US20170304398A1

US20170304398A1 - Polymer conjugates of box-a of hmgb1 and box-a variants of hmgb1

Info

Publication number: US20170304398A1
Application number: US15/581,358
Authority: US
Inventors: Silvio Traversa; Chiara Lorenzetto; Valentina Mainero; Sebastiano Morena; Silvano Fumero; Luca Beccaria
Original assignee: Creabilis Therapeutics SpA
Current assignee: Creabilis Therapeutics SpA
Priority date: 2006-09-15
Filing date: 2017-04-28
Publication date: 2017-10-26
Also published as: EP2068935B8; HK1134779A1; JP2010503630A; US8546547B2; CA2663300A1; US20140065094A1; CA2663300C; AU2007296843A1; MX2009002820A; EP2068935B1; WO2008031612A1; US9707298B2; AU2007296843C1; AU2007296843B2; JP2015193646A; HRP20110487T1; US20090324677A1; IL197441A0; IL197441A; EP2068935A1

Abstract

The present invention relates to novel polymer conjugates of polypeptide variants of the HMGB1 high affinity binding domain Box-A (HMGB1 Box-A) or of a biologically active fragment of HMGB1 Box-A. Further, the invention relates to novel polymer conjugates of polypeptide variants of the HMGB1 high affinity binding domain Box-A (HMGB1 Box-A). Moreover, the present invention concerns the use of said polymer conjugates of polypeptide molecules of HMGB1 Box-A to diagnose, prevent, alleviate and/or treat pathologies associated with extracellular HMGB1 and/or associated with an increased expression of RAGE.

Description

This application is divisional of Ser. No. 13/723,455 filed Dec. 27, 2012, which is a divisional of Ser. No. 12/441,478 filed Mar. 16, 2009, now U.S. Pat. No. 8,546,547, which is a 35 U.S.C. 371 National Phase Entry Application from PCT/EP2007/008029, filed Sep. 14, 2007, which claims the benefit of U.S. Patent Application No. 60/904,776, filed on Mar. 5, 2007, and European Application No. 06019362.0 filed Sep. 15, 2006, the disclosures of which are incorporated herein in their entirety by reference.

DESCRIPTION

The present invention relates to novel polymer conjugates of the HMGB1 high affinity binding domain Box-A (HMGB1 Box-A) or of a biologically active fragment of HMGB1 Box-A. Further, the invention relates to novel polymer conjugates of polypeptide variants of the HMGB1 high affinity binding domain Box-A (HMGB1 Box-A) or of the biologically active fragments of HMGB1 Box-A. Moreover, the present invention concerns the use of said polymer conjugates to diagnose, prevent, alleviate and/or treat pathologies associated with extracellular HMGB1 and/or associated with an increased expression of RAGE.
The HMGB1 protein belongs to the family of high mobility group (HMG) proteins. HMG proteins, so-called due to their high electrophoretic mobility in polyacrylamide gels, are the most ubiquitous non-histone proteins associated with isolated chromatin in eukaryotic cells. These proteins play a generalized “architectural” role in DNA bending, looping, folding and wrapping, since they either distort, bend or modify DNA structures and complexes with transcription factors or histones (Andersson et oh, 2002: Agresti et al., 2003; Degryse et al., 2003). The high mobility group 1 (HMGB1) protein is usually a nuclear factor, in particular a transcriptional regulatory molecule causing DNA bending and facilitating the binding of several transcriptional complexes.
Structurally, the HMGB1 protein is a protein of approximately 25 kDa with a highly conserved sequence among mammals, whereby 2 out of 214 amino acids have conservative substitutions in all mammalian species. HMGB1 is ubiquitously present in all vertebrate nuclei and in particular can be found in fibroblasts, neurons, hepatocytes, glia and in cells derived from hematopoietic stem cells, including monocytes/macrophages, neutrophils and platelets. The HMGB1 molecule has a tripartite structure composed of three distinct domains: two DNA binding domains called HMG Box-A and Box-B, and an acid carboxyl terminus, making it bipolarly charged.
The two HMGB1 boxes are involved in the protein's function as non-sequence-specific architectural DNA-binding elements, conferring the ability to bind DNA into recognized distorted DNA structures and stabilizing nucleosome assembly, remodelling and sliding. Both the A- and B-HMG in boxes are made up of highly conserved 84 amino acid residues, are strongly positively charged and are arranged in three α-helices having a similar L-shaped fold. The long arm of the “L” contains the N-terminal extended strand and helix Ill (Andersson et al. 2002; Agresti et al., 2003; Thomas, J. O. 2001), while the short arm comprises helices I and II. Structure-function analysis reveals that the pro-inflammatory cytokine domain of HMGB1 is the B-Box and in particular the sequence of its first 20 amino acids. The A-Box is an extremely weak agonist of the inflammatory cytokine release triggered by HMGB1 and competitively inhibits the pro-inflammatory activities of the B-Box and of the whole protein. Therefore, from a pharmacological point of view, the A-Box acts as an antagonist of the pathological conditions induced and/or sustained by the B-Box and HMGB1.
The third domain, the carboxyl terminus or acidic tail, is extremely negatively charged since it contains 30 repetitive aspartic and glutamic acid residues, and is linked to the boxes by a basic region of about 20 residues. Mouse and rat HMGB1 differ from the human form by only two substitutions that are located in this continuous C-terminal stretch.
Besides its nuclear location and role as a transcription factor regulator, HMGB1 has also been found in the extracellular medium, actively released by activated cells of the immune systems (monocytes and macrophages) or passively released by damaged or necrotic cells (Andersson et al., 2002; Scaffidi et al., 2002; Bonaldi et a., 2002; Taniguchi et al., 2003; Friedman et al., 2003; Palumbo et al., 2004).
Extracellularly released HMGB1 acts as a potent cytokine and as an extremely potent macrophage-stimulating factor. HMGB1 acts directly by binding to the cell membrane, inducing signaling and chemotaxis, having a chemokine-like function (Yang et al., 2001) and further acting indirectly by up-regulating the expression and secretion of pro-inflammatory cytokines. This makes extracellular HMGB1 protein a potent chemotactic and immunoregulatory protein which promotes an effective inflammatory immune response. Furthermore, other proteins belonging to the family of HMG proteins, and which are able to bend DNA, are released together with HMGB1 in the extracellular medium. These proteins are inter alia HMGB2, HMGB3, HMG-1L10, HMG-4L and SP100-HMG. They share with HMGB1 highly homologous amino acid sequences. Like HMGB1, they trigger/sustain inflammatory pathologies interacting with the same receptors, leading to the same downstream pathways of interaction.
In healthy cells, HMGB1 migrates to the cytoplasm both by passive and active transport. However, all cultured cells and resting monocytes contain the vast majority of HMGB1 in the nucleus, indicating that in baseline conditions import is much more effective than export. Cells might transport HMGB1 from the nucleus by acetylating lysine residues which are abundant in HMGB1, thereby neutralizing their basic charge and rendering them unable to function as nuclear localization signals. Nuclear HMGB1 hyperacetylation determines the relocation of this protein from the nucleus to the cytoplasm (in the fibroblasts, for example) or its accumulation into secretory endolysosomes (in activated monocytes and macrophages, for example) and subsequent redirection towards release through a non-classical vesicle-mediated secretory pathway. HMGB1 secretion by already activated monocytes is then triggered by bioactive lysophosphatidylcholine (LPC), which is generated later in the inflammation site from phosphatidylcholine through the action of the secretory phospholipase sPLA2 produced by monocytes several hours after activation. Therefore, secretion of HMGB1 seems to be induced by two signals (Bonaldi et al., 2003) and to take place in three steps: 1) at first, an inflammatory signal promotes HMGB1 acetylation and its relocation from the nucleus to the cytoplasm (step 1) and storage in cytoplasmic secretory vesicles (step 2); then, a secretion signal (extracellular ATP or iysophcsphatidylchoiine) promotes exocytosis (third step) (Andersson et al., 2002; Scaffidi et al. 2002; Gardena et al., 2002; Bonaldi et al., 2003; Friedman et al., 2003).
Released HMGB1 has been identified as one of the ligands binding to the RAGE receptor. This receptor is expressed in most cell types, and at a high level mainly in endothelial cells, in vascular smooth muscle cells, in monocytes and macrophages and in mononuclear phagocytes. Recognition involves the C-terminal of HMGB1. The interaction of HMGB1 and RAGE triggers a sustained period of cellular activation mediated by RAGE up-regulation and receptor-dependent signaling. In particular, the interaction of HMGB1 and RAGE activates several intracellular signal transduction pathways, including mitogen-activated protein kinases (MAPKs), Cdc-42, p21ras, Rac and the nuclear translocation factor KB (NF-κB), the transcription factor classically linked to inflammatory processes (Schmidt et al., 2001),
According to several experimental evidences, released HMGB1 may also interact with receptors belonging to one or more subclass(es) of the family of the Toll-like receptors. Further, HMGB1 may also interact with the functional N-terminal lectin-like domain (D1) of thrombomodulin. Due to the ability of the functional D1 domain of thrombomodulin to intercept and bind circulating HMGB1, the interaction with the RAGE receptors and the Toll-like receptors is prevented.
When released in vivo, HMGB1 is an extremely potent cytokine and a potent macrophage-stimulating factor. In fact, like other cytokine mediators of end toxemia, HMGB1 activates in vitro a cascade of multiple pro-inflammatory cytokines (TNF, IL-1α, IL-1β, IL-1Ra, IL-6, IL-8, MIP-1α and MIP-1β from human macrophages. Therefore, HMGB1 acts as a late mediator during acute inflammation and participates in an important way in the pathogenesis of systemic inflammation after the early mediator response has been resolved.
Moreover, the observed RAGE upregulation in proinflammatory environments and the proved increased expression of this receptor in a variety of acute and chronic inflammatory diseases provide support for RAGE as an attractive target for future medical interventions related to inflammation.
The observed pro-inflammatory effects of HMGB1 in vitro and the correlation between circulating HMGB1 levels and the development of the pathogenic sequence of systemic inflammation in vivo indicate that therapeutically is targeting of this cytokine-like molecule should be of relevant clinical value, suggesting novel therapeutic approaches by a “late” administration of (selective) antagonists/inhibitors of the extracellular activities of HMGB1.
Therefore, several attempts were performed in order to block this extracellular HMGB1 chemo-cytokine protein. Several important approaches were addressed to the administration of antibodies against HMGB1, of antibodies against HMGB1 fragments (for example HMGB1 Boxes) of antibodies to RAGE, of soluble RAGE (sRAGE), of ethyl pyruvate (Czura et al., 2003; Lotze et al., 2003) and N-terminal lectin-like domain (D1) of thrombomodulin.
HMGB1 A-Box, one of the two DNA-binding domains in HMGB1, has been identified as a specific antagonist of HMGB1 highly purified recombinant A-Box has protected mice from lethal experimental sepsis even when initial treatment has been delayed for 24 hours after pathology induction, further suggesting that HMGB1 antagonists may be administered successfully in a clinically relevant window wider than the one used for other known cytokines (Yang et al., 2004).
Structural function analysis of HMGB1-truncated mutants has revealed that the A-Box domain of HMGB1 competitively displaces the saturable binding of HMGB1 to macrophages, specifically antagonizing HMGB1 activities. As has been already seen with the protective activity of anti-HMGB1 antibodies, the administration of the A-Box rescues mice from sepsis even when treatment has been initiated as late as 24 hours after surgical induction of sepsis (Yang H. et al., 2004). HMGB1 antagonists or inhibitors selected from the group of antibodies or antibody fragments that bind to an HMGB1 is protein, HMGB1 gene antisense sequences and HMGB1 receptor antagonists are known from U.S. Pat. No. 6,468,533, WO 02/074337 and US 2003/0144201.
A promising attempt for inhibiting and/or antagonising the extracellular HMGB1 chemo-cytokine protein is therefore based on the experimental evidence that the two high affinity binding domains for DNA, i.e. HMGB1 Box-A and HMGB1 Box-B, which are present in the HMGB1 molecule, have two opposing roles in the protein released in the extracellular space. The main activity of HMGB1 Box-B is to mediate the pro-inflammatory activities attributed to the HMGB1 protein. On the other hand, HMGB1 Box-A acts as an antagonist competing with the pro-inflammatory activity of the Box-B domain.
The patent application WO 2006/024547 discloses polypeptide variants of the HMGB1 Box-A, or of biologically active fragments of HMGB1 Box-A, which are obtained through systematic mutations of single amino acids of the wild-type HMGB1 Box-A protein. Therefore, WO 2006/024547 provides new agents as selective inhibitors and/or antagonists of extracellular HMGB1 and their use to prevent, alleviate and/or treat the broad spectrum of pathological conditions associated and induced by the extracellular HMGB1 chemokine and/or by the cascade of multiple inflammatory cytokines caused by the extracellular release of the HMGB1 chemokine proteins.
The efficacy of administration of synthetic protein drugs may be hampered in vivo by factors such as solubility at physiological pH, rapid elimination by glomerulal filtration, cellular clearance and metabolism as well as readily absorption. The efficacy of oral administration is, for example, hampered since proteins are digested if taken orally. The efficacy of systemic administration on the other hand is hampered since proteins under 65-70 kDa are cleared rapidly from the body. In many cases, such disadvantagous effects lead to reduced patient compliance and to reduced drug efficacy preventing an effective therapeutic use of such protein agents.
A successful strategy for improving the efficacy and the duration of the protein agent effects and for reducing potential toxicological effects is the covalent binding of a biologically active protein agent to diverse polymers. One of the polymers that is most often used in the art for improving the pharmacologic and toxicologic properties of an active agent is polyethyleneglycol, PEG in short. Polyethyleneglycol (PEG) polymers are amphiphilic, non-toxic and immunologically inert and can be conjugated to pharmaceuticals to manipulate many of the pharmacokinetic and toxicologic properties.
In the art, many covalent modification of therapeutic useful proteins with polyethylenegfycol (PEG) are reported. Covalent attachment of PEG to a protein (“pegylation”) is useful in order to extend the circulation half life of proteins, since it increases the proteins effective size and reduces it rate of clearance from the body. Moreover, PEG modification of a protein increases the protein solubility, stability and decreases the protein immunogenicity.
The problem underlying the present invention was therefore the provision of novel therapeutically useful protein agents, which act as selective antagonist and/or inhibitors of extracellular HMGB1. The scope of the present invention was therefore to exploit the peculiar characteristics of some polymers, in particular of PEG, in order to develop new administration forms of the HMGB-1 high affinity binding domain Box-A (HMGB1 Box-A), which show the same if not an even higher pharmacological activity and moreover, an improved pharmacokinetic and toxicologic performance in comparison to the non-conjugation HMGB1 Box-A polypeptide and which permit to achieve the best availability of 1-HMGB1 Box-A or of a biologically active fragment thereof in various possible administration routes.
The present invention is therefore directed to a novel polymer conjugate of the human and/or non-human wild type HMGB1 high affinity binding domain Box-A (HMGB1 Box-A) or of a biological active fragment of HMGB1 Box-A. The amino acid sequence of human HMGB1 Box-A is shown in SEQ ID NO:1. A preferred non-human HMGB1 Box-A is the Anapheles gambia HMGB1 Box-A, the sequence of which is shown in SEQ ID NO:301.
A further aspect of the present invention is directed to a polymer conjugate of a polypeptide variant of the human and/or non-human HMGB1 high affinity binding domain Box-A or of a biologically active fragment of HMGB1 Box-A, whereby the amino acid sequence of said polypeptide variant differs from the amino acid sequence of the wild type HMGB1 Box-A by the mutation of one or more single amino acids.
In the context of the present invention, “HMGB1” includes the non-acetylated form or/and the acetylated form of HMGB1. Likewise, “HMGB1 homologous proteins” include the non-acetylated form or/and the acetylated form of HMGB1 homologous proteins. Preferred HMGB1 homologous proteins are HMGB2, HMGB3, HMG-1L10, HMG-4L or/and SP100-HMG.
The novel polymer conjugates of the present invention show an increased water solubility, an improved pharmaceutical manageability, an improved pharmacokinetic and bioavailability and/or a decreased toxicity and/or immunogenicity in comparison to the non-conjugated HMGB1 polypeptide or polypeptide variant. Moreover, it was surprisingly found that the conjugation of the HMGB1 Box-A polypeptide or polypeptide variant and/or fragment does not alter the biological activity of the protein.
The polymer moiety according to the present invention has to be biocompatible, can be of natural or semi-synthetic or synthetic origin and can have a linear or branched structure. Exemplary polymers include without limitation polyalkylene glycols, polyalkylene oxides, polyacrylic acid, polyacrylates, polyacrylamide or N-alkyl derivatives thereof, polymethacrylic acid, polymethacrylates, polyethylacrylic acid, polyethylacrylates, polyvinylpyrrolidone, poly(vinylalcohol), polyglycolic acid, polylactic acid, poly (lactic-co-glycolic) acid, dextran, chitosan, polyaminoacids.
In a very preferred embodiment of the present invention, the polymer is polyethylene glycol (PEG) or polyethylene glycol, wherein the terminal OH group can optionally be modified, e.g. with C₁-C₅alkyl groups or C₁-C₅acyl groups, preferably with C₁-, C₂- or C₃alkyl groups or C₁-, C₂- or C₃aryl groups. Preferably, the modified polyethylene glycol is methoxy-polyethylene-glycol (mPEG).
The polymer used according to the present invention has a molecular weight ranking from 100 to 100,000 Da, preferably from 5,000 to 50,000 Da. In a very preferred embodiment of the invention, the polymer is PEG, which preferably has a terminal OH and/or methoxy group, with a molecular weight ranking from 10,000 to 40,000 Da, and preferably from 20,000 to 40,000 Da. In the most preferred embodiment, a PEG, which preferably has a terminal OH and/or methoxy group, with an average molecular weight of 20,000 Da or of 40,000 Da is used in the present invention.
The polymer moiety of the polymer conjugate of the invention is conjugated to the HMGB1 polypeptide or polypeptide variant by a covalent chemical bond in order to provide a stable conjugate.
Preferred conjugation sites on the HMGB1 Box-A moiety are selected from a lysine, cysteine, histidine, arginine, tyrosine, serine, threonine, aspartate and glutamate residue or from the N-terminal amino group of the protein moiety.
The polymer conjugate of the present invention may be mono-, di- and multi-pegylated conjugates. Preferably, the polymer conjugates of the invention are mono-pegylated.
The polymer moiety is usually covalently linked to the HMGB1 Box-A moiety through a linker group. In particular, in the context of the present invention the term linker group means a group which is obtained by the chemical reaction between polypeptide moiety and polymer moiety. The linker group can be any residue known to those skilled in the art of polymer conjugation, obtained by the reaction of the active group on the amino acid residue of the HMGB1 Box-A moiety and the polymer or the polymer activated by a reactive group. Exemplary linker groups include without limitation alkylene, amine, amide, carbamate, carboxylate, carbonyl, ester, ether, thioether and disulfide groups. Preferably, the linker group is an amine bond, which is obtained by the reaction of the N-terminal amino acid residue with the polymer moiety activated with an aldehyde reactive group and subsequent reduction.
Moreover, the linker group may optionally contain one or more spacer groups. In the context of the present invention, a spacer group is defined as a bifunctional group, having on both termini a reactive functional end-group. With the one reactive end-group, the spacer reacts with the polymer moiety or with the reactive group on the polymer moiety. With the further functional group on the other terminus, the spacer group binds to the functional group on the amino acid residue of the HMGB1 Box-A moiety. Suitable spacer groups are known to those skilled in the art. Examples of spacer groups include, but are not limited to hetero-, bi-functional small molecules or polymer. For example, the spacer group may be represented by bifunctional C₆-C₁₂alkyl groups or heterobifunctional alkyl groups containing from 1-3 heteroatoms selected from N, S and O or an intermediary short bifunctional PEG chain.
Covalent attachment of the polymer to the HMGB1 Box-A moiety to obtain the polymer conjugate of the invention may be accomplished by known chemical synthesis techniques. For example, in one exemplary embodiment of the present invention, the polymer conjugation can be accomplished by reacting a N-hydroxy succinimide polymer (es. NHS-PEG) with the free amine groups on the amino acid residues, preferably on the lysine residues, or at the N-terminal amino acid of the HMGB1 Box-A polypeptide. Alternatively, the polymer conjugation is achieved by reaction of a PEG aldehyde to the N-terminus of the polypeptide by reductive amination. Further, the polymer conjugates can also be obtained by reacting a PEG-maleimide to a Cys residue of the HMGB1 Box-A polypeptide or polypeptide variant.
In the context of the present invention, the term “HMGB1 Box-A moiety” indicates within the polymer conjugate compound the polypeptide moiety. Hence, this term refers to the wild-type HMGB1 Box-A and to biologically active fragments thereof as well as to polypeptide variants of HMGB1 Box-A and of biologically active fragments thereof.
In the context of the present invention, where reference is made to the term “HMGB1 Box-A or amino acid sequence of HMGB1 Box-A”, it is referred to both human and non-human HMGB1 Box-A. In a preferred embodiment of the present invention, the HMGB1 Box-A moiety is derived from the wild type of human HMGB1 Box-A protein and from the wild type of Anopheles gambia HMGB1 Box-A protein.
“Biologically active fragments of HMGB1 Box-A” as used herein are meant to encompass parts of the known wild type HMGB1 Box-A protein, for which at least one of the biological activities of the corresponding mature protein is still observable when known tests are being used. Preferably, a fragment of the mature protein is considered as biologically active if an antagonist activity with respect to the pro-inflammatory activity of the HMGB1 B-Box and the HMGB1 protein as a whole can be determined. Biologically active fragments of native HMGB1 Box-A are fragments of at least 20, 25, 30, 35, 45, 50, 55, 60, 65, 70, 75 or 80 amino acids. Preferred biologically active fragments of native HMGB1 Box-A used in the context of the present invention comprises fragments of at least 77 or of at least 54 amino acids, respectively.
The term “mutation” as used in the context of the present invention can be understood as substitution, deletion and/or addition of single amino acid in the target sequence. Preferably, the mutation of the target sequence in the present invention is a substitution. The substitution can occur with different genetically encoded amino acid or by non-genetically encoded amino acids. Examples for non-genetically encoded amino acids are homocystein, hydroxyproline, ornithin, hydroxylysine, citrulline, carnitine, etc.
In a most preferred embodiment of the present invention, the polypeptide variants of HMGB1 Box-A or of a biologically active fragment thereof are obtained by using a directed evolution process, which technology is extensively described in WO 2004/7022593 and in several further patent applications (PCT/FR00/03503, PCT/FR01/01366, U.S. Ser. No. 10/022,249, U.S. Ser. No. 10/022,390, U.S. Ser. No. 10/375,192, U.S. 60/409,898, U.S. 60/457,135, U.S. 60/410,258 and U.S. 60/410,263), all in the name of Nautilus Biotech S.A. (Paris, France), which are herein incorporated by reference.
The polypeptide variants of the present invention obtained by using directed evolution technology are mutant proteins which differ from the amino acid sequence of the wild type HMGB1 Box-A by the mutation of one or more single amino acid. In a very preferred embodiment of the present invention, only one amino acid replacement occurs on the sequence of the native protein. It is, however, encompassed by the subject of the present invention that the native protein can be further optimised by replacement of a plurality, e.g. two or more, of amino acid replacements. The modified polypeptide variants can therefore differ from the wild type protein sequence by amino acid replacements on 1-10, preferably 2, 3, 4, 5 and 6 different amino acid target positions.
In particular, the very preferred polypeptide variants of HMGB1 Box-A or of a biologically active fragment thereof used as HMGB1 Box-A moiety of the polymer conjugates of the invention are those described in the application WO 2006/024547.
Accordingly, in one preferred embodiment of the invention, the HMGB1 Box-A moiety of the polymer conjugate is derived starting from human HMGB1 Box-A. In particular, one group of polypeptide variants is derived from single mutations introduced into the full-length amino acid sequence (84 amino acids) from Human HMGB1 Box-A (SEQ ID NO:1) (FIG. 1a ). These preferred polypeptide variants are defined in sequences SEQ ID Nos:2-116 (FIG. 1b ).
Other preferred polypeptide variants are obtained starting from biologically active fragments of human HMGB1 Box-A of 77 amino acids (SEQ ID NO:117) (FIG. 2a ) and 54 amino acids (SEQ ID NO:223) (FIG. 3a ), respectively. The polypeptide variants of Box-A of human HMGB1 fragment of 77 amino acids are defined in sequences SEQ ID NOs:118 to 222 (FIG. 2b ). The polypeptide variants of Box-A of human HMGB1 fragment of 54 amino acids are defined in sequences SEQ ID NOs:224 to 300 (FIG. 3b ).
In a further preferred embodiment of the invention, the HMGB1 Box-A moiety of the polymer conjugate is derived starting from Anopheles gambia HMGB1 Box-A. In particular, one group of polypeptide variants is derived from single mutations introduced into the full-length amino acid sequence (84 amino acids) from Anopheles gambia HMGB1 Box-A (SEQ ID NO:301) (FIG. 4a ). These polypeptide variants are identified in the sequences SEQ ID Nos:302 to 418 (FIG. 4b ). Other preferred polypeptide variants are generated no starting from biologically active fragments of Anopheles gambia HMGB1 Box-A of 77 amino acids (SEQ ID NO:419) (FIG. 5a ) and 54 amino acids (SEQ ID NO:529) (FIG. 6a ), respectively. The polypeptide variants of Box-A of HMGB1 fragment of 77 amino acids are defined in sequences SEQ ID Nos:420 to 528 (FIG. 5b ). The polypeptide variants of Box-A of HMGB1 Anopheles gambia (XP_311154) fragment of 54 amino acids are defined in sequences SEQ ID Nos:530 to 610 (FIG. 6b ).
In order to identify the most preferred polypeptide variants of HMGB1 Box-A used as HMGB1 Box-A moiety of the polymer conjugates of the invention, studies, have been conducted to determine the polypeptide variants which show both a similar or even improved activity and an increased protease resistance compared to the wild-type HMGB1 Box-A protein. For this purpose, the activity of Box-A polypeptide variants of the human HMGB1 Box-A of SEQ ID NO:1 in inhibiting HMGB1-induced NIH/3T3 cell migration was determined in chemotaxis assays in comparison to the antagonistic activity of human HMGB1 Box-A wild-type itself (Example 1 and FIGS. 7.1 to 7.9). Moreover, for those polypeptide variants, which show a similar or even higher antagonistic activity than the native HMGB1 Box-A protein of SEQ ID NO:1, the in vitro resistance to protease digestion was determined by incubation of each of these polypeptide variants, with a mixture of trypsin, α-chymotrypsin, endoproteinase Asp-N and endoproteinase Glu-C (sigma). This protease resistance test is described in Example 2 and the results of protease resistance profile of said variants in comparison to native HMGB1 Box-A are shown in FIGS. 9.1 to 9.67.
From the results it can be gathered that the preferred polypeptide variants of HMGB1 Box-A useful as HMGB1 Box-A moiety of the polymer conjugate of the present invention are those variants which show a similar or higher antagonistic activity together with an increased protease resistance. In particular, the preferred polypeptide variants are the polypeptides of SEQ ID NOs: 33, 35, 37-39, 42-45, 47-49, 52, 55, 57, 59, 62, 64, 67, 69 and 104. Among these preferred polypeptide variants, the most preferred variants are those defined in SEQ ID Nos:45, 49, 52, 55, 59, 64 and 67. These very preferred polypeptide variants show a dramatically improved proteinase resistance profile compared with the wild type human HMGB1 Box-A of SEQ ID NO:1 (cf. results of Example 2).
It is noted that the amino acids which occur in the various amino acid sequences appearing herein are identified according to their known one-letter code abbreviations. It should be further noted that all amino acid residue sequences represented herein by their one-letter abbreviation code have a left-to-right orientation in the conventional direction of amino-terminus to carboxyl-terminus.
In the present invention, it was surprisingly found that the above described polymer conjugates exhibit an improved pharmacokinetic and toxicologic performance, leading to an improved bioavailability compared to the non-conjugated HMGB1 Box-A polypeptide moiety. A particular advantageous effect of the polymer conjugation of the HMGB1 Box-A polypeptides and variants thereof, and in particular of the pegylation of these polypeptides, in comparison with the non-conjugated form, is the increase of the hydrodynamic volume of the proteins. This leads to a significant and unexpected improvement of the pharmacokinetic properties of the conjugated compounds due to the avoidance of renal clearance, i.e. reduction of glomerular filtration.
Moreover, the polymer conjugates of the invention exhibit increased resistance to the proteolytic activity of proteases and/or peptidases, in particular exhibit an increased resistance to the proteolytic activity of the human proteases and/or peptidases, in particular of the human serum proteases and/or human gastro-intestinal proteases or peptidases.
In particular, the resistance to proteolysis is at least 10%, 20%, 30%, 40%, 50%, 70%, 80%, 90%, 95% or higher compared to the non-conjugated HMGB1 Box-A. Protease resistance was measured at different timepoints (between 5 minutes and 8 hours) at 25° C. after incubation of 20 μg of Box-A wild type or variants with a mixture of proteases at 1% w/w of total proteins. The mixture of the proteases was prepared freshly at each assay from stock solutions of endoproteinase Glu-C (SIGMA) 200 μg/ml; trypsin (SIGMA) 400 μg/ml and α-chymotrypsin (SIGMA) 400 μg/ml. After protease incubation the reaction was stopped adding 10 μl of anti-proteases solution (Roche) and the samples were stored at −20° C. for the biological activity assay.
As a consequence of the increased stability due to the increased resistance to proteases activity, the polymer conjugates of the present invention also exhibit a longer half-life in body fluids compared to the non-conjugated HMGB1 Box A. In particular, the half-life in serum and/or in blood is increased, whereby an increase of at least 10 minutes, 20 minutes, 30 minutes, 60 minutes or even longer, compared to the non-conjugated HMGB1 Box-A is observed.
Due to the increase of the hydrodynamic volume of the proteins and also due to an increased resistance to proteolysis and thus the higher stability, the polymer conjugates of the invention also exhibit improved therapeutic and biological properties and activity. In fact, they show a more favorable pharmacokinetic and pharmacodynamic profile than non-conjugated HMGB1 Box-A protein and protein variants.
The invention is therefore directed to the use of the above-mentioned polymer conjugates of HMGB1 Box-A as an active agent in a medicament.
A still further aspect of the invention is hence the use of the inventive polymer conjugates for the manufacture of a medicament for the prevention and/or treatment of extracellular HMGB1-associated pathologies or pathologies associated with the HMGB1 homologous proteins. In particular, the HMGB1 associated pathologies are pathologies which are mediated by a multiple inflammatory cytokine cascade.
The broad spectrum of pathological conditions induced by the HMGB1-chemokine and by the HMGB1-induced cascade of inflammatory cytokines are grouped in the following categories: inflammatory disease, autoimmune disease, systemic inflammatory response syndrome, reperfusion injury after organ transplantation, cardiovascular affections, obstetric and gynecologic disease, infectious (viral and bacterial) disease, allergic and atopic disease, solid and non-solid tumor pathologies, transplant rejection diseases, congenital diseases, dermatological diseases, neurological diseases, cachexia, renal diseases, iatrogenic intoxication conditions, metabolic and idiopathic diseases.
HMGB1-associated pathologies according to the present invention are preferably pathological conditions mediated by activation of the inflammatory cytokine cascade. Non limiting examples of conditions which can be usefully treated using the present invention include the broad spectrum of pathological conditions induced by the HMGB1-chemokine and by the HMGB1-induced cascade of inflammatory cytokines grouped in the following categories: restenosis and other cardiovascular diseases, reperfusion injury, inflammation diseases such as inflammatory bowel disease, systemic inflammation response syndrome, e.g. sepsis, adult respiratory distress syndrome, etc, autoimmune diseases such as rheumatoid arthritis and osteoarthritis, obstetric and gynaecological diseases, infectious diseases, atopic diseases, such as asthma, eczema, etc, tumor pathologies, e.g. solid or non-solid tumor diseases associated with organ or tissue transplants, such as reperfusion injuries after organ transplantation, organ rejection and graft-versus-host disease, congenital diseases, dermatological diseases such as psoriasis or alopecia, neurological diseases, ophthalmological diseases, renal, metabolic or idiopathic diseases and intoxication conditions, e.g. iatrogenic toxicity and Behçet disease, wherein the above diseases are caused by, associated with and/or accompanied by HMGB1 protein release.
In particular, the pathologies belonging to inflammatory and autoimmune diseases include rheumatoid arthritis/seronegative arthropathies, osteoarthritis, inflammatory bowel disease, Crohn's disease, intestinal infarction, systemic lupus erythematosus, iridoeyelitis/uveitis, optic neuritis, idiopathic pulmonary fibrosis, systemic vasculitis/Wegener's granulomatosis, sarcoidosis, orchitis/vasectomy reversal procedures, systemic sclerosis and scleroderma. Systemic inflammatory response includes sepsis syndrome (including gram positive sepsis, gram negative sepsis, culture negative sepsis, fungal sepsis, neutropenic fever, urosepsis, septic conjunctivitis), meningococcemia, trauma hemorrhage, hurns, ionizing radiation exposure, acute and chronic prostatitis, acute and chronic pancreatitis, appendicitis, peptic, gastric and duodenal ulcers, peritonitis, ulcerative, pseudomembranous, acute and ischemic cholitis, diverticulitis, achalasia, cholangitis, cholecystitis, enteritis, adult respiratory distress syndrome (ARDS). Reperfusion injury includes post-pump syndrome and ischemia-reperfusion injury. Cardiovascular disease includes cardiac stun syndrome, myocardial infarction and ischemic, atherosclerosis, thrombophlebitis, endocarditis, pericarditis, congestive heart failure and restenosis. Obstetric and gynecologic diseases include premature labour, endometriosis, miscarriage, vaginitis and infertility. Infectious diseases include HIV infection/HIV neuropathy, meningitis, B- and C-hepatitis, herpes simplex infection, septic arthritis, peritonitis, E. coli 0157:H7, pneumonia epiglottitis, haemolytic uremic syndrome/thrombolytic thrombocytopenic purpura, candidiasis, filariasis, amebiasis, malaria, Dengue hemorrhagic fever, leishmaniasis, leprosy, toxic shock syndrome, streptococcal myositis, gas gangrene, mycobacterium tuberculosis, mycobacterium avium intracellulare, pneumocystis carinii pneumonia, pelvic inflammatory disease, orchitis/epidydimitis, legionella, Lyme disease, influenza A, Epstein-Barr Virus, Cytomegalovirus, viral associated hemiaphagocytic syndrome, viral encephalitis/aseptic meningitis. Allergic and atopic disease include asthma, allergy, anaphylactic shock, immune complex disease, hay fever, allergic rhinitis, eczema, allergic contact dermatitis, allergic conjunctivitis, hypersensitivity pneumonitis. Malignancies (liquid and solid tumor pathologies) include ALL, AML, CML, CLL, Hodgkin's disease, non Hodgkin's lymphoma, Kaposi's sarcoma, colorectal carcinoma, nasopharyngeal carcinoma, malignant histiocytosis and paraneoplastic syndrome/hypercalcemia of malignancy. Transplant diseases include organ transplant rejection and graft-versus-host disease. Congenital disease includes cystic fibrosis, familial hematophagocytic lymphohistiocytosis and sickle cell anemia. Dermatologic disease includes psoriasis, psoriatic arthritis and alopecia. Neurologic disease includes neurodegenerative diseases (multiple sclerosis, migraine, headache, amyloid-associated pathologies, prion diseases/Creutzfeld-Jacob disease, Alzheimer and Parkinson's diseases, multiple sclerosis, amyotrophic emilateral sclerosis) and peripheral neuropathies, migraine, headache. Renal disease includes nephrotic syndrome, hemodialysis and uremia. Iatrogenic intoxication condition includes OKT3 therapy, Anti-CD3 therapy, Cytokine therapy, Chemotherapy, Radiation therapy and chronic salicylate intoxication. Metabolic and idiopathic disease includes Wilson's disease, hemochromatosis, alpha-1 antitrypsin deficiency, diabetes and diabetes complications, weight loss, anorexia, cachexia, obesity, Hashimoto's thyroiditis, osteoporosis, hypothalamic-pituitary-adrenal axis evaluation and primary biliary cirrhosis. Ophtalmological disease include glaucoma, retinopathies and dry-eye. A miscellanea of other pathologies comprehends: multiple organ dysfunction syndrome, muscular dystrophy, septic meningitis, atherosclerosis, epiglottitis, Whipple's disease, asthma, allergy, allergic rhinitis, organ necrosis, fever, septicaemia, endotoxic shock, hyperpyrexia, eosinophilic granuloma, granulomatosis, sarcoidosis, septic abortion, urethritis, emphysema, rhinitis, alveolitis, bronchiolitis, pharyngitis, epithelial barrier dysfunctions, pneumoultramicropicsilicovolcanoconiosis, pleurisy, sinusitis, influenza, respiratory syncytial virus infection, disseminated bacteremia, hydatid cyst, dermatomyositis, burns, sunburn, urticaria, warst, wheal, vasulitis, angiitis, myocarditis, arteritis, periarteritis nodosa, rheumatic fever, celiac disease, encephalitis, cerebral embolism, Guillame-Barre syndrome, neuritis, neuralgia, iatrogenic complications/peripheral nerve lesions, spinal cord injury, paralysis, uveitis, arthriditis, arthralgias, osteomyelitis, fasciitis, Paget's disease, gout, periodontal disease, synovitis, myasthenia gravis, Goodpasture's syndrome, Behçets's syndrome, ankylosing spondylitis, Barger's disease, Retier's syndrome, bullous dermatitis (bullous pemphigoid), pemphigous and pemphigous vulgaris and alopecia.
In a further preferred embodiment, the polymer compounds of the invention are used as active agents for the prevention, alleviation and/or treatment of RAGE-related pathologies. RAGE-related pathologies are defined as pathological states associated with an increased expression of RAGE.
RAGE (Receptor for Advanced Glycation End-products) is a multi-ligand member of the immunoglobulin superfamily of cell surface molecules. It is composed of three immunoglobulin-like regions (one V-type immunoglobulin domain followed by two C-type immunoglobulin domain), a transmembrane domain and a highly charged short cytosolic tail that is essential for post-RAGE signalling. RAGE was first identified in 1992 as a binding target for AGEs, non-enzymatically glycosylated and oxidated proteins which accumulate in vascular tissue in aging and at an accelerated rate in diabetes. RAGE is expressed on a wide set of cells, including endothelial cells, smooth muscle cells, mononuclear phagocytes and neurons. While it is present at high levels during development, especially in the central nervous system, its levels decline during maturity.
As reported above, RAGE was the first receptor identified for extracellular HMGB1. HMGB1 binding on the cell surface induces the transcriptional up-regulation of RAGE. Examples of RAGE-related pathologies are diabetes and disorders associated with diabetes such as diabetic vasculopathy, neuropathy, retinopathy and other disorders, including Alzheimer's disease and immune/inflammatory reactions of the vessel walls. A very preferred example of RAGE-related pathologies in this context is diabetes of type I and/or of type II.
In a further aspect of the invention, the use of the polymer conjugates HMGB1 Box-A described above is in combination with a further active agent.
The further agent is preferably an agent capable of inhibiting an early mediator of the inflammatory cytokine cascade. Preferably, this further agent is an antagonist or inhibitor of a cytokine selected from the group consisting of TNF, IL-1α, IL-1β, IL-Ra, IL-6, IL-8, IL-10, IL 13, IL-18, IFN-γ MIP-1α, MIF-1β, MP-2, MIF and PAF.
The further agent used in combination with the polymer conjugate, may also be an inhibitor of RAGE, e.g. an antibody directed to RAGE, a nucleic acid or nucleic acid analogue capable of inhibiting RAGE expression, e.g. an antisense molecule, a ribozyme or a RNA interference molecule, or a small synthetic molecule antagonist of the interaction of HMGB1 with RAGE, preferably of the interaction of the non-acetylated or/and acetylated form of HMGB1 with RAGE, or soluble RAGE (sRAGE). The antibody to RAGE is preferably a monoclonal antibody, more preferably a chimeric or humanised antibody or a recombinant antibody, such as a single chain antibody or an antigen-binding fragment of such an antibody. The soluble RAGE analog may be optionally present as a fusion protein, e.g. with the Fc domain of a human antibody. The small synthetic molecular antagonist of the HMGB1 interaction with RAGE preferably has a molecular weight of less than 1000 Dalton. The small synthetic molecular antagonist preferably inhibits the interaction of RAGE with the non-acetylated form or/and with the acetylated form of HMGB1 and with the non-acetylated form or/and with the acetylated form of HMGB1 homologous proteins, particularly HMGB2, HMGB3, HMG-1L10, HMG-4L or/and SP100-HMG.
The further agent used in combination with the polymer conjugate, may also be an inhibitor of the interaction of a Toll-like receptor (TLR), e.g. of TLR2, TLR4, TLR7, TLR8 or/and TLR9, with HMGB1, which inhibitor is preferably a monoclonal or polyclonal antibody, a nucleic acid or nucleic acid analogue capable of inhibiting TLR expression, e.g. an antisense molecule, a ribozyme or a RNA interference molecule, or a synthetic molecule preferably having a size of less than 1000 Dalton. The inhibitor may be a known inhibitor of a Toll-like receptor, in particular of TLR2, TLR4, TLR7, TLR8 or/and TLR9. The inhibitor preferably inhibits the interaction of the Toll-like receptor with the non-acetylated form or/and the acetylated form of HMGB1 and with the non-acetylated form or/and with the acetylated form of HMGB1 homologous proteins, in particular HMGB2, HMGB3, HMG-1L10, HMG-4L or/and SP100-HMG.
In still another embodiment, the further agent is the functional N-terminal lectin-like domain (D1) of thrombomodulin. The D1 domain of thrombomodulin is able to intercept the non-acetylated form and/or the acetylated form of released HMGB1 and of released HMGB1 homologous proteins, in particular HMGB2, HMGB3, HMG-1L10, HMG-4L or/and SP100-HMG, preventing thus their interaction with RAGE and Toll-like receptors. The D1 domain of thrombomodulin may be native or mutated in order to make it resistant to proteases.
The further agent may also be a synthetic double-stranded nucleic acid or nucleic acid analogue molecule with a bent shape structure, particularly a double-stranded bent DNA, PNA or DNA/PNA chimera or hybrid or a double-stranded cruciform DNA, PNA or DNA/PNA chimera or hybrid structure, capable of binding to the HMGB1 protein. Preferred nucleic acids and nucleic analogue molecules are disclosed in a co-owned and co-pending international patent application No. PCT/EP2005/007198 filed on 4 Jul. 2005 (claiming the priority of U.S. provisional application No. 60/584,678 filed on 2 Jul. 2004), which are incorporated herein by reference. The synthetic double-stranded nucleic acid or nucleic acid analogue molecule with a bent shape structure is preferably capable of binding to the non-acetylated or/and to the acetylated form of HMGB1 and the non-acetylated or/and the acetylated form of HMGB1 homologous proteins, in particular HMGB2, HMGB3, HMG-1L10, HMG4L or/and SP100-HMG.
In a still further embodiment, the further agent used in combination with the polymer conjugate is K-252a or/and a salt or derivative thereof or a polymer conjugate of K-252a or/and of a derivative thereof. The use of K-252a or polymer conjugates of K-252a and derivatives thereof is disclosed in a co-owned and co-pending international patent application No. PCT/EP2005/008258 and filed on 25 Aug. 2005, which is herein incorporated by reference.
Therefore, a further aspect of the present invention is a pharmaceutical composition comprising an effective amount of at least one of the polymer conjugates of HMGB1 Box-A polypeptide or polypeptide variant or a biologically active fragment thereof as an active ingredient for the treatment of HMGB1-associated pathologies and pharmaceutically acceptable carriers, diluents and/or adjuvants. The pharmaceutical composition of the present invention is preferably suitable for the treatment of pathologies associated with the non-acetylated or/and the acetylated form of HMGB1 and/or of HMGB1 homologous proteins. In a further preferred embodiment, the pharmaceutical composition of the present invention comprising the at least one polymer conjugate also comprises a further agent as defined above. The pharmaceutical composition of the present invention may be used for diagnostic or for therapeutic applications.
The exact formulation, route of administration and dosage can be chosen by the individual physician in view of the patient's conditions. Administration may be achieved in a single dose or repeated doses at intervals. Dosage amount and interval may be adjusted individually in order to provide the therapeutical effect which results in amelioration of symptoms or a prolongation of the survival in a patient. The actual amount of composition administered will, of course, be dependent on the subject being treated, on the subject's weight, the severity of the affliction, the manner of administration and the judgement of the prescribing physician. A suitable daily dosage will be between 0.001 to 10 mg/kg, particularly 0.1 to 5 mg/kg.
The administration may be carried out by known methods, e.g. by injection, in particular by intravenous, intramuscular, transmucosal, subcutaneous or intraperitoneal injection and/or by oral, topical, nasal, inhalation, aerosol and/or rectal application, etc. The administration may be local or systemic.
In addition, the polymer conjugates of the Box-A of HMGB1 moiety object of this invention can be reversibly immobilized and/or adsorbed on the surface and/or inside medical devices or drug release/vehicling systems (microspheres). Medical devices and microspheres can be reversibly loaded with the polymer conjugates of this invention, through their binding, impregnation and/or adsorption on the surface of the medical device or of the microsphere or on a layer that coats its surface. When the medical device or the microsphere come into contact with biological fluids, the reversibly immobilized polymer conjugate is released. Therefore, the medical device and the microsphere act as drug-releasing tools that elute the molecule object of this invention in such a way that their release kinetics can be controlled, ensuring controlled or sustained release, as required by the treatment. The methods for coating/impregnating the medical devices and loading microspheres are well known by experts in these technologies.
Thus, a further aspect of this invention is the use of the polymer conjugates of Box-A of HMGB1, wherein conjugated molecules are reversibly immobilized on the surface of medical devices or of microspheres or are adsorbed within them. These medical instruments are preferably surgical tools, implants, catheters or stents, for example stents for angioplasty and, in particular, medicated drug-eluting stents.
Another aspect of the invention concerns a medical device reversibly coated with at least one polymer conjugate of the invention. Such a device can he selected from surgical instruments, implants, catheters or stents. Such a device may be useful for angioplasty.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is further illustrated by the following figures:

FIG. 1a displays the amino acid sequence of the native Human HMGB1 Box-A made of 84 amino acid residues (SEQ ID NO:1).

FIG. 1b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the full-length human HMGB1 Box-A. Further, the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs:2 to 116.

FIG. 2a displays the amino acid sequence of the biologically active fragment of Human HMGB1 Box-A made of 77 amino acid residues (SEQ ID NO:117).

FIG. 2b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the biologically active fragment of Human HMGB1 Box-A made of 77 amino acid residues. Further the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs: 118 to 222.

FIG. 3a displays the amino acid sequence of the biologically active fragment of Human HMGB1 Box-A made of 54 amino acid residues (SEQ ID NO:223).

FIG. 3b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the biologically active fragment of Human HMGB1 Box-A made of 54 amino acid residues. Further, the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs: 224 to 300.

FIG. 4a displays the amino acid sequence of the native Anopheles gambia HMGB1 Box-A made of 84 amino acid residues (SEQ ID NO:301).

FIG. 4b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the full-length Anopheles gambia HMGB1 Box-A. Further, the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs: 302 to 418.

FIG. 5a displays the amino acid sequence of the biologically active fragment of Anopheles gambia HMGB1 Box-A made of 77 amino acid residues (SEQ ID NO:419).

FIG. 5b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the biologically active fragment of Anopheles gambia HMGB1 Box-A made of 77 amino acid residues. Further the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs: 420 to 528.

FIG. 6a displays the amino acid sequence of the biologically active fragment of Anopheles gambia HMGB1 Box-A made of 54 amino acid residues (SEQ ID NO:529).

FIG. 6b shows the type of replacing amino acids on the respective target positions selected to generate the polypeptide variant of the biologically active fragment of Anopheles gambia HMGB1 Box-A made of 54 amino acid residues. Further, the specific amino acid sequences of the generated polypeptide variant are displayed in SEQ ID NOs: 530 to 610.

FIGS. 7A to 7I and Tables 7.1 to 7.9 show the results of the chemotaxis assay described of Example 1 performed on the HMGB1 Box-A polypeptide variants of SEQ ID NO:2 to SEQ ID NO:116 used as HMGB1 Box-A moiety of the polymer conjugates of the present invention. In each figure the activity of a set of polypeptide variants in the inhibition of HMGB1 induced NIH/3T3 cell migration is tested compared to the activity of human wild-type HMGB1 Box-A full-length fragment of SEQ ID NO:1. Each figure shows a table reporting the statistical analysis numerical data and a column bar graph showing the chemotaxis assay results.

FIG. 7A and Table 7.1 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGH1 Box-A wild type of SEQ ID NO:1 (CT500) and polypeptide variants of SEQ ID NO:2 to 15 (identified in the Table and Figure with the code CT501, CT566, CT569, CT570, CT571, CT502, CT572, CT503, CT573, CT504, CT575, CT576, CT576 and CT505, respectively).

FIG. 7B and Table 7.2 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT5OO) and polypeptide variants of SEQ ID NOs:16-23 and 25-29 (identified in the Table and Figure with the code CT577, CT578, CT506, CT579, CT580, CT581, CT507, CT582, CT584, CT508, CT509, CT510 and CT585, respectively).

FIG. 7C and Table 7.3 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT500) and polypeptide variants of SEQ ID Nos:30-35 and 37-43 (identified in the Table and Figure with the code CT511, CT512, CT513, CT514, CT586, CT515, CT516, CT517, CT518, CT519, CT520, CT521 and CT522,respectively).

FIG. 7D and Table 7.4 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NOs:44-57 (identified in the Table and Figure with the code CT523, CT524, CT525, CT526, CT527, CT528, CT588, CT529, CT530, CT589, CT590, CT531, CT591 and CT532, respectively).

FIG. 7E and Table 7.5 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT500) and polypeptide variants of SEQ ID Nos:58-67 and 69-71 (identified in the Table and Figure with the code CT592, CT533, CT593, CT534, CT535, CT536, CT537, CT594, CT538, CT539, CT540, CT541 and CT542; respectively).

FIG. 7F and Table 7.6 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT5OO) and polypeptide variants of SEQ ID Nos:72-85 (identified in the Table and Figure with the code CT596, CT597, CT598, CT599, CT600, CT601, CT602, CT603, CT543, CT544, CT545, CT546, CT547 and CT604, respectively).

FIG. 7G and Table 7.7 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT5OO) and polypeptide variants of SEQ ID NOs:86-99 (identified in the Table and Figure with the code CT548, CT549, CT605, CT606, CT607, CT608, CT609, CT610, CT550, CT551, CT611, CT552, CT553 and CT554, respectively).

FIG. 7H and Table 7.8 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT5OO) and polypeptide variants of SEQ ID Nos:100-113 (identified in the Table and Figure with the code CT555, CT556, CT557, CT558, CT559, CT612, CT560, CT561, CT613, CT562, CT563, CT564, CT565 and CT566, respectively).

FIG. 7I and Table 7.9 show the bar graph and statistical data of chemotaxis migration assay results in the inhibition of HMGB1-induced NIH/3T3 cells by human HMGB1 Box-A wild type of SEQ ID NO:1 (CT500) and polypeptide variants of SEQ ID Nos:114-116 (identified in the Table and Figure with the code CT567, CT614 and CT615, respectively).

FIG. 8 shows the image of the Tricine SDS-PAGE gel loaded with human HMGB1 Box-A Wild type of SEQ ID NO:1 (CT500) at different timepoints after protease digestion of the protease resistance testing described in Example 2. The Box-A wild type protein tested for protease resistance is a His-tagged protein. After 5 minutes of digestion CT500 shows two major bands, one corresponding to the original protein in the sample and the second corresponding to the 84-aminoacid protein without the N-term His-tag (indicated on the figure with an arrow). The profile of this second band shows resistance to proteases for 30 minutes. Minor bands present on this and other gels of FIG. 9A to FIG. 9BO correspond to Box-A digested fragments.

FIG. 9A to FIG. 9BO show the image of the Tricine SDS-PAGE gel loaded with the polypeptide variants of the HMGB1 Box-A moiety of the polymer conjugates of the present invention at different timepoints after protease digestion of the protease resistance testing described in Example 2. Box-A polypeptide variants tested for protease resistance are Hig-tagged proteins. After 5 minutes of digestion the SDS-PAGE gel image of the polypeptide variants show two major bands, one corresponding to the original protein variant in the sample and the second corresponding to the Box-A 84 amino acid protein variant without the N-term His-tag.

FIG. 9A shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:2 (CT501) at different timepoints after protease digestion.

FIG. 9B shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:7 (CT502) at different timepoints after protease digestion.

FIG. 9C shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:9 (CT503) at different timepoints after protease digestion.

FIG. 9D shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:11 (CT504) at different timepoints after protease digestion.

FIG. 9E shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:15 (CT505) at different timepoints after protease digestion.

FIG. 9F shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:18 (CT506) at different timepoints after protease digestion.

FIG. 9G Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:22 (CT507) at different timepoints after protease digestion.

FIG. 9H shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:26 (CT508) at different timepoints after protease digestion,

FIG. 9I shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:27 (CT509) at different timepoints after protease digestion.

FIG. 9J shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:28 (CT510) at different timepoints after protease digestion.

FIG. 9K shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:30 (CT511) at different timepoints after protease digestion.

FIG. 9L shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:31 (CT512) at different timepoints after protease digestion.

FIG. 9M shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:32 (CT513) at different timepoints after protease digestion.

FIG. 9N shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:33 (CT514) at different timepoints after protease digestion.

FIG. 9O shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:35 (CT515) at different timepoints after protease digestion.

FIG. 9P shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:37 (CT516) at different timepoints after protease digestion.

FIG. 9Q shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:38 (CT517) at different timepoints after protease digestion.

FIG. 9R shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:39 (CT518) at different timepoints after protease digestion.

FIG. 9S shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:40 (CT519) at different timepoints after protease digestion.

FIG. 9T shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:41 (CT520) at different timepoints after protease digestion.

FIG. 9U shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:42 (CT521) at different timepoints after protease digestion.

FIG. 9V shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:43 (522) at different timepoints after protease digestion.

FIG. 9W shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:44 (CT523) at different timepoints after protease digestion.

FIG. 9X shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:45 (CT524) at different timepoints after protease digestion.

FIG. 9Y shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:46 (CT525) at different timepoints after protease digestion.

FIG. 9Z shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:47 (CT526) at different timepoints after protease digestion.

FIG. 9AA shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:48 (CT527) at different timepoints after protease digestion.

FIG. 9AB shows the Tricine SDS-PAGE of the polypeptide variant of SEQ NO:49 (CT528) at different timepoints after protease digestion.

FIG. 9AC shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:51 (CT529) at different timepoints after protease digestion.

FIG. 9AD shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:52 (CT530) at different timepoints after protease digestion.

FIG. 9AE shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:55 (CT531) at different timepoints after protease digestion.

FIG. 9AF shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:57 (CT532) at different timepoints after protease digestion.

FIG. 9AG shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:59 (CT533) at different timepoints after protease digestion.

FIG. 9AH shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:61 (CT534) at different timepoints after protease digestion.

FIG. 9AI shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:62 (CT535) at different timepoints after protease digestion.

FIG. 9AJ shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:63 (CT536} at different timepoints after protease digestion.

FIG. 9AK shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:64 (CT537) at different timepoints after protease digestion.

FIG. 9AL shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:66 (CT538) at different timepoints after protease digestion.

FIG. 9AM shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:67 (CT539) at different timepoints after protease digestion.

FIG. 9AN shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:69 (CT540) at different timepoints after protease digestion.

FIG. 9AO shows the Tricine SDS-PAGE of the polypeptide variant of ID NO:70 (CT541) at different timepoints after protease digestion.

FIG. 9AP shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:71 (CT542) at different timepoints after protease digestion.

FIG. 9AQ shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:80 (CT543) at different timepoints after protease digestion.

FIG. 9AR shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:81 (CT544) at different timepoints after protease digestion.

FIG. 9AS shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:82 (CT545) at different timepoints after protease digestion.

FIG. 9AT shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:83 (CT546) at different timepoints after protease digestion.

FIG. 9AU shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:84 (CT547) at different timepoints after protease digestion.

FIG. 9AV shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:86 (CT548) at different timepoints after protease digestion.

FIG. 9AW shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:87 (CT549) at different timepoints after protease digestion.

FIG. 9AX shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:94 (CT550) at different timepoints after protease digestion.

FIG. 9AY shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:95 (CT551) at different timepoints after protease digestion.

FIG. 9AZ shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:97 (CT552) at different timepoints after protease digestion.

FIG. 9BA shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:98 (CT553) at different timepoints after protease digestion.

FIG. 9BB shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:99 (CT554) at different timepoints after protease digestion.

FIG. 9BC shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO: 100 (CT555) at different timepoints after protease digestion.

FIG. 9BD shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:101 (CT556) at different timepoints after protease digestion.

FIG. 9BE shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:102 (CT557) at different timepoints after protease digestion.

FIG. 9BF shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:103 (CT556) at different timepoints after protease digestion.

FIG. 9BG shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:104 (CT559) at different timepoints after protease digestion.

FIG. 9BH shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:106 (CT560) at different timepoints after protease digestion.

FIG. 9BI shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:107 (CT561) at different timepoints after protease digestion.

FIG. 9BJ shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:109 (CT562) at different timepoints after protease digestion.

FIG. 9BK shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:110 (CT563) at different timepoints after protease digestion.

FIG. 9BL shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:111 (CT564) at different timepoints after protease digestion.

FIG. 9BM shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:112 (CT565) at different timepoints after protease digestion.

FIG. 9BN shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:113 (CT566) at different timepoints after protease digestion.

FIG. 9BO shows the Tricine SDS-PAGE of the polypeptide variant of SEQ ID NO:19 (CT567) at different timepoints after protease digestion.

FIG. 10 shows a table in which the results of the Tricine SDS-PAGE are summarized. A cross indicates the presence on the gel of the band corresponding to the 84 amino acid long protein fragment of the HMGB1 Box-A wild-type or of the HMGB1 Box-A polypeptide variant.

FIG. 11 shows the mean plasma concentration/time after a single subcutaneous administration of Box A wild type (WT) and Box A variant number 64 (64) at a dosage of 1 mg/kg. Data representation: Mean±SEM.

FIG. 12 shows the mean plasma concentration/time after a single subcutaneous administration of Box A wild type (WT) and Box A variant number 64 (64) at a dosage of: 1 mg/kg and of PEGylated Box A wild type (PEG-WT) and PEGylated Box A variant number 64 (PEG-64) at a dosage of 5 mg/kg. Data representation: Mean±SEM.

EXAMPLES

1. In Vitro Activity Testing: NIH/373 Cell Migration Assay
The purpose of the present study was to evaluate the activity of each of the HMGB1 Box-A polypeptide variants as defined in SEQ ID NOs:2-116 and to compare their activity to that of human wild type HMGB1 Box-A full-length fragment of SEQ ID NO:1 in order to select all the variants with similar or better activity than wild type.
HMGB1 Box-A activity is evaluated in vitro as inhibition of HMGB1-induced NIH/3T3 cells migration.
1.1 Materials
HMGB1 Box-A wild type and variants (Nautilus Biotech)
NIH/3T3 cells (ATCC n. CRL-1658)
D-MEM medium (GIBCO; cat. n. 31966-021)
Foetal Bovine Serum (GIBCO; cat. n. 10270-106)
Penicillin-Streptomycin 10,000 U/ml (GIBCO; cat. n. 15140-122)
L-Glutamine 200 mM (GIBCO; cat. n. 25030-024)
TrypLE Select (GIBCO; cat. n. 12563-011)
Phosphate Buffered Saline (0.138 M NaCl, 0.0027 M KCl, 0.01 M phosphate, pH 7.4)
PVP free filters (8 μm pore size; 13 mm total diameter) (Neuro Probe; cat. n. PFA8)
Human fibronectin (Roche; cat. n. 1080938)
Blind Well Chemotaxis Chambers (Neuro Probe; cat. n. BW1125)
GIEMSA Stain Modified (Sigma; cat. n. GS1L)
1.2 Filters Preparation
Polycarbonate membranes PVP free filters (8 μm pore size, 13 mm total diameter) are prepared about one hour before performing the experiment by coating them with 30 μl/filter of a solution 50 μg/ml of fibronectin dispensed on the opaque side of the filter. The stock fibronectin solution is prepared by diluting the lyophilized fibronectin in ddH₂O to a final concentration of 1 mg/ml and by keeping the solution about 1 hour at 37° C. for complete dissolution. This stock solution can be stored at −20° C.
The filters are then left to dry under the laminar flux of the hood (about one hour).
1.3 Cells Preparation
NIH/3T3 cells are seeded the day before the experiment (approximately 22-24 hours before performing the experiment) 10⁶cells/plate.
When the filters are ready to use, the cells are detached with Trypsin, counted and resuspended 10⁶cells/ml in serum free culture medium.
1.4 Chemotaxis Assay
In each chemotaxis experiment 14 different polypeptide variants of the human HMGB1 Box-A full-length fragment of SEQ ID NO:1 are tested.
Growth cell medium without serum addition (w/o FBS) is used as negative control representing spontaneous migration.
1 nM HMGB1 is used as positive control. HMGB1 Box-A wild type or the tested polypeptide variants 0.5/1 nM are added to 1 nM HMGB1 to inhibit HMGB1-induced NIH/3T3 cell migration.
Negative control (w/o FBS) and positive control (1 nM HMGB1) are tested in triplicate in each experiment.
HMGB1 Box-A wild type (SEQ ID NO:1) activity in inhibiting HMGB1-induced cell migration is tested in triplicate in each experiment.
Each of the HMGB1 Box-A polypeptide variants (SEQ ID NOs:2 to 116) is tested in duplicate.
Blind Well Chemotaxis Chambers are used. The clean, dry lower well of each chamber is filled with 50 μl of DMEM without FBS added with the appropriate chemotactic agent and inhibitors. A slight positive meniscus should form when the well is filled; this helps prevent air bubbles from being trapped when the filter is applied. With small forceps the filter is placed over the filled well (fibronectin treated side up), being careful not to trap air bubbles and not to touch the filter with fingers. The filter retainer is screwed in by hand. Cell suspension (50000 cells/50 μl) is pipetted into the upper well and 150 μl of serum free medium are added to fill the upper well of the chamber. The filled chamber is incubated for 3 hours (37° C., 5% CO₂) to allow cell migration. After incubation the fluid is removed from the filter. The retainer is unscrewed and immersed in cool distilled water. The filter is lifted out with forceps, placed on a clean surface (solid paraffin) (migrated cells side up) and fixed with a needle (placed on the border area).
1.5 GIEMSA Staining of Migrated Cells
The filters are fixed with ethanol once and then washed three times under running water. A working solution of GIEMSA Stain Modified diluted 1:10 in ddH₂O is prepared just before use. After washing of the filters, the staining is added and left to incubate for 20 minutes. Washing of the staining is performed under running water. The filters are then placed on slides with the migrated cells side down, and the non-migrated cells side is gently wiped off with a wet cotton swab (wipe twice, using two swabs or both ends of a double-tipped swab) being careful not to move the filter. After cleaning, a cover slide is placed on the filter and cells are counted under a microscope at 40× in 10 random fields/filter.
1.6 Data Representation and Statistical Analysis
The results of the NIH/3T3 migration assay performed are reported in the tables and bar graphs shown in Figure and Table 7.1 to Figure and Table 7.9.
Data are represented in bar columns as MEAN±95% Cl.
One-way ANOVA followed by Dunnett's post test (control column data: 1 nM HMGB1 sample+HMGB1 Box-A WT sample) is the statistical analysis performed.
When evaluating the results data, HMGB1 Box-A variants data having a post test p value <0.05 are considered significantly different from HMGB1 Box-A wild type. If the mean of the Box-A polypeptide variant is higher than that of Box-A wild type the column is coloured in red in the graph of the experiment shown in FIGS. 7.1 to 7.9. Those red columns represent HMGB1 Box-A polypeptide variants showing less activity than wild type in inhibiting HMGB1-induced cell migration.
If the mean of the polypeptide variant results lower than that of wild type Box-A then the column is coloured in light blue in the graph of the is experiment shown in FIGS. 7.1 to 7.9. Those variants represent HMGB1 Box-A variants showing higher activity than HMGB1 Box-A wild type in inhibiting HMGB1-induced cell migration.
HMGB1 Box-A variants data having a post test p value >0.05 are considered not significantly different from HMGB1 Box-A wild type. The bar column of those variants are coloured in green. Those variants represent HMGB1 Box-A variants showing the same activity of wild type in inhibiting HMGB1-induced cell migration.
1.7 Results
The activity of polypeptide variants of the human HMGB1 high affinity binding domain Box-A of SEQ ID NOs:2 to 116 was evaluated in comparison to human HMGB1 Box-A wild-type of SEQ ID NO;1 as inhibition of HMGB1-induced cell migration, in order to determine the preferred polypeptide variants useful as HMGB1 Box-A moiety of the preferred polymer conjugate of the present invention.
The chemotaxis assays results revealed (FIGS. 7.1 to 7.9) that for 26 polypeptide variants the mutation according to the present invention could lead to a higher activity in cell migration inhibition in comparison with the activity of the wild-type human HMGB1 Box-A. In particular, a higher activity in cell migration inhibition was shown for the polypeptide variants of SEQ ID NOs: 30-32, 35, 38, 40-41, 43, 48, 51, 57, 63-64, 69, 70, 94, 95, 100, 103-104, 106-107, 109-111 and 113.
Moreover, the chemotaxis assays results revealed (FIGS. 7.9 to 7.9) that 41 polypeptide variants showed no changes in their activity in inhibiting HMGB1-induced cell migration compared to the activity of Box-A wild-type polypeptide. In particular, this is the case for the polypeptide variants of SEQ ID NOs: 2, 7, 9, 11, 15, 18, 22, 26-28, 33, 37, 39. 42, 44-47, 49, 52, 55, 59, 61, 62, 66-67, 71, 80-84, 86-87, 97-99, 101-102, 112 and 114.
All these Box-A polypeptide variants which exhibit a similar or a higher activity than the Box-A wild-type were tested for in vitro protease resistance, in order to choose the most resistant ones that are at least as active as Box-A wild-type (see protease resistance test in Example 2).
2. In Vitro Protease Resistance Testing
The purpose of the present study was to evaluate the in vitro protease resistance of to HMGB1 Box-A variants shown in Example 1 and to compare it to that of wild type HMGB1 Box-A of SEQ ID NO:1 in order to identify the variants with improved protease resistance with respect to wild type polypeptide.
2.1 Materials
HMGB1 His-tagged Box-A wild type and selected variants (Nautilus biotech)
Trypsin (Sigma; cat. n, T8658; lot.n, 045K5113)
α-chymotrypsin (Sigma; cat. n. 06423; lot.n. 109H74858),
Endoproteinase Asp-N (Sigma; cat. n. P3303; lot.n. 046K1049)
Endoproteinase Glu-C (Sigma; cat. n. P6181; lot.n. 075K5100)
Complete, Mini EDTA-free protease inhibitor cocktail (Roche; cat. n. 11836170 001)
Trizma base (Sigma; cat. n. T6066)
Acrylamide/bis solution 40% in water (Sigma; cat.n. 01709)
SDS (Sigma; cat. n. 71729)
Glycerol 99% (Sigma: cat. n. G9012)
Temed (Sigma; cat. n. 87689)
APS (Sigma; cat. n, A 3678)
Polypeptide SDS-PAGE Molecular Weight Standards (Bio-Rad; cat. n. 161-0326)
Premixed 10× Tris/tricine/SDS Buffer (Bio-Rad; cat. n, 161-0744)
β-Mercaptoethanol (Sigma; cat. n. M7154)
Methanol (VWR; cat. n. 20864.320)
Acetic acid (VWR; cat. n. 20104.323)
Brilliant Blue R (Sigma: cat. n. B0149)
Bromophenol Blue (Sigma; cat, n. B0126)
Hydrochloric acid (Merck; cat. n. 1.00319.2511)
3× sample loading buffer for Tricine gels (composition: 150 mM Tris-HCl, pH 6.8; 12% SDS; 36% glycerol; 6% β-Mercaptoethanol; 0.04% of bromophenol blue)
2.2. Protease Mixture Preparation
A mixture of proteases containing trypsine, α-chymotrypsine, endoproteinase Asp-N and endoproteinase Glu-C is used.
Table 1 reports specificity of each of the proteases used in this study.

TABLE 1

protease specificity.

Protease	Specificity

Trypsin	C-term of K, R
	(not if P at C-term of cutting site; slower digestion
	if acidic residue on either side of cutting site)
α-chymotrypsin	C-term of T, P, W, L
	(secondary hydrolysis: C-term of M, I, S, T, V, H,
	G, A)
Endoproteinase	N-term of D, C
Asp-N
Endoproteinase Glu	C-term of E, D
	(not if P is at C-term of cutting site)

Each lyophilized protease is dissolved according to manufacturer recommendations to obtain a stock solution that is aliquoted and stored at −80° C.
100 μg of trypsin are dissolved in 100 μl of dH₂O to obtain a 1 μg/μl stock solution. 25 μg of α-chymotrypsine are dissolved in 50 μl of a solution 1 mM HCl, 2 mM CaCl₂to obtain a 0.5 μg/μl stock solution. 2 μg of endoproteinase Asp-N are dissolved in 50 μl of dH₂O to obtain a 0.04 μg/μl stock solution. 25 μg of endoproteinase Glu-C are dissolved in 50 μl of dH₂O to obtain a 0.5 μg/μl stock solution.
Before performing the experiment one aliquot of each protease stock solution is left to thaw on ice.
Trypsin and endoproteinase Glu-C stock aliquots are diluted in dH₂O to obtain a final working solution of 0.1 μg/μl. α-chymotrypsine stock aliquot is diluted in a solution 1 mM HCl, 2 mM CaCl₂to obtain a final 0.1 μg/μl working solution. Endoproteinase Asp-N aliquot is used without dilution just before performing the experiment a mixture of proteases containing 1% (in weight/weight of total Box-A contained in the sample) of each protease is freshly prepared and immediately added to HMGB1 Box-A to be digested.
2.3. HMGB1 Box-A Wild Type and Variants Protease Digestion
18 μg total of each HMGB1 Box-A (wild type or variants) are digested in each experiment.
HMGB1 Box-A to be tested is left to thaw on ice and the volume corresponding to 18 μg is taken. The volume of this solution is then brought with dH₂O to a final volume of 90 μl in order to obtain the same final volume for each HMGB1 Box-A to be tested.
10 μl of this solution (corresponding to 2 μg of HMGB1 Box-A) are taken before adding the protease mixture. This sample corresponds to “time 0” not digested sample.
The remaining sample (16 μg of HMGB1 Box-A) is added with 8.8 μl (corresponding to 0.16 μg of each protease of the freshly prepared mixture; see 2.2) of protease mixture for digestion.
Protease digestion is performed at 25° C. and a volume corresponding to 2 μg of HMGB1 Box-A (originally present in the mixture) is sampled at defined time points. Digestion is stopped adding 4 μl of a solution of complete Mini EDTA-free protease inhibitor cocktail (1 tablet dissolved in 10 ml of dH₂O).
Timepoints for sampling are: 0, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1.5 hours, 2 hours and 4 hours.
Soon after protease inhibition, samples are added with the appropriate amount of sample loading buffer 3× and incubated at 95° C.° for about 3 minutes.
2.4. Tricine SDS-PAGE of Digested HMGB1 Box-A Wild Type and Variants
After protease digestion and samples preparation, timepoints samples of each HMGB1 Box-A are loaded on a Tricine SDS PAGE gel (see for references: Schägger and von Jagow, “Tricine-sodium dodecyl sulphate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa”, Anal. Biochem. 166, 368-379, 1987).
5 μl of Polypeptide SDS-PAGE Molecular Weight Standards (Bio-Rad) are loaded for reference on each gel.
Each well of the gel is loaded with 10 μl of sample (volume corresponding to 1 μg of HMGB1 Box-A before digestion).
Electrophoresis is performed at 30 V until the bromophenol blue has entered the separating portion of the gel, then at 120 V (Mini Protean 3 System; Bio-Rad) till the end of the run.
Gels are stained by soaking in a Coomassie Brilliant Blue R staining solution (0.1% w/v in 50% methanol, 10% acetic acid) for 1 hour and destained overnight in destaining solution (30% methanol, 10% acetic acid).
Gel images are acquired with Gel Doc 2000 (Bio-Rad) imaging system.
2.5 Results
In the above reported assay conditions HGMB1 wild-type protein resisted approximately 30 minutes to complete protease digestion. In FIG. 8 the band corresponding to the 84-amino acid full-length fragment of human HMGB1 Box-A wild-type of SEQ ID NO:1 protein is visible until 30 minutes of protease digestion.
21 Box-A polypeptide variants tested showed an increased resistance to protease (FIG. 10). In the reported assay conditions these variants resist from 1 hour to 2 hours to protease digestion. The polypeptide variants of SEQ ID NOs: 33, 35, 37, 38, 39, 42, 43, 44, 47, 48, 57, 62, 69 and 104 showed a resistance of 1 hour to protease digestion. FIGS. 9.14, 9.15, 9.16, 9.17, 9.18, 9,21, 9.22, 9.23, 9.26, 9.27, 9.32, 9,35, 9.40 and 9.59 show a band corresponding to the not His-tagged protein of 84 amino acids which is visible until 1 hour of protease digestion.
The polypeptide variants of SEQ ID NOs: 45, 49, 52, 55 and 67 showed a resistance of 1.5 hours to protease digestion. FIGS. 9.24, 9.28, 9.30, 9.31 and 9.39 show a band corresponding to the not His-tagged protein of 84 amino acids which is clearly visible 1 hour and a half after protease digestion. The polypeptide variants of SEQ ID NOs: 59 and 64 even show a resistance of up to 2 hours to protease digestion. FIGS. 9.33 and 9.37 show a band of the not His-tagged protein of 84 amino acids which is clearly visible until 2 hours after protease digestion.
3. Pharmacokinetic Study of Box A (Wild Type and Variants) and Pegylated Box A (Wild Type and Variants) after Single Subcutaneous Administration in Mice.
3.1. Aim of the Study
The purpose of the study was to evaluate and to compare the pharmacokinetic profile of Box A (wild type and variants) and PEGylated Box A (wild type and variants) following a single subcutaneous administration of the compounds in mice.
3.2. Materials and Methods

- Test articles: Box A wild type and variants and the corresponding PEGylated molecules. PEGylated molecules were obtained reacting linear mPEG-aldehyde (40 kDa) with the N-terminus of the respective Box A molecule by reductive amination.

3.3 In Vivo Experiment
Animals: mice (Balb/c, males, 7-9 weeks old, supplied by Charles River Laboratories Italia SpA, Calco one week before the experiment) with an average body weight of 22.2-22.4 g at the moment of the experiment.
Animal husbandry: the animals were housed in a ventilated thermostatic container set to maintain temperature and relative humidity at 22° C.±2° C. and 55±15% respectively, with 12 hours light/dark cycle. Mice were housed up to 10 to a cage, in clear polycarbonate cages (Techniplast, Buguggiate, Italy); drinking water via water bottles and a commercially available laboratory rodent diet (4RF21, Mucedola Settimo Milanese, Italy) were supplied ad libitum.
Experimental groups: 4 (four test items), 20 animals/group, randomly grouped.
Administered dose: Box A wild type and variants were administered 1 mg/kg subcutaneously. PEGylated Box A wild type and variants were administered 5 mg/kg subcutaneously. These doses ensured equimolarity of test compounds.
Administration of the test items: test items were administered subcutaneously by using an insulin syringe fitted with a 0.45×12 mm (26 G×½″) needle at a volume of 0.25 mL/mouse (10 mL/kg body weight).
Test article formulation: Phosphate Buffer Saline solution
Animal sacrifice and blood collection:
Blood samples were collected at the following time points after treatment:
Box A wild type and variants: 5, 20 and 40 minutes, 1.5 and 2.5 hours after administration of the test compounds;
PEGylated Box A wild type and variants: 5, 40 minutes, 1.5, 5 and 10 hours after administration of the test compounds.

- Different timepoints for blood sample collection between test compounds were decided on the basis of longer expected permanence of PEGylated molecules in the bloodstream.
- At each sampling time, approximately 0.4 mL blood samples were collected from the ventral aorta of each animal using an insulin syringe, under deep ether anesthesia, and transferred into polyethylene Eppendorf tubes containing 5 μL heparin (5000 UI/mL) to prevent blood clotting. Blood samples were kept in ice until centrifugation at 1400 g for 5 min. in a refrigerated centrifuge (2-4° C.). From each tube plasma samples were then recovered, put in new Eppendorf tubes and frozen at −80° C. until analysis.

3.4 Analytical Determination
Box A (wild type and variants) and PEGylated Box A (wild type and variants) plasma concentrations were determined in mouse plasma by an ELISA method.
Briefly, a coating solution was prepared by diluting a monoclonal antibody against the N-terminal of Box A to 10 ng/ml in 100 mM carbonate-bicarbonate coating buffer. 100 μL were aliquoted to every well of a Nunc Maxisorp ELISA plate, which was incubated overnight at 4° C. The plate was so washed with PBS 0.05% Tween for 6 times and 300 μl of 5% milk in PBS 1% Tween were added to each well to block the remaining binding sites on the plate. The plate was incubated for 1 hour at room temperature at 300 rpm. Samples were diluted 1:20 in PBS 1% Tween.
rpm. Samples were diluted 1:20 in PBS 1% Tween.
The plate was washed 6-fold and 100 μl of standards and diluted samples were transferred to the designated wells of the coated plate and incubated at room temperature for 1 hour at 300 rpm. The plate was washed 6-fold again prior to the addition of the secondary antibody against the C-Terminal of Box A (1:200). After 1 hour of incubation and 6 washes, the biotin-goat anti-rabbit conjugate 1:20000 solution was added to every well (100 μL/well). After 1 hour of incubation (at room temperature, 300 rpm) and six washes, the plate was incubated with 100 μl/well of the streptavidin-HRP solution 1:100000 for 25 minutes at 300 rpm. The plate was washed 6 times and 100 μl of pre-warmed TMB substrate were added to each well. The signal was developed at room temperature on the bench top and, after 30 minutes, 100 μL/well of Stop Solution were added and the plate was immediately read at 450 nm.
3.5 Results
Mean plasma concentrations of Box A (wild type and variants) and PEGylated Box A (wild type and variants) were calculated for each of the previously described PK samples and the pharmacokinetic profile determined. The results are shown in FIGS. 11 and 12.
As example, here below are reported the PK profiles of Box A wild type and variant n. 64 (M51I) and of PEGylated Box A and PEGylated Box A variant number 64 (M51I). The results are reported as mean values±error (SEM) (4 mice for each time point, analysis in duplicate).
In the following table, the calculated AUC_last(Area Under the Curve calculated at the last experimental point) for Box A wild type, Box A variant number 64, PEGylated Box A wild type and PEGylated Box A variant so number 64 curves are reported.

TABLE 2

Calculated AUC_lastdata for Box A wild type (WT), Box A variant
number 64 (64), PEGylated Box A wild type (WT) and PEGylated
Box A variant number 64 (PEG-64) curves.

	AUC_last
	(μM * min)

	WT	8.899
	64	14.97
	PEG-WT	275.5
	PEG-64	332.5

3.6 Discussion
The relative gain in AUC_lastconferred to WT by the mutation is 1.68X (WT vs. 64), most likely due to the higher protease resistance in the sub cute compartment. The relative gain in AUC_lastconferred to WT by PEGylation is 31X (WT vs. PEG-WT), mainly due to impaired renal filtration of the PEG conjugate, but also to protection from protease action. Putting together mutation and PEGylation yields a relative gain in AUC_lastof 37X (WT vs. PEG-64). Unexpectedly, the two modifications together have a positive effect on animal exposure to the protein that is superior to the sum of the contributions of the single modifications (Le., 37X>1.68X+31X). Thus, single point mutation and PEGylation have a synergistic, cooperative effect on the pharmacokinetic profile of the native protein. A possible explanation of this phenomenon could be that PEGylated proteins are not completely protected from proteolysis in the sub cute compartment. The introduction of a single point mutation gives a boost to resistance in this compartment, allowing higher quantities of protein to enter blood circulation and to be then protected from renal filtration by the bulky PEG chain.

Claims

1. A method of alleviating, and/or treating a pathology selected from the group consisting of HMGB-1 associated pathologies, RAGE-related pathologies, auto immune disease, systemic inflammatory response syndrome, cardiovascular affection, dermatological disease, and neurological disease comprising administering to a patient an effective amount of a polymer conjugate of the human and/or non human wild type HMGB-1 high affinity binding domain Box-A (HMGB1 Box-A), a polypeptide variant of the human and/or non human HMGB-1 Box-A, or of a biologically active fragment of HMGB-1 Box-A, whereby the amino acid sequence of said polypeptide variant differs from the amino acid sequence of the wild type HMGB-1 Box-A by the mutation of one or more single amino acid.

2. The method of claim 1, wherein said cardiovascular affection is ischemia.

3. The method of claim 1, wherein said dermatological disease is psoriasis.

4. The method of claim 1, wherein said HMGB-1 associated pathology is a neurological disease.

5. The method of claim 1, wherein said RAGE-related pathology is diabetes of type I and/or type II.

6. The method of claim 1, wherein said polymer is administered in combination with a further agent capable of inhibiting an early mediator of the inflammatory cytokine cascade, wherein said further agent is an antagonist or inhibitor of a cytokine, an antibody to RAGE, a nucleic acid or nucleic acid analogue capable of inhibiting RAGE expression, a small synthetic molecule antagonist of the HMGB-1 interaction with RAGE or soluble RAGE (sRAGE), the N-terminal lectin-like domain (D1) of native or mutated thrombomodulin, a synthetic double-stranded nucleic acid or nucleic acid analogue molecule with a bent, shape structure, K-252a, a salt or derivative of K-252a, a polymer conjugate of K252a, a derivative of K-252a polymer conjugate, or an inhibitor of the interaction HMGB-1 with at least one Toll-like receptor (TLR).

7. The method of claim 6, wherein said nucleic acid or nucleic acid analogue capable of inhibiting RAGE expression is selected from the group consisting of an antisense molecule, a ribozyme and a RNA interference molecule.

8. The method of claim 6, wherein said synthetic double-stranded nucleic acid or nucleic acid analogue molecule with a bent shape structure is selected from the group consisting of bent and cruciform DNA and PNA.