WO2014109658A2

WO2014109658A2 - Peptides

Info

Publication number: WO2014109658A2
Application number: PCT/PT2014/000001
Authority: WO
Inventors: Nuno Dos Santos; Ana Do Vale; Daniela SILVA; Liliana PEREIRA; Jorge AZEVEDO; Pedro Pereira; Sandra MACEDO-RIBEIRO
Original assignee: Instituto De Biologia Molecular E Celular
Priority date: 2013-01-09
Filing date: 2014-01-09
Publication date: 2014-07-17
Also published as: GB201300381D0; WO2014109658A3

Abstract

Peptides which are able to enter mammalian cells by translocating into the cell cytosol are disclosed, as well as nucleic acids encoding such peptides. Further, peptides for inhibiting NF-kB, and the nucleic acid encoding these peptides are also described. These peptides are based on the toxin AIP56 found in the Gram-negative bacterium Photobacterium damselae piscicida (Phdp).

Description

PEPTIDES

Field of the Invention

The present invention relates to peptides which are able to enter mammalian cells by translocating into the cell cytosol, and the nucleic acid encoding the peptides. Further, the invention also relates to a peptide for inhibiting NF-κΒ, and the nucleic acid encoding the peptide.

Background to the Invention

The NF-KB family of transcription factors is evolutionarily conserved and comprises five members (NF- Β 1 (p50), NF-KB 2 (p52), RelA (p65), RelB and cRel), which form different combinations of homo- or hetero-dimers [1]. Under normal physiological conditions, NF-KB complexes remain inactive in the cytosol through association with the ΙκΒ proteins that mask the nuclear localization domains on NF-κΒ subunits. A variety of stimuli, including bacterial and viral products and cytokines, acting via cellular receptors such as Toll-like receptors (TLRs), Inter leukin-1 receptor (IL-1R) and TNF receptors (TNFRs), trigger a signalling cascade that leads to phosphorylation and degradation of the inhibitory ΙκΒ proteins with rapid activation and transport of the NF-κΒ complexes to the nucleus, resulting in the up- regulation of inflammatory and anti-apoptotic genes [2].

NF-KB activation is considered to be the central initiating event of host responses to microbial pathogen invasion [2]. Therefore, it is not surprising that successful microbial pathogens have evolved complex strategies to interfere with the NF-κΒ signalling. A number of pathogenic bacteria were recently found to interfere with NF-κΒ signalling pathway by targeting different intermediates of the NF-κΒ activation cascade [2-4].

Photobacterium damselae piscicida (Phdp) is a Gram-negative bacterium that infects several warm water fish species worldwide causing very high mortality and, therefore, it is recognized as one of the most threatening pathogens in mariculture [5-8]. It is known that Phdp induces massive destruction of the phagocytic cells of the infected host by a process that occurs systemically and culminates in a secondary necrotic process with lysis of the apoptosing cells [9-11]. This leads to the impairment of host immune defences and to the release of the cytotoxic contents of the phagocytes, contributing to the formation of the necrotic lesions observed in the diseased animals. It has previously been shown that phagocyte apoptosis observed in Phdp infections results from the activity of AIP56, a plasmid-encoded protein exotoxin secreted by virulent strains, and that passive immunization with anti-AIP56 rabbit serum protects against Phdp infection [9, 12]. These results implicated AIP56 as a key virulence factor of Phdp.

AIP56 is synthesized as a precursor protein with a cleavable N-terminal signal peptide that is removed during secretion, originating a mature toxin of 497 amino-acids with the conserved ΗΕΓνΉ zinc-binding motif within its N-terminal region [12], similarly to tetanus neurotoxin [13]. The N-terminal region of AIP56 is homologous to NleC [12, 14], a type III secreted effector present in several enteric pathogenic bacteria.

Recently, it was shown that NleC inhibits NF-κΒ activation and represses NF-icB-dependent transcription by cleaving NF-κΒ p65 within its N-terminal region [15-19]. Summary of the Invention

The inventors have shown that AIP56 is a zinc-metalloprotease that cleaves the Rel Homology domain of NF-κΒ, in particular p65, and that its enzymatic and apoptogenic activities are correlated. In contrast to NleC, which is delivered into the host cell's cytosol through a type III secretion system, AIP56 is an A-B-type exotoxin with an N-terminal domain responsible for the proteolytic activity and a C-terminal domain involved in binding and internalisation into target cells. It has surprisingly been found that AIP56 is able to bind and be internalised into mammalian cells despite being an exotoxin from a bacterial pathogen that specifically infects salt water fish. This finding is unexpected. Therefore, in a first aspect, there is provided the use of a protein to translocate a molecule into a mammalian cell, wherein the protein comprises a translocation domain which is able to translocate the molecule into the cell cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1. SEQ ID NO: 1 is the amino acid sequence of the translocation domain of AIP56.

Without wishing to be held to a particular theory, it is thought that the protein binds to the cell membrane of mammalian cells and is internalized by the cell through an endosomal mechanism. Within the endososmes, and while the endosomal pH becomes acidic, the protein suffers a pH-derived conformational change that allows it to insert into the endosomal membrane and translocate into the cytosol. In particular embodiments, the translocation domain may have at least 50% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 60% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 70% sequence identity to SEQ ID NO:

1. The translocation domain may have at least 75% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 80% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 85% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 90% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 92% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 94% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 96% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 97% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 98% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99.5% sequence identity to SEQ ID NO: 1. In some embodiments, the translocation domain has the sequence of SEQ ID NO: 1.

The translocation domain may have at least 40% sequence identity to SEQ ID NO: 2. This sequence includes the linker from AIP56 in addition to the translocation domain of AIP56 (SEQ ID NO: 1). Therefore, in such embodiments, there is provided the use of a protein to translocate a molecule into a mammalian cell, wherein the protein comprises a translocation domain which is able to translocate the molecule into the cell cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 2.

In particular embodiments, the translocation domain may have at least 50% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 60% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 70% sequence identity to SEQ ID NO:

2. The translocation domain may have at least 75% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 80% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 85% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 90% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 92% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 94% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 96% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 97% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 98% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 99% sequence identity to SEQ ID NO: 2. The translocation domain may have at least 99.5% sequence identity to SEQ ID NO: 2. In some embodiments, the translocation domain has the sequence of SEQ ID NO: 2.

In some embodiments, the protein essentially consists of a translocation domain which is able to translocate the molecule into the cell cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1 or SEQ ID NO: 2. The sequence identity to SEQ IS NO: 1 or 2 may be at least 50%, at least 60%, at least 70%, etc. as mentioned above.

In view of the fact that the protein can translocate a molecule into mammalian cells, this allows the production of biological tools. For example, the molecule may comprise the protein and a functional component for carrying out a particular purpose in the mammalian cell. This means that the molecule can be administered to the external environment of the cell and the molecule will be translocated into the cell by the protein so that the functional component can act in the interior of the cell. This avoids complicated procedures for achieving internalisation of the functional component.

The functional component may be any suitable component which carries out a function in the interior of the mammalian cell to allow the cell to be manipulated for a particular purpose. For example, it may be a peptide or a peptide mimetic, a protein (e.g., an en2yme or a catalytically active enzyme fragment, an antibody or an antibody fragment, a transcription factor, a toxin or a fragment of a toxin), a nucleic acid or a nucleic acid mimetic, a lipid or a lipid mimetic, a linear or branched carbohydrate or carbohydrate mimetic, an inhibitor or an inducer, a labelling moiety, a non-enzymatic toxin, a natural or synthetic antibiotic, an intracellularly activatable prodrug, a drug (i.e. an active pharmaceutical ingredient) or an antigen.

In one embodiment, the functional component may be an antigen so that the protein can be used to deliver the antigen to macrophages and dendritic cells for vaccination. The translocation domain has been found to be particularly effective at causing cell entry into these cell types.

The functional component may be attached to the translocation protein in any suitable way such that each can carry out their function effectively. They may be attached covalently or non-covalently. They may be attached directly or non-directly, for example, via a linker. In particular embodiments, the functional component may inhibit NF-κΒ, i.e. it may be a NF- KB inhibitor. Suitable inhibitors are well known to those skilled in the art. The functional component may be a natural or engineered serine, cysteine, glutamic, aspartic, asparagine, N- terminal nucleophile, or metallo-protease with similar specificity to NleC or AIP56, or capable of inactivating NF-κΒ through alternative proteolytic cleavage. The functional component may be a protein which can inhibit NF-κΒ. The functional component may be a protein which can cleave the Rel Homology domain of NF-κΒ, for example, p65. The functional component may be a zinc-metalloprotease that cleaves the Rel Homology domain of NF-κΒ, e.g. p65. The functional component may be the catalytic domain of AIP56 (e.g. amino acids 1 to 262 of AIP56). In particular embodiments, the molecule is an AIP56 toxin. In other embodiments, the molecule is not an AIP56 toxin.

Various conditions are associated with an increase in the activity of NF-κΒ, e.g. NF-κΒ p65. These include inflammatory conditions such as infection (e.g. viral, bacterial or parasitic infection), autoimmune disease, and neoplastic disease such as tumours (e.g. cancer). Therefore, in another aspect, the invention provides a molecule for use in treating a condition associated with increased NF-κΒ activity in mammalian cells, the molecule comprising an NF-KB inhibitor and a protein which comprises a translocation domain which is able to translocate the molecule across mammalian cell membranes and which has at least 40% sequence identity to SEQ ID NO: 1.

The mammalian cell can be any type of mammalian cell such as human cells or rodent cells, e.g. HeLa cells, Caco, J774, bone marrow derived macrophages or dendritic cells. Preferably, the mammalian cell is a bone marrow derived macrophage or a dendritic cell. The translocation domain has been found to be particularly effective at causing cell entry into bone marrow derived macrophages and dendritic cells.

The mammalian cells may be isolated cells in vitro in which it is desired that the activity of NF-KB is inhibited. In this way, the molecule can be used as a biological tool. Alternatively, the mammalian cells may be part of the organism such that the treatment is treatment of a human, a rodent, etc.

The invention also provides use of a molecule in the preparation of a medicament for treating a condition associated with increased NF-κΒ activity in mammalian cells, the molecule comprising an NF-κΒ inhibitor and a protein which comprises a translocation domain which is able to translocate the molecule into mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1. Also provided is a method of introducing a molecule into a mammalian ceil, the method comprising administering the molecule to a mammalian cell, wherein the molecule comprises a protein that comprises a translocation domain which is able to translocate the molecule into the cytosol of the mammalian cell and which has at least 40% sequence identity to SEQ ID NO: 1, wherein the molecule enters the mammalian cell.

The molecule is preferably administered to the outside of the mammalian cell. Further, the presence of a functional component as part of the molecule is also envisaged, e.g. an inhibitor of NF-κΒ such as the catalytic domain of AIP56. Therefore, there is provided a method of treating a condition associated with increased NF-κΒ in mammalian cells, the method comprising administering to a subject in need of such treatment a molecule comprising an NF-κΒ inhibitor and a protein which comprises a translocation domain which is able to translocate the molecule into the cell cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1.

The various features described above with respect to the uses are equally applicable to the above methods. For example, the different embodiments relating to the sequence identity with respect to SEQ ID NOs: 1 and 2.

In a further aspect, there is provided a molecule comprising a protein which comprises a translocation domain that is able to translocate the protein into the cell cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 1, and wherein the protein is not an AIP56 toxin.

The ABP56 toxin is a 56 kDa protein produced by the Gram-negative bacterium Photobacterium damselae piscicida (Phdp) which infects several warm water fish species. For example, the AEP56 toxin is described in WO 2005/014629 and the amino acid sequence is SEQ ID NO: 4. However, the term ΆΙΡ56 toxin' includes all naturally occurring AIP56 toxins, including those which may differ slightly in sequence to the sequence shown in SEQ ID NO: 4.

The term ΆΙΡ56 toxin' includes derivatives thereof. A "derivative" of the protein refers to a variant of the 56kDa protein which has an altered primary, secondary and or tertiary amino acid sequence compared to the naturally-occurring (native) protein; it includes the native 56kDa protein which has undergone one or more chemical or physical processing steps resulting in a reduction in toxicity of the protein to fish. The derivative may lack or may include the signal sequence (amino acids 1-16). Alternatively, the derivative may be native protein or isolated or purified protein which has been subjected to heat treatment, microwaves, light, water treatment, sonication, cold treatment, freezing, freezing and thawing, Iyophilization, denaturation with urea or detergents, formaldehyde treatment, or any other treatment known to cause alterations in the 3D conformation of proteins. In one embodiment the derivative is recombinantly expressed, having an identical amino acid sequence to the native protein (plus/minus signal sequence), but as a consequence of recombinant expression within a host cell the folding, glycosylation or other post-translational processing of the protein differs from that of the protein in the native state. Any differences in conformation or chemical properties can be reflected in reduced toxicity to fish.

The molecule is preferably isolated. An "isolated" molecule is defined as being substantially free of cellular material or other contaminating proteins from the cell or tissue source from which the protein is derived, or substantially free from chemical precursors or other chemicals when chemically synthesized. The language "substantially free of cellular material" includes preparations of the protein in which the protein is separated from cellular components of the cells from which it is isolated or recombinantly produced. In one embodiment, the language "substantially free of cellular material" includes preparations of 56kDa protein having less than about 30% (by dry weight) of non-56kDa protein (also referred to herein as a "contaminating protein"), more preferably less than about 20% of contaminating protein, still more preferably less than about 10% of contaminating protein, and most preferably less than about 5% contaminating protein. When the 56kDa protein is recombinantly produced, it is also preferably substantially free of culture medium, i.e., culture medium represents less than about 20%, more preferably less than about 10%, and most preferably less than about 5% of the volume of the protein preparation.

The protein comprises a translocation domain which is able to translocate the protein into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1. In particular, the translocation domain is able to bind to the cell membrane and cause translocation of the protein into the cytosol of mammalian cells.

In particular embodiments, the translocation domain may have at least 50% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 60% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 70% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 75% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 80% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 85% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 90% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 92% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 94% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 96% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 97% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 98% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99.5% sequence identity to SEQ ID NO: 1. In some embodiments, the translocation domain has the sequence of SEQ ID NO: 1.

The translocation domain may have at least 40% sequence identity to SEQ ID NO: 2. Therefore, in such embodiments, there is provided a molecule comprising a protein which comprises a translocation domain that is able to translocate the protein into the cell cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 2, and wherein the protein is not an AIP56 toxin. As described above, the translocation domain may have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5% or 100% sequence identity to SEQ ID NO: 2. In view of the fact that the protein can translocate a molecule into mammalian cells, this allows the productions of biological tools. For example, the molecule may comprise the protein and a functional component. The functional component is for carrying out a particular purpose in the mammalian cell. The functional component may be any suitable component which carries out a function in the interior of the mammalian cell. Further details are provided above.

In particular embodiments, the functional component may be a NF-κΒ inhibitor. Further details are provided above. In a further aspect, there is provided a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises a translocation domain that is able to translocate the protein into the cell cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 1, and wherein the protein is not an AIP56 toxin. Further, the nucleic acid does not encode for an AIP56 toxin.

The nucleic acid is preferably isolated.

The encoded protein comprises a translocation domain which is able to translocate the protein into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1. In particular, the translocation domain is able to bind to the cell membrane and cause translocation of the protein into the cytosol of mammalian cells.

In particular embodiments, the translocation domain may have at least 50% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 60% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 70% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 75% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 80% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 85% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 90% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 92% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 94% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 96% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 97% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 98% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99% sequence identity to SEQ ID NO: 1. The translocation domain may have at least 99.5% sequence identity to SEQ ID NO: 1. In some embodiments, the translocation domain has the sequence of SEQ ID NO: 1. The translocation domain may have at least 40% sequence identity to SEQ ID NO: 2. Therefore, in such embodiments, there is provided a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises a translocation domain that is able to translocate the protein into the cell cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 2, and wherein the protein is not an AIP56 toxin. Further, the nucleic acid does not encode for an AIP56 toxin. As described above, the translocation domain may have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5% or 100% sequence identity to SEQ ID NO: 2.

In view of the fact that the protein can translocate a molecule into mammalian cells, this allows the production of biological tools. For example, the nucleotide sequence may encode for a protein comprising the translocation domain and a functional component. The functional component is for carrying out a particular purpose in the mammalian cell.

The functional component may be any suitable component which carries out a function in the interior of the mammalian cell. The functional component may be any suitable component which carries out a function in the interior of the mammalian cell to allow the cell to be manipulated for a particular purpose. For example, it may be a peptide or a peptide mimetic, a protein (e.g., an enzyme or a catalytically active enzyme fragment, an antibody or an antibody fragment, a transcription factor, a toxin or a fragment of a toxin), an inhibitor or an inducer, a labelling moiety, a non-enzymatic toxin, a natural or synthetic antibiotic, an intracellularly activatable prodrug, a drug (i.e. an active pharmaceutical ingredient) or an antigen. In one embodiment, the functional component may be an antigen so that the encoded protein delivers the antigen to macrophages and dendritic cells for vaccination.

In particular embodiments, the functional component may be a NF-κΒ inhibitor. Further details are provided above.

In some embodiments, the nucleotide sequence comprises a sequence having at least 50% sequence identity to SEQ ID NO: 8. SEQ ID NO: 8 is the nucleotide sequence of the translocation domain of AIP56. The sequence may have at least 60% sequence identity to SEQ ID NO: 8. The sequence may have at least 70% sequence identity to SEQ ID NO: 8. The sequence may have at least 75% sequence identity to SEQ ID NO: 8. The sequence may have at least 80% sequence identity to SEQ ID NO: 8. The sequence may have at least 85% sequence identity to SEQ ID NO: 8. The sequence may have at least 90% sequence identity to SEQ ID NO: 8. The sequence may have at least 92% sequence identity to SEQ ID NO: 8. The sequence may have at least 94% sequence identity to SEQ ID NO: 8. The sequence may have at least 96% sequence identity to SEQ ID NO: 8. The sequence may have at least 97% sequence identity to SEQ ID NO: 8. The sequence may have at least 98% sequence identity to SEQ ID NO: 8. The sequence may have at least 99% sequence identity to SEQ ID NO: 8. The sequence may have at least 99.5% sequence identity to SEQ ID NO: 8. In some embodiments, the sequence may have the sequence of SEQ ID NO: 8. The invention also provides a vector comprising the nucleic acid.

The invention also provides a cell containing the vector or nucleic acid described above.

It has been found that the translocation domain of AIP56 is composed of two subdomains. One is a receptor binding domain and the second is responsible for entry into the cell across the cell membrane. Therefore, where targeting of mammalian cells is required rather than translocation, the receptor binding subdomain can be used instead of the whole translocation domain. Therefore, in another aspect, there is provided the use of a protein to target a molecule to a mammalian cell, wherein the protein comprises a receptor binding domain which is able to bind to the cell membrane of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9. SEQ ID NO: 9 is the amino acid sequence of the receptor binding domain of AIP56.

In particular embodiments, the receptor binding domain may have at least 50% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 60% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 70% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 75% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 80% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 85% sequence identity to SEQ ED NO: 9. The receptor binding domain may have at least 90% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 92% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 94% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 96% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 97% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 98% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 99% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 99.5% sequence identity to SEQ ID NO: 9. In some embodiments, the receptor binding domain has the sequence of SEQ ID NO: 9. In view of the fact that the protein can target a molecule to mammalian cells, this allows the production of biological tools. For example, the molecule may comprise the protein and a functional component for carrying out a particular purpose at the mammalian cell.

The functional component may be any suitable component which carries out a function at the mammalian cell to allow the cell to be manipulated for a particular purpose. For example, it may be a peptide or a peptide mimetic, a protein (e.g., an enzyme or a catalytically active enzyme fragment, an antibody or an antibody fragment, a transcription factor, a toxin or a fragment of a toxin), a nucleic acid or a nucleic acid mimetic, a lipid or a lipid mimetic, a linear or branched carbohydrate or carbohydrate mimetic, an inhibitor or an inducer, a labelling moiety, a non-enzymatic toxin, a natural or synthetic antibiotic, an intracellularly activatable prodrug, a drug (i.e. an active pharmaceutical ingredient) or an antigen. In one embodiment, the functional component may be an antigen so that the protein can be used to deliver the antigen to macrophages and dendritic cells for vaccination.

The functional component may be attached to the receptor binding protein in any suitable way such that each can carry out their function effectively. They may be attached covalently or non-covalently. They may be attached directly or non-directly, for example, via a linker.

In particular embodiments, the functional component may inhibit NF-κΒ, i.e. it may be a NF- KB inhibitor. Suitable inhibitors are well known to those skilled in the art. The functional component may be a natural or engineered serine, cysteine, glutamic, aspartic, asparagine, N- terminal nucleophile, or metallo-protease with similar specificity to NleC or AIP56, or capable of inactivating NF-κΒ through alternative proteolytic cleavage. The functional component may be a protein which can inhibit NF-κΒ. The functional component may be a protein which can cleave the Rel Homology domain of NF-κΒ, for example, p65. The functional component may be a zinc-metalloprotease that cleaves the Rel Homology domain of NF-KB, e.g. p65. The functional component may be the catalytic domain of AIP56 (e.g. amino acids 1 to 262 of AIP56). In particular embodiments, the molecule is an AIP56 toxin. In other embodiments, the molecule is not an AIP56 toxin.

Various conditions are associated with an increase in the activity of NF-κΒ, e.g. NF-κΒ p65. These include inflammatory conditions such as infection (e.g. viral, bacterial or parasitic infection), autoimmune disease, and neoplastic disease such as tumours (e.g. cancer). Therefore, in another aspect, the invention provides a molecule for use in treating a condition associated with increased NF-κΒ activity in mammalian cells, the molecule comprising an NF-KB inhibitor and a protein which comprises a receptor binding domain which is able to target the molecule to mammalian cell membranes and which has at least 40% sequence identity to SEQ ID NO: 9.

The mammalian cell can be any type of mammalian cell such as human cells or rodent cells, e.g. HeLa cells, Caco, J774, bone marrow derived macrophages or dendritic cells. Preferably, the mammalian cell is a bone marrow derived macrophage or a dendritic cell.

The invention also provides use of a molecule in the preparation of a medicament for treating a condition associated with increased NF-κΒ activity in mammalian cells, the molecule comprising an NF-κΒ inhibitor and a protein which comprises a receptor binding domain which is able to target the molecule to mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9.

Also provided is a method of targeting a molecule to a mammalian cell, the method comprising administering the molecule to a mammalian cell, wherein the molecule comprises a protein that comprises a receptor binding domain which is able to target the molecule to the mammalian cell and which has at least 40% sequence identity to SEQ ID NO: 9, wherein the molecule binds to the mammalian cell.

The molecule is preferably administered to the outside of the mammalian cell.

Further, the presence of a functional component as part of the molecule is also envisaged, e.g. an inhibitor of NF-κΒ such as the catalytic domain of AIP56. Therefore, there is provided a method of treating a condition associated with increased NF-κΒ in mammalian cells, the method comprising administering to a subject in need of such treatment a molecule comprising an NF-κΒ inhibitor and a protein which comprises a receptor binding domain which is able to target the molecule to mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9.

In a further aspect, there is provided a molecule comprising a protein which comprises a receptor binding domain that is able to target the protein to mammalian cells, wherein the receptor binding domain has at least 40% sequence identity to SEQ ID NO: 9, and wherein the protein is not an AIP56 toxin.

The molecule is preferably isolated.

The protein comprises a receptor binding domain which is able to target the protein to mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9. In particular, the receptor binding domain is able to bind to the cell membrane of mammalian cells.

In particular embodiments, the receptor binding domain may have at least 50% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 60% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 70% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 75% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 80% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 85% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 90% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 92% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 94% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 96% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 97% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 98% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 99% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 99.5% sequence identity to SEQ ID NO: 9. In some embodiments, the receptor binding domain has the sequence of SEQ ID NO: 9.

In view of the fact that the protein can target a molecule to mammalian cells, this allows the productions of biological tools. For example, the molecule may comprise the protein and a functional component. The functional component is for carrying out a particular purpose in the mammalian cell. The functional component may be any suitable component which carries out a function in the interior of the mammalian cell. Further details are provided above. In particular embodiments, the functional component may be a NF-κΒ inhibitor. Further details are provided above.

In a further aspect, there is provided a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises a receptor binding domain that is able to target the protein to mammalian cells, wherein the receptor binding domain has at least 40% sequence identity to SEQ ID NO: 9, and wherein the protein is not an AIP56 toxin. Further, the nucleic acid does not encode for an AIP56 toxin.

The nucleic acid is preferably isolated.

The encoded protein comprises a receptor binding domain which is able to target the protein to mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9. In particular, the receptor binding domain is able to bind to the cell membrane and target the protein to the mammalian cells.

In particular embodiments, the receptor binding domain may have at least 50% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 60% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 70% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 75% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 80% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 85% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 90% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 92% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 94% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 96% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 97% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 98% sequence identity to SEQ ED NO: 9. The receptor binding domain may have at least 99% sequence identity to SEQ ID NO: 9. The receptor binding domain may have at least 99.5% sequence identity to SEQ ID NO: 9. In some embodiments, the receptor binding domain has the sequence of SEQ ID NO: 9. In view of the fact that the protein can target a molecule to mammalian cells, this allows the production of biological tools. For example, the nucleotide sequence may encode for a protein comprising the receptor binding domain and a functional component. The functional component is for carrying out a particular purpose in the mammalian cell.

The functional component may be any suitable component which carries out a function at the mammalian cell. The functional component may be any suitable component which carries out a function at the mammalian cell to allow the cell to be manipulated for a particular purpose. For example, it may be a peptide or a peptide mimetic, a protein (e.g., an enzyme or a catalytically active enzyme fragment, an antibody or an antibody fragment, a transcription factor, a toxin or a fragment of a toxin), an inhibitor or an inducer, a labelling moiety, a non- enzymatic toxin, a natural or synthetic antibiotic, an intracellularly activatable prodrug, a drug (i.e. an active pharmaceutical ingredient) or an antigen.

In one embodiment, the functional component may be an antigen so that the encoded protein delivers the antigen to macrophages and dendritic cells for vaccination.

In some embodiments, the nucleotide sequence comprises a sequence having at least 50% sequence identity to SEQ ID NO: 10. SEQ ID NO: 10 is the nucleotide sequence of the receptor binding domain of AIP56. The sequence may have at least 60% sequence identity to SEQ ID NO: 10. The sequence may have at least 70% sequence identity to SEQ ID NO: 10. The sequence may have at least 75% sequence identity to SEQ ID NO: 10. The sequence may have at least 80% sequence identity to SEQ ID NO: 10. The sequence may have at least 85% sequence identity to SEQ ID NO: 10. The sequence may have at least 90% sequence identity to SEQ ID NO: 10. The sequence may have at least 92% sequence identity to SEQ ID NO: 10. The sequence may have at least 94% sequence identity to SEQ ID NO: 10. The sequence may have at least 96% sequence identity to SEQ ID NO: 10. The sequence may have at least 97% sequence identity to SEQ ID NO: 10. The sequence may have at least 98% sequence identity to SEQ ID NO: 10. The sequence may have at least 99% sequence identity to SEQ ID NO: 10. The sequence may have at least 99.5% sequence identity to SEQ ID NO: 10. In some embodiments, the sequence may have the sequence of SEQ ID NO: 10. The invention also provides a vector comprising the nucleic acid.

The invention also provides a cell containing the vector or nucleic acid described above. As mentioned above, the inventors have shown that AIP56 is a zinc-metalloprotease that cleaves the Rel Homology domain of NF-κΒ, in particular p65, and that its enzymatic and apoptogenic activities are correlated. In contrast to NleC, which is delivered into the host cell's cytosol through a type III secretion system, AIP56 is an A-B-type exotoxin with an N- terminal domain responsible for the proteolytic activity and a C-terminal domain involved in binding and internalisation into target cells. Therefore, AIP56 is able to enter the cell on its own whereas NleC requires a secretion system to allow entry into the cell. Further, it has surprisingly been found that NleC is not able to enter the cell when attached to cell delivery systems such as the anthrax delivery system or the AIP56 delivery system (i.e. the translocating domain described above). In contrast, the catalytic domain of AIP56 can be delivered into the cell using the anthrax delivery system or the AIP56 delivery system.

Therefore, in a further aspect, there is provided a molecule for inhibiting the activity of NF- KB, wherein the molecule comprises a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, wherein the molecule is not AIP56. SEQ ID NO: 3 is the amino acid sequence of the catalytic domain of AIP56.

In particular embodiments, the protein may have an amino acid sequence having at least 50% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 60%) sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 70% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 75% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 80% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 85% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 90% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 92% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 94% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 96% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 97% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 98% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 99% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 99.5% sequence identity to SEQ ID NO: 3. In some embodiments, the protein may have the amino acid sequence of SEQ ID NO: 3.

In some embodiments, the molecule comprises a protein which consists essentially of an amino acid sequence having at least 45% identity to SEQ ID NO: 3. The amino acid sequence of the protein may have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5% or 100% identity to SEQ ID NO: 3. The protein and, therefore, the molecule is able to inhibit NF-κΒ. The protein can cleave the Rel Homology domain of NF-κΒ, for example, p65. The protein is a zinc-metalloprotease that cleaves the Rel Homology domain of NF-κΒ, e.g. p65.

The protein and, therefore, the molecule can inhibit NF- Β in mammalian cells.

The molecule may comprise a delivery system to translocate the molecule into the cytosol of cells, e.g. mammalian cells. Suitable delivery systems are well known to those skilled in the art, and include nanoparticles, viral particles (virus bacteriophages), liposomes, etc. The delivery system may be the translocation portion of a toxin such as an AB toxin. Generally, this will include a binding domain and a translocation domain. The delivery system may be the translocation portion of a toxin selected from Streptolysine O, Staphylococcus aureus alpha-toxin, tetanolysin, cholera toxin, pertussis toxin, Pasteurella multocida toxin, C3 exoenzymes, Clostridium difficile toxin A, Clostridium difficile toxin B, Clostridium sordellii lethal toxin, Escherichia coli CNF, Clostridium botulinum C2 toxin, Clostridium perfringens iota toxin, Clostridium botulinum toxins, BoNT/A, BoNT/B, BoNT/C, Diphtheria toxin, Pseudomonas exotoxin A and anthrax toxin. In some embodiments, the delivery system may be the translocation portion of a toxin selected from Strptolysine O, Staphylococcus aurus alpha-toxin, tetanolysin, Diphtheria toxin, Pseudomonas exotoxin A, Clostridium botulinum C2 toxin and anthrax toxin.

The delivery system may be the delivery system of anthrax toxin, i.e. the protective antigen (PA) domain of anthrax toxin, e.g. PA83.

In some embodiments, the delivery system may be the translocation domain described above. As mentioned above, various conditions are associated with an increase in the activity of NF- KB, e.g. NF-KB p65. These include inflammatory conditions such as infection (e.g. viral, bacterial or parasitic infection) and autoimmune disease, and neoplastic disease such as tumours (e.g. cancer).

Therefore, in another aspect, the mvention provides a molecule for use in treating a condition associated with increased NF-κΒ activity, the molecule comprising a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3. The molecule may be for use in treating a condition associated with increased NF-κΒ activity in mammalian cells.

In some embodiments, the molecule is not ΑΓΡ56. The invention also provides use of a molecule in the preparation of a medicament for treating a condition associated with increased NF-κΒ activity, the molecule comprising a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3.

In another aspect, there is provided a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, wherein the molecule is not AIP56.

In particular embodiments, the protein may have an amino acid sequence having at least 50% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 60% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 70% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 75% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 80% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 85% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 90% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 92% sequence identity to SEQ ED NO: 3. The protein may have an amino acid sequence having at least 94% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 96% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 97% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 98% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 99% sequence identity to SEQ ID NO: 3. The protein may have an amino acid sequence having at least 99.5% sequence identity to SEQ ID NO: 3. In some embodiments, the protein may have the amino acid sequence of SEQ ID NO: 3. In some embodiments, the nucleotide sequence encodes for a protein which consists essentially of an amino acid sequence having at least 45% identity to SEQ ID NO: 3. The amino acid sequence of the protein may have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5% or 100% identity to SEQ ID NO: 3.

The protein encoded by the nucleotide sequence is able to inhibit NF-κΒ. The protein can cleave the Rel Homology domain of NF-κΒ, for example, p65. The protein is a zinc- metalloprotease that cleaves the Rel Homology domain of NF- Β, e.g. p65. The protein can inhibit NF-κΒ in mammalian cells. Therefore, the nucleic acid can be delivered into mammalian cells using an appropriate vector where it can be expressed to produce the protein to inhibit NF-κΒ. Suitable vectors and promoters for expression of the nucleic acid are well known to those skilled in the art.

In some embodiments, the nucleic acid is isolated.

In some embodiments, the nucleotide sequence has at least 50% sequence identity to SEQ ID NO: 5. SEQ ID NO: 5 is the nucleotide sequence of the catalytic domain of AIP56. The nucleotide sequence may have at least 60% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 70% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 75% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 80% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 85% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 90% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 92% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 94% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 96% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 97% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 98% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 99% sequence identity to SEQ ID NO: 5. The nucleotide sequence may have at least 99.5% sequence identity to SEQ ID NO: 5. In some embodiments, the nucleotide sequence may have the sequence of SEQ ID NO: 5. In some embodiments, the nucleotide sequence consists essentially of a sequence having at least 45% identity to SEQ ID NO: 5. The nucleotide sequence may consist essentially of a sequence having at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 97%, at least 98%), at least 99%, at least 99.5% or 100% identity to SEQ ID NO: 5.

The invention also provides a vector comprising the nucleic acid. The invention also provides a cell containing the vector or nucleic acid described above.

Therefore, in another aspect, the invention provides a nucleic acid for use in treating a condition associated with increased NF-κΒ activity, the nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3.

The nucleic acid may be for use in treating a condition associated with increased NF-KB activity in mammalian cells. In some embodiments, the encoded protein is not AIP56.

The invention also provides use of a nucleic acid in the preparation of a medicament for treating a condition associated with increased NF-κΒ activity, the nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3.

There is also provided a method of treating a condition associated with increased NF-κΒ, the method comprising administering to a subject in need of such treatment a molecule comprising a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, or a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3.

For the above uses and method, the other features of the molecule and nucleic acid are equally applicable, for example, relating to the identity of the amino acid sequence to SEQ ID NO: 3 and the presence of a delivery system to translocate the molecule into the cytosol of cells. To determine the percent identity of two amino acid sequences or nucleotide sequences, the sequences are aligned for optimal comparison purposes (e.g. gaps can be introduced in the sequence of a first amino acid sequence for optimal alignment with a second amino acid sequence and the intervening non-homologous sequence in the gap can be disregarded for comparison purposes). There is no requirement for the two sequences to be the same length. Unless otherwise specified, the length of sequence across which the sequences are compared is the entire extent of the alignment. When a position in the first (reference) sequence is occupied by the same amino acid residue as the corresponding position in the sequence, the molecules are homologous at that position (i.e. there is identity at that position). The percent identity between two sequences is a function of the number of homologous positions shared by the sequences (i.e., % identity = (no. of identical positions/total no. of positions) x 100. Optionally, the comparison of sequences and determination of percent identity can be accomplished using a mathematical algorithm. Suitable algorithms are incorporated in to the NBLAST and XBLAST programs of Altschul et al. (1990) J. Mol. Biol. 215:430-10.

Detailed Description of the Invention

The invention will now be described in detail, by way of example only, with reference to the figures in which: Figure 1. AIP56 is a zinc-metalloprotease that cleaves NF-κΒ p65 at the Cys³⁹-Glu⁴⁰ peptide bond. (A) Disruption of the zinc-metalloprotease signature abolishes AIP56 apoptogenic activity. Sea bass peritoneal leukocytes collected from 5 animals were incubated with AIP56 or AIP56^AAIVAA, for 4 h at 22 °C. Mock-treated cells were used as controls. Images shown are representative cytospin preparations stained with Antonow's for labelling neutrophils (brown) followed by Hemacolor. Note the presence of several apoptotic cells (arrowheads) in the sample incubated with AIP56 and their absence in cells incubated with AIP56^AAIVAA and in mock-treated cells. The percentage of apoptotic phagocytes, determined by morphological analysis, is depicted in the dot plot. (B) Incubation of cell lysates with AIP56, ΑΙΡδό^1"28502628 and nicked AIP56 resulted in p65 cleavage. Lysates were incubated for 2 h at 22 °C with 1 μΜ of the indicated proteins in the presence or absence of the metalloprotease inhibitor 1,10- phenanthroline (O-phe) and p65 cleavage assessed by Western blotting. (C) ΑΓΡ56 cleaves NF-KB p65 at the Cys³⁹-Glu⁴⁰ peptide bond. Recombinant sea bass p65ReI (7.5 μΜ) was incubated in the presence or absence of 1 μΜ of AIP56 for 3 h at 22 °C and analysed by SDS- PAGE. Edman degradation of the cleaved sbp65Rel (cl-sbp65Rel) identified the sequence E ⁰GRSA⁴⁴ showing that cleavage occurred after the conserved C³⁹. D) Incubation of leukocytes with AIP56, but not with ΑΙΡδβ^^, AIP56'-^285C262S, AIP56 ^8M97C298S, nicked AIP56 (AIP56nic) or reconstituted AIP56 (AIP56rct) leads to p65 depletion. Leukocytes were incubated with 10 μg/ml of the indicated proteins for 2 h at 22 °C and p65 cleavage was assessed by Western blotting. (E) AIP56-mediated p65 cleavage is caspase-independent. Leukocytes were incubated with 2 μg/ml AIP56 in the presence or absence of the pan-caspase inhibitor Z-VAD-FMK for 2 h at 22 °C, and p65 cleavage was assessed by Western blotting. Numbers to the left of the panels refer to the position and mass of the molecular weight markers, in kDa.

Figure 2. AIP56 is composed of two domains linked by a disulphide bridge. (A) Limited proteolysis of AIP56 with chymotrypsin and proteinase K produces two major fragments. AIP56 (0.6 mg/ml) was incubated with 0.25, 1.25, 6.25 or 25 μg/ml of chymotrypsin, trypsin and proteinase for 30 min on ice and digests analysed by reducing SDS-PAGE. The proteases (marked as *) and undigested AEP56 (marked as **) were loaded as controls. (B) The two AIP56 digestion fragments are linked by a disulphide bridge. AIP56 was incubated with or without 25 μg/ml chymotrypsin (Chym) for 30 min on ice and digests analysed under reducing (+DTT) or non-reducing (-DTT) SDS-PAGE. Numbers to the left and right of the panels refer to the position and mass of the molecular weight markers, in kDa. (C) Schematic representation of AIP56. (D) Far-UV CD spectra of AIP56 (thick solid line), AIP56^1"285 (thin solid line), AIP56^286"497 (thin dashed line) and the weighted sum of AIP56¹-²⁸⁵ and AIP56^286"497 spectra (thick dashed line).

Figure 3. AIP56 N-terminal domain plays the catalytic role and the C-terminal domain is involved in binding and entry into cells. (A) AIP56^1_285C262S and AIP56^286"497C298S lack apoptogenic activity. Leukocytes collected from 5 animals were incubated with AIP56, AIP56^]-^285C262S, ΑΠ^^286"49702985, or a mixture of ΑΠ^^1"28502625 and AIP56^286"497C298S (50 _μ8/ιη1 each) for 4 h at 22 °C. The percentage of apoptotic phagocytes was determined by morphological analysis of cytospin preparations stained with Hemacolor. (B) Delivery of ΑΓΡ56 N-terminal domain into the cell's cytosol using B. anthracis LF/PA system reproduces the activity of full length AIP56. Leukocytes from 4 animals were incubated for 4 h at 22 °C with 20 nM LF¹¹-²⁶³-AIP56¹-²⁶¹ or LFⁿ-²⁶³ AIP56²⁹⁹-⁴⁹⁷ in the presence of 10 nM PA. Cells incubated with 2 μg/ml (35 nM) ΑΓΡ56 were used as positive control. Cleavage of NF-κΒ p65 was detected by Western blotting and the occurrence of apoptosis by morphological analysis of cytospin preparations stained with Hemacolor. Note the presence of several apoptotic cells in the samples incubated with PA+LF¹¹-²⁶³^^^1"261. (C) AIP56 C-terminal domain is involved in toxin binding and entry into the target cells. ΑΠ^β * ** and AIP56^286'497C298S but not AIP56^1"28502625 inhibit AIP56-associated p65 cleavage and apoptogenic activity. Leukocytes collected from 7 fish were incubated with ΑΪΡδβ^¹^, AIP56'^"285C262S or _AIP56286-497C298s _{at fma}, _{concentrations 0}f 0.35, 1.75 or 3.5 μΜ for 15 min on ice, followed by further 15 min incubation on ice with 8.75 nM (0.5 μ^ηιΐ) AIP56 in the presence of the competitors. The competitor:AIP56 molar ratios are indicated. Cells incubated with AIP56 in the absence of competitors or with 3.5 μΜ of each competitor alone were used as controls. Cells were washed, transferred to 22 °C and incubated for 4 h prior to determination of the percentage of apoptotic cells by morphological analysis of cytospin preparations stained with Hemacolor. Left panel presents the box plot of percentage of apopotic cells (the middle bar corresponds to the median and the lower and upper side of the boxes, the first and third quartiles; circles signal extreme observations). The inhibitory effect of the highest dose of each competitor upon AIP56-mediated cleavage of p65 was assessed by Western blotting.

Figure 4. The C-terminal B domain of AIP56 has two subdomains. (A) Limited proteolysis of AIP56 with chymotrypsin reveals an additional small fragment. AIP56 (1.25 mg/ml) was incubated with 47 μ τιύ of chymotrypsin for 30 min on ice and digests analysed by reducing SDS-PAGE. Undigested AIP56 was loaded as control. N-terminal Edman sequencing of the smaller fragment identified the peptide G³⁷⁴YGHD. Numbers to the left refer to the position and mass of the molecular weight markers, in kDa. (B) Schematic representation of AIP56 showing the subdomain structure of the C-terminal B domain and in silico secondary structure prediction (under ΑΓΡ56 representation): Red, alpha-helixes; Green, beta-sheets; unstructured. (C) The C-terminus subdomain (ΑΓΡ56^374"497), within the C-terminal B domain, is involved in toxin binding. Sea bass peritoneal leukocytes were incubated with AIP56^1"

285C262_{S> An}>₅₆286-497C298S _{Qr W56}^ ^_{&t concentrations of} 3 5 _{μΜ for} , 5 _mj_{n Qn} followed by further 15 min incubation on ice with 8.75 nM (0.5 μg ml) AEP56 in the presence of the competitors. Cells incubated with AIP56 in the absence of competitors or with 3.5 μΜ of each competitor alone were used as controls. Cells were washed, transferred to 22 °C and incubated for 4 h prior to determination of the percentage of apoptotic cells by morphological analysis of cytospin preparations stained with Hemacolor. Figure 5. AIP56 toxicity requires integrity of the linker but the disulfide bridge is dispensable for intoxication. (A) Nicked and reconstituted AIP56 are not apoptogenic. Leukocytes collected from 3 animals were incubated with nicked or reconstituted ΑΓΡ56 (AIP56nic and AIP56rct, respectively) for 4 h at 22 °C and the percentage of apoptotic cells determined by morphological analysis of cytospin preparations stained with Hemacolor. Cells treated with ΑΓΡ56 and mock-treated cells were used as positive and negative controls, respectively. (B) Nicked and alkylated AIP56 (AIP56alk) display proteolytic activity in vitro in the same dose range as AIP56. S-labeled sbp65Rel was incubated for 2 h at 22 °C with wild type, nicked or alkylated AIP56 and cleavage assessed by autoradiography. (C) Nicked AIP56 competes with intact AIP56 and inhibits its toxicity. Leukocytes collected from 3 animals were incubated with 3.5 μΜ nicked AIP56 (AIP56nic) for 15 minutes on ice followed by further 15 min incubation on ice with 8.75 nM of AIP56. Mock-treated cells, cells incubated with 8.75 nM AIP56, or cells incubated with 3.5 μΜ of ΑΤΡδβ^^, AIP56^1"285 or AIP56^286"497 before incubation with 8.75 nM of AIP56 were used as controls. Cells were washed, transferred to 22 °C and incubated for 4 h. The percentage of apoptotic cells was determined by morphological analysis of cytospin preparations stained with Hemacolor and the p65 cleavage was assessed by Western blotting. (D) Disruption of the disulphide bridge linking Cys²⁶² and Cys²⁹⁸ partially compromises AIP56 toxicity. Leukocytes collected from 5 animals were incubated with AIP56 or AIP56alk for 4 h at 22 °C and the percentage of apoptotic cells determined by morphological analysis of cytospin preparations stained with Hemacolor. Left panel presents the box plot of percentage of apopotic cells (the middle bar corresponds to the median and the lower and upper side of the boxes, the first and third quartiles; circles and diamonds signal extreme observations). When used at the same concentration, AIP56alk resulted in lower percentage of apoptotic cells than AIP56, except for the dose of 0.5 μg/ml, where no statistical differences were observed. Figure 6. Schematic diagram of the primary structure of AIP56 and AIP56-related proteins. Grey: signal peptides (experimentally determined for AIP56 [12] and predicted for the remaining proteins using SignalP at http://www.cbs.dtu.dk/services/SignalP/ [20,21]; Yellow: regions with high identity to NleC and AIP56 N-terminal catalytic domain; Green; regions with high identity with AIP56 linker polypeptide; Orange: regions with high identity to APSE-2 and AIP56 C-terminal domain; Red: zinc-metalloprotease signature HEXXH (SEQ ID NO: 6); White: regions with low identity to AIP56 domains, NleC or APSE-2. Conserved zinc-metalloprotease signature HEXXH (SEQ ID NO: 6), cysteine residues, and other signalled amino acids are represented at their relative positions. AIP56-related proteins were retrieved by Blast analysis of the AIP56 protein sequence against the non-redundant protein sequences database (updated from [14]).

Figure 7. Disruption of the zinc-metalloprotease signature does not induce major structural changes in AIP56. Native-PAGE (A) and size exclusion chromatography (B) of AIP56 and AIP56^AAIVAA showing that disruption of the zinc-binding motif did not affect the monodispersity/stokes radius of the protein. In Native-PAGE, BSA electrophoretic mobility is shown for reference purposes. (C) Far-UV CD spectra of wild-type AIP56 (thick line) and AIP56^AAIVAA (thin line) showing that the secondary structure content of the toxin was also unaffected by the introduced mutations.

Figure 8. The AIP56 concentrations in the plasma of infected fish are in the same range as the ones used in the present work. (A) The presence of AIP56 in plasmas (5 μΐ aliquots) from sea bass infected with a lethal dose of Phdp strain PP3 was determined by Western blotting. Different concentrations of recombinant AIP56 (5 μΐ) were loaded as standards. Numbers at the left refer to the position and mass (in kDa) of the molecular weight markers. (B) Concentrations of ΑΓΡ56 in the plasmas analysed in (A), determined by densitometry, using a recombinant AIP56 standard curve.

Figure 9. Mutations in Cys and Glu of sbp65Rel inhibit proteolytic processing by AIP56 and NleC. ³⁵S-labeled sbp65Rel, sbp65RelC39A, sbp65RelE40A or sbp65CE39-40AA were incubated for 2 h at 22 °C with 100 nM of the indicated proteins and cleavage assessed by autoradiography.

Figure 10. Structural and functional analysis of recombinant AIP56 N- and C-terminal domains. (A) Reducing and non-reducing SDS-PAGE of purified AIP56^1"285 and AIP56^286"497. Numbers at the left refer to the position and mass (in kDa) of the molecular weight markers. (B) Far-UV CD spectra of ΑΓΡ56^1"285 (thick solid line) and ArP56'^'285C262s (thin solid line).

(C) Far-UV CD spectra of AIP56^286"497 (thick solid line) and AIP56^286'497C298S (thin solid line).

(D) AIP56, AIP56^1"28502625 and NleC cleave sea bass p65 Rel homology domain in vitro. ³⁵S- labeled sbp65Rel (Met'-Arg¹⁸⁸) was incubated for 2 h at 22 °C with 100 nM of the indicated proteins in the presence or absence of the metalloprotease inhibitor 1,10-phenanthroline (O- phe) and cleavage assessed by autoradiography.

Figure 1 1. Analysis of nicked, reconstituted and alkylated ΑΓΡ56. (A) Reducing (+ DTT) and non-reducing (- DTT) SDS-PAGE of AIP56, nicked ΑΓΡ56 (ATP56nic) and reconstituted AEP56 (AIP56rct). (B) Nicking of ΑΓΡ56 does not affect its secondary structure. Far-UV CD spectra of AIP56 (thick line) and nicked AIP56 (AIP56nic; thin line). (C) The culture conditions do not reduce the disulphide bridge of nicked ΑΓΡ56. Nicked AIP56 was added to a sea bass peritoneal cell suspension in supplemented L-15 medium and incubated up to 4 h at 22°C. An aliquot of nicked toxin not incubated with cells (0) and aliquots of the cell culture supernatant collected 2 or 4 h after incubation with cells (all containing 50 ng of nicked toxin) were run in reducing and non-reducing SDS-PAGE and subjected to Western blotting using an anti-AIP56 rabbit serum. (D) Reducing (+ DTT) and non-reducing (- DTT) SDS-PAGE of AIP56 and alkylated AIP56 (AIP56alk). Numbers on the left of the panels indicate the mass of the molecular weight markers, in kDa.

Figure 12. AIP56 enters and cleaves NF-kB p65 in several mammalian cell lines. J774A.1, Raw 264.7 and HeLa cells were left untreated or incubated with 10 g ml^"1 of AIP56 for the indicated times, washed, resuspended in SDS-PAGE sample buffer, and analysed by western blotting for the occurrence of NF-kB p65 cleavage. SDS-PAGE was performed using the Laemmli discontinuous buffer system [22]. Prior to loading, the samples were boiled for 5 min in SDS-PAGE sample buffer (50 mM Tris-HCl (pH 8.8), 2 % (w/v) SDS, 0.05 % (w/v) bromofenol blue, 10 % (v/v) glycerol, 2 mM EDTA and 100 mM DTT). For western blotting analysis, the proteins were transferred onto nitrocellulose membranes. The membranes were blocked for 1 h at room temperature with 5% skimmed milk in tris-buffered saline (TBS) containing 0.1 % (v/v) Tween 20 (T-TBS) followed by incubation for 1 h at RT with the anti- NF-KB p65 C-terminal domain (c-20) rabbit polyclonal antibody (sc-372, Santa Cruz Biotechnology). Detection of the immunoreactive bands was done using a sheep anti-rabbit HRP-conjugated secondary antibody (AP311, The Binding Site) and the ECL West Dura Chemiluminescence substrate (Pierce biotechnology). The positions of the full length and cleaved NF-kB p65 (p65 and cl-p65, respectively) are indicated. Blots shown are representative of at least three independent experiments.

Figure 13. Incubation of mouse bone marrow derived macrophages with AIP56 results in NF- kB p65 depletion. Mouse bone marrow derived macrophages (mBMDM) prepared as described [23] were left untreated or incubated with the indicated doses of AIP56 for the indicated times, washed, resuspended in SDS-PAGE sample buffer, and analyzed by western blotting for the occurrence of NF-kB p65 cleavage. SDS-PAGE was performed using the Laemmli discontinuous buffer system [22]. Prior to loading, the samples were boiled for 5 min in SDS-PAGE sample buffer (50 mM Tris-HCl (pH 8.8), 2 % (w/v) SDS, 0.05 % (w/v) bromofenol blue, 10 % (v/v) glycerol, 2 mM EDTA and 100 mM DTT). For western blotting analysis, the proteins were transferred onto nitrocellulose membranes. The membranes were blocked for 1 h at room temperature with 5% skimmed milk in tris-buffered saline (TBS) containing 0.1 % (v/v) Tween 20 (T-TBS) followed by incubation for 1 h at RT with the anti- NF-KB p65 C-terminal domain (c-20) rabbit polyclonal antibody (sc-372, Santa Cruz Biotechnology). Detection of the immunoreactive bands was done using the goat anti-rabbit alkaline phosphatase conjugated secondary antibody (A9919, Sigma Aldrich) and NBT/BCIP (nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate). The positions of the full length and cleaved NF-kB p65 (p65 and cl-p65, respectively) are indicated. Blot shown is representative of at least three independent experiments.

Figure 14. Intoxication of mBMDM with AIP56 leads to NF-kB p65 depletion and apoptosis. (A) Incubation of mBMDM with AJP56 results in apoptotic morphological alterations and cell loss. Phase contrast microscopy of mBMDM incubated with the indicated doses of ΑΓΡ56 or the catalytically inactive mutant AIP56^AAIVAA for the indicated times. (B) Incubation of mBMDM with AIP56 results in the appearance of cells with apoptotic nuclei. mBMDM were left untreated or incubated with the indicated doses of AEP56 or AIP56^AAIVAA for the indicated times and processed for the detection of DNA fragmentation by TUNEL (green). Nuclei were counterstained with PI (red). Note the presence of TUNEL-positive and condensed nuclei in monolayers treated with AIP56. (C) Incubation of mBMDM with AIP56 results in activation of effector caspase-3. mBMDM were incubated with 5 μg ml^"1 AIP56 or ΑίΡδβ^^ for the indicated times and caspase-3 activity determined by fluorimetry using the substrate Z- DEVD-AMC. Results (in RFU/ng protein) were converted to fold increase by comparing to the corresponding values of the caspase activity in mock-treated cells. The figure presents the box plot of the fold increase in caspase-3 activity along time (the middle bar corresponds to the median and the lower and upper side of the boxes, the first and third quartiles; circle and asterisk signal extreme observations). The fold increase in caspase-3 activity following treatment with ΑΙΡδό^^ (light grey bars) is not statistically different from 1 whereas it is significantly different from one at all times following treatment with AIP56 (dark grey bars). One sample t-test for the hypothesis that the fold increase is equal to 1. (D) Incubation of mBMDM with AIP56 results in NF-κΒ p65 depletion. mBMDM were left untreated or incubated with the indicated doses of AIP56 for 30 min on ice, washed to remove unbound toxin and chased at 37°C for the indicated times. Cleavage of NF-κΒ p65 was analyzed by western blotting (chromogenic detection).

Figure 15. Drugs that interfere with endosome acidification block AIP56-mediated NF-kB p65 cleavage in bone marrow derived macrophages. Mouse bone marrow derived macrophages (mBMDM) prepared as described [23] were left untreated or incubated with 5 μg ml^"1 AIP56 in the presence or absence of 10 nM of the potent inhibitors of the endosomal vacuolar ATPase pump concanamycin A or bafilomycin Al, that block the acidification of early and late endosomes as well as Iysosomes [24]. After 90 min incubation, cells were washed, resuspended in SDS-PAGE sample buffer, and analyzed by western blotting for the occurrence of NF-kB p65 cleavage. SDS-PAGE was performed using the Laemmli discontinuous buffer system [22]. Prior to loading, the samples were boiled for 5 min in SDS- PAGE sample buffer (50 mM Tris-HCl (pH 8.8), 2 % (w/v) SDS, 0.05 % (w/v) bromofenol blue, 10 % (v/v) glycerol, 2 mM EDTA and 100 mM DTT). For western blotting analysis, the proteins were transferred onto nitrocellulose membranes. To control the protein loading, the membrane was stained with Ponceau S. Afterwards, the membranes were blocked for 1 h at room temperature with 5% skimmed milk in tris-buffered saline (TBS) containing 0.1 % (v/v) Tween 20 (T-TBS) followed by incubation for 1 h at room temperature with the anti-NF-κΒ p65 C-terminal domain (c-20) rabbit polyclonal antibody (sc-372, Santa Cruz Biotechnology). Detection of the immunoreactive bands was done using the goat anti-rabbit alkaline phosphatase conjugated secondary antibody (A9919, Sigma Aldrich) and NBT/BCIP (nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate). The positions of the full length and cleaved NF-kB p65 (p65 and cl-p65, respectively) are indicated. The results show that both drugs tested block AIP56-mediated cleavage of p65 indicating that the acidification of endosomes is required for the toxin to reach the cytosolic compartment. Blot shown is representative of four independent experiments.

Figure 16. Acidic conditions induce reversible conformational changes in AIP56. A. TNS analysis of pH-induced hydrophobic transitions in AIP56. TNS (2-(p-ToluidyniI) naphthalene- 6-sulfonic acid) is a commercial probe that is essential non-fluorescent in water and becomes fluorescent when bound to hydrophobic regions of a protein. To investigate the pH-induced changes in ΑΓΡ56 hydrophobicity, the toxin (1.5 μΜ) was incubated with 150 μΜ TNS (Life Technologies) for 15 min at room temperature, at the indicated pH values, and the TNS fluorescence was analyzed using an excitation of 366 nm and an emission scan of 380 to 500 nm. The following buffers were used: for pH 4.0, 4.5, 5.0, and 5.5, 150 mM NaCl, 100 mM ammonium acetate; for pH 6.0 and 6.5, 150 mM NaCl, 100 mM morpholineethanesulfonic acid (MOPS); for pH 7.0 and 7.5, 150 mM NaCl, 100 mM HEPES [25]. The depicted spectrum corresponds to the average of four independent experiments. B. TNS fluorescence analysis following a pH shift. ΑΓΡ56 (1.5 μΜ) was incubated with 150 μΜ TNS for 15 min at room temperature at pH 4.0, and the TNS fluorescence was analyzed as described in A.. The pH was then adjusted to 7.5 by gradual addition of NaOH, and the TNS emission spectrum was recorded again. This spectrum corresponds to the average of three independent experiments. C. Intrinsic tryptophan fluorescence of AIP56. Tryptophan is an amino-acid which fluorescence can be quenched in presence of an aqueous solvent and therefore, changes in the tryptophan fluorescence can be used to detect protein conformational changes. ΑΓΡ56 (1.5 μΜ) was incubated at pH 7.5 or pH 4.0 and tryptophan fluorescence was analyzed using an excitation of 270 nm and an emission scan of 300 nm to 400 nm. This spectrum corresponds to the average of three independent measurements. D. Susceptibility of ΑΓΡ56 to V8 protease digestion. To track the pH-induced conformational changes in AIP56, we subjected the toxin to protease digestion by the Staphylococcus wz-ews-derived V8 protease at pH 4.0 or 7.8. The V8 protease was chosen because it has been reported that it maintains the same activity and specificity (peptide bound hydrolysis on the carboxylic site of glutamines) at those pH values [26,27]. Ten micrograms of AIP56 were incubated with the indicated amounts of V8 protease at pH 4.0 or pH 7.8 for 1 h at 37 °C and analyzed by SDS-PAGE using the Laemmli discontinuous buffer system [22]. The results show that AIP56 is more resistant to proteolysis at pH 4.0 than at pH 7.8, confirming that the toxin changes its conformation when exposed to an acidic environment. Numbers on the left indicate the molecular weight markers, in kDa.

Figure 17. Under acidic conditions, AIP56 interacts with artificial lipid bilayer membranes. A. Current recordings of diphytanoyl phosphatidylcholine/n-decan membranes in the presence of AIP56. The applied membrane potential was 50 mV. Initial experimental conditions consist on 150 mM KCl, 2 mM CaCl₂, 1 mM DTT, 10 mM HEPES pH 7.4. Addition of 10 μg of AIP56 mixed 1:1 with cholesterol suspension in water to the cis compartment of the chamber had no effect on membrane conductivity. Acidification of the aqueous at the cis-compartment by addition of 10 mM CH₃C0₂K pH 4.5 triggered membrane activity. A final pH of 4.6 was achieved inside the chamber. B. Under acidic pH, AIP56 membrane activity can be triggered by a 150 mV voltage-pulse. Current recordings of diphytanoyl phosphatidylcholine/n-decan membranes in the presence of 10 μg of AIP56 mixed 1 : 1 with cholesterol suspension in water. Measurements were performed in 150 mM KCl, 2 mM CaCl_2j 10 mM MES pH 6.0 and under these conditions, membrane activity was triggered by a 150 mV pulse. A very stable signal was observed when voltage was lowered after the onset to 50 mV.

Figure 18. An acidic pulse can drive translocation of AIP56 across the cell membrane. Because drugs that inhibit endosome acidification inhibited AIP56 toxicity, we tested if a low- pH pulse could drive the translocation of membrane-bound ΑΓΡ56 into the cytosol, similarly to what has been reported for several other bacterial toxins [28-30] that translocate from endosomes through a pH-dependent mechanism. Concanamycin A was used to block "normal" toxin uptake. Mouse bone-marrow derived macrophages (mBMDM) prepared as described [23] were incubated on ice with 5 g ml^"1 of AIP56 in the presence or absence of 10 nM concanamycin A (Cone A). Cells incubated in the absence of the toxin were used as controls. After 30 min, supernatant was removed and cells were incubated at the different pH for 1 h at 37 °C in a humidified chamber in the presence of 7 % C0₂. Different pH values were obtained by adding sufficient H₃P0₄ to a buffer containing 0.5 mM MgCl₂, 0.9 mM CaCl₂, 2.7 mM KCl, 1.5 mM KH₂P0₄, 3.2 mM Na₂HP0₄, and 137 mM NaCl. After the incubation, supematants were removed, fresh medium was added and cells were incubated for further 2 h in the presence of Cone A. Afterwards, cells were washed, resuspended in SDS- PAGE sample buffer, and analyzed by western blotting for the occurrence of NF-kB p65 cleavage. SDS-PAGE was performed using the Laemmli discontinuous buffer system [22]. Prior to loading, the samples were boiled for 5 min in SDS-PAGE sample buffer (50 mM Tris-HCI (pH 8.8), 2 % (w/v) SDS, 0.05 % (w/v) bromofenol blue, 10 % (v/v) glycerol, 2 mM EDTA and 100 mM DTT). For western blotting analysis, the proteins were transferred onto nitrocellulose membranes. To control the protein loading, the membrane was stained with Ponceau S. Afterwards, the membranes were blocked for 1 h at room temperature with 5% skimmed milk in tris-buffered saline (TBS) containing 0.1 % (v/v) Tween 20 (T-TBS) followed by incubation for 1 h at RT with the anti-NF-κΒ p65 C-terminal domain (c-20) rabbit polyclonal antibody (sc-372, Santa Cruz Biotechnology). Detection of the immunoreactive bands was done using the goat anti-rabbit alkaline phosphatase conjugated secondary antibody (A9919, Sigma Aldrich) and NBT/BCIP (nitroblue tetrazolium/5-bromo- 4-chloro-3-indolyl-phosphate). The positions of the full length and cleaved NF-kB p65 (p65 and cl-p65, respectively) are indicated. The results show that when 10 nM Cone A were added to cells at pH 7.0 the AIP56-dependent cleavage of p65 was blocked. However, the inhibitory effect of Cone A was less evident when the cells were exposed to pH 5.5 and was abolished when the cells with surface-bound AIP56 were exposed to pH 5.0, suggesting that an acidic pulse is sufficient to induce AIP56 translocation across the plasma membrane. Blot shown is representative of three independent experiments.

SUMMARY

The apoptosis inducing protein of 56 kDa (AIP56) is a key virulence factor secreted by Photobacterium damselae piscicida (Phdp), a Gram-negative bacterium that causes septicaemic infections in economically important marine fish species. It is known that AEP56 induces massive destruction of the phagocytic cells of the infected host, allowing the extracellular multiplication of the bacteria and contributing to the genesis of the pathology. Here, it is shown that AIP56 acts by cleaving NF-κΒ p65. The NF-κΒ family of transcription factors is evolutionarily conserved and plays a central role in the host responses to microbial pathogen invasion, regulating the expression of inflammatory and anti-apoptotic genes. Pathogenic bacteria have evolved complex strategies to interfere with the NF-κΒ signalling, usually by injecting protein effectors directly into the cell's cytosol through bacterial secretion machineries that require contact with host cells. In contrast, AIP56 acts at distance and has an intrinsic ability to reach the cytosol due to the presence of a C-tenninal domain that functions as a "delivery module".

RESULTS

The metalloprotease signature of AIP56 is essential for its apoptogenic activity

In order to clarify the role played by the zinc metalloprotease activity of AIP56, a mutant (AIP56^AA1VAA) containing a disrupted putative zinc-binding motif was produced. The oligomerization state and secondary structure content of the toxin were undisturbed by the introduced mutations (Figure 6) and atomic absorption spectroscopy did not detect zinc in AJP56^AMV , while in AIP56 equimolar amounts of zinc (0.93±0.04 mol zinc/mol protein) were present. When tested ex vivo, AIP56^AAIVAA failed to induce apoptosis of sea bass phagocytes, whilst a large number of cells with apoptotic morphology were observed after treatment with AIP56 (Figure 1A). These results indicate that an intact metalloprotease domain is essential for the apoptogenic activity of AIP56.

It is worth noting that the AIP56 concentrations used in the present work are biologically relevant, since they are in the same range of the ones detected in the plasma of infected fish (Figure 8). AIP56 is a zinc dependent metalloprotease that cleaves NF-κΒ p65 at the Cys³⁹-GIu⁴⁰ peptide bond

When incubated with sea bass cell lysates, AIP56 cleaved p65 with the appearance of a lower

MW fragment (Figure IB). Proteolysis of p65 did not occur in cell lysates incubated with _{AI 56}AAivAA _{Qr with AIp56 in the}

presence of the metalloprotease inhibitor 1,10- phenanthroline (Figure IB). The p65 fragment was recognised by an antibody specific for a peptide located at the C-terminal region of p65 indicating that the AIP56-mediated p65 cleavage occurred within the N-terminal region, where the Rel-homology domain is located. To map the cleavage site, recombinant sea bass p65Rel domain (sbp65Rel) was incubated with the toxin. SDS-PAGE analysis showed that AIP56 cleaved recombinant sbp65Rel in vitro (Figure 1C), and N-terminal sequencing of the cleaved fragment revealed that the cleavage occurred at the Cys³⁹-Glu⁴⁰ peptide bond, similarly to what has been described for NleC [15]. Experiments using in vitro synthesised ³⁵S-labeled sea bass p65Rel domain (sbp65Rel) and three sbp65Rel mutants (sbp65RelC39A, sbp65RelE40A and sbp65CE39- 40AA) showed that mutation of the evolutionarily conserved Cys³⁹ had no effect on the ability of either AIP56 or NleC to cleave p65 (Figure 9). However, mutation in Glu⁴⁰ inhibited cleavage and double mutation of Cys³⁹ and Glu⁴⁰ completely abolished p65 proteolysis by AIP56 and NleC (Figure 9).

To determine if cellular intoxication by AIP56 involves cleavage of NF-κΒ p65, sea bass peritoneal leukocytes were incubated with the wild type toxin or with the AIP56^AAIVAA mutant and p65 proteolysis assessed by Western blotting. Wild type ΑΓΡ56 caused NF-κΒ p65 depletion, whilst Al?56^AMVAA was inactive (Figure ID). It has been reported that caspase-3 can cleave p65 [31,32]. To investigate whether caspases are involved in the AD?56-dependent cleavage of p65, cells were incubated with the toxin in the presence or absence of the pan- caspase inhibitor ZVAD-FMK (Figure IE), previously shown to block AIP56-induced apoptosis [33]. In these experiments, ZVAD-FMK was effective in protecting cells from AIP56-induced apoptosis (data not shown), but did not affect NF-κΒ p65 cleavage (Figure IE), indicating that AIP56-mediated p65 depletion is a caspase-independent event. Taken together, the above results demonstrate that the metalloprotease activity of AIP56 is responsible for the cleavage of NF-κΒ p65 at the Cys³⁹-Glu⁴⁰ peptide bond.

AIP56 has two domains

The primary structure of AIP56 suggests that this toxin comprises two functional domains and could be an A-B toxin with the two domains linked by a single disulphide bond (Figure 6) [14]. Therefore, in order to define domain boundaries within the toxin, limited proteolysis experiments were performed. SDS-PAGE analysis of ΑΓΡ56 digested with chymotrypsin, trypsin or proteinase K revealed that the toxin is highly resistant to trypsin digestion, whereas chymotrypsin and proteinase K cleaved AIP56 into two major fragments with approximately 32 and 24 kDa (Figure 2A). These two domains were only detected upon treatment with the reducing agent DTT, suggesting that they are linked to each other by a disulphide bridge (Figure 2B). N-terminal Edman sequencing of the two major fragments obtained with chymotrypsin revealed that cleavage occurred between Phe²⁸⁵ and Phe²⁸⁶, in the amino-acid stretch between the two unique cysteine residues (Cys²⁶² and Cys²⁹⁸) of AIP56 (Figure 2C). Altogether, these results indicate that AIP56 is composed of two domains linked by a disulphide bridge.

An additional smaller fragment (-17 kDa) was also obtained after chymotrypsin cleavage (Figure 4A). N-terminal Edman sequencing of the smaller fragment identified the peptide G³⁷⁴YGHD (SEQ ID NO: 1 1). The N-terminal domain is responsible for the catalytic activity and the C-terminal domain is implicated in binding to target cells

To better understand the function of the two AIP56 domains, constructs corresponding to the N- and C-terminal domains of the toxin (AIP56^1"285 and AIP56^286"497, respectively) were designed, taking into consideration the chymotrypsin cleavage site (Figure 2C). Purification of these two recombinant proteins using the experimental conditions used for the full-length toxin revealed that they display a major propensity to oxidize leading to the formation of DTT-sensitive dimers (Figure 10A), a phenomenon that could have a functional impact and complicate subsequent analyses. Therefore, versions of the constructs where the single cysteines were replaced by serines (AlP56^l'2S5C262S and AIP56²⁸⁶-^497C298S, respectively) were produced. The mutants are undistinguishable from the non-mutant proteins, as assessed by CD (Figure 10B and IOC) with the N-terminal domain composed mainly of a-helices, whereas β- sheet is the predominant secondary structure of the C-terminal moiety (Figure 2D). Furthermore, the weighted sum of the CD spectra of the N- and C-terminal domains reproduces the spectrum of the entire protein (Figure 2D), indicating conservation of the native structure.

To test the catalytic activities of the AIP56 N- and C-terminal domains, AJP56 ^2S5C2(,2S and AIP56^{286 497C298S} were incubated with fish leukocyte lysates (Figure IB) or with in vitro translated ³⁵S-labeled sbp65Rel (Figure 10D). The C-terminal construct did not display catalytic activity, whereas the N-terminal domain cleaved p65, similarly to the full-length toxin. However, neither changes in cellular p65 levels (Figure ID) nor apoptosis (Figure 3 A) were observed in sea bass leukocytes incubated with the N- or C-terminal truncate or with a mixture of both. This indicates that the two AIP56 domains are non-toxic and suggests that they need to be part of the same molecule to elicit a biological effect.

The cytosolic location of NF-κΒ p65 could mean that the lack of toxicity of the N-terminal domain was related to its inability to enter the cells and reach its target. Hence, a strategy to deliver the N-terminal domain into the cell cytosol was designed. Chimeric proteins consisting of the N-terminal portion of Bacillus anthracis LF fused to the ΑΓΡ56 protease domain (LF^11" ^-Α^ό^1"261) or C-terminal domain (LF^n"263-AIP56²⁹⁹-⁴⁹⁷) were produced. Intoxication assays were performed in the presence of PA, the receptor-binding subunit for LF [34]. In cells incubated with LF^H-²⁶³-AIP56^{1 261} the p65 levels were significantly reduced, confirming that

_LFl l-263._AIp56l-261

was successfully delivered into the cell cytosol, while no changes in p65 were observed in cells incubated with ΙΤ^^'^-ΑΙΡδό^299"497 (Figure 3B). Accordingly, LF^11" ⁶³·ΑΙΡ56^299"497 did not display apoptogenic activity, while incubation with LF^U"263-Aff56^1'261 resulted in an increased number of cells with apoptotic morphology (Figure 3B), similar to what was observed in cells incubated with AIP56. Thus, delivery of the AIP56 N-terminal domain into the cytosol reproduces the toxic effect of the full length toxin, confirming that this domain is responsible for the toxin's catalytic and apoptogenic activities. These results also suggest that the C-terminal domain of AIP56 is involved/required for the entrance of the toxin into cells. To investigate this possibility, AIP56^AAIVAA, AlP56^l-^2S5C262S or AIP56^286" were used in competition experiments with AIP56. Both p65 cleavage and apoptosis were monitored in these experiments. ArP56^28M97C29SS and ΑΙΡδό^^, but not AIP56^1" , were able to inhibit the apoptogenic activity of wild type AIP56 in a dose-dependent manner (Figure 3C, left panel). Furthermore, ΑΙΡδό^^ and AIP56^{286~ 97C298s} inhibited AIP56-mediated p65 degradation, whereas no effect could be observed when AIP56'^'285C262S was used as competitor (Figure 3C, right panel). These results indicate that the C-terminal domain mediates binding of the toxin to the cell surface and entry into the cells. In many AB- type toxins the C-terminal domain is composed of a receptor-binding subdomain/region and a translocation subdomain/region. As suggested by the proteolysis with chymotrypsin the same appears to be true for AIP56, because a smaller fragment could be identified after incubation of AIP56 with chymotrypsin (figure 4A). N-terminal Edman sequencing allowed identifying this fragment as the C-terminus portion of AIP56 (AIP56^374"497). Competition assays performed using the C-terminus fragment (AIP56^374"497) as competitor supported that this fragment corresponds to the AIP56-binding region.

Traaslocation of AIP56 requires integrity of the Cys -Cys linker but the disulphide bridge is not an absolute requirement for toxicity

Results obtained in experiments using N- and C-terminal truncates of AIP56 suggested that the two domains must be part of the same molecule to display toxicity. In order to investigate if the two domains of the toxin bound by a disulphide bridge are able to intoxicate cells, we nicked the toxin with chymotrypsin. Nicking of the toxin and integrity of the disulphide bridge linking the two fragments were confirmed by reducing and non-reducing SDS-PAGE (Figure 1 A). Surprisingly, no changes in p65 cellular levels (Figure ID) and no apoptosis (Figure 5 A) were observed upon incubation of sea bass peritoneal cells with nicked toxin. Similar results were obtained using a reconstituted version of the toxin (Figure ID and 5A) consisting of disulphide-bound AIP56^{1" 85}/AIP56^286"497 along with trace amounts of AIP56^1"285 and of AIP56²⁸⁶-⁴⁹⁷ homodimers and monomers (Figure 11A). Although nicking abolished cellular toxicity, it did not induce major structural changes (Figure 1 IB) and only a 1 °C decrease in Tm (39±0.13 °C for AIP56 and 38±0.25 °C for nicked AIP56; mean±SD of 16 measurements in four independent experiments) was measured by DSF. More importantly, nicked AIP56 retained both proteolytic activity against p65 in vitro (Figure IB and 5B) and cell binding ability, as indicated by the partial inhibition of the AIP56-mediated p65 cleavage and apoptosis in competition experiments (Figure 5C). These results suggest that the integrity of the linker region between the two cysteine residues is needed for toxin internalization, in contrast to what is known for the diphtheria, tetanus and botulinum toxins, where nicking of the inter-cysteine loop is required for toxicity [35,36].

In tetanus and botulinum neurotoxin type A, it has been shown that the disulphide bridge is essential for neurotoxicity [37,38]. We found that disruption of the disulphide bridge linking Cys²⁶² and Cys²⁹⁸ of AIP56 by alkylation (Figure 1 ID) did not affect the catalytic activity of the toxin in vitro (Figure 5B), but partially compromised its toxicity (Figure 5D), suggesting that in AIP56 the disulphide bridge plays a role in the intoxication process but is not an absolute requirement for toxicity. DISCUSSION

In this study, we report the structural and functional characterization of AIP56 as an A-B type bacterial exotoxin that cleaves NF-κΒ p65. Considering the anti-apoptotic functions of NF- κΒ, and in particular, of its p65 subunit [39-41], AIP56-mediated depletion of NF-κΒ p65 likely explains the disseminated phagocyte apoptosis observed in Phdp infections that contributes to subvert the host immune response and determines the outcome of the infection [9,12,33]. This adds to a general theme of host-pathogen interaction that has recently emerged, consisting in the induction of apoptosis of the host immune cells to the pathogen advantage [42-44]. Here we demonstrate that, similar to the anthrax lethal factor and to the clostridial neurotoxins [45,46], AIP56 is a zinc-endopeptidase but with a cleavage activity towards NF-κΒ p65. Furthermore, we show that AIP56 is organized into two distinct domains linked by a single disulphide bond. The N-terminal domain of AIP56 harbours the catalytic activity of the toxin and cleaves NF-κΒ p65 at the Cys³⁹-Glu⁴⁰ peptide bond, within the p65 N-terminal Rel homology domain. Several key residues of p65 known to be involved in DNA interaction are located at the Rel homology domain [47-49]. In the last decade, several reports revealed that Cys³⁸ of human p65 (Cys³⁹ in sea bass p65) interacts with the phosphate backbone of NF- B binding sites [47], that its oxidation and nitrosylation inhibit DNA binding [50] and that it is targeted by several inhibitors of NF-κΒ with anti-inflammatory and/or anticancer properties [51-61]. More recently, it was shown that hydrogen sulphide-linked sulfhydration of Cys³⁸ of human p65 plays a key role in regulating the anti-apoptotic actions of NF-KB [41]. Therefore, cleavage of sea bass p65 by AIP56 disrupts a segment crucial for DNA interaction. Considering that the proteolytic activity of AIP56 towards p65 is similar to the one previously described for NleC (both proteases cleave p65 at the same peptide bond), and based on the observation that p65 cleavage by NleC compromises NF-κΒ dependent transcription [15,16,18], it is likely that AIP56 also affects NF-κΒ transcriptional activity.

Although in many cell types down-regulation of NF-κΒ is not sufficient to trigger apoptosis, it is widely recognised that cells with inactivated NF-κΒ are more prone to commit suicide in response to different stimuli, including TNF-a and TLRs ligands [39,62-64], and that the anti- apoptotic actions of NF-κΒ can be largely attributed to its p65 subunit [39-41]. In the context of bacterial infections, inhibition of NF-κΒ function usually leads to impairment of the inflammatory responses [2]. The induction of apoptosis by bacterial effectors through interference with NF-κΒ activity has also been described, but is a far more uncommon scenario. Examples are the Yersinia YopP/J [65-67] and the Aeromonas salmonicida Aop [68], both inhibiting the degradation of the inhibitory ΙκΒ proteins [65,69-72], and the V. parahaemolyticus protein VP 1686 that interacts with and suppresses DNA binding activity of NF-KB [73]. It remains to be determined whether AIP56-mediated depletion of p65 is sufficient to induce apoptosis, in resemblance to what has been suggested for the macrophage apoptosis induced by the V. parahaemolyticus type III secreted effector VP1686 [73], or if it requires an additional stimulus.

Almost all bacterial effectors that have been described to target NF-κΒ signalling are injected directly into the host cell cytosol by type III or type IV secretion systems (see reviews by [2,4]). In contrast, we have found that the AIP56 N-terminal metalloprotease can only act when linked to a C-terminal binding domain that, by analogy with other A-B toxins, may assist the protease domain in its membrane translocation into the cytosol [34,74,75].

Bacterial A-B toxins are often secreted as a single polypeptide chain that is cleaved into the disulphide-bound A and B domains [36,75]. In these toxins, proteolytic nicking and integrity of the disulphide bond linking the A and B domains are essential for toxicity [33,35,37,76- 78]. In contrast, AIP56 toxicity is abolished by proteolytic nicking and only mildly compromised by disruption of the disulphide bridge by alkylation. Considering that nicked AIP56 retains the ability to interact with the cell membrane, these observations suggest that the linker region (between Cys²⁶² and Cys²⁹⁸) is involved in translocating the toxin into the host cell cytosol. The decreased toxicity resulting from alkylation suggests that the integrity of the disulphide bond is important, although not absolutely required, for AIP56 intoxication. The disulphide bond may be involved in stabilizing the spatial relationship between the domains. In addition, that bond is hydrophobic and polarizable and its alkylation can have implications in membrane insertion, as reported for tetanus and botulinum neurotoxins [79]. AIP56 is synthesised as a single polypeptide and, contrary to what has been reported for most A-B toxins, there is no evidence of proteolytically processed toxin in the bacterial culture supernatants or in the serum of infected fish [12]. Furthermore, despite several attempts, we were unable to detect proteolytic processing of AIP56 upon its interaction with host cells. If AIP56 needs to be processed in order to exert its effect, the lack of detection of processed toxin may result from a very small amount of processed toxin (not detectable in our experiments) being sufficient to intoxicate the cells, similarly to what was described for other toxins [80]. Alternatively, after endocytosis, unprocessed AIP56 may be translocated into the cytosol as described for Pseudomonas exotoxin A [81,82] or may localize in an endomembrane (e.g. endosomal membrane) with the catalytic domain facing the cytosolic compartment where it can interact with and cleave p65. Studies aiming at discriminating between these hypotheses will be developed in the future.

It is now recognised that horizontal transfer of entire genes or portion of genes plays a key role in generating diversity in pathogens by allowing them to acquire novel phenotypic characteristics. Indeed, there are several examples of bacterial genes with a mosaic structure, composed of diverse segments with different origins [83-88]. The structure of AIP56 suggests that the toxin has a chimeric structure, having an N-terminal catalytic domain similar to the type III effector NleC. The actual transfer events that originated such a chimeric protein toxin remain to be disclosed.

The AIP56 catalytic domain and NleC have the same NF-κΒ p65 cleavage activity. However, NleC requires a type III secretion machinery for activity, while AIP56 has an intrinsic ability to reach the cytosol, due to the presence of the additional C-terminal domain that functions as a "delivery module". This difference may have relevant implications when considering the use of both pathogen-derived molecules as therapeutic agents in situations associated with uncontrolled activation of NF-κΒ such as inflammatory diseases and cancer. MATERIALS AND METHODS

Fish

Sea bass {Dicentrarchus labrax), were kept in a recirculating, ozone-treated salt-water (25-30 %o) system at 20 + 1 °C, and fed at a ratio of 2 % body weight per day. Fish were euthanized with 2-phenoxyethanol (Panreac; >5 ml/ 10 L).

Production and purification of recombinant proteins

DNA coding sequences were cloned into Ncol/Xhol restriction sites of pET-28a(+) (Novagen) as described in Supporting Information. Mutants were generated by site directed mutagenesis using QuickChange Site-Directed Mutagenesis Kit (Stratagene) following manufacturer's instructions. Recombinant His-tagged proteins were expressed in E. coli BL21(DE3) cells. AIP56, AIP56^1"285, AIP56^286"497, ΑΠ β^286"49702988, AIP56^374"497, NleC, LF^1,-²⁶³-AIP56¹-^2M and

_LF1 1-263. _AIp56299-497

were purified from the soluble fraction of induced bacteria by metal- affinity chromatography. After this step, AIP56, AIP56¹-²⁸⁵, AIP56^286"497 were subjected to anion exchange chromatography, whereas AIP56^28M97C298S, ΑΓΡ56^374"497 and NleC were subjected to size exclusion chromatography. ^S6^PMWfiiA and AIP56^1"28502625 were purified from inclusion bodies by metal-affinity chromatography under denaturing conditions, refolded by dialysis against sea bass PBS (sbPBS; phosphate buffer saline with osmotic strength adjusted to 322 mOsm) with 10 % glycerol and purified by size exclusion chromatography. For reconstitution of AIP56, ΑΓΡ56^1"285 and AIP56^286"497 were mixed in equimolar amounts in 8 M urea, 1 mM DTT and refolded by extensive dialysis against sbPBS. Nicked AIP56 was obtained by limited proteolysis with 25 g/mI chymotrypsin, as described below, followed by metal-affinity chromatography purification. To prepare alkylated toxin, 63 μΜ AIP56 in 20 mM Tris-HCl pH 8.0, 200 mM NaCl, 10 mM DTT was incubated with 5 mM iodoacetamide (Sigma) for 30 min at RT and dialysed against 20 mM Tris-HCl pH 8.0, 200 mM NaCl.

Sea bass NF-κΒ p65 REL homology domain (sbp65Rel) was purified from the soluble fraction of induced bacteria by metal-affinity chromatography. Untagged ³⁵S-labeled sbp65Rel and sbp65Rel mutants (sbp65RelC39A, sbp65RelE40A, and sbp65C39E40AA) were produced using the TNT T7 Quick Coupled transcription/Translation kit (Promega), in the presence of Redivue™ L-[35S] methionine (specific activity of N 1000 Ci/mmol). Limited proteolysis

AIP56 at 0.6 mg/ml in 10 mM Tris-HCl pH 8.0, 200 mM NaCl was incubated with 0.25 to 25 μ§/Γη1 trypsin, chymotrypsin or proteinase K (molar ratios of protease: AIP56 of approximately 1 : 10 to 1 : 1000) for 30 min on ice. Alternatively, AIP56 (1.25 mg/ml) was incubated with 47 §/πι1 of chymotrypsin for 30 min on ice. Proteases were inactivated by addition of PMSF to a final concentration of 250 g/ml. Digests were analysed by reducing and non-reducing SDS- PAGE. The chymotrypsin digestion fragments were subjected to N-terminal sequencing.

Circular dichroism spectroscopy (CD)

Far UV CD spectra were acquired on an Olis DSM 20 circular dichroism spectropolarimeter controlled by the Globalworks software. Each spectrum is the average of three scans collected at 20 °C with a 0.2 mm path length cuvette and with an integration time of 4 seconds. Proteins were dissolved in 10 mM Tris-HCl, 50 mM NaCl, pH 8.0 and concentrations were determined by absorbance measurements. Analysis of the protein secondary structure was performed using the Globalworks software algorithm.

Cells

Sea bass peritoneal leukocytes were obtained as described [33] and used at a density of 2xl0⁶ cells/ml. The peritoneal population of cells consist of approximately 70% macrophages and 20% neutrophils with the presence of small numbers of eosinophilic granular cells, lymphocytes and erythrocytes [33].

Apoptogenic activity assays

Cells were incubated for 4 h at 22 °C with AIP56 or ΑΓΡ56 derived proteins at the indicated doses. Where indicated, the cells were pre-treated for 30 min at 22 °C with 25 μΜ of the pan- caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp(O-Me) fluoromethyl ketone (Z-VAD- FMK). In experiments using LF chimeric proteins, cells were incubated with the indicated concentrations of LF^-AIPSe^1"261 or LF^11"263-AIP56²"- ⁹⁷ with or without 10 nM of anthrax protective antigen (PA) obtained as described [89]. Mock- and AIP56-treated cells were used as controls. Apoptosis was assessed as described [10], by light microscopy morphological analysis of cytospin preparations stained with Hemacolor (Merck) after labelling the neutrophils using the Antonow's technique [90,91].

NF-KB p65 cleavage assays

In vitro: p65 cleavage was assessed using sea bass peritoneal cell lysates, recombinant or in vitro translated ³⁵S-labeled sbp65Rel. To prepare cell lysates, cells were incubated in 10 mM Tris-HCl pH 8.0, 150 raM NaCl, 0.5% Triton X-100 and 10% glycerol for 30 min on ice, briefly sonicated and centrifuged. The supernatant of 2xl0⁶ cells was incubated with 1 μΜ of the indicated proteins for 2 h at 22 °C and p65 cleavage evaluated by Western blotting using an anti-sea bass NF-κΒ p65 rabbit sera (produced using the peptide SIFNSGNPARFVS (SEQ ID NO: 7) located at the C-terminal region of sea bass p65 as antigen). Recombinant sbp65Rel (7.5 μΜ) was incubated for 3 h at 22 °C with 1 μΜ A1P56 and p65 cleavage evaluated by SDS-PAGE. In vitro translated ³⁵S-labeled sbp65Rel and ³⁵S-labeled sbp65Rel mutants (sbp65RelC39A, sbp65RelE40A, and sbp65C39E40AA) were incubated with 0.1, 1, 10 or 100 nM of the different proteins in 10 mM Tris-HCl pH 8.0, 150 mM NaCl and 10% glycerol for 2 h at 22 °C and p65 cleavage assessed by autoradiography. When specified, 1,10- phenanthroline (Sigma) was used at 5 mM.

Ex vivo: peritoneal leukocytes were treated for 2 or 4 h as described above in the section "Apoptogenic activity assays", collected by centrifugation, washed, resuspended in sbPBS and lysed by addition of SDS-PAGE sample buffer. Cleavage of p65 was evaluated by Western blotting, as described above.

Competition assays

_AIp56AAIVAA _AIp56,-285C262S₅ ^^286-49702988 ^ _AIp₅₆374-497 ^ _FOR ^ ^JJ^ _TQ inhibit AIP56's apoptogenic activity and AIP56-mediated p65 cleavage. Cells were pre- incubated for 15 min on ice with different concentrations (350 nM to 3.5 μΜ) of AEP56^AAIVAA,

_AIp56l-285C262S₅ ^^286-49702988^ _{AIp 56 Qr AI}p5₆374-497_{? followed by} i_{ncu at}ion for further 15 min on ice with 8.75 nM of AIP56 in the presence of the competitors. Unbound proteins were removed by washing with ice cold supplemented L-15 medium [33] and the cells incubated at 22 °C for 4 h.

Statistical analysis

Statistical analysis was performed using a randomized block design, where fish are treated as blocks and the concentration of the treatments/competitors as a factor. The data, percentage of apoptotic cells, have been transformed using the arcsine transformation. Post-hoc comparisons were performed using the Tukey^'s Honest Significant Difference test. Significance was defined for p<0.05. SUPPLEMENTARY MATERIAL AND METHODS

Constructs

For bacterial protein expression, DNA coding sequences were cloned into Ncol/Xhol restriction sites of pET-28a(+) (Novagen) in frame with a C-terminal 6xHis-tag. Mutated versions of the proteins were obtained by site directed mutagenesis using QuickChange® Site- Directed Mutagenesis Kit (Stratagene).

AIP56 mutant (MPSG * "^: plasmid ρΕΤΑΒΜβ^^, coding for His-tagged mutated version of AIP56, was obtained using pETAIP56FT [12] as template. The mutations consist on the substitution by Ala of the key residues for zinc ion coordination and water molecule activation: His¹⁶⁵, Glu¹⁶⁶, His¹⁶⁹ and His¹⁷⁰.

AIP56 domains: AIP56 putative A (AIP56^1'285) and B (AIP56^{286 497}) domains and the C- terminus binding subdomain (AIP56^374"497) were designed based on the chymotrypsin cleavage site (see below). pETAIP56H⁺ [12] was subjected to PCR amplification using primers AIP56Fw4NcoI/AIP56Rv7XhoI (Table SI) for AIP56^1"285, and AIP56Fw6NcoI/AIP56Rv5XhoI (Table SI) for AIP56^286"497. PCR fragments were cloned into pET-28a(+), yielding pETAIP56'^"285 and pETAIP56^286"497. Mutated versions of AIP56^1"285 and AIP56²⁸⁶-⁴⁹⁷ with Cys²⁶² or Cys²⁹⁸ mutated to Ser (AIP56^1"285C262S and ΑΠ^^286"497^⁹⁸⁵) were obtained using either pETAIP56'^"285 or pETAIP56^286-497 as template.

NleC: the NleC coding sequence was amplified by PCR with primers EHECFwlNcoI and EHECRvlXhoI (Table SI) using total DNA [92] from E. coli 0157:H7 strain 4462 as template. PCR products were cloned into pET-28a(+).

LF-AIP56 chimeric proteins: Chimeric proteins consist on the amino-terminus of anthrax lethal factor (LF^11"263) fused to AIP56 N-terminal Ο-Ρ^-ΑΙΡδό^{1" 61}) or AIP56 C-terminal (LF^{l lje3}-AIP56^{295 97}). The sequence encoding LF^11-263 was amplified with primers LFFwlNcoI/LFRvl SacI (Table SI) from plasmid pRSET A (Invitrogen) containing LF gene [93] and regions encoding AIP56^1"261 and AIP56^299"497 were amplified from pETAIP56H+ [12] using the primer combinations NtermAIP56FwlSacI/AIP56Rv9XhoI and CtermAIP56FwlSacI/AIP56Rv5XhoI (Table SI), respectively. PCR fragments were digested with Sad, and LF^11"263 ligated either with ΑΙΡ56¹·²⁶¹ or AIP56²"-⁴⁹⁷ Ligations were subjected to PCR using the primer combination LFFwlNcoI/AIP56Rv9XhoI or LFFwlNcoI/AIP56Rv5XhoI (Table SI) and PCR products were cloned into pET-28a(+). sbp65Rel: The coding region (amino acids 1-188) of sea bass p65 REL domain (sbp65Rel) was amplified by PCR from cDNA produced as described in [94] using DLp65FwlNcoI with either DLp65Rv4XhoI or DLp65Rv2XhoI primers (table S I), for cloning His-tagged or untagged sbp65Rel, respectively. PCR products were cloned in pET-28a(+). The vector carrying his-tagged sbp65Rel was used for bacterial expression and the untagged version for production of ³⁵S-labeled sbp65Rel.

Protein production and purification

Recombinant His-tagged proteins were expressed in E. coli BL21(DE3) cells grown in Lysogenic broth (LB) [95] supplemented with 50 μg/ml kanamycin. Protein expression was induced with 1 mM Isopropyl β-D-l-thiogalactopyranoside (IPTG) and carried out overnight at 17 °C, except for ΑΙΡδδ*^^, AIP56'^"285C262S and sbp65ReI, which were expressed at 37 °C for 4 h. Starting cultures were grown overnight at 37 °C, used to inoculate fresh medium at 1 : 100 dilution and grown at 37 °C. For expressions at 17 °C, cultures were equilibrated to this temperature prior to induction. Protein expression was induced when bacterial cultures reached an OD of approximately 0.6. Induced bacterial cells were collected by centrifugation, resuspended in 50 mM phosphate buffer pH 7.4, 500 mM NaCl, 200 μg/ml lysozyme, 250 μg/ml PMSF and lysed by freeze/thaw followed by sonication in the presence of 10 μg/ml DNase I, and 10 mM MgCl₂.

Affinity chromatography was performed in HisTrap HP columns (GE Healthcare) or using HIS-Select® Nickel Affinity Gel (Sigma) for sbp65Rel purification. Anion exchange chromatography was performed in a Bio-Scale™ Unosphere Q cartridge (BioRad) and size exclusion chromatography on Sephacryl S100 HR column (GE Healthcare) except for ΑΠ^ό^^ that was purified in a Sephacryl S200 HR column (GE Healthcare). Purified proteins were concentrated (5-10 mg/ml) using Amicon Ultra-15 Centrifugal Filter Units, frozen in liquid nitrogen and stored at -80 °C. Protein purity was evaluated by SDS-PAGE followed by Coomassie-blue staining. AIP56: production of soluble recombinant His-tagged AIP56 was adapted from [12]. AIP56 was purified from the soluble fraction by affinity chromatography. Elution was carried out with increasing concentrations of imidazole in 50 mM phosphate buffer pH 7.4, 500 mM NaCl. Fractions containing AIP56 were pooled and applied to an anion exchange chromatography using a linear NaCl gradient (50 mM to 1 M) in 20 mM Tris-HCl pH 8.0. Fractions containing AIP56 were pooled, concentrated and stored in 10 mM Tris-HCl pH 8.0, 200 mM NaCl. AIP56 mutant AIP56^AAIVAA): ΑΠ^ό**¹^ was purified from inclusion bodies, solubilized in 8 M urea, 20 mM Tris-HCl pH 8.0, 500 mM NaCl, 1 mM 2-mercaptoethanol, 5 mM imidazole and purified by affinity chromatography under denaturing conditions. The protein was eluted in the above buffer with increasing concentrations of imidazole. Fractions containing AIP56^AAIVAA were pooled, adjusted to 0.1 mg/ml, and subjected to refolding through dialysis against 3 x 50 volumes of sea bass PBS (sbPBS; phosphate buffer saline with osmotic strength adjusted to 322 mOsm) with 10 % glycerol, at 4 °C. The refolded protein was purified by size exclusion chromatography in 20 mM Tris-HCI pH 8.0, 200 mM NaCl, 10 % glycerol. Fractions containing AIP56^AAIVAA were pooled, concentrated and stored in 20 mM Tris-HCl pH 8.0, 200 mM NaCl, 10 % glycerol.

AIP56 truncates: Production and purification of AIP56^1"285 and AIP56^286"497 were done as described for AIP56. AIP56^1"285C262S was purified from inclusion bodies as described for AIP56^AAIVAA. AIP56²⁸⁶-^497C298S was purified from the soluble fraction by affinity chromatography as described for AIP56 but adding 10 % glycerol to elution buffers. Fractions containing Arp56²⁸⁶-^497C298S _were pooled and purified by size exclusion chromatography as described for AIP56^AAIVAA. ΑΓΡ56^1_285 and AIP56^286"497 were stored in 10 mM Tris-HCl pH 8.0, 200 mM NaCl while ΑΙΓ^ό¹-^28502625 and AIP56^286"497C298S were stored in 20 mM Tris-HCl pH 8.0, 200 mM NaCl, 10 % glycerol.

NleC: NleC was purified from the soluble fraction by affinity chromatography as described for ΑΓΡ56 but using 20 mM phosphate buffer pH 8.0, 500 mM NaCl with increasing concentrations of imidazole for protein elution. Fractions containing NleC were pooled and subjected to size exclusion chromatography in 20 mM Tris-HCl pH 8.0, 50 mM NaCl. Fractions containing NleC were pooled, concentrated and stored in 20 mM Tris-HCl pH 8.0, 50 mM NaCl.

LF-AIP56 chimeric proteins: Chimeric LF^-AIPSe^{1" 61} and LF^11"263- ΑΓΡ56^299"497 were purified by affinity chromatography as described for AIP56. In the case _{of LF}l l-263._{AIp 56}l-261 to avoid the formation of protein aggregates, sonication of cell lysates was avoided and glycerol (5% final concentration) was added to cell lysates and affinity chromatography buffers. The chimeras were concentrated and stored in 10 mM Tris pH 8.0, 200 mM NaCl.

Sbp65Rel: His-tagged sbp65Rel was purified from the soluble fraction of induced bacterial cells by affinity chromatography through elution with increasing concentrations of imidazole in 50 raM phosphate buffer pH 7.4, 300 mM NaCl. Fractions containing sbp65Rel were pooled, concentrated and stored in 50 mM phosphate buffer pH 7.4, 300 mM NaCl.

Protein quantification

Reconstituted heterodimeric protein (AIP56^1"285 linked to AIP56^286"497 by a disulfide bridge) was quantified by densitometry analysis of coomassie-blue stained gels using BSA standards. The other proteins were quantified by measuring absorbance at 280 nm and using the extinction coefficient calculated by the ProtParam tool (http://www.expasy.org/tools/protparam.html), using the Edelhoch method [96], but with the extinction coefficients for Trp and Tyr determined by Pace et al [97].

PAGE and Western blotting

SDS- and Native-PAGE were performed using the Laemmli discontinuous buffer system [22] as described in [12]. Proteins were transferred onto nitrocellulose membranes and probed with anti-AIP56 [12] or anti-sea bass NF-κΒ p65 rabbit sera (produced using the peptide SIFNSGNPARFVS (SEQ ID NO: 7) located at the C-terminal region of sea bass p65 as antigen). Reactive bands were detected using anti-rabbit IgG alkaline phosphatase conjugate (Sigma) followed by BCIP/NBT development or using an anti-rabbit IgG horseradish peroxidase-linked secondary antibody (The binding site) followed by detection with SuperSignal^® West Dura Extended Duration Substrate (Pierce biotechnology). Blots shown are representative of at least three independent experiments.

Analysis of Zinc content

The zinc content was determined by atomic absorption spectroscopy with flame optimization in an Atomic Absorption Spectrometer PU 9200X (Philips). All buffers were prepared with chemicals of the highest purity available and using Milli-Q grade water. All material was previously immersed for 24 hours in 10% nitric acid, washed with Milli-Q grade water and autoclaved in paper bags. Before metal determination, protein samples (0.1-0.2 mg/ml) were extensively dialyzed against 10 mM Tris-HCl pH 8.0, 200 mM NaCl using Amicon Ultra-15 centrifugal filter devices (Millipore). Metal content was measured after standardization in the linear concentration range (0-1 ppm). Results are expressed as the mean±SD of three independent measurements carried out in triplicate.

Analytical size exclusion chromatography

Protein samples in 10 mM Tris-HCl pH 8.0, 200 mM NaCl were subjected to analytical size exclusion chromatography in a Superose 12 10/300 column (GE Healthcare) using an AKTA Purifier FPLC system (Pharmacia) at room temperature and a 0.5 ml/minute flow rate. The column was pre-equilibrated in the above buffer and protein elution was monitored by measuring the absorbance at 280 nm. Molecular weights of eluted proteins were estimated based on column calibration with molecular weight/stokes radius standards.

Circular dichroism Spectroscopy (CD)

Far UV CD spectra were acquired on an Olis DSM 20 circular dichroism spectropolarimeter continuously purged with nitrogen, equipped with a Quantum Northwest CD 150 Temperature-Controlled cuvette and controlled by the Globalworks software. Scans were collected at 20 °C with a 0.2 mm path length cuvette between 190 and 260 nm at 1 nm intervals. Three scans with an integration time of 4 seconds were averaged for each measurement. Protein concentration was determined by absorbance measurements. Proteins were dissolved in 10 mM Tris-HCl pH 8.0, 50 mM NaCl. The results are expressed in terms of mean residue molar ellipticity [0]MR_W in deg cm² dmoF¹, according to the equation [0]MR_W = O_obs * MW * 100 / (1 * c * N) where 0_obs is the observed ellipticity in deg, MW is the protein molecular weight in g mol, 1 is the cuvette path length in cm, c is the protein concentration in g/1 and N is the number of residues of the protein. Analysis of the protein secondary structure was performed using the Globalworks software algorithm. Differential scanning fluorimetry (DSF)

The unfolding of ΑΓΡ56 and nicked AIP56 was monitored by following SYPRO® Orange (Invitrogen) fluorescence. Proteins samples in phosphate/citrate buffer, 150 mM NaCl pH 7.0 were mixed 1 : 1 (v/v) with dye solution in the same buffer in a final volume of 30 μΐ and analyzed in white 96-well plates on a iQ5 Real Time PCR System (BioRad) by measuring fluorescence at 585 nm as a function of temperature (scanned from 20 to 95 °C in 0.5 °C/min steps). AIP56 and nicked AIP56 were used at 5 μΜ with 10 x SYPRO® Orange. Controls included no protein and/or no dye. The melting curves were analyzed using CFX Manager (BioRad) and the melting temperature (T_m) was calculated as the inflection point of the curve of 16 measurements in 4 independent experiments.

Determination of the AJP56 concentrations in the plasma of infected fish

Sea bass weighting 16.3±2.4 g were infected i.p. with a lethal dose (1.8xl0⁶ CFU/fish) of Phdp strain PP3 [9]. Growth conditions and inoculum preparation were carried out as described [10, 12]. Plating serial dilutions of the final bacterial suspensions onto TSA-1 plates and counting the number of CFU following incubation at 22°C for 2 days confirmed bacterial concentrations of the inoculum. Plasmas were collected from moribund fish as previously described [9]. Five microliters aliquots of the plasmas or recombinant AIP56 standards (at 50, 25, 10, 2 and 1 μ^ι ΐ) were analyzed by Western blotting. The concentrations of AIP56 in the plasmas were determined by densitometry using a recombinant AIP56 standard curve. TRANSLOCATION ΓΝΤΟ MAMMALIAN CELLS

Available data show that similarly to diphtheria toxin, botulinum and tetanus neurotoxins, among others, AIP56 is an AB toxin, having an N-terminal/A domain responsible for the catalytic activity connected by a disulfide bridge to a C-terminal/B domain responsible for the binding and internalization of the toxin into the cytosol of susceptible cells.

Results recently obtained show that besides inducing NF-κΒ p65 depletion in sea bass cells, AIP56 also enters and cleaves p65 in several mammalian cells lines (e.g., HeLa, J774A.1, Raw 264.7) (Figure 12) and in both mouse bone marrow derived macrophages (mBMDM) (Figure 13) and mouse bone marrow derived dendritic cells (not shown). Cell entry into macrophages and dendritic cells has been shown to be particularly effective. This result is highly surprising and suggests that the toxin unexpectedly recognises a receptor that is evolutionarily conserved. The extent of p65 depletion upon incubation with the toxin is cell- type specific (Figure 12 and 13). In mouse BMDM, incubation with AIP56 results in apoptosis, as confirmed by morphological analysis and detection of active caspase-3 in the toxin-treated cells (Figure 14).

Different toxins have distinct mechanism of internalization, but usually they are internalized by receptor-mediated endocytosis before translocation into the cytosol [98-102]. The entry process is initiated by binding to cell-surface receptors, followed by endocytosis and vesicular trafficking to the site of membrane translocation. In what concerns the cytosolic entry, the toxins fall into two main groups: some enter from endosomes in response to low pH, whereas others are transported all the way to the Golgi apparatus and the endoplasmic reticulum (ER) before they are translocated to the cytosol. For a subset of toxins, that are translocated directly from endosomes (exemplified by diphtheria toxin, anthrax toxins, Clostridium botulinum C2, tetanus and botulinum toxins and C. difficile toxin B), the trigger for the translocation is the low endosomal pH and, as a consequence, those toxins are inhibited by agents that prevent acidification of the endosome/lysosome. [100]. Analysis of the interaction of AIP56 with mouse BMDM indicates that the toxin is internalized by a process that requires endosome acidification, since concanamycin A and bafilomycin Al, known to interfere with this process, protected cells from intoxication (Figure 15).

In the case of toxins that translocate directly from endosomes, the translocation is triggered by the low pH, which induces a conformational change in the toxin molecule that leads to exposure of hydrophobic areas able to interact with and insert into the membrane, leading to the translocation of the enzymatically active part into the cytosol [100]. In our studies, we found that in vitro, under acidic environment, AIP56 undergoes reversible conformational changes resulting in the exposure of hydrophobic residues/regions and, therefore, in increased hydrophobicity (Figure 16). These conformational alterations were only observed at pH 5.5 or lower and conferred to AIP56 the ability to interact with artificial lipid bilayer membranes (Figure 17). The observed current fluctuations were irregular and inhomogeneous, indicating that interaction of AIP56 with bilayer membranes does not lead to the formation of regular channels comparable to the ones formed by other toxins such as Clostridium botulinum C2 or anthrax toxins [103,104]. The current fluctuations induced by AIP56 are similar to the ones reported to occur with the C. botulinum and C. limosum C3 toxins [105] as well as with the C. difficile TcdB and TcdA toxins [30,106] and may be explained by the formation of transient channels.

Our observations further show that low extracellular pH promotes translocation of cell surface-bound AIP56 across the cytoplasmic membrane into the cytosol (Figure 18), similarly to what has been reported for several toxins that translocate from endosomes through a pH- dependent mechanism [28-30].

It is widely known that uncontrolled activation of NF-kB is associated with several inflammatory diseases as well as with human cancers. The observations that AIP56 is able to enter and cleave NF-kB p65 in mammalian cells opens a new area of research involving this toxin, and the nucleic acid encoding it, for its use as an anti-inflammatory or anti-cancer therapeutic. Furthermore, because the delivery of proteins and nucleic acids directly into viable cells has a huge biotechnological interest, a fine knowledge of the ΑΓΡ56 structural determinants allows the use of AEP56-derived fragments as delivery systems able to translocate different cargos into eukaryotic cells. REFERENCES

1. Gilmore TD, Wolenski FS (2012) Immunological Reviews 246: 14-35.

2. Rahman MM, McFadden G (2011) Nat Rev Microbiol 9: 291-306.

3. Le Negrate G (2012) Cell Microbiol 14: 155-167.

4. Neish AS, Naumann M (201 1) Trends Microbiol 19: 596-605.

5. Magarinos B, Toranzo AE, Romalde JL (1996) Annu Rev Fish Dis 6: 41-64.

6. Romalde JL (2002) Int Microbiol 5: 3-9.

7. Thune RL, Stanley LA, Cooper RK (1993) Annu Rev Fish Dis 3: 37-68.

8. Barnes AC, dos Santos NM, Ellis AE (2005) Dev Biol (Basel) 121: 75-84.

9. do Vale A, Costa-Ramos C, Silva A, Silva DS, Gartner F, et al. (2007) Cell Microbiol 9:

988-1003.

10. do Vale A, Marques F, Silva MT (2003) Fish Shellfish Immunol 15: 129-144.

11. Silva MT, do Vale A, dos Santos NM (2008) Apoptosis 13 : 463-482.

12. do Vale A, Silva MT, dos Santos NM, Nascimento DS, Reis-Rodrigues P, et al. (2005) Mol Microbiol 58: 1025-1038.

13. Schiavo G, Poulain B, Rossetto O, Benfenati F, Tauc L, et al. (1992) EMBO J 1 1 : 3577-

3583.

14. Silva MT, Dos Santos NM, do Vale A (2010) Toxins (Basel) 2: 905-918.

15. Baruch K, Gur-Arie L, Nadler C, Koby S, Yerushalmi G, et al. (2010) EMBO J 30: 221- 231.

16. Muhlen S, Ruchaud-Sparagano MH, Kenny B (2011) J Biol Chem 286: 5100-5107.

17. Pearson JS, Riedmaier P, Marches O, Frankel G, Hartland EL (2011) Mol Microbiol 80:

219-230.

18. Yen H, Ooka T, Iguchi A, Hayashi T, Sugimoto N, et al. (2010) PLoS Pathog 6: el001231.

19. Sham HP, Shames SR, Croxen MA, Ma C, Chan JM, et al. (2011) ect Immun 79: 3552-

3562.

20. Bendtsen JD, Nielsen H, von Heijne G, Brunak S (2004) J Mol Biol 340: 783-795.

21. Nielsen H, Engelbrecht J, Brunak S, von Heijne G (1997) Protein Eng 10: 1-6.

22. Laemmli UK (1970) Nature 227: 680-685.

23. Femandes SS, Nunes A, Gomes AR, de Castro B, Hider RC, et al. (2010) Microbes Infect

12: 287-294.

24. Huss M, Wieczorek H (2009) J Exp Biol 212: 341 -346.

25. Qa'Dan M, Spyres LM, Ballard JD (2000) Infect Immun 68: 2470-2474.

26. Drapeau GR, Boily Y, Houmard J (1972) J Biol Chem 247: 6720-6726.

27. Houmard J, Drapeau GR (1972) Proc Natl Acad Sci U S A 69: 3506-3509. 28. Draper RK, Simon MI (1980) J Cell Biol 87: 849-854.

29. Sandvig , Olsnes S (1980) J Cell Biol 87: 828-832.

30. Barth H, Blocker D, Behlke J, Bergsma-Schutter W, Brisson A, et al. (2000) J Biol Chem

275: 18704-18711.

31. Coiras M, Lopez-Huertas MR, Mateos E, Alcami J (2008) Retrovirology 5: 109.

32. Kang ΚΉ, Lee KH, Kim MY, Choi KH (2001) J Biol Chem 276: 24638-24644.

33. Costa-Ramos C, Vale AD, Ludovico P, Dos Santos NM, Silva MT (2011) Fish Shellfish

Immunol 30: 173-181.

34. Collier RJ (2009) Mol Aspects Med 30: 413-422.

35. Sandvig K, Olsnes S (1981) J Biol Chem 256: 9068-9076.

36. Schiavo G, Malleoli M, Montecucco C (2000) Physiol Rev 80: 717-766.

37. Schiavo G, Papini E, Genna G, Montecucco C (1990) Infect Immun 58: 4136-4141.

38. de Paiva A, Ashton AC, Foran P, Schiavo G, Montecucco C, et al. (1993) J Neurochem

61 : 2338-2341.

39. Beg AA, Baltimore D (1996) Science 274: 782-784.

40. Beg AA, Sha WC, Bronson RT, Ghosh S, Baltimore D (1995) Nature 376: 167-170.

41. Sen N, Paul Bindu D, Gadalla Moataz M, Mustafa Asif K, Sen T, et al. (2012) Molecular cell 45: 13-24.

42. Lamkanfi M, Dixit VM (2010) Cell Host Microbe 8: 44-54.

43. DeLeo FR (2004) Apoptosis 9: 399-413. ^'

44. Weinrauch Y, Zychlinsky A (1999) Annu Rev Microbiol 53: 155-187.

45. Klimpel KR, Arora N, Leppla SH (1994) Mol Microbiol 13: 1093-1100.

46. Schiavo G, Rossetto O, Benfenati F, Poulain B, Montecucco C (1994) Ann N Y Acad Sci

710: 65-75.

47. Chen FE, Huang DB, Chen YQ, Ghosh G (1998) Nature 391 : 410-413.

48. Chen YQ, Ghosh S, Ghosh G (1998) Nat Struct Biol 5: 67-73.

49. Hoffmann A, Natoli G, Ghosh G (2006) Oncogene 25: 6706-6716.

50. Kelleher ZT, Matsumoto A, Stamler JS, Marshall HE (2007) J Biol Chem 282: 30667-

30672.

51. Garcia-Pineres AJ, Castro V, Mora G, Schmidt TJ, Strunck E, et al. (2001) J Biol Chem 276: 39713-39720.

52. Garcia-Pineres AJ, Lindenmeyer MT, Merfort I (2004) Life Sci 75: 841-856.

53. Han Y, Englert JA, Yang R, Delude RL, Fink MP (2005) J Pharmacol Exp Ther 312:

1097-1 105.

54. Sethi G, Ahn KS, Aggarwal BB (2008) Mol Cancer Res 6: 1059-1070. 55. Watanabe M, Nakashima M, Togano T, Higashihara M, Watanabe T, et al. (2008)

Biochem Biophys Res Commun 376: 310-314.

56. Laga M, Cottyn A, Van Herreweghe F, Vanden Berghe W, Haegeman G, et al. (2007)

Biochem Pharmacol 74: 579-589.

57. Ha KH, Byun MS, Choi J, Jeong J, Lee KJ, et al. (2009) Biochemistry 48: 7271-7278.

58. Harikumar KB, Kunnumakkara AB, Ahn KS, Anand P, Krishnan S, et al. (2009) Blood

113: 2003-2013.

59. Liang MC, Bardhan S, Pace EA, Rosman D, Beutler JA, et al. (2006) Biochem Pharmacol

71 : 634-645.

60. Sandur SK, Ichikawa H, Sethi G, Ahn KS, Aggarwal BB (2006) J Biol Chem 281: 17023- 17033.

61. Straus DS, Pascual G, Li M, Welch JS, Ricote M, et al. (2000) Proc Natl Acad Sci U S A

97: 4844-4849.

62. Wang CY, Mayo MW, Baldwin AS, Jr. (1996) Science 274: 784-787.

63. Haase R, Kirschning CJ, Sing A, Schrottner P, Fukase K, et al. (2003) J Immunol 171:

4294-4303.

64. Hsu LC, Park JM, Zhang K, Luo JL, Maeda S, et al. (2004) Nature 428: 341-345.

65. Ruckdeschel K, Harb S, Roggenkamp A, Horaef M, Zumbihl R, et al. (1998) J Exp Med

187: 1069-1079.

66. Monack DM, Mecsas J, Bouley D, Falkow S (1998) J Exp Med 188: 2127-2137.

67. Ruckdeschel K, Mannel O, Richter K, Jacobi CA, Trulzsch K, et al. (2001) J Immunol

166: 1823-1831.

68. Jones RM, Luo L, Moberg KH (2012) Cell Microbiol 14: 274-285.

69. Zhang Y, Ting AT, Marcu KB, Bliska JB (2005) The Journal of Immunology 174: 7939- 7949.

70. Zhou H, Monack DM, Kayagaki N, Wertz I, Yin J, et al. (2005) J Exp Med 202: 1327-

1332.

71. Mukherjee S, Keitany G, Li Y, Wang Y, Ball HL, et al. (2006) Science 312: 121 1-1214.

72. Fehr D, Casanova C, Liverman A, Blazkova H, Orth K, et al. (2006) Microbiology 152:

2809-2818.

73. Bhattacharjee RN, Park KS, Kumagai Y, Okada K, Yamamoto M, et al. (2006) J Biol

Chem 281 : 36897-36904.

74. Montecucco C, Papini E, Schiavo G (1994) FEBS Lett 346: 92-98.

75. Murphy JR (2011) Toxins (Basel) 3: 294-308.

76. Knust Z, Blumenthal B, Aktories K, Schmidt G (2009) Infect Immun 77: 1835-1841. 77. Rupnik M, Pabst S, von Eichel-Streiber C, Urlaub H, Soling HD (2005) Microbiology

151 : 199-208.

78. Krieglstein KG, Henschen AH, Weller U, Habermann E (1991) Eur J Biochem 202: 41-

51.

79. Pirazzini M, Rossetto O, Bolognese P, Shone C Montecucco C (2011) Cell Microbiol 13: 1731-1743.

80. Falnes PO, Ariansen S, Sandvig K, Olsnes S (2000) J Biol Chem 275: 4363-4368.

81. Morlon-Guyot J, Mere J, Bonhoure A, Beaumelle B (2009) Infect Immun 77: 3090-3099.

82. Alami Mr, Taupiac M-P, Reggio H, Bienvenue A, Beaumelle B (1998) Molecular Biology of the Cell 9: 387-402.

83. Stavrinides J, Ma W, Guttman DS (2006) PLoS Pathog 2: el04.

84. Mirold S, Ehrbar K, Weissmuller A, Prager R, Tschape H, et al. (2001) J Bacteriol 183:

2348-2358.

85. Krzywinska E, Krzywinski J, Schorey JS (2004) Microbiology 150: 1707-1712.

86. Haake DA, Suchard MA, Kelley MM, Dundoo M, Alt DP, et al. (2004) J Bacteriol 186:

2818-2828.

87. Davies RL, Campbell S, Whittam TS (2002) J Bacteriol 184: 266-277.

88. Baldo L, Lo N, Werren JH (2005) J Bacteriol 187: 5406-5418.

89. Tonello F, Naletto L, Romanello V, Dal Molin F, Montecucco C (2004) Biochem Biophys Res Commun 313: 496-502.

90. Afonso A, Lousada S, Silva J, Ellis AE, Silva MT (1998) Dis Aquat Organ 34: 27-37.

91. do Vale A, Afonso A, Silva MT (2002) Fish Shellfish Immunol 13: 183-198.

92. Sambrook J, Russel DW (2001) Molecular cloning: a laboratory manual. New York: Cold

Spring Harbor Laboratory Press.

93. Zornetta I, Brandi L, Janowiak B, Dal Molin F, Tonello F, et al. (2010) Cell Microbiol 12:

1435-1445.

94. Pinto RD, da Silva DV, Pereira PJ, dos Santos NM (2012) Fish Shellfish Immunol 32:

110-120.

95. Bertani G (2004) J Bacteriol 186: 595-600.

96. Edelhoch H (1967) Biochemistry 6: 1948-1954.

97. Pace CN, Vajdos F, Fee L, Grimsley G, Gray T (1995) Protein Sci 4: 2411-2423.

98. Falnes PO, Sandvig K (2000) Curr Opin Cell Biol 12: 407-413.

99. Sandvig K, van Deurs B (1999) FEBS Lett 452: 67-70.

100. Sandvig K, van Deurs B (2005) Gene Ther 12: 865-872.

101. Sandvig K, van Deurs B (1996) Physiol Rev 76: 949-966.

102. Watson P, Spooner RA (2006) Adv Drug Deliv Rev 58: 1581-1596. kelstein A (1994) Toxicology 87: 29-41.

mid A, Benz R, Just I, Aktories K (1994) J Biol Chem 269: 16706-16711.

rer J, Kuban J, Heine K, Rupps G, Kaiser E, et al. (2010) Cell Microbiol 12: 233-247.semann T, Jank T, Gerhard R, Maier E, Just I, et al. (2006) J Biol Chem 281 : 10808- 10815.

Claims

1. The use of a protein to translocate a molecule into a mammalian cell, wherein the protein comprises a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: l.

2. The use of claim 1, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

3. The use of claim 1 or claim 2, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

4. The use of any preceding claim, wherein the translocation domain has at least 95% sequence identity to SEQ ID NO: 1.

5. The use of claim 1, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

6. The use of any preceding claim, wherein the molecule comprises the translocating protein and a functional component.

7. The use of claim 6, wherein the functional component is a protein which inhibits NF-KB.

8. The use of claim 7, wherein the functional component is the catalytic domain of AIP56.

9. The use of claim 8, wherein the molecule is AIP56.

10. The use of any one of claims 1 to 8, wherein the molecule is not AIP56.

11. A molecule for use in treating a condition associated with increased NF-KB activity in mammalian cells, the molecule comprising an NF-κΒ inhibitor and a protein which comprises a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1.

12. The molecule for use according to claim 11, wherein the condition associated with an increased NF-κΒ activity is selected from: inflammatory conditions such as infection and autoimmune disease; and neoplastic disease such as tumours, e.g. cancer.

13. The molecule for use according to claim 1 1 or claim 12, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

14. The molecule for use according to any one of claims 11 to 13, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

15. The molecule for use according to any one of claims 1 1 to 14, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

16. The molecule for use according to any one of claims 11 to 15, wherein the NF- KB inhibitor is the catalytic domain of AIP56.

17. The molecule for use according to any one of claims 11 to 16, wherein the molecule is AIP56.

18. The molecule for use according to any one of claims 11 to 16, wherein the molecule is not AIP56.

19. Use of a molecule as a biological tool to inhibit NF-κΒ in mammalian cells, wherein the molecule comprises an NF-κΒ inhibitor and a protein comprising a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1.

20. The use of claim 19, wherein the mammalian cells are isolated mammalian cells.

21. The use according to claim 19 or claim 20, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

22. The use according to any one of claims 19 to 21, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

23. The use according to any one of claims 19 to 22, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

24. The use according to any one of claims 19 to 23, wherein the NF-κΒ inhibitor is the catalytic domain of AIP56.

25. The use according to any one of claims 19 to 24, wherein the molecule is AIP56.

26. The use according to any one of claims 19 to 24, wherein the molecule is not AIP56.

27. A method of introducing a molecule into a mammalian cell, the method comprising administering the molecule to a mammalian cell, wherein the molecule comprises a protein that comprises a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1, wherein the molecule enters the mammalian cell.

28. The method of claim 27, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

29. The method of claim 27 or claim 28, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

30. The method of any one of claims 27 to 29, wherein the translocation domain has at least 95% sequence identity to SEQ ID NO: 1.

31. The method of any one of claims 27 to 30, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

32. The method of any one of claims 27 to 31, wherein the molecule comprises the translocating protein and a functional component.

33. The method of claim 32, wherein the functional component is a protein which inhibits NF-KB.

34. The method of claim 33, wherein the functional component is the catalytic domain of AIP56.

35. The method of any one of claims 27 to 34, wherein the molecule is AIP56.

36. The method of any one of claims 27 to 34, wherein the molecule is not AIP56.

37. A method of treating a condition associated with increased NF-κΒ activity in mammalian cells, the method comprising administering to a subject in need of such treatment a molecule comprising an NF-κΒ inhibitor and a protein comprising a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 1.

38. The method of claim 37, wherein the condition associated with an increased NF- KB activity is selected from: inflammatory conditions such as infection and autoimmune disease; and neoplastic disease such as tumours, e.g. cancer.

39. The method of claim 37 or claim 38, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

40. The method of any one of claims 37 to 39, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

41. The method of any one of claims 37 to 40, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

42. The method of any one of claims 37 to 41, wherein the NF-κΒ inhibitor is the catalytic domain of AIP56.

43. The method of any one of claims 37 to 42, wherein the molecule is AIP56.

44. The method of any one of claims 37 to 43, wherein the molecule is not AIP56.

45. A molecule comprising a protein which comprises a translocation domain that is able to translocate the protein into the cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 1, and wherein the protein is not an AIP56 toxin.

46. The molecule of claim 45, wherein the translocation domain has at least 70% sequence identity to SEQ ID NO: 1.

47. The molecule of claim 45 or claim 46, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

48. The molecule of any one of claims 45 to 47, wherein the translocation domain has at least 95% sequence identity to SEQ ID NO: 1.

49. The molecule of any one of claims 45 to 48, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 2.

50. The molecule of any one of claims 45 to 49, wherein the molecule comprises the translocating protein and a functional component.

51. The molecule of claim 50, wherein the functional component is a protein which inhibits NF-KB.

52. A molecule for inhibiting the activity of NF-κΒ, wherein the molecule comprises a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, wherein the molecule is not AIP56.

53. The molecule of claim 52, wherein the protein has an amino acid sequence having at least 70% sequence identity to SEQ ID NO: 3.

54. The molecule of claim 52 or claim 53, wherein the protein has an amino acid sequence having at least 90% sequence identity to SEQ ID NO: 3.

55. The molecule of any one of claims 52 to 54, wherein the protein consists essentially of an amino acid sequence having at least 90% identity to SEQ ID NO: 3.

56. The molecule of any one of claims 52 to 55, wherein the molecule comprises a delivery system to translocate the molecule into the cytosol of cells.

57. The molecule of claim 56, wherein the delivery system is a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 70% sequence identity to SEQ ID NO: 1.

58. The molecule of claim 56, wherein the delivery system is the protective antigen (PA) domain of anthrax toxin, e.g. PA83.

59. A molecule for use in treating a condition associated with increased NF-KB activity, the molecule comprising a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO : 3.

60. The molecule for use according to claim 59, wherein the protein has an amino acid sequence having at least 70% sequence identity to SEQ ID NO: 3.

61. The molecule for use according to claim 59 or claim 60, wherein the protein has an amino acid sequence having at least 90% sequence identity to SEQ ID NO: 3.

62. The molecule for use according to any one of claims 59 to 61, wherein the protein consists essentially of an amino acid sequence having at least 90% identity to SEQ ID NO: 3.

63. The molecule for use according to any one of claims 59 to 62, wherein the molecule comprises a delivery system to translocate the molecule into the cytosol of cells.

64. The molecule for use according to claim 63, wherein the delivery system is a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 70% sequence identity to SEQ ID NO: 1.

65. The molecule for use according to claim 63, wherein the delivery system is the protective antigen (PA) domain of anthrax toxin, e.g. PA83.

66. The molecule for use according to any one of claims 59 to 65, wherein the condition associated with an increased NF-κΒ activity is selected from: inflammatory conditions such as infection and autoimmune disease; and neoplastic disease such as tumours, e.g. cancer.

67. The molecule for use according to any one of claims 59 to 66, wherein the molecule is not ΑΓΡ56.

68. A nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, wherein the molecule is not AIP56.

69. The nucleic acid of claim 68, wherein the nucleotide sequence encodes for a protein which comprises the amino acid sequence of SEQ ID NO: 3.

70. The nucleic acid of claim 68 or claim 69, wherein the nucleotide sequence has at least 80% sequence identity to SEQ ID NO: 5.

71. The nucleic acid of any one of claims 68 to 70, wherein the nucleotide sequence has the sequence of SEQ ID NO: 5.

72. A vector comprising the nucleic acid of any one of claims 68 to 71.

73. A cell containing the vector of claim 72 or the nucleic acid of any one of claims 68 to 71.

74. A nucleic acid for use in treating a condition associated with increased NF-KB activity, the nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3.

75. A method of treating a condition associated with increased NF-κΒ, the method comprising administering to a subject in need of such treatment a molecule comprising a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3, or a nucleic acid comprising a nucleotide sequence encoding for a protein which comprises an amino acid sequence having at least 45% identity to SEQ ID NO: 3..

76. The method of claim 75, wherein the protein has an amino acid sequence having at least 70% sequence identity to SEQ ID NO: 3.

77. The method of claim 75 or claim 76, wherein the protein has an amino acid sequence having at least 90% sequence identity to SEQ ID NO: 3.

78. The method of any one of claims 75 to 77, wherein the protein consists essentially of an amino acid sequence having at least 90% identity to SEQ ID NO: 3.

79. The method of any one of claims 75 to 78, wherein the molecule comprises a delivery system to translocate the molecule into the cytosol of cells.

80. The method of claim 79, wherein the delivery system is a translocation domain which is able to translocate the molecule into the cytosol of mammalian cells and which has at least 70% sequence identity to SEQ ID NO: 1.

81. The method of claim 79, wherein the delivery system is the protective antigen (PA) domain of anthrax toxin, e.g. PA83.

82. The method of any one of claims 75 to 81, wherein the condition associated with an increased NF-κΒ activity is selected from: inflammatory conditions such as infection and autoimmune disease; and neoplastic disease such as tumours, e.g. cancer.

83. The method of any one of claims 75 to 82, wherein the molecule is not AIP56.

84. A nucleic acid comprising a nucleotide sequence encoding for a protein which comprises a translocation domain that is able to translocate the protein into the cell cytosol of mammalian cells, wherein the translocation domain has at least 40% sequence identity to SEQ ID NO: 1, and wherein the protein is not an AIP56 toxin.

85. The nucleic acid of claim 84, wherein the translocation domain has at least 90% sequence identity to SEQ ID NO: 1.

86. The nucleic acid of claim 84 or 85, wherein the protein comprises the translocation domain and a functional component such as an antigen.

87. The nucleic acid of any one of claims 84 to 86, wherein the nucleotide sequence comprises a sequence having at least 90% sequence identity to SEQ ID NO: 8.

88. A vector comprising the nucleic acid of any one of claims 84 to 87.

89. A cell containing the vector of claim 88 or the nucleic acid of any one of claims 84 to 87.

90. Use of a protein to target a molecule to a mammalian cell, wherein the protein comprises a receptor binding domain which is able to bind to the cell membrane of mammalian cells and which has at least 40% sequence identity to SEQ ID NO: 9.

91. The use according to claim 90, wherein the receptor binding domain has at least 90% sequence identity to SEQ ID NO: 9.

92. The use according to claim 90 or 91, wherein the molecule comprises the protein and a functional component such as an antigen.

93. A molecule for use in treating a condition associated with increased NF-KB activity in mammalian cells, the molecule comprising an NF- Β inhibitor and a protein which comprises a receptor binding domain which is able to target the molecule to mammalian cell membranes and which has at least 40% sequence identity to SEQ ID NO: 9.

94. A method of targeting a molecule to a mammalian cell, the method comprising administering the molecule to a mammalian cell, wherein the molecule comprises a protein that comprises a receptor binding domain which is able to target the molecule to the mammalian cell and which has at least 40% sequence identity to SEQ ID NO: 9, wherein the molecule binds to the mammalian cell.

95. A molecule comprising a protein which comprises a receptor binding domain that is able to target the protein to mammalian cells, wherein the receptor binding domain has at least 40% sequence identity to SEQ ID NO: 9, and wherein the protein is not an AIP56 toxin.

96. A nucleic acid comprising a nucleotide sequence encoding for a protein which comprises a receptor binding domain that is able to target the protein to mammalian cells, wherein the receptor binding domain has at least 40%» sequence identity to SEQ ID NO: 9, and wherein the protein is not an AIP56 toxin.

97. A nucleic acid according to claim 96, wherein the nucleotide sequence comprises a sequence having at least 90% sequence identity to SEQ ID NO: 10.

98. A vector comprising the nucleic acid of claim 96 or 97.

99. A cell containing the vector of claim 98 or the nucleic acid of claim 96 or 97.