WO2017199023A1

WO2017199023A1 - Polymersomes

Info

Publication number: WO2017199023A1
Application number: PCT/GB2017/051374
Authority: WO
Inventors: Giuseppe Battaglia; Loris Rizzello; Denis CECCHIN
Original assignee: UCL Business Ltd
Current assignee: UCL Business Ltd
Priority date: 2016-05-18
Filing date: 2017-05-17
Publication date: 2017-11-23
Anticipated expiration: 2018-11-18
Also published as: GB201608737D0

Abstract

The present invention relates to therapeutic polymersomes. The compositions of the invention comprise a high proportion of tubular polymersomes. Methods of producing such compositions and therapeutic applications of such compositions are also described.

Description

POLYMERSOMES

FIELD OF THE INVENTION

BACKGROUND OF THE INVENTION

The interactions of nanomaterials with living systems have recently attracted increasing attention, especially in the context of uncovering previously unforeseen cellular behaviours. Cells are intrinsically characterised by nanoscale events, in which proteins and molecules have evolved to act as the "fingers of nature", guiding physiological processes in the preservation and proliferation of life.

In this respect, nanoparticles may represent a new tool for tailoring, and possibly probing, cellular pathways, because such nanomaterials coexist in the same nano-universe as the molecular machineries of cells, where nanoscale interactions take place. Although to date studies have often been concerned with the bare understanding of general toxicity of nanomaterials (so-called "nanotoxicology"), little focus has so far been placed on their cellular molecular targets.

Polymersomes (vesicles formed from amphiphilic block copolymers) are the polymeric equivalent of liposomes. They are known to be much more robust and stable than their lipid counterparts due to their macromolecular nature. In addition, their macromolecular nature also allows a very effective tuning of the membrane thickness. Polymersomes that are sensitive to pH have previously been developed and shown to be capable of delivering certain types of molecules to the cell cytosol. Selective targeting of tissues of therapeutic significance is also possible.

A range of medical utilities have been proposed for polymersome technologies, including those where the polymersome incorporates an encapsulated active agent (the polymersome functioning as a delivery system for the active agent) and those where the polymersome itself constitutes an active agent, for example by way the substances formed when it degrades in vivo. Polymersome technology has the potential to enhance selectivity and efficacy in treating a broad range of pathological conditions. Nonetheless, improvements in the potency and selectivity of polymersome therapeutics would be desirable. For example, further polymersome compositions having anti-cancer activity would be desirable.

SUMMARY OF THE INVENTION

It has now been found that the surface geometry of polymersomes has a substantial impact on their interaction with biological materials, including on interactions that are of direct therapeutic significance, for example in cancer therapy. Shaped polymersomes have been developed that demonstrate surprising and beneficial therapeutic properties. A production method for obtaining such polymersomes has been found. The resulting compositions have surprisingly been found to be capable of suppressing the replication activity of tumor cells.

In particular, the present invention provides a method of producing a tubular polymersome composition, the method comprising the steps of: providing a mixed polymersome composition comprising a mixture of tubular polymersomes and non-tubular polymersomes; subjecting the mixed polymersome composition to density gradient centrifugation in a centrifuge; and isolating a tubular polymersome composition from the centrifuge.

The present invention also provides a tubular polymersome composition, wherein the percentage of tubular polymersomes in the total population of polymersomes is at least 50%, as well as a tubular polymersome composition obtainable by the method of the present invention.

Furthermore, the present invention provides a pharmaceutical composition comprising: the tubular polymersome composition of the present invention; and one or more pharmaceutically acceptable excipients or diluents.

The present invention still further provides a tubular polymersome composition of the present invention for use as a medicament.

The present invention also provides a tubular polymersome composition of the present invention for use in the treatment of cancer.

The present invention additionally provides a method of treating cancer, the method comprising administering a therapeutically effective amount of the tubular polymersome composition of the present invention to the subject. The present invention further provides use of a tubular polymersome composition of the present invention in the manufacture of a medicament for use in the treatment of cancer.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows TEM characterisations of spherical (A) and tubular (B) polymersomes, after a sucrose-based density gradient centrifugation, as explained in more detail in Example 1.

Figure 2 shows viability assays of FaDu, HeLa, and FIDF cells incubated with spherical polymersomes (A), and tubes (B), as explained in more detail in Example 1. 3 different polymersomes/tubes concentrations and 3 time points (24h, 48h, and 96h) were tested.

Control + = 2.5% DMSO.

Figure 3 shows FtPLC-based quantification of polymersomes uptake in FaDu, HeLa, and FIDF cells: (A) Comparison between the total mass of up taken spheres and tubes in the different cells (value of polymer present in each cell culture); and (B) Normalisation of polymer/cell, during time; as explained in more detail in Example 1.

Figure 4 shows confocal investigations of uptake of spherical polymersomes (A) and tubes (B) in FaDu cells, as explained in more detail in Example 1.

Figure 5 shows a screenshot of the Matlab based software for the quantification of NDI and MM, as explained in more detail in Example 1. The software is able to discriminate cells with one, two, or more nuclei, as well as bi-nucleated cells with a micronucleus.

Figure 6 shows: (A) NDI quantification for FaDu (left side), HeLa (centre), and HDF (right side), after incubation with spheres and tubes; and (B) Micronucleus assay (MM) for addressing the presence of DNA damage; as explained in more detail in Example 1. Control + = H202 20 uM. *: p<0.05, **: p<0.01, ***: p<0.005.

Figure 7 shows a Trypan blue-based proliferation assay for FaDu, HeLa, and HDF cells, incubated with spherical polymersomes (upward pointing triangles), and tubes (downward pointing triangles), as explained in more detail in Example 1. Control + = 2.5% DMSO.

Figure 8 shows real Time qPCR for quantifying the expression of 10 different genes involved in replication activity (p21, p53), oxidative stress (SOD1, CAT), detoxification metabolism (CYPlAl, CYPIBI), Unfolded Protein Response - UPR (ATF4, ATF6), and general shock (HSP27, HSP70), as explained in more detail in Example 1. White histograms: genes expression for cells treated with spherical polymersomes; black histograms: genes expression for cells treated with tubes.

Figure 9 shows a caspase 3/7 assay for analysing the activation of extrinsic apoptosis, as explained in more detail in Example 1.

Figure 10 shows the combinatorial effect of nanoparticles shape and anticancer drug, as discussed in more detail in Example 2. FaDu, HeLa, and FIDF cells were incubated with free DOXO, or with DOXO encapsulated in spheres and tubes, having the same final dose of anticancer drug. T-test was applied with *p<0.05 or **p<0.01.

DETAILED DESCRIPTION OF THE INVENTION

Polymersomes

Polymersomes are synthetic vesicles formed from amphiphilic block copolymers. Over the last fifteen years they have attracted significant research attention as versatile carriers because of their colloidal stability, tuneable membrane properties and ability in encapsulating or integrating other molecules (for one representative review article, see J Control Release 2012 161(2) 473-83, the contents of which are herein incorporated by reference in their entirety).

It is important to emphasise that polymersomes, as a concept, are known in the art. Well established techniques also exist for producing polymersome compositions. However, these known techniques in the art result in the formation of a combination of differently sized and shaped assemblies. Such a composition is sometimes referred to in this disclosure as a "mixed polymersome composition".

Some illustrative and exemplary disclosure is now provided concerning the chemical composition of polymersomes that can suitably be used in the present invention, for example polymersomes that can suitably form a "mixed polymersome composition" for use in the method of the invention and polymersomes that can thus suitably give rise to the "tubular polymersome composition" of the present invention. However, for the avoidance of doubt, this disclosure does not imply that the products and methods of the invention can only be put into practice using polymersomes having the specifically exemplified chemical compositions. The polymersome used in the present invention is typically a self -assembled structure. The polymersome comprises an amphiphilic block copolymer. The amphiphilic block copolymer comprises a hydrophilic block and a hydrophobic block. Such polymersomes are able to mimic biological phospholipids. Molecular weights of these polymers are at least 5 times higher than naturally-occurring phospholipid-based surfactants such that they can assemble into more entangled membranes (J. Am. Chem. Soc. 2005, 127, 8757, the contents of which are herein incorporated by reference in their entirety), providing a final structure with improved mechanical properties and colloidal stability. Furthermore, the flexible nature of the copolymer synthesis allows the application of different compositions and functionalities over a wide range of molecular weights and consequently of membrane thicknesses. Thus the use of these block copolymers as delivery vehicles offers significant advantages.

Polymersomes typically comprise a bilayered membrane. The bilayer is generally formed from two layers of amphiphilic molecules, which align to form an enclosed core with hydrophilic head groups facing the core and the exterior of the vesicle, and hydrophilic tail groups forming the interior of the membrane.

A typical (largest) diameter of a polymersome is in the range 50 to 50,000 nm (for instance 50 to 5000 nm). More typically, the diameter is in the range 50 to 2000 nm. Polymersomes having a diameter in this range are normally termed "nanopolymersomes" or "nanovesicles". The thickness of the bilayer is generally between 2 to 50 nm, more typically between 5 and 20 nm. These dimensions can routinely be measured, for example by using Transmission Electron Microscopy (TEM) and/or and Small Angle X-ray Scattering (SAXS) (see, for example, J. Am. Chem. Soc. 127 8757 2005, the contents of which are herein incorporated by reference in their entirety).

In aqueous solution, normally an equilibrium exists between different types of structures, for instance between polymersomes and micelles. It is preferred that at least 80%, more preferably at least 90% or 95% by weight and most preferably all of the structures in solution are present as polymersomes. This can be achieved using the methods outlined herein.

The polymersome may be capable of dissociating and releasing any contents after it has been internalised within a cell. Dissociation may be promoted by a variety of mechanisms, but is typically promoted by pH sensitivity of the block copolymer. It is thus preferred that the hydrophilic or the hydrophobic block of the amphiphilic copolymer, preferably the hydrophobic block, has a pendant group with a pKa in the range 3.0 to 6.9. The process of endocytosis induces a reduction in the local pH experienced by the polymersome from around pH 7.4 to around pH 5-6. This pH drop is sufficient to trigger disintegration of the polymersome and release of any internalised content.

By pKa, is meant the pH where half of the pendant (side) groups are ionised. pKa can be determined by a variety of methods including pH titration followed by potentiometric titration, UV spectroscopy and Dynamic Light Scattering (DLS). An appropriate method should be selected to measure the pKa according to the copolymer which is being analysed and its solubility in the test media.

DLS is a particularly preferred method for measuring pKa. As indicated in J. Am. Chem. Soc 2005 127 17982-17983, the contents of which are herein incorporated by reference in their entirety, the DLS signal from a copolymer, such as PMPC25-&-PDPA20 copolymer, in water varies with pH. At a certain pH the signal rapidly increases as the copolymer undergoes a transition from being molecularly deassociated to associated. The pKa is taken as the pH of the mid-point of this rapid increase. These experiments are described further in

Biomacromolecules 2006, 7, 817-828, the contents of which are herein incorporated by reference in their entirety. In this reference, the experiments are performed on micelles of PMPC-&-PDPA block copolymer, but the techniques may also be applied when the phase transition involves polymersome formation.

The pKa of a group in a polymer is determined on the basis of a polymer system (and not assumed to be the same as the pKas of similar moieties in non-polymeric systems).

It is preferred that the hydrophobic block of the polymersome comprises pendant cationisable moieties as pendant groups. Cationisable moieties are, for instance, primary, secondary or tertiary amines, capable of being protonated at pHs below a value in the range 3 to 6.9.

Alternatively the group may be a phosphine.

Preferably, the pKa of the pendant groups is in the range 4.0 to 6.9, more preferably 5.5 to 6.9. The polymersomes are correspondingly capable of disassociating in such pH ranges.

Preferably, the hydrophobic block of the polymersome has a degree of polymerisation of at least 50, more preferably at least 70. Preferably, the degree of polymerisation of the hydrophobic block is no more than 250, even more preferably, no more than 200. Typically, the degree of polymerisation of the hydrophilic block is at least 15, more preferably at least 20. It is preferred that the ratio of the degree of polymerisation of the hydrophilic to hydrophobic block is in the range 1 :2.5 to 1 :8. All of these limitations promote polymersome, rather than micelle formation.

The hydrophilic block may be based on condensation polymers, such as polyesters, polyamides, polyanhydrides, polyurethanes, polyethers (including polyalkylene glycols, especially PEG), polyimines, polypeptides, polypeptoids, polyureas, polyacetals and polysaccharides, but preferably the hydrophilic block is based on a radical polymerised addition polymer of ethylenically unsaturated monomers. The hydrophilic block may have zwitterionic pendant groups, in which case the zwitterionic pendant groups may be present in the monomers and remain unchanged in the polymerisation process. It is alternatively possible to derivatise a functional pendant group of a monomer to render it zwitterionic after polymerisation.

In one embodiment, the hydrophilic block is formed from ethylenically-unsaturated zwitterionic monomers. Non-limiting suitable ethylenically unsaturated zwitterionic monomers have the general formula (I)

YBX (I), in which:

Y is an ethylenically unsaturated group selected from H2C=CR-CO-A-,

H₂C=CR-C₆H₄-A¹-, H₂C=CR-CH₂-A²-, R²0-CO-CR=CR-CO-0-, RCH=CH-CO-0-, RCH=C(COOR²)CH₂-CO-0-,

A is -O- or NR¹;

A¹ is selected from a bond, (CH₂)LA" ² and (CH₂)LS0₃- in which L is 1 to 12; A² is selected from a bond, -0-, -0-CO-, -CO-O, -CO- R¹-, - R^CO-, -O-CO- R¹- and - R^!-CO-O-;

R is hydrogen or C 1-4 alkyl;

R¹ is hydrogen, C_1-4 alkyl or BX;

R² is hydrogen or C 1-4 alkyl;

B is a bond, or a straight or branched alkanediyl, alkylene oxaalkylene, or alkylene

(oligooxalkylene) group, optionally containing one or more fluorine substituents; and

X is a zwitterionic group.

Preferably X is an ammonium, phosphonium, or sulphonium phosphate or phosphonate ester zwitterionic group, more preferably a group of the general formula (II)

in which the moieties A³ and A⁴, which are the same or different, are -0-, -S-, - H- or a valence bond, preferably -0-, and W⁺ is a group comprising an ammonium, phosphonium or sulphonium cationic group and a group linking the anionic and cationic moieties which is preferably a Ci-12-alkanediyl group.

Preferably W⁺ is a group of formula -W^N⁺R^, -W^P⁺R^, -W^S⁺R⁴! or -W^He in which:

W¹ is alkanediyl of 1 or more, preferably 2-6 carbon atoms optionally containing one or more ethylenically unsaturated double or triple bonds, disubstituted-aryl (arylene), alkylene arylene, arylene alkylene, or alkylene aryl alkylene, cycloalkanediyl, alkylene cycloalkyl, cycloalkyl alkylene or alkylene cycloalkyl alkylene, which group W¹ optionally contains one or more fluorine substituents and/or one or more functional groups; the groups R³ are the same or different and each is hydrogen or alkyl of 1 to 4 carbon atoms, preferably methyl, or aryl, such as phenyl, or two of the groups R³ together with the nitrogen atom to which they are attached form an aliphatic heterocyclic ring containing from 5 to 7 atoms, or two or more of the groups R³ together with the nitrogen atom to which they are attached form a heteroaromatic ring having 5 to 7 atoms, either of which rings may be fused with another saturated or unsaturated ring to form a fused ring structure containing from 5 to 7 atoms in each ring, and optionally one or more of the groups R³ is substituted by a hydrophilic functional group; the groups R⁴ are the same or different and each is R³ or a group OR³,

Het is an aromatic nitrogen-, phosphorus- or sulphur-, preferably nitrogen-, containing ring, for example pyridine.

Monomers in which X is of the general formula in which W⁺ is W¹N⁺R³ ₃ may be made as described in WO-A-9301221, the contents of which are herein incorporated by reference in their entirety. Phosphonium and sulphonium analogues are described in WO-A-9520407 and WO-A-9416749, the contents of both of which are herein incorporated by reference in their entirety.

The group of the formula II has a preferred general formula (III)

where the groups R⁵ are the same or different and each is hydrogen or C_1-4 alkyl, and m is from 1 to 4. The groups R⁵ are preferably the same, for example they are preferably all methyl.

In phosphobetaine based groups, X may have the general formula (IV)

in which: A⁵ is a bond, -0-, -S- or -NH- (preferably -0-);

R⁶ is a bond or alkanediyl, -C(0)-alkanediyl- or -C(0) H-alkanediyl- (wherein R⁶ is preferably alkanediyl; and wherein alkanediyl is preferably C_1-6 alkanediyl);

W² is SR⁷, PR⁷2 or NR⁷ ₂, wherein the or each group R⁷ is hydrogen or alkyl of 1 to 4 carbon atoms or the two groups R⁷ together with the heteroatom to which they are attached form a heterocyclic ring of 5 to 7 atoms;

R⁸ is alkanediyl of 1 to 20, preferably 1 to 10, more preferably 1 to 6 carbon atoms; A⁶ is a bond, NH, S or O, preferably O; and

R⁹ is a hydroxyl, C_1-12 alkyl, C_1-12 alkoxy, C7-18 aralkyl, C7-18 aralkoxy, C₆-i8 aryl or C₆-i8 aryloxy group.

Monomers comprising a group of the general formula IV may be made by methods as described in JP-B-03-031718, the content of which is herein incorporated by reference in its entirety, in which an amino substituted monomer is reacted with a phospholane.

In compounds comprising a group of the general formula IV, it is preferred that: A⁵ is a bond; R⁶ is a C2-6 alkanediyl; W² is NR⁷ ₂: each R⁷ is CM alkyl; R⁸ is C_2-6 alkanediyl; A⁶ is O; and R⁹ is Ci-4 alkoxy.

Alternatively, X may be a zwitterion in which the anion comprises a sulphate, sulphonate or carboxylate group.

One example of such a group is a sulphobetaine group, of the general formula (V)

I θ Θ

N {CH^SQs (V)

where the groups R¹⁰ are the same or different and each is hydrogen or C_1-4 alkyl and s is from 2 to 4. Preferably the groups R¹⁰ are the same. It is also preferable that at least one of the groups R¹⁰ is methyl, and more preferable that the groups R¹⁰ are both methyl. Preferably s is 2 or 3, more preferably 3. Another example of a zwitterionic group having a carboxylate group is an amino acid moiety in which the alpha carbon atom (to which an amine group and the carboxylic acid group are attached) is joined through a linker group to the backbone of the biocompatible polymer. Such groups may, for example, be represented by the general formula (VI)

in which A⁷ is a bond, -0-, -S- or - H- (preferably -0-); R¹¹ is a bond or alkanediyl,

-C(0)alkanediyl- or -C(0) Halkanediyl- (wherein alkanediyl is preferably C_1-6 alkanediyl; wherein R¹¹ is preferably alkanediyl); and the groups R¹² are the same or different and each is hydrogen or alkyl of 1 to 4 carbon atoms, preferably methyl, or two or three of the groups R¹², together with the nitrogen to which they are attached, form a heterocyclic ring of from 5 to 7 atoms, or the three group R¹² together with the nitrogen atom to which they are attached form a fused ring heterocyclic structure containing from 5 to 7 atoms in each ring.

Another example of a zwitterion having a carboxylate group is a carboxy betaine

-N⁺(R¹³)2(CH2)_rCOO^" in which the R¹³ groups are the same or different and each is hydrogen or Ri-4 alkyl and r is 2 to 6, preferably 2 or 3.

In the zwitterionic monomer of the general formula (I) it is preferred that the ethylenic unsaturated group Y is H2C=CR-CO-A-. Such acrylic moieties are preferably methacrylic, that is in which R is methyl, or acrylic, in which R is hydrogen. Whilst the compounds may be (meth)acrylamido compounds (in which A is R¹), in which case R¹ is preferably hydrogen, or less preferably, methyl, most preferably the compounds are esters, that is in which A is O.

In monomers of the general formula (I), especially where Y is the preferred (alk)acrylic group, B is most preferably an alkanediyl group. Whilst some of the hydrogen atoms of such group may be substituted by fluorine atoms, preferably B is an unsubstituted alkanediyl group, most preferably a straight chain group having 2 to 6 carbon atoms.

A particularly preferred zwitterionic monomer is 2-methacryloyloxyethyl-phosphorylcholine (MPC). Mixtures of zwitterionic monomers each having the above general formula may be used, as can mixtures of other hydrophilic monomers described herein. In another preferred embodiment, the hydrophilic block is formed from ethylenically- unsaturated monomers that comprise a polyalkylene glycol side chain (e.g., a PEG side chain).

For instance, the ethylenically-unsaturated monomers may have the general formula (I)

YBX (I), in which:

Y is an ethylenically unsaturated group selected from H2C=CR-CO-A-,

A is -O- or R¹;

A¹ is selected from a bond, (CH₂)_LA² and (CH₂)_LS03^" in which L is 1 to 12;

A² is selected from a bond, -0-, -0-CO-, -CO-O, -CO- R¹-, - R^CO-, -O-CO- R¹- and R^!-CO-O-;

R is hydrogen or C 1-4 alkyl;

R¹ is hydrogen, C_1-4 alkyl or BX;

R² is hydrogen or C 1-4 alkyl;

B is a bond, or a straight or branched alkanediyl, alkylene oxaalkylene, or alkylene (oligooxalkylene) group, optionally containing one or more fluorine substituents; and

X is a polyalkylene glycol side chain. For example, such monomers may comprise an ethylenic unsaturated group H2C=CR-CO- that is attached to the polyalkylene glycol side chain. The polyalkylene glycol side chain may have the formula -[0(CH2)_n]_POR24 in which n is from 1 to 6, p is from 1 to 100 and R24 is hydrogen or C_1-6 alkyl. Preferably n is 2 (i.e., the side chain is a polyethylene glycol side chain). Preferably p is from 1 to 50, more preferably from 5 to 20. Preferably R24 is hydrogen or methyl, most preferably hydrogen. It will be understood that individual molecules within such a monomer compound may have a distribution of molecular weights owing to a distribution in the extent of polymerisation in the side chain (i.e., a distribution in the value of p). Typical number average molecular weights of the monomers may be in the range 25 to 1000, preferably 50 to 800. For example, the number molecular average molecular weight of the monomers may be from 200 to 800. A particularly preferred hydrophilic block of this nature is formed from oligo(ethylene glycol) methacrylate

(OEGMA) monomers. The OEGMA monomers may have the formula H2C=CR-CO- [0(CH₂)2]pOCH₃ where p is from 2 to 20.

In one particularly preferred aspect of the present disclosure, the hydrophilic block comprises a phosphorylcholine polymer. A phosphorylcholine polymer is a polymer that comprises one or more phosphorylcholine groups. Thus, for example, the phosphorylcholine polymer is optionally formed from monomers of the above general formula (I), in which X is a group of the above general formula (III) (m=2, R⁵=CH₃).

In a preferred embodiment, the hydrophilic block comprises a phosphorylcholine polymer and the hydrophobic block comprises a pendant group with a pKa in the range 3.0 to 6.9.

The hydrophobic block may be formed of polymers such as polyethers (including

polyalkylene glycols), polyesters, polyamides, polyanhydrides, polyurethanes, poiyimines, polypeptides, polypeptoids, polyureas, polyacetals, or polysiloxanes. One example of a suitable hydrophobic block is polyalkylene oxide, usually polypropylene oxide, that is the same type of block as has been used in the well-studied Pluronic/Poloxamer based systems. One type of highly hydrophobic block is poly(dimethylsiloxane). In one preferred

embodiment the type of polymer forming the hydrophobic block is the same as that forming the hydrophilic block. Preferably the polymer is formed by radical polymerisation of ethylenically unsaturated monomers. Suitable monomers from which the hydrophobic block may be formed have the general formula (VII)

Y^Q (VII), in which Y¹ is selected from H₂C=CR¹⁴-CO-A⁸-, H₂C=CR¹⁴-C₆H₄-A⁹-,

H₂C=CR¹⁴-CH₂-A¹⁰-, R¹⁶0-CO-CR¹⁴=CR¹⁴-CO-0-, R¹⁴CH=CH-CO-0-,

R¹⁴CH=C(COOR¹⁶)CH₂-CO-0-,

A⁸ is -O- or - R¹⁵-;

A⁹ is selected from a bond, (CH₂)_qA¹⁰ and (CH₂)_qS03^" in which q is 1 to 12;

A¹⁰ is selected from a bond, -0-, -0-CO-, -CO-0-, -CO- R¹⁵-, - R¹⁵-CO-,

-O-CO- R¹⁵-, - R¹⁵-CO-0-;

R¹⁴ is hydrogen or C 1-4 alkyl;

R¹⁵ is hydrogen, C_1-4 alkyl or ¹Q;

R¹⁶ is hydrogen or C 1-4 alkyl;

B¹ is a bond, or a straight or branched alkanediyl, alkylene oxaalkylene, or alkylene

Q is a cationic or cationisable group of the formula - R¹⁷ _P, -PR¹⁷ _P and SR¹⁷ _r, in which p is 2 or 3, r is 1 or 2, the groups R¹⁷ are the same or different and each is selected from the group consisting of hydrogen, C_1-24 alkyl and aryl, or two of the groups R¹⁷ together with the heteroatom to which they are attached from a 5 to 7 membered heterocyclic ring or three R¹⁷ groups together with the heteroatom to which they are attached form a 5 to 7 membered heteroaromatic ring, either of which rings may be fused to another 5 to 7 membered saturated or unsaturated ring, and any of the R groups may be substituted by amino or hydroxyl groups or halogen atoms; wherein if p is 3, at least one of the groups R¹⁷ is hydrogen.

Preferably Y¹ is H₂C=CR¹⁴-CO-A⁸- where R¹⁴ is H or methyl and A⁸ is O or H. Preferred groups B¹ are alkanediyl, usually with linear alkyl chains and preferably having 2 to 12 carbon atoms, such as 2 or 3 carbon atoms.

Preferably Q is R¹⁷ ₂ where R¹⁷ is Ci-12-alkyl. Preferably both R¹⁷s are the same. Particularly useful results have been achieved where the groups R¹⁷ are C_1-4 alkyl, especially ethyl, methyl or isopropyl.

Either or both the hydrophobic and hydrophilic blocks may include comonomers, for instance to provide functionality, control over hydrophobicity, control over pH sensitivity, pKa or pKb as the case may be, control over temperature sensitivity or as general diluents. For instance comonomers providing functionality may be useful to provide conjugation of pendant groups following polymerisation and/or polymersome formation, to targeting moieties, or to provide for conjugation between the biologically active molecule and the polymer. Alternatively, functional groups may allow for crosslinking of the polymer following polymersome formation, to confer increased stability on the polymersome structure. Examples of suitable comonomers are compounds of the general formula (VIII)

in which

R is selected from hydrogen, halogen, C_1-4 alkyl and groups COOR in which R is hydrogen or C 1-4 alkyl;

R¹⁹ is selected from hydrogen, halogen and C_1-4 alkyl;

R²⁰ is selected from hydrogen, halogen, C_1-4 alkyl and groups COOR²² provided that R¹⁸ and R²⁰ are not both COOR²²; and

R²¹ is a Ci-10 alkyl, a C_1-20 alkoxycarbonyl, a mono-or di-(C_1-10 alkyl)amino carbonyl, a C₆-₂o aryl (including alkaryl) a C7-20 aralkyl, a C₆-₂o aryloxycarbonyl, a Ci-2o-aralkyloxycarbonyl, a C₆-2o arylamino carbonyl, a C7-20 aralkyl-amino, a hydroxyl or a C2-10 acyloxy group, any of which may have one or more substituents selected from halogen atoms, alkoxy, oligo-alkoxy, aryloxy, acyloxy, acylamino, amine (including mono and di- alkyl amino and thalkylammonium in which the alkyl groups may be substituted), carboxyl, sulphonyl, phosphoryl, phosphino, (including mono- and di-alkyl phosphine and tri- alkylphosphonium), zwitterionic, hydroxyl groups, vinyloxycarbonyl and other vinylic or allylic substituents, and reactive silyl or silyloxy groups, such as trialkoxysilyl groups; or R²¹ and R²⁰ or R²¹ and R¹⁹ may together form -CO R²³CO in which R²³ is a Ci-20 alkyl group.

It is preferred for at least two of the groups R¹⁸, R¹⁹, R²⁰ and R²¹ to be halogen or, more preferably, hydrogen atoms. Preferably R¹⁸ and R¹⁹ are both hydrogen atoms. It is particularly preferred that compound of general formula VIII is a styrene or acrylic compound. In styrene compounds R²¹ represents an aryl group, especially a substituted aryl group in which the substituent is an amino alkyl group, a carboxylate or a sulphonate group. Where the comonomer is an acrylic type compound, R²¹ is an alkoxycarbonyl, an alkyl amino carbonyl, or an aryloxy carbonyl group. Most preferably in such compounds R²¹ is a C1-20- alkoxy carbonyl group, optionally having a hydroxy substituent. Acrylic compounds are generally methacrylic in which case R²⁰ is methyl.

Preferably the comonomer is a non-ionic comonomer, such as a C_1-24 alkyl(alk)-acrylate or - acrylamide, mono- or di- hydroxy-Ci-6-alkyl(alk)-acrylate, or acrylamide, oligo(C2-3 alkoxy) C2-i8-alkyl (alk)-acrylate, or -acrylamide, styrene, vinylacetate or N-vinyllactam.

For optimum nanovesicle formation, the block copolymers should have controlled molecular weights. It is preferable for each of the blocks to have molecular weight controlled within a narrow band, that is, to have a narrow polydispersity. The polydispersity of molecular weight should, for instance, be preferably less than 2.0, more preferably less than 1.5, for instance in the range 1.1 to 1.4. Of course, in the preferred embodiment wherein one of the blocks has a pKa in the range 3.0 to 6.9, the blocks should be selected so that they have the requisite pKa value.

In one embodiment of this invention, the monomer from which the hydrophobic block is formed is 2-(diisopropylamino)ethyl methacrylate (DP A) or 2-(diethylamino)ethyl methacrylate (DEA). In another embodiment, the hydrophilic block is PMPC or poly(oligo (ethylene glycol) methacrylate) (POEGMA). Preferably, the copolymer is a PMPC-6-PDPA block copolymer or a POEGMA-PDPA block copolymer.

Preferably, the block copolymer has general formula PMPC_m-£-PDPA_n or POEGMA_m- PDPAn, wherein m is in the range from 2 to 500, or from 15 to 30 (for instance 25), and n is from 6 to 2000 or from 70 to 180, preferably from 100 to 160, more preferably from 120 to 160. For instance, the block copolymer may have the general formula POEGMA_m-PDPA_n where m is from 15 to 30 and n is from 100 to 160 (i.e. a block copolymer comprising a block derived from m OEGMA monomers joined to a block derived from n DPA monomers).

Typically, the hydrophobic block is not formed from 2-(dimethyl)ethyl methacrylate (DMA) monomers.

The block copolymer may be a simple A-B block copolymer, or may be an A-B-A or B-A-B block linear triblock copolymer or a (A)₂B or A(B)₂ star copolymers (where A is the hydrophilic block and B is the hydrophobic block). It may also be an A-B-C, A-C-B or B-A- C block linear triblock copolymers or a ABC star copolymers (blocks linked together by the same end), where C is a different type of block. C blocks may, for instance, comprise functional, e.g. cross-linking or ionic groups, to allow for reactions of the copolymer, for instance in the novel compositions. Crosslinking reactions especially of A-C-B type copolymers, may confer useful stability on polymersomes. Cross-linking may be covalent, or sometimes, electrostatic in nature. Cross-linking may involve addition of a separate reagent to link functional groups, such as using a difunctional alkylating agent to link two amino groups. The block copolymer may alternatively be a star type molecule with hydrophilic or hydrophobic core, or may be a comb polymer having a hydrophilic backbone (block) and hydrophobic pendant blocks or vice versa. Such polymers may be formed for instance by the random copolymerisation of monounsaturated macromers and monomers.

Further details of a suitable process for polymerising the monomers are to be found in WO 03/074090, the contents of which are herein incorporated by reference in their entirety.

Exemplary methods that can be used for polymerising the monomers are atom-transfer radical polymerisation (ATRP) (see, e.g., an exemplary method described in Journal of the American Chemical Society 127, 17982-17983), living radical polymerisation process, functional NCA (N-carboxyanhydride) polymerisation with efficient postpolymerization modification and ring opening polymerisation (ROP). Living radical polymerisation has been found to provide polymers of monomers having a polydispersity (of molecular weight) of less than 1.5, as judged by gel permeation chromatography. Polydispersities in the range 1.2 to 1.4 for the or each block are preferred. The polymersomes may be loaded using a pH change system, electroporation or film hydration. In a pH change system process, polymer is dispersed in aqueous liquid in ionised form, in which it solubilises at relatively high concentrations without forming polymersomes. Subsequently the pH is changed such that some or all of the ionised groups become deprotonated so that they are in non-ionic form. At the second pH, the hydrophobicity of the block increases and polymersomes are formed spontaneously.

A method of forming polymersomes with an encapsulated material (e.g. an encapsulated drug) in the core may involve the following steps: (i) dispersing the amphiphilic copolymer in an aqueous medium; (ii) acidifying the composition formed in step (i); (iii) adding the material to be encapsulated to the acidified composition; and (iv) raising the pH to around neutral to encapsulate the material.

This method preferably comprises a preliminary step wherein the amphiphilic copolymer is dispersed in an organic solvent in a reaction vessel and the solvent is then evaporated to form a film on the inside of the reaction vessel.

Step (ii), of acidifying the composition, typically reduces the pH to a value below the pKa of the pendant group.

Another method of forming polymersomes with an encapsulated material in the core may involve the following steps: (i) dispersing the amphiphilic copolymer, and when needed the material to be encapsulated, in an organic solvent in a reaction vessel; (ii) evaporating the solvent to form a film on the inside of the reaction vessel; and (iii) re-hydrating the film with an aqueous solution, optionally comprising a solubilised material to be encapsulated.

In more detail, polymersomes are typically prepared by dissolving copolymer in an organic solvent, such as a 2: 1 chloroform:methanol mix in a glass container. If a hydrophobic or amphiphilic material is to be encapsulated, it can be added with the copolymer. Solvent can be evaporated under vacuum leaving a copolymeric film deposited on the walls of the container. The film is then re-hydrated with an aqueous solution, for instance using phosphate buffer saline. If a hydrophilic material is to be encapsulated, it can be included in the aqueous solution. The pH of the resultant suspension is decreased to a pH of around 2, to solubilise the film, and then increased slowly to a pH or around 6. The polymer hydration at neutral pH allows the encapsulation of the material. The dispersion may then be sonicated and extruded, for instance using a bench top extruder. UV spectroscopy and HPLC chromatography may be used to calculate the encapsulation efficiency, using techniques well known in the art. An alternative method for forming polymersomes with an encapsulated material may involve simple electroporation of the material and polymer vesicles in water. For instance the drug may be contacted in solid form with an aqueous dispersion of polymer vesicles and an electric field applied to allow the formation of pores on the polymersomes membrane. The solubilised material molecules may then enter the polymersome vesicles though the pores. This is followed by membrane self healing process with the consecutive entrapment of the material molecules inside the polymersomes.

Alternatively, material dissolved in organic solvent may be emulsified into an aqueous dispersion of polymer vesicles, whereby solvent and the material become incorporated into the core of the vesicles, followed by evaporation of solvent from the system.

The polymersomes used in the invention may be formed from two or more different block copolymers. In this embodiment, in the method of forming polymersomes, a mixture of the two or more block copolymers is used.

For example, 0.01% to 10% (w/w) of material to be encapsulated is mixed with copolymer in the methods described above.

Methods such as those described above typically lead to the formation of a composition that comprises a mixture of polymersomes of different shapes. Such a mixed polymersome composition, for example, typically comprises a mixture of tubular and non-tubular polymersomes. Non-tubular polymersomes include substantially spherical polymersomes.

The mixed polymersome composition may, for example, be one in which the percentage of tubular polymersomes in the total population of polymersomes is less than 50%, preferably less than 40% and more preferably still less than 30%.

Tubular polymersomes and tubular polymersome composition

A tubular polymersome is a polymersome that comprises at least a tubular portion. The term

"tubular portion" can be used interchangeably herein with "elongated portion". The tubular polymersome may consist substantially of the tubular portion or alternatively the tubular portion may form only part of the tubular polymersome, with other parts of the tubular polymersome being non-tubular in shape. A skilled person would readily recognize a polymersome having a tubular portion and would not have any difficulty in distinguishing it from a polymersome lacking a tubular portion.

The term tubular polymersome includes a toroidal polymersome. A toroidal polymersome is a tubular polymersome having two ends that are connected to each other (as in a torus or "donut").

A tubular polymersome can be unbranched or branched. A branched tubular polymersome is a polymersome that comprises one or more branching points with one or more arms extending from the or each branching point.

A branched tubular polymersome is clearly non-spherical in shape because a sphere does not comprise branching points or arms.

An unbranched tubular polymersome is also non-spherical in shape in view of its

tubular/elongated portion.

Methods of quantitatively determining whether a polymersome is tubular include determining (a) its aspect ratio (either in 3D or more typically from a 2D projection of the polymersome in, for example, a TEM image) and (b) its sphericity or circularity (most typically the circularity of a 2D projection of the polymersome in, for example, a TEM image).

Preferably a tubular polymersome has an aspect ratio of less than 1, typically 0.95 or less, preferably 0.9 or less, and more preferably 0.8 or less (wherein aspect ratio is the minimum Feret diameter divided by the maximum Feret diameter of the polymersome). A sphere has an aspect ratio of 1.

Polymersomes are of course three-dimensional structures. Maximum and minimum Feret diameters of such a structure are defined as the maximum and minimum distances, respectively, between two parallel planes restricting the structure perpendicular to that direction. However, it will be appreciated that visualization techniques, such as TEM, typically project a polymersome in 2D rather than directly showing its 3D shape. Thus, the aspect ratio as defined herein is typically the aspect ratio of such a 2D projection of the polymersome. Furthermore, the maximum and minimum Feret diameters are the maximum and minimum distances, respectively, between two parallel tangential lines restricting the 2D projection of the polymersome perpendicular to that direction. Preferably an unbranched tubular polymersome has a sphericity of 0.9 or less, more preferably 0.8 or less (wherein sphericity is the ratio of the surface area of a sphere with the same volume as the unbranched tubular polymersome to the surface area of the unbranched tubular polymersome).

As already indicated, common visualization techniques, such as TEM, typically project a polymersome in 2D rather than directly showing its 3D shape. Thus, preferably an unbranched tubular polymersome is one whose 2D projection, for example in a TEM image, has a circularity of 0.75 or less, more preferably 0.7 or less. Circularity, fare, is defined herein as fcirc — p₂ in which A is area and P is perimeter.

A 2D projection of a tubular polymersome may underestimate the extent to which it is non- spherical/non-circular in view of its projection onto the 2D surface. For example, the projection of a perfect cylinder end-to-end onto a 2D surface is a circle. For the avoidance of doubt, however, it should be noted that the characterisations provided herein take this factor into account. For example, where the percentage of tubular polymersomes in the total population of polymersomes is specified to be at least a certain percentage (e.g. at least 50%, 70%, 80%) or 90%) account has been taken that some polymersomes that are in fact tubular may nonetheless be ascribed to be non-tubular owing to the measurement method employed.

Those skilled in the art would easily be able to distinguish between tubular polymersomes and non-tubular polymersomes. For example, a polymersome composition can readily be visualized using TEM.

Figure 1 shows illustrative TEM images of two different polymersome compositions (A) and (B) (two separate images are provided for each composition). In the images (A), the composition clearly mainly comprises non-tubular polymersomes (in particular, spherical polymersomes). By contrast, in the images (B), the composition clearly mainly comprises tubular polymersomes, specifically in the form of a mixture of unbranched tubular polymersomes and branched tubular polymersomes. In the method of the present invention, the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is greater than the percentage of tubular polymersomes in the total population of polymersomes in the mixed polymersome composition. In the tubular polymersome compositions of the present invention, the percentage of tubular polymersomes in the total population of polymersomes is typically at least 50% (e.g. at least 70%, 80% or 90%). These percentages can readily be determined by a skilled person.

For example, the percentage can be determined using TEM. Preferably one or more TEM images of the composition are obtained, such that the total number of polymersomes visible in the one or more images is at least 20 (more preferably at least 30, more preferably still at least 50).

The number of tubular polymersomes (i.e., the total number of branched tubular

polymersomes and unbranched tubular polymersomes) can in one embodiment be determined by visual inspection of the one or more TEM images. The percentage of tubular

polymersomes in the total population of polymersomes can thus easily be determined.

Alternatively the number of tubular polymersomes can be determined by calculating the aspect ratio of each polymersome in the one of more images. The percentage of tubular polymersomes in the total population of polymersomes can again then easily be determined.

In a further alternative, the number of tubular polymersomes can be determined by determining, first, the number of branched polymersomes and second, by determining the number of polymersomes amongst those remaining that have a circularity of 0.75 or less, more preferably 0.7 or less (i.e. those polymersomes that constitute unbranched tubular polymersomes). The percentage of tubular polymersomes in the total population of polymersomes can again then easily be determined.

Optionally the total population of polymersomes is considered to be the total number of polymersomes that are one of: (a) branched polymersomes; (b) unbranched tubular polymersomes; and (c) substantially spherical polymersomes. Substantially spherical polymersomes are for example those whose 2D projection (e.g. in a TEM image) have a circularity of greater than 0.9 and/or an aspect ratio of greater than 0.95. In the tubular polymersome composition of the present invention, the percentage of tubular polymersomes in the total population of polymersomes is typically at least 50%. Preferably, the percentage of tubular polymersomes in the total population of polymersomes is at least 70%, more preferably at least 80% and more preferably still at least 90%. The percentage of tubular polymersomes in the total population of polymersomes may even be as high as 95% or more. The tubular polymersome composition of the present invention is obtainable by carrying out the production method of the present invention.

Method of producing a tubular polymersome composition

As already explained, a method has now been found for obtaining a tubular polymersome composition from the polymersome compositions previously known in the art, which contain a mixture of polymersomes of different shapes.

The method is based on density gradient centrifugation and in particular comprises subjecting a mixed polymersome composition (e.g. a conventionally known and conventionally prepared mixed polymersome composition) to density gradient centrifugation.

Density gradient centrifugation is itself a well known and established technique. For example, this techniques is well known for use in separating cells parts after lysis according to their density. It has also previously been used for the purification of gold nanoparticles and carbon nanotubes with different sizes and shapes. It has not, however, previously been applied to polymersome compositions, e.g. to obtain a tubular polymersome composition.

The density gradient centrifugation is preferably sucrose gradient centrifugation. In sucrose gradient centrifugation, solutions of successively increasing sucrose concentration (e.g. in PBS) are layered in a centrifuge (e.g. from most dense to least dense). The mixed

polymersome composition is then deposited as a top layer, and centrifugation is then effected.

The centrifugation conditions (e.g. RCF and time of centrifugation) can readily be adjusted to those skilled in the art depending on the chemical composition of the mixed polymersome composition. An illustrative example is provided in the Examples section of this

specification.

The centrifugation is carried out in a centrifuge. Subjecting the polymersome composition to density gradient centrifugation gives rise to separate layers of polymersome compositions having different compositions in the centrifuge. The tubular polymersome composition can be isolated from the centrifuge, e.g. as one of the layers formed by the centrifugation.

Targeting moieties and encapsulated materials

The tubular polymersomes optionally comprise one or both of a targeting moiety and an encapsulated material.

The tubular polymersome optionally comprises a targeting moiety on its external surface. By on its external surface is meant that the targeting moiety is located such that it is able to interact with its target (as opposed to being located at an inaccessible position that precludes interaction with the target, for example by being encapsulated within the polymersome).

The targeting moiety is adapted to enable the polymersome to bind to a target. Typically the targeting moiety binds selectively to the target. The target is a chemical substance that is located on or in the vicinity of the tissue of interest (and thus enables the polymersome to be accumulate specifically at the tissue of interest in preference to other sites). The target is preferably a receptor, e.g. a receptor that is present in particularly high quantity at the target tissue of interest.

The targeting moiety can be any moiety that binds specifically to the target. As is well known in the art, for example from the well developed field of bioconjugates, a wide range of substances can be used as targeting moieties, e.g. to target receptors.

In one embodiment, the targeting moiety is a moiety that is attached to the external surface of the polymersome. Examples of suitable targeting moieties include antibodies, antibody fragments, aptamers, oligonucleotides, small molecules, peptides and carbohydrates. Peptide, antibody and antibody fragment targeting moieties are particularly preferred. However, any such moiety can be used as a targeting moiety in the present invention. The suitability of any given moiety to target any given receptor can be determined using routine assay methods, involving testing for the ability of the moiety to bind specifically to the receptor.

One example of a targeting moiety is a targeting moiety that is adapted to enable the polymersome to bind to a cancer cell. Illustrative and non-limiting examples of such targeting moieties include proteins (mainly antibodies and their fragments), peptides, nucleic acids (aptamers), small molecules, vitamins and carbohydrates. It will be appreciated that targeting of the polymersomes of the present invention to cancer cells is of particular interest in view of their capacity to decrease the replication activity of such cells.

The targeting moiety can be attached to the external surface of the polymersome using routine techniques, for example by adapting well known methods for attaching targeting moieties to polymers, drugs, nucleic acids, antibodies and other substances. The attachment may be non- covalent (e.g. electrostatic) or covalent, though it is preferably covalent. For example, the targeting moiety can be attached by reacting a suitable functional group on the targeting moiety (including but not limited to an amine group, a carboxyl group and a thiol group) with a corresponding functional group on at least one of the copolymers that form, or will form, the polymersome. The attachment can be effected either before the polymersome structure is formed from the copolymers, or after the polymersomes have been formed.

It is also possible to provide for attachment of the targeting moiety to the copolymers by first chemically activating either or both of the targeting moiety and the copolymers. For example, a peptide targeting moiety may be activated by adding a reactive species to one of its termini, such as a cysteine moiety (whose thiol group is well known to react readily with functional groups such as the widely used maleimide moiety). Similarly, a copolymer can be activated by functionalising it with a reactive species (e.g. a maleimide moiety when the targeting moiety carries a thiol group). The copolymer may be provided with such a reactive species either by functionalisation of the copolymer itself, or by providing suitable monomers prior to the polymerisation that forms the copolymer, or by providing a suitable initiator for the polymerisation.

The targeting moiety may be attached directly to the external surface of the polymersome or it may be attached via a chemical spacer. The targeting moiety may also be a pendant group of a polymer comprised by the polymersome (i.e. at least one of the copolymers forming the polymersome itself). Clearly in this embodiment it is not necessary to undertake separate synthetic steps to attach the targeting moiety to the copolymer or the resulting polymersome.

The tubular polymersome also optionally comprises an encapsulated material. Suitable encapsulated materials include anti-cancer drugs.

Pharmaceutical composition The tubular polymersome composition of the present invention can be formulated as a pharmaceutical composition using routine techniques known in the art. For example, pharmaceutical compositions already utilised for the formulation of polymersomes or drug- containing liposomes can be adapted to incorporate the tubular polymersome composition of the present invention.

The pharmaceutical composition comprises one or more pharmaceutically acceptable excipients or diluents. The one or more pharmaceutically acceptable excipients or diluents may be any suitable excipients or diluents. The pharmaceutical composition is typically aqueous, i.e. it contains water (in particular sterile water).

A typical pH of the aqueous pharmaceutical composition is 7.0 to 7.6, preferably 7.2 to 7.4. Pharmaceutically acceptable buffers may be used to achieve the required pH. The pharmaceutical composition may be in the form of a sterile, aqueous, isotonic saline solutions.

Typically the pharmaceutical composition is an injectable composition, e.g. it is suitable for intravenous delivery, for example it is suitable for infusion.

Medical use of the polymersomes

The tubular polymersomes of the present invention are particularly useful for treating cancer. As illustrated in the Example, it has surprisingly been found that tubular polymersomes are capable of selectively reducing the replication activity of tumor cells compared with non tumor cells. This reduction of tumor cell division may be related to an up-regulation of the tumor suppressor proteins p53 and p21. Specifically, tumor cell death is believed to occur via activation of the caspase 3/7 extrinsic pathway of apoptosis, suggesting that the tubular polymersomes lead to the activation of the cell death receptor.

Cancers particularly suitable for treatment according to the present invention include those susceptible to amelioration by: up-regulation of the tumor suppressor proteins p53 and p21; and/or activation of the caspase 3/7 extrinsic pathway of apoptosis.

Specific examples of cancers that can be treated include: cancers of the skin, such as melanoma; lymph node; breast; cervix; uterus; gastrointestinal tract; lung; ovary; prostate; colon; rectum; mouth; brain; head and neck; throat; testes; thyroid; kidney; pancreas; bone; spleen; liver; bladder; larynx; nasal passages; AIDS-related cancers; cancers of the blood and bone marrow, such as multiple myeloma and acute and chronic leukemias, for example, lymphoblastic, myelogenous, lymphocytic, and myelocytic leukemias; advanced malignancy, amyloidosis, neuroblastoma, meningioma, hemangiopericytoma, multiple brain metastase, glioblastoma multiforms, glioblastoma, brain stem glioma, poor prognosis malignant brain tumor, malignant glioma, recurrent malignant glioma, anaplastic astrocytoma, anaplastic oligodendroglioma, neuroendocrine tumor, rectal adenocarcinoma, Dukes C & D colorectal cancer, unresectable colorectal carcinoma, metastatic hepatocellular carcinoma, Kaposi's sarcoma, karotype acute myeloblastic leukemia, chronic lymphocytic leukemia (CLL), Hodgkin's lymphoma, non-Hodgkin's lymphoma, cutaneous T-Cell lymphoma, cutaneous B- Cell lymphoma, diffuse large B-Cell lymphoma, low grade follicular lymphoma, metastatic melanoma (localized melanoma, including, but not limited to, ocular melanoma), malignant mesothelioma, malignant pleural effusion mesothelioma syndrome, peritoneal carcinoma, papillary serous carcinoma, gynecologic sarcoma, soft tissue sarcoma, scleroderma, cutaneous vasculitis, Langerhans cell histiocytosis, eiomyosarcoma, fibrodysplasia ossificans progressive, hormone refractory prostate cancer, resected high-risk soft tissue sarcoma, unrescectable hepatocellular carcinoma, Waldenstrom's macroglobulinemia, smoldering myeloma, indolent myeloma, fallopian tube cancer, androgen independent prostate cancer, androgen dependent stage IV non-metastatic prostate cancer, hormone-insensitive prostate cancer, chemotherapy-insensitive prostate cancer, papillary thyroid carcinoma, follicular thyroid carcinoma, medullary thyroid carcinoma, and leiomyoma.

A therapeutically effective amount of the tubular polymersomes is administered to a patient. A typical dose is from 0.0001 to 1000 mg, measured as a weight of the tubular polymersomes, according to the age, weight and conditions of the subject to be treated, the type and severity of the disease and the frequency and route of administration. This dose may for instance be administered once daily. Preferably, measured as a weight of the tubular polymersomes, daily dosage levels are from 0.0001 mg to 4000 mg.

From 0.0001 to 1000 mg/kg, in total, of the tubular polymersomes may be administered. Typically, from 0.01 to 100 mg/kg of the tubular polymersomes may be administered.

Preferably, from 0.01 to 50 mg/kg or from 1.0 to 20 mg/kg of the tubular polymersomes may administered. These are typically daily doses of the tubular polymersomes. EXAMPLES

EXAMPLE 1

Summary

In this Example the biomolecular effects of polymeric nanoparticle shape were explored, with an emphasis on the role played by spheres and tubular vesicles. In particular, the pH- responsive amphiphilic diblock copolymer poly(2-(methacryloyloxy)ethyl

phosphorylcholine)-poly(2-(diisopropylamino)ethyl methaciylate) (PMPC-PDPA) was used, which may form vesicles (known as polymersomes) and tubes, under specific conditions.

First, a method was developed for isolating the two population of polymersomes (namely, spheres and tubes), based on density gradient centrifugation. This enabled work to be carried out with highly pure samples in terms of shape distribution. A cytome assay was then carried out, useful for quantifying the replication activity of cells, through the quantification of the nuclear division index (NDI), with the aim to understand any possible different interaction of polymersomes and tubes with the replication machinery of the cells. In order to have a broad overview of the molecular responses over different cells, all the investigations were carried out on two immortalised cell lines, namely the HeLa and FaDu cells, as well as on primary (non tumor) human dermal fibroblast (HDF).

After quantifying the NDI, a deeper insight into the potential molecular targets of such nanostructures was sought, by investigateing the expression of a panel of genes involved in general stress responses (e.g., oxidative stress and unfolded protein response pathway), with a focus on the tumor suppressor gene p53 and p21. A strong consistency was found between the regulation of specific genes and the outcomes of the cytome assay, especially in terms of cell proliferation activity. This was also correclated with the different uptake of spherical polymersomes and tubular polymersomes, and with the fate of cells coming in contact with such nanostructures. Finally, any potential adverse effects to the genetic materials were explored by means of micronuclei (MM) assays. Notably, the molecular response to polymersome shape was found to be cell-specific, so that tumor cells differently behaved comparing with primary fibroblasts.

All these data shed light first on the significance of using homogeneous samples of polymersomes (in terms of shape) for biological applications, as it is clear that shape at least in part dictates the final biological outcomes. Furthermore, it is evident that cells are able to "sense" different geometries, to actively respond to them in a precise and reliable way, and to regulate such responses even at genetic level in a cell-dependent way. Finally, tumor cells were demonstrated to differently sense and respond to shape stimuli, comparing to non-tumor cells, thus suggesting that the next development of tumor targeting polymersomes should take into account the question of geometry.

Results

The polymersomes were formed by means of film rehydration. Briefly, an organic solution of PMPC-PDPA was placed in a glass vial, and allowed to dry to deposit a polymeric film on the internal surface of the vials. The film was then rehydrated with a PBS solution, enabling the formation of both vesicular polymersomes and tubes (see the experimental section for more details). During this step, the copolymer will create different shaped structures that are strongly dependent on the copolymer/water ratio. This is a complex process from a kinetical viewpoint, and usually leads to the formation of different metastable phases such as spheres, multilamellar aggregates, and tubular polymersomes. Consequently, a sample will inevitably contain a mixture of both spheres and tubes.

In order to purify the two sub-population of vesicles, a protocol based on a density gradient centrifugation was optimised. In this approach, a PMPC-PDPA solution, continuously rehydrated for 8 weeks in PBS, was gently layered on a discontinuous gradient of sucrose (raging from 5% up to 25%, with a step of 5%). After a centrifugation step, two distinct bands were observed (see experimental section). The top band was characterised by spherical vesicles (at a sucrose density gradient of 5%), while the bottom band contains only tubular structures (in the 20% sucrose solution). Figure 1 shows the TEM characterisations of these bands. It is clear that the density gradient purification method successfully enabled separation of the two different vesicles populations, so that spherical polymersomes (Figure 1 A) were completely separated from tubes (Figure IB).

The next steps were then focused on understanding how such different shapes may affect specific cell behaviour, with a specific focus on the possible interaction with the replication machinery of cells. Before further exploring such topic, a preliminary study aiming to understand the best polymersome incubation conditions was carried out, in terms of concentrations, in order to avoid a diffuse toxicity. Specifically, a viability investigation based on the MTT assay was carried out (see Figure 2). Three final polymer concentrations were tested, namely 0.1, 0.5 and 1 mg/mL, for 24, 48, and 96 hours of incubation time. The results showed that polymersomes did not induce a significant cytotoxicity in almost all the condition tested, both as a function of time and concentration. Only a slight decrease in viability of about 20% and 25% was observed for FaDu and HeLa cells, respectively, after incubating them for 96 hours at the highest concentration (1 mg/mL). Primary FIDF were completely unaffected by polymersomes in all the conditions tested. The same test provided different outcomes for tubular vesicles, where a significant decrease in FaDu and HeLa viability (of c.a. 50% and 40%, respectively) was observed at a polymer concentration of 1 mg/mL after 96 hours of incubation time. Also FIDF cells incubated with tubular vesicles depicted a slight decrease in viability, although it was overall lower comparing to tumor cells (about 20%) of decrease). In addition to set the best experimental conditions for the next investigations, these preliminary data suggest the possibility of a different interaction between cells and both spherical and tubular vesicles. Such interactions appeared to be also cell- dependent, so that tumor cells resulted more affected by tubular structures comparing to primary HDF.

To understand if the first level of different interaction relies on the uptake of these structures, cellular uptake of spherical polymersomes and tubes was assessed. Rhodamine B- encapsulated polymersomes were used, and the fluorescence level quantified within the cells at different time points by means of HPLC. In this method, the cell cultures were all incubated with a final concentration of 0.1 mg/mL of both spheres and tubes, in order to be in a safe concentration range in terms of cytotoxicity, according to the preliminary viability assays shown in Figure 2. It is important, in fact, to quantify the final amount as a function of the number of viable cells, otherwise any outcome will be misrepresented because of cell death.

Figure 3 A shows the total mass of spherical polymers present in the different cell cultures over time, quantified by means of HPLC. It is evident that the uptake of polymersomes follows an increasing trend in all the cells, especially between 48 and 96 hours of treatment, where the engulfment process is much more pronounced. The only difference here is the final quantity of the up taken material, so that FaDu cells resulted to be the most effective in the endocytosis of spherical structures after 96 hours of incubation, with an average polymer quantity of 7 μg (figure 3 A, left). The polymersomes uptake was quantified to be c.a. 4 μg for the HeLa cells, and about 2 μg in the primary fibroblasts after 4 days treatments. The results were completely different for tubular structures. In this experimental condition, the quantification of uptake revealed that there is not an increasing trend of internalised material during time, so that after 96 hours of incubation all the tested cells displayed an average quantity of 2 μg of up taken polymer. These results become more interesting when normalised as a function of the total number of cells present for each time point, as shown in figure 3B. In particular, the overall amount of polymer/cell (ng/cell) significantly increases during time in HDF treated with spherical polymersomes, moving from an initial amount of -0.04 ng/cell up to a final one of -0.06 ng/cell, after 96 hours. On the contrary, the trend is completely opposite in the case of tumor cells (HeLa and FaDu), where the initial value is about 0.03 ng/cell and decreases below 0.01 after 4 days of incubation with polymersomes (figure 3B, left). As expected, the results changed when cells came in contact with tubular vesicles. In this case, all the trends expressing the amount of tubes/cells (in ng/cells) are prone to decrease over time in all the cells (figure 3B, right). Despite this tendency, the only difference, again, is represented by HDF, which were found to have an overall higher amount of tubes (from 0.05 ng/cell at 24 hours to 0.02 ng/cell after 96 hours) comparing to both FaDu and HeLa (where the amount was quantified to decrease from the initial -0.02 ng/cell to almost zero after 96 hours). As in the case of viability, also these data strongly support the evidence of the complete different behaviour of cells treated with spherical or tubular vesicles (see the discussion section for more details).

To further confirm the evidence of differential uptake between sphere and tubes, confocal investigations on FaDu cells, considered as representative cells in this approach (and considering that all the HPLC-related trends of engulfment were similar in the three experimental group), were carried out. FaDu cells were incubated with rhodamine B- encapsulated polymersomes (in the same way as for the previous HPLC quantification), and after 96 hours the cells were stained with calcein green, with the double aim to check for viability and internalisation. The confocal images in Figure 4 demonstrate that all the spherical polymersomes are effectively engulfed within the cells (Figure 4A), and that they are spread throughout the cytosol. On the other side, most of the tubular structures were found to be attached outside of the cells, and only small amount of material engulfed (Figure 4B).

As a next step, it was attempted to understand a possible molecular interaction between these nanostructures and the replication machinery of the cells. One of the most reliable ways to explore such a topic is through the cytome assay, better known as "cytokinesis-block micronucleus cytome assay". This is a method for quantifying the (eventual) cytostatic effects of compounds, as well as DNA damage, and general cytotoxicity. In this approach, cells are treated with cytochalasin-B (Cyt-B), which inhibits the microfilament assembly during the final stage of cytokinesis, thus enabling the nuclear division while stopping the division of the cytoplasm. Thanks to this approach, it is possible to finely quantify the number of mono-, bi-, and multi -nucleated cells as a function of the number of viable cells, namely the Nuclear Division Index (NDI). This value will directly reflect the replication activity of cells, upon specific treatments, and provides a refined quantification of the potential cytostatic activity of the compounds tested. Hence, variation of NDI comparing to untreated cells will reflect the intrinsic ability of such compounds to directly and/or indirectly interact with the replication machinery of the cells.

In the present experimental approach, Fadu, HeLa and HDF cells were first synchronised by serum starvation, in order to force them into the stationary phase of the cell cycle (Go), and then treated with both spherical polymersomes and tubular vesicles for 24h at a final concentration of 0.5 mg/mL. Cell cycle synchronisation is a crucial optimisation step in this technique, as the NDI will be otherwise significantly affected. After 24 hours of treatment, Cyt-B was added in the cell culture, and cells were finally fixed and analysed by means of confocal microscopy. For each set of experiments, more than 3000 cells were analysed in order to have an optimal and correct representation of the NDI (see experimental section for details). All the quantifications of mono-, and multi -nucleated cells were carried out with a home made created Matlab script, which was designed ad hoc with the aim of measure the NDI. Figure 5 is a representative picture of the ability of the software to recognise, distinguish, and count the cells having one or more nuclei, scanned by means of confocal microscopy.

Figure 6A shows the NDI found in all the tested cells. With respect to the FaDu, the NDI for the control (untreated) cells displayed a value of -1.65, while the same cells treated with both spherical vesicles and tubes had a lower NDI of -1.55. The positive control (cells treated with H2O2) revealed an increase value up to 1.8. The trend was found to be quite similar also in HeLa cells. Here the NDI in untreated cells displayed a value of -1.7 while the positive control was -1.8. However, the treatment with polymersomes did not induce any statistically significant shift in the NDI (with respect to the control), while this was quite remarkable in

HeLa cells incubated with tubes. In this case, the NDI value decreased down to 1.4. With respect to both the tumor cells tested, it was thus evident that tubular vesicles interact with the replication machinery of the cell, acting as cytostatic compounds. The outcomes completely changed in the case of HDF, where the DI was -1.2 for the control, and -1.25 when treated with polymersomes (hence, the difference was not significant). Remarkably, the NDI considerably rose up to -1.5 for fibroblasts treated with tubular vesicles, in an opposite trend comparing to tumor cells. Dissimilarly to both HeLa and FaDu, the tubes were found to promote HDF replication, a phenomenon likely to occur in stressed fibroblasts. Consistently with the previous observations, also these findings confirmed that polymersome shape leads to different effects on cells, which respond to these stimuli in a cell dependent way. So that, tuning cell behaviour towards slowing or increasing their replication activity is achievable through polymersome shape.

Having in mind that the NDI experiments suggested an interaction between the vesicular NPs and the cell replication machinery, it was next explored whether this event can be harmful in terms of DNA damage. The same method from the NDI experiments was exploited to carry out a micronuclei (MM) assay. In this case, the Matlab script for figure analyses quantified the number of bi-nucleated cells having at least one micronucleus (that results from an eventual DNA damage). The results for the M assay is reported in Figure 6B. In this case, no significant genotoxicity was observed in any of the tested cells, both for spherical polymersomes and tubes. These data suggest that, although an interaction between

sphere/tubes and DNA will occur, this will not lead to any damage.

To further understand whether the tubes-induced decrease level of proliferation is focused only in the first 24 hours (the time in the cytome assay), or alternatively if it lasts for a longer time, a 4 days trypan blu quantification was carried out of viable cells incubated with 0.5 mg/mL of both spherical vesicles and tubes. Figure 7 shows that the number of viable FaDu and HeLa cells tended to be constant during the 96 hours of incubation time. Specifically,

FaDu were in the range of - lxlO⁵ cells, while HeLa grew up to a maximum of 1.8xl0⁵ between 24 and 96 hours of incubation. This confirms the intrinsic cytostatic activity of these nanostructures. At the same time, the number of cells contacting spherical polymersomes are completely similar to the control (~3xl0⁵ and ~6xl0⁵ cells for FaDu and HeLa, respectively), confirming that such vesicles do not possess the ability of inhibit cell growth during time.

Dissimilarly, the number of fibroblasts incubated with tubes showed a significant

proliferation activity after 24 hours of treatment (up to ~3.6xl0⁵), while spherical

polymersomes-treated HDF behaved similar to the control (~1.4xl0⁵) in the first 1 day.

However, the number of viable HDF incubated with tubular structures shifted down to normal values between 48 and 96 hours, demonstrating the ability of these cells to recovery the temporary stress. Also in this case, all the data are consistent with the outcomes of the changing values of the DI (see previous discussion and Figure 4A).

Both the NDI and MM are morphological assays, so that the conclusions came from macroscopic observations, which are represented in this cases by the variation in the number of multi -nucleated cells (for the NDI) and the presence of micronuclei (MM). To go deeper into the molecular aspects of the interactions between the two shapes and the investigated cells, the expression levels of a panel of genes were screened, by means of Real Time quantitative PCR. Similarly to the experimental setup of the cytome assay, cells were incubated with a 0.5 mg/mL solution of spherical polymersomes or tubes for 24 hours, followed by RNA extraction, retrotranscription and finally by quantitative PCR. It was decided to explore general stress-related genes, namely the SOD1 and CAT for oxidative stress, the cytochromes CYPlAl and CYPIBI for detoxification, the chaperones HSP27 and HSP70, and finally the sensors for unfolded protein response (UPR) ATF4 and ATF6.

Besides these markers identifying different levels of cell stress, the regulation of p53 and p21 was investigated, mainly because in order to understand at molecular level the different proliferation activities previously found in the cytome investigations (the increase and decrease in values of NDI). The quantification of genes expression is reported in Figure 8. It was considered biologically significant all those genes having an increase or decrease in expression of at least 0.75 (i.e., 75%) folds comparing to the expression in the control (untreated cells). Interestingly, it was found that FaDu cells significantly over expressed the p21 gene (more than 2 times) upon treatment with both spheres and tubes (the white histograms represent the "spherical polymersomes experimental group", while the black are the "tubes experimental group"). At the same time, the p53 gene was slightly down regulated only in the presence of tubes, while spherical polymersomes did not induce any significant effect over the regulation of p21. All the other tested genes did not display remarkable regulations in the two conditions. Also in HeLa cells, the treatment with both vesicles and tubes resulted in a significant over expression of p21 (-1.5 times), as in the case of FaDu. However, also found was an important up regulation of p53 upon treatment with tubes, while spherical vesicles were not effective towards p53 regulation. In addition, a slight over expression of the CAT gene was observed when HeLa cells contacted tubular structures. Again, spherical polymersomes did not induce any effect on CAT gene regulation. All the other stress related genes remained unaltered. Finally, the same panel was investigated in primary fibroblasts. As expected, the results were rather dissimilar from cancer cells, as in this case all the genes were not significantly regulated compared to the control, when cells were treated with both spherical polymersomes and tubes. Such findings are in strong agreement with the previous data on the decrease of the DI (which reflect a decreasing tendency in proliferation) when tumor cells are treated with tubes, and provides a molecular explanation to this phenomenon (see discussion for more details).

As a final consideration, it was investigated whether this activation of p21 and p53, together with the consequent decrease in the replication activity of cells, may be related to caspase activation, which are markers of apoptosis. In particular, the activity of caspase 3/7 and 9 was investigated. Caspase 3/7 are known to become active as a function of the binding of external ligands to cell death receptors, which then become active in the pathway known as extrinsic apoptosis. On the other hand, the caspase 9 undergoes activation upon intracellular signalling, related to the stress-related release of cytochrome c from mythocondria, a process known as intrinsic apoptosis. Figure 9 shows the activation of capsize 3/7. In particular, it is evident that spherical polymersomes did not induce any significant caspase 3/7 activation in all the tested cells. On the other hand, tubular structures led to a slight activation in FaDu cells, which was more prominent in HeLa. The caspase 3/7 were not active in fibroblasts treated with the same tubular structures. Similar investigations were carried out to address a possible activation of caspase 9. However, no significant activation of caspase 9 was observed upon treating all the cells with either spherical polymersomes or tubular vesicles.

Discussion

In this methodological study, it was explored how the shape of vesicular spheres and tubes affect the cell behaviour by focusing on several molecular and genetic aspects. Before starting the biological investigations, a method was developed for obtaining a complete homogeneous fraction of spheres and tubes, respectively. A typical production process of vesicular structures, in fact, will result in a combination of differently sized and shaped assemblies, as membranes have the tendency to wrap into different structures and high molecular weight block copolymers (that is the PMPC-PDPA in our study) usually lead to the formation of non-spherical structures. Consequently, a post-production purification method was exploited, based on the density gradient centrifugation. It was demonstrated for the first time that the density centrifugation method may be extremely useful also for the purification of a heterogeneous sample containing both spherical and tubular structures. In order to have a good overview of the biological phenomena, two tumor cell lines (FaDu and HeLa), and one primary non tumor cell (HDF), were investigated.

As a first preliminary investigation the uptake level of both vesicles and tubes was quantified, to understand whether there is a difference in the level of engulfment between the cells. It is worth pointing out that the lowest concentration of both spherical polymersomes and tubes was used for quantifying the dynamics of uptake, in order to avoid undesired toxicity that might hinder reliable outcomes (see Figure 2). It was observed that while spheres are up taken with a linear increasing trend during time, with a maximum over 96 hours, the engulfment of tubular structures was rather constant, without any detected increase (figure 3 A). More interestingly, the normalisation of the signal over the number of viable cells provided further details. In particular, although the overall amount of polymersomes is less in HDF comparing to the other cells (Figure 3 A), the amount per cell was found to be higher in fibroblasts comparing to both FaDu and HeLa (figure 3B). This is probably because the replication activity of tumor cells is more pronounced than their concomitant ability of up taking nanoparticles, so that the amount of vesicles/cell decreases over time. Dissimilarly, particle endocytosis in HDF is faster than the replication activity, and consequently the quantity of vesicles/cells tend to increase over time. On the other hand, the respective trends of tubes/cells are prone to decrease in all the tested cells, mainly because the endocytosis of this structures is rather ineffective. The quantification data were further supported by an additional evidences. In particular, confocal microscopy-based imaging investigations demonstrated that spherical polymersomes are effectively engulfed within FaDu, and they distribute randomly throughout the cytosol (figure 4A). Dissimilarly, tubes have the tendency to lie outside of the cell, and most of the material is not effectively up taken, thus in agreement with the non-increasing trend of uptake found by HPLC quantification (Figure 3A).

All these preliminary investigations suggest that the interactions occurring between spheres/tubes and both tumor and non tumour cells are different. As a consequences, it was considered whether cells may respond in different ways to such dissimilar external physicochemical stimuli. In this respect, attention was focused on the replication machinery of cells, as it is a paramount aspect especially in nanomedicine. Thus the cytome assay was exploited, focusing on the DI first, which provides a score resembling the dynamics of cell division. The NDI is hence a refined method to properly quantify the replication activity of cells treated with the material of interest, as it is possible to detect the number of mono- and multi-nucleated cells through confocal imaging. Focus was placed on the effects within 24 hours in this approach. In these experiments, it was observed that the tubes induced a decrease in the DI comparing to the control in tumour cells, thus reflecting a tendency to slow down their replication activity (Figure 6A). In the case of FaDu, also the treatment with spherical polymersomes led to a slight NDI decrease, while the same spherical polymersomes did not elicit any significant effect on HeLa. On the other hand, the treatment with tubes indued an increase of the replication activity of primary fibroblasts (which is a typical sign of stress), while polymersomes again did not display any remarkable effect (Figure 6A). The change in the NDI values highlights that there is an interaction occurring between the replication machinery of cells and polymersomes, and that this is strongly dictated by the polymersome shape. In addition, the cellular responses to those physicochemical stimuli are rather different between tumor and non tumor cells, as fibroblasts behave in a complete opposite way comparing to both HeLa and FaDu, upon interacting with tubes. The cell replication was also analysed for longer incubation time by means of frypan blue vital counting assay, further confirming the outcomes coming from the NDI experiments (Figure 7). To further address whether this interaction might induce any adverse effects, the presence of micronuclei that are a direct evidence of DNA damage were quantified (Figure 6B).

However, no significant change in the number of micronuclei was detected in any of the conditions tested, thus supporting the idea that such interactions are not harmful to the DNA, albeit they were found to tailor the cell division.

Such evidence on the possibility of tailor cell behaviour via NPs shape prompted further analysis as to how the cells respond at genetic level, namely by screening the expression levels of several genes through the real time quantitative PCR (Figure 8). It was decided to focus on the two regulators of the cell cycle p53 and p21, as well as on other stress related genes, to possibly find a link between the morphological observations of NDI variation and the genetic regulation behind such changes. It is, in fact, well established the role of p53 as a tumor suppressor protein, which responds to dangerous stress signals by (/^') arresting cells cycle, and by {if) promoting cell senescence and apoptosis. Similarly, also the cyclin- dependent kinase inhibitor p21 acts as both sensor and actuator multiple anti-proliferative signals. Surprisingly, it was found that the treatment with spherical vesicles and tubes led to a significant up-regulation of p21 FaDu cells, while p53 was slightly down-regulated upon treatment with tubes. This outcome was interesting. It should be considered that usually p21 and p53 are part of a similar pathways, so that activation of p21 usually induce the up- regulation of p53. However, there are several evidences that p21 and p53 may act also in independent ways. Hence, the present data suggest that the treatment with both spherical and tubular nanovesicles leads to the activation of the alternative pathway of p21, independently of p53. This situation is different in HeLa cells. In particular, these cells up-regulate both p53 and p21 upon treatment with tubes (in this case spherical polymersomes do not lead to p53 activation, thus similarly to the case of FaDu). Additionally, the CAT gene, codifying for the catalase (enzyme important for detoxification from oxidative stress), was found to be slightly activated in HeLa, suggesting that those cells might sense a stronger stress upon incubation with tubes. Such evidence indicates that the treatment with spheres and tubes may alternatively lead to the classical or independent activation of the two proteins, confirming once more the possibility to finely tailor cell behaviour at genetic level by just tuning the Ps shape. In addition, it should be mentioned that the over-expression of p21, and of p53 for HeLa, upon incubation with tubes, is strongly consistent with the data coming from the NDI. In particular, the prominent decrease in the value of NDI for HeLa treated with tubes could be attributed to the up-regulation of both p21 and p53. At the same time, the activation of p21 only in FaDu may explain the slighter decrease of their NDI. As expected, primary fibroblasts behave completely different, as none of the investigated genes were differently regulated with significant statistical validity. Once again, the investigated cells depicted opposite ways to respond physicochemical stimuli, in a way that tubular structures look to affect much more tumour than non-tumor cells.

Finally, an effort was made to address whether the evidence of p21- and p53-related arrest of cell division would result in caspase activation. Both caspase 3/7 and caspase 9 were investigated upon incubating cells with spheres and tubes, to further be able to distinguish between the potential activation of extrinsic (caspase 3/7) or intrinsic (caspase 9) pathways of apoptosis. It was observed that the treatment with tubes in both HeLa and FaDu cells induces a light activation of caspase 3/7 (Figure 9), while spherical polymersomes did not have detectable effects on cells. As expected, caspase 3/7 are not regulated in primary fibroblast in any of the condition tested. The tubes-related caspase 3/7 activity in tumor cells confirms the hypothesis that the slowing down of cell division is related to the activation of p21, which in turns slightly promotes the apoptotic pathway. On the other hand, the analyses of caspase 9 confirmed that an intrinsic apoptosis does not occur in all the experimental conditions. The extrinsic pathway of apoptosis is strictly related to the activation of extracellular death receptors, and in our evidences tubular structures are thus most likely to interact with them. Conclusions

Understanding the interactions occurring between nanomaterials and living systems is a critical aspect in the design of biomedical devices with active functional properties. This is because polymersomes are in the same size range of proteins, so that they may interact each other, with a consequent tuning of specific, possibly uncontrolled, cellular behaviours. It is indeed necessary to deeply understand the way nanoparticles tune the physiology of cells. To this respect, the physicochemical properties of Ps, and shape in particular, are important parameters to be considered in the process of biointeractions.

In this work, the interactions between differently shaped polymersomes, namely spherical and tubular nanovesicles, were investigated with tumor and non tumor cells. It was observed that tubular structures differently interact with cells, and that tumor and non-tumor cells differently respond to tubes. In particular, they actively interact with the replication machinery of tumor cells (but not of non-tumor cells), by reducing the DI, which is a direct measurements of the replication activity of cells. On the other hand, spheres did not influence the cell replication. Moreover, this reduction of tumor cell division was related to a tubes- induced up-regulation of the tumor suppressor proteins p53 and p21. We also demonstrated that cell death occurs only via the activation of caspase 3/7 extrinsic pathway of apoptosis, while the caspase 9 intrinsic pathway resulted inactive. This suggests that tubes lead to the activation of the cell death receptor.

Materials and Methods

Preparation of PMPC25-PDPA70 polymersomes

PMPC25-PDPA70 was prepared by atom-transfer radical polymerisation (ATRP) as described in Journal of the American Chemical Society 127, 17982-17983, 2005. In brief, the initiator was mixed with 25 equivalents of MPC, dissolved ethanol, degassed and

CuBr (1 eq.) and bipyridin (2 eq.) added. After one hour the polymerisation mixture was assessed by MR for complete conversion. A degassed solution of 70 eq. DPA dissolved in ethanol were then added and the polymerisation again ran until completion (16 hours). Salts and organic molecules were removed by filtration through silica gel and consecutive dialysis. The final solution was freeze-dried to yield the PMPC25-PDPA70 polymer. GPC was performed on a GPC max (Malvern Instruments) using Novemax Guard and

Analytical coloumn (PSS Polymer) in water with 0.25 vol-% triflouroacetic acid. Self- assembly of polymersomes was initiated using the thin film rehydration method. Briefly, a solution of 10 mg/mL of PMPC25-PDPA70 in 2: 1 chloroform:methanol was transferred in sterile glass vials, and allowed to dry under a vacuum oven for 48 hours. Sterile PBS was then added under stirring for a period of 30 days for the self-assembly of polymersomes and tubes.

Sucrose density gradient for shape -dependent polymer some purification

For the purification of polymeromes by shape, density gradient centrifugation was used. Briefly, solutions of increasing sucrose concentration, dissolved in PBS, were used, namely 5, 10, 15, 20, and 25% wlv. Aliquots of 200 L for each solution were gently layered from most dense to least dense within a 1.5 mL micro-centrifuge tube, while avoiding random dispersion of each aliquot. Finally, 150 L of solution containing a mixture of rhodamine-labeled vesicles and tubes, prepared by film rehydration, was deposited as the final top layer, and the microcentrifuge tube was centrifuged at 20000 RCF for 2 hours. After centrifugation, 20 L of each layer was collected and analysed by TEM.

TEM characterisations

TEM analysis was performed using a FEI Tecnai G2 Spirit electron microscope and/or a JEOL 2100 operating at 200 kV equipped with a CCD camera Orius SC2001 from Gatan.. Copper grids were glow discharged and the sample was adsorbed onto the grid. The sample was then stained with 0.75wt% phosphotungstic acid (PTA) raised to pH 7.4 with NaOH.

Cell cultures

Primary human dermal fibroblasts (FIDF), ovarian cancer cells (HeLa), and oral carcinoma (FaDu) cells were purchased from ATCC^®. FIDF, HeLa, and FaDu cells were cultured and maintained using Dulbecco's Modified Eagle Medium (DMEM) (Sigma-Aldrich^®) containing: 10 (v/v) fetal calf serum, 2 111M L-glutamine, 100 mg/ml streptomycin and 100 lU/ml penicillin (Sigma-Aldrich " ). Cells were cultured at 37 C/95% air/5% CO?.. Cells were periodically sub-cultured using Trypsin-EDTA solution 0.25% (Sigma-Aldrich^®) for the detachment process and centrifuged at 2000 rpm for 5 min for the pellet collection.

Viability assays

For cell viability, the Thiazolyl Blue Tetrazolium Blue (M I T, Sigma) method was used. Briefly, cells were seeded at a concentration of 5 x 10³ cells/well in a 96 well plate O.N.. Increasing concentrations of polymersomes and tubes were then added in the growth media, namely 0.1, 0.5, and 1 mg/mL, for periods of 24, 48, and 96 hours. The medium growth was then removed and an acidified solution of isopropanol was added to dissolve the water- insoluble MTT formazan. The solubilised blue crystals were measured col ori metrically at 570 nm (plate reader ELx800, BioTek).

For the Trypan blue-based proliferation assay, cells were seeded at a concentration of 8 x 10³ cells/well in a six well plate, and then incubated with both polymersomes and tubes at a concentration of 0.5 mg/mL for 24, 48, and 96 hours. Cells were then detached with a Trypsin-EDTA solution 0.25%, and counted with an automated cell counter (TC20, Bio-Rad).

HPLC quantification

For the uptake quantification by HPLC, cells were seeded at a density of 8 x 10³ cells per well in a six-well plate, and incubated with Rhodamine-labeled polymersomes and tubes (0.1 mg/mL final concentration). Cell were then lysed in acidified PBS (pH 2), and stored at -80 °C for 48 hours to promote lysis. The lysate was then centrifuged at 20,000 G for 1 hour and the supernatant was collected for the HPLC analyses (UltiMate 3000 Standard LC Systems, Thermo Scientific).

Cytome assays and confocal and imaging analyses

For the evaluation of the nuclear division index (NOT) and micronuclei (MM), cells were initially seeded in glass bottom dishes (35 mm diameter-IBIDI^®) at a density of 8 x 10³ cells per well, grown O.N. in complete medium, starved for 24 hours to synchronise the cell cycle. Cells were incubated with H2O2 (positive control), spherical polymersomes, and tubes for 24 hours, washed three times with PBS (5 minutes) and incubated with cytocalasinB (Sigma) for 24 hours to avoid the cytoplasm division. Then, all cells were washed with PBS, fixed with formaldehyde 3.7% for 10 minutes, permeated with 0.1% Triton X-100 in phosphate buffered saline (PBS) for 10 minutes, and stained with Hoechst 33258 (Sigma) and SYTO® 9 (Sigma) for nuclear and cytoplasm staining, respectively.

For the uptake imaging, FaDu cells were seeded in the same glass bottom dishes, incubated with either polymersomes or tubes, and finally stained with Calcein (Life Technologies) for vital staining. Cell imaging for both NDI/MNi and uptake was carried out with confocal microscope (Leica TCS SP8), and imaging quantification was carried out with an ad hoc designed Matlab script. For the scoring of DI and MM, the guidelines of Fenech were adopted.

Reverse transcription polymerase chain reaction (RT-PCR), and PCR assays

Cultured cells, incubated for 24 hours with polymersomes and tubes (0.5 mg/mL), were lysed and total RNA was collected by using RNeasy Mini Kit (Qiagen). RNA concentration was measured with NanoDrop spectrophotometer (Thermo). Complementary DNA (cDNA) was synthesised from every 1 μg of total mRNA in 20 μΙ_, volume per tube with QuantiTect Rev. Transcription Kit (Qiagen). The samples were then run in a standard agarose gel (1%) for RNA quality control check. For the PCR analyses, GAPDH and ACTB were used as reference genes, and:

Quantitative analysis was assessed with QuantiTect SYBR Green RT-qPCR Kit (Qiagen). The amplification process was done in 20 μΙ χώε, using the following steps: 95°C for 5 min to make active the DNA Polymerase, followed by 40 cycles of 95°C (10 s) for denaturation, and 60°C (30 s) for combined annealing and extension for all primers. Melting curve was also acquired, to analyse the sample quality, from 55°C to 99°C, by increasing of l°C/min. Data were analysed via AACt value. 2^"ΔΔα was calculated as follows:

Ct GAPDH;

AACt=ACt(treated) -ACt(control). The genes expression was analysed using the following list of primers:

Gene name Primers

CYP1A1 For: GAAGCAGCTGGATGAGAACG

Rev: GACCTGCCAATCACTGTGTC

CYP1B1 For: CTCGAGTGCAGGCAGAATTG

Rev: TCCTTGGGAATGTGGTAGCC

HSP70 For: CCTCAGTCTGATGGCTCCAG

Rev : CTCCTGGTCC ACTTGC ATCT

HSP27 For: CCAAGTTTCCTCCTCCCTGT

Rev : CTTT ACTTGGCGGC AGTCTC p53 For: TGGCCATCTACAAGCAGTCA

Rev: GGTACAGTCAGAGCCAACCT p21 For: GTGAGCGATGGAACTTCGAC

Rev: CAGGTCCACATGGTCTTCCT

Apoptosis assays

For the characterisation of intrinsic and extrinsic apoptosis, the luminesce-based Caspase- Glo® 3/7 Assay Systems (Promega) and the Caspase-Glo® 9 Assay Systems (Promega) were used, respectively. Cells were seeded at a concentration of 8 x 10³ cells per well in a 96-well plate, and incubated with polymersomes and tubes (0.5 mg/mL) for 24 hours. The caspase solution was then directly added to the media to have a 1 : 1 final ratio, and the luminescence was measured (Varian Cary Eclipse).

EXAMPLE 2

The hypothesis that the combinatorial effect of shape and encapsulated anticancer drug could enhance the efficacy of treatment was tested.

Briefly, cells were seeded to a final concentration of 5 x 10³ cells/well in a 96 well plate overnight. 0.1, 0.5 and 2.5 ug/mL of Doxorubicin were then inoculated, either as a free DOXO or DOXO encapsulated within spherical or tubular polymersomes, to cancer (i.e., HeLa and FaDu) and non-cancer primary (HDF) cells. After an incubation period of 24 and 48 hours, the treatments were removed, cells were washed with Dulbecco's PBS, and treated with 0.5 mg/mL of 3-(4,5-dimethylthiazol)-2,5-diphenyl tetrazolium bromide for 2 hours at 37°C. The reduced salt was then solubilised with a solution of acidified isopropanol and read with a plate reader (wavelength: 570nm and 630nm). Signal was normalised and t-test was used for statistical analyses. The results are shown in Figure 10.

Analyses confirmed that both FaDu and HeLa cancer cells were significantly affected by the combination of polymersomes size and encapsulated Doxorubicin. In particular, tubes encapsulated with DOXO were much more effective than the same final dose of free or spheres-encapsulated DOXO in FaDu cells after 24 hours of treatment. After 48 hours, the effect was similar in the three condition tested. Similar results were obtained with HeLa cells, but in this case tubes and spheres encapsulated with DOXO behaved similarly after 24 hours (but still both more effective than the free drug). However, after 48 hours the tubes loaded with the anticancer drugs were more efficient in killing cancer cells compare to either DOXO- loaded spheres or free drug. On the other hand, no significant effect was observed on non- cancer HDF cell, thus confirming that the system is able to discriminate between cancer and non cancer cells.

Claims

A method of producing a tubular polymersome composition, the method comprising the steps of: providing a mixed polymersome composition comprising a mixture of tubular polymersomes and non-tubular polymersomes; subjecting the mixed polymersome composition to density gradient centrifugation in a centrifuge; and isolating a tubular polymersome composition from the centrifuge.

The method of claim 1, wherein the density gradient centrifugation is sucrose gradient centrifugation.

The method of claim 1 or 2, wherein the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is greater than the percentage of tubular polymersomes in the total population of polymersomes in the mixed polymersome composition.

The method of any one of the preceding claims, wherein the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is at least 50%.

The method of any one of the preceding claims, wherein the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is at least 70%.

The method of any one of the preceding claims, wherein the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is at least 80%.

The method of any one of the preceding claims, wherein the percentage of tubular polymersomes in the total population of polymersomes in the tubular polymersome composition is at least 90%.

8. A tubular polymersome composition, wherein the percentage of tubular polymersomes in the total population of polymersomes is at least 50%.

9. The tubular polymersome composition of claim 8, wherein the percentage of tubular polymersomes in the total population of polymersomes is at least 70%.

10. The tubular polymersome composition of claim 8 or 9, wherein the percentage of tubular polymersomes in the total population of polymersomes is at least 80%.

11. The tubular polymersome composition of any one of claims 8 to 10, wherein the

percentage of tubular polymersomes in the total population of polymersomes is at least 90%.

12. A tubular polymersome composition obtainable by the method of any one of claims 1 to 7.

13. A pharmaceutical composition comprising: the tubular polymersome composition of any one of claims 8 to 12; and one or more pharmaceutically acceptable excipients or diluents.

14. A tubular polymersome composition of any one of claims 8 to 12 for use as a

medicament.

15. A tubular polymersome composition of any one of claims 8 to 12 for use in the

treatment of cancer.

16. A method of treating cancer, the method comprising administering a therapeutically effective amount of the tubular polymersome composition of any one of claims 8 to 12 to the subject.

17. Use of a tubular polymersome composition of any one of claims 8 to 12 in the

manufacture of a medicament for use in the treatment of cancer.