Interaction of Draxin and γ-netrins

Abstract

This invention relates to extracellular protein-protein interactions and their possible therapeutic uses. More particularly, this invention describes the interaction between Draxin, particularly fragments binding to γ-Netrins comprising SEQ ID NO.:1, 2 or 3, and variants thereof, with γ-Netrins, and the use of this interaction to disrupt γ-Netrin/Netrin receptor interactions. The invention also relates to diagnostic and/or therapeutic uses of Draxin or fragments or variants thereof, as well as to an antibody against Draxin inhibiting binding of Draxin to γ-Netrins. Further, the invention relates to fragments of γ-Netrins, in particular Draxin-binding Netrin1-fragments comprising SEQ ID NO.: 51 and variants thereof, as well as to an antibody against γ-Netrins inhibiting binding of γ-Netrins to Netrin receptors.

Claims

1. Draxin-binding peptide comprising (i) at least 20 consecutive amino acids from the sequence KACDCHPVGAAGKTCNQTTGQCPCKDGVTGITCNRCANGYQQSRSP IAPCIKIPIAPP (SEQ ID NO.: 51) or (ii) a variant thereof having a sequence identity of at least 90%, or at least 95% to SEQ ID NO.: 51; wherein said peptide has a length of up to about 200 amino acids and is fused to a heterologous peptide or polypeptide.

2. The peptide according to claim 1, wherein said peptide comprises a sequence selected from the group consisting of SEQ ID NO.: 45, SEQ ID NO.: 48, SEQ ID NO.: 65 and SEQ ID NO.: 77, or a variant thereof having a sequence identity of at least 90%, or at least 95% thereto.

3. The peptide according to claim 1, in combination with a carrier suitable for use in medicine.

4. The peptide according to claim 1, wherein said variant contains at least one non-naturally occurring substitution modification relative to SEQ ID NO.:51.

5. The peptide according to claim 1, wherein said peptide is fused to a functional fragment of an immunoglobulin (Ig).

6. The peptide according to claim 5, wherein said functional fragment of an immunoglobulin (Ig) is an Ig Fc fragment.

7. The peptide according to claim 6, wherein said Ig Fc fragment is a human Ig Fc fragment.

8. The peptide according to claim 7, wherein said human Ig Fc fragment is a human IgG Fc fragment.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1. Draxin directly binds Netrin1.

(2) A) AVEXIS screen results showing that the DraxinA prey protein, in addition to the positive control bait Matn4 (+), only binds to Netrin1a in a protein library consisting of 171 bait proteins. B) Interaction network of the zebrafish Netrin1 and Draxin paralogs. Consecutive screen results show that the interaction is also conserved for the paralogs.

(3) FIG. 2. The binding of Draxin to Netrin1 is conserved for the human homologs and detectable across human and zebrafish proteins.

(4) Heatmap showing results from the AVEXIS assay. The screen has been performed in both bait-prey orientations using pentameric prey proteins. Absorbance at 486 nm (A486 nm) has been measured after 1-hour incubation (black: A486 nm>0.1=binding, grey: A486 nm 0.08-0.1=weak binding, 2 repeats). B) Network view of the results.

(5) FIG. 3. Draxin protein alignment using the Clustal W method.

(6) The alignment of human, mouse, chick, and zebrafish DraxinA and DraxinB protein shows that the N-terminal half of the protein is poorly conserved. In the most C-terminal part of the protein a conserved 10-cysteine containing region can be found, which resembles the cysteine-rich region present in Dickkopf proteins.

(7) FIG. 4. A conserved 21aa DraxinA derived peptide is sufficient for the binding to Netrin1a.

(8) A) By using a set of truncated and deletion containing monomeric DraxinA preys the binding interface of DraxinA to Netrin1a has been mapped down to a 21 amino acid region (aa232-aa252). The protein fragments binding to Netrin1a are indicated in black, weak binding ones in grey, and non binding fragments are depicted in white. B) and C): Using a similar approach the DraxinA binding interface in Netrin1a has been narrowed down to the third EGF-domain containing region (aa401-aa458).

(9) FIG. 5. Multiple species alignment of the Draxin derived 21 aa peptide sequence.

(10) The figure shows that the Netrin1 binding peptide is highly conserved in vertebrates.

(11) FIG. 6. DraxinA inhibits the binding of Netrin1a to Netrin receptors.

(12) A) Schematic representation of the AVEXIS based competition assay. B) Purified full-length Draxin inhibits the binding of of the Netrin1a bait to the DCC-prey proteins. C) This effect can also be seen for the Unc5b and Neo1 (D) netrin receptors. Equal amounts of BSA (D) are not able to inhibit the binding between Netrin and Neo1. E) Draxin is not able to interfere with the binding of RGMc to Neo1. RGMc is another known Neo1 ligand. (% binding: binding with inhibitor/binding without inhibitor ×100%, error bars indicate mean±s.d.; n=4)

(13) FIG. 7. A 21 amino acid fragment of Draxin is sufficient to outcompete Netrin/Netrin receptor interactions.

(14) A) Full-length DraxinA-hFc and the 21aa peptide fused to the hFc region are able to outcompete DCC for binding to Netrin1a. In contrast the hFc fusion of full-length DraxinA with a deletion of aa231-252 is not able to compete for binding. None of the 3 DraxinA-hFc fusion protein versions is able to block the binding between other known tested receptor-ligand pairs: (B) Cntn1a/Ptprz1b, (C) Vasn/Islr2, and (D) EphB4a/EphrinB2a. (% binding: binding with inhibitor/binding without inhibitor ×100%, error bars show mean±s.d.; n=3)

(15) FIG. 8. The binding of the 21aa DraxinA derived peptide is highly specific.

(16) A) A pentameric DraxinA aa232-252-prey protein has been screened against a library consisting of 141 bait proteins. We only observed binding to the Netrin1a-bait. The positive control (+) corresponds to Matn4. B) The DraxinA aa232-252-bait has been screened against 191 pentameric prey proteins and only showed binding to the Netrin1a prey.

(17) FIG. 9. Draxin outcompetes receptor bound Netrin1a.

(18) A) Scheme of the experimental design. B) The inhibitory effect of DraxinA is not decreased by already preformed Netrin1a/DCC complexes.

(19) FIG. 10: Protein alignment of the third EGF domain of Netrins.

(20) Protein alignment of the third EGF domain of human Netrin1, Netrin3, Netrin4 and zebrafish Netrin1a and Netrin1b shows that this domain is highly conserved for the γ-chain netrins but not for Netrin4 which belongs to the laminin1 β-chain derived netrins. In agreement with the protein alignment data our AVEXIS binding data indicate that only γ-chain derived Netrins can bind to Draxin.

(21) FIG. 11: In vivo detection of the Draxin-Netrin1a interaction in zebrafish embryos.

(22) A) Schematic illustration of the assay design. mRNAs encoding the indicated fluorophore tagged genes were injected into one-cell stage zebrafish embryos and imaged at sphere stage (4 hours post fertilization (hpf)). The imaging plane corresponded to a region approximately 15 μm beneath the enveloping layer of the embryos.

(23) B) Single section confocal images of the embryos. (Ba, Ba′, Ba″) embryos injected with 100 pg Draxin-super folder GFP (sfGFP) mRNA displayed uniform distribution of Draxin-sfGFP protein in the extracellular space. In contrast injection of 100 pg of Netrin1a-sfGFP mRNA (Bb, Bb′, Bb″) resulted in dense membrane associated speckles positive for Netrin1a-sfGFP protein. In (Ba) and (Bb) memRFP has been used to label the cell surface. Upon co-injections of 200 pg Draxin-sfGFP mRNA and 200 pg of Netrin1a-mCherry mRNA (Bc, Bc′, Bc″) Draxin-sfGFP and Netrin1a-mCherry proteins co-localize into membrane associated spots. (n=7; arrowheads point to examples of co-localization; scale bars correspond to 10 μm).

(24) FIG. 12: In situ detection of the Draxin-Netrin1a interaction in zebrafish using an affinity probe.

(25) A) Schematic illustration of the experiment. A Draxin.sub.aa209-284-hFc-fusion protein was generated in HEK293-6E cells as a probe to detect endogenous netrins. Mildly fixed 48 hpf embryos were incubated with the affinity probe (Draxin.sub.aa209-284-hFc) and the signal detection has been carried out by using a fluorophore tagged anti human IgG antibody.

(26) B) Results of the in situ detection experiments. The signal from Draxin.sub.aa209-284-hFc-fusion probe was detectable in the floor plate region in wild type (also abbreviated as wt) fish (Ba), and was not detectable in netrinla and netrinlb double-knockdown embryos (Bb). The shh:GFP transgenic line has been used for the experiments to visualize floor plate cells. Arrowheads point to the Draxin.sub.aa209-284-hFc derived signal. Scale bar in (Ba″), 20 μm; applies to all panels. (n>10)

(27) FIG. 13: Heatmap depicting binding results between human DRAXIN and Netrin signaling system members.

(28) AVEXIS was used to test pairwise binding events in both bait/prey orientations between human DRAXIN, the derived 21 amino acid peptide, human Netrin family members, and two representative Netrin receptors (DCC, UNC5B). The Matn4-bait served as an internal prey-protein control and was used for normalization of the A486 nm values; conditioned medium (CM) serves as negative control; n=3.

(29) FIG. 14: Surface Plasmon Resonance Analysis of Draxin Binding.

(30) Binding of Draxin to immobilised recombinant human Netrin1, UNC5B, and DCC was monitored using Surface Plasmon Resonance experiments on a Siacore 3000 instrument. See text for details.

(31) The invention will be further illustrated by the following examples.

EXAMPLES

(32) Methods

(33) AVEXIS Based Library Screen

(34) To detect extracellular protein-protein interactions in a high-throughput manner we used the AVEXIS assay as described in Bushell et al. (Bushell et al., 2008) with minor modifications.

(35) In brief: A zebrafish protein library enriched for mainly in neuronal tissues expressed secreted proteins and extracellular domains of cell surface proteins has been assembled. The library consists of prey and bait proteins. Preys are composed of the extracellular domain (ECD) of interest followed by a CD4 tag (rat Cd4d3+4) and a pentamerization domain derived from the rat Comp protein followed by β-lactamase. For the bait proteins the ECDs are fused to a CD4 tag and a biotinylation peptide. All proteins for the screen have been expressed by transient transfection of Human Embryonic Kidney (HEK293-6E) cells (Durocher et al., 2002) grown in Freestyle medium (Invitrogen) containing 1% FCS. Supernatants have been harvested 6 days post transfection. The bait proteins have been dialyzed against HBS (140 mM NaCl, 5 mM KCl, 2 mM CaCl.sub.2), 1 mM MgCl2, 10 mM HEPES, pH 7.4) to remove free biotin. The proteins in the supernatants have been quantified and normalized.

(36) For the AVEXIS screen the supernatant dilution factors have been adjusted to values allowing faithful detection of the interaction between Vasn (Slit-like2) and Islr2 (Söllner and Wright, 2009) with determined KD of 12 μM and a very short half-life (t1/2 0.16 s) (in preparation). The bait proteins have been immobilized on streptavidin coated 96 well microtiter plates (one bait/well) and incubated for 1 hour at room temperature. After 3 wash steps using HBS as wash buffer the baits have been probed by 50 μl of normalized prey proteins. After one-hour incubation the non-bound preys have been washed away by 2 washes with HBST (0.05% Tween) followed by two HBS washes. Then 50 μl of nitrocefin/well (0.1 mg/ml) has been added and incubated for 1 h at room temperature. Then the absorbance at 486 nm of each well has been measured using a μQuant spectrophotometer (BIO-TEK Instruments, INC). As a positive control for the prey proteins we used the Matn4 ECD as mono-biotinylated bait protein. Matn4 has been shown to bind to the coiled-coil pentamerization domain of Comp (Mann et al., 2004), which is present in all recombinant pentameric prey proteins of our AVEXIS library.

(37) In the primary screen interactions were ‘called’ if the absorbance (at 486 nm) of a well after 1 h of incubation was ≥0.1 and 3 Sigma above the standard deviations of the mean of all wells. All interactions detected in the primary screen have been retested in a validation screen by using independently produced batches of proteins.

(38) Domain Mapping Experiments

(39) For the binding interface mapping experiments we used monomeric prey proteins. The concentration of the monomeric preys has been determined as described for the AVEXIS assay (βlactamase enzymatic activity). After normalization, the monomeric preys have been screened against a set of proteins from the library to identify and remove promiscuous binders caused for example through improper domain boundary design. Both prey and bait orientations were tested in the domain mapping screen and two repeats for each orientation had been carried out.

(40) Protein Purification

(41) His-tagged full-length zebrafish Draxin protein (DraxinCD4d3+46×His) has been expressed in HEK293-6E cells and affinitiy purified from tissue culture supernatants using HisTrap HP columns (GE healthcare). The correct size of the purified protein has been checked on a protein gel.

(42) AVEXIS Based Competition Assay

(43) The procedure of the competition assay is based on the AVEXIS assay. Netrin1 receptors were used as prey proteins together with 6×Histagged purified Draxin (potential antagonist), and probed against Netrin1a bait proteins. The indicated concentrations of the potential inhibitors have been added together with prey proteins. For the competition tests with purified Draxin the concentration of the Netrin receptor prey proteins has been adjusted to an identical threshold binding concentration.

(44) hFc Fusion Protein Normalization for the Competition Assay

(45) We determined the concentration of ECD-hFc fusion proteins in tissue culture supernatants by ELISA, using the human IgG Fc fragment (Calbiochem) as a reference. Dilution series of the ECD-hFc containing supernatants have been incubated over night at 4° C. on 96 well Maxisorp plates (Nunc). After 3 PBS washes the plates were blocked with 0.5% BSA containing PBS (1 h). After additional 2 PBS washes and 1-hour incubation with an anti-human IgG (Fc specific) antibody fused to alkaline phosphatase (SIGMA) the plates have been washed 3 times with PBS. The detection has been carried out by addition of 50 μl/well of the AP substrate p-nitrophenylphosphate (SouthernBiotech). The substrate turnover has been determined by measuring the absorbance at 405 nm.

(46) In Vivo Binding Assay

(47) To reveal whether the interaction between Draxin and Netrin1a is detectable in vivo, an mRNA overexpression assay was designed to visualize the localization of the two proteins in zebrafish embryos. Constructs of full-length zebrafish Draxin and Netrin1a fused with fluorescent proteins were generated by using the Gateway® cloning system (Life Technologies).

(48) Before used in the in vivo detection assay, the coding sequences of the generated fusion proteins were cloned into AVEXIS plasmids, expressed as preys and tested for activity against bait proteins of the corresponding binding partner. Following constructs were selected for the in vivo binding test: Draxin and Netrin1a C-terminally fused to superfolder-GFP (Draxin-sfGFP, Ntn1a-sfGFP) and Netrin1a C-terminally fused with mCherry (Ntn1a-mCherry). The corresponding capped mRNAs were synthesized using the mMESSAGE mMACHINE SP6 or T7 Transcription Kit (Ambion) according to manufacturer instructions. For the injections, zebrafish embryos were dechorionated using 1 mg/ml Pronase (Roche, 11459643001) and then injected with 1 nl mRNA into the cell center at one cell stage. Draxin-sfGFP or Netrin1a-sfGFP mRNAs were injected at 100 pg/embryo in combination with 10-15 pg/embryo of membrane-tagged RFP (mRFP) to label the cell membranes. The same amount of mRNA (100-200 pg/embryo) was injected in the Draxin-sgGFP and Netrin1a-mCherry coexpression experiments. The injected embryos have been cultured at 28.5° C. in agarose-coated dishes. At sphere stage (4 hpf), embryos were immobilized in 1% low-melting-point agarose in glass-bottom Petri dishes with the animal pole facing the coverslip. The imaging plane corresponded to a region approximately 15 μm beneath the enveloping layer of the embryos. Single plane confocal images of the embryos were taken using a Zeiss LSM 780 NLO microscope.

(49) In Situ Detection of Draxin Binding Partners

(50) A netrin binding fragment of Draxin (aa209-284) fused to the Fc region of human IgG (Draxin.sub.aa209-284-hFc) has been expressed in HEK293-6E cells and used as an affinity probe to detect binding partners in zebrafish embryos. Draxin.sub.aa209-284-hFc in situ staining has been done in whole mount wild type and netrin-1 knockdown zebrafish embryos. At 46 hpf wild type zebrafish embryos were dechorionated by for 2 hours at RT incubation in 0.1 mg/ml Pronase. At 48 hpf embryos were prefixed for 10 min at RT in 4% (w/v) paraformaldehyde (PFA) containing 1% Triton X-100 (v/v). After 3 washes (each 20 min) with PBS+1% Triton X-100 the embryos were blocked for 4 h at RT in PBS containing 0.2% BSA and 0.5% Triton X-100. Overnight incubation at 4° C. with HEK293-6E supernatant containing the Draxin-hFc fusion protein was followed by 3 short washes (10 min each) with PBST. Subsequently the embryos were post-fixed in 4% PFA (4 h at RT or overnight at 4° C.), rinsed shortly 3 times with PBST and incubated for 4 h at RT with an Alexa Fluor 568 goat anti human IgG antibody (Invitrogen, 1:250 dilution). After 3 washes (30 min each) in PBST the embryos were mounted in glycerol for visualization using a Zeiss LSM 510 microscope. A shorter version of Draxin containing the Netrin1a binding site had to be used because the full-length version of Draxin-hFc caused uniform background staining in zebrafish embryos probably by unspecific binding to glycosaminoglycans (GAGs) present on cell surfaces. A series of additional control ECD-hFc proteins have been tested, only Draxin.sub.aa209-284-hFc displayed binding to the extracellular space of the floor plate.

(51) Knockdown of netrin1a and netrin1b in Zebrafish Embryos

(52) Morpholino antisense oligonucleotides (Gene-Tools) have been used to generate zebrafish with reduced netrin protein expression levels. The following morpholino sequences have been used to knockdown ntn1a ATGATGGACTTACCGACACATTCGT-3′, SEQ ID NO.: 80) and ntn1b (5′-CGCACGTTACCAAAATCCTTATCAT-3′, SEQ ID NO.: 81). In previous studies both morpholinos had been shown to efficiently knockdown the corresponding genes (Kastenhuber et aL, 2009; Suli et al.; 2006).

(53) Surface Plasmon Resonance (SPR)

(54) Surface Plasmon Resonance (SPR) experiments were performed on a Biacore 3000 (GE Healthcare) at 25° C. using a SA sensor chip in 0.01 M HEPES, pH 7.4, 0.15 M NaCl, 0.005% Surfactant P20 (HBS-P) running buffer at a flow rate of 30 μl/min. The instrument was used according to manufacturer's instructions.

Example 1: DraxinA Physically Interacts with Netrin1a

(55) Using a protein-protein interaction screen assay, designed to identify direct interactions within a protein library consisting of secreted proteins and extracellular domains of cell surface proteins (Bushell et al., 2008), we carried out a large-scale screen involving more than 40,000 binding experiments. The library we used for the screen was strongly enriched for zebrafish proteins known to be expressed in the developing nervous system. During this screen we identified a novel interaction between two secreted proteins with known function in axon guidance, Netrin1a and DraxinA. In the primary screen a DraxinA prey protein has been tested for binding against a library consisting of 171 bait proteins, including a positive control. The DraxinA prey protein specifically bound to the Netrin1a bait (FIG. 1A) and did not bind to any additional proteins of the library. The interaction has been confirmed in both bait-prey orientations in a validation screen using new protein samples. Interestingly, both netrin1 and draxin are duplicated in zebrafish. In subsequent binding assays we were able to show that the Netrin1/Draxin interaction is also conserved for the paralogs Netrin1b and DraxinB (FIG. 1B).

(56) The AVEXIS assay is able to detect very transient and weak interactions due to the avidity effect caused by the use of pentameric prey proteins. Hence, in order to test whether the interaction between Netrin1a and Draxin is transient or rather stable we used monomeric prey proteins and probed them against the corresponding binding partners. Using this approach, we confirmed the interaction between Netrin1a and Draxin suggesting that this interaction is based on strong binding between the two proteins.

Example 2: The Interaction Between Draxin and Netrin1 is Conserved for the Human Homologs

(57) Next we asked the question whether the interaction between Netrin1 and Draxin is conserved. By using the corresponding human homologs NTN1 and DRAXIN we were able to show that the interaction is indeed conserved. In addition, we observed that zebrafish Netrin1a was able to bind to human Draxin and vice versa (FIG. 2). This strongly indicates that this newly identified interaction is conserved within vertebrate species underscoring the biological relevance of this interaction.

Example 3: The Netrin1 Binding Region of Draxin has been Mapped Down to a 21Aa Motif

(58) Next, we narrowed down the region in DraxinA required for binding to Netrin1a. Zebrafish DraxinA consists of 360 amino acids (aa). The first 23 aa are part of the signal peptide. This sequence is followed by a poorly conserved N-terminal half of the protein (FIG. 3). In contrast, the C-terminal half of the protein is highly conserved and ends with a 10 cysteine-containing domain (aa285-aa360). In terms of cyteine spacing this domain is similar to domains present in the Wnt antagonist Dkk1 (Glinka et al., 1998).

(59) To map down the Netrin1a binding region in DraxinA we generated a series of DraxinA truncations and deletions and tested them for binding against Netrin1a (FIG. 4). Using this approach we were able to narrow down the binding region to a 21aa DraxinA protein fragment. In addition, a full-length version of DraxinA lacking these 21aa completely lost the ability to bind to Netrin1a. Additional removal of 5aa from the N-terminal or C-terminal end of the 21aa stretch caused a dramatic reduction of the binding ability to Netrin1a. Interestingly, the Netrin-binding 21aa stretch (aa232-252) of DraxinA is highly conserved cross vertebrate species (FIG. 5). It is noteworthy that, this conserved 21aa region is also highly specific for Draxin and cannot be found in other proteins.

Example 4: Netrin1a Domain Mapping

(60) Netrin1a is a multi-domain containing protein composed of 603 amino acids. It consists of a laminin N-terminal domain (LamNT) encoded by amino acid 44-282 followed by 3 laminin-type epidermal growth factor-like domains (aa284-450), and a C-terminal domain (C345C) encoded by amino acid 486 to 594. In order to map the DraxinA binding region in Netrin1a we generated a set of truncated Netrin1a fragments and probed them in the AVEXIS assay for binding against DraxinA. Using this approach we were able to narrow down the binding region to a fragment consisting of amino acid 401-458 (FIG. 4B). This fragment encodes the third laminin-type EGF domain. The third EGF-domain of Netrin1a is highly conserved in vertebrate Netrin1 homologs. For example, only a single amino acid exchange is present in this domain between zebrafish Netrin1a (CDCHPVGAAGKTCNQTTGQCPCKDGV TGITCNRCANGYQQSRSPIAPC; SEQ ID NO: 64) and human Netrin1 (CDCHPVGAAGKTCNQTTGQCPCKDGVTGITCNRCAKGYQQSRSPIAPC; SEQ ID NO: 65) proteins,

(61) Interestingly, the third EGF domain of Netrin1 has recently been shown to be required for Netrin receptor binding (Finci et al., 2014; Xu et al., 2014). These findings offer a mechanistic explanation for our observed competition assay results.

Example 5: DraxinA is Able to Inhibit the Binding of Netrin1a to Netrin Receptors

(62) By using an AVEXIS-based competition assay (FIG. 6 A) we tested whether the binding of DraxinA to Netrin1a has an influence on Netrins ability to bind to Netrin receptors. First we confirmed that we reliably detected the binding of Netrin1a to Netrin receptors of the DCC/Neo1 and Unc5 families with the AVEXIS method. We also tested whether Draxin is able to bind to the corresponding Netrin receptors. Contrary to previous findings (Ahmed et al., 2011), we were not able to detect direct binding between Draxin and any of the tested Netrin receptors using the AVEXIS platform. This has also been confirmed by a recent publication (Haddick et al. 2014). In the competition assay the extracellular domains (ECDs) of Netrin receptors have been used as prey proteins together with purified full length DraxinA and probed for binding against Netrin1a bait proteins. Using this strategy, we observed a DraxinA concentration dependent inhibition of the binding between Netrin1a and Netrin receptors (FIG. 6B,C,D). The inhibition is specific for DraxinA. Furthermore, DraxinA (FIG. 6 E) is not able to block the binding of RGMc to Neo1, another reported ligand of the corresponding receptor (Bell et al., 2013).

Example 6: The 21Aa DraxinA Fragment Fused to the Human Fc Region of IgG is Sufficient to Block the Binding of Netrin1a to Netrin Receptors

(63) Next we assayed whether the 21aa DraxinA fragment fused to the human Fc tag (Draxin.sub.aa232-252-hFc) is sufficient to outcompete Netrin/Netrin receptor interactions.

(64) We compared the effect of Draxin.sub.aa232-252-hFc with full-length Draxin-hFc and a version of DraxinA where the 21aa Netrin1 binding motif has been deleted (DraxinAΔ.sub.aa231-252-hFc) in the competition assay for their ability to interfere with the binding of Netrin1a to Dcc. The results show that Draxin.sub.aa232-252-hFc has a similar efficiency in inhibiting the binding of Netrin1a to Dcc as the DraxinA full-length version (Draxin-hFc). The DraxinA version with the 21 aa deletion (DraxinAΔ.sub.aa231-252-hFc) is not able to compete for binding to Netrin1a (FIG. 7). In addition we used our set of different Draxin-hFc proteins to test whether they have an effect on other known interactions (Cntn1a/Ptprz1b, Vasn/Islr2, EphB4a/EphrinB2a). None of the 3 Draxin-hFc versions was able to inhibit any of the tested interactions (FIG. 7 B,C,D).

(65) Taken together, our results show that the 21aa region is necessary for the competition and that the Draxin.sub.aa232-252-hFc fusion protein is also sufficient to outcompete Netrin receptors for Netrin1a binding.

Example 7: The Binding of the 21Aa DraxinA Fragment to Netrin1a is Highly Specific

(66) In order to determine the binding specificity of the 21aa DraxinA peptide we used the AVEXIS assay and screened the 21aa fragment as bait and prey against proteins from our library. In this screen against a set of 141 bait proteins the DraxinA.sub.aa232-252-prey only bound to the Ntn1a-bait and to the positive control bait Matn4 (FIG. 8A). Similar results we obtained for the DraxinA.sub.aa232-252-bait protein, which only interacted with the Ntn1a-prey in a screen against a set of 191 different prey proteins (FIG. 8B). These findings indicate that the binding of this short 21 aa peptide to Netrin1a is highly specific.

Example 8: DraxinA Outcompetes Receptor Bound Netrin 1a

(67) Netrin1 binds with high affinity (K.sub.d's in the low nM range) to its receptors of the DCC- and Unc5-family (Leonardo et al., 1997). Hence, we asked the question whether already bound Netrin1a could be displaced from the receptors by DraxinA. In order to do so we carried out an AVEXIS based competition assay and tested three different settings (FIG. 9A). In one experimental setting Netrin1a-baits were preincubated with purified DraxinA before addition of the DCC prey. In the second set of experiments the Netrin1a-baits were incubated with a mixture of DraxinA and DCC prey proteins, and in the third set of experiments Netrin1a-baits have been preincubated with DCC preys followed by the addition of purified DraxinA as inhibitor. We did not observe a difference in the %-binding response between these 3 sets of experiments. These findings show that already formed DCC/Netrin1a complexes can be disrupted by the addition of DraxinA. These findings indicate that DraxinA has a higher affinity for Netrin1a than the Netrin receptor DCC.

Example 9: In Vivo Detection of the Draxin-Netrin Interaction in Zebrafish Embryos

(68) To independently confirm the Draxin/Netrin1a interaction and to test whether both proteins are able to interact in vivo, we made use of transient protein overexpression experiments in zebrafish embryos. mRNAs encoding Draxin fused to superfolder GFP (Draxin-sfGFP) and Netrin1a tagged with mCherry (Netrin1a-mCherry) or superfolder GFP (Netrin1a-sfGFP) have been injected into one-cell stage zebrafish embryos. The distribution of the corresponding fluorophore tagged proteins has been analyzed in sphere stage zebrafish embryos (4 hours post fertilization) (FIG. 11A). At this developmental stage the extracellular space width between the cells is very large, ideally suited to visualize the localization of secreted proteins. Upon injection of mRNA encoding Draxin-sfGFP we observed an evenly distributed signal outside the cells in the extracellular milieu of 4 hpf zebrafish embryos (FIG. 11 Ba). In contrast thereto, the distribution of Netrin1a-sfGFP was restricted to cell surface sub-domains (FIG. 11 Bb).

(69) When Draxin-sfGFP was coexpressed with Netrin1a-mCherry, Draxin-sfGFP re-located to Netrin1a-mCherry positive membrane associated densities (FIG. 11 Bc). This indicated that localized Netrin1a-mCherry was able to capture diffusible Draxin-sfGFP.

(70) The mRNA overexpression experiments showed that Draxin and Netrin1a are able to interact with each other in vivo. To further support this we used another strategy aiming to detect the distribution of endogenous Draxin interaction partners at developmental stages relevant for axon guidance decisions. From our binding assay with monomeric prey proteins we already had hints that the interaction between Draxin and Netrin is of high-affinity. Thus, we fused a netrin-binding fragment of the Draxin-E CD (aa209-284) to the human Fc region to generate an affinity probe. First, we tested this probe on zebrafish embryos from different developmental stages. After very gentle fixation the embryos were incubated with HEK293-6E cell supernatants containing the recombinant soluble Draxin.sub.aa209-284-hFc protein.

(71) Using an Alexa-Fluor 568 anti human IgG antibody to detect in situ bound Draxin.sub.aa209-284-hFc we only detect a signal in close proximity to the floor plate (FIG. 12A, 12Ba). Because floor plate cells express Netrin1a and Netrin1b, we had indications that the signal detected by using the Draxin affinity probe indeed corresponds to in the extracellular space localized netrin. To prove this observation, we compared 48 hpf wt embryos with netrinla and netrinlb double-knockdown embryos (FIG. 12Ba, 12Bb). In double-knockdown embryos the signal from the bound affinity probe was barely detectable compared to non-injected siblings, indicating that the Draxin.sub.aa209-284-hFc probe indeed detected netrin. Taken together, the results from our mRNA overexpression and Draxin affinity probe experiments provide strong evidence that Draxin and Ntn1a are able to interact in vivo in zebrafish embryos.

Example 10: Human DRAXIN/Netrin-Signaling Network

(72) To determine the binding specificity of DRAXIN/Netrin interactions, we carried out a pairwise binding screen between human DRAXIN and human Netrin family members. Except Netrin-5, we included all human Netrin family members consisting of two secreted γ-Netrins (Netrin-1 and Netrin-3) and one secreted (Netrin-4) and two GPI-linked β-Netrins (Netrin-G1 and Netrin-G2) in our binding study. Human DRAXIN and a 21 amino acid Netrin binding fragment derived thereof (SEQ ID NO.: 1) bound to Netrin-1 and Netrin-3 but not to human β-Netrin family members (FIG. 13).

(73) These experiments confirm for human proteins the Draxin/γ-Netrin binding specificity within the Netrin family and showed that the human 21 amino acid DRAXIN fragment (SEQ ID NO.: 1), like its zebrafish counterpart (SEQ ID NOs.: 3), is sufficient for binding.

Example 11: Validation of the Draxin/Netrin1 Interaction by Surface Plasmon Resonance

(74) Recombinant human Draxin was purchased from R&D systems. Biotinylated recombinant human UNC5B, Netrin1 and DCC were produced recombinantly using the described mammalian expression system (HEK293-6E). Biotinylated proteins were immobilised on the SA coated sensor chip and Draxin was injected sequentially in increasing concentrations (0 nM, 1.2 nM. 2.3 nM, 4.7 nM, 9.4 nM, 18.8 nM) for 3 min. Dissociation was allowed for 5 min in HBS-P. Binding was monitored and an interaction of Draxin to immobilised Netrin-1 was observed with a binding constant KO of approximately 20 to 100 nM. No binding of Draxin to UNC5B and DCC was detected (FIG. 14).

REFERENCES

(75) Ahmed G, Shinmyo Y, Ohta K, Islam S M, Hossain M, Naser I B, Riyadh M A, Su Y, Zhang S, Tessier-Lavigne M, Tanaka H: Draxin inhibits axonal outgrowth through the netrin receptor DCC. J Neurosci. 2011 Sep. 28; 31(39):14018-23. Arakawa H: Netrin-1 and its receptors in tumorigenesis. Nat Rev Cancer. 2004 December; 4(12):978-87. Bell C H, Healey E, van Erp S, Bishop B, Tang C, Gilbert R J, Aricescu A R, Pasterkamp R J, Siebold C: Structure of the repulsive guidance molecule (RGM)-neogenin signaling hub. Science. 2013 Jul. 5; 341(6141):77-80. Bushell K M, Söllner C, Schuster-Boeckler B, Bateman A, Wright G J: Large-scale screening for novel low-affinity extracellular protein interactions. Genome Res. 2008 April; 18(4):622-30. Castets M, Broutier L, Molin Y, Brevet M, Chazot G, Gadot N, Paquet A, Mazelin L, Jarrosson-Wuilleme L, Scoazec J Y, Bernet A, Mehlen P: DCC constrains tumour progression via its dependence receptor activity. Nature. 2012; 482(7386):534-7. Delloye-Bourgeois C, Brambilla E, Coissieux M M, Guenebeaud C, Pedeux R, Firlej V, Cabon F, Brambilla C, Mehlen P, Bernet A: Interference with netrin-1 and tumor cell death in non-small cell lung cancer. J Natl Cancer Inst. 2009 Feb. 18; 101(4):237-47. Durocher Y, Perret S, Kamen A: High-level and high-throughput recombinant protein production by transient transfection of suspension-growing human 293-EBNA1 cells. Nucleic Acids Res. 2002 Jan. 15; 30(2):E9. Finci L I, Kruger N, Sun X, Zhang J, Chegkazi M, Wu Y, Schenk G, Mertens H D, Svergun D I, Zhang Y, et al.: The Crystal Structure of Netrin-1 in Complex with DCC Reveals the Bifunctionality of Netrin-1 As a Guidance Cue. Neuron 2014; 83(4):839-849. Fitamant J, Guenebeaud C, Coissieux M M, Guix C, Treilleux I, Scoazec J Y, Bachelot T, Bernet A, Mehlen P: Netrin-1 expression confers a selective advantage for tumor cell survival in metastatic breast cancer. Proc Natl Acad Sci USA. 2008 Mar. 25; 105(12):4850-5. Glinka A, Wu W, Delius H, Monaghan A P, Blumenstock C, Niehrs C: Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature. 1998 Jan. 22; 391(6665):357-62. Islam S M, Shinmyo Y, Okafuji T, Su Y, Naser I B, Ahmed G, Zhang S, Chen S, Ohta K, Kiyonari H, Abe T, Tanaka S, Nishinakamura R, Terashima T, Kitamura T, Tanaka H: Draxin, a repulsive guidance protein for spinal cord and forebrain commissures. Science. 2009 Jan. 16; 323(5912):388-93. Haddick P C, Tom I, Luis E, Quinones G, Wranik B J, Ramani S R, Stephan J P, Tessier-Lavigne M, Gonzalez L C: Defining the Ligand Specificity of the Deleted in Colorectal Cancer (DCC) Receptor. PLoS One. 2014 Jan. 6; 9(1):e84823. Kastenhuber E, Kern U, Bonkowsky J L, Chien C B, Driever W, and Schweitzer J: Netrin-DCC, Robo-Slit, and heparan sulfate proteoglycans coordinate lateral positioning of longitudinal dopaminergic diencephalospinal axons. J Neurosci 2009; 29:8914-8926. Kwan K M, Fujimoto E, Grabher C, Mangum B D, Hardy M E, Campbell D S, Parant J M, Yost H J, Kanki J P, and Chien C B: The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs. Dev Dyn 2007; 236:3088-3099. Leonardo E D, Hinck L, Masu M, Keino-Masu K, Ackerman S L, Tessier-Lavigne M: Vertebrate homologues of C. elegans UNC-5 are candidate netrin receptors. Nature. 1997 Apr. 24; 386(6627):833-8. Mann H H, Ozbek S, Engel J, Paulsson M, Wagener R:Interactions between the cartilage oligomeric matrix protein and matrilins. Implications for matrix assembly and the pathogenesis of chondrodysplasias. J Biol Chem. 2004 Jun. 11; 279(24):25294-8. Mehlen P, Delloye-Bourgeois C, Chédotal A: Novel roles for Slits and netrins: axon guidance cues as anticancer targets? Nat Rev Cancer. 2011 March; 11(3):188-97. Moore S W, Tessier-Lavigne M, Kennedy T E: Netrins and their receptors. Adv Exp Med Biol. 2007; 621:17-31. Paradisi A, Creveaux M, Gibert B, Devailly G, Redoulez E, Neves D, Cleyssac E, Treilleux I, Klein C, Niederfellner G, Cassier P A, Bernet A, Mehlen P: Combining chemotherapeutic agents and netrin-1 interference potentiates cancer cell death. EMBO Mol Med. 2013 December; 5(12):1821-34. Paradisi A, Maisse C, Coissieux M M, Gadot N, Lépinasse F, Delloye-Bourgeois C, Delcros J G, Svrcek M, Neufert C, Fléjou J F, Scoazec J Y, Mehlen P: Netrin-1 up-regulation in inflammatory bowel diseases is required for colorectal cancer progression. Proc Natl Acad Sci USA. 2009 Oct. 6; 106(40):17146-51 Söllner C, Wright G J: A cell surface interaction network of neural leucine-rich repeat receptors. Genome Biol. 2009; 10(9):R99. Suli A, Mortimer N, Shepherd I, and Chien C B: Netrin/DCC signaling controls contralateral dendrites of octavolateralis efferent neurons. J Neurosci 2006; 26:13328-13337. Xu K, Wu Z, Renier N, Antipenko A, Tzvetkova-Robev D, Xu Y, Minchenko M, Nardi-Dei V, Rajashankar K R, Himanen J, et al.: Neural migration. Structures of netrin-1 bound to two receptors provide insight into its axon guidance mechanism. Science 2014; 344: 1275-1279.