SAPOSIN LIPOPROTEIN PARTICLES AND LIBRARIES FROM CRUDE MEMBRANES

Abstract

The invention is directed to a method for studying or identifying a biologically active agent that binds to a membrane protein, the method comprising: (i) obtaining one or more 2D and/or 3D structures of a saposin lipoprotein particle comprising the membrane protein, (ii) modeling the binding of said biologically active agent to said membrane protein present in said saposin lipoprotein particle using said one or more 2D and/or 3D structures and/or resolving the binding sites and interactions between said biologically active agent and said membrane protein in the 2D and/or 3D structure.

Claims

1. A method for studying or identifying a biologically active agent that binds to a membrane protein, the method comprising: (i) obtaining one or more 2D and/or 3D structures of a saposin lipoprotein particle comprising the membrane protein, (ii) modeling the binding of said biologically active agent to said membrane protein present in said saposin lipoprotein particle using said one or more 2D and/or 3D structures and/or resolving the binding sites and interactions between said biologically active agent and said membrane protein in the 2D and/or 3D structure.

2. The method according to claim 1, wherein the 2D and/or 3D structure is from a saposin lipoprotein particle that has been obtained according to a process in which first a library of saposin lipoprotein particles is prepared, wherein library means a set of different saposin lipoprotein particles comprising a heterogenic mixture of saposin lipoprotein particles with different membrane lipid and membrane protein compositions, wherein the particles comprise membrane components from a cell or an organelle membrane and a lipid binding polypeptide that is a saposin-like protein belonging to the SAPLIP family of lipid interacting proteins or a derivative form thereof, wherein the process comprises the steps of a) providing a mixture of crude membrane vesicles obtained from an archaeal, eukaryotic or a prokaryotic cell or an organelle membrane and wherein the crude membrane vesicles comprise both membrane lipids as well as membrane proteins from the crude cell or organelle membranes from which they are obtained; b) contacting one ore more of said crude membrane vesicles with the lipid binding polypeptide in a liquid environment; c) allowing for self-assembly of the particles; and wherein the particles of the library differ in their membrane protein composition, and wherein the membrane proteins are embedded in the lipids of the cell or organelle membrane used as starting material in which they are present and active in vivo.

3. The method according to claim 2, wherein the mixture of crude membrane vesicles in step a) is obtained by a.1) providing a cell and/or a cell organelle; and a.2) lysing or disrupting the cell and/or the cell organelle.

4. The method according to claim 2, wherein between steps a) and b), the mixture of crude membrane vesicles is subjected to a purification step, in particular an affinity purification to enrich the crude membrane vesicles containing the membrane protein of interest.

5. The method according to claim 2, wherein the process comprises in between steps a) and b), in step c) or as a subsequent step d), the purification of the particles by at least partial removal of membrane lipids, membrane proteins, lipid binding polypeptide, unsoluble or aggregated matter and/or detergent, wherein, optionally, the purification is performed by chromatography; ultracentrifugation; dialysis; contacting with detergent-binding biobeads; use of concentrators; or affinity purification methods.

6. The method according to claim 2, wherein the preparation method comprises the additional step of f) purifying at least one type of saposin lipoprotein particle from the library, wherein, optionally, this purification of the at least one type of particle in step f) is performed by affinity purification including but not limited to affinity chromatography and/or immunopurification, in particular by using an antigen or tag on a membrane protein present in the particle to be purified and/or wherein, optionally, the purification is performed by chromatography, in particular size-exclusion chromatography; ultracentrifugation; dialysis; contacting with detergent-binding biobeads; or use of concentrators.

7. The method of claim 1, wherein the method additionally comprises the step of (iii) selecting one or more biologically active agents in silico that have been determined to bind to said membrane protein using said modeling step (ii); and/or (iv) confirming said binding by combining said saposin lipoprotein particle with said one or more biologically active agents identified in step (iii), and measuring the binding between said saposin lipoprotein particle and said one or more biologically active agents.

8. The method according to claim 1, wherein the method is a method for screening a library of possible biologically active agents to identify one or more biologically active agents that bind to the membrane protein.

9. The method according to claim 1, wherein the biologically active agent is a chemical compound or a biological molecule such as a (poly)nucleotide, a peptide, protein or antigen-binding portions thereof.

10. The method of claim 9, wherein the biologically active agent is a pharmaceutically active agent, a drug candidate, a herbicide, pesticide, or other plant protection compound or candidate, or an agent for cosmetic, diagnostic or research applications.

11. The method according to claim 1, wherein the one or more 2D and/or 3D structures of step (i) have been obtained using a structural biology method for structure elucidation; preferably wherein the structural biology method is selected from the group consisting of nuclear magnetic resonance spectroscopy (NMR), X-ray crystallography, small-angle X-ray scattering (SAXS), surface-plasmon resonance (SPR), and electron microscopy (EM), in particular negative-stain electron microscopy and/or cryo-electron microscopy (cryo-EM).

12. The method according to claim 1, wherein the membrane protein is selected from the group consisting of an integral transmembrane protein, an integral monotopic membrane protein, a peripheral membrane protein, an amphitropic protein in a lipid-bound state, a lipid-anchored protein, a chimeric protein with a fused hydrophobic and/or transmembrane domain.

13. The method according to claim 1, wherein i) said particles are disc-shaped, ii) said particles generally have a maximum diameter of from 2 nm to 200 nm.

14. The method according to claim 1, wherein the lipid binding protein is saposin A, saposin B, saposin C, saposin D or a derivative or truncated form thereof, and wherein, optionally, the derivative form of the SAPLIP is selected from i) a protein having at least 80% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6; ii) a protein having at least 30%, or at least 40% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6, wherein said protein is amphipathic, forms at least one alpha helix, and is capable of self-assembling together with solubilized lipids into lipoprotein particles when employed in the process; and iii) a protein comprising the sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6 in which 1 to 40 amino acids have been deleted, added, inserted and/or substituted.

15. The method according to claim 1, wherein the 2D and/or 3D structure is of a saposin lipoprotein particle comprising the membrane protein complexed with the biologically active agent.

16. The method according to claim 1, wherein the particles essentially consist of the at least one lipid binding polypeptide, membrane lipids and/or membrane proteins of the cell or organelle membrane stemming from the cell or the organelle membrane that was used to prepare the particles.

17. The method according to claim 1, wherein the lipid binding protein is saposin A, saposin B, saposin C, saposin D or a derivative or truncated form thereof, and wherein, optionally, the derivative form of the SAPLIP is selected from i) a protein having at least 95% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6; ii) a protein having at least 90%, or at least 95% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6, wherein said protein is amphipathic, forms at least one alpha helix, and is capable of self-assembling together with solubilized lipids into lipoprotein particles when employed in the process; and iii) a protein consisting of the sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6 in which 1 to 40 amino acids have been deleted, added, inserted and/or substituted.

18. The method of claim 17, wherein the protein consisting of the sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6 in which 1 to 40 amino acids that have been substituted, are substituted with conservative amino acids.

19. The method of claim 14, wherein the protein consisting of the sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6 in which 1 to 40 amino acids that have been substituted, are substituted with conservative amino acids.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0286] The invention will hereinafter be described with reference to the Figures, which depict certain embodiments of the invention. The invention, however, is as defined in the claims and generally described herein. It should not be limited to the embodiments shown for illustrative purposes in the Figures below.

[0287] FIGS. 1a) and 1b) depict prior art lipoprotein particles. FIG. 1a is a schematic illustration of the shape and molecular organization of the Apolipoprotein A-1 containing nanosdisc particles of the prior art (e.g. EP 1 596 828 B1 discussed above). FIG. 1b is a schematic illustration of the process used to create Saposin lipoprotein particles of the prior art (e.g. WO 2014/095576 A1) discussed above.

[0288] FIGS. 2a) to 2f) are schematic illustrations of Salipro particles according to the invention. In FIGS. 2a) and 2b) a Salipro particle 1c comprising cell or organelle membrane lipids is shown; depicted in a) as side view and in b) as top view. In FIGS. 2c) and 2d) a Salipro particle 1a comprising cell or organelle membrane lipids and a membrane protein 4a is shown; depicted in c) as side view and in d) as top view. In FIGS. 2e) and 2f) a Salipro particle 1b comprising cell or organelle membrane lipids and an oligomeric membrane protein 4b is shown; depicted in e) as side view and in f) as top view.

[0289] FIGS. 3a) and 3b) are schematic illustrations of the process for preparing the Saposin particle library of the invention. The starting materials for the preparation of the library, i.e. saposin A 2 and crude membrane vesicles 5, 5′, are depicted in FIG. 3a). The membrane vesicles 5, 5′ comprise a plurality of membrane lipids 3 and in case of 5 a plurality of membrane proteins, here exemplified in simplified form by 4a, 4b. The obtained library 7 is depicted schematically in FIG. 3b), exemplified in simplified form as the mixture of different Salipro particles 1a, 1b and 1c that differ in at least one of lipid content, protein cargo content and size. Particles 1a and 1b comprise membrane proteins 4a, 4b. Particle 1c is an “empty” lipid-only particle.

[0290] FIGS. 4a) and 4b) are schematic illustrations of the process for preparing the Saposin particle library of the invention, wherein a membrane protein 4c present in the crude membrane vesicles 5 is tagged with an affinity tag 8. The starting materials for the preparation of the library, i.e. saposin A 2 and crude membrane vesicles 5, are depicted in FIG. 4a). The membrane vesicles 5 comprise membrane lipids 3 and a plurality of membrane proteins, here exemplified in simplified form by 4c, 4b. The obtained library 7′ is depicted schematically in FIG. 4b), exemplified in simplified form as the mixture of different Salipro particles 1d, 1b and 1c that differ in at least one of lipid content, protein cargo content and size. Particle 1d comprises membrane protein 4c with the affinity tag 8 appended thereto. Particle 1b comprises oligomeric membrane protein 4b. Particle 1c is an “empty” lipid-only particle.

[0291] FIG. 5 is a schematic illustration of one embodiment of the process for preparing a particle of the invention by conducting step f) as described above. The library 7′ of FIG. 4b) is subjected to a purification step using an affinity column 9. Schematic representation of the content of the 1.) Flowthrough; 2.) Wash; and 3.) Elution fractions exiting from the column 9 are also shown.

[0292] FIG. 6 shows the results of Example 1a. The Fig. shows Fluorescence Size-Exclusion Chromatography (FSEC) analysis of incorporation of fluorescent GFP-GLUT5 into Salipro nanoparticles from crude membranes.

[0293] FIG. 7 shows the results of Example 1b. The Fig. again shows FSEC analysis of incorporation of fluorescent GFP-GLUT5 into Salipro nanoparticles from crude membranes.

[0294] FIG. 8 shows the results of Example 2. The Fig. shows FSEC analysis of incorporation of fluorescent GFP-GLUT5 into Salipro nanoparticles from crude membranes. The “yield control sample” is run in the continuous presence of detergent in the buffer, while all other samples are run in a detergent-free buffer system.

[0295] FIG. 9 shows the results of Example 3. The Fig. shows SEC analysis with increasing amounts of saposin A added. In the presence of saposin A, membrane proteins from crude membranes are incorporated into a library of Salipro particles and remain soluble in a detergent-free buffer system.

[0296] FIG. 10 shows the results of Example 4. The Fig. shows SEC analysis with increasing amounts of saposin A added. In the presence of saposin A, membrane lipids from crude membranes are incorporated into a library of Salipro particles and remain soluble in a detergent-free buffer system.

[0297] FIG. 11 shows further results of Example 4. The Fig. shows SEC analysis of saposin A, prior art Salipro particles, and Salipro particles prepared according to the invention.

[0298] FIGS. 12a) and 12b) are identical reproductions of FIGS. 4A and 4B, respectively, of Bruhn (2005), Biochem J 389 (15): 249-257, which sequences form part of the disclosure of the present invention.

[0299] FIG. 13 shows the results of Example 7 and can be entitled “Production of pure and homogenous Salipro-CXCR4 particles”. FIG. 13A shows SEC analysis of purified Salipro-CXCR4 particles remaining as stable and homogenous nanoparticles after flash freezing (dark blue line) in liquid nitrogen, as well after 24-hour incubation at 4° C. (light blue line). FIG. 13B show SDS-PAGE analysis of the Salipro-CXCR4 particles.

[0300] FIG. 14 shows the results of Example 8 and can be entitled “Salipro-CXCR4 binds to natural ligand CXCL12”. FIG. 14A shows a schematic illustration of Salipro-CXCR4 binding to CXCL12. The experimental data shows a co-elution of Salipro-CXCR4 with CXCL12 followed by SEC and subsequent SDS-PAGE analysis of separate SEC fraction. The results show CXCL12 presence in the CXCR4 and saposin A containing SEC fractions 13-15 as an indication of the formation of a stable Salipro-CXCR4-CXCL12 protein complex. FIG. 14B shows the negative control incubated in the absence of CXCL12 peptide.

[0301] FIG. 15 shows the results of Example 9 and can be entitled “CXCR4 antagonist AMD3100 blocks binding of CXCL12 to Salipro-CXCR4 particles”. Mass photometry experiments were carried out to measure the molecular size of the respective Salipro nanoparticle complexes. FIG. 15A shows the measurement of plain Salipro-CXCR4 particles. FIG. 15B shows the measurement of Salipro-CXCR4 incubated with CXCL12. FIG. 15C shows Salipro-CXCR4 pre-incubated with compound AMD3100. FIG. 15D shows Salipro-CXCR4 pre-incubated with AMD3100, followed by the addition of CXCL12.

[0302] FIG. 16 shows the results of Example 10 and can be entitled “Salipro-CXCR4 CryoEM structure reveals functional CXCR4 tetramers”. CryoEM was used to solve the Salipro-CXCR4 structure at a 2.9 Å resolution. Density for Saposin A in the Salipro scaffold was colored blue, CXCR4 was colored in orange, and lipids resolved in the structure was colored in white. FIG. 16A shows the Salipro-CXCR4 structure from a top view and FIG. 16B shows a side view cross section, to highlight the lipid residues resolved in the structure.

[0303] FIG. 17 shows further results of Example 10 and can be entitled “Salipro-CXCR4 CryoEM reveals structural differences comparing with published X-ray crystal structure”. The figure compares the published crystal structure of an engineered CXCR4 construct (PDB number: 3ODU), with the Salipro-CXCR4 structure. FIG. 17A zooms in on an amino acid side chain change that clearly displays different orientations in the two structures. FIG. 17B indicates that the structural differences for the orientation of this specific side chain was due to the presence of lipids in the Salipro-CXCR4 structure.

[0304] FIG. 18 shows further results of Example 10 and can be entitled “Salipro-CXCR4 structure reveals differences in the IT1t binding pocket providing guidance for IT1t compound optimization by SBDD”. The figure compares the published crystal structure of the (non-Salipro particle bound) engineered free CXCR4 construct in complex with IT1t (PDB number: 3ODU), with the corresponding Salipro-CXCR4 structure, which shows the differences. FIG. 18 zooms in on one side of the IT1t binding site. Here, two amino acids showed distinct differences in their side chain orientation, when comparing the two structures in relation to IT1t.

[0305] FIG. 19 shows the results of Example 11 and can be entitled “Salipro-CXCR4 particles reveals the first structure of CXCL12 activated CXCR4”. The figure reveals the first ever reported structure of the CXCR4-CXCL12 complex which was made possible by the invention. FIG. 19A shows the Salipro-CXCR4 side view structure. FIG. 19B shows the corresponding Salipro-CXCR4-CXCL12 side view structure with electron density corresponding to CXCL12 colored in grey. FIG. 19C shows the Salipro-CXCR4-CXCL12 top view structure.

[0306] FIG. 1a) depicts a prior art Apolipoprotein A-1 containing nanosdisc particle A (see, e.g., EP 1 596 828 B1 discussed above) comprising lipids B and Apolipoprotein A-1 as lipid binding polypeptide C. Contrary to the apolipoprotein-derived nanodiscs of the prior art, the lipid binding polypeptide of the present invention does not enclose the lipids in a double belt-like fashion (cf. C in FIG. 1a) but, rather, the particles of the invention are held together by a core comprising the membrane lipids which is surrounded by two or more approximately V-shaped or boomerang-shaped lipid binding polypeptides arranged in a head-to-tail orientation with substantially no direct protein-protein contacts between the individual lipid binding polypeptides within a given particle of the invention (cf. FIGS. 2a to 2f).

[0307] FIG. 1b is a schematic illustration of the process used to create Saposin lipoprotein particles G in WO 2014/095576 A1 discussed above. Here, a SAPLIP D is incubated with purified lipids E and detergent F solubilized, purified membrane protein 4a to form a particle G. Lipids E are purified and derived of a different source than membrane protein 4a. Membrane protein 4a is not present in a membrane or a vesicle, but is in an artificial, detergent-solubilized stated (see lipids E and detergent molecules F binding to the hydrophobic surfaces of the membrane protein 4a in FIG. 1b, middle). Thus in the particle G, the membrane protein 4a is not embedded in its natural membrane environment, but rather in a mixture of artificial or exogenous lipids E and, possibly also detergent F.

[0308] FIGS. 2a) to 2f) are schematic illustrations of Salipro particles obtained according to certain embodiments of the invention. The particles 1a, 1b, and 1c comprise a plurality of different membrane lipids 3 and optionally a membrane protein, 4a, 4b. Both the membrane lipids 3 and the membrane proteins 4a and 4b stem from the same cell or organelle membrane that was used to prepare the particles. The lipids 3 are not uniform or homogeneous, but differ from each other as this is the typical case in a biological membrane. In addition, depending on the source from which the crude membrane vesicles 5 are obtained (see FIGS. 3a and 3b), the composition of the membrane will vary, and, accordingly, also the mixture of lipids 3 and optionally present membrane proteins in the Salipro particles. In a further embodiment, which is not shown, the Salipro particles may also comprise further components that are typically present in a cell or organelle membrane.

[0309] The particles 1a to 1c are not drawn to scale. Depending on the size of the membrane protein 4a, 4b incorporated in the particles 1a or 1b, the lipid-only particle 1c can be substantially different in size compared to the other particles 1a, 1b. Also particles 1a and 1b can differ in size, lipid and optimally membrane protein composition. Also particles 1c can differ in size, e.g., if parts of a lipid rafts are entirely incorporated into a Salipro particle. Note that the Salipro particles of the invention are flexible in size. For example, particle 1b harboring an oligomeric membrane protein is larger than and contains more Saposin subunits 2 as compared to particle 1a which contains a monomeric membrane protein.

[0310] FIGS. 2a) and 2b) depict—in simplified schematic form—Salipro particle 1c comprising as lipid binding polypeptide 2 a SAPLIP and cell or organelle membrane lipids 3; it is depicted in a) as side view and in b) as top view. FIGS. 2c) and 2d) depict Salipro particle 1a, which differs from 1c in that it additionally comprises membrane protein 4a; it is depicted in c) as side view and in d) as top view. The membrane protein 4a can be an integral transmembrane protein in monomeric form. However, it can also be in an oligomeric state as depicted in FIG. 2e) or 2f) or a peripheral membrane protein, an amphitropic protein in a lipid-bound state, a lipid-anchored protein or a chimeric protein with a fused hydrophobic and/or transmembrane domain, all of which may be in a monomeric or oligomeric state.

[0311] FIGS. 2e) and 2f) depict Salipro particle 1b, which differs from 1c in that it additionally comprises oligomeric membrane protein 4b. it is depicted in e) as side view and in f) as top view. The particle 1b shows flexibility in size and adapts to the size of the oligomeric membrane protein 4b incorporated therein. In the embodiment depicted in FIG. 2e) or 2f), the particle 1b comprises three SAPLIP molecules 2 per particle which are arranged in a head-to-tail fashion. The hydrodynamic radius of a particle comprising three SAPLIP molecules is in the range of from 5 to 20 nm, depending on the membrane protein incorporated therein.

[0312] FIGS. 3a) and 3b) depict—again in simplified and schematic form—the preparation process of the invention for preparing a Salipro particle library 7 from crude membrane vesicles 5. In FIG. 3b) the library 7 is depicted schematically as the mixture of different representative Salipro particles 1a to 1c that differ in their membrane lipid 3 and optionally protein cargo 4a, 4b and/or size. Of course, in reality, library 7 will encompass thousands of different Salipro particles. Similarly, starting mixture 6 will comprise a high number of different crude membrane vesicle.

[0313] As shown in FIG. 3a), the particles 1a to 1c of the invention are prepared by mixing purified SAPLIP 2 with crude membrane vesicles 5, 5′ and allowing the self-assembly X of the particle 1. The composition 6 comprising the crude membrane vesicles 5 and 5′ can be a crude membrane fraction directly obtained after cell or cell organelle lysis. Vesicles 5 and 5′ usually form spontaneously upon lysis or membrane rupture. The starting composition 6 is only depicted schematically as containing representative vesicles 5 and 5′, one of which (5′) is small and may only contain membrane lipids, the other of which (5) is larger and contains membrane proteins 4a and 4b in addition to membrane lipids 3. The crude membrane vesicles employed as starting material 6 in the preparation process of the invention can be very heterogenous in size and content. A homogenization or particular vesicle generation step is not necessarily required, but possible. Of course, also a composition comprising more uniform crude membrane vesicles can be used as starting material 6.

[0314] FIG. 3a) shows the preparation of the library 7 shown in FIG. 3b). The particles 1a to 1c of the invention are prepared by mixing purified SAPLIP 2 with crude membrane vesicles 5, 5′ comprising membrane lipids 3 and optionally one or more membrane proteins 4a and/or 4b, both of which are derived from the crude membranes. Self-assembly X of the particle 1 then occurs, e.g. at a pH of from about 5.0 to about 10.0. The crude membrane vesicles 5, 5′ can optionally comprise or be associated with detergent molecules (not depicted herein). The same can be true for the Saposin molecules 2. The membrane lipids 3 associated with the membrane proteins 4a and/or 4b are preferably exclusively a carry-over from the membrane protein's native lipid environment in the cell or organelle membrane prior to the provision of crude membrane vesicles 5, 5′. In the particles 1a and 1b, the respective membrane protein 4a and 4b is embedded in components of the hydrophobic portion of the membrane from which it is obtained. Preferably, the membrane protein is in the same or a similar conformation as in its native membrane-bound state.

[0315] FIG. 3b) shows a Salipro particle library comprising representative particles 1a, 1b and 1c. Particles 1a and 1b comprise membrane proteins 4a, 4b. Particles 1a to 1c shown as particular embodiments of the invention (also depicted in more detail in FIGS. 2a) to 2f) are approximately disc-shaped, having a flat, discoidal, roughly circular to square-shaped lipid bilayer circumscribed by the amphipathic α-helices of two or three SAPLIP molecules 2. The lipids 3 of the crude membrane vesicles 5, 5′ assemble into a discoidal bilayer-like structure of discrete size in the interior of the particles 1a to 1c. The SAPLIPs 2 define the boundary of the discoidal bilayer in the particles 1a to 1c, the interior of which is hydrophobic, i.e. comprised of lipid fatty acyl chains and lacking a hydrophilic or aqueous core. The particles 1a to 1c are held together mainly by the hydrophobic interactions of the lipids 3 of the crude membrane vesicles 5 within the bilayer core of the particles 1a to 1c and hydrophobic interactions between the lipids 3 of the crude membrane vesicles 5 and the hydrophobic portions of the amphiphilic helices of the SAPLIPs 2 facing the interior of the particle. In its smallest form, the particle 1c is thought to contain two SAPLIP molecules 2 and at least around 2-5 lipid molecules 3 of the crude membrane vesicles 5,5′. However, the particles of the invention 1a to 1c are flexible in size. Depending on the size of the cargo to be incorporated (e.g. lipid domains, membrane proteins etc.) and the molar ratio of components used in their preparation, it can accommodate multiple, i.e. more than two, SAPLIPs 2, many more lipids 3 of the crude membrane vesicles 5 and optionally one or more membrane proteins 4 a and/or 4b. For example, the particle may contain two to twenty, in particular two to ten SAPLIPs 2 and optionally one or more membrane proteins. Depending on the size of the membrane protein 4a, 4b incorporated in the particle 1a, 1b, the particles can be substantially larger than the lipid-only particle 1c. Generally, an increase in particle size will also be reflected by the number of SAPLIPs 2 per particle, which can be more than two. The particle of the invention may for example comprise two to twenty, in particular two to ten SAPLIP molecules 2.

[0316] FIGS. 4a) and 4b) are basically identical to FIGS. 3a) and 3b) with the difference, that a membrane protein 4c with an affinity tag 8 is comprised in the crude membrane vesicle 5 of the starting material 6. As a result of the self-assembly Y following contacting of the crude membrane vesicles 5. 5′ with purified SAPLIPs 2, the obtained library 7′ comprises particle 1d containing membrane protein 4c and the affinity tag 8. The description of FIG. 3a) and b) above equally applies to FIG. 4a) and b).

[0317] FIG. 5 is a schematic illustration of one embodiment for the preparation process of the invention for purifying a particular particle from the library of the invention by conducting step f) as described more generally above. As starting material, library 7′ is used which can be obtained as shown in FIG. 4a) and b) above. The library 7′ is subjected to an affinity purification, for example, using an affinity column 9. The column 9 contains a recognition entity that binds to the affinity tag 8. For example, if affinity tag 8 is a His-tag, then the recognition entity in affinity column 9 is Ni-Nta. Whereas a purification in column form is depicted in FIG. 5, batch-purification processes are also possible.

[0318] Passage of the library 7′ through the affinity column 9 leads to the specific binding of particle 1d via its affinity tag 8 to the recognition entity in column 9. Particles 1b and 1c as well as other components of the library or debris do not bind or only bind unspecifically to the affinity column 9. Thus, the flow-through 1.) obtained is essentially free of particles 1d. Some residual components of the library, such as particles 1b and 1c, which may have bound unspecifically to affinity column 9 are removed by performing at least one wash step 2.). Finally the purified particle 1d can be eluted in an elution step 3.) by disrupting the binding between affinity tag 8 to the recognition entity in affinity column 9. Such disruption can be performed by a high salt wash, by enzymatic cleavage or, in case of a His-tag/Ni-NTA tag/recognition entity pair with imidazole. The Elution fraction obtained after 3.) contains particle 1d of the invention in purified form.

EXAMPLES

[0319] The following examples serve to further explain the invention in more detail, specifically with reference to certain embodiments and Figures which, however, are not intended to limit the present disclosure.

[0320] The following abbreviations will be used: [0321] GF-buffer pH 7.5: 20 mM HEPES, pH 7.4 and 150 mM NaCl [0322] GFP Green Fluorescent Protein [0323] Glut5: Transporter, membrane protein [0324] HEPES: 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid [0325] His Histidine [0326] HN buffer: 20 mM HEPES, 150 mM NaCl, pH 7.4 [0327] HN-D buffer HN buffer containing 0.2% DDM [0328] M: molar [0329] RT: room temperature [0330] SEC: Size-exclusion chromatography [0331] TCEP: tris(2-carboxyethyl)phosphine [0332] TEV: Tobacco Etch Virus [0333] Tris: Tris(hydroxymethyl)aminomethane

[0334] Purified saposin A used in the below experiments was prepared as follows. Saposin A protein expression was carried out using a vector with the coding region for human saposin A (SEQ ID NO: 1) inserted into a pNIC-Bsa4 plasmid and transformed and expressed in E. coli Rosetta gami-2 (DE3) (Novagen) strains. Cells were grown at 37° C. in TB medium supplemented with Tetracycline, Chloramphenicol and Kanamycin and induced with 0.7 mM IPTG. Three hours after induction, the cells were collected by centrifugation at 12.000×g for 15 min. The supernatant was discarded, the cell pellet was resuspended using lysis buffer (20 mM Hepes pH 7.5, 150 mM NaCl, 20 mM Imidazol) and disrupted by sonication. Lysates were subjected to centrifugation at 26.000×g for 30 min, the supernatant heated to 85° C. for 10 min, followed by an additional centrifugation step at 26.000×g for 30 min. Preparative IMAC purification was performed by batch-adsorption of the supernatant by end-over-end rotation with Ni Sepharose™ 6 Fast Flow medium for 60 min. After binding of saposin A to the IMAC resin, the chromatography medium was packed in a 10-mm-(i.d.) open gravity flow column and unbound proteins were removed by washing with 15 bed volumes of lysis buffer. The resin was washed with 15 bed volumes of wash buffer WB2 (20 mM Hepes pH 7.5, 150 mM NaCl, 40 mM Imidazol). Saposin A was eluted by addition of five bed volumes of elution buffer EB (20 mM Hepes pH 7.5, 150 mM NaCl, 400 mM Imidazol). The eluate was dialyzed overnight against gel filtration buffer GF pH 7.5 (20 mM Hepes pH 7.5, 150 mM NaCl) supplemented with recombinant TEV protease. TEV protease containing an un-cleavable His-tag was removed from the eluate by passing it over 2 ml IMAC resin. Cleaved target proteins were concentrated to a volume of 5 ml using centrifugal filter units and loaded onto a HiLoad Superdex™ 200 16/60 GL column using an AKTAexplorer™ 10 chromatography system (both GE Healthcare). Peak fractions were pooled and concentrated to 1.2 mg/ml protein. The protein sample was flash frozen in liquid nitrogen and stored at −80 C.

Example 1a

[0335] Crude yeast cell membrane fractions were obtained from GFP-GLUT5 expressing yeast cells. The membrane fraction, which contained spontaneously formed crude membrane vesicles, was incubated with detergent and Saposin A, followed by removal of detergent-micelles using gel-filtration chromatography/size-exclusion chromatography (SEC) in detergent-free buffer. This lead to the self-assembly of the membrane components present in the initial mixture into a library of nanoscale Salipro particles. In particular, monodisperse Salipro particles comprising membrane lipids and GFP-GLUT5 could be identified within this library.

1. Membrane Preparation

[0336] Crude yeast membranes were obtained from yeast cells expressing rat GLUT5 from a GAL1 inducible TEV cleavable GFP-His8 2μ vector pDDGFP2 known in the prior art. The vector was transformed into the S. cerevisiae strain FGY217 (MATa, ura3-52, lys2Δ201, and pep4Δ) which then overexpressed GFP-GLUT5 in its cell membrane.

[0337] To generate crude membranes, cells were harvested from 12 L. S. cerevisiae cultures, resuspended in buffer containing 50 mM Tris-HCl pH 7.6, 1 mM EDTA, 0.6 M sorbitol, and lysed by mechanical disruption. Membranes were isolated by ultracentrifugation at 195,000 g for 3 h, homogenized in 20 mM Tris-HCl pH 7.5, 0.3 M sucrose, 0.1 mM CaCl.sub.2), frozen in liquid nitrogen and stored at −80° C.

2. Preparation of Salipro Particle Libraries

[0338] 20 μl of crude yeast membranes containing spontaneously formed crude membrane vesicles harboring inter alia GFP-GLUT5 were mixed with 60 μl HN buffer (20 mM HEPES, 150 mM NaCl, pH 7.4) and 20 μl HN buffer supplemented with 5% DDM, followed by incubation at 4° C. for 1 h. The membrane lysate was then cleared from debri and protein aggregation by ultracentrifugation using a TLA-55 rotor at 47 krpm (100,000 g) for 30 min. 5 μl of the cleared membrane fraction was then mixed with increasing amounts (5-10-20-30-40 μl) of Saposin A (1.2 mg/ml, HN-buffer) and incubated 5 min at 37° C. to allow self-assembly of the Salipro particles. The only lipids present in the setup are those derived from the crude membranes and crude membrane vesicles.

[0339] Thereafter, the sample volume was adjusted to 50 ul with HN buffer and centrifuged 10 min at 13 krpm. SEC analysis was performed (using a Shimadzu HPLC system): 35 μl sample was injected to a 5/150 Superose6 increase column (GE Healthcare) with a flow rate at 0.3 ml/min and the presence of the fluorescent GFP tag monitored online. The SEC buffer consisted of HN buffer without any of detergent.

[0340] As a negative control, Saposin A was entirely omitted from the experimental setup (0 ul SapA), the volume adjusted to 50 μl using NH buffer and the sample was then treated as described before.

3. Results

[0341] The results, which are depicted in FIG. 6, demonstrate that it is possible to incorporate and stabilize membrane lipids membrane proteins such as GFP-GLUT5 from crude cell membranes into soluble Salipro particles, displaying a monodisperse peak. Increasing amounts of Saposin improve the incorporation efficiency and monodispersity of the peak. Accordingly, Saposin A, lipids from the crude membranes and GFP-GLUT5 associate in such way as to form water-soluble particles with an incorporated membrane protein. Whereas only GFP-GLUT5 was monitored in the SEC analysis via its fluoresence, the obtained Salipro particle library also contains a plethora of other Salipro particles harboring the remaining membrane proteome and lipidome of the yeast cell that the crude membrane fraction was obtained from. This can, e.g. be confirmed by mass spectrometry, SDS-PAGE and/or by probing with antibodies which bind to other native yeast membrane proteins and lipids present in the obtained salipro particle library.

[0342] As a negative control, in the absence of Saposin from the setup, GFP-GLUT5 (and, accordingly, also the remaining membrane proteins) is not soluble and aggregates (see high void peak).

Example 1b

1. Setup

[0343] Crude yeast membranes containing crude membrane vesicles and rat GLUT5 were prepared as described in example 1a.

2. Variation of the Process for Salipro Particle Library Formation.

[0344] Two different approaches to make Salipro-GFP-GLUT5 particles from crude membrane extract were evaluated.

[0345] For the first approach, 25 μl of saposin A (0.75 mg/ml) were added to 1 μl crude membrane extract containing crude membrane vesicles and GFP-GLUT5 and incubated for 5 min at 37° C. Thereafter 24 ul HN-D buffer was added to the mix and incubated for an additional 5 min at 37° C. The sample was then centrifuged for 10 min at 13 krpm and SEC analysis was performed as in example aa.

[0346] For the second approach, 1 ul crude membrane extract containing crude membrane vesicles and GFP-GLUT5 were first supplemented by adding 24 ul HN-D buffer, followed by incubation at 37° C. for 5 min. The sample was then centrifuged for 10 min at 13 krpm. The lysate suspension was collected and 25 μl saposin A (0.75 mg/ml) were added and incubated for an additional 5 min at 37° C. The sample was then analyzed by SEC as described in example 1a.

3. Results

[0347] The data shown in FIG. 7 demonstrate that both approaches work to obtain Salipro particle libraries with soluble Salipro-GFP-GLUT5 particles, independently if saposin A was added before or after incubation of the crude membranes with HN-D buffer at 37° C. for 5 min. In both instances, the natural lipids of the crude membranes are still present. Interestingly, when adding saposin A directly to crude membranes, only very little protein aggregation is seen in the void volume. Importantly, both methods work, which gives the process a certain flexibility when reconstituting Salipro particle libraries from complex crude membrane extracts.

Example 2

1. Setup

[0348] Crude yeast membranes containing rat GFP-GLUT5 and crude membrane vesicles were prepared as described in example 1a.

2. Salipro Formation, Titration

[0349] 20 μl of yeast crude membranes were mixed and solubilized with 60 μl HN buffer and 20 μl HN buffer supplemented with 5% DDM at 4° C. for 1 h. The membrane lysate was then cleared from debri and protein aggregation by ultracentrifugation using a TLA-55 rotor at 47 krpm (100,000 g) for 30 min. 5 μl of the cleared membrane lysate, which contained crude membrane vesicles, was then mixed with different volumes (12, 20, 30 and 40 μl) of saposin A (4 mg/ml, HN buffer) and incubated 5 min at 37° C. Thereafter the sample volumes were adjusted to 50 μl with HN buffer and centrifuged 10 min at 13 krpm. SEC analysis was performed (using a Shimadzu HPLC system) and 35 μl sample was injected to a 5/150 Superdex 200 increase column (GE healthcare) with a flow rate at 0.3 ml/min and the presence of the fluorescent GFP tag monitored online. Again, the SEC was performed using HN buffer, in the absence of detergent, to facilitate membrane protein reconstitution into the Salipro particles.

[0350] As a control, 5 μl of the cleared membrane lysate was mixed with 45 μl NH buffer supplemented with 0.03% DDM and centrifuged 10 min at 13 krpm. SEC analysis was performed (using a Shimadzu HPLC system) in HN buffer containing 0.03% DDM and 35 μl sample was injected to a 5/150 Superdex 200 increase column (GE healthcare) with a flow rate at 0.3 ml/min and the presence of the fluorescent GFP tag monitored online. The purpose of this control sample (yield control sample) was to act as a reference point to determine the amount of GFP-GLUT5 that can be solubilized in the permanent presence of detergent, in contrast to the yield of GFP-GLUT5 reconstituted in Salipro particles in a detergent-free buffer system.

3. Results

[0351] The results depicted in FIG. 8 show that an increasing amount of saposin A improves the amounts of GFP-GLUT5 reconstituted into the Salipro particles of the library.

[0352] To quantify the reconstitution yields, the percentage of the GFP-GLUT5 peak value from the “yield control sample” was compared to the peak values of the reconstituted samples. This showed that 36% of GFP-GLUT5 were reconstituted for the 12 μl SapA sample, 57% for the 20 μl SapA sample, 65% for the 30 μl SapA sample and 74% for the 40 μl SapA sample. Altogether, this demonstrates that it is possible to increase the incorporation of membrane proteins from crude membranes into Salipro particles with increasing amounts of Saposin.

Example 3

[0353] The solubility of Salipro membrane protein components of the library was analyzed

1. Background

[0354] Crude membranes contain a plethora of different membrane proteins. In the preparation process of the invention, not only fluorescently labeled GFP-GLUT5, but also the various other yeast membrane proteins from crude membranes are incorporated into Salipro particles. In the absence of detergent, the membrane protein fraction is not soluble and aggregates in detergent-free buffer systems, leading to the formation of a large void-peak in SEC analysis. However, once embedded in Salipro particles, it could be shown that the membrane proteins present in the Salipro particle library remain soluble in detergent-free buffer systems.

2. Setup

[0355] In the same experimental setup as in Example 1a, here the SEC signals were analyzed based on UV absorptions at 280 nm instead of GFP fluorescence, with a focus (zoom) on the void peak indicating aggregated membrane proteins.

3. Results

[0356] In the absence of saposin, the membrane proteins from crude membranes aggregate in detergent-free buffer systems, as indicated by a large void peak appearing in the SEC analysis at around minute 4 (see FIG. 9, 0 μl SapA).

[0357] In contrast, by increasing the amount of saposin that is added to the lysate, the amount of aggregated membrane proteins decreases accordingly. This indicates that membrane proteins remained soluble due to incorporation into the Salipro particles of the obtained membrane proteome library. Note that almost no protein aggregates were detected in the sample containing the highest amount of saposin (FIG. 9, 40 μl SapA).

[0358] Altogether this data indicates that after being subjected to the method of the invention, all membrane proteins from the crude membrane vesicles remain soluble upon detergent removal in a detergent-free environment, due to successful reconstitution into a library of corresponding Salipro nanoparticles. Thus it is possible to generate a particle Salipro-membrane protein/proteome and/or membrane lipid/lipidome library originating from crude membrane.

Example 4

[0359] The solubility of Salipro membrane lipid components of the library was analyzed

1. Setup

[0360] In the same experimental setup as in Example 1a, a different SEC analysis performed (note the difference in axis scale and axis intercept in FIGS. 9 and 10), with a focus (zoom) to peaks originating from monomeric Saposin and lipid-only Salipro particles.

2. Results

[0361] The results depicted in FIG. 10 indicate that adding Saposin A to crude membranes leads to the formation not only of membrane protein containing Salipro particles, but also of lipid-only (i.e. “empty”) Salipro particles. Increasing amounts of Saposin (5, 10, 20, 30 and 40 μl) also lead to the increased formation of lipid-only Salipro particles. SEC analysis reveals a peak originating from lipid-only Salipro particles at 7.6 min (indicated), while the corresponding peak of monomeric Saposin (“free Saposin”) appears at 8.2 min (indicated). The data indicate that the lipids from crude membrane vesicles allow for the formation of Salipro lipid-only particles at neutral pH.

[0362] As a negative control, one sample was analyzed in the absence of saposin (0 μl SapA).

[0363] This experiment using increasing amounts of Saposin was performed on a Superose 6 5/150 column (GE Healthcare). Since the SEC separation of the two saposin-related peaks are not well separated, the experiment was repeated with a Superdex 200 5/150 column (GE Healthcare) (FIG. 11)

[0364] One sample was prepared exactly as described in example 2 for the “12 ul SapA” Sample (designated in FIG. 11 as “Sap+crude membrane”). This time, SEC analysis was performed using UV absorbance at 280 nm. As a first control, one sample contained only Saposin A in order to indicate the position of lipid-free, monomeric Saposin in the SEC profile. A second control sample contained Saposin A particles obtained by the method described in WO 2014/095576 A1, i.e. by incubating Saposin A with purified lipids,

[0365] The data presented herein clearly demonstrates that both cell membrane lipidome as well as proteome can be incorporated into Salipro particles to form respective libraries.

Example 5

[0366] In this Example, a library of Salipro particles was prepared as described in the Examples above, however, only HN buffers without detergent were used. This experiment also resulted in successful incorporation of GFP-GLUT5 from crude membranes into soluble Salipro particles.

Example 6: Purification of Specific Saposin Particles from Libraries

[0367] In this Example, a membrane protein library obtained according to Example 3 is prepared. The library Salipro particles covering essentially the entire yeast membrane proteome and also includes Salipro particles comprising GFP-GLUT5. The latter particles are purified by means of the TEV cleavable GFP-His8 tag present in the GFP-Glut5 construct.

[0368] The library is subjected to a Ni-NTA affinity purification (e.g., Qiagen) according to the manufacturer's instructions. After the prescribed wash steps, the GFP-GLUT5 containing Salipro particles are eluted via TEV protease cleavage or imidazole.

[0369] The purified Salipro particles of any of the former examples can be subjected to structural biology methods to determine their 2D and/or 3D structures. These can then be used in step a) of the method of the invention for studying or identifying a biologically active agent that binds to a membrane protein (see Example 7 below). For example, the Salipro particles of Example 6, in particular the GFP-GLUT5 containing SEC fractions in Example 6 are pooled, concentrated to 5 mg/mL, and flash frozen for cryo-EM preparation.

[0370] The flash frozen sample can be prepared using typical cryo-EM preparation. For example the sample can be applied to glow-discharged Quantifoil R1.2/1.3 grids, blotted for 10 s using blot force 20 and plunge frozen into liquid ethane using Vitrobot Mark IV. Right before plunge freezing, 0.005% (w/v) fluorinated Fos-Choline 8 can be added to the sample to overcome a preferred orientation. The data can be collected, e.g. on a Thermo Scientific™ Krios G4™ Cryo Transmission Electron Microscope (Cryo-TEM) equipped with Selectris X Imaging Filter and Falcon 4™ Direct Electron Detector camera operated in Electron-Event representation (EER) mode.

[0371] Data processing can e.g. be performed in Relion single particle analysis suite 12. After motion- and CTF-correction, a certain number of particles are picked from a certain number of micrographs. Following 2D and 3D classification, the best 3D class consisting of a certain number of particles can be subjected to CTF Refinement, 3D refinement and postprocessing, yielding a certain overall resolution, e.g. of 5.0 Å or less. The 2D and/or 3D structure thus obtained can be used directly or stored electronically and used at any later point in time in step (i) of the method of the invention for studying or identifying a biologically active agent that binds to GLUT5.

Example 7

[0372] 1. CXCR4 Construct Design and Expression

[0373] A CXCR4 construct was designed to contain the wildtype human CXCR4 sequence (UniProt accession number, P61073) with an N-terminal hemagglutinin leader signal peptide (HA.sub.ss) followed by a FLAG tag. In addition, C-terminal GFP was added followed by an EPEA affinity tag (C.sub.tag). The CXCR4 sequence was designed to be flanked by two PreScission protease cleavage sites (HA.sub.ss-FLAG-3C-CXCR4-3C-GFP-C.sub.tag) enabling purification of CXCR4 without the presence of GFP and tags in the final preparation. The expression construct was codon optimised for protein expression in human cells (GeneArt, ThermoFisher). Human Expi293 cells were transiently transfected, and cell pellets were collected 48 hours post transfection (GeneArt, ThermoFisher).

[0374] 2. Preparation of Salipro-CXCR4 Nanoparticles.

[0375] Cell pellets expressing CXCR4-GFP were resuspended with 20 mL of 1.2×HNG buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 5% glycerol) supplemented with cOmplete protease inhibitor cocktail (Roche) and 1% digitonin (Calbiochem). After five minutes incubation at 4° C. in a rotating wheel, cell debris was removed by low-speed centrifugation (5000 g for 5 min at 4° C.) and the supernatant was further incubated for 50 min at 4° C. on a rotating wheel. Remaining cell debris and insoluble material was removed by centrifugation at 30,000 g for 40 min at 4° C. The cleared cell lysate comprising spontaneously formed crude membrane vesicles was incubated for 45 min at 4° C. with 500 μL of equilibrated CaptureSelect C-tag affinity matrix (ThermoFisher). The mixture was loaded into a Poly-prep column (Bio-Rad) and the flow-through was discarded. The CXCR4-loaded resin was resuspended with saposin A, transferred to a tube, and incubated in a rotating wheel for 30 min at 4° C. The mixture was loaded again in the Poly-prep chromatography column and the flow-through discarded. After washing with 44 CV of HNG buffer, the resin was resuspended with 580 μL HNG buffer and split into two tubes. PreScission protease (Cytiva) were added to each tube and incubated overnight at 4° C. in a rotating wheel. After on-column cleavage, the suspension was loaded in the Poly-prep column and purified saposin lipoprotein particles comprising CXCR4 (“Salipro-CXCR4 particles”) were collected by elution, while the cleaved GFP-tag remained bound to the affinity resin. The elution was concentrated to 250 μL using an Amicon Ultra Centrifugal Filter with a 100-kDa NMWL (Millipore) and subjected to size exclusion chromatography, loaded into a Superose 6 Increase 10/300 equilibrated in HN (50 mM HEPES, 150 mM NaCl), using an Åkta Pure chromatography system. The fractions containing Salipro-CXCR4 particles were pooled and concentrated to 4.6 mg/mL (OD1) with an Amicon Ultra Centrifugal Filter with a 100-kDa NMWL.

[0376] 3. Results

[0377] The results, depicted in FIG. 13 show the purification of homogenous, stable, and pure Salipro-CXCR4 nanoparticles. Prepared Salipro-CXCR4 particles were flash frozen and incubated differently to investigate nanoparticle thermal stability (FIG. 13A). The flash frozen sample (dark blue line) was analyzed as a control by analytic Size Exclusion Chromatography (SEC) in a Superose 6 Increase 5/150 GL column (Cytiva) equilibrated with detergent free HN buffer, using a Prominence-i LC-2030C high performance liquid chromatography system equipped with PDA and detector (Shimadzu) at a flow rate of 0.3 mL/min. To probe Salipro-CXCR4 particle stability, flash frozen samples were thawed and subsequently incubated for 24 hours at 4° C. followed by analytic SEC showing the formation of homogenous and stable nanoparticles (light blue line). Reducing SDS-PAGE further showed the presence of CXCR4 and saposin A confirming the production of pure Salipro-CXCR4 nanoparticles (FIG. 13B), corresponding to the sample used for downstream analytics and structure determination.

Example 8

[0378] Salipro-CXCR4 Makes a Stable Complex with its Natural Protein Ligand CXCL12.

[0379] To illustrate the invention, CXCL12 was used as an example for a biologically active agent binding to the membrane protein CXCR4. CXCL12 is the natural protein ligand for CXCR4. Salipro-CXCR4 nanoparticles were generated as described in Example 7. One sample was incubated for 2 hours at 4° C. together with 5× molar excess of CXCL12 (ThermoFisher Scientific). A control sample was incubated similarly, although without adding CXCL12. Samples were analysed by SEC and collected SEC fractions were analysed by SDS-PAGE, to investigate if it is possible to observe CXCL12 co-elution with Salipro-CXCR4 as an indication of the formation of a Salipro-CXCR4-CXCL12 protein complex. FIG. 14A shows CXCL12 presence in the CXCR4 and saposin A containing SEC fractions 13-15. The absence of the CXCL12 band was confirmed in the negative control (FIG. 14B). Thus, Salipro-CXCR4 particles can bind to the native ligand of CXCR4 forming a stable Salipro-CXCR4-CXCL12 protein complex.

Example 9

The Antagonist AMD3100 Blocks Binding of CXCL12 to Salipro-CXCR4 Particles.

[0380] 1. Mass Photometry.

[0381] Mass photometry is an analytic method that optically measures the mass of single molecules in solution, making it possible to measure the size of proteins and protein complexes. Here, all samples were measured in HN buffer using the Refeyn OneMP mass photometer (Refeyn Ltd.) with a 60 s acquisition time. The resulting histograms were fitted to Gaussian distributions using DiscoverMP (Refeyn Ltd.) to extract peak contrast and relative amount of each peak (n=3). Contrast-to-mass conversion was achieved by calibration using NativeMark protein ladder (ThermoFisher Scientific). Three protein species (with specified masses) were fitted to corresponding Gaussian distributions to extract a linear relation between mass and contrast.

[0382] 2. Experimental Setup and Results

[0383] Salipro-CXCR4 nanoparticles were prepared as outlined in Example 7. Analysis of the purified Salipro-CXCR4 particles in mass photometry revelead a protein complex at a molecular size of 350 kDa (FIG. 15, panel A). Incubating Salipro-CXCR4 for 1 h at 4° C. with 5× molar access of CXCL12 prior to the measurement resulted in a larger protein complex at about 370 kDa. This confirmed that CXCL12 can bind to and form a stable protein complex with the Salipro-CXCR4 particles (FIG. 15, panel B). Next, we explored if the known small molecular CXCR4 antagonist AMD3100 could block CXCL12 binding in the assay. As control, Salipro-CXCR4 particles were incubated for 1 h at 4° C. with buffer supplemented with 50 mM AMD3100 (FIG. 15, panel C). Adding an 5× molar access of the CXCL12 to Salipro-CXCR4 particles in the presence of 50 mM AMD3100 did not increase the molecular weight of the nanoparticles (FIG. 15, panel D). The results indicate that AMD3100 binds to CXCR4 in the Salipro nanoparticles, resulting in a blockage of CXCL12 binding to CXCR4.

Example 10

Structure Determination of Salipro-CXCR4 Nanoparticles.

[0384] 1. Cryo-EM Sample Preparation and Data Collection

[0385] Quantifoil 200 mesh 1.2/1/3 Cu grids were glow-discharged using 20 mAmp current for 45 sec and charge set to positive (GloQube®, Quorum Technologies). 3 ul of 4.6 mg/ml protein was pipetted onto the grid in Vitrobot Mark IV (Thermo Fisher Scientific) chamber set to 4° C. and 95% humidity. Grids were then blotted for 10 s using a blot force of +20 and 30 s waiting time before plunge-freezing in liquid ethane. Data collection was conducted using a 300 kV Thermo Scientific™ Krios G4™ Cryo-Transmission Electron Microscope (Cryo-TEM) equipped with Selectris X Imaging Filter and Falcon 4™ Direct Electron Detector camera operated in Electron-Event representation (EER) mode. Thermo Fisher Scientific EPU 2 software was used to automate the data collection. The exposure time of 4.16 s with a total dose of 40.24 e.sup.−/Å.sup.2 was used, and each movie was split into 40 fractions during motion correction. The dose rate on the camera was 5.4 e.sup.−/px/s, and the nominal defocus range was specified between −0.5 and −1.5 μm in 0.25 μm intervals.

[0386] 2. Cryo-EM Single-Particle Analysis

[0387] Data processing was performed using Relion 3 and CryoSPARC™ image processing suites. After motion- and CTF-correction (using Relion's implementation of MOTIONCOR and CTFFIND-4.1, respectively), particles were picked using a template-free auto-picking procedure based on a Laplacian-of-Gaussian (LoG) filter. The initial model was generated in cryoSPARC™, performing the initial pre-processing steps described above independently to the Relion workflow. Following one round of 2D and 3D classification (the latter with C1 symmetry), the best 3D class, displaying clear secondary structure features was chosen. Since C1 particle classification clearly showed CXCR4 tetrameric symmetry, further refinement was done using C4 symmetry resulting in an initial structure reconstruction. This stack of particles was subjected to CTF refinements, Bayesian polishing and further two 3D classification rounds without particle alignment, after which the best particles were selected. Subsequent refinement, masking and sharpening yielded the final 2.9 Å structure map. The map resolution was determined based on the gold-standard 0.143 criterion. Figures were prepared using UCSF Chimera.

[0388] 3. Salipro-CXCR4 Structure

[0389] The structure of the Salipro-CXCR4 complex was solved at a 2.9 Å resolution revealing several novel findings of future value in therapeutic development and Structure Based Drug Design (SBDD) workflows. CXCR4 was for the first time visualised as a tetrameric protein complex, revealing the molecular details regarding its subunit interactions that could serve as a target point for therapeutic development (FIG. 16A). In addition, structural details could be revealed for how the CXCR4 complex interact with lipids (FIG. 16B). These CXCR4-lipid and lipid-lipid interaction sites could represent key regions for future therapeutic development.

[0390] 4. Salipro-CXCR4 Structure Reveals Differences Compared to Published Crystal Structure of Engineered CXCR4

[0391] To explore the Salipro-CXCR4 structure in more details we compared it to a published crystal structure of an engineered and detergent purified CXCR4 (PDB: 3ODU). To obtain the crystal structure, CXCR4 was engineered to contain the stabilizing L125W mutation. In addition, the T4 lysozyme (T4L) fusion protein was inserted between transmembrane helix V and VI at the cytoplasmic side of the receptor, the C-terminal cytoplasmic tail was deleted and further stabilization with ligands was needed to facilitate CXCR4 purification and crystallization—

[0392] The overall protein fold and conformation of Salipro-CXCR4 and the crystal structures are similar to each other, although large differences can be observed at the amino acid side chain orientation at several sites across the structures. In FIG. 17A one such area in the transmembrane region is highlighted. The structure in blue (PDB: 3ODU) corresponds to the published crystal structure, highlighting the aromatic rings of a tryptophan. Compared to the Salipro-CXCR4 structure (orange), the tryptophan residue is at a similar position although the side chain has clearly a completely different orientation. In this case, the difference in side-chain orientation could be explained by the presence of endogenous lipids present in the Salipro-CXCR4 (FIG. 17B).

[0393] 5. Salipro-CXCR4 Structure Reveals Differences in the IT1t Binding Pocket Providing Guidance for Compound Optimization by SBDD.

[0394] To further explore differences between the published crystal structure of engineered CXCR4 (PDB: 3ODU) and the Salipro-CXCR4 structures, we zoomed in at the binding pocket of the small molecular antagonist IT1t (FIG. 18). The crystal structure (blue) and the Salipro-CXCR4 structures (orange) show regions with similar amino acid side chain orientations in proximity to the IT1t (purple) binding site. Importantly, parts of the binding pocket show distinct differences, highlighted as side chain 1 and side chain 2 (FIG. 18). Both amino acids show significant differences when measuring the distances of specific side chains from IT1t. For example, the orientation and location of amino acid side chain 1 and side chain 2 in the Salipro-CXCR4 structure, provides novel and valuable information to design new compounds by SBDD using the chemical structure of IT1t as a starting reference. Thus, the unique Salipro-CXCR4 structure made from the wildtype protein together with its endogenous lipids is of great value when developing new or optimizing existing drugs/binders by SBDD approaches.

Example 11

Structure Determination of Salipro-CXCR4-CXCL12 Nanoparticles.

[0395] 1. Salipro-CXCR4-CXCL12 Particle Production.

[0396] The Salipro-CXCR4 particles were prepared as in Example 7, followed by a two-hour incubation at 4° C. together with 7× molar excess of CXCL12 forming a stable Salipro-CXCR4-CXCL12 protein complex. Homogenous Salipro-CXCR4-CXCL12 particles were further SEC purified as described in Example 7 and its structure was solved as outlined in Example 10.

[0397] 2. Salipro-CXCR4-CXCL12 Structure

[0398] Comparing the Salipro-CXCR4 structure (FIG. 19A) with the structure of Salipro-CXCR4-CXCL12 particles clearly shows chemokine binding to the extracellular loops of CXCR4 with electron density for CXCL12 colored in grey(FIG. 19B). The Salipro-CXCR4-CXCL12 structure (FIG. 19C) shows that four CXCL12 peptides can bind to the tetrameric CXCR4 particles.

[0399] The chemokine peptide CXCL12 is the natural activating ligand for CXCR4, its binding mode and CXCR4-CXCL12 complex structure has remained elusive until know. The Salipro-CXCR4-CXCL12 structure reveals the molecular details of the peptide interaction sites with CXCR4 and the conformational changes taking place in CXCR4 upon activation of its natural ligand. All these molecular details are of great value when designing novel, or optimising existing compounds using SBDD approaches.

[0400] The present invention further comprises the aspects defined in the following clauses (which form part of the present description but are not considered as claims): [0401] 1. Process for preparing a library of saposin lipoprotein particles, wherein the particles comprise membrane components from a cell or an organelle membrane and a lipid binding polypeptide that is a saposin-like protein belonging to the SAPLIP family of lipid interacting proteins or a derivative form thereof, wherein the process comprises the steps of [0402] a) providing a mixture of crude membrane vesicles obtained from a cell or an organelle membrane; [0403] b) contacting the mixture of step a) with the lipid binding polypeptide in a liquid environment; [0404] c) allowing for self-assembly of the particles. [0405] 2. Process according to clause 1, wherein the crude membrane vesicles of step a) are prepared by at least one of or all of the following steps: [0406] a.1) provision of a cell and/or a cell organelle; [0407] a.2) lysing or disrupting the cell and/or the cell organelle; [0408] a.3) obtaining a crude membrane fraction; and [0409] a.4) preparing crude membrane vesicles from the crude membrane fraction obtained in step a.3). [0410] 3. Process according to clause 1 or 2, wherein the process further comprises between steps a) and b) the step of [0411] b.1) contacting the crude membrane vesicles with a detergent in a liquid environment; wherein then in step b) the mixture obtained after step b.1) is contacted with the lipid binding polypeptide in step c) and/or [0412] wherein step b) takes place in the presence of a detergent. [0413] 4. Process according to clause 3, wherein the detergent is selected from the group consisting of alkylbenzenesulfonates or bile acids, cationic detergents and non-ionic or zwitterionic detergents such as lauryl-dimethyl amine-oxides (LDAO), Fos-Cholines, CHAPS/CHAPSO, alkyl glycosides such as short, medium or longer chain alkyl maltosides, in particular n-Dodecyl β-D-maltoside, glucosides, maltose-neopentyl glycol (MNG) amphiphiles, amphiphilic polymers (amphipols), macrocycle or cyclic oligomers based on a hydroxyalkylation product of a phenol and an aldehyde (Calixarene), and mixtures thereof. [0414] 5. Process according to any one of clauses 1 to 4, wherein [0415] i) the particles are disc-shaped, in particular wherein the particles are disc-shaped and do not comprise a hydrophilic or aqueous core, [0416] ii) the particles generally have a maximum diameter of from 2 nm to 200 nm, in particular from 3 nm to 150, preferably from 3 nm to 100 nm; [0417] ii) the self-assembly of the particle in step c) is carried out at a pH from 2.0 to 10.0, in particular 6.0 to 10.0, preferably from 6.0 to 9.0, particularly preferred from 7.0 to 9.0, and most preferred from 7.0 to 8.0 and/or [0418] iii) wherein the process comprises in step c) or as a subsequent step d) the purification of the particles by at least partial removal of membrane lipids, membrane proteins, lipid binding polypeptide, unsoluble or aggregated matter and/or detergent, wherein, optionally, the purification is performed by chromatography, in particular size-exclusion chromatography; ultracentrifugation; dialysis; contacting with detergent-binding biobeads; use of concentrators; affinity purification methods including but not limited to chromatography, magnetic beads, immunopurification and/or membrane/filters to remove unbound/non-incorporated lipids and/or hydrophobic compounds. [0419] 6. Process according to any one of clauses 1 to 5, wherein the lipid binding protein is saposin A, saposin B, saposin C, saposin D or a derivative or truncated form thereof, and wherein, optionally, the derivative form of the SAPLIP is selected from [0420] i) a protein having at least 80% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6; [0421] ii) a protein having at least 40% sequence identity to the full length sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6, wherein said protein is amphipathic, forms at least one alpha helix, and is capable of self-assembling together with solubilized lipids into lipoprotein particles when employed in the process; and [0422] iii) a protein comprising the sequence of SEQ ID NO. 1, 2, 3, 4, 5 or 6 in which 1 to 40 amino acids have been deleted, added, inserted and/or substituted. [0423] 7. Process according to any one of clauses 1 to 6, wherein the particles essentially consist of the at least one lipid binding polypeptide and components of the cell or organelle membrane, in particular membrane lipids and/or membrane proteins, stemming from the cell or the organelle membrane recited in step a). [0424] 8. Process according to any one of clauses 1 to 7, wherein [0425] i) the particles comprise membrane lipids stemming from the cell or the organelle membrane recited in step a), wherein the membrane lipids are optionally selected from the group consisting of phospholipids, glycolipids, cholesterol and mixtures thereof, and/or wherein [0426] ii) at least a portion of the particles comprises membrane proteins stemming from the cell or the organelle membrane recited in step a), and wherein the membrane protein is optionally selected from the group consisting of an integral transmembrane protein, an integral monotopic membrane protein, a peripheral membrane protein, an amphitropic protein in a lipid-bound state, a lipid-anchored protein, a chimeric protein with a fused hydrophobic transmembrane domain and mixtures thereof. [0427] 9. Process according to any one of clauses 1 to 8, wherein the cell membrane is a archaeal, eukaryotic or a prokaryotic cell membrane. [0428] 10. Process according to any one of clauses 1 to 9, wherein no additional lipids besides components of the crude membrane vesicles are added in the process. [0429] 11. Process for preparing purified saposin lipoprotein particles comprising the steps of preparing a library according to the process of any one of clauses 1 to 10 and the additional step of [0430] f) purifying at least one type of saposin lipoprotein particle from the library, wherein, optionally, the purification of the at least one type of particle in step f) is performed by affinity purification including but not limited to affinity chromatography and/or immunopurification, in particular by using an antigen or tag on a membrane protein present in the particle to be purified and/or wherein, optionally, the purification is performed by chromatography, in particular size-exclusion chromatography; ultracentrifugation; dialysis; contacting with detergent-binding biobeads; or use of concentrators. [0431] 12. Library of saposin lipoprotein particles obtainable according to the process of any one of clauses 1 to 10, wherein the particles differ in their lipid and/or protein composition, preferably in their protein composition, particularly preferred in their membrane protein composition. [0432] 13. A saposin lipoprotein particle obtainable by the process of clause 11. [0433] 14. A library of particles according to clause 12 or a particle according to clause 13 for use in medicine, in particular for use in preventing, treating or lessening the severity of a disease or for use in a diagnostic method, a cosmetic treatment or for use as vaccination formulation. [0434] 15. Use of a library of particles according to clause 12 or a particle according to clause 13 as a tool for drug development, drug screening, drug discovery, antibody development, development of therapeutic biologics, for membrane or membrane protein purification, for membrane protein expression, for membrane and/or membrane protein research, in particular lipidomics and proteomics, preferably for the isolation, identification and/or study of membranes and/or membrane proteins or creation of a lipidome or proteome database.