WO2017118764A1

WO2017118764A1 - Novel approaches for the in vivo and in vitro visualization of dying cells

Info

Publication number: WO2017118764A1
Application number: PCT/EP2017/050350
Authority: WO
Inventors: Thomas Brocker; Jan KRANICH
Original assignee: Thomas Brocker; Kranich Jan
Priority date: 2016-01-07
Filing date: 2017-01-09
Publication date: 2017-07-13

Abstract

The present invention relates to an Mfge8 fusion protein comprising a (poly)peptide of interest covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and, optionally, an RGD-binding activity. The invention further relates to a composition comprising the Mfge8 fusion protein of the invention, as well as to the Mfge8 fusion protein of the invention or the composition comprising same for use in medicine, in particular for use in the treatment of cancer and for use in the diagnosis or treatment of a disease associated with cell death or with a defect in phagocytosis. Furthermore, the present invention relates to a method of detecting cell death, to a method of analyzing phagocytosis of dying cells, to a method of diagnosing a disease associated with a defect in phagocytosis, and to a method of determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction.

Description

Novel approaches for the in vivo and in vitro visualization of dying cells

In this specification, a number of documents including patent applications and manufacturer's manuals is cited. The disclosure of these documents, while not considered relevant for the patentability of this invention, is herewith incorporated by reference in its entirety. More specifically, all referenced documents are incorporated by reference to the same extent as if each individual document was specifically and individually indicated to be incorporated by reference.

An inevitable by-product of many physiological processes, such as inflammation, wound healing, and lymphocyte differentiation is the death of a large amount of the participating cells. Cell death can be due to apoptosis, necrosis or necroptosis. Most of these cells die by apoptosis. Arguably, one of the most important features of apoptosis is that cell membranes maintain their integrity while the cell is dying. Thus, intracellular components remain shielded from the immune system, which could otherwise induce severe inflammation and autoimmunity (Nagata, S., Hanayama, R. & Kawane, K. Autoimmunity and the clearance of dead cells. Cell 140, 619-630, (2007)). Apoptotic cells are cleared very rapidly in an immunological silent or even anti-inflammatory fashion by professional phagocytes, such as macrophages and immature dendritic cells, or non-professional phagocytes, such as mesenchymal cells and fibroblasts (Gardai, S. J. et al. Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte. Cell 123, 321-334, (2005)). However, if clearance is impaired or delayed, dying cells turn secondary necrotic, which is accompanied by the loss of the integrity of their cell membranes. This results in the release of intracellular components, which have the potential to trigger severe pro-inflammatory responses and can cause chronic autoimmune diseases such as systemic lupus erythematosus (SLE, lupus) (Gaipl, U. S. et al. Clearance of apoptotic cells in human SLE. Curr Dir Autoimmun 9, 173-187, (2006)).

Cell death also plays a crucial role in many other diseases. It is therefor of great interest to reliably detect and analyze dying cells. For most questions it is further important to detect cell death in very early stages, while the cell is still largely intact. Unfortunately, this detection, and the visualization and characterization of cell death in vivo is difficult (Blankenberg, F. G. In vivo detection of apoptosis. Journal of nuclear medicine: official publication, Society of Nuclear Medicine 49 Suppl 2, 81s-95s, 2008). Poulsen er a/. (Poulsen, R. H. et al. Pharmacokinetics of the phosphatidylserine tracers 99mTc-lactadherin and 99mTc-annexin V in pigs. EJNMMI Research 3, 1-1 (2013)) describe the use of technetium-99m-labeled lactadherin to analyse the kinetics of labeling PS in vivo. Similarly, Falbourg et al. (Falborg, L. et al. Biodistribution of 99mTc-HYNIC-lactadherin in mice-a potential tracer for visualizing apoptosis in vivo. Scand. J. Clin. Lab. Invest. 70, 209-216 (2010)) also describe the use of technetium-99m-labeled lactadherin as a potential tracer to visualize apoptosis in vivo. However, only a rather diffuse signal can be obtained by this method, which does not allow the analysis and monitoring of single dying cells in vivo. Histological analysis of phosphatidylserine (PS)-positive cells in different tissues is not possible, because intracellular PS is universally exposed to staining reagents in sectioned tissues. Quantitation of dying cells in homogenized organs by Annexin V staining is associated with the problem that many cells die during the organ preparation process, thus providing incorrect results. Similarly, organ preparation for downstream applications, such as e.g. flow cytometry, also damages cells and often leads to high background when staining dying cells in vitro for FACS.

Another approach, the detection of fragmented DNA by TUNEL staining, is merely possible at late stages of apoptosis and several studies indicate that most DNA fragmentation in vivo is initiated only after the dying cell has been phagocytosed (Nagata, S. DNA degradation in development and programmed cell death. Annu Rev Immunol 23, 853-875, (2005)). Finally, another method currently employed to visualize dead cells is by staining activated caspase-3 (Gown, A. M. & Willingham, M. C. Improved detection of apoptotic cells in archival paraffin sections: immunohistochemistry using antibodies to cleaved caspase 3. The journal of histochemistry and cytochemistry: official journal of the Histochemistry Society 50, 449-454 (2002)), however, there are also caspase-3 independent pathways of cell death initiation (Broker, L E., Kruyt, F. A. & Giaccone, G. Cell death independent of caspases: a review. Clinical cancer research : an official journal of the American Association for Cancer Research 11 , (2005)).

Accordingly, despite the fact that a lot of effort is currently being invested into identifying methods to detect dying cells, there is still a need to provide means and methods for the specific detection of dying cells that died in vivo or under the respective experimental conditions to be analysed, in order to increase the reliability and specificity of dying cell detection within a specific tissue or sample.

This need is addressed by the provision of the embodiments characterized in the claims.

Accordingly, the present invention relates to an Mfge8 fusion protein comprising a (poly)peptide of interest covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity. Optionally, and preferably, the Mfge8 fusion protein further has an RGD-binding activity. Thus, in accordance with this preferred embodiment, the present invention relates to an Mfge8 fusion protein comprising a (poly)peptide of interest covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and an RGD-binding activity.

The term "fusion protein", as used herein, relates to a construct in which (poly)peptides are fused together that naturally occur as separate molecules. Such a fusion is achieved by the joining of two or more nucleic acid molecules that originally coded for separate molecules, i.e. the fusion protein of the invention is produced by recombinant DNA technology, i.e. genetical engineering. Translation of this fusion nucleic acid molecule results in a fusion protein, with functional properties derived from each of the original molecules. Suitable methods for creating such fusion nucleic acid molecules by recombinant DNA technology as well as suitable vectors for expression of the fusion proteins are well established in the art. In accordance with the present invention, the term "fusion protein" does not encompass conjugate proteins obtained by chemically linking two (or more) separate (poly)peptides, i.e. by expressing the separate (poly)peptides and, after their expression, chemically linking them to form a conjugate. The term "comprising", as used herein, denotes that further components and/or steps can be included in addition to the specifically recited components and/or steps. In those embodiments where the Mfge8 fusion protein includes more than the recited molecules, additional molecules, preferably at the C- or N-terminus, may include for example sequences introduced for purification, typically peptide sequences that confer on the resulting Mfge8 fusion protein an affinity to certain chromatography column materials. Typical examples for such sequences include, without being limiting, oligohistidine-tags, Sirep-tags, FLAG-tags, glutathione S- transferase, maltose-binding protein or the albumin-binding domain of protein G. However, this term also encompasses that the claimed subject-matter consists of exactly the recited components and/or steps. In accordance with the present invention, the fusion protein is an Mfge8 fusion protein, i.e. the fusion protein comprises at least an Mfge8 protein. The term "Mfge8" refers to the milk fat globule-EGF factor 8 protein, a secreted protein found in vertebrates, including mammals as well as birds. It is a membrane glycoprotein that possesses a phosphatidylserine (PS)-binding activity as well as an RGD-binding activity and, thus, promotes phagocytosis of dying cells. Mfge8 has also been implicated in wound healing, autoimmune disease, and cancer. Mfge8 can be further processed to form a smaller cleavage product, medin, which comprises the major protein component of aortic medial amyloid (AMA). Alternative splicing has been reported to result in multiple transcript variants.

The term "Mfge8 protein", as used herein, includes the full-length Mfge8 protein, as well as variants thereof. The full-length human Mfge8 protein is shown in SEQ ID NO:1 and the full length mouse Mfge8 protein is shown in SEQ ID NO:6. Variants of the Mfge8 protein include e.g. an Mfge8 mutein or an isoform of the Mfge8 protein, wherein said variants have to have a phosphatidylserine (PS)-binding activity. More preferably, said variants have a phosphatidylserine (PS)-binding activity and an RGD-binding activity.

As used herein, the term "mutein" refers to a protein having an amino acid sequence that differs from the amino acid sequence of a naturally occurring Mfge8 protein. Said difference in the amino acid sequence can e.g. be due to a substitution, an addition, an inversion, an insertion and/or a deletion.

The term "substitution" in accordance with the present invention, refers to the replacement of a particular amino acid with another amino acid. Thus, the total number of amino acids remains the same.

The term "inversion" in accordance with the present invention refers to a kind of mutation in which the order of the amino acids in a section of the amino acid sequence is reversed with respect to the remainder of the amino acid sequence. The term "insertion" in accordance with the present invention refers to the addition of one or more amino acids to an amino acid sequence, wherein the addition is not to the C-terminal or N-terminal end of the amino acid sequence.

The term "deletion" as used in accordance with the present invention refers to the loss of nucleotides. Preferable, the deletion variant of Mfge8 is a C-terminal fragment of Mfge8. The term "C-terminal fragment" refers to any fragment of the Mfge8 protein, in which one or several amino acids are missing at the N-terminal end. It is well known in the art that functional polypeptides may be cleaved to yield fragments with unaltered or substantially unaltered function. Said number of amino acids to be removed may be one, two, three, four, five, six, seven, eight, nine, ten, 15, 20, 25, 30, 40, 50, 60, 70, or 80 or more than 80. Any other number between one and 80 is also deliberately envisaged. In particular, removals of amino acids which preserve sequences and boundaries of any conserved functional domain(s) or subsequences in the sequence of the Mfge8 protein are particularly envisaged. Means and methods for determining such domains are well known in the art and include experimental and bioinformatic means. Experimental means include the systematic generation of deletion mutants and their assessment in assays for activity known in the art and as described in the Examples enclosed herewith. Bioinformatic means include database searches. Suitable databases included protein sequence databases. In this case a multiple sequence alignment of significant hits is indicative of domain boundaries, wherein the domain(s) is/are comprised of the/those subsequences exhibiting an elevated level of sequence conservation as compared to the remainder of the sequence. Further suitable databases include databases of statistical models of conserved protein domains such as Pfam maintained by the Sanger Institute, UK (www.sanger.ac.uk Software/Pfam). In accordance with the present invention, Mfge8 "isoforms" are variants that are formed by alternative splicing. Exemplary isoforms in accordance with the present invention are shown in SEQ ID NOs: 2 and 3 for human Mfge8 protein as well as in SEQ ID NO:5 for mouse Mfge8 protein. The nucleic acid sequences encoding such Mfge8 variants in accordance with the present invention can be prepared by known methods, such as e.g. by site-directed mutagenesis techniques, high throughput mutagenesis, DNA shuffling, or protein evolution techniques.

The fusion protein of the present invention has to have at least a phosphatidylserine (PS)- binding activity. Preferably, the fusion protein of the present invention further has an RGD- binding activity. Accordingly, when an Mfge8 variant is employed in accordance with the present invention, said variant has to have at least a phosphatidylserine (PS)-binding activity, and preferably an RGD-binding activity, in order to impart this ability onto the fusion protein of the present invention.

Whereas the term "phosphatidylserine (PS)-binding activity" relates to the ability of a molecule to bind to phosphatidylserine, the term "RGD-binding activity" refers to the presence of an Arginine-Glycine-Aspartic acid (RGD) motif in the sequence of the Mfge8 protein, which enables the binding to integrins.

Binding to phosphatidylserine is ensured by the presence of the respective PS-binding domain, which in the mouse is for example present at positions 306 to 463 of the Mfge8 protein shown in SEQ ID NO:5 or at positions 269 to 426 of the Mfge8 protein shown in SEQ ID NO:6. The respective mouse PS-binding domain is shown in SEQ ID NO:7. In the human Mfge8 protein, the respective PS-binding domain has not been mapped, but is presumed to be located in the region spanning from position 68 to 387 of the human Mfge8 protein shown in SEQ ID NO:1 , or spanning from position 68 to 335 of the human Mfge8 protein shown in SEQ ID NO:2 or 3.

The RGD motif in the mouse Mfge8 protein is for example located at positions 87 to 89 in the Mfge8 protein shown in SEQ ID N0.5 or 6. In humans, the RGD motif is for example located at positions 47 to 49 in the Mfge8 protein shown in SEQ ID NO:1 and at positions 46 to 48 in the Mfge8 protein shown in SEQ ID NO:3.

By carrying out sequence alignments using well known tools, such as e.g. NCBI blast, the corresponding regions can be easily identified in other Mfge8 proteins or variants thereof.

Whether a variant of the Mfge8 protein, or the Mfge8 fusion protein of the invention, has a phosphatidylserine (PS)-binding activity and/or an RGD-binding activity can be determined by methods well known in the art. For example, one method is the quantification of the amount of integrin expressing cells that bind to the respective Mfge8 protein coated onto microtiter plates using the CyQUANT cell proliferation assay kit as described in Hanayama, R. et al. (Identification of a factor that links apoptotic cells to phagocytes. Nature 417, 182-187 (2002)). In accordance with the present invention, a molecule is considered to have a phosphatidylserine (PS)-binding activity or an RGD-binding activity if it has at least 10%, such as e.g. at least 25%, at least 50%, at least 75%, more preferably at least 80%, such as at least 90%, more preferably at least 95% and most preferably at least 98% of the phosphatidylserine (PS)-binding activity or the RGD-binding activity as observed for the human Mfge8 protein shown in SEQ ID NO:1 or the mouse Mfge8 protein shown in SEQ ID NO:6 when tested by the same method of determining the respective activity, preferably when tested using the assay described by Hanayama et al., Nature, 2002. Preferably, the above recited values are relative percentages of phosphatidylserine (PS)-binding activity or the RGD-binding activity as observed for the human Mfge8 protein shown in SEQ ID NO:1.

In accordance with the present invention, the Mfge8 fusion protein further comprising a (poly)peptide of interest.

The term "(poly)peptide" in accordance with the present invention describes a group of molecules which comprises the group of peptides, consisting of up to 30 amino acids, as well as the group of polypeptides (also referred to herein as proteins), consisting of more than 30 amino acids. Also encompassed by the term "(poly)peptide" are (poly)peptides that form dimers, trimers and higher oligomers, i.e. that consist of more than one (poly)peptide molecule. (Poly)peptide molecules forming such dimers, trimers etc. may be identical or non- identical. The corresponding higher order structures are, consequently, termed homo- or heterodimers, homo- or heterotrimers etc. The term "(poly)peptide" also refers to naturally modified (poly)peptides wherein the modification is effected e.g. by glycosylation, acetylation, phosphorylation and the like. Such modifications are well known in the art.

The (poly)peptide of interest is not particularly limited. For example, the (poly)peptide of interest can be a molecule for detection of the Mfge8 protein, such as e.g. a reporter protein, an enzyme or a recognition sequence for enzymatic modification, or it can be a therapeutic molecule. Preferred (poly)peptides of interest are discussed in more detail herein below.

In accordance with the present invention, the (poly)peptide of interest is covalently bound to the Mfge8 protein. Covalent bonds are typically formed by the sharing of electron pairs between atoms. In accordance with the present invention, a covalent bond is formed between the molecules of the fusion protein by use of a peptide linker of at least 1 amino acid in length, thereby forming a peptide bond between one of the termini of the Mfge8 protein and one of the termini of the (poly)peptide of interest. In a preferred embodiment of the Mfge8 fusion protein of the invention, the (poly)peptide of interest is bound to the C-terminus of the Mfge8 protein.

Preferably, the linker is 1 to 100 amino acids in length. More preferably, the linker is 5 to 50 amino acids in length, such as e.g. 10 to 30 amino acids in length and even more preferably, the linker is 12 to 25 amino acids in length. Even more preferably, the linker is 15 to 20 amino acids in length, such as 15 to 17 amino acids, and most preferably, the linker has a length of 15 amino acids or a length of 17 amino acids. It is preferred that the linker molecule is a linear or a helical linker, even more preferably the linker is a helical linker. It is further preferred that the linker is a flexible linker using e.g. the amino acids glycine and/or serine. In a particularly preferred embodiment of the invention, between 50% and 100%, particularly between 60% and 100%, particularly between 70% and 100%, particularly between 80% and 100%, particularly between 90% and 100%, and especially 100% of the amino acid residues of the linker molecule are glycine and serine residues, preferably forming an alpha-helix structure.

The length and sequence of a suitable linker depends on the composition of the respective Mfge8 fusion protein. Methods to test the suitability of different linkers are well known in the art and include e.g. the comparison of the protein stability or the production yield of the Mfge8 fusion protein of the invention to fusion proteins comprising different linkers as well as to the respective Mfge8 protein without a further molecule fused thereto. Furthermore, and in accordance with the present invention, it has to be ensured that the linker does not interfere with the phosphatidylserine (PS)-binding activity and/or the RGD-binding activity of the resulting Mfge8 fusion protein of the invention. For testing whether a linker fulfils this requirement, the phosphatidylserine (PS)-binding activity and/or the RGD-binding activity of the Mfge8 fusion protein of the invention can be determined as described herein above. A linker is considered to not interfere with the phosphatidylserine (PS)-binding activity and/or the RGD-binding activity of the Mfge8 fusion protein of the invention if the above described preferred amounts of activity as compared to the Mfge8 protein of SEQ ID NO:1 or SEQ ID NO:6, preferably SEQ ID NO:1 , are maintained in the fusion protein. Preferably, a linker is considered to not interfere with the phosphatidylserine (PS)-binding activity and/or the RGD- binding activity of the Mfge8 fusion protein of the invention if at least 80%, more preferably at least 90% and most preferably at least 95%, of the respective activity as compared to the Mfge8 protein of SEQ ID NO:1 or SEQ ID NO:6, preferably SEQ ID NO:1 , is maintained in the fusion protein. As discussed herein above, cell death plays a crucial role in many diseases and, thus, it is of great interest to reliably detect and analyze dying cells, i.e. cells that die due to programmed cell death, such as due to apoptosis, necrosis or necroptosis. Moreover, it is desirable to detect, visualize and characterize cell death in very early stages, while the cell is still largely intact. However, this is a difficult task to achieve, especially in vivo.

Lactadherin has recently been described as a suitable detection agent for PS exposure and as an alternative for the use of Annexin, the marker that is currently most commonly employed in the art for detecting dying cells, and which inhibits phagocytosis of dying cells (Krahling, S., Callahan, M. K., Williamson, P. & Schlegel, R. A. Exposure of phosphatidylserine is a general feature in the phagocytosis of apoptotic lymphocytes by macrophages. Cell Death Differ 6, 183-189 (1999)). For example, Hu et al. (2008) Braz. J. Med. Bio. Res. 41 (9), pp. 750-757 describe the labeling of apoptotic adherent Tca81 3 and ACC-2 cells with lactadherin and conclude that lactadherin permits the detection of PS exposure earlier than Annexin. Also in WO2005/005954, lactadherin is described as a staining tool for detecting and visualizing apoptotic cells by binding to phosphatidylserine. In both works, however, a label is chemically added to lactadherin, which does not result in the covalent binding of a (poly)peptide of interest to lactadherin via a linker. Moreover, whereas both works focus on PS-binding activity, no regard has been given to maintaining an RGD-binding activity. WO2012/173762 describes a further approach for labeling apoptotic cells based on lactadherin, namely the use of cyclic lactadherin peptides (cLac), which are small and do not require any co-factors to bind PS. However, also these cyclic lactadherin peptides are devoid of any RGD-binding activity and, thus, cannot promote phagocytosis of the dying cells.

In accordance with the present invention, on the other hand, Mfge8 fusion proteins are provided that not only enable the binding and, thus, detection of dying cells at a very early stage in vivo and in vitro, but that also facilitate and induce the engulfment of dying cells by phagocytes, such as e.g. macrophages.

Further, due to the capability of lactadherin to bind to PS, the Mfge8 fusion proteins provided herein also enable binding and, thus, detection of extracellular vesicles having PS on the extravesicular surface in vivo and in vitro as evidenced in the example section (cf. examples 12 and 13) and described herein below in detail. Thus, all definitions, embodiments and statements relating to the Mfge8 fusion protein in the context of dying cells apply mutatis mutandis also to the aspect of extracellular vesicles having PS on the extravesicular surface.

The Mfge8 fusion protein provided herein can be used to stain and detect dying cells and extracellular vesicles in vitro and in vivo by e.g. flow cytometry, imaging flow cytometry, fluorescent microscopy, 2-photon microscopy or electron microscopy using for example the fusion protein Mfge8-miniSOG.

Even more, intravenous or intra-tissue administration of Mfge8 fusion proteins allows the labeling of dying cells in vivo. In contrast to state of the art methods, which often have a significant amount of false positive results due to a substantial amount of unspecific cell death that occurs during organ preparation, the labeling method now available based on the Mfge8 fusion proteins of the invention provides an increased specificity. This is because the dying cells are labeled in vivo prior to organ preparation and, thus, only true in vivo cell death events are labeled, as shown in Example 8 below, which confirms that there is a large discrepancy between the in vivo and in vitro labeling.

There is a plethora of in vivo uses that can now be carried out due to the provision of the Mfge8 fusion proteins of the invention. For example, the administration of the Mfge8 fusion proteins of the invention to specific organs or tissues can help to identify mechanisms of cell death in different pathologies, like muscular dystrophy, where, up to now it was very difficult to identify mechanisms of cell death. Furthermore, the specific labeling of dying cells in vivo can also be used to isolate and purify these dying cells, such as e.g. apoptotic cells, by FACS or MACS and process them for downstream analyses, like proteomics or transcriptomics, which could help identifying new factors that are important in the regulation of cell death. Further, analyzing cell death using in vivo administered Mfge8 fusion proteins of the invention, such as luciferase fused proteins, can be used to quantify cell death non-invasively in life organs and tissues using bioluminescence whole body imagers. Alternatively, by using Mfge8 fusion proteins comprising radionuclides or radioisotopes, analyses in nuclear medicine become possible, which can provide information and diagnostics about internal anatomy, organ function and treatment progress. This enables new approaches of assessing the potency of cytostatic agents and cancer drugs.

Moreover, in certain preferred embodiments of the present invention, the Mfge8 fusion protein of the invention possesses an RGD-binding activity. Mfge8 not only binds to dying cells via its PS-binding activity, it also promotes phagocytosis of these cells via its RGD motif, which is in stark contrast to AnnexinV. Waehrens et al. (2009), J. Histochem & Cytochem 57(10):907-914 describe experiments that show that the attachment of chemically labeled lactadherin to PS is not inhibited by soluble RGD peptide, thus indicating that the observed lactadherin binding to apoptotic cells is unrelated to the integrin receptor-mediated cell adhesion mediated by the RGD motif. However, Waehrens et al. (2009) did not investigate whether dying cells having chemically-labeled lactadherin bound can be phagocytosed and, most importantly, subsequently detected. In accordance with the present invention it is shown that the Mfge8 fusion protein of the invention can not only bind to dying cells, but that said labeled cells can efficiently be phagocytosed. The Mfge8 fusion proteins of these preferred embodiments of the present invention are thus a valuable tool to visualize and quantify phagocytosis in vitro and in vivo and can facilitate the identification of defects in phagocytosis, which are associated with autoimmune diseases, such as e.g. systemic lupus erythematosus or rheumatoid arthritis. Moreover, in contrast to the state of the art marker AnnexinV, the binding of Mfge8 to its ligand PS does not require the presence of Ca ⁺, making its use more practical and faster. This also means that cells sensitive to the presence of high Ca²⁺ concentrations can now be isolated more gently in the absence of calcium and the need for extensive changes of buffers and washing steps is obviated.

Finally, the Mfge8 fusion protein of the invention is capable of inducing rapid B cell responses, when the (poly)peptide of interest serves as an antigen. As shown in Example 1 below, an antigen (here EGFP) fused to Mfge8 was capable of triggering higher antigen-specific immunoglobulin titers after primary immunization as compared to the antigen on its own. Without wishing to be bound by theory, it is hypothesized that the fusion to Mfge8 enables the binding of the antigen to follicular dendritic cells (FDCs) in naive mice, thereby making the antigen available in the B cell follicle, where it can activate naive B cells in an immune- complex-independent manner. In more detail, distribution of intravenously administered Mfge8 fusion protein was assessed by immunofluorescence (cf. example 14). 12h after injection, Mfge8-EGFP was only visible in the B-cell follicle where it accumulated on CD21/35+ follicular dendritic cells (FDCs) (Fig 15). Mfge8 is also known as the FDC marker FDC-M1 and histological stainings using anti-Mfge8 antibodies stain the FDC network (Kranich, J. et al. Follicular dendritic cells control engulfment of apoptotic bodies by secreting Mfge8. J Exp Med 205, 1293-1302, (2008)). The results show that FDCs extensively bind Mfge8. Given that B cells extensively capture Mfge8+ EVs (Figure 17), it is plausible that B cells transport EVs from the MZ into the B-cell follicle, where they hand them over to FDCs, as they do with immune complexes (ICs) (Kranich, J. & Krautler, N. J. How Follicular Dendritic Cells Shape the B-Cell Antigenome. Frontiers in immunology 7, 225, doi:10.3389/fimmu.2016.00225 (2016)). FDCs are known to capture and store native antigen for presentation to B cells during the germinal center reaction. Subsequently, the Mfge8 fusion protein is recognized by tingible body macrophages through integrins which recognize the RGD-motif within the Mfge8 protein sequence (Hanayama, R. et al., Nature 417, 182-187, 2002 (doi:10.1038/417182a [pii]); and Hanayama, R. et al., Science 304, 1147-1150, 2004 (doi:10.1126/science.1094359304/5674/1147 [pii])). Tingible body macrophages take up the Mfge8 fusion proteins bound on follicular dendritic cells (see, e.g., Figure 10B). Tingible body macrophages have important immunoregulatory functions (Smith, J. P. et al., Dev. Immunol. 6, 285-294 (1998). Therefore, it can be expected that this is an important process during the B cell response. In more detail, it is shown herein that the Mfge8-EGFP fusion proteins are first bound by FDCs and then accumulate inside tingible body macrophages (Fig. 10B). Consequently, if the RGD motif is lacking, this effect will be strongly reduced. As such, it can plausibly be expected that this has negative consequences on the efficacy of the immune response raised against the antigenic portion of the Mfge8 fusion protein. This is, because being the only macrophages in the germinal center, they are expected to present processed antigen that they have taken up from FDCs to T follicular helper cells. During the germinal center reaction, B cells undergo affinity maturation and class switching. For this it is crucial that they continuously see native antigen presented by FDCs. But they also need to repeatedly receive survival signals from T follicular helper cells (Victora and Nussenzweig, Annu Rev Immunol, 2012;30:429-57). It is thought that tingible body macrophages are the main source of processed antigen that is presented via MHC-II molecules to T follicular helper cells. Therefore, it is plausible that tingible body macrophages keep T follicular helper cells activated, as long as sufficient antigen is present on FDCs. If antigen associated with EVs cannot be recognized by tingible body macrophages, because the RGD motif of Mfge8 protein part of the Mfge8 fusion protein is lacking, a reduced immune response against the antigenic portion of the Mfge8 fusion protein can be expected. This can, e.g., be the consequence of impaired activation of T follicular helper cells, diminished germinal center reaction, a lower antibody production and/or decreased affinity maturation.

Consequently, the Mfge8 protein of the fusion protein of the invention has RGD-binding activity, if employed to induce a B cell and/or T cell response, e.g., when used as a vaccine as described below in detail.

These findings can be exploited to design novel vaccines, e.g. by fusing viral or tumor antigens to Mfge8. This novel strategy is particularly beneficial for pathogens that evade an efficient antibody response, as is the case for example for some viruses. In such cases, antibody-independent accumulation of antigen on FDCs might facilitate the activation of B cells and the production of neutralizing antibodies.

Consequently, the invention also relates to a vaccine comprising an Mfge8 fusion protein as defined herein, namely comprising the (poly)peptide of interest as an antigen covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and an RGD-binding activity. The vaccine according to the invention is a preferred embodiment of the composition of the invention comprising at least one Mfge8 fusion protein according to the invention described herein below. In other words, the composition according to the invention is in this case a vaccine, wherein the (poly)peptide of interest of the at least one Mfge8 fusion protein serves as an antigen and wherein the at least one fge8 fusion protein has a phosphatidylserine (PS)-binding activity and an RGD-binding activity. It is evident that such a composition, namely the vaccine, is for use in medicine. Further, the invention also relates to the use of the Mfge8 fusion protein of the invention, wherein the (poly)peptide of interest serves as an antigen, as a vaccine; in other words, the use of an fge8 fusion protein comprising the (poly)peptide of interest as an antigen covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and an RGD-binding activity, as a vaccine is also part of the invention. As such, the invention also relates to a method for vaccination comprising the administration of a vaccine comprising or consisting of an Mfge8 fusion protein comprising the (poly)peptide of interest as an antigen covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and, optionally, an RGD-binding activity to the subject to be vaccinated. It is understood that the antigen is suitable to elicit an immune response resulting in the generation of antibodies against the (poly)peptide of interest that serves as an antigen of the Mfge8 fusion body. The immune response triggered by the Mfge8 fusion protein results in immunization against the (poly)peptide of interest serving as antigen. Immunization relates to the process of stimulating and sensitizing the immune system towards the antigen(s) within the vaccine. As such, the immunization against the (poly)peptide of interest serving as antigen manifests itself in at least one of the following, e.g., in the case of immunization against an infectious agent or a tumor: preventing infection of the immunized subject with an infectious agent comprising or consisting of the full or partial (poly)peptide of interest serving as antigen or preventing development of a tumor associated with the full or partial (poly)peptide of interest serving as antigen, modifying or limiting the infection or development of a tumor, aiding, improving, enhancing or stimulating the recovery of said individual from infection or a tumor and generating immunological memory that will prevent or limit a subsequent infection with the (poly)peptide of interest serving as antigen or will prevent or limit a subsequent development of a tumor associated with the full or partial (poly)peptide of interest serving as antigen. The presence of any of said effects can be tested for and detected by routine methods known to the person skilled in the art. As evident from the foregoing, the invention encompasses therapeutic as well as prophylactic vaccines. It is known in the art how to develop vaccines that can be used as prophylactic and/or as therapeutic vaccines. Besides comprising the Mfge8 fusion protein, a vaccine in accordance with the invention may further comprise pharmaceutically acceptable carriers which include any carrier that does not itself elicit an adverse reaction harmful to the subject receiving the vaccine. Suitable carriers are typically large, slowly metabolized macromolecules such as proteins, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids, amino acid copolymers and lipid aggregates such as, e.g. oil droplets or liposomes. Further suitable pharmaceutically acceptable carriers are well-known in the art. Also, the vaccine may comprise one or more adjuvants. The term "adjuvant" is used according to its well-known meaning in connection with vaccines. Specifically, an adjuvant is an immunological agent that modifies, preferably enhances, the effect of a vaccine while having few, if any, desired immunogenic effects on the immune system when given per se. In accordance with the present invention it is defined as any substance that is capable of accelerating, prolonging or enhancing antigen-specific immune responses when used in combination with antigens. Suitable adjuvants can be inorganic adjuvants such as, e.g., aluminium salts (e.g., aluminium phosphate, aluminium hydroxide), monophosphoryl lipid A, or organic adjuvants such as squalene or oil-based adjuvants, as well as virosomes. Preferably, the adjuvant is aluminium hydroxide. Further, the vaccine may comprise diluents such as, e.g. water, saline, glycerol, ethanol etc. Furthermore, substances necessary for formulation purposes may be comprised in a vaccine such as emulsifying agents and/or pH buffering substances. Any combination of the above-mentioned substances may be part of a vaccine in accordance with the invention as needed. The vaccine may only comprise or consist of Mfge8 fusion proteins with only one kind of (poly)peptide serving as antigen or it can comprise different Mfge8 fusion proteins with different kinds of (poly)peptides serving as antigens. In both cases, further antigens not part of Mfge8 fusion proteins of the invention may be comprised in the vaccine. The antigens comprised in the vaccine can be formulated to result in immunization against more than one antigen; in other words, also combination vaccines are contemplated by the invention. Antigens are described herein below and include, e.g. viral proteins, such as e.g. HIV-gp120 or Influenza-hemagglutinin (HA), proteins of bacterial or protozoan pathogens, as well as tumor antigens to induce, e.g., anti-tumor immune responses. Preferably, the vaccine is a tumor vaccine or a viral vaccine. In this case, the sequence of the (poly)peptide of interest that serves as an antigen consists of or comprises preferably a viral sequence capable of inducing an anti-viral response or the sequence of a tumor antigen capable of inducing an anti-tumor immune response. Also preferred is that a vaccine in accordance with the invention does not contain exosomes. When referring to the Mfge8 fusion protein comprising the (poly)peptide of interest as an antigen covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and an RGD-binding activity in the context of the vaccine, the use as a vaccine or in a method of vaccination, this means at the same time also that the vaccine can comprise a nucleic acid molecule encoding said Mfge8 fusion protein. In other terms, the embodiments relating to vaccines herein include DNA vaccines. DNA vaccination is a technique that is well-known in the art and described, e.g., in Ferraro et al., Clin Infect Dis. 2011 Aug 1; 53(3): 296-302 and Kutzler, M. A. & Weiner, D. B. DNA vaccines: ready for prime time? Nature reviews. Genetics 9, 776-788, 2008 (doi:10. 038/nrg2432), as well as below.

In a preferred embodiment of the Mfge8 fusion protein of the invention, the (poly)peptide of interest is (i) a reporter protein, preferably a fluorescent or a bioluminescent reporter protein; or (ii) a recognition sequence for enzymatic modification, preferably a recognition sequence for a biotin ligase or a recognition sequence for a sortase. The term "reporter protein", as used herein, relates to a protein that has the ability to generate a detectable signal, preferably a signal detectable from outside of a cell, organ, tissue or even individual. Preferably, expression of the reporter protein provides an optically detectable signal. The term "optically detectable signal" refers to a light signal that can be detected by a photodetector, such as a light microscope, a spectrophotometer, a fluorescent microscope, a fluorescent sample reader, or a fluorescence activated cell sorter, 3D tomographer, a camera, and the like.

Non-limiting examples of reporter proteins that provide an optically detectable signal include proteins that are capable of being excited by a particular wavelength of light and that emit another wavelength of light, which can be detected by the researcher. Exemplary optically detectable proteins include, without being limiting, fluorescent proteins and bioluminescent proteins. Alternatively, the protein can be an enzyme that catalyzes a reaction which results in a light signal. In a preferred embodiment of the Mfge8 fusion protein of the invention, the reporter protein is a fluorescent or a bioluminescent reporter protein.

Examples of fluorescent proteins in accordance with the present invention include, without being limiting, green fluorescent protein (GFP), enhanced green fluorescent protein (EGFP), red fluorescent protein (RFP) such as, for example, iRFP, mCherry, monomeric DsRed, monomeric As Red, mStrawberry, yellow fluorescent protein (YFP), flavoproteins such as, for example, miniSOG (mini Singlet Oxygen Generator)⁷, or cyan fluorescent protein (CFP). Non- limiting examples of bioluminescent proteins include luciferase proteins such as, for example, bacterial luciferase (\uxAB), photinus luciferase, ren/7/a luciferase or firefly luciferase. In a particularly preferred embodiment of the invention, the Mfge8 fusion protein is Mfge8-EGFP, Mfge8-m Cherry or Mfge8-renilla luciferase. In accordance with this preferred embodiment of the fusion protein of the invention, fusion proteins comprising a detectable moiety are provided. Such fusion proteins are particularly suitable for the labeling and detection of dying cells in vivo and in vitro by methods well known in the art. Because the fusion proteins of the present invention not only label dying cells, but are also phagocytosed, they provide a valuable tool to visualize and quantify phagocytosis in vitro and in vivo and can therefore facilitate the identification of phagocytosis defects, which can lead to autoimmune diseases, such as systemic lupus erythematosus. In an alternative preferred embodiment, the (poly)peptide of interest can be a recognition sequence for enzymatic modification. Such recognition sequences for enzymatic modifications are well known in the art and have been described e.g. in Voloshchuk et al. (Voloshchuk, N., Liang, D. & Liang, J. F. Sortase A Mediated Protein Modifications and Peptide Conjugations. Curr Drug Discov Technol (2015)). In a preferred embodiment of the Mfge8 fusion protein of the invention, the recognition sequence for enzymatic modification is a recognition sequence for a biotin ligase or a recognition sequence for a sortase.

A suitable biotin ligase is for example the enzyme "£. coli repressor of biotin biosynthesis" (BirA), which is capable of transferring biotin to a unique lysine residue in the recognition sequence within the acceptor protein. Suitable recognition sequences for BirA are well known and have been described in the art, e.g. in Fairhead and Howarth (2015) (Site-specific biotinylation of purified proteins using BirA. Methods in molecular biology 1266, 171-184, (2015)). For example, when the Mfge8 protein is fused to a BirA recognition sequence, the BirA enzyme then catalyzes the site-specific addition of a biotin. This would, for example, allow the detection of the Mfge8 fusion protein, and therefore of any dying cells or extracellular vesicles bound by said fusion protein, via biotin-based detection systems. Evidently, these systems can also be used for isolation and/or purification purposes of dying cells or extracellular vesicles.

Sortase enzymes are prokaryotic enzymes that have been used extensively for protein engineering and antibody modifications. The use of sortase and its recognition sequence is thus well known in the art and has been described, e.g. in Alt ef al. (Angew. Chem. Int. Ed. 2015, 54, 7515-7519) who describe an approach in which proteins, here antibodies, are modified via sortase-mediated bioconjugation and are subsequently available for the selective incorporation of fluorescent dyes or radiotracers into the target protein. The use of sortase and its recognition sequence also allows for site-specific addition of biotin. This allows, for example and in the context of the present invention, the detection of the Mfge8 fusion protein, and therefore of dying cells or extracellular vesicles bound by said fusion protein, via biotin-based detection systems. Evidently, biotin-based detection systems can also be used/modified for the purpose of isolation and/or purification purposes of dying cells or extracellular vesicles.

Applied to the present invention, the Mfge8 protein can for example be fused to e.g. a sortase A recognition sequence. The sortase A then catalyzes the site-specific addition of e.g. radioisotopes, such as 18F, 64Cu, 68Ga or 89Zr.

Quantitation of cell death in humans is of great interest in monitoring the efficacy of tumor therapy and for monitoring cell death associated with other pathologies, such as e.g. myocardial infarction. The Mfge8 fusion protein of the invention labeled with e.g. a sortase A recognition sequence can be employed to detect dying cells, or phagocytic cells that have taken up dying cells, by e.g. positron emission tomography (PET). For monitoring the efficacy of tumor therapy or cell death in other pathologies, e.g. myocardial infarction, preferably 18F or 64Cu are added to Mfge8 via sortase A.

Because also the treatment of cancer is of great interest, the labeled Mfge8 fusion protein can also be employed to destroy or at least weaken malfunctioning cells, such as e.g. dying cells of a tumor, wherein the radioisotope that generates the radiation can be localized in the target cells via the Mfge8 protein moiety. Preferably 177Lu or 90Y are added to Mfge8 via sortase A for this application.

It will be appreciated that in accordance with this particular embodiment of targeting dying cells of e.g. tumors, the Mfge8 fusion protein of the present invention comprises an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity, wherein the presence of an RGD-binding activity is not necessary and, thus, optional.

In accordance with this preferred embodiment of the fusion protein of the invention, fusion proteins are provided that comprise as the (poly)peptide of interest a recognition sequence for enzymatic modification, which serves as an "adapter", i.e. it is suitable to connect compounds of interest at a later stage to the Mfge8 fusion protein of the invention. The advantage of such an adapter-Mfge8 fusion protein is that only one fusion protein is generated that can be modified into a plethora of different fusion proteins, depending on the intended use. Further, the coupling of compounds that normally cannot be chemically or genetically connected to Mfge8 can be achieved by this adapter technique. In another preferred embodiment of the Mfge8 fusion protein of the invention, the (poly)peptide of interest is an enzyme, preferably an enzyme selected from the group consisting of thymidine kinase, cytosine deaminase, cytochrome P450 (CYP) and nitroreductase. In accordance with this embodiment of the fusion protein of the invention, an enzyme is fused to the Mfge8 protein, thereby enabling the targeting of said enzyme to dying cells and phagocytic cells that take up dying cells. Preferably, the enzyme is an enzyme capable of catalyzing the conversion of a non-toxic substance into a toxic substance. Non-limiting examples of such enzymes include thymidine kinase, cytosine deaminase, cytochrome P450 (CYP), inducible caspase 9 and nitroreductase. Thymidine kinase (TK) and its use as a pro-drug activating enzyme is well known in the art. Briefly, prodrugs, such as e.g. acyclovir or ganciclovir, which in themselves are not toxic, are converted into toxic drugs in the presence of viral thymidine kinase, thereby leading to cell death of the target cells. Preferably, the thymidine kinase is Herpes simplex virus thyimidine kinase (HSV-TK).

Cytosine deaminase and its use as a pro-drug activating enzyme is well known in the art. Briefly, cytosine deaminase converts the non-toxic compound 5-fluorocytosine (5-FC) into the highly toxic compound 5-fluorouracil (5-FU), thereby killing the target cell.

Cytochrome P450 (CYP) and its use as a pro-drug activating enzyme is also well known in the art. CYP converts non-toxic prodrugs, such as cyclophosphamide (CPA) or ifosfamide (IFO) into toxic drugs that induce cell death of the target cell by DNA-alkylation. Further, inducible caspase 9 and its use as a pro-drug activating enzyme is also known in the art. Briefly, inducible caspase 9 is fused to a human FK506 binding protein (FKBP). This allows dimerization after addition of a synthetic dimerizing drug. Then, inducible caspase 9 becomes activated through dimerization and induces apoptosis in the target cell. Also nitroreductase and its use as a pro-drug activating enzyme are well known in the art. Nitroreductase, e.g. from E. coli converts a substrate, which is not toxic for mammalian cells, such as CB1954 ([5-(aziridin-1-yl)-2,4-dinitrobenz-amide]) into a cytotoxic agent, which is toxic for mammalian cells, thereby leading to cell death of the target cell. By targeting such enzymes - via the fusion protein of the present invention - to dying cells and, consequently, to the phagocytes that take up said dying cells, it is possible to specifically induce cell death in the respective phagocytes, thus enabling the depletion of phagocytes especially in vivo. This approach thus provides a valuable and versatile research tool for studying the effects of depletion of phagocytes.

In another preferred embodiment of the Mfge8 fusion protein of the invention, the (poly)peptide of interest is a therapeutic (poly)peptide.

The term "therapeutic (poly)peptide" is used herein to describe any (poly)peptide that has a preventive or curative effect on a disease condition, preferably a disease condition associated with a defect in the regulation of cell death and/or phagocytosis. For example, a therapeutic (poly)peptide can be a (poly)peptide that replaces a (poly)peptide that is deficient or abnormal in the respective disease; it can be a (poly)peptide that interferes with a molecule or organism that is associated with or causative for the disease; it can be an enzyme that activates an otherwise inactive prodrug; it can be a (poly)peptide that activates the immune system of the host, for example by acting as an antigen or triggering an anti-inflammatory effect; or it can be a cytotoxic drug or radioactive compound that directly damages the target cells.

In a more preferred embodiment of the Mfge8 fusion protein of the invention, the therapeutic (poly)peptide is a Dnase, an antigen or a radioactive compound.

Aberrant self-DNA recognition has been proven critical for the initiation of excessive immune responses in autoimmune diseases such as e.g. lupus erythematosus and lupus patients often have low levels of Dnase. Deoxyribonucleases (Dnases) are enzymes that catalyze the hydrolytic cleavage of phosphodiester linkages in the DNA backbone, thus degrading DNA. As such, Dnases remove superfluous DNA and other cellular leftovers and dying cells. Macrophages use Dnase2a to digest DNA from phagocytosed cells inside phagolysosomes. Impaired Dnase2a function leads to accumulation of DNA inside macrophages, triggering an inflammatory response. Thus, the provision of Mfge8 fusion proteins of the present invention wherein the therapeutic (poly)peptide is Dnase provides a tool for increasing the amount of Dnase in such patients at the site of dying cells, thereby providing a promising tool for the prevention and/or treatment of autoimmune diseases associated with the aberrant recognition of self-DNA.

The choice of a Dnase is not particularly limited and thus, well known Dnases such as e.g. Dnasel , Caspase-activated Dnase (CAD) and Dnase2a (see e.g. Kawane and Nagata. Nucleases in programmed cell death. Methods Enzymol. 2008 442:271-87) can be employed. Preferably, the Dnase is Dnase2a. Human Dnase2a is represented in e.g. NCBI database accession number NP_001366.1 (March 15, 2015) and mouse Dnase2a is represented in e.g. NCBI database accession number NP_034192.1 (February 15, 2015)

Further preferred is that the therapeutic (poly)peptide is an antigen. Antigens in accordance with the present invention include e.g. viral proteins, such as e.g. HIV-gp120 or Influenza- hemagglutinin (HA), proteins of bacterial or protozoan pathogens, as well as tumor antigens to induce e.g. an anti-tumor immune responses.

As discussed herein above, the Mfge8 fusion protein of the invention was found to be capable of inducing rapid B-cell responses, when the (poly)peptide of interest serves as an antigen (Example 11 ). Accordingly, Mfge8 fusion proteins can serve as novel vaccines. Such vaccines are particularly beneficial for pathogens that evade an efficient antibody response, such as e.g. some viruses.

Finally, the therapeutic (poly)peptide may also preferably be a radioactive compound.

The radioactive compound (also referred to herein as radioisotope) may be any isotope suitable for short-range radiotherapy; preferably the radioisotope is a beta emitter.

One well known therapeutic radioisotope, which is a strong beta emitter, is lutetium-177 (also referred to herein as 177Lu). This is prepared from ytterbium-176 which is irradiated to become Yb-177 which decays rapidly to Lu-177. Yttrium-90 (also referred to herein as 90Y) is another well known radioisotope suitable for the treatment of cancer, particularly non- Hodgkin's lymphoma and liver cancer, and it is also used for arthritis treatment. Lu-177 and Y- 90 are currently considered the main agents for therapy based on radionuclides. Further non- limiting examples of suitable radioisotopes include iodine-131 , samarium-153, phosphorus-32, rhenium-188, and boron-10 as well as fiuorine-18, copper-64, gallium-68 or zirconium-89. As discussed herein above, targeting dying cells in humans is of great interest in the treatment of tumors. By employing an Mfge8 fusion protein according to the invention wherein the therapeutic (poly)peptide is a radioactive compound, dying tumor cells can be targeted and the surrounding tumorous tissue can be treated by radiotherapy.

Also encompassed by the present invention is the use of the Mfge8 fusion protein of the invention for labeling and detecting dying cells in vitro and in vivo, including for tracking dying cells on a single cell level in vivo, as described in more detail herein below. The term "dying cells", as used throughout the present description, relates to cells that die due to programmed cell death, such as due to apoptosis, due to necrosis as well as due to necroptosis. Preferably, the dying cells are apoptotic cells. In a particularly preferred embodiment of the use of the fge8 fusion protein of the invention for labeling and detecting dying cells in vitro and in vivo, the Mfge8 fusion protein is Mfge8-EGFP or Mfge8-mCherry. In another preferred embodiment of this use of the Mfge8 fusion protein of the invention, the detection of dying cells is carried out by cytometry, imaging flow cytometry, or microscopy. For microscopy, in particular for electron microscopy, Mfge8-miniSOG is a particularly preferred Mfge8 fusion protein of the invention. It is further preferred that the detection is carried out in a buffer solution, which is free or essentially free of Ca²⁺. As discussed above, the detection of dying cells can be of particular importance for e.g. detecting tumor cells and for assessing the severity of myocardial infarction, but also for monitoring the efficacy of treatment of these diseases. The present invention further relates to a nucleic acid molecule encoding the Mfge8 fusion protein of the invention.

In accordance with the present invention, the term "nucleic acid molecule", also referred to as nucleic acid sequence or polynucleotide herein, includes DNA, such as cDNA or genomic DNA, and RNA. It is understood that the term "RNA" as used herein comprises all forms of RNA including mRNA. Both, single-strand as well as double-strand nucleic acid molecules are encompassed by this term. Further included are nucleic acid mimicking molecules known in the art such as synthetic or semi-synthetic derivatives of DNA or RNA and mixed polymers. Such nucleic acid mimicking molecules or nucleic acid derivatives according to the invention include phosphorothioate nucleic acid, phosphoramidate nucleic acid, 2'-0-methoxyethyl ribonucleic acid, morpholino nucleic acid, hexitol nucleic acid (HNA), peptide nucleic acid (PNA) and locked nucleic acid (LNA) (see Braasch, D.A. & Corey, D.R. [2001] Chem. Biol. 8:1-7). PNA a synthetic DNA-mimic with an amide backbone in place of the sugar-phosphate backbone of DNA or RNA. As a consequence, certain components of DNA, such as phosphorus, phosphorus oxides, or deoxyribose derivatives, are not present in PNAs. LNA is an RNA derivative in which the ribose ring is constrained by a methylene linkage between the 2'-oxygen and the 4'-carbon. They may contain additional non-natural or derivatised nucleotide bases, as will be readily appreciated by those skilled in the art. The nucleic acid molecules of the invention can e.g. be synthesized by standard chemical synthesis methods or produced semi-synthetically, i.e. by combining chemical synthesis and isolation from natural sources. Ligation of the coding sequences to transcriptional regulatory elements and/or to other amino acid encoding sequences can be carried out using established methods, such as restriction digests, ligations and molecular cloning.

Representative nucleic acid molecules encoding the Mfge8 fusion proteins described in the appended examples are provided herein as SEQ ID NOs: 4 and 15 to 20 (encoding the Mfge8 fusion proteins shown in SEQ ID NOs: 8 to 14). The Mfge8 isoform used for these exemplary mouse fusion proteins is shown in SEQ ID NO:6 and the Mfge8 isoform used for these exemplary human fusion proteins is shown in SEQ ID NO:1. Further, the present invention also relates to a vector comprising the nucleic acid molecule of the invention.

Usually, the vector is a plasmid, cosmid, virus, bacteriophage or another vector used conventionally e.g. in genetic engineering. Preferably, the vector is a plasmid, more preferably a plasmid based on pcDNA3.1 as employed in the appended examples.

Alternative vectors including, without being limiting, vectors, such as pQE-12, the pUC-series, pBluescript (Stratagene), the pET-series of expression vectors (Novagen) or pCRTOPO (Invitrogen), lambda gt11 , pJOE, the pBBR1-MCS series, pJB861 , pBSMuL, pBC2, pUCPKS, pTACTI and vectors compatible with expression in mammalian cells like E-027 pCAG Kosak- Cherry (L45a) vector system, pREP (Invitrogen), pCEP4 (Invitrogen), pMCI neo (Stratagene), pXT1 (Stratagene), pSG5 (Stratagene), EBO-pSV2neo, pBPV-1 , pdBPVMMTneo, pRSVgpt, pRSVneo, pSV2-dhfr, plZD35, Okayama-Berg cDNA expression vector pcDV1 (Pharmacia), pRc/CMV, pcDNAI , pcDNA3 (Invitrogen), pcDNA3.1 , pSPORT (GIBCO BRL), pGEMHE (Promega), pLXIN, pSIR (Clontech), pIRES-EGFP (Clontech), pEAK-10 (Edge Biosystems) pTriEx-Hygro (Novagen) and pCINeo (Promega). Non-limiting examples for plasmid vectors suitable for Pichia pastoris comprise the plasmids pA0815, pPIC9K and pPIC3.5K (all Invitrogen). Another vector suitable for expressing proteins in Xenopus embryos, zebrafish embryos as well as a wide variety of mammalian and avian cells is the multipurpose expression vector pCS2+. Further suitable vectors include lentiviral and adenoviral expression vectors, such as e.g. lenti-X-vectors or adeno-X-vectors (Clontech).

Generally, vectors can contain one or more origins of replication (ori) and inheritance systems for cloning or expression, one or more markers for selection in the host, e.g., antibiotic resistance, and one or more expression cassettes. In addition, the coding sequences comprised in the vector can be ligated to transcriptional regulatory elements and/or to other amino acid encoding sequences using established methods. Such regulatory sequences are well known to those skilled in the art and include, without being limiting, regulatory sequences ensuring the initiation of transcription, internal ribosomal entry sites (IRES) (Owens, G.C. et al.

[2001] Proc. Natl. Acad. Sci. U.S.A. 98:1471-1476) and optionally regulatory elements ensuring termination of transcription, which are to be included downstream of the nucleic acid molecules of the invention, and elements ensuring stabilization of the transcript. Non-limiting examples for such regulatory elements ensuring the initiation of transcription comprise promoters, a translation initiation codon, enhancers, and/or insulators. Further examples of regulatory sequences include Kozak sequences and intervening sequences flanked by donor and acceptor sites for RNA splicing, nucleotide sequences encoding secretion signals or, depending on the expression system used, signal sequences capable of directing the expressed protein to a cellular compartment or to the culture medium. The vectors may also contain an additional expressible polynucleotide coding for one or more chaperones to facilitate correct protein folding. Suitable bacterial expression hosts comprise e. g. E. coli strains derived from JM83, W31 10, KS272, TG1 , BL21 (such as BL21 (DE3), BL21(DE3)PlysS, BL21(DE3)RIL, BL21 (DE3)PRARE) or Rosetta. For vector modification, PCR amplification and ligation techniques, see Sambrook & Russel [2001] (Cold Spring Harbor Laboratory, NY).

Additional examples of suitable origins of replication include, for example, the full length ColE1 , the SV40 viral and the M13 origins of replication, while additional examples of suitable promoters include, without being limiting, the cytomegalovirus (CMV) promoter, SV40- promoter, RSV-promoter (Rous sarcome virus), the lacZ promoter, chicken β-actin promoter, CAG-promoter (a combination of chicken β-actin promoter and cytomegalovirus immediate- early enhancer), the gaM O promoter, human elongation factor 1 a-promoter, AOX1 promoter, GAL1 promoter CaM-kinase promoter, the lac, trp or tac promoter, the T7 or T5 promoter, the lacUV5 promoter, the Autographa californica multiple nuclear polyhedrosis virus (AcMNPV) polyhedral promoter or a globin intron in mammalian and other animal cells. One example of an enhancer is e.g. the SV40-enhancer. Non-limiting additional examples for regulatory elements ensuring transcription termination include the SV40-poly-A site, the tk-poly-A site or the AcMNPV polyhedral polyadenylation signals. Further non-limiting examples of selectable markers include dhfr, gpt, neomycin, hygromycin, blasticidin or geneticin. Preferably, the vector of the present invention is an expression vector. An expression vector according to this invention is capable of directing the replication and the expression of the nucleic acid molecule of the invention and, accordingly, of the Mfge8 fusion proteins of the present invention encoded thereby. The nucleic acid molecules and/or vectors of the invention as described herein above may be designed for introduction into cells by e.g. non chemical methods (electroporation, sonoporation, optical transfection, gene electrotransfer, hydrodynamic delivery or naturally occurring transformation upon contacting cells with the nucleic acid molecule of the invention), chemical based methods (calcium phosphate, liposomes, DEAE-dextrane, polyethylenimine, nucleofection), particle-based methods (gene gun, magnetofection, impalefection) phage vector-based methods and viral methods. For example, expression vectors derived from viruses such as retroviruses, vaccinia virus, adeno-associated virus, herpes viruses, Semliki Forest Virus or bovine papilloma virus, may be used for delivery of the nucleic acid molecules into targeted cell population. Additionally, baculoviral systems can also be used as vector in eukaryotic expression system for the nucleic acid molecules of the invention.

Preferably, the nucleic acid molecules and/or vectors of the invention are designed for stable transfection of HEK293 or CHO, preferably CHO-DG44, cells by calcium phosphate-, polyethylenimine- or lipofectamine-transfection (Pham, P.L. et al. [2006] Mol. Biotechnol. 34:225-237; Geisse, S. & Voedisch, B. [2012] Methods Mol. Biol. 899:203-219; Hacker, D.L. et al. [2013] Protein Expr. Purif. 92:67-76).

The present invention further relates to a host cell or a non-human host transformed with the nucleic acid molecule or the vector of the invention.

It will be appreciated that the term "host cell or a non-human host transformed with the nucleic acid molecule or the vector of the invention", in accordance with the present invention, relates to a host cell or a non-human host that comprises, and preferably expresses, the nucleic acid molecule or the vector of invention.

Suitable prokaryotic hosts comprise e.g. bacteria of the species Escherichia, Corynebacterium (glutamicum), Pseudomonas (fluorescens), Lactobacillus, Streptomyces, Salmonella or Bacillus.

Typical mammalian host cells include, HEK293, Hela, H9, Per.C6 and Jurkat cells, mouse NIH3T3, NS0 and C127 cells, COS 1 , COS 7 and CV1 , quail QC1-3 cells, mouse L cells, mouse sarcoma cells, Bowes melanoma cells and Chinese hamster ovary (CHO) cells. Most preferred mammalian host cells in accordance with the present invention are HEK293 cells or CHO, preferably CHO-DG44 cells. HEK293 cells as well as suitable media and cell culture conditions are well known and have been described in the appended examples, in particular Example 1. The host cells in accordance with this embodiment may e.g. be employed to produce large amounts of the Mfge8 fusion proteins of the present invention.

Also within the scope of the present invention are primary mammalian cells or cell lines. Primary cells are cells which are directly obtained from an organism. Suitable primary cells are, for example, mouse embryonic fibroblasts (MEF), mouse primary hepatocytes, cardiomyocytes and neuronal cells as well as mouse muscle stem cells (satellite cells), human dermal and pulmonary fibroblasts, human epithelial cells (nasal, tracheal, renal, placental, intestinal, bronchial epithelial cells), human secretory cells (from salivary, sebaceous and sweat glands), human endocrine cells (thyroid cells), human adipose cells, human smooth muscle cells, human skeletal muscle cells, human leucocytes such as B-cells, T-cells, NK- cells or dendritic cells and stable, immortalized cell lines derived thereof (for example hTERT or oncogene immortalized cells). Appropriate culture media and conditions for the above described host cells are known in the art.

Other suitable eukaryotic host cells are e.g. chicken cells, such as e.g. DT40 cells, or yeasts such as Saccharomyces cerevisiae, Pichia pastoris, Schizosaccharomyces pombe and Kluyveromyces lactis. Insect cells suitable for expression are e.g. Drosophila S2, Drosophila Kc, Spodoptera Sf9 and Sf21 or Trichoplusia Hi5 cells. Suitable zebrafish cell lines include, without being limiting, ZFL, SJD or ZF4.

Appropriate culture media and conditions for the above described host cells are known in the art.

The "non-human host", in accordance with the present invention, can be any non-human animal of interest. Non-limiting examples of animals include mammals, such as e.g. rodents, dogs, felides, non-human primates, rabbits, pigs, and ruminants, in particular cattle; avians such as e.g. chickens, turkeys, pheasants, ducks, geese, quails, ostriches, and emus; as well as fish, such as zebrafish.

All of the mammals, avians and fish described herein are well known to the skilled person and are taxonomically defined in accordance with the prior art and the common general knowledge of the skilled person.

Non-limiting examples of "rodents" are mice, rats, squirrels, chipmunks, gophers, porcupines, beavers, hamsters, gerbils, guinea pigs, degus, chinchillas, prairie dogs, and groundhogs. Preferably, the rodents are mice or rats.

Non-limiting examples of "dogs" include members of the subspecies canis lupus familiaris as well as wolves, foxes, jackals, and coyotes.

Non-limiting examples of "felides" include members of the two subfamilies: the pantherinae, including lions, tigers, jaguars and leopards and the felinae, including cougars, cheetahs, servals, lynxes, caracals, ocelots and domestic cats. The term "primates", as used herein, refers to all monkeys including for example cercopithecoid (old world monkey) or platyrrhine (new world monkey) as well as lemurs, tarsiers, apes and marmosets (Callithrix jacchus). Examples of "ruminants" include, without being limiting, cattle, goats, sheep, giraffes, bisons, mooses, elks, yaks, water buffalos, deer, camels, alpacas, llamas, antelopes, pronghoms, and nilgais. Preferably, the ruminants are selected from the group consisting of cattle, goats and sheep. Most preferably, the ruminants are cattle. A number of different strategies are known in the art for providing non-human host animals carrying (and preferably expressing) the Mfge8 fusion protein of the present invention. Such strategies include, without being limiting, the introduction of the nucleic acid molecule encoding the Mfge8 fusion protein of the invention, or a vector comprising same, as (a) transgene(s) into the genome, by homologous recombination (HR) techniques for targeted gene modifications or by the use of gene trapping or of transposon-mediated mutagenesis. Particularly preferred is the use of technologies, such as the CRISPR/Cas system, zink finger nucleases, TAL-nuclease fusion proteins (TALENs) as well as fusion proteins comprising a Ralstonia solanacearum TALE-like protein (RTL) effector protein and a non-specific cleavage domain of a restriction nuclease (RALEN) for gene modifications.

Preferably, the nucleic acid sequence encoding the Mfge8 fusion protein of the invention is inserted into the genome of a recipient non-human host animal by homologous recombination or transgenesis. Further preferred is that the nucleic acid sequence encoding the Mfge8 fusion protein of the invention is inserted into the genome of the recipient non-human host animal such that said nucleic acid sequence is under the control of endogenous regulatory sequences present in the genome of the non-human host animal. Alternatively, the nucleic acid sequence encoding the Mfge8 fusion protein of the invention can be inserted into the genome of the non-human host animal together with regulatory sequences required to ensure their expression. Such sequences are well known in the art. Such animal models expressing the Mfge8 fusion protein of the invention are particularly useful for the study of cell death and phagocytosis in live animals without the need of injecting or otherwise introducing the Mfge8 fusion protein of the invention into said animal.

The present invention also relates to a method for the production of an Mfge8 fusion protein of the invention, the method comprising culturing the host cell of the invention under suitable conditions and isolating the Mfge8 fusion protein(s) produced. In accordance with this embodiment, the vector present in the host of the invention is either an expression vector, or the vector mediates the stable integration of the nucleic acid molecule encoding the Mfge8 fusion protein of the present invention into the genome of the host cell in such a manner that expression of the protein is ensured. Means and methods for selecting a host cell in which the nucleic acid molecule encoding the Mfge8 fusion protein of the present invention has been successfully introduced such that expression of the protein is ensured are well known in the art.

Suitable conditions for culturing prokaryotic or eukaryotic host cells are also well known in the art. For example, HEK293 cells can be cultured in serum-free medium, typically at a temperature of about 37°C. To increase the yield and the solubility of the expression product, the medium can be buffered or supplemented with suitable additives known to enhance or facilitate both. In those cases where an inducible promoter controls the nucleic acid molecule of the invention in the vector present in the host cell, expression of the polypeptide can be induced by addition of an appropriate inducing agent.

Depending on the cell type and its specific requirements, mammalian cell culture can also be carried out in e.g. RPMI, Williams' E or DMEM medium containing 10% (v/v) FCS, 2 mM L- glutamine and 100 U/ml penicillin/streptomycin. The cells can be kept e.g. at 37°C or at 41 °C for DT40 chicken cells, in a 5% C0₂, water-saturated atmosphere. A suitable medium for insect cell culture is e.g. TNM + 10% FCS, SF900 or HyClone SFX-lnsect medium. Insect cells are usually grown at 27°C as adhesion or suspension cultures. Suitable expression protocols for eukaryotic or vertebrate cells are well known to the skilled person and can be retrieved e.g. from Sambrook, J & Russel, D.W. [2001] (Cold Spring Harbor Laboratory, NY).

Preferably, the method is carried out using mammalian cells, such as e.g. HEK293 cells.

Methods of isolation of the fusion protein produced comprise, without limitation, purification steps such as gel filtration (size exclusion chromatography), anion exchange chromatography, cation exchange chromatography, hydrophobic interaction chromatography, high pressure liquid chromatography (HPLC), reversed phase HPLC, immunoprecipitation or affinity chromatography (preferably using a fusion-tag such as a FLAG tag or a His tag). These methods are well known in the art and have been generally described, e.g. in Sambrook, J & Russel, D.W. [2001] (Cold Spring Harbor Laboratory, NY), and are also described in the appended examples, see e.g. Example 1. In accordance with the present invention, the term "isolating the Mfge8 fusion protein produced" refers to the isolation of the Mfge8 fusion protein of the present invention.

The present invention further relates to a composition comprising at least one Mfge8 fusion protein of the invention.

As mentioned, the term "composition", as used in accordance with the present invention, relates to a composition which comprises at least the Mfge8 fusion protein of the invention. It may, optionally, comprise further molecules capable of altering the characteristics of the Mfge8 fusion protein of the invention thereby, for example, reducing, stabilizing, delaying, modulating and/or activating its function. The composition may be in solid, liquid or gaseous form and may be, inter alia, in the form of (a) powder(s), (a) tablet(s), (a) solution(s) or (an) aerosol(s). The composition may further comprise more than one type of Mfge8 fusion protein in accordance with the present invention. In that case, it is particularly preferred that the Mfge8 fusion proteins comprised in the composition comprise different (poly)peptides of interest fused to the Mfge8 protein, such as different (e.g. fluorescent) reporter molecules.

In one embodiment, the composition is a pharmaceutical composition. In accordance with the present invention, the term "pharmaceutical composition" relates to a composition for administration to a patient, preferably a human patient. The pharmaceutical composition of the invention comprises the compound(s) recited above. The pharmaceutical composition of the present invention may, optionally and additionally, comprise a pharmaceutically acceptable carrier. By "pharmaceutically acceptable carrier" is meant a non- toxic solid, semisolid or liquid filler, diluent, encapsulating material or formulation auxiliary of any type. Examples of suitable pharmaceutically acceptable carriers are well known in the art and include sodium chloride solutions, such as phosphate-buffered sodium chloride solutions, water, emulsions, such as oil/water emulsions, various types of wetting agents, sterile solutions, organic solvents etc. Such pharmaceutically acceptable carriers often contain minor amounts of additives such as substances that enhance isotonicity and chemical stability. Such materials are non-toxic to recipients at the dosages and concentrations employed, and include buffers such as phosphate, citrate, succinate, acetic acid, and other organic acids or their salts; antioxidants such as ascorbic acid; low molecular weight (less than about ten residues) peptides, e.g., polyarginine or tripeptides; proteins, such as serum albumin, gelatin, or further immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids, such as glycine, glutamic acid, aspartic acid, or arginine; monosaccharides, disaccharides, and other carbohydrates including cellulose or its derivatives, glucose, mannose, or dextrins; chelating agents such as ethylenediaminetetraacetic acid (EDTA); sugar alcohols such as mannitol or sorbitol; counterions such as sodium; and/or nonionic surfactants such as polysorbates, poloxamers, or PEG. The pharmaceutical composition may comprise further agents depending on the intended use of the pharmaceutical composition, such as e.g. antitumoral agents for use in the treatment of tumors.

Administration of pharmaceutical compositions of the invention may be effected by different ways, e.g., by intravenous, intraperitoneal, subcutaneous, intramuscular, intradermal, intranasal or intrabronchial administration. Accordingly, it is preferred that the pharmaceutically acceptable carrier is a carrier suitable for these modes of administration. Most preferably, the carrier is a solution that is isotonic with the blood or tissue fluid of the recipient. Compositions comprising such carriers can be formulated by well known conventional methods. Generally, the formulations are prepared by contacting the components of the pharmaceutical composition uniformly and intimately with liquid carriers or finely divided solid carriers or both. Then, if necessary, the product is shaped into the desired formulation.

The pharmaceutical compositions can be administered to the subject at a suitable dose. The dosage regimen will be determined by the attending physician and clinical factors. As is well known in the medical arts, dosages for any one patient depend upon many factors, including the patient's size, body surface area, age, the particular compound to be administered, sex, time and route of administration, general health, and other drugs being administered concurrently. The therapeutically effective amount for a given situation will readily be determined by routine experimentation and is within the skills and judgment of the ordinary clinician or physician. The pharmaceutical composition may be for administration once or for a regular administration over a prolonged period of time. Generally, the administration of the pharmaceutical composition, such as an Mfge8 fusion protein with e.g. Dnase, should be in the range of for example 1 pg/kg of body weight to 50 mg/kg of body weight for a single dose. However, a more preferred dosage might be in the range of 10 pg/kg to 20 mg/kg of body weight, even more preferably 100 pg/kg to 10 mg/kg of body weight and even more preferably 500 pg/kg to 5 mg/kg of body weight for a single dose. Alternatively, when radioactive Mfge8 fusion proteins are employed, smaller amounts in the range of 1 to 20 μg per injection, preferably 5 to 15 μg per injection are generally to be employed.

The components of the pharmaceutical composition to be used for therapeutic administration must be sterile. Sterility is readily accomplished for example by filtration through sterile filtration membranes (e.g., 0.2 μητι membranes).

The pharmaceutical composition may be particularly useful for the treatment of tumors and/or autoimmune diseases, as disclosed below.

In another embodiment, the composition of the invention is a diagnostic composition.

In accordance with the present invention, the term "diagnostic composition" relates to compositions for diagnosing whether a patient is suffering from a disease associated with cell death or with a defect in phagocytosis, or for diagnosing individual patients for their potential response to or curability by the pharmaceutical compositions of the invention or other pharmaceutical compositions. The diagnostic composition of the invention comprises at least an Mfge8 fusion protein according to the invention. The diagnostic composition may further comprise appropriate buffer(s) etc.

The components of the pharmaceutical or diagnostic composition can be packaged in a container or a plurality of containers, for example, sealed ampoules or vials, as an aqueous solution or as a lyophilized formulation for reconstitution. Preferably, the components of the composition are packaged with instructions for use. As an example of a lyophilized formulation, 10-ml vials are filled with 5 ml of 1 % (w/v) or 10% (w/v) aqueous solution, and the resulting mixture is lyophilized. A solution for use is prepared by reconstituting the lyophilized compound(s) using either e.g. water-for-injection for therapeutic uses or another desired solvent, e.g. a buffer, for diagnostic purposes. Preservatives and other additives may also be present such as, for example, antimicrobials, anti-oxidants, chelating agents, and inert gases and the like.

The diagnostic compositions of the present invention can be used in in vivo as well as in in vitro or ex vivo diagnostic experimental designs well known in the art. For example, the above described in vivo imaging methods using fluorescent or radioactive labels can be employed to trace the presence of the Mfge8 fusion protein in specific tissues. Furthermore, methods carried out outside the patient's body such as e.g. immunohistochemical staining of tissues or cells obtained from the patient can be employed for grading the severity of a particular cancer. Example methods are described in more detail herein below.

Further encompassed by the present invention is the Mfge8 fusion protein or the composition of the invention for use in medicine. Numerous uses of the Mfge8 fusion proteins of the invention in medicine are envisaged herein. Specifically, the use of the Mfge8 fusion proteins of the invention in any of the below described methods is envisaged herein. For example, the Mfge8 fusion proteins of the invention that carry a detectable label can be used for non-invasively providing information and diagnostics about internal anatomy, organ function as well as treatment progress, for example in cancer and myocardial infarction therapy.

Further, the Mfge8 fusion proteins according to the invention may be used to assess the potency of therapeutic agents such as e.g. cytostatic agents and cancer drugs, for example by determining the amount of dying cells prior to and after treatment with the respective agent. In this manner, diseases that are associated with or that lead to cell death caused by e.g. apoptosis, such as systemic lupus erythematosus, as well as cancer can be monitored. In addition, also the treatment of these diseases is possible by employing Mfge8 fusion proteins of the present invention comprising a therapeutic (poly)peptide. As discussed herein above, for example the use of Dnase as the therapeutic (poly)peptide aids in the removal of DNA present after the death of cells, thereby alleviating autoimmune responses to this self-DNA that are associated with various autoimmune disorders.

Non-limiting examples of cancers for which a treatment progress can be monitored in accordance with any of the embodiments of the present invention include bronchial carcinoma, cancer of the colon, head and neck cancer, pancreatic cancer, breast cancer, esophageal cancer, stomach cancer and lymphoma. Non-limiting examples of cancers for which a treatment with the Mfge8 fusion protein in accordance with any of the embodiments of the present invention is envisaged include bronchial carcinoma, cancer of the colon, head and neck cancer, pancreatic cancer, breast cancer, esophageal cancer, stomach cancer, lymphoma, prostate cancer, liver cancer and renal carcinoma.

Particularly preferred in accordance with the present invention is also the use of the Mfge8 fusion protein of the invention as a vaccine. To this end, the (poly)peptide of interest of the Mfge8 fusion protein serves as an antigen. The Mfge8 fusion protein of the invention has both a phosphatidylserine (PS)-binding activity and an RGD-binding activity. As has been shown in Example 11 below, such fusion proteins in accordance with the present invention are particularly suitable to induce B cell responses. The responses observed in primary immunization were strikingly stronger than those observed for the antigen alone and were independent of an antibody-mediated accumulation of antigen on FDCs. Also, the generation of isotype switched antibodies was accelerated. Methods for immunization are known in the art and are, for example, described in Plotkin, S. A., Orenstein, W. A. & Offit, P. A. Vaccines (Sixth Edition) (W.B. Saunders, 2013). Also, the method of DNA vaccination is known in the art (see, e.g., Kutzier, M. A. & Weiner, D. B. DNA vaccines: ready for prime time? Nature reviews. Genetics 9, 776-788, 2008 (doi:10.1038/nrg2432)) and envisioned in accordance with the invention in relation to a DNA molecule encoding the Mfge8 fusion protein of the invention for use as a vaccine.

The definitions and preferred embodiments provided herein above, in particular with regard to preferred Mfge8 proteins, preferred (poly)peptides of interest, compositions comprising the Mfge8 fusion proteins of the invention and regimens for their administration apply mutatis mutandis also to the claimed use in medicine. The present invention further relates to the Mfge8 fusion protein or the composition of the invention for use in the treatment of cancer and/or for monitoring the success of such a therapy. The definitions and preferred embodiments provided herein above with regard to the previous embodiments, in particular with regard to preferred Mfge8 proteins, compositions comprising the Mfge8 fusion proteins of the invention and regimens for their administration apply mutatis mutandis also to the claimed use in the treatment of cancer.

In accordance with this preferred embodiment, the (poly)peptide of interest is either a recognition sequence for enzymatic modification, most preferably a recognition sequence for sortase A, or a radioactive compound, most preferably 18F or 64Cu where the Mfge8 fusion protein of the invention is intended for monitoring or it is 177Lu or 90Y if the Mfge8 fusion protein of the invention is intended for treatment. As discussed herein above with regard to the therapeutic (poly)peptide being a recognition sequence for enzymatic modification or being a radioactive compound, the Mfge8 fusion protein of the present invention is particularly suitable for the treatment of cancer and for monitoring the success of such a therapy.

In the first case, i.e. where the (poly)peptide of interest is a recognition sequence for enzymatic modification, it is understood that the Mfge8 fusion protein is not employed as it is, but is initially enzymatically treated with the respective enzyme recognizing the recognition sequence included as the (poly)peptide of interest. Thus, where the recognitions sequence is for sortase A, this enzyme is employed to add a therapeutic moiety of interest to the Mfge8 fusion protein, preferably a radioactive compound. Most preferably, the radioactive compound is 18F or 64Cu where the Mfge8 fusion protein of the invention is intended for monitoring or it is 177Lu or 90Y if the Mfge8 fusion protein of the invention is intended for treatment. The aim of a tumor therapy is always the induction of cell death in tumor cells, e.g. by radiation therapy or cytostatic drugs. Therapy will induce profound cell death in the tumor, which can be quantified using PET with 18F- or 64Cu-Mfge8 or, alternatively, can be employed to further target these cells with 177Lu- or 90Y-Mfge8.

The present invention further relates to the Mfge8 fusion protein or the composition of the invention for use in the diagnosis and/or treatment of a disease associated with cell death or with a defect in phagocytosis.

The term "disease associated with cell death" relates to any diseases caused or accompanied by a dysregulation of cell death induction, for example due to alterations of the apoptotic signaling pathways. Such cell death pathways serve to ensure the normal development of tissues in healthy individuals; however, aberrant cell death can be detrimental to the individual, as the balance between cell proliferation and cell death is no longer ensured. The term "defect in phagocytosis" refers to defects in the uptake of dying cells by phagocytes, as well as to defects in the degradation of phagocytosed cells within the phagocytes, for example due to insufficient amounts of Dnase present in macrophages.

Aberrant regulation of cell death can result in autoimmunity, e.g. due to impaired deletion of auto-reactive cells. Also, aberrant phagocytosis of dying cells can result in autoimmunity. If an apoptotic cell is not cleared efficiently, it becomes secondary necrotic. As a consequence, intracellular components, such as DNA and nuclear proteins are released from the cell and can trigger immune responses and autoimmunity. Furthermore, if the degradation of dying cells inside phagocytes is impaired, undigested dying cell derived DNA can accumulate inside macrophages and trigger pro-inflammatory responses that can result in autoimmunity.

It is particularly preferred in accordance with the present invention that the term "disease associated with cell death or with a defect in phagocytosis" relates to an autoimmune disease, even more preferably to lupus erythematosus or rheumatoid disorders, such as e.g. rheumatoid arthritis.

Suitable means and methods to employ the Mfge8 fusion protein of the present invention in diagnosing and/or treating said diseases have been discussed herein above. The definitions and preferred embodiments provided herein above, in particular with regard to preferred Mfge8 proteins, preferred (poly)peptides of interest, compositions comprising the Mfge8 fusion proteins of the invention and regimens for their administration apply mutatis mutandis also to this claimed use in the diagnosis and/or treatment of a disease associated with cell death or with a defect in phagocytosis. Furthermore, the present invention relates to a method of detecting dying cells, the method comprising:

(a) labeling cells suspected to be dying by contacting the cells with the Mfge8 fusion protein of the invention; and

(b) determining the presence or absence of cells labeled with the Mfge8 fusion protein of the invention.

Also, the present invention relates to a method of detecting extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface, the method comprising:

(a) labeling extracellular vesicles suspected to have phosphatidylserine (PS) on the extravesicular surface by contacting the extracellular vesicles with the Mfge8 fusion protein of the invention; and

(b) determining the presence or absence of vesicles labeled with the Mfge8 fusion protein. For this method of detecting extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface, the RGD-binding activity of Mfge8 fusion protein of the invention can be absent.

The term "dying cells", has been defined above and relates to cells that die due to programmed cell death, such as due to apoptosis, due to necrosis or due to necroptosis. Preferably, the dying cells are apoptotic cells.

The term "extracellular vesicles" (abbreviation: EVs) is used in accordance with its well-known meaning to refer to membrane-contained vesicles that are released by cells of plants, prokaryotes and eukaryotes in an evolutionally conserved manner (Yanez-Μό et al., Journal of Extracellular Vesicles 2015, 4: 27066). Presently, three main subgroups of extracellular vesicles have been defined in the scientific literature: a) apoptotic bodies, b) cellular microparticles (also termed "microvesicles" or "ectosomes"), and c) exosomes (cf. Yanez-Μό et al., Journal of Extracellular Vesicles 2015, 4: 27066). Apoptotic bodies usually have a size ranging from about 1 to 5 μιτι diameter and are released when plasma membrane blebbing occurs during apoptosis, while the second group comprises vesicles of different sizes that pinch directly off the plasma membrane and have a size of about 100 to 1000 nm diameter. Exosomes have a size of about 30 to 100 nm diameter and are usually intraluminal vesicles (ILVs) contained in multi-vesicular bodies (MVBs), which are released to the extracellular environment upon fusion of MVBs with the plasma membrane (Colombo et al., Ann Rev Cell Dev Biol. 2014;30:255-89). All three kinds of EVs can have phosphatidylserine exposed on their extravesicular surface and can in this case be bound by the Mfge8 fusion protein of the invention. In this regard, the PS-binding protein AnnexinV was shown to bind to all three kinds of vesicles (see, e.g., Table 1 in Gyorgy, B. et al., CMLS 68, 2667-2688, 2011 (doi:10.1007/s00018-011-0689-3); Dignat-George, F. & Boulanger, C. M., Arteriosclerosis, thrombosis, and vascular biology 31 , 27-33, 201 1 (doi:10.1 161/atvbaha.1 10.218123); or Zwaal, R. F. & Schroit, A. J., Blood 89, 1 121-1132 (1997)). The phosphatidylserine is part of the EV membrane and exposed on the extravesicular surface, where it can be bound by the Mfge8 fusion protein via the PS-binding capability. As such, the method of detecting extracellular vesicles extends only to the detection of EVs that have phosphatidylserine on the extravesicular surface. Preferably, the extracellular vesicle having phosphatidylserine (PS) on the extravesicular surface to be detected are apoptotic bodies and/or cellular microparticles.

The dying cells or the extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface to be detected can be present in vitro or ex vivo, e.g. in a cell culture dish or a tissue or organ explant, or in vivo, i.e. in a living subject in the context of their natural environment. The term "subject", in accordance with the present invention, includes both humans and other animals, particularly mammals, and other organisms. Thus, the methods claimed herein are applicable to both humans and non-human animals. Preferably, the subject is a mammal, and most preferably, the subject is a human. The term "a living subject" or "a live subject", as used herein, relates to living individuals. In vivo detection of dying cells as described herein provides a far increased specificity compared to in vitro labeling of dying cells, where a large amount of unspecific cell death occurs during organ preparation resulting in false positives, as shown below in the appended examples. Also, in vivo detection of or extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface as described herein provides for an increased specificity as compared to in vitro labeling of extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface, where it can be expected that due to the large amount of unspecific cell death occuring during organ preparation also an increased amount of apoptotic bodies are generated which can ultimately result in false positives. Thus, it is particularly preferred that the method is an in vivo method.

In accordance with the detection methods of the present invention, cells (or tissue) suspected of undergoing cell death, or extracellular vesicles suspected to have phosphatidylserine (PS) on the extravesicular surface are brought into contact with the Mfge8 fusion protein of the present invention, preferably a Mfge8 fusion protein comprising a reporter protein or a recognition sequence for enzymatic modification. The terms "contacting" and "bringing into contact", as used herein, are not particularly limited and include all means of contacting cells/tissues/extracellular vesicles with the Mfge8 fusion protein, such as e.g. adding the Mfge8 fusion protein of the invention to the cell culture medium in which the cells/tissue/extracellular vesicles are kept, injecting the Mfge8 fusion protein of the invention into a live subject, for example intravenously, intraperitoneally, subcutaneously and or directly into the respective tissue or organ, as described in the appended examples.

Upon contacting cells/tissue/organs/extracellular vesicles with the Mfge8 fusion protein of the invention, the Mfge8 fusion protein binds to PS expressed on the cell surface of dying cells or, for the alternative embodiment of the detection method of the invention, to the PS on the extravesicular surface of the extracellular vesicles, provided that the extracellular have PS on the extravesicular surface. Consequently, the presence or absence of dying cells or, alternatively, the presence or absence of extracellular vesicles having phosphatidylserine on the extravesicular surface can be determined by measuring whether a labeling with the Mfge8 fusion protein occurred.

To this end, e.g. the reporter protein or the recognition sequence for enzymatic modification can be used to show the presence of Mfge8 fusion protein-labeled cells. For example, cell death or extracellular vesicles can be analyzed in cell cultures or, alternatively, non-invasively in live organs and tissues. To this end, bioluminescence whole body imagers are employed after in vivo administration of Mfge8 fusion proteins comprising e.g. luciferase or others as (poly)peptide of interest. Moreover, by employing two differently labeled Mfge8 fusion proteins in accordance with the invention, methods such as e.g. FRET (F5rster Resonance Energy Transfer) or BRET (Bioluminescence Resonance Energy Transfer) can be employed. In the first case, for example the combination of Mfge8-CFP and Mfge8-YFP fusion proteins could be employed and in the latter case, the combination of Mfge8-luciferase and Mfge8-YFP fusion protein could be employed to detect and, preferably, quantify dying cells or extracellular vesicles having PS on the extravesicular surface. Alternatively, using a Mfge8 fusion protein comprising a sortase recognition sequence, labeling with a radioactive compound such as e.g.

[18]F or [64]Cu becomes possible, which allows for an analyses in nuclear medicine by PET or other detection methods. Cell death or EV monitoring by PET using radiolabeled Mfge8 (e.g. 18F-Mfge8) can, for example, be used to quantify cell death or EVs non-invasively in vivo. This method to detect cell death can, for example, be employed to monitor the success of tumor therapies and therapeutic interventions in myocardial infarctions.

Also, exosomes are currently in clinical trials for use in anti-tumor immunotherapy. These exosomes are produced from dendritic cells (DCs) from patients. These DCs are loaded with tumor peptides and produce exosomes. These exosomes then present tumor peptides via MHC-I and MHC-II (see, e.g., Thery, C, Ostrowski, M. & Segura, E., Nature reviews. Immunology 9, 581-593, 2009 (doi:10.1038/nri2567)). Thus, labeling of exosomes with 18F- Mfge8 prior to administration could be used in monitoring and tracking these labeled exosomes to gain insights into the mechanisms of the described exosome-based anti-tumor therapy. Furthermore, extracellular vesicle detection as described herein can be used to determine the purity of exosome preparations that are used for administration (e.g., injection) to subjects. This can be achieved, for example, by using fluorescently labeled Mfge8 fusion proteins. To this end, an exosome preparation can be labeled with Mfge8 fusion protein itself labeled with the fluorescent EGFP followed by the determination of the amount of labeled particles. The higher the amount of labeled exosomes determined, the purer the preparation of exosomes.

The means and methods for detecting a signal emitted by the reporter protein or by the enzymatically modified Mfge8 fusion protein have been described herein above and preferably include cytometry, imaging flow cytometry, and microscopy. A number of microscopy techniques may be used for detecting dying cells including, without being limited thereto, fluorescent microscopy, 2-photon microscopy or electron microscopy, multiphoton microscopy, confocal microscopy, stimulated emission depletion (STED) microscopy, laser-scanning microscopy, 4Pi-microscopy, confocal laserscanning microscopy, and total internal reflection fluorescence microscopy (TIRFM).

Preferably, detection of dying cells or, alternatively, EVs is by real time live imaging of dying cells or, alternatively, EVs in a living subject, preferably an experimental animal.

In accordance with the present invention, the terms "determining", "measuring", "evaluating", "assessing" and "assaying" are used interchangeably and include determining if an element is present or absent. Any suitable form of analysis can be employed in this regard. These terms further include quantitative determinations. Assessing may be relative or absolute. "Determining the presence of" includes determining the amount of something present, as well as determining whether it is present or absent.

The methods of detection of dying cells and of detection of extracellular vesicles having phosphatidylserine on the extravesicular surface can be performed simultaneously. The skilled person is in the position to differentiate the population of dying cells and the population of EVs having PS on the extravesicular surface, e.g. as described in the example section (cf. example 13). Simultaneous detection can be advantageous, for example, whenever there is a need to increase accuracy of the detection method for detecting dying cells. As shown in the example section (cf. examples 12 and 13), PS-positive EVs were demonstrated to be attached to cells that were, however, not apoptotic, but as a result were nevertheless considered to be labeled with the Mfge8 fusion protein. Removing said population of false positive cells from the total population of dying cells can increase accuracy of the method of detecting dying cells. Example 13 describes an exemplary way how to differentiate dying cells from cells with attached EVs. Briefly, dying cells and living cells that have been stained with, for example, fluorescent Mfge8 fusion proteins can be distinguished using imaging flow cytometry. In general, image analysis tools that can analyze different features of an image can be used to detect differences between dying cells and cells with attached EVs. For example, one such feature is the ratio between the area of the whole cell (using the area of the brightfield image) and the area of the Mfge8 signal. Dying cells have a smaller ratio than cells with attached EVs. Another feature is, e.g., the minor axis intensity of the Mfge8 signal that can be used to distinguish the two subsets. Dying cells have a larger minor axis intensity than cells with attached EVs.

It is evident from the foregoing, that accuracy of each detection method can be improved by also determining the presence or absence of the population that is not to be detected from the target population in those scenarios, where both dying cells and extracellular vesicles are present at the same time in the analyte. For example, when dying cells are to be detected, one also determines in step (b) whether EVs are present and, if so, can account for the presence of the EVs when determining the presence or absence of dying cells labeled with the Mfge8 fusion protein. This will result in an improved detection method with a higher accuracy, namely less risk of false positive results, in particular, if quantification of dying cells is to be performed. Thus, it is preferred that the determination of the presence or absence of dying cells labeled with the Mfge8 fusion protein in step (b) of the method for detecting dying cells also comprises the differentiation between dying cells labeled with the Mfge8 fusion protein and extracellular vesicles labeled with the Mfge8 fusion protein. It is equally preferred that the determination of the presence or absence of extracellular vesicles labeled with the Mfge8 fusion protein in step (b) of the method for detecting extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface also comprises the differentiation between dying cells labeled with the Mfge8 fusion protein and extracellular vesicles labeled with the Mfge8 fusion protein. In other words, it is preferred that step (b) includes removal of stained dying cells from the population of cells in the detection of EVs and includes removal of stained EVs from the population of EVs in the detection of dying cells. The population of cells are the cells present in the analyte with which the detection method is performed, whereas the population of EVs are the EVs present in the analyte with which the detection method is performed. Due to the fact that both dying cells and extracellular vesicles are indiscriminately bound by the Mfge8 fusion protein of the invention in step (a) of the detection methods, if they are present in the sample to be analyzed, differentiation between dying cells and vesicles having phosphatidylserine (PS) on the extravesicular surface is performed in step (b), e.g., by any one of the methods outlined herein above or as described in the example section. This way, it can be determined whether both dying cells and extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface are present in the analyzed sample. In other terms, it can be checked whether the Mfge8 fusion protein is bound (i) only to dying cells or (ii) only to extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface or (iii) both. This can be particularly important for those detection assays, which are used (also) for a quantitative analysis of dying cells or extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface.

The definitions and preferred embodiments provided herein above, in particular with regard to preferred Mfge8 proteins, preferred (poly)peptides of interest, compositions comprising the Mfge8 fusion proteins of the invention and regimens for their administration apply mutatis mutandis also to this claimed method of detecting dying cells and to this claimed method of detecting extracellular vesicles having phosphatidylserine on the extracellular surface. In a preferred embodiment of this method of the invention of detecting dying cells or, alternatively, extracellular vesicles having phosphatidylserine on the extracellular surface, the method further comprises a step (c1) of isolating the cells or extracellular vesicles labeled with the Mfge8 fusion protein of the invention. It will be appreciated that this step is only applicable in case labeled cells or labeled extracellular vesicles are present after the contacting step (a).

General means and methods for isolating a compound or cell of interest have been described in detail herein above. General means and methods for isolating extracellular vesicles are known in the art (see, e.g., Witwer, K. W. et al., Journal of extracellular vesicles 2, 2013 (doi:10.3402/jev.v2i0.20360)). These means and methods apply mutatis mutandis to the methods of the invention. Preferably, dying cells are isolated by flow cytometry, MACS or microscope guided single cell isolation. Preferably, extracellular vesicles are isolated by ultracentrifugation and filtering.

Once dying cells or EVs have been isolated, they can for example be processed by downstream analyses, for example and in the case of dying cells in order to identify mechanisms of cell death in different pathologies, such as e.g. muscular dystrophy. To this end, a biopsy can be obtained from a patient having the pathology of interest after said patient has been administered an Mfge8 fusion protein of the invention. In the case of e.g. muscular dystrophy, the administration has preferably been carried out intramuscularly. In the biopsy sample, it is then analysed which types of cell are affected. These cells can easily be identified, as they are labeled by the Mfge8 fusion protein of the invention. Moreover, the isolated cells can be employed for further analysis, such as proteomics or transcriptomics, in order to identifying new factors that might play an important role in the regulation of cell death. To this end, the data relating to the proteome or transcriptome of a patient suffering from a pathology of interest can be compared with the proteome or transcriptome, respectively, of the same type of cells obtained from subjects not suffering from the pathology under investigation. In the case of extracellular vesicles, purification of, for example, exosome preparations to be used in anti-tumor immunotherapy (as referred to herein above) can provide a contribution in the safety of use of corresponding preparations as therapy. Further to this method of the invention of detecting dying cells, the present invention also relates to a method of diagnosing a disease associated with cell death, wherein the method comprises: detecting the amount of labeled dying cells present in a sample obtained from a subject suspected to suffer from a disease associated with cell death, wherein the dying cells have been labeled with the Mfge8 fusion protein of the invention, wherein an increased amount of labeled cells detected in the subject compared with the amount of labeled cells detected in a control representative of (a) subject(s) not afflicted by the disease indicates that said subject is suffering from or is at risk of developing a disease associated with cell death.

In order to label dying cells with the Mfge8 fusion protein of the invention, the Mfge8 fusion protein of the invention has either been administered to the subject in vivo in a preceding step, for example by intravenous, intraperitoneal, subcutaneous, or intra-tissue injection. Subsequently, the amount of labeled dying cells in a sample obtained from said subject is determined. Alternatively, a sample can be obtained from the subject and can then be labeled and subsequently tested in vitro.

It will be appreciated that it is possible, but not required, that the comparison with the amount of labeled cells detected in a control representative of (a) subject(s) not afflicted by the disease (also referred to herein as the control) can be carried out simultaneously with the detection of the amount of labeled dying cells obtained from the subject of interest. However, control data for subjects known to not suffer from a disease associated with cell death can also be obtained at a different time, such as prior to or after carrying out the experiment for diagnosing a disease associated with a defect in the regulation of cell death. Moreover, if such control data are already available in the art, the comparison can also be with said known controls. In addition, the provision of the Mfge8 fusion protein of the invention also enables a method of determining the effectiveness of a therapeutic treatment of a disease associated with cell death, the method comprising: comparing the amount of dying cells labeled with the Mfge8 fusion protein of the invention as described above in a sample obtained from a subject prior to receiving a therapeutic treatment with the amount of dying cells labeled with the Mfge8 fusion protein of the invention in a sample obtained from a subject during or after receiving a therapeutic treatment; wherein an altered amount of labeled cells detected during or after receipt of the therapeutic treatment compared with the amount of labeled cells detected before the therapeutic treatment indicates that the therapeutic treatment is effective. More specifically, where the therapy consists of inducing cell death, an increase in the amount of labeled dying cells is indicative of an effective treatment. On the other hand, where the therapy aims at preventing cell death, such as e.g. in autoimmune disease, a decrease in the amount of labeled dying cells is indicative of an effective treatment.

The present invention also relates to a method of analyzing phagocytosis of dying cells, the method comprising:

(a) labeling dying cells by contacting the cells with the Mfge8 fusion protein of the invention, wherein the Mfge8 fusion protein of the invention has an RGD-binding activity; and

(b) detecting the uptake of the labeled cells of step (a) by phagocytes. Phagocytosis is a form of endocytosis wherein cell debris, amongst others, is taken up by other cells, the so-called phagocytes. Thus, phagocytosis is the major mechanism for removing cell debris, which could otherwise become problematic, for example by causing autoimmune diseases. "Phagocytes", in accordance with the present invention, include without being limiting macrophages, dendritic cells, neutrophils, monocytes, and mast cells. Preferably, the phagocytes are macrophages or dendritic cells, most preferably macrophages.

In a first step (a), dying cells are labeled with the Mfge8 fusion protein of the invention as described herein above for the method of detecting dying cells. In a second step, the uptake of the thus labeled dying cells into phagocytes is detected, as shown e.g. in Example 3 below. Preferably, visualization and quantification of phagocytosis of dying cells is carried out based on a method selected from flow cytometry, histology, and microscopy, in particular 2-photon microscopy.

The definitions and preferred embodiments provided herein above with regard to the method of detecting dying cells apply mutatis mutandis also to this claimed method of analyzing phagocytosis. In particular, this method of analyzing phagocytosis may be carried out in vitro, ex vivo or in vivo as described above and the Mfge8 fusion protein preferably comprises a reporter protein, most preferably mCherry, or a recognition sequence for enzymatic modification, most preferably a sortase recognition sequence, for radioactive labeling with e.g. 18F or 64Cu.

In those cases where the method of analyzing phagocytosis is carried out in vivo, the Mfge8 fusion protein of the invention is preferably administered to a living subject in step (a) by intravenous, intraperitoneal, subcutaneous, or intra-tissue injection, and an intermediate step after step (a) and before step (b) is included which comprises: (a') obtaining phagocytic cells, preferably macrophages, from the subject. In the subsequent step (b), these phagocytic cells are then analyzed to quantify the uptake of dying cells labeled with the Mfge8 fusion protein, as described above.

The present invention further relates to a method of diagnosing a disease associated with a defect in phagocytosis, wherein the method comprises detecting the amount of labeled dying cells present in phagocytes obtained from a subject suspected to suffer from a disease associated with a defect in phagocytosis, wherein the dying cells have been labeled by administering the Mfge8 fusion protein of the invention to the subject, wherein the Mfge8 fusion protein of the invention has an RGD-binding activity; wherein an altered amount of labeled dying cells detected in the phagocytes obtained from the subject suspected to suffer from a disease associated with a defect in phagocytosis compared with (a) the amount of labeled dying cells detected in the same type of phagocytes obtained from a control representative of (a) subject(s) not afflicted by a disease associated with a defect in phagocytosis; or (b) the amount of labeled dying cells detected in a different type of phagocytes obtained from the same subject, indicates that said subject is suffering from or is at risk of developing a disease associated with a defect in phagocytosis.

As discussed herein above, a "defect in phagocytosis" refers to defects in the uptake of dying cells by phagocytes, as well as to defects in the degradation of phagocytosed cells within the phagocytes, for example due to insufficient amounts of Dnase present in macrophages. Such defects in phagocytosis can lead to an accumulation of dead cells and cell debris, which in turn can, for example, induce an autoimmune response in the subject concerned. Accordingly, a disease associated with a defect in phagocytosis is a disease caused or associated with one (or several) of these aspects. Preferably, the disease associated with a defect in phagocytosis is an autoimmune disease, more preferably a disease selected from lupus erythematosus and rheumatoid arthritis.

In accordance with this method of the present invention, phagocytes obtained from a subject suspected to suffer from such a disease are analysed in order to detect the amount of labeled dying cells present therein. To this end, the Mfge8 fusion protein of the invention is to be administered to the subject in a preceding step, for example by intravenous, intraperitoneal, subcutaneous, or intra-tissue injection. Subsequently, the amount of labeled dying cells within the phagocytes obtained from said subject is determined. Alternatively, phagocytes can be obtained from the subject and can be tested in vitro for their potential to phagocytose labeled dying cells, as described e.g. in Chaka et al. ((1995) Clin Diagn Lab Immunol (6):753-9).

The amount of labeled dying cells detected in the phagocytes is then compared to a control. As discussed herein above, said control can be e.g. a subject known to not suffer from a disease associated with a defect in phagocytosis. It will be appreciated that in that case the same type of phagocytes is preferably compared between the subject suspected to suffer from the disease and a subject known to not suffer from said disease. As also discussed herein above, the control data can be obtained simultaneously or, alternatively, previously established data (either by previous experimentation or based on known control data) can be relied on.

Alternatively, the control can also be the amount of labeled dying cells detected in a different type of phagocytes obtained from the same subject, in those cases where only a certain type of phagocytes is known or suspected to be affected, while other types of phagocytes act normally. For example, if only macrophages are affected in the respective disease, the control phagocytes obtained from the same subject can be monocytes, neutrophils, mast cells, dendritic cells etc.. It is particularly preferred in accordance with this option that the amount of labeled dying cells detected in macrophages obtained from the subject suspected to suffer from a disease associated with a defect in phagocytosis is compared with the amount of labeled dying cells detected in a different type of phagocytes obtained from the same subject, preferably in monocytes.

Preferably, both controls are employed and even more preferably, the control is a subject known to not suffer from a disease associated with a defect in phagocytosis. It will be appreciated that, as discussed above, it is possible, but not required, that the control data are obtained simultaneously with the detection of the amount of labeled dying cells present in phagocytes obtained from the subject of interest. However, control data for subjects known to not suffer from a disease associated with a defect in phagocytosis or from different types of phagocytes that function normally can also be obtained at a different time, such as prior to or after carrying out the experiment for diagnosing a disease associated with a defect in phagocytosis.

In those cases where the amount of labeled dying cells in phagocytes obtained from the subject is altered as compared to such controls, it can be concluded that said subject is suffering from or is at risk of developing a disease associated with a defect in phagocytosis. An alteration can be an increase, e.g. in those cases where the phagocytes are not able to digest the dying cells sufficiently, such that they accumulate in the phagocytes. An alternative can, however, also be a decrease, e.g. in those cases where the phagocytes are defective in uptake of dying cells.

Accordingly, it is not only possible to diagnose whether a subject suffers from or is at risk of developing a disease associated with a defect in phagocytosis, it is further possible to analyze in more detail the nature of the defect in order to provide the subject with customized medication that aims at overcoming the respective deficiency in phagocytosis diagnosed. Moreover, the phagocytes obtained from the subject can additionally be treated in vitro or ex vivo with a therapeutic compound and their capability of taking up dying cells prior to and following such treatment can be analyzed. Such response experiments can further help in identifying a suitable treatment regimen for the individual subject concerned. The present invention further relates to a method of determining the effectiveness of a therapeutic treatment of a disease associated with a defect in phagocytosis, the method comprising:

(a) detecting the amount of labeled dying cells present in phagocytes obtained from a subject suspected to suffer from a disease associated with a defect in phagocytosis before it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein of the invention to the subject, wherein the Mfge8 fusion protein of the invention has an RGD-binding activity; and

(b) detecting the amount of labeled dying cells present in phagocytes obtained from said subject suspected to suffer from a disease associated with a defect in phagocytosis during or after it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein of the invention to the subject, wherein the Mfge8 fusion protein of the invention has an RGD-binding activity;

wherein an altered amount of labeled dying cells detected in phagocytes obtained during or after receipt of the therapeutic treatment compared with the amount of labeled dying cells detected in phagocytes before the therapeutic treatment indicates that the therapeutic treatment is effective.

The term "treatment", as used herein, relates to any treatment in a subject including: (a) preventing a disease from occurring in a subject which may be predisposed to the disease; (b) inhibiting the disease, i.e. arresting its development; or (c) relieving the disease, i.e. causing regression of the disease (d) reversing the disease symptoms, i.e. leading to a partial or full recovery.

The definitions and preferred embodiments provided herein above with regard to the previous embodiments, in particular with regard to preferred Mfge8 proteins, preferred (poly)peptides of interest, compositions comprising the Mfge8 fusion proteins of the invention and regimens for their administration, as well as with regard to "altered amounts", apply mutatis mutandis also to this claimed method of determining the effectiveness of a therapeutic treatment.

In a preferred embodiment of the methods of the invention, the disease associated with cell death or associated with a defect in phagocytosis is systemic lupus erythematosus or rheumatoid arthritis.

The present invention further relates to a method of determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction, the method comprising: (a) detecting the amount of labeled dying cells present in a subject suspected to suffer from cancer or myocardial infarction before it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein according to the invention to the subject; and (b) detecting the amount of labeled dying cells present in said subject during or after it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein according to the invention to the subject; wherein an altered amount of labeled dying cells detected during or after receipt of the therapeutic treatment compared with the amount of labeled dying cells detected before the therapeutic treatment indicates that the therapeutic treatment is effective. As the Mfge8 fusion protein of the present invention is capable of detecting dying cells in subjects it can, accordingly, be used to detect such cells. By analyzing the amount of dying cells in a subject prior to and during/after therapeutic treatment, it can be determined whether said therapeutic treatment is or has been successful. As described in detail herein above, whether an increase or decrease of dying cells is indicative of successful treatment depends on the aim of the therapeutic intervention. In other words, where the therapy consists of inducing cell death, such as e.g. in cancer, an increase in the amount of labeled dying cells is indicative of an effective treatment. On the other hand, where the therapy aims at preventing cell death, such as e.g. in myocardial infarction, a decrease in the amount of labeled dying cells is indicative of an effective treatment. As described herein above in relation to method for detection of dying cells, accuracy of the detection of dying cells can be improved, when also the presence or absence of extracellular vesicles having phosphatidylserine on the extracellular surface is determined. Accordingly, it is also preferred for the method of this embodiment that the detection steps (a) and (b) comprise the differentiation between dying cells labeled with the Mfge8 fusion protein of the invention and extracellular vesicles labeled with the Mfge8 fusion protein of the invention.

The definitions and preferred embodiments provided herein above with regard to the previous embodiments, in particular with regard to preferred Mfge8 proteins, preferred (poly)peptides of interest, compositions comprising the Mfge8 fusion proteins of the invention, the method of detecting dying cells and regimens for their administration, apply mutatis mutandis also to this claimed method of determining the effectiveness of a therapeutic treatment.

In a more preferred embodiment of this method of the invention, the Mfge8 fusion protein comprises a radioisotope, preferably 18F or 64Cu, that has been added site-specifically to the Mfge8 protein via a sortase A recognition sequence. Accordingly, the method comprises in a first step (i) providing a Mfge8 fusion protein of the invention, wherein the (poly)peptide of interest is a recognition sequence for sortase A. In a second step (ii), the Mfge8 fusion protein is enzymatically modified with sortase A to site-specifically add a radioisotope, preferably 18F or 64Cu to the Mfge8 protein. In the third step (iii), the subject then treated with said radio- labeled Mfge8 fusion protein prior to and during/after the therapeutic treatment and the amount of labeled dying cells is determined, as described above as steps (a) and (b).

The determination of the amount of labeled dying cells can be achieved by any of the methods described herein above, preferably by PET scanning.

In accordance with the above, the present invention further relates to the Mfge8 fusion protein of the invention, wherein the (poly)peptide of interest is a recognition sequence for sortase A, for use in determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction. The steps of carrying out said determination are preferably as described for the method of determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction described above. Moreover, all definitions and preferred embodiments provided with regard to the method of determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction apply mutatis mutandis also to this use of the invention.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. In case of conflict, the patent specification, including definitions, will prevail.

In this specification and the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a compound" includes one or more compounds.

All the sequences accessible through the Database Accession Numbers cited herein are within the scope of the present invention and also include potential future updates in the database, in order to account for future corrections and modifications in the entries of the respective databases, which might occur due to the continuing progress of science.

All amino acid sequences provided herein are presented starting with the most N-terminal residue and ending with the most C-terminal residue (N->C), as customarily done in the art, and the one-letter or three-letter code abbreviations as used to identify amino acids throughout the present invention correspond to those commonly used for amino acids.

Regarding the embodiments characterized in this specification, in particular in the claims, it is intended that each embodiment mentioned in a dependent claim is combined with each embodiment of each claim (independent or dependent) said dependent claim depends from. For example, in case of an independent claim 1 reciting 3 alternatives A, B and C, a dependent claim 2 reciting 3 alternatives D, E and F and a claim 3 depending from claims 1 and 2 and reciting 3 alternatives G, H and I, it is to be understood that the specification unambiguously discloses embodiments corresponding to combinations A, D, G; A, D, H; A, D, I; A, E, G; A, E, H; A, E, I; A, F, G; A, F, H; A, F, I; B, D, G; B, D, H; B, D, I; B, E, G; B, E, H; B, E, I; B, F, G; B, F, H; B, F, I; C, D, G; C, D, H; C, D, I; C, E, G; C, E, H; C, E, I; C, F, G; C, F, H; C, F, I, unless specifically mentioned otherwise. Similarly, and also in those cases where independent and/or dependent claims do not recite alternatives, it is understood that if dependent claims refer back to a plurality of preceding claims, any combination of subject-matter covered thereby is considered to be explicitly disclosed. For example, in case of an independent claim 1, a dependent claim 2 referring back to claim 1 , and a dependent claim 3 referring back to both claims 2 and 1 , it follows that the combination of the subject-matter of claims 3 and 1 is clearly and unambiguously disclosed as is the combination of the subject-matter of claims 3, 2 and 1. In case a further dependent claim 4 is present which refers to any one of claims 1 to 3, it follows that the combination of the subject-matter of claims 4 and 1 , of claims 4, 2 and 1 , of claims 4, 3 and 1 , as well as of claims 4, 3, 2 and 1 is clearly and unambiguously disclosed.

The above considerations apply mutatis mutandis to all appended claims. To give a non- limiting example, the combination of claims 10 and 9 is clearly and unambiguously envisaged in view of the claim structure. The same applies for example to the combination of claims 10, 9 and 7, etc..

The figures show:

Figure 1: Mfge8 fusion proteins

The figure illustrates Mfge8 and its functional domains and the insertion sites of different reporter proteins. SS = signal sequence; RGD = tripeptide Arg-Gly-Asp, required for integrin binding; PT = proline/ threonine-rich domain; C1 = factor-VIII-homologous domain.

Figure 2: Expression of Mfge8 fusion proteins in HEK293 cells

HEK293 cells were transiently transfected with Mfge8-EGFP, Mfge8-miniSOG or Mfge8- mCherry constructs. A) After 24h expression of fluorescent fusion proteins was analyzed by fluorescent microscopy. All three Mfge8 fusion proteins were fluorescent and readily detectable by fluorescent microscopy. Scale bar δθμιτι. B) 24h after transfection supernatant and cells were harvested and analyzed for the presence of Mfge8 protein. All three Mfge8 fusion proteins were detected in the cell lysate and in the supernatant, confirming that the Mfge8 fusion proteins were secreted.

Figure 3: Mfge8 fusion proteins stain dying (apoptotic) cells

Untreated and staurosporin treated (1pg/ml, 2h) Jurkat cells were stained with 7AAD and AnnexinV-Cy5 (A, left) or 7AAD and Mfge8-EGFP (A, right) or where co-stained with AnnexinV-Cy5 and Mfge8-EGFP (B) in either a Ca²⁺-buffer or a Ca -free FACS buffer. AnnexinV stained apoptotic cells only in Ca²⁺-buffer, while the Mfge8 fusion protein stained apoptotic cells in both buffers.

C) Treated and untreated thymocytes were stained with Mfge8-biotin and Streptavidin- Alexa647 or with Streptavidin-Alexa647 only (control). Biotinylated Mfge8 was able to stain apoptotic cells.

Figure 4: Mfge8 fusion proteins are taken up by phagocytes

20x10⁶ Mfge8-mCherry-stained or unstained UV-irradiated apoptotic thymocytes were injected into the peritoneal cavity of mice. 1 h later, peritoneal macrophages were flushed out, stained with anti-F4/80 antibodies and analyzed by imaging flow cytometry. Mfge8-mCherry labeled apoptotic thymocytes were detected in almost all F4/80+ macrophages.

Figure 5: Apoptotic B cells

WT (Bcl2tg-) and Bcl2 overexpressing mice (Bcl2tg+) immunized with sheep erythrocytes were injected with 100 pg Mfge8-EGFP. 30 min later, spleens were harvested and analyzed by FACS. Mfge8-EGFP expression was analyzed on marginal zone (MZ) B cells (lgM⁺lgD⁺CD21^hi9hCD23^low), follicular (FO) B cells (lgM⁺lgD⁺CD21^lowCD23⁺) and germinal center (GC) B cells (CD19⁺lgD^l0WCD95⁺GL-7⁺). Mean fluorescence intensity of Mfge8-EGFP (middle panel) and percentages of Mfge8-EGFP⁺ cells (lower panel) was clearly reduced in Bcl2tg mice.

Figure 6: Apoptotic T ceils

WT (Bcl2tg-) and Bcl2 overexpressing mice (Bcl2tg+) immunized with sheep erythrocytes were injected with 100 pg Mfge8-EGFP. 30 min later, spleens were harvested and analyzed by FACS. Mfge8-EGFP expression was analyzed on naive and effector CD4 T cells (CD4⁺CD62L⁺CD44^" and CD4⁺CD62L^"CD44⁺, respectively) and naive and effector CD8 T cells (CD8⁺CD62L⁺CD44^" and CD8⁺CD62L^"CD44⁺, respectively. Mean fluorescence intensity of Mfge8-EGFP (middle panel) and percentages of Mfge8-EGFP⁺ cells (lower panel) was clearly reduced in Bcl2tg mice.

Figure 7: Imaging flow cytometry of apoptotic GC B cells

WT mice previously immunized with sheep erythrocytes were injected with 100pg Mfge8- EGFP. 30min later, mice were sacrificed and splenic GC B cells analyzed by imaging flow cytometry. Mfge8-EGFP+ GC B cells showed morphological signs of apoptotic cells with apoptotic blebs intensively stained with Mfge8-EGFP. Figure 8: Apoptotic cells during LCMV infection

Mice were infected with the acute lymphocytic choriomeningitis virus (LCMV) strain WE. On day 11 post infection, mice were injected with 100 g Mfge8-EGFP. 30min later, spleens were removed and subjected to FACS analysis. Frequencies of live (AnnexinV^"Mfge8^", grey), in vitro stained apoptotic (AnnexinV'Mfgee^", blue) and in vivo stained apoptotic (AnnexinVMfgee^*, red), T cells (upper panel), dendritic cells (middle panel) and macrophages (lower panel) were compared. AnnexinV^"Mfge8^", AnnexinV⁺Mfge8^" or AnnexinV⁺Mfge8⁺ cell populations were gated. Then frequencies of different T cell (upper right panel), CD11c/ MHC-II (middle right panel) or CD11b/ MHC-II (lower right) expressing subsets were determined within these populations. Substantial differences between in vitro and in vivo stained dying cells were detected.

Figure 9: Intrathymic injection of Mfge8-EGFP

Mice were anaesthetized and injected with 5-1 O g Mfge8-EGFP intrathyimcally. 30 and 60 min later, thymi were removed and distribution of live (AnnexinV^"Mfge8-EGFP ), in vitro stained dying cells (AnnexinV⁺Mfge8-EGFP^~) and in vivo stained dying cells (AnnexinV⁺Mfge8- EGFP⁺) among the different T cell maturation stages assessed. Most dying cells were in the CD4⁺CD8⁺ double positive stage (A). Comparison of CD5 expression among life (AnnexinV Mfge8-EGFP^"), in vitro stained (AnnexinV'Mfgee-EGFP ) and in vivo stained (AnnexinV⁺Mfge8-EGFP⁺) cells was compared. Drastic differences between in vitro and in vivo stained dying cells in their CD5 expression levels were detected among DP and DN T cells.

Figure 10: Immunofluorescent microscopy analysis of fge8-EGFP in the lymph node (A) Mice previously immunized with sheep erythrocytes were injected with 30 g Mfge8-EGFP into the footpad. 4h, 8h, 12h, 24h and 48h later, draining lymph nodes were removed, cryosectioned and stained with FITC-conjugated anti-GFP (upper panel) and anti-CD68 antibodies (lower panel). Mfge8-EGFP was detected up to 24h after injection. Scale bar 100 μιτι. (B) The 12h time point is shown in higher magnification. Section was stained with anti- Mfge8 antibody clone 18A2 (left), which fails to detect Mfge8 that has been phagocytosed by macrophages. Additionally, section was stained with a biotinylated polyclonal anti-Mfge8 antibody, which detects extracellular Mfge8 as well as phagocytosed Mfge8 (middle). Tingible body macrophages are stained with anti-CD68 antibodies (right). Mfge8 fusion proteins are clearly taken up by tingible body macrophages, furthermore Mfge8 accumulation is restricted to the light zone of the GC, where it can also be found within tingible body macrophages. Scale bar 50 μιτι. Figure 11 : 2-photon imaging of dying cells

Mice previously immunized with sheep erythrocytes were injected with 30pg Mfge8-mCherry and 10pg FITC-conjugated anti-BP-3 or with anti-BP-3 alone into the footpad. 24h later, intact draining lymph nodes were imaged using 2-photon microscopy. Mfge8-mCherry labeled cells were detected close to BP-3⁺ FDC networks. Scale bar 50 μιη.

Figure 12: fge8-Dnase2a shows intact Dnase activity

1 pg mouse tail DNA was incubated with 270μΜ or 130μΜ of recombinant Dnase2a or recombinant Mfge8-Dnase2a fusion protein for 2.5h at 37°C at a pH of 5.7. Samples were loaded onto a 2% agarose gel. The undigested control shows a clear band of intact DNA, while Dnase2a or Mfge8-Dnase2a treated samples show a smear, indicative of DNA degradation.

Figure 13: Mfge8-HSVTK shows intact thymidine kinase activity

Wild type HEK (HEKwt) and HEK cells expressing Mfge8-HSVTK (HEK-Mfge8-HSVTK) were treated with 2pg ganciclovir (GC) for up to 7 days. Before (dayO) start of the treatment and on day5 and day7 of the treatment total cell number was measured using a CASY cell counter. HEKwt cells with GC and HEK-Mfge8-HSVTK cells without GC showed normal proliferation similar to untreated HEKwt cells. Only HEK-Mfge8-HSVTK cells on GC-treatment died due to the GC-activated HSVTK activity, confirming that the Mfge8-HSVTK fusion protein maintains intact thymidine kinase function.

Figure 14: Immunization with Mfge8-EGFP and EGFP

Mice were immunized ip with 1.2 μΜ recombinant EGFP or Mfge8-EGFP precipitated in Alum (n= 5 per time point). (A) 5 days later serum was collected and anti-EGFP IgM titers were measured by ELISA. Ip injection of Mfge8-EGFP mixed with Alum resulted significantly increased IgM titers compared to immunization with recEGFP/ Alum. (B) Total anti-EGFP IgG, lgG1 and lgG2b titers were determined on day 7 and 14. On day 7 total IgG was virtually absent after immunization with recEGFP/Alum, but clearly detectable after recMfge8- EGF/Alum immunization. Also lgG1 and lgG2b titers were below the detection limit when recEGFP/ Alum was used for immunization, while lgG1 and lgG2b titers of recMfge8- EGFP/Alum immunized mice were 0.38pg/ml and 0.046μg/ml, respectively. On day 14 total IgG and lgG2b titers were still significantly increased in recMfge8-EGFP/Alum immunized mice, while lgG1 titers were similar in recEGFP and recMfge8-EGFP immunized mice.. Figure 15: Mfge8-EGFP but not EGFP accumulates on FDCs

1.2pm recombinant Mfge8-EGFP or EGFP were injected iv into either (upper and middle panel) or immune (lower panel) mice. 1 and 12hs later, spleens were removed and analyzed by fluorescent immunohistochemistry. Spleens were stained with anti-GFP (left), anti-CD21 (follicular dendritic cells, middle) and anti-SIGN-R1 (marginal zone macrophages). No injected recombinant EGFP was detectable in the spleen of naive mice. In contrast, Mfge8-EGFP accumulated on CD21⁺ FDCs (top left, encircled). Injected EGFP only accumulated on FDCs in immune mice. Scale bar 50 pm. Figure 16: Mfge8⁺ germinal center B cells show apoptotic morphology

(A) Mice were immunized with sheep erythrocytes iv. 5, 7, 10 and 15 days later spleens were harvested and CD19⁺lgD^lowCD95⁺GL7⁺ GC cells quantified using flow cytometry. Numbers of GC B cells peaked on day 10 and then rapidly declined.

(B) On day 10 after immunization with sheep erythrocytes mice were injected with 100pg Mfge8-EGFP. 30min later mice were sacrificed and splenic GC B cells analyzed by imaging flow cytometry. Mfge8-EGFP+ GC B cells showed morphological signs of apoptotic cells with apoptotic bodies intensively stained with Mfge8-EGFP.

Figure 17: Mfge8-EGFP stains extracellular vesicles (EVSs) attached to cells

(A) Mice were injected with 100pg Mfge8-EGFP. 30min later mice were sacrificed and splenic B and T cell subsets were stained and analyzed by flow cytometry. Mfge8-EGFP staining was assessed on CD19⁺CD21 ^highCD23^l0W marginal zone (MZ), and CD19⁺CD21^|0WCD23⁺ follicular (FO) B cells and CD4⁺CD8^" and CD4 CD8* T cells. 22% of MZ B cells fell within the Mfge8⁺ gate, while also cells within the Mfge8^" gate showed increased Mfge8-EGFP fluorescence intensity. Other B cell subsets and T cell subsets showed less Mfge8-EGFP staining.

(B) When Mfge8⁺ MZ B cells were analyzed by imaging flow cytometry, attached Mfge8⁺ vesicles that were attached to the cells were observed, indicating that these cells were not apoptotic, but bound EVs that were intensely stained with Mfge8-EGFP. Figure 18: Lymphocytes bind EVs through phosphatidylserine

(A) Uncoated and recMfge8 coated EVs from PKH26-labeled apoptotic thymocytes were prepared by ultracentrifugation and injected iv into WT and Mfgefr'^' mice. 1 h later mice were sacrifized and splenic B and T cells analyzed by flow cytometry. Binding of PKH26-labeled EV by CD19⁺CD21^highCD23^l0W MZ, CD19⁺CD21^l0WCD23⁺ FO B cells, CD4⁺CD8^_ and CD4 D8⁺ T cells was assessed. MZ B cells showed the highest binding capacity in WT and Mfge ^' mice. FO B cells and T cells showed lower binding capacity as assessed by the mean fluorescence intensity of PKH26. No differences were observed between WT and Mfge8 mice. Precoating of vesicles with 300ng/ml recMfge8 completely abolished binding of EVs by B and T cells. (B) Binding of PKH26-labeled EVs by B cells was analyzed by imaging flow cytometry. PKH26-labeled vesicles are clearly visible on the surface of B cells.

Figure 19: Defining truth populations

(A) To identify features allowing to distinguish dying cells from cells with attached EVs, we defined truth populations of dying cells and cells with attached vesicles. Splenic lymphocytes with attached PKH26-labeled vesicles derived from thymocytes that were injected iv served as truth population for cells with bound vesicles. Active Caspase8⁺Mfge8-EGFP⁺ apoptotic thymocytes treated for 2h with I g/ml staurosporine served as truth population for dying cells.

(B) Two features were identified allowing to discriminate dying cells from cells with attached EVs using the IDEAS software. One feature was the area ratio between the BF area using the standard mask and the Mfge8-EGFP or PKH26 staining using the peak mask (Peak(M02, Mfge8, Bright, 1.6) or Peak(M03, PKH26, Bright, 1.6). Cells with attached vesicles had an area ratio >10. The second feature was the minor axis intensity of the Mfge8-EGFP or PKH26 staining. Dying cells had a minor axis intensity value >3, cells with attached vesicles a value <3. (C) Panels show images of cells that have a BF/PKH26 area ratio >10 and a PKH26 minor axis intensity value <3 (left). These cells clearly have attached vesicles on their surface. Active caspase8⁺Mfge8-EGFP⁺ cells that have a BF/Mfge8-EGFP area ratio <10 and a Mfge8 minor axis intensity value >3 exhibit a more global staining pattern and intensely stained apoptotic bodies (right).

Some active caspase8^"Mfge8-EGFP^" cells that exhibit a BF/Mfge8-EGFP area ratio >10 and a minor axis intensity value <3 indeed show small, weakly stained EVs attached to their surface (middle).

Figure 20: Quantification of apoptotic germinal center B cells

To determine the frequency of Mfge8⁺ GC B cells in spleens of mice immunized with sheep erythrocytes that are truely apoptotic we applied the BF/Mfge8-EGFP area ratio feature and the Mfge8-EGFP minor axis intensity feature to CD19⁺lgD^lowCD95⁺GL7⁺ GC B cells using images obtained from imaging flow cytometry. Approx. 37% of the GC B cells had a BF/Mfge8-EGFP area ratio <10 and a minor axis intensity value >3 and were therefore defined as apoptotic. Approx. 44% of the GC B cells had a BF/Mfge8-EGFP area ratio >10 and a minor axis intensity value <3 and hence were life cells with vesicles attached. The images of the cells confirmed the correct classification of the cells (lower panel). Figure 21: Quantification of apoptotic B and T cells

To determine the frequency of B and T cells in spleens of WT mice that are truely apoptotic we applied the BF/Mfge8-EGFP area ratio feature and the Mfge8-EGFP minor axis intensity feature to CD19⁺ B and CD4⁺CD8^" and CD4^"CD8⁺ T cells using images obtained from imaging flow cytometry. Approx. 13% of all Mfge8⁺ B cells were apoptotic, while approx. 59% had vesicles attached. Only approx. 1% of Mfge8⁺ CD4 T cells were apoptotic and 31% had vesicle attached. Only 0.7% of the Mfge8⁺ CD8⁺ T cells could be classified as cells with attached vesicles. All other cells could not be classified due to the very weak Mfge8-EGFP staining.

The following examples illustrate the invention: Example 1 : Materials and Methods Production of Mfge8 fusion proteins

Mfge8 fusion proteins were stably expressed in HEK293 cells, adapted to serum-free medium (EX-Cell serum-free medium, Sigma). Cells were grown in a 7L fermenter (Infors) for 5 days. The cell culture supernatant was harvested and cleared by centrifugation and filtration (0.2 pm) and the supernatant was then passed through an anti-FLAG affinity gel (M2 Agarose beads, Sigma). Bound protein was eluted with FLAG-peptide, quantified by ELISA and stored at -80°C.

Example 2: Mfge8 fusion proteins (Mfge8-FPs) bind dying cells in vitro To generate a versatile staining tool for detecting and visualizing dying cells in vitro and in vivo fusion proteins (FPs) were generated consisting of fge8 and different (poly)peptides, such as EGFP, mCherry, miniSOG, YFP and CFP (Shaner, N. C. et al. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat Biotechnol 22, 1567-1572, (2004)) or the bioluminescent reporter luciferase (Kim, J. E., Kalimuthu, S. & Ahn, B. C. In vivo cell tracking with bioluminescence imaging. Nuclear medicine and molecular imaging 49, 3-10, (2015)). MiniSOG (mini Singlet Oxygen Generator) is a small fluorescent tag that can be used as a label for electron microscopy studies. It is able to catalyze the polymerization of diaminobenzidine (DAB) into a precipitate that can be visualized by electron microscopy (Shu, X. et al. A genetically encoded tag for correlated light and electron microscopy of intact cells, tissues, and organisms. PLoS biology 9, e1001041, (2011)). The recognition sequence of the biotin ligase BirA (Fairhead, M. & Howarth, M. Site- specific biotinylation of purified proteins using BirA. Methods in molecular biology 1266, 171- 184, (2015)) was also linked to Mfge8 to be able to biotinylate Mfge8 to allow detection of Mfge8-biotin labeled dying cells with different Streptavidin conjugates (Fig. 1 ). To minimize sterical hindrance between Mfge8 and the fluorescent reporter, the two were separated by a 15aa helical linker. To aid purification of the FPs, a FLAG-tag was added at the C-terminus (Fig. 1).

Mfge8-FPs were expressed in HEK293T cells and their fluorescence confirmed by fluorescence microscopy (Fig 2A). Wild type Mfge8 is secreted Hanayama, R. et al. Identification of a factor that links apoptotic cells to phagocytes. Nature 417, 182-187, (2002)), and Western blotting was used to confirm that also the Mfge8-FPs are secreted (shown for Mfge8-EGFP, Mfge8-miniSOG and Mfge8-mCherry, Fig. 2B). Next, it was assessed whether the Mfge8-FPs retained the ability to bind dying cells. For this, apoptosis was induced in Jurkat cells by treatment with 1 pg/ml staurosporin for 2h. The cells were then stained with approx. 150 ng/ml Mfge8-EGFP. To compare the binding efficiency of Mfge8-EGFP with that of another PS-binding molecule, AnnexinV, the cells were co-stained with an equimolar concentration of AnnexinV-Cy5 and then analyzed by FACS (Fig. 3). AnnexinV is presently one of the most widely used reagents for staining dying cells in flow cytometric analyses. However, binding of AnnexinV to PS is dependent on high concentrations of Ca²⁺ and, therefore, cells have to be stained in a special, Ca²⁺-containing binding buffer (van Engeland, M., Nieland, L. J., Ramaekers, F. C, Schutte, B. & Reutelingsperger, C. P. Annexin V-affinity assay: a review on an apoptosis detection system based on phosphatidylserine exposure. Cytometry 31 , 1-9 (1998)). To distinguish apoptotic cells from necrotic cells, 7AAD, a DNA- binding dye that can only enter the cells when the cell membrane has lost its integrity, was used (Waters, W. R., Harkins, K. R. & Wannemuehler, M. J. Five-color flow cytometric analysis of swine lymphocytes for detection of proliferation, apoptosis, viability, and phenotype. Cytometry 48, 146-152, (2002)). Cells that are 7AAD+ are considered necrotic.

Staurosporin treatment clearly increased the amount of AnnexinV^* and Mfge8-EGFP⁺ cells by approx. 2-fold. Importantly, not only a similar percentage of cells was stained by AnnexinV and Mfge8-EGFP (Fig 3A), but all AnnexinV stained cells were also positive for Mfge8-EGFP (Fig. 3B) when Ca²⁺-containing buffer was used, indicating that Mfge8-EGFP retained the ability to bind dying cells by binding to exposed PS. However, when conventional Ca²⁺-free FACS buffer was used, AnnexinV failed to bind to dead cells while Mfge8-EGFP bound dead cells very efficiently (Fig. 3A and 3B). Thus, in contrast to AnnexinV, Mfge8-EGFP is capable of binding to PS also in the absence of Ca²⁺. As a consequence, Mfge8-FPs can be employed to stain dying cells also in those cases where high concentrations of Ca²⁺ can be problematic, e.g. in tissue culture experiments or in vivo.

Further, an Mfge8 fusion protein comprising Mfge8 fused with a recognition sequence for the biotin ligase BirA (Mfge8-BirA rec-seq) was generated. Mfge8-BirA rec-seq was purified and then biotinylated using BirA. To show that biotinylation was successful and that Mfge8-biotin is able to bind apoptotic cells, staurosporin-treated apoptotic thymocytes and untreated thymocytes were stained with Mfge8-biotin and Strepatvidin-Alexa647. Flow cytometry confirmed that also Mfge8-biotin stains apoptotic cells very efficiently (Fig. 3C).

Example 3: Mfge8 fusion proteins can be used in phagocytosis assays

In order to test whether apoptotic cells opsonized with Mfge8-FPs are still efficiently engulfed by macrophages, UV-irradiated apoptotic thymocytes were opsonized with Mfge8-m Cherry and injected into the peritoneal cavity of mice. One hour later, peritoneal cells were harvested and the uptake of Mfge8-mCherry-postive cells into peritoneal macrophages was quantified by an AMNIS ImageStream imaging flow cytometer. F4/80⁺ peritoneal macrophages efficiently phagocytosed Mfge8-mCherry opsonized apoptotic thymocytes (Fig. 4). Approx. 80% of all macrophages were positive for Mfge8-mCherry, which is in line with previous studies that addressed the phagocytosis capacity of peritoneal macrophages (Taylor, P. R. et al. A hierarchical role for classical pathway complement proteins in the clearance of apoptotic cells in vivo. J Exp Med 192, 359-366 (2000)). Furthermore, imaging cytometry of individual macrophages clearly demonstrated the presence of multiple phagocytosed Mfge8-mCherry-positive cells inside the macrophages.

Also in vivo, Mfge8 fusion proteins are readily phagocytosed, which is illustrated by its uptake into tingible body macrophages in lymph nodes (Fig. 10).

Taken together, these results strongly indicate that Mfge8-FPs have retained the full functionality of Mfge8. Therefore, Mfge8-FPs cannot only be used to detect apoptotic cells, but also to assess and monitor the phagocytosis capacity of phagocytes. Mfge8-FPs are thus a valuable tool to diagnose and detect phagocytosis defects, which for example can be caused by deficiency of the complement factor C1 q or other genetic defects. Example 4: Detection of dying cells in vivo using Mfge8-FPs

Analysis of cell death in vivo is difficult (Blankenberg, F. G. In vivo detection of apoptosis. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 49 Suppl 2, 81s- 95s, (2008)): Histological analysis of PS-positive cells in tissues is not possible, because also intracellular PS is accessible to staining reagents in sectioned tissue. Detection of fragmented DNA by TUNEL staining only happens at late stages of apoptosis and several studies indicate that most DNA fragmentation in vivo only is initiated after the dying cell has been phagocytosed (Nagata, S. DNA degradation in development and programmed cell death. Annu Rev Immunol 23, 853-875, (2005)). Another method to visualize dead cells is by staining activated caspase-3 Gown, A. M. & Willingham, M. C. Improved detection of apoptotic cells in archival paraffin sections: immunohistochemistry using antibodies to cleaved caspase 3. The journal of histochemistry and cytochemistry: official journal of the Histochemistry Society 50, 449-454 (2002)), but there are also caspase-3 independent pathways of cell death initiation (Broker, L. E., Kruyt, F. A. & Giaccone, G. Cell death independent of caspases: a review. Clinical cancer research : an official journal of the American Association for Cancer Research 11 , 3155-3162, (2005)). Finally, the quantitation of dying cells in homogenized organs by AnnexinV staining harbors the problem that many cells die during organ preparation.

To overcome these problems and to specifically detect only cells that died in situ, labeling of dying cells in vivo by injecting Mfge8-EGFP intravenously was tested. The advantage of this approach is that dying cells are labeled while the mouse is still alive and before the organ is removed and homogenized. The in vivo labeling approach was tested by injecting 100pg Mfge8-EGFP intravenously into mice that were previously immunized with sheep erythrocytes to induce an immune response, which triggers cell death in various immune cells, especially in B and T cell subsets. To control for the specificity of the Mfge8-labeling, mice were used that over-express the anti-apoptotic factor Bcl2. These mice should have significantly reduced numbers of apoptotic B and T cells (McDonnell, T. J. et al. bcl-2-immunoglobulin transgenic mice demonstrate extended B cell survival and follicular lymphoproliferation. Cell 57, 79-88 (1989); Strasser, A., Harris, A. W. & Cory, S. bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67, 889-899 (1991)). 30min after injection, the spleen were removed and Mfge8-EGFP labeled cells were analyzed by FACS.

The analysis was focused on marginal zone (MZ, lgMlgD⁺CD21^hi9hCD23^low), follicular (FO, lgM⁺lgD⁺CD21^hlghCD23^l0W) and germinal center (GC, CD19⁺lgD^l0WCD95⁺GL-7⁺) B cells in Bcl2tg and control mice (Fig. 5, upper panel). Mfge8-EGFP labeled cells could be clearly detected in all injected mice by a strong increase in Mfge8-EGFP mean fluorescence intensity (MFI) compared to un-injected control mice. Strikingly, Bcl2tg showed a markedly reduced MFI for Mfge8-EGFP (Fig. 5, middle panel). Assuming that Mfge8-EGFP specifically labels dying cells, this result was expected. In line with this, also the percentage of Mfge8-EGFP labeled cells was reduced in Bcl2tg B cell populations (Fig. 5, lower panel).

Overexpression of Bcl2 not only affects B cell survival, also T cell apoptosis is strongly reduced in Bcl2tg mice (Strasser, A., Harris, A. W. & Cory, S. bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67, 889-899 (1991 )). Thus, it was next assessed whether this reduction in T cell apoptosis can also be detected by this in vivo labeling approach. Mfge8-EGFP labeling was assessed in naive (CD62L⁺CD44^~) or effector (CD62L⁺CD44 ) CD4 and CD8 T cells (Fig. 6 upper panel). While CD4 T cells generally showed a low Mfge8-EGFP labeling intensity, a much higher labeling intensity was seen in CD8 T cells, indicating increased apoptosis in CD8 T cells compared to CD4 T cells. Mfge8- EGFP staining was stronger in effector T cells compared to naive T cells and strongest in CD8 effector cells (Fig. 6 middle panel). This is to be expected, since effector T cells are known to be rather short lived (Yuzefpolskiy, Y., Baumann, F. M., Kalia, V. & Sarkar, S. Early CD8 T-cell memory precursors and terminal effectors exhibit equipotent in vivo degranulation. Cellular & molecular immunology, (2014)). In Bcl2tg mice, the MFI and percentage of Mfge8-EGFP labeled cells was strongly reduced in all T cell populations (Fig. 6 middle panel and lower panel, respectively).

Taken together, these results show that immune cells can be efficiently and specifically labeled by intravenously administered Mfge8 fusion proteins. To further strengthen the idea that in vivo administered Mfge8 fusion protein specifically labels apoptotic cells, Mfge8-EGFP was analysed by imaging flow cytometry to assess whether labeled cells have the typical morphological features of apoptotic cells (blebbing). In vivo Mfge8-EGFP labeled GL-7⁺CD95⁺ GC B cells of WT mice were analysed by an AMNIS ImageStream imaging cytometer. Mfge8-EGFP labeled GC B cells had clear morphological signs of apoptosis: they lost their rounded shape and show membrane blebbing, with apoptotic blebs being strongly EGFP labeled (Fig. 7). These results further confirm that i.v. administered Mfge8 fusion proteins can be used to stain apoptotic cells in vivo.

Example 5: In vivo labeling of dying cells is more specific than in vitro labeling

Organ preparation for downstream applications like flow cytometry damages many cells and can lead to a high background when staining dying cells in vitro for FACS. To compare in vivo versus in vitro staining of dying cells, dying T cells, dendritic cells and macrophages were analyzed in spleens of lymphocytic choriomeningitis virus (LCMV)-infected mice, since LCMV causes extensive cell death in these populations (Bahl, K., Huebner, A., Davis, R. J. & Welsh, R. M. Analysis of Apoptosis of Memory T Cells and Dendritic Cells during the Early Stages of Viral Infection or Exposure to Toll-Like Receptor Agonists. Journal of virology 84, 4866-4877, (2010)).

To this end, Mfge8-EGFP was injected 30min before sacrifice into LCMV-infected mice and spleen cell suspensions were labeled in vitro with AnnexinV and appropriate antibodies to identify different cell types. An analysis was carried out for the frequency of live cells (Mfge8^" AnnexinV^") versus in vitro stained dying cells, which are only stained after cell preparation and isolation with AnnexinV (AnnexinV⁺Mfge8 ) and in vivo stained dying cells, which are double labeled with AnnexinV and Mfge8 (AnnexinV" Mfge8⁺). Depending on the cell populations, drastic differences were observed between in vivo and in vitro labeled dying cells. While the frequency of in vivo labeled dying CD8⁺ T cells was one third lower as compared to in vitro labeled dying cells, the frequency of in vivo labeled CD4+ T cells was only 27% of those identified after in vitro labeling (Fig. 8 upper panel). This data suggests that in vitro labeling with AnnexinV either generates a high frequency of false positives, or detects cells which might have been damaged during the isolation procedure.

Pronounced differences were also seen when CD11c⁺ and CD11 b⁺ dendritic cells and macrophages were analyzed. While in vivo labeling (Fig. 8, AnnexinV+Mfge8+ populations) almost exclusively stained CD11 c⁺ and CD11 b⁺ cells that were MHC-Ι , in vitro labeling also stained many MHC-ΙΓ cells. Without wishing to be bound by theory, the inventors believe that these differences are mainly caused by cell damage that occurred during the organ preparation and by releasing cells from their natural environment, which causes many cells to be stained in vitro, but not in vivo. Thus, in vivo labeling of dying cells is superior to in vitro staining in quantifying and characterizing dying cells in specific organs or tissues. In a further approach, Mfge8 fusion protein was directly administered to organs. As example, the thymus is shown here. In the thymus, the organ where T cell differentiation takes place, most developing T cells are eliminated by negative selection. Therefore many apoptotic cells can be detected there (Palmer, E. Negative selection-clearing out the bad apples from the T- cell repertoire. Nat Rev Immunol 3, 383-391 , (2003)). To analyze cell death in the thymus, 5- 10 pg Mfge8-EGFP were injected into the thymi of anesthetized mice using a Hamilton syringe. 30 and 60 minutes later, mice were sacrificed and thymi were analyzed by FACS. Cells were stained for CD4 and CD8 to discriminate double negative (DN, CD4 CD8^"), double positive (DP, CD4⁺CD8⁺) and single positive (SP, CD4⁺CD8^" or CD4^"CD8⁺) stages. AnnexinV staining was included to assess the specificity of the in vivo administration of Mfge8 fusion protein: All Mfge8⁺ cells should also label positive for AnnexinV. This was indeed the case (Fig. 9A upper panel). However, from the presence of large numbers of AnnexinV+ Mfge8- EGFP-negative cells it becomes clear, that AnnexinV stains many more cells as compared to Mfge8-EGFP. One reason for this is probably that AnnexinV stains many additional cells that have undergone cell death during organ preparation. Most developing T cells are eliminated during the DP stage. This was confirmed by in vivo labeling of dying cells. After 30 and 60 min, 60-80% of Mfge8-EGFP⁺ cells were in the DP stage.

Next, CD5 expression on live (AnnexinV^"Mfge8 ), in vitro stained dying cells (AnnexinV⁺Mfge8 ) and in vivo stained dying cells (AnnexinV⁺Mfge8⁺) was compared. CD5 expression on T cells is up-regulated upon T cell receptor (TCR) signaling. In the thymus, CD5 expression correlates with the affinity of the TCR towards self (Azzam, H. S. et al. CD5 expression is developmental^ regulated by T cell receptor (TCR) signals and TCR avidity. J Exp Med 188, 2301-2311 (1998)). Since self-reactive T cells are eliminated in the thymus, CD5 was expected to be higher in dying T cells. Strikingly, CD5 was strongly elevated in in vivo labeled dying T cells in the DN and DP stage, compared to live cells. On the other hand, the CD5 upregulation was much less pronounced in in vitro stained dying cells and could only be seen in the DN, but not the DP stage (Fig. 9B).

These results underline the superior specificity of in vivo Mfge8-labeled dying cells compared to AnnexinV-labeling in vitro. Example 6: Tracking of dying cells by use of immunofluorescent microscopy

As discussed above, the detection of dying cells using PS-binding reagents cannot be used in histology, since PS is universally exposed in sectioned organs. Thus, Mfge8-EGFP was administered in vivo followed by cryosectioning, fixation and staining of the organ, whereby Mfge8 was either visualized with fluorescently labeled anti-Mfge8 or anti-EGFP antibodies since fixation quenched the fluorescence of Mfge8-EGFP.

In order to visualize Mfge8-EGFP stained cells by fluorescent microscopy, 30pg Mfge8-EGFP were administered into the footpad of mice, previously immunized with sheep erythrocytes to induce germinal centers. After 4h, 8h, 12h, 24h and 48h, mice were sacrificed and draining popliteal lymph nodes subjected to immunofluoresence staining using FITC-conjugated anti- GFP antibodies to detect Mfge8-EGFP and anti-CD68 antibodies to stain phagocytic macrophages.

After 4h, Mfge8-EGFP was distributed mainly under the lymph node capsule and within CD68⁺ sub-capsulary sinus macrophages (white arrows), which presumably phagocytosed Mfge8- EGFP labeled dying cells (Fig. 10, white arrows). After 8h, Mfge8-EGFP accumulated in certain areas, most likely on follicular dendritic cells (FDCs). It was also seen within CD68⁺ tingible body macrophages surrounding the FDCs (white arrows). There it was detectable for up to 24h. After 48h, no Mfge8-EGFP could be detected any more.

Mfge8 fusion protein accumulation seems to be restricted to the light zone of the germinal center (identified by DAPI staining). To stain injected Mfge8-EGFP, two different anti-Mfge8 antibodies were used: Clone 18A2 fails to detect Mfge8 that has been taken up by macrophages and therefore only stains extracellular Mfge8. A biotinylated polyclonal anti- Mfge8 antibody was used to also stain Mfge8 that was taken up by macrophages. This approach revealed that Mfge8 is taken up by CD68⁺ tingible body macrophages in the light zone, since Mfge8 stained with the polyclonal anti-Mfge8 antibody (middle) co-localized with CD68⁺ macrophages (right, Fig. 10B).

These findings confirm that dying cells can be visualized by immunofluorescence after in vivo administration of Mfge8-EGFP. These results further show that dying cells seem to accumulate on certain hot spots within the lymph node, most likely FDCs. Example 7: Real time live imaging of dying cells

The above results show the great potential to visualize dying cells using Mfge8 fusion proteins. It was thus analyzed whether it is possible to visualize dying cells in real time in live mice using 2-photon microscopy. As a proof of concept, 30pg Mfge8-mCherry were injected, together with 10pg FITC-labeled anti-BP-3 antibody to visualize FDCs, into the footpad of mice. Twenty-four hours later, intact lymph nodes were imaged using 2-photon microscopy. Mfge8-mCherry could be detected close to BP-3⁺ FDC networks. Furthermore, large intensely stained cells were detected, which most likely resemble macrophages that phagocytosed Mfge8-mCherry opsonized dying cells. These 2-photon results show that it is possible to visualize and track dying cells in vivo in real time using the Mfge8 fusion proteins of the present invention.

Example 8: Discrepancy between in vitro and in vivo stained dying cells Organ preparation for downstream application, like flow cytometry, typically damages many cells and can lead to high background when staining dying cells in vitro for FACS. To compare in vivo versus in vitro staining of dying cells, dying T cells, dendritic cells and macrophages were analyzed in spleens of LCMV-infected mice, since LCMV causes extensive cell death in these populations.

The frequency of live cells (Mfge8^" Annexin V^") versus in vitro stained dying cells, which are only stained with Annexin V (Annexin VMfge8^") and in vivo stained dying cells, which are double labeled with Annexin V and Mfge8 (Annexin V⁺ Mfge8⁺) was compared. Depending on the cell populations, drastic differences between in vivo and in vitro labeled dying cells can be observed. While the frequency of in vivo labeled dying CD8⁺ T cells was similar to in vitro labeled dying cells, the frequency of in vivo labeled CD4+ T cells was much lower than after in vitro labeling (Fig. 8 upper panel).

Pronounced differences were also seen in CD11c⁺ and CD11b⁺ dendritic cells and macrophages. While in vivo labeling almost exclusively stained MHC-Ι dendritic cells and macrophages, in vitro labeling also stained many MHC-II^" cells. These differences are most likely mainly caused by cell damage that occurred during the organ preparation and by releasing cells from their natural environment, which causes many cells to be stained in vitro, but not in vivo.

Example 9: Therapeutic potential of enzymes fused to Mfge8 Fusing enzymes to Mfge8 can be a valuable tool, to target enzymes and other therapeutic proteins to macrophages and other phagocytic cells, such as dendritic cells. Macrophages that lack Dnas2a cannot digest DNA from phagocytosed cells. Consequently, this DNA then accumulates inside the macrophages triggering pro-inflammatory responses that can lead to autoimmunity. By fusing Dnase2a with Mfge8, Dnase2a can be targeted to dying cells. The dying cells are taken up by macrophages together with Mfge8-Dnase2a, therefore the Dnase2a activity inside phagolysosomes can be restored with this method. Mfge8-Dnase2a fusion proteins maintain their full Dnase activity (Fig. 12).

Example 10: Depletion of phagocytes using fge8 fused to an inducible suicide gene

HSVTK is a suicide gene that kills a HSVTK-expressing cell upon treatment with ganciclovir. By opsonizing dying cells with Mfge8-HSVTK, phagocytes taking up these cells become HSVTK positive and susceptible to ganciclovir/HSVTK-mediated cell death. Mfge8-HSVTK fusion proteins maintain their full HSVTK activity (Fig. 13). Example 11 : Triggering of rapid B-ceii responses by an antigen fused to Mfge8

To determine whether Mfge8 fusion proteins comprising the fluorescent protein EGFP as an antigen are capable of eliciting an antibody response, mice were immunized intraperitoneally with 1.2 μΜ recombinant Mfge8-EGFP proteins as well as with 1.2 μΜ recombinant EGFP as control, precipitated in the adjuvant Alum (Imject Alum, ThermoFischer Scientific). Injection volume was 100μΙ. Alum and recombinant proteins were mixed 1 :1 and incubated at 4°C for 1 h under agitation prior to injection.

After five, seven and 14 days, serum was collected and the serum titers of the anti-EGFP antibodies were determined by ELISA (Fig. 14A). For this, microtiter plates were coated with 400ng/ml recombinant EGFP. After blocking, serum was added in the appropriate dilution and incubated for 1h at 37°C. After washing, isotype specific HRP-coupled secondary antibodies (total goat anti-mouse IgG (H+L chain) (Abeam, dilution 1 :20000) or lgG1 and lgG2b, all from Southern Biotech, dilution 1 :4000) were added and incubated for 1h. After washing substrate was added. Reaction was stopped with H2S04 and absorbance was read at 450nm in a VersaMax microplate reader (Molecular Devices). When available mouse anti-GFP antibodies were used as standard (mouse anti-GFP lgG1 clone 9F9.F9, Abeam; mouse anti-GFP lgG2b clone GT859, Abeam). For other isotypes, where no monoclonal antibody to be used as a standard was available, a reference serum fornormalization was used. All modes of immunization resulted in an IgM response, however, when Mfge8-EGFP was used, EGFP- specific IgM titers where approx. 6.5-fold higher compared to the immunization with recombinant EGFP. Furthermore, when EGFP-specific total IgG titers were measured, a dramatic difference between Mfge8-EGFP and EGFP only immunization was observed: total IgG titers in mice that received recombinant EGFP were very low and close to the detection limits, whereas EGFP-specific IgG was easily detectable and significantly higher in Mfge8- EGFP immunized mice as early as on day five after immunization.

Similar results were obtained when specific isotypes were quantified. On day seven, lgG1 titers were approx. 0.4pg/ml in Mfge8-EGFP immunized mice, but below the detection limit in EGFP immunized mice. On day 14 after immunization, lgG1 levels were similar in both groups. lgG2b titers were 0.05Mg/ml on day seven after Mfge8-EGFP immunization and below detection limit in mice that where immunized with EGFP. On day 14, lgB2 levels were still significantly higher in Mfge8-EFGP immunized mice compared to EGFP-immunized mice. To determine why the production of isotype-switched antibodies was so dramatically increased when the antigen was fused to Mfge8, histological analyses of the spleen was performed to assess the location of the antigen. Mfge8-EGFP and EGFP was administered i.v. to naive mice. In addition, mice that received an immunization with EGFP precipitated in Alum two weeks earlier (immune mice), again received an EGFP injection i.v.

12 hours later, mice were sacrificed and spleens analyzed by immune fluorescence. While Mfge8-EGFP was readily detectable on CD21⁺ follicular dendritic cells (FDCs) in naive mice, no EGFP could be detected in naive mice when EGFP was injected that was not fused to Mfge8. However, injection of EGFP that was not fused to Mfge8 into immune mice resulted in that EGFP was detectable on FDCs (Fig. 15). It is well known that FDCs efficiently trap antigen in the form of immune complexes, which consist of antibody, antigen and complement. This is important for activation of B cells, germinal center formation and affinity maturation (Aguzzi, A., Kranich, J. & Krautler, N. J. Follicular dendritic cells: origin, phenotype, and function in health and disease. Trends in immunology. 35(3): 105-13 (2013)). In naive mice, no anti-EGFP antibodies are present, hence no accumulation on FDCs can be observed. It is very striking that when EGFP is fused to Mfge8, FDCs can bind the antigen even in the absence of antibodies, since naive mice showed accumulation of EGFP on FDCs. Thus, by fusing an antigen to Mfge8, the antigen becomes rapidly available in the B cell follicle, where it can activate naive B cells in an immune complex independent manner. Without wishing to be bound by theory, it is postulated that antigen fused to Mfge8 binds to apoptotic cells, which are then captured by cells in the marginal sinus and subsequently transported into the follicle, where they are transferred onto FDCs. Subsequently, Mfge8 is recognized by tingible body macrophages through integrins which recognize the RGD-motif within Mfge8 (Hanayama, R. et al., Nature 417, 182-187, 2002 (doi:10.1038/417182a [pii]); and Hanayama, R. et al., Science 304, 1147-1150, 2004 (doi:10.1126/science.1094359304/5674/1147 [pii])). Tingible body macrophages take up the Mfge8 fusion proteins bound on follicular dendritic cells (Fig. 10B). Tingible body macrophages have important immunoregulatory functions (Smith, J. P. et al., Dev. Immunol. 6, 285-294 (1998)), therefore it can be expected that this is an important process during the B cell response.

Example 12: Mfge8-EGFP detects apoptotic vesicles bound to intact viable cells

Prior to FACS analysis mice were injected with 100pg Mfge8-EGFP. 30min later, spleens were removed and single cells supsensions prepared. After erythrocyte lysis, splenocytes were stained with appropriate antibodies and analyzed by imaging flow cytometry. During the analysis by image flow cytometry, Mfge8-EGFP⁺ cells were also detected, which did not show apoptotic morphology. For example in other B cell subsets, such as CD19⁺CD21^highCD23^l0W marginal zone (MZ and CD19⁺CD21 ^|0WCD23⁺ follicular (FO) B cells, where we expected very little apoptosis, many cells nevertheless were Mfge8⁺ (Fig. 17). We found that approx. 25% MZ B cells and 2.5% FO B cells B cells were Mfge8⁺. Approx. 1% and 10% of CD4 and CD8 T cells were Mfge8+, respectively (Fig. 16A).

Older studies suggested that most MZ B cell would exhibit PS exposed on their surface as a consequence of positive selection of B cells, which would not be indicative of cell death, but rather be due to membrane alterations and transient exposure of PS upon B cell receptor triggering (Dillon, S. R., et. al., Journal of Immunology (Baltimore, Md.: 1950) 166, 58-71 (2001 )). However, when Mfge8⁺ B cells were examined by imaging flow cytometry, the intense staining with Mfge8-EGFP on most MZ and some FO B cells was due to association of live B cells with Mfge8⁺ extracellular vesicles (EVs), rather than global Mfge8-staining of B cell membranes (Fig. 16B). Those B cells showed vital morphology in contrast to apoptotic cells (Figure 16B). Therefore abundant binding of EVs by B cells, especially MZ B cells, is responsible for rendering B cells Mfge8⁺.

EVs comprise several different types of membrane vesicles and can either be exosomes (30- 100nm), microvesicles (IOOnm-Ι μιη) or apoptotic bodies ranging (1-5pm) (Gyorgy, B. et al., Cellular and molecular life sciences : CMLS 68, 2667-2688, doi: 0.1007/s00018-011-0689-3 (201 1 )). All three kinds of EVs can have exposed PS on their surface as part of the their membrane and are therefore bound by Mfge8-EGFP. The EVs attached to Mfge8-EGFP⁺ cells are quite large and thus resemble apoptotic bodies or microvesicles, while the general increase of the EGFP-MFI suggests that also smaller EVs, such as exosomes, which would be stained with lower intensity, are bound by B cells.

To confirm that these attached Mfge8-EGFP⁺ signals are indeed EVs, purified EVs were generated from apoptotic thymocytes that were labeled with the membrane dye PKH26 using the PKH26 fluorescent linker kit (Sigma-Aldrich). For this, thymocytes were washed with RPMI and pelleted and supernatant carefully removed. Then cells were resuspended in 1 ml of Diluent C. Then 1 ml of the 2x DyeSolution (1 ml Diluent C containing 4μΙ PKH26) was added and mixed. After 5 min the reaction was stopped by adding 2ml RPMI containing 10% FCS, followed by 3 washes with FCS-containing RPMI. Then apoptosis was induced in these PKH26-labeled thymocytes by adding I g/ml staurosporine for 3h. Then remaining cell bodies were removed by slow speed centrifugation (500g for 10min). To collect EVs, supernatant was subjected to ultracentrifugation an 100,000g for 90min. Pelleted EVs were resuspended in PBS and injected intravenously into mice. One hour later, mice were sacrificed and their splenic B and T cells analyzed by conventional (Fig. 18A) and imaging (Fig. 18B) flow cytometry. Surprisingly, all B and T cells showed strongly increased PKH26 MFI, indicating that virtually all cells bound iv administered EVs in vivo. MZ B cells showed a strongly increased PKH26 MFI (1 146± MZ B cells FO B showed a slightly lower increase in PKH26 MFI (816±0 B cells (Fig. 18A). The binding capacity of CD4⁺ T cells (MFI 797±T cells (MFI⁺ T cells (MFI 846±46cells (similar to the binding capacity of FO B cells. In order to assess whether Mfge8 is involved in the capturing of EVs by B and T cells, EVs were injected into MfgeS^^' mice or coated the EVs with recombinant Mfge8 prior to injection (Fig. 18A). While no differences were observed in EV binding in Mfge&^ animals as compared to controls, coating of EVs with recombinant Mfge8 completely abrogated binding of EVs to cells (Fig. 18A). This strongly indicates that B and T cells bind EVs through PS.

Example 13: Discrimination between apoptotic cells and cells with attached EVs

Having shown that Mfge8-EGFP efficiently labels not only dying cells in situ, but also EVs, the next aim was to reliably discriminate cells with attached EVs from dying cells. For this, the AMNIS imaging flow cytometry software IDEAS was used.

To do so, truth populations were first defined for both, cells with bound EVs and dying cells. The truth population for cells with attached EVs were splenic lymphocytes from mice which were injected with PKH26-labeled EVs (Fig. 19A left) as described in previous experiments (Fig. 18). The truth population of apoptotic cells was defined as thymocytes double positive for active caspase-8 and Mfge8-EGFP after a 2h staurosporine treatment (Fig. 9A right).

Using the IDEAS software (EMD Millipore), two features were then identified that achieved a clear separation of dying cells and cells with attached EVs. One feature was the ratio between the brightfield (BF) area using the standard mask and the area of the staining used to identify EVs (PKH26 or Mfge8-EGFP) or dying cells (Mfge8-EGFP) using the peak mask (Peak(M02, Mfge8, Bright, 1.6) or Peak(M03, PKH26, Bright, 1.6). Cells with attached vesicles had an area ratio >10). The second feature was the minor axis intensity of PKH26 or Mfge8-EGFP. Dying cells had a minor axis intensity value >3, cells with attached vesicles a value <3.

Hence, cells with attached EVs had a large BF/PKH26 area ratio of >10 and a minor axis intensity of PKH26 <3 and apoptotic cells had a small BF/Mfge8-EGFP area ratio <10 and a large minor axis intensity of Mfge8-EGFP >3 (Figure 19B and 19C). Using these features on caspase-8^"Mfge8-EGFP^" non-staurosporine treated thymocytes revealed that also among these cells some had bound small EVs weakly stained with Mfge8-EGFP (Figure 19B and 19C middle). Taken together, using Mfge8-EGFP and specific IDEAS software analysis masks and features allowed the discrimination of true apoptotic cells from those which were live cells, but had bound PS⁺ EVs.

Then these features were used to analyze cell death and EV trapping among B and T cell populations in the spleen. 30min after injection of 100pg Mfge8-EGFP intravenously, spleens were removed, single cell suspensions prepared and stained with appropriate antibodies. Then cells were analyzed by imaging flow cytometry A substantial amount of cell death could only be observed among CD19⁺lgD^lowGL7⁺CD95⁺ GC B cells, were approx. 37% of the Mfge8- EGFP⁺ cells were apoptotic and 44% attached to EVs (Figure 20). This was in stark contrast to the total splenic CD19⁺ B cell population where only about 13% of the Mfge8-EGFP+ cells were apoptotic (Fig. 21 ). Furthermore, only very few dying CD4⁺ and virtually no dying CD8⁺ T cells could be detected. Also vesicle binding was much lower in T cells than in B cells (Fig. 21). Taken together, the results show that fluorescent Mfge8-EGFP fusion proteins can be used to label not only dying cells in situ, but also cells that had captured EVs, as well as EVs not associated to cells. It was further found, that almost all splenic B cells, but only few T cells capture EVs, which renders them PS positive. Example 14: fge8-EGFP accumulates of FDCs

The splenic distribution of iv injected Mfge8-EGFP was analyzed by immunofluorescence at two different time points. One hour after injection Mfge8-EGFP was confined to the MZ and red pulp. Most Mfge8-EGFP staining localized with SIGN-R1+ (Fig. 15) or CD169+ (not shown) macrophages in the MZ, indicating that these phagocytose Mfge8-labeled dying cells or vesicles (Figure 15). Recombinant EGFP that was injected iv could not be detected (Figure 15).

12 hours after injection, Mfge8-EGFP was cleared from the marginal zone and only visible in the B cell follicle where it accumulated on CD21/35+ follicular dendritic cells (FDCs) (Fig. 15). Mfge8 is also known as the FDC marker FDC-M1 and histological stainings using anti-Mfge8 antibodies stain the FDC network (Kranich, J. et al. Follicular dendritic cells control engulfment of apoptotic bodies by secreting Mfge8. J Exp Med 205, 1293-1302, (2008)). The results show that FDCs extensively bind Mfge8. Given that B cells extensively capture Mfge8+ EVs (Figure 17), it is plausible that B cells transport EVs from the MZ into the B-cell follicle, where they hand them over to FDCs, as they do with immune complexes (ICs) (Kranich, J. & Krautler, N. J. How Follicular Dendritic Cells Shape the B-Cell Antigenome. Frontiers in immunology 7, 225, 2016 (doi:10.3389/fimmu.2016.00225)).

Claims

An Mfge8 fusion protein comprising a (poly)peptide of interest covalently bound by a linker to an Mfge8 protein, wherein the Mfge8 fusion protein has a phosphatidylserine (PS)-binding activity and, optionally, an RGD-binding activity.

The Mfge8 fusion protein according to claim 1 , wherein the (poly)peptide of interest is bound to the C-terminus of the Mfge8 protein.

The Mfge8 fusion protein according to claim 1 or 2, wherein said (poly)peptide of interest is

(i) a reporter protein, preferably a fluorescent or a bioluminescent reporter protein; or

(ii) a recognition sequence for enzymatic modification, preferably a recognition sequence for a biotin ligase or a recognition sequence for a sortase.

The Mfge8 fusion protein according to 1 or 2, wherein said (poly)peptide of interest is an enzyme, preferably an enzyme selected from the group consisting of thymidine kinase, cytosine deaminase, cytochrome P450 (CYP), inducible caspase 9 and nitroreductase.

The Mfge8 fusion protein according to claim 1 or 2, wherein said (poly)peptide of interest is a therapeutic (poly)peptide.

The Mfge8 fusion protein according to claim 5, wherein said therapeutic (poly)peptide is a Dnase, an antigen or a radioactive compound.

A composition comprising at least one Mfge8 fusion protein according to any one of claims 1 to 6.

The composition of claim 7, wherein the composition is a vaccine, wherein the (poly)peptide of the at least one Mfge8 fusion protein serves as an antigen and wherein the at least one Mfge8 fusion protein has an RGD-binding activity.

The Mfge8 fusion protein according to any one of claims 1 to 6 or the composition according to claim 7 or 8 for use in medicine. The Mfge8 fusion protein according to any one of claims 1 to 6 or the composition according to claim 7 or 8 for use in the treatment of cancer and/or for monitoring the success of such a therapy.

The Mfge8 fusion protein according to any one of claims 1 to 6 or the composition according to claim 7 or 8 for use in the diagnosis or treatment of a disease associated with cell death or with a defect in phagocytosis.

A method of detecting dying cells or extracellular vesicles having phosphatidylserine (PS) on the extravesicular surface, the method comprising:

(a) labeling cells suspected to be dying or extracellular vesicles suspected to have phosphatidylserine (PS) on the extravesicular surface by contacting the cells or the extracellular vesicles with the Mfge8 fusion protein according to any one of claims 1 to 3; and

(b) determining the presence or absence of cells or vesicles labeled with the Mfge8 fusion protein.

A method of analyzing phagocytosis of dying cells, the method comprising:

(a) labeling dying cells by contacting the cells with the Mfge8 fusion protein according to any one of claims 1 to 3;

(b) detecting the uptake of the labeled cells of step (a) by phagocytes.

A method of diagnosing a disease associated with a defect in phagocytosis, wherein the method comprises detecting the amount of labeled dying cells present in phagocytes obtained from a subject suspected to suffer from a disease associated with a defect in phagocytosis, wherein the dying cells have been labeled by administering the Mfge8 fusion protein according to any one of claims 1 to 3 to the subject;

wherein an altered amount of labeled dying cells detected in the phagocytes obtained from the subject suspected to suffer from a disease associated with a defect in phagocytosis compared with

(a) the amount of labeled dying cells detected in the same type of phagocytes obtained from a control representative of (a) subjects) not afflicted by the disease; or

(b) the amount of labeled dying cells detected in a different type of phagocytes obtained from the same subject,

indicates that said subject is suffering from or is at risk of developing a disease associated with a defect in phagocytosis.

5. A method of determining the effectiveness of a therapeutic treatment of cancer or myocardial infarction, the method comprising:

(a) detecting the amount of labeled dying cells present in a subject suspected to suffer from cancer or myocardial infarction before it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein according to any one of claims 1 to 3 to the subject; and

(b) detecting the amount of labeled dying cells present in said subject during or after it received the therapeutic treatment, wherein the dying cells have been labeled by administering the Mfge8 fusion protein according to any one of claims 1 to 3 to the subject;

wherein an altered amount of labeled dying cells detected during or after receipt of the therapeutic treatment compared with the amount of labeled dying cells detected before the therapeutic treatment indicates that the therapeutic treatment is effective.

6. The method of claim 14, wherein the Mfge8 fusion protein comprises a radioisotope, preferably 18F or 64Cu, that has been added site-specifically to the Mfge8 protein via a sortase A recognition sequence.