Polyoxometalates as membrane carriers

Abstract

The present invention relates to a delivery molecule comprising a carrier molecule and a cargo molecule, wherein said carrier molecule is a polyoxometalate which exhibits a charge status of –2 to –6 and wherein said cargo molecule is a hydrophilic moiety and to the use of said delivery molecule in therapy.

Description

UW014P 1 POLYOXOMETALATES AS MEMBRANE CARRIERS Description Field of the Invention

[0001] The present invention relates to the field of delivering a molecule into a cell via a carrier moiety. Background Art

[0002] Efficient foreign protein delivery into living cells can completely bypass transcription translation processes related to gene expression, reducing the time required for target protein synthesis from days to hours.

[0003] The design of carriers to assist the passage of bioactive therapeutic compounds through the cellular membrane presents a challenge in life sciences. Different approaches have been attempted to date, mainly focusing on the use of synthetic pores, nanoparticles and lipide formulations, cell-penetrating peptides and counterions activators. However, these systems have been deprived of their pharmaceutical application due to limitations such as endocytic entrapment, aggregation propensity, membrane-lytic activity, and related toxicity.

[0004] In WO2011 / 045296A a composition for delivering a peptide or a protein into a cell is disclosed. The composition comprises a polycationic agent and a polyanionic agent, wherein the polyanionic agent is an inorganic polyphosphate or a polyoxometalate. Zhang Z. et al. describe interactions between nanoparticles and cells by the development of polyoxometalates nanoparticle-peptide conjugates and investigation of their intracellular trafficking behaviors [1]. A method for transducing a protein or peptide into a cell, comprising a step of transporting the protein or peptide into the cell by using a conjugate formed by binding the protein or peptide with a polymer having a cation value of more than 2 and no more than 30,000 is disclosed in EP1316318A.

[0005] The systematic understanding of the peculiar effects that different ions exert on the biological matter can be traced back to Hofmeister, who studied their interactions with proteins. This has led to the formulation of the classical Hofmeister scale, where ions were subsequently classified as being kosmotropic if they showed salting-out properties towards proteins and as chaotropic if they displayed salting-in effects. Subsequently, these hydration effects have been generalized, the notion that the effects are a consequence of property changes of the bulk liquid has been abandoned, and, instead, direct interactions between chaotropic ions and dissolvedUW014P 2 organic matter (“chaotropic effect”) have moved into the focus. Although inorganic cluster anions of the boron and polyoxometalate (POMs) type have been intensively researched for many decades, their chaotropicity has moved into the focus of attention when it was found that these large cluster ions behave as “superchaotropic” ions, that is, they exceed the chaotropic behavior of the hitherto described anions by far. Since the utilization of dodecaborates (B12X122‒) as membrane carriers has been introduced [2] and their activity could be correlated with their superchaotropic nature, the exploration of this new application line to POMs is timely.

[0006] Bijelic A. et al. disclose that the combination of polyoxometalates (POMs) with drugs might be a promising strategy in cancer treatment. Disclosed are various POM-drug hybrids, e.g., the combination of different POMs with the anticancer drug 5-FU, with bisphosphonates, with quinoline antibiotics, or with amino acids [3].

[0007] Geisberger et al. disclose nanocomposites comprising a carrier molecule (trimethyl chitosan (TCM) and carboxymethyl chitosan (CMC)) and a cargo molecule (POM) [4]. Geisberger et al. showed that TCM are suitable drug carriers encapsulating bio-active POMs.

[0008] Therefore, there is still the need for a carrier molecule which may act as a new class of chaotropic membrane carrier to facilitate the transport of hydrophilic molecules across the cellular membrane. Summary of invention

[0009] It is the object of the present invention to provide a delivery molecule which facilitates the transport of molecules of interest across the cellular membrane. The object is solved by the subject matter of the present invention. The present invention relates to a delivery molecule which enables a cargo molecule to be translocated into cells, and a method for transducing the cargo molecule into the cells using said delivery molecule.

[0010] According to the invention, there is provided a delivery molecule comprising a carrier molecule and a cargo molecule, wherein said carrier molecule is a polyoxometalate moiety of the general formula [X1M1M211O40](n1)–, wherein X1denotes P, Si, Ge, or Al; M1denotes Mo, V, Ti, or Nb; M2denotes W, or Mo: n1denotes 2–6; or of general formula [X2M3M411O39(L)](n2)–, whereinUW014P 3 X2denotes P, Si, or Ge; M3denotes Mo, V, Fe, Ga, In, Ni, Co, Cr, or Al;, M4denotes W, or Mo; L is a terminal ligand and denotes O2–or H2O coordinated to M3, n2denotes 3–5; or of general formula [X3(OH)6M56O18]3–, wherein X3denotes Al, Ga, Cr, or Fe; and M5denotes Mo, or W; and wherein the cargo molecule is a hydrophilic moiety, an inorganic molecule, or an organic molecule. The carrier molecule and the cargo molecule are preferably directly linked to each other without a linker moiety.

[0011] A further embodiment relates to the delivery molecule as described herein, wherein said polyoxometalate moiety is of general formula [X1M1M211O40](n1)–, wherein X1denotes P, Si, Ge, or Al; M1denotes Mo, V, Ti, or Nb; M2denotes W; and n1denotes 2–6.

[0012] A further embodiment relates to the delivery molecule as described herein, wherein X1denotes P, or Si; M1denotes Mo, or V; M2denotes W; and n1denotes 4.

[0013] A further embodiment relates to the delivery molecule as described herein, wherein said polyoxometalate is of formula [PVW11O40]4–, or [SiMoW11O40]4–.

[0014] According to one embodiment of the invention, the polyoxometalate is stable at a pH of up to 7.5.

[0015] According to one embodiment of the invention, the cargo molecule is a hydrophilic molecule.

[0016] According to one embodiment of the invention, the cargo molecule is a hydrophilic peptide or hydrophilic protein.

[0017] According to another embodiment of the invention, the cargo molecule is a pharmaceutically active compound.UW014P 4

[0018] According to another embodiment of the invention, the pharmaceutically active compound may be a hydrophilic moiety or an organic compound.

[0019] A further embodiment of the invention relates to a pharmaceutically active compound used as cargo molecule as described herein. The pharmaceutically active compound may be a hydrophilic peptide, a hydrophilic protein, a hydrophilic organic compound, or a hydrophilic inorganic compound.

[0020] A further embodiment relates to the delivery molecule as described herein, wherein said polyoxometalate carrier is bound to said cargo molecule through supramolecular interactions.

[0021] A further embodiment relates to the delivery molecule as described herein, wherein said carrier molecule is present in a range of about 0.01 to 100 µM, or of about 0.1 to 10 µM, or of about 0.5 to 5 µM.

[0022] A further embodiment relates to the delivery molecule as described herein for use in treating a disease curable by said pharmaceutically active compound.

[0023] According to another embodiment of the invention, the pharmaceutically active compound is delivered to or removed from a site within a subject.

[0024] According to another embodiment of the invention, said delivery molecule passes a barrier within a subject based on a superchaotropic effect.

[0025] According to another embodiment of the invention, said barrier is a cell membrane.

[0026] A further embodiment relates to in vitro method for delivering a cargo molecule to a cell, comprising the steps of a. providing a cell, and b. contacting said cell with a delivery molecule as described herein.

[0027] One embodiment of the invention relates to the delivery molecule as described herein as a medicament.

[0028] A further embodiment relates to the delivery molecule as described herein, for use in the treatment of a disease curable by a pharmaceutically active molecule.

[0029] Another embodiment of the invention relates to the delivery molecule as described herein, for use in the treatment of a disease curable by a pharmaceutically active molecule. The a pharmaceutically active molecule may be a hydrophilic peptide or protein.

[0030] One embodiment of the invention relates to a pharmaceutical composition comprising the delivery molecule as described herein.UW014P 5 Brief description of drawings

[0031] Fig.1 POM archetypes. The Keggin (left) and the Anderson (right) archetype.

[0032] Fig.2 Ordering of the POMs according to their chaotropicity (projected based on their charge density), their experimental pH stability, and expected membrane lytic propensity; the window of POMs with practical utility as membrane carriers is shown in yellow.

[0033] Fig.3 (a) Schematic representation of the transport of impermeable oligoarginines facilitated by POMs in model membranes. (b) Cellular uptake of the labelled oligoarginine assisted by POMs.

[0034] Fig.4 (a) Changes in CF emission (λex= 492 nm, λem= 517 nm) in EYPC⊃CF vesicles as a function of time during the addition of increasing concentrations of [SiMoW11]4–(0–150 µM), heptaArg (10 µM), and TX100, for calibration. (b) Dose response curves for heptaArg transport for the active POMs and the corresponding Hill curve fits in comparison with the standards. (c) Pictures of the U-tube experiment (conducted with [SiMoW11]4–) at the opening (t = 0) and the closing (t = 24 h) of the experiment. (d) CF fluorescence measured in the trans phase of the U-tube. In the U-tube experiments, CF, POM, and heptaArg were fixed as 100 µM, 50 µM, and 10 µM.

[0035] Fig.5 (a) Confocal micrographs ofFITCArg8 uptake (4 µM) alone (no carrier) and promoted by different carriers (5 µM) in CHO-K1 cells. Inset scale bars: 10 µm. Representative images of two biological replicates. (b) MTT viability assay of CHO- K1 at different carrier concentrations (0 µM, 10 µM, 50 µM, and 100 µM). Error bars refer to standard deviation (n = 3).

[0036] Fig.6 Microcalorimetric titrations in 10 mM Mes, pH 5.5: Raw ITC data (top) for the sequential injections of POM solution into an oligoarginine solution, and apparent reaction heats (bottom) obtained from the integration of the calorimetric traces. POM / oligoarginine concentrations in µM: a, 100 / 8; b, 100 / 5; c, 100 / 1; d, 200 / 10; e, 100 / 3; and f, 100 / 0.9. Error data is 5 % for the n values and ± 0.5 kcal mol‒1for ∆H and T∆S (duplicate measurements). Note that none of these interactions lead to any precipitation or aggregates, confirmed by the absence of any detectable signal by DLS.

[0037] Fig.7 Microcalorimetric titrations in 10 mM Mes, pH 5.5: Raw ITC data (top) for the sequential injections of POM solution into a heptalysine solution, and apparent reaction heats (bottom) obtained from the integration of the calorimetric traces.UW014P 6 POM / heptalysine concentrations in µM: a, 200 / 8 and b, 150 / 10. Error data is 5 % for the n values and ± 0.5 kcal mol‒1for ∆H and T∆S (duplicate measurements). Note that these interactions did not lead to any precipitation or aggregates, confirmed by the absence of any detectable signal by DLS.

[0038] Fig.8 Transport activity, Y, for heptaarginine in (10 µM) EYPC⊃CF liposomes as a function of the concentration of the different evaluated POMs. Solid lines correspond to the resulting fit with the Hill equation. Blue squares are data obtained at neutral pH, and red stars data obtained at acidic pH.

[0039] Fig.9 Transport activity, Y, for protamine (1 µM) in EYPC⊃CF liposomes as a function of the concentration of the different evaluated POMs. Solid lines correspond to the resulting fit with the Hill equation. Blue squares are data obtained at neutral pH, and red stars data obtained at acidic pH.

[0040] Fig.10 Transport activity, Y, for polyarginine (0.1 µM) in EYPC⊃CF liposomes as a function of the concentration of the different evaluated POMs. Solid lines correspond to the resulting fit with the Hill equation. Blue squares are data obtained at neutral pH, and red stars data obtained at acidic pH.

[0041] Fig.11 Size distribution (by DLS) of EYPC⊃CF liposomes before (black trace), and after the addition of POM alone (100 µM, green trace), and in presence of heptaarginine (10 µM, red trace), protamine (1 µM, blue trace), or polyarginine (0.1 µM, purple trace). Higher [PW12]3–and [SiW12]4–lyse the vesicles.

[0042] Fig.12 Transport kinetics (upper panels) and dose response curves (bottom panels) for the transport of heptaarginine (10 µM) mediated by the superchaotropic POM carriers, [PVW11]4‒and [SiMoW11]4‒, vs pyrenebutyrate (PyBu) in dependence of carrier-cargo sequence of addition. a, 200 µM [PVW11]4‒. b, 60 µM [SiMoW11]4‒. c, 120 µM PyBu.

[0043] Fig.13 Successful transport of the less basic lysine derivative with [PVW11]4‒(a) and [SiMoW11]4‒(c). Comparable transport of heptalysine (circles, dash purple line) and heptaarginine (10µM) (squares, solid blue line) activated by [PVW11]4‒(b) and [SiMoW11]4‒(d).

[0044] Fig.14 Heptaarginine transport through anionic vesicles (DMPE / DPPG / CHOL, 1 / 2 / 1 molar ratio) facilitated by [PVW11]4‒(a) and [SiMoW11]4‒(c). Transport activity as function of [PVW11]4‒(b) and [SiMoW11]4‒(d) concentration, solid lines correspond to the resulting fit with the Hill equation.UW014P 7

[0045] Fig.15 Confocal fluorescence microscopy images ofFITCArg8uptake (4 µM) alone (no carrier) and promoted by the different carriers (10 µM) in CHO-K1 cells. Inset scale bars: 10 µm. Representative images of two biological replicates.

[0046] Fig.16 Confocal fluorescence microscopy images ofFITCArg8uptake (4 µM) alone (no carrier) and promoted by the different carriers (50 µM) in CHO-K1 cells. Inset scale bars: 10 µm. Representative images of two biological replicates.

[0047] Fig.17 The toxicity profile of [PVW11]4-, examined in different cell lines in vitro by the AquaBluer™ test. Description of Embodiments

[0048] The present invention provides a delivery molecule comprising a polyoxometalate as carrier molecule and a cargo molecule selected from the group consisting of a hydrophilic moiety, an organic moiety, and an inorganic moiety.

[0049] As used herein, a “delivery molecule” refers to a molecule which comprises a carrier molecule and a cargo molecule, enabling the transport of the latter into a compartment enclosed by a chemical and / or biological barrier, and finally releasing said cargo molecule into said compartment.

[0050] As used herein, a “carrier molecule” refers to any molecule which is able to contacting a second molecule i.e., a cargo molecule, enabling the transport of the latter into a compartment enclosed by a chemical and / or biological barrier, and finally releasing said substance into said compartment.

[0051] As used herein, a “cargo molecule” refers to any substance that is in contact with a carrier molecule and transported by said carrier molecule into a compartment enclosed by a chemical and / or biological barrier. Without a carrier molecule, the cargo molecule is not able to cross the barrier of said compartment.

[0052] The term “linker-free” or “without a linker” as used herein refers to the absence of any intermediate compound which may connecting the carrier molecule and the cargo molecule via covalent and / or non-covalent interactions.

[0053] The term “hydrophilic” is well known in the art and refers to the capacity of a molecular entity or of a residue to interact with polar solvents, in particular with water, or with other polar groups. By extension, the term “hydrophilicity” refers to the tendency of a molecule to be solvated by water.

[0054] The hydrophobicity of a molecule may be quantified using, but not limited to, tables based on empirical data, partitioning methods, accessible surface area methods, experimentally (e.g. reversed phase chromatography), and physicalUW014P 8 property methods (based on physical properties such as but not limited to partial molar heat capacity, transition temperature and surface tension).

[0055] The term “hydrophilic moiety” as used herein refers to a structure that can dissolve in polar solutions (e.g., water). Usually, hydrophilic substances are polar in nature. Hydrophilic substances tend to readily dissolve in water or polar solvents.

[0056] Hydrophilic molecules or hydrophilic moieties are predominantly polar compounds that have ionic groups. The polar nature of these hydrophilic molecules or hydrophilic moieties enables them to readily absorb water or polar solvents and eventually get dissolved in polar solvents like water. Being a polar protic solvent, water is capable of forming a hydrogen bond (-H—OH-). Hydrophilic molecules or hydrophilic moieties are polar in nature and easily form a hydrogen bond(s) with water thereby getting dissolved in water. Notably, these interactions between the hydrophilic molecule or hydrophilic moiety and water are thermodynamically favored. In general, hydrophilic substances can easily form hydrogen bonds with polar solvents like water, and alcohol. The polarity of a moiety usually defines its hydrophilicity. Hydrophilic molecules usually cannot penetrate a lipid biolayer and cell membrane themselves.

[0057] The hydrophilic moiety according to the invention may be a peptide or protein having hydrophilic properties.

[0058] Polyoxometalate (POM) refers to a polyatomic ion, usually an anion, which consists of three or more transition metal oxyanions linked together by shared oxygen atoms to form closed 3-dimensional chemical structures.

[0059] In one embodiment of this invention the metal atoms of the POM may be selected from the elements in groups 4 to 10, or group 13 of the latest released periodic table of the elements by International Union of Pure and Applied Chemistry (IUPAC), hereafter referred to as periodic table. The metal atom may be selected from the group consisting of titanium, chromium, vanadium, niobium, molybdenum, tungsten, iron, cobalt, nickel, gallium, and indium, or from the group consisting of titanium, vanadium, niobium, molybdenum, and tungsten, or from the group consisting of molybdenum, vanadium, iron, gallium, indium, nickel, cobalt, chromium, and tungsten. According to one embodiment of the invention, the metal ion is selected from the group consisting of vanadium, molybdenum, and tungsten.

[0060] According to one embodiment, the main group oxyanion of the POM may be formed with any of the elements in groups 14, or 15 of the periodic table, such as forUW014P 9 example, from the group consisting of phosphor, silicon, and germanium. The hetero- ion of the POM may be selected from the group consisting of phosphate, silicate, and germanate. The ligand of the POM may be selected from the group of molecular oxygen and water.

[0061] The charge of the POM may be -6 to -2, or -5 to -3, or -4 to -3.

[0062] The term “supramolecular interaction” as used herein refers to an intermolecular non-covalent interaction between two molecules. These forces may include hydrogen bonding, metal coordination, hydrophobic forces, van der Waals forces, pi-pi interactions, or electrostatic effects, or combinations thereof. Interaction. Examples of supramolecular interactions include hydrogen bonding, hydrophobic interaction or hydrophobic bonding, electrostatic interaction or ionic bonding, salt bridging, and coordinate bonding.

[0063] As used herein, the term “pharmaceutically active compound” refers to substances that are commonly used in medicine and biotechnology. A pharmaceutically active compound is any ingredient that provides biologically active or other direct effect in the diagnosis, cure, mitigation, treatment, or prevention of disease or to affect the structure or any function of the body of humans or animals. Non-limiting examples of pharmaceutically active compound classes are small molecules, antibodies, antibiotics, peptides, non-coding RNA, and the like.

[0064] Any pharmaceutically active compound which can be connected to the carrier molecule may be used according to the invention. Suitable pharmaceutically active compounds are for example, hydrophilic peptides or proteins such as phalloidin, pancuronium, vitamin B1, cartilage derived retinoic acid sensitive protein (CD-RAP), bone morphogenetic protein (BMP), or kanamycin A. The pharmaceutical active compound may be an organic compound or even an inorganic compound such as cis-platin.

[0065] As used herein, the term “site” refers to a closed compartment which is enclosed by a chemical and / or biological barrier, e.g., to a cell enclosed by a plasma membrane, or to two phases separated by a chemical layer.

[0066] As used herein, the term “delivered to”, when related to a cargo molecule and a site, refers to the facilitated transport of said cargo molecule to a target site by a carrier molecule. In a specific embodiment, the cargo molecule dissociates from the carrier molecule once the target site is reached.UW014P 10

[0067] As used herein, the term “removed from”, when related to a cargo molecule and a site, refers to the contacting of said cargo molecule by a carrier molecule and the subsequent facilitated transport away from said site.

[0068] As used herein, a “superchaotropic effect” refers to a most extreme case of the chaotropic effect, which refers to the propensity of large anions to associate with hydrophobic and neutral polar phases, contrasting the hydrophobic effect as a distinct assembly motif in chemistry. Superchaotropic ions are characterized by enthalpically driven processes with negative entropic contributions. These superchaotropic anions would be introduced in the extended Hofmeister Serie between chaotropic and hydrophobic ions.

[0069] The concentration of the carrier molecule as used herein may range from about 0.01 to 100 µM, about 0.1 to 10 µM, about 0.5 to 5 µM, or about 1 to 300 µM. In a specific embodiment, the concentration of the carrier molecule and the cargo molecule is equimolar. In another specific embodiment, the concentration of the cargo molecule is 2, 3, 4, 5, 6, 7, 8, 9, or 10 fold higher than the concentration of the carrier molecule.

[0070] A “disease curable by a pharmaceutically active compound” as used herein, refers to the treatment of said disease by said pharmaceutically active compound resulting in a full or partial remission of said disease, specifically resulting in the partial or complete disappearance of clinical and subjective characteristics of the disease.

[0071] Examples Example 1

[0072] The Examples which follow are set forth to aid in the understanding of the invention but are not intended to, and should not be construed to, limit the scope of the invention in any way. The Examples do not include detailed descriptions of conventional methods. Such methods are well known to those of ordinary skill in the art. Synthesis

[0073] All reagents and chemicals were of high-purity grade and were used as purchased without further purification. Carboxyfluorescein (CF), cholesterol, CHO-K1, dulbecco’s phosphate buffered saline (DPBS), 1,2-dimyristoyl-sn-glycero-3- phosphoethanolamine (DMPE), 1,2-dipalmitoyl-sn-glycero-3-phospho-(1'-rac-UW014P 11 glycerol) (DPPG), egg yolk phosphatidylcholine (EYPC), Hank´s balanced salt solution, nutrient mixture F-12 Ham, protamine, pyrenebutyrate, were from sigma- Aldrich (Steinheim, Germany). Resazurin was from TCI (Eschborn, Germany). Trp- Arg7and Trp-Lys7were custom-made by GeneCust (Boynes, France) in > 95 % purity as confirmed by HPLC and MS by the supplier. Peptide stock solutions were prepared in water and their concentration was determined by using the extinction coefficient of tryptophan at 280 nm (ε = 5540 cm–1M–1). FITC-Arg8was from GL Biochem (Shanghai, China) Ltd, and its concentration was determined by using the extinction coefficient of 5-FITC at 491 nm (ε = 73000 cm–1M–1). Na3[α-PVWVI12O40]·nH2O ([PW12]3‒for [α-PVWVI12O40]3‒) was purchased from Sigma- Aldrich. K4[α-SiIVWVI12O40]·14H2O ([SiW12]4‒for [α-SiIVWVI12O40]4‒) [5], Na5[α-AlIIIWVI12O40]·12H2O ([AlW12]5‒for [α-AlIIIWVI12O40]5‒) [6], (C4H12N)4[α-HAlIIIGeIVWVI11O39(H2O)]·11H2O ([GeAlW11]5‒for [α- AlIIIGeIVWVI11O39(H2O)]5‒) [7], K4[α-PVVVWVI11O40]·4H2O ([PVW11]4‒for [α-PVVVWVI11O40]4‒) [8], K4[α-SiIVMoVIWVI11O40]·24H2O ([SiMoW11]4‒for [α-SiIVMoVIWVI11O40]4‒) [9], (Et2NH2)8[{α-PVWVI11O39ZrIV(µ-OH)(H2O)}2]·7H2O ([(PZrW11)2]8‒for [{α- PVWVI11O39ZrIV(µ-OH)(H2O)}2]8‒)

[0010] , Na3[AlIII(OH)6MoVI6O18]·2H2O ([AlMo6]3‒for [AlIII(OH)6MoVI6O18]3‒)

[0011] were prepared according to literature procedures. All salts were characterized in solution by NMR spectroscopy and in the solid-state using infrared (IR) spectroscopy and thermogravimetric analysis (TGA). IR bands (cm–1) for the synthesized compounds: [PW12]3–(1074, 971, 888, 769, 574, 511); [PVW11]4‒(1100, 1074, 1058, 972, 876, 760, 596, 515, 473, 428); [(PZrW11)2]8‒(1099, 1053, 956, 883, 767, 698, 656, 594, 513, 487, 408); [SiW12]4‒(1016, 970, 912, 881, 748, 521, 476); [SiMoW11]4‒(1016, 974, 908, 874, 729, 553, 528, 474, 416); [AlW12]5‒(962, 953, 904, 878, 795, 746, 530, 468); [GeAlW11]5‒(950, 874, 820, 764, 677, 517, 447); [AlMo6]3‒(943, 902, 867, 797, 723, 626, 611, 568, 509, 482, 447). The mass-loss for [PW12]3–is 4.7 % that corresponds to 7 H2O; for [PVW11]4‒is 2.5 % that corresponds to 4 H2O; for [(PZrW11)2]8‒is 12.1 % that corresponds to 9 H2O and 8 Et2NH2; for [SiMoW11]4‒is 12.8 % that corresponds to 24 H2O; for [AlW12]5‒is 6.8 % that corresponds to 12 H2O; for [GeAlW11]5‒is 15.7 % that corresponds to 12 H2O and 4 C4H12N; for [AlMo6]3‒is 8.9 % that corresponds to 5 H2O.UW014P 12 Attenuated total reflection Fourier−transform Infrared Spectroscopy

[0074] All FTIR spectra were recorded on a Bruker Vertex 70 IR Spectrometer equipped with a single−reflection diamond−ATR unit. Frequencies are given in cm‒1, intensities denoted as w = weak, m = medium, s = strong, br = broad. Elemental analysis

[0075] Elemental microanalysis of C / H / N / O contents was performed by Mikroanalytisches Laboratorium (University Vienna, Faculty of Chemistry). An EA 3000 (Eurovector) was used for C / H / N / S-analysis. O-determination was performed by high temperature digestion using the HT 1500 (Hekatech, Germany) pyrolysis system in combination with the EA 3000 system. Thermogravimetric analysis (TGA)

[0076] TGA was performed on a Mettler SDTA851e Thermogravimetric Analyzer under N2flow with a heating rate of 5 K min‒1in the region 298−873 K. Isothermal Titration Calorimetry (ITC)

[0077] All experiments were performed with a VP-ITC MicroCalorimeter from MicroCal, Int,. at atmospheric pressure and 298.15 K. Solutions were degassed and thermostatic prior the titration experiments in a ThermoVac accessory. A constant volume of POM ([PVW11]4‒or [SiMoW11]4‒) (10 µL per injection) was injected into the oligopeptide solution (heptaArg, protamine, polyArg or heptaLys) in 10 mM Mes, pH 5.5 to determine the apparent binding affinity of [PVW11]4‒and [SiMoW11]4‒with the oligopeptides. Dilution heats were determined by titration of the POM into buffer and subtracted from the reaction heat. The neat reaction heat was fitted with Origin 7.0 software by using the “one-set-of-sites” model to obtain the complex stability constant (Ka) and molar reaction enthalpy (∆Hº). The free energy (∆Gº) and entropy changes (∆Sº) were obtained according to the relation: ∆Gº = − RTlnKa = ∆Hº−T∆Sº. Vesicle preparation

[0078] Zwitterionic vesicles in neutral media: A thin lipid film was prepared by evaporating a lipid solution of EYPC (25 mg) in CHCl3(1 mL) with a stream of nitrogen and then dried in vacuo overnight. The dry film was rehydrated with 1 mL buffer (50 mM CF, 10 mM Hepes, pH 7.4 or 50 mM CF, 10 mM Tris, pH 7.4) for 30 min at room temperature, subjected to freeze–thaw cycles (7 times) and extrusions (15 times) through a polycarbonate membrane (pore size 100 nm). Extravesicular components were removed by size exclusion chromatography (NAP-25 columnUW014P 13 Sephadex G-25 DNA grade) with 10 mM Hepes, 107 mM NaCl, pH 7.4 or 10 mM Hepes, 107 mM NaCl, pH 7.4. Phospholipid concentration was calculated by the Stewart assay.

[0079] Zwitterionic vesicles in acidic media: Acidic vesicles were prepared analogously but the rehydration and elution buffer were 50 mM CF, 10 mM Mes, pH 5.5 and 10 mM Mes, 107 mM NaCl, pH 5.5, respectively.

[0080] Anionic vesicles in neutral media: Anionic vesicles were prepared analogously except for the lipid composition, DMPE / DPPG / CHOL (4.4 / 10.4 / 2.6 mg, 1 / 2 / 1 molar ratio) in a 1:1 mixture of CHCl3and MeOH (1mL), the hydration (60 mins) and extrusion were performed at 55 ºC, and the employed buffers were 100 mM CF, 10 mM Tris, pH 7.4 and 10 mM Tris, 140 mM NaCl, pH 7.4 for rehydration and elution, respectively. Transport experiments

[0081] Vesicles stock solutions (5-10 µL) were diluted with the corresponding buffer in a disposable plastic cuvette and gently stirred (total volume 2000 µL, final lipid concentration 13 µM). Carboxyfluorescein fluorescence was monitored at λem=517 nm (λex=492 nm) as a function of time after addition of POMs at 60 s, analyte at 120 s and Triton X-100 (24 µL 1.2 % wt / vol) at 600 s to lyse the vesicles, for calibration. Fluorescence intensities were normalized to fractional emission as:

[0082] Where I0= Itbefore POM addition and I0= I∞after lysis. For Hill analysis, Itbefore lysis was defined as transport activity, Y, and plotted against POM concentration, c, and fitted to the Hill equation (2), to afford the activity in absence of POM, Y0,the maximal activity, Ymax, the concentration needed to achieve 50 % of maximal activity, EC50, and the Hill coefficient, n.Activator efficiency

[0083] In the transport measurements, where different carriers (POMs) are tested with the same cargo, the activator efficiency (E) is determined by their capacity to transport the impermeable cargo to the intravesicular region and is characterized by Ymax, its maximal activity, and EC50, the effective activator concentration. An idealUW014P 14 carrier reaches high Ymaxat low EC50. To count both factors, the activator efficiency is defined as:

[0084] were pEC50is the negative logarithm of the EC50, and f a scaling factor, which was set to 20.6 to calibrate the highest (previously known) activation efficiency, E, as 10. U-tube transport experiments

[0085] The U-tubes were house-made, similar to those of Rebek

[0012] and Matile

[0013] consisting in a small beaker with a central glass barrier, separating the two aqueous layers, cis (sampling phase) and trans (receiving phase), but allowing an interface chloroform layer below the cis and trans phases.3 mL CHCl3 are placed into the U- tube and 1 mL of the cis (combinations of carrier (POM), cargo (heptaArg) and CF in buffer) and trans (1 mL buffer) phases were added. The organic phase was stirred at 700 rpm at room temperature. Aliquots (25 µL) from the trans phase were taken at different times, 400 µL of buffer were added and measured by fluorescence. Cell culture and confocal imaging

[0086] For confocal microscopy experiments, CHO-K1 cells were seeded into 12- well plates (Eppendorf) at a density of 1×105cells / well (per 1 mL) and incubated for 24 h in Ham’s F-12 containing 10 % (v / v) fetal bovine serum at 37 ºC in a 5 % CO2. The cells were washed twice with PBS, and then incubated with POMs and FITC- Arg8(4 µM) for 1 h at 37 ºC, cells were washed twice with Hank´s Balanced Salt solution containing 100 µg / mL heparin, twice again with PBS and fixed with 1 mL cold 4 % paraformaldehyde for 20 min. The cells were rewashed twice with PBS and subsequently the nuclei were stained with 2 µg / mL DAPI for 10 min at room temperature before imaging. Cells were imaged by LSM 980 confocal laser scanning microscope. Images were processed with Zen 3.2 software (Zeiss). Cellular viability assay

[0087] For MTT assays in presence of activators, CHO-K1 cells were incubated in Ham’s F-12 medium containing 10 % fetal bovine serum and 1.0 % penicillin streptomycin at 37 ºC in a 5 % CO2for 24 hours. CHO-K1 cells were dispersed in a 96-well flat bottom tissue culture treated plates (Eppendorf) at density of 5×103cells / well (per 100 µL) and incubated in Ham’s F-12 containing 10 % (v / v) fetal bovine serum at 37 ºC in a 5 % CO2incubator for 12 h to allow cell adhesion. Then, cellsUW014P 15 were incubated with POM, B12Br122−, or PyBu (10 µM, 50 µM, 100 µM) for another 24 h. After this incubation time, 20 µL resazurin solution (0.15 mg / mL) was added to cells and incubated for another 3 h. Finally, the plates were read at λem= 590 nm (λex= 560 nm) using a JASCO fluorometer FP-8500.

[0088] The cell viability was evaluated by the following equation: ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^^^ =100 %( 4 )

[0089] Where, FIsampleis the fluorescence intensity of cells incubated with different compounds, FIblankis the fluorescence intensity of medium, and FIcontrolis the fluorescence intensity of cells in absence of any compound. Results and Discussion

[0090] POMs of the globular Keggin and planar Anderson archetype and with varying charge status from ‒3 to ‒8 (Fig.1) were selected. To offset the known [XW12O40]n‒(X = hetero-ion) hydrolysis propensity in aqueous solution, most pronounced at neutral pH, more highly charged mixed-metal derivatives (Fig.1,left- hand side) were also included. The latter are obtained by hydrolytic removal of one W=O4+unit from the Keggin structure and the subsequent incorporation of an addenda metal to fill the cavity of the monolacunary intermediate. In detail, the following POM anions were studied (as acid or sodium, potassium, or tetraalkylammonium salts), see Fig.1: [α-PVWVI12O40]3–([PW12]3‒), [α-SiIVWVI12O40]4–([SiW12]4‒), [α-AlIIIWVI12O40]5–([AlW12]5‒), [α-PVVVWVI11O40]4–([PVW11]4‒), [α- SiIVMoVIWVI11O40]4–([SiMoW11]4–), [α- HAlIIIGeIVWVI11O39(H2O)]5–([GeAlW11]5‒), [{α- PVWVI11O39ZrIV(µ-OH)(H2O)}2]8–([(PZrW11)2]8‒), and [AlIII(OH)6MoVI6O18]3–([AlMo6]3‒). First, their long-term (24 h and longer) stability in aqueous solution (slightly acidic and neutral) was investigated by applying the active NMR nuclei in their composition (see). The Keggin anions [PW12]3‒and [SiW12]4‒were found to be fully hydrolyzed at neutral or even slightly acidic pH to their monolacunary forms (Fig. S29-S31), but they were included as negative controls, to exclude potential carrier activity arising from degradation products. The other selected POMs were found to be stable at neutral pH (7.5), or only very slowly hydrolyzed, as summarized in Table 1.

[0091] The capability of POMs to act as membrane carriers was explored first in large unilamellar vesicles (LUVs), loaded with carboxyfluorescein (CF). We selected three representative cationic peptides, heptaarginine (heptaArg), heptalysine (heptaLys), protamine, and polyarginine (polyArg), as test cargos. Inside the LUVs,UW014P 16 the CF dye was encapsulated at self-quenching concentrations. With a suitable carrier, heptaArg can be taken up into the vesicles, bind with entrapped CF, and shuttle back to the exterior, spreading the previously quenched CF, and thereby restoring its fluorescence (Fig.3). In the time-resolved fluorescence experiments, to test whether a POM acts as molecular carrier, the CF emission was continuously monitored during the addition of the inorganic carrier (t = 60 s) and after injection the impermeable heptaArg (t = 120 s). At the end of the experiment (t = 600 s), the surfactant Triton X-100 was added to release all the entrapped CF and allow normalization of the intensity data (Fig.4a).

[0092] It should be noted that the addition of the POMs alone (without cargo) retained the stability of the liposomes in the studied concentration range (up to 100 µM, see Fig.11), that is, neither an increase in CF fluorescence nor changes in size distribution as obtained by dynamic light scattering experiments (DLS) were observed, which would have signaled membrane rupture. These negative controls allowed us to proceed to the carrier experiments, with added cargo. Amongst the eight inorganic clusters evaluated, we found two hits: The two mixed metal Keggin POMs [PVW11]4‒and [SiMoW11]4‒triggered the desired increase in fluorescence characteristic for cargo transport. Interestingly, the planar Anderson [AlMo6]3‒ showed also incipient transport activity, but only for polyarginine as cargo, while other Anderson-type POMs were inactive.ob bs e erN[etrwro a a] PB[A [( [A [A [S [P [S [PCTav er CE y 1l YB2 BlMPZlG lW iM V iW War blee e(S,p PC u‒1rorWeW12 oWW1 1 r2 2 ie1d na 0D)Hv226‒] 3t‒11 1 ] 15‒11 ] 4 ] 4‒] 3‒ rSts a . 7n A . em [2 ) 1 ] 1 ]‒ac 4 si ] ]85 4‒‒ ‒btrodn 1tg 0iv oarc1l ie lisl .st st / / / / st sty ;a 1ti0o0nm [bac 1] bale ba ale ble bleid %eM f0V .i fcei1I5olylecTµM1Oe []sictprieris 3Hro9 s / eS d.r n. c,1he ] 7‒ ta PN y 0pb O[mM] oE 7d= mtaAr st st st st st st st [P2 tilo / pitM ysea es temYMga a a a a a aN.V 0Hinu c a[xac b] le ble ble ble ble bl be leW%[SVti7WV.5acre ta (db pC 11I1olf [idieEl,µO pMV]38I cb Cy a5H c0Hp 9 ]Oa7‒3ndPti / f) 7. ro 9t] 8‒ nTviS t ,y w4a e / e. [ m utD aPtrXpegM]inHpEa er.1 9 - - - - 6 8 - -Y Hal pCxim[d] 005 - - - - - - - - 6 4 - -- e- m aptHas5s.5 5 a0. [j] [j] [j] [j] [jil [j,kxg [ aAanaya 0 in s l a1 tµ ] ]( ]r ddth ctM % g)b[m.7.e iv4.ni[etyp ] e;oly 10 48174m9(µ ECbrMk]ga±ti 2Ar0 Mg) 50 ah[ neevm e % . ][trblee1]anroagrro0 snaeri r m4t. 6. 5. 5. Epodh (SM8 1 7 3 i [ riD ] tsm)r .M puo e 7 8 7 8YParpf [h] s7 4 2 8 mroamtiEoCE ,f 1n 50ax (taeto 0 f, ec7 % me) i rsb a timns n ve d eMec[atrv f coN3a9111941(µMEC]pHeda ncC) 507asc e l, .4tal nt p .stirnogra Ht5 5nf . 8 6 5 Eaion.5.3.0.0.9glyctaoto [f]9 9 3 9 8Y cridsre e 1a 01 4 3 4 8 mPmax olyic t t co hM (% Arp 5)gH 2.0 0 Hed[][.6 %pe 122 111 1 (µ ECl]atoo .63 2 fs,M5C0 rY 1)raiellrom a0x7acat; 5mct eM8 1 3 8 8 Eiv %.52.1.6.2i EtyUW014P 18 Table 2. Results of stability investigation by NMR spectroscopy. POM anions with % for 31P,51V,183W or27Al parent (active) anion POM Buffer / Media shifts, ppm based on signals integration D2O31P: ‒14.8;51V: ‒555 [PVVVWVI11O40]4‒Mes pH 5.5 freshly31[PVWVIP: ‒10.8 and ‒14.811O39]7‒(1 %) and prepared [PVVVWVI11O40]4‒(99 %) Mes pH 5.5 after 1V31[P WVI11O39]7‒(1 %) and P: ‒10.8 and ‒14.8 day [PVVVWVI11O40]4‒(99 %) Mes pH 5.5 after 5 [PVWVI 7‒31P: ‒10.8 an11O39] (1 %) and d ‒14.8 days [PVVVWVI11O40]4‒(99 %) Tris-HCl pH 7.531P: ‒11.4, ‒14.4 and ‒ [PVWVI11O39]7‒and freshly prepared 14.9 [PVVVWVI11O40]4‒(81 %) Tris-HCl pH 7.5 after31P: ‒11.4, ‒14.4 and ‒ [PVWVI11O39]7‒and 1 day 14.9 [PVVVWVI11O40]4‒(71 %) Tris-HCl pH 7.5 after31P: ‒11.4, ‒14.4 and ‒ [PVWVI11O39]7‒and [PVW11]4‒5 days 14.9 [PVVVWVI11O40]4‒(68 %) VVI 7‒31[P W P11O39] : 0.4 (free phosphate (coordinated to different Nutrient mixture F-12 from medium), ‒10.6, ‒ alkali and earth alka Ham[a]li freshly 10.7, ‒10.8, ‒11.1 and metals leading t prepared .9;51o ‒14 V: ‒509.2, ‒ different shifts) and 558.8 [PVVVWVI11O40]4‒(43 %) 31P: 0.4 (free phosphate [PVWVI11O39]7‒Nutrient mixture F-12 from medium), ‒10.1, ‒ (coordinated to different Ham[a]after 10.8, ‒13.7, ‒14.4 and alkali and earth alkali incubation for 1 h at ‒14.9;51V: ‒509.2, ‒ metals leading to 37 °C 523.1, ‒526.7, ‒554.2, different shifts) and ‒558.8, ‒559.7 [PVVVWVI11O40]4‒(16 %) 183W: ‒99.9, ‒100.2, ‒ Mes pH 5.5 102.4, ‒103.0, ‒103.5, [SiIVMoVIWVI11O40]4‒‒107.4 183W: ‒99.3, ‒99.6 [SiMoW11]4‒, ‒ Tris-HCl pH 7.5 [SiIVMoVIWVIO ]4‒101.8, ‒102.8, ‒106.711 40Nutrient mixture F-12183W: ‒99.4, ‒99.7, ‒ Ham after incubation 101.9, ‒102.5, ‒102.9, [SiIVMoVIWVI11O40]4‒for 1 h at 37 °C ‒106.9 Hepes pH 7.5 freshl [AlMo6]3‒y27Al: 16.3 [AlIII(OH) MoVIO ]3‒prepared6 6 18UW014P 19 Hepes pH 7.5 after 127Al: 16.3 [AlIII(OH)MoVIO ]3‒day6 6 18Tris-HCl pH 7.527Al: 72.6 [AlIIIWVI12O40]5‒5‒freshly prepared [AlW12] Tris-HCl pH 7.5 after27Al: 72.6 [AlIIIWVIO ]5‒1 day12 40Tris-HCl pH 7.5III IV VI27[HAl Ge W Al: 9.411O39(H2O)] freshly prepar4‒115‒ed [GeAlW ] Tris-HCl pH 7.5 after27[HAlIIIGeIVWVIO Al: 9.411 39(H2O)] 1 day4‒[{α-PW11O39Zr(µ- Hepes pH 7.5 freshly31P: ‒13.7, ‒14.8, ‒14.9 OH)(H2O)}2]8‒·(99 %) and prepared [Zr(α-PW11O39)]10‒[(PZrW1)]8‒21 2[{α-PW pH 7.5 after 111O3Zr(µ- Hepes931P: ‒13.7, ‒14.8, ‒14.9 OH)(H2O)}2]8‒·(99 %) and day [Zr(α-PW11O39)2]10‒NaOAc / HOAc pH 431P: ‒12.4, ‒13.1, ‒ [PVWVI12O40]3‒(30 %) freshly prepared 13.7; ‒15.2 and intermediate anions NaOAc / HOAc pH 431P: ‒12.4, ‒13.1, ‒ [PVWVI12O40]3‒(28 %) after 1 day 13.7; ‒15.2 and intermediate anions [PVWVI1O ]3‒(4 %), [PW ]3‒Mes pH 5.5 fresh312 4012ly P: ‒11.0, ‒12.4; 13.1, [PVWVIared ‒13.7; ‒14.711O ]7‒(61 %) prep39and intermediate anions [PVWVIO ]3‒3112 40(5 %), Mes pH 5.5 after 1 P: ‒11.0, ‒12.4; 13.1; [PVWVI11O39]7‒(61 %) day ‒13.3; ‒13.9; ‒14.7 and intermediate anions [SiIVWVI12O40]4‒and [SiIVWVI11O39]8‒(high 183noise to signal ratio, W: ‒99.8; ‒100.8; ‒ co 124‒rrect integration of [SiW ] Hepes pH 7.5 114.5; ‒119.8; ‒127.8; peak cannot be ‒140.4; ‒174.6 performed; but clearly [SiIVWVI11O39]8‒ispredominant) [a] According to the provider Sigma-Aldrich, Nutrient Mixture F-12 Ham for cell culturing contains sodium pyruvate (0.11 g / L), phenol red, L-glutamine, and does not contain NaHCO3and Hepes, for more details please see https: / / www.sigmaaldrich.com / AT / en / technical- documents / technical-article / cell-culture-and-cell-culture-analysis / mammalian-cell-culture / f- 12-ham.UW014P 20

[0093] Within the Keggin series, the membrane carrier potential appears to be related to the chaotropicity of the clusters, which is expected to decrease with increasing charge status, namely [PW12]3‒> [SiW12]4‒, [PVW11]4‒, [SiMoW11]4‒> [AlW12]5‒, [AlGeW11]5‒. When [PW12]3‒and [SiW12]4‒were kept at pH 1.2, already the addition of [PW12]3‒alone to the liposomes caused an increase in fluorescence, showing strong interaction and membrane disruption. [SiW12]4‒showed the desired cargo transport at acidic pH, without membrane disruption. However, above pH 4.5, [PW12]3‒and [SiW12]4‒undergo rapid hydrolysis to afford the highly charged products of hydrolysis[PVWVI11O39]7‒and [SiVWVI8O39]8‒, displaying neither an interaction with the membrane nor affecting cargo transport (Fig.8–10). Accordingly, they have no biologically relevant carrier utility at neutral pH, discouraging biological use. In contrast, the two Al-based clusters [AlW12]5‒and [AlGeW11]5–are stable at neutral pH, but their higher charge appears to adversely affect their transport property. The same applies for the sandwich cluster [(PZrW11)2]8‒, whose total charge appears to be too high. Accordingly, we concentrated on the two mixed-metal clusters [PVW11]4‒and [SiMoW11]4‒, which showed the desired interplay of sufficiently low net charge (–4), good hydrolytic stability near neutral pH, efficient carrier activity, and broad cargo scope.

[0094] To characterize the transport capability of the new inorganic carriers, the resulting normalized fluorescence traces were plotted versus the POM concentration, generating dose-response curves (Fig.4b, Fig.8–10), which could be analyzed by Hill analysis. The resulting membrane transport parameters are the maximal activity (Ymax), the POM concentration that is required to achieve 50 % of maximal activity (EC50), and the activator efficiency (E), which are compiled in Table 1. For the selected cargos, the two POM carriers [PVW11]4‒and [SiMoW11]4‒show comparable E values to the standards in the field, pyrenebutyrate (PyBu), as amphiphilic carrier, and B12Br122‒, as superchaotropic carrier (Table 1). And although the transport in the vesicles does not reach the same high final Ymaxlevels, which results in a lowering of the E values, the onset for transport activity occurs at much lower carrier concentrations, around 1 µM (Fig.4). Accordingly, the required carrier concentrations are desirably low, e.g., the EC50value of [SiMoW11]4‒is about a factor of 3 lower (16.5 µM versus 48 µM) than that of the boron cluster. The efficacy of the clusters did not markedly decrease when the sequence of addition was reversed (Fig.12). This reflects the reversibility of the involved supramolecular interactions, which is notUW014P 21 observed, for example, for PyBu, where (irreversible) cluster cargo aggregation is known to interfere and significantly reduces its activity (Fig.12). Another contrast to the amphiphilic PyBu is that the carrier activity of the POM carriers was found to be transferable to the less basic heptalysine (Fig.13); in fact, the activity characteristics for oligoarginine and oligolysine transport were very similar (Table 3). Note also that the carrier activity of [PVW11]4‒and [SiMoW11]4‒in the model membranes was found to be comparable at pH 5.0 and pH 7.4 for all cargos listed in Table 1 and 3, which showed that the minor hydrolysis observed for the former on long time scales at neutral pH (ca.20 % after 24 h) did not affect its carrier activity significantly. Table 3 Membrane transport parameters[a]of active POMs at neutral[b]and acidic[c]pH. HeptaArg[d]Protamine[e]PolyArg[f]Hydrolysis / carrier pH Stability20 % hydrolyzed 4‒7.5 to 8 49 5.3 88 41 5.9 88 12 8.2 [PVW11] [PVWVI11O39]7‒5.5 stable 79 28 5.9 79 19 6.6 91 5 10.1 [SiMoW11]4‒7.5 stable 66 17 5.7 72 19 6.0 94 13 8.6[k]5.5 stable 66 9 6.6 80 14 7.3 86 5 9.7 [a] EYPC vesicles. [b] 10 mM Tris, 107 mM NaCl, pH 7.4. [c] 10 mM Mes, 107 mM NaCl, pH 5.5. [d] 10 µM heptaArg. [e]1 µM Protamine. [f] 0.1 µM polyArg. [g] Maximal activity; ±2 % error (SD). [h] Effective concentration to reach 50 % activity of Ymax; 5 % error (SD). [i] Activation efficiency E = Ymax(pEC50 / f), were pEC50is the negative logarithm of the EC50, and f a scaling factor set to 20.6 to allocate E between 0 and 10, 10 % error.

[0095] To affect membrane transport, any carrier needs to display an affinity towards the cargo as well as towards the lipid bilayer. Accordingly, the interaction between [PVW11]4‒and [SiMoW11]4‒and the cargos was analyzed by isothermal titration calorimetry (ITC). These experiments showed strong supramolecular interactions (on the order of 107M−1) with affinities governed by the chaotropic effect (enthalpically driven, interaction with the peptide backbones, Fig.6–7 and Table 4)UW014P 22 and Coulombic interactions (interaction with arginine side chains). The strong but reversible intermolecular interactions between the low-charged POMs and the cationic peptides established by ITC present an interesting finding. Previous ITC studies have focused on the binding of highly charged and therefore weakly chaotropic POMs (n < –6) where electrostatic interactions were implied as main driving force. Table 4 Binding constants (Kain M‒1) and thermodinamic data (in kcal mol‒1) measured for the complex formation between the stable and active POMs and oligoarginines in 10 mM Mes, pH 5.5.[a]HeptaArg Protamine PolyArg n Ka / ∆H T∆S n Ka / ∆H T∆S n Ka / ∆H T∆S 107107107[PVW11]4‒0.90 5.7 − − 3.1 25 − − 6.7 3.5 ± − − ± 23.4 12.8 ± 9 23.8 11.3 0.6 22.8 12.5 0.7 1.10 0.6 − − 3.2 1.7 − − 6.1 1.6 ± − − [SiMoW11]4‒± 23.7 14.4 ± 26.8 16.9 0.2 26.5 16.7 0.1 0.3 [a] Error data is 5 % for the n values and ± 0.5 kcal mol‒1for ∆H and T∆S (duplicate measurements.

[0096] U-tube transport experiments (see Methods, Fig.2c-d), confirmed that [PVW11]4‒and [SiMoW11]4‒act as “real” carriers that can seize and transport a hydrophilic cargo even through a hydrophobic bulk phase. In the U-tube experiments, the trans phase contained only buffer, and the cis phase is loaded with combinations of CF, carrier (POM) and cargo (heptaArg). Aliquots of the trans phase are taken and evaluated by fluorescence at different times. The trans phase is initially (t = 0) nonfluorescent, but the transport of CF from the cis phase through the chloroform barrier leads to a fluorescence signal for successful transport. Fig.4d shows that there is no fluorescence in the absence of POM, although an inefficient transport of CF can be sensed when the POM is mixed only with CF. In the ternary mixture, however, as expected, highly efficient transport is observed. This is consistent with the transport of a POM•heptaArg•CF complex through the organic phase. Accordingly, the superchaotropic properties of POMs cannot only be used forUW014P 23 biological membrane transport, but these properties should be transferable to artificial membranes and phase-transfer processes in general, as recently demonstrated for catalysis and dissolution. To mimic biological membranes, which contain anionic components, equivalent transport experiments were performed in anionic liposomes with the prototypal POM carriers, [PVW11]4‒and [SiMoW11]4‒. These experiments showed that, despite the apparently opposing negative charges, both POMs retained their carrier capabilities (Fig.14, Table 5), with only slightly reduced net efficiencies. This encouraged us to transfer the experiments to CHO-K1 cells, after we had confirmed a sufficient stability of the most promising POM carriers, [PVW11]4‒and [SiMoW11]4‒, in the cellular nutrient mixture (F-12 Ham's medium). Table 5 Membrane transport parameters of active POMs towards different analytes and different lipid composition. Lipid[a]POM Cargo[b]Ymax(%)[c]EYPC [PVW11]4‒heptaArg 84 49 5.3 (zwitterionic) heptaLys 62 38 4.3 [SiMoW11]4‒heptaArg 66 17 5.7 heptaLys 54 22 4.4 DMPE / DPPG / [PVW11]4‒heptaArg 45 50 2.8 CHOL [SiMoW11]4‒heptaArg 61 60 3.6 (anionic) [a] 13 µM phospholipids. [b] 10 µM cargo. [c] Maximal activity; ±2 % error (SD). [d] Effective concentration to reach 50 % activity of Ymax; 5 % error (SD). [e] Activation efficiency E = Ymax(pEC50 / f), were pEC50is the negative logarithm of the EC50, and f a scaling factor set to 20.6 to allocate E between 0 and 10; 10 % error.

[0097] POMs are known to cause cellular damage, and their antimicrobial as well as anticancer effects rest on those. Nevertheless, we tried to define a “therapeutic window” in which the cytotoxic properties would be sufficiently low to allow their carrier activity to be investigated (Fig.5b). Indeed, at concentrations of 10 µM and below, [PVW11]4‒and [SiMoW11]4‒, the CHO-K1 cells retained about the 80 % viability after 24 h according to MTT assay (Fig.5b), while the onset of carrier activity in the negatively charged model membranes fell in the same range (Fig.14). Accordingly, the uptake of 5(6)-fluorescein isothiocyanate-labelled octaarginine (FITCArg8) was evaluated by confocal fluorescence microscopy at 5 and 10 µM carrier concentrationUW014P 24 (Fig.5a and Fig.15). In the absence of carrier, the cells showed minor punctate fluorescence, in agreement with (undesirable) endosomal uptake and entrapment ofFITCArg8. In the presence of the POM carriers, strong diffuse fluorescence was observed in both the cytosol and in the nucleus (Fig.5a). Successful experiments were also conducted at 50 µM (Fig.16), when the incubation time was kept much shorter than the 24 h used in the MTT assay. Consequently, while the present cell- biological results are preliminary, and attempted in fixed cells, the transport activity of the POMs is transferable from model membranes to cells. In fact, while the carrier efficiency in the vesicles was comparable (Table 1), theFITCArg8cell staining results for both POM carriers visually exceeded those by the alternative amphiphilic and superchaotropic carriers, PyBu and B12Br122‒. Example 2

[0098] K4[PVW11O40] showed toxic effects in vitro on parental human embryonic kidney 293 (HEK293) cells, Chinese hamster ovary (CHOK1) cells and Cat.a derived (CAD) cells and HEK293 cells. The toxicology assays were conducted by using close-to-physiological conditions; this was done in the original cell type (CHOK1) and two additional cell types (HEK293, CAD cells) to explore whether the already observed toxicity may vary among cell types. All the planned experiments have been done in a concentration dependent manner (three technical replicates each). Methodology

[0099] Cells of interest were seeded at a density of 3.6× 104cells / 0.2 mL into poly- D-lysine coated 96-well plates 24 hours prior to the experiment. On the next day, the culture medium was replaced by test compounds diluted in standard Dulbecco’s Modified Eagle Medium (DMEM) / Ham’s F-12 Nutrient Mixture (F12) to achieve the final concentrations indicated. After 24 hour of incubation at 37 °C, the solutions were aspirated and replaced with AquaBluer™ reagent (MultiTarget Pharmaceuticals LLC, United States) according to manufacturer’s protocol. After four hours of incubation at 37 °C, fluorescence (Ex.544 nm, Em.595 nm) was measured using a Victor plate reader. Data were analyzed using Excel and GraphPad prism, with % cell viability calculated relative to the vehicle control conditions (0.1 % / 100 % DMSO), after making background corrections.UW014P 25 Conclusion

[0100] Polyoxometalates are known to cause cellular damage in a concentration- dependent manner, and their antimicrobial as well as anticancer effects rest on those. A “therapeutic window” was defined in which the cytotoxic properties would be sufficiently low to allow their carrier activity to be investigated. At POM concentrations of ≤ 10 μM the Chinese hamster ovary (CHO) cells retained about 80 % viability after 24 hours according to the MTT assay (MTT is (3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide)), while the onset of carrier activity fell in the same range. Extending the initial observations by application of the more sensitive AquaBluer™ Test encompassing three distinct cell types—Chinese hamster ovary (CHO) cells, human embryonic kidney 293 (HEK293) cells, and Cath.a-differentiated (CAD) cells

[0014] —the variations in the toxicity profile of [PVW11]4‒within different cellular contexts was meticulously examined. Intriguingly, CAD cells exhibited a remarkable resilience with negligible effects at concentrations of POMs up to 300 μM, implying divergent responses of cells to tested POM (Fig.17). Fig.17 shows the effect of [PVW11]4‒on the cell survival of the three different cell types, i.e., CHO, HEK293, and CAD. References [1] Zhang Z. et al., Biomaterials 2015, 65, 56-65 [2] Barba-Bon A. et al., Nature 2022, 603, 637- 642 [3] Bijelic A. et al., Angewandte Chemie International Edition 2018, 58(10), pp 2980- 2999 [4] Geisberger G. et al., Carbohydrate Polymers, 2013, 91(1), pp 58-67 [5] Téazéa, G. H. A. et al., in Inorg. Synth., Vol.27 (Ed.: A. P. Ginsberg), John Wiley & Sons,Inc., 1990, 85-96 [6] Weinstock, I. A. et al., Am. Chem. Soc.1999, 121, 4608-4617 [7] Weinstock, I. A. et al., J. Am. Chem. Soc.1999, 121, 4608-4617 [8] Salvi G. et al., Proteins: Struct., Funct., Bioinf.2005, 61, 492–499 [9] Assaf K. I. et al., Chem. Int. Ed.2018, 57, 13968–13981

[0010] Ly, H. G. T. et al., Dalton Trans.2013, 42, 10929-10938.

[0011] Manikumari, S. et al., Inorg. Chem.2002, 41, 6953-6955.

[0012] Rebek J. A. et al., J. Am. Chem. Soc.1987, 109, 2432–2434

[0013] Sakai N. et al., J. Am. Chem. Soc.2003, 125, 14348–14356 Qi, Y. et al., J. Neurosci.1997, 15, 1217-1225

Claims

UW014P 26 Claims 1. A delivery molecule comprising a carrier molecule and a cargo molecule, wherein said carrier molecule is a polyoxometalate moiety of the general formula [X1M1M211O40](n1)–, wherein X1denotes P, Si, Ge, or Al; M1denotes Mo, V, Ti, or Nb; M2denotes W, or Mo: n1denotes 2–6; or of general formula [X2M3M411O39(L)](n2)–, wherein X2denotes P, Si, or Ge; M3denotes Mo, V, Fe, Ga, In, Ni, Co, Cr, or Al;, M4denotes W, or Mo; L is a terminal ligand and denotes O2–or H2O coordinated to M3, n2denotes 3–5; or of general formula [X3(OH)6M56O18]3–, wherein X3denotes Al, Ga, Cr, or Fe; and M5denotes Mo, or W; and wherein the cargo molecule is a hydrophilic moiety, an inorganic molecule, or an organic molecule.

2. The delivery molecule of claim 1, wherein the carrier molecule and the cargo molecule are directly linked to each other without a linker moiety.

3. The delivery molecule of claim 1 or 2, wherein said polyoxometalate moiety is of general formula [X1M1M211O40](n1)–, wherein X1denotes P, Si, Ge, or Al; M1denotes Mo, V, Ti, or Nb; M2denotes W; and n1denotes 2–6.

4. The delivery molecule of claim 3, wherein X1denotes P, or Si; M1denotes Mo, or V; M2denotes W; and n1denotes 4.

5. The delivery molecule of claim 4, wherein said polyoxometalate is of formula [PVW11O40]4–, or [SiMoW11O40]4–.UW014P 27 6. The delivery molecule of any one of claims 1 to 5, wherein said polyoxometalate is stable at a pH up to 7.

5.

7. The delivery molecule of any one of claims 1 to 6, wherein said polyoxometalate moiety is bound to said cargo molecule through supramolecular interactions.

8. The delivery molecule of any one of claims 1 to 7, wherein said cargo molecule is a hydrophilic molecule.

9. The delivery molecule of claim 8, wherein said hydrophilic molecule is a hydrophilic peptide or hydrophilic protein.

10. The delivery molecule of any one of claims 1 to 9, wherein said cargo molecule is a pharmaceutically active compound.

11. The delivery molecule of claim 10, wherein said pharmaceutically active compound is a pharmaceutically active hydrophilic peptide or protein, a hydrophilic organic compound, or a hydrophilic inorganic compound.

12. The delivery molecule of any one of claims 1 to 11, wherein said carrier molecule is present in a range of about 0.01 to 100 µM, or of about 0.1 to 10 µM, or of about 0.5 to 5 µM.

13. The delivery molecule of any one of claims 1 to 12 for use in treating a disease curable by said pharmaceutically active compound.

14. The delivery molecule of any one of claims 1 to 12 for use according to claim 13, wherein the pharmaceutically active compound is delivered to or removed from a site within a subject.

15. The delivery molecule of any one of claims 1 to 12 for use according to claim 13 or 14, wherein said delivery molecule passes a barrier within a subject based on a superchaotropic effect.

16. The delivery molecule of any one of claims 1 to 12 for use according to any one of claims 13 to 15, wherein said barrier is a cell membrane.

17. An in vitro method for delivering a cargo molecule to a cell comprising the steps of (a) providing a cell, and (b) contacting said cell with a delivery molecule according to any one of claims 1 to 11.

18. A delivery molecule according to any one of claims 1 to 12 as a medicament.

19. A delivery molecule according to any one of claims 1 to 12 for use in the treatment of a disease curable by a pharmaceutically active compound.UW014P 28 20. A pharmaceutical composition comprising the delivery molecule according to any one of claims 1 to 12.

Images