INERT NANOCAPSULES

Abstract

Compounds are presented having the formula [M.sub.4L.sub.6].X.sub.n, where M is a transition metal ion having 6 d-shell valence electrons, X is a counter ion and n is the number of counterions such that the total charge of the compound of formula (I) is zero, and wherein L is a 5-(5-bipyridin-2,2′-yl)-2,2′-bipyridine or derivative thereof. Methods of preparing the compounds and uses of the compounds to retain radiolabels are also presented.

Claims

1. A compound of formula I
[M.sub.4L.sub.6].X.sub.n,  (I) wherein M is a transition metal ion having 6 d-shell valence electrons; wherein X is a counter ion and n is the number of counterions such that the total charge of the compound of formula I is zero; and wherein L has formula II ##STR00013## wherein R.sup.1, R.sup.2, R.sup.3, R.sup.6, R.sup.7, R.sup.8, R.sup.9, R.sup.10, R.sup.13 and R.sup.14 are independently H, OR.sup.15, OCOR.sup.15, OC(O)OR.sup.15, OP(O)(OR.sup.15)(OR.sup.16), OP(O)R.sup.15R.sup.16, SR.sup.15, SCOR.sup.15, SC(O)OR.sup.15, SP(O)(OR.sup.15)(OR.sup.16), SP(O)R.sup.15R.sup.16, NR.sup.15R.sup.16, NR.sup.16COR.sup.15, NR.sup.16CO.sub.2R.sup.15 or NR.sup.17P(O)(OR.sup.15)(OR.sup.16); wherein R.sup.4 and R.sup.5 are H or linked CH groups to form an N,N-phenanthroline ring, and/or R.sup.11 and R.sup.12 are H or linked CH groups to form an N,N-phenanthroline ring; wherein R.sup.15, R.sup.16 and R.sup.17 are independently selected from H and C.sub.1-C.sub.10 alkyl, polyethylene glycol (PEG) and polyether groups; and wherein Z is selected from a chemical bond, acetylene, disubstituted phenylene, naphthalene, anthracene, pyrene or similar chemically rigid spacer.

2. A compound according to claim 1, wherein M is selected from cobalt(III), rhodium(III) iridium(III), iron(II), ruthenium(II), or osmium(II).

3. A compound according to claim 2, wherein M is cobalt (III).

4. A compound according to claim 1, wherein the ligand is according to formula (III), ##STR00014## wherein each of R.sup.18-R.sup.23 is independently selected from H, OR.sup.24, OCOR.sup.24, OC(O)OR.sup.24, OP(O)(OR.sup.24)(OR.sup.25), OP(O)R.sup.24R.sup.25, SR.sup.24, SCOR.sup.24, SC(O)OR.sup.24, SP(O)(OR.sup.24)(OR.sup.25), SP(O)R.sup.24R.sup.25, NR.sup.24R.sup.25, NR.sup.24COR.sup.25, NR.sup.24CO.sub.2R.sup.25 or NR.sup.26P(O)(OR.sup.24)(OR.sup.25), or —Y—R.sup.27; wherein each of R.sup.24, R.sup.25 and R.sup.26 are independently selected from H and C.sub.1-C.sub.10 alkyl, polyethylene glycol (PEG) and polyether groups; wherein Y is a chemical bond or a linker moiety; wherein Z is selected from a chemical bond, acetylene, disubstituted phenylene, naphthalene, anthracene, pyrene or similar chemically rigid spacer, and wherein R.sup.27 is a targeting moiety.

5. A compound according to claim 4, wherein the targeting moiety is a small molecule, peptide, protein or antibody.

6. A compound according to claim 5, wherein the targeting moiety has a high affinity for a target site such that when administered, the compound binds to the target site via the targeting moiety.

7. A compound according to claim 1, wherein L has formula IV ##STR00015## wherein R.sup.28, R.sup.29, R.sup.30, R.sup.31, R.sup.32, R.sup.33, R.sup.34 and R.sup.35 are independently selected from H, OR.sup.36, OCOR.sup.36, OC(O)OR.sup.36, OP(O)(OR.sup.36)(OR.sup.37), OP(O)R.sup.36R.sup.37, SR.sup.36, SCOR.sup.36, SC(O)OR.sup.36, SP(O)(OR.sup.36)(OR.sup.37), SP(O)R.sup.36R.sup.37, NR.sup.36R.sup.37, NR.sup.37COR.sup.36, NR.sup.37CO.sub.2R.sup.36 or NR.sup.38P(O)(OR.sup.36)(OR.sup.37); wherein R.sup.36, R.sup.37 and R.sup.38 are independently selected from H, C.sub.1-C.sub.10 alkyl, polyethylene glycol (PEG) and polyether groups.

8. A compound according to claim 7, wherein R.sup.29 and R.sup.33 are NH.sub.2 and R.sup.29 and R.sup.32 are H.

9. A compound according to claim 1, wherein the compound is configured to capture and retain an anionic target species.

10. A compound according to claim 9, wherein the anionic target species is a radiolabel anion.

11. A compound according to formula VI;
[T⊂M.sub.4L.sub.6].X.sub.n  (VI) wherein T is a target species retained within the cavity of the compound according to formula I; and wherein L, M, X and n are as defined in claim 1.

12. The compound of claim 11, wherein T is an anionic target species comprising a radioactive isotope selected from the group carbon-11 (.sup.11C), fluorine-18 (.sup.18F), copper-64 (.sup.64Cu), gallium-68 (.sup.68Ga), zirconium-89 (.sup.89Zr), gallium-67 (.sup.67Ga), technetium-99m (.sup.99mTc), indium-111 (.sup.111In), copper-67 (.sup.67Cu), rhenium-186 (.sup.186Re), rhenium-188 (.sup.188Re), lutetium-177 (.sup.177Lu), astatine-211 (.sup.211At), and actinium-225 (.sup.225Ac).

13. A pharmaceutical composition comprising the compound of claim 1 and a pharmaceutically acceptable excipient, carrier, vehicle or diluent.

14. A compound according to claim 1 or a pharmaceutical composition according to claim 13 for use as a diagnostic marker or label.

15. A compound according to claim 1 or a pharmaceutical composition according to claim 13 for use in vivo.

16. A method of preparing a compound according to formula I, the method comprising the steps: a. providing a first solution comprising transition metal ions M having 7 d-shell valence electrons, and a second solution comprising ligands according to formula II; b. mixing the first solution with the second solution; c. oxidising the transition metal ions M in the resulting compound, such that the structure of the complex becomes fixed.

17. A method of preparing a compound according to formula VI,
[M.sub.4L.sub.6].X.sub.n.T  (VI) the method comprising the steps: a. providing a compound according to formula I; b. providing a target species T; c. combining the compound according to formula I with the target species to form the compound according to formula VI.

18. The method of claim 17, wherein the compound according to formula I is provided in a solution.

19. The method of claim 17, wherein the target species is provided in a solution.

20. A method of using the compound according to formula VI for diagnosis comprising the steps; providing a compound according to formula VI; administering the compound to a subject; and monitoring the location and/or movement of the compound within the subject.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0090] Embodiments of the present invention will now be described, by way of non-limiting example, with reference to the accompanying drawings.

[0091] FIG. 1. (a) Chemical structure of inert Co.sup.III.sub.4L.sub.6 nanocarrier systems (only one ligand out of six shown for clarity); (b) X-Ray crystal structure of [ClO.sub.4⊂C1].sup.11+, where C1 is Co.sub.4L.sup.1.sub.6 and L.sup.1 is of formula VI below. Associated anions and solvent molecules removed for clarity; (c) X-Ray crystal structure of [ReO.sub.4⊂C2].sup.11+, where C2 is Co.sub.4L.sup.2.sub.6 and L.sup.2 is of formula VII below. Associated anions and solvent molecules removed for clarity; (d) X-Ray crystal structure of [CuCl.sub.4⊂Fe.sub.4L.sup.1.sub.6].sup.10+, where L.sup.1 is of formula VI below. Associated anions and solvent molecules removed for clarity; (e) X-Ray crystal structure of [PF.sub.6⊂Fe.sub.4L.sup.1.sub.6].sup.11+, where L.sup.1 is of formula VI below. Associated anions and solvent molecules removed for clarity.

[0092] FIG. 2. .sup.1H NMR spectra (500 MHz, 298 K) showing the encapsulation of ReO.sub.4.sup.− by C1 in D.sub.2O. (a) C1.12NO.sub.3 only; (b) C1.12NO.sub.3+NH.sub.4ReO.sub.4; (c) C2.12NO.sub.3 only; (d) C1.12NO.sub.3+NH.sub.4ReO (free cage signals within lightly dashed boxes, occupied cage signals within heavily dashed boxes).

[0093] FIG. 3. Radiochemical yield for .sup.99mTcO.sub.4.sup.− for encapsulation as a function of cage concentration (C1, black squares; C2, black circles).

[0094] FIG. 4. Stability of encapsulated complexes to a range of different anions and conditions.

[0095] FIG. 5: SPECT/CT images showing the comparison of “free” [.sup.99mTc]TcO.sub.4.sup.− uptake in naïve animals vs. nanocarrier encapsulated [.sup.99mTc]TcO.sub.4.sup.−.

DETAILED DESCRIPTION

[0096] While the making and using of various embodiments of the present invention are discussed in detail below, it should be appreciated that the present invention provides many applicable inventive concepts that can be embodied in a wide variety of specific contexts. The specific embodiments discussed herein are merely illustrative of specific ways to make and use the invention and do not delimit the scope of the invention.

[0097] To facilitate the understanding of this invention, a number of terms are defined below. Terms defined herein have meanings as commonly understood by a person of ordinary skill in the areas relevant to the present invention. Terms such as “a”, “an” and “the” are not intended to refer to only a singular entity, but include the general class of which a specific example may be used for illustration. The terminology herein is used to describe specific embodiments of the invention, but their usage does not delimit the invention, except as outlined in the claims.

[0098] The method of preparing the compositions of the invention combines the benefits of a dynamic self-assembly reaction, using bridging bidentate ligands and labile d.sup.7 Co.sup.II ions, and yet is able to create multi-component ensembles with covalent-like robustness following simple oxidation (“Fixing”) of the equilibrated species. We now demonstrate the real benefits of developing this method, showing that an inert, prototypical Co.sup.III.sub.4L.sub.6 species can function in vivo as a generic nanocarrier system for radiolabelled anions, thus providing numerous opportunities in nuclear medicine and beyond.

[0099] Validating the in vivo use of these Co.sup.III capsule systems using the encapsulation of small radio-enriched anions presented several attractive features as well as opportunities. From a nuclear medicine perspective, labelling through non-covalent encapsulation is an underdeveloped approach, it also has the potential to address some of the key operational difficulties in this field. In particular, the short half-lives of many commonly used emitting isotopes have hampered the development of labelled bio-probes because of the difficulty in establishing simple synthetic procedures for use in a clinical setting. In contrast, non-covalent encapsulation is not only operationally simple it is also very rapid. This method also separates the bio-targeting and labelling technologies, wherein the interior of the cage that is responsible for hosting the radiolabel is distinct from the cage periphery, which can readily be decorated with different bio-targeting groups. This gives the potential to use a single technology platform for diagnosing and treating various disease states by encapsulating different radiolabelled species inside the same core-cage structure that is externally functionalised with different groups.

Example 1

[0100] Initial investigations focused on the host-guest chemistry of the small tetrahedron [Co.sup.III.sub.4L.sup.1.sub.6].sup.12+, C1 (FIG. 1a), where L.sup.1 is of formula VII.

##STR00005##

[0101] This novel compound was prepared using the “assembly-followed-by-fixing” method described below starting with the known ligand L.sup.1. Furthermore, the synthesis from divalent Co.sup.II(NO.sub.3).sub.2.6H.sub.2O and oxidized with cerium ammonium nitrate (CAN), resulted in a dodeca-nitrate salt, C1.12NO.sub.3, obtained in 82% yield, that was directly soluble in water at concentrations in excess of 1 mM, thus negating the need to functionalize with aqueous-solubilizing groups. Molecular modelling also revealed that while the cavity size is relatively modest, 134 Å.sup.3, this species should be able to accommodate a number of small anions within its interior, which was indeed the case. It was found that more “hydrophobic” anions bind best, with PF.sub.6.sup.− giving the highest K.sub.a (68000 M.sup.−1) while perrhenate ReO.sub.4.sup.− also showed significant affinity (28000 M.sup.−1; for details of K.sub.a determination, see below). Perrhenate binding is particularly easy to observe in solution as it is in slow exchange on the .sup.1H NMR timescale, with separate signals observable for both the empty and occupied capsule at sub-stoichiometric ratios (FIG. 2). Other anions such as BF.sub.4.sup.−, ClO.sub.4.sup.− and even SO.sub.4.sup.2− also bind, albeit more weakly, with association constants ranging from 100 M.sup.−1 (SO.sub.4.sup.2−) to 7000 M.sup.−1 (ClO.sub.4.sup.−). Considering other biological anions that might compete with a radio-enriched anion, we also looked at PO.sub.4.sup.3− which showed no evidence of binding in D.sub.2O. The encapsulation of ClO.sub.4.sup.− has also been confirmed by X-ray crystallography (FIG. 1b) and more recently ReO.sub.4.sup.− as well.

[0102] Moving our attention to the encapsulation of radioenriched anions, while [.sup.18F]BF.sub.4.sup.− and [.sup.18F]PF.sub.6.sup.− present opportunities for Positron Emission Tomography (PET) imaging, we decided to focus on the gamma-emitting .sup.99mTcO.sub.4.sup.− anion as both isostructural ClO.sub.4.sup.− and ReO.sub.4.sup.− anions bind to C1. The use of enriched pertechnatate also does not require any prior radiochemistry as it is available direct from a source Mo generator, therefore ideal for validating this non-covalent approach to radiochemical encapsulation. We started by investigating the concentrations of C1 required to give quantitative radiochemical yield (>95%) of encapsulated .sup.99mTcO.sub.4.sup.− species, as this would determine the dose of cage compound administered to a subject in any subsequent in vivo imaging experiments. The best way to assess this was found to be using thin layer chromatography (TLC) on standard phase silica gel; while unencapsulated .sup.99mTcO.sub.4.sup.− elutes in water with the solvent front, the encapsulated (.sup.99mTcO.sub.4⊂C1].sup.11+ is retained on the baseline. The % radiochemical yield could therefore be easily calculated by comparing and the radioactive counts associated with solvent front and baseline of cut TLC plates using a Wizard automatic gamma counter (Perkin Elmer). Using this method it was found that full encapsulation could be achieved at 100 μM C1, with an EC.sub.50 (i.e. concentration required for 50% encapsulation) value of 14 μM (FIG. 3, black squares). These preliminary experiments also showed that, as anticipated, labelling the cage compound through encapsulation is extremely facile and rapid, with the radiochemical yields invariant of mixing time, with 5 minutes giving the same yield as hours.

[0103] Next the radiochemical yield was similarly assessed under a range of conditions designed to probe the stability of the encapsulated species (FIG. 4, diagonal bars). As can be seen the radiochemical yield is maintained at a good level (40-60%) in the presence of a large excess of competing NO.sub.3.sup.−, Cl.sup.− and PO.sub.4.sup.2−/3− anions relevant to in vivo use, neither does this diminish as a function of time. In contrast, when anions which are known to bind to the cavity of C1 (i.e. ClO.sub.4.sup.− and PF.sub.6.sup.−) were added to the sample of prior to mixing with .sup.99mTcO.sub.4, then the radiochemical yield reduced dramatically to trace levels. While neither ClO.sub.4.sup.−; and PF.sub.6.sup.− represent concerns from an in vivo perspective, both clearly demonstrate that a high radiochemical yield is a result of encapsulation rather than simple ion pairing with the cage periphery. These experiments also show that the equilibrated samples are stable, showing neither an increase or decrease in radiochemical yield as a function of time. In contrast when the same encapsulation experiment was carried out in serum, very good radiochemical yields were obtained immediately after mixing, however, after leaving the sample for just 30 minutes a value of over 90% had dropped to less than 30%, and after 24 hours this had diminished to trace levels. As the encapsulation process takes place on a much shorter timeframe, as evidenced by the rapid uptake and displacement of .sup.99mTcO.sub.4 in the presence of competing anions, we attribute this slower change to probable cage decomposition, and with it release of the enriched anion.

Example 2

[0104] The diamino substituted ligand, L.sup.2 of formula VIII, was synthesised in several steps, starting from 2-bromopyridine-N-oxide by Pd-catalysed coupling with 5-Bromo-2-(tributylstannyl)pyridine, followed by simultaneous reduction of nitro and N-oxide groups and then Ni catalysed homo-coupling (See supplementary materials for details). The difference in the ligand “strength” immediately became apparent from subsequent “assembly-followed-by-fixing” reactions, in particular the self-assembly part of this method required prolonged heating to equilibrate, furthermore, following oxidation we always obtained a mixture of C2 (Co.sub.4L.sup.2.sub.6).sup.12+ and the corresponding helicate, (Co.sup.III.sub.2L.sup.2.sub.3].sup.6+, which we attribute to probable kinetic trapping in the assembly phase of the reaction. Nonetheless, we could separate from the smaller species using size-exclusion chromatography, which gave a pure sample of C2.12NO.sub.3 in 14% yield.

##STR00006##

[0105] It is suggested that cage disassembly could be triggered by partial reduction of the Co.sup.III centers in the presence of biological reducing groups (e.g. disulfides etc.), which would then lower the energy barrier to subsequent ligand exchange with various species. Therefore, the stronger σ-donor ligand L.sup.2 may stabilize the higher oxidation state of the Co. In addition, if in situ reduction of the cage compound was not occurring, rather just decomposition caused by direct ligand substitution with the Co.sup.III state, then the increased donor strength of the L.sup.2 ligand set may also aid in vivo stability.

[0106] Returning to radiochemical labelling experiments, substitution of the cage structure with 12 amino groups had an unforeseen yet very positive impact on .sup.99mTcO.sub.4 encapsulation. In the absence of any competing anions it was observed that the concentration of cage required to achieve >95% radiochemical yield was reduced nearly 50 fold to 1.9 μM, with an EC.sub.50 value of just 0.05 μM (FIG. 3, black circles). Furthermore, repeating the same anion stability experiments as originally tested with C1, also appeared to show that this increase in affinity is specific to .sup.99mTcO.sub.4.sup.−, as all the other anionic species appear to reduce the radiochemical yield less significantly (FIG. 4, C2 24 hrs). When the same radiochemical labelling experiment was performed in the presence of serum, no real drop in radiochemical yield was observed after even 24 hours, indicating that the amino groups have the desired effect of increasing the cage's kinetic robustness.

[0107] In light of the improved stability in serum, as well as the lower dose required to fully encapsulate .sup.99mTcO.sub.4.sup.−, C2 was selected to take forward for Single-Photon Emission Computed Tomography (SPECT) imaging experiments. Furthermore, MTT assays also revealed low to moderate cytotoxicity (EC.sup.50=10.6 μM, see below) suggesting that this compound should be safe for in vivo use at the concentration levels needed for non-covalent labelling. The biodistribution of free .sup.99mTcO.sub.4 is well understood, with this small anion known to localise in the thyroid, lungs, stomach and liver. It can be used to study disruption of the blood brain barrier as it will not accumulate significantly in the brain if the blood brain barrier is intact. It binds weakly to plasma proteins and clears in ca. 30 mins. It is excreted in both urine and faeces hence can be detected in both kidneys and small intestine prior to excretion (FIG. 5a). In contrast, when the same dose of .sup.99mTcO.sub.4.sup.− was administered in the presence of 1 mg/kg C2.12NO.sub.3, it was clearly observable that the cage compound markedly affects the bio-distribution of this label (FIG. 5).

[0108] Discussion

[0109] What is important to take away from these experiments is not where the .sup.99mTcO.sub.4.sup.− anion localizes in the presence of C2, but rather the significant change compared to the free anion alone. The use of a non-covalent encapsulation approach also provides compelling evidence that the capsule remains intact during imaging, as disassembly would remove the well-defined cavity and with it the significant affinity that is required to keep the anion associated in the presence of vast excess of many other competing biological cations.

[0110] Accordingly, the exemplified radiolabels are stable in biological conditions for the time-frames of typical radiolabel diagnostic techniques, and can be created readily in a clinical, rather than laboratory setting.

Synthesis of Ligands

[0111] Fragments 5-bromo-2,2′-bipyridine and 5-Bromo-2-(tributylstannyl)pyridine were synthesised according to a literature procedure.

5-(5-bipyrdin-2,2′-yl)-2,2′-bipyridine (L.SUP.1.)

[0112] ##STR00007##

[0113] A solution of nickel chloride (0.174 g, 1.34 mmol) and triphenylphosphine (1.34 g, 5.10 mmol) in N,N-dimethylformamide (12.0 mL) was heated at 50° C. for 0.5 h under N.sub.2. The resulting blue suspension was treated with zinc dust (0.0903 g. 1.38 mmol) to produce a red brown suspension after 0.5 h, to which 5-bromo-2,2′-bipyridine (0.299 g, 1.27 mmol) in N,N-dimethylformamide (6.0 mL) was added and the mixture stirred at RT for 3 d. To this, Na.sub.2EDTA (0.25 M, 60 mL, 15.0 mmol) was added and the aqueous layer washed with dichloromethane (5×100 mL) and brine added (20 mL). The combined organic layers were dried over MgSO.sub.4, the solvent removed in vacuo and purified by silica flash column (1% triethylamine, 1% methanol in dichloromethane). This was recrystallised in hot acetonitrile to give white plate crystals. Yield=0.126 g (64%).

[0114] m.p. 233-235° C. .sup.1H NMR (600 MHz, CDCl.sub.3): δ 9.03 (dd, J=2.4, 0.8 Hz, 2H, H.sub.G), 8.75 (d, J=4.8 Hz, 2H, H.sub.A), 8.58 (dd, J=8.3, 0.8 Hz, 2H, H.sub.E), 8.50 (d, J=8.0 Hz, 2H, H.sub.D), 8.13 (dd, J=8.2, 2.4 Hz, 2H, H.sub.F), 7.89 (t, J=7.7 Hz, 2H, H.sub.C), 7.38 (dd, J=7.5, 4.8 Hz, 2H, H.sub.B). .sup.13C NMR (151 MHz, CDCl.sub.3): δ 155.8, 155.6, 149.3, 147.5, 137.0, 135.2, 133.1, 124.0, 121.2, 121.2. HR-ESI: m/z 311.12890 (predicted [M+H].sup.+=311.12912), 333.11180 (predicted [M+Na].sup.+=333.11107).

5′-bromo-4-nitro-(2,2′-bipyridine)-N-oxide

[0115] ##STR00008##

[0116] 2-bromo-4-nitropyridine N-oxide (4.039 g, 18.4 mmol), palladium tetrakis(triphenylphosphine) (1.04 g, 0.901 mmol) and 5-Bromo-2-(tributylstannyl)pyridine (8.29 g. 18.5 mmol) were combined in toluene (80 mL) and heated at 100° C. for 24 h under N.sub.2. The mixture was allowed to cool to room temperature, diluted with dichloromethane (30 mL) and filtered. The solvent was removed from filtrate in vacuo. The crude product was recrystallised from hot ethanol and then purified by silica flash column (gradient: dichloromethane to dichloromethane 4% diethylether) to give a colourless powder. Yield=3.07 g (57%).

[0117] .sup.1H NMR (500 MHz, Chloroform-d) δ 9.19 (d, J=3.3 Hz, 1H, H.sub.F), 8.90 (dd, J=8.7, 0.7 Hz, 1H, H.sub.A), 8.84 (dd, J=2.4, 0.7 Hz, 1H, H.sub.C), 8.35 (d, J=7.2 Hz, 1H, H.sub.D), 8.08 (dd, J=7.2, 3.3 Hz, 1H, H.sub.E), 8.01 (dd, J=8.6, 2.4 Hz, 1H, H.sub.B). .sup.13C NMR (128 MHz, Chloroform-d) δ 151.13, 147.58, 146.03, 142.65, 142.25, 139.47, 126.15, 122.94, 122.54, 119.19. HR-ESI: m/z 295.96780 (predicted [M+H].sup.+=295.96653), 317.9480 (predicted [M+Na].sup.+=317.94847).

5′-bromo-(2,2′-bipyridin)-4-amine

[0118] ##STR00009##

[0119] 5-bromo-4′-nitro-2,2′-bipyridine N-oxide (1.66 g, 5.60 mmol) and iron (1.67 g, 29.9 mmol) were combined in acetic acid (50 mL) and heated at 100° C. for 3 h under N.sub.2. The mixture was allowed to cool to room temperature and a solution of NaOH (10.5456 g, 0.268 mol), Na.sub.2EDTA (17.564 g, 60.1 mmol) in water (50 mL) and ammonium hydroxide solution (30%, 100 mL) was added slowly. The mixture was then extracted with chloroform (3×150 mL) and the combined organic phases were dried over MgSO.sub.4. The solvent was removed in vacuo and the crude product was recrystallised from hot toluene to give colourless needle crystals. Yield=1.03 g (74%).

[0120] .sup.1H NMR (600 MHz, Chloroform-d) δ 8.68 (d, J=2.4 Hz, 1H, H.sub.G), 8.32-8.25 (m, 2H, H.sub.A,E), 7.91 (dd, J=8.5, 2.4 Hz, 1H, H.sub.F), 7.64 (d, J=2.5, 0.5 Hz, 1H, H.sub.D), 6.57 (dd, J=5.5, 2.4 Hz, 1H, H.sub.B), 4.24 (s, 2H, H.sub.C). .sup.13C NMR (126 MHz, Chloroform-d) δ 155.98, 155.02, 153.74, 150.07, 139.56, 122.62, 121.06, 110.00, 106.88. HR-ESI: m/z 249.99910 (predicted [M+H].sup.+=249.99744).

5′-(4′-amino-(2,2′-bipyridin)-5-yl)-(2,2′-bipyridin)-4-amine (L)

[0121] ##STR00010##

[0122] Nickel chloride (157 mg, 1.21 mmol) and triphenylphosphine (218 mg, 1.92 mmol) were combined in in N,N-dimethylformamide (10 mL) and heated at 50° C. for 1 h under N.sub.2. The resulting blue suspension was treated with zinc dust (155 mg, 2.37 mmol) to produce a red brown suspension after 0.5 h, to which 5′-bromo-(2,2′-bipyridin)-4-amine (298 mg, 1.20 mmol) in N,N-dimethylformamide (5 mL) was added and the mixture stirred at RT for 3 d. A solution of Na.sub.2EDTA (3.52 g, 12.0 mmol) and NaOH (1.89 g, 48.0 mmol) in ammonium hydroxide solution (30%, 20 mL) was added to the mixture and it was heated at 50° C. for 24 h. The colourless precipitate was collected by filtration and dissolved in tetrahydrofuran, filtered and the solvent removed in vacuo. The crude product was recrystallised from hot acetonitrile to give colourless powder. Yield=151 mg (37%).

[0123] 342° C. decomposition .sup.1H NMR (500 MHz, DMSO-d) δ 9.09 (dd, J=2.4, 0.8 Hz, 1H, H.sub.G), 8.45 (dd, J=8.3, 0.8 Hz, 1H, H.sub.E), 8.33 (dd, J=8.3, 2.4 HZ, 1H, H.sub.F), 8.15 (d, J=5.5 Hz, 1H, H.sub.A), 7.70 (d, J=2.3 Hz, 1H, H.sub.D) 6.56 (dd, J=5.5, 2.3 Hz, 1H, H.sub.B), 6.21 (s, 2H, H.sub.C). .sup.13C NMR (126 MHz, DMSO-d) δ 155.67, 155.20, 154.82, 149.37, 146.97, 134.83, 132.03, 120.36, 109.14, 105.61. HR-ESI: m/z 341.15240 (predicted [M+H].sup.+=341.15092), 363.1339 (predicted [M+Na].sup.+=363.1334).

Synthesis of [Co.SUB.4.L.SUB.6.].SUP.12+ species

[0124] ##STR00011##

[0125] Cobalt nitrate hexahydrate (0.0211 g, 72.5 μmol) and L.sup.1 (0.0337 g, 109 μmol) were suspended in a mixture of degassed water-acetonitrile (9:1, 5.0 mL) and heated at 50° C. for 15 h under N.sub.2. The reaction was cooled to room temperature before cerium(IV) ammonium nitrate (0.0598 g, 109 μmol) in acetonitrile (6.6 mL) was added via syringe pump (6 μL min.sup.−1, 18 h). Dilution with acetonitrile (8.0 mL) was required to precipitate the product, which was isolated by filtration onto celite and washed with acetonitrile. This was eluted with water and the solution was freeze-dried to give the compound as a yellow solid. Yield=0.0420 g (82%).

[0126] .sup.1H NMR (500 MHz, D.sub.2O): δ 8.97 (d, J=8.1 Hz, 12H, H.sub.D), 8.81 (d, J=8.4 Hz, 12H, H.sub.B), 8.62 (dd, J=8.1, 7.0 Hz, 12H, H.sub.C), 7.91 (d, J=8.4 Hz, 12H, H.sub.F), 7.83 (dd, J=7.0, 6.0 Hz, 12H, H.sub.B), 7.62 (d, J=6.0 Hz, 12H, H.sub.A), 7.50 (s, 12H, H.sub.G). .sup.13C NMR (126 MHz, D2O): δ 155.8, 154.6, 151.7, 149.2, 144.5, 144.5 138.2, 131.8, 127.9, 126.5. .sup.1H DOSY NMR (500 MHz, D.sub.2O): D=2.17×10-6 cm.sup.2 s.sup.−1; calculated hydrodynamic radius=11.3 Å. ESI-MS (m/z): 885 (3+), 648 (4+), 344 (7+), 293 (8+).

##STR00012##

[0127] Cobalt nitrate hexahydrate (17.2 mg, 59.1 μmol) and L.sup.2 (30.1 mg, 88.4 μmol) were suspended in a mixture of degassed water-acetonitrile (9:1, 4 mL) and heated in a microwave reactor for 2 h at 80° C. The reaction was cooled to room temperature before cerium(IV) ammonium nitrate (501 mg, 91.4 μmol) in acetonitrile (4 mL) was added via syringe pump (4.2 μLmin.sup.−1). Dilution with acetonitrile (40 mL) was required to cause precipitation, this was collected by filtration onto celite and washed with acetonitrile. The crude product was eluted with water and freeze dried to give an orange solid. It was then purified by size exclusion chromatography on sephadex LH-20 gel in water and freeze-dried to give an orange solid. Yield=6.1 mg (14%).

[0128] In an alternative method, L.sup.2 (28.4 mg, 83.4 μmol) was suspended in a solution of C1 (20.2 mg, 7.04 μmol) in degassed acetonitrile (9:1, 3.9 mL), an aqueous solution of cobalt nitrate hexahydrate (2 mM, 141 μL) was added, the mixture sparged with N.sub.2 and heated at 75° C. in a sealed vial. The mixture was allowed to cool, filtered through celite and the retentate washed with water (5 mL). The filtrate was freeze-dried, redissolved in water (2 mL) and purified by size exclusion chromatography on sephadex LH20 and freeze-dried to give an orange solid. Yield=11.9 mg, 56%). Characterisation identical to previous method.

[0129] .sup.1H NMR (600 MHz, D.sub.2O) δ 8.55 (d, J=8.4 Hz, 12H, H.sub.D), 7.92 (d, J=2.7 Hz, 12H, H.sub.C), 7.78 (dd, J=8.3, 1.8 Hz, 12H, H.sub.B), 7.31 (d, J=1.6 Hz, 12H, H.sub.F), 6.99 (d, J=7.0 Hz, 12H, H.sub.A) 6.81 (dd, J=7.0, 2.7 Hz, 12H, H.sub.B). .sup.13C NMR (126 MHz, D.sub.2O) δ 158.37, 156.49, 153.36, 148.46, 143.17, 137.28, 124.58, 114.68, 111.97, 109.99. .sup.1H DOSY NMR (500 MHz, D.sub.2O): D=2.14×10-6 cm.sup.2 s.sup.−1; calculated hydrodynamic radius=11.5 Å. ESI-MS ESI-MS (m/z): 369.5 (7+), 315.6 (8+).

[0130] “Cold” Anion Binding Experiments: Determination of Association Constants (K.sub.B)

[0131] K.sub.B for cage and guest combinations determined through .sup.1H NMR titration in unbuffered D.sub.2O. A solution of compounds C1 or C2 with a guest compound was titrated into a solution of C1 or C2, thereby maintaining a constant concentration of C1 or C2.

[0132] Fast Exchange Determination

[0133] All observable shifts in the .sup.1H NMR spectra had their peak position plotted against concentration of guest. A global non-linear curve fitting function was then applied to the combined plots using the 1:1 binding model given by:

[00001]$y=y_0+\Delta y\left(\frac{\left(1+K_a\left(P+x\right)\right)-\sqrt{{\left(1+K_a\left(P+x\right)\right)}^2-4K_a{K}_{a}Px}}{2K_{a}P}\right)$ [0134] y.sub.0=peak position with no guest [0135] Δy=peak position with 100% bound [0136] x=concentration of guest [0137] P=concentration of cage

Origin Function=y=y.sub.0+DY*((1+Ka*(P+x))−sqrt(((1+Ka*(P+x)){circumflex over ( )}2)−4*Ka*Ka*P*x))/(2*Ka*P)

Slow Exchange Determination

[0138] Concentration of guest⊂C1 or guest⊂C2 was calculated from .sup.1H NMR integrals and plotted against guest concentration. A non-linear curve fitting function was then applied to the plot using the 1:1 binding model given by:

[00002]$y=\frac{\left(x+P+\left(\frac{1}{K_a}\right)\right)+\sqrt{{\left(x+P+\left(\frac{1}{K_a}\right)\right)}^2-\left(4\mathrm{Px}\right)}}{2}$ [0139] x=concentration of guest [0140] y=Concentration of host guest complex [0141] P=concentration of host

Origin function=((x+h+(1/K))−((x+h+(1/K)){circumflex over ( )}2(4*h*x)){circumflex over ( )}(½))/2

[0142] The resulting binding constants of some example anions are provided in Table 1 below:

TABLE-US-00001 TABLE 1 Binding constants of various anions in unbuffered D.sub.2O. Errors are estimated to be less than 10%. Guest Formula C1 (M.sup.−1) C2(M.sup.−1) Sulfate [SO.sub.4] 100 Tetrafluoroborate [BF.sub.4] 500 Perchlorate [ClO.sub.4] 7100 21000 Hexafluorophosphate [PF.sub.8] 61000 Perrhennate [ReO.sub.4] 20000 22000

[0143] Cage Cytotoxicity Experiments: MTT Assays

[0144] HeLa cells were plated onto a 96-well plate in Dulbecco's Modified Eagle Medium (DMEM) (10% in PBS) (volume) at approximately 8000 cells/mL. The plate was incubated for 20 h to allow equilibrium. The DMEM was removed and replaced with fresh DMEM (100 μL) containing the compound being investigated, at the appropriate concentration; each concentration was added to seven wells. The plate was incubated for another 20 h before the removal of medium and washing of the cells with PBS. Then a 1.2 mM solution 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-2H tetrazolium bromide (5 mg) in media/PBS (7/3, 10 mL) was prepared and 100 μL added to each well. This was incubated for 3 h, at which point crystals were apparent. A solubilising solution (prepared: 5 mL Triton-X 100, 45 mL Isopropanol, 1 drop HCl (12M)) (100 μL) was added and the plate shaken on a plate shaker overnight to ensure complete dissolution of crystals. UV-visible absorbance measurements were carried out at 490 nm. For each concentration of the added compound the absorbances were averaged (discarding those more than two standard deviations from the mean), The absorbances were then normalised against that of the control cells. The resulting concentrations are shown in Table 2 below.

TABLE-US-00002 TABLE 2 Cytotoxic concentration for ligands C1 and C2 Cytotoxicity values C1 (μM) C2 (μM) Cytotoxic concentration 31.4 10.6 50% (CC.sub.50)

[0145] Pertechnetate Binding Experiments

[0146] .sup.99mTcO.sub.4 was eluted from a 2.15 GBq Ultra-TechneKow™ DTE .sup.99Mo/.sup.99mTc generator (Mallinckrodt. UK) in 4 mL 0.9% saline solution. .sup.99mTcO.sub.4.sup.− (1 μL, 1 MBq) was then added to C1 or C2 solutions (100 μl) and incubated at room temperature for 5 minutes. Pertechnetate encapsulation was assessed by silica gel thin layer chromatography, wherein eluting with DI water separates free .sup.99mTcO.sub.4.sup.−, which runs with the solvent front, while bound .sup.99mTcO.sub.4 is complex remains on the baseline. The proportion of bound and free .sup.99mTcO.sub.4 was determined by cutting the eluted TLCs into strips and independently analysing using a Wallac 1480 Wizard 3″ automatic-gamma-counter (Perkin Elmer, USA). All reactions were carried out in DI water. The results are shown in FIG. 3.

[0147] Radiochemical stability was tested under a range of different salt conditions. In a typical experiment, .sup.99mTcO.sub.4.sup.− (1 μL, 1 MBq) was added to C1 or C2 solutions (50 μl, 132 μM) and incubated at room temperature for 5 minutes. Salt solutions (50 μl, 0.1 mM) were added and incubated at room temperature for 24 hours with aliquots taken at the stated time points and then analysed by the method described above (e.g. running on TLC followed detection using gamma counter).

[0148] In Vivo Imaging Experiments

[0149] Mice were injected iv with 25-35 MBq of .sup.99mTcO.sub.4 and 19 MBq of caged .sup.99mTcO.sub.4 under anaesthesia before being transferred to a temperature-controlled imaging bed and attached to an anaesthetic facemask (Minerve, France). .sup.99mTcO.sub.4.sup.− SPECT images (shown in FIG. 4) were acquired at ˜40 minutes pi for 30 min under anaesthesia, whereas .sup.99mTcO.sub.4.sup.BNH.sub.2 SPECT images were acquired at 20 minutes pi for 100 min. In both cases SPECT was followed by CT scan (240 projections with 1 s exposure to 55 kVp X-rays). SPECT and CT Images (FIG. 5) were reconstructed with an iterative algorithm (HiSPECT, Scivis GmbH, Germany) and with exact cone beam Filtered Back Projection (VivoQuant, inviCRO LLC, USA), respectively.