally derived antibody titer wanes, but before a filll repertoire of dy responses develops. This syndrome often results from mutations at several sites in the collagenous portion of MBL, which interfere with proper ion of MBL ers. However, since MBL can function as an opsonin independent of complement, it is not known to what extent the increased susceptibility to ion is due to impaired complement activation.

In contrast to the cal and lectin pathways, no initiators of the alternative pathway have previously been found to fulfill the recognition functions that Clq and lectins perform in the other two pathways. Currently it is widely accepted that the alternative pathway neously undergoes a low level of turnover activation, which can be readily amplified on foreign or other abnormal surfaces (bacteria, yeast, virally infected cells, or damaged tissue) that lack the proper molecular elements that keep spontaneous complement activation in check. There are four plasma proteins directly involved in the activation of the alternative y: C3, s B and D, and properdin.

Although there is ive evidence implicating both the classical and alternative complement pathways in the enesis of non-infectious human diseases, the role of the lectin pathway is just beginning to be evaluated. Recent studies provide evidence that tion of the lectin pathway can be sible for ment activation and related inflammation in ischemia/reperfusion injury. Collard et al. (2000) reported that cultured elial cells subjected to oxidative stress bind MBL and show deposition of C3 upon exposure to human serum (Collard et al., Am. J. Pathol. [56:1549-1556, (2000)). In addition, treatment of human sera with ng anti-MBL onal antibodies inhibited MBL binding and complement activation. These findings were extended to a rat model of myocardial ischemia-reperfusion in which rats d with a blocking antibody directed t rat MBL showed significantly less dial damage upon occlusion of a coronary artery than rats treated with a control antibody (Jordan et al., Circulation [04:1413-1418, (2001)). The molecular mechanism of MBL binding to the vascular endothelium after oxidative stress is unclear; a recent study suggests that activation of the lectin pathway after oxidative stress may be mediated by MBL binding to vascular endothelial cytokeratins, and not to glycoconjugates (Collard et al., Am. J.

Pat/ml. [59:1045-1054, (2001)). Other studies have implicated the cal and alternative pathways in the pathogenesis of ischemia/reperfusion injury and the role of the lectin pathway in this disease remains controversial (Riedermann, N.C., et al., Am. J.

Pathol. [62:363-367, 2003).

Recent studies have shown that MASP-l and MASP-3 convert the alternative pathway activation enzyme factor D from its zymogen form into its tically active form (see Takahashi M. et al., JExp Med 207(1):29-37 (2010); Iwaki et al., J. l. 187:3751-58 ). The logical importance of this process is ined by the absence of alternative pathway filnctional actiVity in plasma of MASP-1/3-deficient mice.

Proteolytic generation of C3b from native C3 is required for the alternative pathway to function. Since the alternative pathway C3 convertase (C3bBb) contains C3b as an essential subunit, the question regarding the origin of the first C3b Via the alternative pathway has presented a puzzling problem and has stimulated considerable research.

C3 belongs to a family of proteins (along with C4 and (x-2 macroglobulin) that contain a rare posttranslational modification known as a thioester bond. The ter group is composed of a glutamine whose terminal carbonyl group forms a covalent thioester linkage with the sulfhydryl group of a cysteine three amino acids away. This bond is unstable and the electrophilic glutamyl-thioester can react with nucleophilic moieties such as hydroxyl or amino groups and thus form a covalent bond with other molecules. The thioester bond is reasonably stable when sequestered within a hydrophobic pocket of intact C3. However, proteolytic cleavage of C3 to C3a and C3b results in exposure of the highly reactive thioester bond on C3b and, following nucleophilic attack by adjacent moieties comprising hydroxyl or amino groups, C3b becomes covalently linked to a target. In addition to its well-documented role in covalent attachment of C3b to complement targets, the C3 thioester is also thought to have a pivotal role in triggering the alternative pathway. According to the widely accepted "tick-over theory", the alternative pathway is initiated by the generation of a fluid-phase convertase, iC3Bb, which is formed from C3 with hydrolyzed thioester (iC3; C3(H20)) and factor B (Lachmann, P.J., et al., Springer Semin. Immunopathol. 7:143-162, ).

The C3b-like C3(H20) is generated from native C3 by a slow spontaneous hydrolysis of the internal thioester in the n (Pangbum, M.K., et al., J. Exp. Med. 6-867, 1981). Through the activity of the )Bb convertase, C3b les are deposited on the target surface thereby initiating the alternative pathway.

Prior to the instant discovery described herein, very little was known about the initiators of activation of the alternative pathway. Activators were thought to include yeast cell walls (zymosan), many pure polysaccharides, rabbit erythrocytes, n immunoglobulins, s, fiangi, bacteria, animal tumor cells, tes, and d cells. The only feature common to these activators is the presence of carbohydrate, but the complexity and variety of carbohydrate structures has made it difficult to establish the shared molecular determinants which are ized. It has been widely accepted that alternative y activation is lled through the fine balance between inhibitory regulatory components of this pathway, such as factor H, factor I, DAF, and CR1, and din, the latter of which is the only positive regulator of the alternative pathway (see Schwaeble W.J. and Reid K.B., Immunol Today 20(1): 17-21 (1999)).

In addition to the apparently unregulated activation mechanism described above, the alternative pathway can also provide a powerful amplification loop for the lectin/classical pathway C3 convertase (C4b2a) since any C3b generated can participate with factor B in forming additional alternative pathway C3 convertase (C3bBb). The alternative pathway C3 convertase is stabilized by the binding of properdin. Properdin extends the alternative pathway C3 convertase half-life six to ten fold. on of C3b to the alternative y C3 convertase leads to the formation of the alternative pathway C5 convertase.

All three pathways (i.e., the classical, lectin and alternative) have been thought to converge at C5, which is cleaved to form products with multiple proinflammatory effects.

The converged pathway has been referred to as the terminal complement pathway. C5a is the most potent anaphylatoxin, inducing alterations in smooth muscle and vascular tone, as well as vascular permeability. It is also a powerful chemotaxin and activator of both neutrophils and monocytes. CSa-mediated cellular activation can significantly amplify inflammatory responses by inducing the release of multiple additional inflammatory ors, including cytokines, hydrolytic enzymes, donic acid metabolites, and ve oxygen species. C5 cleavage leads to the formation of C5b-9, also known as the ne attack complex (MAC). There is now strong eVidence that sublytic MAC deposition may play an important role in inflammation in addition to its role as a lytic pore-forming complex.

In addition to its essential role in immune defense, the complement system contributes to tissue damage in many clinical conditions. Thus, there is a pressing need to develop therapeutically effective complement inhibitors to prevent these adverse effects.

In one aspect, the present invention provides a method of inhibiting MASP dependent complement activation in a subject suffering from paroxysmal nocturnal hemoglobinuria (PNH). The method includes the step of administering to the subject a ition comprising an amount of a MASP-3 tory agent effective to inhibit -dependent complement activation. In some embodiments, the method r comprises administering to the t a composition comprising a MASP-2 inhibitory agent.

In another , the present invention provides a pharmaceutical composition comprising at least one inhibitory agent, wherein the at least one inhibitory agent comprises a MASP-2 inhibitory agent and a MASP-3 inhibitory agent and a pharmaceutically acceptable carrier.

In another aspect, the present invention es a pharmaceutical composition comprising a MASP-3 inhibitory agent that binds to a portion of MASP-l (SEQ ID NO: : full-length) and that also binds to a portion of MASP-3 (SEQ ID NO:8) and a ceutical carrier. 2013/035488 In another aspect, the present invention provides a pharmaceutical composition comprising a MASP-3 inhibitory agent that binds to a portion of MASP-2 (SEQ ID NO: : full-length) and that also binds to a n of MASP-3 (SEQ ID NO:8) and a pharmaceutical carrier.

In another aspect, the present invention provides a pharmaceutical composition comprising a MASP-3 inhibitory agent that binds to a portion of MASP-l (SEQ ID NO: : full-length) and that also binds to a portion of MASP-2 (SEQ ID NO:5) and a pharmaceutical carrier.

In r aspect, the present invention provides a pharmaceutical composition comprising a MASP-3 inhibitory agent that binds to a portion of MASP-l (SEQ ID NO: full length), that binds to a portion of MASP-2 (SEQ ID NO: 5: fiJll-length) and that also binds to a portion of MASP-3 (SEQ ID NO:8) and a pharmaceutical carrier.

As bed herein, the pharmaceutical compositions of the invention can be used in accordance with the methods of the invention.

These and other aspects and embodiments of the herein described ion will be evident upon reference to the following detailed description and drawings. All of the US. patents, US. patent application ations, US. patent applications, foreign patents, foreign patent applications and non-patent publications referred to in this specification are incorporated herein by reference in their entirety, as if each was incorporated individually.

DESCRIPTION OF THE DRAWINGS The ing aspects and many of the attendant ages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein: FIGURE 1 illustrates a new understanding of the lectin and alternative pathways; FIGURE 2 is a schematic diagram adapted from Schwaeble et al., Immunobz'ol 205:455-466 , as modified by ng et al., BBA 1824253 (2012), illustrating the MASP-2 and MApl9 protein domains and the exons encoding the same; FIGURE 3 is a schematic diagram d from Schwaeble et al., Immunobz'ol 205:455-466 (2002), as modified by Yongqing et al., BBA 3 (2012), illustrating the MASP-l, MASP-3 and MAp44 protein domains and the exons encoding the same; FIGURE 4 shows an alignment of the amino acid sequences of the MASP-l, MASP-2 and MASP-3 proteins and indicates consensus regions therebetween; FIGURE 5 shows an alignment of the amino acid sequences of the MASP-l, MASP-2 and MASP-3 Alpha chains; FIGURE 6 shows an alignment of the amino acid sequences of the MASP-l, MASP-2 and MASP-3 Beta ;